Ionic Liquids Facilitate the Development of Absorption Refrigeration

The rapid progress in the development of ionic liquids has generated enthusiasm for their application in many traditional fields and renewed interest in absorption refrigeration. New absorption refrigeration working pairs containing ionic liquids have gained widespread attention in the past decade. In a chapter entitled “The Latent Application of Ionic Liquids in Absorption Refrigeration” [1] that we have published 3 years ago with InTech in the book entitled “Applications of Ionic Liquids in Science and Technology” achieved impressive readership results and has so far been accessed more than 4000 times. Over the past 3 years, progress in this field has been outstanding, and a few commercially competitive new working pairs were discovered. In this chapter, we describe the latest progress in the development of a few mentionable new working pairs containing ionic liquids for absorption refrigeration and a type of completely new conceptual absorption refrigeration working pair that was proposed by us and is expected to lead to a major breakthrough in the development of absorption refrigeration.


Introduction
The rapid progress in the development of ionic liquids has generated enthusiasm for their application in many traditional fields and renewed interest in absorption refrigeration. New absorption refrigeration working pairs containing ionic liquids have gained widespread attention in the past decade. In a chapter entitled "The Latent Application of Ionic Liquids in Absorption Refrigeration" [1] that we have published 3 years ago with InTech in the book entitled "Applications of Ionic Liquids in Science and Technology" achieved impressive readership results and has so far been accessed more than 4000 times. Over the past 3 years, progress in this field has been outstanding, and a few commercially competitive new working pairs were discovered. In this chapter, we describe the latest progress in the development of a few mentionable new working pairs containing ionic liquids for absorption refrigeration and a type of completely new conceptual absorption refrigeration working pair that was proposed by us and is expected to lead to a major breakthrough in the development of absorption refrigeration.

Recent progress in absorption refrigeration working pairs containing ionic liquids
In the past 3 years, enthusiasm for studies on absorption refrigeration working pairs containing ionic liquids seems to have waned. The once preferred ionic liquid working pairs, such as Freon-IL, CO 2 -IL, and NH 3 -IL, do not receive attention from researchers any longer. However, some impressive progress is still being made for working pairs composed of a refrigerant and [RR'Im]DMP (1-R,3-R'-imidazolium dimethylphosphate).

[mmIm]DMP-CH 3 OH
Zhao Jie et al. [2] measured the saturation vapor pressure of [mmIm]DMP-CH 3 OH at T = 303.15-363.15 K and over a low methanol mole fraction range for x including 0.529, 0.558, 0.582 and 0.605. Zhao Jin et al. [3] measured the saturation vapor pressure of [mmIm]DMP-CH 3 OH at T = 280-370 K and over the high methanol mole fraction range for x including 0.8222, 0.9123, 0.9418 and 0.9652. These experimental results were confirmed by Chen et al. [4] using the UNIFAC model and the Wilson model to predict the vapor pressure and the excess enthalpy, respectively. Figs. 1 and 2 show the predicted vapor pressure and excess enthalpy, respectively, at T = 280-380 K and x = 0-1.   The thermodynamic performances of single effect [mmIm]DMP-CH 3 OH absorption refrigeration have been simulated and analyzed. Fig. 3 shows the effects of the operating temperatures [condensing temperature (T C ), evaporating temperature (T E ), generating temperature (T G ), and absorption temperature (T A )] on the circulation ratio, f, of the single system. From Fig. 3, the f of [mmIm]DMP-CH 3 OH is higher than that of LiBr/H 2 O but still acceptable for operation, and the coefficient of performance (COP) of the [mmIm]DMP-CH 3 OH absorption refrigeration will remain good, if the heat transfer areas of the regenerator are designed appropriately. Fig. 4 shows the effects of the operating temperatures on the COP of the single system. From Fig. 4, the COP of the [mmIm]DMP-CH 3 OH absorption refrigeration is lower than that of LiBr/H 2 O absorption refrigeration under the same temperature conditions, but higher than that of The thermodynamic performances of single effect [mmIm]DMP-CH 3 OH absorption refrigeration have been simulated and analyzed. Fig. 3 shows the effects of the operating temperatures [condensing temperature (T C ), evaporating temperature (T E ), generating temperature (T G ), and absorption temperature (T A )] on the circulation ratio, f, of the single system. From Fig. 3, the f of [mmIm]DMP-CH 3 OH is higher than that of LiBr/H 2 O but still acceptable for operation, and the coefficient of performance (COP) of the [mmIm]DMP-CH 3 OH absorption refrigeration will remain good, if the heat transfer areas of the regenerator are designed appropriately. Fig. 4 shows the effects of the operating temperatures on the COP of the single system. From Fig. 4, the COP of the [mmIm]DMP-CH 3 OH absorption refrigeration is lower than that of LiBr/ H 2 O absorption refrigeration under the same temperature conditions, but higher than that of H 2 O/NH 3 absorption refrigeration under most temperature conditions. When the heat source temperature is greater than 400 K, [mmIm]DMP-CH 3 OH absorption is still possible with a high COP close to that of LiBr/H 2 O absorption refrigeration. In general, [mmim]DMP/methanol has excellent potential for application as the working pair in absorption refrigeration.

[dmIm]DMP-H 2 O
Dong et al. [5] studied the thermophysical properties of the [dmIm]DMP-H 2 O system. The vapor pressures of the [dmIm]DMP-H 2 O system at mass fractions of ionic liquids, ω, in the range of 0.10 to 0.90 were measured and correlated using a non-random two-liquid (NRTL) model. The experimental data and the model predictions are presented in Fig. 5.

[emIm]DMP-H 2 O
Ren et al. [6] measured the vapor pressure of the [emIm]DMP-H 2 O binary system at different IL mole fractions, x, ranging from 0.1 to 0.5, and the experimental data were fitted using the NRTL model (Fig. 8).

Summary
All three of the working pairs described above possess good theoretical cycle characteristics that are better than those of H 2 O-NH 3 , but still slightly lower than those of LiBr-H 2 O. Due to the advantages of the negligible vapor pressure of the absorbent, no corrosion, and no crystallization, these three working pairs can be applied in a wider range of operating conditions than H 2 O-NH 3 or LiBr-H 2 O. Therefore, it is expected that these three working pairs have enormous potential in industrial applications and strong possibilities for commercial development.

The proposal
Adsorption refrigeration is a type of environmentally friendly refrigeration that has been studied for many years. The most commonly used working pairs in adsorption refrigeration systems are ammonia-activated carbon, methanol-activated carbon, water-zeolite, ammoniacalcium chloride, and methanol-calcium chloride. The first three are physical adsorption working pairs, and the last two are chemical adsorption working pairs. The following review begins with the NH 3 -CaCl 2 system.
Calcium chloride reacts with ammonia to form coordination compounds: where ∆H 1~∆ H 3 are the enthalpies of the reaction, and T e1~Te3 are the equilibrium temperatures. Benefiting from the reaction, the most impressive advantage of the NH 3 -CaCl 2 system lies in its higher adsorption capacity compared to the others, while the main disadvantages are the low performances of heat and mass transfer and the phenomena of swelling and agglomeration in the process of adsorption [8]. Much effort has been spent attempting to overcome these defects. For example, Kai Wang et al. [9] proposed a new type of compound adsorbent composed of CaCl 2 and an expanded graphite adsorbent, which could mitigate the deterioration of the adsorption capacity that occurs in the long-term adsorption/desorption process. Using the compound adsorbent, Liwei Wang et al. [10] designed a multi-effect heat pipe-type adsorption refrigeration system, and a COP for their system of 0.39 was reported at a low t E of -20 °C. Obviously, these improvements have little effect, and many other similar efforts [11] proved futile. The essence of all these failures can be attributed to the fact that the adsorbent is a solid. Except for CaCl 2 , typical metal chlorides used as an ammonia adsorbent include SrCl 2 , LiCl 2 , and ZnCl 2 , among others. If only the solid metal chlorides could be dissolved in ionic liquids, there would be no problem that could not be solved in the absorption or adsorption systems. Fortunately, a few ionic liquids containing metal chlorides have been synthesized, including [bmim]Zn 2 Cl 5 [12], that offer high hydrothermal stability and negligible vapor pressure to perfectly meet the absorbent criteria for absorption refrigeration. Compared with a solid adsorbent or other ionic liquids, the advantage of the ionic liquid [bmim]Zn 2 Cl 5 is self-evident. The chemical reaction between NH 3 and Zn 2+ will largely enhance the solubility of NH 3 in the absorbent and reduce the pressure of vapor phase as well as in the NH 3 -CaCl 2 system, and no defects in heat and mass transfer, swelling, or agglomeration are a problem. Some other metal cations such as Ni 2+ [13], Cu 2+ [14], and Fe 3+ [15] were also found to dissolve in ionic liquids, and thus, a family of new conceptual chemical absorption refrigeration working pairs consisting of ammonia and metal chloride-containing ionic liquids seems ready to be developed.

VLE behavior of the binary system of NH 3 -[bmim]Zn 2 Cl 5
In order to reveal the promising latent application of NH 3 -[bmim]Zn 2 Cl 5 as a working pair in absorption refrigeration, the vapor pressure data of the binary system of [bmim]Zn 2 Cl 5 /NH 3 are urgently needed. In our previous work [16], VLE data for the binary system of NH 3 -[bmim]Zn 2 Cl 5 were measured and fitted using the modified UNIFAC (Dortmund) model. [16] Table 2.

Experimental data
The uncertainties in the NH 3 mole fraction in the binary solution, which can be due to the random and systematic errors in the experimental method and the calculation accuracy of the ammonia equation of state (EOS), are also presented in the table.

The modified UNIFAC (Dortmund) model [16]
Because of the non-volatilization of the ionic liquid [bmim]Zn 2 Cl 5 , the vapor phase of the binary system [bmim]Zn 2 Cl 5 (1) + NH 3 (2) consists only of NH 3 , and the total pressure p of the binary solution can be given by [17], where x 2 is the mole fraction of NH 3 in the binary solution, γ 2 is the activity coefficient of NH 3 , V 2 L is the liquid mole volume of NH 3 , B 2 is the second virial coefficient in the ammonia EOS, and P 2 S' is the vapor pressure of pure NH 3 , When the temperature T is below the T C , P 2 S' is equal to the saturation vapor pressure, P S , which can be calculated by [18], where T r is the ratio of the solution temperature T and T C , When the temperature T is higher than the T C , the P 2 S' is defined as the pure NH 3 pressure at T and the critical mole fraction V C , which can be calculated by the RK type EOS as follows [19]: where the temperature dependent term α(T) can be written by: The EOS constants for NH 3 β k and the critical parameters T C, V C, and P C are given in Table 3. In the UNIFAC model, the excess Gibbs free energy is composed of two contributing parts, the combinatorial part and the residual part, and the activity coefficient γ i can be given as follows: ln ln ln where γ i C is the combinatorial activity coefficient and γ i R is the residual activity coefficient.
The combinatorial activity coefficient γ i C describes the repulsive interaction attributed to the molecular size and shape, which can be calculated by: where Z is normally set to 10, v k (i) is the number of group k in component i, and R k and Q k are the volume and surface parameters of the group k, respectively. The values of R k and Q k for the group used in our experiment are listed in Table 4.  The residual activity coefficient γ i R accounts for the intermolecular forces resulting from the corresponding group interaction, which is described as the summation of the group activity coefficient Γ for group k of component i, where Γ k and Γ k (i) are the activity coefficients for group k in binary solution and in the component i, respectively, and can be described as: Eqs. 15-17 can also be used to calculate lnΓ k (i) , except that the group composition variable X m is now the group fraction of group k in component i. For the modified UNIFAC (Dortmund) model, the group interaction parameters between groups n and m, φ n,m , is described as: where a nm (K -1 ), b nm , and c nm (K) are the adjustable interaction parameters for correlating the experimental vapor pressure data. The corresponding correlation results are listed in Table 5.  The P-T-x phase diagrams with symbols for experimental data and lines for the UNIFAC model calculations are shown in Fig. 11. It can be seen from the figure that with an increase in the NH 3 mole fraction, the vapor pressure also increases, and the rising trend becomes increasingly more obvious. With an increase in the binary solution temperature, the vapor pressure increases rapidly. When the temperature is below the T C , the rate of increase becomes more rapid, but when the temperature is higher than the T C of NH 3 , the rate of increase tends to slow and the vapor pressure even declines slightly. Figure 12. Comparison of experimental data and the UNIFAC model calculations [16].    Fig. 16(b) shows the c p -T diagram for T = 243.15-383.15 K. The symbols represent the experimental data, and the lines represent the calculated conic curve. It can be seen that variation in the specific heat capacity with temperature is accurately described by the quadratic curve.

Experimental data for excess enthalpy of NH 3 -[bmim]Zn 2 Cl 5
Temperature-component-molar excess enthalpy (T-x-H E ) data for the binary systems  Table 6. Uncertainties in the temperature, ammonia mole fraction, and molar excess enthalpy are also presented in the table. The uncertainties are due to random errors as well as systematic errors for the experimental apparatus and the calculation accuracy of the UNIFAC model for the VLE of NH 3 /[bmim]Zn 2 Cl 5 . With an increase in the ammonia mole fraction, x 1 , the molar excess enthalpy presents an increasing trend after an initial decline.

NRTL model
Based on the local composition representation of the excess Gibbs energy, G E , Renon and Prausnitz [24] proposed the NRTL model. The G E for the NRTL model can be described by: where g ij and g jj are the interaction energy between ij and jj component pairs, respectively, and α is the non-random parameter. The relationship between G E and the activity coefficient is Therefore, the activity coefficients of components 1 and 2 in a binary mixture can be written as: The definition of the activity coefficient for ammonia, γ 1 , is presented in our previous work [16]. For the NRTL model, the interaction energy between the ij and jj component pairs are defined as: The Gibbs-Helmholtz equation for excess enthalpy is: For the five-parameter NRTL model, the excess enthalpy can be calculated by: The correlation results are shown in Table 7.  Table 7. Binary parameters and non-random parameters for NRTL model Figure 17.     Fig. 18 shows the absolute deviations and relative deviations between the experimental data and the values predicted by the NRTL model for excess enthalpy data. The results indicate that all deviations for excess enthalpy data are less than 3.9%. The measurement deviations are mainly produced by the uncertainties in volumes of the high pressure vessels (0.5%), the little tank (0.5%) and the liquid phase of binary system (0.2%); the weights of [bmim]Zn 2 Cl 5 (0.01%), NH 3 (0.05%) and water (0.01%); temperature distributions in the water bath (1.4%) and the bath container (1.2%); and the UNIFAC calculation accuracies (0.9%). Based on the above uncertainties, the total uncertainty of measurement is estimated to be less than 4.8 %.

Enthalpy of [bmim]Zn 2 Cl 5 /NH 3 solution
The enthalpy of a solution of [bmim]Zn 2 Cl 5 /NH 3 at T and a given NH 3 mass fraction ω 1 can be calculated by: The enthalpy of [bmim]Zn 2 Cl 5 , h 1 , can be calculated by: where c p is the specific heat capacity of [bmim]Zn 2 Cl 5 , which can be calculated using Eq. (1), and T 0 is defined as 273.15 K. The enthalpy of NH 3 can be calculated by [19]: ( )

Thermodynamic analysis of an absorption system using NH 3 -[bmim]Zn 2 Cl 5 as the working pair [25]
In our previous work, the modified UNIFAC model was used to describe the VLE properties of [bmim]Zn 2 Cl 5 /NH 3 [16] and the NRTL model was used to predict the excess enthalpic properties of [bmim]Zn 2 Cl 5 /NH 3 . Based on a single-effect absorption refrigeration model, the thermodynamic performance of the [bmim]Zn 2 Cl 5 /NH 3 absorption system was simulated and compared with that of the NaSCN/NH 3 adsorption system. The coefficients of performance for cooling (COP) and heating (COP * ) and circulation ratios under the condition of a subzero evaporating temperature were calculated and analyzed.   In order to simulate the thermodynamic performance of an absorption system using [bmim]Zn 2 Cl 5 /NH 3 as a working pair, several assumptions were made as follows [25]:

System description and simulation
1. The simulation is conducted under steady state; 2. The vapor pressure losses are neglected, the pressure of the evaporator is equal to that of the absorber, and the pressure of the condenser is equal to that of the generator; 3. The refrigerant flowing out of the condenser is in a saturated liquid state, and the refrigerant flowing out of the evaporator is in a saturated gas state;

4.
The heat recovery rate of the regenerator is set to 0.80 [26]; and 5. The thermal losses and pumping work are negligible.
The mass and energy conservation equations for the evaporator are given by: The mass and energy conservation equations for the condenser are given by: For the regenerator, the energy conservation equation is given by: Based on the above assumptions and the conservation equations for mass and energy conservation, the heat flow values of q G , q C , q E , and q A ; mass flow values of m 2 and m 3 ; and mass fractions of ω 2 and ω 3 , can be calculated. The circulation ratio (f) is calculated by: The COP for cooling is defined by: The exergy efficiency (η ex ) for cooling is given by: The COP * for heating is defined as: The exergy efficiency for cooling (η * ex ) is given by:  Fig. 22 shows variations in the COP and η ex of [bmim]Zn 2 Cl 5 /NH 3 absorption refrigeration with variations in t A and t C at a t G = 90 °C and t E = -10 °C. These results show that both the COP and η ex decline with increases in t A and t C . This is because increases in t A and t C lead to a decrease in the mass fraction of solution from the absorber (ω 2 ) and an increase in the mass fraction of solution from the generator (ω 3 ). These changes in both ω 2 and ω 3 result in a decrease of q E . The slopes of both the COP and η ex curves are less steep when t A and t C are lower, and as t A and t C continue to increase, the slopes become increasingly steep. This can be explained by the fact that,with the continuous increases in t A and t C , the difference between ω 2 and ω 3 continues to become smaller. By comparison, the thermal performance of the [bmim]Zn 2 Cl 5 /NH 3 system is better than that of the NaSCN/NH 3 system when t A and t C are low. However, when t A and t C are high, the thermal performance of the NaSCN/NH 3 system is better than that of the [bmim]Zn 2 Cl 5 /NH 3 system, and the upper operating limit of t A and t C for NaSCN/NH 3 is higher than that for the [bmim]Zn 2 Cl 5 /NH 3 system. This can also be explained by the properties of NH 3 solubility in [bmim]Zn 2 Cl 5 and NaSCN. The higher solubility of NH 3 in [bmim]Zn 2 Cl 5 ensures that the [bmim]Zn 2 Cl 5 /NH 3 system possesses better thermal performance than the NaSCN/NH 3 system with operating conditions of low t A and t C . The stronger combination of NH 3 and [bmim]Zn 2 Cl 5 demonstrates that the upper operating limit of t A and t C for the [bmim]Zn 2 Cl 5 /NH 3 system are lower than those of the NaSCN/NH 3 system.  With an increase in t C , the COP presents a trend of first increasing and the decreasing. The reason for this trend is that the increase in t G has both positive and negative effects on the COP. The positive and negative effects are the same as indicated by the analysis of the trend in COP shown in Fig. 2. When t G is lower, the positive effect is predominant, and the COP increases with an increase in t G . With a further increase in t G , the negative effect is gradually enhanced, and the rate at which the COP increases is continually reduced until it finally becomes negative. When t G is higher, the negative effect is predominant, and the COP decreases with an increase in t G . For t E = -10 °C, -20 °C, -30 °C, and -40 °C, the maximum COPs for the [bmim]Zn 2 Cl 5 /NH 3 system of 0.54, 0.48, 0.42, and 0.35 appear at t G =133 °C, 161 °C, 188°C , and 225 °C, respectively. When t E = -10 °C and -20 °C, the maximum COPs for the NaSCN/ NH 3 system occurs at t G = 81 °C and 98 °C, respectively [27]. These results indicate that the required temperature of the heat source for the [bmim]Zn 2 Cl 5 /NH 3 system is higher than that of the NaSCN/NH 3 system.   With an increase in t G , the f declines, because the increase in t G is conducive to desorption of NH 3 in the generator. With an increase in t E , the f grows, because the increase in t E will decrease the absorption pressure of the absorber. Thus, the absorption ability of the absorber will be decreased. The f is an important parameter for absorption refrigeration. An increase in the f will lead to an increase in the amount of energy used to heat the solution from t A to t G . If the f is greater than 10, the COP decreases, even when the efficiency of the regenerator is greater than 0.9. For t E = -10 °C, -20 °C, and -30 °C, the circulation ratios are less than 10 when t G is greater than 115°C, 139 °C, or 187 °C, respectively. Based on the results shown in Figs. 23 and 24, the [bmim]Zn 2 Cl 5 /NH 3 system can be used when t E = -10 to -30 °C. The COP * decreases with increases in t A and t C , because the increase in t A is not conducive to absorption of NH 3 in the absorber. In addition, the increase in t C is not conducive to desorption of NH 3 in the generator. Figure 26. Effects of t G on η * ex for t G = 170-300 °C with t A = t C = 55 °C, 57 °C, 59 °C, or 61 °C and t E = -10 °C [25]. Fig. 26 shows the effects of t G on η * ex for t G = 170-300 °C with t A = t C = 55 °C, 57 °C, 59 °C, or 61°C and t E = -10 °C. With an increase in t G , the η * ex presents a trend of decreasing after initially increasing. For t A = t C = 55 °C, 57 °C, 59 °C, and 61 °C, the maximum values of η * ex of 0.341, 0.368, 0.384, and 0.402 appear at t G =193 °C, 206 °C, 221 °C, and 246 °C, respectively. It can be seen that the optimalη * ex occurs at a lower t G than did the optimal COP * . This is because the increase in t G leads to an increase in the exergy proportion in q E , which induces a decreasing trend in η * ex but has no effect on the COP * . Figure 27. Effects of t G on the f for t G = 170-300 °C with t A = t C = 55 °C, 57 °C, 59 °C, or 61 °C and t E = -10 °C [25].  The theoretical cycle characteristic of the [bmim]Zn 2 Cl 5 /NH 3 absorption system is also compared with that of the LiBr/H 2 O system. Fig. 28 shows the effects of t G on the COP for t G = 75-130 °C with t A = 35°C, t C = 40 °C, and t E = 5 °C and a heat recovery rate of the regenerator of 0.75. For both systems, the COP initially exhibits a significant increase as the t G increases. As t G continues to increase though, the slope of the COP curve for the [

Summary
The vapor pressures of the binary solution of [bmim]Zn 2 Cl 5 /NH 3 with NH 3 mole fraction x 2 = 0.83-0.94 at T = 323.15-563.15 K were measured via a static method with a total uncertainty of measurement below 4.3% [16]. The experimental data were fit using the modified UNIFAC model, and new group interaction parameters between any two of the four tested groups were obtained with a maximum deviation less than 5% [16]. Based on the modified UNIFAC model and the NRTL model, the thermodynamic performance of a single effect absorption system using [bmim]Zn 2 Cl 5 /NH 3 as the working pair was simulated and compared with those of the NaSCN/NH 3 adsorption system [25] and the H 2 O-LiBr absorption system. The thermal performance of the [bmim]Zn 2 Cl 5 /NH 3 system is better than that of the NaSCN/NH 3 system when the t G is high and t A and t C are low and also better than that of the H 2 O-LiBr absorption system in some cases. With an increase in t G , the COP and COP * present trends of declining after increasing, and the circulation ratios show a decreasing trend. When t E = -30 °C and t A = t C =35 °C, the maximum COP of the [bmim]Zn 2 Cl 5 /NH 3 system is still greater than 0.42. When t E = -10 °C and t A = t C = 60 °C, the maximum COP * is still greater than 1.40. Under these two operating conditions, the circulation ratios remain acceptable. Although the COP of the [bmim]Zn 2 Cl 5 /NH 3 system is less than that of the LiBr/H 2 O system in some specific temperature ranges, the overall theorertical cycle characteristic of the [bmim]Zn 2 Cl 5 /NH 3 system is slightly better than that of the LiBr/H 2 O system, especially when t G is high. Overall, these results indicate that the [bmim]Zn 2 Cl 5 /NH 3 absorption system offers good thermal performance for use in both cooling and heating applications.

Conclusions and outlook
Ten years have passed since ionic liquids were introduced in the field of absorption refrigeration, and unfortunately, the research progress pertaining to absorption refrigeration working pairs containing ionic liquids has been disappointing to us. Most of the working pairs proposed by researchers from all over the world are gradually fading from view due to a lack of practical applications.  3 OH are better than that of H 2 O-NH 3 , but still slightly lower than those of LiBr-H 2 O. So far, the new conceptual chemical absorption refrigeration working pairs containing an ionic liquid, with [bmim]Zn 2 Cl 5 /NH 3 as the representative, are the most ideal ionic liquid-type working pairs for absorption refrigeration. The thermodynamic performances of absorption refrigeration using the proposed chemical working pairs are comparable to those achieved with the LiBr-H 2 O system. Additionally, the ranges of operating conditions for the chemical working pairs are wider than those of the conventional working pairs. At present, promoting the industrial application of [bmim]Zn 2 Cl 5 /NH 3 is our next priority. The discovery of chemical absorption refrigeration working pairs containing an ionic liquid is a milestone in the development of absorption refrigeration technology. It is foreseeable that the application of ionic liquids in absorption refrigeration will achieve a major breakthrough in the development of this technology, with the continued discovery of similar ionic liquid working pairs based on the chemical reaction.