CHAPTER V

THERMOCHEMISTRY AND THERMOPHYSICS

THEHMOCHEMISTltY

Chemical changes are in gPrteral accompanied by either liberation

or absorption of heat. ThiR ii-; evidmce of the fact that the total energy
content of the system existing before the chemical change is different
from thP energy contPnt of the systPm Pxisting af ter the chemical change.
The clwmical change is accompaniP<i by a redistribution of energy; and
while the twtion may often be caused to take place in such a manner that
liberation or absorption of electrical or some other form of energy accounts
for part of the total energy change, in the majority of chemical and metal
lurgical processes libenition or abimrption of heat energy accounts for
nearly all of it. 1
As an cxamplP, consider a system composed of 112 kg. of iron and
48 kg. of frpe oxygC'n, both at 18°C. This system may be regarded a8
having a certain total energy content, made up chiefly of latent chemical
energy. Now, if t he two components of the syRtem are caused to unite
chcmirall', there rPsults a system composed of H\O kg. of the iron oxide
Fe20a ; and thi8 8ystC'm, when hrought to a temperature of 18°C., will
have a total ener11:y eontcnt 198,500 Cal. less than the original sy8tem:
2Fe + 30 = Fe20a + 198,500
Cal.
The energy equivalent to Hl8,500 Cal. has, of course, not been destroyed;
it has merely been trarn;ferrc<l to the surroundings. Much of it may be
retained for a short time as sensible heat by the Fe20a but is lost as the
Fc203 cools to 18°. If the chemical reaction takes place rapidly, as in
roasting iron pyrite, thf1 Fc20a becomes very hot; but if the reaction takc>s
place slowly, as in rusting of iron, the heat is dissipated to the surround
ings as fast as it is generated.
Symbols for Change of Heat Energy and Their Signs.-In thermo
dynamics it is customary to think in terms of the energy content of the
original system; heat evolved in a reaction represents a decrease in the
energy possessed by the original system (reactants). Therefore, heat
1 The change of heat content in a reacting system (heat of reaction) is thus equal

to the decrease of total energy of the system. It is not, however, equal to the decrease
of free energy of the system. The significance of free-energy change is discussed in
Chap. XXIV.
'70
80 M ET ALLURGICAL PROBLEMS

evolved is given a negative sign. Change of energy content, represented

here by heat evolved or absorbed, is given the symbol t:.H . When heat
is evolved (an exothermic reaction), llH is a negative quantity; when heat
is absorbed (an endothermic reaction), t:.H is a positive quantity.
In thermochemistry the symbol Q has been used for the heat of
reaction and has been called positive when heat is evolved and negative
when heat is absorbed. Eor example, an exothermic reaction has been
written as follows:
C + 02 = C02 + 97,200 Cal.
And an endothermic reaction has been written aR follows:
C02 + C = 2CO - 38,880 Cal.
Thus, in a reaction taking place under the usual conditions of constant
pressure:
t:.H = -Q.
While the symbol Q and itR accompanying sign convention were used
in the first edition of this book, the use of the thermodynamic convention
has become so much favored in recent years that llll and its accompany
ing sign convention will be adopted in the present edition. The above
equations will be ..written:
Ml = -97,200 Cal.
and
C02 + C = 2CO; t:.H = +38,880 Cal.
Heat of Formation of Compounds.-Evidently the formation of
Fe203 from Fe and 0 liberates heat. The quantity of heat liberated is
called the heat of formation of Fe20s. Its value is Ml = -198,500 Cal.
per molecular weight (in kilograms). 1 Since the molecular weight con
tains 112 kg. of Fe, the heat of formation of Fe20a may also be expressed
as l98,500 or -1,770 Ca1. per k'l ogram of Fe and -1981500 or
-
-1,240
160
Cal. per kilogram of Fe20a.
Tables XXVI to XLIX, pages 404 to 421, give the heats of formation
of various compounds in Calories, given in one ·column per molecular
weight of the compound in kilograms, and in' another column per unit
weight (kilogram) of the metal or corresponding element. In the tables,
for convenience, the values given are those for -t:.H, so that a positive
value in the table represents heat generated in formation of the com-
1 The molecular weight is also called the formula weight or the molal weight. The
beat of formation per molecular weight may be termed the molal heat of formation.
 TH ERMOCH EM ISTRY AN D TH ERMOPH YSICS 81

pound. In a few cases the heat of formation is designated as negative,

meaning that the heat is absorbed instead of generated when the
compound is formed.
The values of heats of formation may be conveniently indicated in
connection with the chemical reaction symbolizing the formation, thus:

.
11H = -198,500 Cal.,

it being understood that the quantity of heat shown is the quantity

corresponding to the number of molecules of Fe203 in the equation. If
the equation be doubled, for the purpose of expressing the oxygen in
number of molecules of 02, we mwit write:

Ml = -397,000 Cal.

An abbreviated manner of indicating the heat of formation, and

the one adopted for the tables, is to writ!' (Fe2,03) = -198,500 Cal.,
the comma indicating union of the elements enclosed in parentheses.
Here it is unden;tood that the quantity of heat is 11H per molecular weight
for the formula indicated between the parentheses.
Often the same compound may be formed from different materials.
Thus FeS2 may result from Fe + 28 or from FeS + S. Similarly CaCOa
may be formed from Ca + C + 30, from Ca + 0 + C02, or from
CaO + C02. The heats of formation naturally differ but are reitdily
differentiated in the sy8tem adopted. We write>:
(Fe,82) -35,500 Cal.
(FeS,S) -12,400 Cal.
(Ca,C,Oa) -289,500 Cal.
(Ca,O,C02) = -195,050
Cal. (CaO,C02) = -43,450
Cal.
Heat of Decomposition.-Let us now consider the reverse process of
decomposition of Fe20a into its elements 2Fe and 30. To perform this
decomposition will require the absorption of a quantity of energy equal
to the amount liberated when Fe20a is formed. Here 198,500 Cal. will
be needed to reduce 160 kg. of Fe20a. This principle may be expressed
by the statement that the heat of decomposition of a compound is equal
to its heat of formation but of opposite sign.
When it is desired in practice to reduce Fe 203, it is not sufficient merely
to heat the Fe208, though relatively weak oxides may be reduced by
heating alone. In general merely supplying the heat of decomposition
will not cause the decomposition to occur. There must be a suitable
reducing agent to combine with the oxygen and remove it, or else the
reduction must be performed by electrolysis, u ppl·in11; the necessary
82 M ET ALLU RGICAL
PROBLEMS
energy in the electrical form. With Fe 203, in practice the cheapest
method is to use carbon or CO as a reducing agent:
(1)
or
(2)

Thus the carbon or CO is oxidized, which lierates energy and thn:-:

supplies part (sometimes all) of the energy needed. An additional
amount of carbon burned as fuel liberates additional heat to supply the
balance, as well aR to produce the neceRRary operating temperature, mrlt
the iron and the :-;Jag, t'te. The thl'rmochcmical cquatiorni of comlrnstion
of carbon are:
C + 0 = CO; Mf = -29,HiO Oal.
C + 02 = tlH = -97,200 Cal.
C02;
Heat of Reaction.-F.vidently rradion (1) above is a combination
of the reaction
Fe203 = 2Fc + tlH = +198,500 Cal. (3)
30;
and the reaction
t:..11 = -3(29, mo) Cal. (4)
3(C + 0) = 3CO;
Adding reactions (3) and (-l):
Fe20a + 3C = 2Fc + 3C< l ill! = + 1 1 1,020 Cal.
;
It is not necessary, however, 1.o H<'parntC' into two reactions and then
recombine. We may write:
Fe20a + 3C = 2Fe + 3CO
ilH = +198,.500 - 3(29, l fiO) = +111 ,020 Cal.,

placing the heats of formation or decomposition under the respective

compounds with the proper sign and adding algebraically. Similarly
Eq. (2) becomes:
Fe20a + 3CO = 2Fe + 3C02
ilH = +198,500 + 3(29,160) - 3(97,200) = -5,620 Cal.
In this manner the heat evolved or absorbed by any chemical rear
tion, called its heat of reaction, may be calculated from the heats of forma-.
tion of the compounds involved.
By applying the same principle, the heat of formation of a compound
may be calculated from the heats of formation of two other compound:-;
from which the compound in question may be derived. For example,
 THERMOCHl!JMlSTRY AND 7'HERMOPHYSICS 83
there was given on page 81 the heat of formation (Ca,O,C0 2) = -195,-
050 Cal. The heat of formation of (Ca,O) = -151,600 Cal. If this
second combination takes place between the Ca and O occurring in the
first expresHion, the first expression becomes (Ca0,C02), and it::; heat
of formation i:; obtained by :<11 h1 r:u·1 i1111; -151,600 from -195,050,
so t hat (CaO,C02) = -4:3,4.50 Cal. The equatiomi may make the relation
morP rll'ar:

SubLrnctm)!;:
Ca + 0 + <'O· = J.ll = -t 95,050 Cal.
ThereforP: CaCO,, Ca + 0= CaO; Ml = -151,600 Cttl.

CO, = CaCO, - CaO;l

J.lT = -43, 150 Cal.
Ca o t co, = caco,)
This principle i:; ext.n·nwly useful m detl'rmining and !'hl'cking heats
of formation and is 11pplied continually by experimenten; in thermo
chemh,try. I t often permits 11lso the calculation of a heat of formation
from an experimental determination of a different heat of reaction.
Heat of Combination.-Heat of formation is often called hrat of com
liination when referring to the romhining ei!'ments rather than to tlw
compoun1l form!'d. Heat of combination of an element with oxygen may
be called heat of oxidation.
Heat of Solution.-An energy change ahm occurs when a compound
dissolves in water or other solvent. It may r!'sult in either liberation or
absorption of heat, causing the temperature of the solution to rise, a:-;
with sulphmic acid in watl'r, or to fall, as with potassium cyanide in
water. The heat of solution is accordingly either negative or positive.
It follows that the heat of formation of a soluble salt when formed in
solution is different from the value when it is formed as an anhydrou8
compound. The difference betwe('Il these two values is equal to the
heat of solution. The value of the heat of solution varies with the
concentra tien of the solution. In general it is larger (positively or
negatively) in dilute solution than in a concentrated solution. After
the solution
has become fairly dilute, there is little change from further dilution. The·
tables give, in the case of the 'more important soluble salts, the heats of
formation both for the anhydrous material and "in dilute solution."
The latter value is applicable whenever the weight of water is several
times the weight of the compound dissolved. It is necessarily indefinite;
but heats of solution, while large enough to be important, are small
enough so that some degree of indefiniteness or variation from the tru<>
value will not cause an appreciable error.
Heat of Dilution.-From the discussion of heat of solution. it follows
that making a solution more dilute will also liberate or absorb heat.
Except when a rather concentrated solution is made dilute, the change is
84 M ET ALLURGICAL PROBLEMS

small. Heat of dilution is of little practical importance to the metal

lurgist and will not be further considered in this book.
Heat of Formation of Alloys.-When two metals form a solid solu
tion, the heat of formation of the alloy is simply the heat of solution of
one metal in the other. It is a small figure and may be either positive
or negative.
When a chemical compound is formed between the alloying elements,
the heat of formation of the compound must also be considered. It
may be either small or large; the t:i.H is usually negative. Solution of
such a compound in the remainder of the alloy, when it occurR, affords
further possibility of heat change--one about which, incidentally, there
is little or no exact knowledge.
When the metal components exist as separate crystals of pure metal,
as in a eutectic, however fine and intimately mixed the crystal8 may be,
there is no heat of formation involved. There is, of course, a heat of
solution of the constituents while molten, but this is counterbalanced by
the same quantity of opposite sign when the alloy solidifieH and the
metals separate from solution.
Few data are available on heats of formation of alloys, though the
question is an important one.

'J'HERMOPHYSICS

Whereas thermochemistry deah; with heat in relation to chemical

changes, thcrmophysicR has to do with heat af! affected by physical
changes, with particular reference to change of temperature and change
of phase.
If the temperature of a body initially at 0°C. is raised to t°C., pro
vided the body has undergone no change of phase, it will have gained a
quantity of heat equal to its weight times its mean specific heat between
0 and t times t. A change of phase, such as is excluded in this statement,
may be melting, vaporization, allotropic transformation, sublimation,
etc., any of which in general may occur at a definite temperature without
temperature change. Such changes involve absorption or liberation of
heat. If there is no such change of phase involved, the heat content of
a body at t°C. may be taken for practical purposes as the product men
tioned. This formulation deliberately, for reasons of convenience,
ignores the heat content of the body at 0°C. (A parallel formulation
may be made in Fahrenheit units.) 'f he subject of specific heat and
sens1ble-heat content has been discussed on pages 29 and 30.
The h<J!'f cavacity of a body is its weight multiplied by its specific
heat. The expr:<"ion "heat capacity per gram," therefore, often used
in tables, has the sam. meaning as that of specific heat in the ordinary
 1'llERMOCHEM IS1'RY AN D TII ERM OPH YSICS 85

weight unit. 1 The mean heat capacity between t 1 and t 2 multiplied by

the change in temperature (t1 - t2) gives the change in sensible-hea <
content in going from ti to t2.
If t 0 is above the melting point of the body, then the relation just
discussel'I. may be applied only up to the melting point. Here the body
changes tu the liquid phase and in so doing receives or takes up an
additional quantity of heat called the latent heat of fusion. After melting,
the above relation may again be applied up to the boiling point, but now
with a different specific heat, viz., that of the liquid, applied to the risP
in temperature above the melting point. At the boiling point an addi
tional quantity is again absorbed, the latent heat of vaporization. Above
the boiling point the Rpecific heat of the vapor mm't be us!'d. When the
body coohi, the same changPs take plare in the reverse direction, the latent
heats of change being liberated instead of absorbed. ·
Some substanceR undergo transformations other than those of fusion
and vaporization, such as the well-known allotropic changeR of sulphur
and the critical chang<'s of iron. There is a ;;imilar absorption or liber
ation of heat accompanying earh of these, usually a small quantity. For
example, th<' changes from pure 0<- to y-iron and from y- to o-iron, taking
place at about 910 and 1.f00°C., respectively, absorb about 3.9 and l.H
Cal. per kilogmm, resp<'ctivcly, whili the latent heat of fusion of -iro11 is
65 Cal. and th<' latent heat of vaporization about 1,515 Cal. per kilogram.
Dulong and Petit's Law.-Dulong and Petit long ago·observed that
the specific heatH of PlPmcnts arr rPlated to thC'ir atomic weightl'. They
formulated a law which may be stated as follows: The product of the
specific heat of an element times its atomic weight is a constant, sub
stantially the same for all elements.
Practically it haR bN'n found that not only are there some exceptions
to this statemPnt hut that in general the word "substantially " must be
interpreted with liberality. For example, the specific heats increase with
temperature, though slowly-in many cascH about -! per cent for each
100°C. Abo, even at 0° (or any other temp<'rature), the product is only
roughly constant for different substances. For solid metalR the value at
0°C. varies between 0 and 7, averaging about 6..f .
From Dulong and Petit's law it follows that elements having high
atomic weights have low specific heats, and vice versa. Accurate experi
mental' detrminations of the specific heats of dements have rendered
the law of theoretical interest only-with one important exception. It
has been found that in the case of monatomic vapors the constant noted
1 Although it seems preferable to differentiate between "heat capacity" and

"specific heat" as hre indwated, in many books the term "heat capacity" is
alwa·" us('d with reference to one unit of the substance and thus has the same
meaning as "specific heat."
86 M ET ALLU RGICAL PROBJ_,E M S

above is indeed practically a comitant and has the value of 5.0. Since
most metals in the vapor state are always monatomic, this relation
enables one to determine easily the specific heat of a metal vapor, a
quantity which can be measured experimentally only with great

vaor, then, :
difficulty and in many cases not at all. The specific heat of a metal

is 5.0 cal. per gram-atom, or 4 = 0.223 Cal. per cuhic meter. These
statements refer to the specific heat at constant preHs11re, c,..
This value apparently does not vary appreciably with temperature.
It has been used in calculations for various metal vapors frequently and
with considerable confidence.
Summary.-Lct it he required as an example t,o ealcnlat<' the
hmt, content of 1 kg. of nickel (vapor) at 3500°C. Nickel undergoes :t
trans formation in ·the solid i;tatc (a- to {1-nickc'l) at ahont 52°C. It
mP!t.s at 1455°C. and boils at about 2730°C. Jn going from a- t.o {1-
nirkPl, l kg. absorbs 1.33 Cal. The latent heat of fu14ion of nirkcl is
72 Cal. per
kilogram. The latent heat of vaporization has not been <iPtPnninrd
directly, but a method of calculation gives l ,-rno Cal. a:-; an approximation.
The mean specific heat of mlid nick<>l bPtwrcn 0 and 200°C. has hr<>n
determined as 0.1032 + (I 00011.;I ' In the ab;;encc of ot.lwr data, thi:-;
eression may be assumed to apply up to the transformation point. The
specific heat o( {1-nickel is 0.128 at 360°C. anc!remains approximatPly
constant up to the melting point The specific heat of molt m nickel is
known to be slightly greater than that of solid nickel. [ts rxart value
is uncertain since different published results re at variam·c'. Tim
most recent determination iR O.Hl. The specific heat of the vapor i:-;
probitbly close to 5.0 Cal. per molecular weight of 58.7 kg From thes<'
valus, the specific heat of nickel vapor per kilogram is calculated as 0.085.
With the above constants, the calculation now proceeds as follows:
Heat in 1 kg. of a-nickel at 352° = (0.1032 + 0.000051 X 352)352 = ·12 (j ( 'al.
Latent heat of transformatmn, a to {:J = 1 ,3
Heat in 1 kg. of {:1-niekPl at 352° = - - 4:f"ii Cal.
Heat gained between 352 and 1455° = 0.13 X (1,455 - 352) = 143
Heat m 1 kg of solid nickel at the melting pomt (14.55°) = -187 Cal.
Latent heat of f t1H10n = 72
' Heat in l kg. of hquid nickel at the melting pomt = 259 Cal
Heat gaind between 1455 and 2730° = 0.16 X (2,730 - 1,455) = 204
Heat in 1 kg. of hqmd nickel at the boiling pomt (2730°) = - ( 'al
Latent heat of vaporization = I , 490
Heat in 1 kg. of nickel vapor at the boiling point = i,953 - Cal.
Heat gamed hetween 2730 and 3500° = 0.085 X (3,500 - 2,730) = 65
Heat in 1 kg. of nickel vapor at 31\00° = 2)Yttf ( 'al

The labor of determining the above quantity would be greatly lesspncd

if Rome of the intPrmt>diak quantities were k nown. For example, if thr
 Tll NNMU<'ll EM 187'/ff AN IJ 7'1/Rll MOl'llY Sl<'R 87
heat content of liquid nickel at the boiling point (463 Cal.) W<'re given,
only two short steps would be necessary to arrive at the result. Simi
larly, if the heat content of liquid nickel at the melting point (259 Cal.)
he given, one short step will give the heat content of the liquid at a
desir<'d temperature. Table XIV, page 391, gives value8 of hC'at
contents for different metals at their melting points an<l at thl'ir boiling
points.
The heat content of l kg. of the solid metal at the mE>lting point
may be designated qM, and that of the liquid metal at the boiling point
is designated qB. It is important to considn the precision of thest·
quantities. The values of qM for the common metals are known with a
satisfactory degree of precision. The values of q,. are less well deter
mined, but in most cases the probable E'n:or doN1 not exceed ± 5 to 10
pC'r cent. The values of all heat contents above q,., however, arc more
uncertain, owing especially to the uncertainty in the value of the latent
hl'at of vaporization, which, a.<1 appc•ars above in the case of nickel,
is always so large a quantity as to outweigh all the othn quantities.
Expressions for Heat Content and Specific Heat and the Use of
Tables.-The .sensible heat eontC'nt above 0°C. of a body at tem
perature' t may be expresHed in general terms aR follows, when no change
of phase' takC's plaet> in thP hodr whilC' its tPmpC'raturP changes from 0
to t:
'f = 1r X Cm X t, (5)

wlwn• JV iR the weight of thC' body in grams or kilogramR and Cm iR the•

mean spC'cific heat (per gram or pt>r kilogram) between 0 and t0• This
c•xprC'l'lsion ha:-1 already bC'C'n givPn on page 30; it follows directly from
the drfinition of 1-1peeific hC'at. The hC'at content of 1 g. of the substance
iH iVC'll by:
q = Cmf.
Thnt. of I mol i1-1 J?,iVC'n
h·:

ThP heat. eapacity of a body is <'qual to W X c, or lV and t.hr molal

heat capacity of a substanrC' i1-1 equal to C, or -

It has been found that the specific heat varies with temperature, and
c., may be written empirically in the form A Bt + Ct2+ +···
,where
A iH the specific heat at 0°, and B, C, etc., are other constants for the
substance in question. Ct2 is often negligible in magnitude and Dt 3
and
1-111ccceding terms usually so.
At 0°: q = At.
At t0 : q = (A + Bt + Ct2)t, (6)
88 M ET ALLU RGJCAJ, PROBLEM S

where B and C apply only to the mean specific heat c,,.. If these values
of B and C are not known, but the values are known of a and {3, repre
senting the corresponding constants in the expression for the actual
("instantaneous") specific heat at temperature t, a being the specific
heat at 0° (a = A ), then the actual specific heat at t0 is given by:
c, = a + {jt + "{t 2• (7)
l\nd the heat content above 0° at t 0 is given by:
q= .£' (a + {jt + "{l 2)
(8)

dt
(9)
= at + Y2f3t 2 + Ya'Y t·1
= ( + Y2f3 t +
%'Yt2 )t.
Equation (9), it will be obsc>rved, is applicable for finding thr heat
content at t 0 , wlwn the exprrssion for the actual specific heat at t0 is given
instead of the mean specific heat.
Equation (8) is a more general form
of Eq. (6), and it i obyious that
the coefficients in tlie 'expression for
mean specific heat between 0 and t 0
are related as follows to the corre
sponding coefficient,; in thr expression
for actual Rpecific heat at t 0 when
the latter is given in the form of Eq.
(7): A = a, B = J,{3, C = 'S"f .
When C i;; negligible, the graph
of Ct = A + Rt is a straight line, and
0 Temperciture t the mean value of B is its value
Fro. 3.-Vanatwn of spec1hc heat with halfway between 0 and t, or a + !1,
{3t ; temperature (empu ical) but when C is not negligible, then
Ct = A + Bt + Ct 2 is a parabolic curve, in which C = !'a'Y (see Fig. 3).
The chnge in sensible lwat when the temperature changes from t 1
to t2 may be expressed as:

dq = J(t ',' (a + {jt + "{ l + · · · ) dt.

2 (10)

For many substances, especially solids, it has been found that varia
tion of specific heat with temperature can best be expressed in the form
A + BT - CT-2, except when the value of T is small, say below 400°.
(Here T is on the absolute temperature scale, degrees Kelvin.) When
this expression is integrated between 0 and T to obtain the heat content
above 0°K., the result is:
q = (A + 72BT + CT- 2) 'l'. (11)
 1'HERM OCHEM IS1'RY AN D TH ERMOPH YSICS 89
The capital letter C is used as the symbol for the molal heat capacity
of a substance. The heat capacity of l mol would be the same quantity
as specific heat per mol. The Amal! letter c is used as the symbol for the
specific heat of 1 g. In theoretical work the molal heat capacity is of the
greater significance; but in practical calculations, where weights are
known in grams or kilogramR, the values of c and c., in calories per gram
( = Calorics per kilogram) arc morn convePient to mm; the latter will
therefore be given in the tables in Chap. XXV. For the same reason of
convenience in u;;e, ;;;ince it is customary in practice to calculate heat
contents from 0°C. as a base line, fae cxpre:-1sions for Hpecific heats in
this book will have Huch valuPs for the constants that a iR the specific
heat at 0°C. Jn many book;;; the c>xpres:,;ions for ;;;pecific heat and heat
Papacity are given in the form l + BT + C T 2 , where T is on the abso
lute scale and 0°K. ( -273°C.) is the ba:,;e line>.
In using given values or exprc'sHions for Hpecific heat and heat capacity,
it is therefore esst>ntial to note (1) whether they are given per mol or per
kilogram (or, in the case of a gaH, per cubic meter) and (2) wheter they
arc for degrees centigrade (t) using 0°C. a8 the ba;;;c line or for degrees
Kelvin (T) ur:ing 0° absolute t<•mperature a;;; the haHf' line. Both methods
of exprer:Hion will give the ;;;amc reHults if correctly used. More rarely,
the cornitant:-; given may h<> for the Engfoih ;.;ystem (B.t.u. per pound or
per pound-mo! and per degree Fahrenheit). The tahlPs in thi;;; book,
as is eustomary, use metric unifai; but values given in calories per gram
or Calories pPr kilogram hold also for pound-calories per pound, and it

TABLE II I -M]<]AN Sr1,c1nc Ht •T>< m· Som CoMMON

SuB,.,TANCEH
(In Calori<'H p<'r C11b1e MPtPr for OaHPH, ( 'alorif'H p<-r K1lo11:rnm for SolidH)
Tf'm pna t1m,I
I i I
Air, 02,
mtcrval,
degr<'es cenhgradr
•. co
( '02 H,O(g) (' I
810,• Fo
- ------
0 - ----- - -- - ------ ----·----- ---- --------
0-500 0 302 0 406 0 373 0 206 0 166 0 105
0-600 0 313 0 4.51 0 398 0 287 0 238 0 131
0-700 0 31.5 0.460 0 403 0.301 0 2.520 137
0-800 0 317 0 469 0 408 0 314 0 255 0 142
0-900 0 320 0 478 () 413 0 327 0 257 0 148
0-1000 0 322 0 487 0 418 0 339 0.259 0 149
0-1200 0 324 0 496 0 423 0 351 0 262 0 149
0-HiOO 0 328 0 514 0 433 0 372 0 265 0 150
0-1800 0.335 0 541 0 448 0 399 0 149
0-2000 0.342 0 568 0 275
0.463 0 419 0.2lt 0.2lt
0 346 0 586 0 473 0 430 ..

* R. B. So•MAN, see Tablr XX, p. 398.

t Ji,or liquid iron only1 above its melting µornt.
90 MET ALLU RGICAL PROBLEM S

is recommended that problems involving specific heats; heats of reaction,

etc., with weights given in pounds be worked in pound-calories instead
of B.t.u. in order that values given in the table8 may be U8ed directly.
Table III gives the mean specific heats of some common sub
stances over frequently used temperature ranges. These have been
calculated from the expressions given in the tables in the back of this
book and are tabulated here for ready reference.

HEAT OF REACTION A'l' HIGH TJ<JMPERA'l'URES

When the heat of a reaction is calculated from thr heats of formation

of the compounds involved, in the manner discussed on page 82, the
result applies to the reaction at room temperatme, i.e., when the
reacting substances are at room temperature at the beginning and the
products formed are allowed to come to room temperature at the end.
In other words, the calculated quantity of heat i8 that corresponding to
a process beginng and ending at room temperature, irrm;pective of
what tem peratures may have been reached sometime during the
proceRs. This follows necessarily from the fact that heats of formation
are determined and tabulated for the conditions of room temperature.
In figuring the values of heats of formation, room temperature is
now usually taken as 18°C.; but for most purpoo:Ps the heat of reaction
SO Calculated zhay be regarded aR identical with that for a temperatlll'C

of 0°C. or for higher temperaturc:o; up to 35 to 40°C. If , however, the

reaction represents a process beginning or ending at Home elevated tem
perature and it iR desired to know the hmt lihcrakd or required at that
temperature, then the heat of the reaction calculated for room temperature
must be suitably corrected.
As an example of a prOCPHH in practicn where it might be desil'Pd to
know the heat of a reaction at a high trmp<'rat.u rn, consid<'r the conver:-1ion
of copper sulphide to copper metal, in accorda1we with the reaction
Cu2S + 02 = 2Cn + S02.
This reaction is carried out in a BeHsemer convPrter; the temperatme
is 1100°C., so that the reaetanti-; (Cu 2S and 02) are at this temperature,
and also the product;; (Cu and S02). The aHHmnption is really not a
good one with respect to the oxygen and will be corrected below, but for
the present let all the reacting Aubstances be at the same temperature,
so that the problem becomes one of finding the heat of this reaction at
1100°c.
First obtain the heat of the reaction at room temperature:
Cu2S + 02 = 2Cu + S02.
AH = +18,950 - 70,940 = -fil ,990 Cal.
 THERMOCHEMIS'l'RY AND '/'HERMOPHYSICS 91
Now if the reactants are at 1100°C., when they are consumed or dis
appear the heat contained in them will be liberated and will serve to
increase the total amount of available heat. We must therefore calculate
the heat content of the Cu2S and 02 at 1100°C. and add it to the 51,990
Cal. generated. (In this summation, since heat evolved is given a nega
tive sign, the heat in the reactantH will also be given a negative sign.)
Similarly, if the products arc at ll00°C. and remain at or leave the
converter at that temperature, they ab1mrb and carry away a quantity
of heat equal to their heat content at l 100°C. This must accordingly be
calculated and subtracted from the above sum.
Heat contents at 1100°:
In 160 'kg. of Cu,S:* 160 X 0.131 X 1,100 23,050 Cal.
=
Latent heat of fusion = 160 X 34.5 = 5,520
8,030
In 22.4 cu. m. of O,: 22.4(0.302 + 0.000022 X 1,100)1,100 = 36,600 ('al.
Total heat in reactants = 14,600 t'al.
In 128 kg. of Cu: 128(0.0916 + 0 0000125 X 1,083)1,083 = 5,360
Latent heat of fusion = 128 X 41 . 8 180
= 12 ,440
1083 to 1100° = 128 X 0.079 X 17 32,-580 Cal.
=
I n 22.4 cu. m. of 802: 22.4(0.406 + 0.00009 X 1,100)1,100 =
Total heat m products
=
Heat of re11ction 11t 1100°: All = -lil,990 - 36,600 + 32,580 = -56,010 Cal. (I).
Actually the oxygen in the above reaction enters the converter in the
form of unheated air. If it is desired, therefore, to learn the net amount
of heat resulting from the actioIJ. in the converter, the 8,030 Cal.
credited as coming from oxygen should be omitted. Furthermore, since
the oxygen is necessarily accompanied by nitrogen, which enters the
procesH cold and goes out at ll00°, the heat contaim!d in 7%1 X 22.4
cu. m. of nitrogen at 1100° should be deducted.
Heat in nitrogen = 84.3(0.302 + 0.000022 X 1,100) 1,100 = 30,250
Cal.
The net heat evolved in the converter as a result of the action in it is
therefore 5(i,010 - 8,030 - 30,250 = 17,730 Cal. (per 160 kg. of
Cu2S oxidized).
From the principles discussed above, the following rule may be formu
lated: To calculate the heat of a reaction at high temperature, add the
heat content of the reactants, and subtract the heat conten of the prod
ucts from the heat evolved in the reaction at room temperature. This
rule is to be taken in its literal algebraic sense whether the reaction is
exothermic or endothermic. If the above reaction had been one which
absorbed 51,990 Cal. at room temperature instead of liberating that
quantity, we should have written in place of (1):
Heat of reaction at 1100°: AH - +51,990 - 36,600 + 32,580 = +47,970 Cal.
 • The specific heat of liquid Cu 1S is unknown. The value 0.131 is for solid Cu,S
at 900°C.
 It might be supposed that calculation of the heat of a reaction at high
temperatures would be a problem with which the metallurgist is com
monly concerned, since so many of his reactions are furnace processes.
Actually, however, he is less of ten interested in this calculation. The
ordinary heat of the reaction at room temperature is the more useful
quantity, and the factors intr_oduced by the furnace operation are handled
separately as souroos of additional heat supply or heat loss as the case
may be. Thus in a smelting process the sources of heat supply might be
the combustion of the fuel and heat contained in preheated blast. The
items of heat loss might be heat contained in the waste gases, in the hot
metal and slag produced, in cooling water, losses by radiation, convection,
and conduction, etc. These would all be in addition to the heats of the
smelting reactions themselves, and the latter would be taken at room
temperature. By computing and enumerating all the items of heat
supply and heat loss, a "heat balance sheet" of the furnace or process
could be made up. This is more often the most profitable manner of
studying the thermal requirements or heat economy of an operation.
The heat balance is taken up in detail in later chapters.
There are, however, cases where it is desirable for the metallurgist
to calculate the heat of the reaction itself at high temperatures, in the
manner which has been illustrated. Such a case, for example, is in the
common retort process for reduction of zinc ore, considered on page 332.
The metallurgist should be aware of the principles involved and deter
.mine when such a calculation would be .instructive as regardR any par
ticular process.
There are in reality five possible cases in calculating heats of reactions,
1>iz.:
1. Reactants cold, products cold.
2. Reactants cold, products hot.
3. Reactants hot, products hot.
4. Reactants hot, products cold.
5. Mixed conditions: some of the reactants or products cold, other8 hot

The procedure in the various cases will be obvious from the foregoing
discussion.
Kirchhoff ts Equation.-The change in heat of reaction with change of
temperature may be expressed as follows:1
dQ = (Gp - Gp') dt,

where GP and Cv' are the. heat capacities at constant pressure of the
reactants and of the products, respectively, of the given reaction. In
this equation all reactants and products are at the same temperature.
1 KIRCHHOFF, G., Ann. Physik [2), 103, 177 (1858).
 THERMOCHEM /81'RY AN D 1'HERMOPH YSICS 93
Integration of the equation and substitution of the proper values of Cp
and C/ lead to the same result as in the foregoing discussion.
Phase Symbols in Reactions.-Since the heat content of any sub
stance taking part in a chemical reaction will depend greatly on whether
it is in the solid, liquid, or gaseous 'state, it is desirable in some cases to
follow the formula for a substance in a reaction with a symbol in paren
theses to show the phase or state in which the substance is present.
This should be done when any doubt can arise. as to the existing state;
when the state is obvious, it is better for the sake of clarity and simplicit y
to omit such symbols. The symbols most commonly used are:
s = sohd.
c = crystallme
l = liquid.
dil = d1fote solution.
g = gas. aq = dilute aqueous solution
OtherH are Rometimes Examples of the way in which these arc
uReful. m ed are aH follows:
Electrolyt1c Pxtraction of copper from solution formed by leaclung: CuSO,
(aq) + H,O(l) = Cu + H ,SO,(aq) + O(g).
Dctmnmg of srrap tin plate with rhlorm<>:
Sn(.•) + 2CI,(g) = SnCl,(l).
Prod11ct10n of ammonia from cyananudP:
CaCN, + 3H,O(g) = 2NH,(g) + CaO + CO,.
In this book, Rince the phase or stat<> will wmally be obvious from
the nature of the process under discussion, it will not of ten be necessary
to employ these symbols.
Problems
96. Calculate 'the heat of the reaction for each of the seven rf'actions given or
required in Probs. 1 to 5. (Place the proper plus or mmus sign before each answer;
for CO and CO, use values for amorphous carbon.)
97. From the tables, Chap. XXV, obtain the following heats of formation: (F•-,
S1, 03), (FeO, 8102), and (S1, 02). Write each of these in theform of a thermochemical
f'QUation. From these equations determine the heat of formation of FeO, per kilo
gram of Fe.
98. a. Givrn the following heats of formation (from t he tables, Chap. XXV):
(Ag,O, N 20,) = -46,100
(Ag, N, O,) = -29,400.
(N,, O.) = + 600.
Required: The heat of oxidation of silver to Ag 20, per kilogram of silwr.
b. Given the followmg heats of formation ( -till):
MnO from Mn and 0: 1,757 lb.-cal. per pound of Mn.
SO, from S and 30: 2,930 lb.-cal. per pound of S.
MnSO, from Mn, S, and 40: 1,664 lb.-cal. per pound of MnSO,.
Required: The molal heat of formation of MnSO, from MnO and S08•
94 M ET ALLU RGICAL PROBLEM S

99, a. Antimony ore containing 27 per cent Sb2Sa is roasted, producing Sb20•
and 801• The weight of ore treated is 5,000 lb.
Required: 1. The volume of oxygen required, in cubic feet.
2. The quantity of heat evolved, in pound-calories.
b. An electric furnace is used to manufacture gaseous carbon bisulphide
through the reaction of sulphur with carbon. The production is 7 metric tons in
24 hours.
Required: The total amount of energy <'volvcd or required (which?) in the reaction,
in Calories, and the power in kilowatts.
100. An ele?tric furnace is.supplied with CaO and C, in the proportions to form
CaCs:
CaO +3C = CO + CaC,.
It produces 0.25 kg. of CaC, per kilowatt-hour. The CaC', reacts with H20
according to the reaction
CaC, + 2H,O = Ca(OHJ, +
C,H,.
Required: 1. The minimum power to produce 1 ton of CaC 2 ppr hour.
2. The electrothermal energy efficiency of th<• furnace operation.
3. If 100 g. of CaC, is treated with 10 kg. of cold H,O, how many calories will
be set free, and how many degrees rise of temperature will rcHult?
101. a. Calculate the heat content (above 0°C.) of 1.5 kg. of tin (a) soh<l, iust at.
the melting point; (b) molten, just at the melting point; (1') at a tPmperature of 1200°C.
b. Calculate the heat content (above 0°C.) of 1 mo! of silicon vapor at 3700°<'.
(the temperature of the electric arc between carbon electrodes).
102. A coal-burning furnace was used for meltmg copper cathodes and required
7 hr. to melt 160-'llletric tons of copper. The calorific power of the coal used was
5,952 Cal. per kilogram. There was used 250 kg. of coal per ton of copper melted.
The heat capacity of the gases resulting from burning 1 kg. of th<' coal is 4.071 +
0.000669t. The furnace is equipped with a waste-heat boiler; temperature of thr
gases at the entrance to the boiler, 1100°C.; at exit from boiler, 500°('.
Required: 1. The quantity of heat required to melt the copper
2. The thermal efficiency of the furnace durmg the melting.
3. The boiler horsepower developed by the boiler, assuming that it utilizes 94 per
cent of the heat dropped by the gases in passmg through it.
103. In converting copper matte, the chemical action in each of the two stages 1s
represented by the following equations:
(a) 2FeS + 30, = 2Fe0 + 280,.
(b) Cu.S + O, = 2Cu + so,.
Assume that the average temperature in the convl'rter <luring the first stage is l 100°C ,
o.nd during the second stage 1200°C. The air, however, is blown in cold ; assume that
its temperature is 0°C. Calculate for the eaction only, disregardmg the nitrogl'n,
which does not appear in the reaction.
Required: 1. The heat of the first reaction, all components except oxygen being at
1100°.
2. The same for the second reaction, all components except oxygen being at
1200°c.
lM. a. Calculate the heat of the reaction for production of silwon 1\arb1de, bPg111-
ning and ending at 1700°C.:
SiO; + 3C = SiC + 2co.
 1'/l /\'U.\/ 0(' /l /:' \f ll'i1'RY AND 'l'HETlllfOPIIYSICS 95
b, Iron carbide ch'c·omposes as follows, forming crystals of pure iron and
graphite:
Fe,C = 3Fe + C.
Required: The heat of this reaction, per pound of iron, (a} when it occurs in solid
cast iron at a temperature of 2000°F., and (b) whrn it occurs in the production of
artificial graphite at a temperature of 4000°F.
106. Calculate the heat of the reaction
Al20, + sc = 2A1 + aco,
I & •
taking place in the Hall procelJs at a ternp"rature of 1000°C. Although this tem-
perature is below the melting point of AJ,O,, the alumina is actually in the liquid
Btate, being dissolved in molten cryolite. Assume, therefore, that the heat content
of the Al,03 includes its heat of fusion.
106. Th<> rc•action for reduction of zinc from roasted zinc ore is:

ZnO +C = Zn + CO.
llequired: 1. The heat of this reaction as it takes place in the zinc retort, beginning
n nd ending at 1000°C.
2. The heat requir('(l for the r<>action, per kilogram of zinc vapor produced, allow i
ng for th<' fact that the zinc oxide and carbon ar<' put into the retort cold (assume
ll°C.).