Sunteți pe pagina 1din 5

SPE 77386

Analysis of Black Oil PVT Reports Revisited

William D. McCain, Jr., Texas A&M University

Bo = (B oE )(B oSb ) at p>pb . . . . . . . . . . . . . . . . . . . . . (2a)

Copyright 2002, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the SPE Annual Technical Conference and
Exhibition held in San Antonio, Texas, 29 September –2 October 2002.

This paper was selected for presentation by an SPE Program Committee following review of and
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
Bo = ( BoD )
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at at p<pb , . . . . . . . . . . . . . . . . . . . . . (2b)
SPE meetings are subject to publication review by Editorial Committees of the Society of BoDb
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petrol eum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous are correct.
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
This paper presents a brief discussion of the laboratory
procedures, which will lead to the derivations of Eq. 2b and a
correct replacement for Eq. 1.
Combinations of data from two laboratory procedures, Laboratory Procedures
differential liberation and separator test, are used to determine
The two laboratory procedures that provide the necessary
values of oil formation volume factors and solution gas-oil data are the differential liberation (sometimes called
ratios for pressures below the bubblepoint pressure of the differential vaporization or differential distillation) and the
reservoir oil. The equation commonly used to calculate the
separator test (sometimes, incorrectly, called flash test). Each
solution gas -oil ratio is incorrect: the correct equation is of these laboratory procedures will be described briefly. What
derived. The equation used to calculate the oil formation will be seen is that each procedure starts with a quantity of
volume factors is correct; however, a derivation illustrating the
reservoir oil in the laboratory cell at its bubblepoint pressure at
underlying assumptions has not been published. This the reservoir temperature. In each procedure, gas is removed
derivation is also examined. in a sequence of flash vaporizations, with the resulting liquid
ending up at atmospheric pressure and 60°F
Introduction (standard conditions).
The bad news is that the equation usually used in the
petroleum industry to calculate solution gas -oil ratios from
Differential Liberation. Theoretically, the differential
black oil PVT reports, liberation starts with a liquid at some pressure and
temperature, then the liquid is partially vaporized, and each
R s = R sSb − (R sDb − R sD ) , . . . . . . . . . . . . . . . . . . (1) small increment of vapor is at once removed from the contact
BoDb with the liquid.1 Thus, the liquid is in equilibrium at any
instant with a small amount of vapor. This procedure is
is incorrectly formulated. somewhat tedious, so the petroleum industry emulates the
The good news is that the errors at high pressures, where differential liberation with a series of flash vaporizations in
solution gas -oil ratios are more often used, are not severe. A which a definite fraction of a batch of liquid is vaporized and
correct formulation of the equation will improve estimates of kept in intimate contact with the liquid until the gas is
solution gas -oil ratio throughout the full range of withdrawn at the end of each step in the series.
depletion pressures. In practice, the process starts with a sample of reservoir
Further good news is that the equations used to calculate oil at its bubblepoint pressure in the laboratory cell with the
oil formation volume factors from black oil PVT reports, temperature controlled at reservoir temperature. A dozen or
so steps (the exact number depends on the starting bubblepoint
pressure) consisting of flash vaporizations are carried out.
Each step starts with a pressure reduction at constant reservoir
temperature. This causes gas to be vaporized. The gas is

allowed to come to equilibrium with the liquid at the lower bubblepoint is the starting point of both processes and is
pressure and reservoir temperature, and then the gas is independent of the process. Table 2 shows this comparison
removed, and its quantity and specific gravity are measured. for the data of Table 1. The quantity of oil resulting from the
The volumes of the liquid remaining at the end of each step separator test is eighteen percent higher than the oil resulting
are determined. The last flash vaporization step ends at from the differential liberation. The quantity of gas resulting
atmospheric pressure. Then the temperature of the remaining from the separator test is fifteen percent lower than the gas
liquid is reduced to 60°F and the liquid volume is adjusted to from the differential liberation. And compositions and
maintain the pressure at 0 psig. This final liquid is called the properties of the resulting oils and gases differ significantly, as
residual liquid from the differential liberation or, simply, represented by the different API gravities of the oils and
residual oil. specific gravities of the gases.
The volumes of gas removed in all steps are added; this is
the amount of gas in solution at bubblepoint pressure. The gas Calculation of Oil Formation Volume Factors
volumes are decremented to determine the gas remaining in The ratio
solution at the pressure at the end of each step.
Finally, the volumes of gas in solution (in standard cubic 1 residual bbl
feet, scf) and the volumes of the oil in the cell (in reservoir BoDb res. bbl @ p b B oSb residual bbl
barrels, res. bbl) at the end of each step are divided by the = . . . . . . . . . . . (3)
volume of the residual oil (at atmospheric pressure and 60°F).
These are presented in the laboratory report as gas in solution, BoSb res. bbl @ p b
scf/residual barrel, and relative oil volume, reservoir
barrel/residual barrel. Other properties, such as reservoir oil provides a convenient relationship between residual barrels
density, incremental gas specific gravity, and gas from the differential liberation and stock-tank barrels from the
compressibility factor, are also measured and reported. separator test. The term reservoir barrel at bubblepoint
pressure (at reservoir temperature), which appears twice on the
Separator Test. The separator test starts with a sample of the left-hand side of Eq. 3, can be eliminated since both laboratory
reservoir oil at its bubblepoint pressure in the laboratory cell at procedures start with the reservoir oil in the laboratory cell at
reservoir temperature (same as the start of the differential its bubblepoint pressure at reservoir temperature.
liberation). A measured volume of the reservoir oil is expelled The change in volume of oil in the reservoir during
through a sequence of two (usually, though sometimes three) pressure depletion can be written as
flash vaporizations. In the first, at separator temperature and
pressure, the gas that was vaporized is removed, and the liquid BoSb
goes to the second flash at stock-tank temperature and BoSb − Bo = ( BoDb − BoD ) . . . . . . . . . . . . . . . . . . . (4)
atmospheric pressure. The volume of resulting liquid is
determined at atmospheric pressure and 60°F; this liquid is
The units of this equation are
usually called stock-tank oil and the volume is reported in
stock-tank barrels, STB, which also could be interpreted as
change in res bbl of oil  change in res bbl of oil  residual bbl 
standard barrels. =  
The volumes and specific gravities of the gases from the STB  residual bbl  STB 
two flash vaporizations are measured. The two gas volumes
are added, and the sum is reported as solution gas-oil ratio at Under the assumption that the change in volume (res. bbl) of
the bubblepoint, scf/STB. The volume of reservoir oil that oil in the reservoir is the result of a differential liberation
was expelled at the start of the test is divided by the volume of process, the “change” units in the numerator of each side of
stock-tank oil (at standard conditions) and reported as oil the equation are identical. Thus, the equation is correct.
formation volume factor at the bubblepoint, res. bbl/STB. The Eq. 4 can be rearranged as
density of the stock-tank oil is measured and usually reported
in °API. BoSb
Bo = BoD , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (2b)
Example Laboratory Data
Table 1 shows selected data from a differential liberation
which is the equation generally used in the petroleum industry
and a separator test for a black oil. The bubblepoint pressure
to calculate oil formation volume factors from differential
is 3043 psig at a reservoir temperature of 262°F.
liberation and separator test data.
The quantity and properties of oil and gas produced by the
Note that the use of Eq. 2b requires that the ratio of the
two procedures are different. However, this is difficult to see
volumes of residual oil to stock-tank oil remains constant (for
in the formats in which the data are presented. A comparison
a particular oil sample), regardless of the starting pressure.
of the results of the two procedures may be made by placing
The limited experimental data that are available show that this
the data on a basis of one barrel of reservoir oil at bubblepoint
is true.2
pressure and reservoir temperature. The oil in place at the

There are very limited data to test the results of Eqs. 2a The correct formulation is as follows. The equation must
and 2b against measured oil formation volume factors. calculate the gas remaining in solution in the reservoir oil at a
However, Dodson, et al.,2 provided one set of data. Dodson, pressure after pressure depletion from p b to some p, Rs ,
et al., proposed a laboratory procedure for determining oil scf/STB. Further, Rs should be the amount of gas to be
formation volume factors (and solution gas -oil ratios) that is liberated through a separator/stock-tank sequence.
generally considered more accurate than the differential RsSb is the gas originally in solution in the reservoir oil at
liberation/separator test procedure usually used and discussed its bubblepoint pressure as measured in a separator test, scf of
in this paper. Unfortunately their “composite liberation,” gas from sep. test/STB. RsDb –RsD is the volume of gas
although considered a superior method,3 requires a large liberated in the reservoir during a differential liberation from
sample of reservoir fluid and is very time consuming in the p b to p, scf of diff. lib. gas/residual bbl.
laboratory. Thus, it is not used in routine laboratory analysis. The ratio
But Dodson, et al., did provide one example of a routine
laboratory report and the results of a composite liberation for scf of gas from sep. test
the same black oil. Fig. 1 shows values of oil formation R sSb STB
volume factor calculated with Eqs. 2a and 2b compared with RsDb scf of gas from diff. lib.
the more accurate results of the composite liberation.
residual bbl
The small differences between the results of Eq. 2b and
the data from the composite liberation are approximately one
takes into account both the difference in the two oils, residual
percent, well within experimental accuracy. The composite
bbl/STB, and the difference in the two gases, scf of gas from
liberation is an entirely different laboratory procedure; the sep. test/scf of gas from diff. lib.
differential liberation/separator test discussed here is the
industry’s (less expensive) method of approximating the
composite liberation. R sSb
( R sDb − R sD )
Calculation of Solution Gas-oil Ratios
Eq. 1 is used commonly in the petroleum industry to combine  scf of gas lib. by diff. lib.   scf of gas from sep. test, residual bbl 
  
data from differential liberation and separator tests to calculate  residual bbl  scf of gas from diff. lib., STB 
solution gas -oil ratios. The validity of this equation can be
examined easily by rearrangement: is the gas differentially liberated converted to scf of sep.
gas/STB. The difference between the gas originally in
R sSb − R s = ( RsDb − R sD ) . . . . . . . . . . . . . . . . . . . (5) solution and the gas liberated during depletion from pb to p is
BoDb the gas remaining is solution at p.

R sSb
The units are R s = R sSb − (R sDb − R sD ) . . . . . . . . . . . . . . . . . . (6)
R sDb
gas liberated in sep test, scf  gas libera ted in diff lib, scf  residual bbl 
=  
STB  residual bbl  STB  Again, there is the assumption that the differential liberation
mimics the depletion in an oil reservoir, i.e., the gas remaining
in solution is that left after gas has been removed by
Eq. 5 shows that the gas volume liberated during the differential liberation.
separator test has been set equal to the gas volume liberated Eq. 6 can be rearranged into a simpler form
during the differential liberation. If the sources of the data are
not taken into account, the units, scf/STB, appear to be R sSb
correct. However, Table 2 shows that the gas liberated during R s = R sD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7)
R sDb
a separator test is significantly different in quantity and quality
from the gas liberated during a differential liberation. The
ratio BoSb /BoDb in Eq. 5 takes into account the differences in The use of Eq. 7 implies that the ratio of the gas liberated
the oils from the separator test and differential liberation, but by separator test to the gas liberated by differential liberation
the differences in the gases are ignored. be constant (for a particular oil sample), regardless of the
Thus, the material balance expressed in Eq. 5 must be starting pressure. The limited available data from Dodson, et
incorrect. It follows that values of solution gas-oil ratio al.2 show this to be true.
calculated with Eq. 1 must be in error! In fact, this is Again, the Dodson, et al.2 data are the only data available
illustrated every time the equation is used because calculated to use in comparing the results of Eq. 1 and Eq. 7 with
values of solution gas -oil ratio are generally negative at experimental solution gas -oil ratios measured at pressures
low pressures. below bubblepoint pressure of the oil.

Fig. 2 shows the results of Eqs. 1 and 7 using the data RsDb = Solution gas-oil ratio at bubblepoint
from the routine laboratory report compared with the more pressure measured in a differential
accurate results of the composite liberation. Eq. 7 gives liberation, scf from differential
values of Rs , which fit the composite liberation more closely liberation/residual bbl
than the results of Eq. 1. Further, the values of Rs from Eq. 7
converge to zero at 0 psig (as required by the definition of Rs ), References
while the values from Eq. 1 are negative at low pressures. 1. Dodge, B.F.: Chemical Engineering Thermodynamics,
The small differences between the results of Eq. 7 and the McGraw-Hill, New York (1944), 592.
composite liberation data in Fig. 2, as well as the good 2. Dodson, C.R., Goodwill, D., and Mayer, E.H.:
comparison of the results of Eqs. 2a and 2b in Fig. 1, show “Application of Laboratory PVT Data to Reservoir
that the simpler, less costly, routinely used PVT procedures Engineering Problems,” Trans., AIME (1953) 198, 287–
give results adequate for reservoir engineering work. 298.
The difference between the results of Eq. 7 and Eq. 1 in 3. Moses, P.L.: “Engineering Applications of Phase
Fig. 2 are not very dramatic. However, notice that the Behavior of Crude Oil and Condensate Systems,” J. Pet.
Dodson, et al.2 oil does not have much dissolved gas, with Rsb Tech. (July 1986) 38, 715– 723.
of about 600 scf/STB. Fig. 3 shows that the oil of Table 1,
which has about 1000 scf/STB originally in solution, shows a TABLE 1. SELECTED DATA FROM DIFFERENTIAL LIBERATION
much larger difference between the results of the two AND SEPARATOR TEST OF A BLACK OIL
equations, about thirty percent at a reservoir pressure of Differential Liberation at 262°F
1100 psig. Incre-
Conclusions Gas in Relative Oil Gas
Eqs. 2a, 2b, and 7 are the correct equations for calculating oil Pressure, Solution, Volume, Specific
psig scf/residual bbl res. bbl/residual bbl Gravity
formation volume factors and solution gas-oil ratios at 3043 1401 2.022
reservoir pressures below the bubblepoint pressure when 2900 1307 1.965 0.822
combining the data from differential liberation with separator 2600 1140 1.866 0.867
test data. 2300 998 1.783 0.858
The use of differential liberation data for these 2000 868 1.709 0.853
1700 751 1.644 0.856
calculations is nearly as accurate as the more costly composite
1400 642 1.584 0.861
liberation data and is certainly adequate for reservoir 1100 537 1.526 0.881
engineering calculations. 800 436 1.468 0.922
500 331 1.407 1.008
Nomenclature 263 235 1.339 1.197
Bo Oil formation volume factors at pressure, 143 176 1.290 1.432
0 0 1.111 2.194
res. bbl/STB 0 at 60°F = 1.000
BoSb = Oil formation volume factor at bubblepoint
pressure measured in a separator test, res. Gravity of Residual Oil = 43.2°API @ 60°F
bbl @ p b /STB
BoD Oil relative volumes at pressures less than SEPARATOR TEST
bubblepoint pressure measured in a Forma-
differential liberation, res. bbl/residual bbl tion
BoDb = Oil relative volume at bubblepoint pressure Volume Stock-
measured in a differential liberation, res. bbl Factor, tank Oil
Flash Pres- Temper- Gas-oil res. Gravity, Gas
@ p b /residual bbl Vapori - sure, ature, Ratio, bbl/ °API @ Specific
BoE = Oil relative volume at pressures above zation psig °F scf/STB STB 60°F Gravity
bubblepoint pressure measured in a constant Separator 100 70 935 0.775
mass expansion, res. bbl/res. bbl @ p b Stock
Tank 0 72 102 1.714** 48.0 1.309
Rs = Solution gas-oil ratios at pressures less than 1037*
bubblepoint pressure, scf/STB
*Solution gas-oil ratio at bubblepoint pressure
RsSb = Solution gas-oil ratio at bubblepoint
pressure measured in a separator test, scf **Oil formation volume factor at bubblepoint pressure
from separator test /STB
RsD = Solution gas-oil ratios at pressures less than
bubblepoint pressure measured in a
differential liberation, scf from differential
liberation/residual bbl

TABLE 2. THE VOLUMES AND PROPERTIES OF THE OIL AND Fig. 2 Comparison of solution gas-oil ratios calculated
GAS RESULTING FROM THE TWO LABORATO RY PROCEDURES from routine laboratory PVT data using Eqs. 1 and 7 with
data from composite liberation showing that Eq. 7 has a
Liberation Separator Test better fit.
Volume of oil at standard residual bbl STB
conditions at end of 0.495 0.583
res. bbl @ P b res. bbl @ P b 700
Gravity of oil at standard
conditions at end of 43.2°API 48.0°API 600
Volume of total gas at scf scf
692.9 605.0 500
standard conditions
liberated during process res. bbl @ P b res. bbl @ Pb

Solution gas-oil ratio, scf/STB

Weighted average specific 400
gravity at standard 1.0931 0.8252
conditions of total gas
liberated during process

Fig. 1 Comparison showing excellent fit of oil formation
volume factors calculated from routine laboratory PVT
data using Eqs. 2a and 2b with data from composite

0 500 1000 1500 2000 2500 3000 3500 4000
Reservoir pressure, psig
data from composite liberation calculated with Eq. 7 calculated with Eq. 1
Oil formation volume factor, res. bbl/STB

1.2 Fig. 3 Comparison of solution gas-oil ratios calculated

from routine laboratory PVT data using Eqs. 1 and 7,
showing that the deviation between the results is greater
1.15 when initial solution gas-oil ratio is large.



1.05 800
Solution gas-oil ratio, scf/STB


0 500 1000 1500 2000 2500 3000 3500 4000
Reservoir pressure, psig

data from composite liberation calculated with Eqs. 2a and 2b 200

0 500 1000 1500 2000 2500 3000 3500
Reservoir pressure, psig

Eq. 7 Eq. 1