Equations of State for Mixtures of R-32, R-125, R-134a, R-143a, and R-152a

Eric W. Lemmona
Physical and Chemical Properties Division, National Institute of Standards and Technology, 325 Broadway, Boulder, Colorado 80305-3328

Richard T Jacobsen
Idaho National Engineering and Environmental Laboratory, P.O. Box 1625, Idaho Falls, Idaho 83415-2214
Received 30 December 2002; revised manuscript received 16 December 2003; accepted 31 December 2003; published online 2 June 2004

Mixture models explicit in Helmholtz energy have been developed to calculate the thermodynamic properties of refrigerant mixtures containing R-32, R-125, R-134a, R143a, and R-152a. The Helmholtz energy of the mixture is the sum of the ideal gas contribution, the compressibility or real uid contribution, and the contribution from mixing. The independent variables are the density, temperature, and composition. The model may be used to calculate the thermodynamic properties of mixtures, including dew and bubble point properties, within the experimental uncertainties of the available measured properties. It incorporates the most accurate equations of state available for each pure uid. The estimated uncertainties of calculated properties are 0.1% in density and 0.5% in heat capacities and in the speed of sound. Calculated bubble point pressures have typical uncertainties of 0.5%. 2004 by the U.S. Secretary of Commerce on behalf of the United States. All rights reserved. DOI: 10.1063/1.1649997
Key words: density; equation of state; heat capacity; HFC-32; HFC-125; HFC-134a; HFC-143a; HFC-152a; R-404A; R-407C; R-410A; R-507; refrigerant mixtures; speed of sound; thermodynamic properties; VLE.

Contents
1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. The Mixture Equation. . . . . . . . . . . . . . . . . . . . . . . 2.1. VaporLiquid Equilibrium VLE Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Comparisons to Data. . . . . . . . . . . . . . . . . . . . . . . . 3.1. The R-32/125 System. . . . . . . . . . . . . . . . . . . . 3.2. The R-32/134a System. . . . . . . . . . . . . . . . . . . 3.3. The R-125/134a, R-125/143a, R-134a/143a, and R-134a/152a Systems. . . . . . . . . . . . . . . . . 3.4. The Ternary Mixtures. . . . . . . . . . . . . . . . . . . . 3.5. Other Thermodynamic Properties. . . . . . . . . . . 4. Accuracy Assessment. . . . . . . . . . . . . . . . . . . . . . . . 5. Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . 6. References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Tables Pure uid equations of state for the refrigerants used in the mixture model. . . . . . . . . . . . . . . . . . . . Coefcients and exponents of the mixture equations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Parameters of the mixture equations. . . . . . . . . . . . Coefcients and exponents of the ideal gas 595 595 597 598 602 604 606 607 610 613 619 619

equations for the pure uids. . . . . . . . . . . . . . . . . . 5. Summary comparisons of mixture properties calculated from the model to refrigerant mixture data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Calculated property values for computer code verication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Figures 1. Comparisons of densities calculated with the mixture model to experimental data for the R-32/125 binary mixture. . . . . . . . . . . . . . . . . . . . . 2. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/125 binary mixture. . . . . . . . . . . . . . . . . . 3. Comparisons of densities calculated with the mixture model to experimental data for the R-32/134a binary mixture. . . . . . . . . . . . . . . . . . . . 4. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/134a binary mixture. . . . . . . . . . . . . . . . . 5. Comparisons of densities calculated with the mixture model to experimental data for the R-125/134a binary mixture. . . . . . . . . . . . . . . . . . . 6. Comparisons of densities calculated with the mixture model to experimental data for the R-125/143a binary mixture. . . . . . . . . . . . . . . . . . . 7. Comparisons of densities calculated with the mixture model to experimental data for the 593

598

599 617

603

604

605

1. 2. 3. 4.
a

596 596 596

606

607

Electronic mail: [email protected] 2004 by the U.S. Secretary of Commerce on behalf of the United States. All rights reserved. 0047-2689200433259328$39.00

608

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

594

ERIC W. LEMMON AND RICHARD T JACOBSEN R-134a/143a binary mixture. . . . . . . . . . . . . . . . . . Comparisons of densities calculated with the mixture model to experimental data for the R-134a/152a binary mixture. . . . . . . . . . . . . . . . . . Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-125/134a, R-125/143a, R-134a/143a, and R-134a/152a binary mixtures. . . . . . . . . . . . . . Comparisons of densities calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture. . . . . . . . . . . . . . . . Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/125/134a and R-125/134a/143a ternary mixtures. . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparisons of densities calculated with the mixture model to experimental data for the R-125/134a/143a ternary mixture. . . . . . . . . . . . . . Comparisons of isochoric heat capacities calculated with the mixture model to experimental data for the R-32/125, R-32/134a, R-125/134a, and R-125/143a binary mixtures. . . . . . . . . . . . . . . Comparisons of isochoric heat capacities calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture. . . . . . Comparisons of the speed of sound in the vapor phase calculated with the mixture model to experimental data for the R-32/125, R-32/134a, R-125/134a, and R-125/143a binary mixtures. . . . Comparisons of the speed of sound in the vapor phase calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparisons of the speed of sound in the liquid phase calculated with the mixture model to experimental data for the R-134a/152a binary mixture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Isobaric heat capacity versus temperature diagram for an equimolar mixture of R-32 and R-125. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparisons of densities calculated with the mixture model developed in this work and the mixture model of Tillner-Roth et al. 1998 to experimental data for the R-32/125 binary mixture at 330 K. . . . . . . . . . . . . . . . . . . . . . . . . . . Excess volumes for the R-32/125 binary mixture at 250 K and 5 MPa: solid line this work, dashed line mixture model of Tillner-Roth et al. 1998 . Experimental data between 243 and 256 K are shown for comparison. . . . . . . . . . . . . . Excess enthalpies for the R-32/125 binary mixture at 250 K and 5 MPa: solid line this work, dashed line mixture model of Tillner-Roth et al. 1998 . . . . . . . . . . . . . . . . . . . . . 609 List of Symbols

8.

610

Symbol a A cp cv d f F g h l m M n p R s t T u v V w x Z

Physical quantity Molar Helmholtz energy Helmholtz energy Isobaric heat capacity Isochoric heat capacity Density exponent Fugacity Generalized factor Gibbs energy Enthalpy Density exponent Number of components Molar mass Number of moles Pressure Molar gas constant Entropy Temperature exponent Temperature Internal energy Molar volume Volume Speed of sound Composition Compressibility factor (Z p/ RT) Reduced Helmholtz energy ( a/RT) Reduced density ( / Molar density Inverse reduced temperature ( T c /T) Chemical potential Reduced density parameter Reduced temperature parameter

Unit J/mol J J/ molK J/ molK MPa J/mol J/mol

9.

611

10.

612

11.

613

12.

614

13.

g/mol mol MPa J/ molK J/ molK K J/mol dm3/mol dm3 m/s mole fraction

615

14.

615

15.

616

16.

c)

mol/dm3

616

17.

J/mol dm3/mol K

616

18.

617

19.

618

Superscripts 0 E idmix r

20.

Ideal gas property Excess property Ideal mixture Residual Saturated liquid state Saturated vapor state

618

21.

618

Subscripts 0 c calc data i,j red

Reference state property Critical point property Calculated using an equation Experimental value Property of component i or j Reducing parameter

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

EQUATION OF STATE OF REFRIGERANT MIXTURES

595

1. Introduction
The need for equations of state capable of accurate representation of thermodynamic properties of environmentally safe uids continues as new applications are developed requiring the use of refrigerant mixtures. These mixtures of refrigerants are used as environmentally acceptable replacements for chlorouorocarbons and hydrochlorouorocarbons in refrigeration, heat pumps, foam-blowing, and other applications. Mixture equations are required to evaluate the performance of possible working uids. A model is presented here for calculating the thermodynamic properties of refrigerant mixtures that supercedes the model reported by Lemmon and Jacobsen 1999 . This model was initially reported by Lemmon 1996 , and general details and comparisons among different implementations of the model were reported by Lemmon and Tillner-Roth 1999 . The model may be used to calculate all thermodynamic properties of mixtures at various compositions, including dew and bubble point properties and critical points; the model and its calculational abilities have been incorporated into the NIST REFPROP database Lemmon et al., 2002 . The mixture model is similar to the model presented by Tillner-Roth et al. 1998 and published by the Japan Society of Refrigerating and Air Conditioning Engineers JSRAE . The work presented here uses generalized equations applicable to several mixtures, whereas separate equations for each binary mixture were developed in the JSRAE equations. The mixture model presented here is based on corresponding states theory and uses reducing parameters that are dependent on the mole fractions of the mixture constituents and critical points of the pure uids to modify absolute values of the mixture density and temperature. This approach allows the thermodynamic properties of the mixture to be based largely on the contributions from the pure uids. Without additional mixing functions, the model is similar to that for an ideal mixture, and only the excess values, or the departures from ideality, are required to accurately model the properties of the mixture. The model uses the Helmholtz energy as the basis for all calculations. The Helmholtz energy is one of the fundamental properties from which all other thermodynamic properties can be calculated using simple derivatives. The Helmholtz energy of the mixture is calculated as the sum of an ideal gas contribution, a real uid contribution, and a contribution from mixing. The Helmholtz energy from the contributions of the ideal gas and the real uid behavior is determined at the reduced density and temperature of the mixture by the use of accurate pure uid equations of state for the mixture components. Reducing parameters, dependent on the mole fractions of the constituents, are used to modify values of density and temperature for the mixture. The contribution from mixing, a modied excess function, is given by an empirical equation. An excess property of a mixture is dened as the actual mixture property at a given condition minus the value for an ideal solution at the same

condition. In most other work dealing with excess properties, the mixing condition is dened at constant pressure and temperature. Because the independent variables for the pure uid Helmholtz energy equations are reduced density and temperature, properties are calculated here at the reduced density and temperature of the mixture. The shape of the modied excess function is similar for many binary mixtures, and relatively simple scaling factors can be used to determine its magnitude for a particular application. While this approach is arbitrary and different from the usual excess property format, it results in an accurate representation of the single phase properties and phase boundaries. The mixtures studied in this work include the constituents R-32 diuoromethane , R-125 pentauoroethane , R-134a 1,1,1,2-tetrauoroethane , R-143a 1,1,1-triuoroethane , and R-152a 1,1-diuoroethane . Three separate models i.e., three separate excess functions were developed to calculate the properties of the refrigerant mixtures. The rst two describe the properties of the binary mixtures R-32/125 and R-32/134a. The shapes of the excess functions for these two mixtures differ from each other and from those of the other mixtures studied in this work, and could not be modeled by a generalized equation. This was also noticed in the work of Lemmon 1996 , which required additional terms in the mixing functions for these two binary mixtures. Laesecke 2004 noted that the electrical conductances of blends containing R-32 were much higher than other HFC blends. The higher conductance may be due to the polar R-32 molecules associating in the liquid phase via H-F bonds as suggested by Lisal and Vacek 1996 . The shapes of the excess functions for the mixtures R-125/ 134a, R-125/143a, R-134a/143a, and R-134a/152a were similar enough that one function could be developed that described the properties of all these systems. Additionally, experimental data for the ternary mixtures R-32/125/134a and R-125/134a/143a showed that only binary pair interactions are required to model these multicomponent mixtures.

2. The Mixture Equation

The equation for the mixture Helmholtz energy a used in this work is a a idmix a E . 1

The Helmholtz energy for an ideal mixture as used in this work dened in terms of density and temperature is
m

a idmix
i 1

xi a0 i

,T

ar i

RT ln x i ,

where and T are the mixture density and temperature, and are the reduced mixture density and temperature, m is the number of components in the mixture, a 0 is the ideal gas i Helmholtz energy of component i, a r is the residual Helmi holtz energy of component i, and the x i are the mole fractions of the mixture constituents. References for the pure uid ideal gas Helmholtz energy and residual Helmholtz energy equations are given in Table 1.
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

596

ERIC W. LEMMON AND RICHARD T JACOBSEN

TABLE 1. Pure uid equations of state for the refrigerants used in the mixture model Temperature range K 136.34 435 172.52500 169.85 455 161.34 650 154.56 500 Maximum pressure MPa 70 60 70 100 60

Fluid R-32 R-125 R-134a R-143a R-152a

Author Tillner-Roth and Yokozeki 1997 Lemmon and Jacobsen 2004 Tillner-Roth and Baehr 1994 Lemmon and Jacobsen 2000 Outcalt and McLinden 1996

The reduced values of density and temperature for the mixture models used here are / and T red /T, 4
red

Three excess functions were developed for the mixtures studied in this work. The excess function for the mixture Helmholtz energy for these three models is expressed as aE RT
E

, ,x
m i 1 dk tk lk

m 1 i 1 j

where and T are the mixture density and temperature, and red and T red are the reducing values
m red i 1

x ix jF i j

Nk

exp

xi
ci

m 1 i 1 j

m i 1

x ix j

ij

and
m m 1 m i 1

T red

i 1

x iT ci

i 1 j

x ix j

ij

The parameters i j and i j are used to dene the shapes of the reducing temperature and density curves. These reducing parameters are not the same as the critical parameters of the mixture and are determined simultaneously in the nonlinear t of experimental data with the other parameters of the mixture model. Additional values of i j as well as a generalized method for predicting these values for refrigerants not covered in this work was given in Lemmon and McLinden 2001 .
TABLE 2. Coefcients and exponents of the mixture equations k 1 2 3 4 5 6 7 8 1 2 3 4 1 2 3 4 Nk 0.007 2955 0.078 035 0.610 07 0.642 46 0.014 965 0.340 49 0.085 658 0.064 429 0.229 09 0.094 074 0.000 398 76 0.021 113 tk R-32/125 4.5 0.57 1.9 1.2 0.5 2.6 11.4 4.5 R-32/134a 1.9 0.25 0.07 2.0 dk 2 5 1 3 9 2 3 6 1 3 8 1 lk 1 1 2 2 2 3 3 3 1 1 1 2 1 1 2 3

Values of the coefcients and exponents of this equation are given in Table 2. The generalized factors and mixture parameters, F i j , i j and i j , are given in Table 3. The coefcients and exponents of Eq. 7 were obtained from nonlinear regression of experimental mixture data using tting techniques similar to those applied to the development of the R-125 and R-143a equations of state given in Table 1. Additional details of the nonlinear tting process are given in the respective papers for these two pure uids. By including the exponents of Eq. 7 as nonlinear tting parameters, the nal equation was given additional degrees of freedom that are not normally available in linear least squares ts. In particular, the noninteger exponent t k is exible enough to decrease the number of terms required to achieve the same accuracy as ts from linear least squares applications. The physical behavior of the equation was carefully monitored during the tting process using graphical techniques. Desirable characteristics of the nal equation include the ability to produce correct calculated properties within the ranges of temperature and pressure dened by experimental data and to extrapolate to reasonable limits outside those experimental ranges. In addition, the equation is designed to exclude data which contain systematic behavior caused by experimental error. For the most part, the mixture model was tted to experimental values of single phase p T and isochoric heat capacity data and a few selected values of the bubble point pressures. Since the calculation of the bubble point pressure

TABLE 3. Parameters of the mixture equations Binary mixture R-32/125 R-32/134a R-125/134a R-125/143a R-134a/143a R-134a/152a
ij

ij

dm3mol

Fij 1.0 1.0 1.0 1.1697 0.5557 2.0

R-125/134a, R-125/143a, R-134a/143a, R-134a/152a 0.013 073 7.4 1 0.018 259 0.35 3 0.000 008 1299 10.0 11 0.007 8496 5.3 2

28.95 7.909 0.4326 5.551 2.324 4.202

0.006 008 0.002 039 0.000 3453 0.000 4452 0.000 6182 0.004 223

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

EQUATION OF STATE OF REFRIGERANT MIXTURES requires an iterative solution, the number of tted values was kept at a minimum, although comparisons were made to all values upon completion of each t. The tted data were carefully weighted such that a balance was obtained among the uncertainty of each data type e.g., density versus heat capacity , the number of data points in a set, and the region where the data are located on the thermodynamic surface. The equations used for calculating pressure p , compressibility factor Z , internal energy u , enthalpy h , entropy s , Gibbs energy g , isochoric heat capacity (c v ), isobaric heat capacity (c p ), and the speed of sound w are given in Eqs. 8 15 . Z u RT h RT s R g RT cv R 1
0 m 0 i 1

597 a0 i ,T RT ,
E

xi

ln x i

19

and
m r i 1

xi

r i

, ,x ,

20

where r is the reduced residual Helmholtz energy of comi ponent i. If equations for the ideal gas Helmholtz energy in the nondimensional form a 0 ( , ) are used rather than equations in i the dimensional form a 0 ( ,T) as indicated by Eq. 19 , the i following reducing variables: / and T c i /T, 22 rather than the reducing values dened by Eqs. 3 and 4 , should typically be used in the ideal gas equation. This does not apply to the residual part of the Helmholtz energy; the residual and excess terms r ( , ) and E ( , ,x) in Eq. 20 i must be evaluated at the reduced state point of the mixture dened by Eqs. 3 and 4 . This complication is avoided through the use of the classical dimensional equations for functions involving the ideal gas heat capacity, such as a0 RT RT ln
T ci

p RT

1
0 r

21

,
r r

1,
0 r 0 r

10

11

r 0 r

,
2 r 2

12

T 0T 0

0 0 h 0i Ts 0i

T T0

c 0 dT pi

2 2

0 2 r

,
2 r 2

13

T
T0

c0 pi T

dT.

23

cp R

cv R

1
r 2 2 2 r 2 2 2 r r

14

1 2 cp 1 2 cv

The following dimensional form with density expressed in moles per cubic decimeter and temperature in Kelvins and the associated coefcients given in Table 4 can be used for this purpose: a0 i RT e1 e2 T ln 1 c 0 ln T bk T
n 0

w2M RT

15

ck

1 1 t Tk tk 1 tk 24

The rst derivative of pressure with respect to density at constant temperature ( p/ ) T , second derivative of pressure with respect to density at constant temperature ( 2 p/ 2 ) T , and rst derivative of pressure with respect to temperature at constant density ( p/ T) are given in p
T 2 r 2 2 2 2 r 3 2 3 r

a k ln 1 exp

Equations of the form ln N 0 ln Ni

i

...

25

i 1

RT 1 2
r

,
3 r

16

p
2 T

RT

, 17

p T

are derived from dimensional equations, and the critical parameters of the pure uids are built into the coefcients of the equations. Additional information on the mixing function and its derivatives, as well as formulas for other thermodynamic properties, can be found in Lemmon et al. 2000 , which presents an equation for mixtures of nitrogen, argon, and oxygen. 2.1. VaporLiquid Equilibrium VLE Properties In a two-phase nonreacting mixture, the thermodynamic constraints for vaporliquid equilibrium VLE are T T T, 26

18

The ideal gas and residual Helmholtz energy required to calculate all single phase thermodynamic properties given in Eqs. 8 18 above are

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

598

ERIC W. LEMMON AND RICHARD T JACOBSEN n r ni

TABLE 4. Coefcients and exponents of the ideal gas equations for the pure uids R-32 c 0 4.004 486 a 1 1.160 761 a 2 2.645 151 a 3 5.794 987 a 4 1.129 475 e 1 7.254 707 84 R-125 c 1 3.063 a 2 2.303 a 3 5.086 a 4 7.300 e 1 29.876 6745 R-134a c0 0.629 789 c 1 0.377 018 08 c 2 0.060 585 489 e1 12.280 8002 R-143a c 1 1.0578 a 2 4.4402 a 3 3.7515 e1 1.577 780 74 R-152a c 0 3.354 952 c 1 0.010 986 49 c 2 2.501 616 10 5 c3 2.787 445 10 e 1 4.360 056

f i x i RT exp

,
T,V,n j i

32

b1 b2 b3 b4 e2 t1 b2 b3 b4 e2

798 4185 1806 115 10 2231.55735 0.1 314.0 756.0 1707.0 3013.2267

where r was dened in Eq. 20 . The partial derivative at constant temperature, constant total volume not molar volume , and constant mole numbers of all constituents except i is generally evaluated numerically.

3. Comparisons to Data
The uncertainties of calculated values of various properties are determined by comparisons with measured values. Statistical analyses are used to determine the overall estimated accuracy of the model, and to dene the ranges of estimated uncertainties for various properties calculated with the formulation. Summary comparisons of values calculated using the mixture equation to data for p T, heat capacity, and sound speed, as well as second virial coefcients and VLE information for refrigerant mixtures are given in Table 5, along with the temperature range of the data and the composition range for the rst component listed. Bubble or dew point densities are included as p T data, with the bubble or dew point pressure calculated from the mixture model. No further distinction is made between single phase densities and saturated densities. Compositions for VLE data are bubble point compositions except for datasets where only the vapor phase compositions were reported. In a few cases, individual data points were eliminated from the comparisons when the deviation for a particular point was much higher than those for other points by the same author in the same region. For density, individual data points were typically deleted when the deviation exceeded 10%. This eliminates the likelihood of including in the comparisons data points that are in error or are reported incorrectly, including obvious typographical errors in published manuscripts. However, when the deviations slowly increased point by point, showing potentially systematic increasing differences in a particular region, these data points were left in the comparisons. The statistics used to evaluate the equation are based on the percent deviation for any property, X, % X 100 30 X data X calc . X data 33

t 1 0.5 t 2 0.75 e 2 3385.257 07 t1 b2 b3 e2 0.33 1791 823 2527.263 78

t1 t2 t3 e2

1 2 3 2654.673 62

p and
i i

p,

27

i 1,2,...,m,

28

where the superscripts and , respectively, refer to the liquid and vapor phases. Equation 28 is equivalent to equating the fugacities of the coexisting liquid and vapor phases for each component in the mixture fi fi . 29

The chemical potential of component i in a mixture is

i

,T

A ni

c i T,V,n j i

RT ln f i ,

Using this denition, the average absolute deviation AAD in Table 5 is dened as AAD 1 ni
n 1

where c (T) is a function of temperature only and the notai tion n j i indicates that all mole numbers are held constant except n i . The chemical potential in an ideal gas mixture is
0 i

% Xi ,

34

A0 ni

c i T,V,n j i

RT ln f 0 , i

31

is the ideal gas partial pressure of constituent i, where x i p 0 x i RT. Subtracting Eq. 30 from Eq. 31 and solving for f i results in
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

f0 i

where n is the number of data points. The comparisons given in the sections below for the various binary and ternary mixtures compare the equation of state to the experimental data by the use of the average absolute deviation as given by Eq. 34 . Discussions of maximum errors or of systematic offsets always use the absolute values of the deviations.

EQUATION OF STATE OF REFRIGERANT MIXTURES

TABLE 5. Summary comparisons of mixture properties calculated from the model to refrigerant mixture data Temperature range K Pressure range MPa Density range mol/dm3 Composition range mole fraction a

599

Author R-32125p T Benmansour and Richon 1997 Benmansour and Richon 1999a Higashi 1997 c Holcomb et al. 1998 Kishizawa et al. 1999 c Kiyoura et al. 1996 Kleemiss 1997 Magee 2002 Magee and Haynes 2000 Oguchi et al. 1995 Perkins 2002 Piao et al. 1996 Sato et al. 1996 Weber 2000 Weber and Debaugh 1994 Widiatmo et al. 1993 Zhang et al. 1996 R-32125VLE Benmansour and Richon 1997 Benmansour and Richon 1999a Debaugh and Morrison 1995 Fujiwara et al. 1992 Higashi 1997 Holcomb et al. 1998 Kato et al. 2002 Kleemiss 1997 Nagel and Bier 1995 Oguchi et al. 1995 Piao et al. 1996 Takagi et al. 1999 Weber 2000 Widiatmo et al. 1993 R-32125Second Virial Coefcient Kiyoura et al. 1996 Sato et al. 1996 Weber and Debaugh 1994 R-32125Isochoric Heat Capacity Magee 2000a Perkins 2002 R-32125Isobaric Heat Capacity Gunther and Steimle 1996 R-32125Sound Speed Hozumi et al. 1995 R-32134ap T Benmansour and Richon 1999b Higashi 1995 c Holcomb et al. 1998 Kleemiss 1997 Magee and Haynes 2000 Oguchi et al. 1999 Oguchi et al. 1995 Piao et al. 1996 Sato et al. 1994 Weber and Debaugh 1994 Widiatmo et al. 1994b Widiatmo et al. 1997 R-32134aVLE Benmansour and Richon 1999b Chung and Kim 1997 Debaugh and Morrison 1995 Fujiwara et al. 1992

No. points

AADb %

12 909 26 777 23 45 34 94 415 235 228 6 411 543 156 90 17 24 124 18 33 10 8 22 30 39 23 34 11 10 47 90 24 23 39 3 111 363 89 178 19 714 27 44 390 461 61 53 643 220 17 30 22 40 34 25 6

253333 253333 323346 279341 339344 330 440 243 413 200 400 200 400 355 430 300398 263393 320 440 295334 338 373 280310 300380 253333 253333 249338 273 283313 280340 318 349 224 333 205345 250350 263283 248 333 295334 280310 330 440 320 440 338 373 208 345 300397 203313 303343 253333 341365 279340 243 413 200 400 310 473 238 473 261393 320 440 338 373 280340 280330 253333 263323 253358 273

0.14 19.1 0.108 20 3.015.07 0.896 4.61 3.90 4.90 1.835.24 0.01917.1 4.0735.3 2.5735.3 6.3116.9 3.8519.1 0.54 15 1.78 5.27 1.453.98 0.304 4.23 0.884 2.31 0.094 4.6 0.3853.83 0.358 3.9 0.348 4.3 0.6910.818 0.906 2.48 0.83 4.58 2.355.27 0.108 3.68 0.038 5.05 0.3615.65 0.54 1.07 0.284 3.93 1.453.98 0.884 2.31

0.066 17.9 0.04918.3 1.98 10.7 0.71919.3 4.47 8.62 0.8291.72 0.00716.2 8.58 19.6 1.06 17.4 8.34 8.37 6.54 14.8 0.286 17.4 0.836 1.72 0.77716.9 0.106 1.91 10.218.2 0.032.02

0.696 0.0930.885 0.4970.776 0.2370.956 0.204 0.700 0.3670.606 0.5030.508 0.698 0.5 0.874 0.698 0.366 0.902 0.698 0.902 0.416 0.885 0.546 0.204 0.902 0.50.698 0.696 0.0930.885 0.763 0.0550.895 0.2250.896 0.3390.948 0.1110.953 0.4830.517 0.2410.951 0.874 0.366 0.902 0.2690.941 0.416 0.885 0.204 0.902 0.3670.606 0.698 0.902 0.546

0.227 0.627 5.01 0.908 5.67 0.134 0.052 0.047 0.040 0.129 0.116 0.277 0.140 0.627 0.186 0.091 0.074 1.27 0.511 0.339 2.03 0.473 0.252 1.85 0.030 0.415 0.288 0.701 0.999 0.270 0.338 2.21 1.95 3.15 0.448 1.74 0.855 0.044 0.354 5.90 1.13 0.049 0.075 0.356 0.237 0.376 0.208 0.905 0.173 0.110 1.62 0.572 0.515 3.14

11.4 17.1 4.1318.3

0.5 0.698 0.434 0.874

0.0390.554 0.09518.8 3.325.47 0.523 4.29 0.01917.1 2.735.5 0.286 16.7 0.13515.3 0.24115 1.97 6.18 0.33 4.3 0.5773.1 13.01 0.1813.6 0.23.14 0.263 4.47 0.384 0.758 0.04219.7 2.54 12.88 0.80517.2 0.008 17.3 1.118.5 0.11512.9 2.0118.8 0.12116.3 12.12 0.1212.18 1118.6 1214.7

0.2010.777 0.1310.889 0.4570.821 0.130.973 0.4970.555 0.3290.5 0.3920.887 0.274 0.71 0.3290.887 0.3290.887 0.508 0.3290.887 0.395 0.1310.889 0.208 0.76 0.496 0.55 0.204 0.922

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

600

ERIC W. LEMMON AND RICHARD T JACOBSEN

TABLE 5. Summary comparisons of mixture properties calculated from the model to refrigerant mixture dataContinued Temperature range K 283313 280340 258 283 223343 203369 243361 238 301 261283 263293 243333 280340 320 440 333398 338 373 205343 203323 303343 253303 334 365 280342 243 413 200 400 303373 280350 298 423 253303 283313 283313 280340 263303 224 343 206 365 280350 323373 206 345 203323 303343 325342 280328 325344 340344 243373 200 400 308 341 333373 280330 305390 273313 280326 223338 205343 0.0410.529 2.44 3.71 0.798 2.56 2.46 3.71 3.453.71 1.6 17.1 2.1335.4 1.653.48 0.218 3.27 0.10.199 0.115 4.76 0.6222.01 0.7672.64 0.086 3.3 0.0323.69 1.99 8.98 0.61211.5 2.29 8.98 2.94 6.89 6.4113.2 0.88114.1 0.9313.03 0.08 1.45 8.16 11.8 0.0372.53 1014 0.0310.241 0.034 20.3 2.703.99 0.5372.55 0.01917.1 2.84 35.5 0.17 4.03 13.02 0.101 6.62 0.1471.5 0.5171.73 0.4122 0.3793.63 0.2011.57 0.066 2.9 0.0173.97 0.4252.97 0.016 13.4 1.85 8.78 0.52911.9 0.008 13.2 1.66 14.1 0.0692.22 8 12.4 0.029 4.49 13.218.4 Pressure range MPa 0.5671.91 0.379 4.56 0.2010.96 0.0733.15 0.0155.42 0.1735.02 0.1351.29 0.2410.93 0.2521.4 0.084 3.34 0.5773.1 Density range mol/dm3 Composition range mole fraction a 0.1210.673 0.1620.783 0.2020.799 0.4190.517 0.2120.772 0.3920.887 0.274 0.71 0.3290.887 0.1350.923 0.184 0.808 0.3290.887 0.3290.887 0.4820.5 0.508 0.5 0.3970.882 0.1550.896 0.1310.942 0.2670.665 0.350.72 0.50.51 0.5 0.495 0.0870.923 0.2510.751 0.1310.942 0.1790.776 0.1790.776 0.2590.649 00.814 0.4610.514 0.254 0.749 0.0870.923 0.495 0.5 0.2220.719 0.3490.694 0.4120.620 0.350.672 0.412 0.412 .504 0.5 0.412 0.509 0.0730.863 0.2730.737 0.1510.759 0.2870.65 0.4610.499 0.4930.503

Author Higashi 1995 Holcomb et al. 1998 Kim and Park 1999 Kleemiss 1997 Nagel and Bier 1995 Oguchi et al. 1999 Oguchi et al. 1995 Piao et al. 1996 Shimawaki et al. 2002 Takagi et al. 1999 Widiatmo et al. 1994b R-32134aSecond Virial Coefcient Sato et al. 1994 Tack and Bier 1997 Weber and Debaugh 1994 R-32134aIsochoric Heat Capacity Magee 2000a R32134aIsobaric Heat Capacity Gunther and Steimle 1996 R-32134aSound Speed Hozumi et al. 1995 R-125134ap T Benmansour and Richon 1999d Higashi 1999b c Holcomb et al. 1998 Kleemiss 1997 Magee and Haynes 2000 Weber and Debaugh 1994 Widiatmo et al. 1997 Yokoyama et al. 2000 R-125134aVLE Benmansour and Richon 1999d Higashi 1999b Higuchi and Higashi 1995 Holcomb et al. 1998 Kim and Park 1999 Kleemiss 1997 Nagel and Bier 1995 Widiatmo et al. 1997 R-125134aSecond Virial Coefcient Weber and Debaugh 1994 R-125134aIsochoric Heat Capacity Magee 2000a R-125134aIsobaric Heat Capacity Gunther and Steimle 1996 R-125134aSound Speed Hozumi 1996 R-125143ap T Higashi 1999c c Holcomb et al. 1998 Ikeda and Higashi 1995 c Kishizawa et al. 1999 c Kleemiss 1997 Magee and Haynes 2000 Uchida et al. 1999 c Weber and Debaugh 1994 Widiatmo et al. 1994a c Zhang et al. 1998 R-125143aVLE Higashi 1999c Holcomb et al. 1998 Kleemiss 1997 Nagel and Bier 1996

No. points 12 48 25 16 50 36 34 10 40 35 30 57 10 3 131 96 193 11 153 30 17 407 268 18 149 341 23 15 25 40 35 24 31 36 4 94 73 81 30 14 14 19 151 281 7 27 37 205 18 36 16 19

AADb % 2.11 0.428 0.500 0.345 0.447 2.45 1.05 0.412 0.493 2.06 1.70 2.91 3.48 7.74 0.311 1.43 0.016 0.252 5.64 0.237 0.046 0.103 0.266 0.086 0.817 1.41 1.10 0.848 0.554 0.552 0.302 0.278 1.28 2.53 0.375 0.877 0.020 2.25 0.997 2.13 3.95 0.037 0.075 0.951 0.329 0.245 0.136 1.60 0.833 0.306 0.136

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

EQUATION OF STATE OF REFRIGERANT MIXTURES

TABLE 5. Summary comparisons of mixture properties calculated from the model to refrigerant mixture dataContinued Temperature range K 308 341 280330 333398 330 400 333373 205344 203318 303343 280343 243 413 273313 280340 263313 223354 278 333 273313 205361 333 423 243 433 353373 248 368 255298 273 313378 233 473 353373 203323 298 423 230350 230336 253333 253303 341359 244 346 313 453 341359 315 440 243 413 200 400 365 430 263393 280340 253333 253303 273323 221345 222353 205362 270326 280340 0.04 0.549 0.5222.82 0.09217.1 0.388 1.6 0.3793.32 0.21.83 0.0593.39 0.352.88 0.294 1.83 0.0213.94 0.28156 0.08916.9 0.268 3.17 0.104 3.43 0.1310.662 0.2710.286 0.926 4.09 0.08312.1 0.028 15.3 0.094 1.69 0.66212.3 0.03213.9 9.8914 Pressure range MPa 1.653.47 0.7732.83 Density range mol/dm3 Composition range mole fraction a 0.412 0.0730.863 0.4930.515 0.412 0.509 0.5 0.1930.671 0.20.803 0.2820.65 0.4920.501 0.294 0.751 0.350.835 0.0790.92 0.5020.522 0.1450.891 0.0810.905 0.504 0.526 0.4850.538 0.248 0.751 0.497 0.777 0.3150.978 0.118 0.758 0.230.75

601

Author Uchida et al. 1999 Widiatmo et al. 1994a R-125143aSecond Virial Coefcient Tack and Bier 1997 Uchida et al. 1999 Weber and Debaugh 1994 R-125143aIsochoric Heat Capacity Magee 2000a R-125143aIsobaric Heat Capacity Gunther and Steimle 1996 R-125143aSound Speed Ichikawa et al. 1998 R-134a143ap T Holcomb et al. 1998 Kleemiss 1997 R-134a143aVLE Higuchi 1997 Holcomb et al. 1998 Kim et al. 2000 Kleemiss 1997 Kubota and Matsumoto 1993 Lim et al. 2002 Nagel and Bier 1996 R-134a152ap T Dressner and Bier 1993 Tillner-Roth 1993 Weber and Debaugh 1994 R-134a152aVLE Debaugh and Morrison 1995 Kleiber 1994 Sand et al. 1994 Tillner-Roth 1993 R-134a152aSecond Virial Coefcient Schramm et al. 1992 Weber and Debaugh 1994 R-134a152aIsobaric Heat Capacity Gunther and Steimle 1996 Tuerk et al. 1996 R-134a152aSound Speed Beliajeva et al. 1999 Grebenkov et al. 1994 R-32125134ap T Benmansour and Richon 1998 Benmansour and Richon 1999c Higashi 1999a c Holcomb et al. 1998 Hurly et al. 1997 Ikeda and Higashi 1995 Kiyoura et al. 1996 Kleemiss 1997 Magee 2000b Oguchi et al. 1995 Piao et al. 1996 Widiatmo et al. 1997 R-32125134aVLE Benmansour and Richon 1998 Benmansour and Richon 1999c Higuchi 1997 Holcomb et al. 1998 Kleemiss 1997 Nagel and Bier 1995 Piao et al. 1996 Widiatmo et al. 1997

No. points 7 34 6 8 5 109 73 142 17 377 9 40 54 18 41 35 12 139 1721 11 13 25 4 23 7 2 32 49 329 120 11 623 4067 16 42 88 16 105 369 352 12 1025 76 18 9 72 58 44 29 31 20

AADb % 0.301 0.805 7.42 4.66 5.29 0.550 0.832 0.011 1.34 0.053 3.80 0.540 0.418 0.144 0.951 0.756 0.680 0.196 0.053 0.160 0.719 0.826 2.44 0.258 11.2 5.40 2.64 0.370 0.265 0.267 0.162 0.276 7.47 1.57 0.213 6.81 0.587 0.083 0.119 0.202 0.282 0.196 2.14 2.21 0.927 0.919 0.356 0.253 0.650 0.951

0.497 0.138 0.72 0.5 0.128 0.688 0.688 0.058 15.5 0.01215 2.64 10.4 0.71115.3 0.086 2.76 2.64 10.4 0.7672.07 0.00915.4 1.58 17.1 5.86 6.22 0.20915.2 10.4 14.9 0.377 0.1050.469 0.381 0.20.676 0.346 0.381 0.3810.515 0.334 0.348 0.334 0.381 0.38 0.471 0.186 0.473 0.3470.464 0.377 0.1050.335 0.1730.54 0.0450.599 0.144 0.661 0.1870.434 0.3170.381 0.3470.464

0.12.5 0.456 16.5 0.5719 0.1215.2 0.028 17.1 3.27 4.64 0.2293.93 0.3217.79 3.27 4.10 1.575.75 0.026 17.1 2.9735.2 5.1912.4 0.44715 0.724 3.24 0.2122.72 0.251.39 0.556 2.73 0.0733.93 0.074 4.2 0.026 4.77 0.448 2.41 0.724 3.24

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

602

ERIC W. LEMMON AND RICHARD T JACOBSEN

TABLE 5. Summary comparisons of mixture properties calculated from the model to refrigerant mixture dataContinued Temperature range K 340 440 340 440 203345 203318 303343 260 400 253333 263 403 243373 253333 273313 224 345 205364 203318 0.0450.537 0.0510.982 0.004 18.7 1.515 1.4 17.1 0.32.88 0.5971.88 0.073.15 0.0173.96 0.00213.1 0.48212.8 6.1513.4 11.517.1 Pressure range MPa Density range mol/dm3 Composition range mole fraction a 0.23 0.381 0.334 0.381 0.346 0.381 0.34 0.346 0.358 0.358 0.341 0.358 0.338 0.356 0.316 0.331 0.1590.172 0.358

Author R-32125134aSecond Virial Coefcient Hozumi et al. 1995 Kiyoura et al. 1996 R-32125134aIsochoric Heat Capacity Magee 2000b R-32125134aIsobaric Heat Capacity Gunther and Steimle 1996 R-32125134aSound Speed Hozumi 1996 Hurly et al. 1997 R-125134a143ap T Bouchot and Richon 1998 Fujiwara et al. 1998 Kleemiss 1997 R-125134a143aVLE Bouchot and Richon 1998 Higuchi 1997 Kleemiss 1997 Nagel and Bier 1996 R-125134a143aIsobaric Heat Capacity Gunther and Steimle 1996
a b

No. points 11 11 147 48 27 361 1644 162 196 16 22 26 13 24

AADb % 8.48 2.76 0.280 0.766 0.016 0.009 0.292 0.301 0.034 0.486 1.23 0.247 0.578 1.69

Composition range is listed for the rst component. Average absolute deviation in density for p T data and in bubble point pressure for VLE data. For second virial coefcients, numbers given are average absolute differences cm3/mol . c Saturated liquid or vapor densities.

The comparisons of the mixture model to experimental data exhibit many general trends, as shown in the gures presented in this section. In these gures, data of a given type are separated into temperature increments of 10 K, where the temperatures listed at the top of each small plot are the lower bounds of these ranges. Comparisons to p T data focus for the most part on deviations in density, given inputs of pressure and temperature. However, in the critical region, deviations in density are generally higher than in the liquid or vapor phase at states away from the critical point, and several of the systems described below include comparisons based on deviations in pressure, given inputs of density and temperature. Such comparisons are typical for equations of state for both pure uids and mixtures, and are not specic to the model presented here. The articially large deviations are an artifact in the calculation of deviations caused by the fact that dp/d is nearly zero in the critical region. For the VLE data, the comparisons given in the following sections focus on the relative deviation in bubble point pressure. There are some VLE systems for which only the vapor phase compositions were reported, and the relative deviation in bubble point pressure is replaced with the relative deviation in dew point pressure in such cases.

3.1. The R-32125 System The R-32/125 system is perhaps the most widely studied system of all mixtures that have ever been measured, even compared to the well measured systems methane/ethane, nitrogen/argon, and dry air. The data span the entire compoJ. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

sition range and were measured at temperatures and pressures that cover nearly the entire range of practical uid states. Further experimental data for the region at temperatures above 380 K would be of use for verifying the accuracy of the mixture model in this region. Comparisons of experimental density data for the R-32/ 125 binary mixture to the mixture model are shown in Fig. 1. For the datasets of Benmansour and Richon 1997, 1999a , only one out of every 50 points is shown due to the very large number of data points published by these authors. All of the temperature, pressure, and composition ranges covered by Benmansour and Richon are shown in the gures, but the smaller set used for plotting allows the symbol shapes to be seen in the plots. In the liquid phase at temperatures below 360 K, the datasets of Kleemiss 1997 , Magee and Haynes 2000 , and Magee 2002 are represented on average to within 0.03%. The equation represents the data of Widiatmo et al. 1993 , Piao et al. 1996 , Perkins 2002 , and Weber and Debaugh 1994 with average deviations of 0.1%. Comparisons with the data of Benmansour and Richon 1997, 1999a show slightly higher deviations about 0.17% . The data of Benmansour and Richon 1999a in the liquid phase agree favorably with the equation, except for the data at 330 K, which have a systematic offset of about 0.3% and do not agree with other data at this temperature. The AAD for this dataset in the liquid is 0.06% if the data at 330 K are omitted. The data of Piao et al. show systematic offsets near 263 and 273 K disagreeing with other data in the same region and composition , but the average differences fall to 0.08% at higher temperatures.

EQUATION OF STATE OF REFRIGERANT MIXTURES

603

FIG. 1. Comparisons of densities calculated with the mixture model to experimental data for the R-32/125 binary mixture.

The scatter between various experimental datasets is much higher in the vapor region than in the liquid. Calculated values from the equation exhibit deviations between 0.02% and 0.18% on average from the data of Kleemiss 1997 0.02% ,

Kiyoura et al. 1996 , Sato et al. 1996 , Weber and Debaugh 1994 , and Zhang et al. 1996 . Differences are greater for other datasets. Above 360 K, deviations in the area near the critical point
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

604

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 2. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/125 binary mixture.

and at higher temperatures tend to increase, with the maximum errors in the datasets of Kiyoura et al. 1996 and Sato et al. 1996 reaching 0.3% in density. Comparisons with the data of Kleemiss 1997 show smaller differences, but even for this dataset, the model shows offsets of 0.15% at the highest temperatures. As stated previously, in the close vicinity of the critical point, it is not useful to compare deviations in density, because a slight error in pressure in this region can be accompanied by large errors in the density, with differences easily exceeding 5%. Deviations in pressure are more meaningful as a measure of the physical behavior of the model. Above 340 K, the average absolute deviation in pressure is approximately 0.1% for all datasets. Even as the critical points of the mixtures at different compositions are approached 339351 K, 3.6 5.8 MPa , the maximum deviations do not exceed approximately 0.3% in pressure. For the commercial mixture R-410A the 50/50 by mass mixture of R-32 and R-125 , there are four datasets within the region 4 10 mol/dm3: Kishizawa et al. 1999 , Magee 2002 , Perkins 2002 , and Piao et al. 1996 , with the data of Kishizawa et al. and of Perkins measured near the critical point. The equation shows close agreement with the data of Perkins, with an average deviation of 0.16% in density including the very near critical region and 0.07% in pressure. Comparisons to bubble point pressures are shown in Fig. 2. Eliminating the data points that fall substantially outside the main body of VLE data in terms of their deviations from the mixture model, bubble point pressures are represented on average to within 0.4%. The data of Kleemiss 1997 , which are represented with an AAD of 0.05%, were the primary data used in the development of the model given here. However, nearly all of the other data points from various authors are represented within a band of 1%. Other datasets that agree well with the data of Kleemiss include those of Debaugh and Morrison 1995 , 0.26%; Holcomb et al. 1998 , 0.27%; Weber 2000 , 0.32%; Oguchi et al. 1995 ,
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

0.36%; and Widiatmo et al. 1993 , 0.4%. No systematic offsets are seen in the comparisons. In the few cases where both bubble and dew point compositions are given, differences between the calculated and experimental dew point compositions are generally within 0.005 mole fraction, where the dew point compositions and mixture pressure were calculated given the mixture temperature and bubble point compositions.

3.2. The R-32134a System Comparisons of calculated mixture densities to experimental density data for the R-32/134a binary mixture are shown in Fig. 3. For the dataset of Benmansour and Richon 1999b , only one out of every 20 points is shown similar to that for the R-32/125 mixture due to the very large number of data points published by these authors. The data of Kleemiss 1997 and of Magee and Haynes 2000 are represented on average to within 0.06%. Between 210 and 360 K, the average representation is 0.024%. The data of Magee and Haynes between 200 and 210 K for the 0.33 mole fraction of R-32 show an offset of 0.3%; similar offsets were seen in other models, including that of Tillner-Roth et al. 1998 published by the JSRAE, and in the earlier model of Lemmon and Jacobsen 1999 . The liquid phase data of Benmansour and Richon are represented with an average difference of 0.09% excluding the data at 330 K, similar to that done for the R-32/125 mixture . The vapor phase data of Kleemiss at 370 and 390 K cannot both be represented simultaneously within the stated experimental accuracy of the data. In this work, the equation is biased towards the data at 390 K, causing the higher deviations of calculated values at 370 K. Excluding the data at 273 and 283 K which appear to be in

EQUATION OF STATE OF REFRIGERANT MIXTURES

605

FIG. 3. Comparisons of densities calculated with the mixture model to experimental data for the R-32/134a binary mixture.

error with deviations greater than 1% at a composition of 0.45 mole fraction of R-32, values from the equation deviate from the data of Piao et al. below 360 K on average by 0.21%. The data of Piao et al. above 360 K show increasing scatter due to the complexity of modeling and measuring the

critical region. Below 330 K in the vapor phase, the data of Oguchi et al. 1995 and Widiatmo et al. 1994b, 1997 show average deviations of 0.1%. Above 330 K, in the area around the critical region, the scatter in the data and the deviations from the equation increase substantially. DeviaJ. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

606

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 4. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/134a binary mixture.

tions between the equation and the data of Oguchi et al. 1995, 1999 , Sato et al. 1994 , and Weber and Debaugh 1994 are about 0.3%, including some systematic differences. Comparisons to VLE data see Fig. 4 for the R-32/134a system show nearly the same trends as those for the R-32/ 125 system. In a similar fashion eliminating the extraneous data points outside the main group of data , VLE data are generally represented with an AAD of 0.6%. All of the datasets appear to be of similar quality. Average differences are 0.38%, 0.41%, 0.50%, and 0.57% for the datasets of Takagi et al. 1999 , Piao et al. 1996 , Kim and Park 1999 , and Chung and Kim 1997 , respectively. For those datasets that reported both liquid and vapor composition, differences for each data point in the dew point composition are generally about 0.006 mole fraction. 3.3. The R-125134a, R-125143a, R-134a143a, and R-134a152a Systems Calculated densities are compared to the experimental data for the R-125/134a binary mixture in Fig. 5. As was done with the R-32/134a mixture, only one out of every 20 points are shown for the dataset of Benmansour and Richon 1999d . The data of Kleemiss 1997 and of Magee and Haynes 2000 are represented on average to within 0.07%. Below 360 K, the average deviation is 0.05%. In the liquid phase at 240 K, there is a systematic offset of 0.06% compared to the data of Kleemiss. This offset decreases quickly with increasing temperature. The model deviates from the data of Benmansour and Richon in the liquid by 0.11%. In the vapor phase, the average absolute deviation of the equation from the data of Widiatmo et al. 1997 is 0.09%. At the highest temperatures above the critical point, differences from the data of Kleemiss increase to a maximum of 0.26%
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

at pressures around the critical pressure. Similar trends are found in the JSRAE model by Tillner-Roth et al. 1998 at the high temperatures, but with a maximum deviation of 0.20%. Comparisons of calculated values to the experimental density data for the R-125/143a binary mixture are shown in Fig. 6. Differences between the equation and the data of Kleemiss 1997 and of Magee and Haynes 2000 are around 0.06%. Below 360 K, differences fall on average to 0.03% for these two datasets. In the vapor phase, comparisons with the data of Widiatmo et al. 1994a , Weber and Debaugh 1994 , and Zhang et al. 1998 show differences of 0.17%. Comparisons for the R-134a/143a system are shown in Fig. 7. Below 360 K, comparisons with the equation show differences on average of 0.03% in both the liquid and vapor phases from the data of Kleemiss 1997 . Above 360 K, the differences increase at pressures near the critical pressure of the mixture, but decrease to an average deviation of 0.1% in density at lower and higher pressures, including the data in the critical region. Similar comments can be made about the R-134a/152a system see Fig. 8 . Differences below 360 K, as well as at conditions above 360 K away from the critical pressure of the mixture, are about 0.06%. As the critical region is approached, differences increase up to 0.5%. Although there are few publications of measurements for this system, it was covered in detail by Tillner-Roth 1993 for a wide range of temperature and pressure, and for several compositions 0.25, 0.50, and 0.75 mole fraction . These data are well represented by the model reported here. The comparisons to VLE data for the R-125/134a, R-125/ 143a, and R-134a/143a binary mixtures see Fig. 9 are similar to those described above for the R-32/125 and R-32/134a systems. The average absolute deviation for each system is approximately 0.5% in bubble point pressure. Comparisons with the dew point compositions are similar to those for the

EQUATION OF STATE OF REFRIGERANT MIXTURES

607

FIG. 5. Comparisons of densities calculated with the mixture model to experimental data for the R-125/134a binary mixture.

other systems previously described. The R-134a/152a system shows similar trends above 270 K, but at lower temperatures there is a systematic offset of calculated bubble point pressures compared to the data of Debaugh and Morrison 1995 and of Kleiber 1994 , with a maximum difference of 2.4% in pressure for both of these datasets.

3.4. The Ternary Mixtures The R-32/125/134a system is unique from a modeling standpoint since it combines the three mixture equations the individual equations for R-32/125 and R-32/134a, and the generalized equation for R-125/134a . No additional paramJ. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

608

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 6. Comparisons of densities calculated with the mixture model to experimental data for the R-125/143a binary mixture.

eters were required to model the ternary mixture, although slight systematic offsets are seen in several locations. Comparisons of the combined mixture model for this ternary mixture are shown in Fig. 10. Similar to the R-32/125 mixture, only one out of every 20 points is shown for the datasets of Benmansour and Richon 1998, 1999c . In the liquid region
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

below 360 K, the equations represent the data of Magee 2000b and Kleemiss 1997 with an average deviation of 0.05%. At temperatures near 260 K, systematic offsets of 0.04% and 0.08% are seen for the datasets of Kleemiss and Magee, respectively. The liquid data of Benmansour and Richon 1998 are represented by average deviations of

EQUATION OF STATE OF REFRIGERANT MIXTURES

609

FIG. 7. Comparisons of densities calculated with the mixture model to experimental data for the R-134a/143a binary mixture.

0.05% when the data at 330 K are eliminated as was done with the R-32/125 mixture . In the vapor region below 360 K , differences are about 0.06% for the data of Kleemiss, but increase to 0.5% for the data of Benmansour and Richon 1998 and of Piao et al. 1996 . Above 360 K, differences continue to increase, with maximum deviations of 0.5% for the data of Kleemiss and higher for other datasets. The scatter among data sets of various authors is greater than 0.5% in density near the critical region as expected. Figure 11 illustrates comparisons of VLE data for the

R-32/125/134a ternary mixture. Bubble point pressures are represented on average to within 0.7% and dew point composition differences are within 0.005 mole fraction of R-32. Comparisons to the data of Nagel and Bier 1995 show deviations of 0.26% and those with Piao et al. 1996 show deviations of 0.66%. Although the ternary mixture R-125/134a/143a has no additional tted parameters, the properties of this system are represented with accuracies similar to those of the binary mixtures. The experimental data of Kleemiss 1997 are repJ. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

610

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 8. Comparisons of densities calculated with the mixture model to experimental data for the R-134a/152a binary mixture.

resented on average by differences of 0.03%. Deviations for the data of Bouchot and Richon 1998 in the liquid phase are 0.09%. Small systematic differences are evident in the comparisons given in Fig. 12, such as the offset of 0.05% at 300 K. Trends above 360 K in the critical region are similar to those described for the binary mixtures above. Figure 11 also includes comparisons of VLE data for the R-125/134a/ 143a ternary mixture. There are very few phase equilibrium data for this mixture, but the data of Nagel and Bier 1996
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

and those of Kleemiss 1997 are in agreement within about 1% in bubble point pressure, with an AAD of 0.35%. 3.5. Other Thermodynamic Properties The isochoric heat capacity has been measured by Magee 2002 and Perkins 2002 for four of the binary mixtures: R-32/125, R-32/134a, R-125/134a, and R-125/143a. Figure 13 compares values calculated from the model to these data.

EQUATION OF STATE OF REFRIGERANT MIXTURES

611

FIG. 9. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-125/134a, R-125/143a, R-134a/143a, and R-134a/152a binary mixtures.

In addition, comparisons to the experimental data for the R-32/125/134a ternary mixture are shown in Fig. 14. In general, the mixture model represents the data with an average absolute deviation between 0.3% and 0.5% for the binary

mixtures, and 0.3% for the ternary mixture. There is very little systematic behavior in the deviations for the systems studied, and the model represents the data to within their experimental uncertainty.
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

612

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 10. Comparisons of densities calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture.

Comparisons to the saturated liquid isobaric heat capacity data of Gunther and Steimle 1996 for the seven mixtures that they studied show comparable deviations, with differences generally less than 1% for most of the mixtures, except at the lowest temperatures 200 K and near the critical reJ. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

gion where c p tends to increase rapidly with increasing temperature . The R-134a/152a system is the only exception, with deviations of less than 1% at the highest temperatures, but with steadily increasing deviations at lower temperatures, with a maximum of 5% at 200 K. This is the only system

EQUATION OF STATE OF REFRIGERANT MIXTURES

613

FIG. 11. Comparisons of bubble point pressures calculated with the mixture model to experimental data for the R-32/125/134a and R-125/134a/143a ternary mixtures.

with vapor measurements, and the model represents these data Tuerk et al., 1996 with an average absolute deviation of 0.37%. Speed of sound measurements in the vapor phase for the R-32/125, R-32/134a, R-125/134a, R-125/143a, and R-32/ 125/134a mixtures were given by Hozumi 1996 , Hurly et al. 1997 , and Ichikawa et al. 1998 . Comparisons of the model to these data are shown in Fig. 15 for the binary mixtures and Fig. 16 for the ternary mixture. The average absolute deviations for these systems range between 0.01% and 0.04% in the speed of sound. In the liquid phase of the R-134a/152a system, the mixture model represents the speed of sound measurements of Beliajeva et al. 1999 and Grebenkov et al. 1994 within an average absolute deviation of about 0.3%, as shown in Fig. 17.

4. Accuracy Assessment
Based on comparisons to experimental data, the uncertainties of the equation are generally 0.1% in density, 0.5% in heat capacity and speed of sound, and 0.5% for calculated bubble point pressures. The model is valid from 200 to 450 K up to 60 MPa as veried by experimental data. Although the equation was developed using mostly binary data, it is accurate in calculating the properties of the two ternary mixtures for which data were available for comparison. It is expected that this result will apply to other ternary and higher-order systems as well. Table 6 gives calculated values from the model for computer code verication. Graphical analyses of various properties were made to verify the behavior of the equations over their ranges of validity, especially for heat capacities in the liquid region, and to test the extrapolation behavior of the equations. Figure 18 shows a typical example of the isobaric heat capacity as applied to the equimolar mixture of R-32 and R-125 or

R-410A . Plots for the isochoric heat capacity and speed of sound showed similar physically correct behavior for different systems at various compositions. Future measurements are needed to conrm whether the equation is valid for predicting properties of other mixtures and for calculating properties of states in regions not covered by the experimental data used in the development of this model. Such data will enable continued evaluation and renement of the model and modeling process. While early measurements of mixture properties were considered to be less accurate than those for pure substances, modern mixture data are now approaching the accuracy of measurements for the pure uids. Renements in the equations of state for both pure substances and mixtures will improve the prediction of properties for uid mixtures as they become more common as working uids in engineered systems. Comparisons were made to determine the sensitivity of the mixture model to the accuracies of the pure uid equations of state used in its formulation. Two highly accurate equations are available for R-143a: the equation of Lemmon and Jacobsen 2000 used in this work and the equation of Li et al. 1999 . The deviations between the equations and the experimental data are similar for both equations, and the behavior of derived properties such as the heat capacities show similar trends. Replacing the equation of Lemmon and Jacobsen with the equation of Li et al. showed virtually no change in the deviations for the R134a/143a binary mixture in terms of density or bubble point pressure. Likewise, differences between the two equations for calculated values of heat capacity and speed of sound are less than 0.2% for the binary mixture. There are also two highly accurate equations available for R-125, the equation of Lemmon and Jacobsen 2004 and the equation of Sunaga et al. 1998 . The equation of Lemmon and Jacobsen uses a new form of the equation of state to eliminate the large calculated pressures both negative and
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

FIG. 12. Comparisons of densities calculated with the mixture model to experimental data for the R-125/134a/143a ternary mixture.

positive values typical of previous equations of state within the two phase region. The two equations represent the liquid and vapor densities for the pure uid with nearly the same deviations, but because recently measured data were not available to Sunaga et al., comparisons in the critical region are better for the equation of Lemmon and Jacobsen. Comparisons of the experimental data for the R-32/125 binary mixture showed similar trends in density in the liquid and vapor phases at temperatures away from the critical point when the equation of Lemmon and Jacobsen was replaced by the equation of Sunaga et al. However, at temperatures be0047-2689200433261428$39.00

tween 310 and 400 K differences ranged from 0.1% in density at pressures greater than 5 MPa to 0.5% in density at pressures between 2 and 5 MPa. At the critical point of R-410A 344.51 K, 4.9026 MPa , the calculated critical density differed by 17%. The equation of state of Lemmon and Jacobsen 2004 for R-125 should be used with the mixture equations presented here to obtain the uncertainties stated earlier for the mixture model. Although calculated densities and heat capacities in the liquid and vapor are generally quite similar between the new mixture model presented here and the JSRAE model of 614
J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

EQUATION OF STATE OF REFRIGERANT MIXTURES

615

FIG. 13. Comparisons of isochoric heat capacities calculated with the mixture model to experimental data for the R-32/125, R-32/134a, R-125/134a, and R-125/143a binary mixtures.

Tillner-Roth et al. 1998 , deviations in the critical region have been substantially improved by incorporating new experimental data that were not available to Tillner-Roth et al. For example, deviations for the older model exceed 0.5% in density at 330 K for the 0.7 R-32/0.3 R-125 mixture for the data of Magee, 2002 as shown in Fig. 19, although the deviations are quite similar at a composition of 0.5/0.5 for the data of Kleemiss 1997 and of Magee and Haynes 2000 . For the data of Zhang et al. 1996 , the equimolar data show similar comparisons, but higher deviations are observed at pressures near 3 MPa for the mixture containing 0.7 R-32/0.3

R-125. In addition, plots of excess volumes reveal several differences between these two models. In particular, for the R-32/125 mixture at 250 K and 5 MPa, Figs. 20 and 21 show excess volumes and excess enthalpies over the full composition range. Although the excess volumes are of similar magnitudes at compositions of R-32 above 0.5 mole fraction, the data of Benmansour and Richon 1999a conrm that the excess volumes should be negative at mole fractions of 0.21 and 0.25 of R-32, as demonstrated by the new mixture model. The model of Tillner-Roth et al. shows systematic deviations for both the excess volume and excess enthalpy.

FIG. 14. Comparisons of isochoric heat capacities calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture.

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

616

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 15. Comparisons of the speed of sound in the vapor phase calculated with the mixture model to experimental data for the R-32/125, R-32/134a, R-125/134a, and R-125/143a binary mixtures.

FIG. 16. Comparisons of the speed of sound in the vapor phase calculated with the mixture model to experimental data for the R-32/125/134a ternary mixture.

FIG. 17. Comparisons of the speed of sound in the liquid phase calculated with the mixture model to experimental data for the R-134a/152a binary mixture.

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

EQUATION OF STATE OF REFRIGERANT MIXTURES

TABLE 6. Calculated property values for computer code verication Isochoric heat capacity J/ mol-K 76.647 14 75.115 26 109.5538 71.035 76 66.892 97 101.7819 96.946 10 88.878 44 127.6533 91.511 91 88.849 67 122.3978 88.066 94 81.911 78 118.3833 85.804 24 76.512 68 113.1712 81.346 85 75.543 10 92.743 34 86.918 85 Isobaric heat capacity J/ mol-K 133.4161 113.2553 8875.064 113.8506 90.353 74 2496.022 151.7337 108.8584 3295.904 141.3994 121.2343 7882.864 132.5006 102.7968 3725.227 127.7686 95.584 40 7407.069 129.4822 97.923 34 148.5144 111.2741

617

Mixturea 50/50 R-32/125

b

Temp. K 300 300 343 300 300 364 300 300 358 300 300 344 300 300 362 300 300 381 300 300 300 300

Density mol/dm3 13. 0.9 5.8 15. 0.5 5.9 11. 0.4 4.9 11. 0.7 4.9 12. 0.4 5.1 13. 0.3 5.2 13. 0.5 11. 0.5

Pressure MPa 3.602 891 1.639 941 4.570 375 11.681 28 1.018 672 4.958 696 3.475 706 0.834 4532 3.992 433 7.371 538 1.303 202 3.756 290 9.548 637 0.826 9971 4.051 228 9.867 019 0.636 7737 4.320 773 7.889 929 1.023 377 2.309 797 0.997 6918

Speed of sound m/s 410.3700 145.0593 99.953 04 615.0602 167.3402 112.4511 438.1132 136.5385 86.673 45 445.2289 130.5114 86.752 97 546.5012 149.4877 94.848 13 644.0166 161.7513 104.4937 508.7768 151.3492 418.3732 138.1841

Fugacity of rst component MPa 0.760 4262 0.664 9367 1.609 648 0.872 4334 0.450 8831 1.955 381 0.631 9712 0.367 5279 1.467 750 0.708 4060 0.526 7233 1.251 023 0.409 5593 0.346 9823 1.175 577 0.403 9289 0.277 9604 1.414 853 0.547 0054 0.299 1716 0.389 4721 0.281 5930

50/50 R-32/134a
b

50/50 R-125/134a
b

50/50 R-125/143a
b

50/50 R-134a/143a
b

50/50 R-134a/152a
b

33/33/34 R-32/125/134a 33/33/34 R-125/134a/143a

a

Compositions are given in mole percent. Calculated state point is near the critical point.

FIG. 18. Isobaric heat capacity versus temperature diagram for an equimolar mixture of R-32 and R-125. J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

618

ERIC W. LEMMON AND RICHARD T JACOBSEN

FIG. 19. Comparisons of densities calculated with the mixture model developed in this work and the mixture model of Tillner-Roth et al. 1998 to experimental data for the R-32/125 binary mixture at 330 K.

FIG. 20. Excess volumes for the R-32/125 binary mixture at 250 K and 5 MPa: solid line this work, dashed line mixture model of Tillner-Roth et al. 1998 . Experimental data between 243 and 256 K are shown for comparison. J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

FIG. 21. Excess enthalpies for the R-32/125 binary mixture at 250 K and 5 MPa: solid line this work, dashed line mixture model of Tillner-Roth et al. 1998 .

EQUATION OF STATE OF REFRIGERANT MIXTURES

619

5. Acknowledgments
This work was funded by the 21-CR research program of the Air-Conditioning and Refrigeration Technology Institute; the Ofce of Building Technology, State and Community Programs of the U.S. Department of Energy. We thank Dr. Mark McLinden of NIST, Boulder, for his assistance and suggestions during the development and documentation of the equation.

6. References
Beliajeva, O. V., A. J. Grebenkov, T. A. Zajatz, and B. D. Timofejev, Int. J. Thermophys. 20, 1689 1999 . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 3, 149 1997 . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 4, 29 1998 . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 5, 9 1999a . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 5, 55 1999b . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 5, 127 1999c . Benmansour, S. and D. Richon, Int. Electron. J. Phys.-Chem. Data 5, 117 1999d . Bouchot, C. and D. Richon, Int. Electron. J. Phys.-Chem. Data 4, 163 1998 . Chung, E. Y. and M. S. Kim, J. Chem. Eng. Data 42, 1126 1997 . Debaugh, D. R. and G. Morrison, Int. J. Refrig. 18, 518 1995 . Dressner, M. and K. Bier, Fortschr.-Ber. VDI 3 1993 . Fujiwara, K., H. Momota, and M. Noguchi, 13th Japan Symp. Thermophys. Prop. A11.6, 61 1992 . Fujiwara, K., S. Nakamura, and M. Noguchi, J. Chem. Eng. Data 43, 967 1998 . Grebenkov, A. J., Yu. G. Kotelevsky, V. V. Saplitza, O. V. Beljaeva, T. A. Zajatz, and B. D. Timofeev, Proc. CFCs: The Day After, IIR Comm B1, B2, E1, E2, p. 21 1994 . Gunther, D. and F. Steimle, Int. J. Refrig. 20, 235 1996 . Higashi, Y., Int. J. Thermophys. 16, 1175 1995 . Higashi, Y., J. Chem. Eng. Data 42, 1269 1997 . Higashi, Y., Int. J. Thermophys. 20, 1483 1999a . Higashi, Y., J. Chem. Eng. Data 44, 328 1999b . Higashi, Y., J. Chem. Eng. Data 44, 333 1999c . Higuchi, M. and Y. Higashi, Proc. 16th Japan Symp. Thermophys. Prop., Hiroshima, 1995, p. 5. Higuchi, M., private communication with Yukihiro Higashi, Dept. of Mech. Eng., Iwaki Meisei University, Japan, 1997. Holcomb, C. D., J. W. Magee, J. Scott, S. L. Outcalt, and W. M. Haynes, Selected Thermodynamic Properties for Mixtures of R-32 diuoromethane , R-125 pentauoroethane , R-134A 1,1,1,2tetrauoroethane , R-143A 1,1,1-triuoroethane , R-41 uoromethane , R-290 propane , and R-744 carbon dioxide , National Institute of Standards and Technology, Tech Note 1397 1998 . Hozumi, T., private communication, Dept. of Mech. Eng., Keio University, Yokohama, Japan, 1996. Hozumi, T., H. Sato, and K. Watanabe, 4th Asian Thermophysical Properties Conference, Japan, 1995, p. 307. Hurly, J. J., J. W. Schmidt, and K. A. Gillis, Int. J. Thermophys. 18, 655 1997 . Ichikawa, T., K. Ogawa, H. Sato, and K. Watanabe, Speed-of-Sound Measurements for Gaseous Pentauoroethane and Binary Mixture of Pentauoroethane 1,1,1-Triuoroethane, Department of System Design Engineering, Faculty of Science and Technology, Keio University, Yokohama, Japan, 1998. Ikeda, T. and Y. Higashi, Proc. 16th Japan Symp. Thermophys. Prop., Hiroshima, 1995, pp. 169172.

Kato, R., K. Shirakawa, and H. Nishiumi, Fluid Phase Equilib. 194, 995 2002 . Kim, C.-N., E.-H. Lee, Y.-M. Park, J. Yoo, K.-H. Kim, J.-S. Lim, and B.-G. Lee, Int. J. Thermophys. 21, 871 2000 . Kim, C.-N. and Y.-M. Park, Int. J. Thermophys. 20, 519 1999 . Kishizawa, G., H. Sato, and K. Watanabe, Int. J. Thermophys. 20, 923 1999 . Kiyoura, H., J. Takebe, H. Uchida, H. Sato, and K. Watanabe, J. Chem. Eng. Data 41, 1409 1996 . Kleemiss, M., Fortschr.-Ber. VDI 19 1997 . Kleiber, M., Fluid Phase Equilib. 92, 149 1994 . Kubota, H. and T. Matsumoto, J. Chem. Eng. Jpn. 26, 320 1993 . Laesecke, A. Int. J. Refrig submitted, 2004 . Lemmon, E. W., A. Generalized Model for the Prediction of the Thermodynamic Properties of Mixtures Including Vapor-Liquid Equilibrium, Ph.D. dissertation, University of Idaho, Moscow, 1996. Lemmon, E. W. and R. T Jacobsen, Int. J. Thermophys. 20, 825 1999 . Lemmon, E. W. and R. T Jacobsen, J. Phys. Chem. Ref. Data 29, 521 2000 . Lemmon, E. W. and R. T Jacobsen, A New Functional Form and Fitting Techniques for Equations of State with Application to Pentauoroethane HFC-125 , J. Phys. Chem. Ref. Data, submitted, 2004 . Lemmon, E. W., R. T Jacobsen, S. G. Penoncello, and D. G. Friend, J. Phys. Chem. Ref. Data 29, 331 2000 . Lemmon, E. W. and M. O. McLinden, Method for Estimating Mixture Equation of State Parameters, Proceedings of the Thermophysical Properties and Transfer Processes of New Refrigerants Conference, Paderborn, Germany, 2001, pp. 2330. Lemmon, E. W., M. O. McLinden, and M. L. Huber, NIST Standard Reference Database 23: Reference Fluid Thermodynamic and Transport Properties-REFPROP, Version 7.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg 2002 . Lemmon, E. W. and R. Tillner-Roth, Fluid Phase Equilib. 165, 1 1999 . Li, J., R. Tillner-Roth, H. Sato, and K. Watanabe, Int. J. Thermophys. 20, 1639 1999 . Lim, J. S., J.-Y. Park, B.-G. Lee, and Y.-W. Lee, Fluid Phase Equilib. 193, 29 2002 . Lisal, M. and V. Vacek, Fluid Phase Equilib. 118, 61 1996 . Magee, J. W., Int. J. Thermophys. 21, 95 2000a . Magee, J. W., Int. J. Thermophys. 21, 151 2000b . Magee, J. W. private communication, NIST, Boulder, Colorado, 2002 . Magee, J. W. and W. W. Haynes, Int. J. Thermophys. 21, 113 2000 . Nagel, M. and K. Bier, Int. J. Refrig. 18, 534 1995 . Nagel, M. and K. Bier, Int. J. Refrig. 19, 264 1996 . Oguchi, K., K. Amano, T. Namiki, and N. Umezawa, Int. J. Thermophys. 20, 1667 1999 . Oguchi, K., T. Takaishi, N. Yada, T. Namiki, and T. Sato private communication, Kanagawa Institute of Technology, Japan, 1995 . Outcalt, S. L. and M. O. McLinden, J. Phys. Chem. Ref. Data 25, 605 1996 . Perkins, R. A. private communication, NIST, Boulder, Colorado, 2002 . Piao, C.-C., I. Iwata, and M. Noguchi private communication, Mech. Eng. Lab., Daikin Industries, Ltd., Japan, 1996 . Sand, J. R., S. K. Fischer, and J. A. Jones, Int. J. Refrig. 17, 123 1994 . Sato, T., H. Kiyoura, H. Sato, and K. Watanabe, J. Chem. Eng. Data 39, 855 1994 . Sato, T., H. Kiyoura, H. Sato, and K. Watanabe, Int. J. Thermophys. 17, 43 1996 . Schramm, B., J. Hauck, and L. Kern, Ber. Bunsenges. Phys. Chem. 96, 745 1992 . Shimawaki, S., K. Fujii, and Y. Higashi, Int. J. Thermophys. 23, 801 2002 . Sunaga, H., R. Tillner-Roth, H. Sato, and K. Watanabe, Int. J. Thermophys. 19, 1623 1998 . Tack, A. and K. Bier, Fortschr.-Ber. VDI 3, 1 1997 . Takagi, T., T. Sakura, T. Tsuji, and M. Hongo, Fluid Phase Equilib. 162, 171 1999 . Tillner-Roth, R., J. Chem. Thermodyn. 25, 1419 1993 . Tillner-Roth, R. and H. D. Baehr, J. Phys. Chem. Ref. Data 23, 657 1994 . Tillner-Roth, R., J. Li, A. Yokozeki, H. Sato, and K. Watanabe, Thermodynamic Properties of Pure and Blended Hydrouorocarbon (HFC) Refrigerants Japan Society of Refrigerating and Air Conditioning Engineers, Tokyo, 1998 . J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004

620

ERIC W. LEMMON AND RICHARD T JACOBSEN

Widiatmo, J. V., H. Sato, and K. Watanabe, Int. J. Thermophys. 16, 801 1994a . Widiatmo, J. V., H. Sato, and K. Watanabe, Fluid Phase Equilib. 99, 199 1994b . Yokoyama, C., T. Nishino, and M. Takahashi, Fluid Phase Equilib. 174, 231 2000 . Zhang, H.-L., H. Sato, and K. Watanabe, J. Chem. Eng. Data 41, 1401 1996 . Zhang, H.-L., S. Tada, H. Sato, and K. Watanabe, Fluid Phase Equilib. 151, 333 1998 .

Tillner-Roth, R. and A. Yokozeki, J. Phys. Chem. Ref. Data 25, 1273 1997 . Tuerk, M., M. Crone, and K. Bier, J. Chem. Thermodyn. 28, 1179 1996 . Uchida, H., H. Sato, and K. Watanabe, Int. J. Thermophys. 20, 97 1999 . Weber, F., Fluid Phase Equilib. 174, 165 2000 . Weber, L. A. and D. R. Debaugh, Int. J. Thermophys. 15, 863 1994 . Widiatmo, J. V., T. Fujimine, H. Sato, and K. Watanabe, J. Chem. Eng. Data 42, 270 1997 . Widiatmo, J. V., H. Sato, and K. Watanabe, High Temp.-High Press. 25, 677 1993 .

J. Phys. Chem. Ref. Data, Vol. 33, No. 2, 2004