i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
available at www.sciencedirect.com
w w w . i i fi i r . o r g
journal homepage: www.elsevier.com/locate/ijrefrig
Experimental evaluation of a cascade refrigeration system
prototype with CO2 and NH3 for freezing process applications
J. Alberto Dopazo, José Fernández-Seara*
Área de Máquinas y Motores Térmicos, E.T.S. de Ingenieros Industriales, University of Vigo, Campus Lagoas-Marcosende No 9, 36310 Vigo,
Spain
article info
abstract
Article history:
A prototype of a cascade refrigeration system using NH3 and CO2 as refrigerants has been
Received 25 August 2009
designed and built. The prototype is used to supply a 9 kW refrigeration capacity horizontal
Received in revised form
plate freezer at an evaporating temperature of 50 C as design conditions. The prototype
14 June 2010
includes a specific control system and a data acq*uisition system. The experimental
Accepted 11 July 2010
evaluation started with the real conditions within the design operating parameters.
Available online 16 July 2010
Subsequently, several tests were performed fixing four CO2 evaporating temperatures (50,
45, 40 and 35 C). At each one of the evaporating temperatures evaluated, the CO2
Keywords:
condensing temperature was varied from 17.5 to 7.5 C and an experimental optimum
Refrigeration system
value of CO2 condensing temperature was determined. The discussions on the experi-
Cascade system
mental results include the influence of the operating parameters on the cascade system’s
Carbon dioxide
performance. In addition, the experimental results are compared with two common double
Ammonia-development
stage refrigeration systems using NH3 as refrigerant.
ª 2010 Elsevier Ltd and IIR. All rights reserved.
Prototype-experiment
Evaluation expérimentale d’un système frigorifique au CO2/
NH3 en cascade utilisé pour les applications congélation
Mots-clés : Système frigorifique ; Système en cascade ; Dioxyde de carbone ; Ammoniac ; Développement ; Prototype ; Expérimentation
1.
Introduction
Currently the refrigeration industry is witnessing important
changes owing to the fact that the synthetic refrigerants,
which have a negative impact on the ozone layer or considered responsible for global warming, have been banned or are
in the process of being outlawed or restricted in the near
future. As better long-term alternatives replace them,
Lorentzen (1994) suggested the use of natural substances such
as ammonia, carbon dioxide and the hydrocarbons as refrigerants, thus causing a renewed interest in this area.
Ammonia is a naturally available old refrigerant used from
the middle of the 1800s in absorption and compression
refrigeration systems. It has excellent thermodynamic and
transport properties as refrigerant, but it brings with it a few
application constraints such as toxicity and flammability. On
the other hand, for applications at temperatures lower than
35 C, the volumetric displacement requirements of the NH3
* Corresponding author. Tel.: þ34 986 812605; fax: þ34 986 811995.
E-mail address: [email protected] (J. Fernández-Seara).
0140-7007/$ e see front matter ª 2010 Elsevier Ltd and IIR. All rights reserved.
doi:10.1016/j.ijrefrig.2010.07.010
258
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
Nomenclature
A
COP
DT
h
m
n
p
Q
RC
T
U
V
v
W
2
area (m )
coefficient of performance
temperature difference (K)
specific enthalpy (kJ$kg1)
mass flow (kg$s1)
compressor’s velocity (rpm)
pressure (kPa)
rate of heat transfer (kW)
compressor pressure ratio (discharge/suction)
temperature (K)
overall heat transfer coefficient (kW m2 K1)
volumetric flow (m3 s1)
specific volume (m3 kg1)
power (kW)
Greek symbol
h
efficiency
r
density (kg m3)
Subscripts
act
actual
atm
atmospheric
compressor works out relatively higher (Pearson, 2001) and
the evaporating pressure of the system reaches values lower
than the ambient pressure which could lead to air leakage into
the system. At the present time, ammonia is the most
commonly natural substance used as refrigerant in low
temperature applications and it has also been proposed in low
power refrigeration and heat pump systems (Palm, 2008).
At the beginning of the last century another natural
substance, CO2, was used as refrigerant. This was replaced in
the middle of the 1900’s by artificial refrigerants presently
banned (Kim et al., 2004). Recently published researches by
Lorentzen (1994) and Pearson (2005) point out that to recommence the employment of CO2 as a refrigerant offers high
potential when used in low temperature refrigeration systems.
Included within the advantages of CO2 is that it is environmentally friendly, non-toxic, non-explosive, easily available
and can be used in refrigeration processes within a wide range
of temperatures (from 50 C). In addition, it is compatible with
the oils commonly used in actual refrigeration systems (Baolian
and Yufeng, 2007). It offers low pressure ratios and low specific
volume values, which when coupled with high pressure levels
allows for reduction in size of refrigeration components
(Cabello et al., 2008). The main disadvantages of CO2 as
a refrigerant are the high work pressures (7.2 MPa at 30 C).
Lorentzen (1994) laid special emphasis on the high potential
of reinitiating the use of CO2 as a refrigerant in transcritical
cycles and in cascade refrigeration systems. With relation to
the first case, experimental results obtained from refrigeration
system prototypes developed by the automotive industry
(Brown et al., 2002; Liu et al., 2005), and in heat pumps and
refrigeration systems for domestic applications (Neksa, 2002;
Hwang and Radermacher, 1999; Hrnjak et al., 2000), showed
similar COP values compared to the COP obtained from
systems that use actual refrigerants. Groll and Kim (2007)
c
condenser, condensation
cas
cascade heat exchanger
carbon dioxide
CO2
CO2/NH3 cascade prototype with CO2 and NH3
dis
discharge
e
evaporator, evaporation
ele
electric
exp
experimental
water
H2O
i
indicated
in
inlet
manuf compressor’s manufacturer
ammonia
NH3
opt
optimum
out
oulet
s
isentropic
sat
saturated
SC
sub-cooling
SH
super-heating
suc
suction
v
volumetric
carried out an extensive compilation of results obtained from
researches developed over the last several years, related to the
use of CO2 in transcritical systems. Regarding the use of CO2 in
cascade refrigeration systems, the pair of CO2eNH3 has been
the centre of attention of much research developed in recent
years, generating high expectations. Eggen and Aflekt (1998),
Pearson and Cable (2003), and Van Riessen (2004) showed
practical examples of the use of cascade refrigeration systems
for cooling in supermarkets. Eggen and Aflekt (1998) focused
their attention on the development of a refrigeration system
prototype for a medium sized supermarket in Norway. Pearson
and Cable (2003) published collected data from a refrigeration
system installed in a local line of Scottish supermarkets
(ASDA). Van Riessen (2004) performed the economical and
technical feasibility of a cooling system to be introduced into
a Dutch supermarket.
On the other hand, Ma et al. (2005) evaluated the performance of a cascade CO2eNH3 cycle using an expander in
substitution for the expansion valve. Lee et al. (2006) analyzed
the optimum condensing temperature in the cascade heat
exchanger. Sawalha et al. (2006) and Likitthammanit (2007)
designed and constructed an experimental cascade refrigeration system with CO2 and NH3, for applications in a medium
size supermarket in Sweden. Belozerov et al. (2007) evaluated
the effect of the evaporating CO2 temperature on the COP of
a cascade system with CO2eNH3. Getu and Bansal (2008)
showed a thermodynamic analysis of a cascade refrigeration
system with CO2eNH3. Dopazo et al. (2009) presented a theoretical analysis of the design and operating conditions, and the
influence of these parameters on the COP system in a cascade
refrigeration system with CO2 and NH3, including the influence
of the compressor’s efficiency. Bingming et al. (2009) reported
experimental data obtained from a cascade refrigeration
system with CO2eNH3 when using screw compressors.
259
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
The main scope of this research was the design,
construction and experimental evaluation of a cascade
refrigeration system prototype using NH3 and CO2 as refrigerants, to supply a horizontal plate freezer with 9 kW of
refrigeration capacity at 50 C of evaporating temperature.
The experimental facility is equipped with a system control to
maintain the operating conditions constant. The prototype
design includes a data acquisition system based on a 16-bits
data acquisition card and a PC which allows real time readings
and storage of the necessary variables to calculate and
analyze the operating parameters that outline the performance of the prototype and it’s main components. The
discussions on the experimental results obtained include the
influence of the operating parameters on the cascade system’s
performance. Finally, the experimental results obtained from
the prototype are compared with two common double stage
refrigeration systems using NH3 as refrigerant, and with the
same operating conditions.
2.
Experimental prototype
The experimental cascade refrigeration system prototype has
been designed to supply a horizontal plate freezer. The
designed refrigeration capacity of the plate freezer is 9 kW, at
50 C of evaporating temperature. The schematic diagram of
the experimental prototype of the cascade system is shown in
Fig. 1, and a real view of the experimental facility can be
appreciated in Fig. 2.
The cascade refrigeration system is made of two single
stage systems connected by a heat exchanger (cascade heat
exchanger). CO2 is used as refrigerant in the low temperature
system and NH3 is the refrigerant for the high temperature
system. In the low temperature system CO2 is evaporated in
a stainless steel horizontal plate freezer. The plate freezer is
overfed with CO2 liquid using an ejector, which also acts as an
expansion device, and allows the liquid coming from the CO2
gaseliquid separator to recirculate. In the plate freezer, the
CO2 at evaporating temperature absorbs the cooling duty (Qe,
CO2) from the cooling tins, commonly used in the food refrigeration industry, and enters the CO2 gaseliquid separator.
The CO2 is suctioned from the gaseliquid separator and
compressed in a semi-hermetic compressor. The discharged
superheated CO2 flows from the compressor unit and enters
into the cascade heat exchanger where it is condensed,
ejecting the heat into the cold refrigerant (NH3), and then
enters into a liquid reservoir tank. From the liquid reservoir,
the flow of CO2 enters into a sub-cooling process through a coil
located inside the gaseliquid separator. The sub-cooled CO2
flows through a valve, and then enters the ejector, in which it
is expanded to evaporating pressure. A secondary flow of
saturated liquid of CO2 coming from the gaseliquid separator,
passes into the ejector and the two flows are mixed inside. The
resultant mixture enters into the plate freezer where it is
evaporated.
In the high temperature system, the flow of NH3 coming
from the gaseliquid separator is compressed in an open type
compressor coupled to an electric motor with a pulley system.
The superheated flow of NH3 enters into the air condenser and
ejects the heat into the atmospheric air. The condensed stream
of NH3 coming from the air condenser is channelled into
a liquid reservoir tank. The NH3 liquid leaves the reservoir tank
Low Temperature system (CO2)
High Temperature system (NH3)
(10)
P01
T01
Air condenser
CO2 separator
T04
(8)
(6)
P02
(4)
M02
CO2
Compressor
(5)
M01
Liquid
reservoir
P07
T07
NH3
separator
(11)
V02
T05
(9)
(2)
Cascade
heat exchanger
P08
(12)
(7)
(13)
Plate freezer
Power controller
230 Vac
P06 T06
(15)
(14)
P04
ECO2
P05
T02
(1)
ENH3
V01
P03
Tin,H20
T03
VH2O
(3)
Liquid reservoir
T Temperature sensor
M Mass flow meter
P Pressure transducer
V Volumetric flow meter
E Ejector
Tout,H20
Fig. 1 e Schematic diagram of the CO2/NH3 cascade refrigeration system.
NH3
Compressor
Frequency
converter
260
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
Fig. 2 e Photograph of the experimental prototype of the cascade refrigeration system with CO2 and NH3 built.
and flows through a valve and enters into the NH3 ejector,
where it is expanded until it reaches evaporating pressure and
then mixed with a secondary flows of liquid NH3 coming from
a gaseliquid separator. The resulting flow mixture enters into
the cascade heat exchanger, in which it evaporates and
returns to the gaseliquid separator. Employing the method
used for the low temperature system, the evaporation process
of the NH3 was also designed using liquid overfeed.
The main components used in the experimental facility
and their characteristic parameters are shown in Table 1.
The experimental cascade refrigeration system prototype
was equipped with a data acquisition system based on a 16bits data acquisition card and a PC. In the low temperature
system with CO2, the suction temperature and the discharge
temperature of the CO2 compressor unit are measured by
using two sensors located in the refrigerant line close to each
compressor (T01 and T02, in Fig. 1). The condensed CO2
temperature is measured by using the sensor T03, (refer Fig. 1),
located in the CO2 liquid reservoir tank. The CO2 sub-cooled
liquid temperature is measured by using the sensor T04
located in the outlet sub-cooler line. The suction pressure of
the CO2 system is measured by using the pressure transducer
P01 located on the gaseliquid separator. The discharge
pressure of the CO2 compressor is measured by using the
pressure transducer P02 installed in the discharge line, close
to the compressor. In addition, the CO2 condenser pressure is
measured by using the pressure transducer P03 (refer Fig. 1).
The pressure at the CO2 ejector inlet is measured by using the
pressure transducer P04. The main mass flow of CO2 is
measured in the outlet sub-cooling line, by using the mass
flow-meter M01, whilst the secondary mass flow of CO2 is
measured by using the mass flow-meter M02, located at the
auxiliary outlet of the gaseliquid separator (refer Fig. 1).
In the high temperature system with NH3, the suction and
discharge temperatures are measured by using the sensors
T05 and T06, located on the suction and discharge lines of the
NH3 compressor, respectively (refer Fig. 1). The NH3 condenser
temperature is measured by using the sensor T07, located in
the NH3 liquid reservoir tank. The suction pressure is
measured by using a pressure transducer located in the
gaseliquid separator (P05). The discharge pressure of the NH3
compressor is measured by using the pressure transducer P06,
located on the discharge line of the NH3 compressor. The
condensing pressure of NH3 is measured by using the pressure
transducer P07, installed in the liquid reservoir tank of NH3.
The pressure at the NH3 ejector inlet is measured by the
Table 1 e Main components of the experimental prototype of cascade refrigeration system.
Component
CO2 compressor
Plate freezer
Cascade exchanger
NH3 compressor
Air condenser
Characteristics
Bitzer, model: 2EC-6.2 K. Displacement with 1450 min1:11.4 m3 h1, No. of cylinders: 2, weight: 82 kg, max.
power consumption: 6.3 kW.
Tucal, plate material: stainless steel, plate dimensions: 1550 1200 mm, plate numbers: 5.
Alfa Laval, model: ANHP76-50H. Max power: 35 kW, heat exchanger surface 4.9 m2. Material: ALLOY. 50 P.
Bitzer, model: W4PA. Displacement with 1450 min1: 47.14 m3 h1, No. of cylinders: 4, weight: 77 kg,
Goedhar, model INAL-G TYPE. Max power: 44 kW, heat exchanger surface: 109 m2, material Stainless steeleAluminium.
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
pressure transducer P08. The main volumetric flow of NH3 is
measured by using the volumetric flow-meter V01, (refer
Fig. 1). The secondary flow of NH3, used to overfeed the NH3
evaporator, is measured by using the volumetric flow-meter
V02, located at the auxiliary outlet of the NH3 gaseliquid
separator (Fig. 1). The NH3 compressor is cooled by circulating
water through it. The volumetric flow rate of the cooling water
is measured by using the volumetric flow-meter VH2O. The
inlet and outlet cooling water temperatures are measured by
using the sensors Tin,H2O and Tout,H2O, respectively (see Fig. 1).
All the temperature sensors used in the experimental
facility are A Pt100 inserted in 3 mm diameter stainless steel
pockets. The pressure transducers are WIKA ECO 1 type, with
an accuracy of 0.5% of the full scale (40 bar). The refrigerant
mass flows of the CO2 system are measured by using Coriolis
flow-meters, with an accuracy of 0.25% of the measured
value. The volumetric flows of the NH3 system are measured
by using electromagnetic flow-meters, with an accuracy of
0.25% of the measured value. The electric power of each
compressor is measured by using watt-meters, with an
accuracy of 2% of the measured value.
The operating control of the cascade prototype is made
using a PLC. The controlled variables are: the condensing
pressures of NH3, the evaporating pressure of NH3 and the
evaporating pressure of CO2. The condensing pressure of NH3
is controlled by modifying the velocity of the fans of the air
condenser. There are three options: fans off, fans at velocity 1
(less velocity) and fans at velocity 2 (high velocity). To control
the evaporating pressure of CO2, a flow line which connects the
CO2 compressor discharge line to its suction line was installed,
thus allowing the recirculation of a small portion of the mass
flow of compressed CO2. The recirculation of the mass flow of
CO2 is controlled using a valve installed in the flow line
previously mentioned. The evaporating pressure of NH3 is
controlled by varying the capacity of the high temperature
system; in the same manner, three values of capacity for this
compressor are available: 100% of the capacity, 50% of the
capacity and compressor when off (0% of capacity). In addition,
the experimental facility was fitted with a frequency
converter, Danfoss VLT 2800, which permits the modification
of the electric motor’s velocity, and consequently the gradual
variation of the NH3 compressor’s capacity.
The mass flow of the liquid refrigerant is controlled by
means of the valves located at the ejectors inlet lines of each
system.
Any time the experimental prototype becomes non functional, the pressure of the low temperature system will be
controlled using an auxiliary refrigeration unit. This auxiliary
unit will cool the CO2 of the low temperature system with
a coil located in the gaseliquid CO2 separator.
3.
Experimental procedure and data
reduction
3.1.
Experimental procedure
The main object of this work was focused on the experimental
evaluation of the performance of the facility designed and
constructed, under stationary flow conditions.
261
In order to keep the refrigeration charge constant, two
immersion electric resistances were installed inside four (4)
aluminium tins similar to those commonly used in the local
refrigerated fish industry, and filled with potable water. The
maximum electric power of each resistance is 1500 kW,
regulated independently using electric power regulators. Each
tin was equipped with one A Pt-100 type temperature sensor,
inserted in 3 mm diameter stainless steel pockets, located at
the tin’s geometrical centre. The sensors permit real time
readings of the water temperature inside each tin and it’s
fluctuations during the test.
The experimental evaluation of the prototype was initiated
fixing the evaporating temperature at 50 C (design condition). The capacity control of each compressor was established at 100%, the fans velocity fixed at “velocity 1” (less
velocity), and the valves positioned to maintain the liquid
levels constant in the liquid reservoir tanks.
Subsequently, the performance of the experimental
prototype was evaluated at several CO2 condensing temperatures, keeping constant the evaporating temperature of CO2,
the condensing temperature of NH3, the super-heating of the
gas refrigerant suctioned by each compressor, and the
cascade heat exchanger temperature difference. The operating conditions of the prototype during the experimental
evaluation carried out are shown in Table 2.
The CO2 evaporating temperature was fixed by controlling
the electric resistances installed in the tins. The NH3 evaporating pressure was established by using the frequency
converter. The saturated liquid temperature (or sub-cooler
temperature) of each refrigerant of the prototype was verified
comparing the measured values against the saturation
temperatures, calculated at the pressures values measured in
each system’s liquid reservoir tanks. The flow of refrigerant in
each system was fixed by using the valves located before
reaching each ejector, keeping the liquid levels constant in the
liquid reservoirs tanks. The level of liquid in each reservoir
tank was checked by direct observation, by means of a device
installed exclusively for this purpose. Once all temperatures,
pressures and flows measurements were stabilized, the
experimental data was collected during a time period of 270 s,
at intervals of approximately 5.4 s.
3.2.
Data reduction
During the experimental evaluation of the cascade refrigeration system, several operating parameters for each of the
main components of the facility, as well as the low and high
temperature systems and global COP were calculated from the
data experimentally measured. The calculations from the
previously mentioned parameters were formulated from the
mass and energy balances applied to each one of the main
prototype components. Thermodynamic properties of CO2
and NH3 were obtained from the Refprop Database (Lemmon
et al., 2004). The calculated parameters are detailed as
follows, according to Fig. 1.
The pressure ratios of the CO2 and NH3 compressors are
established by Eqs. (1) and (2).
RCCO2 ¼ P02 =P01
(1)
RCNH3 ¼ P06 =P05
(2)
262
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
Table 2 e Operating conditions of the prototype during the experimental evaluation.
Te,CO2 ( C)
DTSH,CO2
( C)
Tc,CO2 ( C)
DTSC,CO2
( C)
Te,NH3 ( C)
DTSH, NH3
( C)
Tc,NH3 ( C)
15
15
15
15
7.5, 10, 12.5, 15
7.5, 10, 12.5, 15, 17.5
7.5, 10, 12.5, 15, 17.5
10, 12.5, 15, 17.5
0
0
0
0
11, 13.5, 16, 18.5
11, 13.5, 16, 18.5, 21
11, 13.5, 16, 18.5, 21
13.5, 16, 18.5,21
15
15
15
15
30
30
30
30
35
40
45
50
The volumetric efficiencies of the CO2 and NH3 compressors
are calculated in Eqs. (3) and (4).
hv;CO2 ¼ ðV1 $3600Þ=11:4
hv;NH3
(3)
(4)
The values 11.4 and 47.14 (m3 h1) correspond to the theoretical volumetric flows of the CO2 and NH3 compressors, respectively (Table 1). The suction volumetric flow rates of CO2 (V1) and
NH3 (V9) are determined by means of Eqs. (5) and (6), respectively. In Eqs. (5) and (6), the CO2 mass flow rate M01 and the NH3
volumetric flow rate V01 are experimentally measured and the
vapour specific volumes v1 and v9 are obtained as a function of
the pressure and temperature experimental measurements at
the suctions of the CO2 and NH3 compressors, P01 and T01, and P05
and T05, respectively. In Eq. (6), the mass density r11 is obtained
as a function of the experimental temperature and pressure
measurements, P07 and T07, according to Fig. 1.
V1 ¼ M01 $v1
(5)
V9 ¼ V01 $r11 $v9
(6)
The indicated efficiency of each compressor is obtained in
Eqs. (7)e(9).
hi ¼ ws =wact
(7)
ws ¼ hout;s hin
(8)
wact
( C)
5
5
5
5
(10)
The cooling capacity of the plate freezer is obtained in Eq.
(11). In Eq. (11), the CO2 mass flow rate (M01) is measured in the
experimental facility, h1,sat is obtained as a function of
experimental measurement of P01, and h3 is obtained from the
experimental measurements of P03 and T03, according to the
nomenclature in Fig. 1.
Qe;CO2 ¼ M01 $ðh1;sat h3 Þ
(11)
The condensing capacity in the cascade heat exchanger
with CO2 (the low temperature system) and the evaporating
capacity of the high temperature system with NH3 are calculated in Eqs. (12) and (13), respectively. In these equations h2 is
calculated as a function of T02 and P02, h3 is determined from
P03 and T03, h11 is obtained from P07 and T07 and h9,sat is
calculated as a function of P05.
Qc;CO2 ¼ M01 $ðh2 h3 Þ
(12)
Qe;NH3 ¼ V01 $r11 $ðh9;sat h11 Þ
(13)
The overall heat transfer coefficient is obtained in Eq. (14)
and the reference temperature difference in the cascade
exchanger is established in Eq. (15).
UAcas ¼ Qc;CO2 = Tc;CO2 Te;NH3
¼ ðhout hin Þ þ QH2 O m
NH3
compressor, the heat transfer was assumed zero because it
does not possess its own refrigeration system, and for the NH3
compressor, the heat transfer is calculated in Eq. (10).
QH2 O ¼ VH2 O $rH2 O $ hout;H2 O hin;H2 O
ðV9 $3600Þ 1450
¼
$
47:14
n
DTSC,
(14)
(9)
DTcas ¼ Tc;CO2 Te;NH3
In Eq. (9) QH2O represents the heat transferred by each
compressor’s own system of refrigeration. For the CO2
(15)
The condensation capacity of the air condenser is calculated in Eq. (16).
Table 3 e Experimental prototype cascade refrigeration system at design operating conditions.
Parameter
Te,CO2
Tc,CO2
Pe,CO2
Pc,CO2
DTSH,CO2
DTSC,CO2
Qe,CO2
Qc,CO2
Wele,CO2
hv,CO2
hi,CO2
mCO2
Value
Parameter
Value
50 C
17.48 C
682 kPa
2127 kPa
15 C
0.34 C
9.45 kW
13.2 kW
3.93 kW
0.667
0.579
124.4 kg h1
DTcas
UAcas
nNH3
fNH3
RCCO2
RCNH3
mH2O
Tin,H2O
Tout,H2O
COPCO2
COPNH3
COPCO2/NH3
3.48 C
3.87 kW K1
1401 rpm
50 Hz
3.12
6.36
900 kg h1
26.2 C
28.9 C
2.40
2.14
0.92
Parameter
Te,NH3
Tc,NH3
Pe,NH3
Pc,NH3
DTSH, NH3
DTSC, NH3
Qe, NH3
Qc, NH3
Wele, NH3
hv,NH3
hi,NH3
VNH3
Value
20.96 C
29.72 C
182 kPa
1158 kPa
15 C
4.74 C
13.5 kW
17.2 kW
6.32 kW
0.672
0.588
1.23 l min1
263
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
3.0
Qc;NH3 ¼ V01 $r11 $ðh10 h11 Þ
(16)
COPCO2 =NH3 ¼ Qe;CO2 = Wele;CO2 þ Wele;NH3
2.5
2.0
COP
The cascade refrigeration system COP, the COP of the high
temperature system with NH3, and the COP of low temperature system with CO2 are obtained in Eqs. (17)e(19),
respectively.
1.5
1.0
(17)
0.5
COPNH3 ¼ Qe;NH3 =Wele;NH3
(18)
0.0
-18
COPCO2 ¼ Qe;CO2 =Wele;CO2
4.
Results and discussion
4.1.
Experimental results
In Table 3 the experimental mean values of the measured and
calculated parameters at the design operating conditions are
presented. It can be seen from the operating conditions which
have been evaluated, a refrigeration capacity of 9.45 kW was
obtained. It is 5% higher than the value initially established as
the design condition (9 kW). In addition, the high value
obtained for the NH3 compressor pressure ratio, 6.36, stands
out when compared to the value obtained for the CO2
compressor pressure ratio, 3.12. In the same way, the electric
power experimentally measured in the NH3 compressor was
60.8% higher than the electric power measured in the CO2
compressor. However, the difference between the condensing
capacity of CO2 and the evaporating capacity of NH3 in the
cascade exchanger was only 2.2%, with a measured reference
0.76
-15
-14
-13
T c,CO2 (°C)
-12
-11
-10
temperature difference of 3.48 C, and an overall heat transfer
coefficient (UAcas) of 3.87 kW K1. The compressor velocity
experimentally measured was 1401 r.p.m., at 50 Hz of
frequency, fixed experimentally by means of a frequency
converter. A variation of 3.4% when compared to the manufacturer’s data shown in Table 1 was appreciated.
In Fig. 3, the results of the volumetric efficiency of each one
of the prototype compressors, obtained from experimental
data, are shown. Compressor volumetric efficiencies calculated from manufacturer data are included. In both cases it
can be appreciated that there is a broad consensus between
the experimental data and the manufacturer’s data. The
maximum difference obtained for the mean values of the
volumetric efficiencies was 2.6% for the NH3 compressor and
0.7% for the CO2 compressor.
Figs. 4e6 depict the results obtained from the operating
conditions shown in Table 2.
Fig. 4 shows the experimental results of COPNH3, COPCO2
and COPCO2/NH3 versus the CO2 condensing temperature,
obtained at the CO2 evaporating temperature of 50 C. It can
be seen, that as the CO2 condensing temperature increases the
COP of the high temperature system also increases. In the case
of the low temperature system with CO2, the reverse trend is
30
ηV,CO2,manuf
ηV,NH3,manuf
25
0.74
Qe,CO2
Qc,CO2
Wele,CO2
Qe,NH3
Qc,NH3
Wele,NH3
-17
-15
Te,CO2 = -50 °C
Tc,NH3 = 30 °C
DTcas = 3,5 °C
20
(k W)
0.72
ηV
-16
Fig. 4 e COPCO2, COPNH3 and COPCO2/NH3, as a function of
CO2 condensing temperature, determined from
experimental data.
0.80
ηV,CO2,exp
ηV,NH3,exp
-17
(19)
On the other hand, a detailed analysis of the experimental
uncertainties was carried out according to ISO (1995). This
analysis revealed that the maximum for typical uncertainties
was estimated to be 0.35% for Qe,CO2, 0.43% for Qc,CO2, 1.84% for
Qe,NH3, 1.83% for Qc,NH3, 2.02% for COPCO2, 2.72% for COPNH3
and 1.67% for COPCO2/NH3. Error bars have been included in
Figs. 4e6, showing the estimated typical uncertainties for
each case.
0.78
T e,CO2 = -50 °C
T c,NH3 = 30 °C
DTcas = 3,5 °C
COPCO2
COPNH3
COPCO2/NH3
0.70
15
0.68
10
0.66
0.64
5
0.62
0.60
0
50
100
150
200
250
Time (s)
Fig. 3 e Volumetric efficiencies experimentally measured
and manufacturer’s data of the cascade system
compressors.
0
-18
-16
-14
-13
T c,CO2 (°C)
-12
-11
-10
Fig. 5 e Experimental values of the heat flow rates and the
electric power consumptions versus the CO2 condensing
temperature.
264
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
1.75
Te,CO2
COP
1.50
-35
-40
-45
-50
1.25
1.00
0.75
-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9
-8
-7
-6
-5
Tc,CO2 (°C)
Fig. 6 e COPCO2/NH3 as a function of the CO2 condensing
temperature at several CO2 evaporating temperatures.
observed. These trends are expected, since, as the CO2
condensing temperature increases, the pressure ratio of the
CO2 compressor increases and the pressure ratio of the NH3
compressor decreases. The increase of the pressure ratio in
the CO2 compressor leads to an increase of electric power in
the CO2 compressor and, as a result, the COPCO2 tends to
decrease, while in the system with NH3, the decrease of the
pressure ratio leads to a lowering of electric power in the NH3
compressor and consequently the COPNH3 tends to increase.
In the case of the prototype’s overall COP, the maximum
experimental value obtained was 0.98. This value was reached
at the CO2 condensing temperature of e15 C, approximately.
The results obtained for the CO2 with condensing temperatures
different to 15 C, show reduced COPCO2/NH3. This trend
coincides with the theoretical analysis presented by Lee et al.
(2006), Getu and Bansal (2008) and Dopazo et al. (2009). In the
cited researches, the influence of the CO2 condensing temperature on the COP of cascade refrigeration systems using CO2
and NH3 was analyzed. The outcome of the analysis indicates
that there exists a CO2 optimum condensing temperature that
defines the maximum COP. In this case, for the operating
conditions that were fixed, the experimental results shows that
the CO2 optimum condensing temperature is approximately
15 C, and a maximum COPCO2/NH3 of 0.98 was also obtained.
Fig. 5 depicts the experimental results obtained for the
refrigeration capacity of the prototype (Qe,CO2), the CO2
condensing capacity (Qc,CO2) and the NH3 evaporating capacity
(Qe,NH3) in the cascade exchanger, the NH3 condensing capacity
in the air condenser (Qc,NH3) and the consumption of electric
power measured in each one of the prototype compressors.
The operating conditions evaluated are those previously cited.
The tendency, described previously, regarding electric power
consumption in each compressor of the prototype can be seen
in this figure. In the case of the NH3 compressor, the amount of
electric power fell from 6.32 kW to 4.53 kW, when the CO2
condensing temperature rose from 17.5 C to 10 C. For the
CO2 compressor, the opposite tendency is observed, when the
consumption of electric power is increased from 3.93 kW to
4.43 kW.
The refrigeration capacity of the prototype measured
during the experimental evaluation diminishes from 9.45 kW,
for Tc,CO2 at 17.5 C, to 8.07 kW, for Tc,CO2 at 10 C. When the
optimum value of Tc,CO2 was fixed the refrigeration capacity
observed was 9.31 kW, which is 3.4% higher than the design
condition (9 kW).
The NH3 condenser capacity shows similar behaviour to
that observed in the electric power consumption of the NH3
compressor. In this case, the value drops from 17.17 kW to
14.93 kW. The experimental results obtained for the
condensing and evaporating capacities calculated in the
cascade exchanger show an acceptable energetic balance in
all the cases which were evaluated. The maximum absolute
unbalance measured was 0.58 kW (4.39% of Qc,CO2), and it was
obtained at Tc,CO2 at 12.5 C.
4.2.
Experimental determination of the optimum CO2
condensing temperature. Comparison with theoretical model
results
Fig. 6 shows the experimental results obtained from the constructed cascade refrigeration system prototype for several
values of CO2 evaporating temperatures, with variations of the
CO2 condenser temperature. The operating conditions of the
prototype for the different operating points evaluated are
shown in Table 2. Fig. 6 shows a tendency similar to that
depicted in Fig. 4 for the COPCO2/NH3 with the CO2 evaporating
temperature at 50 C. When the CO2’s condensing temperature increases, the COPCO2/NH3 also increases to a certain
maximum, and from that point the trends change to
decreasing. The previous commentary refers to all of the cases
which were experimentally evaluated. In each case, as the CO2
evaporating temperature increases, the optimum value for the
CO2 condensing temperature measurement increases, from
15 C (with Te,CO2 ¼ 50 C) to 10 C (with Te,CO2 ¼ 35 C).
This increasing trend of the optimum CO2 condensing
temperature was explained by Lee et al. (2006), Getu and
Bansal (2008) and Dopazo et al. (2009). In those researches,
theoretical correlations were proposed to determine the
optimum CO2 condensing temperature in a cascade
Table 4 e Experimental measurements of the optimum CO2 condensing temperature compared to theoretical values given
by published correlations.
Experimental
Te,CO2 ( C)
35
40
45
50
Lee et al. (2006)
Getu and Bansal (2008)
Dopazo et al. (2009)
Tc,CO2 (K)
Tc,CO2 (K)
Diff. (%)
Tc,CO2 (K)
Diff. (%)
Tc,CO2 (K)
Diff. (%)
257.2
259.7
262.7
263.7
254.7
256.7
258.7
260.7
1.0
1.2
1.5
1.1
253.4
255.7
258.0
260.3
1.5
1.5
1.8
1.3
263.4
265.4
267.4
269.3
2.4
2.2
1.8
2.2
265
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
Table 5 e Operating conditions considered for comparison of the CO2/NH3 cascade system with the NH3 two-stage
refrigeration systems.
Suction super-heating
degree ( C)
Condensing temperature ( C)
Liquid sub-cooling
degree ( C)
15
30
5
35, 40, 45 and 50
refrigeration system similar to the one experimentally evaluated in this research. The correlations are given in Eqs. (20)e
(22). In these correlations, Tc, Te and DTcas correspond to the
NH3 condensing temperature, the CO2 evaporating temperature and the reference temperature difference in the cascade
exchanger, respectively. Additionally in Eq. (21), the superheat degree in the evaporator inlet (DTSH) and the sub-cooling
degree in the condenser outlet (DTSC) are included (temperatures in K):
Tc;CO2 ;opt ¼ 40:63 þ 0:4$Tc þ 0:4$Te þ Tcas
(20)
Tc;CO2 ;opt ¼ 7:0992 þ 0:2662$Tc þ 0:4602$Te þ 0:5736$DTcas
þ 0:0215,DTSH þ 0:2945$DTSC
(21)
Tc;CO2 ;opt ¼ 218:78 þ 0:39064$Tc þ 0:3965$Te þ 0:60747$DTcas
(22)
The optimum CO2 condensing temperatures determined in
Eqs. (20), (21) and (22) are shown in Table 4, and the results are
compared with the experimental data measurements. In the
first two cases, the correlations underestimate the value of the
optimum CO2 condensing temperature. The maximum
difference obtained in Eq. (20) was 1.5%, while the maximum
difference observed using Eq. (21) was 1.8%. From the results
given by Eq. (22) it can be appreciated that the optimum CO2
condensing temperature was over-estimated in all of the
cases which were evaluated. In this latter, the maximum
difference, when compared to the experimental result was
2.4%. In this last research, the influence of the compressor’s
indicated efficiency on the theoretical estimated optimum
CO2 condensing temperature was analyzed. Differences of up
to 8.1% were observed using correlations taken from available
technical literature to calculate the indicated efficiencies of
the CO2 and NH3 compressors, which could explain the
differences observed within the theoretical correlations
which were analyzed: Lee et al. (2006) calculate the isentropic
efficiencies of each compressor as a function of its pressure
ratios using second order equations, Getu and Bansal (2008)
assume an isentropic efficiency constant and equal to 0.78,
and Dopazo et al. (2009) estimate the indicated efficiency of
the compressor as a function of the pressure ratio using first
order equations.
In Fig. 6 the experimental results obtained for the COPCO2/NH3
are presented. It can be observed that the COPCO2/NH3 increases
when increasing the CO2 evaporating temperature, whilst
keeping constant the CO2 condensing temperature. The
minimum value of COPCO2/NH3 observed during the experimental evaluation was 0.9 (Te,CO2 ¼ 50 C and Tc,CO2 ¼ 10 C),
and the maximum COPCO2/NH3 was 1.59 (Te,CO2 ¼ 35 C and
Tc,CO2 ¼ 10 C).
4.3.
Cascade refrigeration system with CO2/NH3 vs.
common double stage refrigeration systems with NH3
The experimental results obtained from the cascade refrigeration system prototype evaluation were compared with
a system with two compression stages with flash tank economizer and with a system with two compression stages with
intercooling in a flash tank (intercooler flash tank), using NH3
as refrigerant in both cases. The operating conditions
considered to carry out the analysis are shown in Table 5. In
the cascade system, the COP corresponds to the maximum
value obtained from the experimental tests carried out on
each one of the CO2 evaporating temperatures considered.
The intermediate pressure of the double stage compression
system with NH3 was calculated taking into account that the
pressure ratios of each stage were equal. The details of the
NH3 compressor efficiencies taken from the experimental
prototype were used in the analysis of the common NH3
systems. The pressure drops in the heat exchangers and
connecting lines, as well as the heat losses to the environment
were neglected.
Fig. 7 shows the COP values of the three systems compared
against different evaporating temperatures in the systems.
Also shown are the schemes of the common NH3 refrigeration
systems which were analyzed. It can be seen that the system’s
COP increased when the system’s evaporating temperature
was increased. It can also be appreciated that the NH3 double
stage system with economizer showed the lowest COP within
the system’s evaporating temperature evaluations. In the
range of evaporating temperatures between 50 C and
40 C, the higher COPs were obtained with the CO2/NH3
cascade system. The greatest difference was observed when
the evaporating temperature was at 50 C, 6.5% higher than
2.00
COPCO2/NH3
1.75
1.50
COP
Evaporating
temperature ( C)
COPNH3,double-stage with economizer
1
COPNH3,double-stage with intercooling 2
1.25
Systems with NH 3
Wlow
1.00
Whigh
0.75
Wlow
Qc
Whigh
Qc
Qe
Qe
0.50
Double-stage with economizer
0.25
-52
-50
-48
-46
-44
-42
-40
Double-stage with intercooling
-38
-36
-34
Te (°C)
Fig. 7 e COP of the refrigeration systems analyzed.
Schematic diagrams of the two common double stage NH3
refrigeration systems considered are included.
266
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
V NH3
V CO2
Pe,NH3
Pe,CO2
100
80
40
20
100
0
10
Patm
-55
-50
-45
-40
-35
1
P (bar)
V (m3/h)
60
0
-30
Te (°C)
Fig. 8 e Evaporating pressure and volumetric flows at the
compressors inlet as a function of the system evaporating
temperature.
the NH3 double stage system with intercooler and 19.5%
higher than the NH3 double stage system with economizer.
At a 35 C evaporating temperature a higher COP was
obtained using the NH3 double stage compression system
with intercooler, 6.3% higher than the cascade system using
CO2/NH3 and 6.9% higher than the NH3 double stage system
with economizer. From these results, it can be inferred that for
a refrigeration process at 40 C evaporating temperature or
lower, the use of cascade system with CO2eNH3 provides
a better performance than the common systems which were
analyzed with NH3 as refrigerant. Otherwise, at an evaporating temperature of 35 C or lower, the use of a NH3
common options system performs more advantageously
when compared to the CO2eNH3 cascade system.
Fig. 8 depicts the evaporating pressure of the refrigerants
(CO2 and NH3) as a function of the evaporating temperature.
Also included is the volumetric flow measured at the suction
line of the CO2 compressor and the volumetric flow considered
at the suction line of the low-stage compressor of the NH3
typical double stage refrigeration, of the systems analyzed. In
the first case, it can be appreciated that, within the evaporating
temperature range considered, the NH3 typical systems
present evaporating pressures lower than the ambient pressure, which could lead to air leakage into the system. Otherwise, the experimental measurements of evaporating
pressures within the cascade system with CO2eNH3 were
higher than the ambient pressure. Regarding the volumetric
flow, the differences observed between the cases compared
stand out considerably. The highest values of the volumetric
flow where always observed for the NH3 systems, with values
of up to 840% higher than the cascade system (72.5 m3 h1 vs.
7.7 m3 h1, Te ¼ 50 C).
5.
Conclusions
This paper has dealt with the design, construction and
experimental evaluation in stationary conditions of an
experimental prototype of a cascade refrigeration system with
CO2 and NH3, to supply a horizontal plate freezer of 9 kW of
nominal refrigeration capacity at 50 C of evaporating
temperature.
The results obtained from the experimental evaluation
carried out at the designed operating conditions were satisfactory, obtaining an experimentally measured refrigeration
capacity of 9.45 kW, 5.0% above the plate freezer‘s designed
refrigeration capacity.
In addition, an analysis of the influence of the CO2 evaporating and condensing temperatures on the overall system’s
COP was carried out based on experimental measurements.
The results confirm the tendencies obtained from theoretical
studies published in available technical literature. As such,
the experimental results showed an optimum CO2 condensing
temperature in each and every one of the CO2 evaporating
temperature values considered. The optimum CO2
condensing temperatures experimentally measured were
compared to the values of the optimum CO2 condensing
temperatures given by several correlations published by
different authors. The resulting maximum difference was
2.4%.
Likewise, the COP of the cascade refrigeration system
experimentally measured was compared to the COP obtained
from two common double stage NH3 refrigeration systems:
a system with two compression stages with flash tank economizer, and a system with two compression stages with
intercoling in a flash tank (intercooler flash tank), using NH3 as
refrigerant in both cases. It can be appreciated that for evaporating temperatures lower than 35 C, the systems with
NH3 as refrigerant present evaporating pressures lower than
the ambient pressure, while the evaporating pressures
obtained from the cascade system were always higher than
ambient pressure. From the analysis of the results taken from
the overall COPs of the compared refrigeration systems, at
40 C evaporating temperature or lower, it would appear that
the option of the cascade system with a COP values of up to
19.5% higher, is the most advantageous.
On the other hand, when the CO2 volumetric flow
measured at the suction line of the CO2 compressor of the
experimental prototype was compared to the NH3 volumetric
flow determined at the suction line of the low-stage
compressor of the common NH3 refrigeration systems
analyzed, the results showed that the volumetric flow of NH3
was always higher, reaching differences of up to 840% at
50 C evaporating temperature.
Finally, the facility provided the experimental measurements of the volumetric efficiency of each one of the
compressors installed in the prototype. The experimental
results were compared to the manufacturer’s data. A broad
consensus of agreement was found between the values
measured and those given by the manufacturer, with differences of less than 3%.
Acknowledgements
The authors would like to acknowledge the technical support
provided by the KINARCA S.A.u. Company in designing and
building the prototype. The prototype has been built and
tested at the KINARCA facilities in Vigo (Spain). The authors
also acknowledge the financial support received from the
“Xunta de Galicia” (Project PGIDIT05DPI006E).
i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 4 ( 2 0 1 1 ) 2 5 7 e2 6 7
references
Baolian, N., Yufeng, Z., 2007. Experimental study of the
refrigeration cycle performance for the R744/R290 mixture.
Int. J. Refrigeration 30, 37e42.
Belozerov, G., Mednikova, N., Pytchenko, V., Serova, E., 2007. Cascade
type refrigeration systems working on CO2/NH3 for technological
process of products freezing and storage. In: IIR Ammonia
Refrigeration Conference, Ohrid, Republic of Macedonia.
Bingming, W., Huagen, W., Jianfeng, L., Ziwen, X., 2009.
Experimental investigation on the performance of CO2/NH3
cascade refrigeration system with twin-screw compressor.
Int. J. Refrigeration. doi:10.1016/j.ijrefrig.2009.03.008.
Brown, J.S., Yana-Motta, S.F., Domanski, P.A., 2002. Comparative
analysis of an automotive air conditioning systems operating
with CO2 and R134a. Int. J. Refrigeration 25, 19e32.
Cabello, R., Sanchez, D., Llopis, R., Torrilla, E., 2008. Experimental
evaluation of the energy efficiency of a CO2 refrigerating plant
working in transcritical conditions. Appl. Therm. Eng. 28,
1596e1604.
Dopazo, J.A., Fernandez-Seara, J., Sieres, J., Uhı́a, J., 2009.
Theoretical analysis of a CO2eNH3 cascade refrigeration
system for cooling applications at low temperatures. Appl.
Therm. Eng. 29 (8e9), 1577e1583.
Eggen, G., Aflekt, K., 1998. Commercial refrigeration with
ammonia and CO2 as working fluids. In: Proceedings of the
Third IIR: Gustav Lorentzen Conference on Natural Working
Fluids, Oslo, Norway, pp. 281e292.
Getu, H., Bansal, P., 2008. Thermodynamic analysis of an R744-R717
cascade refrigeration system. Int. J. Refrigeration 31, 45e54.
Groll, E.A., Kim, J.H., 2007. Review of recent advances toward
transcritical CO2 cycle technology. HVAC&R Res. 13 (3), 499e520.
Hrnjak, P., Richeter, M., Song, S., Kim, M., Bullard, C., 2000.
Transcritical CO2 heat pump for residential application. In:
Fourth IIR-Gustav Lorentzen Conference on Natural Working
Fluids at Purdue, pp. 9e16.
Hwang, Y., Radermacher, R., 1999. Experimental investigation of
the CO2 refrigeration cycle. ASHRAE Trans. 105 (1), 1219e1227.
ISO, 1995. Guide to the Expression of Uncertainty in
Measurements. International Organization for
Standardization (ISO), 9e27 pp.
267
Kim, M., Petterson, J., Bullard, C., 2004. Fundamental process and
system design issues in CO2 vapor compression systems. Prog.
Energy Combust. Sci. 30, 119e174.
Lee, T., Liu, C., Chen, T., 2006. Thermodynmic analysis of optimal
condensing temperature of cascade-condenser in CO2/NH3
cascade refrigeration systems. Int. J. Refrigeration 29 (7),
1100e1108.
Lemmon, E.W., McLinden, M.O., Huber, M.L., 2004. Reference
Fluid Thermodynamic and Transport Properties (REFPROP),
Version 7.0. National Institute of Standards and Technology
(NIST).
Likitthammanit, M., 2007. Experimental Investigation of CO2/NH3
Cascade and Transcritical CO2Refrigeration Systems in
Supermarkets. Dissertation, KTH Scholl of Energy and
Environmental Technology, Stockholm, Sweden.
Liu, W., Chen, J., Chen, Z., 2005. Experimental investigation of
a CO2 automotive air conditioner. Int. J. Refrigeration 28,
1293e1301.
Lorentzen, G., 1994. Revival of carbon dioxide as a refrigerant. Int.
J. Refrigeration 17, 292e301.
Ma, Y.T., Ning, J.H., Chen, Q., Ma, L.R., 2005. Study on replacement
and energy-saving of natural refrigerant CO2/NH3 cascade
refrigeration cycle. In: International Conference on
Compressor and Refrigeration, Dalian, China.
Neksa, P., 2002. CO2-heat pump systems. Int. J. Refrigeration 25,
421e427.
Palm, B., 2008. Ammonia in low capacity refrigeration and heat
pump systems. Int. J. Refrigeration 31, 709e715.
Pearson, A., 2001. New developments in industrial refrigeration.
ASHRAE J. 43, 54e58.
Pearson, A., 2005. Carbon dioxide e new uses for an old
refrigerant. Int. J. Refrigeration 28, 1140e1148.
Pearson, A., Cable, P., 2003. A distribution warehouse with CO2 as
refrigerant. In: Proceedings of the international Congress of
Refrigeration, Washington, DC, USA.
Sawalha, S., Suleymani, A., Rogstam, J., 2006. CO2 in Supermarket
Refrigeration. CO2 Project Report Phase I. KTH Energy
Technology.
Van Riessen, G.J., 2004. CO2/NH3 Supermarket Refrigeration
System with CO2 in the Cooling and Freezing Section. TNO
Environment, Energy and Process Innovation, Apeldoorn,
Netherlands. Available from: http://www.energie.nl/nel/
nl05e1210.html.