Drop-In Replacement of R134a in a Household Refrigerator with Low-GWP Refrigerants R513A, R516A, and R1234ze(E)

Abstract

This study proposes the experimental evaluation of alternative refrigerants with low global warming potentials (GWPs) such as R1234ze(E), R513A, and the mixture R516A as a drop-in replacement for R134a in a domestic refrigerator with a volumetric capacity of 513 L. Initially, the optimal charge for each refrigerant was defined based on the minimum energy consumption of the refrigerator, then the thermal and energy performance of the refrigerator was evaluated. Finally, a total equivalent warming impact analysis (TEWI) was performed. The main results indicated that the optimal charge of the alternative refrigerants was below that corresponding to R134a (105 g), of which R516A (86 g) presented the greatest charge reduction. Regarding the average temperature of the refrigerator compartments, very adequate thermal conditions were observed; thus, the alternative refrigerants showed very similar conditions to R134a. For the coefficient of performance (COP) and considering R134a as a reference, it was observed that R513A presented the greatest reduction of around 28%, while R1234ze(E) showed an increase of 13% in relation to R134a. Finally, the TEWI analysis showed R1234ze(E) as the refrigerant with the least impact.

1. Introduction

Systems based on vapor compression technology for heating, air conditioning, and refrigeration (HACR) have worked with a wide range of refrigerants in the past. Given the fact that they have specific thermophysical and safety features suitable for certain applications, these fluids are crucial to the operation and design of refrigeration systems.

Hydrofluorocarbons (HFCs) are among these refrigerants and are widely utilized all over the world. Due to their high global warming potential (GWP) or high permanence in the atmosphere, they significantly contribute to the greenhouse effect [1]. For instance, R134a is one of the most often utilized as a working fluid HFC refrigerants in HACR because of its high efficiency. However, it is a fluid that significantly contributes to the greenhouse effect because of its high GWP of 1300 [2].

Primarily as a result of growing global warming, various agreements and regulations have been established in recent decades to lessen the environmental damage produced by refrigerants. Regarding this, the European Union intends to phase out 79% of HFC refrigerants by the end of 2030 by limiting their use if they have global warming potential values exceeding 150 [3] in HACR. In addition to the aforementioned, internationally speaking, the Kigali Amendment, which called for a decrease of 80 to 85% in HFC consumption and production by the middle of the twenty-first century, was passed in 2016 [4]. The roadmap for the implementation of this amendment was established in Mexico by the Ministry of the Environment and Natural Resources [5], and it defines alternative refrigerants in domestic refrigeration applications such as the use of R600a (a hydrocarbon (HC)), and R1234yf (a hydrofluoroolefin (HFO)), as alternatives to R134a.

Thus, the refrigeration industry has been forced to make significant changes towards the development and use of environmentally friendly refrigerants. In this sense, synthetic fluids such as HFOs have been considered as viable refrigerants to replace R134a. Where R1234yf is a pure HFO widely studied and used in HACR due to its low GWP < 1 and thermophysical properties such as R134a [6,7], there are also new mixtures (HFC/HFO/HC/R744) that have been evaluated and may be very attractive due to their low global warming potential compared to R134a [8,9]. Among the recent alternative mixtures to replace R134a, mainly HFC/HFO, are R513A and R516A [10], the latter being more attractive from an environmental point-of-view due to its low GWP of 131 compared to that for R513A of 573.

As mentioned above, there are several investigations on alternate refrigerants in the recent literature. For instance, an evaluation of a dual-evaporator ejector refrigeration system using R1234ze(E) and R516A was conducted by Iskan and Direk [11]. According to the experimental findings, R516A performed quite similarly to R134a, indicating that it can be employed in this system. Al-Sayyab [12], on the other hand, examined R513A, R516A, and R1234yf in a heat pump (cooling and heating mode) under different operating conditions. Their findings indicated that R513A performed similarly to R134a, whereas R516A using refrigeration systems needs to be modified to perform similarly to R134a. In a mobile air-conditioning system, Li and Tang [13] investigated binary mixes made from pure refrigerants such R134a, R1234yf, R1234ze(E), R32, R227ea, and R152a. In comparison to R134a, they were unable to find a mixture that performed better. Binary mixes can be modified to have better performance, though, by changing their chemical composition. The refrigerants R1234yf, R1234ze, R717, and R600a were used to hypothetically analyze a vapor compression system by Soni et al. [14]. From an energy efficiency standpoint, the results indicated that R600a was the best R134a substitute. In a study employing R1234ze(E) and its binary mixes, Ramirez-Hernández et al. [15] examined the efficiency of a vertical display refrigerator at controlled ambient temperatures. After analyzing the results, it was concluded that R1234ze(E), in addition to demonstrating a 6% energy boost, constituted a strong contender to replace R134a. In a heat pump system, the new refrigerant blend RGT2 (R134a/R1234yf/R161, 54%/43%/3% wt%) was theoretically and experimentally assessed by Liu et al. [16]. The findings demonstrated that RGT2 has a cooling capability comparable to R134a while having a 3.8% lower COP. When Sánchez et al. [17] examined various low-GWP refrigerants to replace R134a in a commercial refrigerator, they discovered that R290 and R1270 offered the greatest energy savings. while R1234yf consumed more energy than R134a. Between the various refrigerants, the refrigerator’s internal temperatures stayed remarkably constant. To determine the appropriate charge of R513A, Yang et al. [18] examined the performance of R513A and R134a in a household refrigerator. In accordance with the findings, the refrigerator’s energy usage when using R513A was 3.5% less than when using R134a. The thermal and energy efficiency of a household refrigerator using R513A was assessed by Belman-Flores et al. in their publication [19]. According to the findings, a refrigerator running on R513A provides the best conditions for food preservation and uses around 9% less energy than one running on R134a.

Based on the above, a continuous effort was observed in the development and evaluation of new low-GWP refrigerants in the refrigeration and air-conditioning industry. This aims to minimize the environmental impact by transitioning to more environmentally sustainable refrigerants without neglecting the energy consumption and thermal conditions for each application. On the other hand, due to the properties of the refrigerants and the design of each refrigerator, different energy and thermal results have been obtained. This allows for further exploration of the compatibility and performance of alternative refrigerants to replace conventional refrigerants, such as R134a.

In the case of Mexico, there is a progressive reduction in halogenated refrigerants according to the Kigali Amendment, in which domestic refrigeration ranked fifth in the consumption of HFCs, with approximately 2.07 MtCO₂eq, using R134a [5]. Despite the fact that R600a is mandated as the working fluid in new refrigerators, it is important to note that 88% of homes have at least one refrigerator and that most of them use R134a [20].

Therefore, the present study explores the use of low-GWP refrigerants as a direct replacement for R134a in a typical Mexican domestic refrigerator. Among the refrigerants to be studied, R1234ze(E) is proposed, which, although it has been evaluated [21], it is important to understand its behavior in domestic refrigeration in more detail. R513A is another one of the refrigerants analyzed in this work with a different refrigerator design from the one evaluated in [19]. The mixture R516A is also introduced as a novelty in the field of domestic refrigeration, where there is no solid knowledge about performance. This study aims to contribute to decision-making in the short and medium term for the replacement of R134a in refrigerators in use.

The objective of the study is to expand the information on the thermal and energy performance of alternative refrigerants in domestic refrigeration. In the different experimental tests carried out in this work under controlled chamber conditions, the effect of the refrigerant charge variation on some characteristic parameters was analyzed, as well as the determination of the optimal charge of each refrigerant. Finally, a comparison was made from the thermal and energy point-of-view between the refrigerants studied and a TEWI projection is presented.

2. Low-GWP Refrigerants Alternative to R134a

Because of their physical and chemical qualities, performance traits, and safety considerations, refrigerants utilized in vapor compression systems are recognized to be less than optimal [22]. However, the development of new refrigerant mixtures has led to improved selection conditions according to existing limitations in environmental regulations. Thus, in this work the azeotropic mixtures R513A (R134a/R1234yf) and R516A (R1234yf/R134a/R152a) were evaluated, the latter being a recently developed mixture whose performance is explored in this work. In addition, this work also evaluates the pure refrigerant R1234ze(E), which is considered as an almost immediate replacement for R134a.

Table 1. Properties of the replacement refrigerants for R134a.

	R134a	R513A	R516A	R1234ze(E)
Composition	-	R1234yf/ R134a	R1234yf/R134a/ R152a	-
Mass percentage	-	56/44	77.5/8.5/14	-
Boiling point [°C]	−26.07	−29.82	−29.35	−18.97
Critical temperature [°C]	101.06	94.91	96.65	109.36
Critical pressure [bar]	40.59	36.48	36.15	36.35
Liquid density [kg m−3]	1206.70	1134.20	1066.81	1163.10
Vapor density [kg m−3]	32.35	37.63	34.58	26.32
Latent heat [kJ kg−1]	177.78	156.35	164.01	166.92
Cp liquid [kJ kg−1 K−1]	1.43	1.41	1.46	1.38
Cp vapor [kJ kg−1 K−1]	1.03	1.05	1.09	0.98
Liquid conductivity [mW m−1 K−1]	81.13	69.93	70.09	74.22
Vapor conductivity [mW m−1 K−1]	13.83	14.03	14.38	13.59
Viscosity liquid [μPa s]	194.90	166.01	154.84	190.51
Viscosity vapor [μPa s]	11.69	11.62	11.42	12.23
GWP100	1300 a	573 b	131 b	1 a
ASHRAE class [24]	A1	A1	A2L	A2L

^a [2]; ^b [25].

defines and compares some of the main thermophysical properties and safety aspects of the refrigerants under study. The properties are estimated using the REFPROP 10 software [23] at a temperature of 25 °C.

Reviewing the table, most of the properties of the refrigerants under study are very similar. Among the properties, the one with the most notable variation is density, which directly affects the amount of refrigerant charge, a relevant aspect in the analysis of this work. The liquid density of alternative refrigerants is lower compared to R134a. For example, the refrigerant R516A presents the smallest decrease in density, approximately 11.6% with respect to R134a. Another important property to consider is the latent heat; the higher its value, the greater the heat absorbed, causing a high cooling production and a lower circulating mass flow. In this sense, it is observed that R134a is the fluid with the highest value in this property. One of the limitations in the use of refrigerants is the value of the GWP; refrigerants R1234ze(E) and R516A can be considered as long-term alternatives to R134a since they have a GWP of less than 150. R513A shows a 56% reduction in GWP compared to R134a, so it would be considered a short-term alternative. On the other hand, R513A is non-flammable (A1), similar to R134a. This is different from R516A, which is moderately flammable (A2L), like R1234ze(E). However, A2L-classified refrigerants are suitable for use in domestic refrigeration. In general, the refrigerants have similar thermophysical characteristics to R134a; in addition, the compatibility with the POE10 lubricant is adequate for alternative refrigerants, which motivates their evaluation and comparison in this work.

3. Experimental Refrigeration System

The experimental facility is equipped with a Mabe refrigerator model RMT1951X No-Frost type (automatic defrost), which has a total refrigerated volume of 513 L, of which 372 L corresponds to the fresh food compartment and 141 L to the freezer. Note that the refrigerator is designed to work initially with an R134a refrigerant and POE10 lubricant. The refrigerator consists of a spine-finned evaporator, where heat transfer is carried out by forced convection through a fan with a constant operating speed of 2500 rpm and a 290 W resistive element. It has a mounted multilayer tube-wire condenser operating by forced convection and, finally, a hermetic reciprocating compressor model EM3Z50HLT. It should be noted that this compressor is different from the one presented in the study [19], with the purpose of increasing the exploration and information on the performance of the domestic refrigerator working with alternative refrigerants. Figure 1 illustrates the refrigerator under study in this work: the freezer is located at the top and the fresh food compartment is at the bottom.

Table 2. Technical specifications of the experimental refrigerator.

Component	Characteristics	Component	Characteristics
Compressor	Hermetic reciprocating 115–127 V Frequency 60 Hz Capacity 164.2 W Displacement 4.5 cm3	Capillary tube	Copper material Length 2.44 m Internal diameter 6.6 × 10−4 m
Condenser	Forced convection Carbon steel material Number of layers 7 Total tube length 8.5 m	Evaporator	Forced convection Aluminum material Inverted-V tube array External diameter 9.5 × 10−3 m

defines the technical specifications of the main components.

3.1. Test Procedure

Determining the ideal refrigerant charge is crucial for the refrigerator’s thermal and energy efficiency. Therefore, the least energy consumption of the refrigerator is used to determine the ideal refrigerant charge in this study [26]. The charging process is carried out for all refrigerants, including R134a, since it works with a refrigerator that operates with a different compressor than the one that leaves the factory. To determine the optimal charge of each refrigerant, a variation of charges with increments of 10 g was established to evaluate the energy consumption of the refrigerator. The measurements of each charge were made with a digital scale with an uncertainty of ±0.01 g. The experimental charging process is specified in detail in [19]. For the charge evaluation, the refrigerator was kept inside a room whose temperature was maintained at 32.2 °C ± 0.6 °C and a relative humidity of 65%. The tests presented a duration until the refrigerator reached thermal stability in both compartments, which varies between each of the refrigerants evaluated, approximately between 8 and 10 h of testing.

Table 3. Refrigerant charge.

	Refrigerants
	R134a	R513A	R516A	R1234ze(E)
Charge [g]	90	70	70	70
	100	80	80	80
	110	90	90	90
	120	100	100	100
	130	110	110	110

defines the charges made for each evaluated refrigerant. Note that for R134a there is a different charge variation; this is because the tests began with this refrigerant, which allowed knowledge of the performance of the refrigerator and a perspective on the charge variation for the other refrigerants. Once the optimal charge for each refrigerant had been defined, the energy consumption of the refrigerator was evaluated under the official Mexican standard NOM-015-ENER-2018.

3.2. Instrumentation and Measurements

The proposed refrigerants have been evaluated using a fully instrumented refrigerator. While temperature was measured with K-type thermocouples, pressure was monitored using transducers with a measurement range of 0–25 bar. Additionally, measurements for the compressor’s energy usage were made using a digital wattmeter.

Table 4. Measured parameters and uncertainty.

Temperature	Pressure	Power	Charge
K-type thermocouples ±0.3 K	Pressure transducers ±1%	Digital wattmeter ±0.4 W	Digital balance ±0.01 g

defines the parameters measured in this work with their respective uncertainty.

Twelve thermocouples were installed inside the refrigerator compartments to evaluate the ideal charge. To prevent density and volume changes throughout the test, four thermocouples were put within wooden blocks to detect the temperature in the freezer. The remaining thermocouples were placed in 0.245 L plastic containers filled with a solution of 80% water and 20% glycol in the fresh food compartment. The thermocouples were properly distributed within the compartments (see Figure 1a). Once the appropriate charge for each refrigerant was defined, the refrigerator was instrumented under the NOM-015-ENER-2018 standard, where Figure 1b illustrates the distribution of the thermocouples, which were fixed in copper cylinders with a diameter of 2.9 cm and the same height. The test room conditions were the same as for the charge determination. It is worth mentioning that the climate room was certified. All tests in this study were performed without opening the refrigerator doors, with no food charge, and with the thermostat position (5/5).

The main components of the refrigerator were instrumented as well with thermocouples installed at the input and output of each piece of equipment to further study the behavior of the appliance. Five uniformly spaced thermocouples were inserted along the length of the tube in the heat exchangers, evaporator, and condenser. Thermal paste and dielectric tape were used to secure these thermocouples to the tube wall. In addition, the refrigerant pressure in the suction and discharge lines was also measured. The signals generated by each of the devices were stored in a computer through the Compact-RIO data acquisition system and processed in the LabVIEW graphical interface. Data collection was performed at two-second intervals for the duration of each test in this work.

4. Results and Discussions

This section shows the main results on the performance of the domestic refrigerator working with the refrigerants R513A, R516A, and R1234ze(E), compared to R134a. First, the effect of the refrigerant charge on some parameters of interest in the refrigerator operation was analyzed. Second, the thermal and energy behavior of the refrigerator was shown with the optimal charge of each refrigerant, and, finally, a TEWI analysis comparing the combined effect refrigerants was carried out. The data presented in this section correspond to the average results of two tests conducted with each refrigerant charge, as well as with the optimal charge.

4.1. The Effect of the Refrigerant Charge

The refrigerant charge is an aspect of great interest and study because of its effect on the performance of any refrigeration system [27]. Therefore, in this work, a study was conducted to determine the optimal charge of each refrigerant; in addition, the effect of the charge on some characteristic parameters of the domestic refrigerator was analyzed. Thus, Figure 2 shows the thermal behavior of the refrigerant along the length of the heat exchanger tube as the refrigerant charge varied. Each point shown in the figure represents the average temperature during the period of thermal stability of the refrigerator (see Section 3.1). In addition, five measurements are indicated in the figure with respect to the length of the tube: zero indicates the refrigerant inlet to the exchanger, half the length to position one, which indicates the refrigerant outlet of the equipment. For illustrative purposes, the figure shows the behavior for refrigerant R1234ze(E) and the behaviors for the other refrigerants evaluated in this work are similar.

Figure 2 shows a greater irregularity in the refrigerant temperature along the evaporator compared to the temperature in the condenser. For the case of the evaporator, in this study the degree of superheating was defined as the difference between the refrigerant temperature at position one (temperature at the outlet of the evaporator) and the temperature at position three-quarters. The higher refrigerant charge ensures that flow is present along the evaporator two-phase and not completely vapor, which makes the process inefficient. Note that with a smaller quantity of mass (70 g), the evaporator is almost empty. Therefore, a greater quantity of mass is required to ensure the correct operation of the evaporator. On the other hand, the degree of subcooling in the condenser is defined as the difference in temperature in position seven-eighths and position one (temperature at the outlet of the condenser), representing a minimum variation between the different charges.

Based on the above, the degrees of superheating and subcooling of each refrigerant with respect to the variation of the charge are shown in Figure 3. It is expected that the increased charge causes an accumulation of refrigerant in both the evaporator and the condenser. Regarding the degree of superheating, it has been found that the highest superheating is displayed at low refrigerant charges because the evaporator is emptier. In contrast, at higher charges, the evaporator overflows, which lowers the degree of superheating as well as the refrigerant temperature in the suction line. Comparing the refrigerants, it is observed that R513A has the highest degree of superheating. As for the degree of subcooling, this increases slightly with the increasing refrigerant charge, which can also cause a minimal increase in operating pressures. In addition, the cooling capacity of the system can be favored. Figure 3 shows that R134a had the lowest degrees of subcooling, even lower than the unit.

The evaporator with low refrigerant charge does not have a sufficient capacity to absorb heat, which can be reflected in the behavior of Figure 4, where the cooling capacity of the refrigerator increases as the refrigerant charge increases. This parameter is calculated based on experimental data (pressure and temperature) to estimate the enthalpy values at the inlet and outlet of the evaporator. This calculation considers an average of two compressor operations cycles during the period of thermal stability of the refrigerator. Note that R134a has the highest cooling capacity among the alternative refrigerants, and, in fact, has the largest latent heat favoring heat absorption (as shown in Table 1). For example, for a charge of 90 g, R134a represents an increase of 8.7% in its cooling capacity compared to R1234ze(E), whose latent heat is 6% lower than that of R134a. R513A has the lowest cooling capacity.

4.2. Methodology to Define the Optimal Charge

In this paper, the determination of the optimal charge for each refrigerant is defined in relation to the minimum energy consumption of the refrigerator. Thus, Figure 5 illustrates the energy performance of the refrigerator for each refrigerant charge variation. Each point represents the energy consumption of the refrigerator, which is calculated considering an ON/OFF cycle of the compressor, where the total time, t_total, is previously defined to estimate the number of cycles per day, N_cycles.

$${\mathrm{t}}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}={\mathrm{t}}_{\mathrm{O}\mathrm{N}}+{\mathrm{t}}_{\mathrm{O}\mathrm{F}\mathrm{F}}$$

(1)

$${\mathrm{N}}_{\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{e}\mathrm{s}}=\frac{1440}{{\mathrm{t}}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}}$$

(2)

From the equation above, 1440 corresponds to the minutes of a day. Therefore, the average consumption for a cycle is calculated by the following equation:

$$\mathrm{E}\mathrm{C}=\left(\frac{{\mathrm{W}}_{\mathrm{O}\mathrm{N}}·{\mathrm{t}}_{\mathrm{O}\mathrm{N}}+{{\mathrm{W}}_{\mathrm{O}\mathrm{F}\mathrm{F}}·\mathrm{t}}_{\mathrm{O}\mathrm{F}\mathrm{F}}}{60000}\right)\left(\frac{{\mathrm{N}}_{\mathrm{c}\mathrm{y}\mathrm{c}\mathrm{l}\mathrm{e}\mathrm{s}}}{\mathrm{d}\mathrm{a}\mathrm{y}}\right)$$

(3)

where 60,000 represents a conversion factor and W is the power consumed by the refrigerator. It should be noted that the power consumption of the refrigerator for each charge is estimated under the Mexican standard NOM-015-ENER-2018, whereby an interpolation is performed between the values obtained for power consumption and temperature at the FF and FZ for the damper positions (1/1, 5/5) of the refrigerator under study. The position 1/1 corresponds to an average temperature of −14.4 °C in the freezer and 7.2 °C in the fresh food compartment, while a temperature of −17.8 °C in the freezer and 2.8 °C in the fresh food compartment correspond to position 5/5. Equations (4) and (5) represent such interpolation, where the reported energy consumption corresponds to the maximum value between both equations.

$$\mathrm{E}{\mathrm{C}}_{3.9°\mathrm{C}}=\frac{\mathrm{E}{\mathrm{C}}_{\left(1/1\right)}-\mathrm{E}{\mathrm{C}}_{\left(5/5\right)}}{{\mathrm{T}}_{\mathrm{F}\mathrm{F},\left(1/1\right)}-{\mathrm{T}}_{\mathrm{F}\mathrm{F},\left(5/5\right)}}\left(3.9°\mathrm{C}-{\mathrm{T}}_{\mathrm{F}\mathrm{F},\left(5/5\right)}\right)+\mathrm{E}{\mathrm{C}}_{\left(5/5\right)}$$

(4)

$$\mathrm{E}{\mathrm{C}}_{-17.7°\mathrm{C}}=\frac{\mathrm{E}{\mathrm{C}}_{\left(1/1\right)}-\mathrm{E}{\mathrm{C}}_{\left(5/5\right)}}{{\mathrm{T}}_{\mathrm{F}\mathrm{Z},\left(1/1\right)}-{\mathrm{T}}_{\mathrm{F}\mathrm{Z},\left(5/5\right)}}\left(-17.7°\mathrm{C}-{\mathrm{T}}_{\mathrm{F}\mathrm{Z},\left(5/5\right)}\right)+\mathrm{E}{\mathrm{C}}_{\left(5/5\right)}$$

(5)

Figure 5 shows the experimental energy consumptions for each charge variation; in addition, there is a curve for each refrigerator that represents a quadratic regression. The minimum degree of correlation is of the 0.94 order and corresponds to the regression of R513A. Therefore, the appropriate charge for each refrigerator is that which corresponds to the minimum energy consumption according to the quadratic regression. Thus,

Table 5. Optimal charge of each refrigerant.

Refrigerant	Optimal Charge [g]
R134a	105
R513A	88
R516A	86
R1234ze(E)	94

defines the optimal charge found for each refrigerator; it is observed that the charge for alternative refrigerants is lower with respect to R134a. For example, a reduction in the refrigerant charge of 16% is obtained for R513A (very similar to that reported in [19]), 18% for R516A, and 10.5% for R1234ze(E). This reduction in the refrigerant charge is due to the lower density of alternative refrigerants in relation to R134a. This is contrary to what was reported in [21], where the refrigerator working with R1234ze(E) had a higher mass than R134a.

4.3. Refrigerator Performance Working with Alternative Refrigerants

Once the optimal charge for each refrigerant is defined, we proceed to evaluate the performance of the refrigerator by comparing the different refrigerants. The tests for this comparison were performed with the door of both compartments closed, without thermal charge, at position (5/5) of the thermostat, and with the chamber conditions defined in Section 3.

4.3.1. Temperature Abatement in Both Compartments

Figure 6 shows the initial thermal behavior of both refrigerator compartments for each of the refrigerants evaluated. The test was performed at a chamber temperature of 32.2 °C ± 0.6 °C and represents the thermal condition from startup to the first stop of the compressor. By a quick inspection it is observed that the refrigerants R513A and R516A have very similar thermal performance in both compartments regarding R134a, while R1234ze(E) shows a warmer condition that, in turn, involves a longer start-up time of the compressor.

In more detail, the abatement time obtained for refrigerant R134a was 5.94 h, in which a temperature of −23.71 °C was reached in the freezer and a temperature 3.26 °C in the fresh food compartment. The refrigerants R513A and R516A show very similar performances to R134a with an abatement time of 6.29 h and 6.25 h, respectively, with a slight decrease in temperature for both refrigerants, in the freezer of 0.5 °C and 1.2 °C in the fresh food compartment, compared to the values obtained with R134a. For the refrigerator R1234ze(E), the abatement period was 7.71 h, which represents an increase in the operating time of 30% compared to R134a. The temperature of the fresh food compartment for this refrigerant is −21.03 °C and 2.5 °C for the freezer.

4.3.2. Operating Pressures

The performance of the refrigerator operating pressures, suction, and discharge are shown in Figure 7. These pressures represent the stable condition of the ON/OFF cycling of the compressor, i.e., the normal operation of the appliance after the first start-up. By a quick inspection of the figure, it can be noted that R1234ze(E) presented a different behavior to other refrigerators. For example, the average suction pressure for R134a was 1.24 bar, while for R1234ze(E) it was about 1.06 bar, representing a 14.5% reduction in suction pressure. As for the refrigerants R513A and R516A, suction pressures very similar to R134a were observed. As for the discharge pressure, R1234ze(E) showed a decrease of about 26% with respect to R134a, while the discharge pressure of the mixtures was very similar and slightly higher than R134a. The behavior presented by the operating pressures of R1234ze(E) with respect to other refrigerants has to do with the lower density that this refrigerant has.

On the other hand,

Table 6. Compressor pressure ratio.

	R134a	R513A	R516A	R1234ze(E)
Pressure ratio	9.06	8.76	8.78	7.87

illustrates the compression ratio of the refrigerants evaluated, concluding that the alternative refrigerants required a lower compression ratio than R134a, which could be reflected in an increase in volumetric efficiency. Similar to the work presented in [18], R513A showed a slight reduction in the compression ratio compared to R134a.

4.3.3. Operating Temperatures

The suitable operation of domestic refrigerators is known to significantly extend the life cycle of perishable foods, including preventing the growth of certain types of bacteria. This performance is linked to the temperature, distribution, and air flow inside the compartments [28]. Thus, Figure 8 illustrates the thermal behavior in the fresh food compartment, TFF, and the freezer, TFZ, for each refrigerator during a thermal stability period of approximately 8 h of testing. In general, it was observed that the refrigerants exhibited adequate thermal behaviors for both compartments (see Figure 8a). The variability in the fresh food compartment was minimal between the alternative refrigerants with respect to R134a, while for the freezer, the refrigerant R1234ze(E) had a higher temperature than the rest of the refrigerants, which is reflected in Figure 6.

More specifically and considering the performance, Figure 8a,b shows the average temperatures in both compartments. The average temperature of the fresh food compartment for R134a was 3.11 °C, slightly higher than the temperatures of the other refrigerants, with R516A representing the lowest temperature with a decrease of 20% compared to R134a. For the freezer, R513A and R516A presented minimal differences of −0.3 °C with respect to R134a. While the refrigerator operating with R513A reflected a less cold condition for the freezer. However, the low-GWP refrigerants proposed in this study reflected very suitable thermal behaviors for domestic application. Note that the thermostat was at position (5/5) where, by calibration, the average temperature in the compartments was expected to be 2.8 °C and −17.8 °C for the fresh food and freezer compartments, respectively [26].

4.3.4. COP Analysis

In this study, an approximate COP was estimated by the ratio of enthalpy differences, as shown in the following equation:

$$\mathrm{C}\mathrm{O}\mathrm{P}=\frac{{\mathrm{h}}_{\mathrm{o}\mathrm{u}\mathrm{t},\mathrm{e}\mathrm{v}\mathrm{a}\mathrm{p}}-{\mathrm{h}}_{\mathrm{i}\mathrm{n},\mathrm{e}\mathrm{v}\mathrm{a}\mathrm{p}}}{{\mathrm{h}}_{\mathrm{i}\mathrm{n},\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}-{\mathrm{h}}_{\mathrm{o}\mathrm{u}\mathrm{t},\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}}$$

(6)

From the previous equation, the numerator represents the cooling capacity which represents the difference in enthalpies between the input and output of the refrigerant in the evaporator, considering that there are no charge losses, while the denominator represents the work per unit mass, which is estimated with the difference in enthalpies between the suction and discharge of the compressor. These enthalpy values were estimated based on experimental measurements of pressure and temperature. Thus, Figure 9 shows the comparative COP between the different refrigerants evaluated, assuming that R134a represents 100%, i.e., the figure reflects the percentage variation of the COP of refrigerants R513A, R516A, and R1234ze(E) with respect to R134a as a reference. R513A and R516A showed a reduction of 28% and 25%, respectively, while R1234ze(E) showed an increase in COP of 13% compared to R134a. These variations in COP were mainly due to the reduction in the refrigerator’s cooling capacity. Although R1234ze(E) has a lower cooling capacity, it also has a lower energy consumption, which is reflected in the increase in COP, showing this refrigerant as a viable option from the energy point-of-view, as well as thermal to replace R134a. The same case as that reported in [21] where the refrigerator working with R1234ze€ showed a decrease in the energy consumption with respect to R134a.

An analysis was conducted to assess the uncertainty associated with the estimation of the parameters from the measurements and to evaluate the reliability of the results. The EES software was used to calculate the relative errors in the output parameters, specifically in the COP calculation, using the methodology provided by the same software. The maximum error obtained was ±2.1%.

4.4. TEWI Analysis

A measurement known as the total equivalent warming impact (TEWI) has been put in place to quantify the combined effect of refrigerant leakages into the atmosphere over the course of an equipment life and the emissions produced during the creation of electrical energy for the activation of the refrigeration system. Such measurements may be made as part of the design process for a refrigeration system or, as in this instance, as part of the diagnostic process for switching from R134a to a low-GWP alternative refrigerant. The TEWI can be assessed using Equation (7), in which the first term represents direct emissions and the second represents indirect emissions [29].

$$\mathrm{T}\mathrm{E}\mathrm{W}\mathrm{I}=\left({\mathrm{G}\mathrm{W}\mathrm{P}}_{100}·\mathrm{m}·\mathrm{L}·\mathrm{n}+{\mathrm{G}\mathrm{W}\mathrm{P}}_{100}·\mathrm{m}·\left(1-\mathsf{\alpha }\right)\right)+\left(\mathrm{E}\mathrm{C}·\mathsf{\beta }·\mathrm{n}\right)$$

(7)

From the Equation (7), m is the refrigerant charge, L is the annual refrigerant leakage rate, n is the operating life system, α is the recovery factor, EC is the energy consumption per year, and β is the emission factor, CO₂. In this study, the TEWI analysis was based on a household refrigerator’s 15-year useful lifespan, an average annual refrigerant leakage rate of 2%, and a 70% refrigerant recovery rate α. We suggested comparing the situations of Mexico and Spain to further investigate the TEWI. Then, it would be possible to see the impact of each nation’s emission factor value in relation to its energy production.

An emission factor for the national electricity system in Mexico of 0.423 kg CO₂-eq kWh⁻¹ was considered [30]. For Spain, a value of 0.150 kg CO₂-eq kWh⁻¹ [31] was considered, reflecting a notable difference between the two countries.

Table 7. Total equivalent warming impact for R134a and low-GWP refrigerants.

Parameter	R134a	R513A	R516A	R1234ze(E)
GWP	1300	573	131	1
L [kg per year]	0.00210	0.00176	0.00172	0.00188
n [years]	15	15	15	15
m [kg]	0.105	0.088	0.086	0.094
α [%]	70	70	70	70
EC [kWh per year]	507.35	494.94	505.89	489.46
βMexico [kg CO2-eq kWh−1]	0.423	0.423	0.423	0.423
βSpain [kg CO2-eq kWh−1]	0.150	0.150	0.150	0.150

shows the values used in this study for the estimation and comparison of the TEWI between the refrigerators evaluated.

Figure 10 shows the contribution of the combined impact of the refrigerator, noting that direct emissions are minimal compared to indirect emissions by type and energy consumption. It was also noted that the situation of the refrigerator in Mexico reflects a higher TEWI value than in Spain; this is largely marked by the type of power generation between the two countries. Therefore, the use of these alternative refrigerants in domestic refrigeration would be more advisable in Spain, specifically R1234ze(E) and R516A. Refrigerants R134a and R516A have a practically similar combined impact, and this is because the R516A presented the highest energy consumption among the alternative refrigerants of 505.89 kWh per year, 3% below the consumption of R134a. R1234ze(E) is the refrigerator that presents lower impacts, reflecting a reduction of 6% compared to R134a.

According to the main results of this work, it can be deduced that the direct replacement of R134a by the R513A and R516A mixtures, and pure refrigerant R1234ze(E), is an acceptable proposal in general terms both energetically and operationally, under the direction of maintaining an adequate thermal behavior in the thermal compartments of the refrigerator. Among the refrigerants evaluated, R1234ze(E) specifically seems to be the best choice as a refrigerant in the long term for its low GWP of one and it also yielded the lowest energy consumption which was reflected in the highest % COP and lower TEWI, while R513A continues to show interesting environmental and energy aspects that allow it to be considered as a fluid in the short term [19]. On the other hand, the expectation of the new refrigerant R516A is limited in this study by the high energy consumption recorded, which leads to the use of this refrigerant in domestic refrigeration as more viable through improvements to the design of the refrigerator and not as a direct replacement. Due to the low GWP of R516A, it is necessary to expand the research to improve its performance in domestic refrigeration.

5. Conclusions

This paper presented the experimental results of the thermal and energy performance of a domestic refrigerator working with low-GWP alternative refrigerants as a direct replacement of R134a. The pure refrigerant R1234ze(E) and the azeotropic blends R513A and R516A, the latter as a novelty in the domestic refrigeration industry, were evaluated. The most relevant aspects are listed below.

The optimal charge of each refrigerant was defined by the proposed methodology, which focused on the minimum energy consumption of the refrigerator. The alternative refrigerants had a lower mass compared to R134a (105 g), with R513A (86 g) presenting the smallest reduction.

The refrigerator was evaluated with each optimal charge under the NOM-015-ENER-2018 standard. The average temperatures of the refrigerator compartments showed adequate conditions regarding the position of the thermostat (5/5). Thermally, the alternate refrigerants presented remarkably similar conditions to R134a, which makes them suitable for domestic refrigeration.

R1234ze(E) showed a COP increase of 13% regarding R134a, while the blends presented reductions of 25% for R516A and 28% for R513A.

According to the TEWI analysis and the Mexico case, R1234ze(E) presented the maximum reduction of 5% in the combined impact with respect to R134a, with the R516A refrigerant being more similar to R134a due to its high energy consumption.

Finally, this study raised options on low-GWP refrigerants in domestic refrigeration due to the need to eliminate R134a. Although some countries have defined their roadmap, as in the case of Mexico, it is important to expand the availability of refrigerants in the short and long term according to a low GWP and thermal and energy performance. Hence, this work showed that R1234ze(E) is the most suitable refrigerant for the direct replacement of R134a; however, R513A and R516A mixtures may present better conditions under a redesign of the refrigerator.