1、GPA RR-338 95 m 382qb99 0037623 bhO CY1 Research Report -138 Thermodynamic Properties of CO2 + CH4 Mixtures A Joint Research Report by the Gas Processors Association and the Gas Research Institute Project 842 GRI Contract No. 5087-260-1449 C-A. Hwang H. Duarte-Garza P. T. Eubank J. C. Holste K. R. H
2、all Texas A and for the infringement of any patent or the violation of any federal, state or municipal law or regulation arising from the use of any information, apparatus, method or process disclosed in this report. GRI DISCLAIMER LEGAL NOTICE: This report was prepared by the Texas A or 2. Assumes
3、any liability with respect to the use of, or for damages resulting from the use of, any information, apparatus, method or process disclosed in this report. . 111 Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or networking permitted without l
4、icense from IHS-,-RESEARCH SUMMARY Title Contractor Texas A! + CH4 mixtures. For each mixture, the Burnett measurements are at 300 and 320 K up to 10 MPa while the pycnometer measurements range from 225 to 350 K (in 25 K increments) up to 35 MPa. The accuracies of the densities are better than 0.1%
5、and are discussed in detail in the report. Energies and entropies derived from P-V-T measurements are about the same order of accuracy as calorimetric measurements (estimated uncertainties about 2x experimental). The properties reported here are: densities, virial coefficients (second, third and cro
6、ss), energies, enthalpies and entropies. The report contains approximately 400 measurements and 800 derived values. Because the number of laboratories capable of measuring fluid properties accurately over extended ranges of temperature and pressure Ras diminished dramatically over the past two decad
7、es, we have adopted the approach of constructing several apparatus capable of rapid, automated measurements. The ranges of the various apparatus overlap to assure internal consistency. Primary measurements agree within 0.1% for these apparatus. We use a Burnett apparatus, a Burnett-isochoric apparat
8、us, a continuously weighed pycnometer, two isochoric apparatus and a flow calorimeter to measure fluid ensities, energies, enthalpies, entropies and phase equilibria. The results from these apparatus are sufficiently accurate and precise to permit stringent testing and development of equations of st
9、ate and predictive correlations. iv Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,- GPA RR-Jn38 95 3824699 0017628 142 FOREWORD Production, processing, and transmission of carbon dioxide-rich
10、 streams for use in enhanced oil recovery require detailed knowledge of high pressure thermal properties. Enthalpies are needed for compressor design, exchanger and cooler design, and for simulation of high temperature and pressure profiles in carbon dioxide wells. Near-critical fluids show large ch
11、anges in thermodynamic properties in response to small variations in temperature, pressure and trace impurities. In many of the proposed applications of carbon dioxide, the fluid will be at near-critical or super-critical conditions. Project 842 has been designed to develop a reliable thermal proper
12、ties data bank. High accuracy experimental thermal properties have been collected for carbon dioxide-rich mixtures containing small quantities of nitrogen, hydrogen sulfide, and light hydrocarbons, e.g., methane and ethane. This particular report presents the results for mixtures of carbon dioxide a
13、nd methane. Experimental measurements of densities for mixtures of carbon dioxide and methane over a wide range of temperature and pressures are collected using four different experimental techniques. Other properties are evaluated using standard thermodynamic procedures. This report presents data f
14、or virial coefficients, energies, enthalpies and entropies. This project has been funded as a joint venture between GRI and GPA. Arild Wilson Chairman GPA Enthalpy Steering Committee V Lyman Yarborough Chairman Technical Section F Technical Data Development Jeff Savidge Project Manager Gas Research
15、Institute V Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,-TABLE OF CONTENTS REPORT DOCUMENTATION PAGE i1 GPA DISCLAIMER 111 GRI DISCLAIMER . 111 . . RESEARCH SUMMARY . iv FOREWORD v LIST OF
16、TABLES v11 . LIST OF FIGURES . vili INTRODUCTION 1 RESULTS AND CONCLUSIONS . 1 DISCUSSION OF RESULTS . 2 EXPERIMENTAL DETAILS 5 Burnett Apparatus . 5 Pycnometer 5 EXPERIMENTAL UNCERTAINTY . 6 REFERENCES 7 APPENDIX . Nomenclature . 8 vi Copyright Gas Processors Association Provided by IHS under licen
17、se with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,-Table I . II.1 . 11.2 . III . IV- 1 . IV.2 . Iv.3 . IV.4 . v.1 . v.2 . v.3 . v.4 . VI . VI1 . VI11 . LIST OF TABLES Page Experimental P-V-T values for CO2 + CH4 mixtures (Burnett apparatus) 9 Derived second a
18、nd third virial coefficients for CO2 + CH4 mixtures . 10 Derived cross virial coefficients for CO2 + CH, 10 Experimental P-V-T values for CO2 + CH, mixtures (pycnometer) . 11 Pressure residual properties and correction terms for 9.83 mol % CO2 + 90.17 mol % CH4 13 Pressure residual properties and co
19、rrection terms for 29.1 1 mol % CO2 + 70.89 mol % CH, 15 Pressure residual properties and correction terms for 66.82 mol % CO2 + 33.18 mol % CH4 17 Pressure residual properties and correction terms for 90.11 mol % CO2 + 9.89 mol % CH, . 19 Total properties for 9.83 mol % CO, + 90.17 mol % CH, . 21 T
20、otal properties for 29.1 1 mol % CO, + 70.89 mol % CH, 23 Total properties for 66.82 mol % CO2 + 33.18 mol % CH4 . 25 Total properties for 90.1 1 mol % CO2+ 9.89 mol % CH, 27 Physical constants. conversion factors. and molecular weights 29 Enthalpies and entropies for CO2 and CH4 in the ideal gas st
21、ate . 30 Parameters for 6Z vs p fitting equation for CO2 + CH, 30 LIST OF FIGURES Figure 1 . Schematic diagram of Burnett apparatus 31 Figure 2 . Schematic diagram of pycnometer 32 vii Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or network
22、ing permitted without license from IHS-,- ! (1984) and CH4 (198 1) comprise Table VII. Finally, Table VI11 lists the parameter values for the model correction term, 6z. The measurements in this project are state of the art and generally accurate within 0.1%. This information is suitable for both str
23、ingent testing and development of models and correlations. In fact, these data formed a significant contribution to the development of AGA-8. The data extension technique developed in this project can be extended to other models and form a general correction term for any equation of state. With this
24、 correction, the equations are nearly as accurate in the near critical regions as they are in any other region. 1 Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,-DISCUSSION OF RESULTS The data
25、 require no special discussion, being self-explanatory. However, the data extension technique does need some explanation. Energies, enthalpies and entropies can be calculated from P-V-T data using the concept of residual functions. The density residual function is: and the pressure residual is: QR(T
26、, P) = (T, P) - IG(T, P Here denotes a property of a real fluid and denotes the property of an ideal gas at the same conditions. All the residual functions are related to two dimensionless integrals involving P-V-T properties of the real fluids: U 1 RT T a(l/T) p O A dp (z- 1)- P RT O Isochoric meas
27、urements provide a direct determination of the derivative required to compute the integral 11. Because most of our data are isothermal P-V-T measurements, we have developed a modified approach to obtain accurate energy, enthalpy and entropy values. We utilize an accurate mathematical model (DDMIX) f
28、or mixtures containing carbon dioxide developed by Ely (1989). We use this mathematical model to calculate the fluid properties and then correct these values using our experimental P-V-T values. The procedure is to define: where 2 is the compressibility factor and represents a thermoynamic property.
29、 The subscript exg denotes an experimental value while the subscript M denotes a value calculated with a mathematical model, in this case DDMIX. The residual internal energy is: U -u; 6U -+- RT RT RT where O 2 Copyright Gas Processors Association Provided by IHS under license with GPANot for ResaleN
30、o reproduction or networking permitted without license from IHS-,- GPA HR-138 75 382Qbq 0037633 50T = The residual enthalpy then becomes: H U U only pressure and temperature of the sample fluid are measured. Figure 1 is a schematic diagram of the Burnett apparatus, which is discussed in detail elsew
31、here (Hwang, 1988). The temperature is measured with a MINCO (model S1059-2) platinum resistance thermometer (PRT) which is inside the Burnett cell between the primary volume VA and the secondary volume VB. Thermopiles measure the temperature gradients across the Burnett cell and the radiation shiel
32、d. The automatic temperature controller maintains the temperature within f 1 mK and f 5 mK of the setpoint temperature for the short and long-term respectively. Pressures above 4 MPa are measured with an automatic DH Instruments (model 50200) DWG, and a Ruska Instrument (model 2465) air DWG is used
33、for pressures below 4 MPa. The pressure of the sample fluid in the Burnett cell is related to these pressure gauges through an external pressure medium (N2) and a diaphragm-type DPI (Ruska model 2416). The technique for operating the diaphragm-type DPI is described in detail by Holste et al. (1977).
34、 Pycnometer A sample cell (pycnometer) of known volume is suspended from an electronic balance (Lau, 1986). A cross-sectional diagram of the balance chamber and its isothermal bath appear in Figure 2. Water and carbon dioxide were used to determine the sample cell volume as a function of temperature
35、 and pressure. The sample cell, fabricated from beryllium copper (Be-Cu 175) to obtain high thermal conductivity and good mechanical strength, is inside a copper compartment. A feed valve allows the entire confined sample to be subjected to the same uniform temperature. A variable volume bellows cel
36、l permits adjustment of sample density when the feed valve is closed. The isothermal bath is inside an aluminum radiation shield and a vacuum chamber to minimize heat transfer. The balance, Arbor Laboratories model 507, measures the force required to maintain the null position of the weighing pan wi
37、th the sample cell attached. This balance has a completely electronic mechanism, a capacity of 500 g, a precision of 0.3 mg, and it maintains vertical position to within 1 pm during the weighing operation. Pressures are measured using either a Rosemount pressure transducer or a Paroscientific pressu
38、re transducer. Both transducers were calibrated against a DH Instruments (type 26000) automatic pressure standard (dead-weight gauge, DWG). Temperatures are measured with a MINCO (model S1059PA5X10) four-lead PRT, which is adjacent to the sample cell on the inside surface of the compartment. 5 Copyr
39、ight Gas Processors Association Provided by IHS under license with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,-EXPERIMENTAL UNCERTAINTY Isothermal Burnett data consist of a series of pressure measurements after expansion of the sample fluid from a smaller (VA)
40、 to a larger volume (VA+VB). The derived densities are calculated from where pm and N, are parameters of the maximum likelihood fit, N, is the zero pressure cell constant, pm is the density at the lowest pressure, i and m denote properties after expansions i and m (last), respectively. The standard
41、error of density, ap), is: We estimate that om) = 10Apm and o(Noo) = 4x1V5 N, at 95% confidence limits, based upon maximum likelihood fit results. To simplify further the above equation, (m-i) is expressed as a function of molar density: m - i 5 2.41n(p/ mol. m-7 - 1 1 Therefore, -=0.01% 1+0.16 2.41
42、n (m:m-)-”Ir P As a result, the estimated accuracy in the Burnett density measurements calculated from the above equation is 0.04 % for our highest density point and is about 0.01 % for our lowest density point. As for the second and third virial coefficients (B and C), we estimate the absolute erro
43、rs in these values to be less than 0.2 cm3/mol and 60 cm6/mo12 respectively (95% confidence limits). The uncertainties in the pycnometric density measurements arise from the uncertainties in the mass determinations and the cell volume calibration. The error in the cell volume calibration, which incl
44、udes random errors introduced by temperature and pressure measurements, uncertainties from using a calibrating fluid whose equation of state is known, and from mass determinations, is about 0.04%. The estimated accuracy (Hwang et al., 1992) in the pycnometric density measurements (95% confidence lim
45、its) is: Ap = (0.15 kg. mW3)2 + (0.004p)2112 or where p has units of kg-m”. From Table IV, we find 8H/Hi to be on the order of O to 3.5% of the HM values. The estimated uncertainty of 6H values is 10 to 30 J/mol, which corresponds to about 1.5 J/g. This 6 Copyright Gas Processors Association Provide
46、d by IHS under license with GPANot for ResaleNo reproduction or networking permitted without license from IHS-,-value is approximately twice the experimental uncertainty of enthalpies obtained using a flow calorimeter (Mller et al., 1993; Castro-Gomez et al., 1990). REFERENCES Castro-Gomez, R. C.; H
47、all, K. R.; Holste, J. C.; Gammon, B. E.; Marsh, K. N. J. Chem. Thermo., 1990,22,269. Ely, J. F. NZST Mixture Propers, Program (DDMIX) Version 9.07. National Institute of Standards and Technology Office of Standard Reference Data, Gaithersburg, MD. Embry, D. L. Analysis of Burnett-lsochoric Data. M.
48、 S. Thesis, Texas A Hall, K. R.; Eubank, P. T. Ind. Eng. Chern. Fundam. 1977, 16,378. Hwang, C.-A. PVTMeasurement of Carbon Dioxide + Methane Mixtures and an Equation of State Development. Ph.D. Dissertation, Texas A Simon, B. P.; Holste, J. C.; Hall, K. R. 1992, To be published. Lau, W-W R. A Conti
49、nuously Weighed Pycnometer Providing Densities for Carbon Dioxide + Ethane Mixtures Between 240 and 350 Kat Pressures Up to 35 MPa. Ph.D Dissertation, Texas A Gammon, B. E.; Marsh, K. N.; Hall, K. R.; Holste, J. C. J. Chem. Thermo. 1993,25, 1273. TRC Thermodynamic Tables - Hydrocarbons, u-1 010, Table 23-2-(1.101)-. Thermo- dynamics Research Center, The Texas A&
