A New Equation of State and Tables of Thermodynamic Properties for Methane Covering the Range from the Melting Line to 625 K at Pressures up to 1000 MPa* U. Setzmann’ and W. Wagner* stint fr Thm und Fidynana, Ruhr Unvesit Bochum. D630 Bochum. Feral Republi of Geman Received Jane 1,19; revised manuscript recived May 29,1991 ‘This work reviews the data on thermodynamic properties of methane which were available up to the middle of 1991 and presents @ new equation of state in the form of a fundamental equation explicit in the Helmholtz free energy. A new strategy for ‘optimizing the structure of empirical thermodynamic correlation equations was used to determine the functional form of the equation. The Helmholtz function cont ing 40 fitted coefficients was fitted to selected experimental data of the followi properties: (a) thermal properties of the single phase (pp) and (6) of the liauid- ‘vapor saturation curve (pip'p") including the Maxwell criterion, (c) speed of sound w, (d) isochoric heat capacity ¢,, (e) isobaric heat capacity, (f) difference of enthalpy A, and (g) second virial coefficient B. Independent equations are also lluded for te vapor pressure, the satusated liquid aitd yap Weasites, the isu- baric ideal gas heat capacity and the melting pressure as functions of temperature. ‘Tables for the thermodynamic properties of methane from 90 K to 620 K for pres- sures up to 1000 MPa are presented. For the density, uncertainties of # 0.03% for pressures below 12 MPa and temperatures below 350 K and + 0.03% to + 0.15% {or higher pressures and temperatures are estimated. For the speed of sound, the ‘uncertainty ranges from + 0.03% to + 0.3% depending on temperature and pres- sure, Heat capacities may be generally calculated within an uncertainty of 196. TO verify the accuracy of the new formulation, the calculated property values are com- ppared with selected experimental results and existing equations of state for ‘methane, ‘The new equation of state corresponds to the new International Temperature Scale of 1990 (ITS-90) and is extrapolable up to pressures of 20000 MPa, “ey wont: oration: data evakaton; equation of ate; fundamental aqton: methane: mating Hine ‘vapo-lgud eoextence cue; propery tbl; thermal aod alre properties metastable Rates Contents, Nomenclature . = 1064 223. Melting Curve. Fed Pols and Ppl Constants for Methane ... 1064 Introduction 2 1065; 1.4, Backgroui 1005 1.2 Prior Correlations of Methane Properties and Need for a New Correlation........- 1065 1.3. Organization of the Paper 1066 2, Experimental Results... + 1056 21. Triple Point + 1067 22 Two-Phase Regions : 22.1, The Normal Bolling Point. 22.2. Critical Point. 1087 4057 1068 ‘Dedeated to Prot DrcIng.K. Stephan on occasion of his Ot bith "Presa aes Hals AO, D470 Ma, Federal Republi of German. “To whom correspondence shoul be addresod. {21991 by the U.S Secretary of Commerce on behalf of the United States. This copyright is asigned to the American Isitte of Physics, and the American Chemical Soe ‘Reprints aval foe ACS see Repeats Lb at back of ee. (047-2609/01/061061-86/810.00 1061 224, Liguid-Vapor Saturation Curve. 224.1. Vapor Pressure ...... 224.2. saturated Liquid Density... 1v09 2243. Saturated Vapor Density. 22.44. Enthalpy of Veporization. Heat Capa Speed of Sound. ingle-Phase Region . 23.1. Thermal Properties. 23.2 Virial Coetncients 2333 Difference of Enthalpy 2.3.4. Isobaric Heat Capacity, 2355 Isochoric Heat Capacity 23.6. Joule-Thomson Coefficient. 23.1. Isothermal Throttling Coefficient. 23.8. Speed of Sound ..... 24. Ideal Gas Properties. i 2.4.1. Isobaric Heat Capacity. 2.4.2. Reference Value for the Entropy. 243. Reference Value for the Enthalny. 23. Phys: Chom. Ret. Dat, Vol 20, No. 6, 19811062 1082 3082 1082 3. Auxiliary Equations for Methane .. 3.1. The Vapor Pressure Equation . 43.2. The Saturated Liquid Density Equation 33. The Saturated Vapor Density Equation .. 1082 3.4. The Melting Pressure Equation.......... 1086 535. The Equation for the Ideal Gas Heat Ca- pacity 1086 4. Development of the Fundamental Eaton for Methane 1086 441, The Helubolte Funion j 1006 4.1.1. The Helmholtz Energy forthe Ideal Gs 1087 4.12, Fitting and Optimizing the Residval Part of the Fundamental Equation... 1087 42. The Data Set for the Optimization and the Fitting Procedure ‘The New Fundamental Dquation + 6. Comparisons of the New Fundamental Equation ‘with Experimental Data and Other Equations of State 6.1. Liquid-Vapor Boundary»... 6.141. Thermal Properties on the Cocristence Gane. 6.1.2. Speed of Bound on the Cocnatonce Curve....... 6.13. Heat Capacity on the Coexistence Curve.......- 6.14. Enthalpy of Vaporization 62. Single Phase Region .. 62.1. Thermal Properties. 62.2. The Second Viral Covffieient :+01 1090 62.3. Speed of Sound . 624. The Isochorie Heat Capacity ..... 1101 5. The Isobarie Heat Capacity Difference of Enthalpy.... . Joule-Thomson Coefficient. 63. Critical Region GA, Metastable Stator 65. Extrapolation Behavior of the New Funda- mental Equation 7. Accuracy of the New Fundamental Equation 8. Conclusions 9. Acknowledgments 10. References. -Appenaix, Thermotiynamic Properties of Methane. List of Tables Available correlations for methane. Summary of the experimental values for the triple point of methane 3. Summary of the experimental values for the normal boiling point of methane. . 4, Summary of the experimental values for the critical point of methane 5. Summary of the experimental values for the melting pressure of methane .. 6. Selected melting pressure data... 4 Phys. Chem. Ret, Data, Vo. 20, No.6, 1961, 10. ML 2 33. 1s. 16. wm. 18, vw. 20. au. U. SETZMANN AND W. WAGNER Summary of the experimental values for the vapor pressure of methane Summary of the experimental vlucs for saturated liquid density of methane Summary of the experimental values for saturated vapor density of methane Calculated saturated vapor densities p* for temperatures T < 120 K. Total experimental uncertainties of the sat. feted vopor densities (P > 120K) pubisied by Kleinrahm and Wagner! and of the caleu- lated densities (T'< 120K, see text. Sommary of the experimental data for the 1069 1070 1071 107 enthalpy of vaporization, 1072 ‘Summary of the experimental data for the heat capacity on the saturated liquid line ......... 1073 Euimnay ofthe experinental date for ic ocd of sound onthe saturated liquid and vapor ine. 1073 Density corrections Ap (Pa = Pu ~ 8) for selected pseudossochores published by Goodwin" 1073 ‘Summary of selected ppT data for methane (Group 1 data) 1075, Bommary of experimental pp data for methane .. 1077 Summary of selected second virial coefficients for methane (Group 1 dats 1077 Summary of second virial coef longing to Group 1 data and third virial coeffi- cients for methane.........++0+0 1079 Summary of experimental data for the enthalpy of methane. Summary of isobaric 1070 ‘methane 1080 Summary’ of isochorie heat 9 methane .... 1080 3. Summary of Joule-Thomson ‘coefficients for methane ...... 1080 ‘Sommary of speed af touad data for mathane -- 1081 . Ideal gas heat capacity values for methane... 1082 5. Coefficients of the auxiliary equations for the thermal properties py 9", and p* along the va- ppor-liquid coexistence curve of methane...... 1086 Coens fr the Heal gas heat eapacy equation (3.8) zi 1086 Relations of thermodynamic properties to the dimensiones Heimboisfecon © comeing of @ and 4, see Eq. (4.1). 1088 2 |. Caleulated p, Contributions of properties to the sum of squares for the optimization and the fitting process. ‘Values forthe parameters of the modified two- dimensional Gaussian terme in the last sim of Eq. (47). 1089 and p” data for taking into account the phase equilibrium condition when developing the residual part ® of the fundamental equation, Eq. (53) ‘Summary of the data used for the linear opt ‘mization procedure and for the nonlinear‘THERMODYNAMIC PROPERTIES OF METHANE 38, Numerical values of the coefficients of the ideal gas part of the dimensionless Helmholtz function, Eq. (52) - |. The ideal gas part of the dimensionless Helmholtz energy "and its derivatives. ..... 5. Coefficients and exponents of Eq. (5.3) ‘The residual part @ of the dimensionless Helmholtz energy and its derivatives......-. ‘Values forthe critical exponent B of the satura~ tion dancitios p! and. pf calevisted from too RB 1093 3 Eq, (63) in the region very close tothe critical speed of Sound on the eritical sochore calcu lated from Eq, (6.1) and the scaled equation of, Kurumov' in the immediate vicinity ofthe crit- ical point haa 39, Thermodynamic properties of saturated methane .. nT 40, ‘Thermodynamic properties of methane in the homogeneous region aus List of Figures 1 Percentage deviation of experimental vapor presure data from values calculated from the fapor presure equation, Eq. (32) 2. Percentage deviation of experimental setu- rated liguid density data from velues caleu- lated from the saturated Tiquid density equation, Ea.) 3. Second virial coefficients of methane cai. Jated from Eq. (2.5) when the vapor pressure and the saturated vapor density equation pub lished by Kleinrahm and Wagner? is used . 4, Percentage deviation of experimental saturated vapor density data of Kleinrahm and Warner? and ofthe data sted in Table 10 from values Calculated from the saturated vapor density equation, Eq. (35).- 5. Poreentage Gevnton of the original and the corrected ppT data of Goodwin™ from values calculated from the fundamental equation, Fo. (53). 17s 6. Distribution ofthe pp data used in establish. ing the residual pat ofthe fundamental equa- tion, Eq, (63), in a pT diagram, 7. Abnolune ovation’ of caper Viral coefcients from values calculated from Eq. (53) . 8. Distribution ofthe isobar heat capacity data ‘sed in the establishment of the residual part of the fundamental equation, Eq. (5.3), in a pT diagram vee 1080 2. Disuvuion ofthe isechore inca capacity da tsed in the establishment of the residual part afte fundamental eputon Ea 53) disgram 1008 1070 171 107 1076 1078 1081 10. nL vw, B. 14 16. 17. 18. 19. 21 26. 2, Distribotion ofthe speed of sound data sed in the establishment of the residual pat of the fun- damental equation, Fa. (84), ina pT dngram Percentage deviation of experimental melting pressure data from values calelated from the ‘ling pressure equation, Eq. (3.7) ....+.--+ Percentage deviation of experimental eal gas isobaric heat capacity data from values caleu- lated from the ideal gos heat capacity equation, Fy GS) Percentage deviation of experimental data on the vapor liquid saturation lin fom values cal culated from Eq, (53) ...... Percentage deviation of experimental speed of Sound data on the vapor-liquid saturation line from values calculated from Ea. (51). Penrentnge deviation of experimental heat capacity data on or along the saturated liquid line from values calculated from Eq, 5.1)...» 109 Percentage deviation of experimental enthalpy of vaporization data from values calculated from Eq. (53) coeeesesssseesessees : Percentage density deviation ofthe experimen- ta) poTr data from values calculated from Eq. (53). For the experimental uncertainties Se6 Table W6..-0eeee00s 1057 Percentage density deviation ofthe experimental pT data from values calculated from Eg. (6.3). For the experimental uncertainties see Table 16 Percentage density dvition ofthe experimental ‘pp data from values calculated from Eq, (53) in a high resoltion with respect to density. For the experimental uncertainties see Table 16. Percentage density deviation ofthe experimental ‘po data from valves calculated from 9. (53) in 2 high resolution with respect density. For the experimental uncertainties see Table 16...» ‘Absolute™ deviation ‘of experimental’ second Viral coefficients from values calculated from Eg. (53) -essee Percentage deviation of tho’ experimental speed of sound data from values calculated from Eq, (5.1). Percentage deviation of the experimental isochoric heat eapacity data from values cal culated from Eq. (5.1). 1103 ne 1085 1098 (. Representation ofthe isochorie heat capacities, in the liquid state below 160 by Eq, (6-1) and F 104 ‘of the experimental iso- baric heat capacity data from values calculated from Bq. (5.1) . ee Absolute deviation of experimental data of differences of enthalpy from values calculated from Eq, (6.)..-+ sees 1106, ‘The plot of the Joule-Thomson coefficient in the region below 11MPa calculated from Eq, (.1) and from other equations ..... os 1107 4. Phys. Chem. Dat, Vol. 20, No. 6, 19011064 28, Percentage deviation of selected experimental data on the vapor-liquid saturation line in the critical repion from values esleulated from E563) Me ocrernrererny 108 29, Pereeniage pressure deviations of the exper mental po data from values calulted from Ee. (62) inthe eal rgion 30, Representation of the speed of sound in the eel ron ye (1) and by ter ee tons : sa Kepresciaion of the sochori eat capacity in the ertial region by Eq. (Sl) and by other esuations, 19, Poreennge dent deviation of caper pT data on six selected ‘wochores in the tmetasable region fom values clelated rom Eq, (53).- 3, Peteentage devin of he experimental speed of sound data in the metastable region from val ‘es ealeulated from Eq, (1) ~ 12 > 1109 110 firey 12 34. ‘The experimental Hugoaiot data in comparizon from with corresponding. valves calculated Eg. (51) and from other equations 35, Tolerance diagram for density. 30. toverance eiagram tor speed of 37, Tolerance diagram for enthalpy... 3, nb mB 113 and» Nomenclature ‘Symbol Deseription 2 Adjustable coefficient ‘A Specinie Heimnoita energy B Second virial coefficient C Third virial coefficient © Adjustable coefficient G Specific isobaric heat capacity Specific isochoric heat capacity Specific heat capacity along the saturated guia tine Specific Gibbs energy Specific enthalpy Specific enthalpy of vaporization Serial numbers ‘Maximum value for the serial number j Number of data, Molar mass Serial number Adjustable coefficient Pressure Specific entrony ‘Absolute temperature Adjustable coefficient Specific internal energy, Adjustable coeticient ( Speed of sound ‘Any thermodynamic property Compression factor f+ = pioRT)] ee eo yhesget, eur) nue Greek eBoy Adjustable coefficients 4 Phys. Chom, Rt. Data, Vo. 20, No.6, 1901 U. SETZMANN AND W. WAGNER Precorrelated value for w data ("= gles) Difference in a quantity Reduced density (8 = p/p.) Isothermal throttling coefficient Partial differenti Precorrelated value for ¢y data Adjustable coefficient Converted coefficient (8° = @/T,) ‘Transformed temperature (9 = 1-T/T2) Joule-Thomson coefficient Mass density Standard deviation Inverse reduced temperature (x — T./T) Dimensionless Helmboltz energy [ = A/(RT)] ‘Weighted sum of squares Superserpte {deal gas property © Residual + Saturated liguid state "Saturated vapor state Denotes a vector meter EegoueroD Subeeripte b At the boiling point Experimental ij Indices m Denotes the melting pressure, Index for experimental data nl Nonlinear Reference state ‘oH Initial state values for the Hugoniot curve ‘measurements ‘opt Optimated orig Original 5 Denotes the vapor pressure on the liquid- vapor saturation curve tAttthe triple point Along the satiration curve Physical constants for methane ‘M——-Molar mass M cf. Ref. 209) Re Molar gas constant R,, = (MASI + 0.00021) Smal, cf. Ref. 210 R Specific gas constant R= ReJM = 05182705 kii/(kg K) Critical temperature 7, = 190-964 K Critical pressure p. = 4.5922 MPa tical density pe = 162.66 kg/m? ‘Triple paint tomporatira T, = 00.6041 K ‘Triple point pressure p, = 0.011696 MPa Normal boiling point temperature T, = 111.668 K 16.0428 + 0.0013) grmol, Re aPPS‘THERMODYNAMIC PROPERTIES OF METHANE Reference temperature T, = 298.15 K Reference pressure p. = 0.101325 MPa Reference enthalpy in the ideal gas state at 52 Reference entropy in the ideal gas state at Tyand pe 83 = OKi/(kg K) 1. Introduction 4.4. Background Methane isthe major component of natural gas In this contest knowledge of the thermodynamic properties of methane is required for the processes of liquefaction, separation, storage, pumping and transport of atural 25, etc. Moreover, methane serves as an important raw material for the production of hydrogen, alcohols and many other products of the chemical industry. Due to in- creasing efforts in the optimization of chemical plants, necessary to provide accurate thermodynamic proper- ties of methane. Furthermore, since methane isthe fest compouinl fe yoouy uf Ue alkanes, sen sfested to as a reference fi In 197%, the International Union of Pure and Applied Chemistry (TUPAC) published monosraph' which reviewed most of the experimental data available up to 1976 and presented extensive tables of the thermal and caloric properties of methane derived from an equation of state Since 1976, many ivestigaions ofthe thermody namic properties of methane have been carried out. A comparison ofthe new experimental data™ with existing, equations of state shows that none ofthese equations is able to represent the new measurements within their experimental uncertainties, f, Seo. 6 In addition tothe increasing amount of experimental data, the correlation techniques have beon significantly improved during the last decade. Up to now, none ofthe existing equations of 1065, state for methane has been optimized with respec to its structure. This means that the terms in these equations have been determined subjectively by experience or by tual and. ersor. Tn 1989, Setzmann and Wagner™ presented a new strategy for optimizing the structure of empirical thermodynamic correlation equations, Based on a comprehensive functional expression fr the conside ered physical dependence, which is celled a “bank of terms, the new procedure optimizes the structure and the lenth ofthe equation simultaneously. The aplication Oo this optimization procedure result in an equation which meets the quality desired for representing th experimental data with the lowest number of fited coefficients. In akon with this optiniation method we used 3e¥- eral thermodynamic properties forthe determination of the new fundamental equation. The use of more than one thermodynamic property i called ‘mult-propert iting which s the state-of the-art procedure for establishing ef- fective wide-range equations of state. Ths technigue was extensively applied when establishing our new fundamen- fet equation for methane. 1.2, Prior Correlations of Methane Properties and ‘Need for a New Correlation A comprehensive evaluation of methane data was reported by Angus etal in 1978, This IUPAC mono- [graph reviewed the experimental data of methane which had been published prior to 1976. Since 1976, and espe- cially during the last five years, many state-of-the-art ‘experiments on the determination of the thermodynamic properties of methane have been carried out. The increasing amount of data stimulated many authors to develop correlation equations for methane. ‘Table 1 summarizes the equations of state for methane which have been developed since 1970, Al ofthese wide "Tam 1 Avaliable coreations for methane ‘Authors Year Temperature Presure Stracure of Number of Dstt wed in range range the eqtion coelems the corelstion Bender? sm sass 0-50 Extended BWR? » pot pes" Goodat 14 snge-s0 0-50 ‘Nonsnaitie a porno Mecano? ism sass40) 0-40 Enaded BWR 2 DoT Bere Aneoreval!QUIPAC) fom anata tna Funded BOE 2 Sythov ee” 1s Sass-1000 O00 Double ower expansion 54 inTandp Ey & Hanley! ss sass 080 tended BWR 2 Sievers & Schule!* 1984 sass-e0s 05D Extended BWR (Bender) 0 reson Leland” 1986 SSS-S00 Oa Damping function for 4432 ‘Youngove & Ey 1967 sass-con 0-200 2 -> Kuruma eal 38 saws Revised and extended 2 betw scaling Law lend era. * 1969 91-00 O10) Developed for omgen 2 pero ty Schmit and Wagner ea Boned. Wesb-Rabin, Detalls om slated dat wl be given by Ely (to be published). 41 Phys. Chem. Ret. Data, Vol 20, No. 6, 19911066 ze equations of state are based mainly on the po (pressure p, density p, temperature 7) published by Goodwin and. Prydz," Bouslin eral,” and Cheng.” Recent experimental investigations of the pp properties of methane measured by Kleinrahm and Wagner? Kleinrahm et al, Handel eal. Pieperbecket a." and ‘Achtermenn ot a indicate thatthe alder experiment values of Goodwin” show systematic deviations - up to (0.2% in density - from the new state-of-the-art ppT data. ‘Therefore, all equations of state, which have been devel- ‘oped up 10 now, contain errors at least of up to + U2% in density in a wide range of temperatures and pressures. In 1988, Kurumov etal. published a revised and extended scaling law equation which represents the thermodynamic properties of methane in the critical region. This nonanalytic function represents most of the experimental data in the critical: region within the estimated experimental uncertainties. It gives correct limiting behavior atthe critical point (weak divergence of tha feachorie haat raparity . andl the reciprocal spond of sound w" , etc.) but the scaled equation is valid only in very small region around the critical point and the agreement with experimental data deteriorates very rapidly as soon as the equation is extrapolated outside the near-critical region. Because of its small range of validity and its mathematically complex structure, the craled equation ie not very convaniant for anginaoring ap plications. For our correlation, we decided to retain a completely analytic form of the fundamental equation which guarantees a fast and convenient evaluation of the ‘equation. 1 wil be shown later tnat our new tundamen- tal equation represents the experimental data in the critical region at least as well as the scaling law equa- maw etal.) althewigh eur equa critical point. ‘The most recent equation of state for methane was Each ofthe existing equation of state for methane has several of the following disadvantages: (a) not able to teprecent the dato within the experimental wncerainia, (©) only valid ina restricted temperature oF pressure range, (6) not able to represent the properties in the enlarged critical region, (2) a rather complicated struc- ture, (e) ot corresponding fo the eurtent international Temperature Seale of 1990 (ITS-90)2* Therefore, we developed a new equation of state which is capable of representing the thermariynamin eurfae of methane fo the range from 90.69 K < T-< 625 K at pressures up 10 1000 MPa within the uncertainty of the available data, ‘Organization of the Paper In Sec. 2, we discuss the available experimental information on the thermal and calorie properties of ‘methane in the fluid region, on the melting line, and on the liquid-vapor saturation curve. The selected data which were used to develop the new fundamental ‘Phys. Chem, Ret, Dats, ol 20, No.6, 1981 U. SETZMANN AND W. WAGNER equation, the auxiliary equations for the saturation line, and the melting pressure equation are discussed in this section. In addition to the fundamental equation, we de- veloped supplementary equations for the vapor pressure Pw the saturated liquid density p’, the saturated vapor density 9, the melting pressure py, and the ideal gas heat rapscity (2which are given in See 3 Sertinn d antaine 9 ‘description of the techniques used when developing the fundamental equation (multi-property-fiting, optimiza- tion procedure). The structure and the coefficients of the new Helmholtz function are presented in Sec. >. Com- parisons of properties calculated from the new equation of state for methane with the selected experimental data are given in See, 6, Finally, the uncertainties of the prop- erties calculated with the new formulation are discussed and tables of the thermodynamic properties of methane are listed in the Appendix. 2. Experimental Results Atl availahla exprimontal information a the thermal ‘and calorie properties, the triple point, the eritcal point, the normal boiling point, and the melting pressure has been reviewed in order to determine the most accurate ‘data set for methane. ‘Throughout this paper the word fata’ is used to refer to experimental measurements. ‘The literature review was based mainly on earlier data compilatinne nf Angnect al !Sytchew et al ana Sievers ‘and Schulz." After our critical evaluation of all the data ‘we concluded that the measurements before 1950 did not ‘meet the present quality requirements. Therefore, this section includes all experimental investigations Which have been carried out during the last 40 years or older works which are of essential importance for the charac- terization of the data ‘of single properties. Ref- ‘erences of experimental results not mentioned can be found in Refs. 1, 10, 25 and in the work of Setzmann.™ ‘To condense the description of the data situation, the enaracteriste information on the single cata sets is sum- marized in the tables for the corresponding property. When the information could be found, these tables also ‘contain valnes for the experimental nncertainting for ech data which were used to establish the new equation of state or, at least, to check its quality by comparison with the corresponding data. Usually, these uncertainty values correspond to the values given by the experimentalist Since we noticed, however, that a few authors had pub- lished overly optimistic estimations of their experimental nnrertainties we sometimes hal tn estimate mare reall tic values of the corresponding experimental uncertain- ties; these new uncertainty values which we ourselves changed are presented in parenthesis in the correspond- ing tables. ‘To simplify the evaluation process regarding the quality of the different data sets for the reader, for some properties the data have heen classified into three groups. In these cases, the eriterion for putting the single data sets into the corresponding group is, clearly defined in these tables.‘THERMODYNAMIC PROPERTIES OF METHANE: ‘The new equation of state should correspond to the ‘now International Temperature Scale of 1990 (ITS-90).* ‘Therefore, all data which had been based on older tem- perature scales (including IPTS-68) and which had been considered to be relevant for the development of the new reqiation of state or for any comparisons were converted to ITS-90, This means that all temperature values inthis * cortespond to ITS-90. The conversion from -68 scale to ITS-90 scale was not cartied out by ‘interpolation within ‘Table 6 of Ket. 24° but with a conver- sion equation given in Table 1.6 ofthe Supplementary In- formation for ITS-902" This conversion equation is accurate to within 1 mK ahove 272K and 1.5 mK Kelow 273K. In spite of this uncertainty, in some cases it was necessary 10 add an additional decimal to the converted temperatures in order to ensure that any recalculation ‘yom 11-90 temperatures to the original 1PTS-08 tem- peratures produces the same figures as given in the o inal source after rounding to the same number of decimal. 2:1, Triple Point ‘The triple point temperature and the triple point pres- sure were measured by numerous workers. The most re- ‘ent sources of triple point data are listed in Table 2. The agreement between the results of Blanke,® Bedford etal. Pavese eral.” and Kleinrahum and Wagner” is excellent. Based on our assessment of these works we se- lected the following values forthe temperature and pres- sure of the triple poi 7 7 onsen = 0009 K = (11696 + 0402) x10 Pa "The wanectaiay gi the temperature conversion of = 0.0015 K. When taking into account the phase equilibrium condition, the solu- tion of the new fundamental equation, Eq, (5.1), at ‘90,6941 K yields a pressure of 116.96% 10" Pa, The vapor pressure equation, Eq. (32), and the melting pressure equation, Eq, (3.7), have been constrained tothe selected Uiple poi parameters. Qa) 2.2. Two-Phase Regions 22:4. The Normal Boling Point ‘The temperature at which methane boils under a pres- sure of 0.101325 MPa (I atmosphere) has been measured ‘extensively. Table 3 is summary of reported experimen tal values forthe boiling point temperature. The selected temperature from the work of Pavese etal” and Klein- rahm and Wagner” is Th = (111.668 + 0.0025) K, 22) “Tae on caepuon te ne wea gs properacs,otane rom sts ‘al thermody, which have becs based on the univers thm Symani temperate vale “Inthe temperatare range considered bere, the diferences between 1087 “Tea 2 Summary ofthe experimental values for the tiple poi of ethane Soaree Yer ik rr) 9s ones = dna ta? 175 $0691 nose Paves eal 19% apsd200mS 116962002 Bonboure & Fel! 1978 90.495 5 Bonhoure& Palo 196) 904847 7 Pres" sl sna 1696002 Blake 1983 sness7sooms = Blake 9s onasass0c0ms | — Bediod eral iat Shesesenaor = Paves ete 19s sussiassani05s — Pave etal” se snesseosanums — lcnr & Wegner? 1955 9068410003 116562020 TAIN emperatres were comveried to TTS90, Ta ome canes I wat ‘neces to giean atonal fgre when converting from IPTS- <8 { TTS-90. The unceraintos in the temperature ales taken fo he temperture comers 1 5-0 ‘The temperature ofthe tiple point was ogy determined onthe TTS scale using the tape pot of argon as the reernes. ‘Te montorement (IPTS-48 sel) ofthe le pat temperate of ‘thane was based on the condensation polo of eaygn. There is + Systematic dierence of (4 mK for measurersents based the the ile pin of argon and these bated upon the condensation poi of tvesn For deta ce Beso al ‘No data the value wa selected by somparizon ofthe exiting mes- ‘urements and taking ito account he qual requirements of he rss ‘Tame 3 Summary ofthe experimental values forthe normal bing pola of methane Soar Your 7K Oise 15 1092 Heaning & Stock wat Lao Keyes tal! 192 miss 1908 us ‘Bloomer & Pareat® 1983 ‘Tinvot & Pach 1959 has eal 1960 ; Be Paves etal? 1976 111.5679 0001 Klcinram & Wapper? 1986 11167520083 where the uncertainty sorrcaponds to the value gin by Pavese etal-® ( (2001 K) including the uncertainty of the temperature conversion (+ 0.0015 K). ‘When taking into account the phase equilibrium condi- tion, the solution of the new fundamental equation for (0.101325 MPa yields a temperature of 111.667 K. The va- por-pressure equation, Eq. (3.2), has been constrained to the selected temperature of the normal boiling point. Phys. Chem, Ret, Data, Vol. 20, No 6, 19011068 2.22. Cleat Point The 1esults of recent detetminations of the critical point parameters of methane ate given in Table 4. Sum- aries of ealir investigations concerned withthe citi cal temperature .. the citical density oan the eritial pressure ps ere given by Armstrong etl." and Sytchev etal.” In 1984, Kleinrahm and Wagner** described the evelopment andthe construction of new apparatus to oblain accurate measurements of the saturated liguid land yapor densities of pure fuis together with the vapor pressure along the whole coexistence curve. The authors bresented new experimental results for the saturation densities and the vapor pressure of methane and ex- tracted the critical point parameters given in Table 4 from their measurements. We took ove their values with ‘wo eacepulons The uneerany tn Tz Was enlarged due 1 the uncertainty in the temperature conversion to ITS-0. ‘The enlargement of the uncertainty in pe was based on four experience with the evaluation of our recent pT ‘measurements. on carbon” dioxide,"™*™ where the Parameters ofthe critical point were also extracted from the experimental densities ofthe saturated liquid and va- por. T. = (190564 * 0.012) K bs = (45822 = 0.002) MPa 9. = (162.66 = 02) kg/m’ 03) “Tasus 4, Summary ofthe experimental ves fr the xt point of| U, SETZMANN AND W. WAGNER ‘Table 5 summarizes the information on the available data sets which are cla rs the data having the smallest experimental uncertainty. Group 2 comprises data which do not differ by more than 3% from the values of Group 1. Group 3 includes data Which have larger systematic deviations from Geonp 1 and Group 2 data which form the selected data, Table 6 reports the uncertainties of the selected data used to es- tablish the melting pressure equation, Eq, (3.7) ‘TaSL 5, Summary ofthe experimental values fr the melting pesures of methane Soarse Year Number Temperate Group ofdota range TK Freeh & Vershovie*™ SSL wars 3 Suyland er wo "8 Tinsisa 3 Reeves eal wor 1 ass 3 Grace & Kennedy wor 21030 3 Cheng? im 7 mizat Pye & Onan in 139-86 1 Cheng ta ms? Hiren etal i 1 mais Wilda ea we 2 sas 2 Kortbeck eal toe es ince the experiential reals were only poled in arabica form, itis df wo das the dats ‘These dta are etal withthe data of Cheng: the adn point, the tiple pola presur, ian extrapolate value. “Tae experimen eults nee only published in graphical frm. Their ‘mec aes were conned to sty Reroeek™ Se MARR OD ae et ig pe ‘Veanix*" 1966 190.8 4.6265 162.50 Sours, ad as Grigor & Steele 19681908 462 162.60 aa a se ee oem eo mie ino eg See ke Ron ae ge ; mS Saat oe ous is i ee Olson” ws = - 180 Wieldraaijer etal” 1983 500 Slam awape? Ske aso Sa, Sil? ayy omen (ae i ‘The values forthe critical parameters given by Kleinrahm sud Wagner? were confirmed by Kurumoy eral. in 1988, Kurumov etal, presented a revised and extended sealing law equation for methane. When determining the con- stants of their scaled equation of state it was possible to treat the critical parameters as adjustable parameters Since, however, a refit of the critical parameters led to values which ‘very good agreement with the values xlven by Sletnrahim and Wagner.” Kurumov er af." also accepted the critical parameters given in Eq. (23) for their equation. 22.3 Melting Curve ‘There are 11 sets of measurements of the pressure- emperavure relationship along te melting curve 4 Phys. Chem, Rf, Data, Vol 20, No.6, 1991 2.2.4, Ligule-Vapor Saturation Curve Based on a detailed investigation of existing data along the saturation curve we concluded that the measure iments of Kleinrahm and Wagner represent the phase boundary of methane’ more accurately than previous ‘measurements. Thus, ow correlation equations for the vapor pressure p, the saturated liquid density p, and the saturated vapor Gensiyp glen in See. 3, have been es tablished based only on the results of Kleinrahm and Wagner. Friend er al." ofthe National Institute of Stan- dards and Technology (NIST), who developed an equa- tion of state for methane in 1988, came to the same conclusion. They strongly emphasized the use of the data of Kleinrahm and Wagner when establishing their corr- fauon equations Tor te vapor-qura pase Dounoary.‘THERMODYNAMIC PROPERTIES OF METHANE 22.4. Vapor Pressure “The vapor pressure of methane has been measured ex tensively from the triple point tothe critical point; within the ast 100 years, more than 30 datasets have been pub- lished. The most recent determinations of the vapor pres- sure are given in Table 7, Since the residual part of the fundamental equation, Eq, (5.3), and the vapor pressure equation, Eq. (3.2), are based on the experimental vapor pleswuies UF Kickualun ound Wages,” Unly thio data oot ‘was assigned to form the Group 1 data, Group 2 data contain experimental results which do not differ by more than # 0.3% in pressure from the values of Group 1. The third group includes data which differ considerably in a systematic way from the selected data. A detailed com- parison of the selected data set for the phase boundary Irilh mcaouieicute uf vile eaperiuentalita waa previ ously given by Kleinrahm and Wagner in Ref. 2 Figure 1 illustrates the percentage deviations of the Group 1 and Group 2 data from values calculated from the vapor pres- sure equation, Eq. (3.2) This comparison shows that the Group 2 data are clearly outside the total experimental ‘uncertainties of Kleinrahm and Wagner's vapor pres- trem, the numeri valves of thets wneertaintion aro ‘given in Ref. 2. Kleinrahm and Wagner” discussed that the effect of impurities on the vapor pressure may explain the disagreement between the data sets. ‘TABLET. Suramar ofthe experimental value forthe vapor presi of ‘methane Source ‘Year Nomber Temperature Group wdat range TK Bloomer & Parent” 08a) Arsuongaa" sss ah Lz Htermans & White 196118081903. Vaniterbecketal”™ 1963 2 SND Van Itetbecketal? ot 3k RIED is 13 OBS 65 3 IDS wo sets 3 m 9 mim 3 wm 9 rim wm 3 mims 2 wm SMS we "2 weiss 2 wth ws 62 7 sims 2 wm 2 joe, 3 1986 ist 907s "Te reference only containesmsated dat the orginal experimental ‘aloes were publaed Uy Ag stan te TUPAC wcrgep ot vnethane 22.42. Saturated Liquid Denety ‘The saturated liquid density has been investigated extensively over the entire temperature range. Table 8 summarizes the information on these data sets. Since the 1069 experimental uncertainties of the saturated liquid densities of Kleinrahm and Wagner are clearly less than {le uncertainties ofthe y' data of other authors, we de cided to use only the 54 data points published by Klei rahm and Wagner for the establishment of our equation of state, These 9 values form the Group 1 data. All data Sets which agree within the combined experimental un- certainties withthe data of Kleinrahm and Wagner are considered to belong to the Group 2 data All the other Sallie saturated liu deity deta deviate aoteaa cally from the selected data and are placed in Group 3. St ae foxde le aa Temperature T/K Snes set Tio Peicnye Sevitn Sp. ~ Gung ~ Paice € pe imental por presure data fom alts calculated frm ‘vapor pressure equation, £9. (32). The plod uncersiny range comesponds lo the data of Kleine and Wapner= "Tate 8 Summary of he experimental valves for sturated liquid 120 K and forthe saturated vapor densities G -380 calculated by using the procedure desribed above for z T < 120 K. The uncertainties given forthe revised p" val ata ‘ves comespond 9 maximum shifts ofp” which would not 8 ‘Traue10.Caclted strated vp dense or emperaturs S -4s0 Tei0k s 7H viigin) 120K) plished by Kleinrahm and Wagner? fd ofthe eluted Sener (7 = 120K see txt), “Temperature TK Unceriny op) 100 waa ry 100.000, ows 120000 3 140.000 bins 160.00 ‘me 180.000 om 139.00 092 000 on 190313 0313, 0568 131 uite produce an absurd behavior of the equation of state the pas region for T < 120K when these shifted o val- vues were used for establishing the equation. The uncer- tainties estimated for the revised p" values are also shown, in Fig. 4. ‘The physteally incorrect venavior of the second virial coefficient calculated from the original equations for the saturated vapor density and the vapor pressure is caused by the correlation equation for 6” and not by the vapor pressure equation, The vapor pressure equation was con- strained at the triple point and the normal boiling point which means that two ofthe four fitted coefficients were fixed. Since the remaining two cveificients were deter- ‘mined by fitting the equation to the whole vapor pressure curve, this equation does not have the degree of freedom to produce a basically wrong course between the triple point and the normal oiling point. Thus, there is no rea- son why the vapor pressure equation might be so incor- rect as to produce these wrong B values. ‘As a result of our detailed investigation of the satu- rated vapor densities we have developed a new correla tion equation for this property, see Sec. 33. 2.244. Enthalpy of Vapotztion ‘There are only four sets of data of the enthalpy of vaporization. The data of Hestermans and White* de- pend an the knowlege of values fr the saturated guid U. SETZMANN AND W. WAGNER and vapor density. The authors used values for these {quantities which were published by Bloomer and Parent® and Matthews and Huid. These data sets are not von- sistent with the data published by Kleinrahm and Wag- ner" forthe saturation densities. Therefore, the values of ‘Hestermans and White" have to be considered to be less reliable. Jones eta." used methane with a purity of the sample of only 99.45 mol%. The characteristics of the four data sets, which were not used to establish the new equation of sate, are summarized in Table 12, 22.45, Hest Capachty ‘There are only a few laboratories which have published experimental measurements of heat capacities in the saturation state. The experimental data of the heat ccapactty along the saturated liquid curve cy and of the heat capacity cj ate summarized in Table 13. However, these data were not used for the development of the new ‘equation of state. 22.46. Speed of Sound ff sound of the saturated liquid and vapor is given in ‘Table 14. The data sets which were used for the develop- ment of the new fundamental equation are characterized by a number which differs from zero in the column ‘se~ lected data points’ of Table 14. The difference between the number of data points of a Uata set and the number ‘of selected data points ofthis data set i based on the fact that we either rejected abundant data or that there were obviously misprints in the publication. ‘The speed of sound measurements of Gammon and Douslin® were performed close to the phase boundary. All data points of their publication for which the measured pressure differs by nore tn 0.002 MPa funn the vapor pressure [ealeu- lated from Eq, (32)] at the measured temperature were not included in Table 14. 2.3, Single-Phase Region 2:34. Thermal Propertion ‘The temperature and pressure dependence of the den- sity of methane has been investigated in more than 50 pa- pers. A detailed description ofthe quality ofthe available ‘Tan 12, Summary of the experimental dat forthe ethlpy of vaporization ‘Source ‘Yer Number “Temperate Uscoraiaty Parr of dat ange TK ofsample Font & Gua 959 7 a fan = OS = Hestermans & White 1961 0 111-165 = 996 molt Jooes eal a 5-189 = 948 mal Cofweler ol” 1964 1 100 Seah, = 130 9998 mol 4 Phys. Chem, Ret. eta, Vol. 20, No. 6, 199‘THERMODYNAMIC PROPERTIES OF METHANE “Tamu 13 Sunmary of the experiments data forthe heat eapacy on the saturated lig ine ‘Source ‘Year Number Temperature Property rk econ & Kareat™ io Cais m6 Wiebe & Bevo 1850-20 Hesemans und White 19619 Cutler and Morison §— 1985S Younpone™™ Wm 6 163 Roer™ data was presented by Angus etal," Sytchev eral,’ ‘Tester, as well as by Sievers and Schulz.» As a result of, recent experimental investigations of Kleinrahm etal. AAchtormann ot 20 and Morrie 4! wo halt eve the assessment of the data situation given in the prior vestigations mentioned above; details were discussed by Setemann.* ‘One ofthe most important results of our investigations for assembling a selected ppT data set was the fact that Goodwin's ppT data’ were not consistent with the ex- ‘perimental saturated liauid densities o' of Kleinrahm and Wagner: If we had used these two data sets together in the selected data, it would not have been possible to es- tablish an equation of state which showed a smooth be- of the derivatives near the saturated ligu On the other hand, we had to use Goodwin's data forthe development ofthe new equation of state because these values had been the only 00d poT data in the liquid re- {Bon up to that time (1988), Therefore, we tried to find ‘the reason for the assumed errors in Goodwin's data but ‘we could not find any hint inthe corresponding papers.” Based on the order ofthe difference between the p’ data ‘of Ref.2 and the p’ values extrapolated from Goodwin's oT data for p < peitwas unlikely thatthe reason forthe difference was an error in the temperature or pressure ‘measurement. Consequently, we assumed an error in the density values. Thus, we concluded 10 ‘correct? Goodivin’s densities by the same correction for all values fon the same isochore but by different corrections from ‘sochore to isochore. For p > ps the corrections Ap were 1073 determined by comparison ofthe p' values extrapolated from Goodwin's isochoric pp7 values*” with Kleinrahm and Wagner’? p values and, for p< by the compar- son of Goodwin's ppT data? with the very reliable ppT data of Kleinrahm eal.‘ Schamp et al." and Roe." The ‘mumerical values of these corrections are listed in “Table 15 forthe most characteristic isochores; the maxi ‘um correction obtained inthis way amounts to 0.14%. Since we had, of course, our doubts as to whether these corrections wore cbjectvaly reasonable we initiated new oT measurements in our group carried out by Hindel tal for pressures up to 8 MPa (pressure limit of the ‘pparatus) with the densitometer based on our “Two: Sinker’ buoyancy method.* Figure 5 shows the compari- son of Goodwin's orginal and corrected data with the new ppT measurements of Hinde eral.? and, for the 200 feothorm, with recent data of other authors ae wll ‘These comparisons illustrate that Goodwin's corrected data agree well with recent experiments, This fact con- firms that the corrections of Goodwin's ppT data were not unreasonable, ‘Thus, Goodwin’s original pp data" probably contain systematic errors of up to 0.2% in den- sity which is about two times greater than the uncertainty tstimated hy Goodwin himerlt Tn the course of the development ofthe new equation of state it became apparent again that Goodwin's cor- rected data are the only acceptable pT dataset for the liguid region at higher pressures. Only the use of these corrected data allows a good representation of the caloric ‘Taaue 15, Densiy sorrestons p(n = fae) for selected pseodosochores published by Goodwin Taochore alka?) ‘Correction rr 410 o 05x10 3 Oy 10 io tasty 0 09% 10-%e¢ 10 12x10" 3 1ax0pe¢ a 86210 "haw “Tam 14. Summary of he experimental data for the sped of sound on te strate gu an vapor tne ‘Source Year ‘Nomber of data Property ‘Uncertainty tot ‘elected Van Wusibock & Verhagen 1089 6 7 * Say = oxi" Van Doct etal? ws 8 % m Jon = Ox10-49 Blagol ea. 0% S " b= 3-5, Sean we so = ” Ban = a yetO“he say 1975 3 3 » Say 1975 3 7 ” (Gammon & Dousia* m6 6 ~ 4 Phys. Chom. Ret. Dat, Vol 20, No.1074 U. SETZMANN aNn W. WAGNER 0.2 138K - 140 K ot A A ape ae ad SS AA ae Ofre-e-o-e eye ay | vv M m0. -0.2' 4 ge 0.2 7 eacre é aSdy Not is AAA AA A A a 4 a) v 2 wv t oleecsBeuSgZys: y y Un e “0.4 g -0.2 a _ 00. K= 310k ee aeae n O.4 a Saas 0 Heo ea ey A -0.4 70-25 5 10 15 20 25 30 Pressure p/MPa 4 Gougwin'™(orig) ¥ GoudMin (curr) © Haendel et al® @ Pieperbeck et a1? 4 Achtermann et al x Kleinrahm et alt + Jaeschke § Hinze” A Mol Tenun"® Fic.5. Percentage deviation fhe original nd the cortected pp data of Godiah” from ales alee fom ‘he fundamental equation, B63). For the dats especially the speed of sound measurements, in this region. Since, however, all equations of state which have ‘been developea aurng the last 10 years were ted {0 ‘Goodwin's original data, they have to differ substantially {rom our new equation of state, see Sec. The selected. pp? data sels which correspond 10 Group 1 data to which the new equation of state has been fitted are summarized in Table 16, Figure 6 shows the distribution of the selected data in a pT diagram. Not all ‘of the setectea data points are presented in Figure © 4 Phys. Chem, Ret. Data, Vol. 20, No. 6, 1991 ‘apeimental uncertainties se Table 16 because the representation of all data points in the critical region and in regions where several data sets overlap each other would have made the plot 100 crowded, All experiments not mentioned in Table 16 are pre- sented and classified in Table 17. Group ? contains poT data which differ from the Group I data by more than their uncertainty given in Table 16. All the other avail able data deviate considerably from the Group 1 date and are only of historical interest‘THERMODYNAMIC PROPERTIES OF METHANE 1075 ‘Table 16 Summary of selected pp data for methane (Groop 1 dats) ‘Source ‘Year Number of dats ‘Temperature Presure Pal of ncorainy? total selected “range TIK" range pIMPa samplenol% ovinK 3a, » Sehamp a” FT TT 2 Ts (x10) Downe.” Woe ADSL M1 aI aI“ ‘ im MeL ataT aS] “Gado Roe wm SSID BD Sax (-5)x10-%) bots wm seta (S=30)x10-) Trappenieseral!” = «19D AMOS 2H“ ($10) More" we SS (0-9) Mallerp!* iss Sk SL on ss Tx10-% ‘Achicrmam eal 1986 35S @x10-) lelarah eel? 19856 206206 T8O9S. 67 © swoS 3TOT Pair Jette & Hine” 1991S} a sxu0- Kiearah etal 1968 HBOS SRS 3 TIO 2x10 Achicrmnneral® = 1991S BST 2 aco, Hinde etal? ior 2100-280 sa95 7x10-% 2x10 Reriboss h Bvte!l# 1000 wie meas. igo tooo SI eto) Piperbeckeral™ 1990 17S 17S— MSDS OIID 989095 STI 2x10 “Uoceraag values given in brackets wee eximated by ounshes ‘Te have a rezone relation tothe dts of eher are hi gon, ‘There are many reported values for the second virial coefficient in the literature, Figure 7 shows the compari= ‘son between values predicted by the new fundamental ‘equation, Eq. (5.3), and experimental results for the sec- ‘ond virial coefficient. Table 18 lists the temperature range, the number of data, and the uncertainty ofthe ex- perimental values of the selected second virial coeffi- cients. These selected data agree closely with each other and form the Group 1 data which were used to establish the new equation of state. All selected virial coefficients were derived from ppT measurements except for those of Lemming" which originated from speed of sound mea- surements. In contrast to the selected data, all the other second virial coefficients show significant deviations from, the Group 1 data and are placed in Group 2 and 3 of ‘Table 19. The Group 2 data consist of virial coefficients ‘whieh gitrer from the Group 1 data vy more than tei uncertainty given in Table 18. All the other data which clearly deviate from the Group1 data are put into Group 3, 23.3, Ditference ot Enthelpy ‘The literature sources which deal with enthalpy mea- surements are summarized in Table 20. The data sets of ‘measurements of enthalpy differences Ak = (Ty,px) — ‘h(T.p2) which were used to establish the new equation of state are marked as selected data in Table 20. The val- ues given by Ayber! and Dawe and Snowdon™ were rived from their measurements of the integral Jou! ‘Thomson coefficient. Jones etal." measured isobaric hheat capacities and presented the results as smoothed and interpolated isoharic heat capacities. The authors in- nly any hed pT dat ne wend cluded 2 table of esthalpy valuee in their publisntion. stead of using the enthalpy data given by Jones etal," we decided to fit our equation directly to the isobaric heat capacities. The additional inclusion of the enthalpy data ‘did not improve the representation of this data set. It should be stressed that Jones ef al® used methane con- taining 0.45 mol% impurities. 23.4 laobarle Heat Capacty Tsobaric heat capacity data have been measured by six laboratories. Figure 8 shows the distribution of the se- lected data on pT coordinates. The available experimen- tal data sets are anmmarized in Table 21. Since we could ‘not find any uncertainty estimation in these papers we es- timated the uncertainty of these data to be about = 2% ‘This estimation is based on our critical review of the is0- bani neat capacity data and consistency tests with other ‘thermodynamic properties. The selected data were used to develop the new equation of state. 235, lnochorle Heat Capachy ‘here are three sets of experimental isochoric heat ca- pacity measurements which are summarized in Table 22. ‘Most of the data points were published by Younglove'® in 1974. In 1976, Roder™ repeated some of Younglove's ¢, measurements on methane using the same apparatus as Younglove did. He found differences of up to 1% in ‘comparison to the data published by Younglove. Roder’s ‘experimental datal™ and the values calculated from the fundamental equation of state agree within + 0.5%, Hence we have to conclude that the uncertainty of ‘+ 059%, which Younglove estimated for the majority of 4: Phy Chem, Re, Data, Vol 20, No.6, 19011076 U. SETZMANN AND W. WAGNER 40 . 2? ; = 40° s = soe © = = & z = 5 : Hu a baa? RRR a rRERRRER © 40% g RRR x 2 Regn eh eS a x s re Rare a Mie tegercicucheteneg= g feaxx reas e é yarn * 2 Vapor pressure curve 400 200 300 400 500 600 Temperature T/K ppT-Data + Goodwin’ © Haendel et al? = Pope™ @ Noo™ vy Kleinrahm et al? = Morris!” @ Pieperbeck et al” + Jaeschke & Hinze” w Kieinnahm at alé 4 Achtenmann et al? @ Trappeniers et al’” Schamp et al" < Kortbeek & Schouten" A Mollepup"® ® Achtermann et al! x Douslin et al” Fi, 6 Distibution ofthe po data used in exabishing the residual part of the fundamental equation, Ea. (53) in 87 igre, 4. Phys. Chem, Ret. Data, Vol 20, No.6, 1991‘THERMODYNAMIC PROPERTIES OF METHANE. 1077 ‘able 17. Summary of experinental pp dat for methane oes Your Neaber Tomportae =— Groep ot data ange TK range p/P a 1 7 ea = 3 ayer otal®) 12 0 man aan 4 Keyes Burke! 7 o man 32.35 3 Frpeth & Verscoyle™ 31 z 29 re 3 Kalo & Gad 1931 166 2473 a-100 3 Michels & Nederorag™ 1935 56 ma 2 Michele & Nedernage™ 1936 119 23 2 ids eral 98 Be sit 3 Kazama & Levcheno!™ ou 2 msn 1510 3 Pavionich & Tigo” 1988 37 03335 1196 3 Date & Kobuya™ 196 196 ma sot 3 Moelle fa 161 5 14423 14s 3 Van Iterbeci ef." 1963 12 15-188 on 3 Dobro a 1964 ‘2 10-195 10 3 Wallace eta 1964 Fa 218-38 aio 3 Grigor eral 2 1966 ® 95.190 ais7 3 Hang 1966 2 13185 3 Mamedov & Mamedn™ 1969 e 3 (MeMath & Edminer®” 1969 2 3 Eppa 10 150 ve) 2 Samoone ea 2% a @ 455 3 Sorokin & Blagoy wm a ome 3 Venn eral. 1970 cy 15.6 3 “Tri oa im 7 2-861 3 Cheng 17 a2 72-1000 2 odosevich & Mile 1973 rt ooeais B Bay Sri 4 9 38 3 ‘Gammon & bows inne a dies 2s ‘Minas ea" im ‘7 159) 3 78 se iz 3 8 3 ssn8 3 1992 a ars 2 1986 0 ates 3 Kortbeck ea 1986 s 100-1000 2 Starman & Gammon 18 1 as 3 Machado etal" 1988 ” sll 3 Baidakov & Guria!™ 1999 165 iss 3 MeEloy a.” 199 2 ae-126 3 ‘Trot 18, Summary of selected second vial oefcent for methane (Group 1 data) Source Year Number Temperaturg Unceriainy ‘fda range KY Soylent) Dowie WA 16 TB 02 Ree sea Kieinabmeral* 1988 «63a ons Lemming wo 7 80 07 his measurements, was too optimistic. The distribution of the selected cy data in pT coordinates is shown in Fig. 9. ‘These selected data were used to develop the new equa- tion of state 2.36, Joule Thomaon Cooticlent ‘There are only two datasets ofthe Joule-Thomson co- cffcient which are listed in Table 23. The data measured are integral Joule-Thomson coefficients ‘could not be used in the fitting process ofthe new equation of state. As already stated by Angus eal," the dataset published by Budenhozer etal" does not seem to be very reliable. The inclusion of the data set of Bu- denholzer et al. into the fitting process did not improve the representation of the Joule-Thomson data, and we decided not to use these measurements for establishing the new equation of state; sce See. 62.7. ‘J. Phys. Chom. Ret. Dat, Vol 20, No 6 19811078 U. SETZMANN AND W. WAGNER 4.5 1.0 2 £ os = oS 0 ~N 70.5 E -1.0 . > Q “1.5 at — 100 200 300 400 500 600 700 Temperature T/K © Thomaes & Steenwinkel™® = Pope™ < Epperly™ > Mueller et a1” @ Roe! + Hahn et al? = Lemming™ x Schamp et al" Michels & Nederbragt™ x Douslin et al® @ Lichtenthaler & Schaefen'® > Lee & Edmister®® ¥ Beli et al’ y Amer x Kleinrahm et alt Michels & Nederbragt™ mFreeth & Verschoyle™ x Sunn Strain et all Mihara et al A Hamman et a] 4 Beattie 6 Stockmeyer” FiG.7. Absolute deviation AB = (Bays ~ Bou) of expetinenal second virial oefcients om values aeuated fom 4, (53)- Al data with deviations of more than = 15 en inol te plated onthe lint ofthe deviation sae. 237, Isothermal Thotting Coetctet In the literature only the paper of Fucken and berger," puoiisnea in 1434, could be rouna wnien deals With integral isothermal throttling experiments. The au- thors published 75 data in the temperature range be- tween 165 K andl 799 K at pressures up to 11 MPa withont giving any uncertainty value for their measurements. ‘These data were not used to fit the new equation of state, 23.8, Speed of Sound ‘There are extensive measurements of the seed of sound for methane. The overwhelming majority ofthese ‘measurements has been performed during the last 20, years. The information on these data is summarized in ‘Table 24. rigure 10 shows the alstsfbution ofthe selected data on pT coordinates, The speed of sound data cover, the temperature range from 100 K to 450 K at pressures ‘upto 1000 MPa and. due to the scarcity of information on J Phys. Chem. Ret Data, Vol 20, No.6, 1961 the heat capacities of methane, the selected speed of sound data were extensively used to develop the new fun- damental equation. These selected data are marked in ‘able 24 1m tne column ‘Sumber of aata el... 2.4. Ideal Gas Properties 2,44. lsobale Heat Capacity “The feat gas neat capacty can be derived from suis tical mechanical models based on spectroscopic data. In 1963, McDowell and Kruse™ presented calculated values of efor the temperature range from 60) K to 5000 K. ‘These calculations were based on a rigi-rotator har- ‘monic oscillator model which was modified to take into account the low-temperature quantum effects, anh romiety, cenrtugal alstorion, ana vibravion-rotation teractions. Although the values of MeDowell and Kruse ‘were published nearly 30 years ago, they still represent the most accurate data for the isobar heat capacity cin‘THERMODYNAMIC PROPERTIES OF METHANE 1079 “Taste 19, Summary of second vit coetcens not belonging 19 Group 1 data and thi vl coeticens for methane Sours Your Ponery Nomber “Temperate rm ot data range 77K (caly 8) Free & Vershope™ wa Ts 7 Mise e Nederbagt® 1935 43 2 Miele f Nederbragt™* 1936 m3 2 Beate & Stockmeyer! 1902 573 2 anman eral. 1955 33383 3 Gin" 1958 msn 3 Schamp eta 138 disaus 2 ‘Thomacs & Stenwinkel™ 190 18-20 3 Mucler eta 1961 146288 3 a 1966 1m 5 er i223 3 Byene eal 1968 m2 3 Hore tal. 1968 2273 2 Lechter & Sehifer™ 198 33 3 Eppes” 1970 se60 3 [Lee & Edmiser' wn 0-38 3 Stein eal in 265i 3 m= 1 ws 2 Belinea is 300-550 5 Haka ra 19 0-23 3 ww 298-348 2 ert & Hauler 194 35305 3 ‘Amer 1986 So 3 Metioy erat 1989 30-323 3 “Tass 21 Summary of eperental dat forthe enthalpy of methane ‘Source Yer ‘Number of data Temperature Prose Uncertaingy Boden eral 19 ® aT ‘oii03 es Janes eal 1963 6 400-083 osu io Sagal eal oe 7 suas 0340 eetortane ‘Ayer 198 2 ieeant aie 4 Dillard eral 1968 4 = 28 54 3x10-7ah Dawe & Snowdon! ws 6H mater ono Steet Kasteren &Zoldnrast™ = 1979 ne) ‘an-ass 325 10ae “Sine the methane wid forthe measurements Bad » py of only 98 mal the oa uncetany of these datas cleanly greater ian given By the thors the ideal gas state of methane, and they were used to determine our cfequation; see Sec. 3.5. Table 25 summa- rizes recent determinations ofthe ideal eas heat capacity, With the exception of Lemming's" and Goodwin's? data, who derived ¢ values from speed of sound measure- ments, all data sets were calculated from spectroscopic ats. ‘2.42, Reference Value forthe Entropy For the calculation of the entropy in the property tables of the Appendix, the state at which the entropy was assunted Iu be sere wa that of the ileal ya at 298.13 K and 0.101325 MPa. Ifthe entropy of the crystalline solid methane at 0 Kis to be taken as zero then the difference ‘of the entropy between this state and that of the ideal gas 4 298.15 K and 0.101325 MPa has tobe added to the val- ‘ues ofthe entropy given inthe tables. The recommended value for this entropy difference is given by McDowell and Kruse'™ a5 11.611 ki(kg K). ‘2.0, neverence vane tor me emmy For calculations of the enthalpy, there are three differ- tent states in common use at which the enthalpy may be taken as zero, The first state is that of the ideal gas at 298.15 K, and this is the reference state used for the en- thalpy values given in the property tables of the Ap- pena. The sevwnd aoumuvnly used state fs duet of Whe ‘enthalpy of the ideal gas at 0 K. The enthalpy difference between the ideal gas at 0 K and at 298.15 K is given as (624.39 kilkx by McDowell and Kruse.™ As discussed by J. Phys. Chem. Ret Data, Vol 20, No, 6, 18911080 U, SETZMANN AND W. WAGNER, 10° cData e’budenhotzer et at, + dunes et 1! vw Kesteren & Zeldenrust” & 2 a seeietecneeteiesineseeetestee 2 stk] stresses tresses ¢ @ 2 8 2 8 a sseece e eosouee a or pressure curve 10° 400 300 00 Temperature T/K Fic. R._Dikrtion ofthe kare hat cay den in the ethene rcp of the fundamental equation, Ea. (3), ina dag, “Taaur 21 Summary of iobri eat cpacies for methane ‘ouree Yer "Romber of dria Prewsre vou selected range pIMPo Hear™ 1919 = 7 a iae™ is a 7 ot Ecken & von Lide™ 1909 3 = at Donlaer ea" 1939 a a ars Sones eo 1963 ous a7 ‘ase 22 Summary of bochorc het capac for methane ore Year Nomier of dn Temgeratare Pree nes foul elected "Tange THK’ ange pa %e, Giacomini ins 7 i eae conor = Youngiove™ is som 5130 3483 (@-3)x10-%) oder 1976 3 % vera Sas Gesa0, "Unesrtainyvalacs given In brake were etimated by ourshes “TAIL 2, Swmmary of Joue-Thomeon cetiets for methane ‘Phys. Chem, Rt, Data, Vol. 20, No. 6, 1901‘THERMODYNAMIC PROPERTIES OF METHANE. 4001 Pressure p/MPa cy-Data » Youngiove™ + Roden Fra. 8. Dis fundamental equation, Eg (53) ina pT dagran. 200 300 Temperature T/K 400 tion of the bochore heat capacity data wed inthe establishment of the residual part of the “Taste 24, Summary of sped of sound data for methane Source Year ‘Number of data ‘Temperature Presure Uncen oat selected © FAME” rangepiMPa San, Dinon al ™ 192 TS ma a 5 singly ss wf 116-253 ct 0%) Van Itarbeok & Verhagen! 1010 < 512 omz.ox (@ 5x10) Len™ 1956 fae) 236-473 10-100 iow Terese” 1851 vee as oor 07 mis Blasi et. 1967 see 01-32 Gp me Singer 199 nf a2) S510 Cerdamone ta. 10 eS 13103 (6x10) Pitneskaya & Bilevh"® 1972 m 450 (@-2)x10-%) Dregs 13 ieee oms (a-21x10-%) Seay! is ri 5 1738 (05~2)x 10-40 sea 1975 a» 1318 5x10-99 (Gsramoa & Dowsit® 1916 mB one (01-10) 10 Baidakv era." 1982 uo 14 gpar™) Kortbeck etal 10, mw TS 100-00 i) Swaraman & Gammoat 1986 sie 1421s (1-19) x10-%) Goodwin? 1988 103108 027 (x10) Lemming’ 1969 Cod o1as 110-4 Kortbeck & Suutew™ 1989, 1585 100-1000 (G-3)x10-%¥) *"Thceraingy ales given in brackets were eximatod by curses, ‘J. Phys. Chem. Ret. Dat, Vol 20, No. 6, 19811082 TABLE 25. Ieal gs heat cacy values for methane ‘dat range 7K MeDovell & Kro™ we SO Giucik™ ma foe) ‘TRCTabes"* 1902 ie ‘soso ‘TRCTabes™ wo Se oes Goodwin 1980 * Bsa Tema! 1869 > Bao These da were clelated on the bass of epecronopi dt Derived from speed of sound measurement ‘Angus etal, its not possible to give an exact value for the dilference between the crystalline slid at zero tem- pesale-e, which might be the third reference state, a the ideal eas at 298.15 K. 3, Auxiliary Equations for Methane New correlation equations for liquid-vapor and soli liquid coexistence properties, consistent with the ITS-50, hnave been developed. To obtain very effective equations for these properties, their functional structure was opt rized using our new optimization procedure." A brief eye 8 BG a Aai2 woe YY Y ++ © soe aa 7 2 40 vy ++ a e y oodonoo oeoo8 eo oeooo08 os avs oo Seacs aa 107 ooeoag + + 4100 200 300 400 500 Temperature T/K w-Data 4 Gammon § Douslin® x Van Dael et al? + Straty” x Straty™ > Van Itterbeek et al’ 4 Kortbeek & Schouten" © Baidakov et a1 ® sivaraman & Gammon” @ Lemming’ y Goodwin” +Terres et al” Fa. 10, Distribution ofthe speed of ound data usd in the extalshent ofthe residual pat ofthe fundamental equation, Eq, (53), ina pT diagram, J. Phy Chom. Ret. Data, Vol 20, No.6, 1891,1084 U. SETZMANN AND W. WAGNER, 3 € ka a ae Kain OSAMA we 20% : 3 a keke S # 50 92 34 96 gi 350 400 200 250 300 Temperature T/K x Prydz & Goodwins 4 Clusius 6 Weigand Nunes DePonte § Staveley” © cheng” o Kortbeek et a: a Wieldraaiser et a2? -~ TUPAC? Fi. 11 Perentge deviation pu = (ange ~ Pact) Pauy Of caperinental melting pressure data fiom ves essed fro the meling presse uation, (:1}. For the experimental ‘cetines we Table 6 200 708 ‘500 ‘goo obo Temperature T/K © sooaan™” s981" 2 McOowel? & Kruse™ x Banno™ © TRC-Tables 1987'* Fo. 12. Percentage vation Sef = (fe ~ nam of xprimentl ideal a abi eat apaiy ‘ata fom valves elclted fom the hel gas heat expo equation, Ea. (38). All dat with Phys. Chom, Ret. Data, Vol. 20, No. 6, 1991Fa, 13. Perentage deviation 4y = Qagt ~ 7 ‘THERMODYNAMIC PROPERTIES OF METHANE ‘Uncertainty “027700 1,120. 140160480 Temperature T/K wkieinranm & Wagner* © calc. values (Tab. 40) ~~~ Auxiliary-eqg. - Tupac! Friend et aly YaeNom a ps of experimental data on the vaporsquld saturation ine fom values exclated frm Ea (5.3) The uncertainty range pleted by soi lines ore- ‘ponds othe original valves of Kleinahm and Wagner. The uncertalgy range plotted by ded fines CUrtesponds fo the unceranes estimated forthe ealclsted values of Table 10, The values elated fiom th aualiary equations (Eqs. (3.2), (84), and (5) are plated by dted Lines which ae very cose tothe 20 ne. ‘1na5 4. Phys. Chem, Re, Dats, Vol. 20, No.6, 19611086 shows that Eq. (3.5) represents the selected experimental density data for temperatures above 120 K within their ‘ety snl seater, see also the lowest diagram of Fi, 13. ‘At temperatures below 120 K, the saturated vapor densi- ties calculated from the new equation of state, Eq. (53), are represented within 0.005% by Ea. (35). Fa. (3.5) has a different form in comparison with the corresponding. equation in Ref. 2 because the data set used to fit the new 5" equation was different for T < 120K; see the discus- Be 22.43. ‘TRILE26, Coefients ofthe auay equation fo the thera prop and along the rapoenud eoeestence came T= ase K Pero m= ~6036209 fm = A083 fy = ~0494si99 mem 1408 ‘Seta gud dena equation 4) 7 = ose k = 16 ig? fs = Omens ‘Satrsed vapor deol equation G3) T= W560 K = 1266 m= ~1580280 sm = ~ 2852653, im = — 3006480, ey =~ 1391889 a= 37589961 3.4, The Melting Pressure Equation ing procenre equation has heen Aouelopnd tn identity the properties ofthe coexisting liquid and solid phase. The following bank of terms, based ona functional form proposed by Simon, was investigated: Pap 1+ Emr =- 1, GH) where p, is the triple point pressure and 7, is the triple Point temperature. As a reslt of applying the optimiza- tion procedure the final melting pressure equation for methane i as follows: Pop = 1 + m{(TiT)S~ 1] + nT" ~ 1) 3-7) with p. = 0.011696 MPa, 7. = 96941 K, me 24756810" and m= ~7:36602% 10°, Comparisons of the melting pressures calculated from Eq, (3.7) with se- lected data and with the melting line equation of the 1U- PAC? are given in Fig. 11, The agreement berween Eq, (3.7) and the results of Prydz and Goodwin" and of ‘Cheng is within = 0.2% in pressure, except in the triple point region. Since the IUPAC equation was established 41 Ph. Chem, Ret, Data, Vol. 20, No. 8, 1901 U. SETZMANN AND W. WAGNER, ‘based on wile point parameters (, = 0.011719 MPaand T, = 506891 K) which differ significantly from the parameters used in this work (Eq. 21)} the disagree: ment between the two equations in the triple point region isnot surprising. 3.5. The Equation for the Ideal Gas Heat Capacity Te equation GR = n+ 3 OG G8) was fited othe ideal gas isobaric heat capacities of Me- Dowell and Krose™ (see Table 25) of which, however, ‘onl the 36 data between 60 Rand 3000 K wore wed The ‘ocficientsm and values for ae given in Table 27. In the emtze range of validity of Eq.) from 60K to 5000 K, the differences between the data of McDowell and Rrse andthe values calculated from his equation donot exceed # 0.015%. A comparison of valves up to 1000 iis shown in Fig 12. The dala reported by Len ming! and Goodwin" hich are bessd on peed of Sound measurement, ate in close agreement with the data of MeDowell and Kruse; the devitions do not ex ceed 0.07% "TaILe 27 Coeficens forthe ideal gas heat capacity equation (38) 7 ™ 8 Se eee 1 oonet9 eK 2 ‘6002 957K 4 issn S105 K 5 nans 15080 4, Development of the Fundamental Equation for Methane ‘The new equation of state developed forthe represen- tation ofthe thermodynamic properties of methane is a fundamental equation explicit in the Helmboltz eneray Since the application of optimization procedures and the ‘mult-property-fiting technique used in the development ‘of such fundamental equations was extensively discussed by sctzmann ana Wagner” as well a by Saul und Wage net™ in recent papers, we have restricted ourselves here to the main features of our strategy for the development of wide-range equations of state 4,4. The Helmholtz Function ‘The fundamental equation is expressed in form of the Helmholtz energy A with the two independent variables, density p and temperature T. ‘The dimensionless Helmholtz energy ® = A/(RT) is commonly spit into a‘THERMODYNAMIC PROPERTIES OF METHANE. part depending on the ideal gas behavior @* and a part Which takes into account the residual fluid behavior &, namely G2) = O'Br) + O62), a) ‘where & = pipe is the reduced density andr = T./T's the inverse reduced temperature with pe and T; as the critical <énkity and temperature, respectively. The equation for {8°(%2) ean he ahtained from 9 careelatinn equation for the isobaric heat capacity in the ideal gas state. 4.1.1. The Houmnona Energy or me teat Gas ‘The Helmholtz energy for the ideal gas is given by 42) AXsT) = bP) = RT ~ T3*G.T). ‘The enthalpy ofthe ideal gas isa function of temperature only, and te entropy ofthe Ideal gas depends on temper- ature and density. Both properties canbe calculate if an equation for the ideal gas heat capacity cf is available. ‘When cf is inserted into the expressions for h*(T) and s*(psT) in Eq. (4.2), te following equation for A°(p,7) i, obtained: 7 -R ear ~ RI (In(ole.)] ~ 138 3) where p, = pJ(RT.) isa reference density. Furthermore, arbitrary reference values for the temperature T. the pressure pq the entropy sz and the enthalpy hz have tobe Selected. For our fundamental equation, the reference enthalpy hg has been taken to be zero at T, = 298.15 K, and the reference entropy 2 has been taken to be zer0 at T, = 298.15 K and p, = 0.101325 MPa, The final equa- n for * = A°/(RT) is given as Eq, (5.2) in Sec. 5. Aton) = flegar + ng-ar— 7 4.1.2. Fung ana opumtang he Restauat Part ‘ofthe Fundamental Equation In this subsection, the procedure to determine the functional form of the residual part of the dimensionless Helmholtz energy 4(8,x71), where 7 represents the vec- tor of the coefficients to be fitted, is summarized, ‘The basie tools are the ‘multiproperty fi" to determine the coefficients m, of the vector i for a fixed form of ‘® (G7) and the ‘optimization’ of the final functional form of the equation for . Since the Helmholtz energy is not accessible to direct measurements, i is necessary to determine the unknown structure and the unknown coefficients, of the residual ‘of the dimensionless Helmholts energy from proper ies which are experimentally available. Some examples of relations of the Helmholtz function to different ther- modynamie properties are given in Table 28, 1087 Dring the optimization process for determining the best structure for #, a lot of different forms of & equa- tions have to be fitted tothe experimental data. There- fore, at firs, ths fitting process is explained using the general relationshipz = z(¢,y), where x,y, andz denote {general thermodynamic variables. In our ease 7 might be DetisM, eyy-. abd and y ate B andr, respectively, see ‘Table 28, This means that we have the experimental data Zeta Yoq) andthe relationship to any correlation equa- fan for @ ae 7 = rR): for the example of ¢. = 6.(82) the equation 2 = 2(0,8,x71) corresponds to the sith equation in Table 28 with ® = @* + accord- {ng to Eq. (4.1), With a known expression for #*8,r), see Fg. (5.2), ¥@.r7) I Hite to the experimental data in such a way thatthe weighted sum of squares SE aw) is minimized, the index denotes one particular property {hat is being considered. In Bq. (44, the weighting factor corresponds to I/ with as the total uncertainty of the experimental data according to the Gaussian error prop- ‘sation formula given in this ease bv ade] abe] ‘obey’ ae [PEL e+ ESL # + FEL a] oo where. a, and o, ae the uncertainties wth respect to the single variable xy, anda; the paral derivatives (28z/ {ros canbe calculated from a preliminary equation of Sale! When "is fited to more than one property, itis called a simultanious or muliproperty fit andthe resul- ing sum of squares ¥=3¥ aA 66) is to be minimized. Table 29 lists the different expres- sions for the sum of squares corresponding to the single properties; for details see below. ‘The structures of most of the existing correlations for the residual part & of the Helmholtz energy have been. determined subjectively based on the experience of the correlator or by trial and error. To improve this situation, ‘Wagner and co-workers developed different optimization strategies""*"*"" For the optimization of formula- i ing two-dimensional problems like vapor ations, Wagner" developed a special ver- Sion of a stoprine rogreaion anelyia, which hea boon adapted by de Reuck and Armstrong™ for the develop- ment of equations of state, Based on the knowledge that the regression analysis does not provide sufficient var ability to optimize complex problems like the determina tion of wide-range fundamental equations of state, Ewers and Wagner"®* developed the evolu tion method (EOM), a random soarch strategy uses some principles from biological evolution. The optimization method used inthis work has recently been developed by Setamann and Wagner Ths new method 4. Phys. Chom, Ret, Data, Vol. 20, No.6, 19611088 combines both the reliability of the EOM and the high speed of convergence ofthe repression analysis of Wagner. ‘The development of the functional form for 4(8,x) is split into three steps. ‘The first step is the formulation of a general expression for the equation which functions asa ‘bank of terms’. For methane, thie hank of terme of the residval part of the dimensionless Helmholtz energy was formulated as £ Singer tet SS noe aan aP-ntewt + Enoive an U. SETZMANN AND W. WAGNER, ‘The last sum of Eq, (4.7), a two-dimensional Gaussian curve in combination with a polynomial of ® and +, was introduced to improve the behavior ofthe equation in the vicinity ofthe ential point. The parameters dt, Bu A and 1 of these terms are given in Table 30. The values for the parameters and > differ substantially from values vied by other correlators = 'In the second step, after selecting a suitable bank of terms, we used our new optimization strategy, * whichis based on a mathematical and statistical analysis, to deter- mine the most effective equation for ®. his equation consists of the particular combination ofthe 393 terms of, Eq, (4.7) which yields the best representation of selected data dieenetod in Sor 7 Tn the ent ization process different forms of, equations & = OG) were ited to the experimental data of different kinds of thermodynamie properties by ‘minimizing x" according to Eqs. (3) and (30). The prod- Jem of simoltaneously minimizing the sum of squares of “Tauce28, Reaion of thermadymamic propertie tthe diensoneat Hell anstion coming of 4° nd 00 Eq (4) Pom Pressure a er Imran 182 Laer en ny BOD Was wots op +00, Entropy ed = (05 + Oy - — Ge nee MD ise seee etreretmcty Bw en 4 0 ete cay Aw an «ny « aa Strate id ofa ie sped ita gesviastes Thome Osi « i ‘coefficient § a 1204- texan ring tp = 1 = BE ‘coefficient ee —.LmrmUmUm ns ee nr “4 Phys. Chem, Ret. Data, Vol 20, No.6, 1861‘THERMODYNAMIC PROPERTIES OF METHANE. 1089 ‘Table 29. Contributions of properties tothe sum of guaes forthe optiniaton and fing process 7 Dae Dawe Daa Sem of oa (wens) (pests) (Gad) ro 1 per) - Por xt [Par - & 49) 2 en) - eon) de S [B+ eens el ost «0 yet stout) fee = te cervan,s(arsens] oc? oe 7 “0 on rman ay Phase eaulloum condton a) a) wy d= $ [OSBED— pctoueen Tost 5 phere) wer) woo at S,[PGfRE- wen | ay ‘ Pen) oe oe) 7 Maswel terion nl] - en + 20% aa ata cy FF 0 + 2001 +0000) oat 416) “o 9 «ty [g- 9 + 205+ on), an nto Before areeissen] «[arsereeiassen] [ a ae [BB [esecemrosam sfareererase] Gun mre) = mre) gfe oy « CARR BOEY Te a ney - Oe) ake 2 [R= 1— a CEBBT act cy m8 me “re 2 ote) - tte) abe ¥ [gs envoy - Getetatees fe aay into) Tuo) +9-114209] +[ereetseassen] 7, 3 Mw . Me) nS [ersersras209] +[acsers+eia+2en] | ‘Wales which have tobe calculated rom «preliminary equation of wate ‘The coupling condition p = pRT(L + 8%) as tobe ten nto acount. 4 Phys. Chem. Re. Dts, Vol. 20, No. 6, 19911090 different thermodynamic properties becomes a problem ‘of solving a system of normal equations with regard to the unknown coefficients m. Depending on the properties to which the equation is ited the system of equations be- comes linear or nonlinear. The optimization procedure of Setzmann and Wagner" is, however, restricted to those properties his Yead ww sptein OF uta eyeations {hat are litear in the coefficients m, If the Teation z = 2(%,8,x71) between any property and the Helmholtz function isa linear combination of O” andlor its dervae tives, then the base requirement fora system of linear ‘equations is met. Ths is tre for the pressure p(,7), the isochoric heat capacity cx(p,7), the internal energy (yD) the second virial coefficient D(T), and the inde {eet application of the phase eat ever, many thermodynamic proper function of pressure p and temperature T and not as @ function of density p and temperature T which are the in- dependent variables of the Helmholtz function A. This yields an implicit relation between the measured state evinblesp and 7 cod the independvnt vsiatos ofthe Heelmholiz fonction and leads to a system of nonlinear equations for the determination of the coefficients, even ifthe relation between the property depending on T and p) and the Helmholtz function is near, e the enthalpy h. Several methods have been proposed which allow fo the linearization of certain ‘nonlinear proper- tie [o(Typ), ep(Typ), w(Fyp)] 0 that thece linearized data can aso be inluded in an optimization proces. This means that the functional structure ofthe resulting best ‘equation was optimized (opt) by minimizing the fol- lowing sum of squares: 8) where the definitions of the single jG = 1, 2, 10), cor- responding tothe linear and linearized data, are given as Eqs. (4.9) to (4.18) in Table 29. ‘However, to take into account the orginal experimen- tal information of all selected data (linear and nonlinear ata), in the third step, the final determination ofthe co- efficients. of the aptimized @ equation was performed in a nonlinear fitting process. This nonlinear fit (al) was, realized by minimizing the following sum of squares: aeaberda (422) where the relations between the nonlinear weighted sum Of squares xb. xb, and xb and the dimencionlese Helmholtz energy @* and @ are elven as Eqs. (4.19) 10 (421) in Table 29. To solve this nonlinear minimization problem, a modified Marquardt method, presented by Fretcner~* was used. ‘Then, the optimized equation with the values ofthe co- efiientsn, obtained from the nonlinear fit [Eq. (422)] ‘was used to recalculate the ‘precalculated data’ (see ‘Table 29) for the fit of the linearized data; see Eqs. (4.16) to (4.18) in Table 29. Based on these new values, the steps ‘optimization’ by taking ito account the linear and | Phys. Chem. Ref, Data, Vol 20, No. 6, 1991 U. SETZMANN AND W. WAGNER linearized data (Eq. (4.8)] and the step ‘nonlinear fit’ of, the optimized equation to the linear and nonlinear data [Ea. (4.22)] were repeated. This entire cycle (optimiza- tion, nonlinear fit, recalculation of the ‘precalculated {or the linearized least square sums) was repeated ‘until convergence was achieved. The convergence crite- Lit the values uf die synlinea data valeulated from the optimized equation before and after the nonlin- car fit have differences wi are at Teast one order of {ainties ofthese data. In this case, the final form ofthe & ‘equation is determined. ‘At this stage we would emphasize thatthe entire ite tion ycle described in the previous paragraph including the ‘creation’ of linearized data was only carried out 10 ‘make the nonlinear data available forthe linear optimi tion procedure. Comparisons (see Sec. 6) were always ‘made with respect to the original linear and nonlinear data, More details on the process of the determination of a fundaimcntal equation of state slag optimisation prove dores and multi-property fitting techniques are given by Saul and Wagner," Setamann,” and Mars:™ the opti- mization procedure is described in detail by Setamann and Wager.” “Tea 2, Value or the parameter ofthe mailed rwodimensonl ‘Guia term fate at sum of Eg (4.7) ™ a 10 107 107 io 107 100 107 tor 10 ior 107 13 13 13 La 1B 13 LB nB 4a u3 LB mn ut SS288e ee ESSE SE EUERERERREES BBLERSBEBERSSESRESEREEERREEE) >‘THERMODYNAMIC PROPERTIES OF METHANE. 4.2. The Data Set for the Optimization ‘and the Fitting Procedure In this subsection, we will give a survey of the selected ‘data which were used for the optimization process and. for the nonlinear fitting of the optimized equation to the experimental data. ‘To take into account the phase equilibrium condition wiih fitting a wide-range equation of state to data, we needed a set of vapar pressures p, saturated liquid densi ties p’, and saturated vapor densities p” in rather narrow temperature steps along the whole saturation curve. ‘Therefore, the required data had to be calculated for temperatures trom the triple point to the critical point with the help of independent equations for the vapor pressure and the saturation densities; these correlation SSquatione are given in See. 2 Taha 21 dofinee the ealew lated data set for the phase equilibrium condition, and ‘Table 32 summarizes the experimental data which were ‘used for the determination of the new fundamental equa- tion. “Taaur 31. Calculated pa 9 and p* data for taking nto account the ‘hase sql eondion when ceveopg te Tends part ofthe fundamental equation, Eq (83) 1091 5. The New Fundamental Equation [As discussed in See.41, the fundamental equation is expressed with the dimensionless Helmholtz energy ® AMT) a8 the ideal gas part and the residual par: AGAMRT) = 402) = VON + HOA) 4) with 8 = plocand-s = TYP see also Ea, (4.1) ‘The ideal gas part of the Helmholtz function was ob- tained according to Eq, (4.3) when the ¢f equation given in Sec. 35 is used: (6,2) = In(®) + ay + oer + as In(x) +Sami-e oa where 8 = plpyt = TJT, Te = 190.564 K, and 162.66 kg/m’. The values of the coefficients a, and 6 ‘ou ae gen in Table 3s, uation (93) sald the temperature region 6 K < T < 3000 K. The eon- slans ay and a, wee adjusted to give zero forthe ideal fa enthalpy a T. = 298.15 K and forthe ieal 38 en- ftopy at 7, = 25015K and py = 0.101325 MPa. In ‘Table 34, al required derivatives ofthe ideal gas part &* ‘Number “Temperature Sep With respect to 8 and + are explicitly given. ota ‘ange wat o Gaal < T< SGN OSK _Tams33, Numerical valve the coefons ofthe del gs pst ofthe a ssisel = T= 1801941 1K » * 3 asszanexid-* 2 20-20 —~a00 3 -oasTisisio 0 00-50 “arenes on in an oen sin 020 0 Te 05K B= OSTA2705 Wilks K) J. Phys Chom. Re. Bata, Vol 20, No.6, 1991, In this entire region, the new fundamental equation, Eq. (5.1), represents the selected data within the estimated experimental uncertainties. Equation (5.3) as been cousrained w dhe critical parameters given in Eq.(23) (T. = 190.564 Kk, Pe = 4.5992 MPa, p, = 162.66 kg/m!) and to the first and ‘second partial derivative of pressure with respect to den- sity being zero at the critical point. In Table 36 all re- quired derivatives of the residual part ®* of the dimensionless Helmholtz function with respect to 8 and + are explicitly given, 6. Comparisons of the New Fundamental Equation with Experimental Data and Other Equations of State In the following subsections, the quality ofthe new fun ‘damental equation is discussed based on comparisons with experimental data, Most figures also show comp: sons with the IUPAC eauatjon of state," the correlation published by Friend etal, and, especially in the en- Targed critical region, wit the revised and extended scal- ing law equation of Kurumov etal. To avoid ‘exceptionally long discussions, the comparisons given be~ low are representative rather than exhaustive; a more de- tailed discussion can be found in Ref. 26, where, however, both the equation and the comparisons are based on the IPTS-68 temperature scale. nll deviation plots, the zero-line represents our new fundamental equation. For caloric properties, the entice fundamental equation, Eq. (5.4), te ncccasery but for thermal properties only Eq, (53) is needed because the ideal gas part forthe thermal properties is already incor- ‘porated in the corresponding relations between the single properties and the dimensionless Helmholtz energy ® consisting of 4 and see Table 28. The new equation and the data presented inthe figures below correspond to the nem tenperotuie scale 17-90. To oor equation with those equations corresponding to the IPTS-68, we first converted the ITS-90 temperatures, used in the figure, into IPTS-68 values. This conversion‘THERMODYNAMIC PROPERTIES OF METHANE 1093, ‘Tle 36, The sesdual prt © ofthe dimensionless Helo energy and its deviates te Bet inet-id—cary] +2 notre n0-4F-Me- 740-208-809] (8-8) + dldy—Dpe-A] = Satter Bergan s £ materernerarcnin rs ae = Satan an + Feta rea -ctai-1-a09)-<09) or = Eagar 8 nests F nates a neentide-2n(e-a3) = Eaaeuanest +E nestenaanet + Spnatre sesraeetuur- wir “uae « B naatnget +E need met = Enns e t+ Enea d ate) + Baaeesoarn from ITS-90 to IPTS-68 was carried out with the equa- tion given in Table 1,6 of Ref. 24° which we also used to convert any IPTS-68 temperature of the experimental data to ITS-90; see the introduction to Sec. 2. In most regions, with the exception ofthe critical region, itis not at all necessary to apply the conversion procedure be- ‘cause the differences are so small that they are not recog- nizable in the figures. 6.1, Liquid-Vapor Boundary 16.1.1. Thermal Properties on the Coexistence Curve Figure 13 shows the deviations between experimental and calculated values with respect tothe vapor pressure, to the saturated liguid aud saturated vapor Seasity. Por @ sven temperature the equation of state yields these properties when the phase equilibrium condition Ti=T",p'=p",andg'=s", (6.1) is applied, which means that its not necessary to use any susiliary equation. The agreement between the new ful damental equation, Eq. (5.3), and the experimental pp, and p data of Kleinrahm and Wagner" is excellent. All data points are reproduced within the claimed small un- certainty ofthe data. Figure 13 demonstrates that neither the IUPAC equation’ nor the equation of Friend eral," both evaluated by applying the phase equilibrium condi- tion, are able to represent the data within their uncer~ tainty. As was stated in Sec. 2.2.43, for temperatures. below 120 K the saturated vapor densities had to be cal- culated by am iterative procedure because the evaluation [i 2euB A Motr—2B0r—W)] of the saturated vapor density equation given by Kile rahm and Wagner yields data which are not consistent ‘vith the various thermodynamic properties of the homo- geneous gas region. Friend etal." probably overfitted their equation to the saturated vapor density data of Kleinrahm and Wagner in the triple point region. Their equation represents these data within + 0.1% although the uncertainty given in Ref. 2 is between * 0.3% to + 0.8% at the triple point temperature. 16.1.2. Speed of Sound on the Coestatence Curve Figure 14 illustrates the comparison between the ex- perimental data of speeds of sound and heat capacities and the corresponding values calculated from Eq, (5.1), Th setuiated liyuid speed uf suid dala uf you Dosh etal.® Straty.""™* and Gammon and Dousln™ are in agreement with the speed of sound values calculated from Eq, (5.1) within = 0.29 except forthe eical re- sion, where the new equation of state fails to reproduce the limiting value w = O m/s when approaching the ci cal point; at the critical point Eq. (5.1) yields fet —nr? 205.95 ms, see alo Table 38, Aidhough the data of Straty®™ and Gammon and Dousln® agree quite ‘well with each other for saturated liquid speeds of sound, the saturated vapor speed of sound values of these two data sets deviate systematically from each other. Bearing {in mind that Gammon and Douslin measured the speed ‘ofsound at pressures which were slightly below the vapor pressure atthe temperature being considesed and taking into account that the pressure dependence of the speed of sound (2 /3p)renyay at temperature T = 165 Kis about ~50(mis)/MP, the w* values calculated from J. Phys. Chem. Ret. Dat, Vol 20, No 6, 19811094 U. SETZMANN AND W. WAGNER. 100* aw T, 120 140 160 180 Temperature T/K + Straty™ x Van Dagi et al? = Blagoi et al™ x Straty™ ‘4 Gammon § Doustin® wa Friend et alt <= qupac! Fo 14, Percentage devistion y= Gat ~ yD = 0s) of experimental sped of sound data on the ‘aporigui saturation lip fom vals clelted fom a, (51). Fr the experimental unertanis see "TnIUE ST Values forthe excl exponent of the stration dense and caleuated from Eq, (3) inthe region very closet the esl “Temperature diferene (e-Timk tothe etal temperature * 0 » 0 A a ° ‘Ciel exponent 8 aw 036 Ost ~—~—oa~—~=CSCTOSCaSSSCS “TAILE 38, Valves forthe fochorc heat capacity and the sped of wound onthe etal ochre clclte from Eq, (1) and theses equation ‘of Karumor! in the immediate vi ofthe el poit ror Property compared 1 wo wo? 0 wo o F451) moms 28.95 28.78 377 377 Scaled equation mn 13.09 sas 20 137% = ey wiry 12 seme see seu se Sealed equation” gy aL 236 soit 11592 - 4. Phys. Chom. Ret. Dat, Vol 20, No. 6, 1901‘THERMODYNAMIC PROPERTIES OF METHANE Eq. (5:1) agree well with the values of Gammon and Douslin, Based on these facts itis logical that the data of Gammon and Douslin show a positive deviation of up to several 0.1% from the w" values calculated from our equation of state For the saturated vapor speeds of sound, the equation of Friend etal." shows systematic deviations from Eq, (5.1) and from the IUPAC equation’ in the temperature region below 130 K. The reason for this probable misbe havior might be the overfttng of the p" values of Klei rahm and Wagner? for temperatures below 120 K; see the end of the preceeding section. Friend er al." adopted the structure of their methane equation from the oxygen ‘equation of state developed by Schmidt and Wagner" which forms the basis of the IUPAC monograph on oxy- gen.” Friend eral. determined the adjustable parame- tore of their methane version of the oxen equation by fitting the parameters to selected data of the following properties of methane: p pT data, pp'p" data forthe lig- ‘uid-vapor coexistence curve (mainly the p.g’p" data of Kleinrahm and Wagner), second virial coefficients, cy ‘data, and linearized ¢, co and w data. A detai gation of their correlation shows that for temperatures 1095 below 120K calculated c, and c, values of the saturated ‘vapor rise to a maximum, whereas a monotonic decrease with falling temperature would be expected (This is at least the experience from corresponding investigations of other well measured substances, e. . oxygen, water, and carbon dioxide.). Such an incorrect representation ofthe heat capacities also influences the representation of the speed of sound values because the calculation of this ‘property from a fundamental equation implies the caleu- Iation of e, and ¢, values, see Table 28 (61.3. Heat Capactty on the Cosslatence Curve Figure 15 shows comparisons between experimental and c. data and the corresponding values calculated from Eq. (2.1). The data of Chace Yaungloue i Reader, and Cutler and Morrison” are represented within their ‘experimental uncertainties, which are between + 0.5% at low temperatures and 5% near the critical point for ‘Younglove's data, for example. Although these data were not included in the simultaneous nonlinear fit, their rep- resentation is fairly good. f ut 7, 400 1, 120 140 Temperature T/K x Eucken & Karwat™ z Younglove™ @ Wiebe & Brevoort™ + Roden Friend et al? y Cutler & Morrison” = Clusius™ AHestermans & White v= TUPAC! Tic. 15, Percentage deviation y = (aye = Yas = 6c) of experimental eat capacity data on or along the ed ligld ine fom values aelated fom Eg (3.1). All data Wi deviations of sore than = 2% are poted on he limit ofthe deviation sale. 4. Phys. Chem. Ret. Data, Vol. 20, No. 6, 19916.1.4 Enthalpy of Vaportzation "The deviation of experimental data of the enthalpy of vaporization 4h, from values calculated from Eq. (5.3) is. shown in Fig. 16, In addition to the experimental Ay data, Fig. 16 also contains Wh, values calculated from the. Clauisius-Clapeyron equation we ar TU) Gn)” (62) where the quantities dpa, p’ and p* were determined ftom Eqs. G2), (34) and (3). The agreement between these data andthe a values caleuated from Eq (53) Shows tha, also for hs combination of the thermal prop- erties of the saturation curve, the consistency between the numa equation for py and 9" and Ey (82) tery good, With the exception of the two data of Frank and Clusivs” and Colwell etal.” the other My data are les reliable because of impurities inthe sample (Jones trai) or the we of older? and pales forthe a ation of the measurements (Festermans and White), a fT] poe ime erate stew ielnelg Temperature T/K ia 16. Percentage deviation M(Ah.) = (Ahan ~ Shans) hon of experimental enthalpy of vaporization data from values cr lated from Eq (53). 6.2, Single Phase Region Since this section only gives a compressed suvey of the overall quality ofthe naw fundamental equation, we wil show a Fepresentative sample from the data set that we used for establishing our equation. The deviation plots do 101 contain any information on the uncertainty ofthe ex- perimental data because such etror bars. would have made the figures overerowded. In general, however, for these selected data the uncertainty is represented by the scatter of the data, Detsile with regard to the uncertain- ties of properties calculated from our new formulation are given in Sec. 7. Most figures also show comparisons withthe TUPAC equation of state! and the correlation of Friend etal.® All the other equations of state for ‘ethane, which are summarized in Table 1, represent the properties of methane more unfavorably than the equa- fion® of the IUPAC and of Friend etal. do. Since the 4. Phys. Chem. Ret. Data, Vol 20, No. 6, 1991 U. SETZMANN AND W. WAGNER stated validity of the equation of Friend et al. i restricted to pressures below 100 MPa, comparisons of our equation ‘of sate with this equation ate given only for pressures be- low 100 MPa, The IUPAC equation and the new funda- ‘mental equation are valid for pressures up to 1000 MPa. (62.1 Thermal Properties Figures 17 and 18 show the percentage density devia- tion ofthe selected experimental ppT data from the data calculated from Eq, (53). The estimated experimental ‘uncertainties of these data and further information ean be taken from Table 16. We chose the low resolution of dolp = # 0.2% for the density deviation, because other- wise it would not have been possible to compare our ‘equation with the TUPAC equation’ and the one of ind of al "The high revolution of the deviation eoale in the additional Figs. 19 and 20 were selected to illus trate the quality of our new fundamental equation in two regons: pp ata for p< 12 MPa and T= 450K and oT data for p > 12MPa and T < 450K. The quality of the representation of poT data for T > 450 K is dis cussed together with Fig 18. The eriical region wil be ioowed Hn detail in S00. 63. Figures 17 and 18 show thatthe new equation of state, Eq. (53), represents the selected experimental ppT data significantly better than all the other equations available for methane do. The sjstematic deviations between the IUPAC equation’ and the equation of Friend eal. and ‘our equation are based on the fact that both correlators relied too much on the data of Goodwin etal 8 As die ‘cussed in See. 23.1, we used Goodwin's corrected data ‘only for the development ofthe new equation of state but the deviation plots contain the original (uncorrected) data of Goodwin etal. It has already been mentioned thatthe data of Goodin deviate by up t0 0.2% in density from recent state-of-the-art measurements. pol data forp < 12MPaandT< 450K Figures 19 and 20 show comparisons of data from state-of-the-art ppT experiments with density values ca culated from Eq. (5.3). The new equation agrees ‘+ 002% in density with the ppT data of Kleinrahm crol 24 Pieperbeck etal 2 Achtermann otal 20 and Handel etal? All ppT data sets are represented within their experimental uncertainties; see Table 16. poll data tor p > 12 MPa and Y'< 490K Figures 19 and 20 illustrate that, also at higher pres- sures, Eq, (53) gives a good representation of the ppT data of Morris", Achtermann et al." Jaeschke,"” ‘Trappeniers etal," and Kortbeek and Schouten." The experimental and calculated densities generally agree Within + 0.05%. The range from 30 MPa to the melting pressure at temperatures below 250 K is not covered by reliable data, Nevertheless, we estimate that the uncer- tainty of density values calculated from Eq. (5.3) in this‘THERMODYNAMIC PROPERTIES OF METHANE 1097 300 x el 194K ~ 196 K ea scmseep iat ° uuu iit 9 2 ne HuK= MEK 9 9 10 - 100 ° i tu 2 th 240 K = 242 K. -0. apa HH =O, pb ui oemeto* 10° 40 ao ae souaua40) Pressure p/MPa Pressure p/MPa © Haendel et al. + Goodwin®* w Rudusevich @ Miller’? = Pope” © Roe™ v Kleinrahm et al? “— Friend et al" Phase, boundary - ~~ TUPAC! easy deviation ofthe experimental pT daa from vals cleloted from Eq (53). For the experiental n= seo Table 16. ia 17. Perens region is less ‘cause the fitting and opti achieved be- ‘There is only one set of Group 1 measurements which tion process included speed covers temperatures above 450K (see Fig. 6). Douslin ‘of sound measusenrents which cover the rage mentioned eral.® published 374 measurements of the compression above. factor z in the temperature range between 273K and 623 K at pressures up to 37.5 MPa, The total uncertainty poT data for T > 450K in z claimed for their measurements increases to about J. Phys. Chom. Ret. Dat, Vol 20, No. 6 1981ro90 U, SETZMANN AND W. WAGNER 250k - 262 K 0.2 0.2 0 oO ~0. Lun i ue ann am 0.2 Fee 0.2 Hh ‘ “o-eito™ 40" 40° 40°. 40° Yo Yor tO 40% 40" Pressure p/MPa Pressure p/MPa © Haendel et al: + Goodwin? = Kleinrahm et al. @ Trappeniers et al’? + vaescnke 6 Hinze" © Roe™ © Pieperbeck et al? a Morris's" ® Douslin et al? & Schamp et al’! — Friend et al” s-o- IUPAC’ Fie 18, Percentage density deviation ofthe experimental pp dt from vals calculated from Eq, (53). For the experimental uncer ‘aindne ve Tae 16 ‘J. Phys. Chem. Ret Data, Vol. 20, No.6, 1901,‘THERMODYNAMIC PROPERTIES OF METHANE. 0.06 1099 178 K = 180 k 100 (Proe™ Pease) / Poot -0.06 “oF 410" 40" 40° Pressure p/MPa + Goodwin*® = Morris" m Pieperbeck et al.” + Jacoohke & Hinzoll? « Schamp et a1’ ~~~ Phase, boundary - TUPAC! © Haendel et al? © Roe"? x Kleinrahm et al’ 1 Aehtonmann ot ai? @ Trappeniers et al! “Friend et al’ Fic. 19. Percentage density deviation ofthe experimental ppT data from valves elated fom Ba. (3) a a high resaution with respect to deny. For the expernental ucetalaes se Table 16, +0204 at the highest temperatura and praccure. The data are well reproduced by Eq, (5.3) within this error limit. Unfortunately, these data already show a system- atie deviation from our primary ppT data at temperatures below 430K. ‘Iheretore, we estimate the uncertainty of, density values calculated from Eq. (5.3) tobe = 0.07% to ‘© 0.15% for temperatures above 450 K at pressures of 20M a Figure 21 illustrates comparisons between selected ex- ‘perimental data ofthe second virial coefficient and B val- tues calculated from Eq. (9.3). The calculated second coefficients are in good agreement with the data of Kieinrahm etal," Roe," and Douslin etal.” 4. Phys. Chem. Ret, Dats, Vol. 20, No. 6, 19011100 0.08 U. SETZMANN AND W. WAGNER. 293 K 5) / Part 100 (brane Pe 407 10° 407 40" 40° Pressure p/MPa + Jaeschke & Hinze" + Achtermann et a1? = Morris" wo @ Trappeniers ct ot x Achtermann et al” Friend et al x Kleinrahm et alt a Pieperbeck et a1? x Schamp et a1!" ; “ Korebeck @ Cohouten’ TUPAC’ Fr. 20, Percentage density deviation ofthe exerinental pp data fom values caked fom Fg (53) in hgh ‘esolaon wih respec o dens. orth experental uncertain see Table 16 6.23, Speed of Sound Figure 22 shows the percentage speed of sound devia- fione along sovera isothorme plotted ageint procoure. ‘The high quality ofthe selected data representation a all temperatures and pressures characterizes the new funda- ‘mental equation, Eq. (5.1), Neither the IUPAC equation’ nor the equation of Friend et al.” is abl to represent the speed of sound inthe liquid region and in the high pres- sure region within the uncertainty of these dats. How- cver, many mearuramente of the propertiee of methene 21 ‘4. Phys. Chom. Ret. Dat, VoL 20, No. 6, have been carried out in the last few years so that these. data had not been available to most of the previous corre- lators. Tn 1986, Kortbook ora publiched pp and epeed ff sound data for methane. At that time we had already started with the development of our fundamental equs tion. Our preliminary equation of state was not able 10 reproduce the w data of Kortbeek although the data were included in the fitting and the optimization process. In 1989, Kortbeek and Schouten" repeated their measure- mente,"The new eet of data.on the epeed of eound dies‘THERMODYNAMIC PROPERTIES OF METHANE Temperature T/K io 21 Absolute dvistion AB = (Bom ~ BB of experiment ‘cond iri oon from ales nelated from Bg (83). For the experimental unceints sce Table 1, vp to ~ 0.7% trom their first measurements. Although the data now agree much better with our equation of, state, we believe that the maximum uncertainty of “+ 1119%, claimed hy Korthenk a af for their new epaed ‘of sound data, might stil be a little bit too optimistic. 1624. The Isochorie Hest Capacty Figure 23 shows the percentage deviations of c, data from our new equation of state, while Figure 24 shows an sholute plot of ein the liauid revion close to the satu- rated liquid line, Figure 2 illstrates that Eq, (5.1) can represent the data of Roder™ better than the data of ‘Younglove" although both experiments were performed with the same equipment, Figure 24 shows that the TU- PAC equation’ as well as the equation of Friend et al. probably exhibit a physically incorrect behavior in the liquid resion at lower temperatures. The values of the ‘sochoric heat capacity on an isochore are expected 10 increase wth decreasing temperature when approaching the saturated liguid line, and the heat capacity on the isnot expected to have a atures. This is atleast the experience from corresponding investigations of other well-measured fuids, e. Os, HAO, and COs 62.5, The Ieobere Heat Copacty Figure 25 gives an impression of the percentage devia- tions of the cy data from values calculated from the equa- tion of state, Eq, (5.1). With the exception of the low temperature ‘region and the critical region, all investi gated equations of state show very similar deviations from the experimental data, Except near the critical point, the ‘maximam deviation between the date and the values cal culated from our equation of state is about = 2% which ‘corresponds to the estimated experimental uncertainty; see See. 23.4 1101 6.26. Ditference of Enthslpy ‘The absolute deviations between experimentally deter. mined differences of enthalpy by Ayber,"” Kasteren and Zeldenrust,"® Dawe and Snowdon,” and Dillard er al."* and Ah values calculated from Eq. (5.1) are presented in Fig. 26. The main part of the enthalpy differences pub- lished by Dawe and Snowdon (Magma = 124.7 Ug) is reproduced within its stated uncertainty of 0.3 J/g. Ayber ed the integral Joule-Thomeon effect (AT. OK, Apau = 8MPa, Ait = Oki /kg) so that the maximum deviation of 1 J/g is acceptable, especially when bearing in mind that Ayber used methane which had contained 1.3 mol% impurities. 62:7. Joule-Thomeon Coeticent Figure 27 shows the Joule-Thomson coefficient caleu- lated from three equations of state and the experimental coefficients of Budenholzer etal." on six isotherms. The equations, which al agree in this egion, indicate thatthe experimental Joule-Thomson coefficients of Budenhotzer etal. are not consistent with the other reliable experi- rental data which the THPAC equation. the eqation of state of Friend eal. and Eq, (5.1) are able to represent. 6.3. Critical Region ‘The opinion has been widely held that analytic equa tions of state covering the whole fuid region are not able to represent the properties in the critical region. In this contert, we define the enlarged critical region of methane by a temperature region between 187K and 205K (T.=1906K) and densities from 90 to 230ky/m? Gye = 162.7 kgin') We wil show that our new equation able to represent the existing experimental information ‘on the thermodynamic properties of methane in the crit- I resion at least as well as the revised and extended scaling model of Kurumov etal." The validity range of tion corresponds tothe region defined Comparisons of the calculated and measured vapor pressure, saturated liquid and vapor densities for temper- stures close to the critical temperature are given in Fig. 28. The scaling law equation has difficulties in repre- senting the sotrated liguid and vapor densities in the critical region. Our new fundamental equation, however, reproduces all experimental pp" data of Kleinrahm and Wagner! within should be mentioned that the new fundamental ‘equation, Eq, (5.3, isnot consistent with Eqs. (3.4) foro" and (3.5) for p° with regard to the critical exponent B. ‘The critical exponent of Eqs. 3.4) and (35) is B = 0384 whereas the critical exponent 6 of Eq. (53) is 05 (for ‘T+. like for any analytical equation of state). How rer a9 Table 7 shows, tha tnconaetoncy cxne calyfor (1-7) € (20 t0 30)mK, and for (7,—T) > 50 mK the three equations completely agree with respect to B. On the other hand, the value has very litle to do with the 4. Phys. Chom, Ret. Data, Vo. 20, No.6, 19911102 U. SETZMANN AND W. WAGNER : st i 273k a 423k fu 107 10" Pressure p/MPa > Van Itterbeek et al" © Baidakov et al'® Gammon & Douslin™ +r Goodwin™ + Terres et al” === Phase, boundary TUPAC! 107 10" 40 10° 1 Pressure p/MPa < Kortbeek § Schouten" # Straty™ © Sivaraman & Ganmont © Leaming™ 10° 10" 40° Friend et al io. 22 Percentage deviation ofthe experimental speed of sound data from values clelte from Ba, (61). For the experimental uncertainties se Table 2 quality of any correlation equation. Although the Kuru- mov equation" contains the so-called “true” value {= 0.325 corresponding to the universality hypothesis of the renormalization group theory, this equation is just not able to represent the saturation densities of the critical region within their experimental uncertainties. ‘A.comparison of Eq. (5.3), the scaled equation of state of Kurumov etal," the equation of Friend eral," and the IUPAC equation' with the ppT data of Kleinrahm 4 Phys. Chem. Ret, Data, Vo. 20, No. 6, 1901 etal and Hindel etal is presented in Fig. 29. Agai ‘only Eq. (5.3) is able to represent the data within the ex- perimental uncertainty Up to now, the anomalous behavior of calorie proper- ties of methane inthe critical region could not have been properly modelled by analytical equations of state. With the help of special terms (modified two-dimensional Gaussian terms) in our equation, we are now able to rep- resent the caloric properties speed of sound and heat ca-‘THERMODYNAMIC PROPERTIES OF METHANE. 1403 1C, erst Sy, cate 100 (Cy, expe B 384 g/t ° Pi L 100 200 300 Temperature T/K © Younglove™ rnase pounaary TUPAC! 400 200 300 Temperature T/K + Roder™ Friend et al ie. 2, Percentage deviation ofthe experimental tochorc hat eapaciy dats fom values calaated from Ea (51)- For pacity in a wide range of temperatures and densities as well as the scaled equation of state of Kurumoy et al. ‘Although, of course, the new fundamental equation does not reproduce the limiting properties when approaching the eritical point, which is however, not of importance for ‘normal’ engineering applications, it is possible to repre- sent calorie properties much better than all the other ex- isting equations of state for methane do. A comparison between the new fundamental equation, Eq. (5.1), the scaled equation of Kurumov etal. the equation of Friend etal, and the experimental speeds of sound of ‘Gammon et al. is presented in Fig 30. Eq (5.1) yields the best overall representation of the w data of Gammon tal, For temperatures (T~T.) > 1K, Bq, (51) repre- sents the data within their experimental uncertainty of = 196. Closer to the critical point (within Te + 0.5K), the scaled equation is superior to our equation but i is ‘obvious that is agreement with the experimental data de- teriorates very rapidly as soon aa the scaled equation is extrapolated outside its very small range of validity. Di rectly at the etical point, Eq, (5:1) yields a speed of sound value of 205,95 mis which i, of course, in confit With the expected limiting value O mis. Table 38 shows in ‘more detal which values for w and ey Eq. (5.1) and the scaled equation of Kurumov' yield in the immediate ity ofthe erties! point. ‘J Phys. Chem. Ret. Dat, Vol 20, No. 6, 19811104 2.2 U. SETZMANN AND W. WAGNER hori: heat capacity cy/ kd/ (kg K)) 1 Youngiove™ + Roger 3 ca. (5.0 8 : 2 Tupac too 720 740 160 160 Temperature T/K ‘ic. 24, Representation ofthe tochoric heat capscies inthe lig stat ow 180 Ky Eg (5.1) and ty other equation ‘The representation of the isochoric heat capacity data fon near crtiealjsochore, p = 163 kg/m? and onthe so chore p = 189 kgm’ by Eq. (5.1) is shown in Fig. 31. The ‘comparison for the 189 kg/m? isochore shows that the ¢. data of Younglove™® are systematically higher than the more recent data of Koaer'™ taken withthe same appara: tus which was also used by Younglove, With the excep- tion of Roder’s ¢ value at T = 19097 and p = 189 kg/m the sealed equation of Kueumew et al hut also the new fundamental equation, Bq. (5.1), are able to repre- sent all data within the experimental uncertainty esti mated to be about + 3% to * 5% (see also Table 2) “This is also valid for the ritcalisochore p = 163 kgm? ‘where, however, the data do not extend close enough to the critical temperature, The plot of the critical isochore len shoves that Fg (6 1) i in agreement with the sealed equation for temperatures (T'—T_) > 1K but at T = T. Eq, (5.1) yields the finite value ¢y ~ 3.6kI(kg K), see also Table 38, The TUPAC equation’ and the equation of Friend erat. are notable to Tepresent the Isoctorc teat capacities in the near critical region. ‘At the end of this critical region discussion, in Table 38, Ea, (5.1) and the scaled equation of Kurumov et al." 4 Phys. Chom. Ret. Data, Vol. 20, No.6, 1991 are compared with regard to their values for w and, very ‘loge to the critical point, Table 38 shows that the resis calculated from the two equations deviate systematically for (7~T.) < 1 K. When approaching T,, the deviations between the two equations increase clearly. For (2=1,) < 10°", nowever, ihe course oF the two equa- tions is only based on the mathematical model and not supported by experimental data. It is also interesting to note that for (T=T.) < 10"'K the values calculated from the scaled equation forw are far away from zero and forc, far away from infinite. However, this equation can- not be evaluated for (T~T,) < 10~*K because this re- gion is excluded from the application range due 1 ‘numerical problems. When we switched off the tempera- ture limit of 10°* Kin the program package of the Kura- mew equation, at (7—T) = 10" K we calculated with this equation w = 50.6 mis and c, = 58.7 kI/(kg K) 6.4, Metastable State Baidakov et al. published measurements of ppT data (Ref. 150a) as well as speed of sound data (Ref. 185) in the metastable region. The poT experiments were only‘THERMODYNAMIC PROPERTIES OF METHANE. 1.7 Wee 1105, 3.0 wee erat” Create) Ce, erpt 100 ( HY che it %00 200-300-400 Temperature T/K a Budenholzer et aly m Kasteren & Zeldenrust’ = Phase, boundary ~ TUPAC! Pro: 3, Pesontage deviation of the anper relative measurements because Baidakov and Gurina’™ determined their densities from density values onthe sat- ‘rated liquid line which had been given by Goodwin and Prydz” and McClune.™ As discussed in Sec. 2.2.4.2., oth data sets for the saturated liquid density used by Baidakov and Gurina deviate systematically from the state-of-the-art data of Kleinrahm and Wagner? and, nec- essary, the ppT data have to show systematic deviations {rom values calculated from Eq. (5.1). Fig. 32 shows the representation of the pT data on six isochores and we conclude that Eq, (5.1) is able to represent the properties in the metastable region as well as the properties in the homogeneous states. This statement is supported by 400 200-300-400 Temperature T/K + Jones et al® Friend et al? el nour hat eapciy date re value elated fom 1 (6. Fig. 33, where the representation of the speed of sound data of Baidakov ef al. ig shown, For both ppT and speed of sound measurements Baidakov etal. used ‘ethane with a purity of only 99.9 mol. It is commonly known that the effect of impurities becomes significant when the meaturements approach near crillal condi- tions. 6.5. Extrapolation Behavior of the New Fundamental Equation To conclude the discussion of the new equation of state, its extrapolation capability is examined with the 4. Phys. Chom. Re. Data, Vol 20, No. 6, 19011108 U. SETZMANN AND W. WAGNER 7 sper tossiik 4 sve 9 ° Fg | : i syoan 10 = 248 aw 3 Sot tm} 0 ~ a nm “4 a gl — 5 v0 : 10164 zoo sear ° Me q “4 ol ea 2104 A 19.4 we 2 ao “s ° -4 - i 2.2 we i 3.000 Se o ° ° a “40 goo a0a a0 “4G GoaS8e eo Temperature T/K Temperature T/K Aybar m Dawe & Snowaon x Kasteren & Zeldenrust™ Dillard et aly ‘Fe 26, Absolute devition of experimental dts of diferenes of enthalpy frm vals cleulatd fom Eq, help of Fig. 34. In 1981, Nellis l.™ caried out shock wave measurements in methane. Based on the conserva tion relations for mass, momentum, and energy, they ob- tained data on the Hugoniot curve f= hat = 03M + pasd(Mpat~ 1/9} (03) where iis the enthalpy, p the pressure and the density after the shock wave experiment and has, pat. and pa are the initial state values. The comparison of Eq, (5.1) t0 the experimental Hugoniot data, which were not used to fit the equation, is shown in Fig. 34, The agreement of the Hyon curve calculated whit he help of Eqs. (3.1) and (63) with the data of Nellis et a.* is excellent. A safe extrapolation of Eq, (6.1) might be possible for pressures ‘up to 20000 MPa, 41 Phys. Chom, Ret, Data, Vol. 20, No.6, 1991 7. Accuracy of the New Fundamental Equation Estimates of the accuracy of the density, speed of sound, and enthalpy calculated from the new fundamen- presented here have been made and are tne wolerance diagrams, Pgs. 33, 30, and 37, respectively. The estimates were derived from compari sons ofthe various sets of experimental results and taking into count our own experience in judeing experimental ata. ‘The calculated values of heat capacity (¢ and c.) are estimated to be accurate to within = 1%, The estimates OOF the unicervalndes OF uke enutalpy shown bi Fg 97 ate considerably less certain than those of the density and speed of sound. Because ofthe scarcity of enthalpy data, the tolerances given are only rough estimates.THERMODYNAMIC PROPFRTIFS OF METHANE 1407 icient p/ (K/MPa) coef @ Budenholzer et al" — ea. 5.4) ~ TUPAC! Joule-Thomson Pressure p/MPa "ia, 27, The pot ofthe Jole-Thomson coefficient inthe region below 11 MPa caleultd rom Eg (5.1 and fom other equations Phy Chem. Ret. Data, Vol 20, No. 6, 19911108 U. SETZMANN AND W. WAGNER 0.03 heerainty 2 8 0.03 > e q } eS 2 Uncertainty -0.5 vd ri el Uncertainty \ 3 i ° ° 0.5 i 196 °~«87°=—~=«CSSSCKUGSSSCtBO| AH Temperature T/K . x Kleinrane 6 Wagner? ~~~ kurumoy et al” —--— Friend et al? Tupac’ enation Ne m a 9d ~ ps af cated ampere et on the vapor liquid stration Inn th ral pion frm ales ealelated om Ey (52) Ths uncertain ranges ofthe experimental data are plated at well, Sine the Karumor equation ‘sony valid for temperatures T'> 187K, the py and p" values calculate for T'< 187 Kcor- Qa, Vol 20, No. 6, 1991‘THERMODYNAMIC PROPERTIES OF METHANE. 1109 190,568 ““d0 50 100 150 200 Density p/ (kg/m) © Haendel et al? 7 Uncertainty 2 Pope* Friend et al" "R29, Percentage presse deviation ofthe experimental data from ales calulated trom Ea, (5 0. ‘0 50 100 150 200 Density p/ (kg/m) y Kleinrahm et al? T uncertainty Roe" ~~~ Kurumov et al? in the xia Fepion. The ploted uncertain range corespond fhe total uncertain pressure forthe data of Keira (ral? and Hindle? 4. Phys. Chem. Ret, Dat, Vol 20, No. 6, 19911110 U. SETZMANN AND W. WAGNER, 400 / | i B30 J i] y i ~ | = |h i 2 2 5 200k 200.6 j 3 e ek j = ~~ 4 2 / 3 ‘if @ 260 Bowne 4 a A 74h WEEKLY Nes 190.86K= 03K ‘Me’ 200 2 80 Too 7 B00 250 Density p/ (kg/m) @ Sivaraman & Gams ‘4 Gamaon § Douslin®™ 9. 0 Friend et a1" 0.3, Resa of th eed fs i the il eo by 8 an by ter ‘funtion Sines the Karome equation nol ali for 90 yh © p © 200 gla, ‘hues clelted from this equnton ote this reglon correspond to etaplated va ‘i The uncetsny of the data nti gre wae estimated foe about = 15 which oresponds rough 10th sas of the sybols pled, 4. Phys. Chem. Ret. Data, Vol 20, No.6, 1981,‘THERMODYNAMIC PROPERTIES OF METHANE. nit i p= 163 kg/m? x pe cy / (ka/ (kd K)) Sagal md Te 7300 erty Temperature / K T = |i ea 3 i 2 | = ah 3 # . e i, Te (p'= 189 kg/m?) r 2 480 a0 Zo Temperature / K @ Young2ove™ + Roder™ io. 31, Representation ofthe lachore heat capacy inthe eal plan by Eq, (5.1) and ty eer equations, The uncertainty of the expeimental dat inthis reioncore- ‘ponds oughly fo te se of te sols potted. 14. Phys. Chem. Ret, Data, Vol 20, No. 6, 19911112 U. SETZMANN AND W. WAGNER: oe 224 ig/s! ae 28 kg/at hee Se ele i 1 2 T 2 1 e i 1 ef 2 0.2 1 SS an 204 g/t on 957 gt 3 ae at Het eee L ° ; on 1 1 B -o.2 -0.2 i Soe ste oe _st8 ual S ttt Hetty pee 7 ° ! 0 ; 1 1 ~O40F 410° 40° 407 1 ‘40° Pressure p/MPa Pressure p/MPa + Baidakov & Gurina’* ~~~ Phase boundary Fic 32 Percentage density deviation of experimental pp data on sx selected isochores in the metastable region rom values eluted fom B ($3). = = a . 3 1 = Ss = 3 Leet " fi a 107 10° 40° 407 10° 40° Pressure p/MPa Pressure p/MPa © Baidakov et al” Phase boundary TUPAC" Friend et alt from values cauated 1,38, Percentage deviation ofthe experimental sped of sound data inthe metastable ep ‘tom E4(51). 4 Phys. Chom. Ret. Data, Vol 20, No 6, 1991‘THERMODYNAMIC PROPERTIES OF METHANE 20000) 1113 Famerren ot at 18000 Pressure p/MPa Density p/ (kg/m) ‘Per M4 The experimental Husoait data in comparison with core ‘ponding values caleuatd fom Eq. (81) and foe ober ‘uations, The hatched area corresponds to the region where Eg (5) has been fied wo experimental deta, " F ¢ g 5 ae 8 = ir wy | bees BE f Sipe Bt 100 ted _30 a5 BOT —T00 Temperature 1/% Fro, 36, Tolerance diagram for dens (in region D the percentage prose unertany ges). ne vale for Ne ences tespond to tie the standard deviation. Pressure p/MPa Ip ji Be x00 Bad 300400500809 Foo Temperature T/K Fi, 36. Tolerance digram for speed of sound. Inthe uncertainty statement forthe regan , the temperature range (7-7) < (3K exuded. The values forthe uncertainties correspond to twee the standard deviation o Pressure p/MP3 Beosay - tS 4000030040050 B0o Too Temperature T/K Fo. 37, Tolerance diagram for eathapy. The valves forthe unertaa 4. Phys. Chem. Ret. Data, Vol 20, No, 6, 19811114 8. Conclusions Based on recent experimental data on the thermody- namic properties of methane, a new fundamental equa- tion for the Helmholtz energy has been developed which covers the temperature range from 90 K to 625 K at pres- sures up 1 1000 MPa, With the help of a new optin tion procedure’, we have established corre ‘equations which offer a state-of-the-art representation of the thermodynamic properties of methane. ‘The new equation of state, Eq. (5.1), which corre- sponds to the new temperature scale ITS-9, represents nearly all experimental thermal and caloric data, which are considered to be reliable, within the experimental certainty. Ths statement is not only valid for the homoge- neous phases, but also for the vapor-liquid coexistence cenrve and even for the crtieal region. Furthermore. the ‘equation allows a meaningful extrapolation for pressures ‘up to 20000 MPa, Based on the high overall accuracy, the ability of area sonable representation of the erlueal region, and the ex trapolation capability, the new formulation i a significant improvement in comparison to existing studies. 9. Acknowledgments ‘The authors are grateful to the Deutsche Forschungs- {gemeinschaft for their financial support of this project. We are indebted to P. J. Kortheek, E. C. Morris, H. J. Achtermann, G. Magnus, G. Handel, and N. Pieperbeck for providing a copy of their experimental data prior to publication, We would also like to thank very much A. russ and H. Mller for their comprehensive help in con- verting all the equations to the ITS-90 temperature scale, for producing the figures on the computer aid checking all the equations and tables. 10. References. 'S, Angus B. Armtron, and K. M.de Revel, Inematonal Tales of ‘he Fd Site = Methane, Intemational Union of Pure and Apptied Chemitny (Penexmon Pree Oxford. 1979). Vol. 5. 2RKleinraen and W. Wagner, J. Chem. Phetmadya. 18,79 (1986) OR Keinahm, W-Dusche and W. Wagner, J. Chem. Thermos. 1, 1103 (1986). “R,Kiinrabm, W. Duschek, W. Wagne “Thermodyn, 3,621 (1988). 5G, Mandel, R, Rlcinrahm, and W. Wapner, Mesurent of he (pe see, deny temperate relation of methane in the homogeneous as ‘nd lid region nthe kmperate rage fo 100 Ko 200 Kt pre Soe ip too tra, Suomntce to. cae. tnerooyR. U9}. A. Sharaman and B. E. Gammon, “Spedfsound menaerens in ‘nal gs fade,” Gat Research Instate Report 860085, 1986. “E-Bender, Equton of tat rpeening he phate ehaviow of pure ab- tunes oecedngs Othe 3h Symposia on Thermophys! Prop- trios (American Society of Mechanical Engineer, New York, 170), pp. 227-25 MUD. Goodwin, The Themoplpsical Properties of Methane, fom 9010 ‘00 Kr Pen np fo 700 bar Nat. Stand (US) Tech. Note 653, 1974 ‘RB. McCarty, Cryogenics 14 276 (17), V. V. Scher, AA. Vsierman, V. A. Zagoruchenk, A.D. Korky, GLA Spirdones, nd V. A. Taymaen, Themodmamie Popes of dM. Jeschke, J. Chem. 4 Phys. Chem. Ret. Dat, Vol. 20, No.6, 1901, U. SETZMANN AND W. WAGNER: Methane, (Wtenisphere Publishing Corporation, Washington, 1987) ‘ignalypobithed by Standards Publishers, Moscow, 19. °F Elyand H.1.M. Hale, Ind. Egy. Chem. Fund 3,323 (1981). "8G, Sven, Sota, VBL Rosca 62 (1044). ‘D, D Erickson and" W. Leland In Ie Thermophys 7, 911 (186) B.A. Younglove and J. FEL, J. Phys Chem. Ret. Data 16, 377 980, SD _. Kurumor.G. A. Okhowy. an J. V. Senger It J. Thermophys 9,75 (588. "DG. Friend, J. Eby and H.lngham, 3 Phys. Chem. Ref Data 18, 583 (98, M, Benedict, GB. Webb, and LC. Rubin, J. Chem. Phys 8 334 (0940, ™U,Setamann and W. Wapner, Int. J. Thermophys 10 108 (1989). RD. Goodwin and R. Prydz, J. Res. Nat. Bur Stand, Sec, A, 81 97), "D.R.Dowsn, RH. Hanson, RT. Moore and F.P. McCullough, 4. Chem, Eng, Data 9, 358 (1962. VM. Cheng. PhD. Thesis, Princeton Unites, 1972 N. Pepereck, R Klein, W. Wagner abd M, Jeschke, J. Chem. cmon. £3,113 (39) PHL. Achtermnn J. Hong, W. Wagner, and A. Pras, Refiactve Inder ‘an Deny Ihe for Methane from 273 Kio 373 Kanda Prsaes 1034 MPa, Sabmited to J. Chem Eng. Data (1991. Sa Teno They Meuslogse 23 200). "Supplementry Tformatin forthe ITS. International Bureau of Weighs and Measure Pavilo Bete, F-92312 Sewes, France, 1960. (0981), Complete Text: Chemie Ing. Techn. MS 93381. 2U, Setmann, Eine neue Method 2 Opinierangthemodymamischer ‘Kanelatonslichungen nd Sie Amend tar ftelg cnr Pu ech Poe. VDF ae Heh 3 t PD. R. Lovejoy, Nature 197,35 (1963. 1,C.G.Calado,G.A. Garcia, and L.A. Staveley, J. Che. Soc,Farar ay Trane 170, 148 (1990, ‘PAL. Kida, KL. Leni JC. G. Cala, and LAK Staveley, Chem. Thernodys. 7, 87 (1979), mp, Paes, G. Cagna, and. Feri, The eile pba of pre methane, Prosedings of the th Ineratonal Cyopenie Conference, TPC ‘Sek snd Tecan 1976 pp. 281-25 23) Bonhoure and R. Pella, Mettloga 14, 15 (1978) 2A. Inaba and K. Mitsu Jp. J. App. Phys. 17, 1451 (1978). 2 Boshoure and R Pella, Metrlopia 16, 95 (198). ME. Povra Metrologia 17,25 (1981). 2 Blank, PTB-Miteltngen 9,250 (1983). RE. Bedford, G. Bonnie, H. Mas, and F. Paves, Metoogi 20, yl 0988). Furulawa, RC Kemp, H. Mass RL Roby, Hi, Sakura and Ling Shan Kang, Metoogi 10,127 (98. 2K. Otzenai, Compt. Rend, Acad, S10, 90 (1885. M.A. Hunter. J. Phi. Cher, 10, 330 (1906), ‘SFr Honing and A Stock, Z Physik 4, 26 (1921. “RG. Keyes, LB. Smith and DB, Joubert, J. Math, Phys, 191 922, 5, Young Proc. Roy. rsh Acad. 38, 6S (1928), OF, 6 Bache sd eB Sit, Data were podished m Ket G.T. Armstong FG, Brckwede, and R. Beat, JRes. Nat But Stan, 85, 391955). “0.7 Bloomer and J.D. Parent, Chem. Eng. Prog. Symp. Se. 49,11, (255. “F.G: Keer an HG ura Arm: Chem Sa, 143 (192. {BR A Puan 1TH. Verto Proc BS A104 19, "=H, vanes a V1, Gna A, Chem Sx 8,994 (1830- "2A. Mies and GW. Nedra, Pays 300 (1955). "5A Mth and GW. Nedeorag Pac 3,369 (1820, "AH. Oi HH Reamer, BH Sag, ad W.N Lacy, Ind ag. Coen 35,92 (1988). vy 5 Kenaoon and GT, Levchenko, Zn kc 16, 580 (940, !PN. V. Palovch and DL Tmo, Teplonergstita 5, 6 (1958). ‘Daten o. none, ut Cném Kents non Aguous 0+ 1867 (98 ‘OW. cle, W. Lend and R. Kobayashi, A LCE. 6, 267 (58). ia Werth ©. Veta aH tan Py a 0, 748 092) AL Delt I Ellin and F Boks Indi chim. big 9 6 (0, 20° A. Dobro. N. Bela, an I F. Gabe, Gaoja ‘Prolene 947 (68). Con Was, Sieber, nd J.J. Meet, Hooetbon Pro- centing 8,17 960. 4 Phye. Chem. Ret, Data, Vo. 20, No, 6, 18811116 "AE. Hooter, Ph D. Tess, Rice Unessy (1960. "ETS. Huang Ph D. Theis, University of Kansas (196). YA M Mamedov and A. Mamedos, lvestiya ish achebuych avedenil Net xr 12.71 (196), ‘SH HL Je Meath and W.. Easter, AT. Ch E31, 370 (196), "5, Roberson and SE, Babb, J. Chem. Phy. $1357 (1968). "A. D. Eppery, Ph D. Thess, University of Mistoos, Columbia sm, "*¥V. A. Sorokin ad Yo. P. Blgo, in: Thermodinamicesie i ther: snothiniesje Konstaniy, Ede KV. Astochoy, Nauka Publishing House, Mesow, 97-10 (1970). WD. 8. sis, LR. Liasis, and. S. Timmerman, DAN SSSR 198, sear 618. Rodosvch and R. C Miler, A. 1. ChE. 19, 729 (1973. A, RButaev and V. G Skip, Oaov. Proms 12, 4 (197, ™S Mihara, HE. Sgar,Y. Ara and, Saito J. Chem. Eng. Japan 19, 35097 °SA.2.Franesconi, PhD. Thess Unies Friderclana Karsh (978. ‘MM, Nunes Da Pont, W.B. Stet, and L.A. K, Staveley J. Chem, “Thera 1 101 (78) 2°. Aehtermann,F. Kick Kral 34, 256 (1982), ‘9H J. Ackermann, F,Klbasa and H, Rogeer, Bremnst-Warme- Kraft 34 311 (198), {SNE M. Amer, PhD. Tess, Unversity of Misou (1986), °°), RS Machado, W.B. Steet, and U: Deters. Chem. Eng, Data 35,148 188), YG Daido and TA Grin, J hem, Themed 2, 109 and H, Rogener,Brenst-Wirme (an ‘W.Lemaing Forsch Be. VDKZ. Rete 19, N32 (156), 2, A Beat a W.H,Stckmeye, J. Chem Pps 1 47 (182). °°5,D, Hamann, A Lambert and PB. Thomas, Aust. Chem. 8 9 a3. “RD. Gann MS. Tei, Unies of California Ree (188 Data ken fom: 3A Hu and TM. Reed, Chom Engne. Da 4,36 (1869) keer and Reven Sein tere 239 (40 ‘newer, Deeminaon of ied a ein, US A Foc Re port AFOSR No. 67-208 (05) MCA. BnesMR Jones snd AK Saeley, rns. Parade Soe ine ase ‘RB Homer, Noga, T. W. Je Leland, apd R Kaba, (Chem. Ps 4,205 (1968) i NL and K. Sor, Be. Bunsnges Py Chem. 7, 2 096), vRSE ee and W.C Eder, A ChE 3.16 1047 (97, "SRE Sin, RCN: Lichelr, Sevan, nd K.SchalyBe Bunseng Fs Cem. 75, 158 (97D, 6} elm W- Rebels K Scaler, and Schramm, er Basen i Chom. 282 (199). VOR Hab, K Shafer sn 8, Schramm, Ber Bansenge. Phys. Chm. 73.27 Gi. "=. Ker an H. Hb, Ber Bunsenges. Phy. Chm 8, 82 (184). “ef Dokl Dy ape W. Lary, ad Egy Chm 3,00 (59. "D-N SatglJ.M. Gus, AJmbbelar and G.M. Won, Ad. Cy Eig 10, 24 (196). sagt, Vl oe Be 4,168) "9D. Dir, W-C Easter JH Er, and RL. Robina, ACh EJ. 14,925 1568. YR ADawe and PN Soon J CRem. Eng. Dats 19,720 (970 "7.1L 0. Kastren and Zelda nd. Chem. Funda 18 353 (i. AV, Hea, An. Py 686 (198) 2°, Wi, Am. Chem. So 4887 (123, "7. ack and K von Le, Pps, Chem. BS 413 (1929) 8 Anco Ph Mag 0,186 (925) 4. Phys. Chom. Ret, Data, Vo. 20, No.6, 1991 U. SETZMANN AND W. WAGNER, "a ocken nd W. Berger, Z. Gesamte Kan 145 1530, ‘HB Dion C. Camp and A Pate, PR Soe, Landon Ate, 1152. 1 Hui, P:R. 67, 298 1985) "Laem, Rech CNRS 36, 26 (1980. "Tern We Jaa, od Rossan, Brennsott hemi 38, 129 (os "Vin Wertek, J. Tae, A. Cops and W. 162 (96, LR Singer 3. Chem, Py, 4729 (196)- °M,Cndamone TT. St, DP. R- Ein pi Soe Amer. 6, 1254 (1970). a Sriacnay aha \ SBD, a, Cem 48190 (9%), EK: Dreguyay, Tepes ja ulead ch sme ji ney ee avec, soo 1,138 (1973).

You might also like

• Fear: Trump in the White House

  

  From Everand

  Fear: Trump in the White House

  Bob Woodward

  Rating: 3.5 out of 5 stars

  3.5/5 (738)
• Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  

  From Everand

  Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  Gilbert King

  Rating: 4.5 out of 5 stars

  4.5/5 (266)
• A Man Called Ove: A Novel

  

  From Everand

  A Man Called Ove: A Novel

  Fredrik Backman

  Rating: 4.5 out of 5 stars

  4.5/5 (4610)
• • Magazines
  • Podcasts
  • Sheet music
• A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  

  From Everand

  A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  Dave Eggers

  Rating: 3.5 out of 5 stars

  3.5/5 (231)
• The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  

  From Everand

  The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  Viet Thanh Nguyen

  Rating: 4.5 out of 5 stars

  4.5/5 (122)
• Grit: The Power of Passion and Perseverance

  

  From Everand

  Grit: The Power of Passion and Perseverance

  Angela Duckworth

  Rating: 4 out of 5 stars

  4/5 (590)
• The World Is Flat 3.0: A Brief History of the Twenty-first Century

  

  From Everand

  The World Is Flat 3.0: A Brief History of the Twenty-first Century

  Thomas L. Friedman

  Rating: 3.5 out of 5 stars

  3.5/5 (2259)
• Shoe Dog: A Memoir by the Creator of Nike

  

  From Everand

  Shoe Dog: A Memoir by the Creator of Nike

  Phil Knight

  Rating: 4.5 out of 5 stars

  4.5/5 (540)
• Principles: Life and Work

  

  From Everand

  Principles: Life and Work

  Ray Dalio

  Rating: 4 out of 5 stars

  4/5 (608)
• The Little Book of Hygge: Danish Secrets to Happy Living

  

  From Everand

  The Little Book of Hygge: Danish Secrets to Happy Living

  Meik Wiking

  Rating: 3.5 out of 5 stars

  3.5/5 (401)
• The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  

  From Everand

  The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  Mark Manson

  Rating: 4 out of 5 stars

  4/5 (5813)
• Never Split the Difference: Negotiating As If Your Life Depended On It

  

  From Everand

  Never Split the Difference: Negotiating As If Your Life Depended On It

  Chris Voss

  Rating: 4.5 out of 5 stars

  4.5/5 (844)
• Her Body and Other Parties: Stories

  

  From Everand

  Her Body and Other Parties: Stories

  Carmen Maria Machado

  Rating: 4 out of 5 stars

  4/5 (822)
• Team of Rivals: The Political Genius of Abraham Lincoln

  

  From Everand

  Team of Rivals: The Political Genius of Abraham Lincoln

  Doris Kearns Goodwin

  Rating: 4.5 out of 5 stars

  4.5/5 (234)
• John Adams

  

  From Everand

  John Adams

  David McCullough

  Rating: 4.5 out of 5 stars

  4.5/5 (2409)
• The Emperor of All Maladies: A Biography of Cancer

  

  From Everand

  The Emperor of All Maladies: A Biography of Cancer

  Siddhartha Mukherjee

  Rating: 4.5 out of 5 stars

  4.5/5 (271)
• A Tree Grows in Brooklyn

  

  From Everand

  A Tree Grows in Brooklyn

  Betty Smith

  Rating: 4.5 out of 5 stars

  4.5/5 (1929)
• The Glass Castle: A Memoir

  

  From Everand

  The Glass Castle: A Memoir

  Jeannette Walls

  Rating: 4.5 out of 5 stars

  4.5/5 (1716)
• Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  

  From Everand

  Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  Margot Lee Shetterly

  Rating: 4 out of 5 stars

  4/5 (897)
• Yes Please

  

  From Everand

  Yes Please

  Amy Poehler

  Rating: 4 out of 5 stars

  4/5 (1898)
• Rise of ISIS: A Threat We Can't Ignore

  

  From Everand

  Rise of ISIS: A Threat We Can't Ignore

  Jay Sekulow

  Rating: 3.5 out of 5 stars

  3.5/5 (137)
• Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  

  From Everand

  Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  Ashlee Vance

  Rating: 4.5 out of 5 stars

  4.5/5 (474)
• The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  

  From Everand

  The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  Ben Horowitz

  Rating: 4.5 out of 5 stars

  4.5/5 (348)
• The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  

  From Everand

  The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  Brene Brown

  Rating: 4 out of 5 stars

  4/5 (1092)
• Angela's Ashes: A Memoir

  

  From Everand

  Angela's Ashes: A Memoir

  Frank McCourt

  Rating: 4.5 out of 5 stars

  4.5/5 (441)
• The Constant Gardener: A Novel

  

  From Everand

  The Constant Gardener: A Novel

  John le Carré

  Rating: 3.5 out of 5 stars

  3.5/5 (104)
• The Perks of Being a Wallflower

  

  From Everand

  The Perks of Being a Wallflower

  Stephen Chbosky

  Rating: 4.5 out of 5 stars

  4.5/5 (2104)
• Wolf Hall: A Novel

  

  From Everand

  Wolf Hall: A Novel

  Hilary Mantel

  Rating: 4 out of 5 stars

  4/5 (3811)
• On Fire: The (Burning) Case for a Green New Deal

  

  From Everand

  On Fire: The (Burning) Case for a Green New Deal

  Naomi Klein

  Rating: 4 out of 5 stars

  4/5 (74)
• The Woman in Cabin 10

  

  From Everand

  The Woman in Cabin 10

  Ruth Ware

  Rating: 3.5 out of 5 stars

  3.5/5 (2521)
• The Art of Racing in the Rain: A Novel

  

  From Everand

  The Art of Racing in the Rain: A Novel

  Garth Stein

  Rating: 4 out of 5 stars

  4/5 (4203)
• The Light Between Oceans: A Novel

  

  From Everand

  The Light Between Oceans: A Novel

  M.L. Stedman

  Rating: 4.5 out of 5 stars

  4.5/5 (789)
• The Yellow House: A Memoir (2019 National Book Award Winner)

  

  From Everand

  The Yellow House: A Memoir (2019 National Book Award Winner)

  Sarah M. Broom

  Rating: 4 out of 5 stars

  4/5 (98)
• The Outsider: A Novel

  

  From Everand

  The Outsider: A Novel

  Stephen King

  Rating: 4 out of 5 stars

  4/5 (1850)
• Sing, Unburied, Sing: A Novel

  

  From Everand

  Sing, Unburied, Sing: A Novel

  Jesmyn Ward

  Rating: 4 out of 5 stars

  4/5 (1104)
• The Unwinding: An Inner History of the New America

  

  From Everand

  The Unwinding: An Inner History of the New America

  George Packer

  Rating: 4 out of 5 stars

  4/5 (45)
• Little Women

  

  From Everand

  Little Women

  Louisa May Alcott

  Rating: 4 out of 5 stars

  4/5 (104)
• Brooklyn: A Novel

  

  From Everand

  Brooklyn: A Novel

  Colm Toibin

  Rating: 3.5 out of 5 stars

  3.5/5 (1947)
• Bad Feminist: Essays

  

  From Everand

  Bad Feminist: Essays

  Roxane Gay

  Rating: 4 out of 5 stars

  4/5 (1018)
• Manhattan Beach: A Novel

  

  From Everand

  Manhattan Beach: A Novel

  Jennifer Egan

  Rating: 3.5 out of 5 stars

  3.5/5 (792)
• Steve Jobs

  

  From Everand

  Steve Jobs

  Walter Isaacson

  Rating: 4.5 out of 5 stars

  4.5/5 (807)