sulfide, 14.5% carbon dioxide respec- liquid, and either solid hydrogen sulfide 3. Kohn, J. P.

3. Kohn, J. P., and Fred Kurata, Univ.

tively (Figure 5 or 7). or solid carbon dioxide are in mutual Kansas, Lawrence (1955).
Isothermal compression of the 60% equilibrium. Those mixtures with more 4. Reamer, H. H., B. H. Sage, and W. N.
hydrogen sulfide mixture from the vapor than 66 mole % carbon dioxide can have Lacey, Znd. Eng. Chem., 43,976 (1951).
phase a t - 133°F. shows that at 4.4 5. Kohn, J. P., Ph.D. thesis, Univ.
vapor and liquid in equilibrium only Kansas, Lawrence (1956).
Ib./sq. in. abs. the vapor is in equilibrium with solid carbon dioxide; those with 6. Ricci, J. E., “The Phase Rule and
with incipient solid hydrogen sulfide and less than 12.5 mole % carbon dioxide Heterogeneous Equilibrium,” D. Van
liquid. Continued compression causes the have triple-point equilibria with solid Nostrand, New York (1951).
system to enter the vapor-liquid region hydrogen sulfide. 7 . Steckel, F., Svensk Kern. Tidskr., 57,
until a t 6.5 Ib./sq. in. abs. solid carbon I n view of the availability of the phase 209 (1945); see (1).
dioxide forms. At the latter condition data at temperatures down t o the solid- 8. Kohn, J. P., and Fred Kurata, Pet.
vapor and liquid in equilibrium with liquid-vapor loci of the three binary Proc., 11, No. 12, 57 (1956).
solid carbon dioxide have compositions systems which comprise the methane- 9. Davis, P. C., Ph.D. thesis, Univ.
of 32.5% hydrogen sulfide, 67.5% carbon hydrogen sulfide-carbon dioxide ternary Kansas, Lawrence (1951).
10. Lange, N. A., “Handbook of Chem-
dioxide and 8670 hydrogen sulfide, 14y0 system, i t is interesting t o speculate istry,”Handbook Publishers, San-
carbon dioxide respectively. about the behavior of the ternary. dusky, Ohio (1956).
If compression from the vapor phase It would be particularly interesting 11. Maass, O., and D. McIntosh, Trans.
for the 60% hydrogen sulfide mixture to know what influence carbon dioxide Roy. Sac. CUP,.,3, No. 8, 65 ( 1914).
now occurs a t - 137”F., solid hydrogen has upon the heterogeneous liquid- 12. Cardoso, E., and E. Armi, J. Chem.
sulfide forms. Upon further compression liquid-vapor behavior reported for the Phys., 10,504 (1913).
liquid forms a t a vapor-liquid-solid methane-hydrogen sulfide system (6). 13. Reamer, H. H., B. H. Sage, and W. N.
hydrogen sulfide triple point a t - 137°F. Lacey, Ind. Eng. Chem., 42,140 (1950).
and 4.7 lb./sq. in. abs. Vapor and liquid 14. West, J. R., Chem. Eng. Progr., 44,
ACKNOWLEDGMENT No. 4, 287 (1948).
compositions a t this point are 45y0 hydro-
gen sulfide, 55y0 carbon dioxide and 91% The authors wish t o thank the Ethyl 15. Moles, E., Bull. sac. chim. France,
hydrogen sulfide, 9% carbon dioxide re- Corporation for their generous support of p. 1006 (Nov.-D~c.1950).
spectively. Vapor and liquid exist a t pres- this work through their sponsoring of 16. Bottomley, G. A., D. S. Massie, and
sures between 4.7 and 5.5 lb./sq. in. abs. fellowships. Thanks are also due to Mr. R. Whythlow-Gray, Proc. Roy. Soc.
Jerome Brewer for his considerable assist- (London),A200, 201 (1950).
a t -137°F. At 5.5 lb./sq. in. abs. and
ance in the laboratory program. 17. Sobocinski, D. P., Ph.D. thesis, Univ.
-137’F. vapor, liquid, and solid carbon Kansas, Lawrence (in preparation).
Phillips Petroleum Company donated the
dioxide coexist where vapor composition hydrocarbons used in the low-temperature 18. Robinson, D. B., and J. A. Bailey,
is 38% hydrogen sulfide and liquid bath. Can. J . Chem. Eng., 35, No. 4, 151
composition is 87% hydrogen sulfide. (December, 1957).
Similar analyses can describe the 19. Thiel, A., and E. Schulte, Z. physilc.
behavior of any mixture in the low- LITERATURE CITED Chem., 96, 328 (1920).
temperature region. As directly indicated 1. Bierlein, J. A., and W. B. Kay, Znd.
on Figures 5 to 7 systems containing Eng. Chem., 45, No. 3, 618 (1953).
between 12.5 and 66 mole yo carbon 2. Donnelly, H. G., and 1). L. Katz, ibid., Manuscript received December 8, 195R; revision
receioed April 67, 1969; paper accepted April 67, 1969.
dioxide have triple points a t which vapor, 46, 511 (1954). Paper presented at A.I.Ch.E. Cincinnati meeting.

The Enthalpy of Water in the Liquid State

ROBERT BYRNE and GEORGE THODOS
Northwestern University, Evanfon, Illinois

The enthalpy of water in the liquid state has been calculated from 32°F. to temperatures
approaching the critical and pressures ranging from saturated conditions to 160,000 lb./sq.
in. abs. (approximately 11,000 atm). The results of this study are presented graphically
and show that the influence of pressure on enthalpy is significant, particularly In the lower
temperature region. At these conditions pressure is found to increase the enthalpy of
liquid water by a s much as 360 B.t.u./lb. above the corresponding enthalpy of the satu-
rated liquid state.
A comprehensive literature search disclosed PVT data for water that permitted the
construction of a density correlation. This correlation expressed in reduced coordinates
extends from the normal freezing point of water to temperatures of 1,870”F.(TR = 2.0)
and pressures ranging up to 10,915 atm. (PR = 50). The recent extensive PVT data
of Kennedy reported in 1950 supplemented with the earlier data of Amagat and Bridgman
allowed the calculation of enthalpies at these elevated temperatures and pressures. For
these calculations basic thermodynamic relationships were adapted which utilized this
reduced density correlation. This approach has made possible the extension of the thermo-
dynamic properties of liquid water above the highest pressure reported by Keenan and
Keys. Below this pressure of 6,000 lb./sq. in abs. good agreement was found to exist
between the enthalpy values presented by Keenan and Keyes and those reported in this
investigation.

ConsiderabIe information is presented substances are composed of electrically The hydrocarbons are characteristic of
in the literature on the thermodynamic symmetrical molecules which tend to this class of compounds for which thenno-
properties of nonpolar compounds. These behave as perfect gases or ideal solutions. dynamic properties can be readily pre-

Vol. 5, No. 4 A.1.Ch.E. Journal Page 551

dicted from generalized thermodynamic The results of this study permit the densities now can be defined in the
correlations. The prediction of these extension of the enthalpies reported by temperature region of 0.6 5 TR 5 1.35
thermodynamic properties for nonpolar Keenan and Keyes (16) from pressures to pressures as high as PR = 10 (2,183
substances depends upon the critical of 6,000 to 160,000 Ib./sq. in. abs. and atm.) and for the higher temperature
temperature and pressure and, as recently temperatures up to 650°F. region of 1.35 5 TR 5 2.00 up to pres-
proposed by Meissner and Seferian (Zd), sures of PR = 6.0 (1,310 atm.).
also upon the critical compressibility In Figure 1 the density correlation has
REDUCED-STATE DENSITY CORRELATION
factor. been presented on rectilinear coordinates
For nonpolar substances such as A comprehensive literature search for to accentuate the regions of interest
argon, nitrogen, and methane the critical density data of water has resulted in a for water in the liquid state. Reduced-
compressibility factors extend up to a compilation showing the dependence of state correlations of this type, as well
value of 0.291. Values of the critical com- density on both temperature and pres- as those expressed on logarithmic co-
pressibility factors for helium, hydrogen, sure. Experimental densities for the ordinates, are receiving considerable
and neon are yet higher (2, = 0.307); liquid and gaseous states including values attention in attempts to generalize the
however these substances fail to follow in the critical region have been con- PVT behavior of substances. In addition
the theorem of corresponding states, as sidered in this study. The critical con- to the correlating parameters of reduced
expected from their identical critical stants for water reported by Kobe and temperature and pressure, Lydersen,
compressibility factors. This abnormal Lynn (21) were used to calculate reduced Greenkorn, and Hougen (23) have
behavior is attributed to quantum effects temperatures, pressures, and densities. utilized the critical compressibility factor
which are significant for these three These critical constantsare T, = 647.4"K., as the third correlating parameter in
substances. Conversely polar substances P, = 218.3 atm., and pc = 0.32 g./cc. their studies involving the thermo-
have characteristically lower z, values; Since the majority of density data dynamic properties of substances.
for example, water and methyl alcohol could not be conveniently expressed
have z, values of 0.231 and 0.222, respec- as a function of reduced pressure, it THERMODYNAMIC TREATMENT
tively. The lowest z, value is reported became necessary to cross-plot the ori-
by Lydersen, Greenkorn, and Hougen ginal density data in order to obtain The enthalpy change of a system as a
(23) as 0.197 for hydrogen cyanide. reduced densities a t the appropriate function of temperature and pressure can
Other Substances reported in the litera- reduced pressures. Figure 1 presents be expressed in differential form as
ture have critical compressibility factors the final correlation of reduced density
ranging from 0.231 to 0.291. for water. In this compilation the satu-
The direct association of polarity and rated density data of Batelli (1, S),
the critical compressibility factor of Mendelejeff (15), Perot (d9), Waterston
substances is apparent, thus leading to (37), and Wullner and Grotrian (39) Equation (1) is equivalent to the thermo-
the qualitative generalization that sub- obtained before the turn of the century dynamic relationship
stances having high dipole moments are were found to be in good agrecnimt
always associated with low critical com- with the data of a number of inwsitg'ttors dH = C, dT
pressibility factors. Since water is charac- (8, 10, 15, 18, 35) who later rchported
teristic of substances having high dipole density data for the saturated ciiivclope.
moments ( p = 1.87 debye units) (SS), The experimental high-pressure data of
its thermodynamic properties cannot be Amagat (1) and Rridgman (5, 6) for
accurately developed from existing gen- liquid water cover the temperature range where
eralized correlations (9, 11). Therefore
water has been selected as typical of
those compounds which must be treated
individually for an evaluation of their
of -20" to 198°C. and account for
densities up to pressures of 160,000
lb./sq. in. abs. Of particular interest are
the data reported in 1950 by Kennedy
(s)= c ,
P

thermodynamic properties. This investi- (17) which permit the extension of the and
gation is concerned with the calculation
of enthalpy for water in the liquid state
a t elevated temperatures and pressures.
reduced density correlation into tempera-
ture and pressure regions where no data
were previously available. Conscquently
(5)v -
T
= T(%)
P

1. ENTHALPY
TABLE DIFFERENCES
FOR COMPRESSED
LIQUIDWATER(LOW-PRESSURE
REGION)
h - h,, Enthalpy difference, B.t.u./lb.
Pressure,
lb./sq. in.
abs. 32°F. 50°F. 100°F. 150'F. 200°F. 250°F. 300°F. 350°F. 400°F. 450'F. 500°F. 550OF. 600°F. 650°F.
1,000 2.9 2.8 2.6 2.4 2.1 2.0 1.9 1.3 1.0 0.3 -0.1
2,000 5.5 5.3 5.1 4.8 4.5 4.0 3.6 2.9 2.0 0.9 -0.6 -1.6 -1.8
3,000 8.1 8.0 7.7 7.1 6.8 6.1 5.6 4.4 3.2 1.5 -0.6 -2.9 -5.3 -7.4
4,000 10.9 10.8 10.1 9.7 9.0 8.2 7.5 6.0 4.4 2.1 -0.3 -3.9 -8.3 -14.6
5,000 13.6 13.5 12.9 12.1 11.3 10.3 9.3 7.7 5.7 3.0 -0.1 -4.7 -10.7 -20.1
6,000 16.2 16.1 15.4 14.7 13.7 12.7 11.2 9.3 7.0 4.1 0.2 -5.2 -12.2 -24.2
8,000 21.8 21.6 20.6 19.6 18.3 17.0 15.2 12.9 9.9 6.2 1.3 -5.3 -14.5 -30.1
10,000 27.1 26.9 25.7 24.4 22.9 21.2 19.1 16.5 13.0 8.9 3.1 -5.6 -15.2 -33.7
12,500 34.0 33.6 32.1 30.5 28.7 26.6 24.1 21.1 17.1 12.2 5.8 -2.9 -15.2 -35.6
15,000 40.6 40.1 38.3 36.6 34.4 32.1 29.1 26.0 21.4 16.1 8.9 -0.6 -14.3 -35.8
17,500 47. I 46.7 44.7 42.7 40.2 37.5 34.2 30.9 25.9 19.9 12.2 2.0 -12.7 -35.3
20,000 53.7 53.0 51.0 48.9 46.2 43.1 39.4 35.7 30.3 24.0 16.0 4.9 -10.2 -34. I
25,000 66.4 65.7 63.1 60.7 57.6 54.0 49.9 45.6 39.6 32.9 23.6 11.2 -4.9 -30.5
30,000 78.9 77.9 75.1 72.3 69.0 65.1 60.7 55.6 49.0 41.9 31.7 18.2 1.2 -25.2
35,000 90.9 89.9 87.1 84.0 80.2 76.2 71.7 65.7 58.8 50.5 39.9 25.9 7.9 -18.5
Reference state: h, = enthalpy of water at saturated-state conditions

Page 552 A.1.Ch.E. Journal December, 1959

 In order to utilize the density informa- 2 for pressures up to 35,000 Ib./sq. in. abs. pressures as high as 160,000 lb./sq. in.
tion presented in Figure l, the following and for temperatures up to 650°F. abs.
substitutions for pressure, temperature, Keenan and Keyes (16) prcsent enthalpy
and density were used: differences for liquid water a t pressures
ENTHALPY OF SATURATED LIQUID WATER
up to 6,000 Ib./sq. in. abs. Values re-
ri
1' = Pep,, 1 = T,YR ported by Keenan and Keyes and those The majority of heat-capacity data
resulting from this study are found to available are restricted to low tempera-
and be in agreement. Enthalpy differences tures. Rossini (28) points out that at
for pressures above 35,000 lb./sq. in. abs. present the data of Osborne, Stimson,
M
'li = - are presented in Table 2 for temperatures and Ginnings (27) are the most reliable
Pc PR ranging from 32" to 250°F. and for for the heat capacity of water up to
to produce Equation (3), which relates
the effect of pressure on enthalpy above
a reference state that has been selected
as the enthalpy of liquid water a t its
corresponding vapor pressure:

Jh. PC

Equation (3) pcrmits the calculation of

enthalpy changes due to pressure at
isothermal conditions. Integration of
Equation (3) requires the evaluation of
the function

f(PR, TR) = ( l / P R )
+ (TR/PR')( ~ P R / ~ T R ) P , *
Graphical differentiation utilizing the
chord-area method has been employed
to produce the slopes ( d p , / d T ~ ) p , from
Figure 1as functions of reduced tempera-
ture for constant parameters of reduced
pressure. The pressures considered ranged
from the saturated state to P R = 50
in the temperature interval 0.422 5
T R 5 0.6 and to PR = 10 in the tempera-
ture interval 0.6 <
TR <
1.20. Utilizing
this information, one could directly
obtain the dimensionless quantity

(~/PR) + (TR/PR')(~PR/~TR)P~
for the temperature interval 0.422 5
T R 5 0.95. The function f(pR, TR) did
not become negative until the reduced
temperature became TR = 0.85. These
negative values can be explained from
the fact that the negative contribution
Of the quantity ( TJPR') ( d P d d T R ) P R
becomes larger than the quantity ( l / p ~ )
for TR 2 0.85. With this information
Equation (3) was graphically integrated
for a number of constant temperatures
between the limits of the reduced vapor
pressure and reduced pressure. The
results of this procedure produced
enthalpy changes due to pressure above
the reference saturation pressure. For
these calculations the vapor pressures
compiled by Keenan and Keyes (16) T
Reduced Temperature, TR= -
have been used. Tc
The results of this study are presented
in Table 1 and also graphically in Figure Fig. 1. Reduced-density correlation for water (rectilinear coordinates).

Vol. 5, No. 4 A.1.Ch.E. Journal Page 553

212°F. Since temperatures up to 650°F. of enthalpies for the corresponding satu- particularly since the saturated enthalpy
have been considered in this study, their rated liquid state. The following repre- values for the liquid state have been
data were of restricted value. Because sentative values are summarized: obtained by an approach different from
of the temperature limitation of their that employed by Keenan and Keyes (1 6).
data, the constant-pressure heat-capacity Koch (20) The calculation of enthalpies for the
data of Koch (20) were utilized to estab- Keenan and 150 300 saturated liquid state with the data of
lish the enthalpy of water in the saturated "F. Keyes ( 1 6 ) atm. atm. Koch used has been carried out in order
liquid state up to 650°F. 200 168.0 167.6 167.2 to establish these values directly from
For this analysis the heat-capacity 400 375.0 374.7 373.5 data for the liquid state. The agreement
data of Koch at 150 and 300 atm. were 600 617.0 615.3 613.8 of these values and those reported by
graphically integrated to produce the Keenan and Keyes appears reasonable.
respective enthalpies above a reference The results produced from Koch's data Therefore the values reported by Keenan
state of 32°F. Enthalpy changes, h - h,, at 150 atm. are in closer agreement with and Keyes have been accepted to produce
for pressure effects, as obtained from the values reported by Keenan and Keyes the saturated enthalpy values presented
Figure 2, made possible the calculation (16).This comparison appears acceptable, in Figure 3.
The enthalpy differences for com-
pressed liquid water presented in Tables
1 and 2 have been consolidated with the
saturated enthalpy of liquid water to
IOC produce Figure 3. The isobars of 10,000,
20,000, and 30,000 lb./sq. in. abs. extend
to 650"F., while those of higher pressures
9c were limited to temperatures of 250°F.
Restrictions have been imposed on the
low-temperature range to account for
8C
the polymorphic behavior of water (IS)
in the formation of the different types
of ices.
7c
NOTATION

6C CP = molar heat capacity at con-

stant pressure, cal./(g.-mole)
(OK.)
5c h = enthalpy for liquid state,
B .t.u ./lb.
h, = enthalpy for saturated liquid
0 state, B.t.u./lb.
\
3 4c H = enthalpy
lil
M = molecular weight, g./g.-mole
u0) P = pressure, atm.
c 3c Pc = critical pressure, atm.
f
u
v PR = reduced pressure, P/Pc
'Tr
a = reduced vapor pressure
-02 2c R
T
= gas constant
= temperature, OK.
fC TC = critical temperature, OK.
w
- v)
IC TR
V
= reduced temperature, T / T ,
= volume, cc./g.-mole
rI
J=
VC = critical volume, cc./g.-mole
2, = critical compressibility factor,
0
P, V J RT,
I.L = dipole moment, debye units
P = density, g./cc.
- IC PC = critical density, g./cc.
PR = reduced density, p/p0

- 2c
LITERATURE CITED

1. Amagat, E. H., Ann. chim. et phys. (6),

- 3c 29, 505 (1893).
2. Batelli, Angelo, Acad. Mem. Torino (a),
41, 55 (1891).
-4c 3. _ _ , Ann. chim. et phys. (7), 3,
408 (1894).
4. Bauer, Gustav, Ann. Phys. Chern., 55,
184 (1895).
-5c
00 5. Bridgman, P. W., Proc. Am. Acad.
Arts Sci., 47, 439 (1912).
Pressure, psi0 6. Zbid., 66, 185 (1931).
7. Callendar, H. L., Phil. Trans. Roy. Soe.
Fig. 2. Effect of pressure on enthalpy of liquid water. (London),A212, l ( l 9 1 2 ) .

Page 554 A.1.Ch.E. Journal December, 1959

 TemDerature. F
Fig. 3. Enthalpy of liquid water.

8. -, Proc. Inst. Mech. Engrs. (Lon- 17. Kennedy, G. C., Am. J . Sci., 248, 26. Osborne, N. S., H. F. Stimson, and
don), 116, 507 (1929). 540 (1950). D. C. Ginnings, iMech. Eng., 57, 162
9. Edmister, W. C., Ind. Eng. Chem., 30, 18. Keyes, F. G., and L. B. Smith, Mech. (1935).
352 (1938). Eng., 53, 132 (1931). 27. ___ , J. Research Natl. Bur. Stand-
10. Havlicek, Jaroslav, and Ladislav Mis- 19. - , and H. T. Gerry, Proc. Am. ards, 23, 197 (1939).
kovsky, Helv. Php. Acta, 9, 161 (1936). Acad. Arts Sci., 70, 319 (1936). 28. Rossini, F. D., “Chemical Thermo-
11. Him, G. A., Ann. chim. et phys., 10, 20. Koch, W., Forsch. Gebiete Ingenieurw., dynamics,” p. 31, John Wiley, New
32 (1867). 5, 138 (1934). York (1950).
12. Hougen, 0. A., and K. M. Watson, 21. Knoblauch, Oscar, R. Linde, and H. 29. Perot, Alfred, Ann. chim. et phvs. ( 6 ) ,
“Chemical Process Principles,” p. 479, Klebe, 2. Ver. deut. Ing. 49, 1697 13, 145 (1888).
John Wiley, New York (1947). (1905). 30. Ramsay, William, and Sydney Young,
13. International Critical Tables, 4, p. 11, 22. Kobe, N. A,, and R. E. Lynn, Jr., Phil. Trans. Roy. SOC.(London), A183,
McGraw-Hill, New Yrok (1928). Chem. Revs., 52, 117 (1953). 108 (1892).
14. Jakob, Max, Engineering, 132, 684 23. Lydersen, A. L., R. A. Greenkorn, and 31. Smith, L. B., Mech. Ens., 50, 153
(1931). 0. A. Hougen, Rept. 4, Eng. Exp. Sta., (1928).
15. -,and W. Fritz, 2. Ver. deut. Ing., Univ. Wisconsin, Madison (1955). 32. ___ and F. G. Keyes, ibid., 52,
73, 629 (1929). 24. Meissner, H. P., and Ralph Seferian, 123 (1630).
16. Keenan, J. H., and F. G. Keyes, Chem. Eng. Progr., 47, 579 (1951). 33. Ibid., 53, 135 (1931).
“Thermodynamic Properties of Steam,” 25. Mendelejeff, D. I., Liebig Annal., 119, 34. - , Proc. Am. Acad. Arts Sci., 69,
John Wiley, New York (1936). 1 (1861). 285 (1934).
35. Trautz, Max, and Hans Steyer, Forsch.
Gebiete Ingenieurw., 2, 45 (1930).
36. Van Nieuwenburg, C. J., and H. B.
TABLE
2. ENTHALPY
DIFFERENCESFOR COMPRESSED
LIQUIDWATER Blumendal, Rec. trav. chim., 51, 707
(HIGH-PRESSUREREGION) (1932).
37. Waterston, J. J., Phil. Maq. (41,. ... 26,.
h - h,, Enthalpy difference, B.t.u./lb. 116 ( 1863j.
~

40,000 60,000 80,000 100,000 120,000 140,000 160,000 38. Wesson. L. G.. “Tables of Electric
Ib./sq. lb./sq. lb./sq. lb./sq. lb./sq. lb./sq. lb./sq. Dipole ’ Moments,” The Technology
t, “F. in. abs. in. abs. in. abs. in. abs. in. abs. in. abs. in. abs. Press, Mass. Inst. Technol., Cam-
bridge (1948).
32 102.7 149.9 194.6 238.0 279.9 320.8 360.2 39. Wullner, F. H. A. A., and Otto
50 101.8 148.8 193.3 236.6 278.3 319.2 359.2 Grotrian, Annal. Phus. Chern., 11, 545
100 99.0 145.5 189.6 232.1 274.0 314.9 355.8 (1880).
150 95.9 141.7 185.2 227.2 269.3 310.3 352.0
200 92.1 136.6 179.9 222.3 264.5 305.9 347.6 Manvscript received September 1 8 , 1958: reuision
250 87.8 131.2 174.1 216.7 259.1 301.1 342.8 received March 30, 1959; paper accepted April 8, 1969.

Vol. 5, No. 4 A.1.Ch.E. Journal Page 555