65 PDF

d ü s t r i a l e d it io n SEPTEMBER, 19 4 4
■: :
m m E w M T m E m m m m m G m m m r n
* H e a t T ra n s fe r O p e ra tio n s
For Y o u r Im p o rta n
Thoroughly experienced in the problem s of heat transfer in
the ^chem ical industry, D evine Engineers are prepared to
design units specifically engineered for any particular heat-
exchanging operation. Built in an up-to-date plant b y expert
p o r t e r workmen with every modern facility at their disposal, Devine
—— Heat Exchangers achieve an unusually high degree of effi-
Ciency and mechanical simplicity. Low initial investment and
minimum maintenance costs are characteristic of Devine
Equipment. If you have a heat transfer problem, Devine En
gineers will help you solve it.
i v isio n
I __________
W iiM i
Portion of new 100-octane aviation gasoline
refinery of The Standard O i l C om pa ny ( O h io ) ,
located on site of the original plant built b y
J. D. Rockefeller 7 5 years ago in Cleveland.
EDITO R: W A L T E R J. M U R P H Y IS S U E D S E P T E M B E R 11, 1944 * V O L. 36, NO. 9 • C O N S E C U T IV E NO. 17

Associate Editor: F. J . V A N A N T W E R P E N
E D I T O R I A L S ...................................................................... 777
Assistant to Editor: N . A . P A R K I N S O N Dehydration of Commercial G ases b y S o lid A d so rb e n ts ...........................
.............................. R. G . C appell, E. G . Hammerschmidt, and W . W . Deschner 779
M anuscript Editor: H E L E N K . N E W T O N
Sulfuryl C hloride in O rg a n ic C h e m is t r y .................................... H erbe rt C . B ro w n 785
M ake-up Editor: B E R T H A R E Y N O L D S V ulcanization of G R - I F. P. Baldw in, L. B. Turner, and R. L. Z a p p 791
M anuscript Assistant: S T E L L A A N D E R S O N M e cha n ical Properties of Films from A m y lo s e , A m y lo p e c tin , and W h o le Starch
Triacetates..................................... ... Roy L. W h istle r and G . E. H ilb e rt 796
Soyb e an Protein P r o d u c t io n ................ P. A . Belter, A . C . Beckel, and A , K. Smith 799
Production of 2,3-Butylene G ly c o l from W o o d H yd ro ly za te s . . . . David Perlman 803
A d v is o r y B o a rd
V a p o r-L iq u id Equilibria of C lose-B oilin g Binary H yd ro c a rb o n M ixtures . . . . . .
W. L. Ba d g e r H. R. M urdock A. R. Richards and E. H argreaves 805
So a p -B o ilin g Equilibria for Sodiu m Stearate. The N e w Phase, Kettle W a x . . . .
H. E. Ba r n a r d A . S. Ric h a r d s o n James W . M c B a in , Kenneth G ardner, and Robert D. V o i d 808
W, H. D o w W . A . SCHMIDT Saccharification of W heat b y Fungal A m y la se s for A lc o h o l P r o d u c t io n .................
M artin Roberts, Stephen Laufer, Earl D. Stewart, and Leonard T. Saletan 811
G a st o n D u Bo is R. N. S h r e v e
Reactivity of S o lid F u e l s .................................................................A . A . O r n in g 813
C. F. Kettering L. V . S teck Solvent Dehydration b y Salting O u t ..............H . P. M e issn e r and Charles A . Stokes 816
Preparation and Purification of H y d ro u s O x id e Sols b y Ion Exchangers . . . . . .
W. S. La n d is C. M . A . S tine
J ohn W . Ryznar 821
O. E. M a y E. C. S u l l i v a n R h eological Properties of A sp h a lt .................................................................
R. N . Traxler, H . E. Schw eyer, and J. W . Romberg 823
C. S. M iner E. R. W eidlein
Thermodynamics of G aseou s Paraffin Kenneth S. Pitzer 8 2 9
Carotene Content of A lfalfa . . . Ralph E. Silker, W . G . Schrenk, and H . H . King 831
Calcium M e t a p h o s p h a t e G . L. Frear, E. F. Deese, and J. W . Lefforge 835
Rate of S e d im e n t a t io n ............................... ........................H a ro ld H . Steinour 840
Printing Inks from C ollo id a l Dispersions of So y b e a n P r o t e i n ...............................
A lfre d F. Schmutzler and Donald F. Othm er 84 7
The American Chemical Socie ty assumes no re
sponsibility for the statements and o p in io n s ad Tartrates from G rape W a s t e s ..............................................................................
vanced by contributors to its publications.
J. R. Matchett, R. R. Legault, C . C . Nimmo, and G . K. N otte r 851
Correlating V a p o r C om positions and Related Properties of S o l u t i o n s .....................
Copyright 1 9 4 4 by Am erican Chem ical Society
Donald F. O thm er and Roger G ilm ont 858
3 1 ,4 0 0 cop ies o f this issue printed Specific Heats at L o w Temperatures of Titanium and Titanium Carbide . . K. K. Kelley 865
A u g u s t's H e a d l i n e s .............................................................................................8 6 7
I. & E. C. Reports, 5. A s W e See It, 51. L ast-M in ute Flashes . . (A d v t. Sect.) 124
i-'-ijP.ublished by the Am erican Chem ical Socie ty at Easton, Pa. Editorial O ffice: 18th of the preceding month and for the A nalytical Edition not later than the 30th
1155 16th Street, N . W ., W a shington 6,.D. C./ telephone, R epublic 5301/ cable, of the preceding month. Claim s for missing numbers w ill not be a llo w e d (1 ) if re
Jiechem (W ashington). Business O ffice: Am erican Chem ical Society, 1 1 5 5 16th ceived more than 6 0 days from date of issue (o w in g to the hazards o f wartime delivery,
Street,/N. -W ., W ashing, i 6 , D. C . A d v e rtisin g Office: 3 3 2 W e st 4 2 n d Street, no claims can be h on ore d from subscribers outside o f N o rth Am erica), ( 2 ) if loss w a s
N e w Y o rk 18, N . V.j telephone, Bryant 9 -4 4 3 0 . due to failure of notice o f change o f address to be received before the dates specified
M L y S i as second-class matter at the Post O ffic e at Easton, Pa., under the A c t of in the p reced ing sentence, o r ( 3 ) if the reason for claim is “ missing from files".
<ywrch 3 ,1 3 7 9 , as 2 4 times a year — Industrial Edition monthly o n the 1st, Analytical A n n u a l subscription — Industrial Edition and A nalytical Edition sold o n ly as a
edition monthly on the 15th. A cce ptan ce for mailing at special rate o f postage p ro unit, members $ 3 .0 0 , nonmembers $ 4 .0 0 . Postage to countries not in the Pan-
vided for In Section 1 1 0 3 , A c t o f O c t o b e r 3, 1 9 1 7 , authorized J u ly 1 3 , 191 8 . Am erican U n io n $2.25/ C an ad ian postage $ 0 .7 5 . S in g le cop ies — current issues,
of Jjem,ttan5cs anc* orders for subscriptions and for single copies, notices o f changes Industrial Edition $ 0 .7 5 , A nalytical Edition $ 0 .5 0 ; back numbers, Industrial Edition
or address and n e w professional connections, and claims for missing numbers should be $0.80 , A nalytical Edition prices o n request; special rates to members.
D C r u ^ merican Chem ical Society, 1 1 5 5 16th Street, N . w ., W a sh in gto n 6, The Am erican Chem ical Society also publishes C hem ical a n d E n gine ering News,
• Changes of address for the Industrial Edition must be received on or before the C hem ical Abstracts, and Jo u rn a l o f the A m e rica n C h e m ica l Society. Rates on request.
Vol. 36, No 9
ou tried
for RAW MATERIAL

VARIATIONS?
O U can d o better than sim ply blame raw
Y material variations o n the war. Many p r o c
essors are actually o v e r co m in g them — m ain
ty p e s. . . and ch eck ed by tests and experim ents in
the w o r ld ’s largest filtration laboratory. T h e re ’ s
n o o b lig a tio n fo r this service.
taining the quality o f their finished products,
getting lo n g e r filter cycles and faster filtering— For m o re in form ation and fo r sam ples and
p rices o f J -M Filter A id s, w rite
sim ply through the use o f a different filter aid
in their present equipm ent. to J o h n s - M a n v ille , 22 East
4 0 th S tre e t, N e w Y o r k 16,
John s-M anville m akes nine grades o f Celite N ew Y ork .
Filter Aids,^ each d esign ed fo r a specific filtra
tion con d ition . O n e o f these m igh t w e ll be the
answ er to your wartim e p ro d u ctio n p rob lem s.
I f f l - F ILTER -CEL HYFLO S U P E R -C E L
T o h elp you find out, J -M offers the services C EL IT E N O . 5 0 5 C E L IT E N O . 501
o f its Filtration Engineers, w h o can recom m en d S T A N D A R D S U P E R -C E L C E L IT E N O . 503
fig h t fiber aid fo r you r op era tin g c o n C E L IT E N O . 5 1 2 C EL IT E N O . 535
C EL IT E N O . 5 4 5
ditions. T h ese recom m en d ation s are based on
years o f experien ce w ith raw materials o f all Celite is available for prompt shipment
JOHNS-MANVILLE FILTER AIDS
give m aximum flow rates with required clarity on every filtration service
Gostunesice a n d 9 n d u it> u i
Economic Problem No. 2. I. & E. C. editors have Problems ever will be recurring in both markets, and there
written voluminously on the problem of cartels, describ are other materials that must be considered in the same
ing them as our se.cond major economic problem (the first way. When the Dutch regain control of quinine, we must
is the national debt). We have commented on cartels correlate our activities in antimalarials. There will be a
in the Reports of January, February, April, and June battle between silk and synthetics, and even air com
of this year, pointing out that mere castigation of cartels merce will require constant negotiation between countries.
by the Department of Justice does not serve as a solution Either these agreements will be approved, or we will not
to the problem. enjoy the benefits of foreign trade.
First and foremost is a simple fact—namely, there are We are at the crossroads, and a clear-cut workable plan
problems in international trade which can be solved only must be promulgated in Washington and agreed to, at
by agreements between oompanies, in much the same way least in .major principle, by the government departments
that international difficulties between governments are and agencies concerned. There appears to be little op
solved by agreements. A second fact to be remembered position to the idea that appropriate government spokes
is that other countries not only approve of cartels but men participate in and even initiate necessary interna
freely foster their development. The two nations that tional agreements. But industry should not be viewed
we are now fighting specialize in this type of thinking, but as a suspicious character and relegated to a bench in the
cartels are not ipso facto signs of fascism, for several demo anteroom while negotiations are in progress. Is it hope
cratic nations adhere to a similar economic philosophy. less to assume that industry and government officials
Projection of our thoughts into postwar serves to em can cooperate constructively? We believe they . can,
phasize the problem. Suppose the German cartel in dye- and much good can evolve from such planning if suspicion
stuffs and an American concern are trying to sell materials is ruled out and certain irreconcilable individuals are no
in the South American market. The German cartel, longer permitted cartc-blanche veto power.
blessed by its government, 11111 have no difficulty in forc
ing the American concern out of business, if the American Chemical Profits. Time was when popular belief held
is unable by government edict to partake in agreements to the proposition that war was a profitable adventure for
about markets, prices, and patent rights. Economic industry. And because chemicals necessarily played an
conditions in the conquered countries will be such that all important role in modern warfare, the chemical industry,
their energies will be devoted to developing world markets; it was alleged, was more than slightly interested in the
if judicious horse trading is not to be allowed, we may as periodic misery of mankind. Cold statistics are available
well step out of foreign trade immediately. But let us to show that the chemical industry for even the most selfish
realize that this question is of greater importance than the of reasons (profits) desires peace, not war. This premise is
mere establishment of a certain amount of foreign trade. supported not only by results obtained in the period following
The lifeblood of the United States in peace and war is World War I but is confirmed by the latest report of the
dependent, to some degree, upon access to certain raw Securities and Exchange Commission reporting on net profits
materials controlled by foreign countries. of forty-one chemical companies in 1942 and for the six pre
The State Department recognizes these facts; the ceding years.
United States has entered into an agreement on petro Net profit in 1942 was reported at 8.7%, as contrasted with
leum and soon will begin talks on rubber. We applaud the 1936, a depression year, when chemical manufacturers showed
action of Cordell Hull in recognizing the practical implica a profit of 21.9%. If abnormal war-swollen profits were ever
tions of this problem and in doing something constructive an incentive leading nations to adjudicate their differences
about such major items in world trade. by the sword, that incentive has disappeared in the face of
More remains on the horizon, however. We will not taxes which make war an unprofitable venture for victor,
have finished when we sign treaties covering oil and rubber. vanquished, and “ munition makers” . (Continued, on page 8)
/ In in tcfip ^ ietcd iu e m o n th ly d iy e d t fjO i ch em ii/ d , ch e m ica l e n fin e e id , a n d

e x e c n tio e d in th e ch e m ica l p w d n cin tj. a n d ch e m ica l con d u -m in y incLudhued
5
INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No. 9
12 th o u san d miles
r o n i c h e a t i n g c o n c e n t r a t e s h. f . * p o w e r
.I
three times the electronic pow er it takes to send short
wave radio messages around the w orld concentrated into a
24-inch stack o f ven eers! In this actual application, a large
T herm ex high frequen cy unit bonds the veneers w ith heat-
setting adhesives into p lyw ood panels in very few minutes.
T od a y many industries are using T herm ex h. f. to heat non
con du ctin g materials in seconds or minutes instead o f hours.
Its speed, u n iform ity and con trolla bility m ay be an answer
to your processing problem s, you r gateway to new products,
new markets, new econom ies. As the p ion eer in high frequency
dielectric heating, this com pa n y offers unequalled application
engineering experience in many fields . . . cou p led with the >
advantages o f invaluable research and highly developed equip-
THE FIRST INDUSTRIAL * HIGH FREQUENCY ment. Call on us fo r p rom pt, practical help. W rite T h e Girdler
DIELECTRIC HEATING EQUIPMENT C orporation, T h erm ex D ivision, Louisville 1, Kentucky.
September, 1944 INDUSTRIAL AND ENGINEERING CHEMISTRY 7
HARDE
v NEED MORE
O u r BOTTLENEC/f
P R O D U C T IO N !
Can You Improve Your Drying Process?

Or— can w e help y o u im prove it? L in k -B e lt’s D ryer Engineers are w ell inform ed about
drying processes. T h e y have found better w ays to dry m a n y food and chem ical products.
T h e y ’v e solved every one of the problem s those fellow s at the top o f this page are shouting
about, and a lot more.
It Might Pay You

to Investigate
W h y not write L in k -B elt today?
T e ll us w hat y ou are drying, and
Here’s W hy the Roto-Louvre Dries So Fast, So Safely how, and w hat your requirements
are. W e can tell you a good deal
WET A IR EXHAU ST
M A TERIA L about m ost drying processes, and
FROM
-\B IN r [ we can tell you ju st w hat w e can
do for you. W e can run samples
of your product in our drying
process laboratory, in full-size
equipm ent. I f w e can’t help you,
© M ATERIAU
MATERIAL! w e’ll tell you so, frankly, and save
your tim e. T h e R o to-L o u vre D r y
DRY
MATERIAL er is one of four different types
DISC H AR G E
o f d r y e r s' and coolers m ade by
A B C
Large heat passages send a As inside shell tapers, heat N ear discharge end, sm all L in k -B elt. Send for Catalogs.
large volum e o f h ot gases passages grow smaller, pass est am ount o f heat enters;
m aterial lies in thickest lay
through a thin layer o f wet ing less heat through a er. N o danger o f burning
material. thicker layer o f m aterial. or overheating.
LINK-BELT C O M P A N Y
Chicago 9, Indianapolis 6, Philadelphia
40, Atlanta, D allas 1, M inneapolis 5,
San Francisco 24, T oron to 8. Offices
in principal cities.
95 33
W at» to C H E C K I. & E. C.
R e p o rts on the
G o + tu fte s ic e
,
Chem ical W orld ñ 71 ,
W E LD E D S E A M S Today Jn & u A P u f,
in Stainless Steel .
The following table eloquently illustrates the futility of
PRO C ESSIN G war even as a profit motive:
Sales
E Q U IP M E N T re a r
1930
1937
(in T h ou sa n d s)1
$ 904,279
1,004,584
Op era tine
Profit, %
N et
P rofit6, %
L. A 1938 849,080
193!) 1.030,440
1940 1,204,217
1941 1,771,470
Before inducting” a new processing vessel into war 1942 2,068,333
w ork examine its joints and w elded seams. For the " Registrant com panies num ber 32 in 1930, 1937, 37 in 1938. 37 in
life o f your equipment depends on the soundness 1939, 38 in 1940, 38 in 1941, and 41 in 1942.
A fter incom e taxes in relation to sales.
o f the welds.
Im proper w elding can often be recognized with

Mineral Plethora. Late in June a representative of the
the naked eye. The diagrams below may guide engi
neers in detecting proper and im proper welds. Way Production Board sounded a warning to the meeting
of the American Society for Testing Materials that this
The most practical way to eliminate the danger of
im proper w elding is to select a fabricator having country's mineral reserves are in danger of depletion. His
specialized experience in w ork in g wich stainless view was not of the immediate future, but he pointed to the
steel. For years, S. Blickman, Inc., has devoted its huge maws of modern industry which were consuming an
large facilities mainly to the fabrication o f stainless ever-growing quantity of metal even in peacetime. A
steel equipm ent in gauges up to Yg" thick. Our k n ow
different problem, however, is wrinkling the brows of miners,
h ow assures you o f processing equipm ent with welded
seams that stand up under wartime production. smelters, and metal importers. A survey in The Wall Street
Journal of Uncle Sam's mineral stock pile shows 13,636,290,-
551 pounds on hand, comprising upward of fifty items.
This includes 5.6 billion pounds of bauxite, 2.1 billion of
manganese ore, 2.1 billion of chrome ore, and 1 billion
of zinc concentrates and ore. It also includes stocks
PO O R. Im p ro p e r i
P io le s a r e n o ! e v e n of beryllium, iridium, osmium, palladium, rhodium, and
o lh e r ruthenium.
Salvage after the last war resulted in a serious glut on the
lead market. Its use in this war is not so diverse; it goes
principally into slugs for small-arms ammunition. The
copper industry, however, will probably be faced with a
tremendous scrap return, and the combined stock pile in
all hands is currently about 1287 million pounds. A lead
scrap return amounting to billions of pounds would not be
surprising. The same is true of aluminum, where scrap is
already beginning to affect the market.
The mercury stock pile is 5.8 million pounds, compared
to a national consumption in 1940 of 2.1 million. Already
the market has suffered a break, although the present price
is still far beyond the normal peacetime figure. This
picture will be further complicated when the mercury
mines of Italy, supplying about 40% of the world’s total,
get back into production. Half of them are now in recaptured
territory; although no report has been made on their condi
tion, it is expected that they will begin to operate as quickly
M l ¿ S . «W K .T O ^ ° ' li as possible. The last time mercury made news was when
Italy dropped out of the domestic market, and sent the price ,
from 76 to 215 dollars a flask.
T h ese diagram« appear as part o f the Blick-
.Returning to the speech of the WPB representative, his
m an brochure " W h a t to L o o k fo r W h e n chief point was a plea to strengthen our mineral reserves
y o u S p e c i f y S ta in le s s S t e e l f o r Y o u r
P r o c e s s in g E q u ip m e n t ” W r i t e f o r the
broch u re o n y o u r com pan y stationery. against increased peacetime consumption with an emergency
margin. He urged a program in which the Government
would exert a beneficial control over industry. In the light
of these figures, however, miners, smelters, and importers
would probably like to peek into the future now and then,
to see what is needed and when. (Continued on page 10)
Experience
tailored to Y O U R gas processing need!
On any gas processing problem, however difficult it

may seem, we urge you to take advantage of Girdler's
long and successful engineering and technical experi
ence and unexcelled laboratory facilities. If you are
concerned with the manufacture or purification of gas
eous mixtures or liquid hydrocarbons, take advantage
of this experience and consult Girdler without delay.
The G I R D L E R
9
I. & E. C.
R e p o rts on the
C hem ical W o rld
T oday
Progress Report No. 6. Bradley Dewey, probably by

now the ex-Rubber Director, issued Progress Report No. 6
on July 25. With it he sent a memorandum to J. F.
Byrnes, Director of the Office of War Mobilization, recom
mending abolition of the special powers granted to the
Rubber Director. Dewey’s request for the termination
was based on the fact that the rubber'supply had been
brought to its required level. Our congratulations to
Colonel Dewey for the devotion he showed to the task
and for his success in completing the program. From
the report we cull the following interesting facts:
During the second quarter of 1944, 209,004 long tons of syn

thetic rubber were made (at the approximate rate of 836,000
tons per year).
I he Celanese Corporation is building a butadiene plant, at
Bishop, Texas, using the aldol process which will employ petro
leum gases as raw material.
Less than six thousand workers stand between the present
shortage of tires and an ability to meet all essential requirements.
The Government has lost $25,952,639 on the scrap rubber
deals, and stands to lose $2,000,000 more.
Several states have removed the wartime ban on automobile
speeds.
The important goal for present research in rubber is a new
that (1) evolves less heat, (2) will not lose strength at
elevated temperatures, and (3)' will have greater resilience for a
soft-riding tire.
Crude supply in 1944 is expected to be 118,000 tons.
Foreign Economic Administration has terminated the de
velopment of cryptostegia in Haiti.
Contracts for tire construction of neoprene plants in Russia
have been granted because neoprene does not ship well in the
unvulcanized state.
Oil, Sand, and the Starry-Eyed. Oil is the modern

The Reed V a lv e o p e n s or c lo se s w ith a q u a rte r-tu rn lubricant to international agreements. We hope that
o f the v a lv e stem a n d e x c lu s iv e fe a tu re s a s s u r e e a se the recent accord between the United States and Great
o f o p e ra tio n , p o sitiv e tig h t sh u t-o ff, a n d tro u b le -fre e Britain on the division of the world’ s petroleum resources
o p e ra tio n . does not prove too slippery an understanding, and that
W A R C O N T R A C T S H A V E BEEN T A K IN G A M A J O R P O R T IO N agreers will not soon find themselves skidding on their
O F REED V A L V E P R O D U C T IO N , BUT DU E TO G REATLY international ears. W e view this as more than a pos
E X P A N D E D FAC ILITIES W E A R E N O W IN A P O S IT IO N
TO G IV E PROMPT D E L IV E R Y O N M O ST IT E M S .
sibility. In essence, the agreement is based merely on
the fact that the United States and Great Britain need
For fu rth e r In f o r m a t io n w rite the c o m p a n y
o r the f o llo w i n g d is trib u to rs :
oil; the Atlantic Charter recommends that all nations
REPUBLIC SU PPLY CO. FR IC K & L IN D S A Y CO. have access to raw materials; therefore, oil-producing
L O S A N G E L E S, CALIF. PITTSBURGH, PA. nations ought to make available to these two countries,
SO U T H ERN SUPPLY CO. TA YLO R SUPPLY CO. and the rest of the world, their oil resources. Q.E.D.
W IC H IT A , K A N S A S DETROIT, M IC H .
But suppose other nations do not wish to share their
CULBERT PIPE & FITTING CO. PEDEN IR O N & STEEL CO.
JERSEY CITY, N. J. H O U ST O N , TEXA S oil resources and do not feel bound by the Atlantic Char
ter then what? Another depressing vista opens up in
the oil picture. The United States which, for the second
UHLUE time in history, has floated her allies to victory on a sea of

oil will not be in a position to do so again, should the need
arise. W e are more than likely to become an oil-import
ing nation, and importation of oil is risky business in
time of war.
er b it l o m p n n v The new oil agreement calls- for proposals to other
H O U S T O N 1, T E X A S __________ governments on establishing {Continued on page 14)
September, 1944 INDUSTRIAL AND ENGINEERING CHEMISTRY
i '.I-' AA--'
Check . . . recheck . . . check again! JNothing Bevels, inside and outside diameters, interior
left to chance. From raw materials to finished surfacesîwall thickness, finish, all must meet
product, every step o f the way, each Tube-Turn Tube Turns' standards of accuracy. Small
welding fitting is subjected to many routine in wonder thatT ube-T urn welding fittings help to
spections that are designed to insure accurate di make piping systems lower in upkeep costs, and
mensions and conformity to American standards. more efficient. W rite for Catalog 111 — now!
Selected Tube Turns Distributors in every principal

/ / you don’t have one, write city are ready to serve you from complete stocks.
today lo r the new booklet
T U B E T U R N S (In c .) Louisville, Kentucky. Branch Offices: New York,
Allowable W orking Pressures.” Chicago, Philadelphia, Pittsburgh, Cleveland, Dayton, Washington, D. C.,
Houston, San Francisco, Seattle, Los Angeles.
TUBE-TURN TRADE M ARK
T i/ e ld in q u n c6
Vol. 36, No. 9
coF. /2ioao%j«'\
UQ SF.6R. /.¿AS
D e v e lo p m e n t s m a d e by S w e n s o n e n g in e e r s h a v e im p r o v e d
m a n y p r o d u c t s . . . m a d e p o s s i b l e th e e c o n o m i c a l r e c o v e r y o f
c h e m ic a ls . . . s o lv e d w a s te d is p o s a l p r o b l e m s . . . s im p lifie d p la n t
o p e r a t io n s a n d p r o c e s s e s .
T y p i c a l o f th e b en efits r e a liz e d t h r o u g h th e u se o f S w e n s o n
P r o c e s s E n g in e e r in g S e r v ic e is th e e fficie n t s y s te m d e v e lo p e d
f o r r e c o v e r i n g d r ie d d is t i ll e r ’ s s o lu b le s f r o m s p e n t m a sh or
d is t ille r y s lo p .
tcvcL. /n run r,
HORMAUV iOH D o u b t le s s , y o u h a v e p r o b l e m s in y o u r o p e r a t io n s w h i c h ca n
TO HAVE STORA
AVttiaeLE tvuj b e s o lv e d w ith th e h e lp o f S w e n s o n e n g in e e r s . W h y n o t p r o fit
OPArouf W K
AOWN FOR CIS
REPAIR. b y th e ir yea rs o f e x p e r ie n c e , a n d r e s e a r c h ? T h e y w i l l b e g la d
t o h e lp y o u p la n p r o c e s s i m p r o v e m e n t s . W r it e , o u t l in in g y o u r
p a r tic u la r p r o b l e m s .
LEGBNO
SWEN5DN
ATORS • FILTERS • CRYSTA
A nalysis o f R equirem ents"-
DOUTHITT GRAY-JENSEN SPRAY DRYERS

September, 1944
(9.30* "SoiW
Swenson - E n g in e e re d System for

R e c o v e r y of D r i e d D i s t i l l e r ’s S o l u b l es
T h e fine stillage from distillery slop o f S w enson L o n g -T u b e V ertical and
contains in solu tion and suspension F orced Circulation Evaporators.
materials that are rich in p rotein , m in T h e Douthitt Gray-Jensen p rocess o f
erals, fats, riboflavin, and other factors s p ra y -d ry in g d is tille r ’s s o lu b le s has
o f the V itam in B C om plex. T hese ma been recently introd u ced by Swenson.
terials are a valuable feed fo r poultry It reduces the syrup to solid s con sist
and anim als... especially im portant dur in g o f un iform size particles. T h e dry
in g the wartim e shortage o f feedstuffs. in g is d on e instantaneously under tem
T o recov er the solids in convenient perature co n d itio n s w h ich eliminate
dried form , Sw enson engineers d esign sco rch in g and assure m axim um reten
ed a continuous system for redu cin g tion o f vitamin values . . . necessary to
the fine stillage to syrup by multiple- secure top market prices.
e ffe c t e v a p o r a t io n , and th en sp ra y - S w e n s o n e n g in e e rs have a ls o d e
drying it to a solid . signed and installed continuous mash
H igh ly efficient evaporation is accom - c o o k in g and mash c o o lin g systems . . .
plish ed by the ju d iciou s com bin ation p ro v id in g increased efficiency.
SWENSON EVAPORATOR COMPANY

D ivision o f W h iting C orporation
15671 Lathrop Avenue Harvey, Illinois
SWENSON PRO VIDES THIS FIVE-WAY SERVICE..
D esign and Layout M anufacture o f Equipment Test O p eration P e rio d ic C heck-ups
SWENSON EVAPORATOR CO
H A R V E Y , IL L ., U . S . A .
Title- F l O W SHEET
MA7ÆM4L BÂIAHCE
Capacity
I. & E. C.
R e p o rts on the
Chem ical W o rld
Today
an international petroleum council; here is the crux of

one portion of the problem. Now or never will coopera
tion between nations be tested, anti the brave new world is
going to receive an oily baptism of diplomatic sophistry.
Upon such plans - and conferences hangs the fate of the
idea of international cooperation in this country, free giver
No . 618. Special heavy-duty w eld ed in times that try men’s souls, but not always accorded re
steel b o d y unit with "H *“ metal vane.
Heavy-duty hand w h eel c o n tr o l for ciprocal treatment when the times and souls are forgotten.
op era tion at 150 psi.
Perhaps in event of failure we will become hardhearted
and enforce the principles for which we have risked much.
We shall have the power to do so if we wish.
But there is another solution to all these problems which
we advocate as being more certain than the dreams of the
starry-eyed. Let us turn loose research teams on the
problem of developing fuels from other plentiful re
sources, as well as motors to utilize the fuels, that will be as
powerful as corresponding gasoline and oil motors. These
__ mlined motors need not be economical, for their avowed use will
resistance-
be only the uneconomical purpose of war. We wager
» S »
' “? » that if the country developed within three years after
hS e pS “ * World War II a jet propulsion motor, for example, utiliz
ing powdered coal, sawdust, or, mixtures of chemicals,
world cooperation in matters concerning petroleum would
become automatic. Let us become practical interna
furtherm ore, rse turns to ^ closed tional realists about the best way to prevent the sand of
necessity from wrecking the intricate peace machinery of
‘Sposition, the' saCCOmplished
n â i® by“ body
the starry-eyed. We often suspect that the dawn of
civilization will begin only when every nation can produce
all its needs with what it has. When power is obtained
from the untapped sources of the atom, when food and
fibers are synthesized from air; sunshine, and water, and
man fears not for his material wants, peace shall be. Let
us insist that the peace conference put a research team
to work on it. They might do so if we (the scientists)
explain to them (the diplomats) that only when such prob
lems are solved will their treaties (seem to) accomplish
what they propose.
Liquefied Gas, Forward March! Although the war

has diverted a great deal of butane into aviation gasoline
and synthetic rubber, the forward march of liquefied pe
troleum gases continues. Latest official figures register a
gain in sales of 19% during the past year. The necessary
diversion of butane for war purposes becomes most evident
by contrast. Sales jumped 40% in 1940 and 47.7% in 1941.
As a result of the demands of rubber and gasoline for
15 t o 900 pari.- butane, propane carried the lion’s share of the commercial
C o n tro l and shut-off N o. 617. E lectric m o to r c o n tr o l fo r ex
o f volu m e and p res trem ely s lo w a ctin g valve. H and w h eel
burden. While the war has curtailed the industrial side of
s u r e f o r a ir, g a s .
s te a m , l i q u id s a n d
fo r manual op era tion o n ftear red uction the picture, it has also contributed in certain channels. For
drive. E xtended lever fo r op era tion o f
s em i-solid s. sister valve in parallel line. instance, shipyards have been using propane to supplant
VALVE DIVISION acetylene for metal-cutting operations. It is also used as a
substitute for fuel oil in preheating of plates.
R-S PRODUCTS CORPORATION Manpower, that universal plaint these days, has forced
G erm antow n Ave. & B erkley St. • Philadelphia 4 4, Pa.
curtailment in all operations—production, service, and sales.
With the diversion of butane, this would indicate that the
lowered volume represents but a temporary halt, and that the
march will proceed on the double-quick when normal condi
BUTTERFLY VALVES tions are restored. {Continued on page 16)
14
September, 1944
G E A R S . . .
INDUSTRY AND MEN

Since 1888...The M en of the D. O. James O rga n i
zation have been a component part of the great
expansion of American industry...Since Pearl
Harbor...D. O. James Gears and Gear Reducers
have been important components of the guns,
ships and aircraft of our armed forces...The
experience derived from our pre-w ar and w ar
time activities w ill be an important asset to
American industry when Victory comes . . .
That the post-war Gear and Gear Reducer re
quirements of American industry w ill be capably
handled by our organization is proven by the
testim ony of the continued perform ance of
D. O. Jam es G ears a n d G e a r Reducers of
every type for every in d u strial a p p licatio n .
T O D A Y ... D. O. James Gears and M en

are ready to serve American industry.
Established 1888
56 Y E A R S M A K E R S OF EVERY TYPE OF G E A R A N D G E A R REDU CER
D .O . J A M E S M A N U F A C T U R IN G CO., 1140 W EST M O N R O E STREET, C H IC A G O 7, ILL

Wh y this leading Manufacturer of Abrasives
I. & E. C.
^ DESPATCH ¿ W R e p o rts on the
Chem ical W o rld
'W g a, a n d
¿ » fi TddanA Today P o -itu A G S i
Magnesium Cutback. Magnesium is the second metal

to be cut back under the War Production Board program.
This removal of allocations is not being accepted so quietly
as was that on aluminum a few months ago. The reason
is obvious. Aluminum was established solidly in hundreds
of commercial uses long before the war, whereas magne
sium was just securing a competitive foothold when the
Government demanded multiplication of the industry's
entire production; in fact, it has jumped from 18,000,000
to 600,000,000 pounds since 1941. A series of cutbacks
already issued has halved magnesium production, and
there is talk in the metal trade that W PB will want no
more than 22,000,000 pounds a month this fall.
Willard H. Dow, president of Dow Chemical Company
which nurtured the industry in prewar days, has com
plained to Donald M . Nelson, chairman of WPB, not
of the cutbacks but of the manner in which they are
being handled. He points out that Order M -2-b is being
represented as removing the material from government
control. Ho claims the order only modifies the control
and, in support of this contention, points out that mag
nesium ingots are still under allocation:
As matters now stand, the industry is entirely capable . . . of

supplying all possible needs of the Government, either for domes
tic use or export, and the stock pile is of such proportions as to
give ample insurance against any kind of shortage . . . The uses
I N T H E B U S Y L A B O R A T O R I E S o f one o f A m erica’s lead of magnesium in war have been impressive. The metal has
ing makers o f abrasives, Despatch ovens are used daily for proved itself. But as yet there has not been the opportunity to
hundreds o f processes. They dry samples, con trol product promote the peacetime use of the metal . . . If the industry now
quality and are extensively used in research. had the opportunity, it could go ahead developing markets for
All these operations require very close con trol o f heat peacetime consumption and, in so doing, would be developing
throughout a w ide range, uniform airflow and adequate opportunities for the employment of our boys, as and when they
return to civilian life.
processin g speed to keep abreast o f each laboratory’s
schedule.
The tremendous upswing in production was occasioned
By providin g this, Despatch ovens meet this manufacturer’s chiefly by the erection of government plants, two of which
exacting demands for dependable laboratory equipment. are operated by Dow Chemical Company. One of these
has been ordered out of production. Just what will
happen to the magnesium industry in the course of the
W R IT E for B ulletin
next few months may be the first clue as to what disposi
105 describing Des
p a tc h la b o r a t o r y
tion will be made of government-owned plants when re
o vens adaptability conversion begins in earnest.
;d air to yo u r plant.
hrough F u lly illustrated. U. S. Nubun. One of the most forward-looking an
noothly Gives complete spec nouncements of.recent months was U. S. Rubber Com
every ifications. . . . F R E E ! pany's introduction of Nubun, a synthetic rubber latex
f s p a ce,
insulation for power, lighting, and communication cable.
t spots. ^ SEE Y O U R
The development, according to C. W. Higbee, manager
LOCAL DEALER of the wire and cable department, is expected to be of
y/ide flexibility- little use in the war effort. It is strictly a postwar pos
sibility as matters now stand. U. S. Rubber researchers
have long sought an insulating material which could be
applied to wire by the latex process. Buna S seemed to
be the best possibility, but it was found necessary to modify
and improve on the basic Buna S polymer through changes
in its composition, through changes in the process of
polymerization, and through other fundamental modi
fications.
16
in eye protection
N E W A O FUL-VUE light-w eight acetate rests lightly o n 'th e

face, w hile com fort cable temples hold the
goggle snugly in position, prevent it from
ACETATE G O G G L E sliding forward on the nose. And the ex

cellent fitting qualities o f this goggle make
it ideal for workers w h o need Super
Armorplate lenses ground to their own
prescriptions.
A O Ful-Vue Acetate G oggle N o. F-8247
This new acetate goggle— the result o f comes in 47 mm. eyesize in Pink Crystal
tw o years’ research in A O 's Development with either white or Calobar 6-curve
Department— ofFers real com fort in eye Super Armorplate impact-resisting lenses.
protection. It is available in three bridge sizes, 22, 24,.
It has a stronger, more massive front and 26. Your nearest A O Representative
than ordinary acetate models. A deep can supply you. Call him today. And if
groove in the lens rim forms a lens seat you need acetate goggles for flash welders
with substantial backing to resist severe and those w orking around welding, ask
impact. And the bridge has a slight amount him about the other new Ful-Vue Acetate
o f face form to give added side protection. G oggle that comes equipped with side
shields and Calobar lenses.
Nose pads are also made o f heavy stock
and have well-rounded edges and an ample
flare to withstand extraordinary jolts and
blows.
Besides greater strength, this new goggle,
American IP Optical COMPANY
is designed for all-day com fort, too. The S O U T H B R ID G E , M A SS A C H U S E T T S
M anufacturers fo r m ore than 1X1 Y ears o f p ro d u cts to A id and P rotect V isio n . Branches in ail p rincipal industrial centers
18 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No. 9
MARBLEHEAD
H i g h -Calc ium
CHEMICAL LIME
It’s a to u gh jo b d eh airin g hides, but an im portant one to the great leather industry. The hair
and surrounding fat ce lls must b e rem oved before tanning.
M a r b le h e a d acts pro m p tly to enlarge the pores and p lu m p up the hides, d isso lv in g the album en
and fat, and th o ro u g h ly lo o se n in g the hair. The skins are then e a sily dehaired b y scraping.
For this service, M a r b le h e a d offers the b ig advan tage o f rem aining lo n g in suspen sion and
sla k in g q u ic k ly to a fine, sm ooth putty. It is o n ly o n e of the m any exacting jo b s on which
M a r b le h e a d C h em ical L im e proves its outstan din g efficiency.
* FO UR F O R M S *
| y | | t| jg p
MARBLEHEAD PO W DERED
Q UICK L IM E PEBBLE LIME HYDRATED
L IM E
L U M P L IM E
160 N. LaSalle St.
LIME GO. Chicago, III.
U.S.I. CHEMICAL NEWS
AD VERTISEM EN T —This entire page, is a paid advertisement. Prepared Monthly .by:H SAM ^hiaiihem iediy iHf
September -fa A Monthly Series for Chemists anti Executives of the Solvents and Chemical Consuming Industries "A 1941
New Foam Insulation U.S.I. Expands Line

,
Saves Space Weight of Modified Alkyd
A new, phenol-resin foam plastic insulating Specification Resins
material, which weighs only two pounds per
cubic foot, is lighter and lower in heat con
ductivity than rock wool, glass fibre or cork. N ew S&W W ar Resins W ell Suited
Its present war uses are secret, but it promises fo r Governm ent A gency Coatings
to have many peace time applications because
of its self-raising and self-curing properties.
Supplementing its announcement of Aro-
The new foam plastic should find special
plaz 1375 in the August issue of C h e m ic a l
favor with designers of modern prefabricated
N ew s , U.S.I. now announces eight addition
houses, refrigerators, automobiles and other
products, where low-weight, high-efficiency al Aroplaz resins in which phthalic anhydride
content has been adjusted to meet current
Postwar Cosmetics heat insulation is desired.
The director of the laboratory which devel War Production Board restrictions. The com
oped the new insulation explained, “ It just parative specifications of all nine resins are
Seen as Big Users grows all by itself. It will foam to 30 times its shown in the box below.
own size, that is, a quart can of the mixture All of these resin vehicles have been de
of U.S.I. Chemicals will expand sufficiently to fill a 7 to 8 gallon signed to help manufacturers to formulate
receptacle in about 10 minutes. What little government agency specification finishes at
heat is required is generated by the mixture minimum raw-material cost, and with no
indalone and BK-5 as w ell as itself.” sacrifice in quality.
standbys lik e ethanol expected The new product, which resembles molasses Army specification 3-183 requires S&W
in appearance, will begin to foam and expand
to find m any new applications Aroplaz 1365 blended with S&W Fused Congo
within a maximum of five minutes after the
No. 5, while Quartermaster Corps specifica
mixing stops. It cures itself without applica
Just about the first postwar idea in every tion of heat or any further attention. tion CQD-200A requires S&W Aroplaz 1205-H.
cosmetic manufacturer’s mind is to get back Specification CQD-65B is met with S&W
to prewar raw materials. Substitutes for ethyl Aroplaz 1130 blended with either urea or
alcohol and other cosmetic ingredients have Longer-lived Catalysts melamine.
rendered indispensibie service in extending
the available supplies of these war-scarce For Nitrile Production 3 1 .5 % P hthalic Content Resins
chemicals. But the inherent advantages of S&W Aroplaz Resins 1323L, 1323D and/or
ethyl alcohol, for example, will make the re In producing nitriles from primary alcohols 1333 will be found eminently satisfactory for
turn of unlimited supplies a significant event and ammonia, copper catalysts rapidly lose producing coatings to meet Army specifica
to the industry. In addition, of course, many their effectiveness. Yields range from 80 down tions 3-177, 3-178, 3-181 and 3-187, Engineer
manufacturers have new products ready, wait to 40 per cent. Two newly patented processes Corps specification T-1760, and Federal speci
ing for ethyl alcohol and other chemicals to are claimed to increase yield by boosting the fication TT-E-485.
produce them. activity and stability of the catalyst.
The new processes involve the use, as 2 4 % Phthalic Content Resins
New P roducts C oining
catalyst, of either reduced silver, reduced A maximum phthalic anhydride content
Among the more interesting new products copper or, better, a mixture of the two de
now in prospect, are modifications of the sun of 24 per cent has been set up for Bureau of
posited on a partially dehydrated amorphous
tan lotions and insect repellents which made Ships specification 52R13 and Maritime Com
oxide of aluminum, zirconium, thorium, or
such rapid strides just before the war. Pre other rare earth metal. The latter serves as a mission specification 52MC21—both for alkyd-
cisely what these products are, manufacturers dehydrating agent. resin solutions. These government specifica
naturally do not reveal. In the field of insect tions are met by S&W Aroplaz 1240 and 1244
While the new catalysts lose effectiveness,
repellents, however, it is certain that wide use they do so much less rapidly than copper respectively.
will be made of the knowledge gained in alone. Further, they may be reactivated read
developing insectifuges for our armed forces. 1 6 % Phthalic Content Resin
ily by passing air over the catalyst at the tem
The government’s new all-purpose repellent, perature of reaction. This is followed by Use of Aroplaz 1375 in the Army and Navy
in which IJ.S.I.’s Indalone is a vital ingredi- treatment with hydrogen to restore the oxides replacement specifications where alkyd res
(.Continued on next page) to the metallic form. ins not exceeding 16 per cent phthalic anhy
Among- the alcohols readily converted to dride content are permitted, was discussed
nitriles by this process are n-butanol and in detail in the August issue of C h e m ic a l
New Solvent Blend Gives ethanol. N ew s.
Better Primer Adhesion

According to a leading process engineer, a
R E S I N S PECI F I C A T I O N S
mixture of hydrocarbon solvents, such as tolu
ene and xylene, with alcohol in the ratio of saw Vise. Value Color Lbs./Gal. Anhydride Oil
AROPLAZ Solution G. H. of Solids (GH 19331 at 20°C. Content Content
two parts to one is most efficient in cleaning
surfaces prior to the application of the zinc 1130 60% in X ylo l Z 3 -Z 5 4 -6 5-8 8.5 38 40.5
chromate primer coat. This mixture flashes 1205H 50% in H SN Z 4 -Z 5 6 -1 2 7 -1 0 8.0 3 0 -3 1 .5 37
off the surface quickly and leaves no residue, 1240 70% in MS Z -Z l 6 -1 0 6-8 8.0 2 3 -2 4 64
while the alcohol content is great enough to 1244 70% in MS Y -Z l 6 -1 0 7 -1 2 8.0 2 3 -2 4 65
remove any moisture. 1 3 2 3 Lt. 50% in MS Z -Z 2 5 -1 0 5 -1 0 7.65 3 0 -3 1 .5 54.5
He also states that this mixture has an 1 3 2 3 Dk. 50% in MS Z 3 -Z 4 5 -1 0 16 -1 8 7.7 0 3 0 -3 1 .5 54
added application in cleaning the primer coat 1333 50% in MS U-W 5 -1 0 5 -1 0 7.65 3 0 -3 1 .5 54.5
before the final finishing coat is applied.
136 5 60% in Xylol Z-Z2 18-2 5 6-9 8.5 38.5 45.5
While it will not remove the primer without
1375 50% in MS S-V 12-1 8 16-1 8 7.65 1 5 -1 6 56
abnormal rubbing, it softens it sufficiently to
imnrnvp tin* orlbAn’Ar. »V.«. *
.ADVERTISEMENT—This entire page is a paid advertisement.
September
U.S.I. CHEMICAL NEWS 1944
Ups Color Fastness Postwar Cosmetics T E C H N IC A L D E V E L O P M E N T S

(.Continued from preceding page)
of Printed Textiles ent, provides one excellent example.
Further information on these items
In the field of sun-tan lotions it is equally
Claimed to be resistant to dry cleaning may be obtained by writing to U.S.I.
certain that wide use will be made of BK-5,
solvents and other detergents, a new pig U.S.I.’s light-screening compound. Bright Four n e w a d h e siv e s a re o ffe re d b y their m an u
mented lacquer-in-water emulsion printing prospects, too, are envisioned for combination facturers for the fo llo w in g a p p lica tio n s: (No. 847)
paste has been patented for use in the print for b a c k a n d la c e o f con ta in er la b e ls ; a d h e siv e is
suntan-insect repellent products utilizing cla im e d to d ry im m ed ia tely to c le a r film that re
ing of textiles. Indalone’s dual properties as a light screening sists w a te r, w e a th e r, o il, b rin e: (N o. 848) for se a l
The new product employs a lacquer-in- in g w a te rp ro o f b o x b o a rd con ta in ers; a d h esiv e
agent and insect repellent. com es as a d ry , sta b le p o w d e r : (N o. 849) for
water emulsion containing a pigmented base. la m in a tin g structural w o o d p arts; this resin a d
The base includes a urea-formaljjeliyde resin W ider Usage o f Ethanol h e siv e is cla im e d to cu re in o n e h ou r at 180° F
a n d to h a v e a w o rk in g life o f four hours at 70° F:
in a solvent consisting of xylol and butanol, There are many reasons why the perfume (No. 850) fo r b o n d in g alum in um to alum in um , iron
resins, and pine oil. Before use, paste is dis and cosmetic industry has, since the earliest to iron, or eith er m etal to non -m etals; con sisting
days, been a heavy user of ethanol. Perhaps o f resin b a s e a n d cu rin g a g e n t, a d h e s iv e acts at»
persed in a neutral base made from water, 300° F.
resins, and methyl cellulose. first is its wide-range solvent power. However, U S I
its astringent action on the skin, its antiseptic A n e w syn th etic in se cticid e is o ffe re d for re p la c
in g sca rce natural products in the con trol of
power, and its rapid evaporation rate which a p h id s, le a fh o p p e rs, a n d other pests. In killing
Stabilizes Halogenated leaves a pleasant, cool sensation are factors certain su ck in g insects, the p rod u ct is cla im e d to
e q u a l nicotin e, roten on e, a n d p yreih rum .
of equal importance. (No. 851)
Compounds with Acetone Still another prominent advantage of etha U S I
nol is its pleasant, neutral odor, which entails C ontinuous d e h y d ra tio n o f co m p re s se d a ir or g a s
is s a id to b e fe a sib le w ith a n e w d e v ic e com p ris
According to a recent patent, halogenated no camouflaging. in g tw o ce ra m ic tubes, o n e w a ter-rep ellen t, the
ketones, such as iodoacetone and hromoace- Medicated lotions, shaving lotions, hair lo other w a te r-p e rm e a b le but im p erviou s to air. De
tone, can he stabilized with a water-iniscible s ig n e d to p rotect a ir tools, g a s g en era tors, etc.,
tions and the rest of the strongly alcoholic the d e v ic e is inserted in the lin e. The s e c o n d tube
solvent so that the halogen is retained until products may safely be expected to he on the acts a s a w ick co n tin u o u sly d ra w in g out w a ter
the compound comes in contact with water. w ithout a llo w in g g a s to e s c a p e . (No. 852)
market in ever increasing variety just as soon
Among the solvents mentioned for this use U S J
as conditions permit. The same applies to the
are acetone and ethyl acetate. It is claimed A n e w neu tralizer perfu m e is cla im e d not o n ly to
mildly alcoholic lotions — skin fresheners, h a v e the hid in g p o w e r to c o v e r the h e a v y odors
that this stabilization increases the utility o f toxic a gen ts, but in a d d ition to im part a light
cleaners, deodorants, hair wavers, bleaches
of these halogen compounds as germicides, refresh in g fra g ra n ce . (N o. 853)
and the rest—and to ethanol-containing sham U S I
fungicides, etc. One use suggested for them
poos, liquid shaving soaps, dentrifices. Yes, A n e w hot-form ing p la stic co m e s in la m in a ted
is the treatment of ships hulls to prevent sheets w h ic h ca n b e bent, form ed o r d ra w n , upon
even bubble-bath preparations!
fouling. ----------------- h ea tin g to 275° F, a cco r d in g to a recen t a n
n ou ncem ent. Products so form ed a re s a id to h a v e
Other U.S.I. Products high tensile, flexu ra l, a n d co m p re s siv e strengths.
Citrus Pectate Pulp Solves A Among the other U.S.I. products destined (No. 854)
U S I
to find widening use in postwar cosmetics are
Synthetic Rubber Problem uretlian (as a patented fixative for hair dyes),
M ild e w p roofin g , o f cotton, jute, sisa l, lin en an d
hem p , is s a id to b e e ffe cte d w ith a n e w p roduct.
amyl acetate (for perfuming and flavoring The p rod u ct co m e s a s a solu tion w h ic h is com
p a tib le w ith sta n d a rd w a te r rep ellen ts. (N o. 855)
Unlike natural rubber, the synthetic prod lipstick), ethyl acetate, amyl acetate and
U S I
uct has a decided tendency to adhere to paper ethyl phthalate (as perfume ingredients), H igh er a d h e sio n o f pain ts, la cq u e rs, a n d en a m el
and to flow. As it is shipped from manufactur nitrocellulose solutions (for nail enamels), to su rfa ces o f co p p e r, b ra ss, or b ron ze is s a id to
ing to process plants in paperboard boxes, or ethyl, butyl, and amyl acetates and dibutyl b e o b ta in e d w h e n the su rfa ce is first treated with
a n e w p rod u ct w h ic h p ro d u ce s a sta b le, ad h eren t
multi-wall paper bags, this tendency to stick phthalate as plasticizers in lacquer and as cu p r ic o x id e co a tin g . (No. 856)
to paper greatly complicates its packaging nail polish removers. U S I
and handling. A n e w stearate, cla im e d to p rev en t w a te r a b s o rb -
A film of citrus pectate pulp in an aqueous tion b y e d ib le h y d r o s c o p ic p o w d e rs , is a ls o o f
dispersion containing some sodium phosphate

Alkyd Resin Used fered fo r u se in e d ib le o il em u lsion s, shortenings,
etc. O ther uses in clu d e em u lsifica tion o f w a x e s,
applied to the paper container, is reported to in Wound Coverings o ils, a n d p o lis h e s , a n d a s a pour-point dep ressa n t
for lu b rica tin g o ils. D isp ersible in hot w ater,
solve this problem. The film adheres lightly to p rod u ct is s o lu b le in a lco h o ls a n d hot h y d ro ca r
paper, strongly to synthetic rubber. When Rayon, impregnated with an alkyd resin, bons. (No. 857)
U S I
package is emptied, the film pulls cleanly has been accepted by the British Pharmaceu A m o la sse s re p la ce m e n t for u s e a s a b in d e r in
from the paper, and carries through with the tical Codex as a substitute for oiled silk to fou n d ry w o rk , b riquetting, th ick en in g a gen ts an d
sim ilar a p p lica tio n s, is rep orted to b e a v a ila b le
rubber as a minute, and entirely innocuous be used in covering wet dressings to inhibit in q u antity, w ith ou t a llo c a tio n lim itations.
constituent. drying. (N o. 858)
II C I n d u s t r i a l C h e m i c a l s , I n c
60 6 A ST 4 ÎN D ST.. NEW YORK

SO U I m
mm /
£ branches in all p r in c ip a l c it ie s
alco h o ls A C E T IC ESTERS IN T ER M E D IA T ES ACETONE

A m y l A lc o h b l A m y l A ce tate A c e to a c e ta n ilid e ‘ C h e m ic a lly Pure
B u io n o l (N o rm a l Butyl A lc o h o ll Butyl A ce tate A c e to a c e t-o rth o -a n isid id e R ESIN S
. Fusel O i l — Refined Ethyl A ce tate A c e to a c e t-o rth o -c h fo ra n ilid e S & W Ester G u m s— a ll types
Ethanot (Ethyl Alcohol) O X A L IC ESTERS A ce to a ce t-o rth o -to lu id id e S & W C o n g o G u m s— raw , fused &
S p e c ia lly D e n a tu re d — a ll r e g u la r D ib u tyl O x d la ie A c e t o a c e t - p a r a c h lo r a n ilid e esterified
an d o n h y d ro u s form ulas D iethyl O x a la t e Ethyl A ce to a ce to te S & W 'A r o p la z — a lk y d s a n d a llie d
C o m p le te ly D e n a tu re d — a ll re g u la r Ethyl B en zoyla ce ta te- m a te ria ls
P H T H A L IC ESTERS Ethyl Sod iu m O x a la c e ta te
a n d a n h y d ro u s form ulas D iam y l Phthalate S & W ‘A ro fe n e — p ure ^ h e n o lic s
Pure— 190 p roof, C.P. 9 6 % , D ib utyl Phthalate ETHERS S & W / A r o c h e m — m odified types
A b so lu te D iethyl Phthalate Ethyl Elher S & W N a tu ra l.R e s in s — a ll sta n d a rd
"S u p e r P yro A n ti-fre eze O TH ER ESTERS Ethyl Elher A b so lu te — A .C .S. grad e s
"S o lo x P ro p rie to ry Solvent FEED C O N C E N T R A T E S O THER PRO DU C TS
•Diotol
• A N SO L S D iethyl C a rb o n a t e ’C u r b a y B -G C o llo d io n s Ethylene
A n so l M Ethyl C h o lo ro fo rm a te • C u rb a y S o e c ia i L iq uid Ethylene G ly c o l "In d a lo n e
A n s o l PR Ethyl fo rm ate •V a c a fo n e 40 N itro c e llu lo s e S o lu tio n s U re th an
wmmMmmmmtt. ... .... _ r-,
« » c ite s s V a lv e s cost le a st in the lon g run I
. The record o f ,Jenkins Valve performance, under the Countless war records like this repeat the many peace
toughest conditions o f wartime plane operations, is begin- time records of Jenkins V alves-, and. add up to convinc-i
f aia8 to come in from busy plants like the Merganthaier ing proof o f their "lowest cost In the long run” .
U norypę', Company, world’s largest maker of typesetting Jenkins Bros,., 80 W hhe Strcety N ew Y ork IS;* Bridgeport.
machinery. ■; v • Adanta; BostonPhiladelphia: Chicago. Jenkins Bros.
'- - 4 ;V - - £ ■ M ontreal; London, . '.
p. . Tins cbmpaQy,has had a;.major part in-the production
.of -Fire, „Control Instruments. Since Pearl Harbor their
t ' e<?UIpbient has been subjected.to the equal of .7 years of
normal stress and wear. N o, other test could so "clearly J E N K I N S VALVES : S I N C E 1864
den »onstr'ate the built-in endurance of .every: Jenkins For every Indiisirsdl, Engineering* '-'Marine'and Com« ’
-, , r„ Steel'had. / ..
y Valve - a direct result o f better materials, better work- tiresuite*
! manship, and better design,
■■1V ’■Iv-'PfphT;’-
September, 1944
—a :: :' INDUSTRIAL AND ENGINEERING CHEMISTRY 23
your extinguishers in shape ... FREE BOOK tells how!

W hen fire breaks out, there’s no time to repair or recharge
extinguishers — they must be ready to hit the blaze fast!
So the logical time to keep the equipment in fighting form
is before the emergency. . . by systematic inspections at
fixed intervals.
A method for setting up an extinguisher maintenance

system in your organization is outlined in a booklet pre
pared by Walter Kidde & Company. It covers all types o f
equipment, tells exactly what to check on each type, how
often to examine and recharge. It shows how extinguishers
should be placed and marked, how records should be kept.
This book "Inspection and Maintenance o f First Aid Fire
Extinguishers” , will be sent you upon request. Write
for your copy.
WALTER KIDDE & COMPANY, INC., 14 0 CEDAR STREET, NEW YORK 6*,. N.
RTAMMei« ,«
5d5»*Êgèsi
■ «i( nlrpa. t-M*, i£r»Vvv

*•tkir-rw
‘^riùTt (.hull, n.'
f'ir luiMliiu f>>Uf.
■ialrwajr (‘wkv iv
Ä S
.uiM«ir
M ;
•t*HK t^W'ÆxA
m,whari vrKit u“ i Â i*
^Jj»n (,
iltouaVu\*M
■►u*1«lui •
»dt ion; u .
-••<rn,,.M... .
Ross’ standard lines of heat exchangers, heat

ers, coolers and condensers are tailor-made.
That sounds like a contradiction, but it isn’ t.

It simply means that each unit is carefully
engineered to cover your own operating
requirements, and then specially assembled
from standard, mass-produced parts.
In this way, problems of continuous perform

ance, maximum heat transfer rate, accessi
bility for maintenance, piping arrangement,
etc., can all be individually overcome to
your own best advantage.
FULL DETAILS
Larger Coolers Sm aller Coolers O il Heaters

Bulletin 5322 Bulletin 4 922 Bulletin 3 6 2 4 -A
REQUEST C O P IE S T O D A Y !
ROSS HEATER & MFG. CO., INC.

Division o f A m erica n R a d i a t o r & S t a n d a r d < $ a m ta itg c o r p o r a t io n
GENERAL OFFICES AND PLANT: 1423 WEST AVENUE, BUFFALO 13, N EW YORK
Struthers Wells
RADIAL PROPELLER
AGITATOR
The Struthefsh W ells Radial Propeller
/ Agitator introduces a new principle of
agitation—rby combining the high speed
V, and peripheral velocity of the conventional
A propeller agitator with the vertical mount-
Phis is an excellent example of Struthers

Wells 50 year old engineering ingenuity
— in pioneering, designing and budding
industrial, equipment— /o r e very industry
— to provide economy of operation,' sim
plicity of design dnd lasting dependability.
Write for informative Bulletin No, 52-W .
C o r p o r a t io n
PROCESS EQUIPMENT DIVISION
INDUSTRIAL AND ENGINEERING CHEMISTRY' Vol. 36, No. 9
Panel board for one of the 3 cracking units, show ing

AT LA K E CHARLES som e of the 283 Foxboro Recording, Controlling and
Indicating Instruments.
a n d f in d 2 8 3
m ore re a so n s w h y
th e in d u s tr y counts
on F o x b o ro S
Aviation luel and butadiene processes in this plant

depend on Foxboro Flow Meters, as w ell as Foxboro
Temperature and Pressure Instruments.
It's not surprising that the Cities S ervice Lake

C harles R efinery is ca lle d "the w orld 's m ost m odern
petroleum p lan t." It's the first n e w w holly-integrated
aviation a n d butadien e plant ever built. E very step
of its co m p le x d u a l p ro ce ssin g is con trolled b y the
m ost a d v a n ce d instrumentation. A n d right through
the plant, these control system s are dom inantly
F oxboro!
C h o ice of m ore than 200 F oxboro C ontrol Instru
m ents for these critical operations w a s entirely log
ica l. Throughout the w artim e ga so lin e and buta
d ien e industry, F oxboro C ontrol b a s e d on the fam ous
HYPER-RESET S tabilog C ontroller h as set standards
for op era tin g stability and a c c u r a c y n ever eq u a lled
b efore. A s a result, these vital plants h a v e been
eq u ip p e d with m ore than 6,000 S ta bilog Controllers
a lo n e !
Full d iscu ssion o f the unique HYPER-RESET Stabi
lo g C ontroller is a v a ila b le on request. The F ox b oro
C om p an y, 40 N epon set A ven u e, F o x b o ro , M ass.
Two of the M odel 30 Stabilog Controllers that form
the basis for the super-accurate integrated control Branches in prin cip al cities.
systems at Lake Charles.
REG. U. S. PAT. OFF.

PROCESS CONTROL
ANY R E S P O N S IB L E E N G IN EER
M A Y H A V E A C O P Y O F THIS
BULLETIN. JU ST A S K F O R IT.
ving
A n d Y O U w ill be o f the same opinion processing plants w ill want to refer to

JL jL after reviewing this new Bulletin Bulletin N o . 1 9 5 8 frequently when new
N o . 1 9 5 8 . . . Into its sixty pages is packed m eth od s or new eq uipm ent is b e in g
the m ost advanced thought on drying tech considered.
niques, scores o f photographs, diagrams
and specifications o f the latest equipment
for the many and varied drying require
ments . . . Engineers in chemical and food

BLAW-KNOX DIVISION
O OF B L A W -K N O X COMPANY
b o s t o n
f
M E l A i f CLOSER CONTROL FOR
r t E w V I ELECTRIC FURNACES
Electric furnaces o f m any types are now— today— NOW OPERATING
heating their charges to much m ore constant tempera ON FURNACES
tures than they did ju st a year or tw o ago, thanks to
FO R:
a new M icrom ax Controller.
Carburizing
This latest member o f 't h e M icrom ax fam ily auto Tempering
matically holds electric-furnace temperature firmly and Nitridlng
steadily in line, in spite o f changes in furnace load, High-Speed Steel
soak temperature, outside temperature or any other
On Lead Pots
upsetting factor— large or small.
On Salt Pots
Developm ent o f a Controller to do this jo b for elec On Muffle Furnaces
trically-heated _ equipment was a long, difficult task. On Lab Furnaces
On Home-Made
But the resulting instrument is extremely simple and Furnaces
foolproof, and has won its spurs in m any a plant during
the past 18 months. W e have withheld public announce
ment o f it until strenuous duty over long periods o f
time has showed its merit.
Easy To Start
T h e operator quickly and easily "sets” the Controller
for the particular furnace. From there on, M icrom ax TECHNICAL FACTS
carries on automatically. T h e equipm ent comes as a . D uration-A djusting T y p e M icrom ax Electric Control
single panel, with M icrom ax Controller and its acces gives (1) Fully-proportioned time action as distinguished
sory control unit interconnected and ready for connec from the m ore familiar fully-proportioned positioning
tion to primary element, power supply and furnace. action; (2) A utom atic droop correction; (3) Overshoot
The primary element m ay be either therm ocouple, control.
Rayotube or Therm ohm , depending on the require
D A I -M E C gives the same close, steady control o f
ments o f the process.
electric furnaces that a valve-positioning M E C gives to
Catalog N -00A (2) "M icrom a x E lectric Control, fuel fired furnaces. B ut it doesn’ t throttle the electric
Duration-Adjusting T y p e ,” answers m ost questions power in order to balance its input against the heat de
about this new autom atic Controller. If, however, you mand. Instead, using an on -off contacting system , the
want individualized inform ation, outline the control Controller regulates the durations o f time during which
situation and an L& N engineer will be glad to help. power is on and off. H ence its nam e, D uration-A djusting
Type M E C .
& N O R T H R U P C O M P A N Y , 4920 S T E N T O N AVE., P H I LA. 44, PA.
For E very A m erica n
Jrl Ad N-00A(5)
H E A T -T R E A T IN G F U R N A C E S
INDUSTRIAL AND ENGINEERING CHEMISTRY
mm
A rriv in g re a d y for In sta lla

tion In a C la ss I, G rou p D,
l o c a t i o n , h e r e 's a r a c k
asse m b ly of G -E o il-im
mersed 4 4 0 -v o lt co m bin a
tion starters. Suitable for
m otors up to TOO hp, these
starters operate with all con
tacts at least six Inches
u nde r oil.
Eleven G -E com bi

n a t io n a lr-b re a k
starters for C la ss I,
G r o u p D, h a z a r d s
controlling G -E e x
p lo sio n -p ro o f pum p
m otors in a g a s-c y -
cllng a n d repressuring
plant.
September, 1944
Explosion-proof motors and suitable

control eliminate the need for special
isolation to meet chemical plant haz
ards— both explosion and corrosion.
U I L D I N G separate vaults or housings to isolate
B electric equipment from explosive and corrosive
atmospheres is costly. It wastes engineering and installa
tion time. It complicates maintenance. So do cumber
some remote-driven arrangements.
G -E builds motors and control that are specially

designed to be “ put on the s p o t " in Class I, Group D ,
locations— without any extra safeguards. T h ey carry
their ow n protection. T h ey are “ at h o m e” indoors or
out in the open. Y o u locate them solely on the basis
Throe e xp lo sio n -p ro o f squirrel-cage motors driving raw -
of convenience, and econom y o f space and machinery. m aterial pum p s in a plastics plant. G -E w ou nd-ro tor an d
so lid - and h o llo w -sh a ft vertical m otors are a lso av a ila b le
in e xp lo sio n -p ro o f construction.
NAME YOUR HAZARD
W hether i t ’ s high-octane gasoline, solvents, mag

nesium dust, acid fumes, or outdoor service, there’ s a
way to meet your conditions w ith a com bination o f G -E ' To G e n e ra l Electric C o m p a n y , Section B 6 6 2 -1 2 ,
S ch e n e c ta d y 5, N . Y.
motors and control. For General Electric builds the
most complete line o f m otors and control devices for use Jan« I ’d h'ice fo h a v e a c o p y o f yo u r
h e lp fu l n e w 2 4 - p a g e bulletin,
in Class I as w ell as Class II locations.
M o to rs a n d Control for H a z a rd o u s Locations/ '
And— assistance in selecting the m otors and control (GEA-4J31)
which best combine security and econom y on your jobs Nam e...__
is right w ith in reach o f your telephone,
C om pany..
specializing in chemical plant applications
to have you call on them for help via the A d d re ss
near you.
32
INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36. No 9
engineers and technical men are of

invpluable benefit on problems of design, construction and
operation of lead lined apparatus. These men can do much to
effect savings of important sums of money on your lead burning
work. They will gladly assist in the preparation of tightly drawn
specifications, which will insure good quality lead work.
Permanently located lead burning shops at Cambridge, Mass.,
Buffalo, N. Y., Garwood, N. J., Baltimore, Md., St. Louis, Mo., and
Houston, Texas, simplify contact. Remember that consultation implies
np obligation on your part*
THE ANDREWS LEAD CONSTRUCTION CORP.

tEAD BURNING CONTRACTORS « §p | M | l 1 2 0 B R O A D W A Y , N E W Y O R K 5, N . Y,
PROPERTIES
B oiling range (760m m , H g ) . . .39.3-40.1 °C.
V isco sity, liquid, (760m m , Hg)
— 2 0 ° C .............................. ......... 0.6 7 6
2 0 ° C ........................................ 0.441
V a p o r Density, 4 0 ° C .......... 0.2 0 6 Ib./cu. ft.
P o u n d s per ga llo n , 2 0 ° C ....................1 1.08
Critical T em perature........................ 2 1 6°C.
Critical P re ssu re . . . ; .................... 7 0 8 psl.
Flam m obility— Underw riters’ Laboratory
R a tin g ............ .Non-flam m able
Freezing P o in t ............................ — 96.7°C.
Dielectric Constant, V a po r, 60-140 °C . .1.17
Heat Conductivity, Liquid,
68°F.......0.0 9 9 BTU/sq. ft. (sq. ft.) (°F/ft.)
Surface Tension, A g a in s t A ir
...................... 26 .5 2 dynes/cm . (18°C.)
C o lo r .....................................water-white
•• - .. m m O d o r ......................... pleasant
Solubility, (20°C.),
Powerful, low -boilin g in w ate r.............. .. .2.000 gm ./lO O gm.
water i n 0.162 gm ./lO O gm.
brom ine i n ..................1.96 gm ./lO O gm.
chlorinated solvent paraffin (m.p.50.5°C.)
(15°C.) i n ................. 1.0 gm ./lO O gm .
The stability, so lve n t p o w e r a n d relative n o n -fla m m a b ility o f M e th y a lc o h o lin ................. infinite
ether i n .................... infinite
lene C h loride are properties w h ic h h a v e m a d e th is ve rsatile product chloroform in ............infinite
ad ap tab le to m a n y lines o f w o rk , a n d h a v e stim ulated d e ve lo pm e n t
of m a n y n e w processes a n d techniques.
M e thyle ne C h loride is extrem ely stable, even in the v a p o r ph ase, a n d can fibrous materials such as paper; and
be e co n o m ica lly recovered from extracted residues b y sim ple, lo w -te m p e ra- as a biological preservative, parasiti
ture distillation. It m ix e s w e ll w ith other c o m m o n o rga n ic solve nts, but o n ly cide, fire-extinguisher m edium with
sligh tly w ith w ater; hence it can be e m p lo y e d in m a n y preparation s w h ere c a r b o n te t r a c h lo r id e , ce m e n t in g
the u se o f m ixed so lve n ts is d esirable. The chlorine a to m s can be replaced medium for C ellophane and other
by a va rie ty o f rad icals, m a k in g p o ssib le n u m e rou s syn th eses. products; m anufacture o f artificial
leathers, patent leather finishes, m o
tion picture film, etc.
s o l v e n t for— nitrocellulose, cellulose fu rth er esté rifica tio n o f cellu lose Consult us now on how Du Pont
acetate and other cellulose esters; esters. R eacts with ethylene mercap- Methylene Chloride can assist you in
polyvinyl acetate and other resins; tans, alkali organom etallic com your work. A t present, its use is re
photographic film; waxes, fats, oils, pounds, diphenyl, and m any others. stricted to war production, with a few
greases, rubber, bitum en, p itch , al exceptions. Limited amounts, how
g e n e r a l for— rem oving paint from
kaloids, and m any other organic ever, are available for research work.
wood, m etal and w ool waste; ■de
com pounds. For further details just write: E . I .
inking newsprint; raising the flash
du Pont de Nemours & Co. (Inc.),
e x t r a c t a n t for— animal fats; leather point o f lacquers; adhesives; de-
Electrochemicals Dept., Wilmington
scraps; w ool; caffeine; vegetable and waxing petroleum oils; solvent for
98, Delaware.
mineral oils; cocoa butter; perfumes quick drying lacquers and varnishes;
and flavors; dyes, lactic, acetic, p ro recovery o f phenol from water; in EVERY B O N D B R IN G S V IC T O R Y N E A R E R —
pionic and stearic acids; and such p h a rm a c e u tic a ls; d is in te g ra tin g B U Y ANO TH ER TO D A Y!
materials as hops, rice bran, rolled
oats, grape seeds.
R e f r i g e r a n t for— heat transfer equip

ment; refrigeration and air condi DU P O N T
tioning systems.
d e g r e a s i n g a g e n t for— use with vari

ELECTROCHEMICALS
ous metals, including aluminum foil
where rapid drying is essential.
for— the
PRO CESS A G E N T a n d R EA G E N T
BDPDNj U.S. PAT. O f f
preparation o f various com pounds, BETTER THINGS FOR BETTER L I V I N G . . . THR O U G H CHEM ISTRY
lJhoto courtesy Lanteen Medical Laboratories, Inc.
A t t h e Lanteen Medical Laboratories, Inc. in Chicago, the century o f experience in filtration engineering stands back
final step in the processing of one of their products is to o f every job.
pass it through the Sperry Filter Press shown above. Con Whatever your filtration problem or needs, call on Sperry.
taining six, 200 mesh stainless steel screens, and equipped If you wish, send a sample o f your material for testing. A
with electric pumps, this press is a vital part in maintaining complete laboratory analysis will be made, and unbiased
the purity and uniformity of their product. Another example recommendations submitted without obligation.
o f Sperry Filter Presses in use ■where results count. Sperry also furnishes filtration media o f all types,
Sperry not only manufacturers filter presses for use in for every purpose.
such laboratories, but designs and manufacturers presses
of all sizes, for every industry and for every use. Sperry
Filtration Engineers analyze each problem separately, make
F R E E . . . V a lu a b le d ata and charts on
specific recommendations, then design and manufacture the industrial filtration in this booklet. W rite
equipment to do the job . . . and do it right. Nearly a half- Sperry fo r your copy.
D. It. SPERRY & COMPANY BATAVIA, ILLINOIS

Filtration Engineers for Over 4 6 Years
EASTERN SA L E S REPRESENTATIVE
W ESTER N S A L E S REPRESENTATIVE
Henry E. Jacoby, M.E. B. M . Pilhashy
205 E. -42nd St., N e w York 17 Merchants Exchange Bldg.
Phone ,• M U rra y H ill 4-3581 San Francisco 4, Calif.
Phone: Do 0375
DIFFERENT
-O LD BE TRIED AND PROVED
R ep u b lic E L E C T R U N IT E T u b es Tubes are SAFE, SOUND, E A S Y matter w here the tube may be cut.
are m ade by the m ost m o d e m p r o c TO IN S T A L L A N D ECONOM ICAL
ess o f tube manufacture—a p rocess Sm ooth surface free fro m c o r r o
TO USE.
which has u n d ergon e constant im sion -in vitin g scale and scale pits.
C on sider what those features can
provement sin ce first developed by U n iform strength insured by the
mean to Y O U :
Republic Steel and T u b es D ivision . m ost rig id tests.
Thus, th e s e p re s s u r e tu b es g a in C onsistent uniform ity o f diameter,
W hat E LE C TR U N IT E T u b es have
distinct advantages over tubes made w all thickness and concen tricity
accom p lish ed in savings f o r others,
by o t h e r m e th o d s . T h e y ’re n ew that enable E LE C TR U N IT E Tubes
they a lso can d o fo r you. W rite us
enough to he d ifferen t. to slide in freely and r o ll rap id ly to
fo r the com p lete story.
tight, n on -leak in g joints.
On th e o t h e r h a n d , th ey’r e o ld
Full n o r m a liz in g th at p r o v id e s REPUBLIC STEEL CORPORATION
enough to he tried and proved. Twelve
uniform ly h ig h ductility at every STEEl AND TUBES D IV ISIO N • CLEVELAND 8, OHIO
years o f service (m o re than 120,- Berger M anufacturing Division
p o in t in the tube— easy expanding, Culvert Division • Nilea Steel Products Division
000,000 feet installed) in all types Union Drawn Steel Division • Truscon Steel Company
r o llin g and b ead in g o f tube ends n o Export Department: Chrysler Bldg., New York 17, N . Y.
of steam generating and heat trans
fer equipm ent in practically every
industry and under all con d ition s
of op eration have dem onstrated
conclusively that E L E C T R U N IT E
-B U S
ELECTRUNITE
i- ItfR. U. ?, fat. OtT
B O I 1ER , C O N D E N S ER A N D
HEAT E X C H A N G E R TUBES
INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No. Ç
BR IST O L MODEL,
2 9 3 TEMPERATUR!
CONTROLLER
B R IS T O L
S Y N C H RO
M P lR I P ^ E E S D IA P H R A G M
CONNECTING
CONTROL ^
HERE VALVE T U B IN G
ST EA M
SUPPLY
TEM PERATURE-
SENSITIVE BULB
BUT h o t
BRISTOL MODEL 293
TEMPERATURE
CONTROLLER
FOR CONTROLLING
T H E TEMPERATURE OF
A LIQ U ID IN A TAN K
-f a / M
ARE THE VARIABLES
TH E M ODEL 9 3
CAN C O N T R O L
Temperatures
from - 1 2 5 ° to 1000°F.
Pressures
from full vacuum to 4000 psi.
Liquid Level
in open or closed vessels.
Humidity
rFAMB.
from 0 to 100% R .H .
A ICOHTROLtf SUPPLY! Flow

o f steam, gas, air and liquids
through pipes.
B ristol’s new M odel 93 A ir-operated Controllers M odel 93 is a small, com pact, sim ple instrument
are designed for those cases w here you want varia having a precision m easuring element and B ristol’s
bles put under accurate control, but w here chart F ree-V ane Control the frictionless system w h ich is
records are either not required or are otherw ise the m ost accurate o f all air-operated designs. Y o u
provided for, get an absolutely true visual indication o f results.
A W rite fo r a fre e copy to 1 1 0 B ris to l Road, W a te rb u ry 9 1 , Conn.
AUTOMATIC CONTROLLING AND RECORDING INSTRUM ENTS
THE B R IS T O L COMPANY, W a t e r b u r y 9 1 , C o n n e c t i c u t
T h e Bristol Co. o f Canada, Ltd. B ristol’s Instrum ent Co., Ltd.
T oronto, Ontario L on don , N. W . 10, E ngland
September, 1944_________________I N D U S T R I A L AND ENGINEERING CHEMISTRY 39
There’s a great deal

more to Houdry catalytic
"know-how” than has
yet met industry’s eye...
M T«r » -■ vv;-' ■; ---- ga»
• - • - - i
Twenty years of catalytic researc!h have limiteid possibilities of catalytic science,

centered 'in Houdry Laboratories <
3 huge Forwc ird looking manufacturers are invited
amount of data on all types or c<italytic to coiisu lt with Houdry's distinguished
operations. Much of this priceless 1knowl- staff of ch e m ists an d e n g in e e rs for
edge is applicable to industries w hich m'utucil exploration of how catalysis m ay
hpye not yet utilized the almo st un- be us<îd to advantage in their opérations.
HOUDRY PROCESS CORP O R A T I O N , \V I L M I N G T O N , DELAWARE
F
BBSS'
H O U D R Y
CATALYTIC
/ fi
'^ ¡ C E S S E S
Houdry Catalytic Processes and the T.C.C. Process are available through the follow ing licens
ing agents to all American refiners, subject to approval by the United States Government.
B EC H T E L-M cC O N E CO RP. ' THE LUM M US COMPANY E. B. B A D G E R & S O N S C O .

L o s A n g e le s , C a lif o r n ia N e w Y o r k C ity, N e w Y o r k Bo ston , M a ss a c h u s e tt s
e h in d th o se words stands a wealth o f
B meaning, for when Graver takes over a
steel plate jo b , i t ’s completed as ordered. N o
Graver plants are equipped to build prac
tically any type o f steel plate equipment your
blueprints call for . . . including pressure ves
matter how intricate, each detail is handled
sels, stills, towers, vacuum vessels, tanks,
to meet the m ost rigid requirements..The con
breechings, and chemical process equipment
trolled circulation drying bin, illustrated
o f a wide variety.
above, and constructed for one o f the coun
try’s leading chemical companies, is but one W e will be glad to submit estimates without
example o f Graver’s adherence to customer obligation. Just mail in prints or specifica
tions covering your requirements.
F a b r ic a t e d S t e e l P la t e D iv is io n
GRA-VER TAKK & M F § CQJNÇ.

4 8 1 1 - 4 5 T o d A v e . , E a s t C h i c a g o , In d .
NEW YO RK C A T A S A U Q U A , PA. C H IC A G O TU LSA
IS THE
OF MANY A PROCESS
• The great chemical industry is probably the Wbrid's largest user o f AIR—
because so many o f its productSvrequire air in process of manufacture.
That's where FA N S come into the picture. Many a process depends entirely on
continuous handling of air (or gas) for its success, and that means R ELIA BLE
fans. Since dependability is essential, it pays to buy fans with extra weight, over
size construction—features for many years o f “ Buffalo” Fans for the chemical
industries. Where one five-minute stoppage may mean disaster—it pays to buy
the fans on which many of America’s largest chemical process manufacturers
standardize.
BUFFALO FORGE C O M P A N Y
153 M O R T IM E R ST BU FFALO, N E W Y O R K
C A N A D IA N B L O W E R & F O R G E C O ., L T D ., K IT C H E N E R , O N T .
‘ First
For Fans’’
FANS FOR PROCESS PI.ANTS
42 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No.
Need Piping Equipment. . .

Any Kind?.. . CALL CRANI
O N E S O U R C E OF S U P P L Y . . . O N E R E S P O N S I B I L I T Y F O R ALL M A TER IA L!
It’s as easy as that! Call Crane w hether you need a in g installations. O n e o rd e r cov ers all you r needs
sin gle fitting, o r all the materials fo r a com p lete at any on e time. M aintenance w o r k — h a n d lin g parts
p ip in g system. Valves — fittings — p ip e — fabricated stock s— getting em ergen cy replacem ents—all such
assem blies— and p ip in g a ccessories— fo r p o w e r o r p roced u res are sim plified by takin g advantage o f
p ro ce ssin g fluids— fo r h igh o r lo w pressure— your Crane complete materials service. But, m ainly, you r
Crane Branch supplies all your requirem ents fo r installations get full benefit o f Crane C o .’s 89-year
any service. In brass, iron , and steel—you w ill find experien ce and leadership in p ip in g equipm ent.
that the Crane line p rovides the w o r ld ’s ^greatest
selection o f p ip in g equipm ent.
C R A N E C O ., G eneral Offices: 8 3 6 South M ich i
O n e repon sibility fo r quality and craftsm anship gan Avenue, C h ica go 5, Illin ois . Branches a n d
o f materials helps assure your gettin g the best p ip Wholesalers Serving A l l Industrial Areas.
ONE STANDARD
OF Q U A L I T Y
From end to end, p ip in g systems
are m o re d ep end able w hen on e
h igh standard o f quality guards
every part. T y p ica l o f Crane qual
ity, h ere’s what Crane Ir o n Gate
Valves add to you r p ip in g : Strong
b o d y section s resist severest line
strains. Straight-through ports
give stream line flow . A deeper
stuffing b o x lengthens p a ck in g
life. Extra lo n g gu ides k eep disc
travel true, w hile finest d esign in
every part assures lo n g life and
sm ooth , positive operation.
S T A N D A R D IR O N BO D Y
W EDG E GATE VALVES
Installation in Chemical
Processing Plant... Piping
M aterials 100% Crane.
VALVES • FITTINGS • PIPE

PLUMBING • HEATING • PUMPS
September, 1944 INDUS
These Alum inas

are suitable for c a ta ly tic p u rp oses
A C T IV E T Y P E S
Activated Alum inas
These Aluminas are catalytically active
materials when used alone or as catalyst
Name your needs for a catalyst and it’ s quite likely they can he m et with supports. The pores o f the granules occupy
50 % o f the particle volume.
the help o f A lorco Aluminas. Y ou can have them as fines, powder, granules
At* K o l l e n n k . . 4 . 1 . ~ ___ 1_____ 1__ __________ 1 TVT . 1
or balls, as indicated b y the photograph. N ote the wide-range o f proper
*1 1* Hydrated A lum ina C-730
This Alumina is active after an original
ties suggested h y the list o f aluminas suitable fo r catalytic purposes. heating to 500° F. It has finer particles and
Alorco Aluminas are serving as catalysts, carriers and auxiliary greater surface area than are obtainable
catalysts in the production o f such vital materials as high octane gaso with ground natural materials.
line, synthetic rubber, plastics and explosives. B eing pure, there is less IN A C T IV E T Y P E S
likelihood o f side reactions and poisoning. T h e y are strong and stable. Tabular Alum inas
They h ave extremely large surface areas. Im pregnated with active These Aluminas have a range o f porosity and
catalysts, th ey help increase yields, perm it lower operating tem pera are a form o f granular corundum produced
tures and aid in reduction o f costs. from pure aluminum oxide. They are highly
resistant to abrasion and are unaffected by
Speed and yield o f catalytic reactions are im proved with dry gases high temperatures.
or dry organic liquids. A lorco Aluminas are helping here, too. Machines Monohyc/rafec/ A lum ina D-50
charged w ith activated aluminas d o this jo b automatically. This is a substantially inactive powdered
Select th e materials m ost likely to m eet you r requirements and we’ll Alumina having crystal structure different
Bend you samples for trial. W rite A L U M IN U M C O M P A N Y OF from the usual Aluminum Trihydrate. The
AM E R IC A (Sales ‘ Agent f o r A l u m i n u m O r e C o m p a n y ) 1911 G ulf average particles are 1 micron in diameter.
Building, Pittsburgh 19, Pennsylvania.
ALUMINUM ORE COM PANY

ALORCO
INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No.
C U R E S O iD - S m E
m v E m u m / iR E s
,;...... ........ . .... ._ .._

S tr e a m lin e d <ri f a /d P a ssa g e
/V o fh in q to C /o q ih e O p era tin g M ec h a n ism -
HI L L S - Me C A N N A SAUNOERS PATENT
NO L E AKS OR D R I P S
NO G A L L I N G — NO “ F R E E Z I N G ’ ’ DIAPHRAGM VALVES
NO S T I C K I N G - NO W E A R OR
C O R R O S I O N — NO W I R E The Lo gical Valve for the Chem ical and Process
Industries— for handling A cid s, A lkalis, A lco h o l,
D R A W I N G — NO R E S E A T I N G
Solvents, F o o d Products, Fruit Ju ices, Sem i-Solids,
P ^ u ic iica iL q S o . M a in te n a n c e Viscous Substances, D yes, H ot or Cold W a te r, G ases,
etc.
W hen you p u rch a se v a lv e s, y o u n a tu ra lly are

l o o k i n g f o r d e p e n d a b le s e r v ic e , c o u p le d w ith th e
east a m o u n t o f r e p a ir s . B u t, d o y o u g e t it? W it h
H ills - M c C a n n a V a lv e s y o u a re s u re o f u n fa ilin g
r v ic e fr o m th e start, b e c a u s e th ey h ave n o t h in g
t o g o w r o n g . T h e i r d e s ig n is b a s e d u p o n art e n tir e
ly n e w a n d d iffe r e n t c l o s i n g p r i n c i p l e as illu s tr a te d .
T h e y are c o m p o s e d o f 3 s im p le u n its ; a b o n n e t —
d ia p h r a g m — a n d b o d y . T h e w o r k i n g p a rts a re in
th e t o p u n it, th e d ia p h r a g m in b e t w e e n . . . s e p a r
a tin g a ll m o v i n g p a rts f r o m th e flo w .
T h a t ’ s th e d if fe r e n c e — a n d th ey c u r e a ll c o n v e n
tio n a l v a lv e n ig h t m a r e s . P r o v e it t o y o u r s e l f o n
that n e x t v a lv e r e p la c e m e n t . Y o u ’ ll b e a m a z e d at
S H U T S T I G H T
th e l o n g , t r o u b le -f r e e s e r v ic e H ills - M c C a n n a
e v e n o v e r a c h u n k o f s c a le V a lv e s w i ll g iv e y o u .
N ote h ow the diaphragm cushions over the C a t a lo g V -4 1 on r e q u e s t
weir (the seat o f the valve) pinching o ff the
flow. Closure is positive and com plete. N o need
for stem packing as with valves depending upon
HILLS McCANNA CO.
metal-to-metal surfaces for a seal. W ork in g 2345 NE LSO N S T R E E T • C H IC AG O 18, I L L .
parts are com pletely isolated by the diaphragm . M a n u f a c t u r e r s of
C h e m ic a l V a lv e s • A i r 8, W ate r V a lv e s • M a rin e V a lv e s • C h e m ica l
P ro p o rt io n in g P u m p s • F o rce -Fe e d lu b r ic a t o r s • M a g n e s iu m C a s tin g s
WHERE D O THESE
^ e ^ / i a c t & u e o
FIT I N ? _
Experience has proved that super-refractories by

Carborundum are a valuable aid to efficient furnace
operation. For consideration we suggest:—
CARBOFRAX" (S ilico n C arb id e) in fur

naces for the reduction and refining o f non-ferrous
metals and in the heat treatment o f both ferrous and
non-ferrous metals. It is so employed because o f its
high thermal conductivity, excellent refractoriness,
minimum tendency to spall and outstanding re
sistance to mechanical abrasion.
"ALFRAX" (E lectrically Fused A lu m in u m

Oxide) for thin wall muffles, retorts, hearths, piers,
pier walls and burner blocks. T his product has a
thermal conductivity approximately three times that
o f fireclay, high hot strength, low electrical con
ductivity and is chemically stable.
"M U L L F R A X " (E lectric Furnace M u llite )

for piers, support arches and other load bearing
constructions and linings for electric furnaces melt
ing both ferrous and non-ferrous metals. It is valu
able for these applications because o f its resistance
to elevated temperatures, to deformation resulting
from heavy loads, to thermal spalling and slag
erosion.
T h e services o f our Refractories Engineers are at

your call to help select the right refractory for your
jo b . T h e C arborundum Com pany, R efractories
Division, P enh A m b oy, N . J.
Salr r B rr d r A CI' i,' aK0’ 0I>lli l“ <,cll'' ,ira’ 11>T
e troit’ a ® « 1“ *» B o s to n , Pittsburgh. D istrib u tors: M c C o n n e ll Sales an d E n gin eering C orp ora tion , B irm in gham
Christy X treb n ck C o m p a n y , S t. L ou is, M o .; H arrison & C om p a n y, Salt Lake C ity , U ta h ; P a cific A b ra sive S u p p ly C o m p a n y, I,o s A ngeles, San Fran cisco C a lif-
D en v er Fire C lay C o m p a n y , E l P a so, T e x a s; S m ith-S harpc C o m p a n y, M in neapolis, M in n.
CARBORUNDUM T R A D E -M A R K
(C a rb oru n d u m , C a rb ofrax, AI fra x and M u llfrax are registered trade m arks o f and in dica te m a n u f a c t u r e ' r b . r — t, t. r •___________t
Stop scale á
w
et istops production!
t. PLANT EFFLUENT W IT H O U T C A L G O N
REFINERY CONDENSER TUBES closed with
scale. C a lgon treatm ent now keeps these clean right
t o the m etal. N o prod u ction tim e is w asted cleaning ^ ^ ^ !^§)
out tubes.
I4 0 'F I5 S ° F I7 6 “F I94-*F 2 I2 ° F
TEMPERATURE O F OPERATION
CROSS SECTIONS o f w ater heater test c o ils show
2. PLANT EFFLUENT WITH CALGON
e ffectiven ess of C algon in preven tin g scale at v a r i
ous tem peratures. (6 m on th s.)
1.
2.
P lant effluent w ith ou t Calgon
Plant effluent w ith C algon.
@ @ p @ ®
HO°F I5 8 ° F 1 7 6 *F I9 A * F 2I3*F
T E M P E R A T U R E OF O P E R A T IO N
M O S T water used in cooling changers, gas coolers, furnace jackets,

systems is either scale-form am m onia scrubbers stay clean and
What you can do ing or corrosive. W ith ou t water free from scale with Calgon-treated
treatment, it is merely a question o f water. Corrosion is reduced to such
with Calgon time as to how soon heat exchangers, a degree that it ceases to be a serious
condensers, pipe lines and valves problem .
1. Prevent scale in cold or hot will clog up and stop production.
water lines, heat exchangers, If you have a scale or corrosion
o th e r p ro c e ss W ith C algon* water treatment problem , give us a call. O ur engi
you can prevent scale form ation and neers can quickly show you what
rlinesS to' »such»an .extenti»that
r;r,it control corrosion. O ne to five ppm. Calgon can d o for you. T h e treat
o f Calgon will stop deposition o f ment is simple and can be started
ceases to be a serious probien^
3. Soften water 100% without calcium carbonate in most types o f quickly w ithout disrupting p rodu c
precipitation of lime soap m process equipm ent. T w o to ten ppm. tion.
any soap and water washing o f Calgon controls corrosion. C on T h e cost o f C algon T reatm ent is
4r°P°revent «red-water" caused densers that formerly plugged with low. In fact, it usually pays for .itself
by either dissolved iron or scale in 30 days’ time, have been and starts saving m oney in a few
corrosion in the P '?e. mf . kept clean indefinitely. Heat ex months. W rite for com plete details.
5 Disperse finely divided met
al oxides and salts m cera m e
processes, paper making, oil-
well drilling mud, etc. *Calgon is the registered trade-mark of Calgon, Inc., for its glassy sodium phosphate products.
Á S U B S ID IA R Y O F H A G A N B U IL D IN G
H A G A N C O R P O R A T IO N PITTSBURGH 30, PA.
WET GRINDING HARDINGE MILLS— INDUSTRIAL

Ball o r
Material Object Capacity Product Pebble REM ARKS
Ground Tons per Size Load '
hour Pounds
As neutralizing Mill operated in closed circuit with.
Limestone agent_______ 9 0 % -2 0 0 M Classifier A
Product mixed with dolom ite to
Furnace Lining 9 5 % -1 0 0 M o f open hearth furnaces
4000
Iron Oxide Balls Mill is ooerated in closed ci
R ecovery of Minus Minus 4M R ecover about 180 0 # ol
Brass Ashes & Slag metal from slag & Fines charged on
A loxite Plate Glass Minus
(Abrasive very hard) Polisher 60 M .
6'x22" W hite Limestone

Pebble (W hiting)_____ Paint Filler
7/x36/
Pebble Barytes Paint F2 A B O V E data taken from Bulletin 41
S'x30' Barytes
visualizes the wide variety of materials
Pebble (Contain.^ the Conical M ill will handle and the
8'x36'
Pebble Calcine* purposes for which it may be em
ployed. There is one best way to grind
to ten mesh, and quite a different way
when a 325 mesh product is wanted.
Some want a granular product, others want

extreme fines. Both are readily obtainable
using the Conical M ill together with aux
iliary equipment that accomplishes this're-
sult without dust and waste of energy.
DRY GRINDING WITH HARDINGE MILL AND AIR CLASSIFIER

Ball o r H . P.
Capacity Pebble Required
Size and M aterial Ground O bject T ons/ Feed Mesh Load REM ARKS
Type hour Size P roduct Pounds
Mill Mill Fan
2'x8"
Ball Lim estone Chemical Reagent .20 1 W 6 5 % -2 0 0 400 Balls 2 .0 1 .5
2'x8" 9 9 .7 %
Ball L im estone | For Filler ■05 1 !4 " -3 25 400 Balls 2 .0 .5 1 8" H ardinge Superfine A ir Cf
3'x2d" I T o make 16C0
Ball Beryllium Frit | Beryllium Oxide .15 W' 9 8 % -2 0 0 Balls 8 2 M aterial hard and abrasive M
3'x24"
Ball Sillimanite (H ard) | R efractory Cement .25 14" 8 0 % -1 0 0 1400
Ball I Fluorspar Chemical M a n u factu re! 1 .1

5'.\36" M aking Filler for
Ball Bank Sand Shingles
6 'x36" I Pulverized Fuel U
Ball I K entucky Coal (Q;
H i R i n i v f i r c o m p a n y
INCORPORATED
Y O R K . PE N N SYLV A N IA , M ain O ffice a n d W orks
N E W Y O R K , 122 E. 4 2n d S t. C H IC A G O , 20S W . W a c k c r D rive SAN F R A N C IS C O , 501 H o w a rd S t. T O R O N T O , 200 B a y S tr e e t

Long and broad experience has given B a d ge r an

invaluable "k n o w h o w " of every phase of process
engineering and chemical plant designing.
• But B ad ge r is not a slave to tradition. B a d ge r is

not content with the application of "established
procedure" in process engineering, design and con
struction. B a d ge r invokes the "k n o w why" of chem
istry and engineering . . . reasons from fundam entals
. . . and through a flexible ap proach improves proc
esses and often evolves entirely new methods and
designs. Such advances m ay translate into lower
plant costs, greater m anufacturing efficiency, a better
product — or perhaps all three.
• Each B ad ge r assignm ent becomes one of individual

study — and usually one of accomplishment that lives
through years of customer satisfaction.
e. b . B a d g e r & s o n s co .
B O S T O N 14 E S T A B L I S H E D 1841
NE W YO RK • P H IL A D E L P H IA . SAN FRAN CISCO • LONDON
PROCESS ENGINEERS AND CONSTRUCTORS FOR THE

C H E M IC A L , PETRO-C H E M IC AL AND PETROLEUM IN D U S T R IE S
...th e Badger plus-service team

N U M B E R FOUR OF A SERIES ON VALVE SELECTION
I i i RELIEF
# VALVES
« ■
W HETHER for power, marine, and production-run quantities make
chemical process, petroleum, Edward relief valves surprisingly low
refining, hydraulic or general indus in initial cost and exceptionally eco
trial services, the most important con nomical over the long pull.
sideration in selecting relief valves is
absolute dependability. Since every valve is factory set for
the required relieving pressure, in
Edward relief valves protect expen
quiries should include information
sive equipment against sudden and
concerning service, frequency of
unusual pressure increases in all kinds
operation, and relieving pressure.
of services. Functional design, with
only a few parts all built to extra Regularly available in hi in. to 2 in.
ordinarily high precision standards, sizes inclusive.
IMPORTANT POINTS
BODY BONNET
Drop forged heat Rolled carbon steel, with
treated steel. Dis body an d locknut joints
tribution of metal carefully ground for tight
fit.
around body flow
areas equalized so B E A R I N G P L UG
t h a t i n h e a t in g H ardened stainless
and cooling there steel. Point contact
is n eg lig ib le dis always permits disk
tortion. to resea t prom ptly
and tightly.
BALL DISK
Hardened corro
SPRING s io n r e s is t a n t
stainless steel, accurately
Fine carbon or stainless steel ground for tight seating.
spring, adequate length for
full area when discharging. SEAT
Design of adjusting mechan Renewable EValloy stainless
ism m a k es p o s s ib le c lo s e steel, with sturdy shoulder
a n d full th reads e n g a g in g
pressure regulation. body. FIG
Is Edward BETTER VALVES Catalog 1 0 2 in your files?

If not, write for your cop y today.
THE EDWARD VALVE & MFG. C O ., IN C . EAST CHICAGO, INDIANA
EDW ARD^VALVES
periment Station, M ethods reported are blanching and the addi-

This Month tion of chemicals, Both are successful in preventing vitamin A
reduction.
P assage of a solution of sodium silicate through an ion ex Production of 2,3-butylene glycol from wood hydrolyzed by
changer may presage a use that will prove as important in sulfuric acid is reported by Perlman. The use of Aerobacler aero-
dustrially as the well known application of ion exchangers for genes on wood sugar solutions of about 17% concentration suc
softening of water. This month Ryznar, of the National Alumí ceeded in converting about 3 5 % of the sugar into the wanted
nate Corporation, reports on the preparation and purification of glycol.
hydrous oxide sols by using ion exchangers. In the example cited
above, the sodium ion, in passing through an acid-regenerated
cation-exchange material, is left behind and the hydrogen ion
takes its place; a silicic acid is formed which is substantially free N ext Month
from other electrolytes. Beds of anion materials m ay also be
used to facilitate the substitution of hydroxyl ion for the anions l V /e fe e l like the man who, upon learning there were vitamins
present. Thus sols of A l(O H )3 can be formed by the passage of W in hash, said he was not surprised because he knew they
an aluminum chloride through an alkali-regenerated anion ex would show up there as soon as the cooks heard about them. We
changer. thus greet vitamins in canned meat and greatly fear the effect
When benzene is mixed with nonaromatics, it is a bad actor in their discovery will have upon the advertising messages as soon
a distillation procedure, for great difficulty is experienced in sepa as the facts become unknown to the copy writers. In October not
rating it properly. Richards and Hargreaves submit more data one but many vitamins will be shown to inhabit the can holding
on this problem with a contribution that covers the behavior of this G .I. staple; niacin, thiamine, riboflavin, and panthothenic
close-boiling hydrocarbons in the range 80° to 81° C. From the acid are crowded into the container along with the meat.
curves submitted, design engineers will be able to pick off the Strangely, the vitamins are not uniformly dispersed but seem to
points they will need when confronted with problems of distilla go hunting the cool corners when the processing starts.
tion in this range. Systems covered are benzene-dimethylpen- Tapioca ceased to come into these United States in its usual
tane, benzene-cyclohexane, and cyclohexane-methylcyclohexane. quantities shortly after the Japs conquered the Dutch East In
The laboratory technique of salting out water from solvents dies. Although for many of us it meant the end of a pleasing des
is little used industrially, but there are possibilities for it. Meiss- sert, the effect on industry was more serious. Some of this nec
ner and Stokes, of M .I.T ., have conducted an intensive investiga essary starch has been coming from Brazil and the Dominican
tion of this problem which is reported here. They show how to Republic, but the real answer seems to be in the development of
predict the maximum dehydrating power of materials used to dry a domestic substitute. A conference of users and government
the solvents. agencies, held two years ago, decided that all but 50 million
This month Kenneth S. Pitzer celebrates his first anniversary pounds o f the annual domestic demand of 300-400 million pounds
as a winner of the A.C.S. award in pure chemistry, presented at annually imported, could be supplied through the development
the Pittsburgh meeting in 1943. He pens a doleful note in his o f domestic starches. October will see reports on the possibility
contribution, for he is no longer working on the thermodynamic of using corn and sorghum starches for textile sizing and finishing,
properties of gaseous hydrocarbons. His recent work, though for paper coatings, envelopes, plywood, and food uses (by which
incomplete, is presented here with the knowledge that the data I hope the authors mean dessert).
will serve industry well. Measurements of viscosities o f coumarone-indene resins in
Continuing the emphasis on ion exchangers in this issue, W est the molten state have proved to be an accurate method of esti
ern Regional Research Laboratory reports on the use of these ma mating the molecular weights of the resin. A simple equation
terials as a means of extracting tartrates from grape wastes. For has been worked out to help in such determinations, and the
merly the United States used 15 million pounds of this material, method applies equally well to mixtures, which is an important
most of it imported from the Mediterranean region. War cut off help in commercial uses.
these sources, and our scientific brains went to work on the de The dehydration of organic solvents is always of paramount
velopment of domestic tartrates. As currently reported by the interest. Continuing a series, we will have an article on the con
Laboratory, the process consists in passing the slop from brandy tinuous countercurrent extraction o f water from methyl ethyl
and wine making through cation exchangers, followed by passage ketone by strong calcium chloride brines. While operating the
through an anion exchanger where, by carefully controlling the packed tower used in these experiments, the operators took time
conditions, tartaric acid is preferentially fixed. Regeneration of off to make a few pertinent recordings o f design data.
the anion bed with sodium carbonate enables the tartrates to be
recovered as the sodium salt.
Ways of preventing the deterioration of alfalfa, important
source of vitamin A through its precursor carotene, is reported in
this issue by a research team from the Kansas Agricultural Ex-
C a r b id e and Carbo n C h e m ic a l s C o r p o r a t io j v
Announces Commercial Production of
Allyl Alcohol
H H H
i i i
H -C = C -C -O H
i
H
Useful Raw Material for Synthesis

Vitally needed for war production,
A llyl Alcohol is now available in com
P H Y S IC A L P R O P E R T IE S
mercial quantities from Carbide and
Carbon Chemicals Corporation at prices M olecu lar W eight 58.05
B oilin g P oin t 96.9°C. at 760 nun.
that make it an important intermediate
V a p or Pressure 17.3 m m . at 20°C.
in industrial chemical synthesis.
Flash P o in t (o p e n cu p ) 85°F.
Although most of the production of Specific Gravity 2 0 ° /4 ° C 0.8520
A llyl Alcohol is for high-priority uses, W eight p er gal. 7.11 lb. at 20°C.
it is expected that substantial amounts W ater S olubility C om plete
will now be available for new users and

for expanded consumption by those al
ready using this chemical. and ease of fabrication at low pressures. Certain
Since it has both a reactive, unsaturated linkage allyl esters are used in perfumes, flavors, and as
and a hydroxyl group, it has unusual applications pharmaceutical intermediates. A llyl halides made
as a chemical raw material. It can be converted to from A llyl Alcohol are used for introducing the
acrolein, propionaldéhyde, glycerin chlorhydrins, allyl group into barbiturate sedatives. The halides
or glycerin itself. are also intermediates in the synthesis of the anes
A llyl Alcohol forms esters which can be polym thetic cyclopropane.

erized alone or with other materials to form resins W rite for samples and the latest information 011
characterized by great clarity, surface hardness, the availability and uses o f this versatile chemical.
C a r b id e and Carbon Ch em icals C o rpo r atio n

Unit o f Union Carbide and Carbon Corporation
3 0 East 4 2 n d Street |t|^ New Y o rk 17, N. Y .
PRODUCERS OF SYNTHETIC ORGANIC CHEMICALS

IN D U STRIAL an d E N G IN E E R IN G CH EM ISTRY
A - PUBLISHED BY THE A M E R IC A N C H EM IC A L SOCIETY
WALTER J. MURPHY, EDITOR
R e c o n v e rtin g T e c h n i c a l B r a in p o w e r assured of a steady flow of young technologists into research

and industry in the years immediately ahead of us.
E VENTS of the past two months have been so favorable It is not too early to develop plans for the speedy return
to the United Nations that it is no longer possible to
of highly skilled and technically trained research workers to
temper the optimistic outlook of the American public in re
our laboratories and of bona fide students and teachers to the
gard to the quick disposition of the German Army. While
Nation’s classrooms. This is not a plea for special privilege
some of our military leaders deplore this trend and warn of for a small minority. Not until the spring of 1944 did the
dire consequences attending partial reconversion of American
Selective Service and the Armed Forces openly recognize the
industry at this time, it is now obvious that the public and
necessity of holding technically trained men in the Produc
Congress are convinced that certain steps in this direction are
tion Army and in research and the advisability of continuance
desirable and will not affect nor even hamper or delay final
of college courses for exceptionally equipped technical stu
military success in both war theaters.
dents. Is it not equally true that such individuals are vital to
An insistent demand exists, now that we are rushing swiftly
the successful reconversion of America from war to peace?
toward the complete annihilation of the Wehrmacht, that
increases in production of civilian goods be permitted, pro
vided such action does not draw upon war workers in critical
Po stw ar E d u c a tio n
manpower areas. To many this may appear to be all there
is to the problem of reconversion. But is it? We can re T HE tragjp situation most of our colleges and universities
convert in a feverish and disorderly fashion, with the goal now find themselves in as the result of the demands of
nothing more than satiated markets of prewar goods and war are widely known but bear repeating. Teaching staffs
models, or we can plan intelligently for a long-term period of have been reduced in numbers and in some instances the qual
full employment and national prosperity. ity is below prewar levels. Graduate instruction and re
To achieve the latter means the use of technological brain search programs have been curtailed. Many of the better
power— lots of it. It will require that such brainpower be and younger instructors and assistant professors have found
mobilized at the earliest possible moment— well in advance it necessary to secure positions in government and industrial
of the final demobilization of our huge wartime army. research. In some instances professors past the retirement
As yet we have had very little information on demobiliza age and not in the best of health have been held at duties
tion plans. Selective Service principles employed in reverse that are strenuous and exhausting.
should guide us to a limited extent in demobilizing more than Will the younger men who are eminently qualified for
10,000,000 men and women from the Armed Forces. Such teaching positions but now engaged in design, plant operation,
plans admittedly would not be popular with large masses of research, and management, return to the colleges and uni
the American public, but they would prove to be a very ef versities in satisfactory numbers? Many of these men have
fective means of assuring a high degree of employment in the become accustomed to positions of authority, a more rapid
postwar period. Early this year our technical colleges and pace of achievement, and last, but certainly not least, im
universities were depleted of most male students, instructors, proved monetary returns.
and younger professors on the plea that a great emergency ex If the proposed plans for returning veterans are carried
isted and the need for young infantrymen transcended all through, the enrollment in our colleges and universities will
other considerations. In the light of developments of the exceed prewar registration by many thousands. Quite
past two months, does this exigency still exist? If it does, likely the build-up given the chemical and engineering fields
will proper provision be made to change present procedure during the war will serve to attract many students who
the moment the need for such individuals in our fighting might otherwise have elected to take courses in liberal arts
forces ceases? schools. The teaching burden in technical schools probably
One cannot read the recent statement of Erie M. Billings, will be proportionately greater.
Secretary of the A m e r i c a n C h e m i c a l S o c i e t y ’ s Committee Men equipped with the desire, ability, and requisite ex
on Professional Training of Chemists, on the pessimistic out perience to teach chemistry and chemical engineering are
look for adequate numbers of trained chemists and chemical relatively few. It will not be an easy matter to recruit them.
engineers in the postwar period (see Chemical and Engineer Certainly they are going to weigh the pros and cons carefully,
ing News, August 25, page 1357) without seriously doubting and our schools must be prepared to make attractive offers in
our ability to continue the expansion of the chemical industry order to secure highly qualified staffs. In the prewar era the
111the next decade. What is unquestionably true as regards general salary levels, especially for instructors and assistant
chemists and chemical engineers in the postwar era is un professors, were scaled well below the salary levels paid in
doubtedly true of all technical professions. Therefore, the industry for individuals with equal training and ability.
Possibility of creating new products, new industries with jobs Salary is not the only consideration but it will assume greater
lor millions, is not so bright as it would be if America were importance in the years ahead, and our colleges and universi-
777
ties will find it advisable to seek more funds for this purpose, Such sound public relation policies should be encouraged.
even at the expense of fancy brick and mortar. We are succeeding at an accelerated pace in portraying the
Considerable enlightenment and a new approach to work of the scientists to the public, but this is only one step in
chemical education in the postwar era should result from the the right direction. Teachers in our colleges and universities
symposium to be conducted under the auspices of the Divi can play important roles, encouraging broader viewpoints in
sion of Chemical Education at the New York meeting of the their students. We are attempting to visualize for our read
S o c i e t y . We a r e informed that this is but the first of a series ers the phases of chemistry which are the links between the
of special symposia dealing with this important problem. laboratory, the plant, and business, for we believe chemists
Representatives of the chemical industry have on numerous and chemical engineers will profit by possessing a broader
occasions criticized both the curricula of our technical schools picture of the field in which they are engaged.
and the products turned out. At New York and at subse In the final analysis, the problem is one in which technolo
quent meetings those who feel that they can contribute con gists must take a personal interest, for certainly present lead
structively will have an opportunity to be heard in a friendly ers of political, social, and economic thought in this country
atmosphere. Industry should prepare to voice at once its are not likely to extend gold-edged invitations to us to join
suggestions during the present transitory period. with them in dictating the master plan for the World of To
morrow other than to assign to us the usual role of supplying
S c i e n t i s t s a s Le a d e rs strictly scientific advances. To put it bluntly, if we are going
to participate, we will have to move in. If we do, let us be
UITE frequently in our scientific publications and meet certain we are prepared to do so constructively.
Q ings, in magazines and newspapers, we are confronted
with the intriguing idea that scientists should participate E n c o u r a g e t h e Y o u n g e r Re v ie w e rs
more actively in the formulation of political, social, and eco
ECENTLY we received a letter from a prominent mem
nomic policies. This stems undoubtedly from the fact that
in wartime the scientist is given greater public recognition.
The scientist personally prefers that this acclaim and recog
R ber of the A m e r i c a n C h e m i c a l S o c i e t y apologizing
for a slight delay in reviewing a manuscript caused by his
nition come to him in the days of peace, for he is directly in desire to permit two of the younger members of his staff to
terested in improving the lot of mankind, nottdestroying it. participate with him in a critical analysis of the contents.
That we favor the idea of chemists and chemical en No apology was necessary, for we would like nothing better
gineers widening the scope of their activities outside the than to see this procedure followed more frequently. While
laboratory and the plant should be apparent, but we do be we do not seek to encourage unnecessary delays in the prepa
come discouraged at times, especially when we are forced to ration of reviews, we do recognize the desirability of initiating
admit that such proposals are made almost exclusively by younger chemists and chemical engineers into the informal
members of the profession; seldom, if ever, by statesmen, group of reviewers who labor unceasingly and usually anony
politicians, clergymen, editors of the lay press. Indeed, we mously to maintain the high standards and international, pres
are forced to report a growing tendency on the part of repre tige of A m e r i c a n C h e m i c a l S o c i e t y journals.
sentatives of these groups to criticize the scientist for “ de
serting” science. Further, if we are to be truthful, we must M a r k e tin g Su rp lu se s
admit that as yet scientists are far from being in complete
HE War Production Board’s request that division di
agreement on this subject.
No one questions that science is powerful, yet the link be T rectors review estimates of reserves is wise. Market
ing war surpluses will be a tremendous problem at best.
tween the acknowledgment of its power and the acceptance of
Older members of the chemical industry remember the
scientists as co-partners in the formulation of political, social,
numerous dislocations which resulted in production after
and economic thought fails to occur to the extent that we
World War I; the task will be infinitely greater at the end
would like to see.
of this war.
Apparently, our rate of progress in this direction will not
It is encouraging to note a determined effort in Washing
be so rapid as we and others desire. We are inclined to
ton to devise ways and means of marketing war surpluses
forget in our enthusiasm for the cause that the chemists and
in an orderly fashion with due regard to the effect such
chemical engineers of this country number not more than
offerings will have on domestic industry and national econ
85,000 men and women out of a total population of more than omy. Momentarily the methods of approach have not j
130,000,000. Include all the technical professions and the been agreed upon. The Senate appears to lean to a
total in relation to the country’s population is still very in board of eight members; the House favors a single ad
significant. Numbers, then, are not the answer. ministrator. While boards and committees often ap
We are not without progress, but it must be recognized pear to be unwieldly and slow to act, there is a feeling
that it is a problem with many facets. At least we see en that the Senate plan would provide greater assurance that
couraging signs that companies are giving increasing recogni all viewpoints will receive proper consideration. On the
tion to individuals in publicity releases. We particularly other hand, one administrator would be able to act more
commend the Polaroid Corporation for the manner in which efficiently and with greater speed. We believe timing to
the work of Woodward and Doering in synthesizing quinine be one of the most important factors. Advice should be
was described. Largely because the personal angle was given sought, and perhaps the present pattern of industry ad
proper recognition, the lay press carried tremendous amounts visory boards now functioning could be employed to good
of publicity, the largest that we have seen given to any scien advantage. As administrator, however, we need a
tific achievement in recent years. Baruch, a Jeffers, a Bradley Dewey, or their equivalent.
R. G . CAPPELL
ration The Floridin Company. Warren, Pa.
E. G. HAMMER SCHMIDT
GASES Texoma Natural Gas Company,

Fritoh, Texas
by Solid Adsorbents W . W . DESCH N ER

J. F. Pritchard & Company,
Kansas City, Mo.
i N U M B E R of engineering operations require gases of (19) indicate that losses in adsorptive capacity are to be expected.
varying degrees of dryness. Hydrate formation (5, 10, Therefore it was decided to determine the adsorptive capacity of
14, 15, 31, 22, 28) and line stoppage of high-pressure several commercial adsorbents under conditions obtaining in a
natural gas transmission lines can be prevented by drying the commercial installation. Test apparatus is shown in Figure 1.
gas as it goes into the lines. The required efficiency o f dehydra Natural gas to be dehydrated was tapped from the inlet header of
tion depends upon the temperature of the ground through which a commercial installation and piped to the inlet packed scrubber
the line passes, internal pressure, and gas composition. Corro of the test units. The gas passed up through the scrubber and
sion and dust formation (2, 24 ) inside the line are also controlled into the top of the four test units, then down through the ad
when gas passing through is dried so that water does not condense sorbent and out through four standard meter runs. Reactivation
in the line. Bright annealing of steel (16) requires an atmosphere gas was supplied to the test unit from the commercial plants
free of water vapor. A number of catalytic reactions require gases reactivation gas heater. Test adsorbers were 73/ j inches inside
which are thoroughly dehydrated before they enter the process. diameter, and adsorbent beds were 34‘/ 2 inches deep.
Gases processed at low temperatures must be thoroughly de Other test conditions were as follows : pressure, 720-740 pounds
hydrated to prevent ice deposits on heat transfer surfaces. per square inch gage; temperature o f adsorbing cycle, 70-90° F .;
Solid adsorbents will dehydrate gases and organic liquids (8, 16) inlet gas saturated with water vapor; gas dehydrated, natural
to advantage, especially if the degree of dehydration must be gas residue from gasoline absorption plant; average molecular
relatively complete. Solid adsorbents are also advantageous weight of absorption oil, 236; carbon dioxide, 0.15% ; hydrogen
because they are relatively inert and do not introduce foreign sulfide, none; mercaptan sulfur, trace.
material into gas or liquid streams which they are dehydrating. The four adsorbents tested were 4 -8 mesh silica gel, activated
Although the adsorptive capacity of solid adsorbents is im alumina grade A, florite (a natural bauxite selected for mechani
portant and adsorptive capacities of now materials are well cal strength and adsorptive capacity), and Hi-fiorite (florite im
known, data are not plentiful on adsorptive capacity as a function pregnated with 12 -14 % manganous chloride, 26). Manganous
of number of reactivations. Plant operation data and literature chloride increases the capacity of the adsorbent to hold water
Operating arid laboratory

data are presented on the
solid adsorption deh ydra
tion o f com m ercial gases.
T h e l o s s in adsorptive
capacity o f several adsorb
ents is given as a fu n ction
o f tim e or n u m ber o f
reactivations. Laboratory
data on accelerated fo u l-
ing o f adsorbents show the
effects o f reactivation a t
m ospheres on th e decline
in adsorptive capacity.
T h e pressure loss for gases
flowing th rou gh beds o f
com m ercial adsorbents is
presented together w ith a
m ethod for estim ating the
I qu antity o f reactivation
gas required for reactivat
ing solid adsorbent beds.
righ t is a n a tu
ral gas dehydration u n it o f
the solid adsorbent type.
779
(W E T G A S IN 2 " P IP E The results of twenty-nine runs on the same

T U B I N G TO D E W material are given in Figure 2. The initial in
/H O T GAS I N - L A G G E D 2 ' P IP E
P O IN T T E S T E R creases in adsorptive capacity are thought to be
due to errors in determining the water dew point
of the exit dried gas, which is sometimes masked
by a hydrocarbon dew point. Although this pro
cedure duplicates plant operations, it is time con
suming and was abandoned in favor of an ac
celerated fouling test.
STATIC TEST
Samples of adsorbent used in test adsorbers

were retained for a laboratory test of adsorptive
C Ü D R Y 8. R E G E N E R A T I N G G A S PIPE capacity. After the runs of Figure 2 had been
completed, samples of adsorbent were removed
from the top and bottom of each test adsorber.
W E S C O T T O R IF IC E M E T E R S -,-
The adsorptive capacity of these samples was
W P IP E ,
measured by a static test procedure as follows:
le i Approximately 20-gram samples of adsorbent were
reactivated by heating in Petri dishes to 360° F.
for 6 hours. The adsorbents were allowed to cool
in a desiccator and weighed. The dishes containing
adsorbent were then placed in a laboratory des
iccator containing aqueous sulfuric acid solution
of 1.35 specific gravity. The desiccator was im
mersed in a constant-temperature bath at 90 ° F.
and evacuated to the vapor pressure of the aqueous
R U N S - LAGGED ' ¡
acid solution. The adsorbent was allowed to come
Figure 1. A pparatus for T esting Adsorptive Capacity to an equilibrium water content in the desiccator
(DimctiHions in inches) and then weighed. The increase in weight was
noted and per cent adsorptive capacity was cal
culated as previously outlined. Results (Table I)
because the combined water of the manganous chloride increases show that silica gel, Hi-florite, and alumina have lost considerable
during the adsorption cycle and is reduced during reactivation. in adsorptive capacity, particularly at the inlet of the adsorbent
All adsorbents were reactivated with inlet moist natural gas bed. It appears that part of the gas stream is adsorbed at the
heated to approximately 350° F. The hot gas was passed up inlet of the bed, and that this material which does not contact the
through the adsorbent until the top of the bed was within ap balance of the bed causes the marked loss in adsorptive capacity.
proximately 5 ° F. of the bottom of the test adsorber. Cold wet Dehydrogenation of saturated paraffin hydrocarbons has been
gas was then admitted into the top of the adsorbers until they reported (£>, IS). Alumina and silica gel with additions of metals
were cooled to 100-120° F. The units were then ready for test. of the fifth and seventh groups are active positive catalysts for
Inlet gas was passed downward through each adsorbent bed at the dehydrogenation of the hydrocarbons. Temperatures at
the same rate until the outlet dew point of the gas from the ad which these reactions are carried out range from 575° to 1650° F.
sorber reacjied 0 ° F. at full line pressure. D ew points of the gas A monomolecular layer of water on the catalyst (6') is reported
were measured by a modified Deaton and Frost pressure dew to promote dehydrogenation. Plant operation shows that un-
point apparatus {IS). Gas quantities were measured by four
standard W escott recording orifice meters. The weight of water
adsorbed was computed from the gas volume passed through the
meter, inlet temperature, and pressure, together with the data
of Deaton and Frost (IS) on the water-carrying capacity of
natural gas:
% adsorptive capacity
(weigh' water adsorbed X 100)/w eight of dry adsorbent
All adsorptive capacities were computed on this basis. The

weight of water in plant operation or dynamic test is the weight
of water retained by the adsorbent when the dew point of the
treated gas has reached 0 ° F.
T able I. D e t e r io r a t io n or A d so rben t
A dsorptive C ap a city b y Static Test % C ap a city of

Adsorber A dsorber New A dsorbent
inlet outlet N ew New N ew
top b ottom adsorbent top b otto m
S ilica gel 8 .5 5 2 1 .5 8 2 4 .7 7 3 4 .5 2 8 7 .1 2
H i-flo r ite 6 .9 8 10.15 12.01 5 8.1 2 8 4 .1 5
A c t iv a t e d a lu m in a 6 .2 1 7 .1 4 9 .3 9 5 5 .4 8 7 6 .0 4
10.5 0 4 9 .7 0 6 7 .8 0
F lo r it e 6.11 5 .3 6 6 .2 5 9 7.7 6 8 5 .7 6
5 .3 7 5 .2 8 5 .61 9 5 .7 2 9 4.1 2
Figure 2; A dsorptive Capacity o f Solid A dsorbents
reactivation period, gas controlled as to composition and heated

T a b le II. E f f e c t o f S a lt A d d itio n on D e te r io r a tio n of to the temperature of the oven was admitted to the bottom of the
A d sorb en t
tube container and flowed through it. After reactivation the ad
% N ew A d sorp tive C ap a city
sorptive capacity of the adsorbent was measured according to the
Adsorbent 51-day exposure 100-day exposure
static test procedure outlined. Results are shown in Table III,
Florite -f* 6 2 .3 41
Florite + m olybdenum 7 4 .3 46 where X indicates the reactivation gas atmosphere. The results
Florite •+• nickel 6 5 .9 53
Florite 4- boron 5 6 .7 show that heating adsorbents in dry air causes them to lose
Florite 4- bism uth 7 2 .6 66 adsorptive capacity slowly. Tim e of exposure appears to
Alumina + 4 5 .0 29 affect loss in adsorptive capacity per cycle. Higher reactivation
Alumina + m olybdenum 4 0 .8 21
Alumina 4* nickel 2 4 .9 20 temperatures cause slightly higher initial adsorptive capacities
Alumina 4- boron 3 9 .7 20
Alumina 4* bism uth 3 3 .7 17 but also increase the rate of decline in adsorptive capacity. Oil
Silica gel 36 27
vapors in natural gas appear to increase the rate of decline of ad
Hi-florite 32 27 sorptive capacity, and this decline seems to be more pronounced
when humid natural gas containing oil vapor is present.
E F F E C T O F O P E R A T IN G V A R IA B L E S
saturated hydrocarbons, such as ethylene and propylene, are
polymerized at 60-350° F. when in contact with a solid adsorb Dehydration of some gases by solid adsorbents is made difficult
ent. Adsorbents which are active as dehydrogenation and poly by impurities or components which they carry. Considerable
merization catalysts may foul themselves by first dehydrogenat- natural gas contains hydrogen sulfide. Adsorbents generally de
ing and then polymerizing hydrocarbons. teriorate more rapidly in the presence of high concentrations of
An attemi t was made to poison the catalytic dehydrogenating hydrogen sulfide which is thought to be particularly injurious in
powers oi the adsorbent in order to lengthen its life and improve the reactivation cycle. Hydrogen sulfide concentrations of 5
its adsorptive capacity. Salts of nineteen elements of groups 1-8 grains per 100 standard cubic feet of gas can be tolerated suc
of the Periodic Table were selected as possible poisons for the cessfully by adsorbents; concentrations of 40 grains have made
dehydrogenating powers of adsorbents. Samples of adsorbent replacement of adsorbent necessary after several months’ opera
were treated with small amounts of the salt poisons and put into tion. Gases containing hydrogen sulfide and oxygen in appreci
porous bags. These bags were placed in the adsorbers of the able concentrations are difficult to dry by means of adsorbents
commercial plant previously mentioned at a point where the wet because the hydrogen sulfide is rapidly oxidized to sulfur and
inlet gas first contacted the adsorbent bed. The test samples water under adsorbent reactivating temperature conditions.
were removed and tasted for adsorptive capacity at the end of 51 The elemental sulfur is partially vaporized and condenses in gas
and 100 days of exposure. Procedures followed the static test coolers. The unvaporized sulfur coats the adsorbent and causes
previously outlined. The main commercial adsorbers were re gradual loss in adsorptive capacity.
activated three times per 24 hours which gives each adsorber Gases containing components that can be polymerized to
one and one half reactivations per day. Results of the tests on high-molecular-weight polymers of low vapor pressure will
adsorptive capacity (Table II) indicate that salt additions do not gradually foul adsorbents unless reactivation temperatures are
retard the deterioration of solid adsorbents under these condi adjusted to remove these polymers during the reactivation cycle.
tions.
The data indicate that the adsorptive capacity of all solid ad
sorbents declines with continued use and reactivation. The
cause of this decline is not thoroughly understood, and reasons
advanced for it are speculative. Adsorbent materials are known
to be highly porous. This fact leads to the assumption that high-
moleeular-weight materials adsorbed but not removed during
reactivation cause a loss in adsorptive capacity through closure
of some of the pores. This assumption is supported by the fact
that adsorbents exposed to heavy hydrocarbon vapors show these
heavy hydrocarbons on extraction with benzol after reactivation.
These heavy hydrocarbons are found throughout the adsorbent
bed, but are generally higher in concentration, and in molecular
weight where the gases first contact the adsorbent. After
thorough extraction with benzene or reactivation with air at
600-800° F., fouled adsorbents have improved adsorptive capac
ity. These treatments will not produce the equivalent of new
adsorbent. Attempts to revivify badly deteriorated adsorbents
by these methods have given poor results, and the methods out
lined have been discarded as impractical. Figure 3. Effect o f Tem perature o f Gas on Adsorptive
Data in Table I indicate that the raw inlet gas Capacity o f F lorite D esiccant
which contains relatively high water concentrations
plus an equilibrium amount of absorption oil vapor a
has a marked effect on the adsorptive capacity of the 3
adsorbent. The efficient removal of this material in
the reactivation cycle, without damage to the adsorb-
wj-c
pL 6
v . a V .V 1
A
- -
ent, is necessary if adsorptive capacity is to be main- «
tained. Small quantities of adsorbent were reac- § h
tivated in controlled atmospheres and temperatures < i 4o 10 20 30 40 50 60 70 80 90
to determine the effects of variables in the reactiva- Jj
T IM E — W EEKS O PERATED
tion cycle. The adsorbent was placed in small tubes
in an oven held at constant temperature. During the Figure 4. Adsorptive Capacity o f F lorite
will raise its temperature 96 ° F . If a natural gas

T able III. E ffect of R e a c t iv a t io n C o n d it io n s o n A d s o r p t iv e C a p a c it y having an average molecular weight of 20 is de
C onditions of R éactivation__________ A dsorp- hydrated under similar conditions, the tempera
Oil N o. of tivo ture of the gas will be raised 6S.5° F. The
Teat H u va C ycle in C apac
D esiccant Hr. Op Gas Air mid D ry pors a Series ity , % same gas saturated with water vapor at 80° F.
N o.
224 Florite 8 700 X X X 1 11.0 under a pressure of 1000 pounds per square inch
233 8 700 X X X 2 9 .8
242a 8 700 X X X 3 7 .2 absolute will contain 0.000632 pound of water
250 8 700 X X X 4 7 .5 per pound of dry gas, and the temperature of the
260 8 700 X X X 5 7 .3
gas will be increased approximately 1.2° F. on
231 A ctivated 8 700 X X X 1 9 .5
240 alumina 8 700 X X X 2 8 .1 removal of the water vapor. Actually, gas tem
248 8 700 X X X 3 7 .8
258 8 700 X X X 4 7 .3 perature increases arc less than the figures given
270 S 700 X X X 5 6 .2 since adsorbent and equipment are also heated.
225 Florite 1 350 X 1 8 .5 The inlet temperatures of gas to be dehy
23G 1 350 X 2 8 .1
244 1 350 X 3 7 .9 drated should be as low as cooling water and
254 1 350 X 4 7 .7
hydrate-forming temperatures for natural gas
243 A ctivated 2 350 X 1 1 3.0
252 alumina 2 350 X 2 1 2.0 will permit, since water removal in the range
262 2 350 X 3 12.7 100-120° F. can be accomplished more eco
298a A ctivated 8 700 X 1 13.63 nomically by cooling than by means of solid
b alumina 32 700 X 2 11.27
c 24 700 X 3 9 .0 3 adsorbents. The effect of contact temperature
d 8 700 X 4 9 .0 3
on adsorptive capacity is shown in Figure 3.
301a A ctivated 8 700 X X 1 9 .2
6 alumina 32 700 X X o 8 .4 These are laboratory data with adsorption and
c 24 700 X X 3 6 .1 reactivation cycles carried out under atmospheric
d 8 700 X X 4 7 .0
pressure.
300a A ctivated 8 700 X X X 1 11.8
b alumina 32 700 X X X 2 1 0.7
c 24 700 X X X 3 9 .5 P E R F O R M A N C E A N D D E S IG N
d 8 700 X X X 4 7 .3
302a A ctivated 8 700 X X X The effective life of adsorbents is an important
1 1 0 .0
b alumina 32 700 X X X 2 7 .9
factor in choosing them and in designing equip
c 24 700 X X X 3 6 .1
d 8 700 X X ment in which they are used. The over-all cost
X 4 5 .9
397a A ctivated 8 700 X 'i 1 12.2 of adsorbent replacement and equipment should
b alumina 32 700 X ) Soaked I 2 1 1.5
c 24 700 X 1 in oil 9.1 be a minimum for drying a unit volume of gas.
d f 3
8 700 X I ) 4 9 .1 T o do this it is necessary to predict adsorbent
304a A ctivated 4 350 X X la 12.03
b alumina 4 350 X X 2 12.60
capacity as a function of time. Figure 4 shows
c 4 350 X X 3 10.23 the adsorptive capacity of florite desiccant in a
d 4 350 X X 4 9 .8 9
e 4 350 X X 5 9 .1 8 commercial plant dehydrating natural gas with
s 4 350 X X 6 8 .2 3 the same composition as that of the work dis
0 4 410 X X 7 8 .5 3
h 4 350 X X 8 7 .7 2 cussed in Figure 2. Adsorbents have been used
t 4 350 X X 9 7 .2 7
5 4 350 X X 10 7 .0 0 continuously for two years under reasonably
305a Activated 4 350 X X lû 11.52 favorable operating conditions and have dehy
b alumina 4 350 X X 2 11.55 drated approximately 2 million cubic feet of gas
c 4 350 X X 3 9 .5 1
d 4 350 X X 4 9 .0 9 per pound of adsorbent. Dried gas dew points
e 4 350 X X 5 8 .3 1
f 4 350 X X 6 7 .4 5 during a large part of the drying cycle are below
0 4 410 X X 7 7 .8 8 —40° F. at full line pressure. A typical drying
h 4 350 X X 8 6 .8 7
i 4 350 X X 9 6 .8 8 cycle is given in Table IV. Table V shows the
3 4 350 X X 10 6 .4 3
performance of a plant drying gas under a pres
308a A ctivated 4 350 X X 1° 12.1
b alumina 4 410 X X 2 11.5 sure of 900 pounds per square inch gage. Con
c 4 350 X X 3 10.5
11 4 350 X X 4 10.0
densable hydrocarbon has not been removed
e 4 350 X X 5 8 .9 prior to the drying operation. Tests on new
/ 4 350 X X 6 7 .9
0 4 350 X X 7 7 .8 adsorbent show that condensable hydrocarbon
* Plus 4 hours in cool humid air. is adsorbed in the drying cycle after water vapor
has been removed. This adsorbed condensable
hydrocarbon is displaced by water vapor as the
adsorbent bed becomes fully saturated with
Relatively nonvolatile compounds in general cause losses in ad water. The data in Table V indicate that adsorbent becomes
sorptive efficiency and capacity of adsorbents. These materials less selective for water vapor with continued use.
may come from gas compressors or may be carried by the gas
stream. Reactivation gas flow through adsorbent beds is usually
D R Y IN G E F F IC IE N C Y
opposite in direction to flow during the adsorption cycle. Such a
flow arrangement removes a large part of these objectionable The efficiency of the dehydrating operation is, in some in
components without having them contact all of the adsorbent bed. stances, of prime importance. The separation of helium from
Temperature, pressure, specific heat, and molecular weight natural gas and the production of oxygen and nitrogen by liquefy
affect the dehydration of gases by means of solid adsorbents. ing air require gases which have been efficiently dried. The
All of these variables affect the temperature of gases after water efficiency and rate of drying by alumina and silica gel in drying
vapor has been removed, and the temperature affects directly air have been investigated intensively (1, .5, 7, 16, 17, S3, 37).
the efficiency and capacity of the adsorbent. For example, air Dew points o f —64° to —104.8° F. are reported for the dried gas.
saturated with water vapor at 80° F. and 1 atmosphere pressure Isopiestics for silica gel indicate that at water concentrations of 6%
carries 0.022 pound o f water vapor per pound of dry air. When and contact temperatures of 70° F., a treated air dew point of
the water vapor is adsorbed, approximately 23 B.t.u. of heat are —60 ° F. can be obtained. Dew points of —100 ° F. for natural gas
given up if it is assumed that the heat of adsorption is equal to dehydrated with florite have been observed in the laboratory
the latent heat. This heat quantity, if applied to the air stream; (.9). Retention of heavy hydrocarbon by adsorbents decreases
their efficiency in the dehydration operation. Combinations of

two adsorbents, one having high adsorptive capacity and the
other giving high efficiency, are sometimes used to reduce the
volume of adsorbent required to dehydrate gases efficiently.
PR ESSU R E DROP TH R O U G H A DSORBEN T BEDS
Accurate prediction of pressure drop through adsorbent beds is

important in the design of solid adsorbent gas-drying installa
tions. This is particularly true if the drying operation is carried
out between two stages of compression or at the discharge of
natural-gas pipe line compressors. The literature discloses some
information on the pressure drop of gases flowing through cata
lyst beds and beds of broken solids ( / / , SO). Pressure drop as es
timated from relations developed by Chilton and Colburn {11)
are higher than measurements made in this work (Figure 5).
These data were measured by flowing natural gas (specific
gravity 0.677) under pressures indicated through a 6-inch i.d.
tower. The adsorbent-packed section was 41 inches deep. Gas
flow was measured with a standard recording orifice meter and
pressure drop was measured on a calibrated differential mercury
meter which could be read to one inch of water. Gas volumes
were corrected for their compressibility in computing velocity.
The temperature of the gas was 70° to 90° F. while pressure drop
data were taken.
Screen analyses of the adsorbents follow:
AJ
2-4 M esk 4 -8 M esh 2 -6 Mesh 2 -8 Mesh w- A *
On 2 3% On 4 1% On 2 0% On 2 0%
2-4 87 4 -8 87 2 -6 98 2 -8 96 *
Below 10 10 B elow 16 12 B elow 6 2 B elow 8 4
Solid adsorbents should be screened to as uniform size as prac PF1ESSUF1E DROP

2 - ? M.ESH FLORI1 "E|
ticable. Large variation in size or excessive fines (4) increase the
pressure drop of fluids flowing through beds of adsorbent. Pres
sure loss varies approximately with the change in density of the
fluid and directly as the square of the velocity. Pressure losses
for conditions other than those presented can be estimated with
reasonable accuracy on this basis.
R E A C T IV A T IO N H E A T R E Q U IR E M E N T S
Solid adsorbents used as dehydrating agents can be reactivated

by heating to 230-250° F. under the most favorable operating
conditions. Gases used for reactivation are heated to higher
.OS .08 .1 .2 .3 .4 .5 .8 .7.8.5 I 2
temperatures in order to furnish heat to the adsorbent bed above
PRESSU RE D R O P IN IN C H E S O F W ATER PER INCH OF BEt>
F ig u re 5. P ressu re D rop th r o u g h A d sorb en t B ed s

Table IV. P erform ance D ata on the A d s o r p t io n C ycle S p e c ific g ra v ity o f ga s, 0-677
(Gas saturated at 8 0 -9 0 ° F .; pressure, 195 lb. per sq. in. gage; absorbent,
1000 lb. of florite and 6000 lb . of H i-florite)
this temperature. Higher reactivation gas temperatures also
Gas D ried, D ried Gas
1000 Standard D ew Point, reduce the quantity of reactivation gas required and also remove
D ato Tim e Cu. Ft. Op
liquids having vapor pressures lower than water more efficiently.
8 /1 /4 1 1 0 :0 0 p .m . < -4 0
8 /2 /4 1 1 0 :0 0 p .m . 213 3 .5 < -4 0 Temperatures of 300° to 500° F. are commonly used in reacti
8 /3 /4 1 1 0 :0 0 p .m . 347 7 .9 < -4 0 vating adsorbent beds. The temperature necessary will depend
8 /4 /4 1 1 0 :0 0 p .m . 504 6 .0 < -4 0
8 /6 /4 1 6 :3 0 p . m . 650 6 .8 -1 1 on the nature of the material ad-
____________________________ sorbed or polymerized during
the adsorption and reactivation
T a b l e V. P l a n t O p e r a t i n g D a t a o n N a t u r a l G a s C o n t a i n i n g 0 .2 G a l l o n o f G a s o l i n e
cycles.
p e r 1 0 0 0 C u b ic F e e t
The heat quantity required for
•--------------------------W ater A dsorbed--------------------------. H yd rocarbon Adsorbed
Measured C ontent C aled. C ontent (M easured) reactivation has been discussed
A d sorp A d sorp A d sorp in the literature {16, SB). H eat
tive tive tive
G al capac capac capac quantities are given as a function
Test D ate A dsorbent lons Pounds ity 0 , % Gallons Pounds ity “ , % Gallons Pounds ity “ , % of the weight of water removed
1 /2 6 -2 9 /1 2 Florite 1032 7 .6 0 945 6 .9 5 N egli N egli N egli
gible gible gible
from the bed to be reactivated.
1 /2 6 -2 9 /4 2 A ctivated 1102 8 .1 0 1146 8 .4 3 N egli N egli N egli The use of such ratios may sup
alumina gible gible gible
1 1/1 7-1 8 /4 2 F lorite 78 650 4 .7 7 8 6 .7 722 5 .3 1 432 2640 19.40 ply sufficient heat in some cases
1 1/1 7-1 8 /4 2 A ctivated 96 802 5 .9 0 9 6 .8 806 5 .9 3 288 1760 12.95
12/7/43 <> Florite 737 6 .4 264 12.33
and be deficient in others.
A ctivated 685 5 .0 246 11.60 Quantitative studies of reac
alum ina
* All capacities to final dew p oint of 0 ° F.
tivation heat requirements have
After thorough reactivation. been made. A complete reacti
vation cycle is shown in Figure 6
water gain between the inlet and

outlet can be estimated and the
latent heat required can be calcu
lated for this gain. Since the ad
sorptive capacity of the adsorbent
is known, a definite weight of ad
sorbent and tower weight will be
associated with the water removed
by one pound of gas. The total
heat required can now be esti
mated and gives the second point
necessary in Figure 7. The outlet
temperature of the reactivation gas
is such that heat required is equal
to heat available. The gas
quantity necessary for the total
E X IT R E A C T IV A T IO N GAS T E M P ./F . reactivation can be estimated
when the outlet temperature of the
Figure 7. E stim ation o f Final gas, for the period during which it
Figure 6. Adsorber R eactivation
T em perature o f R eactivation Gas is saturated at maximum tempera
Cycle
R e a c tiv a tio n p re s s u re , 150 lh . p e r s q . in ,
a h s .; a vera ge m o le c u la r w e ig h t o f n a tu ra l ture, has been computed.
g a s, 19.7 When operating under high pres
sure, equipment weight is large and
sensible heat requirements for it
Heat balances show that the heat required to vaporize the are appreciable. When operating pressures are low, sensible heat
water from the bed approximate the latent heat of vaporization requirements for equipment are negligible. Radiation losses should
corresponding to the temperature at which the bulk of the water be estimated and taken into account if this loss is appreciable.
is removed. This quantity has been determined for water only,
and will be approximately correct for other liquids unless there ACKNOW LEDGM ENT
are large heats of wetting in connection with the adsorption.
The assistance of the Texoma Natural Gas Company, J. F.
Heat requirements for reactivation can be divided into two
Pritchard and Company, Floridin Company, and their employees
parts for estimating the necessary quantity of reactivation gas:
in making this publication possible is acknowledged.
latent heat of vaporization supplied to the adsorbed material
and sensible heat required to heat the adsorbent and apparatus L IT E R A T U R E C IT E D
to the final reactivation temperature. The temperature of the
cu Ahlberg, J. E ., I n d . E n o . C h e m ., 3 1 , 9 8 8 -9 2 (1 9 3 9 ).
reactivation gas supplied to the adsorbers is set in the design ( 2 ) Allyne, A. B., (?as Age, p. 463, N ov. 11, 1933.
conditions. The temperature of the gas leaving the adsorber (3) Anonymous, Chem. News, 130, 340 (1925).
being reactivated can be computed for the constant outlet tem (4) Badger, W . L., and McCabe, W . L.. “ Elements of Chemical
perature period if the following quantities are known: Engineering'-’ , 464, New York, McGraw-Hill Book Co., 1931.
(5) Bechtold, I. C., Refiner Natural Gasoline M fr., 21, No. 4, 71
1. Dew point of inlet reactivation gas (1942).
2. Concentration of adsorbed vapor outlet gas saturated at (6) Berkman, Morrel, and Egloff, "Catalysis” , Part X II , Table I,
reactivation pressure p. 888, Reinhold Publishing Corp., 1940.
3. Latent heat of adsorbed vapor (7) Bower, J. H ., Bureau Standards J. Research, 12, 241 (1934).
4. Specific heat of reactivation gas (8) Capell, R . G., and Amero, R . C., Oil Gas J., 41, No. 6, 37-9
5. Inlet temperature of reactivation gas (1942).
6. Specific heat of adsorbent (9) Cappell, R . G., Am ero, R . C., and M oore, J. W ., Chem. & Met.
7. Specific heat of adsorber construction materials Eng., 50, N o. 7, 107 (1943).
8. Approximate weight of adsorber ( 10) Carson, D . B., and Katz, D. L., "Engineers Technical Publica
tions", 1371 (1941).
The outlet temperature is computed by heat balance and is (11) Chilton and Colburn, In d . E n g . C h em ., 23, 913 (1 9 31 ).
such that the sensible heat loss of the reactivation gas is equal to ( 12) Deaton, W . M ., and Frost, E. M ., Jr., Am . Gas J., 155, 61-4
(1941).
the following heat quantities: (13) D eaton, W . M ., and Frost, E. M ., Jr., Bur. of Mines, Rept.
Investigations, 3399 (1938).
1. Latent heat gain of reactivation gas stream
2. Sensible heat to adsorbent to raise temperature from point
(14) Deaton, W . M ., and Frost, E. M ., Jr., Gas, 16, No. 6, 28-30
(1940).
where reactivation gas is saturated to inlet temperature
of reactivation gas
(15) Deaton, W . M ., and Frost, E. M ., Jr., Gas Age-Record, 81, 33
3. (M ay 26, 1938).
Sensible heat to adsorbent vessel from temperature at
which water is removed to inlet temperature of reactiva (16) D err, R . B ., I n d . E n g . C h e m ., 30, 3 8 4 -8 (1 9 38 ).
(17) Doner, M . V ., and Marden, J. W ., J. Am . Chem. Soc.. 39, 1609
tion gas
(1917).
This computation is a trial and error process which can be (18) Dunstan and Howes, J. Inst. Petroleum Tech., 22, 347 (1 9 36 ).
shortened by plotting several values according to Figure 7. (19) Faber, E. M ., Olson, H. G ., Taylor, W . A., Chem. & Met. Eng.,
28, S05 (1923).
Figure 7 gives the estimated outlet temperature for the reacti (20) Furnas, Bur. Mines, Bull. 307 (1929).
vation cycle shown in Figure 6. Values plotted in Figure 7 were (21) Hammersehmidt, E. G., Am . Gas Assoc. Monthly, 18, 273-6
computed as follows: An outlet temperature for the reactivation (1936).
gas is taken. Since the inlet temperature is known, the quantity (22) Hammersehmidt, E. G ., Oil Gas J., 37, 66 (M ay 11, 1939).
(23) Johnson, J. Am. Chem. Soc., 34, 911 (1912).
of heat available for reactivation is computed from the tempera (24) Larson, D . B., Pacific Coast Gas Assoc. Proc., 30 (1939).
ture drop of the gas and its specific heat. This heat quantity is (25) Lednum, J. M ., Am. Gas Assoc. Proc., 1934, 664-80.
one point necessary in Figure 7. The gas flowing from the bed (26) Martin, H . A., and Spangler, C. V., U. S. Patent 2,284,981
is saturated with water vapor and the weight of water carried (June 2 , 1942).
(27) M unro and Johnson, I n d . E n g . C h e m ., 17, 88 (1925).
per pound of gas is then calculated or determined from experi (28) W ilcox, W . I., Carson, D. B., and K atz, D . L., Ibid., 33. 662-5
mental data. If the reactivation gas before heating is wet, the (1 9 4 1 ).
Sulfuryl chloride is a versatile reagent with great potentialities in
organic syntheses. U nder appropriate conditions it may be used
for the chlorination and sulfonation of both aliphatic and aromatic
derivatives. It may also be ap plied as an acylating and condensing
agent. The sim plicity of these reactions, their ease of control, and
their high yie ld s promise to make the reagent particularly useful in
many applications which require sim ple and unspecialized equipment.
Sulfuryl Chloride
in
Organic Chemistry
cttenbesit C. ßfixuu+i Filling C arboys from the Sulfuryl Chloride
Stabilizer (H o o k e r Electrochemical C om p a ny)
W A Y N E UN IVERSITY, DETROIT, M IC H .
T HE use of sulfuryl chloride as a chlorinating agent for aro nate benzene, Dubois heated the hydrocarbon with sulfuryl
matic compounds dates back to the early days of organic chloride in sealed tubes at 150° C. {23) and Tohl and Eberhard
chemistry. In 1866 the observation was made that phenol used a temperature of 160-170° (70).
is easily chlorinated by sulfuryl chloride at room temperature In the course of a comprehensive investigation of the effect of
and that benzene may be similarly chlorinated at somewhat various catalysts on the reaction between sulfuryl chloride and
higher temperatures. It was subsequently discovered that by benzene, Silberrad found that substances commonly classed as
suitable adaptation o f the experimental conditions sulfuryl halogen carriers (sulfur, iodine) bring about the chlorination of
chloride coidd also be used as a sulfonating agent for aromatic the benzene ring at lower temperatures {65). A mixture of sul
compounds. Means have recently been found for extending fur monochloride and aluminum chloride is a particularly effec
these reactions to the aliphatic hydrocarbons and their deriva tive catalytic agent. A chlorinating reagent composed of 1%
tives. The potentialities of this versatile reagent are not limited of this catalyst dissolved in sulfuryl chloride chlorinates benzene
to its use as a chlorinating and sulfonating agent. The acylation rapidly and sm oothly in the cold {66). W ith this reagent step
of alcohols and amines, the conversion of carboxylic acid salts to wise polychlorination of benzene through the hexachloro deriva
acid chlorides and anhydrides, and the facilitation of various tive takes place under comparatively mild conditions.
condensation reactions are among the better known applications Silberrad also reported that toluene may be similarly chlori
of sulfuryl chloride. nated in stages {67), and that unlike other powerful chlorinating
The ease with which these reactions may be controlled and agents, sulfuryl chloride, catalyzed as described, neither cleaves
directed is noteworthy. For example, in a mixture of an aroma nor attacks the side chain. A practically quantitative yield of
tic and a paraffinie or a cycloparaffinic hydrocarbon such as ben pentachlorotolueno may be obtained:
zene and cyclohexane, it is possible
either to chlorinate selectively or to
sulfonate selectively one or the other
hydrocarbon merely by changing the
experimental conditions. O f particular
interest at this time is the relative
simplicity of the equipment needed.
Some representative reactions of sul
furyl chloride are presented here to
indicate its potentialities in organic
chemistry; no exhaustive review is
attempted.
C H L O R IN A T IO N OF A R O M A T IC NUCLEI
Although sulfuryl chloride readily

chlorinates 'the more reactive ben-
zenoid derivatives, of which the phenols and amines are Silberrad also studied the effects of many substances on the
typical, it reacts with benzenoid hydrocarbons in the absence of ratio of nuclear and side-chain chlorination in toluene {68).
catalysts only under rather vigorous treatment. Thus, to chlori The results are of doubtful value, however, because the toluene
785
786 INDUSTRIA'L AND ENGINEERING CHEMISTRY Vol. 36, No. 9
he used probably contained peroxides (see "Chlorination of

Aliphatic Com pounds'’ ).
The sensitivity of sulfuryl chlorido as a chlorinating agent to A + S O jC l.— > ^
changes in the experimental conditions is well illustrated by its
— Cl
reported action on naphthalene. B y slight alterations in the
reaction conditions, naphthalene can be made to give 1,2,3,4-
h
tetrachloro-l,2l314-tetrahydronaphtlmlene, 1-chloronaphthalene,
o r 1,4-dichloronaphthalene (31, 60, 70):
The chlorination of phenol and its derivatives has been a
common application of sulfuryl chloride (3, 12, 23, 53, 57, 73).
In general, the products obtained are the same as those produced
by the action of gaseous chlorine; the reaction, however, is loss
+ 2SO,Cl, vigorous and more easily controlled. Thus, the stepwise chlorin
ation of resorcinol can be regulated to yield mono-, di-, or tri-
chlororesorcinol as desired (62). Other examples of this appli
cation of sulfuryl chloride follow:
OH
+ SO,Cl2 AlCh, 25
-{- 2SO2CI2
+ 2SO2CI2 100-140° C. ^
•OH
+ S 0 2C1
M any polynuclear hydrocarbons other than naphthalene may

b o chlorinated without the aid of catalysts. Under conditions
similar to those used in the production of 1,2,3,4-tetrachloro-
+ S0,C1; — >■
1,2,3,4-tetrahydronaphthalene, pyrene yields chloropyrene and
perylono yields a mixturo of dichloro derivatives (60). At
room temperatures anthracene is converted to the 9,10-dichloro
derivative (6), whereas acenaphthene is substituted in the 3-
position (SO):
In many instances practically quantitative yields are reported.
Cl The use o f sulfuryl chloride to prepare compounds of high
germicidal properties by the chlorination of alkyl derivatives of
/ \
cresol, resorcinol, and catechol is described in recent patents
+ 2SOjCIs
. (45, 46, 55, 61):
CH a CH ,
Cl
II2C— CH- H3C— c h

+ SO2CI2
-CHa(CH5) t—l
A i °—Cl"
+ SOîCh
Phenol ethers are also chlorinated by sulfuryl chloride, al
though less readily than the phenols themselves (57). Phenol
esters such as phenyl acetate or benzoate, triacetylphloroglucinol,
etc., do not react under mild conditions in the absence of cata
Heterocyclics are chlorinated under similar mild conditions. lysts (57):
F or example, indole has been converted into 2-chIoroindole or OCaHs OC2H s
into 2,3-dichloroindole, depending upon the quantity o f sulfuryl
chloride used (50); pyrarolc at 0 ° C. yields 4-chloropyrazole
{6 1 ); carbazole forms 3-chlorocarbazole (62). Although pyri + SOaClj
dine itself is not readily attacked, pyridine oxide reacts to give
ultimately 2- and 4-chloropyridine (11).
The reaction of aromatic amines and their derivatives with

+ 2SO,Cl,
sulfuryl chloride has also received considerable attention. Ani
line is readily chlorinated in the absence of catalysts to form
the mono-, di-, and trichloro derivatives (25). Only three
chlorine atoms can thus be introduced into the nucleus. An-
-f- SO2CI2
thranilic acid is sm oothly chlorinated to either a m ono- or a
dichloro derivative. Instead of the free amines, the acyl de
rivatives such as acetanilide may be used (26, 78):
NHs N H j.H C l the presence of “ ionizing” catalysts such as carbon (ISA , 49).
Under these vigorous conditions the sulfuryl chloride must be
almost completely dissociated into sulfur dioxide and chlorine
+ S0,C1,
(72), and the chlorination of the hydrocarbon is probably due
to the presence of free chlorine. On the other hand, the re
cently discovered chlorination of aliphatic hydrocarbons and
their derivatives with sulfuryl chloride in the presence of sm all
NIL quantities of organic peroxides (34, 35, 86, 88) takes place a t
moderate temperatures. The mechanism of the reaction prob
COOH ably does not involve prior decomposition o f sulfuryl chloride in
+ SOjCl.— >
the sense just mentioned.
In the absence of catalysts and light, sulfuryl chloride does not
react with saturated aliphatic or alicyclic hydrocarbons such as
n-heptane or cyclohexane at temperatures as high as the boiling
NHCOCH, points of the mixtures. Even in the presence of light, these
hydrocarbons react only very slowly. Substances effective as
catalysts in the chlorination of aromatic hydrocarbons here affect
-j- 2SOiC12
the rate of chlorination only slightly. In comparison, the effect
of a trace of peroxide is striking: The addition of less than 1
mole % of benzoyl peroxide to the reaction mixture causes prac
tically complete reaction in 30-60 minutes.
Several studies have recently appeared of the chlorination of Study of the chlorination of various paraffin hydrocarbons
aryl amides of aromatic acids by sulfuryl chloride (82, 83). The and their halogen derivatives (34) indicates that, generally
authors were interested in the effect of substituents in the basic speaking, the products obtained by the use of sulfuryl chloride
and acidic parts of the molecule on the ease of halogenation. parallel those produced by the photochemical reaction with
Although no detailed discussion of this work can be given here, gaseous chlorine. Thus, the chlorination of »¿-heptane yields
one example will be presented. Benzanilide is readily chlorin 15% of primary and 85 % of secondary chlorides; 1-chloro-
ated, first in the para position of the basic groups, then in the butane is substituted only slightly in the 1-position, 25 % in the
ortho position. However, the anilide of o-nitrobenzoic acid is 2-, 50% in the 3-, and 25 % in the 4-position. Neither chloro
much less reactive, and only one chlorine atom is introduced into form nor sym-tetrachloroethane is further chlorinated by this
the basic part of the molecule even under comparatively vigorous method.
conditions: These and other similar results may be summarized in the
following generalizations: (a) Substitution of hydrogen on a
HN— CO— <( )> H N — CO— / H N — CO— / )> secondary carbon atom takes place in preference to substitution
on a primary carbon atom ; (b) a second substituent chlorine
atom tends to attach itself to the aliphatic chain as far from
the first chlorine atom as possible; (c) substitution of a second
chlorine atom upon a carbon atom already bearing a halogen
substituent is difficult, and a third chlorine atom cannot be
attached to one carbon atom in this way.
The use of sulfuryl chloride as a chlorinating agent has hitherto
H N — CO been restricted primarily to aromatic nuclei in the presence o f
halogen carriers. However, the isolation of benzyl chloride from
n o2 the chlorination product of toluene is reported in several in
stances. D ubois stated that toluene and sulfuryl chloride react
slowly at 115° C. to form benzyl chloride and chlorotoluene (22),
and W ohl patented the preparation of benzyl chloride by the
action of sulfuryl chloride on toluene at 130° (76). Tohl and
Eberhard stated that a refluxing mixture of toluene and sulfuryl
Except in a few instances, the action of sulfuryl chloride on
chloride does not react (70); this statement has been questioned,
aromatic aldehydes and ketones has not been studied. It is
and it has been claimed that toluene and sulfuryl chloride d a
reported that benzophenone does not react with sulfuryl chloride
react in the dark at reflux temperature (68).
in the absence of a catalyst, whereas benzaldehyde reacts to form
It is now established that pure, peroxide-free toluene does n o t
benzoyl chloride rather than any nuclear chlorinated derivative
react at any appreciable rate with sulfuryl chloride in the dark
(24).
even at the boiling point of the mixture (34). A small amount
The use of sulfuryl chloride as a solvent (as well as a chlorinat
of added peroxide induces a vigorous reaction; nearly quantita
ing agent) for processes involving the chlorination and bromina-
tive yields of benzyl chloride are obtained in a few minutes. In
tion of dyes and dye intermediates has been the subject o f a num
the presence of excess sulfuryl chloride the side chain is further
ber of patents (16, SI, 44, 88, 76). In certain instances, as in
chlorinated with the formation of benzal chloride. The intro
the chlorination and bromination of pyranthrone (21), the
duction of a third chlorine atom into the side chain either does
presence of the sulfuryl chloride appears to exert a directive
not take place or else occurs so slowly that, for practical purposes,
influence on the reaction which is not shown by the other halo-
the reaction m ay be neglected. This last observation supports
genating agents. This may be due to the reducing action of
the conclusion previously drawn from the inertness o f sym-
sulfur dioxide liberated during the reaction.
tetrachloroethane and chloroform: The accumulation of tw o
chlorine atoms upon a carbon atom prevents further substitution
C H L O R IN A T IO N O F A L IP H A T IC C O M P O U N D S
(by this method of chlorination) upon that atom.
Until recently the only references to sulfuryl chloride as a In general, derivatives and homologs of toluene readily undergo
chlorinating agent for a saturated aliphatic hydrocarbon were side-chain chlorination by sulfuryl chloride and peroxides;
the patents covering the chlorination of methane at 3000 C. in among the compounds investigated are p-clfforotoluene, ethyl-
benzene, and isopropylbenzene. o-Nitrotoluene and p-nitro- yields 40 % « - and 60% | 8-chloropropionyl chloride, and n-
toluene, however, are exceptions; neither can be chlorinated by butyryl chloride yields 15% «-, 5 5 % 0-, and 30 % y-ehlorobutyryl
this method: chloride.
If, instead of peroxides, halogen carriers such as iodine are
CHClj
used to catalyze the reaction of sulfuryl chloride, substitution
takes place only at the «-position. Hence, the mechanisms in
volved in catalysis by peroxides and by the usual type of halogen
carriers must be of different types:
C H (C H 3)C H 2C1 (10% ) CH3CHCICOCI (40%)

CH3CH2C0C1 + S 0 2C !2 pcr0Xldes< ^
CH2C1CH2COC1(60%)
CH3CH2C0C1 + S 0 2C12 ^ CH3CHC1COC1(100%)

iCC1(CH j) j (90% )
(CH 3)2CHCOOH +
(CH 3) 2C C 1C 00H (15%)
S 0 2C12 peroxides CH 3\
The peroxidc-sulfuryl chloride method finds ail especially < CHoCK
\ C H C O O H (85%)
useful application in the selective side-chain chlorination of aro

matic hydrocarbons of moderately high nuclear reactivity. For Aldehydes and ketones which do not contain an «-hydrogen
example, it is stated that the photochemical chlorination of the atom undergo the usual peroxide-catalyzed reaction {40). Thus,
side chain in m-xylene to form «-chloro-m-xylene is accompanied benzaldehyde is converted by sulfuryl chloride to benzoyl chlo
even under the most favorable conditions by the formation of at ride, and phenyl ierf-butyl ketone is chlorinated in the alkyl group.
42
least 10% of 4-chloro-i«-xylenc { ). Chlorination of «i-xylene Benzophenone does not react (24). If the compound contains
with sulfuryl chloride and peroxides, however, yields pure « - an «-hydrogen atom, substitution in this position readily occurs
chloro-m-xylcne. It has also been reported that the side chain at room temperature in the absence of added peroxide or other
o f ieri-butylbenzene is so unreactive in the absence of peroxides catalysts, probably by a mechanism involving the enol form of
that the side-chain halogen derivatives cannot be directly ob the compounds {2, 17, ): 48
tained; conditions under which reaction with halogens occurs CH3CO CH 3 + S 0 2C12 — ^ CH 3C 0 C H 2C1
lead only to nuclear-substituted products (GJt). However, in
C iI3COCH2COOC2H 6 + S 0 2C12— >- CH3COCHC1COOC2Hc
the presence of a peroxide, sulfuryl chloride chlorinates the side
CH 2(COOC2H 6)2 + 2 S 0 2C12 — >- Cl2C(COO C2I-I6)2
chain selectively and with ease. Unfortunately this type of
selectivity is not universal. The nuclei of fluorene and ff-methyl- The chlorination of 7-acetopropyl alcohol with sulfuryl chloride
naphthalene, for example, are so reactive and nuclear substitu a t 0 0 C. was recently patented as a means of preparing 7-chloro-
tion proceeds so readily, that the peroxide-sulfuryl chloride 7-acetopropyl alcohol, an intermediate in the synthesis of vita
method' cannot be used for the selective chlorination of the m in Bi {16, 59):
side chains of these compounds.
c i i 3c o c i i 2c h 2c h 2o h + S 0 2C12
Sulfuryl chloride reacts readily with most alkenes to give the CH3C0CHC1CH20 H 2CH20H
saturated dichloro derivatives and sulfur dioxide. This reac
tion is the subject of several patents. Among them are the for S U L F O N A T IO N O F A R O M A T IC C O M P O U N D S
mation of dichlorides from amylencs and hexenes by the reaction
The use of aluminum chloride with sulfur monochloride as a
of sulfuryl cliloride with the hydrocarbons below 30° C. {4) and
•catalyst for the chlorination of aromatic compounds by sulfuryl
the reaction of ethylene and propylene (in a mixture with satu
chloride has already been discussed. Under slightly different
rated hydrocarbons) with sulfuryl chloride at 10-40° C. { ). 14 conditions it is reported that aluminum chloride induces a
An investigation of the factors involved in the reaction has
different reaction— sulfonation of the aromatic nucleus. The
revealed that the chlorination is peroxide-catalyzed {35). Many
procedure recommended by Tôhl and Eberhard (70) is the addi
unsaturated compounds (cyclohexene, allyl chloride) usually
tion of small quantities of the catalyst to a cooled reaction mix
contain enough peroxidic material to permit the reaction to
ture of sulfuryl chloride and the hydrocarbon. B y this method
proceed easily without the addition of a peroxide catalyst.
these authors prepared the sulfonyl chloride derivatives of many
However, for the analogous chlorination of other, less oxygen-
aromatic hydrocarbons, among them benzene, toluene, xylene,
sensitive olefins (s)/»i-dichloroethylene, stilbene), the addition of
mesitylene, etc. The familiar orientation rules of oleum sulfona
small quantities of peroxide is essential:
tion are obeyed:
The chlorination of a simple acetylenic derivative with the

s 0 2Cl
•reagent has not been described. It is claimed that divinylacet-
ylene yields an oil which is probably 3,4-dichloro-l,3,5-hexa- S 0 2C1
triene {18). C H 3— t S— CH, I
Sulfuryl chloride, together with peroxides, easily chlorinates ch 3 /\ - c h 3
+ S 0 2C12 AlCUj
aliphatic acids and acid chlorides {36). The hydrogen atoms in
the «-position on the carbon chain are substituted with greater
difficulty than those in the /S- and 7 -positions. That is, the
ch 3
product (except in the instance of acetic acid and acetyl chloride)
consists chiefly of 6- or of (3- and 7 -chlorinated acids, or the cor Besides the sulfonyl chloride derivatives which constitute the
responding acid chlorides. For example, propionyl chloride major part of the reaction product, relatively small quantities
of chlorinated hydrocarbons and sulfones are also formed. With C H 3CH 2CH,OH + SCLCl = CH 3C iI,C H 20 S 0 2Cl + HC1
the highly alkylated benzenes, such as durene and pentamethyl- CH 2OH CH20 S 0 2C1
benzene, Tôhl and Eberhard report that chlorination alone + S 0 2C12 = I + HCI
c h 2o h c IH
i jOH
occurs.
Working at somewhat higher temperatures and using molar <( )>— CH-OH + SO..CI = )>— c h ..o s o 2ci HC1
quantities of aluminum chloride, Booseken observed that, in addi
tion to sulfonyl chloride (IS), sulfonic acids and large quantities
W ith an excess of alcohol, the reaction may proceed to the
of chlorinated derivatives are formed. Since he worked under
formation of (a) dialkyl sulfate or (6) alkyl sulfuric acid (47):
conditions favorable to the dissociation of sulfuryl chloride, it
is highly probable that the formation of these products is due not (а) CHaOSOiCl + CH 3OH = (C H 3)2SO( + HCI
to the action of the reagent itself, but to the effect of its dissocia (б) C H 30 S 0 2C1 + CHaOH = CH 3OSOaH. + CH 3C1
tion products, sulfur dioxide and chlorine.
Another method of introducing the sulfonyl chloride group into Aliphatic amines undergo similar reactions. Secondary
the aromatic nucleus may be mentioned, although its practical amines such as dimethylamine or piperidine form the symmetrical
value in organic syntheses is uncertain. The reaction of the sulfamide derivatives unless special precautions are taken
Grignard reagent with sulfuryl chloride takes place in two stages: (9, 10, 71). The hydrochlorides of these amines react less
vigorously, and the reaction stops after the combination of a
SOs'Cli + CeHsMgBr = C ,H £ O aCl + M gClBr
molecule of sulfuryl chloride with one molecule of the amine
CJicSChCl + CsH,M gBr = C6H 6C1 + C6H 6S 0 2MgBr
hydrochloride:
With the proper precautions it is possible to control the reaction 2(C H 3)2NH + S 0 2C12 — >- (C H ,)SN — S 0 2— N (C H 3)2 + 2HC1
so as to obtain fair yields o f the sulfonyl chloride (19). (CH 3)2N H — HCI + S 0 2C12 — > (C H 3)2N — S 0 2C1 + 2HC1
S U L F O N A T IO N O F A L IF 'H A T IC C O M P O U N D S Although practically all the work on this reaction has dealt with
secondary amines, one primary aliphatic amine is known to re
Sulfuryl chloride also reacts with aliphatic Grignard reagents
act in the same way (27).
to form aliphatic sulfonyl chlorides (19). However, the yields
Under the proper conditions, sulfuryl chloride reacts primarily
are only moderate and the reaction has been but little used.
as an acylating agent, even with aromatic amines. Thus, W ohl
A better method of employing sulfuryl chloride to introduce the
and K och reported that a 60 % yield of sulfanilide is obtained by
sulfonyl chloride group into aliphatic compounds has recently
dropping sulfuryl chloride dissolved in dry ether into a cooled
been developed (41). In the presence of light and a small quan
solution of aniline dissolved in about three times its volume of dry
tity of suitable catalyst (e.g., pyridine or thiophenol) at 4 0 -
ether. Azobenzene is formed as a by-product. p-Toluidine
60 0C., sulfuryl chloride and paraffin hydrocarbons undergo a rapid,
undergoes a similar reaction (77).
vigorous reaction with the formation of alkyl sulfonyl chlorides.
Although acyl derivatives of the aromatic amines do not take
The yields are excellent, in some cases as high as 70% . Sul
part in a reaction analogous to those just mentioned, the metal
fonation is accompanied by the formation of smaller quantities
derivatives of the aryl amides and sulfonamides have been found
of chlorinated products. Among the hydrocarbons which have
to react (54):
been thus sulfonated are cyclohexane, «-butane, ethylbenzene,
and /erf-butylbenzene: CHjGO— N — Na CH 3CO— i\—SOîCl
S 0 2C12
H- SO2CI5 pyridine, ho .
The metal derivatives of the unsubstituted amides do not yield
stable chlorosulfonyl derivatives when treated with sulfuryl
The reaction just mentioned has been extended to include other chloride. Tw o typical reactions (6) are:
types of aliphatic compounds. In the treatment of the lower
CH ,fcO N H N a + S 0 2C12 C H 3GÖNHSO2CI + NaCl
aliphatic acids (propionic, butyric, isobutyric) no catalyst need I
be. added; the sulfonation proceeds rapidly and smoothly when CH 3NCO + S 0 2 + HCI
a warm mixture of sulfuryl chloride and the organic acid is illu
CnHsCONHNa + S 0 2C12 = CtH6C (0 S 0 2Cl)N H + NaCl
minated. The products isolated are not the sulfonyl chlorides,
but a new class of compounds— the inner anhydrides of sulfo- ¿eH 6CN + SO3 + HCI
carboxylic acids (36, 39). These new substances should find
valuable applications: An interesting use of sulfuryl chloride in acylation reactions
was reported by Baumgarten (7). The addition compound of
CH2— COOH h„ •CH.— COOH CH 2— CO, sulfuryl chloride with pyridine, dissolved in an excess of pyridine,
I + S0 2C12^ - I >0
CHn CH..— S 0 2C1 . c h 2— s o / ■ reacts with alcohols and acids to form the alkyl and acyl chlorides,
respectively. If a phenol or an amine is present, it is acylated:
In the sulfonation of the higher aliphatic acids (lauric, palmitic)
the addition of a small quantity of pyridine or some similar cata
lyst is required for optimum results. A large proportion of the
substitution must occur in positions unfavorable to cyclization,
since the products isolated consist primarily of the sulfonyl
chlorides.
R E A C T IO N S A S A N A C Y L D E R IV A T IV E
In many reactions sulfuryl chloride reacts as an acid chloride

of sulfuric acid. For example, with many aliphatic alcohols it
lorms the alkyl chlorosulfonates (5, 10). Under proper condi By the action of sulfuryl chloride on the sodium salts of or
tions good yields can be obtained: ganic acids, acid chlorides and acid anhydrides are formed. Ben
zoyl chloride and benzoic anhydride have been made in this way betw een “ rea ctivity” and “ m echanism ” . T o m en tion one ex
from sodium benzoate (4$), and the corresponding phenylacetyl am ple, a rom atic h ydrocarbon s (benzene, toluene), which are
derivatives have been similarly prepared from sodium phenyl ordinarily considered “ reactive” hydrocarbons, d o n ot react with
acetate (S4 ). sulfuryl chloride in the presence o f peroxides; the so-called un-
The application of the same reaction to the manufacture of reactivo paraffin h ydrocarbon s react under these conditions with
acetic anhydride has received considerable attention (SO). M eth m arked ease; w ith halogen carriers (such as iodine, sulfur, etc.)
ods are described involving the action of sulfuryl chloride on the reverse phenom enon is observed. M oreover, th e reactive
sodium and calcium acetates in varying proportions to yield benzenoid h ydrocarbon s d o n o t take part in th e photochem ical
either the acid chloride or the acid anhydride (74). In one in and sulfonation reaction w ith sulfuryl chloride in w hich the ali
genious process, sulfur dioxide and chlorine are passed into acetic ph a tic h ydrocarbon s so readily pa rticip ate; on the other hand,
anhydride, which catalyzes the formation o f sulfuryl chloride, th e aliphatic h ydrocarbon s are n ot affected' b y th e conditions
and the mixture is then run onto anhydrous sodium acetate. (alum inum chloride and sulfuryl chloride) w hich so readily lead
After completion of the reaction, the acetic anhydride is re to th e sulfonation o f a rom atic hydrocarbons.
covered by distillation (I). These anom alies are readily explained on ce the assum ption is
m ade th a t a rom atic h ydrocarbon s react prim arily b y an ionic
M IS C E L L A N E O U S A P P L IC A T IO N S or polar ty p e m echanism , whereas paraffin h ydroca rb on s react
b y a m echanism in volvin g atom s or free radicals (34, 36, 39).
The treatment of /3-naphthylamine with sulfuryl chloride in
T h is interpretation n ot on ly clarifies the poin ts ju s t raised, but
the presence o f pyridine, followed by addition of calcium oxide
indicates th e directions in w hich th e search for new low -tem pera-
and sublimation of the product, leads to a 98 % yield of a,0-
ture reactions o f the paraffin h ydrocarbon s should proceed (37).
naphthazine (63):
LITERATURE CITED
(1) Akt.-Ges. für Anilin-Fabr., German Patent 244,602 (1909).

(2) Allihn, F., Ber., 11, 567 (1878).
(3) Armstrong, H. E., and Ro3sitor, E. C., Chem. News, 59, 225
r—N H S (1889).
(4) Badische Anilin- und Soda Fabr., French Patent 433,309 (1911).
(5) Barnett, E., Cook, J. W ., and Grainger, H . H., J . Chem. So*.,
121, 2059 (1922).
(6) Battogay, M., and Doniville, L., Bull. soc. chim., 53, 1242 (1933).
(7) Baumgarten, P., Ber., 60, 1174 (1927).
(8) Behrend, P., Ibid., 9, 1334 (1876).
The nature o f the products obtained by the analogous treatment (9) Behrend, R ., Ann., 222, 116 (1884).
o f a-naphthylamine and benzidine has not yet been elucidated. (10) Binkley, TV. TV., and Degering, E. F., J. Am . Chem. Soc., 61, 3250
(1939).
Treatment o f a-trioxymethylene with sulfuryl chloride gives (11) Bobrański, B., Kochańska, L., and Kowalewska, A ., Ber., 71,
chloroacetic acid in good yield (39). 2385 (1938).
Sulfuryl chloride catalyzes the polymerization of cyclopenta- (12) Bocchi, O., Qazz. chim. ital., 26, II, 403 (1896).
(13) Böesoken, J., Rec. trav. chim., 30, 381 (1911).
dieno to “ cyclopentadiene rubber" (69).
(13A) Bosshard, Steinitz, and Strauss, German Patent 413,724
It is claimed that good yields of a CiiHn hydrocarbon (prob (1921).
ably dimethyldicyelopentyl) are obtained by the action of sul (14) Brooks, B. T „ and Smith, D . F., U. S. Patent 1,235,283 (1917).
furyl chloride on cyclohexane in the presence of aluminum chlo (15) Bruck, W ., Ibid., 1,903,891 (1933).
(16) Buchman, E. R ., Ibid., 2,218,349 (1940).
ride (56).
(17) Bülow, C., and King, E., Arm., 439, 211 (1924).
Recently Friese and Djiang (38) reported that in the presence (18) Calcott, W . S., and Carter, A . S., U. S. Patent 1,896,160 (1933).
of a mixture of acetic acid and acetic anhydride, sulfuryl chloride (19) Chorbuliez, E., and Schnauder, O., Helv. ffhim. Acta, 6, 249
reacts with cyclohexene to form the acetyl ester of cyolohexene (1923).
(20) Crom pton, H., and Walker, M ., J. Chem. Soc., 101, 958 (1912).
chlorohydrin:
(21) Dcinet, J., U. S. Patent 1,975,256 (1934).
(22) Dubois, E., Bull. acad. roy. mid. Belg., 42, 126 (1876).
(23) Dubois, E., Z. fü r Chem., 2, 705 (1866).
(24) Durrans, T . H ., J. Chem. Soc., 121, 44 (1922).
Hs [ ||H + S 02C12 + CHjCOOIf + (CH3C 0 ),0 — =-
(25) Eller, W ., and Klemm, L., Ber., 55, 217 (1922).
Hs |H
(26) Elliot, G . H ., and Speakman, J. B., J . Chem. Soc., 1940, 641.
n-2 n
(27) Franchimont, A . P. N., Rec. trav. chim., 3, 419 (1884).
(28) Frieso, H ., and Djiang, D ., Ber., 71, 667 (1938).
Hi f ^ < \ C l
(29) Fuchs, K ., and Kätscher, E., Ibid., 57, 1256 (1924).
Hl J /H (30) Hewitt, J. T ., and Lumsden, C. H ., J . Soc. Chem. Ind., 35, 210
y ^ ococH i (1916).
Xlj
(31) Höchster Farbw., German Patent 286,489 (1915).
(32) Jadhavo, G . V ., and Sukhatankar, D . R ., J. Indian Chem. Soc.,
Neither sulfuryl chloride and acetic acid nor sulfuryl chloride and 15, 649 (1938).
acetic anhydride are capable of inducing the reaction. All three (33) Jadhave, G . V., and Sukhatankar, D . R ., J . Univ. Bombay, 8,
o f the components (sulfuryl chloride, acetic acid, and acetic an Pt. 3, 170 (1939).
(34) Kliarasch, M. S., and Brown, H . C., J . Am . Chem. Soc., 61. 2142
hydride) are necessary. (1939).
The use o f sulfuryl chloride as an antishrink agent for wool (35) Ibid., 61, 3432 (1939).
m ay be mentioned. This process has received considerable at (36) Ibid., 62, 925 (1940).
(37) Ibid., 64, 329 (1942).
tention and recently Elliot and Speakman (36) sought an ex
(38) Kharasch, M. S., and Brown, H . C ., U. S. Patent 2,302,228
planation for this property of sulfuryl chloride. (1943).
(39) Kharasch, M . S., Chao, T . H., and Brown, H. C., J. Am . Chem.
T H EO R E T IC A L IM P L IC A T IO N S Soc., 62, 2393 (1940).
(40) Kharasch, M . S., Chao, T . H ., and Brown, H . C., unpublished
Th e study of sulfuryl chloride reactions has not yielded results work.
(41) Kharasch, M . S., and Read, A. T ., J . Am. Chem. Soc., 61, 3089
o f interest only to the practical chemist, as the discussion might
(1939).
seem to indicate. Far-reaching in its implications to theoretical (42) King, H . S., and Morriam, M . K ., Proc. Nova Scotian Inst. Sei.,
organic chemistry is the light which has been cast on the relation 18, 276 (1933-34).
(43) Köchlin, P., and Heumann, K ., Ber., IS, 1736 (1882). (62) Reinhard, G., J. pract. Chtm., [2] 17, 321 (1 8 78 ).
(44) K u m , M . A., Kooberle, K ., and Berthold, E., U. S. Patent (63) Reitzenstein, F., and Andre, F., Ibid., [2] 87, 101 (1 9 13 ).
1,900,540 (1934). (64) Salibil, I., Chem.-Zto., 35, 9 7 ( 1 9 1 1 ) ; Chem. Zenlr., 1 9 1 1 , 1, 1581.
(45) Kyridos, L. P „ I b id ., 2,171,494 (1939). (65) Silberrad, O., J . Chcm. Soc., 119, 202 9 (1 9 21 ).
(46) Ibid ., 2,171,495 (1939). (66) Ibid., 121, 1015 (1 9 2 2 ).
(47) Lovaillant, R ., and Simon, L. J., Compt. rend., 169, 854 (1919). (67) Ibid., 127, 2 6 7 7 (1 9 2 5 ).
(48) Macboth, A. K ., J. Chetn. S oc., 121, 1116 (1922). (68) Silberrad, O., Silberrad, C. A ., and Parke, B., Ibid., 127, 1724
(49) M cK ee and Sails, U. S. Patent 1,765,601 (1930). (1 9 25 ).
(50) Marrara, G., and Borgo, A., Gazz. chim. ital., 35, IX, 563 (1905). (69) Staudingor, H., and Bruson, H . A., Ann., 4 4 7 , 110 (1 9 26 ).
(51) Ibid., 3 6 ,1, 348 (1906). (70) Töhl, A ., and Eberhard, O., Ber., 2 6 , 2 9 4 0 (1 8 93 ).
(52) Mazzara, G ., and Lamberti-Zanardi, M ., Ibid., 26, II, 236 (1896).
(71) Töhl, A., and Framm, F., Ibid., 27, 2 0 1 2 (1 8 9 4 ).
(53) Ibid., 26, 11,399 (1890).
(72) Trautz, M ., Z. Elektrochem., 14, 534 (1 9 0 8 ).
(54) M eybeck, J., A n n . chim ., 17, 129 (1932).
(73) Ullmann, F., and Conzotti, A., Ber., 5 3 , 8 2 6 (1 9 20 ).
(55) MoncsB, E., U. S. Patent 2,151,137 (1939).
(50) Nenitzescu, C . D ., and Isacescu, D . A ., Ber., 67, 1391 (1934). (74) Verein für Chemische Industrie in Mainz, German Patent 1 61,-
(57) Peratonor, A., and co-workers, Gazz. chim. ital., 2 8 , 1, 197 (1898). 8 8 2 (1 9 0 2 ).
(58) Perkins, M . A., and Deinet, J., U. S. Patent 2,180,836 (1939). (75) Wilke, K ., U. S. Patent 1,8 7 8 ,0 5 0 (1 9 3 2 ).
(59) Perrine, T . D ., Ibid ., 2,216,574 (1940). (76) W ohl, A., German Patent 139 ,5 5 2 (1 9 01 ).
(60) Pongratz, A ., and Eicliler, E., Ber., 69, 1292 (1936). (77) W ohl, A ., and K och, F., Ber., 4 3 , 3 2 9 5 (1 9 1 0 ).
(61) Raiziss.G.W ., and Clemence, L. W ., U. S. Patent2,102,854(1937). (78) W ynne, W . P., J. Chcm. Soc., 6 1, 1042 (1 8 9 2 ).
F. P. B A L D W I N , L. B. T U R N E R , A N D R. L. Z A P P
-Esso Laboratories, Standard O i l Developm ent C om pany, Elizabeth, N. I.
The most rapid sulfur vulcanization of G R - I requires the use of ultra pHE early recognition by the rubber industry of the rela
accelerators. Increase of accelerator concentration above certain tively low cure rate of sulfur-vulcanized Butyl rubber and
points produces inferior original physical properties with no ad its effect on factory production led to considerable work on
vantage in cure rate. A n excess of zinc oxid e over that necessary means of increasing the cure rate. One obvious method is to in
to obtain maximum tensile is required to produce the maximum in troduce more unsaturation into the polym er; but oven though
modulus and vulcanízate stability. For Iow-temperature curing the the cure rate of G R -I has been improved over that of the early
metal salt accelerators should be used, alone or in conjunction Butyl {11), it is still considered a slow-curing polymer. With a
with other accelerators to obtain fast cure rate. Reversion is apparent realization of the limitations imposed upon G R -I because of its
in G R - I vulcanizates if curing times are long enough or temperatures inability to fit into production schedules designed around the iise of
are high enough. Decreasing the rate of vulcanization with less natural rubber, this study presents some important aspects of the
active accelerators delays appearance of the reversion phenom enon. sulfur vulcanization of G R -I. Ever since the discovery and de
velopment of organic accelera
tors, the search has been con
tinued for a simple method
1.5 M E R C A P T O B E N Z O - 1.3 B E N Z O T H IA Z Y L 2 - of relating physical proper
1.0 T E T R A M E T H Y L - 1.27 Z IN C D IM E T H Y L TH 1AZ0LE M ONOCYCLOHEXYL ties to cure state. Early
THRJRAM D IS U L F ID E D IT H IO C A R B A M A T E 0.25 DIPHENYLGUANtDINE S U L F E N A M ID E
attempts to relate combined
1200
M O D U L U S 1000
sulfur to physical properties
AT 300 % BOO had little significance {5,10 ).
L B S / S Q .IN . 600 Recently (9) methods for
400 isolating definite forms of
200 combined sulfur have been
developed, and it is possible
T E N S IL E 1800
that these may lead to
S T R E N G T H 1600
better r e la t io n s b e tw e e n
L B S / S Q . IN . 1 4 0 0
chemically combined sulfur
of a definite type and phys
800
E L O N G A T IO N ical properties.
600
% M any physical tests have
400
been utilized to determine
TEAR 200
state of cure. B y a statis
L B S / IN . 1 00
tical treatment of cross-
linked polymers (4) it is pos
HEAT 80
sible to relate rubberlike elas
B U IL D - U P 60
ticity and swelling capacity to
°C. 40
concentration of cross links.
15 3 0 60 60 120 3 0 6 0 120 24.0 *TŹ0 240
Of these relations, the prop
T IM E O F C U R E - M IN U T E S A T 307°F.
erty of rubberlike elasticity
Figure 1. Effect of A cce lerator T yp e on G R - I V u lcanization in a modified form is the most
tiou, tear, and hysteresis

0.6 S U L F U R 1.2 S U L F U R 2.4- SU LF U R 4.8 SU LF U R properties at a given modu
lus. The greatest discrepancy
occurs in the hysteresis of
the tetramethylthiuram di
sulfide stock and may bi-
due to the fact that the
blocks were given a 15-minute
“ cure lead” in the press
which would tend to favor
T E N S IL E
the fast-curing compounds.
N U M B E R S ON C U R V E S R E F E R T O A C C E L E R A T O R C O N T E N T Neglecting this one dis
MODULUS crepancy, it is obvious that
the time necessary to ob
tain comparable cure for
compounds containing other
than the ultra-accelerators
tends to make factory' use
of these accelerators pro
20 40 60 8 0 100 a 20 40 60 8 0 100 0 20 40 60 80 100 120
T IM E OF C U R E M IN U T E S A T 307°r. hibitive. On a relative basis
the compounds exhibit the
Figure 2. Effect of Tetramethylthiuram D isulfide-Sulfur A cce leration on Tensile and M o d u lu s following cure times:
convenient to measure. For this reason the extension modulus has A ccelerator M inutes
T etram ethylthiuram disulfide
been adopted here as a measure of state of cure. It is realized t hat Zinc dim ethyl dithiocarbam ate
in going from one accelerator to another this criterion is not valid M ercaptobenzothiazole 1
D iphenylguanidine Í
for properties such as aging, creep, vulcanízate stability, etc. B enzoth iazyl 2-m on ocycloh exy l sulfenam ide
Further, it has been indicated (S ) that some of the accelerators
such as tetramethylthiuram disulfide are capable of producing
Thus the need for ultra-accelerators becomes apparent, and
carbon-carbon bonds in addition to the carbon-sulfur-carbon
further work on acceleration was limited primarily to this type of
bonds assumed to be formed during the sulfur vulcanization of
compound.
rubber and rubberlike polymers. However, the formation of such
E f f e c t o f S u l f u r - A c c e l e r a t o r R a t i o . A s a basic study
cross links would probably not change the general properties of
0.25, 0.5, 1, 2, and in one case 4 parts of tetramethylthiuram di
the vulcanízate.
sulfide were used in combination with sulfur dosages of 0.6, 1.2,
2.4, and 4.8 parts in the scinireinforcing furnace black compound.
R O L E O F A C C E L E R A T O R IN V U L C A N IZ A T IO N O F GR-I
Tensile and modulus are plotted against cure time at 307° F.
A c c e l e r a t o r T t p e . Although a wide variety of accelerators in Figure 2.
can be utilized for the sulfur vulcanization of natural rubber and An increase in sulfur concentration up to 2.4 parts produces an
highly unsaturated rubberlikc polymers, the choice of accelerator, increase in cure rate, as judged byr the increasing slope of the
for the satisfactory sulfur vulcaniza
tion of G R -I is relatively' narrow.
This has resulted in the almost ex
2000
clusive use of the so-called ultra T E N S IL E
T E N S IL E
accelerators or combinations of them
1600
with other chemicals. This point 1600
was brought out in previous papers £
(ft“, I I ) , but it is felt that the pres- ^ 1200 1200
entation of more quantitative evi- a

dence is advisable. For convenience, ^ aoo
the work on the effect of accelerators §
MOO. 3 0 0 %
has been confined to one basic com
pound as follows: G R -I, 100 parts;
zinc oxide, 5; stearic acid, 3; sulfur,
2; semireinforcing furnace black,
ELONGATION
54; accelerator as noted. The
,'fc E L O N G A T IO N
G R -I contained 1 part of zinc stear
ate and 0.5 part of phenyl-/3-naph-
thylamine.
NUMBERS REFER TO
Figure 1 shows the effect of P A RT S O F ZINC OXIOE N U M B E R S R E F E R TO
four different accelerators on the P A R T S O F S T E A R IC A C ID
physical properties of this G R -I 0 20 40 60 60 100 120

compound. With modulus as a TIM E OF C U R E -M IN U T E S AT 307 F. T IM E O F C U R E - M I N U T E S A T 3 0 7 ° F .
criterion for measuring state of

Figure 3. Effect of Z in c O x id e Figure 4. Effect of Stearic A c i d on Tetra
cure, all of the compounds show methylthiuram Disulfide Acceleration,
A ctivation on Tetramethylthi
essentially the same tensile, elonga- uram Disulfide A cce leration A ctivate d with 5 Parts of Z in c O x id e
12 0 0 to obtain maximum tensile (Figure 3), but the modulus and elon
z 1000 gation curves show that a low state of cure and an unstable
a
VI cure results. The instability of the vulcanizate, as noted by the
800
\
CD 6 0 0 change in modulus over the cure range, persists at 1 part zinc
oxide concentration and is discussed in detail in a later section.
400 X ! 204
When no zinc oxide is used, the maximum tensile obtained is low
200 and the modulus deteriorates over the entire time range. With
7 ! /
55 110 220 440 25 50 IOO 200 increasing amounts of zinc oxide (from 2 to 10 parts), state of
C U R E A T 2 6 0 ° F . - M IN U T E S C U R E A T 2 0 0 ÖF . - M IN U T E S cure as measured by modulus is increased, and maximum moduli
1200 are obtained with about 5 parts of activator. B y increasing the
Z
IOOO activator to 10 parts, little or no improvement results.
Ü
800 The results of considerable work relative to the effect of fatty
CD
600
acids on the activity of zinc oxide have been published (7, 8, H ).
-I
In general, the conclusions reached are that fatty acids function
400
1 46
r/ ^ » 23
b y converting insoluble zinc oxide to soluble zinc stearate. How
200
1 /
ever, Figure 4 indicates that the addition of stearic acid to this
II 22 45 90 5 10 20 40 com pound has no effect on vulcanization rate but produces
CURE A T 3 0 O ° E - M IN U T E S CUR E A T 3 2 0 V .- M 1 N U T E S slightly lower moduli, possibly through plasticizing action. It is
possible that the small amount of zinc stearate present in G R -I
Figure 5. M o d u lu s at 5 0 0 % Elongation with Z in c D ib u tyl
Dithiocarbamate A cce leration m ay be sufficient to render additional stearic acid relatively in
effective; however, the authors believe that the presence of zinc
oxide as such is necessary when dealing with this type of acceler
ation.
modulus curves in the early stages of cure. An increase over 2.4 A c c e l e r a t o r T im e -T e m p e r a tu r e R e la t io n s . Because of
parts produces little or no change in this respect and results in the complexity of the vulcanization reaction, any time-tempera-
profuse blooming of the cured samples. In all but the 4.8-part- ture relations established must be rather arbitrary. The practice
sulfur stock an increase in cure state and rate of cute accompanies has been to relate the logarithm of cure time to temperature in
incremental increases in accelerator concentra
tion up to 1 part; but in going from 1 to 2 parts ___________
accelerator, there is a tendency toward reduc
tion in cure state (decreased modulus) with Table Effect of V a rio u s A ccelerators on V u lca n iz a tio n Time-Temperature
Relations
little change in cure rate. With 4.8 parts of
sulfur there is an increase in cure state over the _________ Tensile (L b ./S q . In .)-M o d u lu 3 at 5 0 0 % -E lo n sa tio n ( % )
1 part tetra- 1 part zinc 1 part selenium 1 part tetra-
whole range of increasing accelerator content. Cure m ethylthiuram d ibu tyl dithio diethyl d ith io m ethylthiurain
, p _ m onosulfide
W ith respect to tensile strength, 1.2 parts M in. disulfide carbam ate carbam ate
of sulfur are sufficient to obtain maximum 260 55 1 3 8 0 - 350-1000 1370- 270-1060 165 0 - 410-860 1510- 270-940
110 173 0 - 5 1 0 - 860 1800 - 4 9 0 - 860 193 0 - 740-760 180 0 - 580-810
values. The most rapid-curing compounds 220 1 8 6 0 - 7 4 0 - 780 2 0 1 0 - ■ 8 1 0 - 780 2010-1160-700 1 84 0 - 830-710
440 1 05 0 -1 21 0 - 680 1970- -1270- 680 1800-1430-600 1900-1150-680
(those with high sulfur and accelerator) show
the greatest decrease in tensile, after a maxi 280 25 160 0 - 3 9 0 - 880 1330- 210-1060 163 0 - 340 -8 4 0 161 0 - 450-870
50 1 8 2 0 - 6 7 0 - 800 172 0 - 4 5 0 - 860 1900- 630-820 190 0 - 730-790
mum has once been obtained, with increase in 100 1 94 0 -1 01 0 - 720 2 1 2 0 - 8 6 0 - 740 1920-1000-710 2020-1060-710
200 1 82 0 -1 31 0 - 600 181 0 -1 20 0 - 700 1720-1340-610 1800-1310-650
cure time. The tensile plateaus may be in
creased either by .the use of low sulfur and 300 11 1 71 0 - 4 1 0 - 900 118 0 - 170-1100 159 0 - 280-930 172 0 - 400-920
22 1 83 0 - 7 9 0 - 800 177 0 - 4 3 0 - 900 1 8 2 0 - 570-830 2 0 2 0 - 700-820
high accelerator, high sulfur and low accelera 45 1 8 3 0 -U 3 0 - 080 186 0 - 8 0 0 - 800 187 0 - 970-6SO 1880-1130-710
90 1 7 2 0 -1 2 8 0 - 640 1 8 4 0 -1 0 6 0 - 750 1760-1270-640 1710-1280-620
tor, or with both low sulfur and accelerator.
Over-all, then, for the proper balance of cure 320 5 1 62 0 - 4 0 0 - 880 1 3 6 0 - 180-1020 150 0 - 270-930 174 0 - 460-870
10 192 0 - 7 8 0 - 800 171 0 - 3 9 0 - 910 178 0 - 560-800 2 00 0 - 850-770
rate and initial physical properties care should 20 187 0 -1 03 0 - 710 182 0 - 7 4 0 - 790 198 0 - 840-760 2110-1260-700
40 156 0 -1 24 0 - 590 1 95 0 -1 11 0 - 750 1790-1190-670 1850-1460-590
be taken in the choice of accelerator-sulfur
combinations. It should be kept in mind that
the values shown here are not criteria for all
compounds and will vary with pigment type
240 O
and concentration.
180 - \
R o le o f A c tiv a to r s . M etallic oxides, and m = 6.86X10 •
140 ‘ \ m = 6.79 x io -3
specifically zinc oxides, have long played a role
in the sulfur vulcanization of unsaturated IOO ; \
polymers. W ith G R -I, zinc oxide activation is 80

\
necessary for proper vulcanization. N ot only is
O
o
S 31D N
sufficient zinc oxide needed to form a metal salt ; \

*
O
if a nonmetallic dithiocarbamate is used ( / , 8),

■f 3 0 \
but an excess over this amount is required to
TIME
obtain maximum moduli. This is in accord

f\j
o
with observations made on natural rubber

with dithiocarbamate acceleration (18).
15 \
m = 4 .8 3 X 1 0 "
10 . TE TR A M E TH Y L- ZINC DIBUTYL .SELEN IU M DIETHYL . TETR AM ETH YL-

Figure 3 shows the effect of increasing
DITHIOCARBAMATE .THIURAM MONOSULFTd I:
amounts of zinc ox.ide on the physical properties 8 ; t h iu r a m DISULFIDE DITHIOCARBAM ATE
t e m p - “ f. T E M P - F. TE M P - F.
over a cure range at 307° F. In this series no 260 260 300 320 280 300 260 280 3 00 320
stearic acid was used in the compound which R —• ■
2.3 2 .5
included 54 parts of SR F black, 2 parts of sulfur, -4 —XI03
and 1 part of tetramethyltliiuram disulfide.
Only about 0.5 part of zinc oxide is needed Figure 6. V ulcanization Time-Temperature Relations
be measured most accurately was chosen, and

the time required at each temperature to reach
this predetermined modulus was ascertained.
/
J
\ \
o -* -*
I"
- \ 4 These time points were plotted as shown in
B A S E FORMULA
Figure 6, and the slopes of the lines determined.
V -
3, - \
Ö.
"*■ ----- TE N S IL E ----- GR- 100.0 There is considerable difference in slope
FINE THERM AL B L K . 7 2 -0
. ö
SLA B S CUR ED 3 0 / 3 2 0 F.
SULF UR
TMT 3S
2 .0
IX) between the metal salt accelerators and the
others:
-+-AGED IN ST EAM A T 3 2 0 °F .
A ccelerator Slope
UJ >
§ Tetram ethylthiuram disulfide 6 .7 9 X 1 0 -3
a 600 Tetram ethylthiuram m onosulfide 6 .6 6 X 10
Z in c dibu tyl dith iocarbam ate 4 .8 3 X 10“ 3
400 ----- E LO NC A T IO N ----- Selenium dieth yl dith iocarbam ate 4 .7 2 X 10 ~3
Z 600 ----- M O D U LU S AT 300% E —
$ 400 Therefore, b y adjustment of accelerator con

tO 2 0 0
centration, curing at low temperatures can best
k0 3 0 s150 210.2 70 be aecomplished by the use of metallic dithio-
90 030 90 150 210 2 7 0 030 90 150 210 2 7 0 030 90 150 210 2 7 0
carbamatos either alone or in conjunction with
TIM E OF A G IN G -M IN U T E S
other accelerators. Compounds containing both
0 .5 SULFUR 1.0 SULFUR 1.5 SULFUR 2 .0 SULFUR a thiuram and a metallic dithiocarbamate have
been shown to be applicable to these low-tem-
perature cures.
V U L C A N IZ A T E STABILITY
In the vulcanization of natural rubber the

phenomenon of reversion is well known. In
G R -I reversion is obtained only after loiig cure
times. However, since reversion is wholly rela
tive in nature, this is not surprising.
E f f e c t o f Z i n c O x i d e . A s pointed out earlier,
very little activator is necessary to obtain
maximum tensile values (a criterion used to
judge best cure). However, it was shown that
a small quantity is insufficient to produce ulti
4 i r r í 0 ^ ? ?0 0 3 0 9 0 150 210 2 7 0 0 3 0 9 0 5 0 <?¡0 3 7 0 0 3 0 9 0 150 310 2 7 0
mate stability of the vulcanizate.
TIM E OF AGINO - M IN U T E S
The upper graphs of Figure 7 present the ten
Figure 7. Effect of Z in c O x i d e and of Sulfur on G R - I V u lcanízate Stability sile, modulus, and elongation data obtained in an
accelerated reversion test. Compounds contain
ing 2.5, 5, 10, and 20 parts of zinc oxide and ac
degrees Fahrenheit. This m ethod of expression should be con celerated with 1 part of tetramethylthiuram disulfide were cured
fined to rather narrow temperature ranges; further, it does not for 30 minutes at 320° F. W hole tensile slabs were then aged
bring out the differences in scorch characteristics among certain both in the press and in the steam digester for various times at
accelerators having essentially the same cure rate at a definite cur 320° F. Judging reversion by modulus decrease, the compound
ing temperature. In this paper relative cure state is confined containing the lowest amount of zinc oxide shows the greatest
to Btate of cure as indicated by extension modulus, and the rate of reversion. Furthermore the steam-aged slabs show lower
relation of log time to 1 /T° K . is used. moduli than the press-aged slabs. The data indicate that the
The accelerators examined were tetramethylthiuram disulfide reversion tendency is reduced with incremental increase of zinc
and monosulfide, zinc dibutyl dithiocarbamate, and selenium oxide up to 5 parts. An increase beyond this point does not have
diethyl dithiocarbamate. The curing temperatures encompassed an appreciable effect in this compound.
the range 260-320° F. with cure times adjusted accordingly. The fact that an excess of zinc oxide over and above that neces
Tensile, modulus, and elongation data are shown in Table I. sary to activate the accelerator is found necessary to decrease the
M oduli for each of the compounds were plotted as illustrated in reversion tendency o f these compounds suggests that hydrogen
Figure 5; a modulus point on the curve where the changes could sulfide (or in general a mercaptan) is at least partly responsible.
W eyl (13) states that hydrogen sulfide can
decompose disulfides as follows:
R — S— S— R + H2S — 2RSH + S
Table II. Effect of H y d ro g e n Sulfide on Crude and V u lc a n iz e d G R - I
It is also known that monosulfides can undergo
Properties of C rude P olym er Properties of V ulcanised P olym er0 similar reaction:
Tensile, M odulus V ol. swell
Staudinger Intrinsic lb ./s q . in. at 3 0 0 % in cy clo- R — S— R + IRS — 2RSH
T est C ondition m ol. wt. viscosity (elongation, % ) elongation hexane, %
C ontrol 40,600 1.3 0 1620 (630) 620 227 Therefore it is possible for hydrogen sulfide
H eated 2 hr. at 255° F.
in H jS at 50 lb ./s q . in. 40,600 1 .3 0 1640 (820) 150 377 to destroy the sulfur cross links formed whether
H eated 2 hr. at 255° F. they are sulfide or disulfide type.
in vacu o 40,300 1 .2 8 1510 (590) 660 220
In support of this proposed mechanism,
a Basic com pound accelerated with 1 part tetram ethylthiuram disulfide, and cured 60 m in
utes at 307° F.
specimens of crude and vulcanized G R -I were
heated for 2 hours at 255° F., both in
2 .0 M E R C A P T O - shown is relatively unaffected

1.0 T E T R A M E T H Y L ' 1.0 T E T R A M E T H Y L “
1.5 T R IB U T Y LID E N E- 1.3 ZINC DIBU TYL B EN Z O T H IA Z O LE +
THIURAM THIU RAM by changes in sulfur concentra
ANILINE DITH IOCARBAM ATE 0 .5 D IPH EN YL“
D ISU L F ID E M ONOSU LFIDE tion over the time-temperature
GUANIDINE
range used here.
1600 BA SE FORM ULA — T E N S IL E — S L A B S CURED 30'/320°F.
GR-I Í00.0 E ffect of A c c e l e r a t o r
O - AGED IN P R E S S A T 3 2 0 ° F
ZINC O XIDE 5.0
¿1400
FINE TH ERM A L BLK.72.0
+ -A G E 0 IN ST E A M AT 3 2 0 ° F. It was pointed out earlier T ype .
SULFUR 2.0
g '200 that factory usage of accelera
SlOOO tors other than ultra-accelerators
J
800 is probably undesirable; but the
Z 800 — ELONGATION — effect of such accelerators on the
P __% reversion of G R -I are included
-A-"*
a 600
LJ here as a matter of record. The
a 400
accelerated reversion test was
. aoo M ODULUS AT 3 0 0 % E
extended to compounds contain
$ 600 ing several accelerator types,
è <*00- and the results are shown in
_l
200 Figure 8 . The modulus curves
0 J L. i I I I I i J_J L_ show that the use of less potent
0 3 0 9 0 150 210 270 0 30 9 0 150 210 270 0 30 9 0 150 210 270 0 30 9 0 150 210 2 7 0 0 30 90
TIM E OF AGING IN M INUTES accelerators decreases but does
not eliminate reversion tendency.
Figure 8. Effect of A cce lerator T ype on G R - I V u lcanízate Stability During G R -I vulcanization,
many reactions are probably
occurring; but as an oversim
sealed vacuum tubes and in the presence of hydrogen sulfide at 50 plified picture, the vulcanization process may be thought o f as
pounds per square inch pressure to ensure adequate permeation. consisting of tw o major types of reactions. One would be the
The results and control data are presented in Table II. The hy formation of a three-dimensional network, and the other would
drogen sulfide treatment had no noticeable effect on the crude be degradation of both the individual polymer chain and the
polymer, as shown by the Staudinger molecular weight and in network. Therefore the appearance of reversion as indicated
trinsic viscosity data. The vulcanizate, however, when treated by modulus decrease will depend to a large extent upon the ratio
in this manner shows a high degree of reversion, as indicated by of the rates of these general reactions to one another.
the decrease in modulus and the increase in swelling capacity. Of probable interest is the w avy appearance of some of the
Heating in vacuo increased the modulus and decreased the tensile curves in this reversion study. This may be explained
swelling capacity, and both these effects can reasonably be at by an orientation mechanism. Theoretically maximum tensile
tributed to continued cure. should be obtained when maximum chain length is obtained con
We would expect, therefore, that some mercaptans may behave current with maximum orientation. Thus the ideal vulcan
like hydrogen sulfide, according to the following equations: izate from the standpoint of tensile strength would exist when a
continuous molecule is produced in which the polymer units are
R — S— S— R + R 'S II R — S— R ' + R — SH + S hooked up, say, in an end to end fashion. Any increase in sulfur
linkages over this point of maximum chain length would cause
R —S — It + R 'S H —■R — S— R ' + R — SH rigidity and reduce the degree o f orientation.
Thus it is felt that up to the point of maximum tensile effective
According to this reasoning, zinc oxide would reduce this tend molecular weight is being increased. After this point additional
ency toward reversion by reacting with the hydrogen sulfide to linkages cause an increase in modulus but a decrease in tensile
form zinc sulfide, b y virture of their effect on the m obility of the polymer units
with respect to one another. As reversion progresses, whether
H»S + ZnO — ZnS + H ,0 cleavage occurs in the chain proper or at the points of cross link
age, the effect is to remove some of thé excess linkages; and as
or in the caso of mercaptans to form a metal salt, expected, modulus decreases and tensile increases. The decrease
in tensile observed after the second peak can be assumed to be a
2RSH + ZnO - * Zn(S— R ), + H.O progressively effective decrease in over-all molecular weight.
Either hydrogen sulfide or a mercaptan m ay possibly be gener

LITERATURE CITED
ated by many of the reactions taking place during vulcanization.
It is assumed that the degradation which takes place in the pres (1) Bedford and Sebrell. J. I n d . E n g . C hem ., 14, 25 (1922).
ence of a large excess of zinc oxide may be attributed to degrada (2) Clark, LeTourneau, and Ball, Rubber Chem. Tech., 16, 621-33
tion of the chain proper by an oxidative process or some free radi (1943).
(3) Farmer, Ibid., 15, 774-9 (1942).
cal mechanism. (4) Flory and Rehner, J. Chem. Phye., 11, 512-26 (1943).
It seems reasonable to conclude that, in the vulcanization of (5) Hardman and White, I n d . E n g . C hem ., 19, 1037 (1927).
GR-I with this accelerator type and concentration, the major (6) H aworth and Baldwin, Ibid., 34, 1301 (1942).
portion of the cross links formed is composed of sulfur links, since (7) Martin and Davey, J. Soc. Chem. Ind., 44, 317T (1925).
(8) Sebrell and Vogt, I n d . E n g . C hbm ., 16, 792 (1924).
carbon-carbon links would be relatively stable under the condi
(9) Selker and Kemp, Ibid., 36, 16 (1944).
tions of these tests. (10) Shepard and Kroll, Ibid.. 14, 951 (1922).
E f f e c t o f S u l f u r C o n c e n t r a t io n . In a similar test the (11) Thomas, Lightbown, Sparks, Froiich, and Murphree, Ibid., 32,
amount of sulfur was varied, and the data obtained are given in 1283 (1940).
(12) Twis3, Brazier, and Thomas, J. Soc. Chem. Ind., 41, SIT (1922).
the lower graphs o f Figure 7. The compounds show similar
(13) W eyl, “ Methoden der Organische Chemio” , p. 243.
characteristics with regard to rate of reversion. Percentage (14) W hitby and Evans, J. Soc. Chem. Ind., 47, 122T (1928).
modulus decrease from the highest to the lowest point of each
curve is practically constant. This leads to the conclusion that P resen ted b efore the spring m eeting o f the D ivision o f R u b b e r C hem istry,
the reversion characteristic of G R -I compounds accelerated as N ew Y o rk , N . Y .
A m e r ic a n C h e m i c a l S o c i e t t , in
MECHANICAL PROPERTIES OF FILMS FROM
Amylose, Amylopectin, and Whole

Starch Triacetates
ROY L. W H ISTLER AND G. E. HILBERT
Northern Regional Research Laboratory, U. S. Department o f Agriculture, Peoria, 111.
S T A R C H is an abundant Am ylose and am ylopectin triacetatcs differ greatly in triacetate was reported as
and low-cost high poly film -form in g ability. Am ylose triacetate readily form s being soluble in acetone. All
mer, but has not been high -qu ality films having good tensile strength and plia specimens of amylose triace
utilized industrially as a raw b ility; am ylopectin triacetate resembles whole starch tate prepared in this labora
material for the production triacetate in form in g only weak, b rittle film s. Adequate tory, however, have •been
o f films, fibers, or plastics. plasticization o f the am ylose triacetate film s can he found aceton e insoluble.)
Films having tensile strengths accom plished by the addition o f 10-20% o f plasticizer o f T o evaluate the factors in
of 4 -6 kg. per sq. mm. can be the types em ployed with cellulose acetates. In general, fluencing film formation of
form ed from gelatinized the properties o f the am ylose triacetate films are sim ilar starch acetate, a thorough
starch (£?), but such films are to those o f cellulose triacetate films. Because o f their study of the properties of films
unsuitable for most industrial high qu ality and low plasticizer requirem ents, am ylose produced from its individual
applications. Although the triacetate films appear well suited to industrial uses. components seems desirable.
films are pliable when their This article deals with the
moisture content is high, at production and properties of
relatively low humidities they become quite brittle. Starch films prepared from the acetates of amylose and amylopectin frac
films, furthermore, are strongly hydrophilic. This latter dis tions of cornstarch. The fractions were separated according to
advantage is, in general, easily overcome by transforming starch the procedure of Sehoch (11), the ratio of amylose to amylopectin
to its esters. Films or plastics produced from whole starch tri being about 1 to 3. The most important properties of these
acetate or analogous derivatives, however, are very brittle and films and the effects produced by various plasticizers were deter
weak, even when large amounts of plasticizing agent are in mined with particular emphasis on slress-strain measurements.
corporated (2, S). Thus, whole starch, as well as its substitution Films of high tensile strengths and considerable pliability are ob
products, seems to be basically unsuitable for film or fiber pro tained from acetates of the amylose fraction, whereas only very
duction and also for plastics production, unless substitution in brittle nonself-supporting films with low tensile strengths can
volves an increase in polymerization. be obtained from acetates of the amylopectin fraction. The
As a result of the recent advances in the field of high-polymer latter films are even more brittle than those obtained from whole
chemistry it is known that poor film- or fiber-forming character starch acetate. Because of the high qualities of films produced
istics of polymeric materials may be attributed to: (a) relatively from amylose acetate, there seems to be little doubt that the
low molecular weight, (5) tangled or branched molecular con amylose fraction will assume considerable commercial importance
figuration, or (c) the presence of a component having poor film- if an economical method of isolation can be developed.
forming properties with one that is capable of forming films.
M ETHODS
In the case of starch, its poor film-forming properties cannot be
ascribed to too low a molecular weight since this appeals to be F b a c t io n a t io n . Prior to fractionation, cornstarch was
well above (1, 4, 13) the limiting value of 10,000- 20,000 required freed from as much fat as possible by methanol extraction (10).
for film formation (5, 12). Within the last few years it has been Both laboratory-prepared and common commercial starches were
definitely shown that starch consists of at least two com po subjected to fractionation by the use of butanol (11) with sub
nents (6,10 ), one presumed to have a linear configuration and the stantially equal results.
other a branched structure (6). This suggests that the brittle In order to obtain the amylose in a finely powdered form, the
ness of starch acetate films is due to the third cause mentioned centrifuged precipitate (I part) was strongly stirred with ethanol
above, the presence (5 parts), filtered, and
o f the b r a n c h e d th orou g h ly washed
amylopectin fraction. twice with ethanol.
This view receives T h e p r o d u c t w as
som e s u p p o r t in freed from alcohol by
M eyers’ recent state- drying in a vacuum
m c n t ( 7) t h a t desiccator over cal
evaporation of a solu cium chloride; yield
tion of amylose tri of crude amylose was
acetate leads to a 28% .
strong film, whereas After removal of
a poor film is formed amylose precipitate,
from amylopectin tri the amylopectin was
a ceta te. (A m ylose separated by pouring
796
the centrifugate into strongly stirred absolute ethanol (1 part of elastic deformation
solution to 5 parts of ethanol). The precipitate was then washed w hich obeys
and dried as described for amylose. H ooke’s law. When
A c e t y l a t i o n . The crude starch fractions were acetylated, the yield point is
and the products were worked according to the m ethod of Whist r e a c h e d , p la s t ic
ler, Jeanes, and Hilbert (14), using a 3-hour acetylation period. flow c o m m e n c e s .
Acetates prepared b y this procedure contain 44.6-44.8% acetyl. During the period
Acetates of the two starch fractions differ considerably from each of plastic flow, the
other in appearance. As Figure 1 shows, amylose triacetate is curve continually
highly fibrous, while amylopectin triacetate is a fine white powder. bends away from
P r e p a r a t i o n o f F i r m s . The following procedure was used the stress coordin
for preparing films: A definite weight of the amylose, am ylo ate, indicating an o
pectin, or whole starch triacetate (usually 2 grams) was dissolved increase in rate of
in 50 ml. of chloroform, and the solution freed from lint and dust flow as stress is ap
particles by passage through a coarse fritted glass funnel. The plied. Plastic flow
solution was concentrated at about 70° C. in a glycerol bath, and c o n t i n u e s to an
a definite weight of plasticizer was carefully mixed into the elongation of about
thickened solution when required. After concentrating to a 22% , at which point
10% solution, the mixture was allowed to stand several minutes the film ruptures.
to become free of bubbles and was cast on a clean glass plate with The relatively long
a casting knife 6 inches long and a blade set 0.020 inch above the r e g i o n o f plastic
glass plate. After about 4 hours, the film was loosened by mois flow is evidence that
tening the edges with water, carefully stripped from the plate, the film possesses P erce nt E lon ga tion
immediately blotted between filter paper, and placed in a con considerable inher
ditioning room at 21.1° C. (70° F.) and 50 % relative humidity ent" plasticity. Figure 2. Stress-Strain Curves
o f A m ylose and Cellulose T ri
for 12 days. The dried film formed was clear and 0.030-0.040 L i k e film s of acetate F ilm s C ontaining Vari
mm. (0.0012-0.0016 inch) thick. many other high ous A m ou n ts of Dibutyl
During the conditioning period, films rapidly lose weight, mainly polymers, the film P hthalate
as a result of volatilization of chloroform. Equilibrium is attained of amylose triace a . A m y lo s e t r ia c e ta te , c r o s s -s c c -
tio n a l area 0.311 s q . m m .
in about 10 days. Since chloroform exerts strong plasticizing tate decreases in b. C e llu lo s e t r ia c e ta t e , cr o ss -se c*
tio n a l a rea 0.222 s q . m m .
action, the films were subjected to 12-day conditioning to ensure tensile strength on A , IS, C , O , E , a n d F r e p r e s e n t film s
their testing under comparable conditions. The slow rate at which exposure to water. c o n t a in in g 0, 10, 20, 30, 40, an d
5 0 % d ib u t y l p h t h a la t e , r e s p e ctiv e ly .
the solvent evaporates from the film is comparable to the slow W h e n s o a k e d in
rate of solvent evaporation from cellulose acetate films (3). water for 24 hours
St r e s s -S t e a j n M e a s u r e m e n t . A Scott IP2 graphically re the film loses approximately one third of its original strength.
cording, inclined-plane serigraph was used which had a load ca Very little change is noted in the ultimate tensile strength on ac
pacity of 0-2000 grams and a constant loading rate of 36 grams celerated aging o f the film by maintenance at 65° C. for 12 days.
per second. The distance between film clamps was set at 50 Plasticizers m odify the stress-strain properties of amylose tri
mm. for zero load. All tensile strength values reported are cor acetate film. Stress-strain curves of films containing varying
rected for the reduction in area of cross section brought about by amounts o f dibutyl phthalate ¡ire shown in Figure 2a. On in
stretching, the assumption being made that no appreciable volume corporation of the plasticizer, the elastic deformation of the film
change occurred in the film during stretching. The tensile is decreased, the yield value is lowered, the pliability is increased,
strength values, therefore, pertain to the actual film cross section and the character of plastic flow is changed. The last effect is
at the time of rupture. Numerous measurements on a large num most pronounced in film containing about 20% dibutyl phthalate.
ber of films free of mechanical defects show that the reproducibil After the yield value is attained in such film, plastic flow com
ity of results is good. The tensile strength and elongation mences and proceeds at an increasing rate as stress is applied
values recorded are averages of at least ton tests conducted on until the film has stretched 2 0 -3 0 % ; thereupon the stress-strain
at least two different films. curve rises sharply and becomes concave to the stress axis. This
is evidence of increasing resistance to the applied stress and the
F I L M C H A R A C T E R IS T IC S
development of greater strength within the film. This property is
Amylose, amylopectin, and whole starch triacetates yield color
less, transparent, lustrous films differing greatly in strength and
--------------
flexibility. Films of amylopectin and whole starch triacetate are
weak and brittle, even after plasticization. The film of amylose
triacetate, on the other hand, possesses great strength and
flexibility; it is optically isotropic and exhibits the same tensile
strength in all directions. It swells and can be dispersed in pyri
dine, acetic acid, chloroform, and tetrachloroethane, and is in
soluble in acetone, alcohol, ether, and water.
Films were characterized mainly by determination of their
stress-strain relations. This provides a convenient means both • Tensile Strength O f Film
for comparing and evaluating films, and for rapidly determining ° Tensile Stren gth Of Film
the important mechanical effects produced in the films upon in Corrected For Plasticizer Present
corporation of plasticizing agents. Films from whole starch tri

acetate and amylopectin triacetate are so weak and brittle that
no measurements could be made on them. 20 30 40 5
Percent P la stic iz e r (D ibutyl P hthalate)
A typical stress-strain curve (A ) for unplasticized film prepared
from amylose triacetate is shown in Figure 2a. Application of an Figure 3. Effect o f Plasticizer on Tensile Strength o f
increasing stress to the unplasticized film produces a considerable A m ylose Triacetate Film s
desired in commercial films because, by proper stretching during

formation, films can be produced with a considerably increased T a b le II. P r o p e r t i e s o f U n p la s t ic iz e d F ilm s p ro m A m y lo s e
T r ia c e t a t e a n d fro m C e llu lo s e T r ia c e t a t e
strength in one direction. Such an increase in strength is due to
partial orientation of the film molecules and is typical of high Am ylose Cellulose
P roperty Triacetate Triacetate (ßA)
polymers having extreme linearity. Tensile strength, kg ./m m .* 7 .9 8 .6
Variation of tensile strength of the films with increasing dibutyl Elongation at break, % 22 4
R efractive index 1.469 1 .4 8
phthalate content is shown in Figure 3. As the solid line indicates, D ensity 1.341 1.4
a progressive weakening of the film occurs when increasing Hardness, R ockw ell M 6O-05
M odulus o f elasticity, k g ./in m .J ( X 1 0 1) 2 .2
amounts of plasticizer are incorporated. The broken line repre Capillary m elting p oint“ , ° C. 290-300 270-290
sents tensile strength values corrected for the plasticizer present, ° Determ ined on pow dered am ylose triacetate.
and, hence, calculated only for the amyloso acetate in the films.
T he initial slight rise in tensile strength is similar to that produced
by certain plasticizers in cellulose acetate films. tester before breaking. N. C. Schicltz of this laboratory called
Variation of elongation at rupture with increasing dibutyl our attention to tho possibility that the film might exhibit unusual
phthalate content is shown in Figure 4. As expected, the addi electrical properties. Tests carried out by a commercial company
tion of plasticizer causes the film to become softer, more plastic, on the insulating quality of plasticized and unplasticized amylose
and readily extensible. A t plasticizer contents of 50 % or more, triacetate film showed it to have an electrical breakdown value
the films are soft and weak. of 805-880 volts per mil. The plasticized film did not break down
when submitted to 200 volts of direct current for 3 weeks in a
humidity cabinet kept at about 120° F. The film, therefore,
compares favorably with the usual polymers or electric tapes used
for electrical insulation.
Of special interest is a comparison of the properties o f film from
amylose triacetate with those of film from commercial cellulose
triacetate. (Sample of high-grade commercial primary cellulose
acetate containing 44.2% acetyl, obtained from Hercules Powder
Com pany). The general similarity between the two films in such
properties as tensile strength, refractive index, density, and melt
ing point (Table II) is striking. Comparable stress-strain curves
obtained with films o f amylose triacetate and commercial cellu
lose triacetate (Figure 2) reveal some differences. The curves
show that the amylose triacetate film has considerably greater in
herent plasticity— that is, can be stretched to a greater extent.
In general, plasticizers found applicable to cellulose acetates are
F ig u re 4. Effect o f Plasticizer on Elongation o f Am ylose
T riacetate Film s likewise applicable to amylose triacetate, although the latter re
quires less plasticizer. Approximately 10 -20 % of plasticizer is
sufficient to give the amylose triacetate film a useful and lasting
Although most quantitative information was obtained with degree of plasticity, while commercial celluloso triacetates usually
dibutyl phthalate, other common commercial or potentially avail require 20 -40 % . The lower plasticizer requirement is an im
able plasticizers are suitable for plasticizing amylose triacetate. portant advantage in favor of amylose triacetate at present due to
Tensile strengths and elongation at rupture of films prepared from the critical shortage of plasticizing agents.
mixtures of amylose triacetate with various plasticizers are shown
in Table I. Tricresyl phosphate, dicthylcne glycol butyl ether A C K N O tV L E D G M E N T
acetate, pentaerythritol tetraacetate, and the butyl esters give
The writers wish to express their appreciation to R obert A.
results comparable to those obtained with dibutyl phthalate.
Deane and William D . Johnson of this laboratory for aid in carry
These are all excellent plasticizers for amylose triacetate. Films
ing out the work.
containing polyethylene glycol and dimerized methyl linoleate
become slightly hazy after several months. Glycerol is not com L IT E R A T U R E C IT E D
patible with amylose triacetate. (1) Beckman, C. O., and Landis, Q., J. Am . Chem. Soc., 61, 1495
In addition to tensile strength and elongation, a few other (1 9 39 ).
properties of amylose triacetate film are of interest. The film (2) Burkhard, C. A., and Degering, E. F., Rayon Textile Monthly,
23, 80 (1942).
exhibits good fold resistance. Strips 15 mm. wide and 0.03 mm. (3) Durrans, T . H „ and Davidson, D. G., J. Soc. Chem. Ind., 55,
thick, for example, undergo 400-650 double folds on a Schopper 162 (1 9 36 ).
(4) Foster, J. F., and Hixon, R . M ., J. Am . Chem. Soc., 65, 618
(1 9 43 ).
(5) Gloor, W . E., I n d . E n g . Chem., 27, 1162 (1935).
T a b le I. E x t e n s ib ilit y a n d T e n s ile S t r e n g t h o f A m y lo s e (5A) Hercules Powder Co., “ Hercules Cellulose Acetate” , 1941.
T r i a c e t a t e C o n ta in in g V a r io u s P la s t ic iz e r s (6) M eyer, K . H ., “ Advances in Colloid Sciences” , pp. 143-65, N e w
York, Interscience Publishers, 1942.
Tensile Strength, (7) Meyer, K . H „ Bernfeld, P., and Hohenemser, H ., Helv. Chim.
E xtensibility K g ./S q . M m .
P lasticizer (2 0 % ) (=*=5%) (=*=0.3) Acla, 23, 885 (1940).
P olyethylene glycol (C arbow ax 1500) 94 7 .4 (8) Mullen, J. W ., and Pacsu, E., I n d . E n o . Chem., 35, 381 (1943).
D iethylene gly col b u tyl ether acetate 70 8 .2 (9) Neale, S. M ., J. Textile Inst., 15, 443T (1924).
D ib u tyl phthalate 80 7 .9 (10) Sclioch, T . J., J. Am. Chem. Soc.,'64, 2954 (1942).
D ib u tyl sebacate 90 6 .5
D ib u tyl tartrate 78 7 .0 (11) Schoch, T . J., Ibid., 64, 2957 (1942).
D im ethyl phthalate 66 7 .6 (12) Sookne, A. M ., and Harris, M ., J. Research Nall. Bur. Stand
D im ethyl linoleate dim er 54 7 .3 ards, 30, 1 (1943).
Levoglucoaan triacetate 14 5 .3
Pentaerythritol tetraacetate 46 6 .0 (13) Staudinger, H ., Naturwissenschaften, 25, 673 (1937).
vSorbitol hexaacetate 12 5 .6 (14) Whistler, R . L., Jeanes, A., and Hilbert, G . E., J. Am . Chem.
T rib u tyl citrate 56 5 .8 Soc., to bo published.
T rib u tyl phosphate 90 7 .0
Tricresyl p hosphate 44 6 .4 P r e se n t e d before the D ivision of Sugar C hem istry and T e ch n o lo g y at th*
105th M eeting of the A m e m c a x C h e m i c a l S o c i e t t in D etroit, M ich .
Com parison of the U se of
A lc o h o l-E x tra c te d with
Petroleum-Ether-Extracted
Flakes in a Pilot Plant
The pllot-pl«nt production of soybean protein is

reported for the first time. It is shown that a proc
ess consisting of extraction of soybean flakes,
separation of suspended solids from the dispersion,
precipitation, and dewatering m ay be pursued
in a straightforward manner with satisfactory results.
The fractions into which the soybean flakes are
separated during the processing, as well as their
sizes, are given. A com parison is made of the
production of protein from flakes resulting from
the extraction of the oil with petroleum ether in
one case and with alcohol in the other. The
superiority of ethanol-extracted flakes for process
ing soybean protein is n o t e d . ~ ~ T h e photograph
shows discharge of protein from vacuum-drum filter.
P. /}. Better, Q. Becfzel, cuuL /¡. K.

N O R T H E R N R E G ÎO N A .L R E S E A R C H L A B O R A T O R Y , U. S. D E P A R T M E N T O F AG RIC U LT U R E, P E O R IA , ILL,
T H E pilot-piant production of soybean protein is relatively or dilute alkali as desired, are stirred together in jacketed tank 1
simple as is its laboratory preparation; but the steps for the time and at the temperature selected (5, 7). The disper
which lend simplicity to the laboratory process cannot be sion is separated from the coarse particles o f the flakes on gyrating
directly changed to large-scale production methods. The labora screen 2. The dispersion is pumped to storage tank 3, and the
tory process for preparing protein consists essentially of extraction residual flakes are pressed and dried. The dispersion from stor
of solvent-extracted soybean flakes with water or dilute sodium age tank 3 is usually passed through centrifuge 4 for clarification,
hydroxide, removal of the undispersed part by centrifugation, and then pumped to precipitation tank 5 where acid is added to
and precipitation of the protein with acid at a pH between .4 and bring the pH to the desired point for precipitation o f the protein.
5.- Detailed descriptions of this procedure and the properties The precipitated protein is allowed to settle to the bottom o f the
of the protein were described previously (<?, 7). In transferring tank, and the supernatant whey is siphoned off. T o wash the
to pilot-plant production, certain characteristics o f the materials protein, water is added and the mixture stirred and allowed to
become increasingly important. For example, the crude soy settle again after the pH is adjusted to the original precipitation
bean meal mixture contains a mucilaginous substance which defies value. The siphoning and washing m ay be repeated, or the thick
all attempts at filtration; the drying o f large quantities of protein slurry remaining after siphoning o f the wash water m ay be piped
emphasizes its heat sensitivity as evidenced by heat dénaturation to string filter 6. The resultant protein cake from the filter Ls
(1); and under some conditions, foaming is a serious problem. ready for drying.
Factors governing these and related problems are presented in
R A W M A T E R IA L S A N D E X T R A C T IO N
this paper.
This laboratory has been operating a pilot plant (Figure 1) The best raw material for the manufacture of soybean protein
for more than a year, and its description will serve as a guide to is probably the residue resulting from the solvent extraction of
large-scale production of soybean protein. The process consists flaked soybeans. These flakes are used in preference to expeller
of four steps: extraction o f protein, separation of dispersion meal because they have a lower oil content and contain a more
from suspended solids, precipitation, and dewatering. soluble and lighter-colored protein. The greater solubility and
Soybean flakes and the extracting medium, consisting of water lighter color are results of lower and more uniform temperature
799
extracting flakes before their use in making slurry. The meas

urement of the performance o f the screen is the quantity of sus
pended solids passing through. Of equal interest are the volume
and concentration o f the extract passing over the screen. A limit
ing factor in the rate of screening is the formation of foam, which
can be controlled by limiting the rate of feed o f the slurry or by
adding antifoaming agents. In the studies reported here, anti-
foaming agents were not used.
The variations in results attributable to a difference in the
mesh of the screen are given in Table I. The feed rates indicate
the maximum rate for continued operation in our apparatus
without foaming. As would be expected, the feed rate increases
with a decrease in mesh number, but the quantity of suspended
solids passing through the screen also increases. If there were no
Finished further clarification, the protein would be contaminated to the
Product
extent indicated in line 6 (percentage of suspended solids on the
protein). For certain purposes a protein of this quality might
Figure 1. Flow Diagram for Pilot-Plant Production of So y b e an
Protein be satisfactory. When further centrifugal clarification is used,
the increase in recoverable dispersion with decrease o f mesh
number (line 7) and the feed rates (lino 1) become appreciable
trcutnient in the solvent-extraction process. The flakes avail
factors, since the suspended solids indicated in line 3 are in no
able commercially have been extracted by petroleum ether. It
case excessive. The differences in suspended solids would affect
has been known for some time (2) that the conditions of oil
only the time during which a bowl-type centrifuge could be oper
removal have a considerable influence on the character of the
ated without shutting down for cleaning. For a given centrifuge
protein obtained. In addition, we have found that,the method
and a given dispersion, the several times of operation would be
of oil extraction has a profound effect on the properties of the
proportional to the weight values of line 3. These suspended
protein, greatly influences the ease and speed of the protein manu
solids, after recovery by centrifuging, may be combined with the
facture, and consequently determines its economics. During a
wet mash passing over the screen when the wet mesh is to be
study of the effects of the method of oil extraction of the- color,
dried and used as stock feed.
odor, and taste of soybean meal, ethanol was found far superior
to all other solvents studied (including the petroleum ethers)
when used on flakes having a comparatively high moisture con
tent. Furthermore, the production of protein from the ethanol- Table 1. Performance of G yra tin g Screen with Petroleum Ether
Flakes Extracted at p H 9.5
extracted flakes was considerably speeded. For this reason the
Screen Size in Standard M esh 40 80 150
pilot plant study was made on both the ethanol-extracted and the 1. F eed rate, gal. per hour 277 159 128
petroleum-ether-extractcd flakes. The protein content o f the
M a t e r ia l s P a s s in g
ethanol-extracted flakes is about 50% , and that of the petroleum- through Screen %° W O %a wt.fr %a Wt.fr
ether-extracted flakes about 4 5 % on the air-dry basis. Since 2. T otal solids 3 .7 4 0 .6 7 3 .6 8 0 .6 5 3 .4 5 0 .5 3
3. Suspended solids 0 .4 5 0 .0 8 9 0 .3 2 0 .0 5 6 0 .2 9 0 .049
soluble calcium and magnesium salts have a particularly bad 4. D issolved solids 3 .2 9 0 .5 9 3 .3 6 0 .5 9 3 .1 6 0 .5 4
effect (7) on the dispersibility of the protein, the other principal 5. Precipitable protein 1 .8 0 0 .3 2 2 1 .74 0 .3 0 6 1 .74 0 .316
6. Suspended solids on
raw material, water, should have as low a soluble salt content as protein 2 5 .0 ... 18.4 1 6.7
7. Dispersion recovered 6 6 .4 8 7 .5 8 6 .8
possible. 8. W hey solids 0 .2 6 0 .2 9 0 .2 2
Extraction conditions are described in other publications M a t e r ia l s P a s s in g O ver
(», 7). All extractions in this study were made at a water to S creen
flake ratio of 20 to 1 by weight and at a temperature of 30° C. with 9. W ater 89.4 8 9 .7 ... 8 9 .8
10. T otal dissolved pro-
an extraction time o f 30 minutes after reaching the specified te in c 11.6 1 2 ,5 ... 13.7
11. R ecoverable d ry re
temperature. The size of the batch of flakes varied from 60 to sidue 0 .3 3 ... 0 .3 5 0 .41
100 pounds. n Percentage of dispersion.
fr W eight per unit weight o f original flakes.
c T otal dissolved protein passing over screen, as percentage o f total dis
S E P A R A T IO N O F D ISP ERSIO N F R O M FLA K ES solved protein.
In view of the apparent impossibility of filtering the flakes

from the extraction medium by any method which would permit
the residue to be processed for animal feed, the only recourse The effect of hydrogen ion concentration on the separation of
seemed to be the use of some type of centrifuge or of a screening the mash from the dispersion is given in Table II. The extrac
operation. The moisture in the residual flakes from the screening tions at pH 6.6 were made with distilled water; those at 7.3 pH,
operations can be reduced slightly by pressing at a few hundred with 0.3% sodium sulfite; those at pH 9.8 and 9.5, with 0.1%
pounds per square inch, but a considerably higher pressure ap sodium hydroxide; and those at pH 10.4, with 0.2 % sodium
plied for a long time, has only a slight additional effect. In view hydroxide.
of this fact, and the large quantity o f material to be handled, it W ith increasing pH there is an increase in the weight of pre-
would seem that a screening operation followed by continuous cipitable protein per unit weight of flakes (line 11); there is also
pressing constitutes a more desirable procedure than the use of an increase in soluble solids not recoverable as protein (line 12
centrifugal equipment. and suspended solids passing through the screen (line 9). There
In the pilot plant a gyrating screen, 2 X 4 feet, tilted at an is a corresponding decrease in the solids passing over the screen
angle of 6 °, is used to remove coarse solids from the protein ex (line 7). W ith increasing pH o f extraction (line 5), the moisture
tract. Tw o baffles, each 2.5 inches wide, cross the screen at in the mash increases slightly and the removal of liquid by press
about one fourth and one half the distance from the upper end ing becomes more difficult. When the mash is to be used as
and serve to spread the slurry over the entire screen. The stock feed, it would seem advisable to neutralize the alkalinity;
factors affecting the performance o f the screen are: pH of slurry, this would reduce the difficulty of pressing at the same time.
size of screen openings, rate of feeding slurry, and means of Under a gentle pressure of 2 pounds per square inch for 5 minutes,
a water-extracted mash having an original

water content of about 88% was reduced to Table II. Effect of H y d ro g e n -Io n Concentration on W e t Screening of S o y b e a n Flake
81.3%; in another determination, a pressure of (Screen size, 80 m esh; water-fiake ratio, 2 0 -1 ; tem perature, 30° C .; tim e o f extraction, 30
minutes after reaching temperature)
224 pounds per square inch for 10 minutes re
duced the water content to 7 0 % ; a pressure of Flakes E xtracted w ith: A lc .° P.E.& Ale. P .E . Ale. P .E . Ale. P .E .
Feed rate (g a l./h r .)c 720 80 425 225 450 160 500 130
336 pounds per square inch for 90 minutes re
B e fo r e C la r ific a t io n
duced the water content to about 60% . The 1. p H of protein
extraction 6 .6 6 .6 7 .3 7 .3 9 .8 9 .5 1 0.4 10 »4
protein content of the mash on the dry basis is 2. % of dispersion
given in line 15. as total solids 2 .7 9 3 .2 1 2 .6 1 3 .6 3 3 .5 2 3 .6 8 3 .9 1 4 .1 6
3. % of dispersion
C L A R IF IC A T IO N as suspended
solids 0 .2 6 0 .2 4 0 .3 0 0 .2 9 0 .3 1 0 .2 9 0 .4 9 0 .3 6
The dispersion is clarified at 7600 r.p.m. by a 4. % of dispersion
as dissolved
disk centrifuge of the valve-bowl-concentrator 2 .5 3
solids 2 .9 7 2 .3 1 3 .3 4 3 .2 1 3 .3 6 3 .4 2 3 .8 0
type, with automatic control of the solids dis 6. % of m oisture in
mash
8 8 .8 8 8 .6 8 8 .3 8 9 .4 9 0 .0 8 9 .7 9 0 .8 9 0 .4
charge. The sludge discharge is continually 6. % suspended
solids on p ro
recirculated; this operation reduces the dis 1 6.5
tein 14.6 1 9 .0 1 6.5 1 7.5 1 6.7 1 9.4
persion loss to a small quantity. W e ig h t p e r U n it W e ig h t o f S t a r t in g F la k e s
An idea of the contamination of the protein 7. O f solids over
screen 0 .5 6 0 .4 5 0 .5 8 0 .3 6 0 .4 0 0 .3 5 0 .3 1 0 .2 2
due to unremoved suspended solids may be 8. Of total solids
through screen 0 .4 4 9 0 .5 4 7 0 .4 1 8 0 .6 2 4 0 .5 9 9 0 .6 4 8 0 .6 8 9 0 .7 7
obtained b y comparing lines 6 and 14 of 9. Of suspended
Table II. The former gives the contamination solids through
screen 0.0 4 2 0 .0 4 1 0 .0 4 8 0 .0 5 0 0 .0 5 3 0 .0 5 6 0 .0 5 9 0 .0 5 0
when the protein is separated without clarifica 10. Of dissolved
solids through
tion in the centrifuge, and the latter, after clari screen 0 .4 0 3 0 .5 0 6 0 .3 7 0 0 .5 7 5 0 .5 4 6 0 .5 9 1 0 .6 0 3 0 .7 1 0
fication. A t the lower hydrogen-ion concentra 11. Of precipitable
protein through
tions the clarification is satisfactory; at the screen 0 .2 5 2 0 .2 8 0 0 .2 5 3 0 .3 0 1 0.3 0 1 0 .3 0 6 0 .3 1 8 0 .3 4 8
12. Of soluble solids
higher pH values it is not. The clarification (n ot recovered
indicated by comparing lines 3 and 13 is excel as protein) 0 .1 5 6 0 .2 2 6 0 .1 1 7 0 .2 7 4 0 .2 4 8 0 .5 2 8 0 .2 8 5 0 .2 8 1
lent in all cases; only because the ratio o f liquid A f t e r C l a r i f i c a t i o n ( A b o u t 300 G a l . p e r H o u r )

13. % of dispersion
to protein is large does the small quantity of as suspended
solids 0 .0 0 1 0 .0 0 1 0 .0 0 1 0 .0 0 1 0 .0 1 0 .0 0 4 0 .0 8 0 .1 5
suspended matter rise to 8 % when calculated 14. % suspended
as an impurity based on the quantity of protein solids on pro
tein 0 .0 6 0 .0 6 0 .0 6 0 .0 6 0 .5 6 0 .2 3 2 .2 8 8 .0 8
(last figure in line 14).
M ash o v e r S cre e n
The clarification for each extraction was 15. % protein (dry
basis) 5 1 .8 4 3 .8 4 9 .8 3 6 .3 4 4 .4 3 4 .4 4 4 .7 3 2 .7
studied at three feed rates— about 200, 300, and
500 gallons per hour. The removal of suspended a E thanol.
b Petroleum ether.
solids was slightly less efficient at the higher c F or rates w ithout foam ing difficulties.
rates, in which cases the limiting factor was
not the clarification but the formation of
foam in the centrifuge. A satisfactory rate for
our equipment was between 275 and 300 gallons per hour, and ethanol-extracted flakes (curve 1) with those o f proteins ex
the data in fines 13 and 14 o f Table II were obtained at this feed tracted at pH 6.6, 7.4, 10.6, and 11.4 (curves 2, 3, 4, 5) from petro-
rate. leum-cther-extracted flakes, and with that of a protein extracted
at pH 11.4 from ethanol-extracted flakes. The quantity of sol-
P R E C I P I T A T I O N , S E T T L IN G , A N D W A S H IN G
The precipitation of the protein is usually performed by ad

justing the pH to about 4.5. Since the precipitated protein
probably consists of several individual proteins having slightly
different isoelectric points, it is probable that a change in either
direction in the pH of precipitation would cause a change in the
character of the precipitated protein.
The reasons for adopting the elementary process o f settling
are the necessity for removing an appreciable amount of soluble
material from the dilute suspension, and the difficulty of filtering
the precipitated protein.
The percentage of precipitable protein in the dispersions varied
from 1.57 to 1.86%, and the percentage of other soluble solids
(2) varied from 1.0 to 2.0% . The procedure o f settling, de
canting the whey, and washing by decantation is a necessary
preliminary to filtration, since the low concentration o f protein
in the suspension is not readily filtered, and the nature of the cake
is such as to make washing on the filter difficult or impracticable.
The time required for settling and the settling capacity govern
the size of the other units, which can operate more or less con
tinuously.
When the percentage of solids in the settled curd (uniformly
mixed) is plotted against the time o f settling, the slope o f the Figure 2. Settling Rate of Precipitated Protein
curve gives the rate o f settling. These curves (Figure 2) con Curves 1 and 6 from ethanol-extracted flakes at p H 6 .6 and 1 1.4, respec
tively; curves 2, 3, 4 , and 5 from petroleum-ether-extracted flakes at p H
trast the behavior of a water-extracted protein (pH 6.6) from 6 .6, 7.4, 1 0.6, and 1 1.4, respectively.
uble solids, other than protein, increases with increasing pH dition for drying, whereas the scraper causes the curd to bunch up
(line 12, Tabic II), and so may change the viscosity and density into masses which caseharden in the drying operation.
o f the medium through which the protein settles. Since a con Significant data on the filter used for these experiments are:
siderable portion o f the readily soluble carbohydrates have been Diameter of drum, 3 feet; width of face of filtering surface, I
removed from flakes from which oil was extracted with ethanol foot; filtering area, 9.43 square feet; drum speed, 0.375 r.p.m.;
there is an appreciable difference in the quantities of soluble proportion o f drum immersed in the slurry, 1 /4 ; string separa
solids extracted at the lower hydrogcn-ion concentrations from tion, 1 /4 inch; and cake thickness at maximum filtration rate,
the two kinds of flakes. The influence of increasing pH is greater about 5/16 inch.
on the ethanol-extracted than on the petroloum-ether-extracted The rate o f filtration, given as pounds of protein per square
flakes; this effect is shown in line 12 of Table II and in Figure 2, foot of filtering surface per hour for a range of slurry concentra
where curve 6 for protein extracted at pH 11.4 from ethanol- tions in percentage of solids, is presented in Figure 3. Curve 1
extracted flakes is the lowest of all. shows the variation in rate of filtration with slurry concentration
for a protein from ethanol-extracted flakes, the protein being
in the wet condition at the time of discharge. Curve 2 repre
sents the same yield translated to the dry condition. Curves
3 and 4 give the corresponding data on protein from flakes from
which the oil had been extracted wdth petroleum ether. The
difference in the rate of filtration, as well as the lower slurry con
centration at which filtration is possible, is probably a reflection
of the more granular structure of the protein from alcohol-
extracted flakes.
The w’ater content of the protein curd discharged by the filter
varies from 60 to 80% , depending on the nature of the protein,
the thickness of the slurry, the depth of slurry in the filter sump,
and the rate of rotation of the drum. Under normal con
ditions the w'ater content will be about 7 0 % of the curd.
The drying of the protein is carried out in a conventional tray
dryer with the temperature maintained at 120° F. The granular
type of precipitated protein dries rapidly and uniformly, es
pecially when in the condition resulting from the string dis
charge on the filter. The effect of heat and moisture on the
dénaturation of the protein in soybean meal has been studied
( /) , but the specific effects of various methods o f drying the
isolated protein are still under investigation.
D ISC U SSIO N
The described processing of solvent-extracted soybean flakes

for protein was on a scale large enough to indicate the behavior
to be expected in large-scale production. In the case of extrac
tions with water or at a low pH we would expect that (a) about
one fourth of the weight of flakes processed would appear as dry
protein; (6) about half of the flakes would reappear as dried, ex
Fi$ure 3. Filtration Rate of Protein tracted mash with a small additional amount (about one twen
Curves 1 and 2 on basis of w tt and dry protein from
ethanol-extracted flakes; curves 3 and 4 for pro tieth) from the centrifugable solids; and (c) about one sixth to
tein from petroleum-ether-extracted Rakes. one fifth of the starting material would exist in the whey which is
siphoned from the protein after the settling operation as soluble
The considerable difference in the rate of settling of protein carbohydrates, proteoses, albumins, and other soluble substances.
extracted at a low' pH from ethanol-extracted flakes and the rate The relatively large proportion o f the flakes that reappear as
of settling of other proteins, seems to result in part from a more dried, extracted masli could go back into stock feed channels at
granular structure which is known to increase the ease of filtra the price of the starting material and thus reduce by a cor
tion but has not been evaluated further. responding figure the cost of the original flakes. However, this
Curve 1 (Figure 2) show's that the protein from ethanol- mash has approximately the same protein content as the original
extracted flakes reaches a concentration o f 6 % solids in the flakes because, during the extraction of a portion o f the protein,
settled curd in 20 minutes; the nearest approaching curve, 2, a corresponding proportion of the other soluble solids was re
for protein from petroleum-ether-extracted flakes showB that the moved. The removal o f a large proportion of the water-soluble
curd attains this concentration o f solids in 2 hours. In the constituents from the mash yields a product that is useful in
former case the concentration of 6 % solids is reached six times as plastics and adhesives; this type of material has been used in the
fast as in the latter, and above this concentration the factor in soybean plastics developed by this laboratory (5, 4). In any
creases greatly. Furthermore, the protein from ethanol-ex case, the higher protein content should command a premium over
tracted flakes compacts to a concentration much greater than the original price. The whey solids have not been investigated
that ever attained by protein from petroleum-ether-extracted to any extent.
flakes, and this fact translates into economies during filtration. The uses for which the protein is intended and the trade chan
nels through which it moves determine, to a considerable extent,
F IL T R A T IO N A N D D R Y IN G
the details of the process used in its manufacture. For example,
T he protein is filtered by a vacuum drum filter with a string the protein mentioned earlier, which could be prepared by elimi
discharge (shown on page 799). The filter may be operated either nating the clarification operation, can be prepared cheaply and
as a string discharge or as a scraper-type filter. There is little can be used in the adhesive trade, as well as in certain paper
difference in the operational effectiveness of the two, except that coatings. For a superior product the clarification operation is
the string discharge presents a thin layer of curd in excellent con essential, and for the best color, as far as we know now, the
protein must be prepared from flakes from which the oil has L IT E R A T U R E C IT E D
been extracted with alcohol. The color of products prepared (1) Bockel, A . C., Bull, W . C., and Hopper, T . H ., I n d . E ng . C h em .
from this latter type of protein is, for all practical purposes, 34, 973 (1942).
equivalent to those prepared from casein, and the yellowish tinge (2) Boyer, R . A „Ib id ., 32, 1549 (1940).
(3) Brother, G . H ., and M cK inney, L. L., Ibid., 32, 1002 (1940).
is no longer a problem. (4) M cK inney, L. L., Deanin, R ., B abcock G ., and Smith, A . K .,
Any cost estimate would be premature until certain other Ibid., 35, 905 (1943).
factors are known. Among these factors are yield of protein (5) Nagle, R . H., Becker, H. C., and Milner, R . T ., Cereal Chem., 15,
under optimum conditions, effects of water-flake ratio and tem 403(1938).
(6) Smith, A. IC , and Circle, S. J ., I n d . E n g . C h e m ., 30, 1414
perature on the processing of the protein, and optimum condi (1938).
tions for drying both the protein and the mash. These problems (7) Smith, A. IC, Circlo, S. J., and Brother, G . H., J. Am. Chem.
are now being studied in this laboratory. Soc., 60, 1310 (1938).
Production of 2,3-Butylene Glycol from

DAVID PERLMAN
College o f Agriculture,
University o f Wisconsin, M adison, Wis.
WOOD HYDROLYZATES
P roduction o f 2,3-butylene glycol fro m w ood hydrolyzates Erlenmeyer flasks containing 100 cc. of medium. The media
by A e r o b a c te r a e ro g en e s has been investigated. W ood were prepared, divided among the flasks, and autoclaved for 30
sugar solutions in con cen tration u p to abou t 17% have minutes at 120° C. After inoculation and addition of nutrients,
been com pletely ferm ented. A b ou t 35% o f the ferm ented the flasks were placed in a mechanical shaker and shaken con
sugar is converted in to 2,3-butylene glycol, and a sm all tinuously for the duration of fermentation. This shaker, placed
a m ou n t o f acetoin is also form ed. A cclim atization o f the in a 30° C. incubator, had a stroke of 3 inches and a frequency
A e r o b a c te r a e ro g en e s cu ltu re to the hyd rolyzatc simplifies
of 96 cycles per minute. The results obtained with shaken flask
the treatm en t o f hydrolyzatcs for ferm entation purposes. fermentations on a glucose-salts medium have been found to be
comparable with those obtained in 12-liter aerated fermentations
and in pilot-plant operations.
T H E conversion of wood waste into sugar by the Scholler Urea, potassium phosphate, and calcium carbonate (as a slurry)
(3) and similar processes has been extensively investigated
were sterilized separately from the rest of the medium, and were
in recent years. In these processes dilute sulfuric acid added to the fermentation medium at the time of inoculation.
(0.2- 1.0 % ) is percolated through wood chips at elevated tem The levels of these compounds used in the following experiments
peratures (e.g., 160° C.). From a fermentation standpoint these were probably greatly in excess of those required. Preliminary
hydrolyzates contain appreciable amounts of hexoses, pentoses, experiments showed that 2 grams per liter of urea, 1 gram of po
and acetic acid. The hard-wood hydrolyzates are higher in tassium monobasic phosphate, and 3 grams of calcium carbonate
pentose concentrations than those from the softwoods. are usually sufficient.
These sugar solutions have been used in Germany for the pro When received, the liquor contained 3.0 to 6 .0 % reducing
duction of ethyl alcohol and of yeast. Sjolander et al. (5) re material as glucose and 0.1 to 0.3 % organic acids (as acetic), and
ported on the fermentation of hydrolyzates from several kinds of had a titratable acidity of 0.2 to 0.25 N (a corresponding pH of
wood by the butyl anaerobes, Clostridium felsinium and Clostrid- about 1.5). W ood sugar concentration was determined by an
ium bulylicum. alyses for reducing sugar (calculated as glucose) according to
Since 2,3-butylene glycol is formed by Aerobacter aerogenes the method of Shaffer and Som ogyi (4). The samples were
from sugar solutions and hydrolyzed grain mashes, its production heated with the sugar reagent for 30 minutes in order to oxidize
from wood hydrolyzates seemed possible. This organism will a larger portion of the pentoses present. The sugar determina
ferment pentoses as well as hexoses, both of which are found in tions were corrected for the presence of acetoin.
wood hydrolyzates. Hydrolyzates from both hard- and soft 2,3-Butylene glycol was determined on the ether extract of the
woods were tested. Th ey included white spruce, southern yellow neutralized fermented medium according to the procedure out
pine, southern red oak, and Douglas fir. lined by Johnson ((), and also by the “ direct oxidation” pro
M ETH ODS
cedure mentioned b y him. The latter method was applied to a
butyl alcohol extract of the fermented medium. Before butyl
The Aerobacter aerogenes culture was obtained from the
alcohol extraction, the medium was saturated with potassium
Northern Regional Research Laboratory, and is designated in
carbonate. Acetoin was determined b y a modification of the
tbeir collection as A . aerogenes N o. 199. On a glucose-salts
method of Langlykke (£). When the 2,3-butylene glycol values
medium this culture produces mainly the meso form of 2,3-
by these two methods had been corrected for the acetoin present
butylene glycol (6 ); it is presumed that the same form is obtained
in the samples, they were found to agree satisfactorily. In no
on the wood hydrolyzate medium. Other products formed dur
ing the fermentation of glucose b y this organism include acetoin, case did the acetoin formed account for more than 2 % of the
ethyl alcohol, lactic acid, and succinic acid. reducing sugar fermented, and it was usually less than 1 % .
The inoculum for the fermentations was grown (unless other The yield of 2,3-butylene glycol in the tables is the weight
wise indicated) on a medium consisting of 0 .5 % corn-steep yield— i.e., grams of 2,3-butylene glycol plus grams of acetoin
liquor, 0 .5 % calcium carbonate, and 5 % glucose. The inoculum found in the sample, divided by the grams of reducing sugar
was aerated continuously on a mechanical shaker for 12 to 16 present in the sample before fermentation. Both butylene
hours before use. The inoculum size varied between 3 and 5 % glycol and acetoin are included in the yield data; for if the
of the fermentation volume. fermentation is continued too long, a considerable portion of the
The fermentations were conducted in cotton-plugged 500-ml. butylene glycol will be converted to acetoin. This culture has
the follow ing tw o seem ed sim plest: T h e first was a neutraliza

T a b l e I. E f f e c t o f T r e a tm e n t“ o f H y d r o ly z a te s on tion procedu re in w hich the liqu or w as neutralized to p H 6 with
2 ,3 -B d t y le n e G l y c o l a n d A c e t o i n P r o d u c tio n
lim e slurry or alkali. W hen the lim e slurry was used, the pre
G ly c o l an d A c e t o in P r o d u c tio n
cip ita ted calcium sulfate was rem oved b y filtration before
F crm en- Y ield of
Original tation Sugar F er- 2,3-Butylene Supplem ent, sterilization o f the m edium and subsequent ferm entation. The
Sugar, T im e, m cnted, G lycol and 10 G ram s/ second m eth od m a y be called an alkaline procedu re in which
W ood G ./1 0 0 C c. H r. % A cetoin, % Liter
the h ydroly za te was adjusted to pH 10 or 11 w ith alkali or lime
Neutralization Treatm en t6
slurry and filtered. T h is alkaline solution w as then adjusted
Southern
yellow pine 4 .3 18 9 1 .0 3 9 .8 N one to p H 6 w ith sulfuric acid and again filtered. O ther m ethods
Southern red 1 0.0 48 5 5 .0 1 6.5 N one w hich were foun d to be less satisfactory included clarification
oak 1 0.0 48 9 1 .0 3 3 .0 D ifco yeast ext.
1 0 .0 • 48 9 0 .0 3 5 .2 M a lt sprouts
o f the liqu or w ith charcoal a t various pH levels, and treatment
Douglas fir 5 .0 48 9 2 .5 2 8 .5 N one
w ith b oth a cidic and basic exchange m aterials.
7 .5 48 7 7 .8 2 6 .0 N one A s T a b le I show s, h ydrolyzates from variou s w ood s v a ry in
W hite spruce 5 .9 48 1 0 .0 3 .3 N one ease o f ferm en tability b y this organism . In general it m a y be
5 .9 30 8 9 .5 3 2 .8 M alt sprouts
said th at dilute sugar con centration s o f m ost o f the hydrolyzates
Alkaline T rea tm en t0
m a y be ferm ented easily after treatm ent b y the neutralization
Southern 4 .3 18 9 0 .1 3 8 .8 N one
yellow pine 8 .5 38 8 9 .6 3 7 .2 N one
procedure. H ow ever, recovery o f 2 ,3-bu tylen e g ly c o l from
Southern red 1 0 .0 36 9 0 .5 2 6 .6 N one dilute solutions is im practical, and ferm entation o f m ore con
oak 1 0.0 36 9 4 .0 3 0 .0 D ifco yeast ext. centrated h ydrolyzates was attem p ted . T h ese solutions were
1 0.0 36 8 7 .0 2 4 .5 M a lt sprouts
prepared from the dilute h ydrolyzates b y con cen tration a t pH
W hite spruce 5 .9 48 8 7 .6 3 1 .7 N one
1 7.8 75 8 8 .0 3 7 .7 N one 3 in a va cu u m still. A s the sugar con centration was increased
1 7.8 75 8 9 .3 3 6 .9 D ifco yeast ext.
1 7.8 75 8 7 .6 3 1 .7 M alt sprouts ferm en tation o f several o f the h ydrolyzates treated b y the neu
tralization m eth od becam e m ore difficult. T h ese difficulties
0 Nutriente added to m édium per liter: 5 grama CaCOs, 4 grama urea, 2
grama KH tPO* were ov ercom e b y the add ition o f such substances as malt
6 H yd roly zate adju8ted t o p H 6 with lime, filtered to rem ove precipitated
calcium sulfate. sprouts or yeast extract. H ow ever, the addition o f these sub
c H yd rolyzate adjusted to p H 10.5 with lim e and filtered;; adjusted to pH stances was n ot necessary w hen the solutions had been treated
6 and filtered again.
b y the second m eth od o f treatm ent. P relim inary experim ents
w ith D ou g la s fir h yd roly za te (a t a 1 0 % reducing sugar level)
in dicated th at the m edium should be adjusted to betw een pH
been found to utilize acetic acid and it is probable that a small 9 and 10.5 w ith alkali or lim e for best results w hen the alkaline
part of the butylene glycol formed from the wood hydrolyzate m eth od o f treatm ent was used. In all these experim ents the
fermentation comes from this source. However, since the in oculu m was grow n on the glu cose-steep w ater m edium .
organic acid concentration is so small compared to the reducing R attier con centrated w ood sugar solutions were ferm ented
sugar concentration in the medium, the most satisfactory method after treatm ent b y the alkaline procedure. T a b le I I gives
for expressing yield is to base it on the reducing sugar availablo data ob ta in ed from the ferm entation o f southern yellow pine
in the medium. h ydrolyzate a t variou s sugar concentrations. Sim ilar results
were ob ta in ed w ith other h ydrolyzates, shown in pa rt b y T a b le I.
TREATM ENT OF H YD RO LYZATE
Before the wood sugar could be fermented, a preliminary treat A C C L IM A T IZ A T IO N O F C U L T U R E T O H Y D R O L Y Z A T E
ment of the hydrolyzates was necessary. Since the fermentation Experim ents on the ferm en tation o f h yd roly zed grain mashes
proceeds best at pH 6, it was necessary to adjust all media to this b y this culture in dicated th at if the culture w ere grow n for a few
reaction. Several methods of further treatment were tried, and transfers on th e h ydrolyzed grain m ash, im proved ferm entations
resulted. T h is procedu re was repeated w ith several w ood
h ydrolyzates w ith favorable results. T h e advantages are shown
T a b l e II. F e r m e n t a t i o n o f S o u t h e r n Y e l l o w P in e in T a b le I I I , where southern red oak was used as substrate.
H y d r o ly z a t e “ a t V a r io u s S u g a r C o n c e n tr a tio n s B y this procedu re the alkaline m eth od o f treating the h ydrolyzate,
Initial T im e of Yields of w hich had been necessary when rather con centrated sugar solu
R educing Sugar Ferm en 2,3-B utylene tions were ferm en ted w ith ou t the add ition o f auxiliary sub
Sugar, Ferm ented, tation, G lycol and
G ./1 0 0 C c. % H r. A cetoin , % stances, was elim inated and the neutralization m ethod was
4 .0 9 2 .3 15 3 4 .8 satisfactory.
6.0 9 1 .5 24 3 3 .3
8.0 9 2 .0 30 3 5.1 ACKNOW LEDGM ENT
10.0 9 0 .6 48 3 2.1
12. 0 8 9 .8 48 3 3 .8
T h e a d vice and criticism received from R . H . Leon ard are
° H yd roly zate treated as follow s: adjusted to p H 10.5 with lim e, filtered; gratefu lly ackn ow ledged. T h e w ood h ydroly za tes used in this
adjusted to p H 6.0 with HaSOi, filtered; diluted to appropriate sugar con
centration. Nutrients added per liter: 3 grams K H jPO «, 3 grams urea, 5 in vestigation w ere obtain ed from the F orest P rod u cts L aboratory,
grams C aC O j.
M adison, W is. T h e work was sup ported in pa rt b y a grant from
the H eyd en Chem ical C orp oration .
T a b l e II I . A c c l i m a t i z a t i o n o f C u l t u r e t o S o u t h e r n R e d
O ak H y d r o ly z a te by R e p e a te d T r a n s fe r L IT E R A T U R E C IT E D
Yields of (1) Johnson, M . J., to be published.

Transfer0 F erm entation6 . Sugar 2,3-B u tylenc G lycol (2) Langlykke, A. F ., and Peterson, W . H., In d . E n o . C h e m ., A n a l .
N um ber T im e, H r. &'ermentcd and A cetoin , % E d ., 9, 163 (1937).
1 36 27 5 .7 (3) Scholler, H „ Chem.-Ztg., 60, 293 (1936).
8 36 76 2 5 .9
12 36
(4) Shaffer, P . A ., and Somogyi, M ., J. Biol. Chem., 100, 695 (1933).
90 3 2 .5
16 34 93 3 5 .5 (5) Sjolander, N. O., Langlykke, A. F., and Peterson, W . H .t In d .
E n g . C h e m ., 30, 1251 (1938).
“ Transfer m edium : 4 % southern oak h y d rolyza te (neutralized to p H 6 (6) W ard, G . E., Pettijohn, O. G ., Lockwood, L. B., and Coghill.
and filtered), 0 .2 % K H iP O «, 0 .4 % urea, 0 .5 % C aC O ,.
ft Ferm entation m edium : concentrated southern red oak hydrolyzate R . D ., J . Am . Chem. Soc., 66, 541 (1944).
neutralized to p l l 6 with lim e and filtered: diluted to 1 0 % reducing sugar;
0 .3 % K H ,P O i, 0 .3 % urea, and 0 .5 % C aC O j added.
P u b l i s h e d with the ap p rova l o f the D irector o f the W isconsin Agricultural
Experim ent Station.
VAPOR-LIQUID EQUILIBRIA OF
Close-1' ^
' 1 Mixtures
A. R. Richards and E. Hargreaves
TRINIDAD LEASEHOLDS LIMITED, TRINIDAD. S.W.I.
V a p o r - liq u id e q u i li b r i u m d a t a f o r
b in a r y s y s t e m s o f h y d r o c a r b o n s lowed in the apparatus devised by Rosanoff, Bacon, and White
b o ilin g in t h e n e ig h b o r h o o d of
(24). The results were not self-consistent. Experiments were
80° C . a re p re se n te d . A m i n i m u m
c o n s t a n t - b o i li n g m ix t u r e is f o r m e d therefore undertaken in a modification of the apparatus described
by th e syste m b e n z e n e -2 ,4 -d i- by Othmer (22) which ensured better circulation of the liquid in
m e t h y lp e n t a n e , g iv in g a n a z e o tro p e the reboiler. The apparatus in its final form approximated
b o ilin g a t 75.2° C . a n d 757 m m . a n d
closely the design subsequently reported by Othmer in a more
c o n t a in i n g 54.5 m o le % benzene,
a n d b y t h e s y s t e m b e n z e n e -c y c lo - recent communication (23).
h exane, g iv in g a n a z e o tro p e b o ilin g Successive additions of 10-30 ml. of the hydrocarbon were made
a t 77.4° C . a n d 759 m m . a n d to an initial charge o f 200 ml. of benzene in the Othmer apparatus,
c o n t a in i n g 51.5 m o le % be n zen e.
and samples of liquid and condensate were drawn 6 hours after
A n a ly t i c a l d if f ic u lt ie s p re v e n t p re
cise e v a lu a t io n o f d a ta f o r s y s t e m each addition. A series of mixtures of the hydrocarbons was
c y c lo h e x a n e - m e th y lc y c lo h e x a n e . made up b y weighing, and the molal compositions were calcu
lated. After determination of the refractive indices of these
samples on an Abbe refractometer, curves were prepared which
enabled samples obtained from the Othmer apparatus to be
analyzed by a single refractive index determination (Table II).
T H E separation of pure hydrocarbons from petroleum frac
tions by distillation usually requires a high degree of frac
tionation and very efficient columns. However, even under
the best conditions certain separations are found to be impossible;
the separation o f benzene from the low-octane cut boiling in the
range 65-70 ° C. during precise fractionation might be cited as an
example.
Several binary hydrocarbon systems related to this problem
have been investigated, and vapor-liquid equilibrium data have
been published on binary hydrocarbon systems of the following
types: paraffin-paraffin (16), paraffin-aromatic (35), aromatic-
aromatic (15), aromatic-naphthenie (17), and paraffin-naphthenic
(4). In many cases the existence of azeotropes was reported, and
in view of the ever increasing importance of supplies of aromatic
blending agents for high-octane and high-performance aviation
fuel, more information was required. The data are extended in
this communication to the behavior of petroleum hydrocarbons
boiling between 80° and 81° C.
Baker’s Analyzed benzene was refractionated on a still em
ploying 4 feet of 25-mm. Stedman packing (3) at a reflux ratio of
34 to 1. This column was equivalent to about fifty theoretical
plates on total reflux. Eighty-four per cent by volume of the
charge distilling at constant temperature and refractive index was
dried over anhydrous sodium sulfate. The 2,4-dimethylpentane
was obtained during the fractionation of a cold acid alkylate, and 0 0-2 0-4 0-6 08 1-0
the fraction used in this work resulted from five complete series M O L F R A C T IO N B E N Z E N E IN L IQ U ID
of fractionations on stills employing 4 feet of Vs-inch Stedman F ig u r e 1. A c t i v i t y C o e f f ic ie n t s f o r B in a r y

packing at a reflux ratio of 119 to 1. These columns were equiva H y d r o c a r b o n M ix t u r e s
lent to about eighty plates on total reflux. Cyclohexane was pre
pared by the catalytic hydrogenation of a sample of the dry The temperature in the apparatus was determined during the
benzene; methylcyclohexane was prepared in like manner from runs b y a totally immersed Anschutz thermometer previously
toluene. The hydrogenated products were fractionated in stills calibrated to =*=0.05° C. against the platinum resistance ther
employing 4 feet of 3/ s-inch Stedman packing at a reflux ratio of mometer which had been used for the determination of tempera
119 to 1. Comparison of the physical constants of the fractions tures recorded in Table I. Accordingly the temperatures in
used in this work with the literature values is given in Table I. Tables III, IV, and V for the boiling points o f the binary mixtures
Any impurities in the materials used are probably paraffinic, are probably correct to =*=0.1° C.
but in view of the method of preparation it is considered unlikely The data obtained by analysis of samples from the Othmer
that the effect of these impurities on the vapor-liquid equilibrium apparatus are given in Tables III, IV, and V, together with other
data could be detected. data necessary to calculate the activity coefficients as suggested
The method outlined by Cornell and M ontonna (8) was fol- by Carlson and Colburn (7). Conform ity of experimental data
805
y/O 0 /
0/
uj 06
r o
O
_i
o
>-
004
0/ /
0 0-2 0 -4 06 0 -6 1-0
MOL F R A C T IO N C Y C L O H E X A N E IN LIQUID
F ig u r e 3. E q u i li b r i u m D a t a f o r C y c lo h e x a n e -
M e t h y lc y c lo h e x a n e M ix t u r e s a t 760 M m .
Tho curve represents R aoult's law and was calculated
from the following equation, assum ing a = 1.796:
boiling point, a similar equation was used for calculating the

partial pressures of cyclohexane and 2,4-dimethylpentane. In
effect, this method assumes that the pressure-temperature curves
for the three compounds were parallel over the small temperature
ranges under consideration, and the probable errors could not
make any significant change in the conformity or nonconformity
of the activity-coefficient composition relation with a smooth
curve. The relation between these coefficients and the composi
tion of the mixtures is shown in Figure 1; it is evident that the
technique adopted gave satisfactoiy results. The curves in
Figure 2 were drawn from the equilibrium data in Tables I I I and
IV.
T able I. P r o p e r t ie s of H ydrocarbons
Freezing B oiling P oint R efra ctive Index

M aterial Point, ° C. (760 M m .), “ C.
C yclohexane 6.4, (S4) 8 0 .8 (33)
0 .4 (38) 8 0 .5 (2) 1.4204 (11)
0.5 (2 6) 8 0 .8 ( 5 )
0 .2 5 “ (5)
6. 271" 80.'60 4 1 .4 2 1 1 b 1 .4 2 5 8 '
0'6 0 -Ô 1-0 8 0 .0 9 8 (29)
Benzene 8 0 .0 9 4 (37) 1.5012(73)
M O L F R A C T IO N BENZENE IN LIQUID
8 0 .1 2 2 (39)
8 0 .1 2 (28)
F ig u r e 2. E q u i li b r i u m D a t a f o r M ix t u r e s of 80.0 0b 1.49484 l.S Ô il'
B e n z e n e -2 , 4 -D im e t h y lp e n t a n e at 757 Mm. 2 ,4-D im eth yl- 8 0 .7 0 (12) 1 .3920 (73)
pentane 8 0 .0 0 (SO)
(above) a n d o f B e n z e n c - C y c lo h e x a n e (below ) 8 0 .6 0 (31)
-1 2 0 .4 4 l ! 37704 R 3820'
M e th y lçyclo - - 1 2 0 . 4 (IS) 1 01 .2 (S) 1 .4 2 3 0 (77)
hexane 1 00 .8 (34)
with equations of van Laar (SO), Margules (SI), or Scatchard (26) - 1 2 7 .0 1 4 1 0 0 .8 0 b l.'4 Î8 2 4 1 .4 2 3 3 '
is not a necessary criterion of experimental accuracy, particularly * Valnee for r f calculated from values of n o and dn/dt givon in refer-
ences cited.
as it is necessary to choose the correct type of equation to fit the “ From p etroleum . 4 This research. ' C alculated.
data. The data on benzene-2,4-dimethylpentane would, for
instance, be best represented by an equation of the van Laar T able II. R e f r a c t iv e I n d ic e s of H ydrocarbon M ix t u r e s
form; the data on benzene-cyclohexane would require an equa Benaonc—2,4- Benzene— C yclohexane—
tion of the more unusual Scatchard or Margules type. However, D im ethylpentane C yclohexane M e th y lcycloh cxa n e
M ole fra c M o le frac M ole fraction
the continuity of any function such as the activity coefficient tion benzene n 30 tion benzene cyclohexane
D n D* n p°
which depends on temperature and composition, is a valuable 0 .0 0 0 1.3770 0 .0 0 0 1.4212 0 .0 0 0 1.4 1 85
check on the self-consistency of the results. T o obtain these 0 .0 9 5 1 .3832 0 .1 5 4 1.4290 0 .1 4 0 1.4 1 88
0 .2 1 0 1.3914 0 .3 1 0 1.4387 0 .4 5 0 1.4195
coefficients, the formula 0 .2 9 5 1.3 9 86 0 .5 1 3 1 .4542 0 .6 7 5 1.4201
0 .4 1 8 1.4101 0 .7 1 1 1.4693 0 .8 0 6 1.4205
0 .5 2 8 1.4211 0 .8 0 8 1 .4780 1.000 1.4211
I = 80.09 + 0.042683 (p - 760) - 0.00002199 (p - 760)s 0 .6 2 7 1.4330 0 .9 0 4 1 .4860
0 .7 1 5 1.4447 1 .000 1 .4948
0 .7 9 0 1 .4570
given by Smith and Matheson (SO) was used for calculating the 0 .8 7 0 1.4 6 88
0 .9 3 8 1 .4818
partial pressure, p, of benzene at a temperature, I. After making 1.000 1.4944
a suitable correction in the first term for the slight difference in
o f the formation o f an azeotrope, and the devia

T a b le III. V a p o r -L iq u id E q u ilib r iu m D a t a f o r B b n z e n f-2 ,4 -D i- tion from the Raoult law curve is well within ex
METHYLPENTANE (24M5) M IX T U R E S AT 757 M il. perimental error.
Vannr M ole Fraction in : A ctivitv The values of a for each system have been
Pressure, M m . Liquid V apor C oefficient y 'Relative
V olatility
calculated from the equation
Temp., B en B en Ben Ben
0 C. zene 24M 5 zene 24M 5 zene 24M 5 zene 24M 5 « (Caled.)
77.3 6 96 .3 6 8 2 .8 0 .0 2 2 0 .0 7 8 0 .8 7 5 0 .1 2 5 1.032 1 .776 0 .5 9 1 y( 1 - x)
77.0 6 8 9 .8 6 76 .3 0 .8 9 4 0 .1 0 6 0 .8 3 6 0 .1 0 4 1.026 1 .732 0 .6 0 5 x (l - y)
76.5 679.0 6 6 5 .9 0 .8 6 0 0 .1 4 0 0 .8 0 5 0 .1 9 5 1.044 1 .583 0 .6 7 1
76.3 6 7 4 .8 6 6 1 .6 0 .8 3 8 0 .1 0 2 0 .7 8 1 0 .2 1 9 1 .046 1 .547 0 .6 8 9
76.1 6 7 0 .7 6 5 7 .7 0 .8 0 8 0 .1 0 2 0 .7 5 8 0 .2 4 2 1 .059 1.451 0 .7 4 5 where x, y = moles of one component in liquid and
75.9 6 6 0 .5 6 53 .1 0 .7 7 6 0 .2 2 4 0 .7 3 2 0 .2 6 8 1.071 1 .3 8 7 0 .7 8 8 vapor phases in equilibrium
75.7 6 6 2 .2 6 48 .9 0 .7 4 9 0 .2 5 1 0 .7 0 8 0 .2 9 2 1 .0 8 0 1 .357 0 .8 1 3
75.5 6 58 .3 6 44 .9 0 .7 1 9 0 .2 8 1 0 .6 8 2 0 .3 1 8 1.091 1 .328 0 .8 3 8
75.4 6 56 .1 6 4 2 .7 0 .6 6 5 0 .3 3 5 0 .6 4 3 0 .3 5 7 1 .1 1 0 1 .227 0 .9 0 7 and are represented graphically in Figure 4. The
75.4 6 56 .1 6 4 2 .7 0 .6 2 2 0 .3 7 8 0 .6 0 8 0 .3 9 2 1 .1 2 9 1 .2 2 2 0 .9 4 2
75.2 6 51 .9 6 37 .7 0 .5 6 8 0 .4 3 2 0 .5 6 8 0 .4 3 2 1 .1 6 2 1 .3 8 8 1.000 probable errors are indicated by dotted lines.
75.4 6 56 .1 6 4 2 .7 0 .5 2 0 0 .4 8 0 0 .5 3 1 0 .4 6 9 1 .1 7 8 1 .151 1 .0 4 5
75.3 6 54 .3 6 3 9 .7 0 .4 7 5 0 .5 2 5 0 .4 0 6 0 .5 0 4 1 .2 1 0 1.138 1 .087
The data on binary hydrocarbon mixtures con
75.5 6 5 8 .3 6 4 4 .9 0 .4 2 8 0 .5 7 2 0 .4 5 9 0 .5 4 1 1 .2 3 6 1.113 1 .134 sisting of an aromatic with a nonaromatic con
75.5 6 5 8 .3 6 4 4 .9 0 .3 8 4 0 .6 1 6 0 .4 2 4 0 .5 7 6 1 .2 6 8 1.096 1 .1 8 2
75.9 6 0 6 .5 6 52 .2 0 .3 3 8 0 .6 6 2 0 .3 8 6 0 .6 1 4 1 .2 9 8 1 .0 7 7 ' 1.231 stituent were discussed in a recent paper by-
76.1 6 7 0 .7 6 5 6 .7 0 .3 0 0 0 .7 0 0 0 .3 4 8 0 .6 5 2 1 .3 1 0 1 .0 7 4 1 .2 4 6 Griswold and Ludwig (17) where attention was
76.3 6 7 4 .8 0 00 .7 0 .2 0 0 0 .7 4 0 0 .3 1 4 0 .6 8 0 1 .350 1 .0 6 4 1.302
76.6 0 81 .1 6 67 .1 0 .2 1 5 0 .7 8 5 0 .2 6 0 0 .7 3 1 1 .3 8 8 1 .0 5 5 1.343 drawn to the fact that a constant-boiling mixture
77.2 0 9 3 .9 6 8 0 .0 0 .1 6 1 0 .8 3 9 0 .2 1 5 0 .7 8 5 1 .458 1 .042 1.427
or some other abnormality had been encountered
in all cases so far reported. The data offered
T a b le IV. V a p o r - L iq u id E q u il ib r iu m D a t a for B e n z e n e - C y c l o iie x a n e
in the present paper give additional evidence that
M ix t u r e s a t 759 M m . the “ abnormal” behavior of benzene is the rule
V apor M ole Fraction in: A ctiv ity
in such systems.
Pressure. M m . Liquid V apor C oefficient y R elative The data in the literature also indicate that,
Temp., Ben C y clo B en C y clo B en C y clo Ben C y clo V olatility
• C. zene hexane zene hexane zene hexane zene hexane a (C aled.)
while not necessarily ideal, mixtures of two hydro
8 0.2 762 .6 7 50 .9 0 .0 4 5 0 .9 5 5 0 .0 5 9 0 .9 4 1 1 .3 0 6 0.,9966 1.331
carbons in the same homologous series do not ex
8 0.0 758..0 7 4 5 .7 0 .0 6 6 0 .9 3 4 0 .0 8 4 0 .9 1 6 1.275 0..9986 1 .2 9 8 hibit constant-boiling mixtures. Examples have
79.5 746..1 7 3 4 .3 0 .1 0 5 0 .8 9 5 0 .1 2 9 0 .8 7 1 1.2 5 4 1,,009 1.262
7 9.3 741.,5 7 2 9 .5 0 .1 1 8 0 .8 8 2 0 .1 4 5 0 .8 5 5 1 .2 0 2 1..002 1.2 6 8 been cited o f binary paraffinic mixtures and binary
7 8.2 716..4 7 0 4 .7 0 .2 5 4 0 .7 4 0 0 .2 9 0 0 .7 1 0 1.211 1. 027 1.200
7 7.5 700. 8 6 8 9 .2 0 .3 8 3 0 .6 1 7 0 .4 0 7 0 .5 9 3 1.153 1..061 1.106 aromatic mixtures, and it is evident from this work
7 7.4 698..4 6 8 7 .1 0 .4 4 9 0 .6 5 1 0 .4 6 1 0 .5 3 9 1.1 1 8 1. 083 1.050 that at least one binary naphthenic mixture does
7 7.4 698..4 6 8 7 .1 0 .5 0 2 0 .4 0 8 0 .5 0 2 0 .4 9 8 1 .0 8 9 1..107 1.000
7 7.4 098..4 0 8 7 .1 0 .5 5 4 0 .4 4 6 0 .5 4 5 0 .4 5 5 1.070 1,,128 0 .9 0 4 not form an azeotrope. Consequently the difficul
7 7 .5 700. 8 6 8 9 .2 0 .5 9 7 0 .4 0 3 0 .5 8 0 0 .4 2 0 1.053 1..149 0 .9 3 2
7 7.6 703..1 6 9 1 .5 0 .0 2 8 0 .3 7 2 0 .6 0 5 0 .3 0 5 1.041 1..166 0 .9 0 7 ties encountered in achieving a satisfactory sepa
77 .6 703.,1 6 9 1 .5 0 .6 4 5 0 .3 5 5 0.6 2 1 0 .3 7 9 1.042 1..175 0 .9 0 2 ration of pure hydrocarbons from mixtures boil
7 7 .7 705..2 6 9 3 .4 0 .7 2 0 0 .2 8 0 0 .0 8 2 0 .3 1 8 1.020 1 .244 0 .8 3 4
7 7 .8 707. 4 6 9 5 .5 0 .7 3 8 0 .2 6 2 0 .6 9 8 0 .3 0 2 1.015 1..258 0 .8 2 0 ing in a range near to or below that of the aro
7 8 .0 711..9 7 0 0 .1 0 .7 5 3 0 .2 4 7 0 .7 1 7 0 .2 8 3 1 .0 1 5 1..242 0.831
7 8 .2 716..4 7 0 4 .7 0 .7 9 8 0 .2 0 2 0 .7 5 8 0 .2 4 2 1 .0 0 5 1..289 0 .7 9 3 matic compounds cannot be attributed to homol
7 8.3 718. 8 7 0 0 .8 0 .8 3 5 0 .1 6 5 0 .7 9 8 0 .2 0 2 1.008 1 .313 0 .7 8 1 ogous azeotropic mixtures but to azeotropes
7 8 .7 727..6 7 1 6 .0 0 .8 7 9 0 .1 2 1 0 .8 4 7 0 .1 5 3 1 .0 0 6 1 .341 0 .7 6 2
7 8 .9 732. .1 7 2 1 .4 0 .8 9 3 0 .1 0 7 0 .8 6 5 0 .1 3 5 1.004 1 .328 0 .7 0 6 formed by aromatics with paraffins or naphthenes,
7 9 .0 734. 5 7 2 3 .0 0 .9 1 4 0 .0 8 0 0 .8 8 9 0 .1 1 1 1.003 1 .353 0 .7 5 4
7 9 .2 739..2 7 2 7 .4 0 .9 2 8 0 .0 7 2 0 .9 0 7 0 .0 0 3 1.004 1..348 0.7 5 7 as pointed out by Bruun ((>), and also by non
7 9 .3 741..5 7 2 0 .5 0 .9 3 7 0 .0 6 3 0 .9 1 8 0 .0 8 2 1 .0 0 2 1 .354 0 .7 5 3 ideality of some paraffin-naphthenic mixtures.
7 9 .5 746. 1 7 3 4 .1 0 .9 6 0 0 .0 4 0 0 .9 4 8 0 .0 5 2 1.006 1..346 0 .7 6 0
T a b le V. V a p o r - L iq u id E q u il ib r iu m D a t a f o r C y c l o h e x a n e - M e t iiy l -
cyclohexane M ix t u r e s a t 7 60 M m .
V ap or Pressure, M ole F ra ction in: A ctiv ity

Mm. Liquid Vapor Coefficient ■
M eth y l M eth y l M eth y l M e th y l-
Tem p., C y clo cy clo C y c lo cy clo C y clo cy clo C y clo cy cio -
° C. hexane hexane hexane hexane hexane hexane hexane hexane
9 8 .3 1247 7 1 0 .0 0 .0 9 1 0 .9 0 9 0 .1 8 5 0 .8 1 5 1 .2 4 0 0.9609
9 6 .0 1172 6 6 5 .8 0 .1 8 5 0 .8 1 5 0 .2 7 8 0 .7 2 2 0.9751 1.012
94.1 1112 6 3 0 .0 0 .2 7 8 0 .7 2 2 0 .5 6 4 0 .4 3 6 1.3 8 7 0.7 2 90
9 1 .6 1038 6 8 4 .2 0 .4 4 7 0 .5 5 3 0 .6 3 7 0 .3 6 3 1 .043 0.7 2 90
9 0 .5 1005 5 6 5 .2 0 .5 6 4 0 .4 3 0 0 .6 7 5 0 .3 2 5 0 .9 5 0 5 1.002
8 7 .4 916 5 12 .0 0 .6 3 7 0 .3 6 3 0 .7 0 9 0 .2 9 1 0.9 2 34 1.190
8 4 .2 832 4 0 4 .5 0 .8 0 7 0 .1 9 3 0 .9 0 2 0 .0 9 8 1 .020 0.8301
8 3 .6 817 4 5 5 .7 0 .9 5 2 0 .0 6 8 0 .9 6 5 0 .0 3 5 0 .9 0 2 4 0.8 5 77
The relations between the equilibrium data and temperature

for the two series of mixtures are also given in Figure 2. The
minimum constant-boiling mixture formed b y benzene-2,4-
dimethylpentane mixtures contains 54.5 mole % benzene and
boils at 75.2° C. (757 mm.). The minimum constant-boiling mix
ture formed by benzcne-cyclohexane mixture contains 51.5 mole
% benzene and boils at 77.4° C. (759 mm.). Scatchard, W ood,
and M ochcl (87) reported isothermal data for this system, show
ing an azeotrope containing about 50 mole % benzene at a boiling
temperature of 70° C. MOL F R A C T IO N BENZENE IN LIQUID
The refractive indices of cyclohexane and methylcyclohexane
are so close together that an accurate analysis of their mixtures F ig u r e 4. R e la tiv e V o l a t il it i e s o f B in a r y
H y d r o c a r b o n M ix t u r e s
could not be obtained with the Abbe refractometer. The results
A Benzene-cyclohexane
are given in Tables II and V and Figure 3. There is no indication O Benzene-2,A-dlm ethylpentan»
A CK N O W LE D G M EN T (18) Hicks-Bruun, M . M ., and Bruun, J. H., Bur. Standards ./. Re

search, 8, 525 (1932).
Thanks are due to Trinidad Leaseholds Limited for permission
(19) Huffman, H. M ., and Parks, G. S., I n d . E n g . C hem ., 23, 1138
to publish the details of this work. (1931).
(20) Laar, J. J. van, Z. physik. Chem., 72, 723 (1910); 83, 599 (1913).
LITERATURE CITED
(21) Margules, M ., Sitzber. Akad. ll'iss. Wien, Math, naturw. Klasse,
(1) Burtell, F. E., and M ack, G. L., J. Phys. Chem., 36, 65 (1932). II, 104, 1243 (1895).
(2) Bell, G. H., and D avey, W . P., J . Chem. Phys., 9, 441 (1941). (22) Othmer, D . F., I n d . E n o . C iiem ., 20, 743 (1928).
(3) Bragg, L. B., I n d . E nq . C hem ., A n a l . E d ., 11, 283 (1939).. (23) Ibid., 35, 614 (1943).
(4) Bromiley, E. C., and Quiggle, D ., I n d . E n o . C hem ., 25, 1136 (24) Rosanoff, M . A ., Bacon, C. W ., and White, R ., J. Am . Chem.
(1933). Soc., 36, 1803 (1914).
(5) Bruun, J. H ., and Hicks-Bruun, M . M ., Bur. Standards J. Re (25) Sabatier, P., and Senderens, J. B., Compt. rend., 132, 1255 (1901).
search, 5, 612 (1930). (26) Scatchard, G., and Hamer, W . J., J. Am . Chem. Soc., 57, 1805
(6) Ibid., 5, 933 (1930). (1935).
(7) Carlson, I i. C., and Colburn, A. P., I n d . E n o . C hem ., 30, 581 (27) Scatchard, G., W ood, S. E., and M ochel, J. M-, ./. Phys. Chem.,
(1942). 43, 119 (193S).
(8) Cornell, L. W ., and M ontonna, R . E., Ibid., 25, 1331(1933). (28) Smith, A., and Menzies, A., J. Am. Chem. Soc., 32, 1453 (1910).
(9) Doss, M . P., “ Physical Constants of H ydrocarbons” , 2nd ed., (29) Smith, E. R ., J ■ Research Nall. Bur. Standards, 26, 129 (1941).
1939. (30) Smith, E. R „ and Matheson, H „ Ibid., 20, 645 (1938).
(10) Edgar, G ., and Calingaert, G ., J. Am. Chem. Soc., 51, 1540 (31) Smittenberg, J., H oog, H., and Henkes, R . A., J. Am. Chem.
(1929). Soc., 60, 18 (1938).
(11) Egloff, G ., “ Physical Constants of H ydrocarbons” , New York, (32) Timmermans, J., Bull. soc. chim. Bely., 36, 502 (1927).
Reinhold Pub. Corp., 1937. (33) Timmermans, J., J. chim. phys., 23, 760 (1926).
(12) Egloff, G ., and Grosse, A. V., Universal Oil Products Co., (34) Timmermans, J., and Martin, F., Ibid., 23, 733 (1926).
Booklet 217 (1938). (35) Tongberg, C. 0 ., and Johnson, F., I nd . E no . C hem ., 25, 733
(13) Francis, A. W ., I n d . E n q . C hem ., 33, 554 (1941). (1933).
(14) Garner, F. H ., and Evans, E. B., Inst. PetroleumTech., 18, 751 (36) W ibaut, J. P., et al., Rec. trav. chim., 58, 329 (1939).
(1932). (37) W ojciechowski, M ., J. Research Natl. Bur. Standards, 19, 347
(15) Griswold, J., D .Sc., thesis, M .I.T ., 1931. (1937).
(16) Griswold, J., I n d . E n o . C hem ., 35, 247 (1943). (38) Zelinskif, Ber., 34, 2802 (1901).
(17) Griswold, J., and Ludwig, E. E., Ibid., 35, 117 (1943). (39) Zamaczyaski, A., J. chim. phys., 27, 503 (1930).
Soap-Boiling Equilibria for Sodium

Stearate
THE NEW PHASE, KETTLE W AX
JAMES W . M cBAIN , KENNETH GARDNER1, AN D ROBERT D. VOLD2
Stanford University, Calif.
T H E phase rule in its usual form applies, strictly, to equilib salt concentrations (not usually employed) is sodium stearate
ria in soap systems (2, 9). This has been carefully demon curd in equilibrium with lye. The kettle wax phase is dull white,
strated for a number o f binary systems containing water soft, and waxy. On cooling to about 70° C., it turns to a white
with pure and commercial soaps, and for the ternary system so solid curd, which turns again to kettle wax upon heating.
dium palmitate-sodium chloride-water (3, 5, 6, 11). In general, The problem has been approached by a number of independ
the data for commercial soaps in the presence of electrolytes and ent methods; a separate study of the same system was made by
water rather closely conform to the behavior calculable from Lee (7) using the vapor pressure method. Results for binary
their constituents (J)). systems naturally give the starting point for the boundaries in
Heretofore ternary soap systems have been interpreted as in the ternary systems. The simplest m ethod is visual examination
volving equilibria only between curd, neat soap, middle soap, (usually between crossed polaroids), but the precaution must
nigre, and lye. The only existing studies of such ternary systems be taken of first heating the system until it is one limpid isotropic
of a single pure soap are for laurate and palmitate (S, 6, 11), and liquid. In the synthetic method, temperatures at which the
they are now known to be incomplete, since at that time the wax various phase changes occur are observed for systems of known
like island phase, “ kettle wax” , had not been discovered. Data composition, and a boundary may be delimited between pairs on
are given here for sodium stearate. either side of it. The separation of phases is greatly helped by
In 1940 Gardiner (1) found evidence that in the ternary sys centrifugation in an angle centrifuge kept at the equilibrium
tems sodium stearate-sodium chloride-water and sodium pal temperature. Where possible it is highly advantageous to ana
mitate-sodium chloride-water a waxlike form exists at soap boil lyze both layers even though only one phase may separate com
ing temperatures. This new kettle wax phase occupies a promi pletely from the other; usually, however, only the latter phase
nent position in the middle of the phase diagram. It is so pre affords a good sample. It is a requirement that a straight line
dominant that many of the equilibria commonly assumed to be connecting either analysis with the total composition of the sys
the result of "graining out” into soap curd and lye are now shown tem must pass through the actual composition of the other phase
to be equilibria between kettle wax and lye. Only at still higher as well as all mixtures of the two. These methods have been
1 Present address, Lever Brothers C om p a n y, C am bridge, M ass. supplemented by microscopic observations using a heating stage
s Present address, U niversity of Southern C alifornia, Los Angeles, C alif. and by calorimetric determinations which show large heat effects
in the neighborhood of the island phase. Visual and microscopic

examination frequently show not merely the number but the
nature of coexisting phases. Curd is an opaque white fibrous
solid; the waxy phases are much more translucent and waxy look
ing.
The sodium stearate employed was that used and described by
McBain, Void, and Frick (S).
I S O T R O r iC R E G IO N
The most accessible part of a ternary soap system is the iso

tropic fluid phase, or ordinary soap solution containing salt. As
shown in Figure 1 for sodium stearate-sodium chloride-water, the
isotropic solution extends continuously from pure water and is HETEROGENEOUS
bounded on its more concentrated side at each temperature by
the series of compositions known as nigre which can exist in
equilibrium with neat soap. This region of isotropic solutions ex
tends on its lower side into the brine containing almost no soap,
entitled “ lye” . On the lower part of the curve, toward the right-
hand or sodium chloride side, there is a peculiar but characteristic
indentation of the boundary known as the bay region. In the
bay region the tie lines run diagonally downward toward the lye.
Such straight tie lines may be drawn on the composition diagram
to connect any two phases in equilibrium. The outside tie line of
any such group forms one side of a triangle; any point within that TO SATU R ATED
represents the total composition consisting of three phases in
S A L T S O L U T IO N
equilibrium. The actual heterogeneous equilibria are not repre
sented, since they vary greatly over the wide range of tempera
1 2 3 4 5 6 7
tures considered.
The isotropic boundary at 70° C. is represented by dashes as a
PERCENT NaCI
reminder that it is almost wholly supersaturated and therefore Figure 1. B oundaries E nclosing H om ogen eou s Iso
unstable with respect to curd, since the solubility of sodium tropic Soap Solutions W hich C ontain Sodiu m Stearate,
stearate even in the absence of salt is only 2 .5 % at 70° C. The W ater, and Sodium Chloride betw een 70° and 150° C.
stable boundary runs steeply down from this point on the left-
hand side to a lye containing practically no soap. The upper tip
of the 80° boundary is likewise in the region supersaturated with
separated from them; in two cases also the composition of the
respect to curd. Other dashed portions of boundaries at 80°,
incompletely separated mixture floating above the lye is shown.
90°, 100°, 115°, 130°, and 150° C. merely indicate that the bound
The usual observations are partly assembled in Table I ; Ti is
aries are not fixed with sufficient exactitude by observations in
the temperature at which, on cooling, the homogeneous isotropic
the immediate neighborhood.
phase reaches a boundary and begins to separate another phase
such as lye or neat soap. Further changes on cooling, designated
P H A S E D IA G R A M
as Ti and T3, were noted. The significance o f some of these
The completed diagram for 90 ° C. is shown in Figure 2, which changes could be identified by their appearance at those tem
includes the following phases: isotropic solution, curd, neat soap, peratures (indicated by footnotes). By plotting these data on a
middle soap, and kettle working diagram, it was
wax island. (It is not possible to deduce many
known whether at 90° C. of the results shown in
we are dealing with crys Since 1919 M cB ain and collaborators have been reporting Figure 2. One value of
talline curd fibers of so investigations o f phase diagram s o f single pure salts o f the Tc was obtained for the
dium stearate or crystal fatty acids. These have show n th e strict applicability o f temperature at which
line supercurd fibers, the phase rule, first suggested b y M erklen in 1906, and the the last portion of curd
since 90 0is the transition occu rren ce o f true heterogeneous equilibria in all o f these fiber disappears on heat
point between them and system s, in clu d in g even com m ercial soap m ixtures. They
ing. This was 90° C.
they are indistinguish have established the range o f existence o f the fo u r ch ie f
for tube 17 and resulted
able with the naked eye.) recognized phases in soap b oilin g. Ternary system s have
in a three-phase mixture
Most of the triangles been described in terms o f (1) curd fibers, (2) neat soap
of neat-nigre-lye. The
involving three phases phase (liquid crystalline soap solu tion ), (3) a sim ilar bu t
90° centrifuge permitted
in equilibrium are indi separate m iddle soap, and (4) isotropic solutions w ith c o m
cated in Figure 2 so that delimitation of the lower
plicated boundaries extending in to lye or brine con tain ing
the general direction of alm ost no soap. However, it is now proved by a variety o f boundary of the neat
the tie lines can be in m ethods, in clu d in g direct analyses, th at a new waxlikc soap field at 47 % soap
ferred from these alone. form , term ed “ kettle wax phase” , exists in a predom inant for 0.75% sodium chlo
However, the diagram p osition in the ternary diagram at soap boilin g tem pera ride.
in c lu d e s experimental tures even with the highest pure soap, sodiu m stearate. T o save space, micro
points representing, re Only w ith th e highest con cen tration s o f salt are curd scopic observations are
spectively, the composi fibers o f sodium stearate form ed at 90° C. Therefore a n o t r e c o r d e d . They
tions of chosen total sys general description o f ternary system s involves soap, water, comprise descriptions of
tems and the analyzed and electrolyte. the a p p e a r a n c e and
composition of the lye number of phases at a
100
T a b le II. T e m p e r a tu r e s a t W h ic h C a i.o r im e t r ic C h a n ges
O c c o b i n T e r n a r y S o d iu m S t e a r a t e S y s te m s
Sample %
No. N^St & NaCI Transition T e m p .“ , ° C.
22 8 1 .8 1.6 5 7 .7 , 8S.C
31 6 9 .6 6 .1 3 8 .2 (?), 5 6 .5 , 9 1.1
32 6 8 .2 5 .5 5 8 .0 , 7 1 .5 , 8 8 .6
43 8 2 .7 0 .3 7 9 .0 , 1 05 .0, 143.0
48 5 8 .1 1.48 5 8 .0 , 7 9 .1
49 5 4.7 1.52 5 7 .6 , 7 7 .1
45 8 6 .8 13.2 0 .0 0 7 7 .3 , 106, 117 (? ), 145 (?)

* A t which measurable heat effects occur; the largest are underlined.
metric changes may easily detect a phase change before, in these

nontransparent systems, it becomes visible to either the eye or
the microscope; many of these changes occur over a range of
temperature. B y far the greatest heat changes are between 77°
and 91° C.
The analytical results are shown graphically in Figure 2 by
placing as circles (at about 30 % soap) the total composition of the
tubes as made up, together with the composition of the correspond
ing lyes (almost free from soap) as found by analysis; these
pairs of points are connected by tie lines which are extended to
PERCENT NaC ward the other phase. In three cases analyses are shown for the
upper layer (more or less completely separated) in systems con
F ig u re 2. P h ase D ia g ra m fo r th e S oa p S y stem S o d iu m
taining most salt.
S tearate—S o d iu m C h lorid e—W a ter a t 90° C.
The outlines of the kettle wax island phase take into account
T h e w a x lik e k e t t le w ax p h a s e a p p e a rs as a n isla n d in a d o m in a n t
p o s it io n ; tie lin e s s h o w its e q u ilib r ia w ith lyes c o n t a in in g b e tw e e n all the available information, including the important vapor
4.2 a n d a b o u t 21% s o d iu m c h lo r id e . T h e p h a se s p re s e n t a re c u r d
o r s u p e r c u r d , c r y s ta llin e s o d iu m c h lo r id e , k e t t le w a x , n e a t s o a p , pressure isotherms obtained by Lee (7). A ■well known difficulty
m id d le s o a p , a n d is o t r o p ic s o a p s o lu t io n w h ic h in c lu d e s ly es. T r i in certain systems of singlo pure soaps has been the failure to
a n g le s re p re s e n t e q u ilib r ia b e tw e e n th re e c o n d e n s e d p h a ses a n d are
s e p a r a te d b y h e te r o g e n e o u s re g io n s o f tw o c o n d e n s e d p h a ses. D a sh ed obtain good examples of neat soap on lye. An attempted ex
lin e s a re p r o v is io n a l o r in a c c u r a te ly d e t e r m in e d p o s it io n s .
planation is given by the diagram in which only a little room is
left for such systems between the two adjacent triangles neat-
series of temperatures for each of eleven compositions, together nigrc-lye and neat-kettle wax-lye, drawn in heavy dashes. By
with the temperatures at which the changes began or were thus bringing the left-hand boundary of the kettle wax phase
completed. Similarly the calorimetric results for four systems close to the right-hand boundary of the soap-boiler’s neat soap
were paralleled by microscopic observations. The calori phase, it is clear that only over a narrow range of concentration
metric method has been described elsewhere {10). Table II of lye could the system neat-lye be realized.
gives the temperatures at which several systems exhibit unmis The most striking result to be- read from the diagram is the
takable heat changes; the largest are underlined. Inspection wide range of lyes in equilibrium with the newly recognized ket
shows that the change is slight. However, for example in the tle wax phase. This phase occupies a dominant position in the
case o f sample 31, the calorimetric change at 38.2° C. is paralleled diagram. Indeed, a subsequent study o f sodium laurate by
by a faint but distinct microscopic change at 37.5°. The calori- Thorburn {7A ) shows that at 90° C. none o f the tie lines from
lye go to curd, but they converge around kettle wax.
The upper limit of the kettle wax island at 90° C. is fixed, both
by vapor pressure determinations and by the observations that
T able 1. V is u a l O b s e r v a t io n so n C o o l in g H o m o g e n e o u s
I s o t r o p ic S y s t e m s op S o d iu m S t e a r a t e , S o d iu m C h l o r id e , systems containing 70% sodium stearate and 5 or 5 . 5 % sodium
an d W ate r chloride consist definitely of curd on lye; with 4 % sodium chlo
% A ppcarance at ride, 70 % sodium stearate gives wax and lye with a small amount
Tube N atïT N aCl Ti Tx Tx 90° C.
of curd.
1 10.01 4 .0 4 179.5* 1 5 4 .5 « 139«* W ay -lye
2 5 .2 1 ACKNOW LEDGM ENT
10.08 214* 1 8 3 .5 C 173«* W ax-lye
3 9 .8 5 6 .0 0 230 199.5«* 189 * W ax-lye
4 10.00 2 .1 9 • Isotropic aoln. Thanks are due Mrs. M . E. L. M cBain for assistance with the
5 10.00 3 .0 4 133.5* 123« 1ÔÔ«*’ N igre-neat-lye analytical work.
6 10.18 7 .2 1 2 4 7 .5 * 221« 210«* W ax-lye L IT E R A T U R E
7 5 .0 0 3 .0 8 148.5* 1 4 2 .5 « 131.5«* N igre-lyc
8 14.95 3 .0 1 127 * 117« N eat-nîgrc-lye (1) Gardiner, Kenneth, M.A. thesis, Stanford Univ., 1940.
9 19.00 3 .1 1 154* 144.5? 133.5? N eat-lye (2) McBain, in Jerome Alexander’s “ Colloid Chomistry, Theoretical
10 2 9.0 0 3 .0 3 189* 180.5? 167? W ax-lye
and Applied” , Vol. 1, pp. 137-64, Now Y ork, Chemical Cat*-
11 4 .9 1 2 .2 0 111* N igre-lyc log Co., 1926.
12 9 .7 5 2 .6 3 82“ 71*' Isotropic soln.
13 14. S9 2 .6 7 112“ 71« N eat-nigre-lye (3) McBain, Brock, Void, and Void, J. Am. Chem. Soc., 60, 1870
14 19.75 2 .1 4 8 5 .5 * 79«* Isotrop ic soln. (1938).
15 2 0 .8 0 0 .4 S b Isotrop ic soln. (4) McBain, Elford, and Void, J. Soc. Chem. Ind., 59, 243 (1940).
10 2 5 .0 5 0 .5 3 Isotrop ic soln.
(5) M cBain and Langdon, J. Chem. Soc., 127, 852 (1925).
17 2 9 .6 0 1.81 132.5* 113« N eat-nigre-lye (6) M cBain, Lazarus, and Pitter, Z. phyaik. Chem., A147, 87 (1930).
20 2 5 .5 0 0 .8 0 Isotrop ic soln.
27 2 8 .7 0 0 .81 Isotrop ic soln. (7) M cBain and Lee, I n d . E n g . C h e m ., 35, 97 (1943).
28 3 0 .8 0 0 .8 3 9<T5* N eat-nigre (7A) M cBain, Thorburn, and M cGee, Oil & Soap, in press (1944).
29 3 3 .6 0 0 .7 9 104.5* N eat-nigre (8) M cBain, Void, and Frick, J. Phys. Chem., 44, 1013 (1940).
• N igrc-lye. * Neat-nigre. c Neat-nigre-lye. «* W axy-n eat-lye. « W ax y- (9) M cBain, Void, and Void, J. Am. Chem. Soc., 60, 1866 (1938).
lye. (10) Void, Ibid., 63, 2915 (1941).
(11) Void and Ferguson. Ibid., 60, 2066-76 (1938).
Saccharification of Wheat Martin Roberts, Stephen Läufer,
Earl D. Stewart, and Leonard T . Saletan
by Fungal Amylases for S C H W A R Z L A B O R A T O R IE S , IN C ., N E W Y O R K , N . Y .
ALCOHOL PRODUCTION
.Mold bran is com pared with barley m alt fo r the sacchari analysis of the grains is given in Table I. The mold bran was a
fication o f m ashes intended fo r p rod u ction o f industrial commercial product called Polidase-C, prepared by Schwarz
alcohol. Five grains were tested, one corn , tw o types o f Laboratories. On a small scale it is prepared by mixing wheat
wheat, and tw o corresponding granular w heat flours. bran with an equal weight of water and spreading the mixture
.Mashes were m ade by b oth pressure (125° C .) and a tm os in one-inch layers on aluminum triiys. After sterilization in an
pheric cooking (66° C .); they were saccharified b y using autoclave for 30 minutes at 20-pound pressure, the trays of bran
8% m a lt, 2.5 -1% dry m old bran, 3% undried m old bran, are cooled and inoculated aseptically with a culture of Aspergillus
or 1% m a lt for cookin g and 2% dry m old bran fo r sacchari oryzae. Incubation is at 25 ° C. until the mold begins to sporulate
fication. W heat mashes cook ed a t atm ospheric pressure (about 4 days, as a rule). For the purposes o f these experiments,
and saccharified with 2.5-3.0% m old bran yielded 2-14% at this stage the m oldy bran is mixed with I 1/ : volumes of 95%
more alcoh ol than did sim ilar mashes saccharified with alcohol (1500 ml. of alcohol to 500 grams of bran and 500 grams
8% m a lt. M alt was superior for pressure-cooked mashes. o f water) to kill the mold spores. The alcoholic extract is pressed
Mold bran saccharification at 30° C. was equal or slightly out with a hydraulic press. The dried, ground press cake, known,
superior to m old bran at 52.5° in b o th pressure- and a t- as Polidase-C, was used in the experiments to be described.
inosplieric-cooked m ashes. In pressure-cooked mashes Tw o methods o f cooking, pressure and atmospheric, are used by
malt saccharification at 57.5° C. was generally superior to distilleries. Both were used in these experiments, and the pro
mold bran at either 52.5° or 30°, b u t gave low er alcohol cedures followed for wheat and 3 % mold bran are given to illus
yields in atm osph eric cooked m ashes th an m old bran trate the mashing methods:
saccharification at 52.5° or 30° C.
P r e s s u r e M a s h in g . Four hundred milliliters of water are
heated to 50° C., and 97 grams of wheat and 0.5 gram of dry mold
bran are added. The temperature of the mash is held at 50° C.
H E need for increased output of ethyl alcohol from grain for 15 minutes while stirring constantly and then raised over a 15-
for war purposes has made the saccharification of grain minute period to 66° C. where it is held for 30 minutes. During
V mashes by agents other than barley malt o f great practical the next 15 minutes it is raised to 93° where it is held for a second
30-minute period. Finally the mash is autoclaved for one hour at
interest. Such agents would lessen the demands on the supply
20-pound steam pressure before being cooled to the temperature
of barley malt, already somewhat strained. Any improved chosen for conversion.
yield or increased output of grain industrial alcohol through their A tm o s p h e r ic M a s h in g . The procedure is the same as for
use would be highly desirable. Fungal amylases for saccharifi pressure mashing until a temperature of 66° C. is reached. It is
held at 66° C. for 55 minutes and then cooled to the conversion
cation of distillery mashes have been advocated by investigators
temperature within 5 minutes.
(4) who have shown that the yield o f alcohol from grain is in
creased by using mold bran instead o f malt. The combined weight of grain and total saccharifying agent
In the work reported here, the use o f mold bran for the sac used was 100 grams in all experiments. One gram of malt or
charification of wheat mashes was investigated. M old bran wet m old bran was used for liquefaction in experiments involving
percentages from 2.5 to 4.0 were employed, and the alcohol yields these materials.
from these mashes were compared with those obtained when C o n v e r s io n a n d F e r m e n ta tio n . Both types o f mashes are
8% malt was used as saccharifying agents. converted and fermented in the same way. A slurry o f 2.5
In the course of the work it was noticed that malt gave better grams of mold bran is prepared and held at the conversion tem
liquefaction of the mash than did mold bran, whereas the latter perature for 15 minutes before being added to the mash, which is
gave better conversion than malt. T o see if these favorable ef then held at the conversion temperature for another 30 minutes
fects could be combined to advantage, mashes were made using before it is cooled to 30° C. Conversion temperatures o f 52.5°
1% malt for liquefaction in cooking and 2 % mold bran for con and 30° C. were used. The mash is diluted with water to a total
version. '
In the preparation of mold bran, drying is neces- ________
sary to preserve it. It has been suggested (4) that
T a b l e I. A n a ly s is o f G r a in “
mold bran could be used undried, provided it was
W inter Spring
obtained immediately after the growth of the mold W inter Spring Granular Granular A .O .A .C .
was complete. This would save the drying cost and Corn W heat W heat W heat W heat M eth od ( i )
avoid possible losses of enzyme activity during the M oisture 1 0 .7 9 % 1 0 .6 8 % 9 .8 2 % 1 3 .1 2 % 1 3 .8 7 % P. 353, par. 2
Ash 1 .24 1 .67 1.82 0 .5 5 0 .5 5 P. 354, par. 8
drying period. A series of undried mold bran mash- Protein 8 .8 4 13.39 14.35 1 1.6 4 12.69 P. 26, par. 23
C rude fat 4 .1 4 1.74 1 .8 9 0 .9 4 0 .7 1 P. 356, par. 22
ings was therefore made under conditions found C rude fiber 1.84 2 .7 2 2 .6 3 0 .6 3 0 ,5 3 P. 358, par. 27
most favorable for saccharification by dried mold Starch 5 9 .1 3 5 7 .8 7 5 6 .9 3 6 3 .8 6 6 3 .6 8 P. 359, par. 31
R educing sugar
bran. Ccnlc.d. us d ex
trose) N one N one N one N one N one P. 358, par. 28
The grains used were obtained through Com N onreducing sugar
modity Credit Corporation. Corn and wheat were Ccalcd. as su-
crose) 1 .4 8 1 .9 4 1 .6 8 1 .5 5 1 .3 7 P. 359, par.29
supplied by Norris Grain Company, the granular
a D iastatic a ctiv ity of m alt on d ry basis was 194° Lintner a t 20° C . ( / , p . 161, par. 4 8 );
wheat flour from International Milling Company, dextrinogenic a ctiv ity o f m old bran was 64 units per gram (5 ).
and the malt from Pabst Brewing Company. An
811
way until the yeast becomes adapted as

T able II. Y ie l d o f A lco h o l“ in P roof G allons per 100 P ounds G r a in on shown by normal appearance and absence
D r y B a s is of irregular-shaped and thick-walled cells.
W inter Spring About three transfers are necessary.
C on W inter Spring Granular Granular The final culture is kept in the refrigera
version Corn* W heat W heat W heat Flour W heat Flour
tor and used as a stock culture. Stock
T em p ., Pres Pres A tm os Pres- A tm os- Pres A tm os Pres A tm os
Saccharifying A gent °C . sure sure pheric sure pheric sure pheric sure pheric cultures are transferred at least once every
2 weeks. T o prepare the inoculum for a
8 % malt 5 7 .5 10.37 10.05 9 .3 7 9 .6 8 9 .5 3 11.69 1 1.1 9 1 1.5 8 1 0.9 0
1 % malt*» + 2 % 5 2 .5 10.36 8 .9 2 10.45 9 .0 4 10.09 11.89 12.25 11.67 12.17 mash, 1 ml. of stock culture is used to in
m old bran 3 0 .0 1 0.3 4 9 .5 7 10.35 9 .8 0 10.10 11.35 12.39 11.87 12.42 oculate 20 ml. of wheat wort which is then
2 .5 % m old bran 5 2 .5 10.12 7 .7 6 9 .5 2 8 .8 5 10.22 11.02 12.15 11.21 12.13 incubated at 30 0 C. for 20 hours. A volume
3 0 .0 10.43 9 .6 8 9 .8 2 9 .3 8 10.18 11.19 1 2.2 2 1 1.6 6 12.43
3 % m old bran 5 2 .5 1 0 .5 0 7 .9 6 1 0.1 0 9 .2 4 10.35 12.4 2 12.39 of the 20-hour culture equal in milliliters
3 0 .0 10.34 12.24 to 3 % of the weight of the wheat mash
4 % m old bran 5 2 .5 10.17 !!! 9 .9 7 12À0 12.7 0 in grams is centrifuged. The remainder
3 % wet m old b ra n c 5 2 .5 ÓÍ93 10.25 9 .7 3 12.47 1 2.3 5
3 0 .0 10.2 4 9 .8 3 9 .2 0 11.96 1 1.6 8 of the 20-hour stock culture is used to pre
a Yielda are corrected for alcohol produced from malt. A lcoh ol from m old bran is negligible. All pare a new stock culture. In this way the
yields are averages of tw o or m ore results. adaptation is continued and a fresh stock
*» F or prem alting. c Sam ple contained 5 0 % m oisture, therefore 6 grama were used. culture is maintained. After centrifuging,
the supernatant liquid is poured off and
the yeast is resuspended in 10 ml. of
T a b l e III. C o m p a r is o n o f M a l t a n d M o l d B r a n Y i e l d s water. The suspension is centrifuged and
( in P e r C e n t ) the wash water poured off. The washed yeast cells are suspended
W inter Spring in 3 ml. of water and added to the wheat mash. The purpose of
Granular Granular the washing is to eliminate errors due to alcohol or fermentable
W inter Spring W heat W heat
Grain Corn W heat W heat F lour F lour substances present in the yeast inoculum.
Y ield using malt 10.37 10.05 9 .6 8 11.69 1 1.5 8
F or granular wheat flour mashes the yeast is prepared in a
H ighest yield using m old- 10.50 10.45 10.35 12.47 12.70 similar way. For the corn mashes yeast from a tube slant less
bran than a week old is suspended in 3 ml. of water as for wheat, but
Results elsewhere rep td .° the adaptation technique is omitted. The suspension is used as a
Average 10.5 0 10.04 9 .7 7 11.44 11.24*»
Highest 10.71 10.59 1 0.2 0 11.82 11.24 stock culture: the same procedure is followed as for wheat mashes
Theoretical yield c 11.72 11.31 11.17 12.93 12.99 except that the yeast is propagated in barley malt wort.
% o f theoretical yield 90 92 93 96 98
(m old-bran)
° B y other laboratories using the sam e grains; various mashing procedures R E S U L T S A N D C O N C L U S IO N S
we refused b u t the saccharifying agent was 8 % malt.
b Only one la boratory reported yields oni tnis
1 grain.
c Based on the starch and sugar content. The alcohol yield data obtained are given in Table II, and some
calculations derived from the data, in Table III. The results
obtained support the following conclusions:
weight of 550 grams (600 grams for corn), making the mash con
1. Mold-bran-saccharified mashes yield from 2 to 14% more
centration equivalent to 33 gallons per bushel o f wheat or 34
alcohol than malt-saccharified mashes, on all five of the grains
gallons per bushel of corn. A sample o f about 200 grams is re used. Favorable conditions were those in which 3 -4 % dry
moved for testing and analysis. The rest of the mash is inocu mold bran was used on a mash cooked at atmospheric pressure
lated with 3 % yeast and placed in an incubator at 30° C. for 4 8 - and saccharified at 30° C.
2. In most cases mashes saccharified at 30° C. with mold
60 hours, depending on the rate of fermentation. After comple
bran gave alcohol yields equal to or slightly higher than those
tion of the fermentation, 100 grams of the fermented mash are saccharified at 52.5° C. When using fresh, undried mold bran,
distilled, and the remainder is used for testing and analysis. the higher temperature gave better results, probably because at
The alcohol content of the distillate is determined by a pycnome- 30° these mashes are more easily infected than those saccharified
at 52.5°.
ter and corrected for the alcohol yield due to the malt, when malt
3. Mashes saccharified b y mold bran gave higher alcohol
is used. The amount of alcohol obtained from the mold bran is yields with atmospheric cooking, whereas malt saccharification
negligible, and no correction was made when this sacchrifying gave better yields when pressure cooking was used.
agent was used. 4. Mashes premalted with barley malt were more liquid than
corresponding mashes premalted with mold bran, but the alcohol
P R E P A R A T IO N o f y e a st in o c u l u m yields were not im proved over those obtained b y using 3-4 %
mold bran alone.
Early in the experiments it was noticed that the yeast propa 5. While the yields of alcohol obtained b y malt saccharifica
gated in wheat, and especially in granular wheat flour mashes tion of the mashes are in good agreement with those obtained
saccharified by malt, had irregular-shaped and thick-walled cells. elsewhere b y malt saccharification on the same grains, the alcohol
yields from atmospheric cooking and 3 % mold Dran for sacchari
These abnormalities were probably due to toxic substances in the fication were superior to those obtained from mashes saccharified
wort, as the presence o f a heat-labile agent in wheat flour which by 8 % malt. The superiority is particularly evident in the yields
inhibits the growth of yeast but not mold was reported b y Balls obtained from mold bran on granular wheat flour where the in
and Harris (#). Yeast grows normally in wheat mashes cooked crease amounts to about 5 % for granular winter wheat flour and
about 10% for the granular spring wheat flour, over the highest
under pressure or saccharified with mold bran. results obtained elsewhere for 8 % malt.
T o eliminate possible experimental errors due to substances 6. The simplified procedure possible and high yields obtain
that might be toxic, the yeast was adapted to them by successive able b y using 3 -4 % dry mold bran in place of 8 % malt should be
transfers in w ort prepared from the grain for which the yeast was of interest to those engaged in alcohol production from wheat or
granular wheat flour. Elimination of pressure cooking and re
to be later used as an inoculum. Yeast Y567 (obtained from duction of saccharification temperature plus higher alcohol yields
Northern Regional Research Laboratory) was used, and the in- at the same time are the attractive possibilities offered b y the
ocula were prepared as follows: mold-bran saccharification process.
In o c u lu m f o r W h e a t M a s h . The yeast from a tube slant is L IT E R A T U R E C IT E D

suspended in 3 ml. o f sterile water; 1 ml. of the suspension is used
to inoculate 20 ml. of wheat w ort previously only pasteurized in (1) Assoc, of Official Agr. Chem., Methods of Analysis, 5th ed., 1940.
order not to destroy possible heat-labile toxins. (The wheat nutri (2) Balls, A. K ., and Harris, T . H ., Cereal Chem., 21, 74 (1944).
ent medium was made by heating 92 grams of ground wheat in (3) Kneen, E „ Beckord, O. C „ Sandstedt, R . M „ Ibid., 18, 741 (1941).
450 ml. of water to 70-75° C. for half an hour before saccharifica- (4) Underkofler, L. A ., and co-workers, I n d . E no . C h e m ., 31, 734
tion with 8 % malt. On cooling the mash the grains settle out and (1939), 32, 544 (1940), 35, 814 (1943); Brewers Digest, 17,
the supernatant liquid is used for the propagating medium.) N o. 12, 29 (1942).
The seeded wort is incubated at 30° C. for 24 hours, when 1 ml. of
the culture is used to inoculate a second 20 ml. of wheat wort. P b e s e n t e d b efore the D ivision o f Industrial and E ngineering C hem istry at
Successive transfers after 24 hours of incubation are made in this the 107th M eetin g of the A m e r i c a n C h e m i c a l S o c i e t y in C leveland Ohio.
REACTIVITY Ostborg (8). Although the selected 40-60 mesh samples-may
show small differences in other characteristics from tho rejected
finer sizes, they may be expected to have reproducible specific
surfaces and other physical characteristics of tho sample bed.
The reactivity indices are plotted against volatile content on the
o f dry ash-free basis in Figure 1. It should be noted that the re
activity is highest for the lowest reactivity indices. The anthra
cites are obviously the least, and the lignites arc the most re
SOLID FUELS active coals. In the middle-volatile range, however, there is

little variation o f reactivity with volatile content.
The curves in Figure 1 are given by the equations:
T u, =*= 8 .0 ° C. = 229.7 - 0.01041 ( 7 - 27.5)3 (2)
A . A . (MINING
Coal Research Laboratory, Tn ± 11-4° C. = 272.1 - 0.01338 ( 7 - 27.5)’ (3)
Carnegie Institute o f Technology, Pittsburgh, Pa.
The probable errors in these equations are the ranges above and
below the empirical curves within which there is an even prob
ability of finding the reactivity indices for a given coal. Since a
Reactivity indices are presented fo r sixty-eight coals and study of duplicate tests has indicated a precision of measurement
thirty-five cokes. T h e indices, varying inversely w ith re of about =4=2 ° C., there must be factors other than the volatile
activity, are fou n d to correlate w ith the volatile con ten t content which are responsible for the variation in reactivity index.
on the dry asli-frec basis. T hey are highest for cokes, d e 'A study of proximate analysis data has failed to show significant
crease rapidly w ith increasing volatile in th e anthracite correlations other than that with volatile, although there is a
range, are relatively con stan t fo r the m ed ium -volatile possibility that the reactivity indices increase with ash content.
bitum inous coals, and again fall sharply fo r th e h ig h - This would be expected, provided the ash did not exert a catalytic
volatile coals and lignites. O ther factors probably influ influence.
ence th e reactivity, b u t they are n o t disclosed by the IN D IC E S F O R C O K E S
available data. W hile th e precision o f the test is h igh , the
details o f th erm om etric procedure arc shown to influence R eactivity indices are available for thirty-five cokes, but these
the absolute scale o f reactivity indices. data must be broken into two groups, on the basis o f methods
used for determining their volatile contents. Equations 4
through 9 apply to the first group, containing twenty-one cokes
(Figure 2 and Table II), for which the volatile content on the
C O N S ID E R A B L E interest has been shown, both by pro
ducers and consumers of coal, in the Coal Research
Laboratory reactivity test since its first publication in
1937 (2). The test, originally proposed as a "m odified ignition
point m ethod” for the measurement of reaction rates of cokes
with oxygen, has been extended to all ranks of coals in addition
to cokes. Although the test is still undergoing development and
the data obtained may be subject to correction of absolute scale,
they offer the most comprehensive information on the reactivity
of American coals and cokes available at present.
The test procedure {2, 8) is, briefly, to heat the sample of coal
or coke in air, oxygen, or some known mixture o f oxygen and
nitrogen, and to determine the sample temperature and its rate
of rise at the instant when the furnace and the sample are at tho
same temperature. A t this moment it is supposed that there is no
heat flow either to or from the sample. Under these conditions the
rate o f rise becomes a measure of the reactivity of the sample.
By using different rates of power input to the furnace or different
oxidizing gases, it is possible to obtain different rates of rise at
various sample temperatures. These rates may be correlated
by the equation:
where T = absolute temperature

p = partial pressure of oxygen in oxidizing gas
From this equation the reactivity indices, T IS and T 7S, are de
termined as the temperatures at which the rates of rise in pure
oxygen would be 15° and 75° C. per minute, respectively.
IN D IC E S F O R C O A L S
Table I gives the proximate analyses and reactivity indices

for sixty-eight coals ranging in volatile matter on the dry ash-
free basis from 3.5 to 48.2% . The apparatus and procedures
Figure 1. R eactivity Indices, Tu and Tn, fo r Coals as a
for determining the indices, specifically the positioning o f thermo F u n ction o f V olatile M atter C ontents on Dry Ash-Free
couples, are essentially those described b y Sherman, Pilcher, and Basis
813
dry dish-free basis, V, and the ash fusion temperature, F, were

determined by the A .S .T .M . procedure: T able 1. R e a c t iv it y I n d ic e s op C o als
7’is, 7’7S,
Scam C o u n ty V .M .« Asli& B .T .U .r ° C .« i ° C.4
Tu =¿= 15.5° C. = 530.0 -- 29.27 V (4) W harton L uzerne, Pa. 3 .5 8 .7 14,841 405 470
C oop er L uzerne, Pa. 5 .8 9 .8 14,945 312 411»
Tn ds 17.1° C. = 593.1 -- 28.51 V (5) N o. 4 S chuylkill, Pa. 8 .5 7 .4 15,335 333 41G
M errim ac M o n tg om ery, V a. 1 0.7 1 8 .7 271 315
T „ =b 10.6° C. = 498.9 -- 243.9 log .7 (6) Prim rose N o . 6 Lewis, W ash. 1 0 .8 1 7 .9 15,080 292 374
Tn =b 17.0° C. = 562.2 -- 233.6 log 7 (7) M errim ac M o n tg om ery, V a. 1 1.8 1 5 .2 259 298
M cA l ester P o p e , Ark. 1 4 .0 1 0.9 15,742 258 305
Tu =fc 9.2° C. = 000.7 -- 240.4 log 7 - - 0.04261 F (S) Spadra Johnson, A rk. 1 4 .4 1 0 .7 15,832 255 305
L anghorne M o n tg o m e ry , V a. 1 5 .0 1 4 .8 273 317
Tu =fc 11.7° C. = 792.4 -- 225.6 log 7 -- 0.09604 F (9) P oca h on ta s N o . 3 M cD o w e ll, W . Va. 1 6.3 5 .6 15,769 254 285
P ocahon tas N o . 4 M c D o w e ll, W . Va. 1 6 .G 5 .6 15,659 245 295
P ocahon tas N o . 3 M cD o w e ll, W . V a. 1 7 .3 5 .3 15,739 239 282
The dashed curves in Figure 2 are given b y Equations 2 and 3 for U p p er K ittan in g C am bria, P a. 1 7.7 8 .7 15,873 225 269
comparison with the coals; the solid curves are given by Equa P ocahon tas N o. 3 M cD o w e ll, W . V a. 1 8 .2 6 .9 15,787 234 280
P ocahon tas N o. 4 M cD o w e ll, W . Va. 1 8 .2 7 .1 15,752 •238 286
tions 6 and 7. Linear Equations 4 and 5 fit the experimental
B eckley M cD o w e ll, W . Va. 1 8 .9 7 .4 15,645 242 292
points almost as well, but they cannot approximate the curvature U pper F reep ort C am bria, P a. 1 9 .0 7 .3 15,740 228 263
that would be expected in the range of petroleum cokes whose L ow er K ittan in g C am bria, P a . 1 9 .7 7 .0 15,777 235 283
U p p er K ittan in g C am bria, Pa. 1 9 .7 7 .4 15,484 232 272
volatile contents lie in the same range as the anthracites, where L ow er K ittan in g C am bria, Pa. 2 1 .7 5 .8 15,530 236 276
the empirical curves for the coals show their maximum curvature. Sewell R aleigh, W . Va. 2 2 .0 3 .1 15,748 228 264
The logarithmic functions are capable of forming smooth exten Sewell W y o m in g , W . V a. 2 3 .2 5 .8 15,780 216 248
L ow er Freeport C am bria, Pa. 2 3 .6 7 .2 15,534 229 266
sions from the empirical equations given for the coals and yet C arey B uchanan, Va. 2 5 .2 8 .7 15,701 230 277
Carey B uchanan, V a. 2 5 .6 9 .4 15,700 228 267
curve sharply upward, as indicated for the low-volatile fuels.
Sewell F a yette, W . V a. 2 6 .0 5 .2 15,573 226 264
However, the logarithmic equations were fitted to the data on Sewell F a yette, W . V a. 2 7 .0 3 .7 15,619 227 270
cokes with no attem pt to join sm oothly with the data on coals. M iller F a yette, P a. 2 8 .9 10.1 15,507 225 258
L ow er K ittan in g Indiana, Pa. 2 9 .5 8 .0 15,805 225 262
Equations 10 through 15 apply to the second group, containing U p p er F reeport Indiana, P a. 3 0 .9 9 .0 15,567 228 269
fourteen cokes (Figure 3 and Table III), for which the volatile U p p er Banner D ick enson , V a. 3 1 .0 7 .1 15,720 233 271
contents were determined b y a dry distillation test in which the U p p er B anner D ick enson , V a. 3 1 .0 5 .6 15,710 230 272
B u n Creek B uchanan, Va. 3 1 .8 5 .9 15,610 225 260
samples were heated to about 950° C. in a closed tube and the U p p er F reeport Indiana, Pa. 3 2 .9 8 .1 16,055 230 266
L ow er F reeport C learfield, Pa. 3 3 .3 8 .6 15,558 231 268
yields of various gases were determined:
P ow ellton L ogan , W . Va. 3 3 .4 3 .7 15,162 217 269
E agle Logan, W . V a. 3 4 .1 5 .2 15,342 226 271
Tn 13.4° C. = 544.5 - 136.1 7 ( 10) E agle Logan, W . Va. 3 4 .7 4 .0 15,459 221 263
C lin tw ood B uchanan, V a. 3 5 .0 5 .3 15,490 217 256
Tn =b 14.1° C. = 582.9 - 89.1 7 ( 11) C hilton L ogan , W . Va. 3 5 .5 3 .9 15,423 217 257
Tn 13.9° C. = 465.3 - 58.30 log 7 . ( 12) E agle Logan, W . Va. 3 5 .7 6 .0 15,450 229 265
U pper F reeport A rm strong, Pa. 3 6 .4 8 .9 15,296 229 269
Tn =fc 13.9° C. = 529.S - 40.12 log 7 (13) E lk horn N o. 3 F lo y d , K y . 3 7 .0 3 .5 15,101 209 251
Island Creek Logan, W . V a. 3 7 .1 5 .9 15,397 223 265
Tn 13.2° C. = 294.7 - 46.28 log 7 + 0.0697- F (14) Pittsburgh F a yette, Pa. 3 7.1 7 .6 15,353 229 283
Tn rfc 13.4° C. = 374.0 - 29.08 log 7 + 0.0637 F (15) P ow ellton F ayette, W . V a. 3 7 .3 4 .3 15,532 226 272
L ow er Island Creek L ogan , W . V a. 3 7 .4 4 .6 15,380 210 247
C hilton Logan, W . V a. 3 7 .5 3 .1 15,342 215 258
D o ro th y B oone, W . V a. 3 8 .0 6 .0 15,260 215 256
The quoted volatile contents are the total yields of volatile less Island Creek L ogan , W . V a. 3 8 .1 5 .1 15,310 223 264
the carbon dioxide and water, and are given as percentage of C hilton B oon e, W . V a. , 3 8 .2 7 .5 15,160 212 251
dry ash-free coke. The quoted volatile contents for these four D a rb y H arlan, K y . 3 8 .5 4 .7 14.950 200 241
E lkhorn N o . 2 F lo y d , K y . 3 8 .5 5 .4 14.950 202 239
teen cokes are all less than those given for the first group of H igh Splint H arlan, K y . 3 8 .8 3 .5 14,920 201 233
C edar G rove M ingo, W . Va. 3 9 .0 6 .5 15,210 216 258
cokes, but this results only from the differences in analytical
procedure and calculation. The cokes o f the two groups, of B elm ont Kanaw ha, W . V a. 3 9 .2 7 .3 15,080 210 244
C edar G rove Kanaw ha, W . V a. 3 9 .5 6 .9 15,180 211 246
similar reactivity indices, are not fundamentally different. Again Harlan H arlan, K y . 3 9 .7 6 .1 14,940 206 249
Island Creek Logan, W . Va. 3 9 .9 5 .1 15,213 214 262
for comparison with the coals, the dashed curves in Figure 3 are Illinois N o. 6 Franklin, III. 4 0 .3 9 .8 14,572 183 205
given by Equations 2 and 3, and the solid curves are given by B on A ir N o. 2 Fentress, Tenn. 4 0 .9 1 1 .9 211 250
14,950
Equations 12 and 13. The data are represented equally well Island Creek L ogan , W . Va. 4 1 .3 4 .9 15,211 212 265
Illinois N o. 5 G allatin, 111. 4 3.1 1 0 .0 14,734 209 240
by linear Equations 10 and 11; but the logarithmic equations are Illinois N o. 6 H enry, III. 4 4 .0 1 1 .0 14,386 167 180
preferred since, b y analogy with the other fuels, a more extended Illinois N o. G M a cou pin , 111. 4 5 .0 1 1.9 14,025 165 178
range of volatile content would have demanded a curved father N oonan B urke, N . D ak. 4 6 .1 1 2 .8 12,140 171 205
N oonan D iv id e , N . D a k . 4 7 .0 1 1 .0 12,552 174 202
than a linear dependence upon volatile content. C otea u W ard , N . D ak. 4 8 .2 6 .1 11,850 178 207
While Equations 8, 9, 14, and 15, showing the dependence upon ° P ercentage volatile on d ry ash-free basis.
ash fusion temperature F, have been included, there is consider b Percentage on d ry basis.
e B .t.u . per p oun d o f dry ash-free coal.
able question as to their validity. For each group o f cokes the D ifferences less than 5 ° C . are n o t considered significant.
correlation between ash fusion temperature and reactivity in
dices appeared to be significant, and the equations for the in
dividual reactivity indices are consistent with one another.
However, between the two groups of cokes there is an inversion centage in the ash or in the coke, but no significant correlation
in algebraic sign of the coefficient o f ash fusion temperature. was found. This agrees with the finding that ferric oxide added
This would indicate either: (a) One o f the two sets of equations is to coke had no appreciable effect on its reactivity (1). A cata
not valid, the improbable circumstance having occurred that lytic influence of alkali in the ash might be expected, but for the
unknown factors had caused the apparent trend in the limited second group of cokes, for which the requisite data were available,
sample; or (6) there is a fundamental difference between the no significant correlation was found.
two groups o f cokes which is not believed to be true. Since the The significant difference in the two groups of cokes lies in the
most plausible reason for any correlation with the fusion tem methods used for the determination of volatile contents. It is
perature arises from the common influence of iron, both on the probable that the A .S.T .M . procedure wrould give the higher
ash fusion temperature and catalytic effects, a correlation was values when used on the same cokes. It is also possible, in this
sought between the reactivity indices and iron, either as a per- procedure, that an increased loss of weight might be due to leak-
600 650 of oxygen, as indicated in

Equation 1. As applied to
cokes, it is possible with a
5 50 600
single oxidizing gas to obtain
\ the rates o f rise over a consi
500 550 derable temperature range.
This was the basis for the
450 \ 5 0 0 \'
0 original evidence (2) for the
\ \ validity of Equation 1 and

was used to show that, in the
400 450 V . presence of certain added
catalysts on cokes, the as
\ H-
\ sumed first-order dependence
350
h-‘
400
Z \\ upon oxygen concentration
UJ
U
O
d X was not valid (1). The at
d
UJ
Û
ul
Q_ X- tem pt to prove the validity
300 3 5 0 iO __
h
•n of Equation 1 for a bitumin
ous coal b y making a series
250 300 of determinations in different
0
partial pressures o f oxygen
(S) is subject to criticism.
200. 250. Although the experimental
2 4 6 8 2 4 6 8 10 12
V O L A T IL E M A T T E R , PERCEN T V O L A T IL E M A T T E R , P E R C E N T data, plotted according to
Equation 1, gave a straight
Figure 2. R eactivity Indices, Tu and TVs, fo r T w en ty-O n e Cokes as F u n ction o f V olatile
M atter C ontents on Dry A sli-F rcc Basis, as D eterm ined b y A .S .T .M . P rocedure line, the temperature ranges
for individual oxidizing gases
were so limited that it is im
age of air into the crucible and consequent burning in an amount possible to say the individual lines were coincident. The fact
correlated with the reactivity o f the sample. This would affect that the experimental points fell close to a straight line may
the correlation with reactivity in the direction found. be more a characteristic of the apparatus than of the fuel.
On the basis of preliminary data in a new type of furnace now
P R E C IS IO N O F M E T H O D
being developed, it appears that for lignites the order is more
The precision of measurement was somewhat poorer with
cokes than with coals, approaching =t 10° C. for the highest re
activity indices. As with the coals, it is probable that factors
other than volatile content might correlate with the reactivity T able II. R e a c t iv it y I n d ic e s of C okes of G roup 1
indices; but these cannot be disclosed without more accurate V olatile, Ash Fusion % Ash on Tu, Tn,
*Cr
0
%° T e m p ., 0 C. D r y Basis 0 CA
p
data, particularly on volatile content as well as on reactivity
0 .7 3 2365 3 .6 0 514 613
index. The remarkable conclusion is that, for such widely di 0 .7 6 2386 3 .6 5 557 615
0 .8 2 2700 8 .5 0 510 554
v e r g e n t fuels as 0.8G 2365 3 .5 3 493 555
cokes, anthracites, 0 .8 7 2410 4 .3 3 493 556
0 .8 7 2450 6 .5 5 505 556
bituminous coals, 0 .8 8 2277 3 .7 3 512 582
0 .91 • 2263 3 .6 5 516 561
and lignites, at least 0 .9 6 2340 3 .5 8 406 567
to a first-order ap 0 .9 8 2271 3 .8 0 503 581
0 .9 9 2265 3 .7 5 504 562
proximation, the 1.01 2700 8 .2 7 484 542
1 .0 4 2263 4 .1 0 499 567
volatile content de 1 .0 5 2287 4 .4 5 502 573
termines the reac 1 .0 7 2700 8 .4 0 501 547
1 .1 0 2270 4 .4 8 481 556
tivity index. Other 1 .3 0 2340 3 .8 3 4SI 551
2 .1 2 2417 5 .8 3 423 515
f a c t o r s undoubt 7 .9 8 2000 0 .4 3 312 413
edly exist, but they 8 .0 8 2700 0 .3 3 248 301
1 1 .0 6 2700 0 .3 0 235 292
cannot be de
termined without a On d ry ash-free basis.
b D ifferences less than 10° C. n o t considered significant.
further i m p r o v e
ment in the proce
dure and probably T able III. R e a c t iv it y I n d ic e s of Cokes of G roup 2
also in the measure Volatilo, A sh Fusion % A sh on % FeiO j Tu, Tu,

0
d
%“ T e m p ., 0 C. D r y Basis in A sh 0 C.&
ment o f relevant
0 .0 3 3 2669 8 .2 5 1 2.6 6 544 586
physical and chemi 0 .0 3 9 2627 9 .3 0 9 .5 5 531 572
0 .1 1 3 2756 1 0.4 6 1 0.5 6 551 598
cal properties of the 0 .1 7 0 2687 10.47 1 0.2 7 542 581
fuels themselves. 0 .2 5 2 2434 9 .5 0 1 6.0 6 474 534
0 .3 0 5 2672 11.87 8 .4 0 506 569
There is, more 0 .3 5 5 2320 9 .6 3 9 .8 4 503 547
0 .3 5 8 2657 11.29 9 .7 0 487 518
over, considerable 0 .3 7 2 2332 8 .81 1 0.5 7 480 535
difficulty in prov 0 .3 9 8 2597 9 .7 8 9 .4 1 465 528
0 .4 4 9 2553 9 .8 7 9 .9 9 496 559
ing the general va- 0 .5 3 6 2545 9 .3 9 9 .7 0 499 567
0 .5 4 0 2471 8 .5 7 9 .7 7 453 519
Figure 3. R eactivity Indices, Tu fidity of the as-
8 .3 9 9 .2 6 484 549
0 .5 4 9 2434
and T75, fo r Fourteen Cokes as a
F unction o f V olatile M atter Con- u m e firs t-o rd e r n Less carbon d ioxid e and water on dry ash-free basis.
tents on Dry A sh-F ree Basis, as dependence u p o n b D ifferences less than 10° C. n o t considered significant.
D eterm ined by Special T u b e T est the partial pressure
nearly zero and increases toward one with decreasing volatile con of the sample, the reactivity indices do give a measure of the re
tent. Such a change in the character of the reaction may be due activity of the sample at different sample temperatures. Even
either to the differences in rank of the fuels or to the different for the low-rank coals where thero is greatest doubt as to first-
temperature levels within which the rales arc measurable. order dependence upon oxygen partial pressure, Tn does not in
In addition there is a practical difficulty in thermometric pro volve much extrapolation from the experimental points and
cedures. The reactivity indices depend critically upon the gives a measure of the reactivity in oxygen. Likewise, Tu in
determination o f that instant at which there is assumed to be no volves small extrapolations from the experimental points in air,
flow' of heat to or from the sample. In a particular case, the as and since the ratio of 75 to 15 is approximately that o f the partial
sumption that a single thermocouple o f the pair used to determine pressures of oxygen in pure oxygen and air, Tu, could be inter
this conditon W'asin error by 1° C. resulted in a 10° difference in a preted as Tu, measured in air. It is probably for these reasons
calculated reactivity index. The furnace thermocouple is placed that such remarkably good correlations are found between the
in contact with the outer surface of the tube holding the sample. reactivity indices and the volatile contents. The different
Since the temperature is rising, there must be a net flow' of heat slopes found for Tu and for Tn are indicative of the dependence
into the wall of the sample tube at all times. A t the instant when of the mechanism o f reaction upon the rank of the fuel.
the temperatures on the outer surface of the tube and in the mid
dle o f the sample are identical, there must be a minimum tem ACKNOW LEDGM ENT
perature within the tube wall. This must be true, otherwise the
The experimental data presented in this paper were obtained
tube wall could not be rising in temperature. Under these con
by J. J. S. Sebastian, a member of the staff of the Coal Research
ditions a portion o f the heat required to heat the sample tube must
Laboratory prior to August, 1942. Grateful acknowledgment
arise from the heat released from the sample. This is contrary
is here made to those who supplied the samples, and analyses
to the original assumption that identity of temperature of sample
for the work reported in this paper.
and furnace thermocouples corresponded to zero heat flow to or
from the sample. However, the portion of the heat taken from L IT E R A T U R E C IT E D
the sample may be relatively constant. This means that the
procedure is in error due to the inclusion of an unknown portion of (1) Sebastian, J .J .S ., D iv .o f Gas and Fuel Chem., A.C.S. meeting,
Boston, 1939.
the heat capacity of the sample tube together with the heat (2) Sebastian, J. J. S., and Mayers, M . A ., I n d . E n g . C h e m ., 29,
capacity of the sample; the result is an error in absolute scale 1118-24 (1937).
of reactivity indices. (3) Sherman, It. A ., Pilcher, J .M ., and Ostborg, H . N ., A m . Soc.
In spite of these criticisms, the reactivity indices have con Testing Materials, Bull. 112, 23-34 (1941).
siderable value. Unlike ignition points which are as much a P resen ted in p a rt b efore the D iv ision o f Gas and Fuel C hem istry a t the
function of the physical environment as they are of the reactivity 106th M eetin g o f the A m e r i c a n C h e m i c a l S o c i e t y , P ittsburgh, P a.
Solvent Dehydration
by Salting Out ♦ PREDICTION OF MAXIMUM DEGREE
OF DEHYDRATION
H. P. M EISSNER A N D CHARLES A . STOKES
Massachusetts Institute o f Technology, Cambridge, Mass.
T H E problem of removing water from aqueous solutions of where the possibility of forming a water-rich and a solvent-rich
organic solvents is often encountered industrially. Such layer exists.
dehydration is most frequently accomplished by distilla Some methods of dehydration may give greater “ clean-up”
tion, which may be carried out at, above, or below atmospheric than others. In industry, that combination of methods is nor
pressure, and with or without the addition of an entraining mally chosen which accomplishes the desired result at minimum
agent. It is also possible to dehydrate by other methods, such cost. Dehydration by a salting out process, used either alone or
as freezing and filtering out water crystals, reacting the water perhaps in combination with a distilling operation, often shows
chemically with materials such as lime, or adsorption of the water
real cost advantages. The object of this paper is to discuss some
on materials such as silica gel. Another dehydration procedure,
of the principles of dehydration by salting out and to illustrate
often mentioned in the literature (1) and perhaps more frequently
them with the system methyl ethyl ketone (M E K ), water, and
used in the laboratory than in industrial plants, is that known as
salts.
salting out. This involves bringing the wet solvent in contact
with some substance, usually an electrolyte, which has the power IN D U S T R I A L A P P L IC A T IO N
of withdrawing some of the water present to form a second phase
which can then be removed by decantation. The “ salting out” Dehydration by salting out may be accomplished either in
of ether by addition of sodium chloride to a solution of water in batch or continuous-flow operations. Batch operation might in
ether may be cited as an example. The dehydrating substance volve agitating the wet solvent either with the dehydrating sub
may be added either as a solid or as a concentrated aqueous solu stance or with an aqueous solution of this substance (hereafter
tion, this second method being more easily adapted to large-scale referred to as a “ brine” ). Similarly, flow operations would in
continuous countercurrent operations. The addition of such volve passing the wet solvent through a bed of the substance (if
dehydrating substances is advantageous only in those cases a solid) or possibly flowing the wet solvent through a tower coun-
“ Salting o u t” is a tech n iq u e little used industrially for mum possible dehydration of the solvent layer by the given de
dehydrating organ ic solvents. T his m eth od o f rem oving hydrating substance at the specific temperature has now been
water involves adding to th e w et solvent a dehydrating attained. Four phases are now present in equilibrium: the
substance, usually an in organ ic salt in solu b le in the vapor phase, the brine layer, the solvent layer, and the excess
organic liq u id , w h ich will result in the form a tion o f w ater- dehydrating material. For convenience, this type will be called
rich and solven t-rich phases. T he m in im u m water a "saturated three-component system” .
content o f the solvent phase is obtained w hen th e aqueous M e t h y l E t h y l K e t o n e - C a l c i u m C h l o r i d e - W a t e u . Since
phase is saturated w ith the dehydrating substance. A this system will be used to illustrate the points raised, its phase
method is presented for predicting this m axim u m degree diagram is of interest. Table I lists Stanton’s values (10) for
of dehydration for any given substance. T he m eth od is the composition of various equilibrium mixtures. Figure 1
illustrated by data and calcu lation s fo r dehydration o f presents these values graphically, together with a conjugate line,
methyl ethyl ketone by various salts. B, for interpolation purposes. Triangle mgb represents the
three-phase region (although line mb is not drawn into Figure 1
because it falls almost on the edge of the diagram); the two-phase
<§> regions are represented by areas mcb, gfb, and gem. The one-
phase regions are eafg and line dc. The latter is a line rather than
tercurrent to a brine. Only the batch case will be discussed in a region because calcium chloride is insoluble in ketone of this
this paper. low water content. An explanation of the significance of tri
A primary question in designing a salting out operation is the angular diagrams o f this sort may be found in any good text on
choice of a suitable dehydrating substance. In addition to hav physical chemistry or in Sherwood’s book on absorption (8).
ing dehydrating power, this substance must meet other require
ments. T o avoid contamination of the product, it obviously The values presented b y Stanton were obtained b y the usual
must be insoluble in the solvent being dehydrated. For this experimental techniques, as follows: Appropriate amounts of
water, M E K , and calcium chloride were shaken together until
reason sodium hydroxide would not be suitable for alcohol dehy
equilibrium was reached at the temperature in question. The
dration. It must not react with the solvent; therefore sulfuric phases were then separated by settling and decantation, and
acid, for example, could not be used to dehydrate ethyl alcohol. analyzed. Salt in each layer was determined b y evaporating
Similarly, chlorates and nitrates might not be suitable for de a known sample to dryness, and weighing the residue. Water
in the ketone layer was determined ny the method of Smith,
hydrating certain solvents due to explosion hazards. In addi
Bryant, and Mitchell (9). The M E K in the brine layer was
tion, cost questions, corrosion problems, and other considerations analyzed by the method of Marasco (5). The water content in
will narrow down the choice in any given case. the brine layer and the M E K in the solvent layer were deter
In spite of these limitations, dehydration by salting out is mined by difference.
The ivater content of the solvent layer was determined by
quite generally applicable. Normally it is not difficult to find a
Hanak (8) and W erby (11) b y two additional independent
suitable dehydrating substance that is insoluble in the particular methods: titration of a solvent sample with pure water and
solvent to be treated. For example, most inorganic salts of the judging the end point by the first appearance of two phases, and
type discussed in this paper are substantially insoluble in the comparison of the density of an unknown solvent layer sample
higher esters, ketones, ethers, alcohols, and the like. The de with that of known mixtures of ketone and water. These two
methods are applicable only where no salt dissolves in the solvent
gree of dehydration to be expected in all these cases can be pre layer.
dicted by the technique described here. Hence, it is not un
reasonable to expect that the salting out method of dehydration
This system is interesting for several reasons. First, M E K
may find somewhat wider application in the future, especially
and water are not miscible in all proportions, the composition
in those cases where the solvent forms an azeotrope with water,
of the two phases in equilibrium being shown by points d and e
which prevents1separation by ordinary distillation.
of Figure 1. Second, the composition of the vapor in equilibrium
with these two layers is practically the same as that of the sol
D E H Y D R A T IO N O R S A L T I N G O U T
vent phase, as the y-x diagram of Figure 2 shows. Consequently,
When a substance such as sodium chloride is added at constant further dehydration of an M E K -w ater mixture containing over
temperature to a solution o f water in an organic solvent in which
the salt is insoluble, the following occurs: As th'e dehydrating
substance is added in small increments, it progressively with W ATER
FIG URE 1
draws water from the solvent layer and goes into solution, forming SO L U B ILIT Y DATA F O R
a brine layer -which is immiscible with the solvent layer. A little M E K - C A L C IU M CHLOR/DE-
of the solvent also tends to dissolve in this newly formed brine WATER S Y S T E M
Layer. Further additions of salt cause more water to be with 2 3 - 2 6 ’C.
drawn from the solvent layer, which is thereby progressively de N U M B E R S R E T E R TO
hydrated. Finally, a point is reached where no more salt will 7o B E W EK H T
dissolve, and further salt additions result jp no further change in

the composition of the two liquid phases. It is clear that maxi-
T a b le I. E q u i l i b r i u m D a t a o n t h e S y s t e m C a l c i u m
C h l o r i d e - W a t e r - M E K a t R o o m T e m p e r a t u r e (23.5-26° C.)
W ater L ayer, W t. % K eton e L ayer, W t. %
K etone C aC li HiO HiO K eton e
4 5 .0 5 5 .0
i . 96 4 3 .6 0 5 4 .5 6 ! 705 9SL3
2 .2 0 3 3 .6 5 6 4.1 1.87 9S. 1
3 .4 2 2 6 .1 0 7 0 .5 4 .1 1 9 5 .9
5 .1 1 2 0 .3 5 7 4 .5 5 .2 2 9 4 .8
10.6 1 0.4 0 7 9 .0 7 .6 1 9 2 .4
1 6.8 4 .1 2 7 9.1 1 2 .0 0 8 8 .0
2 2 .4 0 .0 7 7 .6 12.60 8 7 .4
10 BO SO AO SO 60 TO 60 90 C*CI2
C E N T M E K IN LIQUID P R E D IC T IO N O F W A T E R V A P O R P R E S S U R E
If the dehydrating substance itself is hydrated as in the case

of calcium chloride, for which the stable hydrate is CaCl2.6H20
at room temperature, then the water vapor pressure over the
saturated three-component systems being considered must be
the same as over the solid hydrate. Inasmuch as the literature
lists the partial pressure of water over most such hydrated sub
stances or over their saturated aqueous solutions, the prediction
% of the vapor pressure of water oyer the solvent layer of the three-
0-
§ component systems presents no difficulties.
I f the dehydrating substance is not hydrated, as in the case of
sodium chloride, then the partial pressure of water over the satu
rated three-component system is not the same as the vapor pres
sure of water over saturated aqueous solutions of this substance.
As a first approximation, however, it may be assumed that the
water vapor pressure is the same for these tw o cases, as demon
strated below. A method for computing the difference between
these two vapor pressures is also presented. Since the literature
lists the vapor pressure over the saturated aqueous solutions of
eo most substances, the estimation of the desired vapor pressure
03 90 95 100
again presents no difficulties.
W EIGHT P E R C E N T M E K IN LIQUID
At constant temperature and pressure the familiar Gibbs
Figure 2. Vapor and L iquid C om position at equation can be written for the brine layer of the saturated three-
One A tm osphere component system as follows:
12.6 % by weight of water by simple distillation at atmospheric «

pressure is not possible.
In other respects this system is characteristic of those under where subscripts S, W , and D refer to the pure solvent, water,
discussion, in that the dehydrating agent is insoluble in the sol and dehydrating substance, respectively. Lewis (4) showed
that if substance S is added in infinitesimal amounts to a saturated
vent but readily soluble in water. For example, consider the solution of D in W , then since d In fn — 0, and since at very low
case of an M E K -w ater mixture containing 10% water, as indi concentrations,
cated by point h on Figure 1. As salt is added, the over-all com
f s = ksxs (2)
position of the system shifts along dotted line bh. Immediately
upon adding a little salt, water is withdrawn from solvent layer
it follows that Equation 1 reduces to:
and a brine layer forms. When enough salt has been added to
reach point j , for example, the composition of the two phases is
d ln/w = - ^ (3)
represented by the terminals of tie line A '. Finally after further , Xw
salt additions, point k is reached, which marks
the limit of salt solubility in the system; from
this point on, three condensed phases are pres
ent— namely, the two liquid phases and the solid
phase, shown by points m, g, and b. Further
additions of salt result in no further change in
the composition of these phases. The maximum
dehydration of the solvent layer, therefore, which ^
can be accomplished by calcium chloride is 5
shown by point m (representing 0.71% water by
weight). It is plain that this maximum de- ^
hydration is obtained when all three condensed ^
phases are present, and that this corresponds ^
with the above described saturated three-com- g
ponent system.
P r e d i c t i o n o f M a x i m u m D e h y d r a t i o n . The
maximum dehydration attained in a saturated ^
three-component system, as described above, O
can be predicted in any given case from the §
fact that the partial pressure o f water must be
the same over the three condensed phases, since
they are in equilibrium. The first problem to
be faced, therefore, involves predicting the par
tial pressure of water over the system. Second,
a relation between the composition of the sol
vent layer and the partial pressure of water
must be found. It is a simple matter to com
pute the composition of the solvent layer if F ig u re 3. Vapor Pressures o f Saturated Salt Solutions at Various
these two relations are known. T eniperatures
■ '
- ’ • ' V ' • I
Similarly, in a solution containing only the two components, o f such liquid-vapor composition data to lower or higher tem
IF and <S, the Gibbs equation for constant temperature and
peratures can usually be accomplished with good precision by the
pressure is:
use of some integrated form of the Gibbs-Duhem equation, such
' d In f s \ / ö In fw \ as the van Laar, Scatchard, Margules equations, etc. These re
xs 0 (4)
k àxs ) lations were discussed in the recent paper of Carlson and Colburn
(2). The van Laar equation, together with the data of Marshall
For the case in which minute amounts of S are added to pure
mentioned above, was used to predict the relation between the
W, S must obey an equation of type 2 (even though constant k
may be different numerically when component D is also present), partial pressure o f water and the composition of M E K -w ater
and Equation 4 reduces again to Equation 3. It follows from mixtures at various temperatures as follows. The van Laar equa
Equation 3 that the fractional lowering of the fugacity (or vapor tions are usually written :
pressure) resulting from adding a small given increment of S
to, say, one mole of pure W is the same as the fractional lowering B
of the vapor pressure of water resulting when this same increment In Vir
t x i + zw R y
(6)
of S is added to a saturated solution of D in one mole of W . The
solubility of S in a saturated brine is always small for the case RBzw1
under discussion, in which the dehydrating substance is quite In Vs (6)
insoluble in the pure solvent but readily soluble in water. Simi 7X1 + zw R )1
larly, the amount of solvent in the brine layer in these saturated
three-component systems is small enough so that it can be con where subscripts W and S refer to water and M E K , respectively.
sidered an increment to which Equation 2 above applies. The Rearranging Equation 5,
fractional lowering of the vapor pressure of water for this case is
never large, due to this small solubility, and may be computed 1 1 I R
from Equation 3 or neglected as a first approximation. (7)
V T In 7 jp V B + V B ZW
This procedure may be illustrated with the system M E K -
water-sodium chloride. It is reported (7) that at room tem
perature' 100 grams of a sodium chloride brine saturated with
both M E K and salt contain about 2.5 grams of M E K and 73
grams of water, corresponding to 0.0086 mole of M E K per mole
of water. The reduction in the partial vapor pressure of water
resulting when 0.0086 mole of M E K is added to 1 mole of pure
water at 25° C. is about 0.2 mm., as is evident from extrapolation
of Marshall's data (6) to this temperature with the help of the
van Laar equation discussed later. This corresponds to only an
0.8% reduction from the vapor pressure of pure water, which at
25° C. is 23.8 mm. The vapor pressure of water over a saturated
aqueous sodium chloride solution at 25° C. is approximately 18
mm., and this must likewise be reduced by 0.8 % when saturated
with M E K , in accordance with the above reasoning. This re
duction is obviously negligible, and so it follows that at this
temperature the vapor pressure of water over a saturated three-
component system of the type discussed is practically the same as F igu re 4. V an L aar C o rre la tio n
that over a saturated aqueous solution of the dehydrating sub fo r M E K —W a te r S o lu tio n
stance in question. It is clear that this vapor pressure reduction
remains negligible at all temperatures for which the solubility
of the solvent in the saturated aqueous solution of the dehydrat Hence the reciprocal of the term V T In yir plotted against *ir
ing substance is small.
should give a straight line on rectangular coordinate paper.
On the basis of the foregoing, it is possible to classify dehydrat Figure 4 presents the data of Table II plotted in this fashion, and
ing substances once and for all with respect to their dehydrating shows that an excellent straight lino can be drawn through the
power in a salting out operation. That is, their relative de first four points. These points were favored rather than the
hydrating power at any given temperature stands in inverse re entire spread of points, since the region of low-water concentra
lation to the vapor pressure over their saturated aqueous solu tion is of greatest interest in this case. The values of R and B
tions. Figure 3 shows the familiar vapor pressure curves of are now readily determined from the slope and intercept of the
saturated aqueous solutions of several salts numbered in order of dotted line of Figure 4, and are found to be 0.907 and 711, respec
increasing dehydrating power. As demonstrated above, these tively. Inserting these values in the van Laar equation, it be
vapor pressures also apply over the type of saturated three-com comes possible to predict the partial pressures of water over M E K -
ponent systems under discussion. Hence, lithium chloride is the water mixtures for any temperature. The solid lines of Figure 5
best and magnesium sulfate the poorest o f the eight salts plotted present this information in graphical form.
in Figure 3, with respect to dehydrating power in a salting out
P R E D IC T IN G M E K D E H Y D R A T IO N
operation of the type discussed.
It is now possible to predict the maximum dehydrating power
W ATER VAPO R PRESSURE OVER SOLVENT LAYER of the salts plotted in Figure 3 with respect to M E K -w ater m ix-
Granting that the partial vapor pressure of water over the
brine layer is known, then the degree of dehydration accomplished
in any given saturated three-component system can be predicted T a b l e II. P a r tia l V a p or P re ssu re s o f W a te r o v e r
if the relation between the partial vapor pressure of water and W a t e r - M E K S o l u t i o n W h e n t =* 73.6° C.
composition of the solvent-water layer is known. Since water l
xw VW P 7V . 71r r.w V T In ri
almost never forms an ideal solution with organic solvents, these
0 0 6 1 9 .7 0
vapor pressures cannot be predicted by Raoult’s law but must be 0 .0 5 9 0 .0 6 8 0 .0 6 ‘. 59 0 .0 5 2 6 0 .0 3 9 3
calculated from laboratory data. Experimental data on vapor 0 .1 0 1 4 6 .0 7 1 4 .0 5 .3 5 0 .1 1 1 0 .0 4 1 4
0 .1 5 1 88 .0 7 3 5 .0 4 .5 9 0 .1 7 6 0 .0 4 3 5
pressure or vapor compositions over solutions of water and most 0 .2 0 2 1 6 .0 7 4 8 .0 3 .9 5 0 .2 5 0 0 .0 4 5 9
0 .2 5 2 3 8 .0 7 5 7 .0 3 .9 9 0 .3 3 3 0 .0 4 7 8
important organic solvents are available in the literature. Usu 0 .3 0 2 5 2 .4 7 6 0 .4 3 .0 8 0 .4 2 9 0 .0 5 0 8
ally, these data are reported for normal boiling temperatures only. 0 .3 5 2 5 9 .3 7 6 0 .9 2 .7 1 0 .5 3 9 0 .0 5 3 8
0 .4 0 2 6 2 .8 7 5 9 .8 2 .4 1 0 .6 6 7 0 .0 5 7 4
For example, Marshall (6) reports the relation between composi 0 .4 1 2 8 2 6 3 .6 7 5 9 .4 2 .3 4 0 .7 0 3 0 .0 5 8 2
0 .9 4 79 8 2 6 3 .6 7 5 9 .4 1 .0 2 1 8 .2 0 .3 8 3
tion and vapor pressure over M E K -w ater mixtures at 73.6° C., 1 .0 2 7 3 .0 2 7 3 .0 1 .0 CO
as shown in the first three columns of Table II. Extrapolation
chloride is capable of the greatest dehydration

of the several salts considered. A t 20° C., for
example, its use results in an M E K -w ater mix
ture containing only about 1.2 mole % (0.3
weight % ) water.
The dehydrating power of salts predicted by
1this method is compared with the results obtained
by experiment in Tables III and IV, which show
the values of Stanton (.10), Hanak (3), and Werby
(11) for various salts. Agreement is good, par
ticularly in view of the comparatively poor pre
cision of the experimental methods used.
C O N C L U S IO N S
%
Organic solvents containing dissolved water
may be dehydrated by adding various substances
2.0 4.0 6.0 10 20 40 60 100 insoluble in these solvent-water mixtures but
P A R T IA L V A P O R P R E S S U R E O F W A T E R . M M . M E R C U R Y soluble in water. A method is presented by
which the maximum dehydrating power of such
Figure 5. Partial Vapor Pressure o f W ater over M EK—W ater Solutions
at V arious C oncen tration s and Tem peratures substances may be predicted. This involves a
knowledge of the relation between the partial
water vapor pressure and composition of solvent-
tures. For example, Figure 3 shows that the vapor pressure over water mixtures and the vapor pressure over the saturated aque
a saturated aqueous calcium chloride solution (solid phace = ous solution of the dehydrating substance. For the case in
CaClj.6H20 ) at 20° C. is 6.1 mm. of mercury. Similarly, which the dehydrating substance forms a stable hydrate, the par
Figure 5 shows that an M E K solution containing 3.6 mole % tial pressure of water over the four-phase system will be the same
water exhibits a partial pressure of water of 6.1 mm. of mercury as over a saturated aqueous solution of the dehydrating material.
at 20° C. Hence the maximum dehydration achieved by use of This is also shown to be nearly true for the ca°e in which the de
calcium chloride on M E K -w ater mixtures at 20° C. must result hydrating substance does not form a stable hydrate.
in a solvent layer containing 3.6 mole % (0.92 weight % ) of
water. The dehydrating power of the other salts can be simi ACKNOW LEDGM ENT
larly predicted.
T o illustrate the effect of temperature on dehydrating power Thanks are due to E. J. Stanton for obtaining the data pre
o f the salts considered, the vapor pressure curves of Figure 3 have sented in Figure 1 and Table I, and to E. W. Hanak and R . T.
been replotted on Figure 5 as indicated by the dotted lines. W erby for the data in Tables I II and IV.
The dehydration obtainable with these salts at any given tem
perature can be read directly from the intersections of the solid NOM ENCLATURE
and dotted lines on Figure 5. This chart makes it clear that the B, R = empirical constants
dehydrating power of lithium chloride is substantially inde / = fugacity
pendent of temperature. On the other hand, the dehydration y = activity coefficient = p/vx
accomplished by calcium chloride and sodium chloride is some k = H enry’s law constant
what affected by temperature. It also shows that the lithium p — partial pressure of given component in a solution
7r= vapor pressure of the pure component
T — temperature, ° Kelvin
T a b l e III. E f f e c t o f T e m p e r a tu r e o n E q u ilib r iu m W a t e r
x = mole fraction
C o n t e n t o f M EK in C o n t a c t w it h S a t u r a t e d B r in e s 2 = mole ratio— in a water-ketone mixture, for example,
_____________% Water hy Weight in Solvent Layer_____________ zw = xw/xs
T cin p ., LiCl C aC h K C H .C O i
» 6. Caled. E xptl. Caled. E xptl. Caled. E xptl.
Subscripts
20 0 .2 9 0 .3 8 0 .9 2 0 .8 3 0 .5 0 0 .6 9
30 0 .2 9 0 .4 0 0 .5 9 0 .7 0 0 .5 5 0 .6 8 D — dehydrating substance
40 0 .3 0 0 .4 2 0 .5 5 0.G5 0 .6 3 0 .6 7
50 0 .3 0 0 .4 3 0 .5 0 0 .5 7 0 .6 5 W — water
GO 0 .3 2 0 .4 5 0 .5 5 0 .6 2 0 .6 2 S — solvent being dehydrated
T a b le IV. E q u ilib r iu m W a t e r C o n t e n t o f M EK in C o n t a c t L IT E R A T U R E C IT E D
w ith S a t u r a t e d B r in e s
(1) Butler, J. A . V ., “ Fundamentals of Therm odynam ics” , pp.
% Water by Weight in Solvent 1G4-75, London, Macmillan C o., 1934.
_______________ Layer________________ (2) Carlson, H . C., and Colburn, A. P ., In d . E n g . C h e m ., 3 4 , 581
T em p ., E xp tl. by E xp tl. b y (1942).
C om pou nd “ C. Caled. density titration
(3) Hanak, E . W ., S.B. thesis, Mass. Inst. Tech., 1941.
M g C li 29 0 .9 8 1 .3 1 .2 (4) Lewis, G. N ., Proc. Am . Acad., 43, 259 (1907).
52 1 .0 1 .6 1.1
M gSO . 25 5 .1 6 .7 5 .3 (5) Marasco, M ., In d . E n g . C h e m ., 18, 701 (1926).
4 9 .5 4. S 5 .7 (6) Marshall, J. Chem. Soc., 8 9 , 1350 (1906).
5 2 .5 5 .0 5.-Í 5 .8 (7) Shell Chemical C o., “ Methyl Ethyl Ketone” , 1938.
LUSO, 25 3 .9 5 .1 “
SrCli 25 2 .7 2.9 2 .8 (S) Sherwood, T . K ., “ Absorption and Extraction” , New York.
51 2 .3 3 .6 M cGraw-Hill B ook C o., 1937.
N a C H .C O i 25 3ÍÓ 3 .3 3 .7 (9) Smith, D . M ., Bryant, W . M . D ., and Mitchell, J., Jr., J. Am.
4 9 .5 3 .4 3 .4
Chem. Soc., 61, 2407 (1939).
“ Determined by titration with Karl Fischer reagent. (10) Stanton, E. J., S.M . thesis, Mass. Inst. Tech.. 1941.
(11) W erby, R .T ., S.B. thesis, Mass. Inst. Tech., 1941.
PREPARATION AND PURIFICATION OF
Hydrous Oxide Sols by Ion

Exchangers
JOHN W . RYZNAR
National Aluminate Corporation, Chicago, 111.
T H E preparation and A new m ethod o f preparing and purifying hydrous oxide was maintained at approxi
formation of sols fall sols is described w hich makes use o f ion exchangers. mately 0.2 gallon per square
into two general classi C ation exchangers m ay he used; fo r exam ple, a solu tion foot of cross-sectional area per
fications, condensation and o f sodiu m silicate is passed through an acid-regenerated minute. One liter of solution
dispersion methods. In the cation-exchange m aterial; the sodiu m ion is replaced by was generally used in forming
former the small particles or a hydrogen ion which leaves the corresponding silicic acid the corresponding sols. In
molecules condense or ag in th e effluent substantially free o f oth er electrolytes. laboratory tests the sols were
glomerate to give a particle of W hen anion exchangers are utilized, m etals form in g in passed through the exchange
colloidal size; in the latter, soluble hydrous oxides can be used to form the corre beds twice in order to remove
large particles are reduced to sponding oxide sols by passage o f a salt solu tion through th e m a x i m u m a m o u n t o f
small ones of colloidal size. an alkali-regenerated anion exchanger. A sol form ed electrolyte.
The method here described is outside o f the exchange bed can he purified by passing it In all laboratory tests ordi
a condensation type but is an through an acid-regenerated cation and then through an nary distilled water was used
improvement, in the ease of ulkali-regenerated anion exchange bed; th e electrolytes in washing out regencrants.
formation of relatively pure arc rem oved and results are sim ilar to those obtained by Washing was stopped on acid
sols, over the older methods dialysis. cycles when the effluent was
which depend on double de alkaline to methyl orange,
com position or h yd roly sis and stopped on the alkaline
followed by dialysis for their formation and purification. cycles when the effluent was colorless to phcnolphthalein. In
None o f the older dialytic methods for the preparation of sols large-scalo plant production of silica sols, Lake Michigan water
substantially free of electrolytes are well adapted to large-scalo was used in the washing process.
production, as the methods o f purification are slow and the proc
esses rather expensive. On the other hand, this new method is C A T IO N E X C H A N G E R S
simple, inexpensive, and adaptable to producing large quantities
of sols in a short time. It makes use of various ion exchangers, In this process N a+, Ca++, M g ++, Al+++, and other metal ions
classified as follows: are removed and replaced with an H + ion. The use of an acid-
regenerated cation exchanger will bo illustrated by its adaptation
I. Cation exchangers to the formation of relatively pure silica sols. These sols have
1. Inorganic: natural and synthetic been made by a number o f different methods.
2. Organic: sulfuric-acid-treated coal, wood, waste petro- Graham (6) is generally credited with the discovery of silica
leum sludge, or lignite; resinsof thepolyhydricphenol-form - sol in 1861 when he prepared it by dialyzing a mixture of water
aldehyde and tannin-formaldehyde types.
glass and 10% hydrochloric acid. Jordis (8) formed sols by
II. Anion exchangers peptization o f silica gels. The formation o f silica sols by action
1. Inorganic: metallic oxides of water on silicon sulfide, chloride, and fluoride has been re
2. Organic: synthetic resins of the m-phenylenediamine- ported (5); clectrodialysis of the sol formed by hydrolysis of
formaldehyde type, as well as other resins of amine-aldehyde methyl silicate (7) or ethyl silicate tetrachloride (10) has given
condensation typo.
very pure sols.
Bird was the first to apply the cation exchangers to the forma
Ion exchangers have been the subject of much study and de
tion o f silicic acid sols as well as sols of other acidic oxides (2).
velopment in recent years. Cation exchangers can be regener
In this process relatively pure silica sols are obtained by passage
ated with acids, so that metallic cations can be replaced in solu
of a dilute solution o f sodium silicate through an acid-regenerated
tion by a hydrogen ion, leaving only the corresponding acids in
cation exchange bed. The sodium ion in the sodium silicate is
the effluent. Similarly, the anion exchangers can be regenerated
replaced b y a hydrogen ion to give a silicic acid sol in the effluent:
with sodium carbonate or sodium hydroxide; on passing the efflu
ent from an acid-regenerated cation exchanger through the anion E xIIj + NajSiOj- E xN aj + HjSiOj
exchanger, all anions such as SO, , Cl- , etc., are taken up and
replaced with a hydroxyl or carbonate ion yielding a deionized In plant operation, solutions o f sodium silicate having as high as
water as the final product. M ore detailed discussion of these 3 .5 % silica have been passed through the cation exchanger with
ion exchangers can be found in the literature (3, 9, 11,12). out difficulty with a flow rate through the exchange bed of about
In small-scale laboratory tests, the exchanger beds were set one gallon per square foot of cross-sectional area per minute. A
up in Pyrex tubes. The depth of the exchanger beds was ap typical analysis o f the composite effluent obtained is 2 .5 % silica
proximately 12 inches, and the surface area was 0.018 square foot. and 0.05% sodium oxide.
This gave a bed of exchange material with a volume of approxi It is possible to obtain sols of lower sodium oxide content, but
mately 0.02 cubic foot. The flow rate, unless otherwise noted, unless certain precautions are taken these sols easily form stiff
821
gels which are irreversible and, for many applications, undesir An interesting example showing the effect of the exchange bed
able. A sample of such a sol taken during a regular plant run at a itself on the final composition of a sol was encountered in the
point o f maximum purity had the following analysis: S i0 2, 2 .7 % ; passage of a solution of sodium molybdate through acid-regen
Na20 , 0.005% ; SO*, 0.0027%. This fraction sets to a stiff gel erated cation exchange beds of two different types. When a
in a few hours. clear colorless solution of sodium m olybdate was passed through
The pH of these purest fractions is generally around 6.0 to 6.3, a sulfonated phenol-formaldehyde resin, the effluent contained
which is near the 6.8 pH of the very pure sols formed by hydroly colloidal yellow molybdic acid; but when a solution of sodium
sis of silicon tetrachloride and subsequent electrodialysis (IS). m olybdate was passed through an exchanger of the sulfuric-acid-
Bradfield (4) reported a pH of 6.5 obtained after repeated wash treated coal type, the effluent contained the colloidal molybdenum
ing and centrifuging of silica. blue complex. Since there is some question as to the chemical
By allowing some sodium silicate to pass into the eflluont after constitution of this molybdenum blue complex, closer study of
the bed capacity is exhausted, the sol can be stabilized by the al the reactions may lead to a more complete understanding of this
kalinity from the sodium silicate; it can then be concentrated by compound as well as the exchange bed itself.
evaporation to a silica content as high as 20% with no danger of
gel formation. Analysis of such a concentrated sol made in large A N IO N E X C H A N G E R S
quantities and used for various industrial applications is: S i0 2,
SO« , C l- , NOa- and other anions are removed in this type of
2 0 % ; NasO, 0.37%.
process. Sols of hydrous aluminum oxides, ferric oxides, and
T he silica sols of maximum purity can be stabilized by other
other metal oxides have been prepared by use of an alkali
methods. For example, ammonium hydroxide, which does not
regenerated anion exchanger. Typical reactions may be ex
increase the mineral content, can be used. Likewise, low con
pressed as follows:
centrations o f aluminum ions will stabilize these sols.
After the exchanger lias become exhausted, it is backwashed E x (O H ), + AlClj — >- ExCla + A l(O H ),
and regenerated with acid in the usual way (S). Any highly E x(O H ), + Fe2(SO ,)a — >- Ex(SO,)3 + 2Fe(O H ),
ionized acid may be used, but sulfuric acid 'is preferred in most E x(O H )s + 3HC1 + Fe;Oj — >- ExCla + Fe20 3 + 3H 20
cases because of low cost. The excess acid is washed out o f the
bed and is then in condition for the next run. Alumina sols prepared in this way are not, on the whole, so free
During the first part of the process, the effluent from the ex of electrolyte or so stable as the silica sols and other sols already
change bed is practically neutral or else slightly acid depending discussed unless very dilute solutions are used.
on: (a) how thoroughly the excess acid used in regeneration was W ith aluminum salt solutions of about 1 % or higher, there is a
removed, (6) purity of the water used in washing out excess acid, strong tendency for gelation or precipitation of the sols in the
(c) purity of the water used in diluting sodium silicate, and (d) anion exchange bed with a resulting hardening or caking o f the
purity of the sodium silicate itself. Any sodium or other salts bed. On the other hand, 82.2% conversion of a 1 % solution of
present other than sodium silicate and sodium hydroxide will aluminum sulfate to the hydrous oxide has been obtained. This
give a corresponding amount of free acid in the effluent. conversion is based on the ratio of sulfate to alumina in the
The effluent behaves in much the same manner as a sol purified effluent. Likewise, with a 0:45% solution of aluminum chlo
by dialysis. At first it is clear and does not give a luminous ride, 99.2% conversion to the hydrous aluminum oxide has been
beam with Tyndall’s optical test, an indication that the silicic obtained, based on the chloride and alumina in the effluent.
acid particles are not large enough to scatter light. On aging Trivalent iron salts behave in much the same fashion as the
there is an increase in the intensity of opalescence, indicating a aluminum salts. A 0.12% solution of ferric sulfate passed
continuous condensation into larger units. The point at which through an anion exchanger showed 95.1% conversion of the
condensation ceases is dependent upon the alkalinity or the pres ferric sulfate to the hydrous ferric oxide sol. Sols from ferric
ence of other stabilizers. salts in concentrations of 1% or greater had a pronounced tend
The aging o f the sol to give larger particles of silicic acid can be ency to precipitate out in the bed.
accelerated by heating. It »may be of interest to mention that in The anion exchanger alone can be used for the purification of a
clarification o f water, silicic acid sols have been used along with sol. If a concentrated solution of ferric chloride is added to a
alum, sodium alumínate, and other coagulants in giving larger larger volume of boiling water, a rich red sol o f hydrous ferric
and tougher floe particles (1). In this process the pure sol hav oxide is formed which is very stable. The reaction may be ex
ing the small particle size is ineffective. It is only the larger pressed as
particle size of the silicic acid formed on aging or heating that FeCl, + 3H20 — Fe( OH) j + 3HC1
gives the highly hydrated, effective floe in clarification. It would
appear from this behavior that the silicic acid undergoes hydra On passing this sol through an alkali-regenerated anion ex
tion as well as condensation in the aging process. changer, the hydrous ferric oxide sol is obtained in the effluent
Silica sols produced by this process have already found numer unchanged, but the chloride ion is removed. A sol obtained in
ous applications in such widely divergent fields as ceramics, tex this manner analyzed as follows: pH, 5.9; Fe20 2, 0.236% ; Cl
tiles, detergents, and insecticides. Important uses for it are in as FeClj, 0.018% . This shows that 97.1% conversion o f the
the molding o f foundry sands and as a thickening agent in inks. ferric chloride to hydrous ferric oxide sol was obtained with the
In the ceramic industry silica sol is used along with such materials corresponding removal of the chloride. The pH of the influent
as magnesium sulfate, lime, and sodium aluminate to form plas was 1.3.
ticizing and bonding gels. In refractories it is used to provide a
C O M B IN A T IO N O F C A T IO N A N D A N IO N E X C H A N G E R S
dry bond, to accelerate the fluxing reaction, and to prevent the-
migration of auxiliary flux constituents. Other applications may Cations and anions are removed and replaced with hydrogen
suggest themselves from a study of its properties. and hydroxyl ions in this process. A combination of cation and
Other sols can be made in the same manner. Tungstic oxide anion exchangers can be used as a medium for the purification of
sols containing 99 % tungstic acid and 0.5 % sodium oxide have ' colloids that are formed by other methods. Colloidal sols
been formed. The soluble salt of any insoluble acidic oxide can formed by double decomposition or hydrolysis can be passed
be used to prepare the corresponding oxide sol. In some cases through the two exchange beds in series with a resultant removal
sols as they are prepared are unstable and may largely precipitate of most o f the electrolytes. This is, in effect, the equivalent of
on the bed. This is true when sodium aluminate or sodium stan- dialysis but is a much simpler and quiqker method of purifying
nate are used in preparing the corresponding sols. the sols.
A silica sol was formed by neutralizing a dilute solution o f so ACKNOW LEDGM ENT
dium silicate with sulfuric acid to a pH of 8.2. Fifty-three grams The writer wishes to express his appreciation to P. G. Bird,
of the neutralized JM brand of 42° B 6. sodium silicate were di J. A. Holmes, and F. K . Lindsay for their helpful suggestions.
luted to 1000 m l.; the resulting solution had a silica content of
1.5%. The electrolyte to be removed was sodium sulfate. The L IT E R A T U R E C IT E D
cation and anion exchangers used were o f the synthetic resin type. (1) Baylis, J „ J. Am. Water Works Assoc., 29 , 1 3 5 5 -9 2 (1 9 3 7 ).
Passage of this sol through the cation exchange bed resulted in (2) Bird, P. G., U. S. Patent 2 ,2 4 4 ,3 2 5 (Juno 3, 1941).
the replacement o f sodium ions with hydrogen ions. The effluent (3) Bird, Kirkpatrick, and M elof, J . Am . Water Works Assoc., ^ 9 ,
1526 (1 9 37 ).
from this bed then contained silica sol and sulfuric acid. Passage
(4) Bradfield, R „ J. Am. Chem. Soc., 44 , 9 6 5 (1 9 22 ).
of this effluent through the anion exchanger resulted in the re (5) Ebler and Fellnor, Ber., 44, 1915 (1 9 11 ).
placement of sulfate ions with hydroxyl ions. The final effluent (6) Graham, T ., Trans. 151, Roy. Soc. (London), 183 (1 8 61 ).
(7) Grimaux, Compt. rend., 98, 1434 (1 8 8 4 ).
contained the silica sol. The following analysis indicates the
(8) Jordis, E., Z. anorg. Chem., 34, 4 5 7 (1 9 0 3 ).
purity o f the sol obtained: S i0 2, 99.3% ; Na20 , 0.51; S 0 2, 0.022; (9) Lindsay, F. K ., Trans. Am. Inst. Chem. Engrs., 37, 5 4 7 (1 9 4 1 ).
Cl, 0.118; Fe20 2, 0.04. (10) Lottormoser and Iiiehn, Kolloid-Beihefle, 3 5, 123 (19 32 ).
To prepare sols of greater purity, it is possible that redistilled (11) M cElroy, K . P., U. S. Patent 1 ,8 1 1 ,5 8 7 (Juno 23, 1 931).
(12) Myers, F. J., I n d . E n o . C hem ., 3 5, 8 5 8 (1 9 4 3 ).
water should be used; also the exchange bed would have to be (13) Radczewski and Richter, Kolloid Z „ 96, 1 (1 9 4 1 ).
washed for a longer period of time to remove the maximum
P resen ted before the D ivision of C olloid C hem istry at tho 107th M eeting o f
amount o f excess regenorant present in the bed. the A m e r i c a n C h e m i c a l S o c i e t y , C loveland, Ohio.
Rheological Properties of Asphalt

R. N. T R AXLE R , H . E. SCHW EYER, AND J. W . ROMBERG
The Texas Company, Port Neches, Texas
The flow characteristics o f tw enty-seven asphalts from type or size o f viscom eter. Data are given w hich show
different sources and processed by various m ethod s arc that certain asphalts arc sim ple liquids at service (at
evaluated in rotary viscom eters o f a type suitable for the m osph eric) tem peratures, while others have the charac
m easurement o f high consistencies at a con stan t rate o f teristics o f com plex liquids. Evidence is given that the
shear. C onsistencies o f each asphalt at fixed tem peratures type o f flow (and i f com plex, the degree) depends on the
were determ ined at two or m ore m ean rates o f shear. I f source o f the asphalt, the m ethod and degree o f processing,
an asphalt is a com plex liq u id , the m easured consistency th e age o f the sam ple, and the tem perature at which the
decreases us the rate o f shear is increased. T he m agnitude evaluation is m ade. It is show n that som e asphalts
or degree o f com plex flow m ay be evaluated by c in the retain their com plex flow characteristics at tem peratures
equation, M = F/Sc; c is unity for asphalts that are sim ple as h igh as their ring and ball soften in g poin t. T h e re
(Newtonian) liquids b u t varies from unity for those that lations betw een the fu n dam en tal rheological charac
are com plex liquids. T his equation is valid over a co n teristics o f asphalts and the em pirical tests com m on ly
siderable range in rate o f shear and is n ot lim ited to one used by b itu m in ou s technologists are illustrated.
E VA LU ATIO N i O F F L O W C H A R A C T E R IS T IC S
M A T E R IA L S of high viscosity when subjected to stress
frequently exhibit many of the rheological proper The most satisfactory instrument for evaluating the flow char
ties of colloidal systems. The principles involved and acteristics of asphalts is the rotating viscometer; one type (.15)
problems m et in these rheological investigations are well illus was used to obtain most of the data discussed below.
trated by studies on asphalt; in such materials the flow may An arithmetical plot of shearing stress against rate of shear is
vary greatly with temperature, it may be simple (Newtonian) or curvilinear for many asphalts. However, such a rheological dia
complex (non-Newtonian), and the flow properties may change gram plotted on logarithmic coordinates is usually a straight line
with age because of the development or increase in amount of (Figure 1) which may be expressed by the relation:
colloidal structure.
These flow characteristics in many instances are the control M = F/S' (1)
ling factor in the applicability o f a given material for a particular where F —shearing stress, dyncs/sq. cm.
use. Rheological studies provide a sound basis for explaining the 5 =rate of shear, reciprocal sec.
c =slope of log S vs. log F plot
behavior of high-viscosity materials under service conditions as
M —value o f F when 5 = 1
exemplified b y asphalts when used in roofing and other water . ft
proofing applications or when applied as road binders and paper Equations of this type are discussed by Barr (5). If the material
plying adhesives. is a simple (Newtonian) liquid, constant c is unity and M is the
This paper discusses the application to asphalts of a complex viscosity in poises. For materials possessing complex flow char
flow equation, the effect of the dimensions of the viscometer and acteristics, constant c may be used to evaluate the degree of com
the rate of shear upon the rheological constants, the factors wlfich plex flow quantitatively. The more this constant varies from
affect the flow characteristics of asphalt, and the relations among one, the greater is the deviation from viscous flow since its nu
consistency in absolute units and data obtained b y certain em merical value is a direct measure of the shearing stress as. rate of
pirical flow tests that are commonly used. shear relation.
For the asphalts discussed in this paper, c lies between 0.45 dii of the inner and outer
and 1.00. However, values between 0.30 and 1.00 have been ob cylinders. A t this selected
tained on certain asphalts, and it is possible that the range will radius the equations for F
be extended by further investigations. Numerical values of c are and S are the same for a
rounded to the nearest 0.05, based on an average precision of viscous or complex liquid
=*=3% for the consistency determination. It is probable that as stated by M ooney and
increments in the value of c of less than 0.05 are not significant Ewart (8).
for*nsphalts.
Obviously an evaluation of c alone is not sufficient to explain
V A L I D IT Y O F F L O W
the flow of a material that behaves as a complex liquid; in addi E Q U A T IO N
tion a consistency value must be given. Constant M in Equation
1 cannot be used to evaluate the consistency (in absolute units) of The validity of Equation 1
complex liquids because of dimensional considerations resulting for a ten- to thirty-fold varia
from taking S to the c power. However, the ratio F/S has the tion in rate of shear may be
dimensions o f poises and may be used to calculate the consistency illustrated by plotting the
in absolute units at any given value of the rate of shear, S. For data in Table I for any one
this purpose a value of S should be selected which is approxi
F t Ii
of the twenty-five asphalts
mately the mean of the range of experimental rates of shear em investigated. The data for
ployed in the Theological measurements. Low rates of shear
(0.001 and 0.0005 sec.-1 ) were suggested previously (18) for con
two of these, A (a simple
liquid) and B (a complex I STAT0R I r
sistency measurements on asphalts at service temperatures such
as 25° C. (77° F.) because the experimental rates of shear used
liquid), are plotted in Figure
1 on logarithmic coordinates.
Sfr. ^ i
must be in tips low range to prevent slippage (IS). Conversely
The resulting straight lines
the viscosity of many soft asphalts is measured in the rotary vis
indicate that Equation 1 is
cometer at rates of shear up to 10 sec.-1 which necessitates con Figure2. R otary Viscom eter
valid over a thirty-fold range
siderable extrapolation in order to obtain the ratio of' F/S at
in rate of shear.
0.001 sec.-1 . Improvements in viscometers and increased knowl
edge of the flow characteristics of asphalts have indicated the The standard size rotary viscometer (No. 5), described in de
suitability of selecting 0.1 reciprocal second as the rate of shear at tail elsewhere (15), was employed in obtaining the rheological
which to compare the consistencies o f both hard and soft asphalts data o f Table I. A sketch of the instrument is shown in Figure 2.
with a minimum of error introduced by extrapolation. This pro Each measurement in Table I represents the equilibrium consist
cedure has been followed in reporting the results in this paper. ency (16) at a constant rate of shear. The upper limits for the
As is well known (6), the calculation of F/S for complex liquids rate of shear used depended on the material under study since
necessarily requires consideration of the effect of complex flow in thixotropy and slippage had to be considered. The lower values
the differential equations for F and S for the particular instru for the rate of shear were fixed by the time alloted for each meas
ment involved. However, complicated equations can be elimi urement since equilibrium viscosities at low rates of shear are
nated by judicious se'ection of the area at which the mean rate of time consuming for Newtonian asphalts of high viscosity as well
shear and shearing stress are measured. In the case of the ro as for those exhibiting complex flow. For soft materials the
tary viscometer, the rate of shear is calculated at a radius which minimum values for rate of shear were selected to produce torques
is the square root of the harmonic mean of the squares of the ra- that could be measured with precision.
Further evidence of the validity of Equation 1 is shown in Fig
ure 3 which gives the data for three different asphalts obtained
in five rotary viscometers having different dimensions. The di
mensions of these viscometers together with those for the stand
ard size instrument (No. 5) are given in Table II. Instruments
5A, 5B, and 5C have the same rotor size as N o. 5, but the stators
have different diameters which result in variable b/a ratios and
annuli of different widths. Instrument 6 has a large-diameter
rotor and a large-diameter stator but the same 6/a ratio as vis
cometer 5.
The rheological data obtained on the three asphalts with the
five viscometers are summarized in Table III, and indicate that
the consistency (at a rate of shear of 0.1 reciprocal second) and
the value of c in Equation 1 are independent of the dimensions
of the viscometers. '
L IM IT A T IO N S O F T H E M E T H O D
The use of Equation 1 as applied to asphalts has the following

limitations: When data for certain asphalts exhibiting a high de
gree of thixotropy are plotted on logarithmic coordinates such as
Figure 1, a slight curvature appears (concave toward the log S
axis) at high rates of shear. It is difficult to ascertain whether
the curvature of the log F vs. log S plot is actual or whether the
lowered equilibrium shearing stress is the result of thixotropic
effects produced during operation of the viscometer. In extreme
cases this curvature or failure to obtain an equilibrium consist
ency may be caused by slippage at the walls of the viscometer.
F igurc 1. R heological Diagram s at 25° C. from The experimental determination of equilibrium consistency for
Instrum ent 5 use in Equation 1 is time consuming for the hardest asphalts be
T a b le I. R h e o l o g i c a l D a t a O b t a i n e d -in R o t a r y V is c o m e t e r N o. 5
Penetrn- Rheolostical
R heoloß ical D ata E valuation
Softening at 2 5 ° C. 2 5° C ., M ean C onsist C onsistency
P oin t, R in g (7 7° F .), 100 T em p , of shearing ency, at S = D egree o
& Ball, 5 C m . per Gram s, m easurem ent, M ean rate of stress P, m ega 0.1 sec. - i , com p lex
Asphalt Process R esiduum S ource 0 C. (° F .) M in ., C m . 5 Sec. ° C . (° F .) shear, S, s e c .-1 d yn e s/cm .* poises m egapoises flow, c
A A ir-blow n G u lf C oast I 4 7 .8 (118) 184 98 25 (77) 3 .2 7 X 10-> 3,760 1 .1 5 1 .16 1 .0 0
9 .8 1 X 10-> 11,700 1 .1 9
3 .2 7 X 10~* 37,300 1 .1 4
9 .8 1 X 1 0 -» 114,000 1 .1 6
li A ir-blow n N ortheast Texas 4 4 .4 (112) 131 149 25 (77) 3 .2 7 X 10"» 3,030 0 .9 3 0 .6 4 0 .9 0
9 .8 1 X 1 0 -» 8,520 0 .8 7
3 .2 7 X 1 0 -» 24,900 0 .7 6
9.81 X 1 0 -» 63,500 0 .6 5
C Steam Pressure still 4 6 .1 (115) 150 + 59 25 (77) 3 .2 7 X 1 0 -» 8,450 2 .5 9 2 .6 5 1 .0 0
9 .8 1 X 1 0 '» 25,300 2 .5 7
3 .2 7 X 1 0 -» 86,500 2 .6 4
D A ir-blow n G u lf C oast II 3 3 .9 (93) 300 + 25 (77) 9 .8 1 X 1 0 -» 5,300 0 .0 5 4 0 .0 5 5 0 .9 0
3 .2 7 X 1 0 -» 15,100 0 .0 4 6
9 .81 X 1 0 -» 40,200 0 .041
E A ir-blow n G u lf C oast II 5 0 .6 (123) 86 80 25 (77) 9 .8 1 X 1 0 -» 7,560 7 .7 4 .0 0 .8 5
3 .2 7 X 1 0 -» 21,000 6 .4
9 .8 1 X 1 0 -» Slippage
F A ir-blow n M ixed 3 9 .4 (103) 126 225 25 (77) 3 .2 7 X 1 0 -» 5,130 0Ü 57 0 .1 4 2 0 .9 0
9 .81 X 1 0 -» 14,100 0 .1 4 4
3 .2 7 X 1 0 -» 40,800 0 .1 2 5
G A ir-blow n M ixed 4 3 .3 (110) 184 139 25 (77) 9 .8 1 X 1 0 -» ■6,000 0 .6 2 0 .4 9 0 .9 0
3 .2 7 X 1 0 -» 18,400 0 .5 6
9 .8 1 X 1 0 -» 48,500 0 .4 9
H A ir-blow n E ast Texas 5 8 .3 (137) 31 64 25 (77) 3 .2 7 X 1 0 - ' 8,160 2 5 .0 6 .8 0 .7 5
9 .8 1 X 1 0 -» 19,200 1 9 .6
3 .2 7 X 1 0 -» 47.800 1 4.6
I A ir-blow n N ortheast Texas 7 2 .2 (162) 4 .7 49 25 (77) 3 .2 7 X 1 0 -» 44,000 135 7 .9 0 .5 0
9 .8 1 X 1 0 -» 78,600 80
3 .2 7 X 1 0 -» 139,000 43
J V acuum G u lf C oast II I 4 8 .3 (119) 200 65 4 8 .3 (119) 0 .3 2 7 3,310 0 .0 1 0 2 0 .0 1 0 3 1 .0 0
0 .9 8 1 10,300 0 .0 1 0 5
3 .2 7 33,100 0 .0 1 0 2
K V acuum G u lf C oast II I 6 6 .7 (152) 11 11 6 6 .7 (152) 0 .3 2 7 3,380 0 .0 1 0 3 0.0 1 04 1 .0 0
0 .9 8 1 10,100 0 .0 1 0 3
3 .2 7 34,200 0 .0 1 0 5
L A ir-blow n M exican 4 4 .4 (112) 200 140 4 4 .4 (112) 0 .3 2 7 6,380 0 .0 1 9 5 0 .0 2 2 5 0 .9 0
0 .9 8 1 17,300 0 .0 1 7 7
3 .2 7 49,100 0 .0 1 5 0
M A ir-blow n M exican 6 7 .8 (154) 46 32 6 7 .8 (154) 0 .3 2 7 7,050 0 .0 2 1 6 0 .0 2 6 5 0 .8 5
0 .9 8 1 10,900 0 .0 1 7 2
3 .2 7 47,800 0 0146
N Steam E ast Texas 5 8 .3 (137) 200 30 25 (77) 9 .8 1 X IO "» 16,300 17.1 1 4 .3 0 .9 5
3 .2 7 X 1 0 -» 53,600 1 6.4
9 .8 1 X 1 0"» 153,000 1 5 .0
0 Steam G u lf C oast II I 5 9 .4 (139) 200 17 25 (77) 3 .2 7 X 1 0 -» 12,000 37 38 1 .0 0
9 .8 1 X 1 0 -» 3.3,300 39
3 .2 7 X 1 0 -» 121,000 37
P A ir-blow n C alifornia 5 2 .8 (127) 150 + 61 25 (77) 1 .0 X 1 0 -» 7,390 4 .6 3 .6 0 .9 5
3 .2 X 1 0 -» 14,100 4 .4
3 .2 X 1 0 -» 122,000 3 .8
Q A ir-blow n M id con tin en t 5 1 .1 (12) 170 57 25 (77) 1 .6 X 1 0 -» 11,500 7 .2 4 .3 0 .8 5
3 .2 X I 0 - » 21,500 0 .7
3 .2 X 1 0 -» 153,000 5 .0
U Steam E a st Texas 3 9 .4 (103) 116 176 25 (77) 9 .81 X 1 0 -» 4,080 0 .4 2 0 .4 0 1 .0 0
3 .2 7 X 1 0 -» 13,400 0 .4 1
9 .8 1 X 1 0 -» 39,600 0 .4 0
8 A ir-blow n E ast Texas 4 1 .1 (106) 113 196 25 (77) 9 .81 X I 0 - » 4,280 0 .4 4 0 .3 2 0 .8 5
3 .2 7 X 1 0 -» 12,500 0 .3 8
9 .8 1 X 1 0 -» 31,800 0 .3 2
T Fluxed G u lf C oast II 5 3 .9 (129) 1 7 .5 94 25 (77) 3 .2 7 X 1 0 -» 5,320 1 6 .3 3 .0 0 .7 0
9 .8 1 X 1 0 -» 12,200 12.4
3 .2 7 X 1 0 -» 27,100 8 .3
9 .8 1 X 10 * Slippage
U A ir-blow n G u lf C oast I 5 4 .4 (130) 200 52 25 (77) 9 .8 1 X 1 0 -» 8,260 8 .4 7 .2 0 .9 5
3 .2 7 X 1 0 -» 26,000 8 .0
9 .8 1 X 1 0 -» 76,600 7 .8
V A ir-blow n G u lf C oast I 6 1 .7 (143) 38 32 25 (77) 3 .2 7 X 10 -< 28,100 80 36 0 .8 5
9 .8 1 X 10-* 69,000 71
3 .2 7 X 1 0 -» 196,000 00
45 (113) 9 .8 1 X 1 0 -» 3,190 0 .3 3 0 .2 9 0 .9 5
3 .2 7 X 1 0 -» 10,300 0 .3 1
9 .8 1 X 1 0 -» 28,100 0 .2 9
65 (149) 0 .3 2 7 2,550 0 .0 0 7 8 0.0 0 82 0 .9 5
0 .9 8 1 ■ 7,220 0 .0 0 7 4
3 .2 7 22,900 0 .0 0 7 0
W Steam Venezuela 5 1 .1 (124) 150 + 57 25 (77) 3 .2 7 X 1 0 -» 12,100 3 .7 2 .4 5 0 .9 0 .
9 .8 1 X 1 0 -» 32.000 3 .3
3 .2 7 X 10-* 93.000 2 .8
9 .8 1 X 1 0 -» 238,000 2 .4
45 (113) 9 .8 1 X 1 0 -* 4,850 0 .0 4 9 0 .0 5 0 0 .9 5
0 .3 2 7 15,300 0 .0 4 7
0 .9 8 1 43,400 0 .0 4 4
65 (149) 0 .9 8 1 2,100 0 .0 0 2 1 5 0 .0 0 21 6 1 .0 0
3 .2 7 7,100 0 .0 0 21 7
9 .8 1 21,100 0 .0 0 2 1 6
X A ir-blow n G u lf C oast II 5 3 .9 (129) 52 77 25 (77) 3 .2 7 X 1 0-* 3,500 1 0.7 5 .4 0 .8 5
9 .8 1 X 1 0-* 9,480 9 .7
3 .2 7 X 10 -» 26,000 8 .0
Y A ir-blow n M exican 5 7 .2 (135) 125 64 25 (77) 3 .2 7 X 10-* 3,830 11.7 5 .0 0 .8 5
9 .8 1 X 1 0 - ' 9,570 9 .8
3 .2 7 X 1 0 -» 27,200 8 .3
F A C T O R S A F F E C T IN G F L O W P R O P E R T IE S
cause of the low rates of shear that must be used. This limits
the application of the equation to research studies of a funda Experienced asphalt rheologists recognize that the difficulties in
mental nature as described below. However, other methods of evaluating the flow characteristics of asphalt are caused by the
analyzing the rheological data on asphalts are being studied and different types of complex flow which may be encountered. T o o
may provide the means for rapid absolute consistency measure frequently these effects have been overlooked or ignored and
ments for use in control and specification work. general conclusions have been drawn from meager data on an in
sufficient number of asphalts. The necessity for adequate in

formation on a wide variety of asphalts is illustrated. M any of T a b le II. D im e n s io n s an d C o n stan ts for R otary
V is c o m e t e r s
these variations in the flow properties of asphalts apparently are
closely related to their colloidal nature. V iscom eter D im ension 5B 5C 5 5A 6
S ou rce. The nature of the hydrocarbons in the petroleum ? — sta tor radius, cm , 0 .9 5 2 1.111 1 .2 7 0 1 .5 8 8 1.905
from which an asphalt is prepared profoundly affects the colloidal i — ro to r radius, cm . 1.905 1 .9 0 5 1 .9 0 5 1 .9 0 5 2.858
* — rotor length, cm . 2 .5 4 0 2 .5 4 0 2 .5 4 0 2 .5 4 0 2.540
structure and thus the flow properties of the finished asphalt. — length of cylindrical
This is illustrated by the data in Table I for asphalts A, B, P, and section of atator, cm . 1.397 1.4 3 3 1 .5 8 8 1 .9 6 5 1.077
a — stator angle, degrees“ 59 6 3 .5 09 80 09
(j, all made by air blowing petroleum residua from widely differ 3 — rotor angle, degrees 90 90 90 90 90
w id th o f annulus, cm . 0 .9 5 3 0 .7 9 4 0 .6 3 5 0 .3 1 7 0.953
ent sources. Asphalt A is a simple liquid whereas B, having a b/a 2 .0 0 0 1 .7 1 4 1 .5 0 0 1 .2 0 0 1.500
lower consistency but air-blown from a different residuum, is a ° F o r each viscom eter, a (angle o f stator cones) was selected bo that the
complex liquid. Asphalts P and Q (complex liquids), air-blown follow ing equation was satisfied; b y using this angle the m ean rate o f shear
a t the ends is approxim ately the sam e as the m ean rate in the cylindrical
to about the same consistency, show different degrees of complex regions o f the viscom eter (5 ):
flow. P _ a » e gin d9 -f- eZ ( a /g gin> Oq - ł/ j sin Ög)
M e t h o d o f P r o c e s s in g . The way an asphalt is processed of where « — (6* — a*)/(6* -f- a*)
ten has a definite effect on the flow characteristics. Tw o com
monly used processes are steam or vacuum distillation and air
blowing. In general, air blowing a residuum will result in an as when fluxing hard asphalt with residuum), the blended prod
asphalt having more complex flow characteristics than is ob uct may exhibit more complex flow than straight processed resid
tained by steam or vacuum reduction. This is illustrated by the uum of the same consistency. This is illustrated by the data in
data on asphalts H , N , R , and S prepared from the same East Table I on asphalts T and X . Sample T was prepared by fluxing
Texas residuum (Table I). Steam-refined asphalt R is essen a hard asphalt (softening point, 121.7° C. or 251° F .), air-blown
tially a simple liquid, whereas air-blown asphalt S, of the same from Gulf Coast residuum II, with an equal weight of the parent
consistency, exhibits complex flow with a value for c of 0.85. residuum. Asphalt X was made by air blowing the same resid
Similar relative variations in c are indicated for N and H. uum to a ring and ball softening point of 53.9° C. (129° F.).
In asphalt manufacture many materials are prepared by blend D e g r e e o f P r o c e s s i n g . It appears that the degree o f procees-
ing. If the components are greatly different in consistency (such ing required to cause the appearance of detectable complex
flow varies with the method of manufacture and the source (na
ture) of the material. The effect of extent of air blowing on the
flow characteristics of asphalts is illustrated by the rheological
data in Table I for asphalts A, U, and V which were manufactured
from Gulf Coast residuum I. Asphalts R and N illustrate the
effect of extent of steam reduction on flow characteristics.
JO' Certain types of residua may be steam- or vacuum-refined to
fairly hard asphalts (with consistencies above 2 megapoises) which
are essentially simple liquids at atmospheric temperatures. This
is indicated by the data for asphalts C, 0 , and BB.
The transformation from Newtonian to complex flow as con
sistency increases is not very definite for a given series of as
phalts. Previously (IS) this was indicated to occur at 3.5 to 7.0
megapoises at 25° C .; the more extensive data given in this paper
o show that complex flow may be exhibited by asphalts of much
l IO" lower consistency (D . F, and S, Table I).
o
£ T e m p e ra tu re o f M e a su rem en t. As the temperature of an
asphalt is raised, any complex flow characteristics tend to dimin
ish and may disappear entirely if the temperature is increased
sufficiently. This behavior may be attributed to a gradual
change in the asphalt from a gel to sol structure. Examples of the
effect of temperature are given in Table I on asphalts V and W at
V IS C O M E T E R b/a. 25°, 45°, and 65° C.
/ A g e H a r d e n in g . All asphalts show an increase ih consistency
o 5 1. 5 0 0
* A 5A 1. 2 0 0
with time not caused by loss of volatile components; the magni
A SB 2 .0 0 0 tude of this effect is dependent on the source of the asphalt and
X 5C 1. 71 4 the method by which-it was processed. The phenomenon of age
• 6 1. 5 0 0 hardening is probably a manifestation of the colloidal nature (sol-
gel structure) of asphalt, since the increased consistency occurring
with time may be wholly or partially destroyed b y heat or me
10* 10» chanical working.
M EAN SHEARING STRESS, F , O W E S / SO. CM. The data on asphalts C, I, P, and Q, of widely different origins,
show that increased rates of age hardening (higher values for the
F igure 3, Data O btained in Various R otary asphalt aging index, A A I) are usually associated with higher de
Viscom eters at 25° C. grees of complex flow (lower values of c). The age hardening of
T a b le III. R h e o lo g ic a l E v a lu a tio n s at 25° C. (77° F.)

•Asphalt C ---------------------- * <■---------------------- A sphalt-G ----------------------* --------------- A sphalt E -
Instrum ent N o. 5B 5C 5 5A 6 6B 5C 5 5A 6 5C 5 5A
C on sisten cy at 0.1 s e c .“ 1, megapoisea 2 .3 0 2 .5 0 2 .6 5 2 .5 5 2 .5 5 0 .4 4 0 .4 4 0 .4 9 0 .4 9 0 .4 4 3 .5 4 .0 3 .7 3 .8
D egree o f com p lex flow, c 1 .0 0 1 .0 0 1 .0 0 1 .0 0 1 .0 0 0 .9 0 0 .9 0 0 .9 0 0 .9 0 0 .9 0 0 .8 5 0 .8 5 0 .8 5 0 .8 5
investigators (11, IS, IS), who have shown that the viscosity at
- that temperature may vary from 8000 to 30,000 poises. Funda
/ mental reasons for this variation, with non-Newtonian materials,
/ / were given elsewhere (18).
The data in Table I on asphalts H, N, O, and Y show how the
! / penetration and flow properties may vary for different materials
< io-> / of similar ring and ball softening points when the consistency is
7“ evaluated at 0.1 reciprocal second. Although low consistency in
I
A. ------ v, y - megapoises at 25° C. is associated with high penetration at 25° C..
V js attention is directed to the high degree of complex flow (low
S ' ¿ V value of c) possessed by asphalt H . Asphalt O with a lower pene
v ,
j / tration and high consistency for the same softening point is a
/ viscous liquid at 25° C.
/
/ P e n e tr a tio n T e s t. A t recurring intervals the penetrome
/ ter (S) has been proposed as an apparatus for measuring consist
/ ency in absolute units, but its use for this purpose is highly contro
/ versial. The most serious criticism of the penetrometer method
/ /
can be directed at assumptions concerning the method of shearing
/ which provides the basis for the theoretical equations, such as
7 those of Pendleton (9). The penetrometer needle moves to dis
3 10-«
3x10 10 3x10 place a relatively small volume of fluid. The flow under such
M EA N SHEARING S T R E S S ,F , O YN ES/ SQ . CM. conditions is not the same as in a coaxial cylinder viscometer
Figure 4. R h eological Diagram s at 25° C. from where the displacement of the m oving member results in no
In stru m en t 5B movement of the material at the wall unless slippage occurs.
Furthermore, the theoretical equations make no correction for the
conical depression around the needle. This depression was rec
asphalts was discussed and A A I was defined elsewhere (Id). The ognized by Rhodes and Volkmann (10) and was shown in photo
data were obtained in falling coaxial-cylinder viscometers (17). graphs by Traxler and M offatt (14). Another rheological fault
Five or six viscometers were filled with each asphalt and placed of the penetrometer is the end effect of the needle point passing
in a cabinet maintained at 25° C. At intervals up to about 1000 through the material. The use of an equivalent cylindrical sur
hours a viscometer was removed from the constant-temperature face for the surface of the needle may be a satisfactory assump
cabinet, and several consistency measurements were made at tion for calculating the surface, but it does not correct for the end
25° C. The A A I for each asphalt was obtained b y determining effect. In general, the needle point corrections are minimized by
the slope at 100 hours of the log time vs. log consistency curve: employing greater depths of penetration or other arbitrary cor
rection factors, but these expedients do not correct for the the
Asphalt Q oretical limitations.
AAI 0 .0 1 7 0 .0 2 3 0 .0 7 3 0 .1 8 3 Empirical equations for converting .penetration readings to
1.00 0 .9 5 0 .8 5 0 .5 0
consistency in absolute units are subject to the same objections
as the theoretical equations unless their validity can be proved.
Because of the occurrence of age hardening it is essential that
Thus, the needle point effects were indicated by Fair and Volk
evaluations of consistency and complex flow of asphalts be made
mann (6) to be of sufficient magnitude to limit the applicability
at the same age if values are to be comparable. Consistency is
of the Saal and Koens relation to penetration depths greater than
usually determined one hour after the viscometer is filled with hot
55 for Newtonian materials.
asphalt. During this time the asphalt is brought to the desired
The presence of complex flow further complicates these pene
temperature for testiijg.
trometer equations, with the result that the theoretical and em
pirical equations are validated for only a few asphalts when ade-
R H E O L O G IC A L A S P E C T S O F T E S T S
Various investigators have attempted to evaluate rheological

properties by empirical flow tests such as the A.S.T.M . ring and
T a b le IV. C o m p le x F lo w * M e a s u re d in D iffe r e n t
ball softening point, A. S .T. M . penetrometer at 25° C., and A. S.- A p p a ra tu s
T. M. ductility test at 25° C. These attempts have been unsuc
Falling
cessful for the most part, largely because of the combination of
V iscom eter Viscom eter P enetrom eter <50 Gram s)
flow and other properties measured by each empirical test.
A sphalt M p. c M p. c M p. c° ck
R in g a n d B a l l S o f t e n in g P o in t . The distinctly different
A 1 .1 6 1 .0 1 .8 0 1 .1 7 1 .0 9
flow characteristics of asphalts at their ring and ball softening B 0 .0 4 0 ,9 0^81 o!o 0 .5 0 1.04 1 .0 8
C 2 .6 5 1 .0 2 .7 5 1 .0 7 .5 1 .3 0 1 .1 3
points (4) are illustrated by the data in Table I for samples J and H 6 .8 0 .7 5 5 .0 0 .7 5 3 .4 0 .6 7 0 .6 0
K, prepared by vacuum distillation of a Gulf Coast residuum, and L I 7 .9 0 .5 7 .2 0 .5 5 .8 0 .5 1 0 .5 2
I 7 .9 0 .5 7 .2 0 .5 4 .9 0 .4 7 4 0 .4 9 4
and M, air-blown from a Mexican residuum. These data refute
O 38 1 .0 34 1 .0 2 « 0 .8 9 *
the concept that all asphalts at their ring and ball softening points T 3 .0 0 .7 1 .2 0 0 .5 6 0 .5 2
are simple liquids having the same viscosity. Of course, by U 7 .2 0 .9 5 1 1.3 1.29 1 .0 1
X 5 .4 0 .8 5 1 .7 5 0 .6 7 0 .6 0
judicious selection of the shearing stresses used in the measure y 5 .0 0 .8 5 3 .1 0 .7 7 0 .7 0
ments, it is possible to alter the consistency values o f the complex * C onsistency in m egapoises at 2 5 ° C . and 0.1 s e c ." 1, and degree o f co m
liquids to a value similar to those for asphalts showing Newtonian plex flow c.
a C alculated from c ■» 1 /n — k/( 1 — k), where k is the slope o f log
flow, as illustrated by Figure 4. A t a mean shearing stress of p enetration p lotted against lo^ tim e (sec.), as discussed b y Pendleton ( 9);
corrected penetrations em ploying d ata greater than 100 penetration only
35.000 dynes per sq. cm., the consistencies of asphalts AA and BB were used.
are identical (250 megapoises), but at a mean shearing stress of & C alculated from E qu ation 1, where F and S were com p u ted accordin g
to P en d leton ’s equations (0 ).
100.000 dynes the viscosity of BB is 250 megapoises while that of 100-gram load.
• Im possible to obtain checking results on this m aterial with a 50-gram
AA is only 59. The viscosity of bituminous materials at their lo a d ; results are for 200-gram load. A sphalt did n o t adhere to needle.
ring and ball softening points has been determined by a number of
quate data are ob result from an attem pt to explain the flow properties of asphalts
tained. This con by results of the standard penetration test'(S). They indicate a
clusion is based on general trend of decrease in penetration with increase in consist
data obtained in the ency. However, there are exceptions to this trend as shown by
rotary viscometer asphalts Y and Q. Their low penetration values may be caused
by their relatively high adhesion for the steel needle (as compared
of to asphalt X ) which retards the rate of penetration and gives low
successive penetra results. Other asphalts (see note on asphalt O in Table IV ) mani
tions. Certain of fest little tendency to adhere to the needle, as shown by the fact
the available data that the needle can be withdrawn from the sample with a clean
re s h o w n in surface. Accordingly, thé penetrometer must be considered as
Table IV. measuring a combination of consistency and adhesiveness of as
Although the fall phalt to steel.
ing coaxial-cylinder D u c t ilit y T e s t. Although the A .S.T.M . ductility test (2)
viscometer yields does not evaluate consistency, it has been shown (.18) that duc
log 5 vs. log F plots tility values at various temperatures and at different rates of de
that are parallel to formation are readily explained by the rhcological properties of
those for rotary the asphalt at the temperature of test. The ductility test result
viscom eter data, is actually the length at which a thread of the material breaks be
e lines do not cause the shearing stress at the area of the break exceeds the co
necessarily super hesive strength of the material when it is deformed under per-
impose because scribed conditions. For many soft materials at 25° G. this
and thixo- breaking point is not attained under the conditions of the test,
tropic effects may and high ductility values result; for very hard materials at 25° C.
interfere. T h e s e the break point is attained soon after the start of the test because
complications were the cohesive strength is exceeded before the material can flow ap
noted by Pendleton preciably, with the result that zero ductility values are obtained.
and their pres Between these two extremes there is a range of ductility where
ence precludes the the results vary with different asphalts and the degree of complex
use of data obtai ned flow exhibited by them. This point is illustrated by the data on
th e f al l i n g asphalts H, V, and Y", O (Table I):
al-cylinder A sphalt H V Y O
viscometer for con- D u ctility (2 5 ° C .), cm . 31 38 125 200 +
rm ing other c 0 .7 5 0 .8 5 0 .8 5 1 .0 0
methods of meas In the case of Y and H, with the same penetration and softening
urement on non- point, the decrease in ductility is in direct agreement with the
tonian ma decrease in the value of c. The importance of the degree of
terials. T h e ap complex flow on ductility is also illustrated by the data for
paratus is limited asphalts V and O which have about the same consistency in
to Newtonian fluids m.egapoises and about the same softening point.
or lo comparative An explanation of these results lies in the fact that the materials
studies on complex with the greater degree of complex flow (lower values of c) de
liquids (e.g., age crease more in consistency with increase in shearing stress. As
hardening) where a the cross section of the thread decreases during the ductility test,
strictly arbitrary the shearing stress increases rapidly and induces, a greater flow
procedu re is fol of material. In the case of complex liquids, such as V, the re
lowed. For com gion of lowest consistency is localized at the point of highest
plex liquids widely shearing stress, with the result that “ necking” occurs. This
different consist rapid reduction in cross-sectional area produces low ductility val
ency results can be ues because the shearing stress exceeds the cohesive strength of
obtained, depend the material in a short time after the test is started. Where the
ing upon the shear- material is a simple liquid, such as asphalt 0 , the nature of the
stress and the flow is independent of the shearing stress and necking does not
distance through occur. Such materials will deform readily into a long thread,
which the inner cyl- under the conditions of the test, before the cohesive strength of
movcs. the material is exceeded. This elongation results in high duc
The data in Table tility values. The upper photograph of Figure 5 (taken at an
I for asphalts Q, S, elongation of 11 cm. for asphalts H, V, and 0 ) illustrates the ef
V, X , and Y', having fects of complex flow.
the same degree of
o m p 1e x f l o w ,
illustrate the con T a b le V. C o m p le x F l o w and D u c t ilit y T e s t
fusion which may Asphalt AA BB
Residuum source G ulf C oast II G ulf C oast II I
Process A ir-blow n Steam
R m K and ball, ° C. ( ° F .) 6 7 .2 ( 1 5 3 ) 0 5 .6 (150)
P enetration at 2 5 ° C ., 100 g., 5 sec. 44 10
D u ctility at 25° C „ 5 cm ./m in ., cm . 5 .5 12
F i g u r e 5. C onsistency at 25° C ., S ■=0.1 s e c .* 1, m ega- 6 .0 250
Illustrations poises
Ne» k in g Degree of com plex flow, c 0 .4 5 1.00
Effect
The necking effect for materials with a high degree of complex (2) Am . Soe. for Testing Materials, Standard M ethod of Test for
flcrw, as distinguished from the elongation of the viscous materials Ductility of Bituminous Materials (D 113-39), Part II, p.
466 (1942).
is accentuated in materials of high consistency; the two lower (3) Ibid., Test for Penetration of Bituminous Materials (D o-2 5 ),
photographs of Figure 5 show asphalts AA and BB which have the Part II, p. 483 (1942).
characteristics given in Table V. The center picture was made (4) Ibid., Test for Softening Point of Bituminous Materials (Ring
at an elongation of 3.5 cm., and the lower one was taken after as and Ball M ethod) (D 36-26), Part II, p. 488 (1942).
(5) Barr, G ., “ M onograph on Viscom etry” , London, Oxford Univ.
phalt BB failed. The high consistency of these materials made it Press, 1931.
necessary to use instrument 5B for the rheological data given in (6) Fair, W . F., Jr., and Volkmann, E. W ., I n d . E n o . C hbm .,
Figure 4. A n a l . E d ., 15, 2 4 0 -2 (1 9 4 3 ).
Since elongation under standard test conditions is the only (7) Grant, F. R ., and Pullar, H. B., P toc. A s s o c . Asphalt Paving
Tech., Jan., 1936, 124.
property measured by the ductility test, it is futile to deduce other (8) M ooney, M ., and Ewart, R . H „ Physics, 5, 350-4 (1934).
interpretations from the results. If any fundamental informa (9) Pendleton, W . W „ J. Applied Physics, 14, 170-80 (1943).
tion concerning asphalt is to be obtained from the ductility test, (10) Rhodes, E. O., and Volkmann, E. W ., Ibid., 8, 492-5 (1937).
it is necessary to measure the force required to deform the ma (11) Rhodes, E. O., Volkmann, E. W ., and Barker, C. T ., Am . Soc.
Testing Materials, Symposium on Consistency, 1937, 30-46.
terial as is done in tensile strength tests. This was done by (12) Saal, R . N . J., Proc. World Petroleum Congr., London, 2, 515-23
Abraham (1) and by Grant and Pullar (7) with apparatus of spe (1933).
cial design. It is obvious that extremely sensitive equipment (13) Traxler, R . N., I n d . E n o . C hem ., 30, 322-4 (1938).
would be necessary for materials having ductilities o f 100 cm. or (14) Traxler, R . N., and M offatt, L. R ., I n d . E n o . C hem ., A n a l .
E d ., 10, 1 8 8 -9 1 (193S ).
higher and for those soft materials with low ductility. It is pos (15) Traxler, R . N., Rom berg, J. W ., and Schweyer, H. E „ Ibid.,
sible that even such special tests would be no more informative 14, 3 4 0 -3 (1942).
than the rheological data obtained over a range of shearing (16) Traxler, R . N „ and Schweyer, H. E., Proc. Am . Soc. Testing
stresses in viscometers designed on scientific principles. Materials, 36, 544-51 (1936).
(17) Ibid.. 36, 518-30 (1936).
(18) Traxler, R . N., Schweyer, H. E., and Rom berg, J. W ., Ibid.,
L IT E R A T U R E C IT E D 40, 1182-1200 (1940).
(1) Abraham, H., “ Asphalt and Allied Substances” , 4th ed., New P resen ted at the annual meeting of the Society o f Rheology, in Now York,
York, D . Van Nostrand C o., 1938. N . Y ., 1943.
Thermodynamics of Gaseous
Paraffins
SPECIFIC HEAT AND RELATED PROPERTIES
KENNETH S. PITZER
University o f California, Berkeley, Calif.
F O R several years the writer has been interested in the When this treatment was first presented, the data necessary
specific heat of gaseous paraffin hydrocarbons. The ac to fix certain parameters was inadequate. There is still much to
companying figure and tables present a convenient summary be desired, but considerable improvement has led to the follow
of the present status of results. ing revised values. The vibration frequencies of the CHi group
are now taken as 827, 1170, 1375, 1460(2), 2950(3) cm .-1 (de
N O R M A L P A R A F F IN S generacy), which are the exact ethane values and are sub
stantially those used previously. Although the propane fre
The values for methane (4), ethane (17), and propane (IS)
are based on reasonably definite molecular-structure param quencies (IS) cannot be definitely classified between CH 3 and
eters (vibration frequencies, internal rotation potential bar CH 2 groups, certain frequencies group closely around the CH,
riers, etc.). For the heavier normal paraffins a structural values and leave 940, 1278, 1338, 1460, and 2950(2) cm .-1 to be
picture was assumed which is uniform throughout the series ascribed statistically to the CH 2 group. The use of these fre
and which, evidence now available indicates, is substantially quency sets gives good agreement with the experimental gaseous
correct (10, 12). While approximations are made both in the heat capacity data on the higher paraffins.
basic picture and in the mathematical analysis, their nature is The potential barrier and reduced moment for the end
such as to lead to small and uniform errors which can be com methyl groups of a longer chain are now assigned exactly the
pensated by the use of adjusted structural parameters. propane values (IS), 3400 calories per mole and 4.51 X 10_<0
Available experim ental gas specific-heat values fo r the corresponding results for heat con ten t and the free energy
normal pnraflins are in excellent agreem ent w ith curves fu n ctio n are also presented. T he data for branched paraf
calculated by m ethods previously published by the writer. fins are too m eager to allow generalizations except that the
Certain param eters in these calculations are revised on the change in specific heat w ith isom erization is sm all. En
basis o f recent spectroscopic studies« Calculated entropies tropy differences calculated previously have heen c o n
are still in excellent agreem ent with m easured values. The firm ed, in so far as additional data are available.
60 *
T able I. E n t r o p ie s of N orm al P a r a f f in H ydrocarbons
C n l./° K ./G ra m M ole

y^>?
T. ° K . ¿c a le d . Soxptl.
E thane (19) 298 .16 5 4.8 5 54 .85 ± 0 .1 5
y
Propane (£) 231 .09 6 0 .4 6 6 0 .4 6 * 0 .1 , / A
ü
n-B utane ( / ) 272 .06 7 2 .0 6 72 0 5 ± 0 . 2 50
n-Pcntane (0) 20S.16 8 3 .2 7 8 3 .1 3 * 0 .2
7i-Heptane (11) 3 7 1 .5 111.5 1 1 1 . 7 7 * 0 .3
T a b le II. T h e r m o d y n a m ic P r o p e r t ie s o f G
P a r a f f in H y d r o c a r b o n s
aseous N orm al
40
AA / y "
A
r ,° K ; 298 .16 400 500 600 800 1000 1500 y
M olal H ea t C apacity, G r a m -C a l./° K . Cp
M ethane
E thane
8 .5 3 6
12.585
9 .7 3 6
1 5.6 8
11.133
18.66
12.546
2 1 .3 1
1 5.1 0
2 5 .8 3
17.21
2 9 .3 2
2 0 .7 1
3 4 .9 0
/
Propane 17.57 2 2 .5 4 2 7 .0 4 3 0 .8 4 3 7 .0 8 4 1 .8 3 4 9 .2 6 30 cf
Butane 2 3.6 1 2 9 .8 0 3 5 .5 4 4 0 .3 2 4 8 .2 3 5 4 .2 0 63 51
Pentane 2 9 .3 0 3 6 .9 1 4 3 .9 6 4 9 .7 6 5 9.3 7 6 6 .5 7 7 7 .7 6
Hexane 35 06 4 4 .0 4 5 2 .3 9 5 9.2 1 7 0 .5 1 7 8 .9 4 9 2 .0 2 X
Heptane 40 82 5 1 .1 8 6 0 .8 3 6 8.6 6 8 1 .6 5 9 1 .3 1 106.27
O ctane 4 6 .5 8 5 8.3 1 6 9 .2 7 7 8.1 1 9 2 .7 9 103.68 120.52
A per CH* 5 .7 6 0 7 .1 3 4 8 .4 3 8 9 .4 5 11.14 12.37 14.25
E n trop y, G r a m -C a l./0 K ./G r a m M ole 20

M ethane 4 4 .5 0 4 7 .1 6 4 9 .4 8 5 1 .6 4 5 5.61 59..21 66. 93
E thane 5 4 .8 5 5 8 .9 8 6 2 .7 9 6 6 .4 2 7 3 .2 3 79.,39 92. 45
Propane 64.51 7 0 .3 7 7 5 .8 9 8 1.1 3 9 0.9 5 99..77 118. 28
Butane 7 4 .1 0 8 1.9 1 8 9 .2 0 9 6 .0 1 108.87 120..31 144.,22
Pentane 8 3 .2 7 9 2 .9 5 101.93 110.33 126.21 140.,27 169.,59
Hexane 9 2 .4 5 104.03 114.73 124.73 143.60 160 ,30 195..03
Heptane 101.64 115.10 127.53 139.12 161.00 180,.32 220, 45 10
Octane 110.82 126.18 1 40.34 153.52 178.40 200, 35 245..88 300 400 500 600
A per CH* 9.1 8 6 11.075 12.803 14.396 17.40 20,.03 25.,43 T,°K.
H eat C ontent ( H £ - H°0), K g .-C a l./G r a m M ole
Figure 1. M olal Heat Capacity o f Gaseous Norm al
M ethane 2 .3 9 7 3 .3 2 3 4 .3 6 5 5 .5 4 9 8 .3 2 1 11 .56 21..13 Paraffin H ydrocarbons from Propane to Heptane
Ethane 2 .8 5 6 4 .2 9 5 6 .0 1 0 8 .0 0 3 12.75 18 .28 34,.50
Propane 3 .5 1 2 5 .5 5 8 8 .0 4 0 10.925 17.77 25 .67 48 64 K is tia k o w a k y a n d R ic e (7 ); O B a ile y a n d F e lsin g (3)-.
Butane 4 .6 4 6 7 .3 7 0 10.636 14.41 2 3 .3 4 33 .58 63 .27 9 P itz e r (12, 2 2); ¿ E u c k c n a n d S a r s te d t (5 ).
P entane 5 .6 6 4 9 .0 4 6 13.081 17.73 2 8 .7 4 41 .34 77 .76
Hexane 6 .6 8 9 10.7 30 15.540 2 1 .0 8 3 4 .1 6 49 .12 92 .25
Heptane 7 .7 1 3 12.414 17.998 2 4 .4 2 3 9 .5 9 56 .89 106 .74
O ctane 8 .7 3 7 14.0 98 20.4 56 2 7 .7 7 45.01 64 .66 121 .23
A per CH* 1 .024 1 .6 8 4 2 .4 5 8 3 .3 4 6 5 .4 2 3 7 .77 14 .49 molal heat capacity curves are shown in Figure 1 together with
F ree E nergy F un ction s, (I /q — F f ) / T , G r a m -C a l./0 K ./G r a m M ole experimental results. Table II also gives the specific heat over
the range 298° to 1500° K . or 77° to 2240° F. The accuracy of the
M ethane 3 6 .4 6 3 8 .8 6 4 0 .7 5 4 2 .3 9 4 5 .2 1 4 7 .6 5 5 2 .8 4
E thane 4 5 .2 7 4 8 .2 4 50 77 5 3 .0 8 5 7.2 9 6 1.1 1 6 9.4 6 results can be judged from Table I and Figure 1. The specific
Propane 5 2 .7 3 5 6 .4 8 5 9.8 1 6 2 .9 3- 6 8 .7 4 7 4 .1 0 8 5 .8 5 heat curves should not err by more than 2 % except at the highest
Butane 5 8.51 6 3 .4 9 6 7 .9 3 7 2 .0 0 7 9 .7 0 8 6.7 2 102.03
P entane 6 4 .2 6 7 0 .3 3 7 5 .7 7 8 0 .7 8 9 0 .2 8 9 8.9 2 117.75 temperatures and are probably more accurate than this.
Hexane 7 0.0 1 7 7 .2 0 8 3 .6 5 8 9 .6 0 1 00.90 111.18 133.52
H eptane 75 76 8 4 .0 6 9 1 .5 4 9 8 .4 2 111.52 123.43 1 49.29 The values are based on the thermochemical defined calorie of
O ctane 8 1.5 1 9 0 .9 3 9 9 .4 2 107.24 122.14 135.69 1 65.06 4.1833 international joules and R = 1.98718 calories per degree.
A per CH* 5 .7 5 6 .8 6 5 7 .8 8 5 8 .8 2 1 0.62 1 2.2 5 15.77
Otherwise in so far as the difference is significant, the physical
S pecific H eat, G ram -C al./® C ./G ra m or B .t.u./® F ./L b . constants of Birge {2) (1941) are used rather than those from
T, 0 K . 298 .16 400 500 600 800 1000 1500 International Critical Tables used heretofore.
(, ° F. 7 7 .0 0 2 6 0 .3 4 4 0 .3 6 2 0 .3 9 8 0 .3 1340.3 2 2 4 0 .3
M ethane 0 .5 3 2 0 .6 0 7 0 .6 9 4 0 .7 8 2 0 .9 4 1 1.073 1.291
E thane 0 .4 1 9 0 .5 2 1 0.621 0 .7 0 9 0 .8 5 9 0 .9 7 5 1.161 B R A N C IIE D -C H A IN P A R A F F IN S
Propane 0 .3 9 8 0.511 0 .6 1 3 0 .6 9 9 0 .8 4 1 0 .9 4 9 1 .117
Butane 0 .4 0 6 0 .5 1 3 0 .6 1 2 0 .6 9 4 0 .8 3 0 0 .9 3 3 1.093
Pentane 0 .4 0 6 0 .5 1 2 0 609 0 .6 9 0 0 .8 2 3 0 .9 2 3 1 .078 Both experimental data and theoretical development for
Hexane 0 .4 0 7 0.511 0 .6 0 8 0 687 0 .8 1 8 0 916 1 .068 branched-chain paraffins are less complete than for the straight
H eptane 0 .4 0 7 0.511 0 .6 0 7 0 .6 8 5 0 .8 1 5 0 .9 1 1 1.061
O ctane 0 .4 0 8 0 .5 1 0 0 .6 0 6 0 .6 8 4 0 .8 1 2 0 .9 0 8 1.055 chains. The entropies (and free energy functions) calculated in
Coo 0 .4 1 1 0 .5 0 9 0 .6 0 2 0 .6 7 4 0 .7 9 4 0 .8 8 2 1 .016
1940 have been confirmed by all experimental values obtained
since. M ost notable was the value of Schumann, Aston, and
Sagenkahn {16) for isopentane which agrees exactly with the
calculated entropy value of 82.0 calories per ° K . at 298.16° K.
gram cm .5 The potential barrier for skeletal internal rotations D ata were also obtained for 2,3,4-trimethylpentane {14) with
is determined, as before, from the experimental entropy values. satisfactory agreement.
These revisions shift this parameter from 3600 to 3300 calories I t now appears that the heat capacities implied b y the 1940
per mole. Although a value smaller than the 3400 calories per tables are somewhat too high for branched-chain paraffins just as
mole o f the end methyl groups may seem improbable, it should be they are for the normal series. W ithout doubt the cause is the
remembered: (o) that this is an adjusted parameter which ab same— namely, the use of too low values for the hydrogen-
sorbs all inaccuracies of independently obtained parameters and bending vibration frequencies. However, the spectroscopic
approximations in the method of calculation; and (6) that the studies have not yet progressed to the point where reliable sets
steric repulsion term which is independently introduced in this of frequencies can be assigned (even in a statistical sense) to the
method effectively raises the barriers in all actual cases. Ex various groups of atoms. N or are there enough reliable gas heat
cept as stated above, the parameters and formulas given in 1940 capacity data to allow generalizations. Available data show
{10) are used in the present calculations. isobutane (8), isopentane (5), and neohexane (5) to have molal
Table I compares the observed and calculated entropies of heat capacities a fraction of a calorie per ° K . lower than their
gaseous normal paraffins. Table II presents the calculated heat normal isomers; the four trimethylpentanes {8) all have slightly
capacity, entropy, heat content, and free energy functions. The greater heat capacity than iV-octane.
L IT E R A T U R E C IT E D
Thus it appears that the best available values for the dif
ferences in entropy or in the free energy function between iso (1) Aston, J. G ., and Messerly, G . H ., J. A m . Chem. Soc., 62, 1917
mers are still those published b y the writer (10) in 1940; the (1940).
(2) Birge, R . T ., Rev. Modern Phys., 13, 233 (1941).
absolute value of the heat capacity of branched paraffins not yet
(3) D ailey, B. P., and Felsing, W . A ., J. A m . Chem. Soc., 65, 42, 44
studied experimentally may best be assumed to be the same as (1943).
the normal isomer. In particular there is no need to revise the (4) Dennison, D . M ., Rev. Modern Phys., 12, 175 (1940).
calculations of isomerization equilibria made by Rossini, Prosen, (5) Eucken, A ., and Sarstedt, B., Z . physik. Chem., B50, 143 (1941).
(6) Kem p, j . D ., and Egan, C. J., J. Am . Chem. Soc., 60, 1521 (1938).
and the writer (15).
(7) Kistiakowsky and Rice, J. Chem. Phys., 8, 610 (1940).
N o attem pt is made in this brief note to compare these results (8) Loininger, R . F., unpublished measurements, Univ. Calif.
with the numerous empirical or semitheoretical calculations (9) Messerly, G . H ., and Kennedy, R . M ., J. Am . Chem. Soc., 62,
which have been published nor with old experimental values 2988 (1940).
where more accurate ones exist. The recent papers of Stull (10) Pitzer, K . S., Chem. Rev., 27, 39 (1940); J. Chem. Phys., 8, 711
(1940).
and Mayfield (18) give many references to such work. (11) Pitzer, K . S., J . Am . Chem. Soc., 62, 1224 (1940).
(12) Ibid., 63, 2413 (1941).
A CK N O W L E D G M ENT (13) Pitzer, K . S., J. Chem. Phys., 12, 310 (1944).
(14) Pitzer and Scott, J. A m . Chem. Soc., 63, 2419 (1941).
The results of the present paper are being combined with (15) Rossini, F . D ., Prosen, E. J. R ., and Pitzer, K . S., J. Research
recent heat of combustion values of Prosen and Rossini for the Natl. Bur. Standards, 27, 529 (1941).
normal paraffins to give free energies of formation and related (16) Schumann, S. C., Aston, J. G ., and Sagonkahn, M ., J . Am.
Chem. Soc., 64, 1039 (1942).
data, in connection with Research Project 44 of the American
(17) Stitt, F., J. Chem. Phys., 1, 297 (1939).
Petroleum Institute, at the National Bureau of Standards. The (18) Stull, D . R ., and M ayfield, F . D ., I n d . E n o . C h e m ., 35, 639,
writer wishes to thank E. J. R . Prosen for carrying out some of 1303 (1943).
the calculations here presented. (19) W itt, R . K ., and Kem p, J. D ., J . Am . Chem. Soc., 59, 273 (1937).
CAROTENE CONTENT OF ALFALFA

Retention on
Dehydration and Storage
Various treatm ents are reported w h ich stabilize the caro RALPH E. SILKER, W . G . SCHRENK,
tene in alfalfa during the process o f drying. B lanching AND H. H . KING
the fresh alfalfa w ith steam , prior to drying, furnishes
Kansas Agricultural Experiment Station, Manhattan, Kans.
com plete protection for the carotene, and considerable
protection is afforded w hen fresh ground alfalfa was
treated with certain chem icals before it was dried. Tw o
types o f chem icals are used— nam ely, antioxidants and was avoided. (The effect of certain metals will be reported in a
substances w hich are know n to inactivate enzym es. later paper.) The dehydrated alfalfa was ground in a Wiley mill
Diphenylam inc and h ydroquinon e are th e m ost effective using a 2-mm. mesh screen; samples were taken for analysis, and
o f the first type; thiourea and sodiu m cyanide arc m ore the remainder was stored in glass jars with tight-fitting covers.
effective than any oth er substances tested o f the second A greater loss of carotene accompanies this slow method of de
type. T he carotene con ten t o f alfalfa m eal decreases as hydration than that of the rapid commercial dehydrators, but
the tem perature o f storage is increased. However, essen the different samples of a given experiment which are compared
tially n o change has been n oted in th e carotene con ten t were subjected to the same conditions.
of alfalfa m eals stored at 3° C.
' M E T H O D O F A N A L Y S IS
M O RE and more attention is being directed toward the The method used for determination of carotene was essentially
retention of natural nutritional constituents in feeds as that of W all and Kelley (18), which is a modification of the
well as in foods since the practice of fortifying them method of M oore and Ely (10). Duplicate samples were run
with various food concentrates has come into common use. The except for a short period when help was not available. Five
primary purpose of this study on the carotene content of alfalfa grams of the fresh material (stemmy portions were broken into
is to find a practicable method to stabilize this precursor of vi small pieces to avoid excessive splashing when the Blendor was
tamin A during dehydration and storage. started) was placed in .a Waring Blendor with 60 ml. of a 1-1
The alfalfa used was stripped from the plants in the field. In Skellysolve B -alcohol mixture, and sufficient alcohol was added to
this way, principally leaves were collected since the bulk of the form a foaming mixture. The moisture content of the alfalfa
stems were avoided, and more nearly uniform samples were avail determines the amount of alcohol needed; 125-150 ml. were
able for analysis. Stripping the plants has an additional ad usually required. Blending was carried out for 10 minutes, and a
vantage in that the carotene content of the material is much nearly colorless pulp remained on filtration. The Blendor and
greater than if the whole plant is used, and thus the effect of a pulp were carefully washed with Skellysolve B, and the extract
particular treatment can be noted more readily. and washings were treated with 100 ml. of water to remove most
The dehydration of the various samples was carried out in of the alcohol and cause the Skellysolve B layer to separate.
a Freas circulating air oven at 65° C. The fresh and blanched Addition of a small am ount of anhydrous sodium sulfate to the
samples were placed on towels on trays so that contact with metal extract eliminated the formation of an emulsion. After separa-
A dehydrated sample, usually 1 gram, was reconsti

tuted by allowing it to stand in a small beaker with
5 to 7.5 ml. of water for 10 minutes before blending.
The volume of alcohol needed to form a foam
ing mixture was usually 90 to 100 ml. The method
gave satisfactory checks and avoided, to a large
degree, the isomerization of /3-carotene to the
“ neo-/S-carotene” of Beadle and Zscheile (2). The
method of calculating the percentage of isomeriza
tion of /3-carotene assumes the presence of only
two pigments, and neglects, of course, the pres
ence of the additional isomer which Kemmerer
and Fraps (7) reported since this study was under
taken. The carotene values obtained when refiux-
ing is used are usually somewhat higher than those
with other methods unless special precautions are
taken (17) This is due to the presence of noncar-
otenoid pigments which M oore (0), Fraps and Kem
merer (6), Wiseman et al. (20), and others have
Figure 1. Effect o f B lan chin g, D ehydration, and Storage on
C arotene C ontent o f Alfalfa pointed out. Passing the carotene extract ob
F •» fr e s h ; F B *» fre s h b la n c h e d ; D “ d e h y d r a te d ; A 513 b la n c h e d ; O - tained b y the Peterson, Hughes, and Freeman
c o n t r o l. A y By C s to r e d a t 3 ° C .; D s to r e d a t r o o m te m p e r u tu r e .
method (13) through a magnesia column gave
values that compared favorably with those reported
here, but the percentage of /3-carotene was usually slightly lower
T a b l e I. E f f e c t o f B la n c h in g o n C a r o t e n e C o n t e n t o f (17). The values obtained with soda lime as adsorbent, accord
A lfa lfa
ing to Kernohan (3), were also high since some of the xantho
Total Carotene, M g./lOO G. Alfalfa (Dry W t.)
phylls were eluted with the carotene. A simplified method,
Sample Blanching Fresh, Fresh, Blanched,
No. Time, Min. Fresh dehydrated blanched dehydrated which avoids refluxing or the use of a Waring Blendor in ex
300 0 4 5 .3 tracting carotene from dehydrated samples, has been described
300 5 4 7 ‘. 0 46. G
(17).
456 0 4 3 .7 3 4 .4 Moisture samples were taken at the time of the carotene de
456 10 4 2 13
termination so that all results are reported on a dry-weight basis.
405 0 5 3 .0
405 5 5 5 .8 5 8 .8 It was most convenient to use an air oven at 105° C .; although
405* . 5 5 5 .5 5 3 .7 the values were not the same as if a vacuum oven had been used,
321 0 3 9 .0 16.8 results are comparable.
321 7 4 5 .7 3 9 .2
321 10 4 2 .9
E F F E C T O F B L A N C H IN G
436 0 4 8 .9 2 8 .6
436 7 4 3 ’. 9
The value of blanching certain vegetables with hot water be
438 0 4 4 .0 2 3 .7 fore home canning has been recognized for many years. The
433g** 7 35 *.4
advent o f quick frozen and dehydrated foods has been responsible
471 0 3 0 .3 1 7.8
471 7 2s ! o for much research on the length of time, on various techniques,
* Ground after blanching. ** Ground before blanching. \
tion, the water layer was extracted three times with

35-ml. portions of Skellysolve B. The combined extracts
were then concentrated on a steam cone, dried with an
hydrous sodium sulfate, and finally drawn through a column
containing a 2 to 1 mixture of H yflo Super-Cel and mag
nesia (M icron brand 2641) to separate the carotene from
the chlorophylls and xanthophylls. Heating is one of the
methods which Polgdr and Zechmeister (14) used to isomer-
ize /3-carotene, but this change was largely avoided if the
carotene extracts were not concentrated below 30-40 ml.
The columns, approximately 75 mm. in length, were pre
pared by adding the adsorption mixture under suction. A
layer of anhydrous sodium sulfate (about 1 cm.) was placed
above the adsorbent, and the column washed with Skelly
solve B before the extract was poured in. A column may
be used for several separations if it is carefully washed with
Skellysolve between samples. The carotene was eluted
from the column with a 4 % acetone-Skellysolve B mixture.
The eluate was made up to a volume of 250 ml. (100 ml.
for the low-carotene samples), and the optical densities
Figure 2. Effect o f G rinding, D ehydration, C hem ical
were measured at 436 mg and 478 m g with a Beckman (S) T reatm en t, and Storage at 3° C. on C arotene C ontent o f
spectrophotometer. The two readings were made so that A lfalfa
it would be possible to calculate the percentage of /3-caro- F ■ fr e s h ; G =* fre s h g r o u n d ; D *= d e h y d r a te d ; A = t r e a te d ; O “ c o n tr o l.
A *■ t h io u r e a ; B «*» s o d iu m c y a n id e ; C =• d ip h e n y la m in e ; I) h y d ro-
tenc according to the method of Beadle and Zscheile (2). q u in o n c
takes place and is correlated with the conversion

T a b le II. E f f e c t o f G r in d in g a n d D e h y d r a t io n o n C a r o t e n e C o n t e n t of chlorophyll to pheophytin.
of A lfalfa
A sticky film which coated the alfalfa was formed
Fresh, Fresh, Loss on Fresh, D e- Loss on G rou n d . D e Loss on
Sample Stripped, G round, Grind- hydrated, D eh yd ra hydrated, D eh yd ra during the process of blanching. Reeve (15) sug
No. M g .« M g .« ing, % M g .« tion, % M g .« tion, % gests that retention of carotene in dried sweet po
350 4 1 .5 3 5 .7 1 3 .9 2 4 .8 4 0 .2 16.6 5 3 .5 tatoes is due to a sealing effect exerted by the gelled
360 6 0 .6 5 6 .0 7 .6 2 7 .3 5 5 .0 11.3 7 9 .8
370 5 4 .8 4 6 .5 1 5 .2 1 7.8 6 1 .8 starch. H e also pointed out (16) that blanching of
380 5 3 .2 4 2 .3 2 0 .8 12.5 7 0 .4
390 5 2 .4 27\ 3 47] 9 1 2.8 spinach liberates an oil from the lipo-protein which
435 4 8 .9 44'. 0 1CL0 2 7 .5 4 3 .8 2 3 .7 4Q.2 dissolves and preserves the carotene. It seems
470 3 0 .3 2 4 .8 18.2 1 7 .8 4 1 .3 12.3 5 0 .5
probable that the above-mentioned film which hard
0 T otal carotene, mg./IOO grama alfalfa (dry-w eight baals).
ens during dehydration serves the same function
with alfalfa. In any event, extraction studies have
T a b le III. E ffect of A dded C h e m ic a l s o n C arotene C ontent shown that it is more difficult to remove the caro
T otal Carotene, M g ./lO O G. tene from blanched alfalfa meal than from the regu
Fresh S tored« for:
Sample Chem ical A dded A lfalfa 0 1 m o. 2 m o. 3 m o. 4 m o. 5 m o. 6 m o.
lar meal. Figure 1 points out the typical changes
353 N one N one 1 6.4
in carotene values during blanching, dehydration,
1 5.6 1 6.2 1 6.0 1 5.6
353 Urea 0 .5 1 1.0 9 .8 9 .4 9 .5 and storage.
353 Thiourea 0 .5 2 3 :3 2 4 .7 2 3 .a 2 3 .4 2 3 .4
353 A cetam ide 0 .5 9 .5 9 .1 9 .7 9 .6
372 N one N ono 1 7.8 15.1 17.9 1 5.3 16.4 E F F E C T O F G R IN D IN G
372 Thiourea 0 .5 2 8 .4 2 6 .4 2 7 .7 2 7 .5 2 7 .4
372 Thiourea 0 .2 5 2 7 .8 2 6 .4 2 6 .4 2 5 .7
372 Thiourea 0 .1 2 5 2 5 .7 2 3 .6 2 4 .3 2 3 .3
The grinding of fresh alfalfa caused a decided loss
372 Thiourea 0 .0 6 25 2 9 .1 2 8 .2 2 8 .0 2 6 .5 of carotene (Table II). The loss may be accounted
382 N one N one 1 2.5 1 2.2 13.2 1 2.5 for by the increase in activity of the enzyme sys
382 D iphenylam ino 0 .5 2 0 .6 2 0 .6 2 2 .4 2 2 .2 (. tems and the more favorable conditions for oxi
332 Olive oil 0 .5 1 1.2 1 1.3 11.1 1 2.3
382 Mineral oil 0 .5 1 1.9 1 0.4 9 .0 1 0.7 dation. Some of the samples were analyzed im
438 N one N one 2 3 .7 2 4 .6 2 4 .3 2 4 .6 mediately after grinding; others were permitted to
433« N a cyanide 0 .5 3 6 .2 3 0 .9 3 2 .9 3 2 .1
438 N a cyanide 0 .2 5 3 2 .1 2 9 .7 3 0 .8 3 0 .7 stand at room temperature during the time needed
438 N a cyanide 0 .1 2 5 3 3 .2 2 6 .1 2 7 .4 2 6 .1 to treat other portions with chemicals. This, un
438 Na cyanide 0 .0 6 25 3 1 .6 3 0 .9 3 0 .3 2 9 .3
doubtedly, explains the large differences in the per
473 N one N one 1 2.3 1 2.7 1 6 .5 16.1 1 7 .0
473 H ydroqu in one 0 .5 2 1 .8 2 4 .1 2 2 .9 2 2 .9 2 3 .5 centage loss fcaused b y grinding. The loss of caro
1 A t 3 ° C . for approxim ate times in m onths. tene on dehydration of a ground sample was
________ usually much greater than the loss for the corre
sponding fresh sample. A continuation of the
increased enzyme activity which accompanied grind
and, of greatest importance, on the effects produced b y blanching ing and the greater opportunity for oxidation also seem probable
different types of raw foods before processing. Blanching in hot explanations for these differences. The lack of uniform losses on
water has been very satisfactory for most fruits and vegetables, dehydration can be explained since no attem pt was made either
but blanching with steam has been found more advantageous to use the same quantity of alfalfa in the different runs or to re
for most leafy vegetables. An autoclave was used as a steam m ove it from the oven after a given time.
cabinet in treating various samples. Table I shows that, in each
case where the blanched sample was analyzed immediately after E F F E C T O F C H E M IC A L S
the blanching process, the carotene content was somewhat higher
than that of the fresh material. This might be explained as due T o ensure most uniform distribution, the alfalfa was ground
to errors involved in sampling or of analysis, but appears to be through a food chopper before being mixed with the chemical.
significant in view of the values obtained for the blanched and T he loss of natural juices from the treated sample was avoided by
dehydrated samples. Sample 456, which was blanched and de allowing the mixture to dry partially in a large evaporating dish in
hydrated, gave a slightly lower value than that of the control; the circulating air oven before spreading it on the towels which
but it was allowed to stand at room temperature for 10 days be covered the oven trays.
fore being analyzed and, in view of findings (Figure 3), would be ex M ost of the substances added to the ground alfalfa were
pected to lose some of its carotene. Sample 438g was ground selected because of their known characteristics. Some of them
through a food chopper before blanching; in view of the losses have been used in food technology to inactivate enzyme systems;
due to grinding of fresh alfalfa which are reported later in this others are classed as antioxidants. The results are shown in
paper, 438g would be expected to behave differently from the Table III. Denny (4) pointed out that thiourea is capable of
unground samples. Sample 471 had been severely frosted in the inactivating the browning systems in peaches, pears, and apples.
field and undergone some changes before blanching. A red band, I t was found to stabilize the carotene when mixed with ground
not previously noted in our work, was always seen on eolumniza-
alfalfa, and the effectiveness was relatively constant over a wide
tion of an alfalfa extract from material that had been frosted,
range of concentration. Figure 2A shows that, even after
but time has not permitted us to study the changes which have
several months in storage, the carotene content of sample 372,
been suffered by the pigments. The other exception where the
which contained 0.5 gram thiourea per 10 grams alfalfa, was ap
value on the blanched and dehydrated material was less than
proximately 7 0 % greater than that of the corresponding un
that for the fresh material was sample 436; this definitely appears
treated sample. W ith one exception, the samples with a lower
to be in error since all of the later analyses of the stored sample
(Table IV ) were higher. The rest of the blanched and de concentration of thiourea give almost identical curves.
hydrated samples contained more carotene than the correspond The ability of cyanides to inactivate enzymes has been known
ing fresh one, in spite of the small loss which accompanied the for a long time. Figure 2B shows the stabilizing effect exerted by
dehydration of all analyzed samples except one. the highest concentration of sodium cyanide used. Less effect
The blanching process altered the pigments in alfalfa and was noted when smaller amounts were used, but all concentra
caused a very deep green color to develop. D utton, Bailey, and tions of sodium cyanide produced an intense green color in the
Kohake (5) showed in studies on spinach that a similar change alfalfa which persisted after dehydration. Pepkowitz (12)
time. However, several similar irregularities have

T a b le IV. E ffect of Sto rao e on C aroten e C ontent of B lan ch ed
since been noted, and it no longer appears that
and Unblanched A lfalfa
errors in analytical procedure were responsible.
Sam ple Storage T otal C arotene, M g./lO O G . A lfalfa (D r y W t.) after Storage for: 1
N o. T em p ., ® C.
The dehydrated samples which showed this
0 mo. 1 m o. 2 m o. 3 m o. 4 mo. 5 m o. 6 m o.
302* R oom 4 6 .6
anomaly had stood at room temperature for several
2 1 .6 1 9 .7 1 8 .8 1 7.1
466 R oom 3 4 .4 29 !o 27.2 2 3 .5 2 3 .1 2 1 .3 days before the initial analysis; apparently they
456* R oom 4 2 .3 3 1 .4 2 8 .4 2 4 .8 2 2 .8 1 9.2
405* 3 6 8 .8 5 1 .2 5 6.9 5 7 .1 5 4 .8 had undergone some change, possibly isomerization,
405** 3 5 3 .7 5 0 .6 5 0 .2 4 9 .0 4 6 .6 which reversed itself on storage at the low tem
321 3 16.8* 1 3.5 1 7 .0 1 4.6 17.1 17.5
321* 3 3 9 .2 4 0 .3 4 3 .0 4 2 .2 4 3 .6 4 1 .0 perature.
321* 3 4 2 .9 4 0 .3 4 1 .6 4 1 .3 4 0 .6 4 0 .0
430 3 2 8 .6 28^8 3 2 .7 3 3 .6 3 1 .7 The effect of different storage temperatures on
436* 3 4 3 .9 4 5 .2 4 9 .8 4 7 .1 4 7 .3 blanched and unblanched alfalfa meal is shown
438 3 2 3 .7 2 0 .1 2 4 .6 2 4 .3 2 4 .6 23! 4
438g*** 3 3 5 .4 3 2 .3 3 3 .1 2 9 .6 2 6 .8 2 4 .1 in Figure 3. These data were obtained with sample
471 3 1 7.8 2 1 .0 2 3 .2 2 2 .6 2 3 .0
471* 3 2 5 .0 2 8 .6 3 0 .9 3 1 .4 2 9 .2 number 321 which had been blanched 7 minutes
*
* Blanched with stehm before drying. and its control over a period of nearly 2.5 months.
** Blanched with steam, then ground before drying.
*** G round before blanching. The carotene values at the storage temperature of
_____ 3° C. are some lower than those noted in Table IV.
The reason.is that moisture determinations were
not made in this special study, and the values
recently pointed out that the addition of sodium cyanide, im plotted in Figure 3 represent milligrams of carotene per 100
mediately before extraction of carotene from raw vegetables, grams of sample rather than per 100 grams of alfalfa on the dry
partially inhibits the photochemical destruction of the pigment weight basis.
and prevents its enzymatic destruetion in the extract.
Diphenylamine (Figure 2C) had about the same stabilizing C O N C L U S IO N S
effect on the carotene in alfalfa as did thiourea. Williams,
. Inactivation of the enzymes by blanching furnished complete
Bickoff, and Van Sandt (19) studied the behavior o f about a hun
protection for the carotene in alfalfa during the process of de
dred antioxidants with pure carotene, which was incorporated
hydration. Addition of a suitable antioxidant or additidn of a
with white mineral oil and the stabilizer in oil-free rice bran.
chemical which is known to suppress enzyme activity was found
T hey report that diphenylamine had the greatest stabilizing ef
to furnish partial protection for the carotene during the same
fect of all the substances used.
Alfalfa which had been treated witfc hydroquinone contained,
after 4-month storage, about 4 0 % more carotene than the cor
responding untreated sample. This would indicate a lower
protection of the carotene than that furnished by diphenylamine.
However, as comparison of C and D (Figure 2) shows, the loss which
accompanied the dehydration process was much less when hydro
quinone was used as stabilizer. (All other samples, except those
which were blanched, exhibited an appreciable loss during de
hydration.) It is unfortunate that different portions of the same
alfalfa were not treated with hydroquinone and diphenylamine.
If this had been done, we would know whether the differences
were real or whether the conditions during dehydration were
responsible for the apparently anomalous results. Morgal,
Byers, and Miller (11) report the stabilizing effect o f hydro
quinone toward solutions o f crystalline carotene in Wesson oil.
Williams, Bickoff, and Van Sandt (19) also showed that the caro
tene in specially prepared pellets is stabilized to a considerable F igure 3. Effect o f Storage T em perature on Carotene
extent by hydroquinone. C ontent o f Alfalfa
It was thought that coating of the alfalfa with an oil might O - 3 °( □ - 1 5 °; A - 2 6 ° i X - 3 7° C.
impede the rate at which the carotene was oxidized. Neither
light mineral oil nor olive oil had any appreciable effect, how
ever. process. Thus at least part of the carotene loss appears to be
Both urea and acetamide had a deleterious effect on the caro due to the synergistic effect of some enzyme and some natural
tene in alfalfa. In fact, there was about 67 % more carotene oxidizing agent in the alfalfa.
in the untreated sample than in those which were treated with Storage at 3° C. of blanched and dehydrated alfalfa protected
either of these chemicals. the carotene for long periods. Loss of carotene took place,
however, when the meal was removed from cold storage and al
EFFECT O F STORAGE TEM PERATU RE lowed to stand at higher temperatures. Ball (1) pointed out
Table IV shows the effect of storage on blanched and un that certain enzymes regain their activity after the blanched
blanched samples. W ith a single exception, a temperature of material is cooled. It seems doubtful whether this is an ex
3° C. held the carotene values at the same level that existed when planation for our results, but our enzyme studies must be ex
they were placed in storage. Sample 438g was the only one tended before a definite statement can be made.
stored at this temperature which showed any appreciable loss. Further, it has been shown that treated alfalfa meals suffered
As pointed out previously, this sample was ground prior to the no additional loss of carotene when held near the freezing tem
blanching and probably should have been subjected to a more perature. Present studies must be supplemented to determine
severe heat treatment. behavior of meals at higher temperatures.
B oth blanched and unblanched samples 471 showed lower L I T E R A T U R E C IT E D
values when they were placed in storage than at any subsequent
(1) Ball, A. K ., Proc. Inst. Food Tech., 1943, 165.
time. A t first it was believed that an analytical error had been (2) Beadle, B. W ., and Zscheile, F . P., J . Biol. Chem., 144, 21
made since it was not possible to run duplicate samples at that (1943).
(3) Cary, H. H ., and Beckman, A. O ., J. Optical Soc. A m ., 3 1 , 692 (14) Polgiir, A ., and Zechmeister, L., J. Am . Chem. Soc., 64, 1856
(1 9 4 1 ). (1942).
(4) Denny, F. E ., Contrib. Boyce Thompson Inst., 12, 3 0 9 (1 9 4 2 ). (15) Reeve, R . M ., Food Industries, 14, N o. 12, 51 (1942).
(5) D utton, H . J., Bailoy, G . F., and Kohako, Eleanor, I n d . E n g . (16) Reeve, R . M ., Food Research, 8, 137 (1943).
C h e m ., 3 5 , 1173 (1 9 4 3 ). (17) Silker, R . E., Sehrenk, W . G ., and King, H. H ., I n d . E n g .
(6) Frapa, G . S., and Kommerer, A. R ., Assoc. Official A gr. Chem., C h e m ., A n a l . E d ., 16, 513 (1944).
2 2 , 190 (1 9 3 9 ). (18) W all, M . E „ and Kelley, E. G „ Ibid., 15, 18 (1943).
(7) Kemmerer, A . R ., and Fraps, G . S., I n d . E n g . C h e m ., A n a l . (19) Williams, K . T ., BickofT, Emanuel, and Van Sandt, Walter,
E d ., 15, 7 1 4 (1 9 4 3 ). J. Biol. Chem., 91, 105 (1931).
(8) Kernohan, George, Science, 90, 623 (1939). (20) Wiseman, H. C., Kane, E. A ., Shinn, L. A., and Cary, C. A .,
(9) M oore, L . A., I n d . E n g . C h e m ., A n a l . E d ., 12, 726 (1940). J. A gr. Research, 57, 635 (1938).
(10) M oore, L. A ., and Ely, R ay, Ibid., 13, 600 (1941).
(11) Morgal, P . W ., Byers, L . W ., and Miller, E . J., I n d . E n g . P r e s e n t e d b e fo r e th e D iv is io n s o f B io lo g ic a l a n d o f A g r ic u ltu r a l a n d F o o d
C h e m ., 3 5 , 7 9 4 (1 9 4 3 ). C h e m is t r y a t th e 107 th M e e t in g o f th e A m e r i c a n C h e m i c a l S o c i e t y ,
(12) Pepkowitz, L. P., J. Biol. Chem., 149 , 465 (1 9 4 3 ). C le v e la n d , O h io . C o n tr ib u t io n 286, D e p a r t m e n t o f C h e m is t r y , K a n s a s S ta te
(13) Peterson, W . J., Hughes, J. S., and Freeman, H. F., I n d . E n g . C o lle g e . T h is w o rk is s u p p o r t e d b y th e K a n s a s In d u s tr ia l D e v e lo p m e n t
C h e m ., A nal. E d ., 9, 71 (1 9 3 7 ). C o m m is s io n .
CALCIUM METAPHOSPHATE
Effect of Impurities on Fusibility,
Citrate Solubility, and Hygroscopicity
G. L. FREAR, E. F. DEESE1, A N D J. W . LEFFORGE2

Tennessee Valley Authority, Wilson Dam, Ala.
F E R T IL lZ E R -grade calcium metaphosphate is produced by F ertilizer-g ra d o c a lc iu m m c la p h o s p h a te , w h ich is now

the Tennessee Valley Authority in full-scale units by burn p r o d u c e d in fu ll-s c a le u n its a t th e T V A F ertilizer W ork s
ing elemental phosphorus with a moderate excess of air and b y r e a c tin g P 20 6 w ith r o c k p h o s p h a te a t h ig h te m p e ra
bringing the hot products of combustion into contact with rock tu res, h as a p p ro x im a te ly th e fo llo w in g c o m p o s it io n :
phosphate. The high-temperature reaction of the P 20 t and rock 66 .3% P 2Os, 25 .2% C aO , 4 .5 % S i 0 2, 1 .8 % A120 3, 2 .1 % Fe 2Oa
phosphate yields molten calcium metaphosphate which collects 0 .4 % F. T h e p r o d u c t is v itreou s u n less it c o n ta in s an
in the bottom of the furnace. After tapping and chilling, the excess o f CaO in w h ic h case it is p a rtly cry sta llin e , p re
product is obtained in the form of vitreous lumps. The com po su m a b ly as a resu lt o f th e fo r m a t io n o f so m e Ca 2P 20 ?.
sition of the product depends upon the composition of the rock W h e n th e p r o d u c t carries an excess o f P 20 6, it is so m e w h a t
phosphate used and upon the conditions within the furnace. A h y g ro s co p ic, a n d in p r a c tic e g ro u n d lim e s to n e is added
typical analysis is: 66.3% P 20 3, 25.2% CaO, 4 .5 % S i0 2, 2.1 % to p rev en t c a k in g . L a b o ra to ry stu d ies are rep orted o n
F e A , 1.8% A120 3, and 0.4 % F ; practically the entire PjOs con ch a n g es in fu s ib ility , c itr a te s o lu b ility , h y g ro s co p icity ,
tent is soluble in neutral ammonium citrate. Numerous pot a n d te n d e n cy t o c ry sta llize w h e n a d d itio n s o f P 2Os, CaO ,
tests and field tests by state agricultural experiment stations have S i0 2, A120 3, a n d F c 20 3 are m a d e to p u re c a lc iu m m e t a
demonstrated that vitreous calcium metaphosphate is a valuable p h osp h a te.
phosphatic fertilizer on most soils.
An account of the early developments in the manufacture of
calcium metaphosphate by T V A, a description of a full-scale cal <S>
cium metaphosphate furnace, and a report of some of the solu
bility characteristics of the product have been published (5 , 6, P 20 3 content that confers fusibility and, consequently, the com po
10). Later publications presented further developments in proc sition of the product depend upon the temperature of the fur
esses for producing metaphosphates and a study of the rate of re nace. T o establish fusibility-compositjion relations useful in the
action of phosphorus pentoxide with rock phosphate (4, 7). The manufacture of calcium metaphosphate, the fusion temperature
present paper gives results of more detailed laboratory studies of o f compositions in the binary system C a 0 -P 20 5 was deter
some of the properties of calcium metaphosphate that relate to mined over the range from 17.5% C aO -82.5% P206 to 49 %
its manufacture and use as a fertilizer. C a O -5 1% P 20 6, or in terms of mole fractions, from 0.35 C aO -
0.65 P 2O 5 to 0.71 CaO-O.29 P 2Os. A few determinations were
F U S IB IL IT Y made to ascertain the fusibility o f compositions obtained by add
ing S i0 2 to calcium metaphosphate.
When gases containing phosphorus pentoxide are brought into
contact with rock phosphate in a high-temperature furnace, the M a t e r ia l s . The calcium metaphosphate used in melting
absorption of PjOi proceeds on the exterior portion of the rook point determinations was generally prepared by fusing in plati
until a composition is reached which fuses at the prevailing tem num crucibles the dehydration product of monocalcium phos
perature. The molten material flows toward the bottom of the phate monohydrate that had been made from Iceland spar (56.0%
CaO) and reagcnt-grade phosphoric acid, and that had been re
furnace; it m ay absorb more P 20 6 on its way. The minimum crystallized from the phosphoric acid solution. Other C a 0 -P 20 5
1 P resen t a d d re s s , S c h o o l o f M e d ic in e , E m o r y U n iv e r s it y , A t la n t a , G a . compositions were obtained b y fusion of resublimed P 20 6 with
: P resen t a d d ress, T e n n e ss e e V a lle y A u th o r it y , G o d w in P la n t, C o lu m b ia , calcium metaphosphate or Iceland spar. The compositions con
Tenn. taining S i0 2 were obtained by fusing rock crystal (100.0% S i0 2)
410.4° C., sodium chloride 801°, gold 1063°, and diopside 1391°.
The samples were usually held in the furnace 15 to 60 minutes be
fore quenching. Owing to the possibility of volatilizing P 20 s,
long exposure o f P 2Os-rich compositions to high temperatures was
avoided.
The results of fusion temperature determinations in the system

C a 0 -P 20 5 are shown in Figure 1. The curve represents the
points of complete fusion in the composition range studied. The
liquidus data, except for the compositions that contained more
than 4 5 % CaO, have been revised to conform with the recent
measurements of Hill, Faust, and Reynolds ( 8 ), whose results
were kindly made available to the authors prior to publication.
The tw o sets of determinations of liquidus temperature generally
differed by less than 10° C. except for the compositions that con
tained more than 7 7 % P 20 6 ( A , Figure 1), as in the determination
of the Ca 0 .2 P2 0 6 liquidus where the exclusion of moisture is im
portant. Information concerning the solid-phase equilibria is
given by Hill, Faust, and Reynolds (8).
Extending from 1720° C., the melting point o f tricalcium phos
phate (11 ), through the portion of the Ca 0 - P 206 system shown
in Figure 1, there is a general decrease in the fusion temperature
with increasing P 2O 5 content, except in comparatively small in
tervals between each eutectic and the pure compound of next
higher P-Ch content. In the temperature range 1292 ° to 984 ° C.,
which includes the temperatures most suitable for the formation
of metaphosphate, the liquid compositions formed by reaction of
phosphorus pentoxide with the solid calcium phosphates do not
become solid upon further increase in P 20 6 contents. Thus, in a
C » 0 , W EIG H T PER C E N T calcium metaphosphate furnace the liquid formed on the surface
of the rock phosphate can, in flowing to the hearth, absorb ad
F ig u re 1. T e m p e ra tu re o f C o m p le te F u sion o f P u re
C a O -P 2Oi C o m p o s itio n s ditional P 20 5 without danger of solidification, provided its tem
perature is maintained.
The fusion diagram indicates that, if phosphorus pentoxide re
with a C a 0 -P 205 composition that had a ratio of P 20 6 to CaO acts with tricalcium phosphate at temperatures above 1200° C.
slightly greater than that sought in the product, since some P 2O 7 ( B , Figure 1), products that contain less than 64 % P 2O 5 (P 2O5/
was lost by volatilization while making the product homogene OaO mole ratio 0.70) may be obtained. These compositions are
ous. B y observing the weight loss as heating progressed, it was
possible to stop the heating when the desired composition had
been attained. Prolonged heating was necessary since each com
position was alternately fused and ground until microscopic ex
amination indicated that a homogeneous glass had been ob
tained. Each glass was analyzed in duplicate; the P 20& was de
termined gravimetrically as magnesium pyrophosphate, the CaO
was determined by permanganate titration of the sulfuric acid
solutions o f the calcium oxalate precipitates, and the S i0 2 was
determined by dehydrating with fuming perchloric acid and
weighing the residue as S i0 2.
Prior to the fusibility measurements, most of the compositions
were crystallized by prolonged heating at temperatures slightly
below the fusion range. Observations of the weight loss indi
cated that substantially no P2Os was volatilized during crystal
lization.
M ea su rem en ts. The liquidus temperatures, or points of
complete fusion, were determined in all cases by the method of
quenching. Charges of a few milligrams of each composition
were suspended in platinum foil envelopes in a constant-tempera-
ture furnace and then quenched in mercury. The lowest temper
ature at which the quenched samples, according to microscopic
examination, were vitreous was regarded as the point o f complete
fusion. The temperature of the specimen was measured by a
platinum to platinum-rhodium ( 1 0 % Rh) thermocouple mounted
with its junction close to the platinum envelope and connected to
a Leeds & Northrup Type K potentiometer. The thermocouples
were compared frequently with a standard thermocouple cali
brated at the melting points of the following substances: zinc
T a b le I. F u s ib ility op F e r t iliz e r -G r a d e C a lc iu m M eta p h osp h a tes“

R e f r a c t iv e L iq u id u s
S a m p le ✓---------------------------------------------- ------ —— - C o m p o s i t i o n , W e ig h t % -------------------------------------------------------------- * In dex o f T e m p .,
N o. PaO* CaO S iO i AlaOa FeaOa M nO K aO N a iO F T ota l G la s s ° C. P r im a r y C r y s t a l
A 6 5 .6 1 2 2 .3 5 5 .3 5 2 .5 9 2 .0 5 0 .3 0 0 .0 7 0 .2 8 0 .3 7 9 8 .9 7 1 .5 4 2 900 P r o b a b ly 0 -C a O .P a O *
B 6 5 .6 2 2 4 .8 7 4 .0 4 2 .4 4 1 .5 7 0 .1 2 0 .0 4 0 .2 0 0 .3 9 9 9 .2 9 1 .5 4 6 958 P r o b a b ly /S -C aO .P aO t
C 6 3 .4 1 2 5 .9 3 4 .2 7 2 .7 2 1 .8 8 0 .1 0 0 .0 9 0 .3 5 0 .5 2 9 9 .2 7 1 .5 4 6 1021 ........................................
° D a t a s u p p lie d b y c o u r t e s y o f U . S. D e p a r t m e n t o f A g r ic u ltu r e , B u r e a u o f P la n t I n d u s t r y , S o ils, a n d A g r ic u ltu r a l E n g in e e rin g .

undesirable since, upon cooling, they readily form crystalline cal

cium pyrophosphate which is not citrate soluble. Calcium py
rophosphate appeared in the product of a semiworks-scale fur
nace if the temperature was much above 1200° C. (5). A t high
temperatures equilibrium between the vapor pressure o f phospho
rus pentoxide from the product and the partial pressure o f phos
phorus pentoxide in the gas phase also may limit the P 20 3 con
tent of the product. The vapor pressure equilibrium was not
studied in the present work.
Below 1200° C. there is a considerable range of temperature
within which fusions of approximately the composition of cal
cium metaphosphate may be obtained. I f the temperature is too
low, however, the proportion of P 20 3 may be excessive and the re
sultant product hygroscopic. Thus, at 950° C. in the system
CaO-P-Oj (C , Figure 1) only compositions that contain more than
74.5% P 20 3 (P 2Os/CaO mole ratio 1.15) are completely fused.
Previous publications (5 , 7) indicated that temperatures of at
least 1100° C. are desirable in a calcium metaphosphate furnace,
since the increased viscosities of the products at lower tempera
tures are conducive to excessive P2Os contents. On the other
hand, the high viscosity of calcium metaphosphate fusions at the
temperature of melting is advantageous for, upon cooling, these
compositions crystallize so slowly that devitrification and conse
quent loss in citrate solubility rarely occur in large-scale opera
tion.
The influence of the impurities S i0 2, Fe 20 3, and AI20 3 upon the
fusibility of calcium metaphosphate was not studied extensively.
The liquidus temperatures of three samples o f fertilizer-grade cal
cium metaphosphate, whose compositions are given in Table I,
and the liquidus temperatures of a few C a 0 -P 20 3- S i 0 2 composi
tions are shown in Figure 2. The data for the C a 0 -P 20 6-S i 0 2
compositions indicate that the temperature of complete fusion
of calcium metaphosphate is lowered' between 30° and 40° C. by
the addition of 5 % S i0 2. Little further change in fusibility F ig u re 3. C itra te S o lu b ility o f C a O -
was observed as the silica content was increased up to 19.5% ; P 2Os G lasses as A ffected b y In co r p o r a tio n
o f A120 3 a n d Fe 20 3
at that composition, evidence of a second isotropic pliase with a
refractive index o f 1.46 was noted. Occurrence at 1400° C. o f a
second liquid phase with this refractive index at approximately the influence of variations iii proportions of constituents upon
this composition was reported recently by Barrett and M cCau- the citrate solubility of the product. T o systematize the results,
ghey {2). the effect of variations in the proportions of the principal ingre
Data not included in this paper indicate that the presence of dients in crude calcium metaphosphate upon its solubility were
A120 3 raises the fusion temperature of compositions approximat studied b y fusing pure calcium metaphosphate with different
ing calcium metaphosphate, and that the presence of S i0 2 coun proportions of added CaO, P 20 6, S i0 2, A120 3, and Fe 20 3. The
teracts the effect of A120 3. The effect of Fe 20 3 upon the fusion effect of crystallization upon the citrate solubility o f several
temperature appeared to bo less than that o f A120 3. compositions also was determined.
In preparing the compositions for solubility studies, reagent-
C IT R A T E S O L U B IL IT Y
grade chemicals were used. CaO, S i0 2, Fe 20 3, and A120 3 gener
The development o f a new fertilizer material would be extremely ally were added to pure calcium metaphosphate in the propor
slow if the availability o f the material as plant food were gaged tions desired in the final product; P 20 3 was used in excess of the
solely by plant growth tests. O f the chemical tests o f availability desired proportion, and the final proportion was approached by
of phosphatic fertilizers that are not readily soluble in water, the following the weight loss during prolonged heating in platinum
solubility of P2Os in neutral ammonium citrate seemed more ap crucibles at 1200 ° C. or higher, as explained in the section on fusi
propriate for calcium metaphosphate than the solubility in citric bility. When the necessary quantity of P 20 5 had been volatil
acid; the ammonium citrate solubility was known to be appli ized, the crucible was withdrawn from the furnace and air-cooled
cable to the water-insoluble fraction of concentrated superphos to room temperature without removing the contents. In this
phate, whereas the citric acid method is applicable to basic slag manner there were obtained in homogeneous glassy form about 8
which contains a much higher proportion of CaO. The choice of grams of each composition in three series having P 2Os/CaO mole
the citrate solubility test as a criterion of availability was sup ratios of 0.8, 1.0, and 1.2. Some attempts were made to prepare
ported by the following consideration: Whereas the form of cal the compositions with the P 20 5/C a 0 mole ratio of 0.8 from cal
cium metaphosphate obtained by slowly heating monocalcium cium metaphosphate without adding CaO. The volatilization
phosphate to 600-650° C. was reported to be both citrate in of so much P 20 5 required exceptionally long heating, and some
soluble and ineffective as a phosphate fertilizer (S), vitreous cal of the charge crept over the edge o f the crucible. Since the ma
cium metaphosphate, although largely insoluble in 2 % citric acid, terial outside the crucible lost P 20 5 rapidly, calculation of the
was found to be citrate soluble and effective in supplying P 20 3 composition of the crucible contents from the weight loss was in
to plants on neutral or acid soils (9). exact. The P 20 3/C a 0 mole ratio in the latter preparations va
The composition of fertilizer-grade calcium metaphosphate ried in the range 0.8 to 0.95. The compositions of most of the
varies with the composition of the rock phosphate used in its man other preparations, with indicated exceptions, were calculated
ufacture and with operating conditions in the furnace, especially from the initial weight of the ingredients and the weight loss.
temperature. Information was desirable therefore concerning Analytical determinations o f P 20 3 and CaO in representative
CaO mole ratios of 0.8 to 0.95, 1.0, and 1.2 are shown in Figure 3.
T a b le II. C i t r a t e S o l u b i l i t y o p P 20 5 i n V i t r e o u s C o m p o s i Alumina was fused with the P 20 5 -C a 0 glasses without depressing
t i o n s o f Pa0 4, CaO, SiOj, Fe 20 3, a n d A120 3 the solubility, provided the proportion did not exceed 0.03 to 0.04
-C o m p o s i t i o n — - % T ota l mole A120 3 per mole CaO. Further increase in the proportion
M o le R a t io P jO j
W eiK iit % P a O i/ A l,O a / F oaO a/ S iO j/ C it r a t e - of A120 3 caused the citrate solubility of P 20 5 to decrease in the
P iO l C a O A laO j F cjO i SiOa CaO CaO CaO CaO S ol. proportion 2 to 3 moles of citrate-insoluble P 20 5 per mole of ad
67 33 0 0 0 0 .8 0 0 0 100 ditional A120 3. Substitution of one third or one half the AI2Os
64 32 4 .3 0 0 0 .8 0 .0 7 5 0 0 89
63 31 3 .4 2 .6 0 0 .8 0 .0 6 0 .0 3 0 83 by an equimolar quantity of Fe20 3 did not appreciably affect the
63 31 4 .5 0 2 .6 0 .8 0 .0 8 0 0 .0 8 96 citrate solubility of P 2Os, but complete replacement of AI20 3 by
0 .8 -0 .9 5 0 0 0 100
0 .8 -0 .9 5 0 .0 4 0 0 97
Fe20 3 appeared to cause further decrease in solubility.
0 .8 -0 .9 5 0 .0 7 5 0 0 87 The desolubilizing influence o f A120 3 and Fe20 3 upon C aO -P 2Ot
0 .8 -0 .9 5 0 .1 0 0 0 81
0 .8 - 0 .9 5 0 .1 3 0 0 71 glasses was offset largely b y incorporating silica into the glass, as
0 .8 -0 .9 5 0 .1 5 0 0 67
0 .8 -0 .9 5 0 .0 6 0 .0 3 0 83
shown by Figure 4. Usually 2 to 3 moles of S i0 2 were required
0 .8 -0 .9 5 0 .0 9 0 .0 4 5 0 70 per mole o f A120 3 or Fe 20 3. Thus, in terms of weight percentage,
0 .8 -0 .9 5 0 .0 8 0 0 .0 4 94
0 .8 -0 .9 5 0 .0 8 0 0 .0 8 97 if more than 2 % R 20 3 is present, the combined percentages of
0 .8 -0 .9 5 0 .0 8 0 0 .1 6 98
0 .8 -0 .9 5 0 .0 8 0 0 .2 4 98
A120 3 and Fe 20 3 should not exceed the percentage o f S i0 2 in the
sample. In the crude calcium metaphosphate the percentage of
72 28 0 0 0 1 .0 0 0 0 100
70 28 2 .0 0 0 1 .0 0 .0 4 0 0 99 S i0 2 generally exceeds that o f R 20 3, and the citrate insolubility
69 27 3 .7 0 0 1 .0 0 .0 7 5 0 0 93
68 27 4 .9 0 0 1 .0 0 .1 0 0 0 85
observed is probably due to variables in manufacture other than
67 26 7 .2 0 0 1 .0 0 .1 5 0 0 70 the ratio of RiO» to S i0 2.
DO
O ft o .1
y i n
u
n
u i .n
1 u n on n
U U AO
66 27 3 .9 3 .0 0 1 .0 0 .0 8 0 .0 4 0 75 The citrate solubility of substantially pure vitreous calcium
65 20 5 .6 4 .3 0 1 .0 0 .1 2 0 .0 6 0 58
metaphosphate (Table II) was unaffected by the incorporation in
67 27 4 .8 0 1 .4 1 .0 0 .1 0 0 0 .0 5 90 the glass of 14% S i0 2, or 0.5 mole of S i0 2 per mole of CaO.
66 26 4 .7 0 2 .8 1 .0 0 .1 0 0 0 .1 0 94
65 26 4 .7 0 4 .1 1 .0 0 .1 0 0 0 .1 5 96 The effect of crystallinity upon citrate solubility is indicated
65 25 4 .6 0 5 .4 1 .0 0 .1 0 0 0 .2 0 99
63 25 4 .5 0 7 .9 1 .0 0 .1 0 0 0 .3 0 100
in Table III. All of the C a 0 -P 20 5 glasses showed about the
61 24 4 .4 0 1 0 .0 1 .0 0 .1 0 0 0 .4 0 99 same high solubility, whereas the solubility of the crystalline
67 27 2 .4 3 .8 0 1 .0 0 .0 5 0 .0 5 0 89 samples generally was lower and decreased progressively as the
66 26 2 .3 3 .7 2 .8 1 .0 0 .0 5 0 .0 5 0 .1 0 97
64 25 2 .3 3 .6 5 .4 1 .0 0 .0 5 0 .0 5 0 .2 0 99 P 20 6/C a 0 ratio decreased. The depression in solubility due to
62 24 2 .2 3 .5 7 .8 1 .0 0 .0 5 0 .0 5 0 .3 0 100 crystallization is apparent from comparison of fertilizer-grade
61 24 2 .2 3 .4 1 0 .0 1 .0 0 .0 5 0 .0 5 0 .4 100
samples B and C. It is fortunate that calcium metaphosphate,
75 25 0 0 0 1 .2 0 0 0 100
74 24 1 .8 0 0 1 .2 0 .0 4 0 0 98 even in a high state of purity, can readily be cooled from the mol
72 24 4 .3 0 0 1 .2 0 .1 0 0 0 84 ten state without crystallization. This property is in marked
71 23 6 .3 0 0 1 .2 0 .1 5 0 0 69
69 23 8 .2 0 0 1 .2 0 .2 0 0 0 56 contrast to the behavior of the pyrophosphate which tends to
70 23 4 .2 3 .3 0 1 .2 0 .1 0 0 .0 5 0 71
Crystallize partially, even upon drastic quenching.
69 23 6 .2 0 1 .8 1 .2 0 .1 5 0 0 .0 7 5 92
68 22 6 .1 0 3 .6 1 .2 0 .1 5 0 0 .1 5 94
66 22 5 .9 0 6 .9 1 .2 0 .1 5 0 0 .3 0 100
63 21 5 .7 0 1 0 .0 1 .2 0 .1 5 0 0 .4 5 99
5 9 .2 « 3 0 .0 0 1 0 .8 0 0 .7 8 ' 0 0 .1 3 0 23 — t —
6 8 .6 “ 2 8 .2 0 3 .6 0 0 .9 6 0 0 .0 4 0 97 M O L E R A T IO P 2 O 5 / C o O ñ ,6 0 -.9 5
6 6 .1 “ 2 6 .8 0 6 .7 0 0 .9 8 0 0 .0 9 0 82 M O L E RATIO A l 2 0 3 / C<xO z.Ofi
6 5 .0 “ 2 6 .0 0 9 .4 0 0 .9 9 0 0 .1 3 0 64
6 2 .2 “ 2 5 .4 0 6 .4 6 .6 0 .9 7 0 0 .0 9 0 .2 4 95
6 8 .6 “ 2 3 .3 0 8 .4 0 1 .1 6 0 0 .1 3 0 70
6 6 .5 “ 2 9 .0 0 0 5 .1 0 .9 1 0 0 0 .1 6 100
6 3 .7 2 5 .6 0 0 1 0 .7 0 .9 8 0 0 0 .3 9 100
6 1 .1 “ 2 5 .2 0 0 1 3 .6 0 .9 6 0 0 0 .5 0 100
5 5 .5 « 2 3 .0 0 0 2 0 .1 0 .9 5 0 0 0 .8 2 99
° C om p osition determ ined b y analysis.
glasses agreed with the calculated compositions within 3% . The

degree o f oxidation of the iron in the products was not deter
mined. Calculations showed, however, that even if all the iron
were reduced to the ferrous state, failure to correct the apparent
PjOs loss by the resultant loss of oxygen in unanalyzed samples in
no caso would cause an error o f more than 1 % in the calculated
P 20 6 content. The agreement between the results o f calcula
tions and the analyses was commensurate with the degree of re
producibility of the citrate solubility tests.
In grinding the glasses to —80 mesh, an effort was made to
treat all samples uniformly so as to obtain substantially the same
size distributions in the compositions. The citrate-insoluble
PjOs was determined by using 1 -gram samples of —80 mesh ma
terial to 100 ml. of neutral ammonium citrate solution. The citrate
digestion was made with inclusion of filter paper pulp.
Largo groups o f citrate solubility determinations were made at a
time to minimize the possibility of irregularities in the analytical
procedure.
The results of citrate solubility determinations on the vitre
ous products are given in Table II. The influence of additions of F ig u re 4. C itra te S o lu b ility o f C a 0 -P 20 t - R 20 3
A120 3 and Fe 20 3 upon the citrate solubility of glasses with P 20 6/ G lasses as A ffected b y In c o r p o r a tio n o f S i 0 2
’ If exposed to the weather, calcium metaphosphate slowly dis

T a b le III. C it r a t e S o l u b il it y as A ffected by solves in rain water. Tests of crude lump calcium metaphosphate
C r y s t a l l iz a t io n
exposed in the open in 2 -3 inch layers showed that a sample with a
M o le % T ota l
R a t io , PaOj
P 20 6/ C a 0 mole ratio of 1.26 was dissolved several times as rap
W e ig h t %
Com pn. P sO # / C it r a t e - idly as one with a ratio of 0.94. During the period of a year, in
N o. CaO P 2O 5 S iO î CaO P h a se s P r e s e n t S ol.
which the rainfall was 39 inches, the sample with the P 2Os/CaO
F-101A 2 0 .6 7 9 .4 0 1 .5 2 C r y s t a ls : 2 C a 0 . 3 P a 0 » 100
D-12 2 4 .7 7 5 .3 0 1 .2 0 G la ss 100 mole ratio of 0.94 lost about 11% o f its weight and 12% of its
D-12 2 4 .7 7 5 .3 0 1 .2 0 C r y s t a ls : 2 C a 0 . 3 P j 0 * &
/3 -C a O . PaO j 82
P 2Oa content; the lumps became coated with an easily friable
19 2 6 .0 7 4 .0 0 1 .1 2 C r y s t a ls : 2 C a 0 . 3 P s 0 6 & layer but did not disintegrate. The sample with the P 20 6/C a 0
^ -C a O .P a O s 60
D-8 2 8 .2 7 1 .8 0 1 .0 1 G lass 100 mole ratio of 1.26 was reduced in the same time to a porous,
BD 30 70 0 0 .9 C r y s t a ls : /3 -C a O . PaOs
s o lid s o ln . 15 crusty mass.
R -l 3 1 .8 0 8 .2 0 0 .8 5 G la ss a n d trô m e lite ® 79 The caking of impure calcium metaphosphate glasses in which
D-10 3 3 .0 6 7 .0 0 0 .8 0 G la ss 100 .
R-2 3 8 .8 6 1 .2 0 0 .6 2 G la s s , tr ô m e lit e , a n d /9- the ratio of P 20 6 to CaO is not too high can be offset by the ad
2 C a O . P *O i s o lid s o ln . 42
R-3 4 4 .6 5 5 .4 0 0 .4 9 C r y s t a ls : 0 -2 C a O .P a O s 7
dition of substances like calcium carbonate, which react with the
Ab 2 2 .4 6 5 .6 5 .4 1 .1 6 G la ss 100 products of hydrolysis of calcium metaphosphate to form non-
B* 2 4 .9 6 5 .6 4 .0 1 .0 4 G la ss 100
O 2 5 .9 6 3 .4 4 .3 0 .9 7 G la ss a n d c r y s ta ls 83 hygroscopic compounds. Previous work (1 0 ) had shown that
a T r o m e lit c is a series o f s o lid s o lu t io n s id e n tifie d b y H ill, F a u s t, an d mixtures o f one part of — 10 0 mesh calcium metaphosphate with
R e y n old s (8 ) ; it lie s b e tw e e n m e t a p h o s p h a te s o lid s o lu t io n a n d c o m p o s it io n
3 C a 0 .2 P a 0 5. three parts of 100 mesh limestone could be stored for 75 days in a
* F e r tiliz e r -g r a d e p r o d u c t s : a n a ly s e s b y c o u r t e s y o f U . S. D e p a r t m e n t o f moist atmosphere without gain in weight or loss of free-flowing
A g ricu ltu re, B u r e a u o f P la n t I n d u s t r y , S o ils, a n d A g r ic u ltu r a l E n g in e e rin g .
property. Figure 5 gives the results of hygroscopicity tests upon
crude —80 mesh calcium metaphosphates, with and without the
addition of 5 % of —100 mesh limestone as conditioner. The
H Y G R O S C O P IG IT Y
limestone-conditioned sample having the P 2Os/CaO mole ratio of
Although lumps of vitreous calcium metaphosphate having a 1.13 in 28 days gained only 0.02% in weight and remained free
PsOi/CaO mole ratio of 1.0 give little evidence o f absorbing mois flowing; another sample of the same calcium metaphosphate
ture or caking upon exposure to the atmosphere, compositions of without limestone gained 1 1 % in weight and became gummy.
high P 20 5/C a 0 ratios are markedly hygroscopic, more so if the Laboratory tests o f longer duration at higher humidity indicate
product is finely ground. T o obtain further information in re the desirability o f using 10% limestone. The effectiveness of
gard to the influence of composition upon hygroscopicity, com ground limestone as a conditioner has been checked with 35-mosh
parisons were made of the hygroscopicity of representative com calcium metaphosphate in storage tests under warehouse condi
positions that had been prepared for the determinations of tions. I f the PjOs/CaO mole ratio did not exceed 1.0 and the
citrate solubility. For this purpose 3-gram samples of —80 mesh particles o f limestone were uniformly distributed, the materials
materials were spread in thin layers on 9-cm. watch glasses. The generally remained granular for at least a year.
watch glasses were mounted on racks in 1 -gallon cans which were The nature of the hydrolysis of calcium metaphosphate has not
tightly closed with friction tops. The cans were placed in a 30 0 C. been investigated completely, but some preliminary observations
constant-temperature air bath. An atmosphere of 59 % rela have been made. Well-defined crystals o f monocalcium phos
tive humidity was maintained b y including in each can shallow phate monohydrate, accompanied b y small amounts of appo-
dishes containing 'saturated aqueous solutions of ammonium ni
trate. The rate of absorption of moisture by the different com
positions was observed by noting periodically the increases in
weight o f the samples during a 45-day period. Comparisons of
the weights gained in 9 days were typical o f those in other inter
vals. In a 9-day exposure to the humid atmosphere, pure C aO -
P20 5 compositions having P 20 6/C a 0 mole ratios of 0.8-1.0 and
of 1 .2 gained, respectively, 1 - 2 mg. and 1 0 m g.; that is, the
weight increased 0.03 to 0.3% . When no silica was present, the
introduction of 0.10-0.15 mole of R 2Oa per mole of CaO reduced
the moisture absorption about half, the reduction being even
greater in the compositions having the P 20 6/C a 0 mole ratio of
1.2. The addition to compositions containing R 2Oa of 2-3 moles
of Si0 2 per mole of It 2Oa increased the moisture absorption to
about that o f pure C a 0 -P 20 6 compositions of the same P 20 6/ C a 0
mole ratio.
The extent of caking after 45-day exposure to the humid atmos
phere showed parallelism to the weight gain. None of the com
positions with P 2Os/C aO mole ratios of 0.8-0.95 caked appre
ciably, whereas the caking at the mole ratio 1 . 2 was more pro
nounced than at 1.0. Introduction of 0.15-0.20 mole o f RiCh per
mole of CaO eliminated caking at the P 20 6/C a 0 mole ratios of
1.0-1.2 under the conditions of test. The introduction of 2-3
moles of S i0 2 per mole of R 20 3 in the latter compositions in
creased the tendency to cake.
Measurements of the crushing strengths of cakes formed from
different fertilizer materials containing up to 1 % moisture upon
storage at pressures of 12 poimds per square inch for 7 days were
reported by Adams and Ross (I). Their tests showed that the
hardness of cake formed from calcium metaphosphate was greater
F ig u re 5. E ffect o f G ro u n d L im e s to n e ( —100 ra csli) on
than that from superphosphate, but much less than from potas H y g r o s c o p ic ity o f F c r liliz e r -G r a d c C a lc iu m M e ta p h o s
sium chloride, urea, ammonium sulfate, or sodium nitrate. p h a te ( —80 m e s h ) a t 30° C . a n d 59 % R ela tiv e H u m id ity
rently amorphous substances, were found on the surface of beads physical Laboratory, Carnegie Institution of Washington. The
of vitreous compositions of ILOs/CaO mole ratio 1.4—1.9; these generous cooperation of S. B. Hendricks, K . D . Jacob, and W. L.
compositions had been made by reaction of rock phosphate with Hill of the U. S. Department o f Agricult ure is acknowledged.
PjOt in excess and had been stored in capped bottles in the labora
tory for five years. Autoelaving of pure, finely ground calcium L IT E R A T U R E C IT E D
metaphosphate with small amounts of water for a few minutes
(1) Adams, J. R .. and Ross, W . H., I nd . E no . C hem ., 33, 121-7
at 180° C. and then cooling gave practically complete conversion
(1 9 4 1 ).
of metaphosphate to monocalcium phosphate. In boiling dilute (2) Barrett, R . L., and M cCaughey, W . J., Am. Mineral., 27, 680-
acids, complete hydrolysis of calcium metaphosplmte to ortho 95 (1 9 4 2 ).
phosphates requires several hours, and in aqueous extracts at (3 ) Bartholomew, R . P., and Jacob, K . D ., J. Assoc. Official Ayr.
Chem., 16, 5 9 8 -6 1 1 (1 9 3 3 ).
room temperature the hydrolysis may continue for months (10).
(4 ) Copson, R . L., Pole, G. R ., and Baskervill, W . H., I n d . E n g .
In the presence of limestone the hydrolysis of calcium metaphos C h e m ., 3 4 , 2 6 - 3 2 (1 9 4 2 ).
phate is followed by formation of dicalcium phosphate, which (5 ) Curtis, H. A., Copson, R . L., and Abrams, A. J., Chem. A- Met.
was identified in the solids obtained by boiling an aqueous slurry Eng., 4 4, 1 4 0 -2 (1 9 3 7 ).
(6) Curtis, H. A ., Copson, R . L., Abrams, A. J., and Junkins, .1. N.,
of the stoichiometric proportions of pure calcium carbonate and
Ibid., 4 5, 3 1 8 - 2 2 (1 9 3 8 ).
calcium metaphosphate. (7) Frear, G . L., and Hull, L. H ., I nd . E no . C hem ., 33, 1560-6
(1 9 4 1 ).
ACKNOW LEDGM ENT (8) Hill, W . L., Faust, G. T ., and Reynolds, D . S., Am. J. Sei., 212,
4 5 7 (1 9 4 4 ).
The authors acknowledge the encouragement and aid of It. L. (9) Jacob, K . D ., and Ross, W . H ., J. Agr. Research, 6 1, 539-60
Copson, J. W . H. Aldred, E. H. Brown, and other members of (1 9 4 0 ).
(1 0 ) M aclntire, W . H ., Hardin, L. J., and Oldham, F. D ., I n d .
the T V A Chemical Engineering Staff in obtaining the data and
E no . C hem ., 29, 2 2 4 - 3 4 (1 9 3 7 ).
preparing this paper. They are grateful also for the helpful (1 1 ) Trömel, G., Mitt. Kaiser-Wilhelm Inst. Eisenforsch. Düsseldorf,
advice of J. F. Schairer and other members of the staff o f the Geo 14, 2 5 - 3 4 ( 1 9 3 2 ) ; Stahl u. Eisen, 6 3 , 2 1 - 3 0 (1 9 4 3 ).
RATE OF SEDIMENTATION
Suspensions of Uniform-Size Angular Particles
H ates o f se d im e n ta tio n are rep o rte d fo r su sp en sion s o f c lo se ly sized em ery p a r
ticles, b o t h flo c cu la te d a n d n o n flo c c u la te d . E x cep t fo r th e v alu e o f an ex
p e rim e n ta l c o n s ta n t, o n e r a te e q u a tio n ap p lies to b o t h sta tes, p rovid ed the
flo ccu la te d su sp en sion s are h ig h ly c o n c e n tr a te d . C o m p a ris o n w ith p rev iou s tests
o n u n ifo r m sp h eres in d ic a te s th a t a p o r tio n o f th e liq u id su sp e n s io n m e d iu m is
carried d o w n w ith th e a n g u la r e m e ry p a rticle s d u rin g th eir fa ll, w h e th e r the
HAROLD IL STEINOUR su sp en sion s are flo c cu la te d o r n o t. T h e q u estion as to w h e th e r this liq u id is h ou n d
Portland Cement Association, to th e pa rticles o r sim p ly sta g n a n t is stu d ie d , an d evid en ce is sh ow n to su p p ort
Chicago, 111. th e la tter view.
N T H E first article of this series ( I B ) equations were de S E D IM E N T A T IO N T E S T S O N E M E R Y P O W D E R S

I
veloped for the sedimentation rates of dispersions of uniform A commercial emery powder was fractionated by air separa
spheres. Under condit ions where the Stokes law gives the ve tion; two fractions called A and B were used in sedimentation
locity of a single isolated sphere, the velocity at any concentration tests. Their densities were 3.79 and 3.77 grams per cc., respec
tively. The appearance of emery A is shown by Figure 1.
of spheres is given by Equation 24 of the first paper; Size analyses (Table I) were obtained for both products by an
Q = adaptation of the Wagner turbidimeter m ethod (16).
The sedimentation tests were made in water. A t first, in
where the function lO-i-aad-e js empirical. A t values of < tests in which flocculation was to be avoided, no dispersing
between 0.3 and 0.7, Equation 24 is practically equivalent to agent was added, for the fresh powders did not flocculate. H ow
Equation 23 ( I B ) : ever, as a safeguard 0 . 1 2 % sodium hexametaphosphate was used
in later testa. Absence o f flocculation was shown by direct
Q = 0 .1 2 3 7 , y -—— observations with a binocular microscope, and by the "firmness
1 —« and constant density of the sediments produced by different
which was derived, except for the value o f the proportionality initial concentrations of solid. In all tests in which flocculation
was desired, zinc sulfate was used, chiefly at 0.12% . All sus
constant, by using the hydraulic radius to denote the size of the
pensions were mixed for 2 minutes with an electrical mixer and,
flow space around the spheres and by assuming that no addi except as noted in the tables, were tested in a cylindrical glass
tional variable shape factor was needed. jar about 100 mm. in diameter, filled to a height of 40 to 60 mm.
In the present article the sedimentation of angular particles is These dimensions were chosen in order that effects of wall fric
tion would be negligible at the center o f the jar ( H ) . A microm
considered, starting with the equations for spheres. The experi
eter microscope reading to 0 .0 0 1 mm. was used to follow the
ments on which the study is based embraced both the disperse change in level of the suspended solids. Readings wore timed
and flocculated states. Low Reynolds numbers (Table IV) en to 1 second. Temperatures were 24° =*= 1 ° C.
sured viscous flow in the displaced liquid. Particles of very small In all tests on flocculated suspensions the readings were taken
size were used to permit flocculation; they were closely sized in on a float placed centrally in the jar (Figure 2). The float was
like one used by Powers (14 ), a thin disk of Bakelite with a
order to obtain uniform settlement in the disperse or non
thread-like glass stem attached to its upper face. The densities
flocculated state. Flocculated suspensions in which many were such that the float rode at the plane o f separation of the
sizes are present will be treated in a third article. suspension and a layer of water which was placed on top in
The manner of its occurrence suggested that some of the water

was not taking part in the general flow of liquid relative to the
solid particles which occurs during sedimentation. Hence,
Powers assumed that a small quantity of water remained with
each falling particle. Symbol w i which he applied to the sub
tractive term in his equation stood for “ immobile water” ; the
same symbol will be used here. Kozeny (1 1 ) and Carman
(S, 5 ) also found that a term equivalent to wt was needed in a
few instances in an equation, similar in principle to 23 (1 6 ),
which K ozeny (11, 12) derived for laminar flow through granular
beds. The need for the term arose only in a few water perme
ability tests on clays.
The three authors cited above concluded that, in tests re
quiring the Wi term, the need probably arose because of liquid
which in some way became bound to the particles. However,
Fair and Hatch (7 ) who derived an equation like that of Kozeny,
believed that stagnation prevented some of the liquid from con
tributing to the flow. Such an occurrence would be analogous to
the separation from laminar flow observed by Johansen (9) at
sharp-edged orifices in circular pipe.
In nonflocculated suspensions any liquid that fails to take
part in the general flow must be in separate increments asso
ciated with the individual solid particles, irrespective of whether
it is bound to the particles or stagnant at angularities in their
contours. If the amount of such liquid per particle is inde
F ig u re I. o f E m ery A (X 500)
quantity sufficient to cover the disk. Readings were taken by T a b le I. S iz e A n a ly s e s o f E m e r ie s A a n d B f r o m S e d im e n

sighting on the tip of the stem. ta tio n in W a t e r , U s in g S o d iu m H e x a m e t a p h o s p h a t f, a s
In suspensions that were not flocculated, wall friction was D is p e r s a n t
not appreciable; hence, in tests on these suspensions the float
D ia m e t e r , E m ery A , E m ery B, D ia m e t e r , E m ery A, E m ery B,
method was eventually abandoned and readings were taken M ic r o n s % b y W t. % b y W t. M ic r o n s % b y W t. % b y W t.
directly on the level at the wall of the vessel. In all cases the
suspensions settled with satisfactorily sharp upper boundaries. 50 - 4 0 2.0 1 0 -9 5 .2 2 3 .3
40 - 3 0 1 .7 9 -8 0 .1 1 7 .6
30 - 2 0 2.2 8 -7 0 .1 1 1 .1
20 - 1 8 . 5 1 .7 7 -6 0 .0 2 .1
Curves of subsidence of the surface of a suspension against
1 8 .6 - 1 7 1 .4 6 -5 0 .1 0 .4
time are shown in Figure 3. Their shapes are fairly typical for 17 - 1 6 1 .9 5 -4 0 .0 0 .3
16 - 1 5 3 .3 4 -3 0 .1 0 .0
most of the tests, but at values of e below 0.65 the nonflocculated 15 - 1 4 6.8 3 -0 0 .4 0 .5
suspensions gave curves which sagged a little in the central por 14 - 1 3 1 2 .5
13 -12 22.0
tion (Figure 4). The effect was distinctly noticeable in tests on 12 -11 2 4 .1 E m ery E m ery
emery A at e = 0.60 and 0.577, and those tests were omitted from 11 -10 1 4 .4 A B
S p . su r fa c e , c a lc d . as f o r s p h e re s , s q . c m . / c c . 4930 625 0
further study. Liquid content e = 0.577 was the lowest at which A v . d ia m . c a lc d . fr o m s p . s u r fa c e , m icro n s 12.2 9 .6
a good mixture could be made.
Results of sedimentation tests are shown in Table II, arranged
in order of increasing dilution. The rates of settlement are for
the initial constant-rate periods.
R A T E S O F S E D IM E N T A T IO N O F N O N F L O C C U L A T E D
S U S P E N S IO N S
Rate data that conform to Equation 23 of the first paper fall

along a straight line through the origin when [<2 ( 1 — «)]■/> is
plotted against e. In Figure 5 this method of plotting has been
applied to the data for the nonflocculated suspensions o f emeries
A and B.
The dashed lines in Figure 5 represent Equation 23 (16) as
applied to the systems under study. In determining the values
of V , the average diameters reported in Table I were used. It is
evident that the experimental points do not conform to Equation
23. However, up to f = 0.75 the data for both emeries are rep
resented by
Q = 0.176 V . (e~ ^ 1f6 8 )3 (1)
an equation similar to 23 of the first paper but differing, in the

term on the right, both in the magnitude o f the initial factor and
in the nature o f the cubed quantity.
The function of c shown by Equation 1 is the same form as
that found by Powers (1 4 ) for flocculated pastes of portland
cement and water. For those pastes the subtractive term was
considerably larger than 0.168 and varied with the cement used.
pendent of the nearness of other particles, then

T a b le II. S e d im e n t a t io n o f E m e r ie s A and B i n 1 0 0 -M m .-D ia m e t e r for a given powder the total amount should be pro
V essel portional to tho volume of the powder. Hence, if
F lu id P o r o s it y
C on cn . of C o n te n t, R a te o f S e t t le - of a is the proportionality constant, the immobile
R e g u la tin g % o f F lu id S e t t le m e n t in e n t, S e d im e n t. liquid per unit of total volume is «(1 — «)• T o cor
T est N o. A g en t, I n itia l in T o t a l ( C m ./ S e c . ) % of % of
(C h ro n o % of R e a d H e ig h t, V o l. X 10« In itia l S e ttle d rect Equation 23 {1 5 ) for such immobile liquid,
lo g ic a l) W a ter W t. in g Mm. <€ X 10*) (Q X 10«) H e ig h t V o l.
"a (l — e) should bo subtracted from each e in the
E m e r y A , N o n flo c c u la t e d
2 0 F lo a t 55 6 5 .0 1390 32 48 equation. The limiting velocity, F., needs no cor
1 0 F lo a t 50 7 0 .0 2155 42 48 rection since it is found experimentally and should
3 0 F lo a t 41 „ 7 5 .0 339 0 52 48
4 0 S id e 57 7 5 .0 3355 53 47 therefore embody the effect of any immobile liquid.
5 0 S id e 56 8 0 .0 541 0 63 46
E m e r y .A, F lo c c u la te d
The modified equation after some rearrangement is:
6 0 .1 2 F lo a t 51 6 0 .0 566 9 56
12 0 .1 2 F lo a t 53 6 2 .5 736 14 56
7 0 .1 2 F lo a t 6 3 .7 806
9 0 .1 2 F lo a t 56 6 5 .0 932 Í7 58
8 0 .1 2 F lo a t 52 6 9 .6 1452 27 58 Q - 0.123 ( 1 + a ) 2 V, ~— (3)
13 0 .0 6 F lo a t 58 7 0 .0 1520 26 59
10 0 .1 2 F lo a t 42 7 5 .0 250 0 35 61
11 0 .1 2 F lo a t 44 8 0 .0 470 0 43 04
14 0 .1 2 F lo a t 73 8 5 .0 9930 60 62
If Wi is used in place of a / { 1 + a ), a equals
E m e r y A , , N o n flo c c u la t e d
— w i ) , and tho equation becomes:
W i/ ( 1
23 0 .1 2 F lo a t 57 6 2 .5 1108 32 44
21 0 .1 2 F lo a t 56 6 5 .0 1371 36 45
20 0 .1 2 F lo a t 56 7 0 .0 2185 45 45
22 0 .1 2 F lo a t 56 7 0 .0 2190 45 46 _ 0.123 V . (« - W iy (3)
15 0 .1 2 S id e 42 7 5 .0 3280 54 45
16 45
H “ (1 - Wi)2 1 - 6
0 .0 6 S id e 43 7 5 .0 3270 55
17 0 .2 0 S id e 42 7 5 .0 3210 54 45
19 0 .1 2 S id e 42 8 0 .0 513 0 63 45
18 0 .1 2 S id e 42 8 5 .0 755 0 72 46 When W i = 0.168 as in Equation 1, 0.123/( 1 — t« )*
E m e r y B , N o n flo c c u la t e d equals 0.178, and Equation 3 becomes practically
0.12 Side® 6 0 .0
0.12 S id e * 6 2 .5 identical with Equation 1 . Hence the experi
0.12 Side® 6 5 .0 mental results may be regarded as wholly consist
0.12 S id e 7 0 .0
0.12 F loat 7 0 .0 ent with the theory that a quantity of liquid pro
0.12 S id e 7 5 .0
0.12 S id e 8 0 .0 portional to the volume of the emory remains with
0.12 S id e 8 5 .0 the particles during their fall. For both emeries A
E m e r y B , F lo c c u la te d
11 0 .1 2 F lo a t 32 6 7 .5 658 20 59
and B the quantity of this liquid is indicated to
9 0 .1 2 F lo a t 36 7 0 .0 841 24 61 be 0 .2 0 2 ( 1 — é ) , or one fifth o f tho volume of solid.
10 0 .1 2 F lo a t 43 7 5 .0 1475 35 61
° V essel 6 8 m m . in d ia m e te r.
According to this theory, « — a ( l — e ) is the
* V essel 63 m m . in d ia m e t e r . general expression for the mobile liquid. There
° T h is v a lu e w as o m it t e d fr o m th e fig u res. I t is w id e ly d iffe r e n t fr o m th e o th e r v a lu e
f o r e — 0 .7 0 a n d a p p e a r s t o in v o l v e a n e rro r s u ffic ie n tly la r g e t o w a r ra n t r e je c t io n . fore, since the linear relation illustrated by Figure
5 was previously indicated {1 5 ) to apply up to
N o .l - Test of one preparation. Readings token at the

center of the surface by observation of a float,
i i i i ' i
No. 2 - Test of a duplicato preparation. Readings taken at the
surface by observation through the glass wall. Xp300
0 _______
0 5 10 15 20 25 30 35
Time, Minutes Time, Minutes
F ig u re 3. S e d im e n ta tio n C urves fo r E m ery A a t F ig u re 4. S e d im e n ta tio n C urves fo r N o n flo c cu la te d

e = 0.63 S u sp en sion s o f E m ery A in P la in W a ter a t e = 0.60
a was zero, it should apply up to «

is 0.2. This agrees satisfactorily
S T A G N A N T L IQ U ID
-Testsin 0.12% Sodium Hexai
The fact that the two emery powders of different -Tests in Plain Water.
size particles have the same value of a helps de

termine whether the immobile liquid is bound or
merely stagnant. If this liquid is stagnant, then
a, the quantity of such liquid per unit volume of
solid, should bo determined by the shapes of the par
ticles and be independent o f particle size, as was
found. But if the liquid is bound to the particles,
the thickness of the layer should probably be about
the same for particles o f different sizes, and the
total quantity would not then remain a constant
proportion of the volume of solid, 1 — e, when the
size of particle was changed. It would be more
nearly proportional to the exposed surface, the
specific surface in sq. cm. per cc. times ( 1 — e). On
this latter basis wt should be 1 . 2 times as large for
emery B as for emery A. The data in Figure 5
conform much better to the one common value for
wi; indeed, as will be shown later, equality of the
wi values is also indicated by a more precise method
of evaluation. However, to obtain more evidence,
tests were made on another emery powder of much
larger particle size.-
A relatively large-grained product o f fairly uni
form particle size was obtained by sieving a coarse
emery powder on screens having square openings,
62 and 74 microns on a side, respectively. The prod
uct, emery E, had a density of 3.93 grams per cc.
and was testod for sedimentation characteristics in
F ig u re 5. l/« v s . e f o r S e d im e n ta tio n o f N on floceu lated .
diethyl phthalate (density 1.115 grams per cc. at S u s p e n s io n s o f E m eries A a n d B
2 2 ° C., viscosity 0.091 poise at 27.5° C.). N o dis
persing agent was needed. The first sedimenta
tion tests were made at 27.5° C., later ones at 23.5° C. All The rates obtained at higher values of t arc shown in Table II I ;
readings were taken on the level of the suspension at the wall those determined at 27.5° C. were used in Figure 6 .
of the vessel. Curves of height of suspension against time were The value o f w i indicated by Figure 6 is 0 .2 2 , a figure distinctly
similar to those for the finer emeries. The porosities of the larger than the value 0.168 found for emeries A and B. This re
sediments wore practically constant at 58 % o f the settled volume. sult is the opposite of what would be expected of bound liquid, if
Rates of settlement obtained at e = 0.60 were exceptionally low the diethyl phthalate is assumed to build about the same thick
and are not reported; the fluid content was judged to be too ness of adsorption layer as the water. The large particles of
close to its lower limit to permit free orientation of the particles- emery E would require much lower values of a and w< than do
the particles of A and B if the adsorption layers were of the same
thickness. It is estimated that the thickness would have to be
T a b l e III. S e d i m e n t a t i o n o f E m e r y E i n D i e t h y l P h t h a l a t e about twenty times as great on emery E particles to account for
Testa in 6 8 - M m .- D i a m . J a r a t 9 2 - M m .- D i a m . J a r a t the results obtained. Against the possibility of this occurrence
2 7 .5 ° C . 2 3 . 5 ° C .°
is evidence obtained by the writer in tests on portland cement
F lu id
con ten t, in various liquids, including diethyl phthalate; this evidence
% o f fluid
T est N o. in t o t a l indicates that there is little change in w i upon changing liquids.
(ch ro n o v o l.
logical)
Furthermore, even the estimated layer thicknesses for emeries'
(« X 10«)
1-1 6 2 .5 A and B are so great as to constitute strong independent evidence*
1-2 6 2 .5 against the existence of tho layers. Assuming that the volum e
2-1 6 5 .0 451 o f bound liquid is approximately equal to the thickness o f layer
2-2 6 5 .0 455
1574
times the area of surface exposed to liquid, thicknesses of 0.30
3 -1 6 7 .5 570 1430 and 0.24 micron are indicated for emeries A and B, respectively.
3 -2 6 7 .5 572 1405
1418 |The surfh.ee areas were calculated from air permeability data
4 -1 7 0 .0 739
4 -3 7 0 .0 767 9 -1 8 0 .0 1806 ( I S ) which indicated values about 1.4 times those obtained by
4 -4 7 0 .0 730 9 -2 8 0 .0 1858 sedimentation analysis.] Y et Bulkley (3 ) found that in flow
9 -3 8 0 .0 1848
5 -1 7 2 .5 925 9 -4 8 0 .0 1876 o f oils the thickness of stationary film does not exceed 0 .0 2 to
5 -2 7 2 .5 926
1 0 -1 8 2 .5 2290 0.03 micron; Bastow and;Bowden (1) studied the flow of various
6-1 7 5 .0 1181 1 0 -2 8 2 .5 2310 liquids and found that liquid 0 .1 micron from a solid surface
6-2 7 5 .0 119 8 1 0 -3 8 2 .5 2300
1 0 -4 8 2 .5 2310 takes part in the movement.
Q R a te s a d ju s t e d t o 2 7 .5 ° C .
b y m u ltip ly in g b y 1.16 ( fa c t o r 1 1 -1 8 5 .0 28 9 0 Still further evidence appears in the previous article (1 5 ) in
based on tests a t e — 0 .7 5 ). 1 1 -2 8 5 .0 284 0 which Equation 23 was found to be adequate for suspensions of
, * T esta m a d e in 6 8 -m m .-d ia m . 1 1 -3 8 5 .0 291 0
jar. 1 1 -4 8 5 .0 2870 glass spheres that averaged 13.6 microns in diameter. Those
spheres were only a little larger than the particles of emery A
average diameters reported in Table I. The wi values as origi

nally estimated were adjusted slightly to make the slopes
strictly 1.82; this was the basis for the statement made earlier
that the equality o f the w i values for emeries A and B did not
rest solely on the evidence from Figure 5. Figure 7 shows that
when these Wi values are used, the data conform closely to the
straight lines even at the high dilutions; as corrected for
stagnant liquid, they conform to the equation found to hold for
spherical particles.
The stagnant liquid may not be strictly immobile, for a
difference simply in the patterns of the streamlines near the
surfaces of angular and spherical particles could conceivably cause
the difference in results. However, since the flow patterns must
be similar in their more general aspects, the close agreement of
the theory with the data seems to indicate that any motion in the
liquid called “ stagnant” must be relatively small.
Miscellaneous data are assembled in Table IV to aid in com
paring conditions in the suspensions of the various emery powders.
The Reynolds numbers are given only for infinite dilution and
are calculated from the formula commonly used for spheres. They
are not strictly comparable when the particles differ in shape.
It should be realized also that a Reynolds number representative
of the general flow will decrease with e (4), since both the velocities
and the distances between particles decrease.
C O M P A R IS O N W IT H W ORK OF OTH ERS

F ig u re 6 . [( ? ( !“ *) ]'/* v s - e f ° r S e d im e n ta tio n
o f N o n flo ccu in te d S u sp en sion s o f E m ery E
Equation 4 has been applied to data obtained by Kermack,
M ’Kendrick, and Ponder (1 0 ) on the sedimentation of red blood
cells, discoidal in shape, apparently uniform in size, and not
and might have been expected to bind a comparable quantity of flocculated. The tests were made on the cells o f man, rabbit, ox,
water, yet no w i term was needed. and sheep. Gaudin (8 ) averaged the results obtained for the
The theory that the W i values found for the nonflocculated sus different types of cells, after converting them to percentages of
pensions of emery particles arc indicative of adsorbed liquid must the estimated velocities at infinite dilution, and then compared
be rejected; the theory that there is stagnant liquid which ac the averages with values calculated from several different equa
companies the angular particles in amounts determined by their tions. Table V reproduces part of Gaudin’s table, including the
shapes is consistent with all data considered thus far. The quan values calculated from the equation of Kermack et al. and those
tity per unit volume o f solid is evidently greater for emery E than from Gaudin’s best equation. Values calculated from Equation 4
for emeries A and B, but the difference in particle shape which are shown in comparison. The method of applying Equation 4
this indicates is in accord with the fact that emery E gives sedi was graphical, and no weight was given to Gaudin’s 1 0 0 % 'rate
ments of relatively high porosity.
R E S U L T S F O R A L L C O N C E N T R A T IO N S
T able IV. D a t a f r o m T e s t s o n E m e r ie s A , B, and E in
Up to this point the concept of stagnant liquid has been tested N o n f l o c c u l a t e d S u s p e n s io n s
only with data obtained at concentrations high enough so that E m ery B E
Equation 3 can be applied. The general Equation 24 (1 5 ) pre S to k e s v e l o c it y V s, c m ./s e c . 0 .0 2 5 0 .0 1 5 0 .0 9 8
R e y n o ld s N o . a t in fin ite d iln .° 0 .0 0 3 3 0 .0 0 1 6 0.0091*»
sented at the beginning of this article provides the opportunity wi 0 .1 6 8 0 .1 6 8 0.22
to make a more extensive test of the concept. If the quantity 0.202 0.202 0 .2 8
of stagnant liquid remains, a fixed proportion of the volume of p a r t ic le d ia m . X liq u id d e n s it y X S to k e s v e l o c it y

a D e fin e d a s:
solid at all concentrations, then Equation 24 should represent v is c o s it y
b P a r t ic le d ia m e t e r o f e m e r y E w a s ta k e n as 0 .0 0 7 6 c m ., c o r r e s p o n d in g to
the data for the emery if e is replaced by a quantity represent lim it in g v e l o c it y in d ic a te d b y F ig u r e 7.
ing actual mobile liquid per unit of total volume. This quantity
is c — a (1 — e), equivalent to (c — u x )/(l — wt). Substitution
T a b le V . A p p lic a t io n o f V a r io u s E q u a tio n s t o D a t a o f
in Equation 24 gives: K e r m a c k , M ’ K e n d r i c k , a n d P o n d e r (1 0 ) o n S e d i m e n t a t i o n
o f R e d B lo o d C e lls
E x p t l. R a t e o f S e d im e n
(4) t a t io n , % o f V a lu e E s t d .
«■ 0 - 3 f o r I n fin ite D iln . T h e o r e t ic a l R a t e o f S e d im e n ta tio n
C o r r e s p o n d in g t o E q u a t io n o f :
R a n g e fo r
A v era g e teste o n K erm a ck , T h is article,
c a lc d . b y v a r io u s M ’ K o n d r ic k , E q . 4 w ith
1 - l - e G a u d in (8 ) c e lls “ P on der G a u d in wi — 0 .1 6 7
where ^ = volume of solid plus immobile liquid per unit of
0 .0 0 2 6 97 9 7 -9 7 98 97 95
total volume. If the equation represents the data, a straight 0 .0 0 6 94 9 3 -9 5 96 95 94
0.01 89 8 6 -9 2 93 93 90
line having the slope 1 .S2 should be obtained when logioQ 0.02 83 7 8 -8 7 86 86 84
0 .0 4 73 6 7 -7 9 72 76 72
(t — tu‘ ) 2's P it te d against (e — iO i)/(l — w i), the complement 0 .0 8 51 4 5 -6 0 43 60 53
0 .1 6 20 2 2 -3 4 N e g .1 35 28
Of (1 — t ) / ( l “ thi).
0 .3 2 13 7 -2 2 N e g .6 8 7
This method of plotting was used in Figure 7 which presents ° C a lc u la t e d in te r m s o f G a u d in ’s a v e r a g e p e r c e n ta g e s b y a r b itra rily
a s s u m in g th e o b s e r v e d r a te a t 1 — < » 0 .0 0 2 5 to b e 9 7 % V t in e a c h case.
the data for the three preparations of emery. The points shown b T h e e q u a tio n w as d e r iv e d f o r lo w c o n c e n t r a t io n s o n ly .
for emeries A and B at (« — ia ,)/(l — » . ) = 1 were based on the
why Equation 3 applies as well as it does, for the equation is

based on the premise of flow past the individual particles. The
increase in w< may mean simply that the contacts or virtual con
tacts between particles which are evidently a feature o f the
flocculated state cause the amount of stagnant liquid per particle
to be somewhat greater than that in a dispersion.
A B S E N C E O F S T R U C T U R A L R E S IS T A N C E
The floe formation assumed above for concentrated suspensions,

is essentially a three-dimensional network extending fairly uni
formly throughout the whole volume o f the suspension, but
offering initially almost no mechanical stru ctu ral resistance to
collapse under the action of gravity. The assumption of negli
gible structural resistance seems necessary because flocculation
causes no change in the numerical factor o f Equation 3; support
for the assumption is provided by determinations of hydrostatic
pressure. That the hydrostatic pressure is an index to the
nature of the resistance is readily shown. When particles in a
liquid form a completely self-supporting structure, they do not
contribute to this pressure which is therefore determined by the
density o f the liquid in the usual way. But in the absence of
structural resistance, particles which settle without acceleration
are supported entirely by the liquid, and the hydrostatic prcssuro
is determined by the density of the mixture.
N o tests for hydrostatic pressure were made on suspensions of
0.3 OA 0 .5 0 .6 0.7 0.6 0.9 1.0
the sized emery powders, but fifteen determinations were made on
(Pf) other concentrated flocculated suspensions, most of which were
Portland cement pastes in which « was 0.58. Structural resistance
F igure 7. Logio ( ? [ ( ! — ««*')/(e — *»>') P v s . (e — u ),)/(l — u><) was thought to be more probable in such suspensions than in
for S e d im e n ta tio n o f N o n flo e cu la te d S u sp en sion s o f
those of inert, closely sized particles.
E m eries A , B , an d E
The hydrostatic pressures were determined by pouring the sus
pensions into a vessel in which a manometer tube, partly filled
with water and stoppered, was supported with its lower end well
since it was not experimental. It is concluded from Table V
above the bottom of the vessel. This manometer was a single
that the data can be represented as well by Equation 4 as by either
tube having an enlarged lower end fitted with a fritted glass disk.
of the others with which comparison is made.
Each suspension was brought to a level such that the head of
water in the manometer was a little less than that calculated for
S E D IM E N T A T IO N R A T E S O F F L O C C U L A T E D S U S P E N S IO N S
equilibrium. In a few minutes after the manometer was un
Emeries A and B were also tested when flocculated. Those stoppered, the water rose to a height which remained practically
tests made possible a direct comparison of effects in the two constant for some minutes and then declined at a rate that sug
states, but flocculated suspensions of a much finer emery powder gested development of boundary effects' at the tube. However,
were also tested to extend the basis for determining the effect of the maximum heights usually prevailed as long as 5 or more min
particle size. This finer emery (D ) was a commercial prepara utes after the suspensions were in place, which was long enough
tion. Its density was 3.38 grams per cc. The results of a size to ensure that the constant rates of sedimentation had been es-
analysis are shown in Table VI. In the tests for rate of sedimen
tation the 0 . 1 2 % solution of zinc sulfate was used as flocculant.
The tests were made at. 25° C. in the same way as those on emeries
T a b l e VI. S iz e A n a ly s i s o f E m e r y D f r o m S e d im e n t a t io n
A and B, except that they were discontinued before completion
in W a t e r , U s in g S o d iu m H e x a m e t a p h o s p h a t e a s D i s p e r s a n t
of settlement. The data are presented in Table V II.
D ia m e t e r , W e ig h t, D ia m e t e r , W e ig h t , D ia m e t e r ,
In Figure 8 , [<2(1 — e)]‘/i is plotted against e for the flocculated M ic r o n s % M ic r o n s % M icron B
•suspensions of emeries A, B, and D . As was true of the data for 5 0 -4 0 1 .5 1 2 -1 1 0 .0 7 -6
4 0 -3 0 2 .3 1 1 -1 0 0 .3 6 -5
the nonfloeculated suspensions, the points for each product can 3 0 -2 0 2 .5 1 0 -9 1 .0 5 -4
be represented by a single straight line, except for a few points at 2 0 -1 5 1 .1 9 -8 1 .4 4 -3
1 5 -1 2 0 .8 8 -7 1 .1 3 -0 «
the highest values of e. The slopes of the lines were calculated
S p . su r fa c e , e a le d . as f o r s p h e r e s “ , sq . c m . / c c . 1 2,950
from Equation 3, the best values of wi; being determined by trial. A v . d ia m ., c a lc d . fr o m s p . s u r fa c e , m ic r o n s 4 .6
On the whole, there is fair agreement with the equation. Since a A v e r a g e p a r t ic le d ia m e te r o f th is p o r t io n ta k e n as 2 m ic r o n s s in c e m ic r o
Equation 3 was derived for nonflocculated suspensions, this s c o p i c e x a m in a t io n in d ic a te d a b s e n c e o f e x tr e m e fines.
agreement indicates a considerable degree of similarity in the

sedimentation of the flocculated and nonflocculated suspensions.
T a b le V II. S e d im e n t a t io n o f F l o c c u l a t e d S u s p e n s io n s o f
Nevertheless, it is evident from Table II that the flocculation E m ery D
materially reduced the rate o f settlement at most concentrations. F lu id F lu id
This is reflected in increased values o f w>. C o n te n t, R a te of C o n te n t, R a te of
% o f F lu id S e t t le m e n t % o f F lu id S e t t le m e n t
This effect of flocculation on the rate of settlement in concen T est N o. m T ota l ( C m ./ S e c . ) T est N o. in T o t a l ( C m ./ S e c . )
(C h ro n o - V o l. X 10« (C h r o n o V o l. X 10«
trated suspensions is the opposite of the effect in dilute ones. In lo g ic a l) (« X IO*) (Q X 10«) lo g ic a l) (« X 10*) (Q X 10«)
contrast, to dilute suspensions, highly concentrated ones may 1 6 7 .5 95 8 7 5 .0 246
develop so continuous a floe structure that all liquid displaced by 2 7 0 .0 112 4 7 7 .5 326
3 7 2 .5 183 5 8 0 .0 438
the settlement must make its way through the actual floe texture 7 7 2 .7 171 6 8 5 .0 956
where the resistance is high. Such an assumption helps explain
leave w P) / ( l — w P) as the proportion o f the actual floe space

occupied by liquid. This quantity has the same form as (e — wi) /
( 1 — wi) which was previously shown to represent the mobile
liquid when Wi results from stagnant liquid.
Liquid in pockets is mentioned merely as a possibility, for the
rate equation is explained more satisfactorily entirely on the basis
of the stagnant liquid which, in any case, appears to be responsible
for most of w%.
The basic assumption that there are no channels for liquid to
escape without passing through the fioc texture is in accord with
direct observations. Powrers, in his work on cement pastes,
usually found no evidence of significant breaks or channels in the
paste structure in tests in which his data were in agreement with
his rate equation; at higher dilutions, channels and “ boils” could
easily be seen, but then the rates were higher than those pre
dicted by the equation. These observations have been con
firmed b y the writer. Also, many indications o f boundary resist
ance which have been obtained in the work on cement pastes are
regarded as further ovidonce of the continuous nature of the floe
structure. In none of the tests on the emery suspensions wore
any channels seen. However, it is believed that the points in
Figure 8 that were obtained at high dilution and that trend up
ward from the straight lines representing the rest of the data do
so because the less concentrated suspensions were too dilute to
prevent breaks in tho structure, breaks too fine to be seen readily.
E F F E C T O F P A R T IC L E S I Z E IN F L O C C U L A T E D S U S P E N S IO N S
Particle size affects the value o f V ,; that it also affects the

value o f W i when the powders are flocculated is suggested by
Figure 8 . However, judgment as to the generality of this
relation should bo reserved until additional data are presonted in
the third article of this series. The fine emery D , for which wi
is much the largest, had relatively low density and was evidently
less pure than the other emeries. It was also loss uniform in
particle size, on a percentage basis. The w i values for flocculated
F ig u re 8. [(K l-e)l1/* os- 6 fo r S e d im e n ta tio n o f F lo c suspensions of the various emery powders follow:
cu la te d S u sp en sion s o f E m eries A , B , a n d D
S p . S u r fa c e by
S e d im e n ta tio n A v e ra g e ioi fo r
M eth od , D ia m e te r , F lo c c u la te d
E m ery S q . C m . /C o . M ic r o n s S u sp e n sion ?
tablished. In general, the equilibrium heights were within =*=1.5% A 4,930 12.2 0 .2 6 8
o f those calculated by assuming no structural resistance. As B 6,250 9 .6 0 .2 8 8
D . 12,950 4 .6 0 .3 5 0
the probable experimental error was estimated to be about the
same magnitude (1 .5 % ), the absenco of appreciable structural re
sistance was considered to have been demonstrated. Among recent investigations of flocculated suspensions, those
o f Egolf and M cCabe ( 6 ), Ward and Kammermeyer (17), and
W ork and Kohler (1 8 ) relate to lower concentrations than those
FLOC STRU CTU RE dealt with hero. The work of Powers (1 4 ) on the sedimentation
or “ bleeding” of portland cement pastes is apparently the most
The assumption that a rather uniform network of particles
extensive previous study of highly concentrated flocculated sus
fills all tho space in concentrated flocculated suspensions makes
pensions with which comparison can be made. Powers’ rate
no allowanco for any liquid outside the floe space proper. It is
equation differs in some respects from Equation 3, but the differ
presumed that concentrations for which the data can bo rep
ences can be reconciled, as will be shown in the third article of
resented by Equation 3 are so great that there is no tendency for
this series.
the floe to occupy less than the total available volume. How
Although the data o f Egolf and M cCabe on 16-micron silica
ever, the data might still bo in approximate agreement with
are for more dilute suspensions, they can also be represented
Equation 3 if the fine structure were broken by pockets of liquid
linearly by plotting [Q (l — e)]l / ’ against e. Apparently, after a
o f relatively large dimensions. The existence of random iso
break in slope like that shown for emery A in Figure 8 , another
lated pockets would not conflict with the apparent requirement
linear relation often obtains.
that liquid escaping from the settling mass should pass through
the floe meshwork and contact the individual particles. The C O N C L U S IO N S
pockets would tend to produce nonuniformity in the flow, but
there would probably be readjustments in floe texture, and the The effect o f particle concentration on rate of sedimentation
rate of sedimentation might not be far from that of a suspension in nonflocculated suspensions settling under conditions of viscous
wholly occupied b y floe of the same concentration as that in the flow' is not the same for angular particles as it is for spheres.
floe space proper. Accordingly, it is o f interest that liquid in However, the results for the two kinds o f particles can be brought
pockets could contribute to u>< in the same way as liquid stagnant into complete agreement b y assuming that the angular particles
at angularities in the individual particles. That is, if liquid in carry with them a volume o f liquid w'hich is a constant proportion
pockets amounted to w r per unit of total mixture, this would of the volume of the solid at all concentrations. The quantity
N OM ENCLATURE
per unit volume of solid appears to be a function only of particle
shape. For different preparations of emery the indicated quan Q = initial rate of settlement of top surface of a suspension,
tities were 20 and 2S% of the volume of solid. Several lines of cm ./sec.
V. = rate of fall of an isolated sphere as given by Stokeslaw,
evidence are presented to show that this liquid is not adsorbed
cm ./sec.; used also to represent rate o f fall of any iso
by the particles. Apparently it is simply liquid that has re lated particle when Reynolds number is such that a
mained stagnant at angularities in the particles. sphere would fall in accordance with Stokes law
Flocculation of concentrated suspensions of the same emery Wi = dimensionless constant used by Powers (1 4 )
i 0p — dimensionless constant analogous to Wi
powders that were tested in the nonflocculated state materially
a = dimensionless constant denoting volume of fluid, pe
reduced the rates of sedimentation. However, the same rate volume of solid, that appears to remain with angular
equation could be applied with fair success when modified only particles during their fall
by an increase in v k which, for the nonflocculated suspensions, e = proportion of total volume of suspension occupied by liquid
(analogous to porosity in beds of particles)
corrects for the liquid that is assumed to be stagnant. It is con
cluded that in highly concentrated flocculated suspensions there
is no opportunity for liquid to by-pass floe structure in escaping
from the mass during sedimentation, but that in general this (1) Bastow, S. H., and Bowden, F. P., Proc. Roy. Soe. (L on d o n ).
A151, 220-33 (1935).
liquid must flow past the individual particles in much the same
(2) Bulkloy, R ., Bur. Standards J. Research, 6, 89-112 (1931).
way as in a nonflocculated suspension. T w o possibilities aro (3) Carman, P. C ., J. Agr. Sci., 29, Pt. 2, 262-73 (1939).
seen for the increase in w;: (1) The quantity of stagnant liquid (4) Carman, P. C., J. Soc. Chem. Ind., 57, 225-34T (1938).
may be increased by reason of interparticle contacts caused by (5) Carman, P. C ., Trans. Inst. Chem. Engrs. (London), 16,1
(1938).
flocculation. (2) There may be isolated pockets of liquid dis
(6) Egolf, C. B., and M cCabe, W . L., Trans. A m .In st. Chem. Engrs.,
tributed through the mass, pockets whose dimensions are ma- 33, 620-42 (1937).
teriallj' greater than those of the meshes in the floe structure (7) Fair, G . M ., and Hatch, L. P., J. Am . Water Work 3 Assoc., 25,
itself. How such pockets can produce an increase in Wi has been 1551-65 (1933).
(8) Gaudin, A . M ., ‘ ‘Principles of Mineral Dressing” , 1st ed.,
shown.
Chap. 8, New York, McGraw-Hill Book Co., 1939.
The possibility that structural (nonviscous) resistance pro (9) Johansen, F. C., Proc. Roy. Soc. (London), A126, 231-45 (1930).
duced by the flocculation may have been a factor in decreasing (10) Kermnck, W . O., M ’Kendrick, A. G., and Ponder, Erie, Proc.
the rates of settlement was investigated by determining hydro Roy. Soc. Edinburgh, 49, 170-97 (1929).
(11) Kozeny, Josef, Kullurtechniker, 35, 478-86 (1932).
static pressures in flocculated suspensions of various powders,
(12) Kozeny, Josef, Sitzber. Akad. W iss.W ien, Ila , 136,271-306(1927).
but principally in aqueous pastes of portland cement. It was (13) Lea, F. M ., and Nurse, R . W ., J . Soc. Chem. Ind., 58, 277-83T
found that there is no appreciable structural resistance during (1939).
the initial stage of the sedimentation when the constant rate is (14) Powers, T . C., Research Lab., Portland Cement Assoc., Bull.
2 (1939).
established.
(15) Steinour, H. H ., I n d . E n o . C hem ., 36, 618-24 (1944).
(16) Wagner, L. A., Proc. Am . Soc. Testing Materials, 33, Pt. II,
ACKNOW LEDGM ENT 553-70 (1933).
(17) W ard, H . T., and Kammormeyer, Karl, I n d . E n o . C hem ., 32,
The writer was assisted by Richard G. Brusch and Herbert W. 622-6 (1940).
Schultz in the experimental work reported in this article. (18) Work, L. T „ and Kohler, A. S., Ibid., 32, 1329-34 (1940).
PRINTING INKS
from Colloidal Dispersions of
ALFRED F. SCHMUTZLER AN D
DONALD F. OTHMER
SOYBEAN PROTEIN
Polytechnic Institute, Brooklyn, N. Y.
was found that some features of the hardening of casein could
be applied to soybean protein.
P R O T E IN dispersions in diethyleno glycol were described in In order to improve the hardness and durability o f casein
a previous article (¡S3). Printing inks made from these dis fibers (artificial wool, lanital), in some instances small amounts
persions are nondrying on the printing press, but when of salts o f heavy metals are added to the spinning solution (10,
printed films of the inks are exposed to steam, they harden im 12, 13, 14 , 2 0 ) ) in others the dispersed casein is reacted with
mediately. They are not sufficiently water resistant for uni isocyanates (5, 15, 17, 18), isothiocyanates (11, 16 ), and carbon
versal use. T o eliminate this defect, reactions designed to lower disulfide (6). Although isocyanates may undergo side reactions
the water absorption of proteins were investigated. in the presence o f hydroxyl groups, isothiocyanates and carbon
In aqueous dispersions, soybean protein will not precipitate disulfide combine solely with amino and amide groups. The
or gel when reacted with formaldehyde (2 6 ). Plastics made from initial step is the formation of thiourea derivatives, followed
it have relatively high water absorption (1 ~ 4 ). In contrast to by condensation reactions with aldehydes, with the formation
soybean protein, casein and blood albumin gel almost immediately of modified thiourea resins (9, 19, 2 4 ). In the reaction of pro
upon the addition of even small amounts of aldehydes to their teins with carbon disulfide, hydrogen sulfide is liberated and can
aqueous dispersions, an indication that the behavior of the former combine with aldehydes, with the formation of thioaldehydes,
differs considerably from that of the latter. Nevertheless it which are much more reactive than the aldehydes from which
P rin tin g in k s w h ich h ard en im m e d ia te ly b y ex posu re to The; tests for hardness and water resistance were made with
ste a m w ere m a d e w ith c o llo id a l d isp ersion s o f a ld e h y d e - printing inks consisting of 30 grams of Lithol red and 70 grams of
reacted soy b ea n p ro te in in d ie th y le n e g ly c o l. T h e “ h a r d dispersion. The two were mixed uniformly and milled on a three-
e n e d ” p r o te in h as a relatively h ig h w a ter a b s o r p tio n . T h is roller laboratory paint mill. Printing was done on a Vandercook
d isa d v a n ta ge w as e lim in a te d b y r e a c tin g th e dispersed proofing press (N o. 219) equipped with split rollers, so that two
p ro te in in th e p resen ce o f c a ta ly sts su ch as a lu m in u m inks could be printed simultaneously for comparative testing.
c h lo r id e a n d z in c n itra te , an d o f p r o te in -p r e c ip ita tin g The prints were passed over steam issuing from a steam bath in
reagents s u c h as p h e n y l is o th io e y a n a te a n d p h e n o lic res order to set the ink to a hard film.
in s. T h e resu lts in d ic a te th a t m a terials so form ed m a y The relative hardness of the printed films were determined on
he u sefu l in o th e r fields. the scratch tester (Figure 1) designed by J. G. Curado for the
research laboratories of General Printing Ink Corporation. A
<3> bead of Pyrex, 1/ l6 inch in diameter, is attached to the lower
end of a vertical rod; at the other end is a small platform for
they are derived; these thioaldehydes are also subject to spon holding weights. By means of a counter-balancing spring at
taneous condensation with themselves (.9). tachment, the bead exerts practically no pressure upon the print
Although the mechanism of the hardening of casein and other when the platform is not weighed down. Tw o rollers, whose
proteins with aldehydes is unknown, reactions o f urea (7, 8, 2 2 ) rate of rotation can be adjusted b y a rheostat, pull the printed
and melamine (1 6 ) with formaldehyde suggest that the amino sheet from underneath the bead, which scratches the printed
groups m ay form monomethylolamines, dimethylolamines, or film if it is weighed down sufficiently.
methylene linkages. The gelling of casein suggests the forma Thus the appearance of the scratched surface indicates the
tion of a larger complex, with adjacent polypeptide chains sup relative hardness of the film; but if the film is so hard that suffi
posedly joined together b y methylene linkages. cient weight may not be used without tearing the paper, the bead
will usually pick up some pigment from the print and deposit it
E X P E R IM E N T A L P R O C E D U R E
on the unprinted part of the sheet; this also serves as a means of
The object o f these experiments was to obtain dispersions of visual comparison of the relative hardness of several inks.
aldehyde-reacted proteins in diethylene glycol, suitable for The water resistance of the dry ink films was tested by pulling
hard-drying and water-resistant printing inks. The dispersions a wet piece of filter paper over the dry print. Whatman filter
(Table I) and additions were heated in an Erlenmeyer flask under paper No. 1, 3 inches in diameter, was soaked in water, and the
a reflux condenser. The contents were shaken vigorously at 10- excess water was allowed to drain off. The reaction of the
minute intervals in order to duplicate occasional stirring, custom printed proof to this test was regarded as a measure of its relative
ary in the preparation o f varnishes. After completion of the resistance to water.
reaction and cooling, 2 grams of thé mixture were diluted with
100 cc. o f distilled water. The dilution was allowed to settle P R E C IP IT A T E F R O M W A T E R D IL U T IO N
for 24 hours and filtered through a tared Gooch crucible; the It was found that the amount of precipitate obtained by
bottom of the crucible was covered with a small circular piece diluting the protein dispersions with water was a measure of the
of acid-washed filter paper. The collected precipitate was dried relative water resistance of the binder in the printing ink. A c
at 90° C. for 20 hours. The amount of precipitate obtained cording to Figure 2, the maximum amount of precipitate in both
was accepted as a measure of the degree of water resistance of
the binder in the printing ink vehicle.
T o define the consistency o f the dispersions, the relative vis
cosity was determined in a Stormer viscometer; the readings in
seconds were converted to poises by a calibration chart (2 3 ).
These values differ considerably from those of the viscosities of
oils and varnishes; their importance depends upon the fact that,
with the same concentration of the same pigment, the Iow-
viscosity dispersions give softer inks than the more viscous dis
persions.
As a general rule, the weight of the plummet for the Stormer
viscometer depended upon the consistency of the dispersion—
e.g., plummet of 100 grams, up to 5 poises; 200 grams, up to 15
poises; 500 grams, up to 60 poises; and 10 0 0 grams, above 60
poises.
The pH was read on a Beckman meter, to define whether the
dispersions were acid or alkaline with respect to diethylene glycol
(pH 6 .8 6 ). The readings on colloidal dispersions of proteins in
diethvlene glycol do not show the true hydrogen-ion concentra
tion; small additions of either guanidine carbonate or acetic
acid to diethylene glycol cause greater deviations from the pH
of 6.86 than the same amounts cause from a pH of 7.0 in double F ig u re 1. C u ra d o S cra tch T ester
distilled water.
Color and appearance depend upon
pigments, while hardness and water
resistance of the printed films are T a b l e I. S o y b e a n P r o t e in D is p e r s io n s
imparted by the binder; in this in Soybean G u a n id in e D ic y a n - P h th a lic D ie t h y le n e R e a c t io n R e la t iv e B e ck m a n

E xpt. a -P r o t c in , C a r b o n a t e , d ia m id e , A n h y d r id e , G ly c o l, T e m p ., V is c o s it y , pH
stance, they are due to the aldehyde- N o. G ra m s G ra m s G ra m s G ra m s G ra m s ° C. P o ise s R e a d inn
reacted protein, which occludes 172 (£3) 20 6 7 4 .0 0 135 1 8 .9 1 0 .4
the pigment and precipitates from 216 20 6 3 .7 0 7 0 .3 0 135 2 5 .5 8 .7
225 20 6 6 8 .0 0 135 2 8 . 0 1 1 .4
its colloidal dispersions upon ex 286 20 3 3 7 4 .0 0 140 2 1 .3 9 .8
posure of the printed film to steam.

alkaline and acid dispersions is obtainable when between 4.5

T a b le II. A c t io n o f P a r a fo r m a ld é h y d e o n A lk a lin e
and 5 grains of paraformaldehyde had been reacted with 100
D i s p e r s i o n 172
grams of 2 0 % soybean protein dispersion; with either more or
(100 gra m s o f d is p e rs io n 172 a n d v a r y in g a m o u n ts o f p a r a fo r m a ld e h y d e
h e a te d 2 h ou rs a t 110 ° C . u n d e r re flu x co n d e n s e r ) less paraformaldehyde the amount of precipitate is less. Tables
P r in tin g In k s* II, III, and IV show that it is somewhat greater in acid than in
D is p e rs io n s R e d im
R e la t iv e A v . "w a te r- p ression
alkaline dispersions but relatively low with respect to total
P a r a fo r m
E x p t. a ld e h y d e , ( if c c k - v is c o s it y , in 8 o l.’ \ g . /g . In k w ith lo a d solids in the dispersions, even in the presence of protein-precipitat
N o. G ra m s m an ) p o is e s 0 o f s a m p le no. m gram s
ing aluminum chloride and zinc nitrate.
172 0 1 0 .4 1 8 .9 0 .0 0 1 9 172R 0
202 3 .0 0 8 .9 • 4 3 .7 0 .0 2 3 7 202R 5 According to Table V, a larger amount is obtainable from
203 4 .0 0 8 .9 5 5 .7 0 .0 2 4 3 203R 10
5 .0 0 8 .4 5 0 .4 0 .0 2 5 5 204R 20
dispersions in which the protein is reacted with carbon disulfide
204
205 6 .0 0 8 .2 4 2 .6 0 .0 2 6 1 205R 20 before the reaction with paraformaldehyde, but the disagreeable
206 7 .0 0 8 .0 4 0 .3 0 .0 2 3 7 206 R 10
207 8 .0 0 8 .0 3 6 .8 0 .0 2 1 5 207R 5 odor freed during the reaction forbids its use in printing inks.
208 9 .0 0 8 .0 3 1 .6 0 .0 1 8 3 208R 0
209 1 0 .0 0 8 .0 3 0 .9 0 .0 1 4 3 209R 0
Ammonium nitrate slightly improves the odor but not sufficiently.
210 3 .5 0 8 .8 5 3 .7 0 .0 2 4 1 2 1 0 R 10 I t shifts the maximum amount of precipitate toward the 6 -gram
4 W ith 5 0 0 -g r a m p lu m m e t. & B u tt e r y a p p e a r a n c e , p o o r w a te r r e sista n ce .
range on Figure 2.
Table V I shows that relatively large amounts of precipitate
T a b le III. A c t io n o f P a r a fo r m a ld e h y d e on A lk a lin e
D i s p e r s i o n 216 are obtainable in experiments made with dispersion 286, in which
( 1 0 0 'g ra m a o f d isp e r sio n 216 a n d v a r y in g a m o u n ts o f p a r a fo r m a ld e h y d e ,
a mixture of diçyandiamide and guanidine carbonate served as
h e a t e d f o r 2 h o u rs a t 1 1 0 ° C . u n d e r reflu x) the peptizing agents for soybean protein. In contrast to the
P a r a fo r m R e la t iv e A v.
E x p t. a ld e h y d e , pH V is c o s it y , “ W a t e r -I n s o l.” , other dispersions of Table I, which foam profusely during acidi
N o. G ra m s (B e c k m a n ) Poises® G . / G . o f S a m p le fication, dispersion 286 foams but slightly during the addition
216 0 8 .7 2 5 .5 0 .0 0 1 7 of acetic acid. After reaction with paraformaldehyde in the
217 3 .0 0 7 .8 3 6 .7 0 .0 2 6 3
218 4 .2 5 7 .4 3 8 .8 0 .0 4 4 9 presence of phenyl isothiocyanate, it yields large quantities of
219 5 .5 0 7 .3 3 8 .9 0 .0 4 6 7
220 6 .6 0 7 .2 3 9 .0 0 .0 3 5 1
precipitate on dilution with water; still larger amounts are ob
221 6 .8 0 7 .1 3 9 .6 0 .0 3 4 3 tainable when the reaction with paraformaldehyde is carried out
222 7 .2 0 7 .1 3 1 .6 0 .0 3 3 3
223 7 .6 0 7 .0 2 9 .3 0 .0 3 1 8 in the presence of a small percentage of phenolic resin Amberlite
224 8 .0 0 7 .0 2 8 .1 0 .0 3 0 5
K 6 S. This resin was selected from several diethylene-glycol-
W it h 5 0 0 -g r a m p lu m m e t.
soluble phenolic compounds listed in Table V II.
T a b le IV. A c t io n o f P a r a fo r m a ld e h y d e o n A c id Tables V I and V II demonstrate the relation between the
D i s p e r s i o n 225 amount of precipitate from water dilutions of the protein disper
(94 gram a o f d is p e r s io n 2 2 5 a n d 6 g ra m s o f g la c ia l a c e t ic a c id h e a te d fo r 2 sions and the hardness and water resistance of the printed films
h ou rs)
made with corresponding printing inks.
P a r a fo r m A nhy R e la t iv e “ W a ter-
E x p t. a ld e h y d e , d ro u s A lC lj, V is c o s it y , I n s o l.” , G ./ G .
t, P H
N o. G ra m s G ra m (B e c k m a n ) P o ise s o f S a m p le T E S T S O N P R IN T IN G IN K S
R ea cted a t 110 ° C .
In Table I I dispersions 204 and 205 yield the greatest amount
225A 0 0 6 .0 2 7 .0 0 .0 4 1 8
226 4 .0 0 6 .0 1 6 .9 0 .0 4 8 5 of precipitate; the corresponding inks, 204R and 205R, gave the
227 5 .0 0 5 .9 1 5 .8 0 .0 5 0 3
228 5 .8 0 6 .0 1 6 .4 0 .0 4 5 2
229 6 .8 0 6 .0 1 8 .1 0 .0 4 4 0
230 4 .5 0 6 .0 1 6 .6 0 .0 4 9 3
231 5 .3 0 5 .9 1 6 .0 0 .0 4 7 7
R e a cte d a t 130 ° C.
232 4 .5 0 6 .0 1 6 .8 0 .0 4 9 9
233 5 .0 0 5 .9 1 6 .4 0 .0 5 1 2
234 5 .3 0 6 .0 1 6 .7 0 .0 5 0 1
235 5 .6 0 6 .0 1 6 .9 0 .0 4 8 5
236“ 4 .5 1 .0 5 .8 1 9 .5 0 .1 0 7 6
237° 5 .0 1 .0 5 .7 1 9 .8 0 .0 9 9 8
238“ 5 .3 1 .0 5 .7 2 0 .1 0 .0 5 7 6
4 A sm a ll a m o u n t o f s c u m w h ic h fo r m e d w as r e m o v e d , a n d tests f o r w a te r-
in so lu b le s u b s ta n c e s w ere c a r r ie d o u t w ith t h e cle a r d is p e rs io n .
T able V. A c t io n o f C a r b o n D is u lfid e a n d P a r a fo r a m ld e h y d e
o n D i s p e r s i o n 225
A d d it io n s t o 94 G ra m s o f D is p e r s io n 2 2 5 , G ra m s "W a te r-
D ie t h y l R e la t iv e I n s o l.” ,
E x p t. ene A c e t io Z n (N a O )r P a r a fo r m (£ e ck - V is c o s it y , G ./G . of
N o. g ly c o l a c id N H «N O * 6 H jO CSi a ld e h y d e m an) P o is e s S a m p le
239° 6 0 10 0 7 .4 3 1 .6 0 .0 7 1 7
240“ 5 1 10 5 6 .7 1 7 .1 0 .0 3 8 1
241 ° 3 3 10 5 6 .5 1 2 .2 0 .0 5 7 2
242 ° 2 . 4 10 5 6 .4 1 7 .6 0 .1 2 7 5
243b 1 5 10 5 6 .1 1 7 .6 0 .1 1 7 5
244 b 3 3 10 5 6 .5 2 1 .2 0 .1 3 9 0
245b 6 0 10 5 7 .3 4 6 .3 0 .1 2 5 0
246b 5 1 10 5 6 .5 4 1 .8 0 .1 6 0 6
247 b 4 2 10 5 6 .5 3 7 .3 0 .1 4 8 2
263 & è’ * 5 8 4 .7 6 .2 7 0 .0 4 1 0
264b 6 8 8 4 .7 5 .9 2 0 .0 3 9 8
265b 6 11 8 4 .7 5 .7 7 0 .0 4 0 3
266 b 6 15 8 4 .8 5 .6 3 0 .0 4 3 8
267& 5 .5 0 .5 5 8 4 .8 ' 9 .3 1 0 .0 4 5 6 Addition of Poroformoldehyde to
268b 5 1 5 8 4 .7 1 0 .1 5 0 .0 4 8 2
269 b 4 2 5 8 4 .7 1 2 .9 0 0 .0 5 8 6 2 0 g. P r o t e i n in Dispersion
270b 3 3C 5 8 4 .9 1 7 .7 3 0 .0 7 2 4
271b 4 2 5 8 4 .7 1 2 .9 0 0 .0 5 8 6
272b 4 2 5 5 5 .0 1 0 .7 2
273 b
0 .0 8 5 0 F ig u re 2. E ffect o f P a ra fo rm a ld e h y d e (G ra m s
4 2 5 6 4 .9 9 .3 0 0 .0 9 0 8
274 b 4 2 5 7 4 .8 1 6 .6 3 0 .0 7 0 8
A d d e d ) o n W a ter In s o lu b ilit y o f S oy b ea n
275b 4 2 5 9 4 .7 1 9 .2 2 0 .0 5 2 0 P ro te in in D ie th y le n e G ly c o l D isp ersion
. “ C S j, a c id , a n d d isp e rsio n 2 2 5 r e a c te d f o r 1 h o u r a t 4 0 ° C .: r e m a in d e r o f ite m s r e a c te d f o r 2 W e ig h t o f w a t e r -in s o lu b le r e s in f o r m e d in 50 m l . o f
hours a t 130 ° C .
w a te r p e r g r a m s a m p le o f d is p e r s io n . N u m b e r s r e fe r
b A ll ite m s m ix e d a n d h e a te d f o r 1 h o u r a t 4 0 ° C ., fo llo w e d b y 2 h o u r s a t 1 3 0 ° C . t o e x p e r im e n t s lis t e d i n ta b le s . A s te r is k in d ic a te s
c A sm a ll a m o u n t o f s c u m fo r m e d . e ffe c t o f s h i ft o f m a x im u m p o in t b y a c t io n o f carbon
d is u lfid e .
T a b l k V I. A c t io n o f P a r a fo r m a ld é h y d e o n D is p e r s io n 286
A d d it io n s t o 100 G ra m s o f D is p e r s io n s P r in t in g In k s
D is p e r sio n z ö ö , u r a m s » "W a ter- R e d im
Zn Phenyl A m b e r lite R e la t iv e in s o l.” , pre ssio n
E x p t. n itr a te is o th io K 6S P a r a fo r m pH ‘ v is c o s it y , g - /g . o f In k B e h a v io r w ith lo a d W a te r
N o. 8 o ln .a cy a n a te s o i n .1* a ld e h y d e (B e c k m a n ) p o is e s s a m p le no. A p p earan ce o n press in g ra m s resistan ce
286 0 9 .8 2 1 .3 0 .0 0 6 9 286 R 1 5 P oor
287 e 5 6 6 .2 3 2 .3 0 .0 7 0 4 287R 10 P oor
288 e 5 5 6 .2 3 9 .5 0 .0 8 9 5 288R S lig h tly F a ir 20 Poor
2 89 e 5 4 6 .2 3 8 .0 0 .0 9 4 8 289R ta ck y 20 P oor
200e 5 4 .5 6 .2 4 6 .4 0 .1 0 0 2 290R 20 P oor
291e 5 5 .5 6 .1 3 8 .1 0 .0 8 7 2 291R 20 P oor
2 92 e 5 7 6 .1 3 7 .5 0 .0 4 8 8 292R 10 P oor
2934 5 3 » .. 4 .4 7 .0 1 1 1 .0 0 .1 2 0 6 293R 30 F a ir
294d 5 3 4 .7 6 .7 1 1 5 .1 0 .1 2 2 6 294R M o d e r a t e ly F a ir 30 F a ir
295d 5 3 5 6 .7 1 1 1 .6 0 .1 1 7 2 295R ta ck y 30 F a ir
296* 5 3 4 .2 6 .7 1 1 2 .0 0 .1 1 9 8 296R 20 F a ir
297 e 5 14 4 6 .5 9 8 .5 0 .1 8 6 4 2 9 7 II G ood, 50 G ood
298 e 5 14 5 6 .4 1 5 7 .2 0 .1 8 5 8 298R V e ry ta ck y b e lo w 4 5 % 50 G ood
2 99 e 5 14 5 .5 6 .5 2 7 2 .0 0 .1 8 7 4 299R h u m id ity 50 G ood
300 e 5 14 6 6 .2 2 7 5 .0 0 .1 9 9 0 300R 50 G ood
a C o n s is tin g o f 5 0 % zin c n itra te h e x a h y d r a t e a n d 5 0 % g la c ia l a c e t ic a c id b y w e ig h t.
b C o n s is tin g o f 6 0 gra m s A m b e r lite K 6 S a n d 8 0 gram a d ie t h y le n e g ly c o l.
e I te m s m ix ed a n d h e a te d f o r 2 h o u rs at 1 3 0 ° C .
A P h e n y l is o t h io c y a n a t e a n d d is p e r s io n 2 8 6 m ix e d a n d h e a te d a t 1 3 0 ° C . f o r 10 m in tu e s ; z in c n itra te s o lu t io n a n d p a r a fo r m a ld e h y d e a d d e d , a n d m ixtu re
h e a te d f o r 2 h ou rs a t 1 3 0 ° C .
T a b l e V II. A c tio n o f P a r a fo r m a ld e h y d e a n d P h e n o ls o n The behavior on the printing press of ink 324R made from this
D i s p e r s i o n 22 5“
R e la t iv e vehicle was good up to 7 0 % humidity. Scratch tests showed im
E x p t. P h e n o lic pH V is c o s it y , “ W a t e r -I n s o l.” , pressions with a load of 60 grams, and the water resistance was al
N o. C om pound (B e c k m a n ) P o is e s G . / G . o f S a m p le
most as good as that of inks 297R to 300R. This example indi
27 6 U .S . P . phen ol 6 .2 1 7 .6 0 .0 9 4 8
277 A m b e r lite K 6 S 6 .0 1 2 5 .1 0 .1 8 0 6 cates the possibilities in the application of soyboan protein
27 8 S u p c r -B e c k a c it e 3 0 0 0 * 6 .1
279 N e v illa c hard® 6 .1 3 6 .9 0 .1 6 2 6 dispersions to inks also containing dispersions of other proteins.
a M ix tu r e : 94 g ra m s o f d is p e rs io n 2 2 5 , 6 o f d ie t h y le n e g ly c o l, 6 o f g la c ia l
C O N C L U S IO N
a c e t ic a c id , 6 o f p a r a fo r m a ld e h y d e , a n d 6 o f p h e n o lic c o m p o u n d . R e sin
d is so lv e d in d ie th y le n e g l y c o l; r e m a in d e r o f ite m s a d d e d , m ix e d , a n d h e a te d In the preparation of printing ink vehicles, best results can be
fo r 2 h o u rs a t 130 ° C .
b R e s u ltin g p r o d u c t u n s ta b le ; r e s in o u s m ix tu re s e p a r a te d fr o m disp e rsio n . obtained when an optimum amount of paraformaldehyde is
e A m o d ifie d c o u m a r o n e -in d e n e resin w ith s e v e r a l p h e n o lic h y d r o x y l reacted with the dispersed soybean protein in diethylene glycol.
g r o u p s (2 1 ). According to the types of dispersions used, the optimum amount
lies within the range 4.5 to 5 grams of paraformaldehyde per 100
grams of 2 0 % dispersions.
The characteristics of these dispersions suggest a variety of
hardest printed films, an indication o f the importance of the maxi other applications, such as specialty paints, paper sizing, and
mum range of paraformaldehyde additions o f Figure 2 . The de artificial fibers. It might be necessary to change to lower-boiling
crease of the precipitate with an excess of paraformaldehyde alcohols, such as ethylene glycol, but the capacity of the disper
sions of aldehyde-reacted soybean protein and corn protein to
may be explained by the rule formulated for the solubility of
liquefy casein-formnldehyde gels, appears to be of varied indus
organic compounds; i.e., greater structural similarity between trial importance.
solute and solvent is accompanied by greater solubility (1 9 ).
It is assumed that with an excess of paraformaldehyde many
(1) Arnold, L. K ., and Quackenbush, A. D ., Proc. Iowa Acad. Sd..
methylol groups are formed and thus cause relatively greater 47, 231-4 (1940).
solubility of the aldehyde-reacted protein in water; conversely, (2) Beckel, A. C., Brother, G . R ., and M cK inney, L . L., I n d . E n q .
such a protein binder retains larger proportions of diethylene gly C hem ., 30, 1236-40 (1938).
col after “ steam-setting” of the printed films. (3) Brother, G . H ., and M cK inney, L . L., Ibid., 31, 8 4 -7 (1939).
(4) Brother, G . H ., and M cK inney, L. L., U. S. Patent 2,262,422
The dispersions of Table VI give the m ost satisfactory printing (1942).
inks. Phenyl isothiocyanate imparts good water resistance and (5) Dickson, J. P., Brit. Patent 523,759 (1940).
hardness; still greater hardness and water resistance are obtain (6) Donagemma, Giusseppe, Ibid., 505,756 (1939).
able with the phenolic resin Amborlite K 6 S. The inks made (7) Einhorn, A., A nn., 343, 207 (1905).
(8) Einhorn, A ., and Hamburger, A ., Ber., 41, 24 (1908).
with dispersions containing this resin are sensitive to humidity (9) Ellis, Carloton, “ Chemistry o f Synthetic Resins” , Now York,
above 4 5 % ; at high humidity the ink fails to distribute on the Reinhold Publishing Corp., 1935.
printing press. Upon the addition o f small amounts of diethyl- (10) Elod, Egon, and Schmitt, Gustav, Kolloid-Z., 92, 105-12 (1940).
eno glycol monophthalate, these inks can be used at a humidity (11) Esselmann, Paul, and Dusing, Josef, U. S. Patont 2,220,441
(1940).
greater than 45 % . (12) Ferreti, Antonio, Brit. Patent 483,731 (1938).
(13) Gohda, Tadashi, French Patent 767,874 (1934).
IN K V E H IC L E S F R O M M IX T U R E S O F T W O D IS P E R S IO N S (14) Gould, S. P., and Whittier, E. O., Ibid., 841,503 (1939).
(15) Groves, W . W ., Brit. Patent 509,852 (1939).
Dispersions of casein and blood albumin in diethylene glycol (16) Hodgins, T . S., H ovey, A . G ., H ewett, S., Barrett, W . R ., and
turn to gel-like masses on intimate contact with paraformalde Meeske, J. C., I n d . E n c j. C hem ., 33, 767-79 (1941).
hyde. When either of these gelled dispersions is mixed with an (17) I. G . Farbonindustrie, A .-G ., French Patent 840,773 (1939).
equal amount of a liquid dispersion of aldehyde-reacted soybean (18) Ibid., 844,289 (1939).
(19) Kamm , Oliver, “ Qualitative Organic Analysis” , New York,
protein or corn protein, liquefaction of the gel takes place. John W iley & Sons, 1932.
The resulting mixtures are Suitable printing ink vehicles, as (20) L eR oy, M . C. A., Bull. soc. ind. Rouen, 60, 25 (1932).
illustrated by one example: (21) R ivkin, Joseph, and Sheehan, W . E ., I n d . E n g . C hem ., 30,
1228-32 (1938).
E x p e r i m e n t 324. One hundred grams of 20 % casein disper (22) Scheibler, H ., Trostler, P., and Scholz, E., Z. angcw. Chem., 41,
sion in diethylene glycol, with a relative viscosity of 25 poises (23) 1305-6 (1928).
was reacted with 3 grams of paraformaldehyde at 100° C. It (23) Sclimutzler, A . F., and Othmer, D . F., I n d . E n g . C hem ., 35,
gelled; the gel was mixed with 103 grams o f dispersion 290 (Table 1196-1202 (1943).
VI) and stirred for 5 minutes until the mixture became liquid and (24) Shriver, R . L „ and Fuson, R . C., "System atic Identification of
homogeneous. It had a relative viscosity o f 129 poises and a Organic Com pounds” , New Y ork, John W iley & Sons, 1940.
Beckman pH meter reading of 6.7. Its precipitate on dilution (25) Smith, A . K ., and M ax, H. J., I n d . E n g . C hem ., 32, 411-16
with water was 0.144 gram per gram of sample. (1940).
TARTRATES from GRAPE WASTES
Use of Ion Exchangers in
Acid-Carbonate Cycle
A su b sta n tia l p a rt o f th e a p p ro x im a te ly 15,000,000 p o u n d s

o f ta rta ric a cid a n n u a lly re q u ire d in o u r d o m e s tic e c o n o m y
a n d h ith e r to im p o r te d is p o te n tia lly a v ailab le in th e -wastes
J. R . M ATCH ETT, R . R . LEGAULT,
o f th e g ra p e p rocessin g in d u str y . L a b o ra to r y -s ca le e x
C. C. N IM M O , AND G. K . NOTTER p e rim e n ts in d ic a te th e c o m m e r c ia l fe a s ib ility o f u tilizin g
Western Regional Laboratory, s y n th e tic io n -e x c h a n g e m a teria ls fo r recov ery o f th is an d
U. S. Department o f Agriculture, Albany, Calif. possib ly o th e r v alu ab le c o n s titu e n ts o f th e w astes.
I N R E C E N T years the United States has annually consumed sulfate ion. In addition the material is very susceptible to the
about 15 million pounds of tartaric acid, most of which has action o f tartar-destroying bacteria.
been imported in the form of argols and crude calcium tar There is a possibility of recovering valuable constituents of the
trate from wine-producing areas bordering the western Mediter •slop in addition to tartaric acid. The most obvious is malic acid
ranean Sea. As a result of the war these sources have been sub which occurs in substantial amount in the grape, and nitrogenous
stantially eliminated, and it becomes important to examine our materials of undetermined character. Research along these
potential domestic sources. Principal among these are the wastes lines will proceed as soon as conditions permit.
from the grape-processing industry, which include pomace from The various phases of the tw o exchange cycles concerned will be
grape juice manufacture and, in vastly larger amount, the pom discussed. Emphasis is placed on the action of the anion ex
ace and still slop from wine and brandy making. The meager changer, since selective adsorption must occur on this bed and
data at hand indicate that perhaps 1 0 million pounds of tartaric the tartrates are recovered from its regeneration liquors. The
acid are potentially available from the latter wastes, divided be findings of preliminary experiments involving both loading and
tween them in approximately equal amounts. regeneration, where solutions of pure substances known to occur
The recovery problem is one of extraction and concentration in grape wastes were employed, will be related to those obtained
in the case of pomace and of concentration in the case of still slop. in the actual labofatory-scale recovery of tartaric acid from still
Research on both phases is in progress at this laboratory. The slop.
present paper presents preliminary results of an investigation of
tartaric acid recovery from still slop. The conclusions may be ap A T P A R A T U S A N D M A T E R IA L S
plied with equal force to pomace extracts of comparable composi
A standardized routine was followed for pretreatment of ion
tion. The process takes advantage of the remarkable efficiency exchangers. The resinous exchanger was first preconditioned ac
of the newly available synthetic ion-exchange materials. It con cording to the manufacturer’s directions ( I ). In the case of the
sists in passing the slop, partially clarified by settling, through anion exchanger, three preliminary cycles (loading with 1 % hy
drochloric acid and regenerating with 4 % sodium carbonate)
a bed of acid-regenerated cation exchanger, thence through a bed
were carried out. After a thorough upwash to remove fines, the
of anion exchanger where through suitable control the tartaric whole bed was washed into a Büchner or sintered glass funnel
acid is preferentially fixed. When the capacity of the beds has equipped for filtration at a reduced pressure. Gentle suction was
been exhausted, the cation exchanger is regenerated by hydro maintained for 2 minutes after the water had drained below the
upper surface o f the bed. The drained resin was transferred to a
chloric or sulfuric acid and the anion exchanger by sodium car
tared beaker and weighed. (The weight obtained will be referred
bonate solution. The tartaric acid is concurrently recovered as a to as the drained weight o f the resin.) A t this time two samples
solution of its sodium salt, concentrated from ten to twenty times were weighed out for moisture determinations which were con
that of the original slop and substantially free from the many im ducted over anhydrous calcium chloride. With reasonable care
samples of uniform moisture content for a given batch of drained
purities that would accompany it through a similar evaporative
resin are obtained. Between batches the moisture content sel
concentration. The final step has thus far been to accumulate dom varies b y more than 2 % ; different lots o f resin varied in
the product as the insoluble calcium tartrate. It would be ad moisture from 49 to 56% .
vantageous to recover it as the free acid, and consideration is be Tartaric acid solutions, 0.02 M (0.3% ), were employed in pre
liminary studies; this approximate tartar concentration is
ing given this possibility.
frequently found in still slops. Adsorptions were conducted col
Brandy still slop is a turbid dicpersion-solution of a complex umnwise in 30-mm. glass tubes provided with suitable openings
mixture o f organic and inorganic materials representing constitu for drain, upfiow washing, and overflow. The resins were sup
ents of the original grape, those arising through fermentation, ported on sealed-in sintered glass plates.
and those added in the course of processing. The tartaric acid
content (mostly potassium bitartrate) may vary from 0 .1 to 0.4% , A N A L Y T IC A L M E T H O D S
depending on the variety and maturity of grape, the soil and cli When possible, tartrate was determined b y simple titration,
mate where it was grown, the weather during the season, and the otherwise by an adaptation of the procedure of Underhill and
plant treatment with particular reference to the number of washes others ( 6', 8) which is based on the characteristic reddish color
resulting from the interaction o f sodium metavanadate and tar
given the pomace in the preparation of the distilling material. taric acid in dilute aqueous acetic acid solution. In the original
Other components of the mixture, among which m ay be men procedure, color intensities were compared in Nessler tubes; for
tioned acetic, malic, and other organic acids, ash constituents, our purposes the method was adapted to photoelectric colorim
nitrogenous matter, and colloidal matter o f indefinite character, etry.
In the visible region the pervanadyl-tartaric acid complex
vary in similar fashion. It is a common winery practice to add exhibits an absorption maximum at about 515 nw (Figure 1).
about 300 parts per million of sulfur dioxide at the time of crush A t approximately this wave length and under the other conditions
ing the grapes, and a substantial proportion appears in the slop as incident to the test, aqueous solutions of the colored complex
851
852 INDUSTRIAL AND ENG I NE E R I N G CHEMISTRY Vol. 36, No. 9
obey Beer's law when the initial concentration of tartaric acid mination except in the case o f malic acid, which gives rise to a
varies from 0.004 to at least 0.02% (Figure 2). slight negative error when its concentration exceeds that of tar
Prior to tartaric acid determination, colored samples must be taric acid by a factor of at least 2. In view of the fact that in
decolorized. In a typical case 20 ml. of still slop are diluted with grape materials malic acid is ordinarily present in lesser amount
about 50 ml. of water; 2 to 4 ml. of normal hydrochloric acid than tartaric acid, this interference does not constitute a serious
are added, and the solution is decolorized by boiling for 2 to 4 problem. In anion-excbanger effluents the malate-tartrate ra
minutes with about 0 .1 to 0 .2 gram of activated carbon, diluted tio sometimes exceeds 2. Approximations can be made in such
to 100 ml., and filtered through a No. 2 Whatman or similar paper. cases b y adding a suitable known amount of tartaric acid to the
In many instances boiling is unnecessary, but in other cases the sample and correcting the final result.
filtrate is turbid regardless of the original pH, the amount of car The metallic ions commonly found in grape wastes have not,
bon, or the duration of the boiling period. This difficulty can be in the concentrations encountered, been observed to cause inter
overcome b y replacing the paper with a Seitz or asbestos filter. ference.
Organic nonvolatile acids were roughly determined as follows:
A measured volume of anion-exchanger effluent, free of mineral
acids, was evaporated to dryness by leading a gentle stream of air
over the solution heated in a glycol bath kept at 130° C. The
residue was dispersed in water and titrated electrometrically to
H 8.4. On known solutions composed of tartaric, acetic, and
E ydrochloric acids, the error of the method does not exceed 5% .
(Erratic results are obtained in the presence of sulfuric acid.)
Sulfuric acid was determined turbidimetrically on samples of
decolorized slop. All pH measurements were made with a glass
electrode.
FLOW RATE
The optimum operating rate of flow through exchanger beds

must be determined within limits on the basis of compromise be
tween the high break-through capacity at low rates and the larger
amount of desired constituent fixed per unit time which charac
terizes fast ones. Preliminary experiments disclosed that, with
0.02 M tartaric acid solution, 20 volumes of solution per volume
o f anion exchanger (backwashed and drained) per hour (W H ) is
an expedient rate and that the adsorption picture characteristic
of this influent and a given rate is, with fresh exchanger at least,
sensibly independent of bed shape. The latter was established
through identity of leakage curves when equal volumes of resin
were employed and the ratio of height to diameter varied as 0 .8 ,
1.9, and 5.1. This finding agrees with that of Meyers and others
WAVELENGTH
(#), who reported that "the results on large-diameter and small-
diameter columns were almost identical” .
F ig u re 1. A b s o r p tio n in V isib le
R e g io n o f V a n a d ic A c id -A c e t ic
A c id jS o lu tio n s , w ith a n d w ith o u t
T a rta ric A cid
The colorimetric estimation is made as follows: T o 0.5 ml. of

glacial acetic acid in a 25-ml., graduated, stoppered cylinder or
volumetric flask is added a measured volume (not exceeding 20
ml.) of a decolorized and neutralized (pH 3 to 9) solution contain
ing from 1 to 5 mg. of tartrate, calculated as tartaric acid. The
mixture is diluted to exactly 23 ml. when a graduated cylinder is
used and nearly to that volume when a volumetric flask is em
ployed. Tw o milliliters of a clear, aqueous, 5 % sodium meta-
vanadate solution is added and the final volume adjusted to 25
ml. The sodium metavanadate is conveniently dissolved in wa
ter at 50-60° C., made up nearly to volume, permitted to cool,
and then filtered. In case of persistent turbidity, the fresh solu
tion is permitted to stand for at least 24 hours, after which filtra
tion is repeated. It is important to check the concentration-
transmission curve at frequent intervals, especially when fresh
solutions of metavanadate are prepared.
The solution is set in a darkened place for 30 minutes for color
development. D irect sunlight induces a bleaching o f the color.
This effect may not be noted unless the room is well lighted.
However, it is clearly desirable to place the tubes in the dark for CONCENTRATION IN PERCENT XIOOO
color development. Under these conditions, the color is stable
and shows no change over a period of 24 hours. When the color F ig u re 2. T y p ic a l C o n c e n tr a tio n -
has developed to full intensity, it is measured with the colorim T ra n sm iss io n (F ilte r M 515) C urve
eter in the usual manner, using the M515 filter. From the per fo r T a r ta ric A cid D e te r m in a tio n
centage of transmission obtained and the concentration-trans
mission curve, the tartaric acid content of the original sample is
readily computed.
The appropriate flowr rate will clearly vary writh concentration
A precision o f =*=2.5% is attainable (transmission measurements
in matched 18-mm. Pyrex test tubes), but in routine determina and character of the various constituents of the influent and by
tions on still slop, ranging in concentration from a few hundredths the adsorption rate of the resin, for although rapid the.reaction is
to a few tenths per cent tartaric acid, the precision is nearer —5% . not instantaneous. This latter factor, in turn, will be influenced
In a series o f fifteen routine determinations, conducted over a
by the particle size o f the exchanger and anything which may have
period of two weeks on a certain sample of still slop, the average
value found was 0.3 3% with an average deviation of 4 % . affected its surface— for example, retention o f certain adsorbates
The other acids found in grapes d o not interfere with the deter such as grape pigments, etc.
90 80 A D S O R P T IO N O F
Ha M ft EF FLU INFL DENT 0.3? AS

V
Fi 2t a
T A R T A R IC A C ID
F R O M M IX T U R E S O F
30
80 / 70 •M 0.03 % AS h 2so PU RE SU BSTAN CES
_ oo __
7
2
X
efflue IT
Still slops are com
70 r J60
V
:2 0
/ plex m ixtures con
taining, in consider
O 60 / A6 0H IN INFLUEf
T7
uj
LU
° 50
\ / able quantity, acetic,
malic, sulfuric, phos
//
ir
UJ
I- p h o r i c , a n d other
__ J
-Sri
Zj 10
3'
/ :n t
q: 50 o' 40 n H*T IN EF FLUE NT
acids free or in the
U
J H, T It INFLUÍ: n t
a. form of salts. The
h
40 r H¡¡T IN INFLUENT w
Í 30 y extent of adsorption
of a given constituent
> r h1EF FLUÍ :n t p — is influenced by the
5
o / 7 / : 20 presence of others (S ,
UJ 6
□
-J FIG. 3 / ) INFLLIENT 0 .3 1 AS tjT a 9 ). Therefore, solu
-J - a o 5 % AS HjS tions were made up
* 20 3E 10 / /
i in distilled water of
y /
UJ
7
10 1 F t T IN E : FLUEN
/
3 4
■
5
3 10
H.T IN EFFL UENT
/ t a r t a r ic a c id and
potassium bitartrate,
LITERS OF EFFLUENT alone and adm ixed
J _ ^
/ w ith oth er s u b
FIG. 4 1
1 2 3 4 5
0.3 % H ZT AS NFLU : n t stances. These solu
LITERS OF EFFLUENT tions wercpassed over
80 1 1 //
H .T 1N EFFLUENT anion-exchanger beds
T DTAlpAcJlD fT lN F L l EN FIG. 5 1 at a f l o w r a t e of
T >TA . AC;id 1i_Ei FLL EN' 17 V V H ; the results
70 LITERS OF EFFLUENT
S' are given in Table I
100
r — — — i i— i i— / T
and Figures 3 to 7.
50 > -TOTAL ACID IN INFLUENT-
Acetic acid should
/ 90 not substantially in
/ r iv>. o
50 < f terfere w ith tartar
/ adsorption even if
80
IN rLU ENT AS present, as would
40 / H,J1 FIG. 7 j
rarely be the case, in
0TA. AC10 1t EF'FU r
U
z
À
J
i 70
- large excess in the in
30 / / fluent slop (Figure 3).
/ 60 / M alic acid would
f \ appear in the effluent
f £10 IN EFFLUEh
EFF LUE NT AS H*T before tartaric, but
/ 1 150
/ S ' y| would be present on
10
F •J N i H,T N EFFLUEt T-\
the bed in quantity
/ / INFL.UE tr h justifying recovery at
. / the time of incipient
y
[\ tartar leakage
IN : LU ENT AS H21 5 30 ?
EFF LUE NT ÍXS 1 3 K M IN EfF LUENT
(Figure 4). Sulfuric
A O / / / ■4S acid has a definite
1 / U
J
30
3 20 / U It INI ¿Sx -o- adverse effect on bed
I / ------
- j f~ capacity for tartar
RE SEN ERA rED RE‘ UN
/
i
20 but will not grossly
10 I /
1 f in terfere w ith the
EFF LUENT .... /
1
y projected operation
10 •*"2, tOtlXt^nAI t U jHfcOlN y
I / when present in the
4 6 8 10 2 3 4 5
LITERS OF EFFLUENT
am ounts norm ally
LITERS OF EFFLUENT found in still slops.
F ig u r e 3 . A d s o r p t i o n , b y A n i o n E x c h u n g e r , o f T a r t r a t e a n d A c e t a t e I o n s f r o m a S o l u t i o n o f M ix e d A c i d s ( D r y W e i g h t Sulfate ion is com
o f B e d 23 G r a m s , V o l u m e A b o u t 90 M l . , F l o w R a t e 17 V V H ) pletely adsorbed from
mixtures of tartaric
F igu re 4 . A d s o r p t io n , b y A n io n E x c h a n g e r , o f T a r t r a t e a n d M a la t c I o n s f r o m a S o lu t io n o f M ix e d A c id s (D r y W e ig h t
o f B e «l 23 G r a m s , V o l u m e A b o u t 90 M l . , F l o w R a t e 17 V V H ) and sulfuric acids,
and at the tartar
F ig u r e 5 . A d s o r p t i o n , b y A n i o n E x c h a n g e r , o f T a r t r a t e I o n f r o m T a r t a r i c A c i d S o l u t i o n i n P r e s e n c e a n d A b s e n c e o f
break-through point
S u l f u r i c A c i d ( E f f l u e n t F r e e o f S u l f a t e ; D r y W e i g h t o f B e d 23 G r a m s , V o l u m e A b o u t 90 M l . , F l o w R a t e 17 V V H )
the total acid fixed
F ig u re 6 ( a b o r e ) . A d s o r p t i o n , b y A n i o n E x c h u n g c r , o f T a r t r a t e I o n a n d T o t a l A c i d f r o m a S o l u t i o n 0 .0 2 M i n P o t a s on the bed is less than
s iu m B i t a r t r a t e a n d 0 .0 6 M i n A c e t i c A c i d ( D r y W e i g h t o f B e d 45 G r a m s , V o l u m e A b o u t 160 M l . , F l o w R a t e 19 V V H ) ;
from ta rta ric acid
(b e lo ic ) A d s o r p t i o n o f T a r t r u t e I o n f r o m 0 .0 2 M P o t a s s i u m B i t a r t r a t e S o l u t i o n o n A n i o n E x c h a n g e r R e g e n e r a t e d w i t h
S o d iu m C a r b o n a t e a n d w i t h S o d i u m H y d r o x id e S o l u t i o n s ( D r y W e i g h t o f B e d 23 G r a m s , V o l u m e A b o u t 9 0 M l . , F l o w alone (Fig ure o).
R a t e 17 V V H ) This som ew hat
unexpected observa
lig u r e 7. A d s o r p tio n , o n A n io n E x ch a n g e r , o f A c id s f r o m “ S k e le t o n S lo p ” (E fflu e n t F re e o f S u lfa t e ; D ry W e ig h t o f
B e d 2 3 G r a m s , V o l u m e A b o u t 90 M I ., F l o w R a t e 17 V V H ) tion is particularly
. T a b j . b I. A d s o r p t io n o p T a r t a r i c A c id (H jT ) fr o m A q u e o u s S o lu t i o n s
____________ H « T a n d T o t a l A o id A d s o r p tio n C a p a c it y 0 , M illic q u iv a le n t a p e r G ra m D r y R e sin

—Co----------- ^ ----------- v - C » ----------- - “ C<o----------- > - C ío----------- - Figure
C o m p o s it io n o f In flu e n t H jT T ota l I I jT T ota l . H jT T ota l H jT T ota l H jT T ota l N o.
0 .0 2 M H iT 8 .0 8 .0 8 .4 8 .4 8 .8 8 .8 5
0 .0 2 M K H T & 0 0 0 0 0 .8 0 .8 6 .0 6 .0 6
0 .0 2 M K I I T C 0 0 0 0 0 0 0 0 7 .8 7 .8 6
0 .0 2 M K H T + 0 .0 6 M A c O H 4 .6 6 .0 5 .0 6 .3 6 .6 7 .2 6
0 .0 2 M H jT + 0 .0 0 1 5 M H aSO* 6 .5 7 .1 7 .3 7 .9 7 .8 8 .5 8 .2 8 .9
0 .0 2 M H jT - f 0 .0 0 3 M HjSO< 5 .7 6 .6 Ö .5 7 .6 7 .1 8 .3 7 .3 8 .5 7 .7 9 .0 5
0 .0 2 M H j T + 0 .0 6 M A c O H . 6 .4 8 .3 7 .1 8 .6 8 .7 9 .5 3
0 .0 2 M H jT + 0 .0 2 M H sM 6 .5 1 0 .4 6 .7 1 0 .5 7 .5 1 0 .3 8 .1 1 0 .7 8 .6 1 0 .8 4
0 .0 2 2 M H jT + 0 .0 3 4 M A c O H - f
0 .0 0 1 8 M IIjSO * + 0 .0 0 8 8 M H sM 6 .2 9 .9 6 .2 9 .9 6 .7 1 0 .3 7 .1 1 0 .2 7 .4 1 0 .6 7
° S u b s cr ip ts (Co, C i, e t c .) refer to p e r o e n t o f in flu e n t ta r ta r le a k in g a t p o in ts w here c a p a c it ie s are c a lc u la te d ,

t P r e v io u s re g e n e r a tio n o f b e d e ffe c t e d w ith 2 % N a jC O j s o lu tio n .
c P r e v io u s r e g e n e r a tio n o f b e d efTcoted w ith 2 % N a O H s o lu t io n .
striking becnuse in the case of mixtures of malic or acetic acids as the normal salt. In the second case the tartrate leakage is
svith tartaric, the total acid capacity at the tartar break substantially reduced and an equivalent amount of carbonate ion
through point was at least equal to that when tartaric acid alone appears in the effluent. From this pertinent though somewhat
was employed. The experiments, however, were not carried out limited example it may be inferred that both regeneration and
under comparable conditions, and definite conclusions should loading are definite anion-exchange phenomena. Beds regener
await accumulation of further evidence. ated with carbonate retain an equilibrium load of carbonate ion,
and this can be replaced under the conditions outlined by tartrate
ions. Beds regenerated with caustic alkali solutions would not
retain anions (except that the free base could obviously be formu
lated as a hydrated amine, in which case the loading reaction
would be written as an anion exchange involving hydroxyl ion
/ which is replaced in the present instance by bitartrate ion but not
/
/ by tartrate), and acid adsorption would proceed through fixation
TOTAL ACID If* EFFLUEN r \
I- of hydrogen ion with consequent retention of an equivalent of
z anion.
S. I
This concept of bed action as one of anion exchange, which ap
pears to be somewhat at variance with current views ( 1 , 2, 3, 5,
\
7), is supported by substantial evidence, to be presented in a sep
arate paper. It will be shown there that a definite replacement
\
3 series of anions can be set up and that the course of exchange will
\
or be fixed by the relative positions of the ions involved and their
nj \
a activity in the solution. Thus, through suitable choice of re
<n
actants, a number of loading-regeneration cycles can be formu
N
lated, some of which m ay prove to be better adapted to recovery
TO" AL ACI ) IN INF LUE NT \
o of constituents from natural products than the usual acid-carbon-
o ate cycle. For example, experiments now underway indicate
that tartrate recovery through regeneration with chloride solu
pH OF INF _UENT tion and loading from unacidified slop (potassium bitartrate solu
----- tion) may be a more economical process than the one described
here.
The final experiment of this series involved a solution contain
/
pH OF EFF LUE NT ing acetic, tartaric, malic, and sulfuric acids in concentrations
similar to those found in a rough analysis of a typical slop. It
will be referred to here as a skeleton slop. The sequence of events
2 3 4 5 is shown by Figure 7 and Table I to be essentially what would
LITERS OF EFFLUENT have been predicted from the other experiments involving the
same acids (Figures 3, 5, 6 , and Table I). Its agreement with
F ig u re 8 . A c id ifica tio n o f a T y p ic a l S till S lop by
C a tio n E x ch a n g er (B ed V o lu m e A b o u t 520 M l., results obtained with an actual slop sample will be discussed
F low R a te 6 W H ) below.
a c id if ic a t io n w it h c a t io n exc h an g er
The potassium bitartrate results are of particular interest.

The obvious function o f the cation exchanger is that of replac
They indicate first that suitable yields of tartaric acid would not
ing the cations in still slop with hydrogen ions and thereby effect
result through passage of untreated slop over the anion exchanger
ing an increase in total acidity. The exchanger employed,
(Figures 6 and 9). A reasonable load can ultimately be attained
but only at the expense of excessive losses through leakage, al Amberlite IR-100 (hydrogen cycle), performed this function
though in the presence of relatively large amounts of acetic acid normally. Its regeneration has already been described (f, t , 4)>
the picture is more favorable. for economical operation it is advisable to retain the partially
The results shown for passage o f potassium bitartrate solution spent acid solution from the latter phases of the process for use as
over beds previously regenerated by sodium hydroxide and by so the first portion in the succeeding cycle.
dium carbonate (Figuro 6 ) are in significant contrast. In the Acidification of still slop by the cation exchanger is illustrated
first instance tartrate ion is fixed to an extent equivalent to pre in Figure 8 . It is evident that pH measurements may be used in
cisely half of the total present while the other half passes through the control of the process. The point of incipient pH rise denotes
the break-through capacity of the exchanger. The first effluents to a relatively high tartaric acid leakage and providing a second
are low in total acid as a result o f dilution by water present on the bed, operated in serios, to scavenge the effluent after tho tartaric
bed at the beginning. The constancy o f pH while the total acid acid break-through has been reached. This technique was
ity is increasing is accounted for by the buffer capacity of the tested on acidified slop; the results are plotted in Figure 1 0 .
effluent. The lower curves (scavenger bed) are offset, consistent with the
The loading process involves exchange adsorption of nitroge fact that the first 6 liters of prime bed effluent were not scavenged.
nous as well as metallic ions. In one case the total nitrogen The second 6 liters served as the first scavenger influent; there
(Kjcldahl) content of the still slop dropped from 250 to 40 mg. after completely acidified slop was used.
per liter as a result of passage through the cation exchanger.
This condition was maintained until the break-through point of
the bed was reached. Thereafter it rose rapidly to a maximum
of 360 mg. per liter before subsiding, an indication of selective
exchange adsorption of the type reported for the anion exchanger.
Similar experiments on other samples o f still slop were indicative
of the adsorption o f nitrogenous compounds.
The break-through capacity o f the cation exchanger may con
veniently be related to the capacity of tho anion exchanger in
terms of volumes o f a typical still slop. The capacity of the
cation exchanger was 10 volumes of still slop per volume of ex
changer; on a similar basis the anion exchanger had a capacity
of about 30 volumes, as will be shown later. These values vary
somewhat with the type of still slop processed. In the course of
one season, the still-slop-volume capacity o f the anion exchanger
varied between three and four times that of the cation exchanger.
A D S O R P T IO N O F T A R T A R I C A C ID F R O M S T IL L S L O P
Still slops normally contain varying but substantial concentra

tions of free acids in addition to the acid and normal salts com
monly present. It appeared worth while, therefore, to deter
mine whether complete acidification of a typical slop was a neces
sary prelude to efficient tartaric acid adsorption (compare Figure
6 and Table I, K H T -a cetic acid mixtures).
Three experiments were run on the early-season slop: (a) A
portion was acidified by passing it through an acid-regenerated
cation exchanger; ( 6 ) equal portions of slop treated as in (a) and
untreated slop were mixed; (c) a portion of the slop was untreated.
The three will be referred to as acidified, half acidified, and un
acidified slops. Each was passed separately over a carbonate
regenerated anion-exchanger bed at a flow rate of 19 W H . The
results, expressed in terms of tartaric acid leakage, are shown in F ig u re 9. A d s o rp tio n o f T a rtra te Io n fr o m T y p ic a l
Figure 9. It is plain that loading with unacidified slop is im S till S lop (0.022 A f) (D ry W e ig h t o f B ed 45 G ra m s,
practical. The loading characteristics o f acidified and half acid V o lu m e 160 M l., F low R a te 19 W H )
C u rv e 1 , a c id ifie d ; 2 , h a l f a c id ifie d ; 3, u n a c id ifie d .
ified still slops, expressed as amounts of tartaric acid adsorbed
at various stages, are compared in Table II.
Complete acidification offers an advantage from the standpoint
Comparison of Figure 7 with the upper curves of Figure 10
of maximum loading with minimum tartaric acid leakage. In
brings out striking similarities between the respective leakage
either case there is a practical advantage in capacity b y operating
curves of the skeleton and the still slops on which the former was
patterned. From the areas included under each curve the data
of Table II I are calculated; they illustrate the correspondence
T a b le I I . D e g r e e o f S t i l l S l o p A c i d i f i c a t i o n a n d H jT
from a slightly different point of view and tend at the same time
A d s o r p t i o n C a p a c i t y a t V a r i o u s L e v e l s o f H2T L e a k a o e
to substantiate the analytical methods, which were admittedly
.------------------------------------ A d s o r p t io n C a p a c it y ----------------------------------
H a lf A c id ific a t io n C o m p le t e A c id ific a t io n rough in the case of sulfuric, acetic, and malic acids. The only
H iT M .e . H i T / L b . H j T / c u . f t . M .e . H , T / L b . H , T / c u . ft. significant difference lies in earlier onset o f leakage for each meas
L e a k a g e“ g. d r y reain resin (a p p r o x .) g . d r y re sin resin (a p p r o x .)
ured constituent o f the actual still slop. This could be accounted
Ci 4 .2 5 .0 4 .9 5 .9
Cm 4 .9 5 .9 8 .3 7 .6 for, at least in part, by tho presence of mineral acids other than
C» 5 .1 6 .1 6 .5 7 .8
Cm 5 .5 6 .6 7 .0 8 .4
sulfuric, and of pigments, colloids, etc., in the slop. The greater
“ S ee f o o t n o t e “ , T a b le I.
total leakage (displacement) of tartaric acid in the ease of still
slop may be considered as further evidence of the action of min
eral acids not included in the skeleton slop.
T a b l e III. C o m p a r a t iv e C o m p o s itio n o f E f f l u e n t s fr o m The lower curves of Figure 10 demonstrate the efficiency at
S k e l e t o n S l o p , S “ , a n d A c t u a l S t i l l S l o p , A 6, i n M i l i . i -
which the scavenger adsorbs the tartaric aeid leakage of the prime
EQUIVALENT PER 12 LlTEUS OF EFFLUENT
bed. The earlier onset of malic and tartaric acid leakage is due
H jM A cO I I H ,T T o t a l A c id
8 A. 8 A 8 A 9 A
in part to the smaller capacity of the scavenger bed (about 25 % )
168 176 370 400 150 200 700 760
although the effect of enhanced influent ratio of volatile acids is
“ D r y w e ig h t o f b e d , 23 gram a. R e s u lts (6 lite rs e flfu e n t) a re m u ltip lie d
manifest also. In practice the prime bed would have been taken
b y 2 t o alig n w ith t h o s e o f th e la r g e r b e d u se d fo r a c tu a l s t ill s lo p . out o f service at 25 to 30 % tartar leakage and the scavenger pic
b D r y w e ig h t o f b e d , 4 4 .5 g ra m s . In flu e n t c o n c e n t r a t io n s a r e in d ic a te d
in F ig u res 7 a n d 9. ture revised accordingly. Tho peculiar breaks in the malic and
tartaric acid curves (between tenth and twelfth liters of effluent)
E F F IC IE N C Y O F P R O C E S S
The over-all efficiency of the tartar recovery process was

tested on a laboratory setup that consisted of three 44.5-gram
(dry weight) beds (160 ml.) operated in the prime bed-scavenger
fashion. The influent flow rate was maintained at approximately
19 W H . A total of 72.3 liters of still slop (tartar content, 0.33%
as tartaric acid) was processed after complete acidification. Each
bed was operated to about 3 0 % leakage. Regeneration of the
loaded beds was effected with sodium carbonate solution. The
yield of sodium tartrate in the combined regeneration effluents
was 222 grams, calculated as tartaric acid, or 9 3 % of the total
throughput. The over-all yield of calcium tartrate was 90% .
From the regeneration effluent acidified to pH 5.0, calcium
tartrate is readily precipitated in coarse crystals of excellent pur
ity by adding a 10% excess of calcium chloride solution. The
yield will be affected to a considerable extent b y the concentra
tion o f tartrate ion in the precipitation solution, since the concen
tration in the mother liquor may be expected to range from 0 .1
to 0 .2 % .
R E G E N E R A T IO N O F A N IO N E X C H A N G E R
Regeneration of the anion exchanger serves the dual purpose of

preparing the bed for the succeeding cycle and recovering the
tartar in a concentration suitable for precipitating a calcium salt
of high purity in maximum yield. A satisfactory process will
result in (a ) a regeneration effluent of 5 -6 % concentration (calcu
lated as tartaric acid) and (b ) an almost completely regenerated
bed since otherwise tartar losses will be sustained in the washing.
For economical operation the time required must be kept to a
minimum and the regenerant reagent utilized efficiently.
Under the conditions of the experiments described here, these
requirements are not met by regeneration through a single down-
flow pass of sodium carbonate solution even if a portion of the
partially spent reagent is reserved for use in the following cycle.
Regardless of flow rate or regenerant concentration, tartar solu
tions above 3 % were not obtained, apparently as a result o f chan
neling due in part at least to carbon dioxide evolution as regener
ation proceeded. Evolution of gas bubbles would presumably
be avoided by carrying out the operation under pressure. Grav
ity flow, however, is generally more convenient for the tartar re
covery process.
The simple counterflow procedure outlined below readily at
tains effluent tartar concentrations of 6 % with complete regener
ation and efficient utilization of reagents; the time required is
about 2 hours: Three vessels are set up, each containing enough
LITER S OF EFFLUENT
regenerant solution so that, when the acid load of a single bed
F igu re 10. A d s o rp tio n o f A c id ic C o n s titu e n ts o f A cidified has been dissolved, the resultant tartrate concentration will be
Slop b y A n io n -E x ch a n g e r P rim e B ed ( a b o v e ) and
5 -6 % . A fourth vessel containing a similar volume of water is
S caven ger ( b e l o w )
provided. The contents of the four vessels are successively cir
T h e s c a v e n g e r w a s a hetl o f u b o u t 2 5 % l e s s c a p a c i t y t h a n p r i m e b e d i
d r y w e ig h t o f b e d s 45 g r a m s , v o l u m e a b o u t 160 11) 1 . 1 (lo w r a t e 19 V Y I I . culated through the bed at a rapid rate for about 2 0 minutes.
The high fluid velocity (about 20-40 W H ) induces turbulence
with resulting excellent contact between bed and solution; gas
bubbles are rapidly entrained and removed. During the 20-
arc o f no significance; they correspond to a point where 600 ee. rninute circulation, sensible equilibrium with the bed is estab
of upwash liquor from the prime bed served as influent to the lished. The liquor of the first circulation proceeds to the precipi
scavenger. This liquor was low in tartaric acid content and of tation vessel, that from the second becomes the first for the
undetermined but evidently high malic acid concentration. following cycle, and so on. T o the water after circulation is
From the example cited, it is evident that considerable flexi added sodium carbonate in amount equivalent to slightly more
bility in operating technique is possible. The advantage of than the total acid adsorbed on a single bed, and the resulting
operating a prime bed in series with a scavenger lies in the solution then becomes the third regenerant while the fourth con
greater loads of tartaric acid obtainable on a given quantity of tainer is again supplied the requisite volume o f water.
resin. For the case illustrated in Figure 10 the load, at the tar In the case of the 44.5-gram (dry weight) beds used with ac
taric acid break-through, amounted to 5.9 pounds of tartaric acid tual still slop in the experiments described above, a volume of
¡>er cubic foot of resin ; the load at 1 0 0 % tartaric acid leakage was 300 ml. was placed in each of the regenerant flasks and 24 grams
approximately 9.1 pounds per cubic foot o f resin. Operation to of sodium carbonate were added at the indicated point for each
so high a leakage is questionable because the efficiency of tartaric bed regenerated. After a few cycles to establish equilibrium in
acid adsorption drops rapidly after the 2 5 % leakage point has the counterflow operation, the tartar concentration (as tartaric
been reached (curve 1, Figure 9). acid) in the four flasks was about 3 % , 1% , 0.1 % and 0, respec-
tiveiy, before regeneration was begun. After regeneration the upon the length o f useful life of the expensive exchange materials
corresponding concentrations were 6 , 3, 1, and 0.1% , at which when contacted with the complex grape solution-suspension.
time the first was sent out of the system for precipitation. Re One anion exchange column has been through 80 cycles in this
generation was followed by downfiow water wash to phenolphtha- laboratory during which it has been contacted with winery ma
lein neutrality, after which the beds were upwashed and drained terials from a number of sources. It has dropped in capacity ap
for the next cycle. proximately 25 % . . From the results presented here, it may be
assumed that recovery from slop of average composition should
be at least 5 pounds per cubic foot of anion exchanger per cycle
when the material is fresh. If the process is carried out sub
PERCENT EFFICIENCY
100 50 25 10 5
stantially as described, the volume requirement of cation ex
10 changer will be three to four times that of the anion exchanger
1
(in the case o f the materials studied). Capital investment in the
plant itself may be nominal. The necessary equipment can be
>-
cc built largely of noncritical materials. Labor will be an item of
a - REGE NEF ATK)N DF F E S \l
u. L OAD ED WIT ^ a 3% H2T greater or lesser consequence in inverse ratio to the magnitude of
y
I • the operation. One man with some technical supervision should
<
tr be able to handle a well designed plant. The chemicals are all
o
e 7
inexpensive.
Results of experiments involving a chloride-tartrate ion-ex
■ i“ "*”* change cycle on the anion exchanger bed thus far indicate that in
>
o / OF RE SIN _0A )ED this cycle the exchanger retains its efficiency longer in contact
s r e GEN ERA TlOf'
Wl TH STII 1. fj 1 OF with the complex grape substance; in addition the process would
require only about half the plant setup demanded by the proce
dure outlined here. The cation exchanger would not be required.
■ —
In either cycle, thorough defecation and decolorization of the
influent solutions appear certain to prolong effectively the use
ful life of exchangers.
< ACKNOW LEDGM ENT
It is a pleasure to acknowledge the unfailing interest and wise

4 8 IE 16 20 counsel of L. B. Howard of this laboratory throughout the course
MILUEQUIVALENTS N a 2C 03
of the investigation. A typical sample of slop produced during
PER MILLIEQUIVALENT H2T
early-season operation at the winery of Bear Creek Vineyard As
F ig u re 11. E fficie n cy o f C a rb on a te sociation was furnished for the work through the generous co
U tiliza tio n in R e g e n e ra tio n o f operation of L. K . Marshall and Allen Pool. Ion-exchange ma
A n io n E x ch a n g er a n d C o n c u r re n t terials used were Amberlite IR-100 (cation exchanger) and Am-
T a r ta r R ecov ery , C a lcu la te d fr o m
M illie q u iv a le n ts o f R e a g e n t R e berlite IR -4 (anion exchanger), a portion o f which was kindly
qu ired p er M illie q u iv a le n t o f T a r provided by Resinous Products and Chemical Company.
ta ric A c id R e m o v e d fr o m B ed s
L oa d ed f r o m S o lu tio n o f P u re A cid L IT E R A T U R E C IT E D
( u p p e r c u r v e s ) a n d fr o m S lop ( l o w e r )
(1) Meyers, R. J., and Eastes, J. W ., I n d . E n g . Chem., 33, 1203
(1941).
The efficacy of the process described is emphasized by the (2) Meyers, R . J., Eastes, J. W ., and Meyers, F. J., Ibid., 33, 701
curves of Figure 11. The high regeneration efficiency of the (1941).
early part of the process results in effective utilization of the car (3) Meyers, R . J., Eastes, J. W ., and Urquhart, D ., Ibid., 33, 1270
(1941).
bonate when the first rcgenerant is being cycled. The last in
(4) Resinous Products and Chemical Co., “ The Amberlites” , 1941.
crements of acid are removed from the bed at low efficiency but (5) Schwartz, M . C., Edwards, W . R ., Jr., and Boudreaux, Grace,
in the presence of the required large excess of reagent, substan I n d . E n g . C h e m ., 3 2 , 1 462 (1 9 4 0 ).
tially all of which will ultimately be used. The upper curve por (6) Snell, “ Colorimetric M ethods of Analysis” , Vol. II, p. 126,
New Y ork, D . Van Nostrand Co., 1937.
trays the single pass regeneration (4 % sodium carbonate solu
(7) Tiger, H. L., and Sussmann, S., I n d . E n g . Chem., 35, 186 (1943).
tion) of a 44.5-gram (dry weight) bed loaded with pure tartaric (8) Underhill, F. P., Peterman, F. I., and Kraus, A. G ., J. Pharm acol.,
acid. The lower curve is similar but for a bed loaded from slop. 43, 351 (1931).
The difference is due to adsorbed acids other than tartaric. (9) Wilson, J. N „ J. A m . Chem. Soc., 6 2, 1583 (1 9 4 0 ).
The tartar from a bed loaded with the pure acid could appar
ently be recovered in about 6 % concentration when 4 % sodium
carbonate is used as regenerant. (In practice this is not realized
on account of dilution due to water present on the bed after drain
ing but prior to regeneration.) From slop-loaded beds the re Perform ance o f A gitated G as-Liq u id
quired concentration of sodium carbonate to produce a 6 % tartar Contactors—Correction
solution is substantially higher and a correspondingly lower flow
rate would be required for high regeneration efficiency. Low The authors wish to call attention to an error in this paper
flow rates aggravate the contact difficulties due to bubble forma which appeared in the June issue of I n d u s t r i a l a n d E n g i n e e r
tion and render it difficult to obtain tartar concentrations greater in g C h e m is t r y . On page 509 the first sentence of the first para
than about 3 % in single-pass operation. graph below Figure 10 should read: “ It should be noted that, in
scaling up a design, the volume of absorbent liquid and the agi
tator power will increase as the cube of the scale factor. . .
D IS C U S S IO N
C . M . C o o p e r , G. A. F e r n s t r o m , a n d S. A. M i l l e r
The economic success of the ion-exclianger procedure for tar
E . I. d u P o n t d e N em ours & C om pany, Inc.
taric acid recovery would appear to depend to a large extent W il m in q t o n , D e l .
C o r r e la t in g V a p o r C o m p o s itio n s
and
R e la t e d P r o p e r t ie s o f S o lu tio n s
V a por c o m p o s itio n s , e q u ilib r iu m c o n s ta n ts (K ), a ctiv ity T h is m e t h o d is c o n v e n ie n t fo r c o rr e la tin g experim en tal
c o e fficie n ts ( 7 ), an d relative vola tilities (a) as fu n c tio n s o f d a ta w h ic h are u su a lly taken a t c o n s ta n t to ta l pressures.
pressure a n d te m p e ra tu re are p lo tte d d ir e c tly to give T h e slop es o f lin es 0 11 b o t h types o f p lo ts are fu n c tio n s o f
s tra ig h t lin es b y th e m e t h o d p reviou sly su g gested (#) fo r la te n t h ea ts a n d p a rtia l h e a ls o f so lu tio n o f th e c o m
vapor pressures. A m in im u m o f d a ta m a y th u s b e used p o n e n ts . T h u s , ex p e rim e n ta l d a ta o n h eats, vapor p res
to d efin e a w h o le sy ste m , a n d th e c o n s is te n c y o f experi su res, v a p or c o m p o s itio n s , a n d a ctiv itie s, h ow ever d eter
m e n ta l d a ta m a y b e ev alu ated rea d ily . A new a n d sim p ler m in e d , m a y be u sed fo r m u tu a l cross c h e c k in g , co rr e ctio n ,
p lo t a n d th e c o rr e sp o n d in g e q u a tio n s are p resen ted for a n d fillin g in w h ere in c o m p le te . A ctiv ities o f so lu tio n s,
pa rtial pressu res, vapor c o m p o s itio n s , e q u ilib r iu m c o n o n ly th e solv en t o f w h ich is v ola tile, m a y b e correla ted .
sta n ts , a ctiv ities, an d rela tive v ola tilities as d ir e c t lo g a F u rth e r m o r e , critica l pressures a n d tem p era tu res m a y be
r ith m ic fu n c tio n s o f th e to ta l pressure o f th e sy stem . u sed (9) fo r e ith e r, esp ecia lly n ea r th e critic a l ran ge.
DONALD F. OTHMER AND ROGER GILMONT

Polytechnic Institute, Brooklyn, N . Y.
A PO R pressuro and latent heats have been correlated by a same pressure unit does not have to be used on both the X and Y
V simple plot or the use of the equation, axis, since there is a constant ratio between any two units; this
ratio would merely m ove the line up or down on the plot without
log P - ( L / L ') log P ' + C changing its form or slope.) (c) Points so obtained are connected
by a line, usually straight.
where at the same temperatures, P and P ' arc vapor pressures
When applied to solutions of two or more volatile liquids, it is
and L and L ' molal latent heats, respectively, of two compounds
convenient but not necessary to use one of the liquids as the refer
(the latter in each case being that of a standard or reference sub
ence material. Figure 1 shows the total vapor pressure of vari
stance) and C is a constant. Log P ' really serves as the tempera
ous solutions of constant composition of the system acetic acid-
ture variable and is obtained directly from vapor pressure data
water from other data { / ) ; here, as in the other logarithmic plots,
of the standard substance. L / L ' is nearly independent of tem
data are plotted at corresponding temperatures. Figures 2 A and
perature. A logarithmic plot gives a line which is substantially
3 show partial pressures o f water (reference substance) and acetic
straight, the slope of which { L / L ') gives latent heat data at any
acid. (Some slight curvature of the total pressure lines and of the
temperature for the compound in question from that of the refer
lines of some other functions should result theoretically from addi
ence substance. This vapor pressure plot was extended ( 9 ) to be
tion of the partial pressure lines near the middle range of liquid
used with reduced pressures at reduced temperatures in order to
composition when the molal latent heats of the two components
increase the precision in use. Further applications were made to
are widely different. In the usual case, as in Figure 1, a plot does
gas solubilities and partial pressures ( I S ) , and to the pressures of
not show this very small deviation which may, for practical pur
adsorbed materials from adsorbents { 1 1 ). A recent application
poses, be neglected).
of the method correlated equilibrium constants of hydrocarbons
P l o t t e d a g a in s t T o t a l P r e s s u r e s o f S y s te m . Another
b y means o f a nomogram {10).
simple m ethod of plotting which is useful in handling experi
The method has now been extended to systems of tw o or more
mental data taken at constant total pressure is shown in Figure
volatile components; in addition to temperature and pressure
2 B . The partial pressure of water from acetic acid solutions is
variables, the variables of compositions of both tho liquid and
plotted against the total pressure existing on tho system. The
vapor phases have been included; the partial heats of solution
slopes of the lines of Figures 2A and 1 are constant and are equal,
calculated b y the method satisfactorily correlate with thermal
respectively, to d log p i/ d log p ” and to d log P / d log p[. The
data.
quotient, d log ]h / d log P, must also be constant and is the value
P A R T IA L A N D T O T A L P R E S S U R E S O F B IN A R Y S O L U T IO N S of the slopes of the lines o f Figure 2 B ; hence these linos of partial
pressure directly plotted against total pressure must be straight,
P lo tte d a g a in s t V apor P ressu res ' o f R e fe re n ce Sub
as they are shown to be.
sta n ce . In most cases (&“) the equation is not needed and vapor
pressures are plotted on logarithmic graph paper (or logarithms
V A P O R C O M P O S IT IO N
are plotted on ordinary graph paper) by three steps: (a) Cor
responding temperatures and vapor pressures o f the reference The vapor composition y lt percentage of more volatile com po
substance are read from a table; temperatures are indicated on nent in the vapor in equilibrium with the liquid of composition X i,
the X axis at appropriate values of pressures, with ordinates is defined by D alton’ s law:
erected accordingly. ( 6) Pressure is plotted from the logarithmic
yi = P i / P or log yi = log pi - log P
scale of the Y axis on the respective temperature ordinates. (The
858
of the slopes of lines of partial pressure and of the lines of total

pressure for the same liquid composition. That these slopes are
constant (i.e., straight lines are obtained for the vapor composi
tion plot) is shown in Figure 4 for aqueous acetic acid solutions.
This algebraic derivation to explain the previously observed
straightness of the lines was suggested by Buffington (S).
A plot of vapor compositions against total pressures also gives
straight lines, as shown in Figure 5 where the data of K eyes ( 6 )
are included. This new plot against total pressures is particu
larly useful in correlating experimental data, which are usually
taken at constant total pressures and are thus plotted directly an
a standard sheet of logarithmic paper.
E Q U IL IB R IU M C O N S T A N T S , R E L A T IV E V O L A T IL IT IE S ,
A C T IV IT Y C O E F F IC IE N T S
By simitar treatment of algebraic quantities representing the ■

slopes of lines, other properties give straight lines, both on the
1000 1500 earlier plot against a reference substance and on the now direct
plot against total pressures. While these functions may bo con
sidered for both components, only the more volatile are included
F igure 1. T o ta l Pressures o f A q u eou s A cetic in Table I, which also gives the slopes of the lines.
A cid S o lu tio n s v s. V a p or Pressures o f W a ter a t
S a m e T e m p era tu res T he equilibrium constant K = y/x may be plotted more readily
than y since, as y approaches zero, the plot extends indefinitely
L in e » o f c o n s t a n t c o m p o s it i o n , in d ic a t e d in m o le % w a te r
while K approaches a constant. The K plots have the same
At a given liquid composition, this may be differentiated with slopes as the corresponding y plots; and the slopes of both may
respect to log j>i (the logarithm of the vapor pressure of pure vary somewhat if y shows very large variations with pressure and
reference substance at the same temperature) : temperature (at constant x) which it usually does not.
Plots of the relative volatility a = K i / K i — y \X s/ x \y i do not
d log y i d log p¡ _ d log P have these disadvantages, since this function does not approach
d log p j d log p? d log zero, and the very slight curvature of the lines is not present.
A useful plot is that of activity or activity coefficient, 7 1 =
The last term is the slope of the lines in the plot of total pres V i/ v ° X \- Figure 6 presents the activities of water and acetic
sures (Figure 1), and the middle term is the slope in the partial acid. Because of the small range of values, the vertical scale is
pressure plot (Figure 2 A ) . Both terms on the right are constant expanded considerably b y plotting logarithms on the ordinary
since the lines of these slopes are straight; the left-hand term vertical ruling of semilog paper; and this expansion could be car
must, therefore, also be constant. It represents the slopes o f lines ried as far as might be desirable to accentuate the spread of the
on a log plot of vapor composition, and is equal to the difference lines or their slope.
“C.
3 500
o
CO
1500
Va por P re ssu re Water, mm.Hg
Vopor Pressu re Water, mn».Hg
F igure 3. P artial Pressures o f A c e tic

Figure 2. P artial Pressu res o f W a ter fr o m A c e tic A cid S o lu tio n s vs. A c id fr o m A q u eou s S o lu tio n s v s.
Vapor ^Pressures o f W a ter ( A ) a t S a m e T e m p e ra tu re s a n d v s. T o ta l V a p or Pressures o f W a ter a t S a m e
Pressu res o f S y stem (B ) a t S a m e T e m p e ra tu re s T em p era tu res
T able I. L og P lots for C onstant V alues of a; (I sosteres ) A P P L IC A T IO N T O H Y D R O C H L O R IC

A C ID -W A T E R
T he first twelve o f these functions give straight lines on logarithm ic paper when plotted
In the application of these methods to gas
against the total pressures on the system or the va p or pressures of a reference substance at
the sam e tem peratures. K / T rs. pressure gives straight lines on sem ilog paper, and E vs. solubilities {1 2 ), the plots of partial pressures
tem perature gives straight lines on standard paper. T he slopes of these lines are the heat of hydrogen chloride over aqueous solutions
ratios indicated in the last colum n. O nly equations and plots for one com p on en t are were straight lines. Similarly the plot of total
indicated, those for a second or subsequent volatile com p on en t are sim ilar; similar equa
pressures gives substantially straight lines, as
tions and plots m ay b e m ade using constant values o f va p or com position where desired.
does the plot of partial pressures of water.
Figure 8 shows that the lines for vapor com
Pr o p Pl o t t e d He a t
Lo g a r it h m ic F o r m E x p o n e n t ia l position are straight within the accuracy of
erty A g a in s t D i f f e r e n t i a l E q . R a t io
J loo P L l
experimental data. A maximum C .B .M . ap
P P ' Log P ■jT L o g P ' * C P ’ C P 'U
d lo g P 1 ' U L' pears which is reflected in the intersections of
dhgp, ./// HP > the lines of total pressure. The C.B.M . is
P, P .!” d lo g p ; V / Logp, ’( u - ^ L o g t f 'C P t ’ C p F
" i also indicated in Figure 8 , where a line of
dfog pi Li D IL Li given weight per cent of hydrogen chloride
P. P L o g P r jy / L o g P > C P p c P c*
d log P L'j Ld (in the liquid) intersects the abscissa of the
d log P Lv Lx Li
P P i Log P ' L o g p/t- C P -- C p / T T corresponding mole per cent of water (in the
d lo g p ; L,° l;
¿/-¿V vapor). The temperature corresponding to
d lo g g i . d logPi _ dim P. Li - L i
y. p; d/ogp,’ d/ogp/ dlogp’ L o g y r ^ —jd -L o g p / i-C y r c p ; V l; the constant boiling point is the ordinate.
d log Hi _ d/og p! , Akerlof and Teare (2) carefully determined the
y, p L o g y p (± -)L o g P t C g p c P ^ '^
d log P d log P T i" ' activities of water in hydrochloric acid solu
dlogK d togP/ d log P L i-L i tions by measurements of electromotive force
K Pi L o g K ••y J ^ L o g p ’t C K 'c p ; ^
d logp ! dlogp/ d log p/ l; rather than that of pressure. The lines rep
d log K d/og p, , 1 1 -,
K P L o g K -( j $ - ) L o g P iC K - c p f e ~ ') resenting their data (Figure 9) are straight,
dJogP d tog P Lg '
which corroborates this method of correlation.
d/oga J logg* J logg, L ,-L t
a P i L. W ^ # LOyfdj r(/ a - c p ; ^ )
d/ogp/ d/ogp/ dlogp/ L°
dbga. dloqyi J h o lt L i- Lx P A R T IA L H E A T S O F S O L U T IO N
a P Logo. ' ^ j ~ - i_ 0g p t C a - c P
d/ o g P d l o g P d/ogp Ly
d/og 7, d/ogp, , -4 The slope of a line in a vapor pressure plot
H,
x p : dlogp/ d/ogpT L ° s 7 ' j f L o g p i< -C 7 - c p ; p L]
is equal to the ratio of the heat quantity in con
dloqTi .dloqpi d/ogp!
_ _A
p Log7p L o g P r C 7 , - c P Ly IL verting to the vapor state one mole of the given
7 d tog P d lo g P d log P Lg
material compared to the molal latent heat at
' E. d -f R dhoT A -.JL
T P d/ogp,’ N F d/ogp/ j r ■ f i g ’i % L ° 9 P P C N F L; the given temperature of the reference sub
stance. The heat required to vaporize one
£ T £ - ¿ T C “T
mole of one component ( L ) from a solution is
«• S e m i - L o g P l o y w i t h f u n c t i o n F E p l o t t e d d i r e c t l y equal to the latent heat of the pure compo
• * S t a n d a r d P l o t o n a r i t h m e t i c g r a p h Pa r e r nent (L °), plus its partial heat of solution ( II) ,
which is equal to and opposite in sign to the
5 / m / l a p E Q U A T IO N S m a y ó e d e r i v e d u s i n g c o n s t a n t v a l u e s o f
heat evolved in dissolving one mole in a
y in s t e a d o f X , E.G.
volume of solution so large that its composi
d/og P Lx • Aa Lx. tion will not change. From the equations of
p p; Log P - L o g P ' t C P 'C p , T p
d/ogp/' l; l; Table I the slopes of the lines of successive
plots are shown constant b y interrelation
with the slopes of lines of previous straight-
line plots. Furthermore, the slopes are shown
A P P L IC A T IO N T O C H L O R O F O R M -E T IIA N O L
successively to be ratios of heat quantities which m ay also be
The careful measurements of Scatchard and Raym ond ( I S ) shown to be constant.
on the system chloroform-ethanol are used to illustrate the appli
cation of some of the correlation plots developed above. Vapor
and liquid compositions were given for isothermal distillation at Tem perolure, °G.
55 65 75 85 95 105 115
three temperatures instead of the more usual isobaric distillation.
Plots were constructed on Figure 7 against ethanol for total pres
sure (P ), equilibrium constants (Ki = yi/xi and K-. = yi/xj),
and relative volatility (a = Ki/Ki). Straight lines result in each
case, even though there is a constant boiling mixture (C .B .M .)
and considerable deviation from ideality. The curved line in the
relative volatility plot represents a constant total-pressure line of
300 mm. of mercury; its shape (which would be straight if
Raoult’s law held) illustrates the irregularity of the system. The
presence of the minimum C .B .M . (maximum total vapor pres
sure) may also be noted. On the plot of total pressures some of
the lines intersect the line for 1 0 0 % chloroform; in others K i = K i
= a = 1 for the C .B .M . These plots offer a simple means for ob
taining the change of C .B .M . with temperature and pressure.
Composition lines can be extrapolated in the a or K plots until Vapor P ressu re W oler, roraHg
they intersect the line for which a — K = I, giving the cor

F igure 4. C o m p o sitio n o f V apors over A q u eou s
responding temperature; the pressure is then obtained from the A c e tic A cid S o lu tio n s v s. V a p or Pressures o f
plot of total pressures. W a ter at S a m e T em p era tu res
uncertainty was also determined. The length of the arrows

indicates the uncertainties involved in the lower direction only,
since it appears that these values are too high. From Figure 9
the slopes o f the lines of activities were determined by the method
of least squares to give the ratio directly. Finally, the values of
the ratio as calculated from the partial heats of solution (7.3) and
the integral heat of solution (1 4 ) were plotted. Excellent agree
ment is shown in Figure 10 among the data obtained by these
independent types of measurement and backgrounds of theory.
O T H E R A l’P L IG A T IO N S
Numerous other systems, some of three or more components,

have been examined, and the results have corroborated the re
sults above. A few studies of the three systems discussed above
were selected because they illustrate the mutual correlation of:
(a) usual vapor composition data at constant pressures (water-
acetic acid and hydrogen chloride-w ater); (b) vapor composition
F igu re 5. C o m p o s itio n s o f Vupors over data taken at constant temperature (chloroform-ethanol); (c)
A q u eou s A c e tic A cid S o lu tio n s its. T ota l activity coefficients obtained from c.m .f. measurements (hydro
Pressures o f S y stem a t S a m e T e m p e ra tu re s gen chloride-w ater); and ( d ) heats of solution directly obtained
T h e s q u a r e s r e p r e s e n t d a t a o f K e y e s (<S) (water-acetic acid and hydrogen chloride-water). Furthermore,
the systems represent the three usual types of binary mixtures—
The lines of both Figures 4 and 5 are nearly horizontal, an i.e., no C .B .M . (w ater-acetic acid), minimum C .B .M . (chloro
indication that the vapor composition does not change greatly form-ethanol), and maximum C .B .M . (hydrochloric acid-water).
with a moderate change of pressure (the normal case). In the One of the most useful features of these plots is the ease of work
particular case where L \ = L \ and Raoult’s law holds, the slopes ing back and forth between constant pressure and. constant tem
reduce to zero; and the lines are horizontal. The vapor composi perature conditions, depending on the method of determining or
tion ( x , y) diagrams are then identical at all pressures; and all of using the data.
the lines of partial pressure and of total pressure (including those
of the pure components) are parallel at 45° on a log plot such as
Figure 1, since the molal latent heats are equal for the pure sub
stances and every possible mixture at the same temperature.
Furthermore, the heats of solution would be zero, as is neces
sarily true whenever Raoult’s law holds. (The coastancy of the
molal latent heat is assumed in the familiar M cCabe-Thiele
method of calculation of distilling columns, which thus assumes
that the x, y diagrams are identical at all pressures and the vapor
composition lines on the log plot are all horizontal.) As indicated
above, the slopes of the equilibrium constant ( K ) plots are the
same as the corresponding vapor composition plots; hence the
heat terms are also the same.
The relative volatility function is useful because its equation is
free of the term for the total latent heat of the mixture. The
slopes of the lines on its respective plots are equal to the differ
ences o f the slopes o f the vapor composition lines for the two com
ponents, and also to the differences of the slopes of the partial
Vapor P ressu re Water, mm.Hg
pressure lines. Even more interesting is the plot for activities or
Temperature,°C.
activity coefficients; the slopes are the same for both and give
directly the ratios of heats of solution to latent heats. Since
latent heats are usually known (or may be determined b y means
of the basic log plots or other methods quite accurately), the
heats of solution may thus be obtained from this plot. When de
sirable, the slopes may be increased for more accurate reading by
increasing the scale of the Y axis. If Raoult’s law holds, the lines
all have zero slope and coincide with the X axis. I f it does not
hold, deviations are indicated by the heat of solution; on the
plot of activity coefficients, this shows up directly by the relative
slopes of the lines.
1 1 K A T Q U A N T I T I E S IN H Y D R O G E N C H L O R I D E - W A T E R
This system was selected to test the correlation of heats since

data are available for partial pressures (5), vapor compositions
(7), activity coefficient (2), and heats of solution (73, 14), as
Vapor P re ssu re Waler, mmHg
functions of temperature and liquid composition. Figure 10
shows the ratio of partial heat of solution to the heat of vaporiza
F igure 6 . L o g a r ith m s o f A c tiv ity C oefficien ts o f W a ter
tion of pure water determined from these three sources. The
in A c e tic A cid S o lu tio n s an d o f A c e tic A cid in A q u eou s
slopes of the lines in a partial pressure plot for water (.5) were S o lu tio n s ag a in st V a p or Pressures o f W a ter a t S a m e
calculated b y the method of least squares from which the range of T e m p e ra tu re s
calculations for distilling columns. It if

more advantageous to use y as the par
ameter in the plot of log P vs. log p°, and
it is necessary to have the specific heats of
the pure components in the vapor state,
the latent heat of the pure component
used as reference in the temperature range,
a log plot of P against p° at constant
yi, and the x, y diagram at the pressure
desired. First a suitable enthalpy chart
for the vapor is constructed b y making
the usual assumption of no heat effect in
mixing vapors at the same temperature,
and assuming convenient datum points
Vapor P ressu re Ethanol, mm.Hg
for each pure component. Combining this
latter chart with the log plot of P vs. p]
and making use of the slopes of the lines,
the Ponchon diagram is readily completed
100 200 300 400 for the chosen pressure at which the x, y
Vapor Pressure Ethanol, mm.Hg diagram is known.
S O L U T IO N S O F N O N V O L A T IL E
M A T E R IA L S
Some of these plots, particularly those

for activity or activity coefficient, are use
ful with liquid solutions o f nonvolatile
materials, and with hydrates and similar
types of associated materials. The activity
coefficient plot is merely the ratio of the
partial vapor pressuro to the partial pres
sure according to R aoult’s law, plotted on
log paper against the vapor pressure of
the pure liquid. The line for the vapor
pressure of the pure liquid is, of course,
horizontal; and the variation from zero
slope indicates directly the heat o f solu
tion, of association, and 'of chemical com
too 200 300 400
Va po r P re ssu re Ethan ol, mm.Hg
bination or hydration (for solid crystals
Vapor P re ssu re Eth a n o l, mm,Hg
with water of hydration). A useful
graphic tool is thus provided. It has
F igu re 7. P lots fo r C h lo r o fo r m -E th a n o l S o lu tio n a g a in st V a p or Pressures been applied to the data for various
o f E th a n ol a t S a m e T e m p e ra tu re s : T o ta l P ressu res, E q u ilib riu m C on sta n ts
K \ o f C h lo r o fo r m , E q u ilib riu m C on sta n ts K i o f E th a n ol, a n d R elative systems and gives excellent correla-
V ola tilities a o f C h lo r o fo rm
T em perature (*C .)
If partial pressure data are available, heats of solution may be
calculated; not only may the methods of correlation between
heats and other quantities be used.to check the respective values,
but in many cases few data of different sources may be used if
necessary to predict with some assurance of accuracy the entire
ramifications o f the system. Thus, the activity data of Akerlof
and Teare ( 2 ), coupled with data on the vapor composition of
water (7) would serve to build the whole hydrogen chloride-
water system throughout an entire range of temperature, pres
sure, partial pressure, vapor composition, heat of solution and
vaporization, and other dependent quantities. Other examples of
extrapolating data taken at only one temperature or one pressure
throughout an entire range of temperatures and pressures might
be given; in some cases the use of reduced temperatures and pres
sures (9 ) may allow further expansion of a paucity o f data or more
precise predictions of desired quantities from other quantities.
Enthalpy charts may bo constructed b y the use of these
methods for use in such work as the Ponchon method of plate
Figure 8. C o m p o s itio n s o f V apors over A q u e o u s H y d ro 200 300 500 8 0 0 1000

c h lo r ic A cid S o lu tio n s v s . V a p o r Pressu re o f W a ter at
S a m e T e m p e ra tu re s Va p o r P r e s s u r e of Wa t e r (mm)
t.ion. Thus Figure 11 shows the activities of water from aqueous centrations). The plot is an excellent correlation, however,
sodium hydroxide solutions. The data of Akerlof and Kegeles where the precision of measurements was good.
(f) for concentrations below 14 molality are plotted, since there A corollary follows in tho basic relation of activity coefficient
were experimental difficulties above that concentration. Errors (y ) and the e.m.f. of an electrolytic cell ( E ) :
in data at the highest temperature (70° C.) are also referred to;
„ R T , E II .
and some of the points at higher concentrations and at 70° C.
N F 7 Xl or T = N F 7 Xl
fall away from the curves (as well as those at 0° C., and high con-
If this is differentiated at constant x 1 with respect to the log of the
vapor pressure of a reference substance:
T e m p e ra tu re ,°C . d (E / T ) R d in y _ R II
d In p i “ N F d In p i N F L \
thus ^ = ( J - p , dog p i + C
Thus, if E / T is plotted for each temperature and concentration of

sodium hydroxide against the log of the vapor pressure of water,
20 30 40 60 100
Vapor P re ssu re Water, mm.Hg
Figure 9. A ctivities o f W a ter in A q u e o u s H y d ro c h lo ric

Acid S o lu tio n s v s. V a p or Pressu res o f W a te r a t S a m e
T em p era tu res (fr o m E le ctro c h e m ica l D ata )
• F r o m pa rtial p re ssu re s
oF rom p a rtia l h e a ts at 2 0 * C .
■Fro m a c t iv it ie s by E.M.F.
° From * p a rtia l h e a ts at 7 0 *C .
e From p a rt ia l heats at 2 5 *C.
15
Weight % H CI
Figure 10. R a tio o f Partial Ilc a ts o f S o lu tio n o f W a ter in H y d ro c h lo ric A c id S o lu tio n s us. W e ig h t Per C en t H y d ro c h lo ric
A cid in th e S o lu tio n
—Ui/Li = s lo p e o f lin e s o n lo g p l o t o f a c tiv itie s (F ig u r e 9 ). D a ta f r o m t h e lite r a tu r e as f o l l o w s : P a rtia l p re ssu re s (5 )t a rro w s s h o w lo w e r l i m i t
o f u n c e r t a in ty . P a rtia l h e a ts at 20° a n d 70° C . (13). P a r tia l h e a ts a t 23° C . (14). A c tiv itie s b y e le c t r o m o t iv e f o r c e m e a s u r e m e n ts ( 2 ) ; d a sh e d ]
lin e p a sses th ro u g h th e s e p o in t s s in c e th e y a re o f m o s t p r o b a b le a c c u r a c y .
Temperature (Absolute)
12.1980
11.0200
10.0434
9.3814
8 .0 9 6 3 „
7 . 0 4 4 8 -2
I CDo
5 .0 2 3 9 x
o
A.0 4 8 4 °
o
o
2
1 .0 0 4 5
1.5 2.0 2.25

Vapor Pressure Water, mm.Hg
Figure 11. A ctivities o f W a ter in A q u eou s

S o d iu m H ydroxide S o lu tio n s v s. V a p or Pressures L o g Vopor Pressure of Woter
o f W a ter at S a m e T em p era tu res
F ig u re 12. E le ctro m o tiv e F orces o f S o d iu m H ydroxide
C ell (1) D ivided b y A b so lu te T e m p e ra tu re s , us. V apor
Pressures o f W a ter a t S a m e T e in p cra I ures
12.1980
11.0200
1 0 .0434
In the Clausius-Clapeyron equation for a single component
9.3814
8 .0 9 6 3 log pi | g j, + C '
7.0448 £
6.0351 1 L / R may be regarded as roughly constant over narrow tempera
o
tp tures ranges. If the last two equations are combined,
4 .0 4 8 4 g
3 .3 2 3 4 z ? = b ^ + C "
or E — b + C " T (approximately)
1.8718 o
1.4191 I where 6 is a constant. Thus, a plot of e.m.f. for any cell vs. tem
1.0045
perature should be almost linear but would not be so close to
0 .7 5 6 2
linear as a plot of E / T vs. log p\. The data for aqueous sodium
0 .5 0 5 8
hydroxide (I) are plotted in Figure 13. Approximately straight
0 .3 5 2 3 lines result, an indication of the utility of this relation. This
0 .2 5 0 5
straight-line relation has long been used empirically as a tempera
ture coefficient for cells; this derivation shows how the constants
may be evaluated and indicates that its use over somewhat wider
Temperature, °C. ranges would be justified.
Similar plots have been made of the variation of electromotive
F ig u re 13. P lo t o f E le ctro m o tiv e F orces o f force of other cells, and the conclusions have been substantiated
S o d iu m H y droxide Cell (I ) ns. T em p era tu res in each case.
(S ta n d a rd C oord in a tes)
ACKNO W LEDG M ENT
Th e authors wish to thank Edward G. Scheibel for suggestions,

straight linos result (Figure 12), since all other quantities are Robert F. Benenati for drafting the figures, and Arthur H. Luley
constant. In this cell there are several electrode potentials; for help in the plotting.
and as over-all E and y are considered, each individual potential
N OM ENCLATURE
or any sum would also give straight lines. As in Figure 11, the
experimental errors which were greater for higher concentrations b = constant
are reflected in the upper lines. The correlation represents an c = constant (different for each equation and usually the anti
log of C )
interesting method of handling experimental data, by comparison
C = constant of integration (different for each equation)
with the relatively tedious way which must be used when iso E = electromotive force
therms arc considered rather than isostercs. F = Faraday constant
Hi = partial heat of solution of com ponent 1 (4) Gihnont, R-, a n d Othmer, D . F., I nd. E ng. Chem., to be pub
Hi = partial heat of solution of component 2 lished.
K\ — iji/ xi or equilibrium constant for com ponent 1 (5) International Critical Tables, Y ol. I l l , p. 301, New York,
Ki = y-i/xi or equilibrium constant for com ponent 2 M cG raw -H ill B ook Co., 1928.
L = molal latent heat of compound in question (6) Keyes, D . B., I n d . E n g . C hem ., 25, 569 (1933).
L' = molal latent heat of reference compound (7) Othmer, D . F „ Ibid., 20, 743 (1928).
hi = molal latent heat of component 1 from solution (8) Ibid., 32, 841 (1940).
Li = molal latent heat of component 2 from solution (9) Ibid., 34, 1072 (1942).
LI = molal latent heat of pure component 1 (10) Ibid., 36, 669 (1944),
LI = molal latent heat of pure component 2 (11) Othmer, D . F., and Sawyer, F. G „ Ibid., 35, 1269 (1943).
Lx = molal latent heat of liquid in equilibrium with vapor of (12) Othmer, D . F „ and White, R . E ., Ibid., 34, 952 (1942).
composition iji (13) Payn, R . C., and Perman, E . P., Trails. Faraday Soc., 25, 599
L.j — molal latent heat of vapors in equilibrium with liquid of (1929).
composition X i (14) Rossini, F . D ., Bur. Standards J. Research, 9, 679 (1932).
jV = valence number (15) Scatchard, G., and Raym ond, C. L., J. Am . Chem. Soc., 60,
Pi = partial pressure of component 1 at given temp. 1278 (1938).
pi = partial pressure of component 2 at given temp.
P resen ted before the D ivision o f Industrial and E ngineering C hem istry a t
p° = vapor pressure of pure component 1 at given temp.
the 107th M eeting o f the A m e r i c a n C h e m c a d S o c i e t y in C leveland, Ohio.
p i = vapor pressure of pure com ponent 2 at given temp.
P = total pressure at given temperature
P ' = total pressure of reference substance at given temp.
R = gas constant
T = absolute temperature Correlating Vapor Pressure and
ii = mole % of component 1 in liquid
Xi = mole % of component 2 in liquid E q u ilib riu m C on stan t D ata—Correction
y, = mole % of com ponent 1 in vapor
yi = mole % of com ponent 2 in vapor The author wishes to call attention to an unfortunate error
a = relative volatility = K i / K i which he has just discovered in Figure 3 o f this paper which ap
7 i = activity coefficient of com ponent 1 peared in the July issue of I n d u s t r i a l a n d E n g i n e e r i n g
it = activity coefficient of component 2 C h e m i s t r y , pages 669-72. The calibrations of the middle scale,
for butane, should all be divided b y 10. That is, the top of
the scale starts at 0.4 and the bottom ends at 10.0.
(1) Akerlof, G., and Kegeles, G., J . A m . Chem. Soc., 62, 620 (1940). D. F. O t h m e r
(2) Akerlof, G., and Teare, J. W., Ibid., 59, 1855 (1937). P o ly te c h n ic In s titu te
(3) Buffington, R. M., Servel, Inc., private communication. B r o o k ly n , N. Y.
SPECIFIC HEATS AT LOW TEMPERATURES OF

T ita n iu m a n d T ita n iu m C a r b id e
K . K . K ELLEY
Pacific Experiment Station, U. S. Bureau o f Mines, Berkeley, Calif.
L ow -tem p era tu re sp ecific h e a t d a ta are r e p o rte d fo r tita n sisted of spheroidal particles with diameters up to 4 mm. A c
iu m m e ta l a n d tita n iu m c a r b id e in th e te m p e ra tu re ra n g e cording to analyses furnished by the company, the purity was
51 ° to 298° K . T h e sp ecific h e a t cu rves o f b o t h su b sta n ces 98.75%, the principal impurities being 0.50% silicon, 0.27% iron,
are n o r m a l. T h e e n tro p ie s o f th e m e ta l a n d c a r b id e h ave and 0.15% vanadium. Spectrographic analyses for other metals
been c o m p u te d as S ,,,.,! = 7.24 =*= 0.07 a n d S 2i,.n = 5.8 increased the total to 99.83% . The sample used in the present
=*= 0.1, resp ectiv ely . measurements contained 241.74 grams.
The titanium carbide was made from rutile of 99 % purity and
petroleum coke in a resistance furnace, the final temperature
T H E determination of low-temperature specific heats and being near 3000° C. The purity o f the carbide was 96.08%.
entropies of titanium-containing compounds is a field that This is not as high as would be desirable, but it was the most
has been neglected. Only two compounds have been suitable sample available. Spectrographic analyses for impuri
studied directly, titanium dioxide (rutile) by M cD onald and ties raised the total constituents accounted for to 99.25% ; the
Seitz (7) and titanium tetrachloride by Latimer (5). The en remainder was presumably combined oxygen and nitrogen. For
tropy of titanium gas was obtained by Kelley (4) from spectro the present purpose it was assumed that the material contained
scopic data, and that of titanium tetrachloride gas by Y ost and 1.82% titanium dioxide and no nitride. This assumption brings
Biair (8 ) from molecular constant data. The only previous value the total accounted for to 99.94% . A 370.65-gram sample was
for the entropy o f solid titanium is the result calculated by Lewis used in the specific heat measurements.
and Gibson (6') from a single low-temperature specific-heat Specific H eat Measurements. The apparatus em ployed has
measurement. As a first step toward remedying this condition, been described (£). The results are expressed in defined calories
determinations of the specific heats o f titanium metal and of (1 calorie = 4.1833 international joules) per gram molecular mass
titanium carbide have been conducted throughout the tempera of material. In accordance with 1941 International Atomic
ture range 51 0 to 298 ° K . The results obtained and the entropies Weights, the atomic weight of titanium is taken as 47.90 and the
computed from them are the subject o f the present paper. molecular weight of titanium carbide as 59.91.
The titanium and titanium carbide were furnished b y the The specific heat results were corrected for the impurities pres
Titanium Alloy Manufacturing Company. The titanium, made ent on the assumption that the specific heats are additive. This
by sodium reduction of the tetrachloride, was malleable and con correction ranged from 0.08 to 0.20% for the titanium, depend-
The results of entropy calculations follow:
C a !./° K ./M o le
Ti T iC
0 — 5 0 .1 2 ° K ., e x t r a p o la t e d 0 .4 0 1 0 .0 8
5 0 .1 2 ° - 2 9 8 .1 6 ° K ., g r a p h ic a l 6 .8 4 4 5 .7 1
3 » 8.i# 7 .2 4 * ± « 0 .0 7 5 .8 =±* 0 .1
Extra allowance of error has been made in the entropy of

titanium carbide to account for possible uncertainty in the cor
rection for impurities.
The entropy of titanium previously in use is 6.6 ( 4 , 6 ). This
figure is based upon a single mean specific heat measurement
made thirty years ago and deserves no weight in comparison
with the present result. N o previous low-temperature specific
heat or entropy data are available for titanium carbide, but an
estimate of S 235.16 = 5.5 had been made (3) by analogy with sili
con carbide. The present result, of course, rests upon firmer
ground.
(1) Kelley, K . K „ J. Am . Chem. Soc., 62, SIS (1940).

0 100 200 300 (2) Ibid., 63, 1137 (1941).
T 'K . (3) Kelley, K . K ., U. S. Bur. Mines, B u ll. 4 0 7 (1937).
F i g u b e 1. S p e c i f ic H e a t o f T it a n iu m (A ) (4) Ibid., 434 (1941). ,
a n d T it a n iu m C a r b id e (B ) (5) Latimer, W . M „ J. Am. Client. 8 0 c., 44, 90 (1922).
(6) Lewis, G. N., and Gibson, G . E., Ibid., 39, 2554 (1917).
(7) M cD onald, H. J „ and Seitz,' H „ Ibid., 61, 2405 (1939).
(8) Y ost, D . M „ and Blair, C „ Ibid., 55, 2010 (1933).
ing on the temperature, and from 0.23 to 12.97% for the titanium
carbide. The corrected specific heat values are given in Table 1 P u b l i s h e d b y p e rm issio n o f th e D ir e c t o r , U . S. B u re a u o f M in e s.
and graphically in Figure 1.

N o irregularities were found in the specific heats of either sub
stance. The specific heat per mole of titanium carbide is lower
than that per gram atom of titanium metal at temperatures be
low 189° K . This decrease in the specific heat of a carbide be
low that of the corresponding metal is noted also in the data for
Synthesis o f Racem ic M enthol from
silicon carbide (2) and tantalum carbide (/ ) . T h ym ol— Correspondence
Entropies at 298.16° K. The entropies were computed by
graphically integrating under C p vs. log T curves between the S ir : A. L. Barney and H. B. Hass have stated [Ind. E ng.
limits of 60.12° and 298.16° K ., and adding the portions obtained Chem ., 36, 85 (1944)] that di-menthol “ will liquefy on prolonged
by extrapolating the measured specific heats below 50.12° K . standing at 30° C., due to the ease of interconversion of at least
In the case of titanium the D ebye function, Z )(358/T ), was found two crystalline forms” .
to fit the measured results to 120° K . and was employed in the Experiments in our laboratory have established the fact that
extrapolation. For the titanium carbide the specific heat at the lowering of the congealing point and melting point of di-menthol
lowest temperature studied is so low that no significant error is due, not to an undefined interconversion, but rather to the
can be introduced by employing the T 3 law for extrapolation. hygroscopic nature of racemic menthols. A sample of racemic
menthol in our laboratory showed a double congealing point of
27.5° and 31.3° C., and a melting point of 33.0° C. After stand
ing 72 hours exposed to air, the congealing points were 25.0°
T able I. S pecific H eats of T itanium and of T itanium and 29.0° C., and the melting point was 31.5° C. The di-mentho!
C arbide was then dried under vacuum, and the dry product showed con
j» OK . T, gealing and melting points identical with those taken before it
Cp ° K. Cp T, ° K. Cp
T it a n iu m 0 (*4tom ic W e ig h t => 47.9 G ra m s )

was exposed to air.
53 .5 1 .2 7 8 1 1 2 .7 3 .7 6 5 2 1 4 .6 5 .4 4 5
Because of the hygroscopic nature of di-menthol, in contrast
57 .1 1 .4 6 4 1 2 3 .2 4 .0 6 1 2 2 4 .9 5 .5 3 8 to the stable and nonhygroscopic character of 1-menthol, careful
61 .4 1 .6 7 9 1 3 3 .3 4 .3 0 6 2 3 5 .0 5 .6 0 7
65 .9 1 .9 1 5 1 4 3 .4 4 .5 1 6 2 4 4 .9 5 .6 8 3 drying should be required before the melting and congealing
70 .5 2 .1 4 2 1 5 4 .0 4 .7 0 7 2 5 5 .5 5 .7 5 4
75 .1 2 .3 5 7 1 6 3 .9 4 .8 7 8 2 6 5 .7 5 .8 2 8
points are taken. This matter becomes particularly important
80 .1 2 .5 8 2 1 7 4 .3 5 .0 1 5 2 7 6 .1 5 .9 0 1 If racemic menthol is recognized as a U.S.P. product.
S3 .8 2 .7 3 7 1 8 4 .2 5 .1 4 2 2 8 5 .8 5 .9 5 1
92 .7 3 .1 0 1 1 9 4 .4 5 .2 5 9 2 9 5 .1 5 .9 8 4
102 .9 3 .4 6 0 2 0 4 .7 5 .3 5 5 A lfred O fnek
G iv a u d a n -D elaw ann a, Inc.
T it a n iu m C a r b id e 1» ( M o le c u la r W e ig h t «= 5 9 .9 1 G ra m s )
New Y ork 18, N. Y.
55 .1 0 .3 1 2 1 1 5 .6 2 .3 8 9 2 1 6 .8 6 .0 9 7
58 .6 0 .3 S 4 1 2 5 .7 2 .7 8 9 2 2 6 .3 6 .3 8 0
62 .3 0 .4 7 3 1 3 5 .6 3 .1 9 0 2 3 6 .0 6 .6 3 9
65 .9 0 .5 7 1 1 4 6 .0 3 .6 0 6 2 4 6 .2 6 .9 2 5 S ir : It is true that our sample which showed lowering of melt
69 .8 0 .6 8 6 1 5 6 .2 4 .0 1 0 2 5 6 .0 7 .1 6 8
74..2 0 .8 2 1 1 6 6 .5 4 .4 0 2 2 6 6 .4 7 .4 2 0
ing point was in an unsealed bottle and not perfectly protected
81 ,.2 1 .0 5 5 1 7 6 .2 4 .7 5 0 2 7 6 .3 7 .6 4 9 from the air. Accordingly we feel that Ofner is correct and be
86 ,.5 1 .2 4 2 1 8 6 .3 5 .1 1 3 2 8 5 .7 7 .8 3 0
95 ,,7 1 .5 8 7 1 9 6 .8 5 .4 7 4 2 9 4 .9 7 .9 8 7 lieve no further comment is necessary.
105 .3 1 .9 6 0 2 0 6 .5 5 .7 7 8
° Cp ex p r e s se d as c a lo rie s p e r g r a m a t o m .
H. B. H ass
* Cp ex p r e s se d as c a lo rie s p e r g r a m m o le c u le . P C K D U E U N IV E R SIT T
L afatette, I nd.
Events during tlie Month, of Interest to
Chemists, Chemical Engineers, Executives
evtewe A lu th e C _ Â it a u
7
If A u g u s t 2. J. D . M aleohnson, technical director o f R o b gram s for reconversion o f industry, where an y questions o f

ert Gair C o., In c. (m anufacturers o f containers) and form erly violation o f antitrust laws are concerned, m ust b e subm itted
chief o f the technical section o f paperboard division o f the to him for a d vice and individual clearance before being put
War P roduction B oard, says in du stry has now passed the into operation 1. ~ ~ U n i t e d States and U nited K in gdom an
“hump” o f the crisis in shipping containers. nounce agreem ent w hereby w orld oil resources are to be pooled
If A u g u s t 3. Further cutback in alum inum p rodu ction in for the use o f all peaceful nations. A n international con
volves closing tw o governm ent-ow n ed plants operated b y ference will be held a t w hich all oil-produ cing and consum ing
Aluminum C o. o f A m erica a t R iverban k , Calif., and B urling nations will be in vited to subscribe to the policies form u
ton, N . J., and reducing production o f a plant a t L os Angeles lated2. ~ ~ C i r c u i t C ou rt o f A ppeals upholds A rm y, N a v y ,
by one th ird‘ . ~ ~ S e n a t o r H arry S. Trum an, vice-presidential and other governm ent agencies in their right to reduce or
nominee o f the D em ocra tic Party, resigns as chairm an of the cancel privately set roya lty rates to patent owners on m a
Senate W ar Investigating C om m ittee. ~ ~ W P B places terials needed in prosecution o f war.
cyanide chem icals used in blueprint and ph otographic dyes If. A u g u s t 9. N ew Y o rk State F ood M erchants A ssociation
and in textile industries under allocation, effective Septem ber ad opts resolution calling on those adversely affected b y State
l.~ ~ H a n s W . H uber, president o f J. M . H u ber C orp., A ttorn ey ’ s recent ruling restricting sale o f vitam ins to licensed
states th at although carbon black has been placed under pharm acies and registered stores to ’ cooperate in seeking an
allocation b y W P B , there will be no rationing or q u ota re in ju n ction restraining enforcem ent o f ruling.
strictions for printing inks.
If A u g u s t 10. T ariff Com m ission discloses that sales o f all
If A u g u s t 4. W ar M obilization D irector Byrnes issues synthetic organic chem icals, exclusive o f explosives and other
directive giving W a r M an pow er Com m ission virtual v eto chem icals under ordnance control, top ped $1,400,000 last
power ov er plant reconversion to civilian good s ou tp u t1/-'-""-' year, an increase o f 5 4 % ov er the previous year and an all-
Senate M ilitary C om m ittee approves bill setting u p a super tim e high.
government agen cy to direct reconversion and establishing
federal standards for postw ar unem ploym en t paym en t to If A u g u s t 11. W PB orders shipm ents o f nitric acid to
discharged war w ork ersr--'~ S en a tor G u y M . G illette states plants m aking nitrogenous fertilizers to cease after O ctober 1.
that Senate Su bcom m ittee on Farm C rop U tilization w ill soon If A u g u s t 12. M agnesium plants will b e slashed fu lly 5 0 %
resume investigations o f alcoh ol su pply and prospective pro as o f Septem ber 1, according to reliable report. T h is drastic
duction. cu tback w ill in som e instances determ ine w ho is to rem ain in
1f A u g u s t 6. B . F . G ood rich C o. announces plans to spend the m agnesium produ cin g field after the w a r .~ ~ O f f ic e o f
upward o f $1,000,000 to build a research and developm ent Price A dm in istration sets specific cents-per-pound ceiling
laboratory after the war. prices on all types o f boiled-dow n soap stock m ade from
vegetable oil foots now being offered for sale, b oth dom estic
If A u g u s t 7. D epartm ent o f Justice announces that dis
and im p o r t e d /~ ~ R e d u c t io n s in tetraethyllead going in to
cussions betw een industry ad visory com m ittees and W P B on
civilian gasolines are estim ated to save enough lead to produce
reconversion and industrial dem obilization problem s while
210,000,000 gallons o f 100-octane aviation gasoline each
war continues d o n ot constitute v iolation o f antitrust laws.
m on th , according to a statem ent b y D ep u ty Petroleum A d
'''""■'British foreign m ission licensing exports from this coun
m inistrator R a lph K . D avies.
try will be liquidated O ctober 1, with con trol vested in U . S.
agencies, it is confidently forecast in foreign trade circles. If A u g u s t 13. Petroleum In du stry W a r C ouncil says a b ou t
Rayon produ ction b y A m erican producers in the second 8 0 % o f the N a tion ’s high-octane gasoline ca p a city is prim ed
quarter o f 1944 sets a new high. to sw itch over to new supergasoline d? -eloped for giant
bom bers w ith on ly m inor changes in facili is. N ew prod u ct
A m erica’ s w ar-born $750,000,000 synthetic
H A u g u s t 8.
is said to have “ v astly im proved octan e rc ig” ov er the 100-
rubber plant m a y p a y for itself w ithin 4 years after Far East
octan e fuel now used in battle planes.
rubber plantations resume norm al operation, James J. N ew
man, v ice president o f B . F . G oodrich , told business leaders in If A u g u s t 14. M on san to Chem ical C o. m ounees th at its
A t la n ta .~ ~ A tto r n e y G eneral B iddle announces th at pro- St. Louis plant will begin volum e prod u ction o f D D T in -
1 Chem . E n g . N ew s , 2 2 , 1393 (A u g . 25 , 1 9 4 4 ). » Ib id ., 2 2 , 139 2 (A u g . 25 , 1 9 4 4 ).
867
secticide Septem ber l 1. ~ ~ N a t i o n a l A ssociation o f M an ufac plans big expansion program w hich would in volve borrowing
turers asks Senate M ilita ry Affairs C om m ittee to delegate in excess o f 320,000,000 to finance research and for additional
disposal o f surplus g ood s, w orth up to 3100,000,000,000, to a plant fa c ilit ie s .~ ~ W P B officials announce there will be
com m ission o f seven in d u str ia lis ts.~ ~ F ir e sto n e T ire & R u b im m ediate return o f distillers to the war alcohol production
ber C o., announces it has developed a process to produce program Septem ber 1, and th at decision for further re
foam rubber from syn th etic latex. sum ption o f beverage distillation will n ot be m ade until late
fa ll.~ ~ S e n a t o r O ’D aniel (T exas) predicts oil firms Mil
1f A u g u s t 15. YVPB ad opts new casein allocation program
b u y U nited States pipe lines at 10 cents on the dollar. ~ ~
to encourage dom estic production, and authorizes dom estic
President subm its A n glo-A m erican oil agreem ent to Senate
processors to im port a pou nd o f acid casein for each pound
for ratification as a treaty. D eclared purpose o f agreement
o f rennet casern produced during first h alf o f 1944 a t W P B ’s
is fair utilization o f w orld’s petroleum resources.
request. ~ ~ N e w sulfuric acid facilities are under construction
a t Searsport, M e., B altim ore, M d ., W ilm in gton, .N . C., if A u g u s t 25. J. A . K ru g, w ho left W P B last A pril to
F ron t R oy al, V a., H ou ston , Texas, B a ton R ou ge, La., a ccep t a com m ission as Lt. C om dr. in N a v y , takes over as
D om ingues, Calif., and G arfield, U tah, to aid production A ctin g Chairm an o f W P B in absence o f D on a ld M . Nelson
o f superphosphate fertilizers. W P B Chem icals Bureau o f on trip to C h in a .~ ~ S e n a t e passes surplus p roperty disposal
ficials p oin t o u t th a t m an y petroleum refineries form erly bill in volvin g upward o f S100,000,000,000 o f postw ar govern
using virgin siflfuric acid for uses other than 100-octane m otor m ent surpluses under an 8-m em bered board w ith tigh t Con
fuel production w ill be obliged to a ccep t fortified spent acid. gressional c o n t r o ls .~ ~ O il agreem ent between the United
States and Britain will be the su b ject o f pu blic hearings by
If A u g u s t 16. W P B places allocation control on uranium. the Senate Foreign R elations C om m ittee, Chairm an Con-
if A u g u s t 18. W ar L ab or B oard appeals to officers o f naliy announces. ~ ~ W P B rem oves allocation controls from
U nited M in e W orkers D istrict 50 to term inate a strike a t civilian produ ction o f styrene and butadiene, and states that
O ldbu ry E lectro-C h em ical C o., N iagara Falls, N . Y . th ey are now in sufficient su pply because o f im proved pro
du ction techniques.
if A u g u s t 20. T w en ty-th ird m on th ly report on chem icals
allocations continues to reflect shortages in individual prod If A u g u s t 26. Chrysler C o r p . announces new machine
ucts. W P B rem oves control over beryllium except beryllium w hich will perm it continuous in jection m olding and extrusion
c o p p e r . ~ ~ J . H . Pew, president o f Sun Oil C o., assails pe o f an y kind o f plastics or rubber. ~ ~ S i d n e y J. Weinberg.
troleum p a ct as step to “ superstate cartel in all parts o f the W P B ’s “ peacem aker” , is resigning as W P B v ice chairman
w orld for the postw ar years” in open letter to Senator Torn in charge o f “ special problem s” . ~ ~ B a s i c rubber regulations
C onnally, chairm an o f Senate Foreign R elations C om m ittee. have been revised b y the R u bb er D irector to reflect progress
in conversion from crude to syn th etic rubber (R evised Rubber
if A u g u s t 22. S ocon y V a cu u m Oil C o. announces com ple
Order R - l and appendixes).
tion o f nationw ide 100-octane construction program when a
new unit using syn th etic bead catalyst (the first east o f the If A u g u s t 28. G eneral T ire and R u bb er C o. announces new
R o c k y M oun tains) was placed in operation a t Paulsboro, m ethod o f processing synthetic rubber b y adding carbon
N . J. N ew unit will turn ou t thousands o f barrels o f aviation black to rubber while still in liquid s t a t e .~ ~ C h e s t e r Bowles,
gasoline d a ily .~ ~ H o u s e o f R epresentatives passes surplus O P A adm inistrator, in address before Seattle C ham ber of
w ar p rop erty disposal bill w hich will give one man responsi C om m erce, offers 11-point program for postw ar reconversion,
b ility for 375,000,000,000 surplus w ar m atériel; Senate designed to provide m axim um em ploym ent and econom ic
counters w ith proposal o f com m ittee o f eight. P roblem goes stability. ~ ~ J . A . K ru g, acting W P B chairm an, tells Senate
to c o n fe r e n c e .~ '~ A tto r n e y G eneral B iddle tells civic organi W a r Investigating C om m ittee he is “ in com plete accord”
zation in L os Angeles th at antitrust prosecutions, held up b y with D on a ld M . N elson ’s plan for a gradual shift to civilian
agreem ent w ith A rm y and N a v y , w ill be resumed within ou tpu t in noncritical labor areas.
n ext 6 m onths.
If A u g u s t 29. W P B orders further cutbacks in aluminum
1f A u g u s t 2 3 . Petroleum A dm inistration for W ar is com production, am ounting to approxim ately 30,000,000 pounds
pleting a refinery expansion su rvey to determ ine am ou n t o f o f ingots a m o n t h .~ ~ D e p a r t m e n t o f Justice charges most
refinery expansion and im provem ents necessary during post o f w ood alcohol industry w ith violation o f antitrust laws by
w ar p e r i o d .~ ~ R i e e Grow ers A ssociation o f California an organizing conspiracy to elim inate com petition w ithin indus
nounces that it w ill con stru ct by -p rod u cts plant to convert try. ~ ~ A t t o r n e y G eneral Biddle, speaking before K ilgore
rice hulls in to alcohol, glucose, w allboard, lignin, and a filler Senate C om m ittee on W a r M obilization , says A llies should
for p la s t ic s .~ ~ W P B changes aniline control arid reduces take over direction of large G erm an m onopolies after w ar to
sm all order exem ptions from 500 to 450 pounds. learn how th ey operate and then break them up.
1f A u g u s t 24. Charles E , W ilson, executive v ice chairm an
if A u g u s t 30. Office o f R u bber D irector postpones curtail
o f W P B , resigns; his letter to the President charges that he
ing production o f butadiene from alcohol, originally set for
was the v ictim o f unfair attacks and criticism s b y aides o f
Septem ber 1, until O ctober 1 because o f continued demand
Chairm an D on a ld M . N elson. President accepts resignation
for butylenes for the 100-octane aviation gasoline program .
w ith regret. ~ ~ O r o n i t e Chem ical C o. announces plans for
construction o f 31,000,000 plant in R ich m on d, Calif., to m ake If A u g u s t 31. Foreign E con om ic A dm inistration issues
phthalic a n h y d r id e .~ ~ B a r r e tt D ivision , A llied Chem ical and order term inating v eto pow er exercised b y purchasing mis
D y e C orp., plans to con stru ct one, or possibly tw o, naphtha sions from Allied and friendly nations over shipm ents of
lene plants in M id d le W e s t . ~ ~ A i r R edu ction C o., In c., Am erican goods to their hom elands, effective O ctober 1.
W h a t about c a ta ly tic crack in g for the s m a ll
refiner?
W s h ere n o w
T h e fluid catalytic cracking unit shown above
has a charge capacity of 2600 barrels per stream
day It w as developed, designed and licensed by
Universal and is in successful operation in the new
plant of Frontier Refining Co., Cheyenne, W y o m in g
Universal has designed other fluid flow catalytic
cracking units ranging in capacity from 1500 to
16000 barrels per day— several are now in operation
Universal Oil Products Co. |C||n n Petroleum Process Pioneers

Chicago 4, 111., U . S .A . P °r ^ Refiners
The Refiners Institute o f Petroleum Technology
GASOLINE POWERS THE ATTACK — DON’T W ASTE A DROP

HERCULOY
insynthetic rubber processes
P o s s e s s in g the strength of mild steel with the

corrosion resistance of copper, Herculoy fre
quently makes it possible to combine in a single
part both structural and process functions, with
resulting economies in time and cost of erection,
and in maintenance. It is a pleasure to note the
use of Herculoy in speeding up the synthetic rub
ber program.
Herculoy — a copper-silicon alloy — can be
welded readily by gas or electricity, which makes
it especially attractive for the construction of Welding 30" pipe made
from Herculoy Sheet—for
pressure vessels, stills and similar chemical plant use in famous company's
synthetic rubber plant.
apparatus. Heat treatment after working relieves
internal stresses without sacrificing strength,
hardness, or resistance to abrasion. Write
for new Manual, "Revere Copper and
Copper Alloys—Technical Information
foir Product Designers.” Contains data
on Herculoy chemical and physical char
acteristics, as well as on other Revere
products.
TYPICAL APPLICATIONS
of Revere Copper and Copper-Base Alloys:
Manifold, 20" with 12" branches, fabricated from Herculoy Sheet
by Gerstein & Cooper, South Boston, Mass. H e a t E x c h a n g e rs a n d C o n d e n se rs M ix e r s
E v a p o ra to rs S t o r a g e Tanks
Pressure V e sse ls R e a c to r V e sse ls
R e ceivers Fraction atin g Colum ns
C a t a ly s t Tubes a n d the like
W e su p p ly c o p p e r a n d c o p p e r - b a s e a llo y s in: SHEET, STRIP,

P L A T E ,R O D , B A R a n d S H A P E S , T U B E a n d PIPE, W E L D IN G R O D
COPPER AND BRASS INCORPORATED

B ounded b y Paul R ev ere in 1801
Executive Offices: 230 Park Avenue, N ew Y ork 17, N . Y .
Your order for M O N S A N T O D I C Y C L O H E X Y L A M I N E

will be filled immediately, whether you’re in a war
industry or not. Currently, supplies are ample and
indications point to continued availability of this inter
mediate to all industries. W hy not check into the
possible applications of Monsanto Dicyclohexylamine
to your present products, or think of it in connection
with your long-range planning? Samples and prices
sent promptly upon request on your company letter
head. The coupon below is for the convenience of those
wishing a copy of our new booklet on the properties
and uses of Monsanto Dicyclohexylamine. M o n s a n t o
C h e m i c a l C o m p a n y , Organic Chemicals Division,
1700 South Second Street, St. Louis 4, Missouri.
D istrict O ffices: N e w Y o rk , C h icag o, B oston , D e tro it, C har
lotte, B irm ingh am , L os Angeles, San Fran cisco, M on trea l,
T o ro n to .
N— H
M O N S A N T O D IC Y C L O H E X Y L A M IN E M O N S A N T O D IC Y C L O H E X Y L A M IN E
H A S THESE CH A R ACTERISTICS OFFERS FUTURE P O SSIBILIT IES
C o lo r: C lear colorless liquid. W h ile M on sa n to D icy cloh ex y la m in e is a th or
P u r i t y : 99 % approx. ou gh ly p ro v e d interm ediate in various indus
C r y s t a ll iz in g P o i n t : 0.1°C . tries, its possibilities are b y n o m eans exhausted.
B o i l i n g P o i n t : 255.8°C .
Sp e c ific G r a v i t y : 0.9 16— 0.920 a t 1 5 .5 7 1 5 .5 °C .
N e w ap plications, new processes, new products
F l a s h P o i n t : 100°C.
m a y b e revealed b y research con d u cted around
i F i r e P o i n t : (Sustained C om b u stion ): 160°C. D icy cloh ex y la m in e an d its uses. O ur tech nical
service m en w ill b e glad to w ork w ith y o u on
S o lu b ilit y : S ligh tly soluble in w ater. S olu b le in
all com m on orga n ic solvents. such p rojects.
M O N S A N T O D IC Y C LO H E X Y LA M IN E
SERVES IN THESE A P P LIC A T IO N S
S tarting P o in t o r C om p on en t in:
1. M an u fa ctu re o f In secticides
2. C orrosion In hibitors
3. O il A dd itives
4. E m u lsifyin g or D em u lsifyin g A gen ts
5. M an u fa ctu rin g D yestu ffs
6. O rganic Syntheses
These suggested uses
construed as
56 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No. <
F o r RESISTANCE TO HCl, H2S 0 4, A lC l3-

U se H a st ello y C o r r o s io n -R é s is t a n t A llo y s
Agitators
Autoclaves
Evaporators
Fractionating Columns
. . . are typical examples.

H astello y a llo y s can be s u p p lie d in c o m p le t e ly
fa b r ic a te d p r o c e s s in g u n it s . . . by a num ber of
e q u ip m e n t m a n u fa c tu r e r s .
Sheets
Plates
Bars
Pipe
T ubing
Castings
Or, if you build equipment, H a s t e l l o y alloys are

Welding Rods
available in standard mill forms as well as in
castings to meet your specifications.
F or m ore inform ation w rite to us at K okom o, Ind. for the booklet, “ H a STBLLOY C orrosion -R esistant A lloy s,” and price list.
BUY UNITED STATES WAR BONDS AND STAMPS
H a y n e s St e l l it e C o m p a n y
U nit o f U nion C arbide and C arbon C orporation
New York, N . Y . IHM Kokomo, Indiana
C h icago— C levelan d— D etroit— H o u s to n — L os A n geles— San F ra n cis co — Tulsa
HIGH-STRENGTH NICKEL-BASE ALLOYS FOR C O R R O S I O N RESISTANCE

1 he word “ H astelloy” is a registered trade-mark o f H aynes Stellite C om pany.
September, 1944
Highly resistant to a t t a c k . . . P r e v e n t s c o n t a m in a tio n
. . . S p e e d s y o u r p r o c e s s e s . . . Cuts u p k e e p c o s t . . .
r h n i IN G W ATER
PURIFIED
GAS
PUMP
ST EAM
IMPURE
- gas“
exch an g er
Yes, aluminum is available for other-than-war no corrosion products to cause contamination.

purposes, the manpower situation permitting. Because of the superior heat conductivity of
Processing equipment, now being designed, can aluminum, heating and cooling operations are
include it and gain the many advantages offered speeded tip. It’s easier to hold materials at even
by aluminum. Old equipment, already serving temperatures. Process efficiencies arc improved.
far beyond its normal life, can have its original Product quality is better.
efficiency restored with new aluminum parts. Processing equipment, storage tanks, ship
The gas purifying system diagramed above, ping containers, made of Alcoa Aluminum
points up these advantages. There, the ab Alloys before the war, have given an excellent
sorber and reactivator are both exposed to account of themselves. Proof that it pays to
hydrogen sulphide. Vicious as this gas is in the employ aluminum. Our engineers will be glad
presence of other metals, it’s like a lamb when to discuss the availability of aluminum for
it meets up with aluminum. The equipment present use. Write A l u m i n u m C o m p a n y o f
has long life, so upkeep is low. There are A m e r i c a , 2 1 5 4 Gulf Building, Pittsburgh 1 9 , Pa.
A L C O A © KtO . v T. M .
A
INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 36, No. !
HACKNEY EXPERIENCE
INFLUENCES PROFIT CURVES
H a c k n e y e n g i n e e r s c o m b i n e u p - t o - t h e - m in u t e r e A lso , all ra w materials are carefully selected.
se a rc h w it h the k n o w le d g e g a in e d in 4 0 y e a r s o f A n d r ig id in sp ection s are m ade after every
v o lu m e p r o d u c t io n . A s a r e s u lt th e u s e r is a s s u r e d stage o f p ro d u ctio n . A s a result, H ack n ey cy l
o f e c o n o m i c a l h i g h v a l u e — a n d p r o f it c u r v e s a r e in ders are u niform in size, w eigh t, strength
k e p t h ig h . and capacity.
T h e w ar effort is dem an d in g a la rge share o f
T HE e co n o m ie s p r o v id e d by H ack n ey cy lin Pressed Steel T a n k C om p a n y’s p ro d u ctio n fa cil
ders have benefited users fo r years. Pressed ities. H ow ev er, as w ar restriction s are relaxed,
Steel T a n k C om pany design and m anufacturing m o re material w ill be available fo r civilian
facilities have m ade p o ss ib le ligh tw eigh t cy lin needs. T h en , H ackn ey p ro d u ct develop m en t
ders; yet they p ro v id e the strength sufficient to w o r k and volu m e m anufacturing w ill be ben e
withstand the abuse o f m od ern sh ip p in g. fiting every co n ce rn that is interested in
O n e reason w hy H ackney cylinders help keep m ak in g im portant savings, sp e e d in g up
profit curves h igh is the heat-treating o f the shipm ents and assuring m axim um
finished p rod u ct by sp ecia l advanced m ethods. p ro te ctio n fo r their products.
Pressed Steel Tank Company

M ANUFACTURERS OF HACKNEY PRODUCTS
G E N E R A L O F F IC E S A N D F A C T O R Y • 1451 S O U T H 66th S T R E E T
Milwaukee 14, Wisconsin
CO N T A IN E R S FOR G A S E S , LIQUIDS A N D SOLIDS

THERE ARE ,
4 M ODELS
WHICH MEETALL MANNER
OF INSTALLATION
, CONDITIONS
For h eating and agitating highly cor

rosive solutions, consider th e new je ts
m ade in D U R IM E T .
The four m odels, w hich m eet practi
M * ! t cally an yîcon dition , are:
Suspended Model 441— for wall or corner in
stallations.
Floor Model 442— for wall or corner installa
tions.
Suspended Model 443— for center of tank in
stallations.
Thru-the-Wall Model 444— for steel, steel-lined,
wood or concrete tanks.
Note how little space is required in the tank
for these jets. Rugged in design and b uilt of
D U R IM E T , they will give long service under
the m ost difficult conditions.
The M odel 441 jets are em ployed in

tanks of greater depth th an 31/2 ft. to
4 ft. The M odel 442 floor-based jets
are best adapted to shallower tanks.
Jets of these sam e designs are also
available in D U R IR O N (Models 421, 422,
423 and 424).
If you are heating acid solutions, such as

pickling solutions or wash liquors, or if
you are dissolving chemicals or digesting
ores, separating sludge acids, etc., you
'■‘"I* will be interested in these new jets.
Allow us to quote on your application.
'/ L L O r S * E Q U IP M E N T
THE DURIRON COMPANY, Inc. DAYTON, OHIO

Why Should YOU
Furnish the Guinea Pigs?
• T ryin g out pum ps— experimenting with them —
is risky business.
You just d on ’t d o it! A nd you shouldn’t d o it.

That’ s our business.
W e d o the experimenting. W e d o the research. W e

m ake sure that F airbanks-M orse Pum ps m eet all
your requirements and m ore—lo n g before they are
installed.
BUY WAR BONDS
FAIRBANKS - MORSE
DIESEL ENGINES WATER SYSTEMS
PUMPS SCALES
MOTORS STOKERS
GENERATORS FARM EQUIPMENT
RAILROAO EQUIPMENT
September, 1944
H iR I?S
LAPP PORCELAIN
CAN IM P R O V E PURITY,
INCREASE SECURITY
AND REDUCE
COSTS
K - - j
• P o p -typ e va lve . A ll- p o r c e la in

co n stru ctio n a v o id s c o n ta m in a -
tton j r o m c o n ta c t o/ co rro siv e
g a se s o r l i q u id s w it h m e ta ls o r
gaskets. B lo w -o f f ad ju stab le .
• S t a n d a r d a t ta c h m e n t to t a n k outlets.
Poppet lin e s u p w it h t a n k bottom, e lim i
n a t in g p o c k e t a t outlet.
PORCELAIN PIPE
• If your processing system in
cludes m odem glass-lined, • S t r o n g u n ifo rm -se c tio n
pip e, g l a z e d i n s id e a n d
rubber-lined, lead-lined or tile . . |
out. P r e c is io n g r i n d i n g o f
processing tanks, you have in p ip e e n d s p e r m it s re c o m
stalled this equipment because m e n d e d u se w it h t h in
h a r d gaskets, o r w h e re
of the corrosive action of necessary, w it h no g a sk e ts
chemicals handled, or because a t a ll. Sm ooth, g l a z e d
surface, in s id e a n d out,
of the need for purity. If you p e r m it s e asy cle a n in g t
are experiencing trouble with valves p ro m o te s p u rity.
or piping, or if there is unexplained

trouble which might be attributable
to valves, pipe or gaskets, you may
need the special properties of Lapp
Chemical Porcelain. This rugged,
dense, non-porous body defies attack
from any acid (except HP) j is smooth,
easily cleaned, and gives long, low-;
, cost service. Lajpp insulator Co.. Inc.,
: . Chemical Porcelain D iv., LeRoy, N .Y.
■ B K r a B & T ' mm m m m m m m
:i
.&
Vol. 36, N o.!
"Secret ingredient"
makes synthetic rubber
better and faster!
Bu + Na + S + Taylor Accuracy
[ = better synthetic rubber... on schedule
"D E C E N T L Y we showed you Taylor Instruments

•A-Y. at work in the huge butadiene and copolymer
plants at Port Neches, Texas. Here are typical photo
graphs from the Los Angeles and Velasco, Texas,
plants of the Dow Chemical Company, principal
supplier of styrene for the synthetic rubber program.
We're proud of the experience and knowhow that
produced the Taylor Instruments capable of perform
ing this exacting task.
But we re proudest of the people here in our plant
who took on the impossible job of delivering the in
struments on time— a job that inspired the W PB to
tell us, “ The age of miracles has T a y lo r T h e r m o s p e e d separable

not yet passed!” If you could visit well constructions.
these Dow styrene plants, you TAYLOR MERCURY TYPE FLOW
would see: RECORDERS
TAYLOR RA T E -O F -F L O W PNEU TAYLOR FULSCOPE RECORDING
MATIC TRANSMISSION SYSTEMS PRESSURE CONTROLLERS with
involving the following . . . Automatic Reset
Taylor A n e ro id Flow Recorders TAYLOR MERCURY-IN-GLASS IN
and C o n tro lle rs (hundreds of DUSTRIAL THERMOMETERS, with
them). The controllers with Binoc Tubing
Automatic R eset; some with In short, you’ d see better than we
Taylor Pneumatic Set. can tell you how Taylor can do a
TAYLOR TEMPERATURE PNEU job for you! Taylor Instrument
MATIC TRANSMISSION SYSTEMS Companies, Rochester, N . Y . and
with Toronto, Canada. Instruments for
Taylor Fulscope Recording T em indicating, recording, and controll
p erature C o n t r o lle r s (scores of ing temperature, pressure, humid
them), with ity, flow and liquid level.
Automatic R e se t, and
Pre-Act (corrects according to rate BUY MORE BONDS
of control point deviation). The
. . . A N D KEEP THEM!
mercury actuated temperature
systems having
Taylor A c cu ra tu s Tubing (com
pensated for ambient tempera
tures), and
IN HOME AND INDUSTRY

Vol. 36, No. 9
• « i n larQB refinery
A *ect,on p etro leu m
w here com p .«-K te d
r / h .e r ®_C. u ^ are se g re g a te d
“ m p° -
S r t . of syn th o '“ p u n t“ « ’ -
E Q U IP M E N T | H E A T -T M A T E D :
■„ „1 P r o c e s s i n g E q u ip m e n t d0? l T u ^ c te d to the expi

fin a l c h e c k , is s I , n , M$ w a y , no
' e rY “ v e s s e ls ^ l a n b . t o w e r s - i o r h .g h - i „ g e y e ol the ,h e m etal or ffi
r| e r o , u r e . ' h i g h - ^ a^ tr^ f l a w i „ the structur u n d e (e cte d .
re m ica l, * Y ^ eê ™ re , u||’Y m a ch in e d

Th e
so e,rletctio
« n&ot its p ro d u cts. o f ■,n su rin 3 the
■then a n n e a le d in a sp e-
THE BABCOCK & W IL C O X CO. 85 LIBERTY STREET. N E W YORK 6, N . Y.

September, 1944
‘Wlities upon the
The intm * Pr09ra m ‘
je m o n w — . B a b c o c k & W il
. \ i-nne a a s o l i n e , s y n e x p e n e n c e in - m ent, h a s taken a
C h e m i c a l s , b>9 ' ° t Q oVVter p r o d u c t s
turing p ro c e _ |v ;„ g m uch ol it-
th e tic r u b b e r , a n ss{u\ c o n d u c t o
le a d in g p a rt »
J ^ n u t a c t u r in g
n e c e ssa ry lor t e s ¿rem endous The B&W lacil.tiesj^ ^ towe
* e ^
burden on
Tthe m a n u fafctu rers
b e number
ol p roc-
o
h ig h - p r e s s e r e „„su rp assed .
t a n k s ,° n d ,stee .¡ „ a t e c h n iq u e , a u g -
e ssV ng ^ U' prTienSorea p fa ^ s built at
G o v e r n m e n t-s p o th e couptry The T e a t treatm ent a n d X - R a V .
S t r a , e 9 ''Cn ° l C e q5 - n t- - Dn d . o r e ment a le s at sa te , efficien t, an d

h a v e c a lle d to q o( neW and a re 9 ^ ° " p ro d ucts.
e c o n o m ic a l p ro u 5^ n
e q u i p m e n t — 1m “ " 0
compHcote^ design-
C H EC K ®
precision, versatility and

In X-Ray Diffraction, Industry has at hand
sp ee d in. chem ical and
an indispensable means for quick, accu
physical determination by rate, chemical and physical determinations,
X -R A Y DIFFRACTION b y gra p h ic d e m on stration of the actual
microscopic structure of materials.
The Picker Diffraction X-R ay Unit offers to

the scientific and engineering professions
an apparatus of the utmost precision, con
venience a n d versatility. Four types of
Diffraction Cam eras are shown, mounted
w ays with perfect reference edges to
insure consistently accurate alignment. The
Unit is also readily adaptable for use with
any special cameras such as the Weissen-
berg for single crystal analysis. Transformer
a n d tube are w a te r cooled, perm itting
continuous operation.
X -R A Y
DIFFRACTION A PPA R A TU S
T w o p o r t a l l o w filtra tio n b e r y lliu m w i n d o w
C o n t in u o u s o p e r a t io n — tu b e a n d t r a n s f o r m e r
w a t e r c o o le d
O p e r a t io n f u lly a u t o m a t l c - c o n t r o l 5 K V to 5 0 K V
in 2 0 s t e p s
X - R a y t u b e s w ith t a r g e t s o f d iffe re n t e le m e n t s
a r e a v a i l a b l e a n d in t e r c h a n g e a b le
C o n c e a l e d In s u l a t e d w a t e r lin e s to m in im iz e
c o n d e n s a t io n
C o n c e a l e d s t o r a g e c o m p a r t m e n t fo r X - R a y tu b e s,
c a m e r a s , a n d a c c e s s o r ie s
P ic k e r X - R a y C o rp o ra tio n m a n u fa c tu re s in d u stria l
X - R a y e q u ip m e n t c o v e rin g e v e ry p h a s e o f a p p li
ca tio n . . . ra d io g r a p h y , flu o ro sc o p y , diffraction.
PICKER X-RAY corporatioi

: 300 F ou rth A v e n u e , N e w Y o r k 1 0 , N . Y
WAITE MFG. DIVISION • CLEVELAND 1 2 , OHIC
BRANCHES AND SERVICE DEPOTS IN PRINCIPAL CITIES OF U.S.A. AND CANADA

'I
The Laue C a m e ra is u se d fo r the o b s e rv a tio n

of preferred o rie n ta tio n o f g r a in s o r fiber
textures a n d the stu d y o f' cold w o r k a n d
recrystallization.
T y p ic a l L a u e C a m e r a D if f r a c t io n
Pattern; sho w s p referred orientation.
The Back Reflection C a m e ra is u se d fo r

precise m e a su re m e n t o f lattice co nstants,
measurement o f su rfa c e stresses, o b s e r v a
tion of d isto rtio n in c ry sta ls a n d the d eter
mination o f c o m p o sitio n o f s o lid so lu tio n
alloys. S a m p le a n d film m a y be rotated
by motor.
T y p ic a l B a c k R e f le c t io n C a m e r a
Diffraction Pattern
For routine id e n tific a tion o f su b sta n c e s o r

unknowns the 7 0 m m . P o w d e r C a m e ra is
used. S p e c im e n can be rotated b y m otor.
typical 7 0 mm. P o w d e r C a m e r a
Diffraction Pattern.
7 0 m m . P O W D E R C AM ERA
for more p re cise m e a su re m e n t o f lattice
constants a n d the d e te rm in a tio n o f a llo y
constitution d ia g ra m s , a n d fo r stu d ie s w h e re
•fie high est re so lu tio n is re q u ire d , the 2 0 0
mm. P o w d e r C a m e ra is used. S p e c im e n can
oo oscillated o r rotated b y m otor.
yp'ca/ 2 0 0 mm. P o w d e r C a m e ra Diffraction Pattern,

ole the greate r resolution of lines a s co m p a re d
/he 70 mm. pattern of the some ar e a above.
YOUR l o c a l p ic k e r e n g in e e r is a s n e a r
as YOUR PHONE - O R U S E T H IS C O U P O N
This is the Microscope
You are Waiting for
These are the lenses that make up the heart
of a microscope. Assembled and mounted on
the stand they become a B&L Microscope . . .
the microscope you have been planning to purchase when
restrictions are lifted.
the B&L low temperature lens cement and other new
It will represent the best and the latest in the applica
techniques. All these things, and more, will be combined
tion of optical science to microscopy. From the initial
to give you the finest optical equipment that can be built
mathematical calculations to the final inspection each
B&L Microscope will always measure up to this standard
and each is the product of the experience that has gone
before.
BA U SC H & LO M B
There are the facilities of America’s first and finest op OPTICAL CO., ROCHESTER, N. Y. jf e k
tical glass plant, new methods of manufacture such as the
diamond milling of optical parts, new materials such as ESTADUSHIO ;
18 5 3
Makers o f Optical Glass and a Complete Line o f Optical Instruments fo r Military Use, Education, Research, Industry and Eyesight Correction and Cornered* 1
C L A R K 300 H .P . Angles, parti o f refrigeration system a» alkylate reactors, have
capacity o r 150,000 cu. ft. o f gas per hour, at 14.7 lbs.
0 C O N T IN U O U S
operation
Production A B O V E
R A T E D C A P A C IT Y
0 Q U A L I T Y of I
p roduction f
One o f the Dresser Industries

INDUSTRIAL AND ENGIN
WeldELLS alone
combine these features:
• S e a m le s s — gre ater stre ngth
+ As far as ive know, no one else ever forged a fitting like and uniform ity.
the one pictured above— or any welding elbow or return • Tangents.— keep weld aw ay from
bend having a wall thickness so heavy in proportion to the zone o f highest stress— simplify
size of the fitting. lin in g up.
• Precision quarter-marked ends
Such fittings simply can’t be made by the processes ordi
— sim p lify layout and help insure
narily employed in the manufacture of welding fittings. But accuracy.
a special method, developed and patented by Taylor Borge, • Selective reinforcement — pro
makes them "duck soup” for us . . . . An excellent example, vides unifo rm strength.
this, of the problems that are so often put up to us by engi Permanent and complete identi
neers who feel we are a little better qualified to solve them fication m arking— saves time and
than any other organization. eliminates errors in shop and field. -:=
• W all thickness never less than
specification minimum— assures full
strength and lo ng life.
• It isn’t likely that you will ever run into a condition calling for a
• Machine tool beveled ends— pro
fitting of this kind. W e simply cite such problems to show you— far
vides best welding surface and ac
better than we can tell you— how much specialized knowledge has gone curate bevel and land.
into our standard line of WeldELLS and other Taylor Forge welding • The most complete line of Weld
fittings. in g F it tin g s and F o rg e d Steel
Flanges In the W orld — insures
Do not underestimate the importance of this broad experience. In complete service and undivided re
WeldELLS the "know-how” acquired in solving every conceivable kind sponsibility.
of forging problem is reflected in features that are combined in no
other make of welding fittings!
A good example of this is the selective reinforcement— extra metal
properly distributed where stresses are greatest. Another is the tangents;
another, the final truing operation which produces those extremely
accurate dimensions. These, and the other features listed opposite, give
you a better job of pipe welding at lower cost. Check over the list
and we believe you will agree that
WeldELLS l l t l A ’C C V C i ' l | t l t i l l
TAYLOR FORGE & PIPE WORKS, G e n e r a lO f fi c e s & W o r k s :C h i c a g o , P . O . B o x 4 8 5

N ew Y o r k O ffic e : 5 0 C h u r ch Street • P h ila d e lp h ia O ffic e : B r o a d Street S ta tio n B ld g .
H o w g o o d a job a re y o u m a k in g o f y o u r m otor a p p lic a t io n s ?
is e ach job re a lly " a s w e e t h e a rt " or d o th e y lo o k m o re like the
results of a s h o tg u n w e d d in g ?
It's e a s ie r to m a k e each job a sw e e th e a rt . . . if y o u ta ke a d
v a n t a g e o f the w id e fle x ib ility o f M a s t e r m o to r d rive s. Fo r in
stance, io o k at the a p p lic a t io n s h o w n b e lo w . B y in c o rp o ra tin g
a m otor, a n eiectric b ra k e , a m e c h a n ic a l v a r ia b le sp e e d unit a n d
a g e a r re d u ctio n . . . a ll d e s ig n e d a n d built into o n e co m p a ct
in té g ra i u n it . . . this M a s t e r m o to r p ro v id e s v a r ia b le sp e e d at
e x a c tly the righ t ra n g e a n d in a d d itio n c a n b e q u ic k ly a n d a c
cu rate ly s to p p e d a t e x a c tly the p o in t d e sire d . It m o u n ts n e a tly on
the m a ch in e , s a v e s sp a ce , s a v e s m o n e y a n d g re a tly im p ro v e s the
output, safety, a p p e a r a n c e a n d co n v e n ie n c e o f the e q u ip m e n t.
P r o b a b ly y o u w ill not n e e d e x a c tly the sa m e c o m b in a tio n o f
m o to r fe a tu re s illu stra te d b e lo w , but the M a s t e r line in c lu d e s
m o to rs for e v e ry current sp e cifica tio n , e v e ry typ e o f en closure,
a n d e v e ry typ e o f m o u n tin g a rra n g e m e n t . . . in fact, is the m o st
fle xib le , the m o st v e rsa tile line o f m o to r d riv e s in the w o rld .
In v e st ig a te M a s t e r 's u n u s u a l a b ility to se rv e y o u e c o n o m ic a lly
w ith m o to rs that re a lly fit the job in s iz e s from 1 / 1 0 to 1 0 0 H P.
THE MASTER ELECTRIC C O M P A N Y « DAYTON 1, O H IO

ABSORBS VIBRATION...PROVIDES MOVEMENT
in the four corners of the earth
On every industrial and transportation frontier, Barco Flexible
Joints stand guard over fluid-conveying pipes . . . helping to
minimize the destructive action of vibration and shock. Thirty
years of continuous use have only strengthened engineering endorse
ment of the Barco principle and design. Barco Manufacturing Co.,
Not Inc., 1823 Winnemac Ave., Chicago 40, 111.
In Canada: The H olden Co., Lid., Montreal, C anad a
FLEXIBLE
BARCO
swivel
com-
swivel
JOINTS with
and
move-
ever,y
T H E F R E E E N T E R P R I S E S Y S T E M IS TH E SAL VAT I ON OF AMERI CAN BUSI NESS
Not just a swivel jo in t. .,

hut a combination o f a swivel and ball joint
"M O V E IN DIRECTION'
with rotary motion and responsive movement through every angle.
Septem ber, 1944 INDUSTRIAL AND ENGINEERING CHEMISTRY 73
pmauiiBMPniMMMiinn
F iv e l iie w H o o k e r P r o d u c t s Y o u
Among Hooker's thirteen new chemicals announced at the Chemical Exposition, highly chlorinated propane derivative
December 1943, some have shown especially promising futures. Many chemists all and therefore nearly saturated. It is
over the country have been trying out these new products. Their work is of course insoluble in water and soluble in alcohol,
confidential, but we feel that you may find, as they have, chemicals to assist in doing ether, and most chlorinated solvents. It
an old job better or to do an entirely new job. is resistant to oleum, mixed acids, fuming
Whether your research or production problems are concerned with present or post nitric acid and hydrogen fluoride.
war needs, we invite your inquiries about these new products. The Hooker Labora Chlorpropane Wax 130 has been found
tories have prepared additional data and will be glad to send you such information to be compatible with the following types
when requested on your letter head. of plastics:
Methacrylate Interpolymcr Rosin
plastics. The nature of the resultant ma Modified Phenolic Resin Gilsonite
. H E X A C H L O R B U T A D IE N E terial is indicated as soft (S), rubbery (It), Methyl Abietate Resin Ester Gum
Phenol Formaldehyde Polystyrene
Cl Cl Cl Cl tough (T), and brittle (B): Chlorjnated Piccolyte Alkyd Resin
1 1 .1 Modified Phenolic Resin (S) Rosin (S) Chlorinated Rubber Piceoumaron
c C = c Phenol Formaldehyde (S) Ester Gum (S) Urea Formaldehyde Natural Rubber
1 ■1 Chlorinated Piccol.vte (S) Polystyrene (It) Polyvinyl Chloride „ Terpene
Poly __ Resin
Cl Cl Benzyl Cellulose (T) (B) Piceoumaron (S) Chlorinated Diphenyl Resin
D escription : Hexaehlorbutadiene is a Chlorinated Rubber (R) Natural Asphalt (S)
Poly T erpenc Resin (S) Poly vinyl Chloride (R) S uggested Uses: Plasticizer; dielectric
clear, colorless liquid with a mild char Methacrylate Interpolymer (T) (R) wax; chemical resistant lubricant; or
acteristic odor. It is insoluble in water ganic systheses to make synthetic rubber
and soluble in alcohol, ether and most S uggested Uses: Hydraulic fluid, trans
former, and dielectric fluids, anti-freeze and plastics; ingredient of pyrotechnic
chlorinated solvents. It is highly stable compositions.
and is resistant to hydrolysis by water or additive for hydraulic, dielectric and heat
transfer fluids, solvent for fats, oils, Physical Data:
mild alkalies. Molecular Wt. (aver.) 811
Hexaehlorbutadiene has been found to be waxes, coal tar dyes, asphalts, solvent,
Melting Range °C 110 to 135
compatible with the following types of diluent, and plasticizer for protective Boiling Range °C 210 to 270
coating and insulating compositions; ex Analysis* (typical)
plastics. In most cases the resultant ma Octachlorpropane 85%
terial tends to be soft (S) or rubbery (It). tractant for phenols, etc. from liquids
Heptachlorpropane 15%
In a few cases, tough (T) or even brittle such as waste waters; ingredient of insec
(B) materials are formed. ticidal compositions, paint and varnish C H L O R P A R A F F IN R ES IN 70
removers, paints, solvents and plastic
Clilorinntvd Piccolyte {S) (II) Rosin (S)
Modified Phenolic Resin (B) Polystyrene (R) compositions. C24H20C121 (aver.)
Polyvinyl Chloride (R) (T) Aikyd Resin (S) P hysical D ata:
Phenol Formaldehyde (S) Benzyl Cellulose (S) D e s c r ip t io n : Hooker Chlorparaffin
Molecular Weight (pure Trichlor Resin 70 is a highly chlorinated paraffin.
Chlorinated Rubber (B) Natural Asphalt (T) cumene) 223.5
Poly Terpene Resin (S) Vinyl Polymer (S) (R) Freezing Range, °C -30 to —15 It is a brittle resin having an amber color
S uggested Uses: Solvent for natural and Boiling Range. °C 245 to 265 in lump form. It may be crushed readily
synthetic rubber, and other polymeric Refractive Index, n20/D 1.535 to 1.560 to a white powder which exhibits little
Snnnifip
Specific Gravity l15.5°/15.5°C
i S °/1 S 5*1” 1,20. t o ] 32
substances; high boiling, non-flammable * tendency to agglomerate on standing. It
solvent for degreasing and extraction C H L 0 R P R 0 P A N E L IQ U ID 170 is more stable than chlorinated paraffins
processes; dielectric fluid for switches and containing lesser amounts of chlorine. It
transformers; hydraulic fluid; heat trans C 3 H i . 5 C16 . 5 *(aver.) is soluble in most aromatic solvents, but
fer medium; chemical reagent in manu D escription : Chlorpropane Liquid 170 is difficult of solution or insoluble in
facture of synthetic rubber; plasticizer. is a clear, colorless liquid with a char mineral spirits. It begins to soften meas
Physical Data: acteristic odor, is insoluble in water and urably at about 90°C and is still very
Molecular Weight 261 soluble in alcohol, ether and most chlor viscous at 120°C.
Melting Range, °C -19 to - 2 2 Chlorparaffin Itesin 70 has been found
Boiling Range, °C 210 to 220
inated solvents. It becomes quite viscous
Refractive Index, n20/D 1.551 to 1.554 at temperatures below-o0°C. It is re to be compatible with the following types
Specific Gravity 15.50/15.56C 1.65 to 1.70 sistant to oleum, mixed acids, fuming of plastics:
nitric acid and hydrogen fluoride. Chlorinated Diphenyl Resin Rosin
T R IC H L O R C U M E N E Chlorpropane Liquid 170 has been found
Modified Phenolic Resin
Chlorpropane Wax 130
Gilsonite
Ester Gum
(CH3)2— CH— C6H2C13 to be compatible with the following types Methyl Abietate Resin Polystyrene
of plastics: Phenol Formaldehyde Alkyd Resin
D escription : Trichlorcumene (Isopropyl Chlorjnated Piccolyte Piceoumaron
Methacrylate Interpolymer Polystyrene Chlorinated Rubber Vinyl Polymer
Trichlorbenzene) is a colorless liquid with Phenol formaldehyde Piceoumaron Urea Formaldehyde Benzyl Cellulose
a mild aromatic odor. It is a mixture of Methyl Methacrylate Vinyl Polymer Poly Terpene Resin Polyvinyl Chloride
Urea Formaldehyde Natural Rubber Methacrylate Intcrpolymer
Trichlorcumene isomers. It is insoluble Poly Terpene Resin Polyvinyl Chloride
in water and soluble in alcohol, ether, S uggested U ses: Plasticizer, rubber S uggested Uses: Ingredient of fire,
and most common solvents. It is highly water and mildew retardant paints, fabrics
and plastic modifier; sealing liquid; paint
stable, being resistant to oxidation and softener, insecticide. and other protective coatings. Its stability
hydrolysis. and chemical inertness relative to most
Trichlorcumene has been found to be- Physical Data: substances make it of particular interest
compatible with the following types of Molecular Weight (aver.) 268 for the impregnation of combustible ma
Boiling Range °C 100 to 260
Refractive Index n20/D 1.520 to 1.523 terials to give fire retardant action. In
Specific Gravity ]5.5°/I5.5°C 1.70 to 1.75 protective coatings and paints, it docs
Analysis,* (typical) not affect the rate of drying.
Pentachlorpropanc Oto 5%
Hcxachlorpropane 40 to 50% Physical Data:
Heptachlorpropane 40 to 50% Molecular Weight (aver.) 10G3
Octachlorpropane 0 to 5% Analysis (typical) Weight %
Chlorine 68 to 72
C H L O R P R O P A N E W AX 130
H ooker
Free HC1 (max.) 0.05
Heavy metals (max.) 0.01
^ 3 ^ ( M 5 ^ 7 . 8 5 * ( a v e r*) Softening Range, °C 90 to 100
Flash Point None
D escription: Chlorpropane Wax 130 is Fire Point None
a tough white crystalline wax, possessing Specific Gravity, 15.5°/15.5°C, (ruin.) 1.600
Acid Number (mg. KOH/gm.) (max.) 0.50
a mild camphor-like odor. It is a very Stability, mg. H Cl/25 gms. 60 to 70
C H E M I c a ls H O O K E R E L E C T R O C H E M I C A L COMPANY
9 Forty-Seventh Street . Niagara Falls, N. Y .
NEW Y O R K , N. Y. » TACO M A, W ASH . • W IL M IN G T O N , C A L IF .
For every C O R R O S IO N problem in flow control,

there’ll always be a P O W E L L Valve to lick it I
G
- . . . •• «> -
P o w e ll C o r r o s io n -R e s is ta n t V a lv e s h a v e m o t j u s t b e e n “ b o r n ” —
th e y ’ re a p ro d u c t o f e v o lu tio n — th e r e s u lt o f n e a r ly a c e n tu ry o f
p a tie n t s tu d y , re s e a rc h "and e x p e r im e n ta tio n b y P o w e ll’ s s ta ff o f
E n g in e e r s a n d M e ta llu r g is ts . T h e r e fo r e y o u c a n b e s u re T O D A Y
th a t T O M O R R O W y o u ’ ll a lw a y s b e a b le to c o n tro l e ffic ie n tly th e
flo w o f a n y a n d a ll c o rro s iv e m e d ia . F o r P o w e ll h a s e v o lv e d a
co m p le te lin e o f v a lv e s o f e v e r y ty p e — G lo b e s , G a t e s , C h e c k s ,
Y s , e tc .— in a w id e v a r ie ty o f c o rro s io n -re s is ta n t m a te r ia ls . T h e s e
in c lu d e su ch p u re m e ta ls a n d spe cial a llo y s as A l u m in u m , A l u m i
n u m B r o n z e , E v e r d u r , H a r d L e a d , H e r c u l o y , In c o n e l, M o n e l
M e t a l , N i c k e l ( P u r e ) , N i c k e l I r o n , N i - r e s i s t , S ilv e r ( P u r e ) ,
H a s t e llo y A llo y s “ A ” , “ B ” , “ C ” a n d “ D ” , D u r im e t “ T ” a n d
20 ” , I l i u m , M is c o “ C ” , s ta in le s s I r o n A llo y s a n d m a n y m o re .
O t h e r s w ill b e a d d e d as n e w r e q u ire m e n ts a ris e .
The Wm. Powell Company

Dependable Valves Since 1846
Fig. 2453— Large size Stainless Steel Cincinnati 22, Ohio
Gate Valve for 125 pounds W . P. Has
flanged ends, outside screw rising stem
threaded through bronze bushing in
upper yoke arms, bolted flanged bonnet
and taper
wedge solid
d isc. Sizes,
2 to 30",
inclusive.
Write Dept. A,
The W m. Powell Co.,
for our new booklet—
“ Powell Valves for Cor
ro sion R e s is ta n c e .”
F ig . 2559— Large size Stain

less Steel Swing Check Valve
F ig . 2309— Stainless for 150 pounds W . P. Has
Steel Flush Bottom flanged ends and bolted flang
Tank Valve. For draining ed cap. Disc is hung on a 5
metal tanks and auto degree angle and has suffi
claves. In this type, the F ig . 1979— Stainless Steel Globe Valve fo
cient lift t o ' provide full 150 lbs. W . P. Has flanged ends, outsid
disc rises into the tank straightway, unobstructed
when the valve is open. screw rising stem threaded through bronz
flow area through valve body. yoke bushing, bolted flanged bonnet an(
Sizes, M " to 8 ", inclusive. Sizes, 2" to 12", inclusive. plug type disc. Sizes, M " to 3 ", inclusive
Septem ber, 1944 INDUSTRIAL AND ENGINEERING CHEMISTRY
I: fIi
THE MAGIC WORD IN MIXING
Alm ost like m agic, ingredients mixed in Baker
Perkins Dispersion m ixers are thoroughly and
uniform ly dispersed and diffused with one
another. B aker Perkins Dispersion M ixers
give a homogeneous mix in the m anufac
ture of rubber, paint, enam el, ink or
plastics. They are w id ely ,used to effi
ciently disperse "in so lu bles" of small
and uniform particle size in a h eavy
viscous or plastic vehicle. B-P Dis
persion M ixers are being used in
an ever increasing number of
chem ical processes to replace
grinding mills, roll mills and
b all mills
Above: A 150 gallon w orking capacity H eavy Duty

Dispersion M ixe r equipped with air-operated com
pression cover; suitable for rubber, plastic, paint
an d enamel dispersers. This unît is rapidly replac
ing roll mills for the dispersion of pigments. 100-
150 H.P. drive. Right: A size 15 Type V Baker
Perkins Disperser in tilted position. Note the pow
erful dispersion blades and compact arrangem ent
of motor and drive. For further facts about Baker
Perkins Dispersion units and other efficient equip
ment for Chemical processing, write Baker Per
kins Inc., Chemical M achinery Division, Saginaw ,
Michigan.
B A K E R P E R K I N S e q u i p m e n t GI VES MORE YEARŚ OF E F F I C I E N T PRODUCTI ON

T U R B O -M IX E R S,
A G IT A T O R S, AERATORS
ONE
ORGANIZATION
ONE TA N K S AND PRESSURE VESSELS
RESPONSIBILITY
C O N K E Y VACUUM AND
STANDARD OF PERFORMANCE PRESSURE FILTERS
Now you can secure virtually every kind o f pro

cess equipment you need, from one completely
dependable organization---General American
Process Equipment. This relatively new- combi
nation o f well-known equipment manufacturers
means far more to you than the convenience o f LOUISVILLE DRYERS, COOLERS, KILNS
one source, one contract, one billing. It means
that each piece o f equipment you buy will be
designed and built to integrate most efficiently
with the rest—that your process will be dealt
with as a whole rather than as a series o f piece
meal problems—that you -w'ill be assured o f the
utmost economics in both the purchase and
operation o f your new equipment.
Write for an account o f our experience, our facili

ties, our responsibility. Address:
GENERAL AMERICAN
CONKEY VACUUM EVAPORATORS
PROCESS EQUIPMENT
A Division of General American Transportation Corporation
A s s o c ia t e d w ith P la t e a n d W e ld in g D iv is io n
Executive Offices: :
433 BAXTER AVE., LOUISVILLE 4, KY,

N e w Y o r k O f f ic e : 4 2 0 L E X IN G T O N A V E N U E
Septem ber, 1944 INDUSTBIAL AND ENGINEERING CHEMISTRY 77
New C h e m ical Contours for In d u s t r ie s

In Oronite Research Laboratories, chem ical''contours” roles in the progress o f m odern industry. Others will
are being changed. Out o f Petroleum, Oronite Research becom e available on a commercial scale only when re
is leading the way toward the creation o f important in duction o f war-time dem ands releases equipment and
dustrial chemicals. Out o f Petroleum, new materials are materials for their manufacture. Meanwhile, O r o n ite ’ s
emerging on the hori research staff continues to develop new ways in which
These Oronlte 'Products are N ow
in Commercial Production: zon o f t o m o r r o w ... in du stry may utilize these ch em ical p r o g e n ie s o f
O F O N J T E D K Y IN G O IL S chemical forms which petroleum.
O RO PLA5T*
I Rubber Plostlcizers and Dispersing Agents) will give the industrial Y ou r nearest Oronite Chemical Company office will,
W A T E R P R O O F IN G C O M P O U N D
PO LYBUTENE5
chemist newtools with upon request, furnish a catalog o f our present products.
C A L O D O R A N T *. ' which to accomplish Some o f these are today available only in limited quanti
(Natural Gas Odorant)
C R E S Y L IC A C ID S his objectives. ties, but in due course this situation w ill change. Tom or
N A P H T E N IC A C ID S
A L IP H A T IC A C ID S
Some o f these Oronite row , new O ro n ite Chemicals w ill be at your disposal
S O D IU M S U L F O N A T E S p ro d u cts are already and some may prove o f value in shaping chemical con
W E T T IN G A G E N T S
p l a y i n g im p o r ta n t tours in your particular industry.
O R O N IT E C H E M IC A L COMPANY
' R U S S B U IL D IN G , S A N F R A N C I S C O 4 , C A L IF .
3 0 R O C K E F E L L E R P L A Z A , N E W Y O R K 2 0 , N. Y.
W H IT E -H E N R Y -S T E W A R T BLDG.,SEATTLE 1, W A SH .
S T A N D A R D O l l BLDG., LOS A N G E L E S 15, CALIF. * Reg. U. S. Pat. Off.
LEUIIS VERTICRI REID PURIPS
can be depended on for safe, eco
nomical, trouble-free delivery of sul
furic acid in any concentration.
Built by men who understand the

difficulties of pumping H2SO 4,Lewis
Vertical Acid Pumps a llow none of
the liquid p u m p ed to contact the
shaft p acking . . . thus the hazard
of stuffing box leakage is entirely
avoided.
V A L W A Y S PRIM ED
NO STUFFING
BOX LEAKAGE
V MAXIMUM SAFETY
6
CHUS. S. L UIIS & CO., II C. 1 ST. LOUIS, mo
k u u m
T rad e M ark
t : PIPES, TUBES, FITTINGS
W it h s t a n d A l l C o r r o s iv e C h e m ic a ls
Can Be A s s e m b le d
P ip e , tu b e s, f it t in g s , made o f "N a tio n a l” carbon
and graphite, or o f "K a rba te” m aterials. (carbon or W rite National Carbon
Company, Inc., Cleve
graphite made im pervious to fluid seepage under land 1, O hio, Dept.
pressure) are available in standard sizes for the 25-1, for Catalog Sec
tio n M -8 8 0 2 —a n ew
chemical and process industries w herever corrosion bulletin on "Karbate”
Heat Exchange Equip
problems exist.
ment.
These products, plus "K arbate” centrifugal pumps,
Globe and Saunders valves, and heat exchangers, are
fabricated into com plete conveying and heat-transfer
systems o f practically any design.
This is because only "K a rba te” materials have all
these properties: H igh resistance to the action o f
most acids . . . notably hydrofluoric acid . . . and o f
alkalies'and other corrosive chem icals; high mechan * BU Y UNI TED STATES WAR B O N D S *•
ical strength, light w eight, resistance to thermal
shock, g o o d m achineability, and therm al conductiv
N A TIO N AL CARBON COMPANY, INC.
ity exceeding that o f many metals. Unit of Union Carbide and Carboti Corporation
General Offices: 30 East 42nd Street, N e w Y ork 17, N . Y .

The w ord s " N a t io n a l” and "K a r b a te ” are registered trade-m arks Division Sales Offices: Atlanta, Chicago, Dallas, Kansas City,
o f N a tio n a l C a rb o n C om pany, Inc. New York, Pittsburgh, San Francisco
80
Vol. 36, No. 9
• Let Penn Salt cooperate with you

on all problems of handling and using
this important new catalyst in the
alkylation process.
Penn Salt pioneered in the com

mercial production of Anhydrous HF
at its Easton, Pa., plant (formerly the
Sterling Products Co.). Consequently,
this company’s technicians have un
usual experience which they will gladly
put at your service without obligation.
To meet the demands of the petro

leum industry in its program to in
crease greatly the output of 100 octane
gasoline, Penn Salt has expanded its
facilities for producing Anhydrous HF.
For complete information, write at once.
LT
»S THE KEYSTONE A N Y
<w U v v iu c N t K B U IL D IN G , P H IL A D E L P H IA 7,
New York . Chicago . St. louis . Pittsburgh . Cincinnati • Minnec
Wyandotte • Tacoma
W Ê SSÊ B
....................
PROVIDING ENLARGED AND IMPROVED
F A C IL IT IE S FOR PIPE F A B R IC A T IO N
AND ERECTION IN NEW ENGLAND
T h e Lumsden & Van Stone Com pany—fabricators

.and erectors o f piping since 1 8 9 8 —has been purchased by Midwest Piping & Supply
Go's', Inc."As a result, the plant in South Boston has been enlarged, new equipment
installed, and the organization expanded. Also -available-th rough this subsidiary is
the accumulated experience o f the nationwide Midwest organization In the fabrication
and erection o f pow er and industrial piping for ail operating conditions.
M ID W E S T P IP IN G & SUPPLY C O , IN C .
Main O ffices 1450 SpüiÜ Second S is Sir* Louts 4f Mo*
Plants: St*-Louis, Passaic (N. J.);pnd tas'A n ge la s
Subsidiary? tumsden Si V a n Stone Co., South Boston, Mass.
Safes Offices: N ew York (Eastern .Division)— 30 Church St, * C h ica go — 6 45 Marquette Bidg.
Ip s A n ge le s— 520 A nderson St. * Houston— 229 Shell Bidg. * Tuisa— 533 M a y o Bidg.
A tlanta— Red Rock Bidg. ■ South Boston— 426 First S b
PIPING FABRICATORS AND CONTRACTORS . . . MANUFACTURERS OF W ELDING-FITTINGS

' ■- .i V ^ ; / ■/' - 'À::.
V o l. 36, No. c
is needed r
RAYMOND
ROLLER MILL W ith
for Capacities up to W h i z z e r
40 Tons per hour Separator

•
W ith
W r it e fo r H igh Side Roller M ill wicn aouoie whizzer separator.
W id e range fineness control b y changing speed of
C A TA LO G #51 revolving whizzer blades through variable speed trans
mission.
W ith the chemical and process industries running at new high levels o f pro-
problem s ym° ° ler M llls are effectively solving today’s pulverizing
Available in Super These m odern machines are capable o f delivering huge tonnages o f finely
Mills and Midget ground materials at record lo w costs and with minimum use o f manpower.
RollerMills forlarge fi t7!.5 6" ° / “ * “ * -time by handJinS the product automatically from
and sm all plant re teed hopper to finish bin in a continuous, dustless system drying grind
quirements. ing, classifying, rejecting impurities and conveying the material to storage.
bey provide close product control, enabling you to meet exacting specifica-
ar fP dryness and fineness in producing chemicals, pigments, fillers,
resins, fullers earth, iron oxides, talc, barytes, and many other non-metallic
.V’'TU; ¡SINCE 18 8 7 1
O u t of the 2 5 ,0 0 0 different fabrics w e distribute,

m a n y a re specially m ade for use with plastics, or
with products m ade with a fabric base. There m a y be textiles on this hu ge
list w hich though unfam iliar to y o u w o uld e n ab le y o u to pro duce a better
or more econom ical product.
O u r technical k n o w le d g e of fabrics, a n d o ur w id e facilities for textile re

search m a y be a b le to help you find the m aterial that will help speed y o u r
production schedule. W e are one of the la rgest sources of su p p ly in A m e rica
BUY M O R E a n d are constantly e n la rg in g our line of fabrics for the plastics industry.
WAR B O N D S
W e shall be g la d to study y o u r in dividual needs in o rd e r to p ro vid e the
p ro p e r fabric for a n y specified product.
W E L L I N G T O N S E A R S C O M P A N Y
65 W orth Street N ew Y ork 13, N. Y.
INDUSTRIAL AND ENGINEERING CHEM I S TRY Vol. 36, No. 9
CHAPM AN’S
N E W !
G iv e s y o u th ese f e a t u r e s . . .
Replacing Chapman standard List 960 small forged steel gate valve o f the rising outside
screw type, the new List 960 is designed for still greater strength and longer service.
Specifically, here are a few outstanding features:
k Bonnet now fitted with a substantial forged k Stems, plugs, and seat rings on all valves are
steel yoke. A flanged forged steel packing heat treated stainless steel to insure greater
gland is provided, thereby eliminating exposed working qualities.
threading on valve yoke, thus giving immunity k Swivel bolts are used for holding the pack
from rusting and corrosion when placed in ex ing gland.
posed locations. k N o gaskets to blow , as in valves with flanged
bonnet connections.
JSpecify
i'k '- iii A-isw
List96y u 0u uoutside
u jw u c sscrew
tr e w ty p e in
type in ccarbon
a m o n ste e i to
steel r w oo rK
forw ingg ppressures
rkin re ssu re s 1to 800 lbs. at 750° F. O r in
alloy steel for service at 1000 lbs. w orkin g pressure at 750° F. ~ Both valves available in screw and
w elding ends. W here service conditions make it preferable the List 960i ri rising stem inside screw type
is available.
For greater pressures— get Chapman List 990 with heavier walls and extra strength. In sizes up to and
including 1 % , with screw or welding end.
The Chapman Valve Mfg. Co., Indian OrchardMass.

T W IN 0 - F I N . . . the only
F IN N E D -T U B E H EAT E X C H A N G E R
with these im porta n t points
| p o in t 1 ^_ 40,000 IN ST A LL E D U N IT S
By far the greatest num ber of installations of an y finned-tube unit . . .
an d successfully used on a greater variety of heating, coolin g and
conden sing services than an y other design of heat transfer apparatus
on the market.
| p o in t 2 ,_ PROVEN B Y 12 Y E A R S OF S E R V IC E
Durability as w ell as effectiveness of design and construction proven
b y lengthy operating records^ on a w ide variety of fluids, pressures
and temperatures . . . frequently in difficult services cau sin g trouble
with other designs of heat transfer apparatus.
ANY M A T E R IA L FOR ELEMENTS

A N D FIN S
G-Fin construction is not limited to steeL but m ay b e applied to an y
procurable non-ferrous metal as w ell as an y ferrous alloy for both
elem ents and fins.
NO PACKED JO IN T S
The exclusive patented hairpin design of the elem ents u sed in Twin
G-Fin Sections eliminate interior and exterior p a ck ed joints w hich are
a cause of trouble b eca u se of the possibility of lea k age.
GREATEST IN T E R C H A N G E A B IL IT Y
Twin G-Fin Sections are built in a small num ber of standard types and
sizes, w hich h ave b een carefully determ ined for greatest adaptability.
This method of standardization greatly simplifies the stocking problem
and provides maximum intercharigeability of purchased sections.
Write for
BULLETIN 1613 THE GRISCOM-RUSSELL CO.
285 M a d i s o n A ve nu e N e w York 17, N. Y
G R IS C O M -R 0 S S E L L
T W IN G - F I N S E C T IO N
*7h e fy n ia e à A c U ¿ t e a t Z x c U c w a e S i
FFERS M INIM UM RESISTANCE

ERS
F ib e r g l a s T o w e r P a c k in g , with its low percentage

of glass by volume (2.25% ) when packed at its nor
mal density of 3 ^ pounds per cubic foot of material,
offers little resistance to the travel of gases. Thus, its
free volume of 97.75% makes it an exceptionally
desirable packing medium for scrubbers.
Another outstanding F iberglas feature: the disper
sion o f its fibers throughout a given volume
the gases in the scrubber into a finely divided state.
Fiberglas is glass in fiber form . Consequently,
individual fibers are incom bustible . . . do not absorb
m oisture . . . will not rot or decay . . . are resistant to
m ost ch em icals...an d they remain unchanged b y time.
The glass fibers in Fiberglas Tower Packing aver
age approximately .0080" in diameter. They are ar
ranged in a jackstraw manner and are bonded together
with a fugitive binder to give sufficient rigidity of
form for easy handling and application.
FO R S C R U B B E R S
This ease of handling is a highly desirable feature
Fiberglas Tower Pack
of Fiberglas Tower Packing. It can be installed up to ing comes in the con
venient pack illustrated
70% faster than other materials. Shutdown time is
■at left. Packs are bond
held to a minimum. And because of its low cost it ed with a water-soluble
binder. Standard size
is not necessary to clean it; simply replace it with I" V 1«
new, clean Fiberglas packs.
If your process requires a packing material having
large fiee volume that provides high dispersion,
write, giving details: Owens-Coming Fiberglas Corpo
ration, 1951 Nicholas Building, Toledo 1, Ohio. In
Canada, Fiberglas Canada, Ltd., Oshatva, Ontario.
*?*ee S A M P L E S
W rite for folder contain
ing small sam ples of sev
en F ib e rg la s B a s ic . Fibers
ran gin g in diameter from
F i b e r g
*T . M. Reg. U.S. Pat. Off.-
l a s
.0 0 0 2 2 " to .0080".
... A BASIC MATERIAL

B R O W N F IN T U B E
SECTIO N A L HEAT E X C H A N G E R S
w it h t h i s n e w
U A N 6 E V H E A P
SEAL AND
E L A N G E D _
Offer These .Outstanding Advantages

Flanged construction throughout, — e lim i This new flanged construction permits in
nating screw fittings and screw unions, — terior o f fintube to be inspected and cleaned
and avoiding torsional strains on any member. w ithout disturbing shell side seal.
Gasketed seals eliminate ground joints, and Sturdy rugged design, e m p lo y in g fo r g e d

all maintenance troubles com m on to ground steel flanges with stud bolting. A ll joints
joint construction. have been tested to several times normal
operating pressures.
Gaskets are renewable, fu lly retained, and
joints may be inspected easily before sealing. These exchangers are designed and built to
Seals may be made or broken with a small give you a maximum o f lon g, efficient, satisfac
standard wrench, — from the fron t — per tory service, — and to insure easy installation,
m itting close side by side installation in easy inspection and easy maintenance by your
m inim um space. workm en . . . Let us quote on your requirements.
the B R O W N F IN T U B E «>
1 2 0 FILBERT STREET • ELYRIA, OHIO
MANUFACTURERS O F IN T E G R A L L Y B O N D E D F IN T U B E S A N D F IN T U B E H EAT E X C H A N G E R S
88
INDUSTRIAL AND ENGINEERING JHEMISTRY Vol. 36, No. 9
The use of Lïndol as the hydraulic fluid in this die-casting

operation—which involves high pressures and the pres
ence of molten metal—is insurance against serious fires
or explosions in case of leaks. This is one of many
applications where Celanese organic phosphates render
a vital service in reducing industrial hazards.
C E L A N E S E C O R P O R A T I O N O F A M E R I C A
89
INDUSTRIA' AND E-N G I N E E R I N G CHEMISTRY
September, 1944
T h e C e l a n e s e c o n c e p t
o f c h e m i c a I s e r v i c e
The s c i e n c e of synthetics has transposed ity, stability and electrical properties.

A t Celanese, research is flexible and con
many deficiencies in “ natural” substances to
“ built-in” advantages precisely suited to in tinuous. N o material is regarded as having
reached the ultimate in development.
dustrial needs.
Problems of corrosion, flammability, oxi Greater precision is always the goal— not
just adequacy. This attitude toward research
dation, heat resistance, volatility, and many
others, have steadily yielded to the chemist — required in such large measure by the
very nature of synthetics— is an asset of the
working in synthetics.
Celanese Chemical organization that will
The Celanese organic phosphates are ex
become increasingly important in meeting
amples. Since originating Lindol*, a pioneer
the problems and opportunities of postwar
product in this field, Celanese has created
prod u ction . T h a t is w hy
many “ custom-made” vari
Celanese welcomes inquir
ations of organic phosphates
ies that present needs in
to meet industry’ s require
term s of properties and
ments with greater preci P L A S T IC IZ E R S characteristics required.
sion.
Celanese Chemical C or
“ Celluflex” * and “ Cellu- O R G A N IC PHOSPHATES
p oration, a division o f the
lu b e” * for instance, have
L U B R IC A N T A D D IT IV E S Celanese Corporation of
opened up a wider range of
America, 180 Madison A v
variations in solubility, spe IN T E R M E D IA T E S
enue, N e w Y o r k 16, N . Y .
cific gravity, higher viscos
DYE-STU FFS • r c e . U . S . P a t. o rr.
T E X T I L E S • P L A S T I C S • C H E M I C A L S
90
Vol. 36, No. c
A FEW FACTS ABOUT
CORHART PRODUCTS
Corhart Electrocast Refractories are
high-duty products manufactured by
melting selected and controlled refrac
tory batches in electric furnaces, and
casting the molten material into molds.
After careful annealing, the finished IF Y O U N E E D
shapes are ready for shipment. Dense,
high-melting refractories, they are es
pecially designed for resistance to cor
rosive action.
PRODUCTS
CORHART STANDARD ELECTROCAST is a
A B E T T E R
high-aluminous refractory.
CORHART ZED ELECTROCAST is Zirconia-
bearing.
CORHART ELECTROPLAST is a high-temper-
ature plastic refractory made from Standard
Electrocast which has been ground and crushed.
Especially designed for ramming. Furnished dry.
R E F R A C T O R Y .
CORHART MORTAR is a high-quality cement
for laying up Electrocast, clay brick, or any
aluminous refractory.
C H A R A C T E R IS T IC S O F
ST A N D A RD ELÊCTRO CA ST
POROSITY : Less than 0.5%—therefore virtually ® Corhart Electrocast Refractories are high-duty products
no absorption.
FUSION POINT : Cone 38 without any appre which have proved considerably more effective than con
ciable softening below that point.
HARDNESS: 8, Mineralogist’s scale. ventional refractories in certain severe services. I f your
SPECIFIC GRAVITY: Blocks weigh approxi
mately 183 lbs. per cu. ft.
processes contain spots where a better refractory is needed
COEFFICIENT OF EXPANSION: 0.000006
between room temperature and 900° C.
SPECIFIC HEAT: 0.25 cal. per gm. per 'C. at to provide a balanced unit and to reduce frequent repairs,
980° C.
THERMAL CONDUCTIVITY: 25 BTU per sq. Corhart Electrocast Refractories may possibly be die answer.
ft. per hour for gradient of 1°F. per inch.
COMPOSITION : Standard Electrocast is of an The brief outline at the left gives some o f the basic facts
aluminous crystalline nature.
CORROSION : Because of low porosity and about our products. Further information will be gladly
inherent chemical make-up, Corhart products
are highly resistant to corrosive action.
sent you on request.
A P P L IC A T IO N S
Most heat processes present spots where a better
refractory material is needed in order to provide Corhart Refractories Company, Incorporated, Sixteend
a balanced unit and reduce the expense of re
peated repairs. It is for such places of severe
service that we invite inquiries regarding Corhart and Fee Streets, Louisville 10, Kentucky.
Products as the fortifying agents to provide the
refractory "balance” desired.
The following is a partial list of applications " C o r h a r t ” i s a t r a d e - m a r k , r e g is t e r e d U . S . P a t e n t O ffic e .
for which Corhart Products are suggested:
ELECTROLYTIC CELLS — for production of
Magnesium and other light metals.
SILICATE OF SODA FURNACES — sidewalls,
bottoms, and breascwalls.
HEARTHS AND SMELTERS — for non-ferrous
metals.
ALKALI AND BORAX MELTING FURNACES ENDURANCE
— fast-eroding portions.
GLASS FURNACES — the entire installation of
flux walls and bottoms, breastwalls, ports, tuck*
stones, forehearths, recuperators, etc.
RECUPERATORS—tile, headers, separators, etc.
ENAMEL FRIT FURNACES — flux walls and
bottom.
BRASS FURNACES • - metal contact lining.
ELECTRIC FURNACES — linings for rocking
type, and rammed linings of Electroplast for
this and other rypes.
BOILERS — clinker line.
STO CK SH A PES AND S P E C IA L S

Standard and Zed Electrocast are made in stock
shapes and in many special shapes. The weight
of individual pieces may run to 3500 pounds.
September, 1944
n u
m m
In R e s e a r c h , A n a l y s i s a n d C ontrol,
Serves In These D e v e lo p m e n t s
C o n s u lt le a d in g L a b o r a t o r y S u p p ly H o u s e s
th r o u g h o u t t h e U n ite d S t a t e s a n d C a n a d a
f o r K im b le p r o d u c ts t o m e e t y o u r n e e d s .
KIM BLE GLASS C O M P A N Y • • • . VINELAND, N. J.

H EW YORK • CHICAGO • PHILADELPHIA • DETROIT • BOSTON • INDIANAPOLIS • SAN FRANCISCO
IN D U S T R IA L A N D E N G IN E E R IN G C H E M IS T R Y
Vol. 36, No.
FO AM G LAS* CORKBOARD
for permanent mois for highest insu
ture and vapor seal lating efficiency
Armstrong Offers a New Insulation Construction

A R M S T R O N G has developed a new type o f
out all moisture. This insulation material is glass
x low-temperature insulation construction
in cellular form, with each tiny cell wholly iso
that combines Armstrong’s Corkboard with
lated from the others. Hence, Foamglas is mois-
Foamglas- and takes advantage o f the best
tureproof, vaporproof, and fireproof.
properties o f both materials. This new construc
tion furnishes a positive, permanent seal against The new construction is easy to erect since
infiltration o f moisture and vapor— thus assuring both materials are erected in accordance with
permanently high insulating efficiency. standard specifications. Interior wall finishes are
Since it is basically standard corkboard the same as for standard corkboard construction
construction, the new combination retains cork F O R C O M P L E T E IN F O R M A T IO N about
board s natural moisture resistance and excep the new combination construction— including
tionally low thermal conductivity. This assures engineering drawings and erection specifications
high efficiency for heavy-duty work. — write today to Armstrong Cork Co.,
However, the outer shell is Foamglas— to seal Building Materials Division, 9709 Con- fw A
cord Street, Lancaster, Pennsylvania.
*Reg. C. S. Pat. Oil. Product Mfg. by Pittsburgh Corning Corp.
Cotkboflrd
Mlne,al W °°> Bo»rd Foamslm*
Cork Covering
ARM STRONG CORK COM PANY

Insulation (Q) Headquarters
Over the telephone he directs the

firing o f b ig guns on one o f the
fronts. This is a vast war and
communications are vital.
So the' telephone industry’ s man

ufacturing capacity has had to be
devoted almost exclusively to p ro
ducing electronic and telephone
equipment fo r our armed forces.
That explains w hy there are de
lays in filling orders fo r home
ones,
All o f us in the telephone busi

ness are grateful fo r y ou r fine
spirit o f co-operation in this war
emergency.
BELL T E L E P H O N E SYSTEM
Please tr y to k e e p th e Lo n g
Distance circuits clear from 7 to
10 each night fo r the service men
and women.
The b ig guns sta rt boom ing when th e se ctio n c h ie f o f a cannon com pa ny g ives th e o rd e r to " fire 1
Compressors furnish clean compressed air, free from dust, heat or oil.
Nash Compressors perfectly meet the need for clean compressed air in transferring
beer or wort by pressure displacement, for maintaining pressure displacement, and
for supplying clean compressed air to bottling machines.
Nash Compressors furnish clean air without supplementary filters or air washers.
They are simple, efficient and economical. One moving part, rotating on ball bear
ings. No internal wearing parts in metallic contact, and no internal lubrication. Ask
for Bulletin D-236.
N A S H E N G I N E E R I N G C O M P A N Y
SOUTH N O RW ALK, CONNECTICUT, U. S. A
September, 1944
Columbium-Bearing STAINLESS STEEL

Serves the Process Industries
GATX33060
mÊm
This 8 0 0 0-g allon tank car, m ade of coium bium -bearing stainless
steel w elded with colum bium -bearing w e ldin g rod, keeps chemical
shipments iron-free to the extent of one part per m illion.
• The a u ste n itic chro m ium -nickel sta in le ss ste els

containing colum bium re ta in th e ir fu ll co rro sio n
resistance e v e n a f t e r lo ng e x p o s u re to t e m p e r a
tures in th e r a n g e o f 7 5 0 to 1 6 5 0 d e g . F . If w e ld e d
with c o lu m b iu m -b e a rin g w e ld in g ro d the w e ld s
also a r e fu lly c o rro sio n -re sista n t w ith o u t h e a t t r e a t
C o lum bium -bearing stainless steel in this 12-in. blow pit p ip ing for a paper mill
ment su b se q u e n t to w e ld in g . Y o u should co n sid e r assures high corrosion resistance regardless of temperatures encountered during
fabrication and service.
these a d v a n t a g e s w h e n p u rch a sin g c h e m ic a l p ro c
essing e q u ip m e n t.
W e d o not m a k e s te e l— but fo r o v e r 3 7 y e a r s
we h a v e m a n u fa c tu re d “ E le c tro m e t” f e r r o - a llo y s
and m e ta ls used in m a k in g s te e l. Ferro chro m iu m
and fe rro co lu m b iu m , u se d f o r a d d in g chrom ium
and colum bium to s te e l, a r e tw o o f th e se f e r r o
a llo y s . . . F u rth e r in fo rm a tio n a n d m e ta llu rg ic a l
assistan ce b ased on this w id e e x p e r ie n c e a re
a v a ila b le to h e lp yo u m ake p r o f ita b le use o f
stain less a n d o th e r a llo y ste e ls a n d iro ns.
B U Y UNITED STATES W A R B O N D S A N D STAM PS Need for further heat-treating is elim inated w he n these large chem ical tanks are
constructed of colum bium -bearing stainless steel with co lum bium -bearing w e ld
in g rod.
E lectro M eta llu r g ic a l C ompany

Unit of Union Carbide and Carbon Corporation
3 0 E a s t 4 2 n d S tre e t l*t < H N e w Y o rk 1 7 , N . Y .
E le c tro m e t Trode Mark
In Canada: Electro Metallurgical Company of Canada, Limited, Welland, Ontario Ferro-Alloys &Metals
The w ord Electromet" *s a reaistered trnde-m nrk o f Fl*»rfro AA^tnllnmimi Ct>rr
D R Y ing benzol is a case in point— used in the production of plastics. Other Lectro
Water in benzol has always been objection dryers dry air, gases and various organic liquids,
able, and users dried it in a perfunctory' sort of working at atmospheric pressures or as high as
way. Then along came new uses for benzol 3,000 pounds per square inch. They dry to
which called for really drying it D R Y . Chemists extremely lowr dewpoints or operate in connec
now' talk in such terms as “ 3 to 7 parts o f water tion with automatic controls to maintain con
per million” . stant moisture contents.
Finding that amount o f water in benzol is like Whatever your D R Y ing problem, it is quite
trying to spot one Jap in seven acres of Chinese. likely that our engineers have previously solved
But that’s the kind o f moisture-seeking-out similar ones. No need to waste the time o f y'our
jobs that Lectrodryers arc handling. Activated men on unfamiliar tasks. And, often, there’s a
alumina, the drying agent in Lectrodryers, dries standard Lectrodryer ready to step out o f the
benzol and other hydrocarbons within the pres catalog and handle the jo b for you. P it t s b u r g h
ent limits o f measuring their moisture content. L ectro d ryer C o r p o r a t io n , 395 32nd Street,
The Lectrodryer pictured above dries benzol Pittsburgh 30, Pennsylvania.
LECTRODRYERS DRY W IT H A C T IV A T E D A L U M IN A S
L E C T R O D R Y E R
. .... 8ei.U.S.PaL0ll.
C O R P O R A T I O N
MORE ECONOM ICAL

AND E FFI
Pritchard Atmospheric Towers feature maximum air flow

and drift control
u ', u r i n g the 15 years J. F. Pritchard and Company now in operation are p ro o f that this policy has been
have designed and con stru cted coolin g towers three rigidly adhered to.
equally important factors have been kept in mind . .. There is a Pritchard T ow er for every job with the
Improved C o o lin g Efficiency . . . R ed u ced O p era tin g capacity to meet every load dem and. Y ou r cooling
C o s t. . . and Prolonged T ow er Life. The hundreds o f tower problem can well be placed in the hands o f
Pritchard Tow ers— both mechanical and atmospheric— Pritchard engineers with the full assurance that it will
receive individual study and result in a properly de
signed and constructed tower.
Atm ospheric Cooling Towers are designed to give the greatest

amount o f air flow, contacting the water where concentration is
the heaviest. Construction is such that water break-up is maximum,
thus giving more actual water surface for air to pass over.
Mist extractors prevent excessive moisture loss due to drift, the
trapped water b ein g returned to circu la tio n by means o f drift
retrievers. These are exclusive Pritchard features.
P ritch ard M ech a n ica l Cooling Tow ers are con

Above — One of Ihe Non-corro- structed fo r induced draft or forced draft opera
sivo “Featherw eight" fans in
stalled in a Pritchard forced draft tion. Both are built upon the principle o f counter
mechanical tower flow, a method making fo r efficiency. Use o f the
exclusive non-corrosive "Featherweight” fans as
sures low er cost and longer life. J. F. Pritchard
Right — Pritchard induced
and Company, C ooling T ow er Division, Fidelity
draft mechanical cooling
lowers offer patented Building, Kansas City, Missouri.
protective features
K A N S A S CIT Y HOUSTON
NEW Y O R K
C H IC A G O P ITTS B U R G H
ENGINEERS A N D C O N S T R U C T O R S
gFOR THE C H EM ICAL • PETROLEUM • G A S A N D P O W E R INDUSTRIES!
Vol. 36, No. 9
cLLLOTT e n g i n e e r i n g a i d s p r o d u c t i o i
o f b l o o d p l a s m a a n d p e n i c i l l i n
In the processes used for producing the two life- Elliott engineers have m ade a continuing study
giving substances, blood plasm a and penicillin, of the steam jet ejector for over 25 ye a rs, and have
extrem ely high vacuum is essential. The m ainte pioneered in m any of its applications to process
nance of such high vacuum w as im m ediately recog w ork. The steam jet ejector itself is b asically simple,
nized as a job for steam jet ejectors. The engineers has no moving p art, requires no lubrication or other
working on the processes turned to Elliott Com pany, attention than mere supervision, occupies small
whose engineers have successfully ap p lie d multi space, can be installed an yw h ere.
stage steam jet ejectors to m any high-vacuum •
processes. Elliott four-stage and five-stage ejectors
W rife fo r B ulletin G -7 on steam je t ejecto rs a n d th e ir w ide
a re very successfully operating in these processes
ra nge o f a p p lica tio n . I f y ou h a v e a sp ecific vacuum prob lem ,
and others such as magnesium production. Some take it up w ith Elliott en g in eers.
of these ejectors a re the largest five-stage ejectors
yet built, and some are m aintaining the highest E L L I O T T C O M P A N Y
Dept., J E A N N E T T E , PA.
vacuum (lowest absolute pressure) ever obtained Heat Transfer
D ISTRICT O F FIC E S IN P RIN C IP A L C ITIES
A n Elliott engineer explains

the use of four-stage steam
jet ejectors for b lood plasma
production, to a group of
Red C r o s s w o rk e r s , who
incidentally are wives of
Elliott Com pany men.
STEAM JET
EJECTO RS U O T T
S T E A M JET EJECTORS
D ESUPERHEATERS • FILTERS
An Important
p ^ I
A N N O U N C E M E N T
P p' ' *v- *' ^ ^ 1* ^ t " '7 ; " v ■ ' U'.U " ■ : U'- A . ' -v 'U AN'.. • < ‘ig.;
to Wectrode and Anode users
o t h a t OUR product development men Mr. Nelson brings to the work both aca
will be better able to keep pace with devel demic and practical training. For fifteen
opments in your industry, National Carbon years he guided quality control operations
Company, Inc., has just reorganized and for many o f the carbon and graphite prod-
expanded its Electrode Service Engineer xicts o f the company. He lias spent much
ing Department. time in the field on actual application prob
This department concerns itself con lems involving the use o f elcctric-furnace
stantly with the improvement o f carbon electrodes.
and graphite electrodes and graphite an Dr. Chappell’s education and training
odes. It also consults with and advises cus also make him exceptionally qualified as
tomers in the electrometallurgical and supervisor of clcctrolhcrm ic work o f the
electrochemical industries on the applica new department, while Dr. Johnson, out
tion technique, and the efficient use of standing research and process-engineering
electrodes and anodes. scientist, is thoroughly familiar with prob
The key personnel o f the reorganized lems arising in the electrolytic division.
department is headed by T. L. Nelson, and Many trained technical assistants are on
will consist o f Dr. Charles H. Chappell, in the staff o f the department, the entire facil
charge o f the electrothermic service divi ities of which will immediately be placed
sion, and Dr. Neal J. Johnson, in charge of at your service upon notice to any member
the electrolytic service division. of our nation-wide sales department.
BUY UNITED STATES W A R B O N D S
T h e “N ational"’ and “Achcson” Seals, as

well as the words “N atio n al” and “Achcson”
a r c r e g iste re d tr a d e -m a r k s o f N a t io n a l
Carbon Com pany, Inc.
NATIONAL CARBON COMPANY, INC.

U nit o f Union C a rb id e a n d C arbon Corporation
nsa
G e n e r a l O ffices: 3 0 East 4 2 n d Street, N e w Y o r k 17, N. Y.
D iv is io n S a le s O ffice s: A t la n t a , C h ic a go , D a lla s , K a n s a s City, N e w Y o r k , Pittsbu rg h , S a n Francisco
THE ULTI MATE

I N I N T I M A T E
DUSTLESS BATCH
BLEN D IN G
STURTEVANT ROTARY BATCH BLENDERS are THE

designed to produce to the most exacting specifica
STURTEVANT
tions. These efficient self-cleaning mixers blend any
ROTARY DUSTLESS
number of ingredients with varying weights, densi
BATCH BLENDER
ties, fineness and other physical properties — with
no loss of dust or finely ground materials. Operation
is simple, automatic and fast. The drum is heavy the bottom of the drum, while a third action forces
steel, trunnion mounted. Drive shaft and rear this material from both ends to the middle, adding
trunnion have anti-friction bearings. another mixing action with no separating effect.
Finally, the chute, swinging in a blending position,
Four separate actions assure perfect blending. As produces an important fourth lateral action and
the w eighed and measured material enters the drum assures a thorough and perfect mix.
through the opening in the center, it is constantly
lifted in a series of revolving buckets and cascaded Discharge, with the chute in the discharge position,
into an intimate mix. (See illustration “ R eceiving" (see illustration below) is through the same center
below.) opening and is completely automatic.
The steadily revolving action of the drum makes a STURTEVANT MILL COMPANY
continuous blend or mix of the material remaining in 2 HARRISON SQUARE, BOSTON 22, MASS.
R E C E IV IN G D IS C H A R G IN G
• Send today for further details! •
T U R T E V A N T
I
I
I
O u t o f a te s t g t u b e
in to h e r e ...I
For Y our
E hear a lo t a b ou t m ira cle p rod u cts that
P o s t w a r P ro d u cts
w ill co m e fr o m in du strial laboratories.
General American engineers
are ready now to consult O n e o f th em m ay b e y ou rs— presen tin g n ew
with you— to plan new tank p ro b le m s in safe tran sp ortation — r e q u irin g cars
cars with every feature need
w ith n e w lin in g s o r coatin gs u n u sed today.
ed to transport your products
safely. Call or write our
G en era l A m e rica n w i ll m eet the ch a lle n g e —
general offices — 135 South
just as w e p r o v id e cars w ith
LaSalle St., Chicago 90, 111.
carry h eliu m gas, m u ria tic acid, butane
oth er h a rd -to-h a n d le com m od ities.
GENERAŁ AMERICAN TRANSPORTATION

CORPORATION
IN D U S T R I A L AND ENGINEERING CHEMISTRY Vol. 36, No. 9
OLDBURY
E L E C T R O C H E M IC A L
S
G R A N U L A T IN G COMPANY
WET MILLING
PULVERIZING
Comparisons with ordinary m illing meth
ods prove that MIKRO-PULVERIZER
"feed-line” efficiency actually saves more PHOSPH ORU S
than half the processing time — more than
half the manpower and fuel costs. O X Y C H L O R ID E
In grinding many products, it’s just one
step from feed to processed product when
you use the MIKRO-PULVERIZER! Y ou
get closer particle size control in pulver-
izi°g> granulating, or wet milling . . .
coarse, fine, or superfine grind.
And that goes for chemicals, food, dye-
stuffs — crystalline, amorphous, smeary, or A y fA N Y years of manu-
hygroscopic materials!
facturing experience,
together w ith continuous
technical improvements, ena'
ble us to produce a water'white
Phosphorus Oxychloride re'
markably free from impurities.
Shipping containers are re'
turnable drums and tank cars.
The M IK R O -P U L V E R IZ E R blends
and pulverizes in one operation/
FR E E TEST G R I N D )
• Eliminates screening
Just submit a sample of
• Keeps air dust-free youf material to our test
grind laboratory. There’s
• Reduces cleaning time up no obligation and all tes»
to 90% data are held in strictest
confidence.
• Saves power from 25to75%
• Requires a minimum o f
attention o f unskilled Informative new32-page
help — mesh controlled catalogue showing the
mechanically complete MIKRO -PU LVER
IZER line sent FREE on
• Needs little floor space — request. P la n t a n d M a i n Office:
no steel or concrete
foundations N IA G A R A FA L LS, N E W Y O R K
p u l v e r iz in g m a c h in e r y c o m p a n y
4 0 Chatham Road • Summit, New Jersey h{ew T o r \ Office:
2 2 EAST 4 0 t h ST., N E W YORK 1 6 , N .Y .

POLYETHYL
A " C a r b id e C h e m ic a ls " P ro d u ctio n .
A c h ie v e m e n t f o r th e Nqj|ry
Im portant N e w Plastic Has M a n y “ One year ago’ vour plant com m enced the production .oft
polyethylene, ;u com ponent o f radio cable essential to the
Unusual Properties
efficiency o f electronic communications units and, therefore,
A Tittle -over tw o years ago, the U. S. Navy learned that vital to the success of naval operations. Production for the ,.j
Carbide and Carbon Chemicals Corporation in collaboration year has equalled 240 per cent o f the rated output for the
-with an associate com pany, The Linde Air Products C om facilities. Everyone engaged in developing the pi’b’cjuct, ~
pany, had developed a high pressure synthesis o f a new planning, engineering, and managing the plant,,jinêach o f ,
material, Polyethylene, and it was found that this material you engaged in producing polyethylene may. bb.justly proud J
was exactly suited to meet the Navy’s requirements o f an o f a valuable contribution to the war effort.? I' : r '■Lb'S-*'
-'insulation for coaxial cable used in radar equipment. Today, approximately two years after Carbide was given
At the N avy’s request, these two companies, working ’ the assignment, this plant is producing polyethylene at 600
together but entirely independent o f anyone else, designed per cent o f rated capacity and is providing the N avy’s re
in their own Engineering Department and built with their ig quirements o f this material for use in coaxial p a b le ,,'
own Construction Organization a plant to produce polyethv- pi ' i-\i if/-: '
lene by a process different from any other comm ercial poly-
THE IMPORTANT N EW PLASTIC! Polyethylenei'esins
ethylene process. ;
have the most favorable electrical characteristics of-'ahy plas
W ithin thirteen months from the date the project was r?
tic material for use in this electronic application. In addition,
authorized, this plant was producing at 180 per cent o f rated ,
this new plastic material has many other,.exceptional char
capacity.
acteristics, which suggest widespread application in many 1
At the conclusion o f the first full year o f production the ■
different fields. Polyethylene plastics are tough and impact-
Navy Department told the plant;
resistant. T hey are inherently flexible and extensible. They
have an extremely low water vapor transmission coefficient
and will absorb an unusually low percentage o f water. Their
chemical resistance is outstanding. Polyethylene is one of
the lightest plastics, so light that it will float in water. It
maintains its valuable properties over a w ide range o f tem
perature. It remains usable at temperatures lower than 90
degrees below zero, Fahrenheit, and is sufficiently rigid for
use in temperatures up to 230 deg. F.
Polyethylene is colorless and translucent as originally
manufactured but it can be formulated to produce colored
products o f exceptionally high lustre. These products can be
fabricated b y standard processes on existing plastics equip
ment. ' ■ '
M olded and extruded products, cloth coatings, flexible
sheeting and film, and monofilaments are among the poly
ethylene plastic products which will be available in the
future. .A
Polyethylene plastics are now restricted to applications
covered b y W P B Limitation Order No. 348. Technical data
and samples for controlled end uses can be obtained by
manufacturers with plastic-processing equipm ent by writ
ing Department 2.
Plastics Division
CARBIDE AND CARBON CHEMICALS C O R PO R A TIO N
Unit of Union Carbide and Carbon Corporation
HU
30 E a st 42nd S tr e e t, N ew York 17, N. Y*
( This advertisement has been rev iew d and approved by the U. S. Navy Department")
Vol. 36, No. 9
Hey Bill, the

like the one
we
KINNEY
1/O Û U 444H ,
PUNPS
Since you re not an angel and can’t w o rk on such a high
plane, you ’d better keep both feet o n the grou n d and
use a K IN N E Y Pum p! T h e vacuum sintering o f tanta
lum and alloy metals, and the produ ction o f magnesium
A s illu s t r a t e d , th e lo u v re s
by the ferrosilicon process are som e regular uses for
fo rce the larger bubbles to
K IN N E Y Pumps. A n oth er o f many new applications
c o llid e . T h is is th e a c t io n
o f these dependable pumps is maintaining the lo w ab so
so u g h t by o p e ra to rs o f lute pressures necessary in the electric furnace refining
Fractio nating C o lu m n s who o f rare metals.
w anted to use a pressed type
c a p , b u t w e re u n a b le to K IN N E Y H igh Vacuum Pumps, w ell kn ow n fo r their
o b ta in th e d e s ire d re s u lts ultimate vacuum, volum etric efficiency, and trouble-free
, with the s t r a i g h t s lo t s used operation, are available in tw o types: K IN N E Y Single
in the conventional pressed c a p . T h erefo re, most Stage Pumps fo r lo w absolute pressures d ow n to 10
operators o f Fractio nating Colum ns continued to use m icron s; K IN N E Y C om poun d Pum ps fo r lo w absolute
the c a s t bubble cap because this was the only w ay pressures d ow n to 0.5 m icron . Bulletin 18 on request.
r d ic - u t
S m uUbb,'C c o llid i" S ^ t i o n . T h is new Let us know your requirements.
h ,i d e v e lo p m e n t in c o r p o r a t in g lo u v re s ,
allows in e use o f the pressed cap .
n, Stase Compound
Dry Vacuum Pump Dry Vacuum Pump
Pressed from 14 gauge m aterial. A v a ila b le in Stainless
Steel or C o p p e r. W h ile the standard size cap used
to d a y ,s 3 diam eter, B R IG H T O N can m anufacture
the new L O U V R E B U B B LE C A P in an y size desired.
Prices on ap p licatio n .
vm
JJÜL
-rrTT
*,r
k-
K I N N E Y M A N U F A C T U R IN G ,CO.
3549 W A S H I N G T O N S T . , B O S T O N 30 M ASS
NEW Y O R K C H IC A G O P H IL A D E L P H IA
< LOS ANGELES S A N F R A N C IS C O
We also manufacture vacuum tight valves, liquid pumps, dutches and
bituminous distributors.
September, 1944
WHEN
BEACHES
BECOME
PORTS
« ^ .A J N V IN G m e n , tan k s, trucks, guns and e q u ip m e n t directly

o n e n e m y beach es is all in th e d a y ’ s w ork fo r U ncle Sam s
L S T s. F or m o re th a n tw o years ou r p la n ts have b een fa b r ica t
in g steel fo r the h u lls o f th ese “ fig h tin g Lruck h orses.” Every
th in g fr o m sm a ll u n its to p refa b rica ted steel box section s, are
sh ip p ed fr o m ou r fa b rica tin g p la n ts b y rail to ou r sh ipyard . . . .
h o iste d in to p la ce w ith craw ler cranes and w elded i n lo L STs.
F or o u tsta n d in g p r o d u ctio n o f L S T s, all o f ou r fa b rica tin g
p la n ts an d o u r shipyards have h een aw arded th e A rm y-N avy
“ E .” D espite this d irect c o n tr ib u tio n to the w ar effort we have
b e e n d esig n in g and fa b rica tin g all types o f storage tanks fo r
th e p e tro le u m in d u stry , th e s y n th e tic ru b b er in d u stry and
ch e m ica l an d process p la n ts. I f y ou have a storage p rob lem
... Atcarded to all oj
fabricating plantsy write o u r n earest office, o u tlin in g y ou r req u irem en ts. Y o u r
our shipyard and dry ~ in q u iry w ill receive ou r im m e d ia te a tte n tio n .
dock yards.
C H IC A G O B R ID B E & IR O N C O M P A N Y
Chicago 4 2482 McCormick Bldg. Washington S ...........................................¡• 3 3 0 B ?’»«» Bid».
New York 6 3374— 165 Broadway Bldg. Houston 1............................................... 5639 Clinton S 3 " *
Philadelphia 3 1636-1700 Walnut St Bldg. Tulsa 3 .........................................................1636 Hunt Bldg
Cleveland I S .......................................... 2253 Guildhall Bldo. Greenville York Street;
Birmingham 1 1574 North Fiftieth St. San Frlncisco '5............................ 1067RialtoB
In Canada: HO R TO N STEEL W O RKS, LIM ITED, FORT ERIE, ONT.
Plants in BIRM IN G H A M , C H IC A G O and GREENVILLE, PA.
A ce Hard Rubber-lined and
covered steel tanks in all sizes Single and double acting, cen
and capacities to specification. trifugal and rotary gear pumps
These may be connected with rendered immune to chemical
entire circulating systems of attack with full hard rubber
hard rubber or rubber-lined protection over all parts in con
pipe and fittings. tact with corrosive solutions.
Gate, diaphragm and check Solid HARD RUBBER, flanged

valves with fully bonded hard and threaded — Hard Rubber
rubber linings over all inner lined flanged iron pipe and fit
surfaces. tings—SARAN pipe and fittings,
in a range of standard sizes.
RECONVERSION
A LREADY, in hundreds of RECONVERSION dem ands quick
(LS jI m a n u f a c t u r i n g plants revision of schedums, realignment
throughout the country, the pres of facilities, planned replacements
sure of war production is easing of obsolete facilities, new process
off; giving ground to problems of ing methods. All of which m eans
pent-up civilian dem and. dependable equipment that will
Long-range planned economies stand up and deliver uninterrupted
now — well ah ead of, or even at the service for years ahead.
time of re-conversion — are bound American Hard Rubber Com pany
to have a definite effect on post invites you to look into the! a d van
war profit margins. tages offered you by converting—
w i t h
H A R D R U B B E R P R O T E C T IO N
ACE HARD RUBBER will provide economies in plant maintenance
the logical answer to m any chem costs, such as no forward looking
ical processing problems, where plant executive or engineer can
adequate protection against costly afford to overlook.
corrosion is an economic "m ust.” W h ere both heat and chemical
Savings effected with A.ce Hard resistance are factors in industrial
Rubber-lined Tanks, Pumps, Pipe processing operations, new synthe
Fittings and Valves, are a matter tic hard rubber compounds offer
of recorded performance. A ce pro further protective opportunities. In
tects against da m age to valuable this direction, too, A ce research
and solutions and goods in process.

Result: substantial long range
has already m ade significant pro
gress.
SARAN This rem arkable new plastic

has materially increased our
ability to extend non-corrosive
protection into new fields. Its resistance to
W R IT E FOR
YOUR
FREE COPY
A. reference b ook of
m any rubber solvents, as well as to almost 64 p a g e s for plant
all active chem icals, is remarkable. e x e c u tiv e s, e n g i
neers an d labora
Saran pip e and fittings are presently tory iedhnicians.
available in standard iron pipe sizes from
AMERICAN HARD
1/ 2 to 4 inches. A simple electric heat butt- RUBBER COMPANY
w elding operation m akes it possible to 11 IVe’xer Street,
New Ycrk 13, N.Y.
assem ble long Saran pipe lines on the job. Akron 4, Chio
Circular on Saran sent on request. 111 W. Washington St.,
Chicago 5, III.
m e a n s f l C F protection
fo r your chem ical storage and circulating system
Vol. 36, No. 9
0í¡jST C O Œ ; m ill v o u r p u m p s
»?£* PROBLEM
DUS T be o b s o le t e ?
D u st even in sm aJla m ou n ts
Many plants will one day face costly "one-purpose”
s h o u ld be c o n t r o lle d b e
pump waste. T o reduce this waste, Taber Pumps are
cau se it is d estru ctive in
designed for wide adaptability... merely changing
s o m any w ays. N o m atter
how serio u s your dust u d ù tf Impeller and Casing fits them for various services.
hazard m ay b e D R A C C O
has the r ig h t e q u ip m e n t
DRACCO
PNEUMATIC CONVEYORS
This feature, plus time-proved, high practical per
formance, make Taber Pumps more suitable for the
to c o n t r o l dust fr o m an processing industry. For detailed, helpful information
in d iv id u a l m a ch in e to an
The cost o f m oving chem on this new Taber series "General-Use” Horizontal
en tire plan t. W hy not
icals, grains and granular Centrifugal Pumps— write for Bulletin CLVS-339.
c o n s u lt D R A C C O E n g i
materials can be greatly 6409
n eers; they have o v e r 30 Ja T A B E R "G E N E R A L -
reduced in plants using
~ U S E ”C E N T R IF U G A L
y ea rs’ e x p e rie n ce . obsolete methods. It P U M P . N e v e r obsolete.
Easily adapted to many
might pay to check into jobs. H a s over-size ball
bearings, extra shaft diam
your handling problems. eter, deeper stuffing box.
W hy not let DRACCO TA BE R P U M P CO .

(Est. 1859)
Engineers do it for you — S93 E lm S t.,B u ffa lo ,N .Y .
they have reduced han
dling costs to the M IN I
M U M for many. W rite
for further details.
(&***“DRAnnn ^ metal
^ Lm n u u u f a b r ic a t io n ^
i
The D RACCO Metal
Fabricating Division
is helping many in
solving intricate fab
ricating problem s.
W e can handle the
w ork from light
gauge up to 14" plate
and are n ow prod u c
T O P S IN SIFTING EFFICIENCY
Generally considered tops in sifting efficiency in
ing gear guards, ma
chine stands and
★24 vital war plants where they have been d oin g heavy,
h o “ r duty. Invariably additional units have been
ordered whenever greater sifting capacity was required,
bases, cabinets o f all l hese sturdy, com pact, self-contained machines crow d
types and many other products. Our Engineers are m ore sifting capacity into less floor space than you ever
thought possible. Offers com plete Rotary M otion
exceptionally w ell qualified to help you. mechanically con trolled. D oes an extremely thorough
jo b on single o r multiple separations. P recision-built o f
finest materials foryears o f econ om ical, trouble-free service.
F o r F u r t h e r I n f o r m a t i o n W r it e
DRACCO C O R P O R A T I O N B. F. Gump Co.

4055 E- 116th St., Cleveland 5, Ohio New Y o rk Office: 130 W. 42nd St
Established 1872
I DUST CONTROL EQUIPMENT 415 South C lin to n Street, C h ica go 7, III.
^PN EU M ATIC CONVEYO RS • METAL FABRICATION r . ... . Makers of Equipment for
jrindmg, Sitting, Mixing, Feeding and Weighing of Dry Chemical Products.
September, 1944
Heat Exchangers in the foreground have monel metal

shells, admiralty metal tubes and naval bronze tube
sheets, floating head covers, and baffles. V ogt Ell Bolts
are used on floating head covers.
Reactors for catalytic polymerization units being assem
bled and tested. Each is of welded construction and has
bolted-on hinged covers.'
Group of Diesel Oil Exchangers for a visbreaker. Used
to exchange heat between diesel oil and corrosive crude
containing salt and sulphur compounds.
62"x30'-6" long Gas Cooler for a catalytic unit in a large
refinery. Tube bundle at right is for a second unit.
HENRY VOGT M ACHINE COMPANY

Incorporated
L O U IS V I L L E , K E N T U C K Y
Branch Offices: New York. Philadelphia. Cleveland, Chicago, Dallas.
D O W T H E R M
FIRfT UNIT
4 ,0 0 0 ,0 0 0 Bto. per hr. capacity
9 The development of new processes in the
early 1930's emphasized the limitations of the high
temperature heating methods then available. In
that period Dowtherm Vapor heating systems were
developed by Foster Wheeler. Utilizing the latent
heat from condensing vapor, Dowtherm combines
the constant temperature advantage of steam heat
ing with high temperature processing at low pres

sure.
Precision control of product heating from 400 to
700 degrees F.; ability to maintain uniform surface
temperatures; to supply heat at different and vary
Original Dowtherm Vaporizer installed in 1932 furnishes ing temperatures from one source of heat; to heat
process heat for chemical production.
and cool in the same cycle . . . these characteristics
of Dowtherm heating have gained the unqualified

acceptance of Industry.
Foster Wheeler is a four
star winner under the
U. S. Navy Board of
Awards for production.
FOSTER WHEELER CORPORATION

S ptember, 1944 IN
H E A T I N G
I 6 4 '"UNIT
3 5 ,0 0 0 ,0 0 0 Btu.
• Since the installation of the original
Dowtherm unit for a relatively simple high tem
perature chem ical operation, the indirect heating
system has b een successfully adapted to a variety
of processes in the chem ical, petro-chem ical and
petroleum industries. A m ong these applications
are:
Synthetic Rubber Production

Edible Oil Deodorization
Bodying of Drying Oils
Synthetic Resin Production
Phthalic Anhydride Purification
Fatty Acid Distillation
Estérification
Fat Splitting
Special Petroleum Processing
Plastic M oulding and M anufacture
Asphalt Processing and Coating
Wire and Cable M anufacture
Further information on. Dowtherm, including its

use in Power Plants, is available in bulletin
ID-43-5.
Large new Dowtherm installation supplying process heat for butadiene

production.
165 BRO A D W A Y, NEW YORK 6, N. Y.

Vol. 36, No. 9
Industrial Brownhoist Material Handling Equipment

Industrial Brownhoist manufactures equipment for handling all
types of materials. W hether you need small clamshell buckets or
the largest coal and ore bridges, Industrial Brownhoist will design
and build standard or special equipment to do the. job ..with the
greatest' speed and efficiency.
Above: Three Industrial Brownhoist high-lift electric car dumpers,

capable of handling sixty cars per hour, each car carrying up to
120 tons of coal. Right: A typical Industrial Brownhoist gasoline
or diesel locomotive crane for use with magnet, hook or bucket.
Patented Monitor-type cab allows 360° visibility and increases
operator efficiency.
• BAY CITY, » MICHIGAN

DISTRICT OFFICES: New York, Philadelph. lia, Cleveland, Pittsburgh, Chicago • Agencies: Detroit, Birmingham, Houston,
Denver, Los Angeles, San Francisco, Seattle, Vancouver, B.C., W innipeg, M anitoba, C anadian Brownhoist, Ltd., Montreal, Quebec.
ACCURATELIQUID LEVEL INFORMATION

AT A GLANCEf WHEN YOU WANT IT - RECLAMATION STILL
FOR
BY-PRODUCT COKE \
ff
or — :------
CHEMICAL
INDUSTRY
CO LU M N
LIQUIDOMETER
" THEY
Tjy r\S*Dc ALW
A ! 1A/AYS
AK/C* D
nEr PEN
n r D ABLE
Mn # r-"* *
RE
100% automatic.
No pumps, valves, or auxiliary units
n eeded to read them.
M any vital products of the chem ical and
M odels available for either remote by-product Coke industries are ordinarily
or direct readings. lost in processing. Be consistent with
A c c u r a c y u n a ffe c te d b y sp ecific TO CO KE those aiding the national em ergency by
gravity of tank liquid. in d u s t r y using Hicks reclam ation stills to prevent
these wastes.
A pproved b y Underwriters' Labora
W rite fo r further information.
tories for gau gin g hazardous liquids.
Write for complete details.
LIAUIPOMETER CORP. EST A BLISH ED 1841
I MAM A\/C I n u A ICI Au n / * iv i PROCESS DIV. 51 E. 42 ST., N. Y. C. Phone: M U. 2-5132

September, 1944
The V a p o r R e co ve ry S y ste m s C o m p a n y
is p ro u d to be a vital link in the su c c e ss
ful o p e ra tio n o f the C ities S e rvice 1 0 0
O c t a n e A v ia t io n G a s o lin e Plant a t Lake
C h a rle s, L o u isia n a . A s t h o u s a n d s of
u s e r s k n o w , " V A R E C " C o n s e r v a t io n
U nits a ssu re p o sitive protection to v e s
sels sto rin g v o la tile liq uid s. " V A R E C "
d e s ig n a n d co nstru ctio n h a v e bu ilt a
re p u ta tio n fo r d e p e n d a b ilit y . This g u a r
a n te e s " V A R E C ’S " a b ility to m atch
p ro m ise w ith p e rfo rm a n c e a n d to meet
y o u r m ost rig id sp ecificatio ns.
"V A R EC " A PPRO V ED IN S T A L L A T IO N S

V A R E C " C onservation vent unit, here illustrated, is listed
by U nd erw riters’ Laboratories, A p p ro ve d by A ssociated
Factory M u t u a l’s Laboratories, A p p ro ve d by U. S. Coast
G u ard (form erly U. S. Bureau of M a rin e Inspection and
N a v ig a tio n ), A p p ro ve d by L lo yd 's Bureau of S h ip p in g ,
an d A p p ro ve d by Am erican Bureau of S hip p ing.
Lake Charles Installations include "V A R E C " Conservation

Vent Units, Flame Arresters, Free Vent Units, Weather-
hoods, Hillside Flanges, Gauge Covers, Safety Tank
Winches, Swing Line Nozzles, Swing Joints, Swing Pipe-
Clamps, Sheave Brackets and Tank Gauges.
Puce Setter THE V A P O R R E C O V E R Y SYSTEM S C O M P A N Y

C o n s u l t a n t s . D e s i g n e r s & M a n u f a c t u r e r s o f C a s C on tr ol & T a n k E q u i p m e n t
Since 1928
Branch Offices ând Stocks carried at

New York City— New Orleans, La.— Houston. Texos— Tulsa, Okla.
Agencies Everywhere
INEERING CHEMISTR Y Vol. 36, No. 9
S e r v ic e
for la b o ra to rie s exp an d in g K E E P S ’ EM

to handle Army-Navy training
LIVING!
T R I-C L O V E R T ransfusion Valves w ith
p r e c i s i o n - f i t t i n g p a rts ( t w o t e n - t h o u
sandths o f a n inch accu racy) con trol b lo o d
flo w b y h o ld in g the vacuum in Transfuso-
V a c B ottles i n co lle ctin g b lo o d fo r plasm a
and fo r transfusion. BLOOD PLASMA IS A LIFE
SAVER OF WORLD WAR II
O th e r T R I -C L O V E R p recision -m ad e
item s co n trib u tin g t o saving lives are Wounds in this war* ore greater
— but loss of life is definitely
P um ps, V alves and Fittings fo r P enicillin less, due largely to the use of
and S erum ; A n ti A ircraft P rojectiles; Land Blood Plasma.
M in e Strikers; A n c h o r C hain L inks fo r Upper Left: Valve used in Blood
all types o f b oats; as w e ll as fo r m u n i ■ Transfusion
tion s fo r o ffe n siv e warfare. Above: Infusion of Plasma
apparatus.
T R I -C L O V E R sk ill and expert w ork
m anship d e v e lo p e d w h ile m ak in g V alves,
Sanitary F ittings and M ilk Pum ps fo r the
D airy and F o o d Industries is a idin g
A m erica's E ffort.
A n d — T R I -C L O V E R is also p ro d u cin g
its fu ll q u o ta o f these supplies f o r the
D airy, F o o d and C hem ical Industries.
O rd er in advance w h en p ossible.
TRI-CLOVER
T yptcal A ib eren e-eq u ip p ed co lleg e laboratory.
MACHINE CO.
K E N O S H A ,
W ISCONSIN “th e G zw p & ESi
Currently we are active in fitting-out college

laboratories to meet tjie influx of students
under the Arm y-Navy war training program.
Whether you require a few additional pieces
of equipment, or must m ove to new quarters,
or are planning a complete new laboratory, we
are equipped to serve you. Alberene Stone’s
durability and econom y have been proved for
more than 50 years in the country’s leading
educational, industrial and research laborato
Sifters, Crushers, Cutters, Dry and liquid Mixers,
ries. Our specialized, technical experience in M ills, G rinde rs, P u lve rize rs, C onveyor System s,
Complete Installations.
the design, fabrication and installation of table-
tops, fume hoods, sinks, tanks, shelving, is at
your immediate service. Reasonably prompt The handling equipment construction "kno w
how” of the Mercer Engineering Works, Inc.,
deliveries can be made. Clifton, N. J . .. „ The more than 40 years pro
cessing equipment experience of Robinson
Mfg. Co., Muncy, P a .. . . All are embodied in
and represented by
Our engineers are experienced in designing, re M E R C E R -R O B IN S O N , CO ., IN C .
assembling and installing laboratories o f every 3 0 C H U R C H ST., N E W Y O R K 7, N. Y.
type. Call on us to d o the com plete jo b . Please

address A lberene Stone Corporation o f Virginia,
419 Fourth Avenue, N ew Y ork, 16, N. Y . Quarries
and M ills at Schuyler, Virginia. Sales offices in
HANDLING
principal cities.
LBERENE STONE f# EQUIPMENT

A LABORATORY EQUIPMENT
Trailer Trucks (All Types) Wheel Tractor Cranes
(3 to 7 ton) Fork Lilt Trucks, Lift Platforms, Hoists,
Live Skids, Wheels, Casters. <
_ t
■
iia g — i
NALISIS
5
wl 'i f '
¡sOP&r' ■■.
EAN ASSURED . B A S IL IT Y
.........
S Ü ! ■ '• : ■■ i-J C i d
• In the Ladish Metallurgical Labora

tories, trained technicians w ork with
the finest equipment available to insure
that your Ladish Quality Forged Steel
Flanges w ill d o the jo b for which they
are intended . . . will have assured
weldability.
The heat identity code, permanently

V t 'S 'i A ^
stam ped on every Ladish Quality
LA D isrr. Forged Steel Flange, means that cer
i Ç O N T B O L L E D Q U A L IT Y
tified reports o f the w ork o f the Ladish
Metallurgical Engineering Laborato
ries attesting the chemical and physical
properties o f the steel are always avail
able upon request.
s y e e
Steel identity stamped on each flange
w a s pioneered by Ladish and has been
L A D I S H D R O P r O B O E CO
a Ladish standard for m any years.
L A D IS H DROP FORGE CO.

CUDAHY • WI SCONSI N
TO MARK PROGRESS M IL W A U K E E SU BU RB
PERFORMANCE
AND ECONOMY
. . . two important factors of
PRACTICAL DESIGN
A D ow ningtow n designed unit must not o n ly
work but it must give trouble free, reliable per
formance over a reasonably long period, and it I* » ™ * ™ ?
must be econom ical both in first cost and in
maintenance expense.
H ere is an example of D ow ningtow n efficient y&Bffo S lS I

design, the purpose of which is to break up the ’— ---------- '
flow of liquids outside the tube bundles. This
result is attained b y utilizing two separate sets of sealing strips
C ross S e c tio n v ie w o f H e a t E x ch a n ge r sh o w in g S e a lin g Strips for
on each side, in order to eliminate thoroughly any possibility of b la n k in g o ff p o te n tia l b y -p a ssin g where tube b u n d le d o e s not
en tirely fill the shell.
channeling of the shell side fluid.
Our heat exchanger K now le d ge is backed b y over thirty years

DOWNINGTOWN IRON WORKS
consecutive experience in the field of pressure vessel fabrication. Repeat
orders in H e a t Exchanger Equipm ent were over 70 per cent of D ow n in gto w n ’s
DOWNINGTOWN, PA.
1 9 4 2 -4 3 business. HEAT E X C H A N G E R S
NEW JOBS
TOUGH JOBS
IIIT E R PRESSES
IN
FILTRATION
CLARIFICATION SHARP
THICKENING
WASHING
EXTRACTION
TEMPERATURE
CHANGES
Portable filtration unit,
all stainless steel, for
filtration and washing; GLASS TANKS by "Pittsburgh" are made of specially-tem
complete with pump and pered glass. It's 4 times as strong as ordinary glass. And
piping. Built in any size it can withstand the shock of sharp temperature varia
from laboratory to pro tions.
duction scale. GLASS TANKS are imprevious to almost all acids, alkalis,
chemicals, liquids. They can even handle hot chromic
acid.
If you're developing a product the filtration charac And you can get glass tanks now . . . promptly . . . in
teristics of w hich have not yet been determined; if types, sizes and colors to suit your needs.
you're stymied in your process because the material
makes filtration, clarification, washing or extraction S en d th e c o u p o n f o r c o m p le t e fa c ts on —'
G L A S S T A N K S B Y " P IT T S B U R G H "
slow or uneconom ical with your present equipment • —
Pittsburgh Plate Glass Company
Let the Shriver Laboratory show you — free — how 2268-4 Grant Building, Pittsburgh 19, Pa.
a Shriver Filter Press, versatile as to application, Please send me, without obligation, free literature on Glass
Tanks by "Pittsburgh. ”
design and capacity, can b e made to tick your prob
lem — no matter how tough. Write for catalog. Name................. ...........................................................
Address. .............................. ...............................
T. SHRIVER & COMPANY, Inc. O '

8 6 2 H a m ilt o n St. • H a r r is o n , N . J .
PITTSBURGH PLATE GLASS COMPANY
September, 1944
LA Y O U T AND
DESIGN OF P LA N T
Those w ho think of chem ical plant design in terms of process
research, developm ent and- recom m endations should see the
CHEMICO Engineering Department at work . . . a large corps
of skilled specialists w ho develop plant layouts and design
structures and equipm ent for most effective and efficient
operation.
It is this phase of CHEMICO service that results in astonishingly
low labor and m aintenance costs of CHEMICO plants . . . for
it provides automatic operation, centralized control, elimination
of w aste motion, and high overall plant efficiency.
The performance of the m any CHEMICO plants all over the
world testifies to the skill of these layout and design specialists,
and is your assurance of the reliance that can b e placed on
the authoritative recom m endations of the CHEMICO engineers.
C H E M IC A L C O N S T R U C T IO N CORP.
30 ROCKEFELLER PLAZA • NEW YORK 20, N. Y.
European Representatives, Cyanam id Products, Ltd.,
Berkhamsted. Herts, England Cables, Chomiconst, New York
The 6 - Point Chemico Service

1. A n a ly s is o f In d iv id u a l p la n t c on d itio n s
2. Se le c tio n o f p ro ce sse s
3. L A Y O U T A N D D E S IG N O F PL A N T
4. Erection o f p la n ts
CHEMICO
5. T ra in in g a n d in itia l s u p e r v isio n o f crew
6. G u a r a n te e o f re su lts
SoftstrÆ,
CHEMICAL PLANTS ARE

PROFITABLE INVESTMENTS
Vol. 36, No. 9
W e A r e Doing Our
YOU GST fo r UNCLE
We'll Do the Same for
LONGER O u r business is b uild in g pum ps. O u r
O N L Y p o licy is to build fine pumps.
W h e n you install pum p s " b y A u ro r a ,"
SERVICE
you know that lasting efficiency and
d e p e n d a b ility are at work on yo ur
im portant liquids handling jobs.
Aurora Deep W e ll Turbines
for all conditions— 4 " to 2 4 "
WITH ROPER PUMPS

N S A Aurora
...and h e r e a r e th e r e a s o n s w h y Type A D Hor. Split Case,
Centrifugal
Sum p Pump
In a R op er Pump, the only m ovin g parts are tw o equal size Two Stage Centrifugal
p u n v in g gears. O perating in a case with p roper clearance,
these gears actually "float” in operation . . . there is no per Type O D Hor. Spllt-Case Double Suction
ceptible wear on case or gears. Because R op er pumps are so Single Stage Centrifugal
sim ple in design, and because internal pressure is equalized
at all points, you can count on a R op er for sm oother, quieter A P C O TURBINE- M J O M J
TYPE PU M P S— the m S B l
operation and lon ger service. All sh ock and thrust is ab simplest of all pumps. Ideal for A P C O Horizontal
so rb e d ' by the slidin g joint w hich connects the gears. small capacity, high head duties» Condensation
A n oth er reason for the popularity o f R op er Pumps is the Silent, compact and lasting. Return Unit
fact that gears and bearings Type G M C Close-
can be inspected w ithout dis C ou p led Centrifugal W rite fo r
turbing p ip in g o r pow er. CONDENSED CATALOG
In every in s ta n ce , R o p e r H D K & M ü 'í í OR
Pumps mean greater efficiency
and greater econom y.
C A T A L O G IM
SW T5
Type G G U Side Suction A P C O Single Stage
Sin gle Stage Centrifugal Turbine-Type
D ISTRIBUTO RS IN P R IN C IP A L C ITIES
Write fo r
Catalog 934 l^ crtcu iM P p u y u x ¿.
GEO. D. ROPER CORP., ROCKFORD, ILL.

64 Loucks Street, A U R O R A , IL L I N O I S
OT TER
E n g in e e re d p rim a rily fo r use in heaters, boilers, air-co nd itio n
ing a p p a ra t u s a n d w here it is nece ssary to im m erse the
unit in the m edium to be controlled. W ith protective
¡unction box a n d enclosed d ial, this sensitive
T herm osw itch a d d s to both efficiency a n d looks
o f m anu factu re d p ro d uct o r process.
SARCO
No. 9
W it h ± 1 / 1 0 ° s e n
s itiv ity and a range
from m in u s 100°
THERMOSTATIC to p lu s 6 0 0 ° F. the
T h e r m o s w it c h is
to d a y 's m o st e f f i
STEAM TRAPS cien t h ea t
cold) control.
(and
R a te d 10a. 11 5v.
W h e n y o u w ant m axim um heat extraction a n d
q u ick venting, the S a r c o N o . 9 is the steam tra p to
W r it e fo r new C a t a l o g
use. It w o rk s at all p re ssu re s w ithin its ra n g e w ith
out c h a n g in g seats. It is sm all, in exp en sive, e a sy
to install a n d cannot fre eze. S iz e s from Vz to 2 ", T h e Im m e rsio n T h e rm o sw itc h
0 to 2 5 0 lbs. A s k fo r C a t a lo g N o . 2 5 0 .
SARCO SARCO com pany,

475 fifth Avenue, New York 17, N. Y..
in c.
SAVES STEAM SABCCCmOA, LTD., es Richmond S tW .,T 0H0NT0, OHT.

' ■■ '..181
T y p i c a l A m e r i c a n Ixi&ensgiiy
m ad e p a ssible . .
The valves shown ab o ve, fo r use in caus+Ic process lines in o r e

p rep aratio n plants w ere o rig in ally designed w ith c a st steel bodies. W h e n it
was im possible to get the more than 200 tons o f steel castings required fo r
them , the valve s w ere re-designed so th a t the bodies could be fa b ric a te d
from steel tu b ing .
H o w e ve r, the tubing w as not prom ptly availab le and the Ja y - C e e o rg ani
zation m ade it from steel plate and then produced the bodies from the
fa b ric a te d tu b in g . The only steel castings used w ere th e yokes and the
sm allest size bodies.
E xce p t fo r th e few steel castings, these valves, ranging in sizes from 6 " x 4 "
to I 8 " x l 6 " w e r e com pletely fabricated, machined, assem bled and
tested by us.
Produced by "J A Y -C E E "

If it's m a d e of I R O N or ST EE L
A L U M IN U M PLANTS
d r a w a p ic tu r e of it — we'll build it. H oppers & Chutes Annealing Baskets
Stacks & Breeching Special M achinery
POW DER PLANTS

Tanks Fabricated Piping
Pressure Yessels Special Machinery
OTHER PLANTS
Structural Steel Iron C a stin gs
Steel and Plate W o rk Special Machinery
JO H N S O N C IT Y FOUNDRY & MACHINE WORKS, J O H N S O N C IT Y , T E N N E S S E E e st. issal
GRAY IRON and ALLOY IRON CASTINGS STRUCTURAL STEEL — MISCELLANEOUS

and ORNAMENTAL IRON WORK - SPECIAL MACHINE WORK
120
|| F. B. Porter R. H . Fash
T H E B IR D -A R C H E R C O . ■ B.S., Ch.E., Pres. B.S., Vice-Pres.
Established 1891
E n gin e e rs a n d C o n su ltan ts T H E F O R T W O R T H L A B O R A T O R IE S
o n W a t e r C o n d it io n in g Problem s C o n su ltin g , A n a ly t ic a l C he m ists and j SAM UEL P. S A D T L E R & S O N , IN C .
C h e m ica l E n gin e e rs
i Surveys Plant Studies Analyses
W hen you have propositions in the Southwest
j 4 0 0 M a d is o n A v e . , N e w Y ork, N . Y. consider using our staff and equipment to save C o n su ltin g and A n a ly t ic a l C hem ists
time and money, C h e m ical E n gin e e rs
i 8 2 8 K M o nroe St., Fort Worth, Texas
1
W e render many chemical services
F R O E H L IN G & R O B E R T S O N , IN C . for industrial clients.
BORN E N G IN E E R IN G CO. Established 1881
Petroleum and N atural G a s C o n s u ltin g and A n a ly t ic a l C hem ists | 210 S. 13th St., Philadelphia 7, Pa.
an d In sp e c tio n E n g in e e rs
i A n a ly s e s an d Tests | Richmond Virginia "N o th in g Pays Like Research"
11 A p p r a is a ls
D e sig n & C o n stru c tio n E. W . D. H U F F M A N , Ph.D . F O S T E R D . S N E L L , IN C .
In sp e c tio n Se rv ic e M ic ro a n a ly tic a l Laboratories O u r staff o f chemists, e n gin e e rs and
b a cte rio lo gists w ith la b o ra to rie s for
E n g in e e rin g R eports i O rganic, Inorganic — Quantitative, Qualitative I analysis, research, p h ysical testing and
Precision Analyses Special Analytical Problems b a c te rio lo g y are p re p a re d to re n d e ry o u
Tulsa, O k la h o m a E v e ry Form o f C h e m ic a l S e rv ic e
!j M a je s tic Bldg. D e nve r, C o lo r a d o
! 305 Washington Street Brooklyn, N. Y.
Patent Law Offices

L A N C A S T E R , A L L W IN E & R O M M E L
THO M AS F. H E A L Y
C H E M IC A L D E V E L O P M E N T S
Booklet — "G eneral Information Concerning B.S. C h e m ical E n g in e e rin g
C O R P O R A T IO N Inventions & Patents" and "Fee Schedule"
sent without obligation. R e gistered Patent A tt o rn e y
Established 1915 A tto rn e y -A t-L a w
C on su ltin g Engineers
Patents — C op yrigh ts — T rad e -M arks
C H E M IC A L PATENT PROBLEM S
Suite 446, 815 — 15th St., N. W.
Research D e ve lop m e n t and Washington 5, D. C. M u n s e y B ld g. W ash., D . C,
Plant D e sig n
L A W A L L & H A R R IS S O N
EDW ARD T H O M A S
1773 Sp rin gfie ld Street D ayton 3, O h i o Biological A ssays — Clinical Tests
Chemical — Bacteriological — Problems | Attorney-At-Law
Organic Synthesis Registered Patent Attorney
Pharmaceutical and Food Problems
Write for Descriptive Brochure
Research C hem ical Patent Problem s
" When Peace Comes"
1921 W a ln u t St. Ph ila de lp h ia 3, Pa.
Woolworth Bldg. N ew York, N. Y.
E D W A L L A B O R A T O R I E S , IN C .
L O U I S E. L O V E T T U N IF IE D L A B O R A T O R I E S , IN C .
E. W . Lowe, Ph.D. W . S. Guthmann, Ph.D.
W. B. Hendrey, Ph.D. W . M . Malisoff, Ch.E., Ph.D., Pres.
C on su ltin g Engineer
Process Development, Research and Umpire I
A n a ly t ic a l and C o n su ltin g C h e m ists i Biological Assays, Staff specialists on Hormones,
Research o n Part-time o r F e llo w s h ip Basis Rayon and Cellophane Process, Plant Design. Reports. Vitamins, Synthetic Drugs, the A lc o h o l Industry,
Commercial Applications of Osmosis the Sugar Industry, and the Petroleum Industry,
732 Federal Street C h ic a g o 5, III. j Chemical Warfare.
Descriptive booklet No. 21-J available M adison, O h io
1 7 7 5 Broadw ay N e w York
C . L. M A N T E L L H E R B E R T W A T E R M A N , Ch.E., Ph.D.
G U S T A V U S J. E S S E L E N , IN C . C on su ltin g Engineer C on su ltin g Engineer
Electrochemical Processes and Production Problems

Plant Design Equipment and Plant Design
Research and Process Development
Chemical Research 601 W . 26th St., N e w Y ork, N . Y. 2 10 0 Roselln Place, Los Angeles, Calif.
Telephone O lym p ia 3780
and Developm ent
M E T C A L F & E D D Y T R U M A N B. W A Y N E
E n gin e e rs
C o n su ltin g C h e m ica l E n g in e e r
857 Boylston Street Boston, M ass. Airfields
Emulsion Processes and Problems.
Investigations Reports Design O rganic Polymers & Resins.
Supervision of Construction and Operation Chemical Synthesis and Production.
"T o d a y's Research Is Tomorrow's Industry" Management Valuations Laboratory I Patent Matters, Surveys & Reports.
Bankers Mortgage Bldg. Houston, Texas
Statler B u ild in g Boston, M a ss. Telephone Fairfax 7708
F R A S E R -B R A C E Harvey A . Seil, Ph.D. Earl B. Putt, B.Sc.

E N G I N E E R I N G C O . , IN C . S E IL , P U T T & R U S B Y T H E W EST P O R T M IL L
Incorporated
D e sign , co n stru ctio n and C on su ltin g Chemists W estport, Con n .
Specialists in the analysis of
installation o f com plete Foods, Drugs and O rganic Products
plants a n d projects 16 E. 34th Street, N ew York, N. Y.
L a b o ra to rie s a n d Testing Plant of
Mechanical, Heavy Industries, Shipbuilding, T elephone— M U rra y H ill 3-6368
| ilydro-EIectnc Development, Power Plants The D O R R C O M P A N Y , Inc.
C hem ical and Helming Plants, Process Industries, i C O N S U L T IN G
i Metallurgical Developments and Processes, E x - j
ii plosives, Plastics, Water Supply and Treatment Rubber Technologist
Sewago and Industrial Wastes Treatment. |' Chemical, Industrial, Metallurgical and Sanitary
N a t u ra l a n d S ynth e tic R u b b e r
Engineers
RE P O R T S— A P P R A IS A L S — R. R. O L I N L A B O R A T O R I E S
Complete Rubber Testing Facilities Consultation — Testing — Research — Plant Design
C O N S U L T IN G Established 1927
1 0 East 40th Street N e w Y o rk 16, N . Y . j P. O . Box 372, A kro n 9, O h io Descriptive brochure, "Testing that Pays Dividends" '
Telephones: H E 3724, FR 8551 upon request
J............. ............. .. ....- '................

September, 1944
..2 processing operations in a single cylinder

• B artlett-Sn ow d ry er-coolers consist o f a to 110°F. or less, at the p oin t o f discharge.
Bartlett-Snow Style "H ” in d ire ct-d ire ct heat Two fans, o f ex ceed in gly lo w horsepow er
dryer, and a Bartlett-Snow rotary air cooler com requirement, are employed, both discharging to a
bined into a single shell. Material is introduced single dust co lle cto r. A m inim um o f fu el is
into the feed end o f the dryer,— and gives up all required,— and because the drying and cooling
but a very sm all am ount o f its m oisture. The sections are built integrally the power requirement
material then passes through a specially designed is only slightly more than would be needed for
lock —w hich—without involving any m oving or either the dryer or cooler used singly. But send
complicated parts—permits the material but not for a copy o f our Bulletin N o. 89- It gives com
the hot air and flue gases to pass into the "cooler” plete details about Bartlett-Snow dryers, coolers,
end. Here, in the cooling section, the material calciners, autoclaves, kilns and other heat process
loses the balance o f its moisture, and also enough ing equipment,— and contains m uch technical
sensible heat to b rin g the temperature down data o f interest to engineers and operating men.
the C. 0. BARTLETT & S N O W co.

6 2 0 7 HARVARD AVENUE • CLEVELAND 5, OHI O
Engineering and Sales Représentatives in Principal Cities
D RY E RS • C A L C I N E R S • K I L N S • P R E S S U R E V E S S E L S
• C O M P L E T E M A T E R IA L H A N D L IN G F A C IL IT IE S T O M E E T A N Y R E Q U I R E M E N T
O N E C O N T R A C T • O N E G U A R A N T E E O F S A T IS F A C T O R Y P E R F O R M A N C E • U N I T R E S P O N S I B I L I T Y
A L P H A B E T I C A L L I S T OF A D V E R T I S E R S
Acme Coppersinilhing & Machine C o 123 Foxboro C o .......................... 26 Patterson Foundry it Machine C o........... 1
Alberene Stone Carp.............. 114 Fraser-Brace Engrg. Co., In c................... 120 Pennsylvania Salt M fg. C o ..................... so
Aluminum Co. of Am erica ............. 57 Froehling & Robertson, Inc.................... 120 Picker X -R a y Corp................................ 66 G7
Aluminum Ore C o .. ...................... 43
Pittsburgh Lectrodryer Corp.................. yg
American Hard Rubber C o ....................106-107 General American Process Equipm ent 76 Pittsburgh Plate Glass C o. ................. ng
American Optical C o ............................. 17
General American Transportation C orp.... 101 Porter Co., Inc., H. K ............................ 2
American Smelting & Refining Co............ 32 General Electric C o .................. ; ........... 3 0 -3 Í Powell Co., W m .............. 74
American Telephone & Telegraph C o 93 Girdler Corp......................................... g ¡9 Pressed Steel Tank C o ........................... . 5$
Andrews Lead Construction C'orp.......... 32 Graver Tank & Mfg. Co., Inc................. 40 Pritchard it Co., J. F ... 97
Armstrong Cork C o ... .......................... 92 G riscom-Russell C o ............................... 85 Proctor it Schwartz, Inc......................... 125
Aurora Pump C o .............................. 118
Gum p Co., B. F ........................ 10 s Professional Directory............................ 120
Proportioneers, In c ................................ 126
Babcock & Wilcox Co. . .......................... 64 65 Hardinge Co., In c.................................. 47 Pulverizing Machinery C o ...................... 102
Badger & Sons Co., E. B 48- 49 Haynes Stellite C o ’. . . . * .............. 56
Baker Perkins In c.. .................. 75
Healy, Thomas F ............................... .. 120
Barco M fg. Co...................................... 72 Raymond Pulverizer D iv ........................ £2
Hicks it Son Co., Inc., S. D .................. 112
Barnstead Still & Sterilizer .Co., In c 28 Reed Roller Bit C o ...................... 10
Hills-M cCanna C o ................................. 44
Bartlett <fc Snow Co., C. O . ........... 121 Republic Steel Corp............................... 35
Hooker Electrochemical C o ..................... 73
Bausch <fc Lomb Optical Co ..................... 68 Revere Copper it Brass, In c .................... 54
Houdry Process Corp.......................... ;. 39
Bird-Archer C o ........... 120 Richardson Scale Co.............................. 124
Huffman, E. W. D ................................... 120
Blaw-Ivnox C o .. .......................... 27 Roper Corp., Geo. D .............................. n$
Blickman, Inc., S ................................... 8 Ross Heater <t M fg. Co., In c . . A ............. 24
Industrial Brownhoist C o rp ..................... 112
Born Engineering C o .............. 120 R - S Products C orp ................................ 14
Brighton Copper W o r k s............... 104 James Mfg. Co., D. 0 ............................ 15
Bristol Co...................................... 36 Jenkins Bros......................................... 2I
Sad tier it Son, Inc., Samuel P ................. 120
Brown Fin tube C o ................................. 87 Johns-Mnnvillc Corp.............................. 4
Sarco Company, In c.............................. 118
Buffalo Forge C o . . .. . . .............. ,•........... 41
Johnson C ity Foundry it Machine W orks.. 119
Seil, Putt it ltu sb y................................. 120
Shriver it Co., Inc., T ............................ 116
Calgon, I n c . . . . . . . . . . . . . . .................... . 40 ICidde it Co., Inc., W alter....................... 23
Snell, Inc., Foster D ........................... 120
Carbide & Carbon Chemicals Corp..........52:103 Kimble Glass Co.............. .................... 91
Sperry it Co., D. R .......................... 34
Carborundum C o ... ......................... 45 Kinney M fg. C o ........................... 104
Stokes Co., F. J.................... 22
Celanese Corporation of America..............88-89
Struthers Wells Corp.............................. 25
Chapman Valve Mfg. C o ........................ 84
Ladish Drop Forge C o ............................. 115 Sturtevant M ill C o ................................ 100
Chemical Developments Corp................. 120
Lancaster, Allwine it Rom m el................... 120 Swenson Evaporator C o .......................... 12-13
Chemical Construction Corp... ............... 117
Lapp Insulator Co., In c............. 01
Chicago Bridge & Iron C o ...................... 105
La W all it Un rrisson.................. . ......... . 120
Clark Bros. Co., In c ..................... 69 Taber Pump C o . ................. 10S
Leeds it Northrup C o ....................... . 29
Combustion Engineering Co., Inc............ 82 Taylor Forge it Pipe W o rk s. ....... 70
Lewis & Co., Inc., Chas. S ...................... 78
Corhart Refractories Co., Inc. ..... 90 Taylor Instrument C o ...................... 62-63 '!
Link-Belt C o . . . . . . ................................ 7
Crane C o ... ........................................ 40 Thomas, E d w a rd ................ 120
Liquidometcr Corp., T h e ........................ 112
Lovett, Louis E . ............ 120 Tri-Clover Machine Co .......................... 114
Despatch Oven Co.. . ........... ............... 10 Tube Turns, In c ................................... n
Devine Div., J. P.................. ................ 2
M an tell, C. L ......................... 120
Downingtowh iron W orks. . ......... 1 iff
Marblehead Lime C o ............................. 18 Unified Laboratories, In c.. ......... 120
Dracco Corp. 108 '
Master Electric- C o ............................. 71 Union Carbide & Carbon Corp.52:56:79:95:99:103
D u Pont de Nemours & Co.,Inc., E. 1____ 33
U. S. Industrial Chemicals, In c ..................19-20
Mercer-Robinson Co., Inc....................... 114
Duriron Co., Inc.......................... 59
U. S. Stoneware C o................................ 12S
Merco Nordstrom Valve C o .................... 37-3S
Metcalf it E d d y . ............................. 120 Universal Oil Products C o ...................... 53
Edwal Laboratories, In c ......................... 120 Midwest Piping Co., In c ........... 81
Edward Valve & M fg. Co., In c............ 50 Monsanto Chemical C o .......................... 55
Vapor Recovery Systems C o ................... 113
Electro Metallurgical C o ........................ 95
Vogt Machine Co., Inc., H enry................ 109
101 liott Com pany ......................... 98
Nash Engineering Co............................. 94 Vulcan Copper it Supply C o ................... 127
Esselen, Inc., Gustavus J....... ....... 120
National Carbon Company, In c .. ............7 9 :9 9
Waterman, Herbert... ......... ; .......... 120

.Fairbanks, Morse it C o .......................... 60
Oldbury Electro-Chemical C o ................. 102 Wayne, Trum an B .......................... 120
Fenwal, Inc........................................... Hg
Olin Laboratories, R. R .......................... 120
Wellington Sears C om p an y ............. S3
Fort Worth Laboratories........................ 120
Oronite Chemical C o ....................... 77
Westport M ill (Dorr Co., In c .)................ 120
Foster Wheeler Corp............... 110 i l l
Owens-Corning Fiberglas Corp................ $6 W hiting Corp......................... 12-13
lilla :G«
1 REA S OWH Y . . .
ACME patented MULTI-JET BAROMETRIC CONDENSERS
OPERATE WITH MAXIMUM ECONOM Y/
A C M E PATENTED
MULTISET BAROMETRIC
CONDENSER
Designed, fo r vacuum processes where the condensate is not

SECO ND THIRD
EFFECT EFFECT
retained, the Acme Patented Multi-Jet Barometric Condenser

VAPOR
PIPE
combines the economies of both wet and dry operation. This

patented unit offers the following exclusive features:
, 0 1 COUNTER-CURRENT FLOW that produces higher vacua
(hitherto feasible only with dry operation).
0 CONVERGING JETS that expel the incondensible gases
from the top of the condenser without the use of vacuum
HOTW ELL j
CONDENSER FOR EVAPORATORS pump or ejector (the principal advantage of wet operation).
VAPO R
A C M E PATENTED
© A PATENTED SEAL that prevents the escape of vapors until
CWM HEADER
PIPEs
MULTI-JET BAROMETRIC
CONDENSER they reach the top of the condenser, thus making it possible
•«05PHERIC
H
EAD
ER \ to combine advantages (1) and (2) In one piece of equipment.
UR HEADER ' BAROMETRIC
»4 COOKERS LEG
© POSITIVE CONTROL of. both involute and converging jets
that attains maximum condensation at a minimum expendi
M A N H O LE
ture of water, even under fluctuating loads.

C § LO W INITIAL COST of equipment and installation.
SPARGER
STEAM
HEAOER
© . SIMPLE DESIGN includes no moving parts, will not clog,
HOTWELL and requires Virtually, no maintenance cost. .
© CONVENIENT OPENINGS for quick, easy inspection.
VACUUM CO O IIN G SYSTEM
' -V M
i .B u lle tin M j - 4 4 VAcm e P a te n te d M u l i i - j e * Yj '
A C M E PATENTED
MULT1-JET BAROM ETRIC
is C O N D E N SE R f
V AP O R
PIPE/
ÎCULATING
PUMP
J la it- M ittuie. rf-laiJteA.
WEIGH
F R O M TH E E M f O r S DESK
<TTiie ground troops are making glorious history in France and
ACCURATELY Í PROPORTION • will continue to do so until Germany crumbles. The RAF
and our own Air Force, long holding the spotlight, are now some
what overshadowed in news reports by the armored divisions
and the infantry; but we should not forget that the devastating
BLEND attacks on German production have made success possible.
Ib is is especially true of the German oil position. According to
Leo T. Crowley, Foreign Economic Administrator, Germany's
output of petroleum products, as recently as April of this year,
was still high enough to permit stock-piling. Today production
BULKY AND LUMPY CHEMICALS does not meet minimum needs even though these needs have
been drastically cut by elimination of the least essential war uses.
Meanwhile, with Romania now thoroughly penitent, our heavy
bombers are switching their attention from the Ploesti area to
hitherto untouched refineries in Poland, Slovakia, and Hungary.
Eliminate the element of On August 20 we dropped calling cards on the former Vacuum Oil
human error in propor refinery at Czestochowa, the I. G. synthetic oil and rubber plant
tioning several chemicals at Oświęcim in Polish Silesia, and a refinery at Dubova, Slovakia,
for mixes b y w eighing near Budapest. H ie three plants are said to have an annual
each material accurately capacity of 287,500 tons. On the same day the Petroleum Ad
ministration for War authorized September oil production at a
w ith two or more Rich rate of .5,051,000 barrels daily, the highest output ever certified
ardson Conveyor Feed to the various oil-producing states. If Napoleon were alive, he
Automatic Hopper Scales. would have to modify his famous statement about which side
The continuous, auto
RICHARDSON CO N V EYO R FEED
God is on in war.
matic operation of a bat ★ It is not generally known that the State Department is spon
AU TO M ATIC HOPPER SCALES
soring a special mission abroad to study the possibility of

tery of these scales will increase production of the restoring trade to commercial channels. The group will first
finished mix — prevent spoilage — save time. When the conduct a survey in North Africa and probably will extend activi
proportions of mixes are changed, any one of the battery ties to other areas. On this committee is John L. Gillis, vice
of Richardson Scales may be cut out of the line at will. president of Monsanto Export Company, but currently represent
An interlocking system scops all scales if material in any ing the Foreign Economic Administration. It is possible that this
mission is hut the forerunner of others.
one runs low. ★ Additional reconversion a few hours after Germany’s collapse
The accuracy of Richardson Scales helps to assure product is possible, according to L. It. Boulwnre, Operations Chairman
of War Production Board. He reports that WPB has a full pro
uniformity, for accurate weighing makes possible close gram set up to remove Ł and M orders from daily business
control of operations. The exactness of weighing me routine, and is awaiting the green signal.
chanically and autom atically prevents spoiling of mixes. ★ One of the questions most frequently asked in Washington
today is whether distillers will get another “ holiday” soon. Just
There are Richardson units for weighing, proportioning how soon is “ soon” ? Official circles are mum on the subject,
and blending any bulky or lumpy chemical accurately to but many are willing to wager that, come early spring, there will
any predetermined w eight; adaptable for single or be not only a holiday but considerable easing of restrictions.
multiple batch weighing. ★ 1 his is not a book review column, but A. Cressy Morrison’s
“ Man Does Not Stand Alone” and Ben Brooks’ “ Peace, Plenty
In the wide Richardson line, there are scales designed for and Petroleum” are on the “ must” list. Your editor became so
powdered, ground, crushed and lumpy materials such as enthralled with “ Man Does Not Stand Alone” that he finished
the last ten pages amid the grime of South Station in Boston
salt, soda ash, potash, sodium sulphate, alum, borax, while suburbanites pushed him around and made uncomplimen
barytes, ores and similar materials; other units for tary remarks about individuals who block traffic. Your editor
weighing and proportioning liquids; others for weighing has maintained a dislike for reading aloud acquired in childhood,
and bagging products before shipment. but he could not resist the temptation of regaling his staff and
family' with some of Ben’s salty' (and oily) philosophy.
For complete details on the Richardson Conveyor Feed ★ Output of rayon y'arn
vY t ,
Automatic Hopper Scale, send for Bulletin 2140-1. in the second quarter of
this year totaled 135,-
700,000 pounds, accord
ing to Rayon Organon, &
new record. A slight de
cline is noted in the
RICHARDSON SCALE COMPANY second quarter from the
CLIFTON, N. J. first quarter in the pro
Atlanta N ew York Chicago M inneapolis Montreal d u c tio n o f v is c o s e -
Philadelphia Boston Wichita Omaha Toronto euprammonium ya rn ,
San Francisco
but acetate showed an
increase.
124
September, 1944
For tonnage production of wet-solid or

plastic materials we recommend the Proctor depth on a constantly moving conveyor.
Aero-Form Continuous Drying System. The On completion of the cycle the material
continuous operation of this system over is discharged at the delivery end— ready
comes all of the costly disadvantages of for subsequent processing. This system
batch drying. It brings many benefits from makes the ideal link between preliminary
the standpoint of the ultimate quality of and subsequent processing. It takes uni
the product being dried. Last, but far from form quantities, dries the material thor
least, this system makes tonnage produc oughly and quickly, to the desired mois
tion a reality . . . at new low costs. Coming ture content, and automatically passes on
in the form of a filter cake, from a plate uniform quantities to subsequent process
and frame filter, coming from a continuous ing equipment. If you have a product
centrifugal filter or from a continuous which requires tonnage production— it
rotary filter— the product is first pre will pay you to carefully investigate the
formed into the most ideal shape for con advantages of this Proctor Aero-Form
tinuous drying. It is thpn carried through Drying System.
[H E m ic m m o u sT R iE S
m anufiactuted ¿y P R O C T O R & S C H W A R T Z • IN C - P h ila d e lp h ia • P a
’ l n c.
ol* y e d
>en ic ili
/o °P ort W ar f
Cess f u JJy (
t f c S O J in « . <
: ■ 85 <w u
P r e v e « t i 0 il i
? u P p l i e d tu
^ « e s fQr
2 ? § f
r r*«-« :r e a s > ^ A,
' f^ ta J b J e
s f°r - Lng " «

Jypochic
teriliz * tio:
e *t
WRITE TO %PROPORTIONEERS, INC.% 27 CODDING ST., PROVIDENCE I, RHODE ISLAND

^ R E S E A R C H • • •
B A S I C E C O N O M I C P L A N N I N G
P I L O T P L A N T O P E R A T I O N
PLANT and EQ U IPM ENT DESIGN
F A B R I C A T I O N • I N S T A L L A T I O N
A b o v e Photograph courtesy of
INITIAL O P E R A T IO N C A R B ID E and C A R B O N C H E M IC A L S C O R P O R A T IO N
THE VULCAN COPPER & SUPPLY CO., CIN CIN N A T I. OHIO

'J/
Jd a 'g a l ad ¿ o n e , a i ad fle c c -
¿¿/ e ad a / ite c e d iitn g —
S/ygfO ti a
/ if/ a b ig tn eeU tvn i f o t grad , a-e'l
a n t/ / tgait/ d .
by a notew orthy
h a r a c t e r iz e d and almost all acids and alkalies. It is not and surgical tubing, transmission lines
C chemical inertness and a marked
resistance to aging, Tygon rubber-like
affected by oil, soap or perspiration.
Tygon Tubing is, however, attacked by
for beverages, electrical insulation sheath
ing (its dielectric strength is equal to
Tubing is formulated to meet a wide some solvents. that o f rubber), air lines for automatic
range o f physical, chemical and electrical machines (Tygon Tubing may be braid
Physically, T y g o n T u b in g may be
requirements. reinforced for high pressures), channel
made in varying degrees o f flexibility,
Compared to rubber, Tygon Tubing semi-rigid, or rigid. Certain formulations, strips, gasketing, bumpers, bushings,
offers advantages of substantial impor sight glasses, etc.
for example, retain their flexibility at
tance. Unlike rubber, Tygon shows no temperatures as low as 80 °F below zero; Tygon Tubing is now available in
tendency to oxidize, to develop cracks or others are so rigid they may be readily diameters up to l'/i" , in any desired wall
fissures, or to become brittle with age. In turned, sawed or cut. Tygon Tubing, thickness, in clear, black, white, gray,
its flexible forms it has s-t-r-e-t-c-h with while thermoplastic, functions satisfac and in a wide range of colors.
out snap, shows a flex life ten to twelve torily at tem peratures up to 175°F.
times that of rubber. It is highly resistant Certain surgical formulations withstand
to abrasion. autoclaving at 20 lbs. o f steam.
U. S . S T O N E W A R E
Chem ically, T y g on T u b in g shows Applications for Tygon Tubing include S in c e 1 8 6 5
excellent resistance to water, alcohols such widely varying uses as laboratory NEW YO RK AKRON, 0.
TYGON TANK L IN IN G S AND G A S K E T IN G M A T E R IA L S . TYGON S H E E T IN G S FOR S U R G IC A L AND H O S P IT A L U SE , TYGON F L E X IB L E TUBIN I

TYGON M OLDED ITE M S . TYGON L IQ U ID F O R M U L A T IO N S , TYGON D IP P IN G F O R M U L A T IO N S . TYGON TAPE . TYGON M A S K IN G M ATERIAL!

65 PDF

Uploaded by

Copyright:

Available Formats

65 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

65 PDF

Uploaded by

Copyright:

Available Formats

d ü s t r i a l e d it io n SEPTEMBER, 19 4 4

refinery of The Standard O i l C om pa ny ( O h io ) ,

located on site of the original plant built b y

J. D. Rockefeller 7 5 years ago in Cleveland.

EDITO R: W A L T E R J. M U R P H Y IS S U E D S E P T E M B E R 11, 1944 * V O L. 36, NO. 9 • C O N S E C U T IV E NO. 17

for RAW MATERIAL

JOHNS-MANVILLE FILTER AIDS

/ In in tcfip ^ ietcd iu e m o n th ly d iy e d t fjO i ch em ii/ d , ch e m ica l e n fin e e id , a n d

Can You Improve Your Drying Process?

It Might Pay You

Im proper w elding can often be recognized with

tailored to Y O U R gas processing need!

On any gas processing problem, however difficult it

Progress Report No. 6. Bradley Dewey, probably by

During the second quarter of 1944, 209,004 long tons of syn­

Oil, Sand, and the Starry-Eyed. Oil is the modern

UHLUE time in history, has floated her allies to victory on a sea of

Selected Tube Turns Distributors in every principal

TUBE-TURN TRADE M ARK

DOUTHITT GRAY-JENSEN SPRAY DRYERS

Swenson - E n g in e e re d System for

SWENSON EVAPORATOR COMPANY

SWENSON PRO VIDES THIS FIVE-WAY SERVICE..

D esign and Layout M anufacture o f Equipment Test O p eration P e rio d ic C heck-ups

an international petroleum council; here is the crux of

Liquefied Gas, Forward March! Although the war

INDUSTRY AND MEN

T O D A Y ... D. O. James Gears and M en

56 Y E A R S M A K E R S OF EVERY TYPE OF G E A R A N D G E A R REDU CER

D .O . J A M E S M A N U F A C T U R IN G CO., 1140 W EST M O N R O E STREET, C H IC A G O 7, ILL

¿ » fi TddanA Today P o -itu A G S i

Magnesium Cutback. Magnesium is the second metal

As matters now stand, the industry is entirely capable . . . of

N E W A O FUL-VUE light-w eight acetate rests lightly o n 'th e

ACETATE G O G G L E sliding forward on the nose. And the ex­

New Foam Insulation U.S.I. Expands Line

Better Primer Adhesion

Ups Color Fastness Postwar Cosmetics T E C H N IC A L D E V E L O P M E N T S

dispersion containing some sodium phosphate

60 6 A ST 4 ÎN D ST.. NEW YORK

alco h o ls A C E T IC ESTERS IN T ER M E D IA T ES ACETONE

your extinguishers in shape ... FREE BOOK tells how!

A method for setting up an extinguisher maintenance

■ «i( nlrpa. t-M*, i£r»Vvv

Ross’ standard lines of heat exchangers, heat­

That sounds like a contradiction, but it isn’ t.

In this way, problems of continuous perform­

Larger Coolers Sm aller Coolers O il Heaters

ROSS HEATER & MFG. CO., INC.

Phis is an excellent example of Struthers

Panel board for one of the 3 cracking units, show ing

Aviation luel and butadiene processes in this plant

It's not surprising that the Cities S ervice Lake

REG. U. S. PAT. OFF.

A n d Y O U w ill be o f the same opinion processing plants w ill want to refer to

and specifications o f the latest equipment

for the many and varied drying require­

ments . . . Engineers in chemical and food

& N O R T H R U P C O M P A N Y , 4920 S T E N T O N AVE., P H I LA. 44, PA.

During the second quarter of 1944, 209,004 long tons of syn

ACETATE G O G G L E sliding forward on the nose. And the ex

Ross’ standard lines of heat exchangers, heat

In this way, problems of continuous perform

for the many and varied drying require

A rriv in g re a d y for In sta lla

Eleven G -E com bi

W hether i t ’ s high-octane gasoline, solvents, mag

R e f r i g e r a n t for— heat transfer equip

d e g r e a s i n g a g e n t for— use with vari

CARBOFRAX" (S ilico n C arb id e) in fur

A llyl Alcohol forms esters which can be polym thetic cyclopropane.

Solid adsorbents should be screened to as uniform size as prac PF1ESSUF1E DROP