Density of Crude Oils Saturated with Natural Gas

By MARSHALL B. STANDING* AND DONALD L. KATZ,t MEMBER A.I.M.E.

(Los Angeles Meeting, October 1941)

DENSITY data are reported on IS saturated upon the gas-oil ratio, gas analysis, and crude
hydrocarbon liquids in the range of 35° to gravity is outlined.
250°F. and 1000 to 8220 lb. per sq. in. The ap-
parent liquid densities of methane and ethane INTRODUCTION
are shown to vary with the density of the sys-
tem in which they are present. A method is The densities of naturally occurring
proposeo whereby the densities of liquid hydro- liquid hydrocarbon mixtures are important

0.08

1&1 0.07

~
::;)
0
0.06

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

0
1&1
0
0
< 0.05 t'-.... "'" "'-- ~
.... 1-
!!l-
,:
i'-... ~ ......... '-
0-
I-~
..J

VI
0.04
I'--- r--- ' -r-- r--
-- --- ~ lBS,/SQ.IN.

-
0.03
z ....VI ............. t--- !---looo lBS./SQ. IN .
00:
-Q,
1-::1
~O
o:U
II:
00
UI-
0.02

0.0 ,r--
l-
-
r-- t---
I-- r--.
f--
l-
r- .s 000 l BS./50 .....
I
3000 lB:>./SO.IN.
2000 lBS./SO.IN. ~
1000 LBS. /SO.IN
~OO LB'SJSQ.IN.1 -
o
0.60 0.'5 0.70 0.75 0.80 0.85 090
DENSITY OF SYSTEM 1l0·F a. ATMOSPHERIC PRESSURE
FIG. I.-CORRECTION FOR COMPRESSmlLITY OF LIQUIDS.

carbon mixtures containing both methane and in many petroleum engineering computa-
ethane in solution may be computed at elevated tions. Calculation of the shrinkage of a
temperature and pressures within the accuracy subsurface sample of crude oil as the
of usual engineering computations. A method of natural gas is evolved is one example of
computing the shrinkage of crude oils based
the use of liquid-density data.
Sage, Hicks, and Lacey 5 have presented
Partial abstract of thesis submitted by M. B. a method of computing the density of
Standing to the Horace H. Rackham School of Gradu-
ate Studies, University of Michigan. Manuscript hydrocarbon liquid mixtures based on
received at the office of the Institute July 26, 1941.
Issued as T.P. 1397 in PETROLEUM TECHNOLOGY, partial molal volumes. Katz 1 has indicated
November 194I.
* University of Michigan; present address, Stand- a method based on the principle of additive
ard Oil Company of California, La Habra, California.
t Assistant Professor of Chemical Engineering,
University of Michigan, Lnn Arbor, Michigan. • References are at the end of the paper.

159
160 DENSITY OF CRUDE OILS SATURATED WITH NATURAL GAS

volumes of the components and using natural gas was transferred by mercury
apparent densities for methane and ethane. displacement at constant temperature and
The former method does not cover the pressure from the equilibrium cell into
necessary pressure and composition range, pycnometers of 25 c.c. capacity. The con-

0.0 9
~ t--..
a
'"Uz
0.0 8~
0.0 7t---..
~
""
~
-.......
-........
I'---- r-... ~
-- ---
r--!!~ --- ---
I'---- r-... ~
I- .

:2
1-111
0.0
&1-- " r--.... -........ ~O.' I-- I--- r-

--
mZ I'-- I--
::I~
111)( 0.0
51'--
"""---- t---

--- -r-- ~-- - - - r--

- --
'"m '"...
--r-- ~
-
0.0 t-- I--
e~
!I: 0.03
r-- I--
-
r-- ~.". -

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

~~
-l-
I-
0.02
:-- - 100·
u O
",I-
a: I
a:", 0.0 I eo"".
0::1
ua I
0 60·'"
0.80 0.85 0.70 0.75 0.80 0.85 0.90
DENSITY 0". SYSTE~ 60·'" & PRESSURE P
FIG. 2.-CORRECTION FOR THERMAL EXPANSION OF LIQUIDS.

while the latter is inaccurate because of the tents of the pycnometers were discharged
values of the apparent densities used. into a small topping column from which
This paper reports the density of 15 the components lighter than heptane could
saturated crude oils in equilibrium with
natural gas at pressures up to 8220 lb. per TABLE I.-Analysis of Typical Liquid
sq. in. The apparent-density method of 5330 Lb. per Sq. In. Abs., 120°F.
computing liquid densities has been revised Mol Weight
Component
l!nd applied to the data presented, giving a Fraction Fraction
maximum error of 4 per cent and an aver- ------------ ---
age error of 1.2 per cent of the experimental Methane... . . . . . . . . . . . . . . . . . .. 0.5641 0.1091
Ethane .......... "... .... ...... 0.0351 0.0127
liquid density. Propane................. . ..... 0.0137 0.0073
Normal butane ................ 0.0244 0.0171
Isopentane .................... 0.0247 0.02 1 5
DATA OBTAINED Normal pentane ................ 0.0326 0.0 28 4
Isohexane. .. . . . . . . . . . . . . . . . . .. 0.0188 0.0195
Normal hexane ................ 0.0246 0.0256
In the process of obtaining equilibrium Heptane and heaviera. . .. . . . . . .. 0.2620 0·7588
data on mixtures of natural gas and crude I. 000 I. 000

oils, the densities and analyses of the co-

a Density, grams per c.c., 0.881; molecular weight.
existing phases were obtained. The data 240.
were in the temperature and pressure
range of 35° to 250°F. and 1000 to 8220 lb. be distilled into a low-temperature column
per sq. in. The gas densities and methods for fractionation. The residue from the
of computation were reported in an earlier topping column and any material boiling
paper.7 above the hexanes were combined as
The liquid phase in equilibrium with the "heptanes and heavier." The pycnometers
 MARSHALL B. STANDING AND DONALD L. KATZ 161

TABLE 2.-Properties of the Saturated Liquids

Mol Fraction Density. Grams per C.C.

Run Pressure, Tempera- Mol

No. Lb. per Sq. ture. Wt .• Experimental,
In. Abs. Deg. F. C,+ C,+ C,+ at Tempera-
CR,
at 60°F. ture and
Pressure

A-4 1,600 120 0.2965 0.3468 196 0.850 0.702

A-I 3.185 120 0.4385 0.3186 240 0.863 0.696
A-2 5.270 120 0·5558 0.2760 279 0.904 0.745
A-3 8,220 120 0.6076 0.2565 356 0·954 0. 81 4
B-1 2.915 250 0.4158 0.3874 231 0.870 0.697
C-2 1,010 120 0.2234 0.4200 186 0.836 0.716
C-l 2.880 120 0.4850 0.2592 186 0.844 0.652
C-3 5.330 120 0.5641 0.2620 240 0.881 0.712
F-l 3.185 250 0·4470 0.3500 216 0.862 0.654
F-2 4.3 15 250 0.5 124 0.2590 256" 0.888 0.657
F-3 5.530 250 0.5766 0.2370 290 G 0·912 0.677
G-2 3.485 35 0·5700 0.2000 2IS/J 0.863 0.679
G-3 4.970 35 0·5799 0.2182 3 1 5" 0·903 0.766

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

D-l 4.330 120 0.5260 0.2560 268 0.893 0.731
E-I 4. 195 120 0·5178 0·3312 242 0.877 0.753

" Estimated from plot of specific gravity vs. mol weight.

were weighed before and after the liquid determined with small glass pycnometers
was discharged and any mercury accom- and the molecular weight was determined
TABLE 3.-Calculation of Density of Rio by the cryoscopic method, using water-
Bravo Trap Liquid saturated benzene as a solvent.

Density. Cubic TABLE 4.-Comparison Experimental and

Weight Grams Centi- Calculated Densities, Rio Bravo Trap
Component Fraction per C.C .• meters
60°F .• per Gram
Sample
Liquid
I Atm.
Density. Grams per C.C.
Methane. . . . . . . .. 0.01499 Pressure, Tempera-
Ethane ........... 0.01448 Lb. per Sq. ture, Deg. F.
Propane. . . . . . . . .. 0.03395 0.5 10 0.0665 In. Abs. Experi-
i-butane .......... Calculated
0.01151 0.5 64 0.0204 mental'
n-butane. . . . . . . .. 0.04860 0 . .5 84 0.0832
i-pentane ......... 0.01334 0.625 0.0213
n-pentane. . . . . . . .. 0.01649 0.631 0.0261
1,000 100 0.770 0.765
Hexanes and heav- 2,000 roo 0.778 0.770
ier. . . . . . . . . o. 84664 0.849 0·9972 3.000 roo 0.785 0·775
0·97053 1.2147 1,000 190 0·731 0.729
2,000 190 0·732 0·734
3.000 r90 0.745 0.745
Density of propane and heavier 0·97053 = 0.799
1.2 1 47
Weight fraction ethane in ethane and heavier =
0.0 1448
0.9850r = 0.or47 Analysis of a typical saturated liquid is
Fig. 4 for 0.799 density at 1.47 wt. per cent ethane given in Table 1. The absence of isobutane
gives 0.791 gram per c.c. for ethane and heavier.
Fig. 4 for 0.791 density at 1.50 wt. per cent methane is due to the method of blending a butane-
gives 0.776 gram per c.c. as density at 60°F. and one
atmosphere. Correction of 0.776 gram per c.c. by free natural gas with the Arkansas crude 2
Fig. I to density at 1000 lb. gives 0.781 gram per and varying quantities of the natural
c.c. Correction of 0.781 grams per c.c. from 60°F. to
100°F.• Fig. 2. gives 0.765 gram per c.c. gasoline,2 both of which were devoid of
panying the fluid to the topping column isobutane. Data on the density and proper-
was collected and weighed. The density of ties of the IS saturated liquids are given in
the heptanes and heavier fraction was Table 2.
 DENSITY OJ.' CRUDE OILS SATURATED WITH NATURAL GAS

EFFECT OF TEMPERATURE AND PRESSURE rate relationship covering most hyrlrocar-

ON LIQUID DENSITY hon liquids.
Any method of computing the density
TABLE 6.-Example Calculation of
of hydrocarbon liquids requires corrections
Shrinkage Factor Crescent Crude
to the density for pressure and temperature
Given:
changes. Sage, Lacey and coworkers on GaS liberated. . .. .• . . . • .• SI4 cu. ft. per bbl.
residual oil
Crude gravity (from U.S.
TABLE 5.-Comparison of Experimental and Bureau of Mines)...... 41.7° A.P.I.
Saturation temperature... 149°F.
Calculated Densities of Saturated Crude Saturation pressure. . . . .. 2575 lb. per sq; in. abs.
Oils GAS ANALYSIS

Density. Grams per C.C. Volume.

Component Per Cent
Pres- Tem- Devia·
Run Lb~r;~r pera.. tion Cal- Methane ........... , ............. . 75·33
No. culated Ethane .......................... .
Sq. In riure'F Experi- Ca\cu- 10.09
Abs. eg.. mental lated from Propane .......... ............... . 6.96
Experi- Butanes ............... .......... . 5·58
mental. Pentanes ..................... ... . 1.56
Per Cent Hexanes plus ..................... . 0.48

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

A-4 1.600 120 0.702 0.709 +r.oo CALCULATION OF SHRINKAGE
A'I 3.185 120 0.696 0.722 +4. 02
A-2 5.270 120 0.745 0.751 +0.80
A-3 8.220 120 0.8 14 0.8I! -0·37 Liquid.
Lb.~er
B-1 2.915 250 0.697 0·700 +0·43 Com- Cu. Mol. Cu. t •• I;ii'!-
ponent Ft. Wt. Pounds 60°F.
Cu.
C-2 1,010 120 0.716 0.713 -0.42 andI4·7 Ft.
C-I 2.880 120 0.652 0.649 -0.46 Lb. per
C-3 5.330 120 0.712 0.708 -0.56 Sq. In.
F-I 3.185 250 0.654 0.668 +2.16
--- --
F-2 4.315 250 0.657 0.664 +1.06 Methane .. 612. 16 25.8
F-3 5.530 250 0.677 0.681 +0.59 Ethane ... 82.0 30 6.5
Propane .. 56.6 44 6.6 3r.S 0.208
G-2 3.485 35 0.679 0.676 -0·44 Butanes .. 45·4 58 6.9 36.1 0.191
G-3 4.970 35 0.766 0·756 -1·30 Pentanes .. 12·7 72 2·4 39.2 0.061
Hexanes+ . 3.9 86 0.8 4r.5 0.019
D-l 4.330 120 0.731 0.713 -r. 64 Crude oil.. 5.61 286.0 51.0 5.610
E-I 4.195 120 0.753 0.736 -2.26 Total
---
335·0
--
6.089
- - - --- - - - - - - - - - - - - - Propane+ 302.7
Algebraic average ..................... +0.28 Ethane+ 309·2
Numerical average . ................... r. 16

Density of propane plus = 302

6 8. 7 = 49.81b. per cu. ft .
.0 9
A.P.I. Project No. 37 have amassed con- = 0.799 grams per c.c.
siderable data on the effect of temperature Weight per cent ethane = ~ = 2.1
309·2
and pressure on the specific volume of a
Weight per cent methane = 25·8 = 7.7
great variety of liquids. A survey of these 335·0
and other data on liquid densities indi- From Fig. 4:
Density of ethane plus = 0.789 gram per c.c.
cated that pressure corrections at 60°F. Density of methane plus = 0.712 gram per c.c.
From Figs. I and 2 :
were essentially functions of the liquid Density of system at 60°F. and 2575 lb. = 0.727
gram per c.c.
density and that the thermal expansion of Density of system at 149°F. and 2575 lb. = 0.686
gram per c.c. = 42.8 lb. per cu. ft.
liquids was little affected by pressure. Volume of one barrel of crude and accompanying gas
Fig. I was prepared from the mass of at 149OF . an d 2575 lb . = 335·0
42.S = 7.82 cu. f t.
data in the literature, and gives an addi- Volume of dissolved gas = 7.82 - 5.61 = 2.21 cu. ft.
tive correction to the density for increases Shrinkage = 2'621 = 39.4 per cent based on residual
5. I
in pressure. Minor differences occurred oil.
Shrinkage reported by Lindsly = 40.7 per cent.
between the density changes with pressure
for materials of the same density but were For computing the thermal expansion
not regular enough to permit a more accu- under pressure, the atmospheric pressure
 MARSHALL B. STANDING AND DONALD L. KATZ

corrections may be used,4 or Fig. 2, which methane and ethane dissolved in hydro-
is based primarily upon Sage and Lacey's carbon liquids is a function of the system.
data. Calculation of the apparent liquid density

...,
z O. 6
-<
...,...
J:

...o .-dd-
O. S
1,;-11
)0
... . -:Ii-
---
-u p
"'u
~ ~!---'"'+
Z,
1&.1 •
0 2
I!I 0.4
... ....r""-
...,z ..c""
a:
:
II.

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

-< 0.3

1&.1
Z
0.4
.,
'Q.
«
...,... 9~~
J:
iQ.
2
...o ./
V Q Q
'Q.
0.3 0
o~
)ov ...... 0
II

~o...
... u
iii': .J.. ~
~
Z ::l e
~I!I
0.2
...Z V.6A 6 .J.
/ .6
1&.1
a: 7
"
• ETHANE: - N BUTANE
~
II.
< O. I
A
t ETHANE - HEPTANE
ETHANE- CRYSTAL. OIL.
e METHANE-CYCL.O HEXANE
Q METHANE -BE NZENE
"0- METHANE-PENTANE
0 METHANE-HEXANE
~ METHANE-HEPTANE
METHANE- PROPANE
"
'Q. METHANE-CRYSTAL. OIL.
9 METHANE- CRUDE OIL.

0.3 0.4 0.5 0.6 0.7 0.8 0.9

DENSITY OF SYSTEM 8"°,,. & ATMOSPHERIC PRESSURE
FIG. 3.-VARIATION OF APPARENT DENSITY OF METHANE AND ETHANE WITH DENSITY OF THE SYSTEM.

ApPARENT DENsrrmS of methane and ethane after correcting

m' METlIAN~; AN)) ETHANE the density of the mixture to 60°F. and one
It is known that the rule of additive atmosphere pressure gave the relationship
volumes applies very well for pentane and shown in Fig. 3. Since the systems did not
hexane, an!! that the apparent density of contain both methane and ethane, the lise
164 DENSITY OF CRUDE OILS SATURATED WITH NATURAL GAS

of the liquid densities of the other constit- assumes that the presence of the ethane in
uents at 6o°F.l and one atmosphere left systems containing methane does not
the density of either methane or ethane change the methane curve of Fig. 3.

O.P ~ V/
,~ t\t " / 0 /
./
~~
t ~~p.~ / 10/
~
<0
..,
~t
~ '1''' / / V
Cit~,,~t%
I O.IJ
/' /
/' ~ . /
::)
~ IIt~ f, II ' /
...
t\f>~\'~~0~~V30
./
~
0.7
/ ./
~
....
... ~~~~~ /
/" ~
... .....-
I:)

IA ~ ~ ~ ~ 40 .....
....... V .-

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

~
0.6
..,!:::
~ ld ~ ~ ~ ~ -:::: so- l---

~ ~ ~ l8 t:::-
Q V
0.5
0.4 0.5 0.6 0.7 0.8 0.1/ 1.0
O£NSITY OF PROPAN£ PLUS - GM. / CC.

0.9
~
,....
~
<0
0.8
I
~ ~~
.,~ "t
.,
)...
0.7
tf/.
(t
10.. oJ5
I:)

)... ~
fl
,....
VI 0.6
~
:g

D£NSITY or £THAN£ PLUS - GM. /CC.

FIG. 4.-DENSITIES OF SYSTEMS CONTAINING METHANE AND ETHANE.

as the only unknown. Fig. 4 is a plot of CALCULATION OF LIQUID DENSITY

the densities of systems containing methane The method of computing a liquid den-
and ethane as a function of the density of sity is illustrated in Table 3, using the Rio
the remainder of the system and concen- Bravo trap liquid analysis. 6 The weights of
tration of the volatile hydrocarbon. This components heavier than ethane are con-
 MARSHALL B. STANDING AND DONALD L. KATZ

verted to volumes by dividing by the passed from the reservoir to the stock tank.
specific gravity or density of the component Table 6 presents an example calculation
at 60°F. and one atmosphere. The volume of the shrinkage factor for the Crescent
and weight of this propane plus fraction pool crude, using the data presented by
gives its density at 60°F. and one atmos- Lindsly.3
sphere. Fig. 4 gives in turn the density REFERENCES
including the ethane and then the methane. 1. D. L. Katz: Trans. A.I.M.E. (1938) 127. 159.
The resulting density at 60°F. and one 2. D. L. Katz and C. C. Singleterry: Trans. A.I.M.E.
(1939) 132. 103.
atmosphere is a fictitious value, since the 3. B. E. Lindsly: Petro Eng. (1936) 7. series.
4. National Standard Petroleum Tables. Nat. Bur.
liquid cannot exist at these conditions. Stds. eire. C-410.
5. B. H. Sage, B. L. Hicks and W. N. Lacey: Amer.
The correction for pressure must be made Petro Inst. Drill. and Prod. Practice (1938) 402.
6. B. H. Sage and H. H. Reamer: Trans. A.I.M.E.
at 60°F. by Fig. r before correcting for (1941) 142. 179.
thermal expansion by Fig. 2. 7. M. B. Standing and D. L. Katz: p. 140, this volume.
Table 4 shows the computed and experi-
mental densities of the Rio Bravo trap DISCUSSION
liquid. 6 Table 5 gives similar information (Carlton Beat presiding)

Downloaded from http://onepetro.org/TRANS/article-pdf/146/01/159/2177640/spe-942159-g.pdf/1 by guest on 19 April 2025

on the gas-saturated crude oils of Table 2. M. G. ARTHUR, * Santa Fe Springs, Calif.-
It should be noted that these liquids con- The accurate prediction of the volumetric
tained from 22 to 60 mol per cent methane. behavior of naturally occurring hydrocarbon
The maximum deviation between the mixtures has been restricted by a lack of an
calculated and experimental densities is abundance of published data on the properties
4.0 per cent, with a numerical average of such systems at high temperatures and
deviation of 1.r6 per cent. Since none of pressures. Although the densities of the gas and
these data were used in the apparent liquid phases were predictable to a fair degree
density correlation for methane and ethane, of accuracy at states well below the critical
temperatures and pressures, the methods and
this agreement should be typical for crude
data did not permit an accurate prediction
oils.
of densities at conditions near the criticals.
Since many of the recently discovered petro-
ApPLICATION TO SHRINKAGE CALCULATION
leum reservoirs, and probably many of those
Among the calculations requiring a to be discovered in the future, will exist at
knowledge of the density of the liquid these high temperatures and pressures, any
phase, or reservoir fluid if no gas cap is data correlating the volumetric behavior at
such conditions are a valuable addition to the
present, are those involving the estimation
literature.
of reserves. The usual method of relat- The methods reported by Messrs. Standing
ing the volume of the crude oil as measured and Katz give results that check well with
at the stock tank to the volume that it experimentally determined values and the
occupied with its dissolved gas at the methods are especially valuable if the results
reservoir conditions is by means of a are as accurate for crude oils of different sources
shrinkage factor. This factor may be meas- having different compositions of the hydro-
ured on a subsurface sample but may be carbons of greater molecular weight than the
computed from the analysis of the gas hexanes.
evolved from the reservoir liquid as it * Union Oil Company of California.