Int. J. Environ. Sci. Technol.

DOI 10.1007/s13762-014-0502-6

ORIGINAL PAPER

An experimental study on charge optimization of a trans-critical

CO2 cycle
C. Aprea • A. Greco • A. Maiorino

Received: 13 September 2013 / Revised: 4 December 2013 / Accepted: 11 January 2014

Ó Islamic Azad University (IAU) 2014

Abstract The phasing out of hydrochlorofluorocarbon overall exergy destruction. At the optimal refrigerant
and chlorofluorocarbon fluids and further environmental charge, the compression losses attain a minimum.
problems arising from new, synthetic working fluids
stimulate a continuously rising interest on natural candi- Keywords CO2  Trans-critical cycle  Refrigerant
dates. The nontoxic and nonflammable CO2 impacts nei- charge  Coefficient of performance  Optimization
ther on ozone depletion nor on global warming if leaked to
the atmosphere. The critical temperature of CO2 (31.1 °C)
is almost ambient and, therefore, it undergoes a trans- Introduction
critical refrigeration cycle. In cooling mode operation, a
trans-critical CO2 system, as compared to conventional air- The traditional refrigerants, i.e., chlorofluorocarbon (CFCs)
conditioners, has a lower performance which, contrary to and hydrochlorofluorocarbon (HCFCs) fluids, were banned
conventional systems, strongly depends on the refrigerant by the Montreal Protocol (Montreal Protocol 1987) because
charge. In this paper, the performance of the trans-critical of their contribution to the disruption of the stratospheric
system was evaluated experimentally under different ozone layer (Molina and Rowland 1974). Indeed, according
refrigerant charge amounts. The influence of charge on to the Kyoto Protocol, most hydrofluorocarbons (HFCs)
coefficient of performance (COP), cooling capacity, com- have large global warming potentials (GWPs) and there-
pression ratio, suction line superheat was analyzed in fore provide a non-negligible, direct contribution to global
detail. The experimental results indicate that the COP of warming when leaked to the atmosphere.
the trans-critical cycle attains a maximum at optimal As a consequence, considerable research effort was
refrigerant charge. By varying the charge the cooling devoted to the development of environmentally safe
capacity attains a maximum, as well, that corresponds to refrigeration systems using natural working fluids (Bhatkar
the optimal charge. In order to understand the effect of et al. 2013). CO2 was considered as a promising alternative
refrigerant charge on the performance of each device of the of HFCs. Indeed, carbon dioxide is a component of air,
plant, an exergetic analysis based on the experimental data obtained from atmosphere itself by fractionation. Thus, it
was carried out. The analysis shows that the exergy flow would have no impact on global warming, apart from for
destroyed in the compressor is one of the major causes of the energy consumption associated with the fractionation
process.
CO2 has many excellent advantages in engineering
applications, such as no toxicity, no inflammability, higher
C. Aprea  A. Maiorino
Dipartimento di Ingegneria Industriale, Università di Salerno, via volumetric capacity that enables compact systems, lower
Ponte Don Melillo, 84084 Fisciano, Salerno, Italy pressure ratio, better heat transfer properties, complete
compatibility with normal lubricants, easy availability,
A. Greco (&)
lower price and no recycling problems (Lorentzen and
DII, Università degli Studi di Napoli Federico II,
P.le Tecchio 80, 80125 Naples, Italy Petterson 1993; Lorentzen 1994, 1995; Liao et al. 2000;
e-mail: [email protected] Bansal 2012).

123
 Int. J. Environ. Sci. Technol.

The critical temperature of CO2 (31.1 °C) is usually by means of an expander. A detailed parametric analysis
lower than typical heat rejection temperatures of air-con- (Robinson and Groll 1998) revealed that COP can increase
ditioning and heat pump systems. This results in a trans- if the expander efficiency is 30 %, but decreases when the
critical vapor compression cycle in lieu of a conventional expander isoentropic efficiency is 60 %.
one in water heating and comfort cooling and heating Heidelck and Kruse (2000) discussed a new expander
(Rebora et al. 2006; Tagliafico et al. 2012). design based on a modified, reciprocating machine. Nickl
Application of a trans-critical CO2 cycle to domestic et al. (2002) suggested a new expander–compressor design
water heating systems (Neksa et al. 1998; Neksa 2002) has that provides a further 10 % increase in COP and a 50 %
advantages over conventional systems in terms of power improvement over the same system with a throttle valve.
consumption and heating efficiency (Hwang et al. 1999, Ma et al. (2002) investigated the performance
2001). The performance of trans-critical CO2 cooling sys- improvement of a CO2 cycle by applying an expander with
tem, however, is lower than that of conventional air-con- a reciprocating, rotary or screw compressor with equivalent
ditioners, due to large expansion losses and high power consumption. Baek et al. (2005a, b: Part I and Part
irreversibility during the gas-cooling process (Aprea and II) reported theoretical and experimental studies on a pis-
Maiorino 2009a, b; Pérez-Garcı́a et al. 2013). Therefore, ton–cylinder work recovery.
many researchers analyzed the performance of the trans- In order to reduce expansion losses, an experimental
critical CO2 refrigeration cycle, in order to identify investigation of the valve dynamics has been carried out by
opportunities to improve the energy efficiency of the sys- Ma et al. (2012). A semi-hermetic reciprocating compres-
tem (Kim et al. 2004). sor was developed with a test system incorporated into the
Among the improvement methods, many researchers compressor performance test rig, with a focus on investi-
analyzed the effect of an internal heat exchanger (IHX) to gating the dynamics of the discharge valves.
reduce throttling losses. Boewe et al. (2001) indicated up to The performance deterioration of a basic, single-stage
25 % COP improvements, as compared to the cycle with- CO2 cycle can be improved using a multistage compressor
out IHX. Mu et al. (2003) showed that larger IHX capacity and intercooling of liquid and vapor refrigerant. Inagaki
can lower the optimum high-side pressure. Chen and Gu et al. (1997) found that, for a two-stage CO2 air-condi-
(2005) derived a practical expression for IHX effectiveness tioning cycle, capacity and COP are improved by 35 and
with a theoretical study and found that the latter is an 20 %, respectively, at moderate ambient temperature and
important factor in achieving high cycle performance with by 10 and 5 %, respectively, at high ambient temperature.
a trans-critical cycle. Aprea and Maiorino (2008) showed Halozan and Rieber (2000) and Kim and Kim (2002)
that using the IHX, the COP is increased by 10 %. Tao studied the performance of an auto-cascade refrigeration
et al. (2010) proposed that the IHX can reduce the throt- cycle using CO2 as a working fluid. Huff et al. (2002)
tling loss of the trans-critical cycle. Xu et al. (2011) compared the performance of a two-stage CO2 cycle with
investigated experimentally the effect of the IHX on a that of a conventional one by means of a simplified model.
trans-critical CO2 ejector system. Experimental results Hwang et al. (2000) carried out experimental studies on
showed that the IHX reduces the ejector contribution to the the performance of a two-stage CO2 cycle consisting in an
system performance. intercooler and a flash tank. Groll et al. (2002) performed a
Cabello et al. (2012) presented an experimental analysis thermodynamic analysis, focusing on the compression ratio
of the energetic performances of a modified single-stage in a trans-critical CO2 cycle with and without an expansion
CO2 trans-critical refrigeration plant with an IHX based on turbine.
vapor injection in suction line. Three different injection Eskandari et al. (2012) proposed a theoretical model of a
points have been evaluated experimentally. The measure- new two-stage multi-intercooling trans-critical CO2
ments showed that the cooling capacity and the COP can be refrigeration cycle with ejector-expansion device. In this
enhanced in 9.81 and 7.01 %, respectively. cycle, the vapor compression line includes two intercool-
Zhang et al. (2013) reported a theoretical study based on ers, the first with external coolant and the second with
a thermodynamic analysis to study the effect of an IHX on cycle refrigerant. The overall performance of the new cycle
the performance of an ejector-expansion trans-critical was compared with that of the conventional ejector
cycle. They found that adding an IHX does not necessarily refrigeration cycle (15.3 % higher) and with that of an IHX
improve the energetic performances in the ejector-expan- ejector refrigeration cycle (19.6 % higher).
sion cycle, contrary to what occurs in a conventional To improve the COP of the cycle, an ejector can be used
throttling valve cycle. instead of a throttling valve to recover some of the kinetic
Further improvements to system performance in reduc- energy of the expansion process. Liu et al. (2002) per-
ing expansion losses are achieved by extracting and mak- formed a thermodynamic analysis of the trans-critical CO2
ing use of the work potentially available from the process vapor compression/ejection hybrid refrigeration cycle. Li

123
Int. J. Environ. Sci. Technol.

and Groll (2005) proposed a theoretical model showing

Electrical
that the COP of a CO2 cycle can be improved by more than air heater
16 % with an ejector.
The performance of heat pumps and refrigeration systems
Oil
with different refrigerant charge amounts was studied in many Separator Gas
experimental and theoretical analyses (ASHRAE 1993; Palm P,T Cooler
2007). Choi et al. (2002) studied the effects of improper
refrigerant charge on the performance of a water-to-water heat P,T P,T
P,T 3
pump using R22 as the refrigerant, with an electronic expan-
sion valve and a capillary tube. In a further study (Choi et al. Auxiliary
2004), they tested the performance of R407C under various Compressor
refrigerant charge and outdoor temperature conditions. Main
Cho et al. (2005) compared experimentally the perfor- Compressor
mance of a CO2 cycle with that of other refrigerant sys- 2’
tems, by varying the refrigerant charge amount. P,T
Corberan et al. (2008) presented the experimental results Flowmeter IHX
regarding the optimization of the charge with a reversible P,T 1’ P,T
water-to-water heat pump using propane as a refrigerant.
Poggi et al. (2008) reported a review of refrigerant charge
studies in refrigerant plants. P,T 3’
The considerable body of literature presented clearly
underlines the problems related to the low-energetic per- BPV
Suction
formance of CO2 systems. Overall optimization of the plant P,T
Buffer
performance is, therefore, mandatory. The performance of 4
the trans-critical CO2 system strongly depends on refrig- Flash
erant charge, much more than that of a conventional air- Tank
conditioner. Surprisingly enough, whereas, in the literature, Evaporator EEV
many papers deal with the influence of charge in traditional
vapor compression plants: just one does in trans-critical 5’
ones. Therefore, the present paper intends to reduce the gap 1
by presenting an ample database of experimental results on
the performance of a CO2 trans-critical cycle at different P,T P,T
refrigerant charges, in order: (1) to indentify the optimum P,T
amount of refrigerant maximizing cycle COP; (2) to eval-
uate the effect on cycle performance of undercharging or Fig. 1 A sketch of the experimental plant
overcharging the plant; (3) to investigate the effect of
refrigerant charge on the exergetic performance of each two single-stage, semi-hermetic, reciprocating compres-
device of the plant. sors, an oil separator, an air gas cooler, a liquid receiver, an
The experimental tests discussed in this study are carried air evaporator, an electronic expansion valve (EEV) and an
out with a prototype R744 system working as a classical electronically regulated back pressure valve (BPV).
split system to cool air in a trans-critical cycle. An auxiliary compressor installed in a by-pass (Elbel
The experimental prototype has been set up at the and Hrnjak 2004) enables the analysis of the ‘‘flash vapor’’
Cooling Techniques Laboratory of the University of Sal- influence. In a trans-critical cycle, the refrigerant at the
erno. The experimental tests have been carried out from inlet of the expansion device is typically a single-phase
June to July 2012. fluid with its high pressure exceeding the critical one. The
refrigerant at the outlet of the expansion device is in a two-
phase condition, provided that the fluid crosses the satu-
Materials and methods rated liquid line during the isenthalpic expansion process.
As a result, some fraction of the refrigerant flow enters the
Experimental plant evaporator in vapor state and thus does not provide a sig-
nificant cooling effect. Because of the flash gas by-pass, the
The experimental plant is described in Fig. 1 (Aprea and evaporator is fed with pure liquid and the vapor by-passes
Maiorino 2009a, b; Aprea et al. 2012). Basically, there are the evaporator through the auxiliary compressor circuit. A

123
 Int. J. Environ. Sci. Technol.

flash tank is used downstream of the back pressure valve. Pressures are measured by piezoelectric sensors whose
The flash tank acts as an accumulator storing refrigerant in current output is recorded directly on the data logger. They
excess for test conditions requiring smaller mass flow rates. were calibrated by the manufacturer in the range of
Furthermore, it also acts as a separator where the liquid 0–100 bar with an accuracy of ±0.8 % F.S. Refrigerant mass
phase accumulates at the bottom, the vapor phase occu- flow rate is measured using a Micro Motion mass flow meter
pying the top. with an accuracy of ±0.2 %. Two-watt transducers are used
The main compressor is a semi-hermetic one, and its to measure the electrical power supplied to the compressors
working pressure range is 15–120 bar. The intermediate with an accuracy of ±0.2 %. A balance is used to measure
pressure is of approximately 65 bar. Thus, the working the refrigerant charge with an accuracy of ±0.2 %.
pressure of the auxiliary, semi-hermetic compressor ranges An analysis was carried out, according to the procedure
from 55 to 120 bar. The nominal compression power for suggested by Moffat (1985), to evaluate the accuracy of the
both compressors is 2.5 kW. experimental data discussed in this paper.
At evaporation temperatures of 5 and of 30 °C at the gas The refrigerant power is evaluated as:
cooler exit, where the pressure is 80 bar, the refrigerating

Q_ ev ¼ m_ r hr;out;ev  hr;in;ev ð1Þ
power is of about 3,000 W.
As discussed in (Aprea and Maiorino 2008), the use of with an uncertainty range of ±1.7–3.5 %.
the IHX in a carbon dioxide refrigeration split system The compression ratio is evaluated as:
improves the coefficient of performance (COP) by 10 %. pin:gc
Therefore, an IHX on the refrigerant line between com- b¼ ð2Þ
pout;ev
pressor suction and the exit of the gas cooler was provided.
The lamination occurs thanks to the back pressure valve with an uncertainty range of ±2.1–2.7 %.
and to the electronic expansion one. Indeed, if a back The COP is evaluated as:
pressure was employed as the only throttling device, it Q_ ev
would not be possible to control the refrigerant mass flow COP ¼ ð3Þ
W_
rate, which is necessary to maintain the balance between
refrigeration power and thermal load. Thus, in order to The uncertainty of W_ is ±0.2 %. The COP uncertainty
regulate the evaporation temperature, an auxiliary circuit range is ±1.8–3.8 %.
can be used to by-pass the back pressure valve. The volumetric efficiency is defined as:
 
The air temperature on the gas cooler is modulated by m_ r vr;suc
some electrical resistances and an air flow driven by a gv ¼ ð4Þ
V_ suc
blower located in a thermally insulated channel. This
simulates variable external conditions, as well. where the displacement rate (V_ suc ) is the volume swept
through by the pistons in their suction strokes per unit time
and is specified by the manufacturer. The uncertainty range
System monitoring of the latter parameter is ±2.3–3.9 %.
Temperature and pressure at key points of the plant are
The plant is fully instrumented, in order to evaluate its continuously monitored, to verify that the apparatus is
performance as a whole, as well as that of each single indeed operating in steady-state conditions.
component. Carbon dioxide pressure and temperature are A software application (ISAAC) has been developed in
measured at the inlet and at the outlet of each component. Labview (Labview ver.7.1) using REFPROP subroutines
The working fluid mass flow rate is monitored at the main (REFRPROP ver.9.1). It provides real-time evaluations of
compressor suction (see Fig. 1). the COP and of the enthalpies at all points of the thermo-
Temperatures are measured by means of four-wire dynamic cycle. Furthermore, it depicts the whole cycle on a
100-X platinum resistance thermometers with an accu- p–h diagram and checks that steady-state conditions actu-
racy of ±0.15 K. The sensors are located outside the ally hold. Figure 2 shows as an example a p–h diagram of
pipe, with a layer of heat transfer compound (aluminum the trans-critical cycle with the IHX.
oxide plus silicon) placed between the sensor and the
pipe to provide good thermal contact. The whole pipe is
insulated with 25-mm-thick flexible insulation tape. The Results and discussion
temperature measurement system is checked against a
sensor positioned in a pocket in a similarly insulated The heater provides a fixed ambient temperature of 30 °C.
pipe. The difference between the two measurements was Tests have been carried out by varying the refrigerant
always \0.3 °C. amount from 5.7 to 9 kg by progressively adding

123
Int. J. Environ. Sci. Technol.

initially increases and then decreases with increasing

refrigerant charge. The optimum refrigerant charge is
0.269.
As the refrigerant charge increases, the system pressure
increases in both the gas cooler and the evaporator. The
pressure ratio decreases with the addition of refrigerant
charge. When the system operates at low refrigerant
charge, i.e., in undercharging conditions, the specific work
of compression increases as a consequence of the ther-
modynamics properties and because the electric motor’s
windings are not properly cooled. As the refrigerant charge
increases beyond its optimal value, COP decreases with the
vaporization enthalpy.
Figure 4 reports the pressures in the gas cooler and in
the evaporator, respectively, as a function of the NC. It is
Fig. 2 A p–h diagram of the trans-critical cycle with the internal heat apparent that increasing the refrigerant charge increases
exchanger both pressures. The pressure increase in the gas cooler
under overcharging conditions (NC [ 0.269) can pose a
refrigerant into the system in 100-g increments. In each security problem and the blowout disk might enter in
test, the opening degree of the back up valve (BPV) is action.
fixed. In Fig. 5 the compression ratio is reported as a function
CO2 refrigerant charge (RC) amount was normalized by of refrigerant charge. The compression ratio decreases with
the maximum refrigerant charge (RCmax), i.e., by the the increasing refrigerant charge. As the refrigerant charge
maximum mass of saturated liquid at a room temperature approaches the optimum value (NC = 0.269), the slope of
of 25 °C: the compression ratio rapidly decreases.
RC RC Figure 6 shows the effect of NC on mass flow rate and
NC ¼ ¼ ð5Þ indicates that the latter increases with refrigerant charge.
RCmax V ql ð25  CÞ
Beyond the optimal value, the increase is quite limited.
where V is the volume of the whole plant (0.0393 m3). Indeed, by increasing the refrigerant charge, the compres-
Figure 3 reports COP as a function of the normal- sion ratio decreases, thus increasing the volumetric effi-
ized refrigerant charge (NC). It can be seen that COP ciency of the compressor.

120

2
undercharging overcharging

100

1.8
Pressure (bar)

80

1.6
Gas cooler
COP

Evaporator
60

1.4

40

1.2

20
0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34
1
0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34 NC optimal value
NC
Fig. 4 Pressures in the evaporator and in the gas cooler as a function
Fig. 3 COP as a function of normalized refrigerant charge of normalized refrigerant charge

123
 Int. J. Environ. Sci. Technol.

3 0.5

undercharging overcharging
undercharging overcharging

2.8 0.4

Vapour quality
Compression ratio

2.6 0.3

2.4 0.2

2.2 0.1

2 0
0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34 0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34

NC optimal value
NC optimal value

Fig. 5 Compression ratio as a function of normalized refrigerant Fig. 7 Vapor quality at the outlet of the back pressure valve as a
charge function of normalized refrigerant charge

28 4

undercharging overcharging
undercharging overcharging
3.5
26
CO2 mass flow rates (g/s)

3
Power (kW)

24

2.5
Cooling power
Compression power
22
2

20 1.5

1
18
0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34
0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34
NC NC optimal value
optimal value

Fig. 6 CO2 mass flow rate as a function of normalized refrigerant Fig. 8 Cooling capacity and compression power as a function of
charge normalized refrigerant charge

at the outlet of this device decreasing the vapor quality at

Figure 7 indicates that vapor quality at the outlet of the the evaporator inlet.
back pressure valve decreases with increasing charge, Figure 8 shows the variations in both cooling capacity
eventually reaching very low values (near the conditions of and compression power with increasing NC. Initially, the
saturated liquid) at overcharging. Indeed, the gas cooler compression power slightly increases and then decreases.
pressure increases with refrigerant charge. The external Indeed, when the system operates in undercharging con-
temperature is fixed, and therefore, the temperature at the ditions, the specific work of compression slightly increases
gas cooler exits too. This leads to a decrease in the enthalpy as a consequence of the corresponding increase in

123
Int. J. Environ. Sci. Technol.

compressor suction pressure. This yields higher specific Table 1 Gas cooler and evaporator pressures, compressor volumetric
volume and mass flow rate. This increase is also because efficiency and temperature at the inlet of the compressor as a function
of refrigerant charge
windings of the electric motor are not properly cooled.
Under overcharging conditions, the power decreases Charge (kg) pev (bar) pgc (bar) gv Tin,cp (°C)
because of the compressor malfunctioning. Indeed, at an
5.7000 29.677 92.000 0.65568 13.230
excessive charge, liquid-phase refrigerant can reach the
5.8000 31.828 92.300 0.66780 12.800
compressor. Furthermore, it should be noted that the
5.9000 32.091 92.100 0.67900 12.300
cooling capacity increases with refrigerant charge, reaching
6.0000 33.357 92.400 0.68000 12.000
a maximum. The maximum in cooling capacity corre-
6.1000 33.514 92.500 0.68800 11.900
sponds to maximum COP. Indeed, by increasing the
6.2000 33.891 93.200 0.69300 11.700
charge, refrigerant mass flow rate and the evaporating
6.3000 35.113 93.400 0.69800 11.500
pressure both increase. For undercharged conditions, the
6.4000 35.434 93.900 0.69900 11.400
cooling capacity increases with increasing charge, due to
6.5000 35.602 93.990 0.70010 11.330
the increase in refrigerant mass flow rate and to the
decrease in vapor quality at the inlet of the evaporator. For 6.6000 35.741 94.000 0.71200 11.000
overcharged conditions, the cooling capacity decreases 6.7000 36.409 94.300 0.72300 10.920
with increasing refrigerant charge, since the effect of the 6.8000 36.744 94.800 0.73100 10.410
increase in evaporating pressure prevails. Indeed, beyond 6.9000 36.962 94.993 0.73600 9.8000
the optimal value, the increase in refrigerant mass flow rate 7.0000 37.698 95.000 0.74400 9.6600
is very limited and the refrigerant enthalpy at the evapo- 7.1000 38.404 95.626 0.75500 9.3300
rator outlet increases with increasing the evaporating 7.2000 39.630 96.300 0.76900 9.1200
pressure. This leads to a reduction in the specific variation 7.3000 39.755 96.208 0.77900 8.7000
of the vaporization enthalpy. 7.4000 40.251 97.005 0.79900 8.5000
Table reports gas cooler pressure, evaporator pressure, 7.5000 40.606 97.850 0.80300 8.3300
compressor volumetric efficiency, temperature at the 7.6000 40.738 98.171 0.80600 8.2200
compressor inlet, as a function of the refrigerant charge. 7.7000 40.868 98.481 0.81100 8.1100
The results reported in Table 1 clearly show that the tem- 7.8000 41.076 98.982 0.81900 8.0200
perature at the inlet of the compressor always decreases 7.9000 41.119 99.093 0.82200 7.9800
increasing refrigerant charge. Indeed, at overcharging 8.0000 41.368 99.300 0.82900 7.8200
conditions, the superheating degree at the compressor inlet 8.1000 41.436 99.460 0.83400 7.7900
is very low and liquid refrigerant might reach the com- 8.2000 41.460 99.520 0.83900 7.5900
pressor. The volumetric efficiency always increases with 8.3000 41.506 99.614 0.84200 7.4300
refrigerant charge. Indeed, by increasing refrigerant charge, 8.4000 41.619 99.928 0.84900 7.4200
the compression ratio decreases thus increasing the com- 8.5000 41.841 100.00 0.85300 7.3800
pressor’s volumetric efficiency. 8.6000 41.785 100.20 0.85500 7.3100
In order to understand the effect of refrigerant charge on 9.0000 42.143 100.30 0.86200 7.2200
the performance of each device, an exergetic analysis
based on experimental data was carried out. This analysis
indicates the total plant irreversibility distribution among

_ des;gc ¼ m_ r exr;in;gc  exr;out;gc  Q_ gc sgc
Ex ð8Þ
the plant components, pinpointing those that contribute
most to overall plant-inefficiency. where the dimensionless exergetic temperature of the air
The exergy flow destroyed in the evaporator is evaluated flow in the gas cooler can be defined as:
as: To

sgc ¼ 1  ð9Þ
_ des;ev ¼ m_ r exr;in;ev  exr;out;ev  Q_ ev jsev j
Ex ð6Þ TTMT;a;gc
where the dimensionless exergetic temperature of the air The exergy flow destroyed in the compressor, neglecting
flow in the evaporator can be defined as: the heat transfer with the environment, is evaluated as:


To _ des;cp ¼ m_ r exr;in;cp  exr;out;cp þ W_
Ex ð10Þ
sev ¼ 1  ð7Þ
TTMT;a;ev
The exergy flow destroyed in each valve (BP and EEV)
The exergy flow destroyed in the gas cooler is evaluated: is evaluated as:

123
 Int. J. Environ. Sci. Technol.

the compressor power consumption increases because the

increase in suction pressure in the compressor results in
higher specific volume and mass flow rate.

Conclusion

The performance of the trans-critical system is more

sensitive to refrigerant charge amount than that of a
conventional air-conditioner. System performance stem is
critically linked to an appropriate mass charge. In the
present paper, the effects of refrigerant charge were
analyzed by means of an experimental prototype, working
as a classical spit system to cool air in a trans-critical
cycle.
The COP of the trans-critical cycle attains a maximum
value corresponding to the optimal refrigerant charge
(7.5 kg).
In order to characterize the effects of refrigerant charge,
a detailed analysis was carried out on the major parameters
affecting cycle performances by varying the charge from
-32 to ?17 % of the optimal value. In this range, plant
Fig. 9 Destroyed energy in the devices of the experimental plant as a COP varies between -21 and -7.5 % of the optimal value.
function of refrigerant charge Therefore, the COP reduced more significantly at under-
charged conditions than at overcharged.

As the refrigerant charge increases, the system pressure
_ des;v ¼ m_ r exr;in;v  exr;out;v
Ex ð11Þ increases in both gas cooler and evaporator, thus decreas-
The exergy flow destroyed in the IHX is evaluated as: ing the compression ratio. The volumetric compressor’s

efficiency decreases with increasing refrigerant mass flow
_ des;ihx ¼ m_ r exr;in1;ihx  exr;out1;ihx
Ex 1

rate. The vapor quality at the inlet of the evaporator
þ m_ r exr;in2;ihx  exr;out2;ihx ð12Þ decreases due to a reduction in the enthalpy at the exit of
the gas cooler. Initially, the compression power slightly
Figure 9 reports the exergy flow destroyed in each
increases and then decreases with increasing charge.
component of the plant as a function of the refrigerant
Indeed, at an excessive charge, liquid refrigerant can reach
charge.
the compressor. The cooling capacity increases with
Figure 9 clearly shows that the exergy loss in the gas
refrigerant charge reaching a maximum and then decreases.
cooler predominates and accounts for more than 30 % of
The maximum in cooling capacity corresponds to the
the total exergy loss. The exergy losses in this device are
maximum COP.
not affected significantly by refrigerant charge.
In order to understand the effect of refrigerant charge on
The exergy flow destroyed in the compressor is one of
the performance of each device of the plant, an exergetic
the major causes of exergy destruction in the plant. The
analysis based on the experimental data was carried out.
refrigerant charge strongly affects the exergy losses in this
This analysis shows that the exergy flow destroyed in the
device. Under both overcharging and undercharging con-
compressor is one of the major causes of exergy destruc-
ditions, the compressor losses amount to approximately
tion in the plant. The refrigerant charge strongly affects the
22 % of the total losses. At the optimal refrigerant charge,
exergy losses in this device. Under both overcharging and
the compression losses are 15 % of the total. When the
undercharging conditions, the compressor losses amount to
system operates at low refrigerant charge, the specific work
about 22 % of the total. At the optimal refrigerant charge,
of compression increases as a consequence of the ther-
the compression losses are 15 % of the total. Therefore, at
modynamic properties and because the electric motor’s
the optimal refrigerant charge, the compression losses
windings are not properly cooled. When the refrigerant
attain a minimum.
charge yielding a correct cooling of the electric motor’s
windings, the compression power attains a minimum. As Acknowledgments The financial support of University of Salerno
the refrigerant charge increases beyond the optimal value, was gratefully acknowledge. (FARB UNIVERSITY OF SALERNO

123
Int. J. Environ. Sci. Technol.

NR. ORSA072593,’’Indagine sperimentale su un impianto frigorifero References

a compressione operante con l’anidride carbonica’’).
Aprea C, Maiorino A (2008) An experimental evaluation of the
transcritical CO2 refrigerator performances using an internal heat
exchanger. Int J Refrig 31:1006–1011
Aprea C, Maiorino A (2009a) Transcritical CO2 refrigerator and sub-
List of symbols
critical R134a refrigerator: a comparison of the experimental
results. Int J Energy Res 33:1040–1047
Symbols Aprea C, Maiorino A (2009b) Heat rejection pressure optimization for
a carbon dioxide split system: an experimental study. Appl
Energy 861:2373–2380
BPV Back pressure valve
Aprea C, Greco A, Maiorino A (2012) An experimental evaluation of
COP Coefficient of performance the greenhouse effect in the substitution of R134a with CO2.
ex Specific exergy (J/kg) Energy 45:753–761
_
Ex Exergy flow (W) ASHRAE Standards 116 (1993), Methods of testing for seasonal
efficiency of unitary air-conditioner and heat pumps. ASHRAE
EEV Electronic expansion valve Standards 116. Atlanta (GA), ASHRAE
h Specific enthalpy (kJ/kg) Baek JS, Groll EA, Lawless PB (2005a) Piston cylinder work
IHX Internal heat exchanger producing expansion device in a carbon dioxide cycle. Part I:
m_ r Refrigerant mass flow rate (kg/s) experimental investigation. Int J Refrig 28:141–151
Baek JS, Groll EA, Lawless PB (2005b) Piston cylinder work
NC Normalized charge producing expansion device in a carbon dioxide cycle. Part I:
Q_ Heat transfer rate (kW) theoretical investigation. Int J Refrig 28:152–164
RC Refrigerant charge (kg) Bansal P (2012) A review-status of CO2 as a low temperature
refrigerant: fundamental and R&D opportunities. Applied Therm
SC Specific charge (kg/kW) Eng 41:18–29
T Temperature (°C, K) Bhatkar VW, Kriplani VM, Awari GK (2013) Alternative refrigerants
To Ambient temperature (°C, K) in vapour compression refrigeration cycle for sustainable
v Specific volume (m3/kg) environment: a review of recent research. Int J Env Sci Technol
10(7):871–880
V_ Volumetric flow rate (m3/s) Boewe D, Bullard C, Yin J, Hrnjak PS (2001) Contribution of internal
W_ Compression power (kW) heat exchanger to transcritical R744 cycle performance. Int J
HVAC&R Res 7(2):155–168
Cabello R, Sànchez D, Patiño J, Llipis R, Torrella E (2012)
Experimental analysis of energy performance of modified
Greek symbols single-stage CO2 transcritical vapour compression cycles based
on vapour injection in the suction line. Appl Therm Eng
b Compression ratio 47:86–94
gv Volumetric efficiency Chen Y, Gu JJ (2005) The optimum high pressure for CO2
transcritical refrigeration systems with an internal heat exchang-
s Dimensionless exergetic temperature ers. Int J Refrig 28(8):1238–1249
Cho C, Ryu Y, Kim H, Kim Y (2005) Effects of refrigerant charge
amount on the performance of a transcritical CO2 heat pump. Int
Subscripts J Refrig 28:1266–1273
Choi JM, Kim YC (2002) The effects of improper charge on the
performance of a heat pump with an electronic expansion valve
a Air and capillary tube. Energy 27:391–404
des Destroyed Choi JM, Kim YC (2004) Influence of expansion device on the
cp Compressor performance of a heat pump using R407C under a range of
charging conditions. Int J Refrig 45:378–384
ev Evaporative Corberán J, Martı́nez IO, Gonzálvez J (2008) Charge optimization
ev Evaporator study of a reversible water-to-water propane heat pump. Int J
gc Gas cooler Refrig 31:716–726
in Inlet Elbel S, Hrnjak P (2004) Flash gas bypass for improving the
performance of transcritical R744 systems that use microchannel
IHX Internal heat exchanger evaporators. J Refrig 7:724–735
max Maximum Eskandari NF, Yavari MA (2012) Performance of a new two-stage
TMT Mean thermodynamic multi intercooling transcritical CO2 ejector refrigeration cycle.
out Outlet Appl Therm Eng 40:202–209
Groll EA, Baek S, Lawless PB (2002) Effect of pressure ratio across
r Refrigerant fluid compressors on the performance of the transcritical CO2 cycle
suc Suction with two-stage compression and intercooling, The International
v Expansion valve Refrigeration and Air Conditioning Conference, Purdue, R11-7

123
 Int. J. Environ. Sci. Technol.

Halozan H, Rieberer R (2000) CO2 as a refrigerant possible Moffat RJ (1985) Using uncertainty analysis in the planning of an
application, The Fourth IIR-Gustav Lorentzen conference on experiment. Trans ASME J Fluids Eng 107:173–178
natural working fluids, Purdue, pp 43–50 Molina MJ, Rowland FS (1974) Stratospheric sink for chlorofluo-
Heidelck R, Kruse H (2000) Expansion machines for carbon dioxide romethanes: chlorine atom catalyzed destruction of ozone.
bases on modified reciprocating machines, The Fourth IIR- Nature 249:810–812
Gustav Lorentzen conference on natural working fluids, West Montreal Protocol on substances that deplete the ozone layer (1987).
Lafayette, pp 455–462 New York, NY
Huff HJ, Hwang Y, Radermacher R (2002) Options for a two-stage Mu JY, Chen JP, Chen ZJ (2003) System design and analysis of the
transcritical carbon dioxide cycle, The Fifth IIR-Gustav Lorent- transcritical carbon dioxide automotive air-conditioning system.
zen conference on natural working fluids, Guangzhou, China, J Zhejiang Univ Sci A 4(3):305–308
pp 143–149 Neksa P (2002) CO2 heat pump systems. Int J Refrig 25:421–427
Hwang Y, Radermacher R (1999) Experimental investigation of the Neksa P, Rekstad H, Zakeri P, Schiefloe P (1998) CO2 heat pump
CO2 refrigeration cycle. ASHRAE Trans 105(1):1219–1227 water heater: characteristics, system design and experimental
Hwang Y, Celik A, Radermacher R (2000) Performance of CO2 results. Int J Refrig 21:172–179
cycles with a two-stage compressor, The Fourth IIR-Gustav Nickl J, Will G, Kraus W, Quack H (2002) Design considerations for
Lorentzen conference on natural working fluids, Purdue, p 105 a second generation CO2-expander, The Fifth IIR-Gustav
Hwang Y, Huff H, Preissner R, Radermacher R (2001) CO2 Lorentzen conference on natural working fluids, Guangzhou,
transcritical cycles for high temperature application, Proc. China, pp 189–196
ASME Int Congr 2001, New York, pp 255–260 Palm B (2007) Refrigeration systems with minimum charge of
Inagaki M, Sasaya H, Osakli Y (1997) Pointing to the future: two refrigerant. Appl Therm Eng 27:1693–1701
stage CO2 compression, heat transfer issues in natural refriger- Pérez-Garcı́a V, Belman-Flores JM, Navarro-Esbrı́ J, Rubio-Maya C
ants. Int Inst Refrig 24:131–140 (2013) Comparative study of transcritical vapour compression
Kim SG, Kim MS (2002) Experiment and simulation on the configurations using CO2 as a refrigeration mode base on
performance of an autocascade refrigeration system using carbon simulation. Appl Therm Eng 51:1038–1046
dioxide as a refrigerant. Int J Refrig 25:1093–1101 Poggi F, Macchi-Tejeda H, Leducq D, Bontemps A (2008) Review:
Kim MH, Petterson J, Bullard CW (2004) Fundamental process and refrigerant charge in refrigerating systems and strategies of
system design issues in CO2 vapor compression systems. Prog charge reduction. Int J Refrig 31:353–370
Energy Combust Sci 30:119–174 Rebora A, Senarega M, Tagliafico LA (2006) Influence of some
Labview ver. 7.1, National Instruments Software design parameters on the thermal performance of domestic
Li D, Groll EA (2005) Transcritical CO2 refrigeration cycle with appliances. Heat Mass Transf 42:803–811
ejector-expansion device. Int J Refrig 28:766–773 REFPROP ver.9.1 (2010) NIST Thermodynamic properties of
Liao SM, Zhao TS, Jakobsen A (2000) A correlation of optimal heat refrigerants and refrigerant mixtures database. US Dept. of
rejection pressures in transcritical carbon dioxide cycles. Appl Commerce, Gaithersburg
Therm Eng 20(9):831–841 Robinson D, Groll E (1998) Efficiencies of transcritical CO2 cycle
Liu JP, Chen JP, Chen ZJ (2002) Thermodynamic analysis on the with and without an expansion turbine. Int J Refrig
transcritical R744 vapor-compression/ejection hybrid refrigera- 21(7):577–589
tion cycle, The Fifth IIR-Gustav Lorentzen conference on natural Tagliafico LA, Scarpa F, Tagliafico G (2012) A compact dynamic
working fluids, Guangzhou, China, pp 184–188 model for household vapour compression refrigerated systems.
Lorentzen G (1994) Revival of carbon dioxide as a refrigerant. Int J Appl Therm Eng 35:1–8
Refrig 17(5):292–301 Tao YB, He YL, Tao WQ, Wu ZG (2010) Experimental study on the
Lorentzen G (1995) The use of natural refrigerants: a complete performance of CO2 residential air-conditioning system with an
solution to the CFC/HCHC predicament. Int J Refrig internal heat exchanger. Energy Convers Manage 51(4):64–70
18(3):190–197 Xu X, Chen G, Tang L, Zhu Z, Liu S (2011) Experimental evaluation
Lorentzen G, Petterson J (1993) A new efficient and environmentally of the effect of an internal heat exchanger on a transcritical CO2
benign system for air-conditioning. Int J Refrig 16(1):4–12 ejector system. J Zhejiang Univ Sci A 12(2):146–153
Ma Y, Wei D, Za S, Li M (2002) Thermodynamic analysis and Zhang Z, Ma Y, Wang H, Li M (2013) Theoretical evaluation on the
compression expander for CO2 transcritical cycle, The Fifth IIR- effect of internal heat exchanger in ejector expansion transcrit-
Gustav Lorentzen conference on natural working fluids, Gu- ical CO2 refrigeration cycle. Appl Therm Eng 50:932–938
angzhou, China, pp 292–297
Ma Y, He Z, Peng X, Xing Z (2012) Experimental investigation of the
discharge valve dynamics in a reciprocating compressor for
trans-critical CO2 refrigeration cycle. Appl Therm Eng 32:13–21

123