Recent Advances in Ejector-Enhanced Vapor Compression Heat Pump and Refrigeration Systems—A Review

Abstract

The incorporation of ejectors into heat pump and refrigeration cycles has been the subject of growing interest, largely due to their simple structure, high reliability, and cost-effectiveness. This paper investigates the recent advancements in novel design concepts of ejector-enhanced vapor compression heat pump and refrigeration cycles. An overview of novel single-stage and two-stage compression cycles utilizing a single or multiple ejectors is provided. First, the system setup, operational principles, description, and figures of the existing schemes are provided. Second, the main results, such as the coefficient of performance (COP), volumetric heating capacity and exergy destruction, are discussed. In conclusion, the paper presents a coherent summary of the current developments, future prospects, and the current knowledge gap. A plethora of research is present in developing theoretical systems with high efficiency. However, experimental tests for real-life implementations are limited. This review aims to provide the reader with an overview of recent theoretical and experimental studies.

1. Introduction

In a world where energy supplies are shrinking and consumption rates are rising rapidly, driven by factors such as economic development, industrialization, population growth, and technological innovation, the necessity for efficient and sustainable solutions has reached an unprecedented level [1,2]. Traditional energy sources, such as fossil energy sources (FES) including oil, natural gas, and coal, are struggling to keep up with the growing demand, resulting in higher costs and an increased negative environmental impact [1]. As a result, the world is confronted with the problems of climate change, environmental risks, and resource limitations [3], thus urging a rapid transition from traditional fossil fuels to clean, sustainable, and efficient renewable energy to achieve carbon neutrality. [1,4]. Taking the initiative to achieve carbon neutrality by 2050 is an urgent matter to limit global warming to 1.5 °C this century and thereby reduce the negative impacts of climate change [5]. In order to achieve the desired temperature reduction, the International Renewable Energy Agency (IRENA) has set a target of tripling global renewable energy generation capacity by 2030 [6].

Policymakers have indicated that the transition to a renewable and sustainable energy economy is of key importance [5]. Renewable energy sources have a vital role to play in mitigating climate change [6]. The combination of innovation and experience has resulted in renewable energy becoming one of the most cost-effective and widely utilized energy sources in numerous countries worldwide [4]. Renewable energy production is of critical importance to sustainable development due to its inherent sustainability and absence of greenhouse gas (GHG) emissions [7]. The Sustainable Development Goals (SDG) agenda, which was established to promote sustainable development, includes a total of seventeen goals, each with its own specific purpose. The seventh goal is to promote the supply and use of affordable, reliable, accessible, and sustainable energy. The goal for 2030 is to achieve a significant increase in the availability of affordable, reliable, and modern energy services while increasing the share of renewable energy in the global energy mix [7,8]. Considering the circumstances, a transition towards renewable energy production and energy-efficient technologies is imperative [9]. According to Alabbasi [10] a sustainable energy system may be defined as a system capable of generating sufficient clean, safe, and reliable power to meet the energy needs of all members of society at an affordable price. In order to achieve the goal of a net-zero carbon economy, the transformation of the energy sector, including continued growth in the share of renewable energy is necessary.

Heat pumps have emerged as a promising solution for sustainable heat production and recovery, with a range of implementation options. High-temperature heat pumps play a pivotal role in the decarbonization process [11,12], as well as in the hot water supply and a range of industrial sectors, including plastics, textiles, wood, construction, beverages, cheese production, spray drying, and other chemical industries. [13,14]. The electrification of urban residential buildings is a crucial step in reducing carbon emissions [15]. A clear path for the electrification of process heat has not yet emerged, but many scientists believe that large-scale industrial heat pumps will play an important role in this transition [14]. Zhong et al. [15] demonstrated that, when selecting electrification appliances, a high-efficiency electric appliance like a heat pump can already reduce carbon emissions in most areas. Jensen et al. [16] investigated an integrated heat pump system used to electrify an amine CO₂ capture unit for a biogas upgrading process using aqueous monoethanolamine. In their study, the researchers found that the electrification of biogas upgrading with heat pump systems can reduce the total energy input, which can be even further reduced when coupled with a vacuum-operated stripper. Khosravani et al. [17] observed that the utilization of an air-source heat pump instead of a conventional heating, ventilation, and air conditioning (HVAC) system for heating purposes slightly reduces the summer electricity load, but is not profitable in the winter. Williams et al. [18] conducted a state-by-state comparison of building electrification in multifamily buildings. They found that, by 2021, due to improvements in heat pump technology, all states performed better with electrified buildings rather than the traditional mixed-fuel buildings.

Low-temperature heat pumps for domestic use have limited use in industrial processes due to a larger heating capacity and high-temperature heat demand [19]. The utilization of a two-stage compression enables the compressors to operate at a reduced compression ratio, consequently reducing the discharge temperature and enhancing the heating performance [20]. Kosmadakis et al. [21] investigated the financial aspects of high-temperature heat pumps and found that the two-stage cycle is more cost-effective for higher-temperature lifts. Additionally, the efficiency of heat pumps can be enhanced through the implementation of a two-stage compression cycle [22]. This enables the two-stage compression cycle to be used in high-temperature heating systems [19] and in cold climates [23].

As heat pump technology progresses, hybrid systems are being developed with the objective of enhancing their efficiency. The combination of heat pumps with fuel cells can be utilized as a combined heat and power system. As demonstrated by Bendaikha et al. [24], the proton exchange membrane fuel cell (PEMFC) can serve as the heat source in such a system. Similarly, in our previous research, we coupled a high-temperature heat pump with a solid oxide fuel cell (SOFC) with integrated carbon capture [25]. The SOFC generated electricity and heat, which were used in a Rankine cycle and then in the heat pump. Zhang et al. [26] investigated a fuel-cell backup and heat pump-assisted fuel-cell battery electric vehicle, wherein they discussed various crucial operational conditions and enhancements in equivalent efficient battery capacity. In addition to fuel-cell-coupled hybrid systems, heat pumps are currently being studied for use with solar energy [27,28,29] and geothermal energy [30,31,32].

In addition to integrating the heat pump into a hybrid system, the design of the heat pump technology itself can be altered to achieve better performance. The ejector is frequently used to improve the performance of vapor compression heat pumps by recovering the expansion work via the ejector [33]. Ejectors are an attractive option for refrigeration and heat pump systems due to their simple structure, high reliability (as a result of the lack of moving parts), and low cost [34]. A simple schematic of an ejector is shown in Figure 1. The ejector was first invented and proposed in 1858 by Henry Giffard. The basic structure of an ejector consists of two inputs and a discharge port [35]. The high-pressure primary flow enters the ejector and is expanded and accelerates through the nozzle. It flows with high speed and creates a low-pressure zone entraining the secondary flow, which is at low pressure [36]. In the ejector, both flows are mixed in the mixing chamber and exit the ejector at the diffuser. The ejector heat pump (EHP) represents an efficient alternative for energy production, offering a potential replacement for conventional vapor compression heat pumps. Moreover, EHPs have emerged as a viable alternative in recent cooling applications and refrigeration-cycle studies [34,37]. Single- and dual-ejector compression cycles are introduced in this study. The two ejectors can be operated alternatively or simultaneously. The former concept is introduced to enable alternating cooling or heating operations. As one ejector operates, the second one is bypassed, and the system works as a single-ejector cycle. In the second alternative, both ejectors are working simultaneously and enable better expansion work recovery. This review article examines and assesses the various ejector-based compression cycles. The objective of this study is to provide an overview of the currently available ejector-enhanced systems. The focus of the study is the review of different designs and their performance, rather than the operation of the ejector itself. A more detailed description of the systems is provided in the respective articles.

2. Ejector-Enhanced Single-Stage Vapor Compression Cycle

In this section, the single-stage compression heat pump and refrigeration cycles are presented. The systems are grouped into subsections that share the same operating principles. A brief overview of the relevant novelties of each system is given, including a discussion of the results.

2.1. Single-Stage Ejector-Enhanced Vapor Compression Cycle

This subsection presents an introduction to the advancements of the fundamental ejector-expansion vapor compression cycle. The operational principle is illustrated in Figure 2. At point 1, the working fluid or refrigerant is compressed to a high-temperature and high-pressure state (stream 2). As it passes through the condenser, it is cooled and condensed, emitting heat. Subsequently, the refrigerant at the condenser outlet (stream 3) is directed into the ejector as the primary ejector fluid. The ejector outflow is separated in a flash tank or separator into the top vapor phase and bottom liquid phase. The latter (stream 7) is expanded in an expansion valve and enters the evaporator (stream 8). In the evaporator, a heat source evaporates the working fluid into the vapor phase (stream 9) before it enters the ejector as the secondary flow. To conclude the cycle, the upper vapor phase at the flash tank is directed back into the compressor as stream 1.

In a study conducted by Fingas et al. [38], a single-stage air-to-water R290 heat pump system was evaluated with two distinct expansion devices, namely a throttling valve and a two-phase ejector followed by a liquid separator. The schematic representation of the system is presented in Figure 2. The ejector-enhanced system consists of a compressor, two heat exchangers to simulate the condenser and evaporator, an ejector, a separator, a throttling valve, and two auxiliary glycol loops for heat addition and rejection. The objective was to assess the suitability of these devices for domestic applications. The study was conducted experimentally on a small-scale 8 kW unit. The results show that the ejector-based heat pump performed significantly better than the throttle valve design due to the recovery of the throttling losses. In the ejector mode, the compressor could be operated at lower pressure ratios, and higher mass flow rates could be achieved. The latter is independent of the evaporation pressure and ambient conditions, allowing the heat pump to achieve higher heating capacities. The overall COP of the ejector-mode heat pump ranged from 2.6 to 3.0, an increase of up to 38.2% at the lowest ambient temperature compared to the direct expansion mode.

Singmai et al. [39] conducted a similar experiment on a water-to-water two-phase ejector heat pump with a cooling load up to 2500 W and compared the results to a vapor compression heat pump. The heating COP is increased by 5.7–11.6% depending on the operating conditions. The researchers found that increasing both the heat sink temperature and the heat-source temperature increases the COP of the system. However, it should be noted that a lower heat-source temperature results in higher COP improvements.

The system was also tested in transcritical conditions. In transcritical conditions, the refrigerant does not condense, so a gas cooler is used instead of a condenser. Taslimi Taleghani et al. [40] developed a model for the two-phase ejector and incorporated it into an existing experimentally validated model of a CO₂ (R744) heat pump system. The researchers reported a 17% and a 20% improvement in heating COP and heating capacity, respectively, under the given operating conditions. The results show that the effect of the gas cooler outlet temperature on the system is greater than that of the evaporator outlet temperature, as the increase in evaporator pressure exceeds the increase in gas cooler outlet temperature. The effect of the heat transfer area was also investigated, and it was found that an increase had a positive effect on system performance but a negative effect on ejector efficiency. To achieve further improvements in efficiency, the authors propose the use of innovative ejector designs that optimize pressure lift or enhance the configuration of the heat pump system with multiple ejectors [38]. Further development of heat pumps is needed to investigate the increase in both system performance and COP. The results of the studies can be used for future reference cases with more accurate design models to predict the fluid properties in two-phase flows to increase ejector-expansion heat pumps [39].

In 2013, Zhang et al. [41] proposed a design for an ejector-enhanced heat pump-boosted district heating system, with the objective of enhancing existing heating systems through the integration of combined heat and power. An updated, novel modified design was proposed by researchers [42], integrating an ejector heat pump cycle with waste heat recovery, with the objective of replacing the traditional compression heat pump cycle. The system facilitates the recovery of heat in flue gases. The simplified basic structure is presented in Figure 3. The system comprises a condenser, an evaporator, a pump, an ejector, a valve, and a boiler/boiling device. The refrigerant at the ejector outlet (stream 4) passes through a condenser and exits at high pressure and low temperature (stream 5). Subsequently, the stream is divided into two distinct streams, designated as stream 6 and stream 1. The former is then expanded in a valve and stream 7 enters the evaporator. The refrigerant (stream 8) is directed into the ejector as the secondary fluid. Meanwhile, stream 1 is directed into a boiling device via a pump. Stream 3 at the boiler outlet is directed into the ejector as the primary flow. The system is set up in two stages. In the first stage, the working fluid circulates in an air-source heat pump, while the second stage uses the flue gases’ heat for the evaporator and boiling device. Several parameters were examined in order to investigate the heating COP. It was observed that, as the evaporation temperature increased, the COP also rose. However, an inverse relationship was identified between the COP and the condensation temperature, whereby an increase in condensation temperature resulted in a decrease in COP. The system demonstrates an effective novel cycle for the recovery of waste heat. However, further studies are required to enhance the cycle, with the objective of improving system efficiency and adaptability to a range of operational conditions. It is important to assess the feasibility of manufacturing the necessary equipment for practical implementation.

A steam-ejector heat pump for domestic water heating was designed and constructed [37]. The cycle configuration is similar to that described by Zhang et al. [42]. After the condenser, the refrigerant is led into a reservoir and then split into a high-temperature evaporator (boiler) and a low-temperature evaporator via a pump and throttle valve, respectively. The primary and secondary flow are injected into the ejector and condensed in the condenser, before circulating back into the reservoir. For this study, the COP evaluation at different operating conditions was considered. The researchers observed that, as previously reported by Zhang et al. [42], the experimental COP value increases with the evaporation temperature. Nevertheless, this is only the case for the low-temperature evaporator, as a reduction in COP was observed with the increase of the high-temperature evaporator. In the sub-critical operation mode, the following results were obtained. The highest COP for the system was 2.42, achieved for a high-temperature evaporator of 130 °C and a low-temperature evaporator of 30 °C, while the condenser capacity was approximately 5.6 kW. The study is supported by experimental data and is not entirely based on computer simulations or modeling. Although the system did not achieve the desired water outlet temperature, the study significantly contributes to the comprehension of a gas-fired ejector heat pump design.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A1. Main findings of ejector-expansion single-stage heat pump cycles (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Spitzenberger et al. [37]	E	Ejector-enhanced heat pump system with an integrated boiler	A reduction in COP was observed with the increase of the high-temperature evaporator.
Fingas et al. [38]	E	Ejector-expansion vapor compression heat pump cycle	Rise in COP by 38.2% (compared to conventional), lower pressure ratio, and high mass flow could be achieved.
Singmai et al. [39]	E		COP improvement of 5.7–11.6%.
Taleghani et al. [40]	T		COP and volumetric heating capacity improvement of 17% and 20%, respectively. Gas cooler outlet temperature has bigger influence than evaporation temperature.
Zhang et al. [42]	E	Ejector-enhanced heat pump system with an integrated boiler	Evaporation temperature increases COP, whereby condensation temperature decreases it.

. In comparison to conventional compression systems, single-stage ejector-enhanced systems showed an increase in COP, which rose with the evaporation temperature and decreased with the condensation temperature. Furthermore, the incorporation of an ejector resulted in a reduction in the pressure ratio. The greatest observed COP increase was 38.2% in comparison to a conventional heat pump. The results demonstrated that the temperature of the gas cooler outlet had a greater impact on system performance than that of the evaporator outlet in transcritical operation (R744). The impact of the heat transfer area was also examined, revealing that an increase had a favorable impact on system performance but a detrimental effect on ejector efficiency. Some of the experimental tests did not achieve the desired results, and therefore, further research is needed when operating in different conditions. It is evident that the systems require further enhancements and modifications before they can be effectively deployed in industrial settings.

2.2. Ejector-Enhanced Vapor Compression Cycle with an Internal Heat Exchanger

In this subsection, the studies of heat pumps enhanced with an internal heat exchanger (IHX) are presented.

Researchers [43] conducted a study to compare a transcritical CO₂ ejector-expansion cycle incorporating an internal heat exchanger with a standard regenerative cycle and a vapor injection cycle. The IHX-enhanced system is presented in Figure 4. The study investigates the possibility of using the system for electric vehicles. The cycle is designed to operate in cooling, heating, dehumidifying, and defrosting modes. The proposed scheme adds an ejector between the separator and the out-cabin internal heat exchanger compared to the classic vapor-injection system. The refrigerant is directed into the compressor and compressed into a high-temperature and high-pressure state (stream 2). As the refrigerant passes through the in-cabin gas cooler/condenser, it emits heat and is consequently directed into the IHX (stream 3) before being injected into the ejector as the primary flow (stream 4). The ejector outlet is separated into the liquid and vapor phases in the separator. The former (stream 6) passes an expansion valve (stream 7) and is then evaporated in the in-cabin evaporator. The vapor phase (stream 8) is directed into the ejector as the secondary fluid. The vapor phase from the separator (stream 9) is heated in the cold IHX inlet before entering the compressor to complete the cycle. The highest achievable cooling and heating COP is 2.00 at the temperatures of 35 °C/27 °C (out-cabin/in-cabin air temperature) and 1.85 at 0 °C/20 °C, respectively. A comparison of the obtained results with those of the basic regenerative and vapor injection systems reveals a maximum COP improvement of 21.6% and 31.0%, respectively. The study offers valuable insights into the knowledge gap in the field of heating and cooling performance for ejector-expansion systems in electric vehicles. The study is significant, as the results are obtained on an experimental bench with test conditions that reflect real-life situations. However, further research is required to achieve an optimal design point for performance under the entire temperature range needed for electric vehicles.

The cycle has been studied in other studies with minor adjustments. Ameur and Aidoun [44] presented a study that tried to maximize the overall efficiency of the system by adjustments to the ejector geometry in varying operating conditions. The results support the conclusion of Yang et al. [43] that an increase in COP is present in the IHX system. In addition, they found that the operation of the whole cycle at optimal conditions does not align with that of its components. Wang et al. [45] found an increase in exergy efficiency due to the reduction in exergy destruction in the throttling valves of the new system. An experimental study was conducted to evaluate the performance of the transcritical R744 IHX cycle heat pump in a water heating system with a built-in oil separator [46]. The researchers focused on the effects of the hot water outlet temperature, compressor speed, expansion valve opening, and compressor discharge pressure on the ejector and overall system performance. The system demonstrated high efficiency, reaching a COP of 4.6 at the tap-water outlet temperature of 70 °C. Comparing the results with a basic refrigeration cycle, the enhanced system shows a COP improvement of 10.3%.

Further investigation was undertaken to determine the performance of the system when utilizing alternative low global warming potential (GWP) refrigerants [47]. In addition to the ejector system, a parallel compression configuration with an economizer and an IHX was also introduced. The results of the newly implemented system were presented in comparison to a single-stage IHX cycle without an ejector. The conventional ejector-enhanced IHX system using HFC-245fa presents an improvement in the COP and volumetric heating capacity of up to 36% and 40%, respectively. The novel system with parallel compression achieves an improvement in the COP of up to 72.5% and a notable volumetric heating capacity improvement of 80%, also compared to the single-stage cycle. Environmentally friendly refrigerants were tested, and it was found that HC-601 showed the highest COP for each configuration. Meanwhile, HCFO-1224yd(Z) and HCFO-1233zd(E) resulted in a compromise between COP and volumetric heating capacity. It is, therefore, evident that the selection of appropriate configurations and refrigerants is crucial for achieving optimal performance in different application conditions and requirements.

An alternative to the IHX system, presented in Figure 5, was suggested by Qin et al. [48]. At the IHX outlet, the stream is divided into two distinct flows: the primary ejector flow and the expansion valve inlet. The expanded fluid is evaporated and fed into the ejector as the secondary fluid. In this configuration, the ejector outlet is not led into a separator. Rather, the fluid is evaporated directly in an air-source evaporator before returning to the IHX. The results show that this novel system outperforms the conventional ejector IHX system and yields a 12% higher exergy efficiency. Furthermore, the heating COP decreased to 4.27 compared to the 4.58 of the conventional system, but the cooling COP is significantly higher, at 6.11 compared to 1.83.

Pardiñas et al. [49] expanded on the studied IHX alternatives and suggested some modifications. The researchers showed that putting an ejector in between the evaporation process and splitting it into two stages reduces the evaporation pressure and simultaneously allows for the increase in compression suction, in some cases. The study is supported by experimental data and offers valuable data to overcome the issues of the actual application of the ejector-enhanced cycles.

An experimental study [50] evaluated the ejector performance in an ejector CO₂ system of 33 kW cooling capacity, with and without an IHX, at different gas cooler outlet temperatures. The cycle introduces a heat-recovery heat exchanger and a glycol tank to the conventional ejector IHX system components, as shown in Figure 6. The system comprises a compressor, which is followed by the placement of a glycol heat-recovery heat exchanger. An additional ambient air-source gas cooler is present for the purpose of cooling the refrigerant. The refrigerant, now at a low temperature and a high pressure (stream 4), enters the internal heat exchanger. The gas that has undergone partial cooling represents the primary ejector flow (stream 5). The ejector outlet is led into the flash separator (stream 6), where the upper vapor phase (stream 10) functions as the cold internal heat exchanger inlet. The heated vapor, presented by stream 1, is then led back into the compressor. The liquid from the flash (stream 7) is throttled in a valve, and stream 8 is fed into an evaporator, where it is evaporated by the glycol solution. The secondary ejector flow is then entrained into the ejector, represented by stream 9. The glycol solution circulates back into the glycol tank and is heated in the heat-recovery heat exchanger, which is placed after the compressor. The study experimentally evaluates the performance of the system for simultaneous heating and cooling applications. The researchers found that the inclusion of an IHX resulted mainly in lower ejector efficiency, pressure lift, and entrainment ratio. The system COP at its peak value reached almost six and was, on average, 13.4% higher than without the IHX. The cooling COP was, on average, improved by 11.5%. To support the previous statement [42], the increase of the evaporator temperature increases the COP, while the gas cooler temperature decreases it. Accordingly, a water-cooled gas cooler must be selected for the heat pump system to avoid lower evaporator temperatures and higher gas cooler outlet temperatures. The system has good implementation options in simultaneous heating and cooling applications and air conditioning.

Fan et al. [51] proposed a novel design of an ejector-enhanced internal auto-cascade heat pump cycle for district heating in cold regions. The scheme introduces a cascade-heat exchanger to an ejector-based IHX compression cycle, as illustrated in Figure 7. The selected refrigerant is a zeotropic mixture of R32 and R290, which is directed into the compressor (stream 1). As a high-temperature and high-pressure vapor (stream 2), it is condensed in the condenser and releases heat for the supply of water. Stream 3 is led into the separator via an expansion valve (stream 4). At this point, the mixture is split into a bottom liquid (rich in R290) and an upper vapor phase (rich in R39). Stream 10, which is rich in R290, is throttled in a valve and then partially evaporated in the cascade heat exchanger. The partially evaporated R290-enriched flow, designated as stream 12, functions as the primary flow in the ejector. The R39-rich vapor phase (stream 5) is condensed in the cascade heat exchanger and then additionally cooled in the internal heat exchanger. Subsequently, the stream undergoes throttling in the upstream valve before entering the evaporator (stream 8). It then enters the ejector as the secondary flow. The refrigerant exiting the ejector (stream 13) is evaporated in the internal heat exchanger and then enters the compressor as stream 1, thereby completing the cycle.

The findings of this study support the findings of previous research, which indicated that an increase in evaporation and a decrease in condensation temperature would result in a higher COP of the system [42,50]. The COP improvement between the cycles decreases with the evaporation temperature and increases with the condensation temperature. The comparison of this cycle with a conventional heat pump shows that, for the evaporation range from −25 °C to 5 °C, there is a 9% improvement in COP at the upper end of the range, rising to 19% at the lower end of the evaporation temperature range. A similar trend is observed in the volumetric heating capacity, with an increase of 37% at the evaporation temperature of −25 °C and a 12% increase at 5 °C. In general, the novel ejector-enhanced internal auto-cascade heat pump cycle shows significant performance enhancement for district heating applications, with a higher COP and volumetric heating capacity and a lower compression ratio.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A2. Main findings of ejector-enhanced vapor compression cycle with an internal heat exchanger (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Yang et al. [43]	E	Ejector-expansion vapor compression cycle with an IHX (transcritical)	System has up to 31.0% improvement in COP and was developed for use in electric vehicles.
Wang et al. [45]	T		System majorly decreases exergy destruction in throttling valves and enhances exergy efficiency.
Zhu et al. [46]	E		High efficiency with a COP reaching 4.6, showing an improvement of 10.6% compared to the corresponding conventional cycle.
Mateu-Royo et al. [47]	T	Parallel compression configuration with an economizer and an IHX	Proposed an enhanced IHX scheme with the possibility of using environmentally friendly refrigerants.
Qin et al. [48]	T	Ejector-expansion vapor compression cycle with an IHX	Proposed system for simultaneous heating and cooling with an overall higher COP.
Guruchethan et al. [50]	E	Ejector-enhanced compression IHX cycle integrated with heat recovery heat exchanger	The inclusion of an IHX resulted mainly in lower ejector efficiency, pressure lift, and entrainment ratio. Potential implementation in simultaneous heating and cooling applications.
Fan et al. [51]	T	Ejector-enhanced compression IHX cycle with an implemented cascade heat exchanger	Developed system using a zeotropic mixture has significant performance enhancement for district heating applications.

. This subsection investigated the impact of the addition of an IHX into the system. A significant reduction in exergy destruction was complemented by a lower entrainment ratio and pressure lift but also lower ejector efficiencies. Nevertheless, the overall exergy efficiency and COP reached higher values (up to 31.0%), thereby enabling the utilization of environmentally friendly refrigerants. Some systems were tested for use in electric vehicles and for simultaneous cooling and heating applications with positive and encouraging results. It is important to note that the majority of the tests were conducted theoretically. Therefore, a decrease in COP improvement should be taken into account when planning real-life applications.

2.3. Ejector-Enhanced Multiple-Heat Sink/Source Vapor Compression Cycle

To improve the efficiency of heat pump systems, the change from single-stage to multiple-stage evaporation or condensation needs to be considered. Studies show that two-stage evaporation systems improve the efficiency of heat pumps [52]. In this section, the impact of multiple-stage evaporation/condensation on ejector-based compression cycles will be evaluated.

Jing et al. [53] modified a conventional flash tank vapor injection heat pump. Figure 8 shows the proposed design of the ejector-enhanced flash tank vapor injection heat pump cycle with two evaporators for dryer applications. A similar cycle without the drying drum was already suggested by Cao et al. [54]. The cycle introduces an additional high-temperature evaporator, an expansion valve, and a drying drum. The addition of the high-temperature evaporator reduces the irreversible loss caused by the lower temperature difference by utilizing the ejector’s entrainment ability. The refrigerant (stream 1) enters the compressor and is compressed to a high-temperature and high-pressure state, represented by stream 2. As it passes the condenser, it emits heat to the air used in the drying drum. The refrigerant at the condenser outlet (stream 3) is directed into the ejector as the primary flow. Similarly, as in the systems in Section 2.1, the refrigerant is directed into a flash tank (stream 4) where it is separated into the bottom liquid (stream 5) and the upper vapor phase (stream 1). The liquid refrigerant is split into two distinct streams and fed via expansion valves into two different evaporators. Stream 6 is throttled to a lower pressure than stream 8. The higher-pressure evaporator feeds the secondary flow of the ejector. The lower-pressure evaporator is followed by a compressor, represented by stream 7. The warm humid air at the drying drum outlet is, consequently, led through both evaporators before returning to the condenser. A comparison of the novel cycle with a conventional flash tank vapor injection heat pump shows an improvement in the COP and the volumetric heat capacity of 18.88% and 66.99%, respectively. The COP decreases with the condensation temperature. As the temperature increases from 55 °C to 75 °C, the COP moves from 5.81 to 4.39. Cao et al. [54] found a COP improvement of 12.1% and 22.0% compared with a conventional ejector cycle and a basic refrigeration cycle, respectively. Additionally, it was found that increasing the temperature in any of the evaporators also increases the COP. This was also found in [42,50,51]. The exergy efficiency analysis showed that the application of ejectors can effectively reduce the exergy destruction in the expansion valves. A 75.98% reduction in exergy destruction was noted in the new cycle compared to a conventional flash tank vapor injection cycle. In the new cycle, the compressor contributed the most towards exergy destruction and, therefore, has the highest potential for optimization. This claim is supported by the exergy analysis study on a similar cycle with a single-stage evaporation [55].

Lu et al. [56] studied a similar cycle to the ejector-based heat pump cycle presented by Fingas et al. [38] and proposed a modification. Three cycles were compared: the basic vapor compression heat pump cycle with a throttle valve and a basic ejector-expansion heat pump cycle. Figure 9 shows the modified dual-ejector-expansion heat pump cycle. The refrigerant (stream 1) is directed into the compressor where it is compressed into a high-temperature and high-pressure state (stream 2). It passes through the condenser and enters two distinct ejectors as the primary fluid (stream 3). The two-phase refrigerant at the outlet of both ejectors (stream 4) is mixed and, as stream 5, directed into the first evaporator. Subsequently, the refrigerant is led into a separator, where the vapor phase moves back into the compressor. The bottom liquid phase (stream 7) is separated into stream 8 and stream 10 via corresponding valves. Both streams are throttled and eventually evaporated in evaporators 2 and 3, respectively. The evaporator outlets represent the secondary flow into the ejectors. The main findings of this study suggest that the newly implemented modified dual ejector-expansion heat pump cycle with triple-evaporation outperforms the basic cycles. Due to the use of dual ejectors, a lower compression ratio and outlet temperature could be used because of the expansion work recovery. The COP achieved by the novel cycle at T_c = 50 °C and T_E = −20 °C was 3.57, which is a 26.1% and 10.5% improvement compared to the basic compression and single-ejector cycle, respectively. The volumetric heating capacity was likewise improved with the suggested cycle modification, namely a 40.5% and 13.6% increase regarding the basic compression and single-ejector cycle, respectively. Ejectors contribute significantly less towards exergy destruction than a throttle device, as already found by Jing et al. [53]. So, a more sustainable compression ratio could be used. Consequentially the exergy destruction in the dual-ejector cycle is reduced by 28.6% and 13.6% compared to the basic compression and single-ejector cycle, respectively. Considering the novelty of the design, further research needs to be conducted on the behavior of the cycle under different conditions, real-life implementations, etc.

The self-optimization of two-stage evaporation has been studied on a transcritical CO₂ ejector-expansion heat pump [57]. It was utilizing an extremum-seeking control-based approach to obtain the optimal discharge pressure. The system efficiency could be raised by 5.7% compared to the conventional system. The implementation of such techniques is crucial for the further improvement of systems.

Research has been carried out on a simultaneous separated condensation and cascaded evaporation-ejector compression heat pump cycle using zeotropic mixtures with low global warming potential [33]. The schematic representation of the system is presented in Figure 10. The system consists of 3 condensers—a sub-condenser, a high-temperature condenser and low-temperature condenser, and a high-pressure and low-pressure evaporator. As the refrigerant passes through the compressor, it is elevated to a high-temperature and high-pressure state, represented by stream 2. The zeotropic mixture is then partially condensed in the HT condenser, transferring the final heat to the hot water. The refrigerant at the outlet (stream 3) is directed into a flash tank, where it is split into two streams: the upper vapor phase (stream 4) and the bottom liquid phase (stream 9). The former passes through the LT condenser heating the city water entry. Subsequently, it is directed into the expansion valve (stream 5) and, as stream 6, leads into the LP-evaporator. The vapor phase refrigerant (stream 7) is fed into the ejector as the secondary flow. The bottom liquid refrigerant from the flash tank (stream 9) is used for the secondary city-water heating in the sub-condenser. The outlet (stream 10) is fed into the ejector as the primary flow. The ejector outlet (stream 8) is fully evaporated in the HP evaporator before entering the compressor. The heat source in both condensers is wastewater. The highest achievable COP of the proposed system is 6.67 and 6.73, with the mass fraction of CO₂ reaching 0.18 and 0.22 for CO₂/R1234ZEE and CO₂/R1234YF, respectively. A comparison with a conventional heat pump and a conventional ejector-compression heat pump shows an improvement of 69.0% and 30.7%, respectively, for CO₂/R1234YF as the refrigerant. The proposed system provides an alternative to support and enhance temperature matching between the condensation and evaporation processes. With this, the irreversibility associated with heat transfer is reduced. The large glide temperature zeotropic mixture introduced can extract more low-grade heat from the heat source, namely wastewater, and, therefore, produce a large amount of temperature-lifted hot water.

A modification to the separated condensation system was proposed by Wang et al. [58] and later by Zhao and Yu [59]. The former cycle shown in Figure 11 comprises a high-temperature and low-temperature condenser, an ejector, an evaporator, a flash tank, and expansion valves. The compressed superheated vapor (stream 2) is divided into the primary ejector inflow and the high-temperature condenser. At the ejector outlet (stream 6), the refrigerant is condensed in the LT condenser (low-temperature heat sink) and subsequently expanded in an expansion valve. The superheated vapor entering the high-temperature condenser forms the high-temperature heat sink and is expanded in an expansion valve before entering the flash tank (stream 4). The top vapor phase is fed back into the ejector as the secondary flow (stream 5), while the bottom liquid phase (stream 10) is expanded and mixed with the throttled low-temperature condenser outlet to form stream 9. This forms the evaporator inlet, where the refrigerant is evaporated and compressed to conclude the cycle. The working fluid for the cycle was the environmentally friendly R1234yf. The modified flash tank system working at standard operation conditions showed an improvement in COP from 2.79 to 3.49 compared to a conventional flash tank cycle. In addition, the suggested system outperforms the conventional cycle in COP, volumetric heating capacity, and exergy efficiency by 24.93%, 24.92%, and 38.84%, respectively. The system provides a solution when a higher temperature source is needed, since the cycle saves more energy due to the application of the ejector.

The sub-cooler system of Zhou and Yu [59] comprises a vapor-injection cycle with an ejector for air-source heat pump applications augmented with sub-coolers, as shown in Figure 12. In addition to the standard components, the system incorporates a high-temperature and low-temperature condenser, as well as two sub-coolers instead of a flash tank, as proposed by Wang et al. [58]. The refrigerant at stream 1 enters the compressor and is compressed into a high-temperature and high-pressure vapor. The compressed superheated vapor, presented by stream 2, is divided into the primary ejector inflow and the high-temperature condenser. The ejector outlet (stream 11) is directed into the low-temperature condenser, where it creates a low-temperature heat sink. As the refrigerant enters the high-temperature condenser, it creates a high-temperature heat sink and is condensed. Stream 3, at the outlet, is throttled in an expansion valve where it is subsequently combined with the low-temperature condenser outlet. The expansion valve allows the low-temperature condenser pressure to be maintained. The combined stream (stream 5) represents the inflow to the first sub-cooler. The outlet is divided into two streams. The first passes through an expansion valve and represents the cold fluid inlet of the sub-cooler (stream 9). As it leaves the sub-cooler as the vapor phase, it enters the ejector as the secondary fluid (stream 10). The second stream then passes through a second sub-cooler (stream 6). At this stage, the outlet is also divided into two streams, with the first stream passing through a valve and representing the cold stream of the sub-cooler (stream 13). The second stream then passes through a final valve and then enters an evaporator (stream 8). The evaporator outlet (stream 1) and the cold sub-cooler outlet (stream 14) are connected to the compressor. The system was tested with different refrigerants, including R134a, R1234yf, R1234ze(E), and R32, at the same temperature of evaporation. It was found that R1234(E) is the best replacement for R134a. The results demonstrate that, when the proposed modified sub-cooler cycle is compared to a basic sub-cooler vapor injection, an increase in COP and volumetric heating capacity of 9.2–11.2% and 5.7–7.11%, respectively, is observed. The COP of the novel system ranged from 3.48 to 4.76, with a condensation temperature span of 60–80 °C and an evaporation temperature range of 2–18 °C.

The studies provided a good basis for further experimental studies to further improve the efficiency of the proposed heat pump cycle. Experimental verification and validation are still needed.

Jing et al. [60] designed an ejector heat pump cycle with the two parallel condensers presented in Figure 13. The cycle adds a generator, a booster(compressor), and a low-temperature condenser to the usual ejector-based heat pump. In the generator, the high-pressure refrigerant absorbs heat and enters the ejector as the primary flow (stream 5). At the ejector outlet (stream 1), the refrigerant is split into two streams. The first part passes through the low-temperature condenser, where it is cooled and condensed. The second part of the refrigerant is pressurized in the booster and enters the high-temperature condenser as superheated vapor (stream 2). The cooled refrigerant at the high-temperature condenser outlet (stream 3) is divided into two streams. The first is expanded in a valve and, as stream 6, combined with the low-temperature condenser outlet after passing through an expansion valve (stream 8). The combined streams form the evaporator inlet (stream 9) and then the ejector secondary fluid (stream 10). To complete the cycle, the second part of the high-temperature condenser outlet circles back to the generator via a pump (stream 5).

Comparing the proposed novel system with a basic booster-assisted ejector heat pump, it shows improvement in all aspects. The researchers provided two COP values: COP_hw and COP. The former includes the heat absorbed in the generator, while the latter does not. Operation under typical conditions showed that the modified cycle improved the COP_hw and COP by 50.34% and 16.97%, respectively. In addition, the total exergy destruction was reduced by 12.70%, but the greatest potential still lies in the optimization of the ejector.

Zou and Yu [61] studied the basic ejector-enhanced heat pump cycle with different refrigerants. They also introduced the sub-cooler vapor injection heat pump cycle and a combined ejector-enhanced sub-cooler vapor injection heat pump cycle for water-heating purposes in cold regions. The latter represents a novel design illustrated in Figure 14. The refrigerant (stream 1) is compressed into superheated vapor. The saturated refrigerant at the condenser outlet (stream 2) is fed into the ejector as the main driving flow. The ejector is followed by a flash tank, which separates the refrigerant’s vapor phase (stream 5) and uses it to cool the refrigerant compressed by the compressor to decrease the discharge temperature of the compressor. The flash tank’s bottom liquid phase (stream 6) is additionally cooled in the sub-cooler. At the sub-cooler outlet (stream 7), the refrigerant is split into two streams. It is partially expanded and led into the sub-cooler (stream 9) to evaporate and function as the secondary ejector fluid (stream 10). Partially, it is expanded and directed into the evaporator (stream 8). The evaporated refrigerant is cycled back into the compressor (stream 1). To compare the novel system to the basic ejector-enhanced heat pump cycle and the sub-cooler vapor injection heat pump cycle a thermodynamic model based on the energetic, exergetic, economic, and environmental (4E) analysis was developed. Two different refrigerants, R134a and R290, were compared. The latter was proposed as a more sustainable alternative. The systems were compared at different operating conditions within a condensation temperature range of 50–80 °C (75 °C), evaporation temperature range of −40–0 °C (−30 °C), and temperature difference of sub-cooler cold end of 5 °C. The proposed system showed improvements in heating COP and volumetric heating capacity. Compared to the basic ejector-enhanced heat pump cycle and the sub-cooler vapor injection heat pump cycle, there is an increase in COP of 27.6% and 5.3%, respectively, and an increase in volumetric heat output of 42.0% and 21.2%, respectively. Furthermore, the exergy destruction of the expansion valves is lower in the ejector-based heat pump cycles, as already discussed by Lu et al. [56]. The novel system cost per unit of exergy production is 21.6% and 5.1% lower than that of the basic ejector-enhanced and the sub-cooler vapor injection cycle, respectively. The use of R290 resulted in a 9.9% reduction in carbon emissions for the life cycle’s climate performance, but its flammability must be considered. The authors confined themselves to simulations and modeling and did not carry out experiments on an actual heat pump. Hazard assessment and validation of the system will be essential for future real-world applications.

Another similar study was conducted by Wang et al. [62], comparing an ejector-enhanced vapor injection cycle with a basic vapor injection cycle for low ambient temperatures. The system is shown in Figure 15. This study considered a design similar to Zou and Yu [61] with the addition of a second flash tank, but without a sub-cooler. Three different refrigerants were considered, namely R22, R290, and R32. The former is known for its negative impact on the environment, in particular, its high global warming potential and ozone-depleting characteristics [63]. Thus, two alternatives are considered. The results indicate that the proposed system outperforms the basic injection cycle with each refrigerant, with an increase of 6.0–8.4%, 7.3–10.2%, and 6.7–8.2% for R22, R290, and R32, respectively, in the volumetric heating capacities under given working conditions observed. Furthermore, the COP follows a similar trend of outperforming the basic vapor injection cycle, with R290 being the favored refrigerant, reaching a COP value above five. For R22, R290, and R32, an increase in COP of 2.6–3.1%, 3.2–3.7%, and 2.9–3.1% was compared with those of the basic cycle. Although these improvements seem small, they are compared to another injection cycle. Comparing the results achieved by Zou and Yu [61], a similar improvement in comparison to the injection cycle was noted. It must be stated that the results are based on a model and were not obtained on a real-life unit.

Two novel designs utilizing two condensers operating at different temperatures were proposed, as shown in Figure 16 [64,65]. The cycle comprises a compressor, double condensers, an evaporator, an ejector, an expansion device, and a tee valve. The refrigerant (stream 1) is compressed and passes through the initial condenser, which serves as a high-temperature heat sink, and thereafter enters the ejector as the primary flow (stream 3). The ejector outlet (stream 4) serves as the inlet for the second condenser, providing a medium-temperature heat sink. Subsequently, the condensed refrigerant (stream 5) is expanded through an expansion valve and split into the secondary ejector flow (stream 8) and evaporator inlet (stream 7). As it passes through the evaporator, it is evaporated and cycled into the compressor inlet (stream 1). In the second alternative, the stream is divided into two parts after entering the evaporator, namely the secondary ejector flow and the evaporator inlet. For the working fluid, pure R134a and two ternary azeotropic mixtures were considered. The first one was a blend of R600a + R1234ze(E) + R13I1, and the second one was R1234yf + R134a + R152a [64]. A comparable system to the second alternative was previously examined by Liu et al. [66]. The researchers introduced an additional flash tank between the first condenser and the ejector, as presented in Figure 17. The superheated vapor (stream 2) passes condenser 1, creating a high-temperature heat sink. At the outlet (stream 3), it is directed into the newly introduced flash tank. The upper vapor phase (stream 4) is introduced into the ejector as the primary fluid, while the lower liquid phase (stream 11) is expanded through a valve (stream 12) and mixed with the ejector outlet (stream 5). The mixed stream (stream 6) represents the second condenser inlet. As the refrigerant passes through the second condenser, it is partially condensed. At the outlet (stream 7), it is expanded with an expansion valve and directed into the evaporator. The vaporized refrigerant at the evaporator outlet (stream 9) is then partially directed into the ejector (stream 10) as the secondary flow and partially led into the compressor (stream 1). The refrigerants considered were R717, R1234yf, R134a, and R290.

A comparison made of the heating COP and exergy efficiency indicates that the second alternative performed the best of the studied alternatives [64,65]. The best working fluid in all systems of hot water production was the R600a + R1234ze(E) + R13I1 blend, followed by R134a and the blend R1234yf + R134a + R152a. The COP for the different refrigerants in comparison to the first alternative and conventional cycle was improved by 30.30–52.38% and 53.85–92.75%, respectively. The exergy efficiency of different refrigerants was enhanced by 14.95–15.49% and 20.76–21.58%, respectively [64]. In the case of Liu and Lin [65], the first alternative produced a 20.2% higher COP and 19.5% higher volumetric heating capacity compared to the conventional cycle. The second alternative outperformed the first one, with a 36.5% increase in COP and a 37.2% increase in volumetric heating capacity.

Researchers conducting the flash tank alternative found similar improvements [66]. The comparison was made to a conventional dual-temperature air-source heat pump and found that, with R1234yf as the working fluid, the COP of 2.80 could be increased to 4.17 using the proposed novel scheme. A positive increase in the volumetric heating capacity of 67.22% was also found. The researchers concluded that the evaporation temperature raises the COP but also has a bigger impact on the system than the condensing temperature. To verify and validate the results for real-life implementation, a practical system needs to be considered.

Li et al. [67] proposed a novel system to solve some of the problems dual-temperature air-source heat pumps are facing. The proposed system introduces a dual-temperature air-source heat pump cycle with a method for self-defrosting capable of simultaneous heat-source production at different temperatures. The system can be operated in two alternatives, as shown in Figure 18. A thermodynamic model was used to analyze the system with four possible refrigerants, namely R134a, R600a, R290, and R1234yf. The operating principle of the system is based on the compression of the working fluid in the compressor (stream 1 to stream 2). At the three-way valve, the superheated vapor is split into stream 3 and stream 13. The former is condensed in condenser 1, creating a high-temperature heat sink. The outlet (stream 4) is expanded and is, as stream 5, led into another three-way valve. Meanwhile, stream 13 enters the ejector as the primary flow and is, as stream 12, mixed with stream 5 in the valve. As stream 6 passes condenser 2, it creates a lower-temperature heat sink in comparison to condenser 1. The exiting refrigerant (stream 7) enters a four-way valve and, in alternative 1, enters evaporator 1 first. It is led into evaporator B via an expansion valve. The refrigerant exiting the second evaporator (stream 10) is split into the secondary ejector flow (stream 11) and compressor inlet (stream 1). In this alternative, evaporator 1 is in the defrosting process, and evaporator 2 is in the heat absorption process. In the second operation mode, evaporator A is in the heat absorption process, and evaporator B is in defrosting mode. The researchers found that, in comparison to a traditional single-temperature heat pump, the system outperforms in all measured aspects. The heating COP and exergy efficiency were improved by 20.72–44.47% and 29.70–49.19% respectively. The best performance was measured with R600a as the working fluid.

The study demonstrates the potential for the implementation of the novel ejector-enhanced dual-temperature air-source cycle, due to a significant improvement in the energy utilization rate and a reduction in temperature fluctuation during defrosting. However, further validation on a practical, real-life unit is still required.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A3. Main findings of ejector-enhanced multiple-heat sink/source vapor compression cycles (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Jing et al. [53]	T	Two-stage evaporation	Significant exergy destruction reduction and performance improvements. Compressor optimization is required for further reduction of exergy destruction.
Lu et al. [56]	T	Dual ejector-expansion heat pump cycle with triple evaporation	Performance improvement for air-source heat pump system application in cold regions. Dual ejectors allow a lower compression ratio and outlet temperature because of the expansion work recovery.
Liu et al. [33]	T	Separated condensation and cascaded evaporation	The large glide temperature zeotropic mixture introduced can extract more low-grade heat from the heat and therefore produce large temperature-lifted hot water.
Wang et al. [58]	T	Dual-temperature air-source heat pump	Due to energy savings, the system provides a good alternative for when a high-temperature heat source is needed.
Zhao and Yu [59]	T	Parallel condensers and sub-cooling	Performance improvement and implementation of environmentally friendly refrigerants.
Liu et al. [66]	T	Dual-temperature air-source flash tank heat pump cycle	The evaporation temperature raises the COP but has also a bigger impact on the system than the condensing temperature.
Li et al. [67]	T	Dual-temperature air-source heat pump with self-defrosting	Significant improvement in energy utilization rate and reduced temperature fluctuation during defrosting.

. The introduction of multiple heat sinks/sources improved the overall heating and cooling COP of the proposed systems. The systems demonstrate an optimal energy utilization rate and are capable of extracting low-grade heat due to the utilization of large temperature glide zeotropic mixtures as refrigerants. A notable reduction in exergy destruction was observed. However, further optimization of the compressor is necessary. A dual-ejector configuration with triple evaporation and a dual-temperature air-source heat pump were identified as a promising single-stage solution for high-temperature heat demands and utilization in cold climates, due to their energy savings and expansion work recovery. A decline in experimental studies is noted, due to the increased complexity of the cycles.

2.4. Ejector-Enhanced Solar-Assisted Vapor Compression Cycle

In this section, solar-assisted ejector cycles are discussed.

Zeinad Sajjadi et al. [68] proposed a new configuration for a solar-assisted ejector-enhanced vapor compression-based cycle. Different refrigerants are discussed and compared to a classic ejector-expansion refrigerant cycle. The novel concept is presented in Figure 19. An evacuated tube solar collector (ETSC) was utilized to superheat the refrigerant between the compressor outlet (stream 2) and condenser inlet (stream 3) with solar energy. The rest of the cycle follows a conventional ejector-enhanced injection cycle. Stream 4 enters the ejector as the primary flow and is then led into a flash tank. The vapor phase (stream 1) is circulated back into the compressor, while the liquid phase (stream 6) is expanded and evaporated before being directed into the ejector as the secondary flow (stream 8). Under the same conditions, the basic and solar-enhanced cycles were compared, whereas for the latter, the researchers found a cooling COP improvement of 21.06%. The solar-enhanced cycle operated at a COP of 4.016 compared to 3.408 for the basic cycle. However, the thermal leakage coming from the ETSC needs to be given attention. The rise in solar radiation energy induces a decrease in exergy efficiency and a bigger thermal leakage. When comparing two ejector-based cycles, the improvement in exergy efficiency or exergy destruction will not be as evident as when compared to a throttle device cycle. The ETSC contributes to 51.46% of the exergy destruction. Additionally, in the modified cycle, the condenser exergy destruction increased by 59.53% compared with the regular ejector cycle. This is the result of the rise in heat dissipation from the condenser to the environment to cool the refrigerant from the higher temperatures

Further research on solar collectors and their decrease in exergy destruction and experimental studies developing the practical implementation of the system need to be conducted.

A novel ejector-enhanced solar-air dual-source heat pump cycle was investigated, as shown in Figure 20. Li et al. [69] utilized solar energy for both electricity and heat generation. The proposed design allows for operation at a higher heating capacity during weak solar radiation and low ambient temperature. The solar input leads to an increase in the enthalpy of the ejector, thereby improving the expansion processes and mixing that occur in the ejector. In addition, the ejector booster function reduces the required refrigerant mass flow rate, allowing the compressor to operate at lower power consumption. The PVT can be utilized to generate power for use in the compressor, while simultaneously transferring heat from the solar collector. This enables operation at higher temperatures and increases system efficiency. The refrigerant at the condenser outlet (stream 3) is divided into two parts, the high-pressure part of the solar PVT collector (stream 4) and the low-pressure part of the air-source evaporator (stream 8). The pressure is regulated by corresponding expansion valves. After the PVT collector, part of the flow is fed into the ejector as the primary flow (stream 5), while the secondary flow is fed partially from the other evaporator outlet (stream 9). The combination of the ejector outlet (stream 6) and the partial flows after both evaporators (streams 5 and 9) are fed into the compressor as stream 1. The inverter has the capacity to store the generated electricity in batteries or to convert the direct current into an alternating current. This enables the inverter to drive the compressor or to export the electricity to the power grid. For the given operating conditions and refrigerant R134a, the proposed system achieved a COP and exergy efficiency of 4.271 and 20.46% respectively, an increase of 25.96% and 25.83% compared to an air-source heat pump. A reduction in electricity demand from the grid was also observed, specifically 51.96% when using 20 m² PV. The use of more environmentally friendly refrigerants is of paramount importance, so different refrigerants were investigated. The results show that the refrigerant R152a has the highest COP of 4.489 and exergy efficiency of 21.05% while being environmentally friendly.

The intermittency of solar energy continues to be one of the main problems for system stability. The system cannot be regulated by adjusting the heat absorption of the collector or evaporator if there is insufficient solar energy. It is very important to further research adjustable ejectors for the case of stable operation. A reliable practical implementation of such systems still poses massive challenges due to the limitations of different climatic conditions and the time of limited solar energy.

Zou and Yu [70] developed a solar-assisted ejector-enhanced vapor injection heat pump cycle. Figure 21 shows the novel solar-assisted system. To a conventional system, it introduces a pump and generator at the condenser outlet to increase the pressure of the refrigerant and raise the enthalpy at the primary flow of the ejector. In the previous study [69], this was discussed, as it reduces power consumption. The superheated vapor at the compressor outlet (stream 2) is condensed. The liquid condenser outlet (stream 3) is directed into the generator via a pump (stream 4). The pump and generator raise the pressure and enthalpy of the refrigerant, enhancing the work it can perform as the primary flow as it enters the ejector (stream 5). The ejector outlet is then separated into the vapor phase (stream 7) and liquid phase (stream 8) in the flash tank. The liquid phase is expanded and led into the evaporator as stream 9. The evaporator outlet (stream 1) is split into the secondary ejector flow and compressor inlet, while stream 7 is added for intermediate compressor cooling. The system operating at the optimum intermediate pressure achieves a COP of up to 5.44, compared to 3.93 of a normal flash tank vapor injection cycle. This is an improvement of 38.4%. The effective use of solar energy with the introduction of an ejector allows the heating capacity to be increased by 89.2%.

The intermittency of solar energy remains a major negative aspect of solar-based systems. If there is not enough solar energy, the collector cannot directly regulate the system. It is impossible to adjust the heat absorption of the collector/generator. In this case, an adjustable ejector must be installed, as discussed by [69]. Further research into practical implementation with current limitations and challenges is required.

Zou et al. [71] proposed a similar design to Zou and Yu [70]. After the condenser, there is a pump and then a solar collector acting as a generator, as shown in Figure 22. The system employs warm and humid air from the drying chamber to facilitate the evaporation of the refrigerant. Subsequently, the high-temperature heat sink from the condenser is utilized to both dry and heat up the air before it is recycled back into the drying chamber. The study compared the solar-enhanced cycle with a conventional ejector-expansion heat pump cycle. The cycles were compared with the R134a refrigerant, and an increase in COP of 23.64–39.82% was observed over the entire condensing temperature range. The heat capacity increase was found to be from 32.79 to 33.96%. Due to the high GWP of R134a, alternative refrigerants were considered, including R32, R1234ze(E), R1234yf, and R1233zd(E). The highest COP of 5.40 was achieved with R1234yf, which is an environmentally friendly refrigerant. This COP value is in agreement with Zou and Lu [70], who achieved 5.44 with a similar cycle. The exergy analysis shows that, in agreement with Zeinad Sajjabi et al. [68], the solar collector/generator is the main problem causing exergy destruction. When the solar radiation energy reaches 1750 W, the collector is responsible for 84.8% of the total exergy destruction. In this sense, the solar collector has the greatest potential for optimization and heat leakage and should be considered in practical applications.

Experimental research studying the transient response characteristics of a similar system using a double-pipe exchanger for simulating the actual solar collector was conducted [72]. The results of the comparison demonstrate that the replacement of R134a with R1234yf and R1234ze(E) has a minimal effect on the system’s dynamic characteristics. In addition, a subcooling control method on the adjustable ejector with the PID (proportion integration differentiation) was conducted [73]. The researchers managed to reduce the power consumption of the subcooling controller by 6.24% and 6.99% compared with those with two different ejector throat areas. The proposed subcooling method achieved satisfactory system-improvement results. Due to a lack of experimental studies, this research provides data to support the control-oriented design and practical application of the presented system [72]. The incorporation of ejectors has the potential to greatly enhance the utilization of clean energy from solar sources in solar-enhanced heat pumps, thereby providing a novel direction for research in the field of heat pump drying. [71].

Yu and Yu [74] compared three cycles to evaluate the impact of a hybrid solar air source for drying applications. A classic vapor injection cycle with a flash tank was used as the base. The second cycle was modified by adding an ejector and a solar collector. The latter is added between the condenser and the ejector, as already proposed by Zou and Lou [70] and Zou et al. [71]. The third and novel cycle presented in the study suggests that the solar-assisted collector/evaporator is placed upstream of the secondary flow inlet of the ejector, as shown in Figure 23. The working principle is similar to a standard ejector-expansion vapor compression cycle shown in Figure 2. The difference is that the liquid phase from the separator (stream 6) is split into two evaporators via expansion valves. Stream 7 is evaporated and then, as stream 1, circled into the compressor as usual. Meanwhile, stream 8 is evaporated using solar-assisted collectors to feed the secondary ejector flow (stream 9). The advantages of this arrangement are that the entrainment ability of the ejector increases the vapor quality at the flash tank and at the same time the input solar energy is transferred from the solar collector to the condenser, thus improving the cycle performance. The proposed system achieved a COP of 5.08 under the given conditions. The researchers found that the COP of the novel cycle is improved by 36.6% and 2.0%, respectively, compared to the conventional vapor injection cycle and the solar-assisted ejector-enhanced injection cycle when operating at the optimum intermediate pressure. Furthermore, the exergy destruction cost of the proposed cycle is 13.2% lower compared to the vapor injection cycle, while [68] found that the solar collector in this cycle contributes to 51.46% of the exergy destruction. The cycle is favorable when operating with environmentally friendly refrigerants, such as R32, R1234yf, and R1234ze(E), as an alternative to the high GWP R134A.

A similar layout of the enhanced vapor injection cycle heat pump discussed above was previously studied by the same researchers [75]. The solar collector is placed between the condenser and primary flow into the ejector, as evident in Figure 24. The superheated vapor (stream 2) passes through the condenser and is, at the outlet (stream 3), split into two streams. Partially it is expanded (stream 4) and directed into the solar collector via a pump (stream 5). Stream 6 enters the ejector as the primary flow. The other part of the condenser outlet is directed into the sub-cooler. At the outlet, it is split into streams 9 and 10 via two expansion valves, respectively. Stream 10 circulates back into the sub-cooler and, as stream 11, enters the ejector as the secondary fluid. Stream 9 evaporates and, at the outlet as stream 1, enters the compressor. As in the previous study, the solar-enhanced cycle outperformed the basic sub-cooler vapor injection and the ejector-enhanced sub-cooler vapor injection cycles, with an improvement in COP of 23.8% and 19.0%, respectively. In contrast, exergy destruction is most evident in the solar collector, which is expected [68].

Comparing the two studies from Yu and Yu [74,75], it is evident that the former compared exergy destruction costs and the latter just exergy destruction. This means more work needs to be conducted studying the cost of solar collection exergy destruction properties and how they affect the overall system.

Fan et al. [76] proposed a modified heat pump water heater using a zeotropic mixture of R290/R600a. The study introduced an additional sub-cooler and a zeotropic mixture of two refrigerants as opposed to a single refrigerant in the whole cycle. The system is presented in Figure 25. The refrigerant at high-temperature and high-pressure (stream 2) is condensed in the condenser. Stream 3 enters the flash where the top vapor phase (stream 8) circles back into the condenser. Depending on the solar intensity available during operation, the system has two operating alternatives, namely Mode A/B for high/low solar radiation intensity conditions, respectively. In operation mode A, the solar collector is operating, while in mode B, stream 4 bypasses the collector. In both operating modes, stream 5 enters the ejector as the primary flow. The ejector outlet flow is fully evaporated in the sub-cooler using stream 9. The sub-cooled refrigerant at the sub-cooler outlet (stream 10) is expanded and directed into the evaporator (stream 10). The vaporized stream 11 is fed into the ejector as the secondary fluid. The alteration of the design comes after the ejector outlet, which is heated by a sub-cooler that simultaneously cools the condensed refrigerant before it enters the evaporator.

The results of the simulations show favorable results with the suggested cycle modification, which is expected due to the similarity with other evaluated solar modified cycles. The system performs best when in mode-A operation. As the condenser inlet refrigerant saturated temperature is increased from 45 °C to 80 °C, while maintaining a constant 10 °C evaporator outlet temperature, an increase in COP in the range of 28–33% and 22–24% is observed in comparison to the conventional heat pump and mode-B operation, respectively. Similarly, mode-A and mode-B operation could increase the volumetric heating capacity by 22–47% and 3–13%, respectively, in comparison to the conventional heat pump. The exergy destruction of the system has not been evaluated.

A waste heat solar-driven ejector-compression heat pump for simultaneous heating and cooling applications has been investigated, as illustrated in Figure 26 [77]. The ejector is included in the cooling applications and uses a PV/T panel as a waste heat source for the generator. A waste heat exchanger is placed to improve the system performance by using the waste heat from the condenser. The refrigerant at the condenser outlet (stream 8) is partially led into the generator (stream 13) via a pump and partially into the flash tank (stream 9). After the generator, two heat exchangers are placed. The heat exchangers in the system are separate and operate depending on the lift temperature. At a certain lift temperature or too low solar intensity, the second heat exchanger actuates (stream 15). The primary ejector flow (stream 16) is the outlet at the second heat exchanger and the secondary flow comes from the vapor phase of the flash tank (stream 12). Three different cities with different climates and two refrigerants were considered for the experiments. In the absence of solar intensity, the COP increased from 3.7 to 4.0 by using waste heat recovery. The overall cooling COP is improved by 7% using R450a compared to a conventional vapor compression system using R134a as the working fluid.

Figure 27 presents the concept proposed by Al-Sayyab et al. [78] that also utilizes waste heat recovery with a PV/T generator. At the condenser outlet (stream 3), the refrigerant is led into a separator, where the refrigerant leaves as a saturated liquid (stream 5) and a saturated vapor (stream 4). The saturated liquid expands through an expansion valve into a mixture of liquid and gas (stream 6). Thereafter, streams 6 and 4 enter the evaporative condenser. The primary stream leaves the evaporative condenser in a mixture state (stream 7) at generator pressure. In the generator, it is evaporated by the cooling water heat provided by the PV/T waste heat recovery coil and then injected into the ejector as the primary flow (stream 8). The secondary stream is provided as the bottom outlet from the evaporative condenser (stream 9) and is expanded through a valve (stream 10) and then evaporated before being fed into the ejector (stream 11). An extensive exergy destruction analysis was carried out. The compressor showed the largest exergy destruction source of the whole system. This is not in line with previous research findings, where the solar collectors were the primary exergy destruction source. The researchers argue that, with further optimization, 59.4% of the system’s exergy destruction can be avoided.

An ejector-enhanced solar-driven cascade heat pump cycle with an ejector sub-cycle and vapor compression sub-cycle is introduced to improve air-source heat pump cycles in low ambient temperature conditions [79]. Figure 28 represents the schematic overview of the system. The system can be split into two sub-cycles, namely the vapor compression sub-cycle, using R1234yf as the refrigerant, and the cascade sub-cycle, using R134a, R1234yf, or R141b as the working fluids. The refrigerant is initially compressed (stream 2) into a high-temperature and high-pressure state, subsequently undergoing condensation in the condenser (stream 3), thereby generating a high-temperature heat sink for water heating. The condensed refrigerant is then expanded through an expansion valve prior to entering the intercooler (stream 4), where it absorbs heat from the fluid produced by the ejector. This process results in the complete evaporation of the refrigerant into a vapor state, which is then circulated back into the compressor (stream 1), concluding the compression sub-cycle. As the ejector outlet vapor undergoes partial condensation in the intercooler, it is split into two streams. One portion is expanded in an expansion valve before passing through an evaporator (stream 8). Subsequently, the vapor phase is fed into the ejector as the secondary flow (stream 5). The other part of the intercooler outlet passes through a pump (stream 9), whereby the high-pressure liquid refrigerant is evaporated in the following collector–generator. The high-temperature and high-pressure vapor at the outlet represents the primary ejector flow (stream 10).

The results indicate that the proposed system significantly increases the COP and system exergy efficiency. Under the same conditions, the cascade system reached a COP of 4.20, which is a 45.8% increase compared to a conventional air-source heat pump cycle. Furthermore, an improvement of 12.0% in exergy efficiency was observed. The temperature of evaporation has a positive influence on the COP and exergy destruction. Despite the potential for energy savings afforded by the cascade cycle, the researchers emphasized the necessity of studying the system characteristics under actual operating conditions prior to real-life implementation.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A4. Main findings of ejector-enhanced solar-assisted vapor compression cycles (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Zeinad Sajjadi et al. [68]	T	Solar ejector-enhanced compression refrigeration cycle	The rise in solar radiation energy also leads to greater thermal leakage and a decrease in exergy efficiency.
Li et al. [69]	T	Solar-assisted dual-source air heat pump cycle	A reduction in electricity demand from the grid was observed, specifically 51.96% when using 20 m2 PV. The system cannot be regulated by adjusting the heat absorption of the collector or evaporator if there is insufficient solar energy.
Zou and Yu [70]	T	Solar-assisted vapor injection heat pump cycle	The intermittency of solar energy remains a major negative aspect of solar-based systems. If there is not enough solar energy, the collector cannot directly regulate the system.
Zou et al. [71]	T	Solar ejector-enhanced compression heat pump cycle	The system is suitable for drying applications, but the heat leakage should be considered in practical applications. The solar collector has the greatest potential for optimization.
Chen et al. [72]	E		This research provides data to support the control-oriented design and practical application of the presented system.
Yu and Yu [74]	T	Ejector-enhanced flash tank vapor injection heat pump cycle with hybrid solar–air source	The entrainment ability of the ejector increases the vapor quality at the flash tank, and at the same time, the input solar energy is transferred from the solar collector to the condenser, thus improving the cycle performance.
Al-Sayyab et al. [78]	T	Solar-assisted ejector-enhanced flash tank vapor injection cycle with an integrated evaporative condenser	The compressor showed the largest exergy destruction source of the whole system. The researchers argue that with further optimization 59.4% of system exergy destruction can be avoided.
Li et al. [79]	T	Ejector-enhanced solar-driven cascade heat pump cycle with two sub-cycles	Increase of 45.8% in COP and a 12.0% decrease in exergy destruction in comparison to a conventional air-source heat pump.

. Solar-assisted cycles significantly outperformed the conventional cycles. When solar energy can be fully utilized, the COP is increased by over 40% compared to a conventional heat pump system. Furthermore, the utilization of solar panels can diminish the necessity for electricity from the grid. One of the primary challenges associated with solar energy is its intermittency. Additionally, the increase in solar radiation energy results in elevated thermal leakage and a reduction in exergy efficiency. Nevertheless, the overall exergy destruction of solar-assisted systems is lower than that of conventional cycles. It is important to note that the majority of research in this field is theoretical or uses alternative methods to simulate solar panels. The existing literature indicates that solar-assisted systems can provide an effective enhanced compression cycle, provided that the challenges associated with real-life implementations can be adequately addressed.

2.5. Dual-Ejector Vapor Compression Cycle

In this subsection, systems utilizing dual ejectors will be presented.

A transcritical R744 refrigeration cycle with a dual parallel ejector heat pump system for electric vehicles is proposed [80]. The heating cycle is similar to the ejector heat pumps studied previously [38]. Represented by the red streams in Figure 29, the heating mode functions as follows. The vapor phase refrigerant (stream 1) is compressed into a high-temperature and high-pressure (stream 2) state and then split in a three-way valve into the in-cabin gas cooler. The refrigerant is then injected into the ejector for heating as the primary flow, represented by stream 3. The ejector outlet (stream 4) is separated in a separator where the top vapor phase from the separator is sent to the compressor (stream 1). The bottom liquid refrigerant (stream 5) is expanded and directed into an out-cabin heat exchanger (stream 6), which functions as an evaporator. The vapor phase working fluid is, as stream 7, supplied to the ejector as the secondary fluid. In the refrigeration cycle, the refrigerant is passed through the three-way valve at the compressor outlet (stream 2) to the out-cabin heat exchanger, where it is cooled. Afterward, it is supplied into the cooling ejector as the primary flow, designated as stream 3. Likewise, as with the heating cycle, the ejector outlet is separated into the bottom liquid (stream 5) and top vapor phase (stream 1) in a separator. The liquid bottom phase is expanded (stream 6) and led into the in-cabin evaporator and consequently, as stream 7, cycled into the ejector as the secondary flow.

The study provided important insights into ejector-enhanced cycles for electric vehicles. The cooling COP for the optimized cooling ejector, in comparison to a conventional transcritical CO₂ refrigeration cycle and single-ejector transcritical CO₂ cycle, is improved by 17.32–23.42% and 7.31–9.47%, respectively. In the heating mode, the COP with the optimized heating ejector, in comparison to the single-ejector system, is improved by 18–19.79%, while the unoptimized single-ejector system shows a decrease of 0.07–2.43%. It is evident that the introduction of the dual-ejector system improves the performance and adaptability of the cycle for implementation in electric vehicles. Fixed ejectors are limited by working conditions, which hinders their use under various operating conditions.

Fingas et al. [81] proposed a two-phase ejector alternative to their ejector-based heat pump presented above [38]. The system in Figure 30 operates on the same principle as an IHX cycle [43]. The difference between the systems is that this alternative has two parallel ejectors, a low-temperature and a high-temperature one.

The results of the system were compared to a conventional IHX heat pump. The proposed two-phase ejector system improved the COP from 4% to 38% in comparison with the conventional IHX system working under similar conditions. A more significant increase was observed in the heating capacity. The introduction of the ejector led to a reduction in the pressure ratio of the compression, resulting in an increase of up to 90% in the heating capacity. The researchers argue that, with next-generation ejectors, like the bypass-type or micro-channel heat exchangers that are designed for this application, the COP improvements will be significantly increased.

Liu et al. [82] proposed a system based on a conventional ejector-enhanced flash tank vapor compression cycle. It introduces a dual heat-source air-source vapor compression cycle; the heating mode operation is presented in Figure 31. The refrigerant (stream 1) is compressed in the compressor into a high-temperature and high-pressure state (stream 2). If the system is in heating mode, it is split into the condenser and ejector inlet. In the condenser, a high-temperature heat sink is created, and the refrigerant is condensed. Simultaneously, it passes through an ejector, and both outlets (streams 5 and 11) are mixed. Stream 6 passes another condenser to create the second heat sink. It is then led through an IHX and, as stream 9, into the air-source evaporator. At the outlet, the stream is partially injected into the ejector as the secondary flow and partially in the four-way valve back into the IHX. There it completes the cycle by being sucked into the compressor. In cooling mode (b), the four-way valve leads the refrigerant through the evaporator (stream 3). At the evaporator outlet (stream 4) the stream is split into the primary ejector flow and IHX inlet. There, it subsequently undergoes condensation in two stages. An energetic, exergetic, economic, and environmental (4E) analysis performance was conducted comparing the novel system to a conventional dual heat-source air-source heat pump system. The results show that the novel system significantly outperformed the conventional. The cooling COP was improved by 17.7–28.9%, the cooling capacity per unit volume rose by 18.1–36.7%, and the exergy efficiency was increased by 13.4–35.1%. Among other things, the maintenance and operating costs, as well as the CO₂ emissions, have been reduced. The study provides a good base for future optimization and application studies.

Figure 32 illustrates the dual-ejector refrigeration cycle proposed by Monda and De [83]. A low-grade heat-driven ejector is introduced to reduce the power input to the compressor. Two environmentally friendly refrigerants, R32 and R1234yf, were considered. The working principle of the system can be described as follows. The refrigerant (stream 1) is compressed in the compressor into a high-temperature and high-pressure state (stream 2) and directed into the ejector as the secondary fluid. At the ejector outlet (stream 3), it is condensed and split into two distinct streams. One part is directed into the heater via a pump (stream 5) and is vaporized before entering the ejector as the primary flow (stream 6). The continuation of the bottom half of the scheme is a classical ejector-enhanced vapor compression cycle with a flash tank, as discussed before. The results show that by utilizing the heat-driven ejector and the compressor, the COP improvements are 25.7% and 37.2% compared to a conventional ejector-expansion refrigeration cycle for R32 and R1234yf, respectively. It was found that an optimum compressor ratio for each working fluid exists corresponding to which COP is at maximum.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A5. Main findings of dual-ejector vapor compression cycles (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Zoe et al. [80]	T/E	Parallel dual-ejector transcritical R744 refrigeration cycle	Improved COP for implementation in electric vehicles.
Fingas et al. [81]	E	Parallel dual-ejector vapor compression	Significant COP improvement of up to 38% compared to a conventional IHX heat pump.
Liu et al. [82]	T	Dual-ejector dual-heat source/sink vapor compression heat pump cycle	The maintenance and operating costs as well as the CO2 emissions have been reduced. The study provides a good base for future optimization and application studies.
Mondal and De [83]	T	Dual-ejector vapor compression refrigeration cycle	Utilizing a heat-driven ejector improves the COP. In addition, an optimum compressor ratio for each working fluid exists corresponding to which COP is a maximum.

. The literature on dual-ejector systems is very limited compared to other single-stage systems. The implementation of a dual-ejector cycle enhances the recovery of expansion work and enables the system to operate in a cooling or heating mode. Theoretical and experimental studies have demonstrated that the incorporation of a second ejector can enhance the system by up to 38% compared to a conventional cycle. The potential use of such systems in electric vehicles was presented, given that fixed ejectors are constrained by operational parameters, which hinders their application across a range of operating conditions. It was observed that the proposed dual ejector enhances the cooling and heating cycle performance. Furthermore, an economic analysis showed that such systems provide a good base for future optimization and application studies.

3. Ejector-Enhanced Two-Stage Vapor Compression Cycle

This section is dedicated to reviewing the latest novel concepts of two-stage compression-ejector-based heat pump cycles. Conventional two-stage vapor compression heat pumps are designed to operate more efficiently at a greater temperature differential between the evaporator and condenser, thereby enhancing their performance in cold climates. [84].

Three simple modifications (System E1, E2, and E3) to the conventional two-stage compression heat pump cycle (System C1) were introduced and compared [85]. The modified cycles are presented in Figure 33. Each system’s efficiency and performance were evaluated for heating and cooling applications. For the working fluid, the refrigerants R152a, R1234yf, R32, and R290 were considered. The initial alternative, (a.) E1, introduces an ejector between the flash tank and the condenser. The primary ejector flow is the condenser outlet (stream 5), while a split valve is installed at the evaporator outlet (stream 1) with the purpose of partially feeding the low-pressure compressor and simultaneously providing the secondary ejector flow. The ejector outlet (stream 6) is led into the flash tank, where the top vapor phase (stream 9) mixes with the low-pressure compressor outlet and enters the high-pressure compressor as stream 3. The bottom liquid phase (stream 7) is throttled and evaporated in the evaporator. The second system, (b.) E2, introduces the ejector at the same position and with the same primary and secondary flows. The alternation comes in the flash tank and low-pressure compressor. The low-pressure compressor outlet is being directed into the flash tank. The flash tank bottom liquid phase is likewise expanded and evaporated, while the top vapor phase goes directly into the high-pressure compressor without mixing. The last system, (c.) E3, introduces an internal heat exchanger. As the condensed refrigerant is injected into the ejector as the primary flow, the outlet is led into the internal heat exchanger. There it enters as the cold fluid and is led into the flash tank. The bottom liquid phase is throttled and evaporated in an evaporator before being injected into the ejector as the secondary fluid. The top vapor phase is sucked into the low-pressure compressor. The compressed refrigerant is cooled in the internal heat exchanger and fed into the high-pressure compressor.

For the presented system, the refrigerating COP and heating COP were calculated. This review is focused on the heating application of the systems. For the refrigerating mode, the R32 applied in system E3 is recommended. For the heating applications with the conditions of T_c = 70 °C and T_e = −5 °C, a maximum COP of 3.48 is obtained in systems E1 and E2 using R152a. For the same systems, a COP of 3.35 is obtained with R134a, 3.31 with R290, and 3.16 with R1234yf. In addition, the exergy efficiency was the highest in systems E2 and E3, while the exergy loss was the lowest. In conclusion, the proposed systems have good potential in heating applications when using R152a, R1234yf, and R290 refrigerants. Compared to C1 and E3, the heating COP of systems E1 and E2 using R152a is improved by 3.0% and 4.5%, respectively.

Cao et al. [86] developed a novel two-stage compression heat pump system that is enhanced with a booster and an ejector. A more recent study developed a mathematical model to analyze the system from a thermodynamic point of view and subsequently optimized it using an elitist multi-objective non-dominated sorting genetic algorithm [87]. The system shown in Figure 34 consists of a condenser that receives the superheated working fluid at high pressure and temperature (stream 4). The working fluid considered for the study was a zeotropic mixture of R600 and R143a. At the condenser outlet (stream 5), it is divided into two streams. It, then, partially passes through an expansion valve (stream 12) and enters the sub-cooler as the cold fluid. There it is heated (stream 13) and mixed with stream 2. The other part of the condenser outlet passes through the sub-cooler as the hot fluid before entering the ejector as the primary fluid (stream 6). The ejector outlet is then separated into liquid (stream 8) and vapor phases (stream 1) in the separator. The liquid phase is expanded and vaporized in an expansion valve before passing through the newly introduced booster and entering the ejector as the secondary fluid (stream 11). The vapor phase, which leaves the separator with a different mass fraction than the liquid phase, is compressed in the lower compressor (stream 2) and mixed with the cold sub-cooler outlet. The mixed stream forms the high compressor inlet (stream 3).

The system shows improvements in all aspects of the observed operating conditions. An improvement of 25% in COP was noted when compared to the conventional two-stage system. An exergy analysis was conducted and showed that the highest exergy destruction was within the ejector, which also was already commented on by Jing et al. [60]. As already mentioned in previous studies, the condenser temperature decreases the COP, but it was found that the exergetic efficiency and heating load increase. Considering the multi-objective NSGA-II [87], the optimal values of COP, exergy efficiency, and unit cost of the product are 2.78, 25.9%, and 113.57 USD/GJ respectively.

A study was conducted by Shifang et al. [88] to investigate the potential of a multi-heat sink heat pump based on a two-stage ejector-compression cycle. The proposed system, presented in Figure 35, operates with two distinct types of heat sinks: the high-temperature hot water ranging from 60 °C to 90 °C and the low-temperature water ranging from 30 °C to 40 °C. The results were obtained with a validated theoretical model and compared to a conventional two-stage compression multi-heat sink heat pump cycle. The scheme comprises two condensers, a cooler, an evaporator, two compressors, a valve, and an ejector. The low-temperature and low-pressure refrigerant is sucked into the low-pressure compressor, where it is compressed to a higher pressure and temperature (stream 2). Subsequently, the refrigerant is cooled in the cooler to the saturated vapor state, producing a low-temperature heat sink. At the cooler outlet (stream 3), the fluid is divided into two streams, one that is mixed with the ejector outlet and one that is directed to the high-pressure compressor inlet. The latter is compressed into a high-pressure and high-temperature state (stream 4) prior to condensation in the first condenser, whereby it produces a high-temperature heat sink. Following this, the refrigerant is injected into the ejector as the primary flow (stream 5). The partially split stream at the high-temperature compressor inlet is mixed with the ejector outlet (stream 6) and fed into the second condenser (stream 7), where it produces a low-temperature heat sink. Furthermore, the refrigerant is throttled in a valve (stream 9) and subsequently undergoes evaporation in the evaporator. A portion of the outlet forms the secondary ejector flow, while the remaining portion is sucked into the low-pressure compressor, thereby completing the cycle.

A comparison between the modified and conventional systems is presented. Three different refrigerants were considered, including R134a and two environmentally friendly alternatives, R290 and R1234yf. With the obtained results, it is evident that the proposed cycle outperforms the conventional two-stage compression cycle. The COP values for the novel system are 4.13, 4.37, and 4.44 for R1234yf, R290, and R134a, respectively. Meanwhile, the conventional system reached 3.83, 4.11, and 4.20 for R1234yf, R290, and R134a, respectively. The exergy analysis was also favorable for the ejector-enhanced system, with exergy destructions of 0.119kW, 0.109kW, and 0.105kW for R1234yf, R290, and R134a, respectively, as the conventional system reached 0.144kW, 0.127kW, and 0.122kW for R1234yf, R290, and R134a, respectively. The newly researched refrigerants were found to be good environmentally friendly alternatives with similar performances.

Yu et al. [89] proposed a new closed-loop drying system, demonstrated in Figure 36. The experimental investigation of a solar-assisted ejector-enhanced heat pump drying system utilizing the R134a refrigerant is presented herein. In order to simulate a solar collector within the experimental environmental chamber, a hot water cycle system was employed. The experimental setup is capable of operating in two distinct working modes, namely the modified ejector-enhanced and the conventional heat pump drying mode. The modified system comprises a low-pressure compressor and a high-pressure compressor, an ejector, a condenser, an evaporator, valves, and a double-pipe exchanger, which is used to simulate the solar panel. As the saturated vapor enters the low-pressure compressor (stream 1), a rise in temperature and pressure occurs. Subsequently, the mixture is combined with the ejector outlet (stream 8) and enters the high-pressure compressor (stream 3). The high-pressure and high-temperature vapor phase is directed to the condenser, where it undergoes condensation. The condensed working fluid (stream 5) is then split and throttled in a corresponding electronic expansion valve and fed into the double-pipe exchanger (stream 6) and the separator (stream 9). The refrigerant at the heat exchanger outlet is subsequently passed into the ejector as the primary fluid (stream 7). The remaining portion of the stream is separated in the separator into a liquid (stream 11) and a vapor phase (stream 10). The upper vapor phase represents the secondary fluid flow within the ejector, whereas the lower liquid phase is expanded and then evaporated in the evaporator. The refrigerant, which has undergone evaporation, completes the circuit by being drawn into the low-pressure compressor. When the system is operating in conventional heat pump mode, the solenoid valves bypass the ejector.

Under the given operating condition, the proposed new system always yields a higher moisture extraction rate and COP compared to the conventional system. Under identical operational parameters, the modified cycle demonstrated a COP of 2.95, whereas the conventional system yielded a mere 2.80. Furthermore, the modified cycle resulted in a higher moisture extraction rate with a reduced drying time. Nevertheless, higher exergy destruction has been noted in the modified system due to the higher input energy and drying temperature.

Figure 37 shows a modification of the conventional two-stage transcritical CO₂ compression heat pump proposed by Liu et al. [90]. The study introduces a modified transcritical CO₂ ejector-enhanced two-stage compression cycle with an IHX for the production of domestic hot water. The cycle comprises a low-pressure and high-pressure compressor, an internal heat exchanger, a heat exchanger, a condenser/gas cooler, an ejector, and an evaporator. The low-pressure and low-temperature refrigerant, CO₂, is sucked into the low-pressure compressor prior to supplying heat in the internal heat exchanger (stream 2) for intermediate heating of city water. Subsequently, the CO₂ is compressed into a high-temperature and high-pressure state (stream 3) within the high-pressure compressor (stream 4). As it passes through the condenser, heat is provided for the hot city water. The condensed CO₂ (stream 5) enters the ejector as the primary fluid. The ejector outlet (stream 6) is cooled in a heat exchanger, which serves as the first stage of the city-water heating process. The refrigerant (stream 8) is then expanded and enters the evaporator, where it undergoes the process of evaporation. At the outlet (stream 1), the flow is split into two distinct streams, namely the secondary ejector flow and the low-pressure compressor inlet.

The novel system surpassed the conventional two-stage compression cycle in all evaluated aspects, including the COP, system heating capacity, exergy destruction, and exergy efficiency. The COP under the same operating conditions of the new cycle reached 3.7, with the conventional system operating at 3.3. The proposed system improved the heating COP, system heating capacity, and exergy efficiency by 12.1%, 15.7%, and 15.3%, respectively, and the total system exergy destruction was reduced by 5.1%. The largest exergy consumer was in both systems the throttling valve. It amounted to 46.87% and 26.7% for the conventional and ejector-enhanced systems, respectively. This study supports the knowledge gap on ejector-enhanced two-stage compression heat and promotes further improvement of the transcritical CO₂ two-stage compression cycle for domestic hot water production.

As discussed in Section 2.3, multiple-stage evaporation systems improve the efficiency of heat pumps. Furthermore, multiple-stage evaporation in the two-stage compression cycle needs to be considered. Due to the lack of two-stage compression ejector cycles, a major contribution was made by Sarkar [91]. The study from 2017 is still one of the only papers on this type of system with a dual-ejector multiple-evaporation setup. The researchers presented two conventional three-evaporation two-stage compression refrigeration systems and four ejector-enhanced alternatives. In Figure 38, the most efficient of the studied alternatives is presented. As the refrigerant in stream 3 is compressed, it is fed directly into the condenser. The condensed working fluid (stream 5) is fed into the ejector as the primary flow. The ejector outlet (stream 6) is separated into the vapor phase (stream 3) and liquid phase (stream 7). The former cycles back into the high-temperature compressor, while the bottom feed is split into evaporators 1 and 2 through the corresponding valves. Stream 8 is evaporated and mixed.

The analyses and optimization results are consistent with the previous findings that the ejector-enhanced system shows better performance. The presented, best-performing alternative gives a 19.3% cooling COP and 24.3% volumetric cooling capacity improvement compared to the conventional two-stage compression cycles, a 66.8% cooling COP and 18.6% volumetric cooling capacity improvement compared to the ejector-enhanced single-stage compression cycle, and a 116.7% cooling COP and 69.1% volumetric cooling capacity improvement compared to a basic single-stage compression cycle. The difference between the proposed ejector-enhanced two-stage compression cycles was small, with a maximum deviation of 4%. The results clearly favor the implementation of the proposed cycles. Heating application analyses of the systems were not carried out.

A brief overview of the main findings of this subsection is presented in Appendix A,

Table A6. Main findings of Ejector-Enhanced Two-Stage Vapor Compression Cycle (E—experimental; T—theoretical).

Source	Method	Design Concept	Findings
Lie et al. [85]	T	Ejector-enhanced two-stage vapor compression heat pump cycles	The proposed systems have good potential in heating and cooling applications and can use environmentally friendly refrigerants: R152a, R1234yf, and R290. Compressors contributed the most towards exergy destruction.
Cao et al. [86]	T	Ejector-enhanced two-stage compression heat pump with sub-cooling	An improvement of 25% in COP was noted when compared to the conventional two-stage system. An exergy analysis was conducted and showed that the highest exergy destruction was within the ejector.
Shifang et al. [88]	T	Ejector-enhanced multi-heat sinks two-stage compression cycle heat pump	The system shows lower, but similar COPs with environmentally friendly refrigerants compared to R134a.
Yu et al. [89]	E	Solar-assisted ejector-enhanced two-stage heat pump dryer system	The modified cycle resulted in a higher moisture extraction rate with a reduced drying time. Nevertheless, a higher exergy destruction has been noted in the modified system due to the higher input energy and drying temperature.
Liu et al. [90]	T	Ejector-enhanced multi-heat sinks two-stage compression cycle heat pump with an IHX	The proposed system improved the heating COP, system heating capacity, and exergy efficiency by 12.1%, 15.7%, and 15.3%, respectively, and total system exergy destruction was reduced by 5.1%.
Sarkat et al. [91]	T	Ejector-enhanced three-evaporator two-stage compression refrigeration cycle	The presented, best-performing alternative gives a 19.3% cooling COP and 24.3% volumetric cooling capacity improvement compared to the conventional two-stage compression cycles.

. Two-stage compression heat pumps are almost exclusively used in high-temperature heat pump systems. The systems are constructed around an LP and HP compressor. The utilization of two compressors allows for a reduction in the compression ratio, which, in turn, enhances the heating performance. The studies are just as limited as the dual-ejector vapor compression cycles. The COP, heating capacity, and exergy efficiency are all enhanced, and higher temperatures at the condenser outlet can be achieved. A 25% increase in COP is observed when compared to a two-stage compression heat pump, which would be even higher if compared to a single-stage. The highest exergy destruction is observed in the compressors, which represent the primary optimization target within the system.

The following section will present a concise analysis of the observed results, along with an examination of future prospects.

4. Discussion and Future Prospects

All gathered data are presented in Appendix A in the form of tables. Overall, in every study addressed in this review, an increase in performance was observed when an ejector was added to the system. It is very important to state the interpretation of the results. The biggest improvement in percentages was observed when comparing the ejector-enhanced system to a conventional heat pump or refrigeration cycle. In order to gain comparable improvement percentages, a comparison to the corresponding modified cycle without an ejector should be provided.

Despite the positive effects, the implementation of systems into real-life applications is very limited. Most experimental studies are based on basic single-stage ejector-expansion vapor compression heat pump and refrigeration cycles introduced in Section 2.1. A big decrease in experimental studies is found in the following subsections due to the increased complexity of the systems. Nevertheless, the experimental studies showed good results, even though the theoretical models provided better overall performance. Studies involving pilot plants highlighted some issues, such as higher exergy destruction in two-stage compression cycles, insufficient heating capacity, or lower ejector efficiency. Yet, they provide a good base to build on further research until the systems are ready for industrial application.

Two potential areas of research have been identified as offering the potential to address the current challenges. The first is the study of more advanced and efficient novel systems, with a view to identifying a suitable solution that can be implemented in real life. The second research option is the optimization of the current existing compression system and the step from theoretical models into pilot plants. As the majority of the reviewed literature is concerned with the theoretical investigation of novel design concepts, a more thorough and detailed testing procedure on existing systems is required. In practical applications, the system will not operate under optimal conditions, and therefore, a decrease in COP is almost certain. It is important to research the theoretical models on pilot plants for the whole operating condition range and its impact on performance.

By focusing on these areas of research, the gap between theoretical models and practical applications can be bridged, ensuring systems perform efficiently in industrial applications.

5. Conclusions

The impact of the implementation of ejectors in vapor compression heat pump and refrigerant cycles to enhance their performance and efficiency is presented. In the paper, the exploration of novel design concepts, rather than the optimization and improvement of existing systems was prioritized. The reviewed studies demonstrate that the incorporation of one or multiple ejectors into a compression heat pump or refrigeration system has a positive effect on performance, efficiency, working conditions, and, in most cases, exergy destruction. In more detail, the COP, volumetric heat capacity, and exergy efficiency and destruction were highlighted.

The improvement in the COP ranges from an incremental 3% to 4% improvement to 40%, with some cases demonstrating an increase of over 50%. The increase in the volumetric heating capacity reached slightly higher percentages on average. Moreover, the substitution of a throttling valve for an ejector significantly reduced the exergy destruction of the systems. It should be noted that most of the studies did not optimize exergy destruction, so further improvements in this field, as well as in COP and heating capacity improvements, are possible. Many studies found the single-stage vapor compression cycle to be a good alternative under transcritical operation. Additionally, the novel systems can utilize low-GWP refrigerants while still providing high efficiency. Potential applications for this technology include electric vehicles, air conditioning, drying processes, and simultaneous cooling and heating. When using solar collectors, the thermal leakage and intermittency of solar energy are causing major problems in their possible application. In addition, the exergy destruction of solar collectors needs further research.

Two-stage compression cycles have been demonstrated to exhibit high efficiency and outperform single-stage cycles, particularly in cold climates. Yet, the theoretical and experimental studies are far more limited compared to single-stage compression cycles. Further research is, therefore, needed to identify the accurate application of ejector-enhanced multi-stage compression cycles.

In general, while novel designs are being constantly developed and optimized, the technology still requires further development. The majority of studies on novel research are theoretical. While the models are refined and validated, they typically assume steady conditions. Although they can demonstrate the profitability of a system, further experimental validation is required before an ejector cycle can be implemented in real-life applications. Moreover, the impact of operating conditions needs to be considered when designing real-life implementations of the technology. System performance (COP, volumetric heating capacity, and exergy efficiency) could deteriorate when exposed to operating conditions outside of the theoretical model. Research focusing on the implementation of validated theoretical models into actual implementation is therefore needed.

This review demonstrated the superiority of ejector-enhanced compression cycles and hopes to pique interest in the further development of novel concepts and important research into the actual application of the systems.