JOURNAL OF APPLIED POLYMER SCIENCE VOL. 10, PP.

201-209 (1966)

Pressure-Volume-Temperature Behavior of High

Density Polyethylene

GEORGE N. FOSTER III,* NATHAN WALDMAN, and

RICHARD G. GRISKEY , Chemical Engineering Department, Virginia
Polytechnic Institute, Blacksburg, Virginia

Synopsis
Equilibrium pressure-volume-temperature behavior in both the solid and molten
regions was determined for a high density ( p = 0.958 g . / ~ m . ~polyethylene.
) Data
were measured with a recently developed compressibility device capable of obtaining
precibe and accurate data. Residual curve treatment showed that the data were true
equilibrium data. Compressibilities calculated from the data of this work compared
favorably to existing data which were limited to 205°C. The present work extended
the compressibility behavior to 250OC. It was also found that differences in com-
pressibility of low and high density polyethylenes were not eliminated in the molten
region, indicating that the effect of differences in morphology was not eliminated.
The Spencer-Gilmore equation was fitted to the data of the present work. The internal
pressure (a) term of the equatiou showed a definite relation to polymer morphology.

Lack of fundamental physical data is a major deterrant to the progress

of polymer science. A particularly serious instance is the need for com-
pressibility behavior of high polymers.
A number of investigators, among them Bridgeman,' Parks and Rich-
ards,2 Spencer and Gilmore13p4 Matsuoka,8-10 Hellwege et al.,ll
arid Heydemann and Guicking12have measured such data. These studies
have, however, been limited both in the range of temperature and/or
pressure investigated and the types of polymers studied.
A particular example is the case of polyethylene. A number of investi-
g a t o r ~ ~ ,have
~ , ~ studied
- ~ ~ polyethylenes having low densities a t ambient
pressure and temperature (i.e., po = 0.920 g./cme3or less). However,
only two studieslO7l 1 were found that considered the higher density material
(po = 0.973 and 0.981 g . / ~ r n . ~ ) .Both these studies were limited with
respect to temperature.
The present study was undertaken to satisfy the need for higher tempera-
ture data and also to determine the behavior of a polyethylene having a
density intermediate (po = 0.958 g . / ~ m . ~to) those previously studied.
The apparatus used in the study is shown schematically in Figure 1.
The compressibility chamber is depicted in Figure 2. Polymer samples
were placed in the chamber which was evacuated after it had been immersed
* Present address: Union Carbide Plastics Company, Bound Brook, New Jersey.
201
202 G. N. FOSTER 111, N. WALDMAN, AND R. G. GRISKEl'

Compressibility Chamber
and
Oil Thermostat

- n

A
-
- I 1

~ 8 Vacuum
" Line :losure- I/Z" Bolt

Seal

Compressibility
Chamber
.
Chamber Holder.

,Bottom Closure
m
Fig. 2 . Compressibility chamber and holder.
 HIGII DENSITY POLYETHYLENE 203

-02

- 04

3
\

2 -06

s
J -08
0
>
J
a
2
- -10
v)
W
a

-12

- 14
I I I I I
0 100 200 300 400 500 600
PRESSURE, ATMOSPHERES

Fig. 3. Yolume residuals for solid polyethylene.

PRESSURE. ATMOSPERES

Fig. 4. Pressure residuals for molten polyethylene.

 G. N. FOSTER 111, N. WALDMAN, AND R. G. GRISKEY

PRESSURE, ATMOSPHERES

‘i;

Fig. 5. Pressure-volume-temperature data for solid polyethylene.

in the oil bath thermostat. The device was then thoroughly seasoned.
A careful annealing procedure was followed for solid-phase measurements.
Readings were then taken by holding temperature constant, changing
volume, and then balancing the pressure for a long enough period of time
to assure the proper equilibrium pressure-volumetemperature combina-
tion. A detailed description of apparatus and procedure is given else-
where.I3
Temperatures were controlled to +O. 10°C. Volume accuracy was
0.25y0or better while the accuracy of pressure nieasurements were 1%
or less than 1%.
Data were measured for both the solid and molten regions froin pres-
sures of 1 to 618 atm. and temperatures from 23 to 250°C. These raw
data were smoothed by using volume and pressure residual curves. The
technique of smoothing with residuals involves first calculating either a
volume or pressure from an equation of state (at a selected temperature
and pressure or a selected volume and pressure). In this work the Spencer
and Gilmore3s4equation of state,
(P + r)(V - W ) = (R/M)T (1)
was used, where P is pressure, V is volume, T is absolute temperature, R
is the gas constant, ilf is the molecular weight of an interaction unit, and
 HIGH DENSITY POLYETHYLENE 205

i
600

500

400
v)
W
K
W
r
n
v)
0

\
s
+ 300
a
w'
K
2
v)
v)
w
K 200
n

100
xP
0 -
1.2 1.3 I .4
SPECIFIC VOLUME, CC/GM.

Fig. 6. Pressure-volume-temperature data for molten polyethylene.

T and W are constants of the Spencer-Gilmore equation. Next the pressure

or volume residual 1' is calculated:
~p = P c a ~ c- P e x p t l (2)
rv = Vcalc - Vexptl (3)
These residual values can then be plotted as isothermal curves versus
either volume or pressure. Figures 3 and 4 represent such plots. Next
smoothed data points can be obtained by using the residual together with
calculated values :
f'exptl = P c a ~ e- r p (4)
Vexptl = V e n l c - 1'0 (5)
The smoothed data are presented in Tables I and I1 and Figures 5 and 6.
Residual curves also provide a means of checking the consistency of
experimental data as well as the attainment of equilibrium during the
experiments. Smooth curves show that data are consistent and that
l49l5
206 G. N. FOSTER 111, N. WALDMAN, AND R. G. GRISKEY

TABLE I
Pressure-Volume-Temperature Data for High Density Polyethylene
( p 0 = 0.958 n./cm.9

Tem-
pera- Volume, cm.S/g.
ture,
"C. 1 atm. 79.3 atm. 158.5 atm. 232 atm 316 atm. 474 atm. 618 atm.
23 - - - 1.0424 1.0423 1.0422 1.0419
50 - - 1.0542 1.0542 1.0540 1.0537 1.0528
75 1.0680 1.0676 1.0669 1.0660 1.0650 1.0640 1.0621
100 1.0861 1.0857 1.0853 1 . 0852 1.0843 1.O840 1.0827
125 1.1162 1.1157 1.1152 1.1140 1.1137 1.1119 1.1101
130 1.1549 1.1530 1.1513 1 1490 1.1446 1.1446 1.1405

TABLE I1
Pressure-Volume-Temperature Data for Xlolteii High Ileiisity Polyet,hylene
Tem-
pera- Volume, cmm.3/g.
ture,
"c. 1 atm. 79.3 atm. 158.5 atm. 232 atm. 316 atm. 474 atm. 618 atm.
140 1.2852 1.2793 1.2725 1.2675 1.2607 1.2493 1.2393
175 1.3211 1.3132 1.3047 1.2979 1.2899 1.2759 1.2679
210 1.3526 1.3430 1.3336 1.3254 1.3147 1.2983 1.2887
250 1.3925 1.3784 1.3653 1.3549 1.3431 1.3282 1.3187

equilibrium is attained. The relative smoothness of the plots in Figures 3

and 4 indicate that this indeed seems to be the case for the present work.
No direct comparison was made between the pressure-volume-tem-
perature data of the present work and that of the previous investigations.
The reason for this was that the morphology of the polyethylene of the
present study differed from that of the polyethylenes studied previously.
These morphological differences were reflected in the pi) values for the
various polymers.
Instead, the data were compared on the basis of compressibilities
) , molten polymer behavior. It was considered that the
( ~ / V ) ( A V / A Pfor
compressibility behavior would give a truer reflection of similarities or
dissimilarities than a comparison of specific volumes a t a given temperature
and pressure. I n addition, such a comparison in the melt region would be
more meaningful since morphological differences should be minimized.
The comparison is shown in Figure 7. As can be seen, there is excellent
agreement between the data of the present work and that determined for
high density polyethylene by Hellwege et aLl1 I n addition, the data for
low density polyethylene of Hellwege et al." and those of Parks arid Rich-
ards2 shows fair agreement. However, all of the data of Matsuoka'"
deviates considerably.
A number of conclusions can be drawn from Figure 7. First, the ex-
cellent agreement of the data of this work with those of Hellwege et al.
 HIGH DENSITY POLYETHYLENE 207

I30 150 I 70 190 210 230 250

TEMPERATURE, O C

Fig. 7. Comp:lrison compressibility data for molten polyethylene: (A) PO = 0.919

g . / ~ m . ~ ; (A) (a) pa = 0.981 g./cm.3;8-10
l l PO = 0.973 g . / ~ m . ~ ; (" 0 )PO = 0.913 g./~m.~;*-lO
(0) PO = 0.920 g . / ~ m . ~ ; ~)po = 0.958 g . / ~ m . ~ .
(0

establishes the compressibility behavior of high density molten polyethyl-

ene up to 250°C.
Next, differences in the compressibility behavior of high and low density
polyethylene are still apparent even in the molten state. Thus, apparently
the effect of morphological diff erences between high- and low-density
polyethylenes is not completely eliminated in the molten region. Finally,
it appears that the data of Xatsuoka disagree drastically with those of
other invehgators.
A 1)ossible reason for this last occurrence might be the degree of tem-
perature control attained in the variow studies. For example, Matsuoka
reportq a =tl.O"C. control as compared to =tO.l"C. for the present work.
This in itself could account for the divigreemerit.
208 G. N. FOSTER 111, N. WALDMAN, AND R. G. GRISKEY

The Spencer and Gilmore equation of state was fitted to the data of the
present work in both the molten and solid regions. Values of constants
obtained were M = 28.1, a = 6769 atm., and W = 0.917 cm.Y/g. for the
solid polymer. ill and a values were the same for the molten polymer and
W was 1.11~ m . ~ / g Spencer
. and Gilmore4report values of 28.1, 3240 atm.
. M , a, and W , respectively for low density poly-
and 0.875 ~ m . ~ / gfor
ethylene. Differences in the a (internal pressure) values between the
present work and that of Spencer and Gilmore can be attributed to two
reasons. First, the value reported by the latter investigators was derived
from nonequilibrium data while that of the present work was obtained
from equilibrium data. The second reason is that polymer morphology
influences the a value. This can be seen from Spencer and Gilmore’s4
work which reported a higher a value for low density polyethylene than for
polystyrene, the former polymer being more crystalline in nature than the
highly amorphous polystyrene. On this basis it is not surprising that a
still higher a value was obtained from the even more crystalline high density
polyethylene.
Grateful acknowledgment is made to the National Science Foundation for support
of this research by means of grant GP-2099, and to E. I. du Pont de Nemours and Co.,
Inc. for the polyethylene samples.

References
1. Bridgeman, P. W., Proc. Am. Acad. Arts Sci., 76, 71 (1948).
2. Parks, W., and R. B. Richards, Trans. Faraday Soc., 45,206 (1949).
3. Spencer, R. S., and G. D. Gilmore, J . Appl. Phys., 20, 502 (1949).
4. Spenrer, R. S., and G. D. Gilmore, J. Appl. Phys., 21, 523 (1950).
5. Weir, C. E., J . Res. Natl. Bur. Std., 45, 468 (1949).
6. Weir, C. E., J . Res. Natl. Bur. Std., 50, 153 (1953).
7. Weir, C. E., J . Res. Il’atl. Bur. Std., 53, 245 (1954).
8. Matsuoka, S., J . Polymer Sci., 42, 511 (1960).
9. Matsuoka, S., J . Polymer Sci., 57, 567 (1962).
10. Matsuoka, S., and Maxwell, B., J . Polymer Sci., 32, 131 (1958).
11. Hellwege, K. H., W. Knappe, and P. Lehmann, Kolloid-Z., 183, 110 (1962).
12. Heydemann, P., and H. D. Guicking, Kolloid-Z., 193, 16 (1963).
13. Foster, G. N., and R. G. Griskey, J . Sci. Znstr., 41, 759 (1964).
14. Li, K., and L. N. Canjar, Chem. Eng. Progr. Symp. Ser., 49, No. 7, 147 (1953).
15. Canjar, L. N., paper presented a t the meeting of the Division of Petroleum
Chemistry, American Chemiral Society, New York, September 9, 1957.

Resume
Le comportement B 1’6quilibre pression-volume-temperature dam les regions B 1’6tat
solide et B 1’Ctat fondu des polykthyknes a C t B dCtermin6 pour le poly6thykne de densit6
6levCe ( p = 0,958 g/cc). Les resultats ont C t C mesures en employant une methode de
compressibilit6 recemment decrite e t capable de fournir des rCsultats recis et reproducti-
bles. Le traitement ultkrieur de la rourbe montre que les resultats sont des veritables
rksiiltats d’equilibre. Des compressibilitk calculees au depart des rCsnltats de ce travail
se comparent favorablemerit aux donnCes existantes, qni 6taient limitCes B 205°C. Ce
present travail 6tand le domaine de compressibilite jusque 250°C. On a trouve que les
diffCrences de compressibilite des polykthylbnes de basses et hautes densites ne sont pas
4liminPes dans la region B 1’Ctat fondu, ce qui indique qiie les effets des differences de mor-
 HIGH DENSITY POLYETHYLENE 209

phologie ne soiit pas dliminees. L’dquatioii de Spencer-Gilmore s’adapte aux r6sdtats

du present travail. La pression interne, T , de 1’Cquationmontre une relation definie avec
la morphologie du polymhre.

Zusammenfassung
Die Abhangigkeit des Gleichgewichtsvolumeiis von Druck und Temperatur in deli
festeri und geschmolzenen Bereichen eines Polyathylens hoher Dichte ( p = 0,958 g/cc)
wurde bestimmt. Die Messergebnisse wurden mit einer neu entwickelten Kompressi-
bilitatsvorrichtung erhalten, welche zuverlassige und genaue Daten liefert. Die Ilesi-
dualkurveribehandlung zeigt, dass die Daten wahre Gleichgewichtsdaten sind. Die aus
den 1)aten der vorliegenden Arbeit berechneten Kompressibilitiitswerte stimmeii gut mit
vorhandenen, bei 205°C erhaltenen Daten uberein. In der vorliegenden Arbeit wird das
Kompressibilitatsverhalten bis zu 250°C untersucht,. Weiters wurde gefunden, dass die
Kompressibilitatsunterschiede zweier Polyathylene niedriger und hoher Dichte im
geschmolzenen Bereich nicht eliminiert werden, was beweist, dass der Einfluss der
morphologischen Unterschiede weiter besteht. Die Gleichung von Spencer-Gilmore
wurde zur Wiedergabe der erhalterien Daten verwendet. Der Term fur der inneren
Druck, T , der Gleichung zeigte eine definierte Beziehung znr Polymermorphologie.

Received September 7, 1965

Prod. No. 1283