Energy Conversion and Management: X 20 (2023) 100407

Contents lists available at ScienceDirect

Energy Conversion and Management: X

journal homepage: www.sciencedirect.com/journal/energy-conversion-and-management-x

New zeotropic CO2-based refrigerant mixtures for cascade

high-temperature heat pump to reach heat sink temperature up to 180 ◦ C
P. Ganesan *, Trygve M. Eikevik
Process and Power Research Group, Department of Energy and Process Engineering, Norwegian University of Science and Technology (NTNU), Trondheim, Norway

A R T I C L E I N F O A B S T R A C T

Keywords: There is a great demand for clean, high-temperature heating in industries. The present work demonstrates the
Cascade system modeling and analysis of a high-temperature heat pump (HTHP) using a two-stage cascade refrigeration system
High-temperature heat pump with newly developed refrigerant mixtures. Environment-friendly CO2/butane and CO2/pentane mixtures were
Natural refrigerants
introduced in the Low-stage (LS) and High-stage (HS) cycles, respectively. The model was built in MATLAB and
Zeotropic mixtures
MATLAB
explored the possibilities of high-temperature heating. A heating source temperature of 10 to 50 ◦ C was
considered as the system’s input. The HTHP was designed to deliver hot water at a temperature above 115 ◦ C.
The performance of the system was investigated in terms of the heat sink and hot water temperatures, tem­
perature lift, heating capacity, LS, HS, and total system’s coefficient of performance (COP), volumetric and
isentropic efficiencies, and discussed. The analysis indicated that the system could deliver high-temperature
water with a temperature lift of 70 ◦ C. Mixtures of 5% CO2/ 95% butane and 5% CO2/ 95% pentane exhibit
excellent thermo-physical properties making them suitable for high-temperature heating applications with high
COP. For instance, such mixtures obtain the system’s COP and heating capacity of 3.6 and 201.5 kW, respectively
for the heat sink temperature of 116.7 ◦ C. It shows an improvement of 20% compared to pure fluids. Also, the
heat sink temperature could reach up to 181 ◦ C when the CO2 composition in the refrigerant mixtures is 20%.
The volumetric (0.72–0.87) and isentropic (0.61–0.71) efficiencies of the compressor were studied and
displayed.

when it is operating at a low source temperature and is capable of

1. Introduction supplying high temperatures [8,9]. Many industrial, commercial, and
urban applications are proposing the use of HTHP as a feasible tech­
The energy crisis and greenhouse gas emissions are growing nology for decarbonization [10,11]. The HTHPs with an output tem­
increasingly serious [1]. It is important to find new ways to generate and perature above 100 ◦ C attract more interest in industrial heat retrieval
effectively use energy [2,3]. Space heating and many industries like food and have remarkable potential for development [12,13].
and oil need high-temperature water [4]. Some applications need more An analysis of the performance characteristics of a water-to-water
hot water to carry out the processes when the environmental tempera­ cascade HP designed to provide heating and hot water simultaneously
ture is low, accounting for a huge amount of energy consumption [5]. during the winter season was conducted based on variations in sec­
Vapor compression heat pump (VCHP) has been a valuable technology ondary fluid temperature. Reported that the system was able to provide
for heating and cooling and is widely adopted. A heat pump system better energy efficiency and a much more stable high-water heating
typically includes single-stage, multi-stage, cascade, and parallel heat potential [14]. The thermodynamic performance of the CHTHP was
pumps. There are some limitations to single-stage heat pumps, such as investigated. Under different heat source inlet temperatures, a mathe­
their high-pressure ratio and their lower COP at high water outlet matical model has been developed considering energy, efficiency, and
temperatures, and their temperature promotion is typically limited to the environment. Based on simulation results, the proposed system
50 ◦ C [6]. Alternatively, the cascade HTHP (CHTHP) has been taking showed good stability and feasibility under various operating conditions
broad attention from researchers due to its comparatively high COP, the [15].
flexibility of working fluids, and significant temperature lift [7]. Cascade HTHP is an attractive option to upgrade waste heat to a high tem­
HTHP has better overall efficiency compared to a single-stage HTHP perature for industrial applications. To achieve such a high temperature,

* Corresponding author.
E-mail addresses: pganesanmech@gmail.com, ganesan.palanichamy@ntnu.no (P. Ganesan).

https://doi.org/10.1016/j.ecmx.2023.100407
Received 11 May 2023; Received in revised form 9 June 2023; Accepted 10 June 2023
Available online 13 June 2023
2590-1745/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Nomenclature PR Pressure ratio

PRHS Pressure ratio at high-stage
COP Coefficient of performance PRLS Pressure ratio at low-stage
COP Total Total Coefficient of performance Q Heating/cooling capacity (kW)
COPHS High-stage coefficient of performance QC Heating capacity (kW)
COPLS Low-stage coefficient of performance Qe Cooling capacity (kW)
Cp hw Specific heat capacity of hot water (kJ/kgk) QHS High-stage heating/cooling capacity (kW)
Cpw Specific heat capacity of water (kJ/kgk) QLS Low-stage heating/cooling capacity (kW)
h Enthalpy (kJ/kg) Thw, in Hot water inlet temperature (◦ C)
HS High-stage Thw, out Hot water outlet temperature (◦ C)
LS Low-stage Tlift Temperature lift (◦ C)
ṁ Mass flow rate of the refrigerant (kg/s) Tsource, in Heat source inlet temperature (◦ C)
ṁHS Mass flow rate of the refrigerant at high-stage cycle (kg/s) Tsource, out Heat source outlet temperature (◦ C)
ṁLS Mass flow rate of the refrigerant at low-stage cycle (kg/s) Vs Stroke volume (capacity of the compressor) (m3/s)
mw Mass flow rate of water (kg/s) W Power consumption of compressor (kW)
Ƞvol Volumetric efficiency of the compressor (%) WHS Power consumption of high-stage compressor (kW)
PC Condenser pressure (heat sink) (bar) WLS Power consumption of low-stage compressor (kW)
Pe Evaporation pressure (bar) WTotal Total Power consumption of compressors (kW)
PHSE Pressure at high-stage evaporator (bar) X Pressure ratio
PLSC Pressure at the low-stage condenser (bar) ρ Density of refrigerant at the compressor suction (kg/m3)

the selection of suitable refrigerants for the HTHP is crucial [16]. A 152a/600 and R-1233zd(E)/161 resulted in up to a 14% COP rise over
cascade HTHP using low GWP refrigerants in both HS and LS configu­ the baseline [10]. Four parallel mixtures with suitable composition and
rations was investigated. The authors found that based on the COP, R245fa were explored for different high-temperature conditions [24].
isobutane, and butane appear to be the most convenient working LS HTHP was designed using two parallel compressors and two condensers
fluids. The peak performance was obtained with pentane and 1336mzz using ammonia as the refrigerant to generate hot water at 96 ◦ C and
(Z) in the HS cycle. Interestingly, pure natural fluids such as butane, investigated. Reported that the system was able to provide a tempera­
pentane, and isobutane performed well compared to other fluids. Also ture lift of 60 ◦ C with a COP of 3.56 [25]. A simulation model was
indicated that the use of pentane/butane achieved maximum COP [17]. developed for a cascade HTHP and analyzed using R134a and CO2 as
A cascade HP system designed for use in high-temperature heating working fluids. Further, the simulation results were compared with the
systems has been studied thermodynamically. Using different working experimental results. The simulation and experimental results were in
medium combinations in the cycles of the cascade HP has been shown to good agreement. The results indicate that the COP improved with an
affect the COP of the system. An analysis was performed on the influence increase in heat source temperature. The maximum COP of the system
of the compressor’s subcooling, superheating, and isentropic efficiency was 3.07 for a hot water supply of above 50 ◦ C [26].
on COP [18]. The working fluid component is inevitably explored during the
Water source cascade heat pumps were established, which could development of zeotropic mixtures, which has been happening at an
provide heat at 170 ◦ C at high temperatures using BY3B/R245fa, BY3A/ unprecedented pace in recent years [27]. Zeotropic mixtures have
R245fa, and BY3B/BY6 and found that heating capacity, input power, higher heat transfer resistance than a weighted mean of the resultant
COP, and discharge pressure were in good agreement [19]. An experi­ resistances of their pure components due to higher heat transfer re­
mental investigation on a CHTHP was carried out using BY-3(A&B) and sistances in the vapor and liquid phases [28]. The key desirable of the
R245fa on the LS and HS cycles. The results proved to heat the water to R744/natural refrigerant blends are associated with their GWP, ODP,
142 ◦ C with a COP of 1.66 [6]. Thermodynamic analysis of high- and flammability properties lower than those of pure fluids [29]. The
temperature heat pumps with low global warming refrigerants such as high-side pressure can be reduced when pure CO2 is mixed with low-
R1233zd(E), R1336mzz(Z), and R601 was carried out and compared GWP working fluids. Heat pump water heaters were evaluated using
with R245fa. Results indicate that these refrigerants have better energy CO2 blends with 10 low-GWP working fluids. A study was conducted to
efficiency and reduced environmental effects compared to R245fa [20]. determine how discharge pressure, compression ratio, hot-water outlet
It has been proposed to use BY-3 as the working fluid in the LS refrig­ temperature, and chilled water inlet temperature impact the heat
erant cycle of a CHTHP and R245fa as the working fluid in the HS pump’s COP. As compared to a pure CO2 cycle, CO2/R41 and CO2/R32
refrigerant cycle. Simulated results indicate that the HTCHP system exhibit high COP and low high-side pressure, making them suitable for
using BY-3 and R245fa can produce hot water at 142 ◦ C and achieve a heat pump water heaters. The high COP of the system was a result of
100 ◦ C temperature lift [21]. HTHP which can produce heat at 150 ◦ C great thermal matching in the heat exchangers [30].
from a source temperature of 100 ◦ C or below was designed to operate When R290 was added to CO2 with a small fraction, the system
with low GWP refrigerants [22]. In a cascade heat pump, nine potential performances of heat pumps using mixtures of CO2 and R290 were
refrigerants were investigated including R600, R245a, and R601, to experimentally examined and compared with those of pure CO2. Based
achieve a heat sink temperature of 90 ◦ c from the evaporator tempera­ on the results, heat pump systems based on CO2-based mixtures could be
ture of 20 ◦ c. The highest COP of 3.08 was achieved with R601 in the LS designed and operated in a more efficient manner [31]. Three different
cycle and R245fa in the HS cycle of the cascade system [23]. combinations of mixed refrigerants were evaluated in a cascade refrig­
Based on 25 ◦ C and 35 ◦ C water flow, the work evaluates a semi- eration system that employs two cycles of vapor compression: R744 +
empirical two-stage cascade cycle for HTHP applications. Two proto­ R1270, R744 + R717, and R744 + RE170. Reduced compressor power
type single-stage HP with different temperature lifts (R-1234ze(E) and and increased COP were observed compared to pure fluids [32].
R-1336mzz(Z)) were used as baseline inputs. To evaluate its energy Research has been conducted to analyze cascade HPs, but they
performance, various novel mixtures were included for both stages in usually focus on low temperatures, while very little information is
the proposed two-stage cascade HTHP model. For the LS and HS, R- available about high water outlet temperatures. The researchers are

2
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

heating.

2. Modeling of cascade high-temperature heat pump

The system used, refrigerants under consideration, operating con­

ditions, algorithm flow chart, and the equations used for modeling and
optimization of cascade HTHP are presented in detail in this section.

2.1. System description

This section presents an explanation of the system and the refrigerant

mixtures considered for the study. A schematic diagram of the proposed
system is presented in Fig. 1 and the corresponding pressure-enthalpy
and temperature-entropy diagrams with all state points indicated are
presented in Fig. 2. A two-stage cascade vapor compression refrigeration
system is intended to deliver hot water at a temperature of more than
115 ◦ C for space heating purposes. The system is formed by connecting
two basic vapor compression cycles. LS condenser and HS evaporator
form the cascade heat exchanger. Water is used as the secondary fluid to
act as the heating source for the evaporator. The Cascade system allows
a higher temperature lift when compared with a single-stage system. The
pressure ratio can be easily controlled using two compressors. As this
study focuses on an HTHP, maintaining the system stability is one of the
important aspects. Hence, the cascade system was chosen for this study.
It also helps improve performance in terms of pressure, temperature,
and compressor efficiencies.
Fig. 3 presents the temperature vs the heat transfer of the high-
temperature heat pump system for the new refrigerant mixtures
considered. The temperature of the refrigerant mixtures is indicated in
Fig. 1. Schematic diagram of the Cascade High-Temperature Heat Pump. black color. The heat source temperature line is indicated in green color,
and it exchanges the heat with the LS evaporator. The heat source
working on different kinds of refrigerants to identify the suitable re­ temperature (7–50 ◦ C) is higher than the LS evaporator (2.5 to 42.5 ◦ C)
frigerants/mixtures for high temperature applications. The previous to exchange the heat from the source to the evaporator. There is no cross
studies indicate that mostly Hydrofluorocarbon (HFC), and Hydro Fluro- in temperature, so heat exchange is not reversed. The degree of super­
Olefins (HFO) based refrigerants/mixtures were explored and these are heat in the LS evaporator is 5 ◦ C. In the cascade heat exchanger, the
potential threats to the environment. In some cases, CO2 and hydro­ degree of superheating and subcooling are 5 ◦ C. The temperature level of
carbons were investigated. The past investigations revealed that, for the HS evaporator is maintained lower than that of the LS condenser to
higher heating temperatures, the COP was very low and vice-versa. exchange the heat. It can be noted that there is no cross in temperature
Hence, exploring suitable environment-friendly refrigerants or mix­ in the cascade heat exchanger and it helps the heat exchange in the
tures with high performance for heating temperatures 180–200 ◦ C and proper way as it should from the LS condenser to the HS evaporator and
above is challenging. Considering the above facts, this work tries to ensured that there is no reversed heat transfer. The HS condenser ex­
address the challenges of high-temperature heating. The natural zeo­ changes the heat with the heat sink fluid (water). The heat sink fluid
tropic mixtures are superior to pure fluids in terms of heat source temperature is indicated in red color. The heat sink temperature of the
matching, and environmental protection, and exhibit excellent ther­ system was in the range of 119 to 181.5 ◦ C. Again, there is no temper­
modynamic performance. The thermodynamic application of zeotropic ature cross between the heat sink fluid and the refrigerant mixtures of
mixtures is not as widespread as that of pure fluids, and a concise the HS condenser and the HS condenser rejects the heat to the water
summary of heat pump systems is lacking. Zeotropic mixtures have a key which can be further transferred to the point of heating applications.
property called temperature glide, which could contribute to the The hot water delivery temperature was in the range of 61 to 124 ◦ C
increased energy efficiency of the system. In this work, modeling, and with a maximum temperature lift of 74 ◦ C.
analysis of an HTHP using a two-stage cascade refrigeration system were
proposed to deliver hot water above 115 ◦ C. The model of the CHTTP 2.2. Screening of refrigerants and consideration
was established in MATLAB. A uniqueness of this study was that natural
zeotropic mixtures were used in both the LS and HS cycles of the cascade The choice of refrigerant is one of the foremost imperative compo­
system for the first time in the literature to the best of the author’s nents in deciding the secure, reliable, and performing operation of the
knowledge. CO2/butane and CO2/pentane mixtures were introduced for HTHP. Potential refrigerants were screened to select suitable re­
the first time as refrigerants in the system’s LS and HS cycles. The main frigerants for HTHP applications. The selection was based on many
objective was to explore and optimize both the high-temperature oper­ criteria such as critical temperature, environment friendliness, and the
ation and COP of the system. The system’s volumetric and isentropic performance of the refrigerants. Thermos-physical properties of the re­
efficiencies were also modeled and analyzed. Most of the researchers fix frigerants were analyzed using REFPROP 9.1 [33]. After an assessment,
the compressor efficiencies as 0.75 during the simulation, however, in it was decided to use natural refrigerant mixtures as the availability
practice, it varies. In this article, Compressor efficiencies were also number of natural refrigerants is very limited. Moreover, refrigerant
simulated with respect to the pressure ratio of the heat pump systems. mixtures have an important property called temperature glide. Tem­
This investigation is the need of the hour in cold climate countries where perature glide refers to the difference between dew point and bubble
the high temperatures heating for residential heating and industrial point temperatures. These mixtures are called zeotropic mixtures. The
process heating is dominant. It could be a step towards zero-carbon condensation and evaporation processes are no more isothermal in this
case. In recent times, carbon dioxide is becoming one of the most

3
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 2. Ph and TS diagrams.

attractive refrigerants. Also, it was found that butane and pentane re­ high critical temperature were suitable for this study. The main interest
frigerants were suitable for high-temperature applications because of of this study is to achieve a temperature lift of 70 ◦ C with higher per­
the high critical temperatures. Hence, this study proposed and intro­ formance of the system. So, it was decided to investigate low CO2 mass
duced a combination of carbon dioxide, butane, and pentane. The use of fractions of up to 20% starting from 5% and high butane (80–95%) and
CO2 + butane and CO2 + pentane was proposed in the LS and HS of the pentane (80–95%) mass fractions to balance both temperature lift and
system, respectively. The properties of the proposed refrigerant mixtures performance of the system as presented in Table 1.
were collected from the REFPROP version 9.1 [33].
The saturation temperature vs CO2 mass fraction diagram for the
proposed mixtures is presented in Fig. 4. It can be noted that the level of 2.3. Operating conditions
saturation temperature is higher for CO2 + pentane than for CO2 +
butane. The bubble and dew points of the mixtures are also mentioned. Heat source and sink temperatures are the key parameters for the
The temperature glide of the proposed mixtures over the CO2 mass specific application of heat pumps. The heat source was the input
fraction percentage is presented in Fig. 5. The temperature glide in­ parameter of the entire system that influences the system’s performance.
creases with an increase in CO2 and drops from 30%. In the end, it The supply and outlet temperatures of the heating source (water) for the
reaches 100 percent CO2 where the glide reaches zero. CO2 mass fraction evaporator were in the range of 10–50 ◦ C and 7–39 ◦ C, respectively. The
percentage vs critical temperature is shown in Fig. 6. The critical tem­ temperature lift of the cascade system was in the range of 70 ◦ C. The
perature reduces for both mixtures as the CO2 mass fraction increases. It inlet and outlet temperatures of a secondary fluid (water) in the HS
was found that the Zeotropic mixtures with low-temperature glide and condenser (heat sink) were in the range of 51–71 ◦ C and 65–124 ◦ C,
respectively. Subcooling and superheating in both stages were

4
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 3. Temperature vs heat transfer of the system.

Fig. 5. CO2 mass fraction vs Temperature glide.

Fig. 4. CO2 mass fraction vs Saturation temperature.
2.4. Equations for system modelling
considered for the operation of the system. All the details of the simu­
lation model are listed in Table 2 below. The following assumptions The equations used for the calculation of the mass flow (ṁ)(Eqn. (1);
were made for developing and analyzing the model: power consumption of the LS (WLS) (Eqn. (2); and HS compressors (WHS)
(Eqn. (3); and total power consumption (WTotal) (Eqn. (4) of the com­
• LS condenser exchanges the same heat capacity with the HS evapo­ pressors are presented below [25]:
rator in the cascade heat exchanger.
ṁ = ρ × (Vs /3600) × ηvol (1)
• Dry saturated refrigerant is available at the outlet of the evaporator.
• Wet saturated refrigerant is available at the outlet of the condenser. WLS = ṁLS × (h3 − h2 ) (2)
• Expansion process is isenthalpic.
• The pressure drops and heat losses were neglected. WHS = ṁLS × (h9 − h8 ) (3)
• The evaporator and condensers are counterflow heat exchangers.

5
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 6. CO2 mass fraction-Critical temperature.

Table 1
Refrigerant mixtures with different mass fractions under consideration.
Case Low-stage cycle(CO2 (R744) + High-stage cycle(CO2 (R744) +
Butane(R600)) Pentane (R601))

1 5% + 95% 5% + 95%
2 10% + 90% 10% + 90%
3 15% + 85% 15% + 85%
4 20% + 80% 20% + 80%

Table 2
Simulation Parameters.
Parameter Value

Heat source temperature (Tsource, in) 10–50 ◦ C

Heat sink temperature (Tsink) >115 ◦ C up to 181.5 ◦ C
Hot water outlet temperatures (Thw, out) >100 ◦ C up to 124 ◦ C
Subcooling degree in both stages 5 ◦C
Superheating degree in both stages 5 ◦C
Temperature lift (Tlift) 60–70 ◦ C
Hot water inlet at the heat sink (Thw, in) 51–71 ◦ C

WTotal = WLS + WHS (4)

The equations used for the calculation of the heating capacity of the
cascade heat exchanger (QLS = QHS) (Eqn. (5); the heating capacity of
the HS condenser (QC) (Eqn. (6); the cooling capacity (Qe) (Eqn. (7) are
presented below [25]:
QLS = QHS = ṁLS × (h3 − h5 ) = ṁLS × (h8 − h12 ) (5)

QC = ṁHS × (h9 − h11 ) (6)

Qe = ṁLS × (h2 − h6 ) (7) Fig. 7. Flowchart of MATLAB Simulation procedure.

The equations used for the calculation of the hot water outlet (Thw,
out)(Eqn. (8) and the heat source outlet (Tsource, out) (Eqn. (9) tempera­ COPLS = QLS /WLS (10)
tures are presented below [25]:
( ) COPHS = QC /WHS (11)
Thw,out = Thw,in + Qc /Cphw × mhw (8)
( ) COPTotal = QC /WTotal (12)
Tsource,out = Tsource,in + Qc /Cp × mw (9)
The equations used for the calculation of LS (PRLS) (Eqn.13); and HS
The equations used for the calculation of low stage (COPLS)(Eqn. (PRHS) (Eqn.14) pressure ratios are listed below:
(10); HS (COPHS)(Eqn. (11); and total COP (COPTotal)(Eqn. (12) of the
PRLS = PLSC /Pe (13)
system are presented below [25]:

6
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

50
48
46
44
42
40
38
36
34

Tsource ( C)
32
30
28
26
24
22
20
18
16
14
12 20% CO2 15% CO2
10 10% CO2 5% CO2

10 30 50 70 90 110
Thot water ( C)

Fig. 8b. Result of heating source inlet temperature on Hot water outlet
temperature.

Fig. 8a. Result of heating source inlet temperature on Heat sink temperature.

PRHS = PC /PHSE (14)

The isentropic (Ƞisen) (Eqn.15); and volumetric (Ƞvol) (Eqn.16) effi­
ciencies were calculated using the LS and HS pressure ratios (x) and the
equations below. We developed these equations based on the tests and
measurements conducted at our lab to calculate the efficiencies of the
compressor.

ηisen = − 0.00000461x6 + 0.00027131x5 − 0.00628605x4 +

(15)
0.07370258x3 − 0.46054399x2 + 1.40653347x − 0.87811477

ηvol = 0.0011x2 − 0.0487x + 0.9979 (16)

2.5. Algorithm flowchart

The cascade HTHP model was established in MATLAB - R2021b [34].

The flow chart of the algorithm used for modeling is presented in Fig. 7.
The algorithm starts with the inputs of guessed values of initial, sub­
cooling, superheating values, and heat source temperatures. It calculates Fig. 8c. Result of heating source inlet temperature on Heat Capacity.
the condensation and evaporation temperatures and then it calculates
the LMTD of all the three heat exchangers in the system. If the LMTD of temperature has risen with the increase in heat source temperature. It
all three heat exchangers is not optimized, then the initial temperatures can be noted that the heat sink temperature has shown a similar increase
will be adjusted until it reaches the required limit. If it is within the limit, in trend with the enhanced source temperature for all the cases 5% (case
then it proceeds to calculate the parameters such as the power con­ 1), 10% (case 2), 15% (case 3), and 20% (case 4) CO2 mass fraction
sumption of the individual compressors as well as the total power percentages of the refrigerant mixtures). In case 1, for the lowest source
consumption. temperature of 10 ◦ C, the heat sink temperature can be reached up to
It also estimates the isentropic and volumetric efficiencies of the 119 ◦ C. A maximum of 132 ◦ C heat sink temperature was observed for
compressors, heating capacities, and COPs. Finally, it provides all the the source temperature of 50 ◦ C. In case 4, heat sink temperatures of 167
output results and plots the required graphs. Then, the algorithm ends. and 181.5 ◦ C were reached for the source temperature of 10 and 50 ◦ C
respectively. A 15.25% boost in heat sink temperature was found for
3. Results and discussion case 2 compared with case 1. Similarly, the heat sink temperatures
increased by 8.89% and 5.58% for cases 3 and 4 respectively compared
3.1. Effect of heat source inlet temperature on the system performance to the previous cases. It is perceptible that the heat source temperature
has influenced the heat sink temperature to a great level.
In this work, water was used as the heating source to provide heat to Fig. 8b presents the relation between the heat source temperature
the LS evaporator. This section studied and presented the effect of heat and the hot water temperature delivered by the system. The hot water
source inlet temperature on the heat sink temperature, hot water outlet temperature increased with an increase in source temperature for all the
temperature, heating capacity, refrigerant mass flow rate, total power cases. For the source inlet temperature of 10 ◦ C, the hot water temper­
consumption, and total COP of the system. The heat source inlet tem­ ature reached 65 ◦ C with a temperature lift of 55 ◦ C. For the source
perature was kept in the range of 10–50 ◦ C. The influence of the heating temperature of 50 ◦ C, the hot water temperature reached 124 ◦ C with a
source inlet temperature on the heatsink temperature is presented in temperature lift of 74 ◦ C in case 4. The system was able to deliver the hot
Fig. 8a. The CO2 mass percentage in the refrigerant mixtures varied water at 100 ◦ C while the source temperature was 38–40 ◦ C. The hot
between 5 and 20% both in LS and HS cycles. The LS evaporator

7
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

0.6

Refrigerant mass flow rate (HS)

0.5 Refrigerant mass flow rate (LS)

Refrigerant mass flow rate (kg/s)

0.4

0.3

0.2

0.1
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Tsource ( C)

Fig. 8d. Result of heating source inlet temperature on Refrigerant mass flow rate.

Fig. 8e. Result of heating source inlet temperature on Total power consumption.

water delivery of the system was induced by the heat source tempera­ development specifies that the heating capacities are significantly sha­
ture. For the case 3 and 4, initially, the hot water delivery was 2–3◦ lower ped by the heat source inlet temperature.
than in cases 1 and 2 and then it shifts up higher than in cases 1 and 2. Fig. 8d indicates the relationship between the refrigerant mass flow
The tendency of heating capacity corresponding to the variation in rate of the LS and HS cycle and the heat source inlet temperature. The
heat source inlet temperature was studied in detail and is indicated in refrigerant mass flow is boosted with the increase in heat source inlet
Fig. 8c. It was clear that the heating capacity increased with the increase temperature. It indicates that the heat source inlet temperature in­
in heating source inlet temperature. A similar trend was observed in all fluences the refrigerant mass flow rate to a significant level. In the LS
the cases. In all the cases, a 3.3 to 6% difference in heating capacity was heat exchanger, the heat exchange takes place between the water
noted compared to previous cases. The system’s minimum and flowing through the secondary circuit and the refrigerant flowing
maximum heating capacities were 44 and 232 kW, respectively. The through the LS evaporator. As the heat source inlet temperature

8
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 8f. Result of heating source inlet temperature on variation in Total COP.

increases, the refrigerant absorbs the heat from the heat source. This source temperature. Investigating the mass flow rate of the refrigerant is
causes the pressure of the LS evaporator to increase to a considerable very essential because it affects the design and volumetric efficiency of
level. In turn, it affects, the density of the refrigerant to increase. Hence, the compressor of the heat pump to a great level. The LS and HS cycle
the mass flow rate of the refrigerant increases with the increase in heat mass flow rates were increased from 0.11 to 0.49 and 0.14 to 0.5 kg/s

Adrian Mota-Babiloni et al. 2018

3.6

3.5 Present work

3.2

3 2.88
2.83
COP

2.61

2.5 2.35

1.5
25 30 35 40 45 50 55
Tsource ( C)
Fig. 8g. Comparison of COP of present work with published work.

9
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Pressure (kPa)
LS Evaporator
LS Condenser
HS Evaporator
HS Condenser

Tsource,in (C)

(a)

2500

2000

LS Evaporator
Pressure (kPa)

1500
LS Condenser
HS Evaporator
1000 HS Condenser

500

0
0 10 20 30 40 50 60
Tsource (C)

(b)
Fig. 9. Pressure vs Tsource (a) 5% CO2 (b) 10% CO2 (c) 15% CO2 (d) 20% CO2.

10
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 9. (continued).

respectively. There is a difference of 3.7 to 15.97% was found between power consumption for different CO2 mass fraction percentages of the
LS and HS cycles. refrigerant mixtures (cases 1–4). It was observed that there is a uniform
The variation in total power consumption of the system concerning increase in power consumption for cases 1 and 2. Whereas, for cases 3
different heat source temperatures is presented in Fig. 8(e). The system’s and 4, initially, the power consumption was lower than in the other two
total power consumption increased with the increase in heat source cases and then it goes up above other cases. This is due to the high
temperature. Also, the diagram indicates the variations of the total pressure and temperature glide of cases 3 and 4. The maximum total

11
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 10a. LS, HS, and Total power consumption for the different mass fractions of refrigerant mixtures.

power consumption of the system was 50.2, 68.9, 81.7, and 96.64 kW for evaporator, and HS condenser, and pressure at each stage could
5, 10, 15, and 20 CO2 mass fraction percentages, respectively. The sys­ contribute to the performance of the system. For instance, these pres­
tem’s total power consumption was increased by 18.4, 15.6, and 15.4% sures could affect the efficiency of the compressor to a great level and in
for cases 2, 3, and 4 respectively when compared with the previous turn COP. Fig. 9 (a-d) presents the pressure level for all the stages such as
cases. As the mass fractions increased in the LS and HS cycles of the LS evaporator and condenser, HS evaporator and condenser for the in­
system, power consumption rises steadily. crease in CO2 mass fraction in the refrigerant mixtures from 5 to 20%
The impact of variation in heat source temperatures on the system’s with respect to the heat source temperature increase from 10 to 50 ◦ C.
total COP is given in Fig. 8f. It can be noted that for the cases 1 and 2, the R744/R600 and R744/R601 refrigerant mixtures were used in the LS
system’s total COP has steadily improved significantly with the increase and HS of the HP cycle. Hence, the pressure levels of each stage will be
in heat source temperature. For the cases 3 and 4, it decreased slightly different. As the heat source temperature increased, the pressure in all
and then boosted up. This trend is well aligned with the power con­ the stages increased in turn increasing in heat sink temperature. The LS
sumption of the system. The system’s total COP was in the range from evaporator pressure range was from 121 to 554 kPa for the heat source
2.2 to 3.6, 1.85 to 3, 2.28 to 2.7, 1.8 to 2.4 for cases 1 to 4, respectively. temperature of 10–50 ◦ C as the CO2 mass fraction in the refrigerant
The COP levels are at the maximum when the CO2 mass fraction per­ mixtures was from 5 to 20%. The LS condenser, HS evaporator, and HS
centage of the refrigerant mixture is lower and the minimum when it is condenser pressure range were from 1044 to 3531, 252 to 1749, and 983
higher. A notable trend is that as the system’s power consumption to 4021 kPa. An increase in pressure of 21%, 59%, 75%, and 68% was
increased when the CO2 mass fraction of the refrigerant increased, the observed for the LS evaporator, LS condenser, HS evaporator, and HS
COP has shown a decrease in trend when the CO2 mass fraction of the condenser respectively for cases 1 to 4. The pressure range of the system
refrigerant mixture is raised. The decrease in percentage was found as is between 121 and 4021 kPa. The pressure level of the system has
15.6, 11.8, and 10% respectively when compared with the previous increased significantly as the CO2 mass fraction goes up in the refrig­
cases. erant mixture. It was noted that the system pressure for cases 1 and 2
The present work was compared with the published work of Adrian was well below 3000 kPa and very suitable for the smooth operation of
Mota-Babiloni et al., Energy, 2018 [17] and presented in Fig. 8g. The HTHP with excellent performance as most of the compressors available
graph indicates the relationship between the heat source inlet temper­ in the market were designed for this pressure level. whereas the pressure
ature and the system’s total COP. Adrian Mota-Babiloni et al. has used level was very high for the cases 3 and 4 with 15 and 20% CO2 mass
butane and pentane as the working fluids in the LS and HS of the cascade fractions.
system, respectively. Whereas in this work addition of CO2 with these
two fluids was suggested. The COP of their system increased from 2.35 3.2. Effect of different mass fractions of the mixtures on the system
to 2.88 with the rise in source temperature from 30 to 50 ◦ C. It can be performance
noted that for the same source temperature, the COP of the present work
has risen from 2.83 to 3.6. According to the present work, the COP was This section examined and displayed the effect of variation of the
well above when compared to the published work. Significant CO2 mass fraction of the refrigerant mixtures on the LS, HS, and total
improvement (20% increase) in system performance was noted when power consumption, maximum COP, and the level of heat sink tem­
using the proposed zeotropic mixtures due to the improved thermo- perature that can be obtained in the system. Fig. 10(a) presents the LS
physical properties. and HS compressor and total power consumption concerned with the
Investigating the pressure at all stages of the cascade HTHP is key to different CO2 mass fractions. In the LS cycle and total power consump­
optimizing the better performance of the system. The cascade system tion rises gradually with the increase in CO2 mass fractions. The power
includes several pressure stages such as LS evaporator, LS condenser, HS consumption of the system was in the range of 56 to 96.65 kW. High

12
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 10b. LS, HS, and Total maximum COP for the different mass fractions of refrigerant mixtures.

Fig. 10c. Variation in heat sink temperatures for the different mass fractions of Fig. 11a. Effect of hot water temperature outlet on Compressor power
refrigerant mixtures. consumption.

power consumption of 96.65 kW was observed for case 4 which in turn to 20%), there was a reduction in COP of 33.33% was noted. An anal­
the system COP reduced significantly. By comparison, the power con­ ogous trend was found for the LS cycle of the system. The maximum COP
sumption increased by 18.4%, 15.67%, and 15.4% respectively when decreased from 6.06 to 2.95 with a reduction of 51%. The maximum
the CO2 mass fraction goes from 5 to 20%. The HS compressor power COP of the HS was increased from 7.33 to 8.54.
consumption was slightly increased. For case 1, as seen from the figure, The influence of the CO2 mass fraction of the mixture on the mini­
both compressors were loaded equally and indicates that the case 1 is mum and maximum heat sink temperatures that can be achieved was
better performing. studied and presented in Fig. 10c. It is important to analyze the heat sink
The maximum COP of the LS, HS, and total system is represented in temperature for cases 1 to 4, as the hot water delivery temperature is
Fig. 10b. The maximum COP of the total system has shown a decrease in very much influenced by the heat sink temperature. It was observed that
trend from 3.6 to 2.4 concerned with the increase in CO2 mass fractions. the maximum heat sink temperature was 181.6 ◦ C and 118.9 ◦ C when
This trend gives a clear indication that the system can deliver higher using the 20% and 5% CO2 in the refrigerant mixture. This work aimed
COP for the lower level of CO2 mass fraction in the refrigerant mixture. to obtain a hot water temperature of more than 115 ◦ C as the same was
Overall, when the CO2 mass fraction of the mixture increased by 25% (5

13
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

Fig. 11b. Effect of hot water temperature outlet on the heating capacity for different mass fractions of refrigerant mixtures.

Fig. 12a. Result of heat sink temperature on LS COP for different heating Fig. 12b. Result of heat sink temperature on HS COP.
source temperatures.
studied and presented in Fig. 11a. As given in the figure, the power
reflected in the figure mentioned. It can be noted that the system can consumption of both the LS and HS compressors increased progressively
deliver a hot water temperature of more than 115 ◦ C when using 5% CO2 when the hot water outlet temperature increased from 64 to 116 ◦ C for
in the refrigerant mixture. Further, an increase in mass fraction will the case 1. The LS compressor power consumption from 18.25 to 28.7
boost the heat sink temperature to a higher level. kW with a rise of 36% was seen. The HS compressor power consumption
from 9.45 to 27.5 kW with an improvement of 66% was noted. It can be
noted that the power consumption of both LS and HS compressors was
3.3. Effect of hot water outlet temperature on the system performance the almost same when the hot water temperatures reach 116 ◦ C. It in­
dicates that both compressors were loaded evenly, and the system was
The variation in LS and HS compressor power consumption related to effectively employing both LS and HS compressors during the operation.
the outlet temperature of the hot water from the HS heat exchanger was The effect of hot water outlet temperature on the variation in heat

14
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

4. Conclusion

The present work demonstrated the modeling and optimization of a

cascade HTHP for clean heating with new natural zeotropic mixtures.
For the first time, CO2/butane and CO2/pentane mixtures were intro­
= 0.0011x2 - 0.0487x + 0.9979
vol
duced as refrigerants in the system’s Low-stage (LS) and High-stage (HS)
cycle, respectively. The model was built in the MATLAB platform and a
thorough analysis was carried out. The following conclusions were
drawn from the current work:

• The refrigerant mixtures revealed excellent properties in terms of

= -0.00000461x6 +
isen best thermal matching to meet the requirements of high-temperature
0.00027131x5 - 0.00628605x4 + heating with very high performance. It showed an improvement of
0.07370258x3 - 0.46054399x2 +
20% COP, compared to those of pure fluids. The temperatures of all
1.40653347x - 0.87811477
the individual heat exchangers were not crossing with each other
making the refrigerant mixtures suitable to effectively exchange
heat.
• The results indicate that the heat source temperature affected the
Fig. 13. Result of Pressure ratio on Compressor efficiencies.
system performance to a great level, and it is possible to achieve
high-temperature heating.
• The hot water temperature and heat sink temperatures and power
capacity of the cascade HTHP was studied and presented in Fig. 11b.
consumption were elevated, as the CO2 mass fraction percentage of
Almost a linear approached curve was detected between these two pa­
the refrigerant mixtures boosted.
rameters. The heat sink temperature was the responsible parameter to
• Increases in heat sink temperature affected the COP to lower. The
increase the hot water outlet temperature. The heating capacity of the
pressure level of each stage of the system was better optimized to
system was enhanced as the heat sink temperature grew. For a raise in
make it adaptable to the global standard.
hot water temperature from 64 to 116 ◦ C, the heat capacity has risen
• Volumetric efficiency of compressors showed a decrease in trend,
from 60 to 201.5 kW (69.8% rise) (case1). The maximum heat capacity
while the isentropic efficiency was enhanced up to some point and
of the system was 232 kW (case 4). There is a difference in heat ca­
then dropped gradually, to the pressure ratio increment.
pacities of 13% was observed for cases 1–4.
• The comparison of the performance of cases 1 to 4 indicates that,
case 1 with a 5% CO2 mass fraction, used in both stages was a well-
3.4. Effect of heat sink temperature on the individual system COP performing mixture as it generates a high COP, suitable pressures,
temperature glide, and high-temperature heating possibilities. Also,
The effect of heat sink temperature on the LS and HS cycle COP for possible to use a 10% CO2 mass fraction for better temperature lift at
case 1 was studied and presented in Fig. 12a and Fig. 12b. The source the cost of a 16.6% reduction in the system’s total COP.
temperatures were kept constant, and the heat sink temperatures were • This investigation is the need of the hour in cold climate countries
varied to get the variation in COP of LS and HS cycles. The COP of the LS where the high temperatures heating for residential heating and in­
and HS declined as the heat sink temperature rose. It is perceptible that dustrial process heating is dominant. It could be a step towards zero
the level of COP is low for low source temperature and elevated for the carbon world.
upper source temperature. A decrease in LS COP of 30% was observed
while the heat sink temperature increased. A raise in source inlet tem­ CRediT authorship contribution statement
perature influenced the LS COP to a great level (51%). The HS cycle COP
has shown a decrease in trend with an increase in heat sink temperature. P. Ganesan: Conceptualization, Methodology, Data curation,
The COP has reduced from 8.5 to 6.44 when the heat sink temperature Writing – original draft, Investigation, Writing – review & editing.
was increased from 95 to 123 ◦ C. The same trend and values were Trygve M. Eikevik: Supervision.
observed for all the source temperatures as the source temperature does
not influence the HS COP. The reduction in HS COP was 17.92%. Declaration of Competing Interest

3.5. Effect of pressure ratio on efficiencies of the compressors The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
The pressure ratio is one of the key components in the performance the work reported in this paper.
of cascade heat pump systems. High pressure ratio leads to decrease in
compressor efficiencies. The cascade systems are efficient in handling Data availability
the system pressures and can be operated with maximized performance
compared to single stage systems. The evaporation and condenser The data that has been used is confidential.
pressures are constant whereas there is a temperature glide during the
condensation and evaporation processes. The effect of the pressure ratio Acknowledgment
on the isentropic and volumetric efficiencies of the compressor is pre­
sented in Fig. 13. The compressor’s volumetric efficiency decreases The authors gratefully acknowledge the financial support from the
gradually when the pressure ratio is increased from 3.3 to 8.6. The Research Council of Norway under the Chinese-Norwegian collabora­
volumetric efficiency was decreased by 22% (0.85 to 0.66). The isen­ tion project on Energy (Project number: 304191) - Key technologies and
tropic efficiency of the compressor increased from 0.55 up to 0.75 and demonstration of combined cooling, heating, and power generation for
then started to decline gradually as the pressure ratio goes up. The low-carbon neighborhoods/buildings with clean energy (ChiNoZEN).
maximum isentropic efficiency of the compressor was 0.75 at a pressure
ratio of 3.3 whereas the higher volumetric efficiency was 0.85 at a
pressure ratio of 3.3.

15
P. Ganesan and T.M. Eikevik Energy Conversion and Management: X 20 (2023) 100407

References internal heat exchangers using refrigerants with low global warming potential.
Energy 2018;165:1248–58.
[18] Rudonja N, Gojak M, Zlatanović I, Todorović R. Thermodynamic analysis of a
[1] Ganesan P, Thirugnanasambandam M, Rajakarunakaran S, Devaraj D. Specific
cascade heat pump incorporated in high-temperature heating system. J Mech Eng
energy consumption and CO2 emission reduction analysis in a textile industry. Int J
2020;66(11):677–83.
Green Energy 2015;12:685–93. https://doi.org/10.1080/15435075.2013.829479.
[19] Li X, Zhang Y, Ma X, Deng Na, Jin Z, Yu X, et al. Performance analysis of high-
[2] Mengxiao Yu, Liu X, Liu Z, Yang S. Energy, exergy, economic and environmental
temperature water source cascade heat pump using BY3B/BY6 as refrigerants. Appl
(4E) analysis of a novel power/refrigeration cascade system to recover low-grade
Therm Eng 2019;159:113895.
waste heat at 90–150 ◦C. J Clean Prod 2022;132353. https://doi.org/10.1016/j.
[20] Sulaiman AY, Cotter DF, Le KX, Huang MJ, Hewitt NJ. Thermodynamic analysis of
jclepro.2022.132353.
subcritical High-Temperature heat pump using low GWP Refrigerants: a theoretical
[3] Thirugnanasambandam M, Ganesan P, Soundrapandian E, Ragulkumar K. An
evaluation. Energ Conver Manage 2022;268:116034. https://doi.org/10.1016/j.
approach to energy, emission and economic analysis in a coffee industry.
enconman.2022.116034.
International Journal of Mechanical and Production Engineering Research and
[21] Ma X, Zhang Y, Fang L, Xiaohui Yu, Li X, Sheng Y, et al. Performance analysis of a
Development 2018;8(7):392–403.
cascade high-temperature heat pump using R245fa and BY-3 as working fluid. Appl
[4] Hamid K, Sajjad U, Ulrich Ahrens M, Ren S, Ganesan P, Tolstorebrov I, et al.
Therm Eng 2018;140:466–75. https://doi.org/10.1016/j.
Potential evaluation of integrated high temperature heat pumps: a review of recent
applthermaleng.2018.05.052.
advances. Appl Therm Eng 2023;230:120720.
[22] Kosmadakis G, Arpagaus C, Neofytou P, Bertsch S. Techno-economic analysis of
[5] Xu L, Li E, Xu Y, Mao N, Shen Xi, Wang X. An experimental energy performance
high-temperature heat pumps with low-global warming potential refrigerants for
investigation and economic analysis on a cascade heat pump for high-temperature
upgrading waste heat up to 150 ◦C. Energ Conver Manage 2020;226:113488.
water in cold region. Renew Energy 2020;152:674–83.
https://doi.org/10.1016/j.enconman.2020.113488.
[6] Ma X, Zhang Y, Li X, Zou H, Deng Na, Nie J, et al. Experimental study for a high
[23] Uusitalo A, Turunen-Saaresti T, Honkatukia Jh, Tiainen J, Jaatinen-Värri A.
efficiency cascade heat pump water heater system using a new near-zeotropic
Numerical analysis of working fluids for large scale centrifugal compressor driven
refrigerant mixture. Appl Therm Eng 2018;138:783–94.
cascade heat pumps upgrading waste heat. Appl Energy 2020;269:115056. https://
[7] Dong S, Meng X, Hu X, Sun Z, Wang H, Luo Y. Investigation of cascade high
doi.org/10.1016/j.apenergy.2020.115056.
temperature heat pump optimal design theory based on experiment supporting
[24] Chang Xu, Yang Ha, Xiaohui Yu, Ma H, Chen Mo, Yang M. Performance analysis for
multi-objective optimization. Energ Conver Manage 2022;267:115873.
binary mixtures based on R245fa using in high-temperature heat pumps. Energy
[8] Le KX, Huang MJ, Shah NN, Wilson C, Artain PM, Byrne R, et al. Techno-economic
Conversion and Management: X 2021;12:100123. https://doi.org/10.1016/j.
assessment of cascade air-to-water heat pump retrofitted into residential buildings
ecmx.2021.100123.
using experimentally validated simulations. Appl Energy 2019;250:633–52.
[25] Ganesan P., Ryssdal S.T., Eikevik T.M., Wang R. Bin Hu, Performance assessment of
[9] Boahen S, Anka SK, Lee KH, Choi JM. Performance analysis of cascade multi-
high-temperature heat pump in an integrated energy system. 19th International
functional heat pump in summer season. Renew Energy 2021;163:1001–11.
Refrigeration and Air Conditioning Conference at Purdue; 2022. https://docs.lib.
[10] Navarro-Esbrí J, Fernández-Moreno A, Mota-Babiloni A. Modelling and evaluation
purdue.edu/cgi/viewcontent.cgi?article=3493&context=iracc.
of a high-temperature heat pump two-stage cascade with refrigerant mixtures as a
[26] Zhang H, Geng X, Shao S, Si C, Wang Z. Performance analysis of a R134a/CO2
fossil fuel boiler alternative for industry decarbonization. Energy 2022;254:
cascade heat pump in severe cold regions of China. Energy 2022;239:122651.
124308.
https://doi.org/10.1016/j.energy.2021.122651.
[11] Mateu-Royo C, Navarro-Esbrí J, Mota-Babiloni A, Amat-Albuixech M, Molés F.
[27] Mengjie B, Li Z, Ruikai Z. Review on applications of zeotropic mixtures. J Therm
Theoretical evaluation of different high-temperature heat pump configurations for
Sci 2022;31:285–307. https://doi.org/10.1007/s11630-022-1569-x.
low-grade waste heat recovery. Int J Refrig 2018;90:229–37. https://doi.org/
[28] Fronk BM, Garimella S. In-tube condensation of zeotropic fluid mixtures: a review.
10.1016/j.ijrefrig.2018.04.017.
Int J Refrig 2013;36(2):534–61.
[12] Wu Di, Hu B, Wang RZ, Fan H, Wang R. Rujin Wang, The performance comparison
[29] Di Nicola G, Polonara F, Stryjek R, Arteconi A. Performance of cascade cycles
of high temperature heat pump among R718 and other refrigerants. Renew Energy
working with blends of CO2 + natural refrigerants. Int J Refrigeration 2011;34:
2020;154:715–22.
1436–45. https://doi.org/10.1016/j.ijrefrig.2011.05.004.
[13] Zou H, Li X, Tang M, Jiang Wu, Tian C, Butrymowicz D, et al. Temperature stage
[30] Dai B, Dang C, Li M, Tian Ha, Ma Y. Thermodynamic performance assessment of
matching and experimental investigation of high temperature cascade heat pump
carbon dioxide blends with low-global warming potential (GWP) working fluids for
with vapor injection. Energy 2020;212:118734. https://doi.org/10.1016/j.
a heat pump water heater. Int J Refrig 2015;56:1–14. https://doi.org/10.1016/j.
energy.2020.118734.
ijrefrig.2014.11.009.
[14] Boahen S, Anka SK, Lee KH, Choi JM. Performance characteristics of a cascade
[31] Zhang X, Wang F, Fan X, Duan H, Zhu F. An investigation of a heat pump system
multi-functional heat pump in the winter season. Energ Buildings 2021;253:
using CO2/propane mixture as a working fluid. Int J Green Energy 2017;14:
111511.
105–11. https://doi.org/10.1080/15435075.2016.1253577.
[15] Wu Z, Wang X, Sha Li, Li X, Yang X, Ma X, et al. Performance analysis and multi-
[32] Massuchetto LHP, Nascimento RBCd, Carvalho SMRd, Araújo HVd, d’Angelo JVH.
objective optimization of the high-temperature cascade heat pump system. Energy
Thermodynamic performance evaluation of a cascade refrigeration system with
2021;223:120097.
mixed refrigerants: R744/R1270, R744/R717 and R744/RE170. Int J Refrig 2019;
[16] Abedini H, Tomassetti S, Di Nicola G, Quoilin S, Arteconi A. Zeotropic Mixtures
106:201–12.
R1234ze(Z)/acetone and R1234ze(Z)/isohexane as refrigerants in high
[33] Lemmon EBIHM, McLinden M. NIST Standard Reference Database 23: Reference
temperature heat pumps: influence of the accuracy in thermodynamic properties
Fluid Thermodynamic and Transport Properties-REFPROP, Version 9.1, National
evaluations. Int J Refrig 2023. https://doi.org/10.1016/j.ijrefrig.2023.05.008.
Institute of Standards. Gaithersburg: and Technology; 2018.
[17] Mota-Babiloni A, Mateu-Royo C, Navarro-Esbrí J, Molés F, Amat-Albuixech M,
[34] MATLAB. Version 2021b, The Math Works, Inc, 2021b.
Barragán-Cervera Á. Optimisation of high-temperature heat pump cascades with

16