Int J. Refrig. Vol. 20, No. 4, pp.

256 266, 1997

© 1997 Elsevier Science Ltd and IIR
All rights reserved. Printed in Great Britain
ELSEVIER PII:S0140-7007(97)00009-01 0140-7007/97/$17.00 + .00

Working fluids for an absorption system based on

R124 (2-chloro-l,l,l,2,-tetrafluoroethane) and
organic absorbents
I. Borde
M e c h a n i c a l E n g i n e e r i n g D e p a r t m e n t , B e n - G u r i o n U n i v e r s i t y o f the
N e g e v , P.O. B o x 653, B e e r - S h e v a 84105, Israel

M. Jelinek and N. C. Daltrophe

T h e I n s t i t u t e s f o r A p p l i e d R e s e a r c h , B e n - G u r i o n U n i v e r s i t y o f the
N e g e v , P.O. B o x 653, B e e r - S h e v a 84105, Israel
Received 4 June 1996; revised 2 January 1997; accepted 26 January 1997

The possibility of using R124 (2-chloro-l,l,l,2,-tetrafluoroethane, CHC1FCF3) and

organic absorbents as working fluids in absorption heat pumps was investigated. Vari-
ous classes of organic compounds, all commercially available, were tested as absorbents
for possible combination with R124; the absorbents included DMAC (N',N'-dime -
thylacetamide, CaH9NO), NMP (N-methyl-2-pyrrolidone, CsH9NO), MCL (N-methyl
e-caprolactam, C7HI3NO), DMEU (dimethylethylene urea, CsH10N20), and DMETEG
(dimethylether tetraethyleneglycol, C~0H2205). To evaluate the performance of a
candidate refrigerant-absorbent pair in a refrigeration or heat pump cycle, the
thermophysical properties of the pure components and the mixture and the equilibrium
and transport properties have to be determined, either from experimental data or by
prediction methods. The thermal stability of the refrigerant-absorbent must also be
tested. A method for the calculation of the concentration in the liquid and gas phases and
the excess thermodynamic properties of the mixture as a function of the system
temperature and pressure based on our experimental setup is described. On the basis of
vapor-liquid equilibrium measurements, density and viscosity measurements and
thermostability testing, enthalpy-concentration diagrams were constructed. The
performance characteristics of the investigated working fluids in terms of the coefficient
of performance (COP) and the circulation ratio ( f ) were calculated for a single-stage
absorption cycle. In terms of overall performance (COP, f and stability) R124-DMAC
was found to be the superior combination, followed by R124-NMP, RI24-DMEU and
R124-MCL (the three pairs for which stability problems were found at high
temperatures), and finally by R124-DMETEG. © 1997 Elsevier Science Ltd and IIR.
(Keywords: absorption machine; absorbant; chemical composition; thermo-dynamic property;
measurement;COP)

Couples frigorig6nes 5. base de R124

(2-chloro- 1,1,1,2-t&rafluoro6thane) et d'absorbants
organiques pour systemes ~ absorption
On a ~tudiO la possibilitO d'utiliser le R124 (2-chloro-l,l,l,2-tOtrafluoro~thane,
CHCIFCF3) et des absorbants organiques comme fluides actifs dans des pompes it chaleur
it absorption. On a essay~ diffOrentes catOgories de composants organiques, tous
commercialis~s, comme absorbants pouvant £tre associ~s au R124. I1 s'agissait des produits
suivants: le D M A C (N, NP-dim~thylacdtamide, C4HgNO), le N M P (N-m~thyl-2-
pyrrolidone, CsHgNO), le MCL (N-mdthyl e-caprolactam, CzH13NO) le DMEU (de
dimdthyldthyl~ne urge, CsH~oN20) et le DMETEG (dim~thyl~ther tdtraOthylbne-glycol,
CWH2205). Pour ~valuer la performance d'un couple frigorigbne-absorbant dans un cycle
frigorifique ou de pompe ~ chaleur, il faut d~terminer les propri~t~s thermophysiques des

256
 Working fluids for an absorption system based on R124 and organic absorbents 257

Nomenclature
A Parameter in the Van Laar equation Greek letters
B Second virial coefficient, parameter in the
Van Laar equation 3' Activity coefficient
C o Specific heat capacity r/ Viscosity (10 -3 Pa s) or (cP)
f Circulation ratio (rhs/rhr) Weight fraction in liquid phase
G Weight ~b Weight fraction in vapor phase
G E Excess molar Gibbs energy p Density (10-3kgm 3) or (gcm 3)
h Specific enthalpy (kJ kg- 1) /~ Correction factor
H E Excess molar enthalpy
hE Specific excess enthalpy Subscripts
M w Molecular weight
rhr Mass flow rate of the refrigerant a Absorbent
rhs Mass flow rate of the strong solution b N.B.P
P Pressure (bars) c Critical point
Qe Heat rejected from the evaporator e Evaporator
Qg Heat supplied to the generator g Gas
T Temperature (K) r Refrigerant
t Temperature (°C) s Solution
V volume
V0 The total free volume of the reactor Superscripts
V Molar volume
v Specific volume 1 Liquid
Wp Energy supplied to the pump s Saturation
x Mole fraction in liquid phase
y Mole fraction in vapor phase

corps purs et du mOlange, et les propri~tds d~quilibre et de transfert, sait par des mesures
expOrimentales, solt par des m~thodes de provision. Il faut aussi tester la stabilit~ thermique
du mdlange frigorigbne-absorbant. On dOcrit une mOthode de calcul de la concentration en
phases liquide et gazeuse et des autres propridtOs thermodynamiques du m~lange, en
fonction de la temperature et de la pression du systOme, sur notre dispositif exp~rimental.
On a Otabli les diagrammes enthalphie-concentration par des mesures it l~quilibre vapeur
liquide, des mesures de densitO et de viscositd et des essais de thermostabilit~. On a calculO
les caractOristiques de performance des frigorigbnes ~tudiks exprimdes par le coefficient de
performance (COP) et le rapport de circulation (f) pour un cycle it absorption monoOtagO.
Pour la performance g~n~rale (COP, f et stabilitY), on a trouvO que R124-DMAC est la
meilleure combinaison, suivie par R124-NMP, R124-DMEU et R124-MCL (trois couples
pour lesquels on a rencontrO des problbmes de stabilitO it haute temperature), et enfin le
R124-DMETEG. © 1997 Elsevier Science Ltd et IIR.
(Mots cl6s: machine ~i absorption; R124; absorbant; composition chimique; propri&+ thermo-
dynamique; mesure; COP)

The possibility of using the refrigerant R 124 (2-chloro- tested as absorbents for possible combination with
l,l,l,2,-tetrafluoroethane, CHCIFCF3) as an alter- R!24.
native refrigerant in absorption heat pumps and In order to evaluate the performance of the
refrigeration units was investigated. Various classes candidate refrigerant-absorbent pairs in a refrigera-
of organic compounds, all commercially available, tion or heat pump cycle in terms of the coefficient
such as DMAC (N, N '-dimethylacetamide, CaH9NO), of performance (COP) and the circulation ratio
NMP (N-methyl-2-pyrrolidone, CsH9NO ), MCL (f), the thermophysical properties of the pure
(N-methyl e-caprolactam, C7HI3NO), DMEU components and the mixtures, the equilibrium and
(dimethylethylene urea, CsHIoN2 O) and DMETEG transport properties and the thermal stability of the
(dimethylether tetraethyleneglycol, C10H2205) , were refrigerant-absorbent pair have to be determined
258 I. Borde et al.

from experimental data or by prediction methods. ducer was calibrated with a Beamex pressure cali-
The method of calculation of the concentration in the brator type PCI05. Weight was determined on a
liquid and gas phases as a function of temperature METTLER electronic balance type PM4000 (resolu-
and weight fraction based on our experimental tion of 0.01 g, accuracy ±0.02 g).
setupl 3, the procedure for calculating the excess For each set of measurements, the reactor was
thermodynamic properties of the mixture, and the weighed so that the total weight of the refrigerant G~
construction of enthalpy-concentration diagrams are and the absorbent Ga in the vessel were known. The
presented. In addition, the density and the viscosity temperature was increased and at equilibrium the
and the stability tests of the investigated working pressure was measured. At the end of each set of
fluids were evaluated. Finally, the performance measurements, the reactor was cooled down to the
characteristics of the investigated working fluids in ambient temperature, a certain amount of gas was
terms of the COP and f were calculated with the aid released, and the reactor was weighed again. From the
of a computerized simulation program based on a amount of gas released, the amount of absorbent
single-stage absorption cycle containing a jet ejector vapor released could be estimated. This procedure
mixer4-6. was repeated up to the end of the experiment.
The results of the investigation are presented for the
pair R124-DMAC, as a model working fluid.
Model for calculation of the concentration
The mixtures in the liquid phase and in the vapor
Vapor-liquid equilibrium of mixtures
phase were treated separately, where the equilibrium
between the two phases was defined as the equality of
Experimental setup for the equilibrium measurements the fugasities of the components in the two phases,
With the aim of screening of large numbers of f~' = f l . The fugasity equality of the refrigerant and
potentially interesting working fluids for absorption the absorbent in the two phases may be expressed as:
cycles, we designed and built the experimental setup
yrP = Xr%~rrFr (1)
presented in Figure 1 and developed an evaluation
methodology that permits measurements of large ya P = xa%pSaFa (2)
amounts of data with sufficient accuracy for the while the pressure correcting factor Pi is equal to:
generation of equilibrium properties of refrigerant-
absorbent pairs 1. The temperature at equilibrium up F i = exp [ ( V' - Bi)( P - P~i (3)
to 150°C was measured in the liquid phase with a
PT100 RTD with DORIC type 410A temperature
indicator (resolution of 0.1°C, accuracy of ~0.2°C), where V] and Bi are the molar volume and second
and in the gas phase above the solution by a TC type virial coefficient of component i in the liquid phase,
T with YEW type 2575 digital thermometer (resolu- respectively.
tion of 0.1°C, accuracy of +0.4°C). The absolute The total free volume of the reactor V0 is the sum of
pressure in the vessel was measured up to 20 bars with the volumes occupied by the solution Vs and the gas
a Druck pressure transducer type PDCR 961 with a vg.
DPI 261 digital pressure indicator (resolution of V0 = Vs + Vg (4)
0.001 bar, accuracy ± 0.1 BSL). The pressure trans-
The weight of the refrigerant in the reactor Gr is the
sum of the weights of the refrigerant in the liquid
phase Grs and in the gas phase Grg.
,1 G r = Grs q- Grg (5)
The weight of the absorbent in the vessel Ga is the sum
3 of the weights of the absorbent in the liquid phase Gas
and in the gas phase Gag.
1. Equilibrium measurement
ISVapo r " reactor. G a = Gas + Gag (6)
I~ phase 2. Electrical heater
3. RTD for solution temperature The volumes of the gas phase Vg and the liquid phase
,oil , 4. Thermocouple for gas temperature
5. Pressure transducer
6. Valve
Vs are given by Equations (7) and (8):
[[~Liquid 2
7. Safetyvalve Vg = Grgvrg q- Gagvag (7)
8. Magneticstirrer
9. Stirring and heating plate Vs = Grsvrs + Gasvas (8)
~r====9====~ 10. Insulationaround the reactor
[ ® m ® I where Vrs, Vas, 'Org and 'Uag are the specific volumes of
Figure 1 Experimental setup for the equilibrium measure- the refrigerant and the absorbent in the liquid and
ments vapor phases as function of the temperature and/or
Figure 1 Dispositif exp&irnental de mesures il l'~quilibre de the pressure of the system.
phase By rewriting Equations (4) to (8), the weight of the
 Working fluids for an absorption system based on R 124 and organic absorbents 259

refrigerant in the solution Grs as a function of the The relationship between the activity coefficients can
u n k n o w n weight of the absorbent in the gas phase Gag be obtained by dividing Equation (13a) by Equation
can be found from Equation (9): (13) to give:
V0 -- vrgGr -- vagGa -- (G a - aag)(Vas -- Vag ) g (1 + 1//3"]2
Grs = ln% = ~ \~-ff-j InTr (14)
Vrs -- Vrg)
(9) where/3 = (B/A)(xa/Xr) and B/A = Va/Vr, in which
The weight fraction and the mole fraction of the Va and Vr are the molar volume of the absorbent and
refrigerant and the absorbent in the solution are given the refrigerant in the liquid phase, respectively. By
by the equations: solving Equations (1), (2), (9)-(11) and (14), the weight
fractions of the refrigerant and the absorbent in the
Grs ~rMa liquid phase and in the gas phase can be calculated.
~r = ?Grs + Gas ) ; Xr (~rMa + ~aMr )
It should be noted that Vrs - the specific volume of
~a = l --(r ; Xa = 1 - x r (10) the refrigerant in the liquid phase - was obtained as a
function of the pressure rather than as a function of
and the weight fraction and the mole fraction of the the temperature of the system.
refrigerant and the absorbent in the gas phase are
given by the equations:
Thermophysical properties of the pure components
Grg ~rMa
~r = (arg + Gag ) Yr = (/PrMa _}_~)aMr) The basic data required for the pure components
comprise molecular weight, normal boiling point,
~a=l--~r ya= 1--yr (11) critical temperature, critical pressure and critical
Since there is one u n k n o w n variable in Equation (9), volume (Table 1).
the relationship between the activity coefficients was
used. By adopting the Van Laar model for the excess Vapor pressure of the pure components
Gibbs energy, the relationship between the activity
The vapor pressures of the refrigerant and the
coefficients can be found:
absorbents as a function of temperature are given by:
gE Axrx a (12) B
RT Xr(B/A) + Xa In(Ps)=A+~-~+DT+EIn(T)+FT 2 (15)

In% = A l l + ( A ) (~aa)]
Xr -2 (13) where Ps is the saturated vapor pressure in bars and T
is the temperature in K. The coefficients A to F were
obtained by regression of the vapor pressure of each
ln"/a=B[l+(B) (x-2axl]-2kxr/j (13a) component using the Statistical Package for Social
Sciences (SPSS) (Table 2).

Table l Basic data for the pure components

Tableau 1 Donn~esde base pour les corps purs
Formula Mw tb (°C) tc (°C) Pc (bar) vc (m3 kg i)
R 124 CHCIFCF3 136.50 - 11.0 122.2 35.74 0.001770
DMAC CaHgNO 87.12 165.0 382.4 42.11 0.003254
NMP CsHgNO 99.13 203.0 453.1 45.96 0.003183
MCL CyH13NO 127.19 235.8 485.0 38.25 0.003235
DMEU CsHIoN20 114.15 223.0 472.3 44.44 0.003211
DMETEG C10H2205 222.29 275.3 440.2 19.53 0.003084

Table 2 Coefficients A to F in Equation (1) for the saturated vapor pressure of the pure components
Tableau 2 Coefficientsde A g~F dans I'~quation (1) pour la pression de vapeur saturOe des corps purs
,4 B C D E F
R124 - 119.33446 - 1351.341 0 -0.12497 27.21197 8.32656E-5
DMAC 6.01834 -1800.240 - 138.5 0 0 0
NMP 8.50639 -3063.357 - 115.4 0 0 0
MCL 8.09382 -3078.393 - 130.0 0 0 0
DMEU 26,80333 -25627.39 459.0 0 0 0
DM ETEG 24,56738 -21564.48 328.4 0 0 0
260 I. B o r d e et al.

25
R 124-DMAC
RI24-DMAC

20
"~ -2
-10
e~
ca -3 2~
15
-4
iooJ / / / / / -15 ~.

i
t~ -20 ea
ea -5 60
u~
,~ 10 ca
-6 40
-25 ~

2O
-30

-35

-90
I I t I
0.2 0.4 0.6 0.8 1.0
Mole fraction [X]
0 0.2 0.4 0.6 0.8 1.0
Weight fraction [~] Figure 3 Excess enthalpy for R I24-DMAC
Figure 2 PT~ for R124-DMAC Figure 3 Enthalpie suppl~mentaire pour le R124-DMAC
Figure 2 PT( pour le R124-DMAC

Pressure, temperature, concentration relationships The U N I F A C model 7 was used to evaluate the
The relationship linking concentration, temperature activity coefficient -y in the liquid phase of nonelec-
and pressure under equilibrium conditions can be trolytic substances.
expressed as a polynomial of the form: The specific excess enthalpy h E may be expressed as
a polynomial of the form:
7 7
P=Z E Pid~iTj (16) 7 7
j=0 i--0 h E = 4.1868 Z Z hijxiTJ (19)
j=0 i-0
where P is the pressure of the system in bars, ~ is the
weight fraction of the refrigerant in the liquid where h E is given in kJ kg -1 and x is the mole fraction
solution, and T is the temperature in K. The of the refrigerant in the liquid solution. The coeffi-
coefficients Pzj were obtained by regression of all the cients hij were obtained by regression of all the points
points determined by equilibrium measurements calculated by the U N I F A C model, using SPSS.
and the additional pure component vapor pressure, The coefficients hij for R124-DMAC are presented
using SPSS. The coefficients Pzj for R 124-DMAC are Table 4, and the specific excess enthalpy h E as a
presented in Table 3, and the pressure-temperature- function of the temperature and the molar fraction is
weight fraction relationships are shown in Figure 2. shown in Figure 3.
The large number of measured points over a broad It should be noted that the U N I F A C model was
temperature-concentration-pressure field is evident used solely in order to evaluate the excess properties
from these figures. required for constructing enthalpy-concentration
diagrams.
Excess thermodynamic properties
The excess molar Gibbs energy of mixing G E is the Enthalpy-concentration diagrams
point of departure for calculating other excess
The evaluation and design of absorption systems
properties. G E of a solution expressed in terms of
requires a knowledge of the enthalpy of the mixtures
the molar concentration and the activity coefficient is
in the liquid phase and of the enthalpy in the vapor
given as:
phase as a function of temperature, pressure and
G z = R T ~ ( x i In("//)) (17) concentration. To construct enthalpy-concentration-
pressure-temperature diagrams for solutions, the
From Equation (17), the expression for the molar specific enthalpy must be obtained as a function of
excess enthalpy H E of the mixture can be derived as the temperature of the pure components in the liquid
follows: phase.
The specific enthalpy of the absorbents in the liquid
HE ~O[GE/T]~ (18) phase as a function of the temperature can be derived
= t oE- 7 Tj : from the specific heat by means of the following
 Working fluids for an absorption system based on R 124 and organic absorbents 261

F r r I -F I I I

r.-- i I

-F I i
-F f

I I

I I
•FI i i
oo

I
M

+ i I -F I f

-F+ t II -F-F I I

I I

L~
-F-F I I I -F-F ~ I

< < ~-

I I
4~

+...~

Ill I i
NN~ N NN

0
i @
~i I I i

[., [.-
262 I. Borde et al.

Table 5 CoefficientsCpo and Cp~ in Equation (20) for the specific enthalpy of the absorbents in the liquid phase
Tableau 5 Coefficients Cpo et Cpl dans l'~quation (20) pour l'enthalpie sp~cifique des absorbants dans la phase liquide
DMAC NMP MCL DMEU DMETEG
Cpo 0.4578 0.3732 0.3980 0.3286 0.5276
Cpl 0.000839 0.000234 0.001203 0.0003286 0.000294

Table 6 Coefficientshrl to hr7 in Equation (21) for the specific enthalpy of R124 in the liquid phase
Tableau 6 Coefficients h, h h~7 dans l'dquation (21) pour l'enthalpie spdcifique du R124 dans la phase liquide
h~t hr2 hr3 hr4 hr5 hr6 hr7
R 124 0.250767 0.347753E-3 0.100977E-5 0.342458E-8 0 0 0.699612E- 14

2~ From the specific enthalpy of the pure components

R124-DMAC 600 and the excess specific enthalpy of mixing, it is
possible to calculate the enthalpy of the solution in
18G the liquid phase, which consists of the sum of the
weighted contributions of the two components, as a
400
function of the temperature and concentration, as
161] follows:

300 h =hr,+ha(1 -~) q-h E (22)

140
z= The enthalpy-concentration diagrams for R124-
"60
200
DMAC as a function of the weight fraction of the
120 refrigerant for a set of isobars and isotherms in the
state of equilibrium are presented in Figure 4.
Since the thermal effects of the mixtures were
100 I00 evaluated with the help of theoretical models for the
excess molar Gibbs energy of mixing, and the
behavior of the excess enthalpy of mixing is a product
I I I I
0.2 0.4 0.6 0.8 1.0 of the model, different values and shape relationships
Weight fraction [~] can be obtained. It is important to understand that
the behavior of the excess enthalpy of mixing as a
Figure 4 Enthalpy-concentration diagram for R124-
DMAC function of the temperature and weight fraction
directly influences the enthalpy-concentration dia-
Figure 4 Diagramme enthalpie-concentration pour le R124-
DMAC grams and hence the thermodynamic calculation of
the absorption cycle, i.e the COP. This can cause
equation: changes in the COP up to 15%.

ha = 100 + 4.1868 [Cpol + 1 Cp I/2] (20)

Density and viscosity of the pure components and
where ha is given in kJkg -1 and t in °C. The mixture
coefficients Cpo and Cpl for the absorbents are
Simultaneous measurements of density and viscosity
given in Table 5.
of pure absorbents and their mixtures with the
The specific enthalpy of the refrigerant in the liquid
refrigerant were carried out in our experimental
phase as a function of the temperature can be derived
setup 2 at temperatures up to 120°C and pressures up
from published experimental data or from predicted
to 16bars. Since the system was not designed for
data s by means of the following equation:
measuring the density and the viscosity of the pure
refrigerants, the density and viscosity of the pure
hr = 100 + 4.1868 hrit i (21) refrigerant were taken from the literature.
.=

where hr is given in kJkg -1 and t in °C. The

Pure components
coefficients hrl for R124 are given in Table 6. The
enthalpies of the pure components at 0°C were taken The measured density of the pure absorbents may be
to be 100kJkg 1. expressed as a function of temperature by the use of a
 Working fluids for an absorption system based on R 1 2 4 and organic absorbents 263

Table 7 Coefficients Po to P3 in Equation (23)

Tableau 7 Coefficients Po it P3 dans l'~quation (23)

P0 Pl P2 P3

R 124 1.4660 -0.3394E-2 -0.16063E-5 -0.20040E-6

DMAC 0.9528 -0.5856E-3 -0.33632E-5 0
NMP 1.0387 -0.3891 E-3 -0.46746E-5 0
MCL 1.0150 -0.4408E-3 -0.30513E-5 0
DMEU 1.0615 -0.3216E-3 -0.44295E-5 0
DMETEG 1.0157 -0.2281E-3 -0.58465E-5 0

Table 8 Coefficients % to ~/2 in Equation (24) properties of the pure components, a complete set of
Tableau 8 Coefficients 71oh ~2 dans l~quation (24) data was obtained for density and viscosity as a
function of temperature and weight fraction.
~o ~1 ~2 The density as a function of temperature and
R124" -8.9301 5094.90 357.4 weight fraction may be expressed as:
DMAC - 16.8400 1348.80 -223.2 3 3
NMP - 1.3984 202.08 - 196.3
MCL - 1.4098 256.99 -207.3 p= F, piStJ (25)
DMEU - 1.4610 239.99 - 193.1 j=0 i=0
DMETEG -3.1570 844.54 - 103.2 where p is given in 10 -3 k g m -3. The coefficients pq
* Based on predicted data were obtained by regression of all the data, using
SPSS. The coefficients Pij for R I 2 4 - D M A C are pre-
sented in Table 9, and the densities of refrigerant-
polynomial equation: absorbent mixtures as functions of the temperature
3 and the weight fraction at equilibrium, as calculated
p = ~_, pi ti (23) by means of Equation (25), are shown in Figure 5.
i=0 The viscosity as a function of temperature and
where p is expressed in l0 -3 kgm -3 and t in °C. The weight fraction may be expressed as:
coefficients P0 to P3 for the refrigerants and the 3 3
absorbents are given in Table 7. In(r/) = Z Z r/ij~ iT -j (26)
The measured viscosity of the pure absorbents may j=0 /-0

be expressed as a function of temperature by the use where r/is expressed in 10 -3 Pa s. The coefficients r/~j
of the following equation: were obtained by regression of all the data, using
r/I (24) SPSS. The coefficients r/ij for R 1 2 4 - D M A C are pre-
In(r/) = r/0 q T + 772 sented in Table 10, and the viscosities of refrigerant-
absorbent mixtures as functions of the temperature
where r/ is expressed in 10 3 P a s and T in K. The
and the weight fraction at equilibrium, as calculated
coefficients r/0 to r/2 for the refrigerants and the
by means of Equation (26), are shown in Figure 6.
absorbents are given in Table 8.

Thermostability tests of the working fluid

Absorbent-refrigerant mixture
The stability of the working fluids may be defined as
The densities and viscosities of the a b s o r b e n t - their ability to retain constant thermophysical proper-
refrigerant pairs were measured at equilibrium con- ties under working conditions so as not to signifi-
ditions as functions of the temperature and the cantly affect the performance of the system over long
pressure of the system. By using Equation (16) for periods of service. This is a broad definition, which
the calculation of the weight fraction and the includes the various aspects of stability that are

Table 9 Coefficients p~j for R I24-DMAC

Tableau 9 Coeffic&nts Pid pour le RI24-DMAC
i

j 0 1 2 3
0 0.95521E+01 0.43672E+00 -0.28160E+00 0.35610E+00
1 -0.15533E-03 -0.37079E-03 0 -0.29230E-02
2 -0.13981E-04 0 0.13550E-04 0
3 0.60941 E-07 -0.97547E-07 0.33531 E-06 -0.50554E-06
264 I. Borde et al.

1.6 R124-DMAC ] 1"6

R124-DMAC

1.4 ~fc " ~ 1.4

1.2
~ . _ , . / 0°C
E '7
g.
'= i.a 0 1.0 ~---

-I- - , C , to..
to.°
0.8 ~z
~7

0.6 \\
0.4 I I I I 0.4
0 0.2 0.4 0.6 0.8 1.0
Weight fraction [{l
I I I I
Figure 5 pT~ for R124-DMAC 0 0.2 0.4 0.6 0.8 1.0
Figure 5 p T ( pour le R 1 2 4 - D M A C Weight fraction [~,]
Figure 6 r/T~ for RI24-DMAC
Figure 6 tiT ~ pour le R 1 2 4 - D M A C
generally mentioned and sometimes studied sepa-
rately, such as the thermal stability of the pure
components, the chemical stability of the working The results of the preliminary short-term stability
fluid, the corrosiveness of the fluid towards the tests for the working fluid pairs are given in Table 11.
equipment, and the catalytic action of the equipment As can be seen from the table, no changes were found
on the fluid. in the gas phase, but in the liquid phase the following
The stability of the working fluids was tested using changes were noted: for R124-NMP the color
the method previously described by Daltrophe et al. 3 changed to deep brown, and a light precipitate
These preliminary short-term tests were carried out at formed; for RI24-DMEU the color changed to
200°C over 6h in a stainless steel (SST316) vessel. black, and a dense precipitate formed; and for
Changes in color, formation of precipitate or crystals R124-MCL the color changed to deep brown. By
and turbidity in the liquid phase were followed for adding small amount of stabilizer, such as OLOA260,
several days in a glass tube. and improvement was obtained for R124-DMEU and

Table10 Coefficients r/ia for RI24-DMAC

Tableau 10 Coefficients Oij pour le R 1 2 4 - D M A C
i
j 0 1 2 3
0 0.76992E+01 0.60007E÷00 -0.89875E+01 -0.17031E+01
1 -0.67692E+04 0 0.45554E+04 0
2 0.13434E+07 0 0 0
3 0 -0.64446E+08 0 -0.10631E+09

Tablel l Results of the preliminary short-term stability tests at 200°C with R124 and several absorbents
Tableau 11 ROsultats des essals pr~liminaires de stabilit~ h court terme it 200°C avec le R124 et plusieurs absorbants
Absorbent Gas analysis (GC) Liquid analysis: GC/color/precipitate/turbidity Stabilizer OLOA260
DMAC Nothing Light color No improvement
NMP Nothing Deep brown, light precipitate Improvement
MCL Nothing Deep brown No improvement
DMEU Nothing Black, strong precipitate Improvement
DMETEG Nothing Light color No improvement
 Working fluids for an absorption system based on R 124 and organic absorbents 265

G ,• RI24-NMP, and for Rl24-DMAC and

DMETEG a small color change was noted.
R124-

Single-stage absorption cycle

The performance characteristics of the investigated
H~ working fluids in terms of COP a n d f w e r e based on
the thermodynamic calculation of a single-stage
absorption cycle containing a jet ejector mixer4 6,
)
presented schematically in Figure 7.
M

Theoretical performance
A
The absorption cycle (Figure 7) was calculated in
Figure 7 Schematic representation of a single stage terms of the heat and mass balance for each
absorption cycle: G, generator; A, absorber; C, condenser; component with the aid of a computerized simulation
E, evaporator; HS, solution heat exchanger; P, solution program based on the thermophysical properties of
pump; M, jet ejector the refrigerant-absorbent pairs measured and evalu-
Figure 7 Schema d'un cycle mono~tag~ ~ absorption." G, ated in our laboratory. As a result of this cycle
gdndrateur; A, absorbeur," C, condenseur; E, dvaporateur," HS, analysis, COP a n d f c a n be calculated under different
Ochangeur de chaleur pour la solution; P, pompe de la solution,"
temperatures of the components. COP (based on the
M, ~jecteur
first law of thermodynamics) andfwere calculated by
the equations:
0.60 rhs
COP= Qe ; f=7- (27)
Qg + Wp mr
0.55
where Qe, Q~ and Wp are the heat rejected from the
0.50 _ o Q B • • • • • • • evaporator, the heat supplied to the generator and
q
the energy supplied to the pump, respectively; rhs is
0.45' O [] 0 O 0 0
0 [] [] O O
O
the mass flow rate of the strong solution (from the
[]
absorber to the generator); and rhr is the mass flow
0.4C - o DMAC
o NMP rate of the refrigerant.
• MCL As an illustration, the absorption cycle in Figure 7
035 • DMEU
[] DMETEO was calculated under the following operating condi-
0.30 I I I I I I tions: generator temperature of 80-150°C, evapora-
80 90 100 110 120 130 140 150 tor temperature of -5°C, condenser temperature of
Tg °C 32°C and absorber temperature of 28°C. COP a n d f a s
Figure 8 Variation of COP with generator temperature functions of the generator temperature for the five
Figure 8 Variation du COP en fonction de la temperature au working fluids are given in Figures 8 and 9.
g~n~rateur As can be seen from Figure 8, the highest values of
COP were obtained for R124-NMP and RI24-
DMEU; the lower values were found for RI24-
DMAC and R124-MCL; and the lowest values was
71 0 given by RI24-DMETEG. Figure 9 shows that the
O DMAC lowest values f o r f w e r e given by R124-DMAC and
6 O
~ o NMP RI24-NMP. Slightly higher values were obtained by
[] • MCL
5 • DMEU RI24-DMEU and RI24-MCL and the highest values
I o [] DMETEG were produced by R124-DMETEG.
0
f o In terms of overall performance (COP, f and
| o
O D stability), R124-DMAC was the superior combina-
I [] []
o II O O tion, followed by R124-NMP, R124-DMEU and
O
R 124-MCL (pairs exhibited stability problems at high
temperatures), and finally by R124-DMETEG.

I I I t I I
80 90 100 110 120 130 140 150 Discussion and conclusions
Tg °C
In this paper, a model for the calculation of the weight
Figure 9 Variation of f with generator temperature fraction of the refrigerant in the solution based on
Figure 9 Variation de f e n function de la temperature au equilibrium data is presented in detail. The obtained
g~nOrateur calculated data (PTxy) were used to construct the
266 I. Borde et al.

enthalpy-concentration diagrams for the investigated References

working fluids. By using our methodology, a large
1 Borde,I., Jelinek, M. Enthalpy and exergy-concen-
number of measured points were obtained with suffic- tration diagrams for the mixture of R22 with N,N-
ient accuracy over a broad spectrum of temperature- dimethylacetamide. Proceedings of the XVllth
pressure-concentration ranges. With the aid of a International Conference of Refrigeration, Vienna,
computerized simulation program for a single-stage Austria (1987) 172-177
absorption cycle, the performance characteristics were 2 Borde, I., Jelinek, M., Daltrophe, N. C. Development
of advanced absorption systems driven by low tem-
calculated at the same conditions for all investigated perature heat sources. Energy Efficiency in Process
working fluids and could thus be compared. Technology, Commission of the European Commu-
The investigation showed that in absorption units nities, Elsevier, Applied Science, London (1993)
R124 can be used as an alternative refrigerant in 521-530
3 Daitrophe, N. C., Jelinek, M., Borde, I. Stability test
combination with some of the absorbents tested by us. for working fluids for absorption systems. Thermo-
In terms of overall performance, RI24-DMAC was physical Propert&s of Pure Substances and Mixtures
found to be the best combination. In terms of COP Jot Refrigeration, International Institute of Refrigera-
and f, R124-NMP and R124-DMEU showed similar tion Commission BI, Herzlia, Israel (1990) 233-240
values to RI24-DMAC, but these two combinations 4 Borfle, I., Jelinek, M., Daltrophe, N. C. Absorption
system based on the refrigerant R134a. Int. J. Refrig.
exhibited stability problems at high temperatures. (1995) 18 387-394
These stability problems can be improved by using 5 Borde, I., Jelinek, M., Daltrophe, N. C. Working sub-
stabilizers such as OLOA260, but more stability tests stances for absorption heat pumps based on R32.
have to be carried out at lower temperatures. Proceedings of the 19th International Congress of
A comparison of the performances of working fluids Refrigeration, The Hague, The Netherlands (1995)
80-87
based on R124 with those based on the refrigerants 6 Daltrophe,N. C., Jelinek, M., Borde, I. Heat and
R229-1t and R1342'4'5 showed that the working fluids mass transfer in a jet ejector for absorption systems.
based on R 124 had lower COP values than those based A ES- VOl. 31, Proceedings of the International A bsorp-
on R22, but higher values than those based on R134a. tion Heat Pump Conference ASME 1993, New
Orleans, LA (1994) 327-332
The working fluids based on R124 had lower fvalues 7 Fredenslund, A., Gmehling, I., Rasmussen, P. Vapor-
than those based on R22 and R134a. Liquid Eq~uilibriausing UNIFAC. Elsevier, Amster-
Since the refrigerant R22 will be phased out in the dam (1977)
not-too-distant future, the RI24-DMAC combina- 8 Jelinek, M., Daltrophe, N. C., Borde, I. Prediction of
tion can thus be used as an alternative working fluid thermophysical properties of CFC-free refrigerants.
Energy Efficiency in Refrigeration and Global Warm-
in absorption cooling machines driven by low ing Impact, International Institute of Refrigeration
potential heat sources for subzero temperatures, at Commission B1/2, Ghent, Belgium(1993) 199-204
which the combination of ammonia-water and 9 Borde, I., Jelinek, M. Absorption heat pumps with
lithium bromide-water cannot be used. organic refrigerant-absorbent. Proc. H. Int.
Workshop on Research Activities on Advanced Heat
Fluid Pairs Pumps, Graz, Austria (1988), Vol. 2,
Acknowledgements 89-101
10 Borde, I., Jelinek, M. Development of absorption
The research was supported by a grant from G.I.F., refrigeration units for cold storage of agricultural
the German Israeli Foundation for Scientific products. Int. J. Refrig. (1987) l0 53-56
Research and Development, and by the Anne 11 Borde, I., Jelinek, M. Evaluation of thermodynamic
properties of binary fluid mixtures. Heat and Mass
Tanenbaum Chair in Engineering Thermodynamics Transfer in Refrigeration and Cryogenics. Hemi-
- incumbent Professor I. Borde. sphere, New York (1987) 381-392