Performance of Air-Conditioning System with Different Nanoparticle Composition Ratio of Hybrid Nanolubricant

To reduce fuel consumption, the automotive air-conditioning (AAC) system’s coefficient of performance (COP) needs to be improved. The use of a diverse selection of hybrid nanolubricant composition ratios is expected to improve the properties of single nanolubricants, resulting in improved AAC system performance. The goal of this study was to find the best combination of hybrid nanolubricants for the best performance of the AAC system. Al2O3-SiO2/PAG hybrid nanolubricants at 0.06% volume concentrations with various composition ratios (20:80, 40:60, 50:50, 60:40, and 80:20) were investigated. An initial refrigerant charge of up to 155 g and a compressor speed of up to 2100 rpm were used in the experiment. The cooling capacity, compressor work, and COP of the AAC system were measured to determine its efficiency. The COP enhancement and compressor work reduction were recorded up to 16.31% and 18.65% for the 60:40 composition ratio, respectively. The maximum cooling capacity up to 75.84% was recorded for the 80:20 ratio, followed by 60:40. The maximum COP value of 8.81 for 155 g of hybrid nanolubricants was obtained at 900 rpm with a 60:40 composition ratio. Therefore, for optimal performance in the AAC system, a 60:40 composition ratio of the Al2O3-SiO2/PAG nanolubricant combination is strongly recommended.


Introduction
In today's society, an air-conditioning system in a vehicle is a necessary and indispensable piece of technology [1] as it is used to provide vehicular thermal comfort for the driver and passenger. The effect of cooling the passenger compartment on the vehicle energy consumption and emissions is highly dependent on the thermal comfort [2] in real conditions. There are a few issues with the current automotive air-conditioning (AAC) system including the fact that it increases fuel consumption while cooling the compartments to the desired level of thermal comfort [3,4]. First, vehicles are in a low-efficiency region when operating with the air-conditioning system during idle conditions for an extended period of time [5]. Consequently, increments in the workload, power, and fuel consumption of the vehicles are required to maintain the operation of the AAC system. The next issue is the increase in harmful gasses emitted by vehicles such as carbon dioxide (CO x ), hydrocarbons (HC), nitrogen oxides (NO x ), and carbon monoxide (CO) [6,7] during idling and slow moving traffic. The gases released into the atmosphere will cause air pollution [8] and increase the average temperature on Earth. Therefore, these issues are being addressed in viscosity class, is approved for the R134a refrigerant [38]. Equation (1) is used to calculate the concentration of nanolubricants in each volume. φ = m p /ρ p m p /ρ p + m L /ρ L × 100 (1) where φ is the volume concentration of nanolubricants (%); m p is the nanoparticle mass (g); ρ p is the nanoparticle density (kg/m 3 ); m L is the lubricant mass (g); and ρ L is the lubricant density (kg/m 3 ). Nanoparticles added at volume concentrations of more than 0.06% greatly enhanced the coefficient of friction and wear rate, even to values higher than pure PAG [34]. Thus, a maximum volume concentration of 0.06% for various composition ratios was studied. Table 1. Properties of the nanoparticles [37].

Properties PAG 46
Density, g/cm 3  For a volume concentration of 0.06%, the Al 2 O 3 -SiO 2 /PAG was prepared using a two-step preparation method. The composition ratios of the hybrid nanolubricants investigated in the current study were 20:80, 40:60, 50:50, 60:40, and 80:20, which were determined using the volume percentages of the mono nanolubricants. The Al 2 O 3 -SiO 2 hybrid nanolubricant composition ratios were prepared in volume percentages in accordance with Table 3, which outlines the composition ratios used in the current study. In preparing Al 2 O 3 -SiO 2 /PAG with a 50:50 ratio, 52.5 mL of Al 2 O 3 /PAG nanolubricant was mixed into 52.5 mL of the SiO 2 /PAG nanolubricant. The nanolubricants were then combined for up to 30 min using a magnetic stirrer before being sonicated for up to two hours. The Al 2 O 3 -SiO 2 /PAG had an optimum stability at a sonication time of 2 h [39,40]. The dispersion of the hybrid nanolubricants was observed using TEM, as shown in Figure 1. After preparation, the dispersion stability of the hybrid nanolubricants was visually examined as well as thirty days later. As seen in Figure 2, no sedimentation appeared in the samples. The average absolute zeta potential reading for a 0.1% volume concentration of Al 2 O 3 -SiO 2 /PAG was up to 61.1 mV. The zeta potential for Al 2 O 3 -SiO 2 /PAG was found to be above the stable limit, suggesting an outstanding stability condition. The sonication duration, stability, and thermal physical features of Al 2 O 3 -SiO 2 /PAG have all been confirmed and verified in previous studies [39].  1  20  80  2  40  60  3  50  50  4  60  40  5  80  20   Micromachines 2022, 13, 1871   3  50  50  4  60  40  5 80 20

Automotive Air-Conditioning System Testing Equipment
The experimental configuration consisted of the original components of the car's AAC system. The driver, with its control system, the water bath system for t orator, the piping system, and other measurement equipment, comprised the be figuration of the refrigeration system. The compressor, condenser, evaporator, ex valve, and receiver drier are the essential components of the AAC system. The fi placement compressor is commonly used in a compact car type AAC system. T pressor has maximum refrigeration capacities of up to 1.2 kW. The water bath sys

Automotive Air-Conditioning System Testing Equipment
The experimental configuration consisted of the original components of the compact car's AAC system. The driver, with its control system, the water bath system for the evaporator, the piping system, and other measurement equipment, comprised the bench configuration of the refrigeration system. The compressor, condenser, evaporator, expansion valve, and receiver drier are the essential components of the AAC system. The fixed displacement compressor is commonly used in a compact car type AAC system. The compressor has maximum refrigeration capacities of up to 1.2 kW. The water bath system was a properly insulated 60 liter water tank with a water inlet and outlet for the purpose of the evaporator calorimetric system. The insulated expansion valve and the evaporator were placed inside the tank. The cooling capacity and the mass flow rate of the system were determined using the evaporator calorimetric method in accordance with ASHRAE Standard 41.9 (2000). The inverter frequency controller and the electrical motor were used

Automotive Air-Conditioning System Testing Equipment
The experimental configuration consisted of the original components of the compact car's AAC system. The driver, with its control system, the water bath system for the evaporator, the piping system, and other measurement equipment, comprised the bench configuration of the refrigeration system. The compressor, condenser, evaporator, expansion valve, and receiver drier are the essential components of the AAC system. The fixed displacement compressor is commonly used in a compact car type AAC system. The compressor has maximum refrigeration capacities of up to 1.2 kW. The water bath system was a properly insulated 60 liter water tank with a water inlet and outlet for the purpose of the evaporator calorimetric system. The insulated expansion valve and the evaporator were placed inside the tank. The cooling capacity and the mass flow rate of the system were determined using the evaporator calorimetric method in accordance with ASHRAE Standard 41.9 (2000). The inverter frequency controller and the electrical motor were used by the driver with the control system. A 4 kW capacity frequency inverter was employed to run the electrical motor and control the compressor revolution speed. A 2.2 kW 3-phase induction motor drives the compressor via a belting system. The maximum speed of the motor was 2850 rpm. A 3-phase power analyzer from Lutron (model DW-6092) was used to measure the electrical power characteristics and power consumption of the induction motor to drive the compressor. The instrumentation was controlled by a datalogger, which included flow rate sensors, thermocouples, and pressure sensors. The datalogger consisted of a datalogger integrated with 16 channels of the temperature measurements. The 16 channels datalogger represent 16 thermocouple points of measurement as shown in Figure 3. All of the readings were stored automatically in the datalogger. Figure 3 depicts the schematic design of the AAC system bench and Figure 4 shows the AAC system configurations. The Sanden SWJ-7B08 compressor runs on the R134a refrigerant and is lubricated using PAG lubricants. The nanolubricants lubricate the compressor bearings and other moving parts and are carried through the compressor by the refrigerant, where they circulate with the refrigerant throughout the entire system. motor to drive the compressor. The instrumentation was controlled by a datalogger, which included flow rate sensors, thermocouples, and pressure sensors. The datalogger consisted of a datalogger integrated with 16 channels of the temperature measurements. The 16 channels datalogger represent 16 thermocouple points of measurement as shown in Figure 3. All of the readings were stored automatically in the datalogger. Figure 3 depicts the schematic design of the AAC system bench and Figure 4 shows the AAC system configurations. The Sanden SWJ-7B08 compressor runs on the R134a refrigerant and is lubricated using PAG lubricants. The nanolubricants lubricate the compressor bearings and other moving parts and are carried through the compressor by the refrigerant, where they circulate with the refrigerant throughout the entire system.   Prior to the AAC system trials, the thermocouple, pressure gauge, and flow were calibrated in the current investigation. Calibration of the thermocouple, p gauge, and flow meter were performed in the present study prior to the AAC sys periments. In this experiment, a K-type thermocouple with a diameter of 0.3 mm temperature range of −40 to 375 °C with a tolerance of 1.5 °C was used to mon temperature. This was calibrated using the Standard Platinum Resistance Therm (SPRT). A K-type thermocouple was chosen because it can tolerate a large temp range (0 and 1260 °C ) and has great oxidation and acidity tolerance due to its type (chromel and alumel). Similar sensor calibrations were also performed by p researchers [29,41,42]. The thermocouples (T1-T8) were used to measure the enter exiting temperature of the refrigerants for each of the four main components of t system. In addition, the thermocouple monitored the temperature of the inlet ( outlet (Tout) of the water bath and the ambient temperature (Tamb). Table 4 summar list of sensors and equipment with their uncertainty values. The overall flow cha AAC system test run using hybrid nanolubricants with different composition shown in Figure 5.  Prior to the AAC system trials, the thermocouple, pressure gauge, and flow meter were calibrated in the current investigation. Calibration of the thermocouple, pressure gauge, and flow meter were performed in the present study prior to the AAC system experiments. In this experiment, a K-type thermocouple with a diameter of 0.3 mm and a temperature range of −40 to 375 • C with a tolerance of 1.5 • C was used to monitor the temperature. This was calibrated using the Standard Platinum Resistance Thermometer (SPRT). A K-type thermocouple was chosen because it can tolerate a large temperature range (0 and 1260 • C) and has great oxidation and acidity tolerance due to its material type (chromel and alumel). Similar sensor calibrations were also performed by previous researchers [29,41,42]. The thermocouples (T 1 -T 8 ) were used to measure the entering and exiting temperature of the refrigerants for each of the four main components of the AAC system. In addition, the thermocouple monitored the temperature of the inlet (T in ) and outlet (T out ) of the water bath and the ambient temperature (T amb ). Table 4 summarizes the list of sensors and equipment with their uncertainty values. The overall flow chart of the AAC system test run using hybrid nanolubricants with different composition ratios is shown in Figure 5. Table 4. Uncertainties of the AAC system sensors and measurement devices.

Experimental Procedure
The AAC system was flushed with nitrogen (N2) gas to remove pollutants an ardous compounds while also preventing component degradation. The system wa subjected to a leakage test by attaching a vacuum pump to it. Both the flushing and age test were performed each time prior to the AAC system experiments. Througho experiment, the atmospheric temperature and humidity were measured and reg between 45.5 and 55.5 °C and 50 and 60%, respectively. It should be noted that the t was designed in compliance with the SAEJ2765 (2008) specification and was develo a control room. The PAG lubricants in a volume of 105 mL were introduced into the compressor for the benchmark study of the AAC system running with PAG lubr The system was then put through a leakage test to ensure that no leaks occurred. F ing that, 95 g of the R134a refrigerant was added to the system, which charged the s Water was then pumped into the bath tank until the evaporator was complete mersed. The water heater regulated the water bath's intake (Tin) and outlet (Tout) te atures until the water flow temperature was uniform. At all times, the inflow flo was fixed at 8 liters per minute (lpm). The temperature of the inflow as well as o water baths was generally maintained within an allowed range of 30 to 31 °C.
Once all the previous steps had been accomplished, the motor was started, a compressor's magnetic clutch was engaged. The frequency meter was adjusted to

Experimental Procedure
The AAC system was flushed with nitrogen (N 2 ) gas to remove pollutants and hazardous compounds while also preventing component degradation. The system was then subjected to a leakage test by attaching a vacuum pump to it. Both the flushing and leakage test were performed each time prior to the AAC system experiments. Throughout the experiment, the atmospheric temperature and humidity were measured and regulated between 45.5 and 55.5 • C and 50 and 60%, respectively. It should be noted that the test rig was designed in compliance with the SAEJ2765 (2008) specification and was developed in a control room. The PAG lubricants in a volume of 105 mL were introduced into the AAC compressor for the benchmark study of the AAC system running with PAG lubricants. The system was then put through a leakage test to ensure that no leaks occurred. Following that, 95 g of the R134a refrigerant was added to the system, which charged the system. Water was then pumped into the bath tank until the evaporator was completely immersed. The water heater regulated the water bath's intake (T in ) and outlet (T out ) temperatures until the water flow temperature was uniform. At all times, the inflow flow rate was fixed at 8 liters per minute (lpm). The temperature of the inflow as well as outflow water baths was generally maintained within an allowed range of 30 to 31 • C.
Once all the previous steps had been accomplished, the motor was started, and the compressor's magnetic clutch was engaged. The frequency meter was adjusted to obtain the required compressor speed of 900 rpm. The datalogger and power analyzer were both turned on at the same time in order to measure the temperature, pressure, power usage, and water mass flow rate. The AAC device was left running for 20 min (to achieve a steady-state condition), while the experimental data were recorded at the same time. The datalogger and power analyzer were turned off at the end of the 20-min experiment period to stop the data collection. The heater, motor, and compressor were all shut off afterward. The system was then run for a second cycle, with the compressor speeds gradually increased up to 2100 rpm and variable initial refrigerant charges of up to 155 g were added. The AAC system was then flushed out before additional experimental testing could begin. Afterward, all of the procedures were duplicated using Al 2 O 3 -SiO 2 /PAG. The experiment was carried out at a constant compressor speed to study the efficiency of the AAC system with various Al 2 O 3 -SiO 2 /PAG composition ratios. The AAC system was then tested with different compressor rates of speed and initial refrigerant loads to find the best composition ratio of Al 2 O 3 -SiO 2 /PAG.

Experimental Analysis
The temperatures at each position in Figure 3 as well as the minimum and maximum pressures (P low and P high ), which were all recorded in the datalogger, were all obtained to assess the performance of the AAC system. Using information gathered from the datalogger, the performance of the AAC system's parameters including the cooling capacity, compressor work, COP, and AAC power consumption, was analyzed. Equations (2)-(5) were used to evaluate key parameters such as cooling capacity, compressor work, COP, and AAC power consumption. The enthalpy difference between the water inlet, T in , and outlet, T out , was used to calculate the cooling capacity ( . Q L ) of the AAC system. Equation (2) yields the cooling capacity, while Equation (3) calculates the heat absorbed (Q L ). The heat capacity of water, C p,w , was held constant at 4.2 J while its mass flow rate ( . m w ) was kept at 8 lpm. Equation (4) can be used to obtain the compressor work (W in ) of the system. Equation (5) can be used to determine the COP of the AAC system. The ratio of heat absorbed to the work undertaken by the compressor to remove the heat is known as the COP. By determining the value of pressure and temperature from the experimental data, the enthalpy of the AAC system is computed. The obtained results were used in conjunction with the properties tables [43] and the psychrometric chart in order to formulate an estimate for the enthalpy.

AAC Performance Using PAG Lubricants
The refrigerant charge is a crucial aspect in the evaluation of the refrigeration system performance. This is a result of insufficient refrigerant charge reducing the COP and increasing the energy consumption [44]. Thus, the refrigerant charge will affect the performance of the system. This is due to the increase in compressor work along with the increase in the refrigerant charge amount. It is important to study the optimal charge of the refrigeration system to optimize the performance and efficiency of the refrigeration system. Consequently, the refrigerant charge will influence the performance of the system. This is due to the increase in compressor work and the rise in the amount of refrigerant charge. It is essential to examine the appropriate charge of the refrigeration system in order to enhance its performance and efficiency. Following a setup evaluation of the AAC system, the AAC system studies were commenced. The AAC system performance, specifically cooling capacity, coefficient of performance (COP), and compressor work were studied using a range of R134a refrigerant charges from 95 to 155 g. The performance of pure PAG lubricants employed in the AAC system acted as the reference data for the relative performance validation of hybrid nanolubricants. The cooling capability of pure PAG lubricants at various refrigerant loads is shown in Figure 6. The cooling capacity rose as the starting refrigerant charge and speed increased, as seen in the graph. The conservation of mass equation was demonstrated by the cooling capacity increasing as the mass flow rate increased. The mass flow rate tended to rise with an increasing compressor speed, thus increasing the cooling capacity [29]. The compressor speed increment will increase the system's high side pressure, thus increasing the enthalpy of the saturated refrigerant. Additionally, when the system's mass flow rate increases, the evaporator's temperature rises [45]. This is due to the fact that when the system's mass flow rate increases, the evaporator's temperature rises [45]. Due to the two-phase refrigerant flow from the condenser to the expansion valve, a reduced cooling capacity was observed at a lower refrigerant charge [29].
Micromachines 2022, 13, 1871 9 of 19 studied using a range of R134a refrigerant charges from 95 to 155 g. The performance of pure PAG lubricants employed in the AAC system acted as the reference data for the relative performance validation of hybrid nanolubricants. The cooling capability of pure PAG lubricants at various refrigerant loads is shown in Figure 6. The cooling capacity rose as the starting refrigerant charge and speed increased, as seen in the graph. The conservation of mass equation was demonstrated by the cooling capacity increasing as the mass flow rate increased. The mass flow rate tended to rise with an increasing compressor speed, thus increasing the cooling capacity [29]. The compressor speed increment will increase the system's high side pressure, thus increasing the enthalpy of the saturated refrigerant. Additionally, when the system's mass flow rate increases, the evaporator's temperature rises [45]. This is due to the fact that when the system's mass flow rate increases, the evaporator's temperature rises [45]. Due to the two-phase refrigerant flow from the condenser to the expansion valve, a reduced cooling capacity was observed at a lower refrigerant charge [29].  Figure 7 shows the graph of the compressor work of PAG against refrigerant charges with a variation in the compressor speeds. From the graph, the compressor work of pure PAG lubricants showed a small decrease in the compressor work with an increase in the initial refrigerant mass. However, the compressor work increased with the increase in speed. This was due to increments in the discharge temperature and pressure at the compressor along with the increasing compressor speed. Therefore, increasing the enthalpy difference will result in an increment in the compressor work. When operating at higher speeds, compressors typically exert a much greater amount of effort than when operating at lower speeds. Thus, this will also lead to a decrease in the efficiency of the system. A similar trend was observed for the compressor work in the literature [46,47]. Figure 8 shows the graph of the COP for the PAG based lubricants against the initial refrigerant charge. The most commonly analyzed main parameter for measuring the air-conditioning system performance is the COP [48]. With an increasing refrigerant mass and compressor speed, the system COP increases considerably. This is because of the increase in the cool-  Figure 7 shows the graph of the compressor work of PAG against refrigerant charges with a variation in the compressor speeds. From the graph, the compressor work of pure PAG lubricants showed a small decrease in the compressor work with an increase in the initial refrigerant mass. However, the compressor work increased with the increase in speed. This was due to increments in the discharge temperature and pressure at the compressor along with the increasing compressor speed. Therefore, increasing the enthalpy difference will result in an increment in the compressor work. When operating at higher speeds, compressors typically exert a much greater amount of effort than when operating at lower speeds. Thus, this will also lead to a decrease in the efficiency of the system. A similar trend was observed for the compressor work in the literature [46,47]. Figure 8 shows the graph of the COP for the PAG based lubricants against the initial refrigerant charge. The most commonly analyzed main parameter for measuring the air-conditioning system performance is the COP [48]. With an increasing refrigerant mass and compressor speed, the system COP increases considerably. This is because of the increase in the cooling capacity, which overweighs the increment of the compressor power at the higher compressor speed and refrigerant charge. At a compressor speed of 900 rpm and a refrigerant load of 155 g, the COP reached its maximum of 7.09. However, due to limitations in the AAC setup and to avoid compressor operation overload, the AAC system was charged up to 155 g only. Overload compressor operation can lead to performance degradation within the compressor [49]. Hence, Al 2 O 3 -SiO 2 /PAG hybrid nanolubricants in this study were evaluated up to 155 g of refrigerant charge of the AAC system performance for further investigations. ing capacity, which overweighs the increment of the compressor power at the higher compressor speed and refrigerant charge. At a compressor speed of 900 rpm and a refrigerant load of 155 g, the COP reached its maximum of 7.09. However, due to limitations in the AAC setup and to avoid compressor operation overload, the AAC system was charged up to 155 g only. Overload compressor operation can lead to performance degradation within the compressor [49]. Hence, Al2O3-SiO2/PAG hybrid nanolubricants in this study were evaluated up to 155 g of refrigerant charge of the AAC system performance for further investigations.   Figure 9 depicts the cooling capacity of various composition ratios of Al 2 O 3 -SiO 2 /PAG against the initial refrigerant charge at 900 rpm. The cooling capacity of hybrid nanolubricants increased along with the increase in the initial refrigerant charge. It was discovered that Al 2 O 3 -SiO 2 /PAG lubricants had a better cooling capability than the pure PAG lubricants. This is because the nanoparticles in hybrid nanolubricants help to improve heat transmission, causing the heat transfer in the evaporator to increase. Moreover, with the addition of nanoparticles in the lubricants, the decrement in the heat transfer coefficient can be avoided in the evaporator [50]. The Al 2 O 3 -SiO 2 /PAG hybrid nanolubricants have better thermal physical properties than the PAG lubricant [51]. The nanoparticles in hybrid nanolubricants improve heat transmission, resulting in an increase in heat transfer in the evaporator. Consequently, there is a correlation between the thermal conductivity and the composition ratios of the hybrid nanolubricants, and cooling capacity enhancements can be attributed to this correlation. The highest cooling capacity enhancement was achieved for the 80:20 ratio with 75.84% improvement for the hybrid nanolubricants compared to the pure PAG lubricants. The cooling capacity at the 60:40 ratio was enhanced up to 70.94% and higher than the 20:80 ratio. However, the 50:50 ratio showed an improvement of 45.40% and found the lowest cooling capacity enhancement. The composition ratio with more than 50% Al 2 O 3 nanoparticles contributed to high cooling capacity enhancements up to 70%. Initial Refrigerant Charge (g) Figure 9. Cooling capacity of hybrid nanolubricants at various composition ratios.

Nanolubricants with Composition Ratios
The compressor work of different composition ratios was plotted against the initia refrigerant mass, as shown in Figure 10. The highest compressor work was noticed for th PAG lubricants. This indicates that the compressor using pure PAG lubricants require more work compared to the nanolubricants under similar working conditions. The com pressor work using hybrid nanolubricants at the 60:40 ratio exhibited the highest reduc tion of 17.95% and the lowest reduction of 7.59% at a ratio of 50:50. The composition ratio of hybrid nanolubricants can greatly influence the compressor work of the AAC system Due to the optimal composition ratios of hybrid nanolubricants, the compressor workload is minimized as the friction coefficient decreases. Therefore, reducing the friction coeffi The compressor work of different composition ratios was plotted against the initial refrigerant mass, as shown in Figure 10. The highest compressor work was noticed for the PAG lubricants. This indicates that the compressor using pure PAG lubricants requires more work compared to the nanolubricants under similar working conditions. The compressor work using hybrid nanolubricants at the 60:40 ratio exhibited the highest reduction of 17.95% and the lowest reduction of 7.59% at a ratio of 50:50. The composition ratios of hybrid nanolubricants can greatly influence the compressor work of the AAC system. Due to the optimal composition ratios of hybrid nanolubricants, the compressor workload is minimized as the friction coefficient decreases. Therefore, reducing the friction coefficient is crucial to cutting the energy consumption of the compressor. However, the 50:50 ratio can only slightly reduce the compressor work when compared with other ratios due to high viscosity, especially for high refrigerant charges. Less work due to lower viscosity gives an advantage to the compressor operation. Viscosity of the lubricants is one of the major factors for compressor performance in the refrigeration system [37]. From Figure 11, the COP of Al 2 O 3 -SiO 2 /PAG with different composition ratios increased with a refrigerant charge. All composition ratios showed enhancements in COP performance compared to the pure PAG. The COP performance for the 60:40 ratio exhibited the highest enhancement of up to 16.31%. Meanwhile, the lowest COP enhancement of 5.7% was attained for the 50:50 ratio. A high percentage of SiO 2 nanoparticles in the hybrid nanolubricants greatly contributed to the COP enhancement. Meanwhile, a high percentage of Al 2 O 3 nanoparticles can lower COP enhancement. The 60:40 ratio exhibited better overall performance in terms of cooling capacity, compressor work, and COP. The average enhancement and reduction in the AAC system for all composition ratios are summarized in Table 5. In this study, the experimental setup was used to identify the optimal hybrid nanolubricant design for the AAC system performance. The optimization analysis process can then be further validated using any available techniques such as the Taguchi method, response surface technology methods (RSM) [40], machine learning (ML) based intelligent techniques, metaheuristic algorithms [52], and other optimization techniques using the experimental data that were obtained.    Figure 11. Coefficient of performance of hybrid nanolubricants at various composition ratios. Figure 11. Coefficient of performance of hybrid nanolubricants at various composition ratios.

Nanolubricants with Compressor Speeds
The AAC system performance, namely cooling capacity, compressor work, and COP were investigated for various compressor speeds at a 0.06% volume concentration and 60:40 composition ratio of Al 2 O 3 -SiO 2 /PAG hybrid nanolubricants. The performance of the AAC system was investigated at different compressor speeds varying from 900 to 2100 rpm. Figures 12-14 indicate the performance parameters of the Al 2 O 3 -SiO 2 /PAG composite nanolubricants at different compressor speeds as a function of the initial refrigerant charge. Figure 12 shows the cooling capacity for hybrid nanolubricants with 60:40 at different compressor speeds. As the compressor speed rises, the system's cooling capacity also rises. The Al 2 O 3 -SiO 2 /PAG lubricant had a significantly larger cooling capacity than the pure PAG lubricants at all compressor speeds. The incorporation of nanoparticles to PAG lubricants increased the system's cooling capability. A system with a large cooling capacity has superior heat transfer performance and can cool down the system more effectively. Enhancement in the thermal conductivity of nanolubricants improves the heat absorbance and utilizes the system to operate at the utmost efficiency. The improvement in the cooling capacity is in line with the thermal conductivity evaluation previously reported [39,53]. Figure 13 represents the effects of compressor work at different compressor speeds for Al 2 O 3 -SiO 2 /PAG at the 60:40 ratio. When compared to the pure PAG lubricants, the compressor work for hybrid nanolubricants was significantly less. In addition, when the refrigerant charge rises, the compressor's workload diminishes. Low compressor speed is thought to result in less load on the compressor, resulting in less compressor effort compared to greater compressor speed. When compared to the pure PAG, Al 2 O 3 -SiO 2 /PAG contributes more to lowering the compressor effort. This is because of a decrease in the tribological behavior of hybrid nanolubricants, specifically the friction coefficient and the percentage of wear. Thus, in order to reduce the compressor friction losses, which lower the pressure drop and compressor effort, hybrid nanolubricants are required in the refrigeration systems. The hybrid nanolubricants are being considered since a higher coefficient friction results in greater friction losses. Figure 14 shows a graph of COP against the initial refrigerant mass of the system at different compressor speeds for the 60:40 ratio. The graph indicates that COP improves as the refrigerant charge increases, but declines as the compressor speed increases. The graph depicts a low COP for the pure PAG lubricants. This is because the pure PAG lubricants have low cooling capacities but high compressor work. The maximum COP achievable was 8.81 for Al 2 O 3 -SiO 2 /PAG of the 60:40 ratios at the 155 g refrigerant charge and 900 rpm. Sabareesh et al. [54] and Kumar and Elansezhian [32] both found that this pattern persisted. Al 2 O 3 -SiO 2 /PAG was observed to have a similar pattern in COP increments with the pure PAG lubricants. High compressor speeds result in more work being done by the compressor, resulting in a decrease in COP. In order to overcome this, the compressor's workload must be reduced. Consequently, the efficiency of the compressor is improved. position ratios. The AAC system performance, namely cooling capacity, compressor work, and COP were investigated for various compressor speeds at a 0.06% volume concentration and 60:40 composition ratio of Al2O3-SiO2/PAG hybrid nanolubricants. The performance of the AAC system was investigated at different compressor speeds varying from 900 to 2100 rpm. Figures 12-14 indicate the performance parameters of the Al2O3-SiO2/PAG composite nanolubricants at different compressor speeds as a function of the initial refrigerant charge. Figure 12 shows the cooling capacity for hybrid nanolubricants with 60:40 at different compressor speeds. As the compressor speed rises, the system's cooling capacity also rises. The Al2O3-SiO2/PAG lubricant had a significantly larger cooling capacity than the pure PAG lubricants at all compressor speeds. The incorporation of nanoparticles to PAG lubricants increased the system's cooling capability. A system with a large cooling capacity has superior heat transfer performance and can cool down the system more effectively. Enhancement in the thermal conductivity of nanolubricants improves the heat absorbance and utilizes the system to operate at the utmost efficiency. The improvement in the cooling capacity is in line with the thermal conductivity evaluation previously reported [39,53]. both found that this pattern persisted. Al2O3-SiO2/PAG was observed to have a similar pattern in COP increments with the pure PAG lubricants. High compressor speeds result in more work being done by the compressor, resulting in a decrease in COP. In order to overcome this, the compressor's workload must be reduced. Consequently, the efficiency of the compressor is improved.

Conclusions
The performance of the AAC system was tested at various composition ratios, compressor speeds, and initial refrigerant charges. The cooling capacity, compressor work, and COP were plotted and discussed in relation to the initial refrigerant charges. The op-

Conclusions
The performance of the AAC system was tested at various composition ratios, compressor speeds, and initial refrigerant charges. The cooling capacity, compressor work, and COP were plotted and discussed in relation to the initial refrigerant charges. The optimum AAC system performance using Al 2 O 3 -SiO 2 /PAG was investigated using hybrid nanolubricants with various composition ratios to achieve the optimum AAC system performance efficiency. The COP is highlighted as the main priority to determine the optimum composition ratio of nanolubricants for the performance of the AAC system. The highest average enhancement of 16.31% for COP was recorded with the 60:40 ratio at constant volume concentration. The lowest average enhancement of 5.70% was at the 50:50 ratio. Al 2 O 3 -SiO 2 /PAG exhibited the highest COP value compared to other composition ratios. The effectiveness of the system can be improved by using nanolubricants with the appropriate composition ratio, which has a greater potential for enhancing the COP performance, automotive cooling rates, and a reduction in the compressor work. Due to their remarkable improvement in thermal, physical, and heat transfer capabilities to increase the efficiency and dependability of the refrigeration and air conditioning systems, hybrid nanolubricants have additional advantages over conventional fluids and singlecomponent nanolubricants. Therefore, for the best performance in refrigeration systems, Al 2 O 3 -SiO 2 /PAG hybrid nanolubricants with a 60:40 ratio are advised. Performance optimization of Al 2 O 3 -SiO 2 /PAG refrigeration systems is required to validate the existing findings. The current study was a short-term laboratory experiment that ruled out the AAC system durability using hybrid nanolubricants. To extend the present work, additional research is needed to optimize the performance parameters and durability effects of Al 2 O 3 -SiO 2 /PAG composite nanolubricants on the performance and application of an AAC system.