Kobayashi 1953 VLE Propane Water
Kobayashi 1953 VLE Propane Water
Kobayashi 1953 VLE Propane Water
sn-water Systems
RIKI KOBAYASHIl AND DONALD L. KATZ
University of Michigan, Ann Arbor, M i c h .
W O R
water
K systems was initiated by Scheffer in 1913 on the hexme-water ( 3 1 ) and the pentane-water ( 5 2 ) systems. His studies
were primarily confined to the phase relations of the systems in
the region bounded by the vapor pressure curve for the pure
hydrocarbon and the conditions under nhich three phases coexist
-a vapor phase, a lighter liquid phase rich in hydrocarbon, and a
heavier liquid phase rich in water. Scheffer found that as the
temperature of the hydrocarbon-water mixtures under the coexistence of three phases was increased, a temperature was reached a t
which the intensive properties of the vapor phase and the lighter.
liquid phase became continuously identical. The conditions
under which this critical phenomenon occurs have been referred to
as the three-phase critical conditions and the composition of
the identical phases as the three-phase critical composition ( 1 9 ) .
The solubility of methane in water has been reported a t 25' C.
by Michels, Gerver, and Bijl ( 2 1 ) up to a pressure of 6800
pounds per square inch and a t the same temperature by Frolich
et al. ( 1 2 ) up to a pressure of 2060 pounds per square inch.
The studies of the saturated water content of hydrocarbon gases
were undertaken in connection with the dehydration of natural
gases to prevent the formation of natural gas hydrates. Lauhlere
and Briscoe ( 1 6 ) and Deaton and Frost ( 9 ) determined the saturated concentration of water in natural gases up to 100" F. and
600 pounds per square inch. These data on natural gases were
supplemented by those of Russell, Thompson, Vance, and Huntington (26) for pressures up to 2000 pounds per square inch.
Sage, Lacey, and coworkers have determined the dew point
composition of the methane-water (a$) and the ethane-water
( 2 4 )systems from 100" to 460" F. up to 10,000 pounds per square
inch. Reamer e l al. (25) have also determined the concentration
of water in the n-butane-water system in the vapor and the lighter
liquid phases in the three-phase region. Similar data on the
concentration of water in the vapor and the lighter liquid phases
for the propane-water system have been reported by Poettman
and Dean ( 2 3 ) . Data on the water content of a natural gas
saturated with water and the solubility of the same natural gas
in water and in brine solutions up to 250' F. and 5000 pounds per
square inch have been presented by Dodson and Standing (10).
McKetta and Kat2 (19, 2'0) made an extensive study of the
methane-n-butane-water system in the two -phase and threephase regions, and developed general plots for the estimation of
the solubility of water in hydrocarbon gases and liquids (20).
Culberson, Horn, and McKetta ( 6 ) reported data on the solubility of ethane in water up t o 1200 pounds per square inch and
methane in water a t 77" F. up to 10,000 pounds per square inch.
Culberson and McKetta (7, 8) have recently extended their data
on the solubility of methane in water and ethane in water over the
temperature range from 100' t o 340' F. up to 10,000 pounds
per square inch.
I n this investigation vapor-liquid equilibria data were obtained
in the two-phase and three-phase, regions for the propane-water
system a t temperatures from 54' to 300' F. and pressures from
100 to 2800 pounds per square inch. The system has been studied
with relative thoroughness, especially in the region surrounding
205.7' F. and 637 pounds per square inch, the three-phase critical
Present address, Department of Chemical Engineering, The Rice Instit u t e , Houston, Tex.
1
440
0-150 OR
0-1000 PSI
TO T R A I N
TO BURETTE
F R O M MERCURY S O U R C E
February 1953
priate positions in the cell were uced for charging, sampling, and
discharging the contents of the cell. In Figure 1, V , L,,and F . 1
designate the vapor, lighter liquid phase, and the heavier liquid
phase sampling ports, respectively. The difierent sampling
ports provide a means for sampling each phase through a different
port located in its own phase to avoid contamination of the
sample lines by a second phase.
The vapor phase sampling line, V-T4-16, was wrapped with
a n electrical rasistance heater in order to superheat the vapor
sample leaving the cell. It is especially important that the con?
densation of the vapor phase be prevented when sampling in the
three-phase region, as complete condensation of that phase could
occur with a very slight cooling below the equilibrium temperature. The resulting change in phase would cause the loss of
water from the sample in the sample lines. Heating the vapor
s a m p h g line also hastened the purging of any excess water remaining in the lines from previous runs.
i'
I
A
D
R
D
Figure 2.
E
G
H
NEEDLE VALVE,VAPCR
NEEDLE VALVE, LIQUID
ASOARITE TOWER
D E H m R l T E TOWERS
WATER-PROOFED
CONNECTIONS
TRAP
SETTLING BULB
V
VACUUM
MERCURY BUBBLER
MERCURY MANOMETER
M CALIBRATED BOTTLE
N
THERMOMETER
0
G L A S S CASE
P
WIRE GUARD
U
DRYING TUBES
PUMP
J
_ _
44 1
mosphere, it was necessary to prove the effectiveness of Dehydrite as a drying agent a t reduced pressures. No loss in
weight of previously used drying tubes could be detected after
the pressure within the drying tubes had been maintained a t
less than 0.1 mm. of mercury absolute pressure for 12 hours.
Figure 3 shows the apparatus utilized in the analysis of the
water-rich liquid phase.
t- I'I
EXPERIMENTAL MEASUREMENTS
The vapor and liquid phases were expanded from the equilibrium pressure to the pressure of the analytical train a t valves 16
and 12, respectively. The expanded sample was passed through
two U-tubes in series and the dehydrated propane gas was
measured in a calibrated glass bottle. Dehydrite (magnesium
perchlorate) was used as the dehydrating agent. The gain in
weight of the U-tubes, the pressure rise of the gas in the bottle
and the temperature of the gas collected in the bottle were recorded. The compressibility factors of propane gas a t atmospheric and lower pressures were applied in the calculation of the
phase composition from the observed data.
Because the dehydration of the propane was conducted a t
pressures ranging from less than 0.1 mm. of mercury to 1 at-
'
442
00040
PROPANE-WATER
,012
1
SYSTEM
140
160
,010
00036
W
4
z
008
5 ,006
$00032
goo028
IL
Io
E
& 00024
004
-I
1
0
,002
.00020
40
40
60
Figure 4.
TABLEI.
100
120
140
160
180
200
EXPERIMEXTAL
DATA IN THREE-PHASEREGION
63.0
64.1
83.1
87.0
102.5
111.6
111.8
127.7
128.2
144.1
144.1
170.0
179.0
179.2
187.6
188.6
190.3
192.8
193.0
201.6
201.7
Pressure
Lb./Sq. Inch' Abs.
Composition of
Vapor Phase, Mole
Fraction Water
60
80
100
120
TEMPERATURE,
2;
Temp.,
O F.
'
80
Figure 5.
'E
I80
200
TABLE
11. EXPERIMENTAL
DATAI N TWO-PHASE
REGION
(Concentration of water in propane-rich phases)
Propane-Rich
CompositionPhase
of
Temp.,
Pressure
O F.
Lb./Sq.Inoh' Abs.
Mole Fraction Wate;
102
100
0.00954
141
100
0.00492
0.00696
533
100
818
100
0.000587
1015
100
0.000540
0.000673
2018
100
2023
100
0.000591
0 000643
2798
100
2798
100
0.000623
I
150
150
150
150
150
150
150
146
217
289
856
1564
2484
2803
0.02649
0.01702
0.01200
0.002328
0.002224
0.002060
0.002046
190
190
190
190
190
190
190
309
465
1342
2003
2803
0.04967
0.02910
0.01722
0.00502
0.00475
0.00463
205.7
205.7
205.7
205.7
205.7
205.7
205.7
205.7
205.7
207
445
603
637
696
755
1217
2023
2803
0.0637
0.02639
0.01568
0.00998
0.00813
0.00779
0.00726
0.00674
0.00640
230
230
230
230
230
230
230
230
252
446
630
718
823
1530
2023
2704
0.08094
0.04325
0.02659
0.02028
0.01366
0.0133
0.00959
0.00919
260
260
260
260
260
260
260
260
439
675
812
1411
2023
2023
2755
0.07856
0.04660
0.03599
0.02499
0.01843
0.01646
0.01490
0.01649
300
300
300
300
300
300
622
928
1217
1441
2023
2803
996
~ .
0.10146
0.06360
0.04496
0.03875
0.03105
0.02684
'
February 1953
443
2.303 R T
Figure 6.
PRESSURE, PSlA
Figure 7.
MATERIALS USED
EXPERIMENTAL RESULTS
The experimental vapor-liquid equilibria data for the propanewater system are presented in the three-phase region from 54' F.
to the three-phase critical of 205.7' F. and 637 pounds per square
inch absolut,e in Table I and Figures 4 and 5. The experimental
data in the two-phase region from 100 to 2800 pounds per square
inch are presented at isotherms of loo", 150, 190', 205.7',
230, 260, and 300' F. for the propane-rich phases in Table I1
and Figure 6, and at isotherms of 54", loo', 133", 190', 205.7",
230', 260, and 300' F. for the water-rich phase in Table I11
and Figures 7 and 8. The graphically smoothed data in the
three-phase region for all three phases are presented in Table IV.
INDUSTRIAL A N D E N G I N E E R I N G C H E M I S T R Y
444
TABLE
111.
EXPERIMEXTAL
DATAI N TWO-PHASE
REGION
54.0
54.0
54.0
54.0
617
617
1222
1988
0.0002963
0.0003029
0.0002906
0.0002948
100
100
100
100
100
100
100
133.0
133.0
133.0
133.0
72
117
428
622
1531
2012
2687
0.0000863
0.0001552
0.0002046
0.0002178
0.0602151
0.0002239
0.0002304
188
1199
1810
2787
0.0001499
0.0002249
0.0002267
0.0002364
181
307
O.OOO1146
0.0001763
190
190
190
190
131
224
359
990
1523
2012
2787
205.7
205.7
205.7
205.7
205.7
205.7
230
400
478
910
1810
2787
230
230
230
230
230
230
222
504
810
1128
1810
2787
2 60
260
260
260
260
260
170
332
511
751
1232
1810
2787
0 . WOO796
0.0001330
0.0001960
0,0002580
0.0002703
0.0002745
O.M)02880
0.0001261
0,0002162
0.0002449
0.0002842
0.0003008
0.0003134
0,0001301
0.0002633
0.0003243
0.0003414
0,0003542
0.0003765
0.0001001
0.0001964
0.0002892
0.0003665
0 0004197
0.0004393
0.0004766
300
300
300
300
300
300
300
265
471
694
987
1565
1810
2787
0.0001666
0.0003031
0.0004130
0.0004876
0.0005801
0.0006078
0 0006861
Temp.,
170
170
100
190
190
260
PRESSURE, PSlA
February 1953
TEMPERATURE, DEG.
Figure 9.
445
0
Figure 10.
TABLE
VI. GRAPHICALLY
SMOOTHED
DATAI N
Pressure,
Lb./Sq.
Inch. Abs.
100
200
300
400
500
600
637
700
800
900
1000
1500
2000
2500
3000
446
60
100
130
0.0002873
O.OOO2670
0.0002679
0.0002687
0.0002696
0.0002703
0.0002706
0.0002711
0.0002717
0,0002722
0.0002728
0,0002748
0.0002760
0.0002768
0.0002775
0.0001273
0.0002045
0.0002060
0.0002076
0.0002091
0.0002106
0.0002112
0.0002122
0.0002137
0.0002152
0.0002166
0.0002221
0.0002264
0.0002287
0.0002302
0.0000876
0.0001603
0.0002003
0.0002043
0.0002075
0.0002090
0.0002108
0.0002124
0.0002145
0.0002164
0.0002182
0.0002228
0.0002276
0.0002307
0.0002355
TWO-PHASE
Temperature, O F.
160
190
205.7
Mole Fraction Propane in Water-Rich
0.0000609
0.0000698
0.0000582
0.0001303
0.0001172
0.0001197
0.0001700
0.0001782
0.0001696
0.0002132
0.0002138
0.0002107
0.0002400
0.0002210
0.0002468
0.0002530
0.0002680
0.0002261
0.0002550
0.0002271
0.0002723
0.0002295
0.0002578
0.0002767
0.0002319
0.0002602
0.0002815
0.0002623
0.0002341
0.0002846
0.0002358
0.0002642
0,0002865
0.0002415
0.0002715
0.0002957
0.0002462
0.0002783
0,0003043
0.0002507
0.0002835
0.0003116
0.0002542
0.0002877
0.0003160
REQIOY
220
Phase
0.0000562
0.0001180
0.0001708
0.0002173
0.0002550
0.0002798
0.0002878
0.0002952
0.0003053
0.0003115
0.0003155
0.0003257
0.0003357
0.0003445
0.0003526
250
280
310
0.000513
0.0001170
0.0001768
0.0002307
0.0002767
0.0003107
0.0003222
0.0003375
0.0003562
0,0003695
0.0003787
0.0003995
0.0004144
0.0004293
0,0004442
0.0000415
0.0001160
0.0001857
0.0002480
0.0003000
0.0003453
0.0003602
0.0003828
0.0004087
0.0004287
0.0004444
0.0004927
0.0005246
0.0005502
0.0005753
0.0000220
0.0001147
0.0001923
0.0002677
0.0003277
0.0003803
0.0004007
0.0004290
0.0004627
0.0005133
0.0004886
0.0006160
0.0006858
0.0007332
0.0007703
(Vapor-Liquid Equilibria f o r
Binary Hyd rocarbon-Water Systems)
CORRELATION OF DATA
ROCEDURES for correlating data are useful when they permit the prediction of the behavior of systems for which data
are not available or when they may be used to extrapolate experimental data. The water contents of hydrocarbon-rich
phases are correlated, whether the phase be gaseous, liquid,
or in the fluid region.
The hydrocarbon-rich portion of the hydrocarbon-water system
may consist of a liquid phase and a gas phase, or of a single phase.
The proposed correlation uses temperature and the molal volume
of the hydrocarbons in the phase as variables and in this manner
avoids reference to the number of phases in the system and to
pressure.
Several correlations have been made of the saturated water
content a t high pressures of gases such as nitrogen, hydrogen,
methane, ethane, and complex hydrocarbon mixtures ( I , 20,
22, 24). These methods, however, do not treat the vapor phase
and the hydrocarbon-rich liquid phase as continuous phases.
The simplest relation expressing the saturated water content
in the vapor phase, which is generally valid a t low pressures for the
nonpolar gases, is that
Y = P/P
(1)
where y = mole fraction of water in the vapor a t the temperature of the system
p o = vapor pressure of water at the temperature of the
system
P = total pressure of the system
Equation 1 results from the use of Daltons law of partial
pressure and Raoults law, and the liquid phase is substantially
pure water. Under conditions of temperature and pressure
where the vapors no longer behave ideally, Equation 1 leads to
serious error. Saddington and Krase ( 2 7 ) have found that a t
high pressures the saturated water content of nitrogen greatly
exceeds the value predicted from the application of Poyntings
relationship (18),and this has been found to be true for aqueous
hydrocarbon systems ( 2 2 ) . That the relationship cannot explain the increased volatility of water with increased pressure