Gases OF: A New Compres, Sibility Correlation For Natural and Its Application To Estimates Gas - In.Place
Gases OF: A New Compres, Sibility Correlation For Natural and Its Application To Estimates Gas - In.Place
Gases OF: A New Compres, Sibility Correlation For Natural and Its Application To Estimates Gas - In.Place
2642
ABSTRACT of natural gas and condensate reserves. psi a and temperature range from 80 to
It is used in the decline curve, the vol- 400°F. A summary of this comparison
This paper presents an evaluation of
umetric, and the material balance meth- is presented in Table I. The numerical
compressibility factor data and a dis- average deviation of the chart com-
ods of gas reserves estimation.
cussion of their application to the esti- pressibility factors from the experi-
mation of gas reserves. The behavior of gases at high pres- mental compressibility factors is 1.05
sures has been investigated extensively per cent, and the algebraic deviation is
A correlation is presented which pro-
in recent years and considerable data -0.28 per cent.
vides compressibility factors for use in
have been assembled on the compressi- The Katz correlation appears to be
both two-phase and single-phase hydro-
bility characteristics of a number of quite accurate for single-phase hydro-
carbon systems. Accuracies comparable
gases. Notable among the investigators carbon systems. It was based on single-
to those obtained previously for single- of high pressure gas relationships are phase systems and does not necessarily
phase systems only can be expected. A Kvalnes and Gaddy', Kay', Sage and provide a means of accurately predict-
simple means of predicting the presence Lacey" and Brown" Standing" and
4
ing the behavior of two-phase systems
or absence of a liquid phase in a con- Katz • Other contributors are Smith and which might be encountered in conden-
densate system of known composition is Watson", Roland and Kaveler', and sate work. Data published recently by
illustrated. Stevens and Vance". The data on com- Sage and Olds lO have enlarged the
pressibility factors assembled by these knowledge of phase behavior in con-
The correlation is based on 1,030
investigators have been presented and densate systems by covering extensively
compressibility determinations from 21
correlated in numerous ways; particu- the compressibilities of systems in the
hydrocarbon samples taken from eight larly noteworthy is the method first pre- two-phase as well as single-phase re-
oil fields. Of the data used, 75 per cent sented by Kay' which relates compres- gions for a wide range of temperatures,
were from California, 15 per cent were sibility factor "Z" to pseudo· reduced pressures and gas-oil ratios. An evalu-
from the Mid-Continent area, and 10 temperature and pseudo-reduced pres- ation of these data indicates that con-
per cent were from South America. The sure. Other investigators have used this ditions can arise. especially in conden-
average numerical deviation of the ex- empirical relationship successfully and sate systems having fairly low gas-oil
perimental data from this compressi- numerous charts covering a large num- ratios, in which two phases occur and
bility chart is 1.22 per cent. ber of gases have been constructed on the usual gaseous compressibility fac-
Charts and tables are included and this basis. tors do not accurately apply.
discussed which illustrate the errors in- It is the purpose of this paper to The measured PVT data of Sage and
volved through the misuse or nonuse evaluate compressibility factor data and Olds covering two-phase systems have
of compressibility factors in estimates to illustrate the importance of their been correlated with additional such
of gas-in-place. proper use in the estimation of gas-in- data from the Mid-Continent' area and
place for both gas and condensate South America; these are presented in
INTRODUCTION reservoirs. Fig. 1. This chart is the result of 1,030
The compressibility factor is a coeffi- compressibility* determinations from
cient which expresses the deviation of a EVALUATION OF COMPRESSI- 21 hydrocarbon samples taken from
gas of given composition from the Per- BILITY FACTOR DATA eight oil and condensate fields. Of the
fect G_as Laws. A factor such as this is The compressibility chart presented data used, 74.75 per cent were from
necessary in any calculation involving by Standing, Katz, Brown and Holcomb' California, 15.45 per cent were from the
volumes of gaseous mixtures, and finds in 1941 is prob~bly the most recent and Mid-Continent area, and 9.80 per cent
extensive application in the estimation reliable chart for natural gases. The were from South America. The chart
reliability of the Katz chart has been construction is based on the original
Manuscript received at office of Petroleum investigated by the writers for a fairly
Branch January 28. 1949. Paper presented at
AIME Annual Meeting in San Francisco, Feb- large number of gases (single-phase) * In thi~ llap€r the term "compressibility"
ruary 13-17, 1949. l'fiel's to the term "z" in the equation YV ==
1 References are given at end of paper.
in the pressure range from 500 to 10,000 ZNRT. fol' either sing-Ie- or two-phase systems.
compressibility correlation for natural Quite often in the use of the volu- recti on. The gas-in-place can be over-
gases developed by Katz et al'; jt pre- metric equation some of the terms and estimated by not using the equivalent
sents compressibility factors for both factors are not used or are overlooked. ratio. This is illustrated in Fig. 2 for
single-phase and two-phase systems (as It has been the practice in some cases systems of various gas-oil ratios. This
to use the compressibility of the pro- over-estimation can be as great as 15
determined by PVT measurements) .
duced gas at surface conditions and to per cent for systems of low gas-oil
Pseudo-reduced temperature and pres-
neglect any correction for the liquid ratio (4,000-5,000 cu. ft. per barrel).
sure and produced gas-oil ratio are re-
present. In the above equation, the The errors resulting from the com-
quired to use the chart. An evaluation
equivalent ratio (E.R.) makes this cor- plete omission of the compressibiUty
of the accuracy of this compressibility
chart is presented in Table II. The
numerical average deviation of experi-
TABLE I
mental data from chart values is 1.22
Comparison of Calculated and Experimental Compressibility Factors
per cent, and the alegbraic deviation is
(Katz Chart)
-0.31 per cent.
It is necessary to be cautious in the
No. of Total Average Maximum
application of any empirical correla- Points of
Per Cent Per Cent Per Cent
Gases Investigated Comparison
Error Error Error
tion involving systems as complex as --------------._------------_._.- - - - - - - - - - - - - - - ---
Kettleman Hills gas by Sage and Lacy9. + 3.28 +066 + 1.0
natural gases and condensates. The - 3.69 -0.53 -1.6
application of the correlation presented Rio Bravo gas by Sage and Reamer ll . . . o
21 - 15.20 -0.72 -1.5
here should be restricted to systems
Eight gases in Phillips Laboratories7 . 66 + 68.95 +1.04 +2.5
similar to those used in the correlation 72 - 56.84 -0.78 -2.5
until more extensive checks with other Four gases in Texas A. & M. Laboratories l5 .•. 24 + 30.14 +1.26 +38
45 - 9904 -2.20 -6.7
systems can be made. The properties _._-----------------------._--------- - - - - - - - - - - - - - - - - - - - - - -
l4
Sixteen gases by Standing and Katz . + 9.95 +1.42 +4.2
of the hydrocarbon systems on which - 15.23 -1.69 -5.6
the correlation is based are presented San Joaquin gas by Sage and OldslO 26 + 7.92 +030 +06
33 - 19.13 -0.58 -1.6
in Table III. Also, this correlation does 31 Gases total:
not cover pressures below 800 psia for Positive errors. . .................. . 128 +120.24 +094
-1.12
N egati ve errors ... 187 -209.13
systems having gas-oil ratios less than Numerical average .. 315 329.37 1.05
Algebraic average ... 315 - 88.89 -0.28
10,000 cu. ft. per barrel.
APPLICATION OF COMPRESSI-
BILITY FACTORS IN THE TABLE II
ESTIMATION OF GAS
Agreement of Experimental Data with Correlation (Fig. 1)
RESERVES
As stated before, the compressibility
No. of Total Average Maximum
factors of natural gas and condensate Points of Per Cent Per Cent Per Cent
Gases Investigated Comparison Error Error Error
systems are used extensively in the de-
San Joaquin Valley" (California)
cline curve, the volumetric, and the Wheatville Field (2 gases) . 67 + 52.67 +079 +203
93 -102.33 -1.10 -2.67
material balance methods of gas re-
North Belridge Field (4 gases) . 76 + 81.56 +1.07 +4.46
serves estimation. The misuse or non- 149 -163.69 -1.10 -2.81
use of compressibility factors in any of Helm Field (2 gases) .. 41 + 31.05 +0.76 +2.02
these methods results in sizeable errors 82 - 89.48 -1.09 -2.72
of estimation. No attempt will be made Paloma Field (5 gases) . ............. 108 +158.31 +1.47 +4.60
154 -238.63 -1.55 -3.98
here to cover the use of compressibility
Mid-Continent
factors in all methods of reserves esti- Cayuga Fiold' (3 gases). 40 + 48.29 +1.21 +3.95
105 -169.91 -1.62 -3.01
mation; instead this paper will deal
Field B (2 gases) .. 14 + 28.78 +2.06 +5.54
primarily with the volumetric method.
South America
Other authors3 have covered in some Field A (3 gases) . 59 + 68.86 +1.17 +5.05
measure the use of compressibility fac- 42 - 27.49 -0.65 -1.67
factor from the volumetric calculation Fig. 3 in the gas.oil ratio range from does not necessarily follow that all of
can be quite large. As illustrated in 6,000 to 20,000 are partly liquid at the gas·oil ratio systems shown occur at
Fig. 3, the per cent under·estimation of some of the depths shown; nevertheless, all the depths shown.
the gas· in· place becomes greater as the accurate results can be obtained for The use of compressibility factors in
gas·oil ratio of the system is decreased these systems by using Fig. I, if the
the volumetric equation does not com·
and can be as large as 50 per cent in a total system composition is known. The
pletely eliminate all sources of error.
6,000 gas·oil ratio system existing as curves presented here are primarily for
illustrative purposes; and although the As mentioned earlier, the Katz com·
two phases at 3,000 feet. Errors as large
as 20 per cent may exist for single· analyses of actual systems were used pressibility factors for gaseous systems
phase systems. The systems plotted in to calculate the data for the curves, it cannot always be applied to systems of
low gas.oil ratio where conditions are
such that a liquid phase is present at
the temperature and pressure consid·
TABLE III ered. Fig. 4 shows the per cent over-
Properties of Systems Used in Correlation (Fig. 1)
estimation of gas·in·place resulting
from using gas (single·phase) com·
Mol Mol Methane pressibility factors for systems that are
Gravity I Fraction Fraction Plus MollY!.
two· phase. Such errors would be mini·
System ~ umber Air=1 Methane Heptanes+ Heptanes+ Heptanes+
-----
-~-------
1. ...... 0.8611 0.8238 0.0292 0.8530 125.0 mized by using the chart presented in
2. .. 0.8181 0.8376 0.0228 0.8604 124.9
3 . .. .. . .. 0.9603 0.7964 0.0973 0.8937 107.7' Fig. l.
4. ......... 0.7520 0.8642 0.0353 0.8995 107.7'
5 .. ... 0.7089 0.8782 0.0225 0.9007 107.7' Fig. I has advantages over the nor-
6 .... .... 0.6720 0.8903 0.0115 0.9018 107.7'
7 ... ........ 0.8441 0.8074 0.0237 0.8311 125 8 mal single·phase compressibility charts
8 .. 0.8310 0.8119 0.0217 0.8336 125.7
9 .. 1.0170 0.7582 0.0576 0.8158 146.7 in that it can be used safely in both
10 .. .... 0.9585 0.7749 0.0476 0.8225 146.6
11 . ..... 1.1160 0.7155 0.0919 0.8074 137.1 single- (infinite gas·oil ratio) and two-
12 . .. . ..... 0.9955 0.7477 0.0659 0.8136 137.1
13 ••.. 0.8267 0.7930 0.0295 0.8225 137.1 phase regions. This chart can also be
14 .. . .... 1.3812 0.6058 0.1265 0.7325 137.6
15 .. 1.1750 0.6673 0.0882 0.7555 137.6 used with the Katz chart (included as
16. 1.0520 0.7041 0.0653 0.7694 137.6
17. .. . .. ... 1.4745 0.6340 0.1449 0.7789 139.8 part of Fig. I) to indicate the presence
18. ... .. ... 1.0700 0.7562 0.0749 0.8311 139.8
19 .. 0.8201 0.8316 0.0317 0.8633 139.8 of liquid in a system of known com·
20 .. ..... 0.7317 0.8788 0.0225 0.9013 131.0
position, at reservoir conditions. Any
21 ... . ..... 0.6996 0.8894 0.0143 0.9037 1350
marked difference in compressibility
factors from the two charts is a satis·
factory indication of liquid being pres·
ent in the system at the state point
Typical Analysis chosen. This is illustrated by example,.
given in Tables IV and V. It is believed
Mol that this usage of Fig. I will be of
Component Fraction
considerable value in indicating whether
Methane ........... .. 0.8238
Ethane. 0.0428 a condensate system is at or below dew
Propane .... 0.0351 point at reservoir conditions, in the abo
Isobutane. 0.0161
n-Butane. 0.0303 sence of complete PVT data. In the
Isopentane. 0.0060
n-Pentane. 0.0068 actual determination of dew points by
Isohexane, . 0.(034
n-Hexane .. 0.0065 this method, the disagreement between
Heptanes 0.0071
Octanes+ ... 0.0221 the two compressibility factors will not
1.0000 necessarily be zero at the dew point,
but will be within the accuracy of the
*Hexanes+
TABLE IV
Comparison of Calculated and Measured Specific Volumes for a Condensate Mixture
T=1600 F.
p T R=1.273
Gas·Oil Ratio = 6241 cu. ft./bbl
two correlations. It has been found for The authors wish to express their for their helpful suggestions and com-
a temperature of 150°F. this disagree- appreciation to the Magnolia Petroleum ments. Appreciation is also extended to
ment is 1.5 per cent at the dew point Company for permission to publish this Dr. D. L. Katz for use of certain por-
and for a temperature of 250°F. the paper, and to operating department tions of his graphical correlation of
disagreement is 0.5 per cent. The rela- personnel of Magnolia and its affiliates, pseudo-reduced properties vs. com-
tionship between temperature and per Socony-Vacuum and General Petroleum, pressibility'.
cent disagreement is essentially linear
between 150 and 250°F.
A sample calculation of gas-in-place
for an actual condensate system, using TABLE VI
Fig. 1, is presented in Table VI. Com-
parison of the result of this calculation
Example of Gas-in-Place Estimation Using Fig. 1 and the Katz Correlation
Given:
with one using single-phase compressi-
bility values shows that in the latter Reservoir Pressure, P = 2,500 psia
case a sizeable over-estimation of the
Reservoir Temp., T 150°F. =
Porosity, ¢ = 0.30
gas-in-place would have resulted. Ap- Interstitial Water Content, Sw 0.23 =
propriate adjustments must, of course, "PR (Total Product at Reservoir Cond.) 3.89 =
be made in the final 'reserve estimates p T R (Total Product at Reservoir Cond.) 1.25 =
for effect of unrecoverable liquid phases "PR (Produced Gas at Std. Cond.) 0.02179 =
I'Tn (Produced Gas at Std. Cond.) = 1.259
left in the reservoir, for two-phase Produced Gas·Oil Ratio = 6,241 cu ft/bbl
systems. Trap Gas/Day = 2,754,000 cu ft
Vent Gas/Day =
223,116 cu ft
CONCLUSIONS Tank Oil/Day = 477 bbl
Specific Vol of Tank Oil 0.021314 cu ft/lb =
The present work demonstrates the Molecular Wt of Tank Oil 120.06 =
feasibility of applying two-phase com- A. Using Fig. 1
pressibility factors to two-phase systems 43,560 x ¢ x (l-Sw) x 379.3 X Z, x P x E.R.
for purposes of gas-in-place estimations, Q= Z, RT
where representative composition data
are available. Accuracies comparable to Z, from Fig. 1 = 0.996
Z, from Fig. 1 = 0.720
those obtained in single-phase systems
E.R. = -;:;-2-;;-9;:;;::;-:;-:;-::-,+_2--,'(,:;::97;:;;o7-,-,I--:;I-c:C:;-:~=-:::-;-)
using single-phase compressibility fac- 6
tors can be expected. A simple means , 77,116 • 477 x 5.615 x 379.3
= 0.8823
of predicting the presence or absence (0.021314 x 120.06)
of a liquid phase in a condensate reser- 43,560 x 0.30 x 0.77 x 379.3 x 0.996 x 2,500 x 0.8823
voir is indicated, and the proper use of Q= 0.720 x 10.73 x 609.6
compressibility factors has been eval- = 1,781,300 cu ft gas/acre·ft
uated and illustrated.
B. Katz Chart
ACKNOWLEDGMENTS Z, from Katz Chart = 0.623
This paper originated as part of the 43,560 x 0.30 x 0.77 x 379.3 x 0.996 x 2,500 x 0.8823
work of the Subcommittee on Gas Re- Q= 0.623 x 10.73 x 609.6
serves of the AIME Production Tech-
nology Committee; helpful comments = 2,057,475 cu ft gas/ acre·ft
of members of that committee are here- . . 2,057,475 - 1,781,300
by acknowledged. Per Cent Over·EstlmatlOn = 1,781,300 x 100 = 15.6
TABLE V
Comparison of Single-Phase and Two-Phase Compressibility Factors for Condensate Systems in Which Liquid
Was Present at the State Point Chosen
•
,~
F
, Ii-'- Pt,....- f-t"'~
, , !-,' .... ~" Iq,;:~ k:c,
,
J.- ::.-- v~ f ~;§
,
I I I
rc:rc:
r:::: +- I i IJ. K .....("" i 1 ,~: I : H-
"';1 ! ,\'\~
t-f.-
t-
I t- .l~
'.J.-
I-t
,
'"T
...;.--
V-
/ V V-
I
1"1 ... I
I
T rtill Sifrl (IN
I l- ~
lII'll'llc
If
Lc: ~.~
!'t.
If
r-Zll
1
["F
,~-,
, ..-J--" , V
~-
~ I ~~ I : I
f'f'
,
'1= [c-;;jI ,'\ t'--.' Vi hi.", i I I
:~~ t'--. ......;. .l.- V-:, V~ I ! ! I i i l' 1
U I V "/' Wi' I i ! 1 I I bQh. I-n I
.: /: (IlL IIA 10,
" "- /v
~,
I
!" i
I _i_
!
I
' : "
I
I
;i
i I 1
Ii'
i H ; ,
.,'"
I I
: I ~~- ,
!"fIT' I i : ! I 1
I- i t--- i [., 0
h , , 1'-,
,
IT I I
~;. . I''''
1
:~
..,-.,; : ,.;;;
.~r j -:r. _....O':.l'.':r! ,~ I'''u r ..
::'!'~ !"
r
, 1--,.,.
I I
... 1
!
,
I
1
1 lC
i Ii,; I'", .T!
1
rR ,
; !
I
, I
I>-
I""P( jE'JUllE 1 I : I i : I ~
/I~I NG,A ~1'11 (Of 10"- ;Hi 1HI1-R ,"" ~"t. ')' fL ALI~N LH .~ t-:t--= Rt:- ~
r; ~"P 'r"t ',~,,: ).It ". "'fI:.~1J '/-" '''''F I ! lUI IK K t- t-r-:t-
! ,
" ...
,.
"
1"
I
FI.: +- t-
b. ,~I.-< br";j,,, i,,"~ r." ~';6 I""r. Or , f l" r- t-
tfh,
l- ,
i I
1
-r f I
, r-. t-- ;;
; I i
I 5
i I ' I
'"~ i I" ~.:>
i r
I' I I ,~~ ~Ot8 \
FIG. 1 - COMPRESSIBILITY OF HYDROCARBON SYSTEMS
(BASED ON ORIGINAL CHART BY D. L. KATZ)
. W:i:+'=
. r;:::T~' 'Li:S~: :Ji~~;±,~
Ie::
r~ r== ~¥~, FC
.~ -
.~
-+-.-<..-
.T...;::'