Documente Academic
Documente Profesional
Documente Cultură
net/publication/235961260
CITATIONS READS
0 3,994
1 author:
Lateef A. Kareem
Cyphercrescent
15 PUBLICATIONS 12 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Halley Class solvers for non linear system of equations View project
All content following this page was uploaded by Lateef A. Kareem on 19 September 2016.
a r t i c l e i n f o a b s t r a c t
Article history: Natural gas engineering entails production, processing, storage and transportation of natural gas. A good
Received 26 November 2013 handling of the gas requires a sophisticated understanding of how its density, compressibility, pseudo-
Received in revised form pressure and specific heat capacity vary with the gas condition. A variety of methods have been pre-
15 April 2014
sented in petroleum and gas journals and a host of scholastic materials to evaluate other properties over
Accepted 17 April 2014
Available online
a wide range of temperatures, however, the available correlation for isobaric specific heat capacity is for
only 150 F. We generated 200 samples of natural gas mixture with methane component ranging from
0.74 to 0.9985 using normally distributed experimental design. The variations of the respective specific
Keywords:
Specific heat capacity
heat capacity of the components and the effect of composition on the specific gravity and overall specific
Isentropic expansion heat capacity of the gas were taken into consideration. The developed correlation reads in the specific
Processing gravity and temperature to generate the ideal gas specific heat capacity of the sample. The result yielded
Storage and transportation system design 99.75% accuracy at 150 F when compared to experimental data, as against the result from isentropic
coefficient method which overestimated the ideal gas specific heat capacity by 25% at the same tem-
perature. The ideal gas specific heat capacity developed compared to 6000 data points generated from
mixing rule at different temperatures resulted in correlation regression coefficient of 0.9999. To account
for the deviation from ideal gas behaviour, this work presents 99.7% R squared value for dimensionless
residual specific heat capacity as a function of reduced temperature and pressure compared to that
calculated from Starling Carnahan equation of state. This model is the first explicit correlation for the
residual specific heat capacity of natural gas to be derived.
Ó 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jngse.2014.04.011
1875-5100/Ó 2014 Elsevier B.V. All rights reserved.
L.A. Kareem et al. / Journal of Natural Gas Science and Engineering 19 (2014) 74e83 75
DH ¼ nG C v DT (3) Table 1
Constants of the degree three fit of Cp as a function of temperature (Cengel and
Boles, 2002).
or
Compounds a b c d
ZT2 Methane (C1) 19.89 5.024 102 1.269 105 11.01 109
DH ¼ nG Cv dT (4) Ethane (C2) 6.900 17.27 102 6.406 105 7.285 109
Propane (C3) 4.04 30.48 102 15.72 105 31.74 109
T1 n-Butane (C4) 3.96 37.15 102 18.34 105 35.00 109
i-Butane (C4) 7.913 41.60 102 23.01 105 49.91 109
The expression remains the same except for the isochoric spe- n-Pentane (C5) 6.774 45.43 102 22.46 105 42.29 109
cific heat capacity which replaces its isobaric counterpart. The i-Pentane (C5) 6.774 45.43 102 22.46 105 42.29 109
majority of petroleum and gas journals and textbooks seem to have n-Hexane (C6) 6.938 55.22 102 28.65 105 57.69 109
neglected the necessity of a correlation for specific heat capacity of Heptaneþ (C7þ) 9.118 73.20 102 37.98 105 76.47 109
Carbon 22.26 5.981 102 3.501 105 7.469 109
natural gas. The only available correlation is for isentropic exponent dioxide (CO2)
k as a function of specific gravity gg at 150 F from which the Nitrogen (N2) 28.90 0.1571 102 0.8081 105 2.873 109
isobaric specific heat can be calculated as follows:
2:738 log gg heat capacities of the two mixtures over a temperature range of
k150 F
¼ ; 0:55 < gg < 1 (5)
2:328 50e1550 C as shown in Fig. 1.
But,
3. Methodology
Cp
k ¼ (6)
Cv Certain properties, including pseudo-critical temperature and
pseudo-critical pressure, are calculated as the sum of partial
also, pseudo-critical pressures and temperatures of the components of
natural gas sample (Abou-kassem & Dranchuk, 1975; Farzaneh-
Cp ¼ Cv þ R (7)
Gord et al., 2010; George and Brown, 1944). These are elucidated
in the following mathematical expressions:
Where R is the universal gas constant ¼ 8134 J/kmol K
X
n
Cp Ppc ¼ yi Ppci (9)
k ¼
Cp R i¼1
1 R X
n
¼ 1 Tpc ¼ yi Tpci (10)
k Cp
i¼1
R k1 Moreover, the specific gravity can be expressed in the same way.
¼ For instance, consider:
Cp k
X n Xn
MWgas 1 MWi
so that, gg ¼ ¼ yi MWi ¼ yi
MWair MWair i ¼ 1 i¼1
MWair
kR
Cp ¼ (8) Therefore,
ðk 1Þ
For the fact that Eq. (8) only gives the value of the isobaric X
n
Table 2
Constants of linear regression of ideal gas isobaric specific
heat capacity degree three coefficients against specific
gravity.
Constants Estimate
a1 10.9602
a2 25.9033
b1 2.1517 101
b2 6.8687 102
c1 1.3337 104
c2 8.6387 105
d1 3.1474 108
d2 2.8396 108
Cp ¼ 10:9602gg þ 25:9033 þ 2:1517 101 gg 6:8687
102 T þ e1:3337 104 gg þ 8:6387 105 T 2
kJ
þ 3:1474 108 gg 2:8396 108 T 3 ;
kmol K
0:55 < gg < 1 and 100 K < T < 1500 K
(19)
Fig. 1. Comparison of the specific heat capacities of (80% CH4, 20% C2H6) and (90% CH4,
In field units,
10%C3H8).
specific gravity, because it will carry the information on the mole Cp ¼ 8:0211gg þ 3:3359 þ 2:0744 102 gg 4:2441
fraction of the constituent gases.
The following procedures were applied in carrying out this: 103 T þ 8:1528 106 gg þ 4:8536 106 T 2
þ 1:2887 109 gg 1:1626 109 T 3 ðBTU=lbmol RÞ;
1. Qualitative and quantitative analyses of the natural gas samples
to identify individual constituents and their respective mole 0:55 < gg < 1 and 280 F < T < 2240 F
fractions. (20)
2. Evaluation of specific gravity of the natural gas samples.
3. Evaluation of the coefficients (a, b, c and d) of the degree three fit It is important to note that Eq. (19) has been presented in
for the samples using the degree three fit coefficients of the graphical form in Fig. 6 and this has made it easier than calculation
constituents and their mole fractions. for a quick estimate.
4. Generation of linear fit for the coefficients as a function of
specific gravity of the samples.
5. Combining the results to form the required expression.
3.1. Results
a ¼ a1 gg þ a2 (14)
b ¼ b1 gg þ b2 (15)
c ¼ c1 gg þ c2 (16)
d ¼ d1 gg þ d2 (17)
X
n X
n
a ¼ yi ai ¼ 18:126; b ¼ yi bi ¼ 8:3407 102
i¼0 i¼0
X
n X
n
c¼ yi ci ¼ 7:9480106 ; d ¼ yi di ¼ 6:1263109
i¼0 i¼0
T ¼ 150 F ¼ 324:33 K
Cp ¼ a þ bT þ cT 2 þ dT 3
Cp ¼ð18:126Þþ 8:3407102 ð324:33Þ
þ 7:9480106 ð324:33Þ2 þ 6:1263109 ð324:33Þ3
kJ
¼44:133
kmolK
Therefore,
To test the validity of the result, the Cp obtained at 150 F using
the correlation in Eq. (19) was compared with the result obtained kR 1:17612 8:314 kJ
from Eq. (13) for the test sample shown in Table 3. Cp ¼ ¼ ¼ 55:499
k1 1:17612 1 kmol K
The specific gravity of the sample is evaluated thus:
Substituting 0.708451 for gg and 324.333 K for T in Eq. (19),
therefore:
MWgas 1 X n kJ
gg ¼ ¼ y MWi ¼ 0:708451 Cp ¼ 44:242
MWair MWair i ¼ 1 i kmol K
This result implies an error of 0.25%.
The constants (a, b, c and d) of the ideal gas isobaric specific heat
capacity are calculated as follows:
Fig. 4. Least square fit for specific heat capacity constant c. Fig. 5. Least square fit for specific heat capacity constant d.
78 L.A. Kareem et al. / Journal of Natural Gas Science and Engineering 19 (2014) 74e83
Table 3 130
Test sample data.
1400
Compounds Mole fraction 120
Methane (C1) 0.8833
Ethane (C2) 0.0270 110
Propane (C3) 0.0203 1200
n-Butane (C4) 0.0174
i-Butane (C4) 0.0106 100
n-Pentane (C5) 0.0096
1000
Temperature (K)
i-Pentane (C5) 0.0087
90
n-Hexane (C6) 0.0082
Heptaneþ (C7þ) 0.0068
Carbon dioxide (CO2) 0.0052 80
Nitrogen (N2) 0.0029 800
70
By using the chart in Fig. 6, the specific heat capacity of the ideal 600 60
gas is also estimated as 44 (kJ/kmol K).
50
400
4. Validation of the ideal gas specific heat capacity correlation
40
To validate this correlation, the plot of the ideal isobaric specific 200
heat capacity obtained from the mixing rule shown in Fig. 7 was 30
compared with that obtained from Eq. (19) shown in Fig. 8. In Fig. 9, 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95
a cross plot of isobaric specific heat capacity by mixing and corre- Specific gravity (Air = 1)
lation shows the nearly unit correlation of the regression coeffi-
Fig. 7. Contour plot of the specific heat capacity by mixture law.
cient. The statistical analyses in Table 4 denote that the standard
deviations of the constants are of 102 order less than the value of
the constant. The narrow 95% confidence intervals and the P-values
indicate that the probabilities that these constants could have been
Fig. 6. Specific heat capacity chart as a function of temperature and specific gravity.
L.A. Kareem et al. / Journal of Natural Gas Science and Engineering 19 (2014) 74e83 79
vHreal
Cp; real ¼
vT
vðHideal þ Hresidual Þ
Cp; real ¼
vT
vHideal vHresidual
¼ þ
vT vT
vHresidual
¼ Cp; ideal þ
vT
vHresidual
Cp; residual ¼
vT
From thermodynamics of residual properties, we know that:
The ratio of the residual specific heat capacity to the molar gas
5. Correction for the effect of pressure constant is the dimensionless residual specific heat capacity.
Using change of variable by putting:
An ideal gas maintains a constant specific heat capacity
regardless of its pressure as long as temperature is constant. But
P ¼ Ppc Ppr
natural gas is far from ideal (Abou-kassem & Dranchuk, 1975; Abou-
Kassem and Dranchuk, 1982; Dranchuk and Quon, 1964; Farzaneh- Therefore,
Gord et al., 2010; George and Brown, 1944; Goodwin, 1961; Dran-
chuk et al., 1976; Seifarthi and Joffe, 1952; Weissl and Joffe, 1957;
0 1
ZPpr
Yorizane et al., 1983). Therefore, the deviation is accounted for by Cp; residual v B 2 vZ dPpr C
¼ @T A
the residual specific heat capacity which is the temperature de- R vT vT Ppr
rivative of the residual enthalpy. To obtain the residual specific heat 0
capacity, the widely accepted Hall and Yarborough z factor Performing the derivative, we have:
ZPpr ZPpr 2 !
Cp; residual vZ dPpr 2 v Z dPpr
¼ 2T T
R vT Ppr vT 2 Ppr
0 0
T ¼ Tpc Tpr
Therefore,
ZPpr ZPpr !
Cp; residual vZ dPpr 2 v2 Z dPpr
¼ 2Tpr Tpr (21)
R vTpr Ppr vTpr
2 Ppr
0 0
APpr
z ¼
y
2
Fig. 9. Cross plot of the mixture Cp against the correlation Cp for an ideal gas. where A ¼ 0:06125te1:2ð1tÞ and y is the root of:
80 L.A. Kareem et al. / Journal of Natural Gas Science and Engineering 19 (2014) 74e83
Table 4
Statistical parameters of the constants of correlation of the ideal gas isobaric specific heat capacity.
Ppc 10.960 1.516 101 {11.259, 10.661} 72.289 8.849 10144 0.9637
a2 25.903 1.123 101 {25.681, 26.125} 230.617 1.294 10241
b1 2.151 101 7.779 104 {2.136 101, 2.167 101} 276.144 5.567 10257 0.9974
b2 6.869 102 5.772 104 {6.983 102, 6.755 102} 118.991 1.970 10185
c1 1.333 104 5.033 107 {1.344 104, 1.324 104} 264.539 2.564 10253 0.9972
c2 8.639 105 3.727 107 {8.565 105, 8.712 105} 231.286 7.329 10242
d1 3.147 108 1.343 1010 {3.121 108, 3.174 108} 234.249 6.028 10243 0.9964
d2 2.839 108 9.953 1011 {2.859 108, 2.820 108} 285.275 9.315 10260
dA 2 2
y þ y2 þ y3 y4 ¼ 0:06125e1:2ð1tÞ þ 0:147tð1 tÞe1:2ð1tÞ
APpr þ By2 þ CyD ¼ 0 (22) dt
ð1 yÞ3
dB
¼ 14:76 19:52t þ 13:74t 2 ;
where dt
dC dD
B ¼ 14:76t 9:76t 2 þ 4:58t 3 ; C ¼ 90:7t 242:2t 2 þ 42:4t 3 ; ¼ 90:7 484:4t þ 127:2t 2 ; ¼ 2:82
dt dt
D ¼ 2:18 þ 2:82t
Tpc 3. Using the values of y and vy/vt from Eqs. (24) and (25), v2y/vt2 is
t ¼
T computed from:
ZPpr !
Cp; residual v2 Z dPpr
¼ t 2 d2 A d2 B 2 d2 C D d2 D D dC dD D
R vt 2 Ppr P y þ 2 y þ 2 Cy ln y þ 2 y ln y
0 pr
dt 2 dt 2 dt dt dt dt
2
We have: dD dB dC D1
þ ln y CyD þ 2 2 yþ Dy
! dt dt dt
"
vZ dA y
dt
A vy
vt dD D1 vy 8 þ 20y 4y2
¼ Ppr þ Cy ðD ln y þ 1Þ þ
vt y2 vt
dt ð1 yÞ5
0 #
2 2 1 vy 2
vy
y A vvt2y 2 dA vy
2
d Ay2 þ 2A 2B þ CDðD 1ÞyD2
2
v Z B dt 2 vt dt vt C vt
¼ B
@
CPpr
A " #
vt 2 y3 2
1 þ 4y þ 4y2 4y3 þ y4 D1 v y
þ 2By þ CDy ¼0
ð1 yÞ4 vt 2
Hence,
0 (26)
2 2 1
d Ay2 þ2A vy y A vvt y2 þ 2 dA vy
2
ZPpr
Cp; residual B dt 2 vt dt vt C
¼ t 2 B CdPpr d2 A
R @ y3 A 2
¼ 0:294ð1 tÞe1:2ð1tÞ 0:147tð1 tÞe1:2ð1tÞ
2
0 dt 2
2
2 3
2 6
a1 ea2 ð1tÞ Ppr t a3 Ppr t 7
7
2 3 3 5;
dA dB dC dD D a7 þ a6 Ppr t þ a5 Ppr t þ a4 Ppr t
Ppr y2 þ yD þ Cy ln y
dt " dt dt dt # 1:2 Tpr 3 and 0:01 Ppr 15
1 þ 4y þ 4y2 4y3 þ y4 D1 vy
þ 2By þ CDy ¼ 0 (27)
ð1 yÞ4 vt
(25) where
L.A. Kareem et al. / Journal of Natural Gas Science and Engineering 19 (2014) 74e83 81
The values for the constants in Eq. (27) are contained in Table 6. 6.2. Example 2
The excellent performance of Eq. (27) is reflected in the comparison
of its result to that of Sterling Carnahan as indicated by 99.9% cor- To calculate the amount of heat to be extracted from a 0.6
relation of regression coefficient (Table 7). gravity natural gas at 5000 psi in order to bring its temperature
The specific heat capacity of natural gas can thus be written as: down from 600 F to 200 F.
In order to demonstrate how the model can be applied, the ZT2 "
following examples are provided. Please note that the solutions to 50 X
4
DH ¼ Cp; real dT ¼ 10:6121 þ 4 Cp; real ð2iÞ
Example 1 are provided after each problem, while the solution to 3
i¼2
T1
Example 2 is provided in subsection 6.2.1. #
X
3
þ2 Cp; real ð2i þ 1Þ þ 18:0886
6.1. Example 1
i¼1
Table 6
Statistical parameters of the constants of correlation of the residual isobaric specific heat capacity.
7. Conclusions with those obtained from natural gas enthalpy chat (George and
Brown, 1944) and can be used in the design of gas transportation
A robust correlation for isobaric specific heat capacity has been system or any other application where the isobaric specific heat
developed. This correlation is in two parts. The first is the ideal gas capacity of natural gas is a requirement.
isobaric specific heat capacity given by Eqs. (19) and (20). The
second part is the residual of the isobaric specific heat capacity
given by Eq. (27). Eq. (19) has a correlation of regression of 0.9999
and a maximum relative error of 0.04. It is applicable to samples
with specific gravity in the range of [0.55, 1.0] and temperature in
the range of [100, 1500] K. Eq. (27) has a regression coefficient of
0.997 and maximum relative error of 0.05 compared to the Starling
Carnahan equation of state. It is applicable when the pseudo
reduced pressure and temperature fall within [0.01, 15] and [1.1, 3],
respectively. An exhaustive regression analysis of the constants of
the correlations developed was also conducted. The P-values
(max ¼ 0.022628) and 95% confidence interval show that the
probability that these constants could have been derived by chance
is negligible. This correlation gives results in excellent agreement
Table 7
Result of statistical analysis of least square fit for the residual specific heat
capacity.
References
Abou-kassem, J.H., Dranchuk, P.M., 1975. Calculation of Z factors for natural gases
using equations of state. J. Can. Pet. Technol 14 (3). http://dx.doi.org/10.2118/75-
03-03.
Abou-Kassem, J.H., Dranchuk, P.M., 1982. Isobaric heat capacities of natural gases at
elevated pressures and temperatures. Proc. SPE Annu. Tech. Conf. Exhib. http://
dx.doi.org/10.2523/10980-MS.
Canale, R.P., Chapra, S.C., 2010. Numerical Methods for Engineers, sixth ed. McGraw
Fig. 12. Plot of real gas isobaric specific heat capacity with temperature. Hill, New York.
Cengel, Y.A., Boles, M.A., 2002. Thermodynamics: an Engineering Approach, fourth
Table 8 ed. McGraw Hill.
Specific heat capacity of a sample gas at different temperatures (200 T 600). Dranchuk, B.P.M., Quon, D., 1964. A General Solution of the Equations Describing
Steady State Turbulent Compressible Flow in Circular Conduits, pp. 60e65.
i Temp F Cp, real i Temp F Cp, real Dranchuk, P.M., Purvis, R.A., Robinson, D.B., 1976. Computer calculation of natural
BTU R1 lbmol1 BTU R1 lbmol1 gas compressibility factors using the standing and Katz correlation. Pet. Soc.
CIM.
1 200 14.5496 6 450 13.0952 Farzaneh-Gord, M., Khamforoush, A., Hashemi, S., Namin, H.P., 2010. Computing
2 250 13.7025 7 500 13.2611 thermal properties of natural gas by utilizing AGA8 equation of state. Int. J.
3 300 13.2430 8 550 13.4827 Chem. Eng. Appl. 1, 20e24 http://dx.doi.org/10.7763/IJCEA. .V1.4.
4 350 13.0402 9 600 13.7430 George, B., Brown, G., 1944. A Series of Enthalpy-entropy Charts for Natural Gases.
5 400 13.0094 Goodwin, R.D., 1961. Apparatus for Determination of Pressure-Density-Temperature
Relations and Specific Heats of Hydrogen to 350 Atmospheres at Temperatures
Above 14 K, p. 65.
Acknowledgement Seifarthi, J.H., Joffe, J., 1952. Isobaric heat capacity of propane. J. Ind. Eng. Chem. 44,
2894e2897.
Sutton, R.P., 1985. Compressibility factor for high- molecular- weight reservoir
The authors wish to thank the two anonymous reviewers for gases. In: 60th Annual Technical Conference and Exhibition of Society of Pe-
their constructive criticism, suggestions and advice. troleum Engineers.
Trube, A., 1957. Compressibility of natural gases. J. Pet. Technol 9. http://dx.doi.org/
10.2118/697-G.
Nomenclature
Weissl, A., Joffe, J., 1957. Generalized correlation for effect of pressure on isobaric
heat capacity of gases. J. Ind. Eng. Chem. 49 (1), 120e124 http://dx.doi.org/
a Constant of degree 3 fit for Cp (kJ/kmol K) 10.1021/ie50565a039.
b constant of degree 3 fit for Cp (kJ/Kmol2 K) Wolberg, J., 2006. Data Analysis Using the Method of Least Squares: Extracting the
Most Information from Experiments, second ed. Springer.
c constant of degree 3 fit for Cp (kJ/Kmol3 K) Yorizane, M., Yoshimura, S., Masuoka, H., Yoshlda, H., 1983. Thermal Conductivity of
d constant of degree 3 fit for Cp (kJ/Kmol4 K) Pure Gases at High Pressures by Use of a Coaxial Cylindrical Cell, pp. 454e458.