Comprehensive performance analysis and optimization of 1,3-dimethylimidazolylium dimethylphosphate-water binary mixture for a single effect absorption refrigeration system

Abstract

The energy and exergy analyses of the absorption refrigeration system (ARS) using H₂O-[mmim][DMP] mixture were investigated for a wide range of temperature. The equilibrium Dühring (P-T-X_IL) and enthalpy (h-T-X_IL) of mixture were assessed using the excess Gibbs free non-random two liquid (NRTL) model for a temperature range of 20°C to 140°C and X_IL from 0.1 to 0.9. The performance validation of the ARS cycle showed a better coefficient of performance (COP) of 0.834 for H₂O-[mmim][DMP] in comparison to NH₃-H₂O, H₂O-LiBr, H₂O-[emim][DMP], and H₂O-[emim][BF4]. Further, ARS performances with various operating temperatures of the absorber (T_a), condenser (T_c), generator (T_g), and evaporator (T_e) were simulated and optimized for a maximum COP and exergetic COP (ECOP). The effects of T_g from 50°C to 150°C and T_a and T_c from 30°C to 50°C on COP and ECOP, the X_a, X_g, and circulation ratio (CR) of the ARS were evaluated and optimized for T_e from 5°C to 15°C. The optimization revealed that T_g needed to achieve a maximum COP which was more than that for a maximum ECOP. Therefore, this investigation provides criteria to select low grade heat source temperature. Most of the series flow of the cases of cooling water from the condenser to the absorber was found to be better than the absorber to the condenser.

Graphical abstract

Keywords

ionic liquid driven absorption cycle / H₂O-[mmim][DMP] / coefficient of performance (COP) / exergy analysis / thermodynamics mixture property

1 Introduction

Global warming and climate alteration have diverted toward environment friendly, renewable and sustainable cooling processes to fulfill the rising demand of humanity. The current cooling technology has been dominated by the electricity-driven system, wherein more than 75% of the electricity is obtained via nonrenewable fossil fuel sources like coal and petroleum liquids [1,2]. The thermal driven cooling technology involves a significant commitment in the advancement with the use of low-grade thermal energy, as well as the saving of high-grade electricity [3,4]. The low-grade energy (low temperature) is accessible through waste heat recovery from an industrial process or utilization of renewable thermal energy accessed from solar [5], biomass, and geothermal energy [6]. Nowadays, the cooling demand is rising worldwide, which is fulfilled mostly by the mechanical driven vapor compression system and the thermal powered absorption refrigeration system (ARS) [7]. The ARS system utilizes low-grade heat more efficiently to generate cooling effect and becomes an alternative to the high-grade energy-driven mechanical vapor compression system. The thermal driven ARS indirectly reduces primary energy consumption and simultaneously cuts the emissions of harmful CO₂ into the atmosphere. Such a thermal driven ARS could be a particular solution for Middle East countries with a high solar intensity (2113 kWh/(m²·a) with more than 70% of energy consumption for building cooling) like Qatar to fulfill rising air conditioning demand in summer [8–10].

Commercially, two working mixtures (H₂O-LiBr and H₂O-NH₃) have been employed in the ARS for air conditioning and refrigeration application [11]. In these working pairs, H₂O-LiBr (absorbent: lithium bromide LiBr, refrigerant: H₂O) is efficient for a temperature of above 5°C and widely used for centralized air conditioning with a capacity of more than 50 TR. The two main concerns of H₂O-LiBr are crystallization and high corrosion with metal over long period use and a rise in temperature. Therefore, alternative potential working pairs to replace H₂O-LiBr are necessary to overcome the crystallization and corrosion problem and to utilize low-grade energy more efficiently.

Recently, ionic liquids (ILs) as absorbent with water, alcohol, and ammonia has been extensively investigated due to their good thermal and chemical stability, presence in a liquid state at a wide range of temperatures, tunable properties with adjustment cation and anion, and nonvolatile with zero vapor pressure [12–16]. For these reasons, ILs can be the most prospective absorbent in combination with various types of refrigerants such as water [17–19], hydrofluorocarbons [20–23], ammonia [24–26], and alcohol [27–29] as the alternative working pair for the ARS with applications like air conditioning, heat transformer, heat pumping, and refrigeration purpose [30–32].

Kim and Kohl [33] have simulated two ILs [hmim][PF6] and [hmim][Tf₂N], with refrigerant R134a for the ARS and estimated mixture thermodynamics properties by using Redlich-Kwong as an equation-of-state. Their simulation results have suggested that [hmim][Tf₂N]-R134a exhibits a better performance compared to [hmim][PF6]-R134a. Sujata and Venkatarathnam [34] have investigated five imidazolium-based ILs with ammonia as an alternative working pair to vapor absorption heat transformer using the PC–SAFT equation of state. They have stated that ILs with ammonia can be considered as a possible alternative in absorption heat transformers for heating. Popp et al. [31] have categorized most promising IL-H₂O based working pairs, i.e., H₂O-[mmim][C₂H₅COO], H₂O-[mmim][HCOO], H₂O-[Me₄N][OAc], H₂O-[mmim][OAc], H₂O-[Me₄N][HCOO], and H₂O-[Me₄N] [C₂H₅COO] as solar heat pump purpose. Again, the absorption heat transformer performance has been studied by Chen and Liang [29] using CH₃OH-[mmim][DMP] and H₂O-[mmim][DMP] wherein the heating COP of both pairs are found to be lower than the conventional LiBr-H₂O. They have reported a maximum exergy loss for [mmim][DMP]-H₂O in comparison to [mmim][DMP]-CH₃OH and LiBr-H₂O.

Chen et al. [35] have studied the use of [mmim] [DMP] with refrigerant CH₃OH as working fluids for triple effect compression of the ARS with a phosphoric acid fuel cell. They have concluded that the exergy losses, fuel cell efficiency, and heat transfer characteristic determine the overall performance of the proposed hybrid system. Wu et al. [36] have studied six variants of hydrofluorocarbons/hydrofluoroolefins based refrigerant with IL [HMIM][Tf₂N] as a working pair for a hybrid compression ARS. Their simulation outcomes demonstrated an enhanced COP from 0.191–0.463 to 0.366–0.670 for the proposed compression-ARS in comparison to a normal ARS. They concluded that the low-pressure compression-ARS is more efficient with a reduced compressor exit temperature.

Zheng et al. [30] have reviewed potential imidazolium type IL refrigerant combination as working pairs for the ARS based on the thermophysical property of mixture, the status of evaluation with selection methods, performance modeling and assessments, in which, the working pair [mimm] [DMP]-H₂O reveals a maximum COP of 0.829 for a single effect ARS at a T_g, T_e, T_a, and T_c of 80°C, 10°C, 30°C, and 40°C. Thermo-physio-chemical properties such as vapor-liquid equilibrium, heat capacity IL, excess enthalpy, density, and heat capacity of the mixture of H₂O-[mimm][DMP] mixture have been investigated by He et al. and Dong et al. for X_IL (mass fraction) from 0.1 to 0.9 [28,37].

According to Dong el al [37], the proposed H₂O-[mimm][DMP] mixture can be the most promising alternative to LiBr-H₂O. As per the authors’ understanding, this is the only published result for the performance simulation which is available for limited conditions. Therefore, it is necessary to explore the energy and exergy-based efficiency of the ARS for cooling for a wide range of operating conditions. Hence, comprehensive energy and exergy analysis with optimization of H₂O-[mimm][DMP] mixture is essential. The proposed study involves the use of a wide range of temperature along with the evaluation of thermodynamics properties and performance evaluation using Frst Law and Second Law of Thermodynamics, wherein the exergy performance analysis of a single effect ARS can provide irreversibility and entropy generation in each component [33,38–41]. Therefore, the main aim of this paper is to assess the performance H₂O-[mimm][DMP] based ARS and the activity coefficient of water obtained through the regressed NRTL model. The separate computer algorithm was generated in Scilab (mathematical open source software) to evaluate the equilibrium mixture properties like total bubble pressure (P-T-X_IL), composition X_IL, enthalpy (h-T-X_IL) of the proposed H₂O-[mimm][DMP] binary mixture, the thermodynamic properties of refrigerant water in the pure state and to compute the performances parameters (COP, ECOP, and CR) of the ARS. For this reason, the energy and mass balance (total and component) across the ARS and each component of the ARS were employed to compute the energy and exergy-based COP. The performance calculations of the proposed IL-based absorbent [mmim][DMP] with the LiBr, NH₃ and three [emim][DMP], EMISE, and [emim][BF₄] were performed. Finally, the effect T_g, T_c, and T_a on the COP and ECOP were comprehensively studied and optimized for a wide operating temperature like T_e of 5°C to 15°C, T_g of 50°C to 150°C, T_a and T_c of 30°C–50°C for efficient use of low-grade heat.

2 Methodology

2.1 Absorption refrigeration system

Figure 1 presents the general scheme of a single effect absorption refrigeration cycle used for the assessment of the proposed H₂O-[mimm][DMP]- as a working pair.

The main items of the single-effect ARS comprise of an absorber and a condenser to dissipate heat to an ambient generator to gain heat from low-grade resources and an evaporator to generate cooling at the required temperature. A heat exchanger (SHE) is installed to recover heat between strong and weak solutions and help to reduce the heat input to the generator. This SHE enhances the performance of the ARS with a reduction in the total input low-grade heat obtained from renewable sources. At a steady-state, mixture composition within the generator and the absorber are in equilibrium and the absorption refrigeration cycle operates in two pressure levels. The absorber and the evaporator operate at a low-pressure (P_a and P_e) whereas the generator and the condenser at a high pressure (P_g and P_c). In the absorption cooling cycle, the cooling effect is generated with the evaporation of liquid H₂O into vapor within the evaporator as the exchange of heat from the space to be cooled. The generated low-pressure vapor is absorbed into the strong solution (There is more X_IL in the generator than in the absorber.) and returns to the bottom of the generator. The heat of the absorption due to the refrigerant vapor absorption into the strong X_IL solution is dissipated by supplying cooling water- or air-cooled type heat exchanger. The absorber strong IL-H₂O mixture is propelled via electrically driven pump to the high-pressure generator as a feed wherein the mixture separates into a strong IL-H₂O mixture and a pure H₂O vapor since the IL is nonvolatile and has a very low vapor pressure. The low-temperature heat obtained from the sun or waste heat can be used to input the heat load to the generator. The produced pure refrigerant H₂O vapor is converted into the liquid state with the removal of latent heat of condensation via a cooling water circuit flowing through the condenser. Overall, the condenser and the absorber dissipate the heat of absorption and condensation to the ambient temperature. Therefore, the performance analysis of the ARS based on the direction of the flow of cooling water is critical. The absorption cooling cycle continues with the expansion of pressurized liquid refrigerant through the isentropic expansion valve.

2.2 Thermodynamic properties of H₂O(1)-[mmim][DMP](2)

The widely accepted regression-based IAPWS formulations are used to determine vapor-liquid equilibrium properties of pure H₂O which is refrigerant [42]. The activity coefficient based non-random two liquid (NRTL) model has been widely considered to correlate vapor-liquid equilibrium H₂O-[mmim][DMP] mixture due to its proven accuracy [18,20,23,32]. Due to the zero vapor pressure of the IL, the total pressure of the binary mixture is expressed by water vapor or partial pressure of water. The relation among the equilibrium bubble pressure, composition, and temperature of H₂O-[mmim][DMP] mixture is expressed as

(1)p_1=P=P_1^{\mathrm{s}\mathrm{a}\mathrm{t}}{x}_1{\gamma }_1,

where {{\displaystyle P}}_{\mathrm{1}}^{\mathrm{s}\mathrm{a}\mathrm{t}} is the vapor pressure of refrigerant H₂O for loaded composition x₁ and the water activity coefficient ({{\displaystyle \gamma }}_{\mathrm{1}}) is correlated by the local composition-based activity coefficient NRTL model. The excess Gibbs free energy (g^{\mathrm{E}}) using NRTL for the proposed H₂O(1)-[mmim][DMP](2) mixture with activity coefficient are expressed as

(2)g^{\mathrm{E}}=RT{x}_1{x}_2[\frac{G_{21}{\tau }_{21}}{x_1+x_2{G}_{21}}+\frac{G_{12}{\tau }_{12}}{x_2+x_1{G}_{12}}],

(3)\ln {\gamma }_1=x_2^2[{\tau }_{21}{\left(\frac{G_{21}}{x_1+x_2{G}_{21}}\right)}^2+\frac{G_{12}{\tau }_{12}}{{(x_2+x_1{G}_{12})}^2}],

(4)\ln {\gamma }_2=x_1^2[{\tau }_{12}{\left(\frac{G_{12}}{x_2+x_1{G}_{12}}\right)}^2+\frac{G_{21}{\tau }_{21}}{{(x_1+x_2{G}_{21})}^2}],

(5)G_{12}=\exp (-{\alpha \tau }_{12}),G_{21}=\exp (-{\alpha \tau }_{21}),

(6){\tau }_{12}={{\tau }^0}_{12}+\frac{{{\tau }^1}_{12}}{T},{\tau }_{21}={{\tau }^0}_{21}+\frac{{{\tau }^1}_{21}}{T},

where α is a non-randomness factor; {{\displaystyle {\tau }^{\mathrm{0}}}}_{\mathrm{12}}, {{\displaystyle {\tau }^{\mathrm{1}}}}_{\mathrm{12}}, {{\displaystyle {\tau }^{\mathrm{0}}}}_{\mathrm{21}}, and {{\displaystyle {\tau }^{\mathrm{1}}}}_{\mathrm{21}} are model parameters computed by fitting the experimental VLE data and the NRTL model with the minimum percentage average relative deviation (ARD) of 1.44% [37] as listed in Table 1.

The excess enthalpy in terms of excess Gibbs free energy and enthalpy of the mixture is computed as

(7)h^{\mathrm{E}}={-RT}^2{\left[\frac{\mathrm{d}(g^{\mathrm{E}}/RT)}{\mathrm{d}T}\right]}_P,

(8)h=x_1{h}_1+x_2{h}_2+h^{\mathrm{E}},

where x is the composition of each component in the solution; h₁ is the enthalpy of refrigerant H₂O, which is the function of temperature and respective saturated pressure; the value of h₁ of pure water is calculated from the IAPWS formulation [42]; and h₂ is the enthalpy of [mmim][DMP] at pure state, whose value is calculated from the heat capacity (C_p2) of IL (C_p is heat capacity, and 2 indicates the type component i.e., IL) and temperature difference as reported in Ref. [28].

(9)h_2=C_{{\mathrm{p}}_2}\mathrm{\Delta }T\mathrm{.}

2.3 Performance modeling and assessment

Figure 2 depicts the model flowchart applied to compute the energy and exergy analysis of a [mmim][DMP]-H₂O driven single stage absorption system. A complete computer program to solve the mathematical model to compute performances is established in mathematical programming software SCILAB 6. The detailed computer program comprises of evaluating thermodynamics properties of refrigerant H₂O by using well known IAPWS formulation, computing the thermodynamics properties of the mixture of [mmim][DMP]-H₂O by using the thermodynamic activity coefficient models, and solving overall and component mass and energy balance equations of single effect ARS and calculating heat load, COP, ECOP, and CR based on the state properties in terms of T, P, X, h, and m.

The mathematical modeling and simulation are based on the assumption that the ARS is in a steady-state; the pressure drop within the ARS component and the connecting pipe are neglected; the sink or reference temperature (T_ref) used for calculation of exergy-based COP (ECOP) calculation is 25°C; the power required for pumping refrigerant strong solutions from low to the high-pressure side is assumed to be zero [8,19,36,43,44]; the state points 1–10 are in thermodynamic equilibrium; the binary mixture within the low-pressure absorber and high-pressure side generator is in a vapor-liquid equilibrium state; and the refrigerant within the condenser and the evaporator is in a saturated condition based on input temperature.

With these assumptions, the ARS analysis is performed based on the refrigerant mass flowrate of 1 kg/s and with consideration of mass, energy, and exergy balance based on the thermodynamics principles.

The expression used for total and component mass balance across each unit (the absorber, evaporator, condenser, generator, and SHE) of the ARS are

(10){\dot{m}}_{\mathrm{i}\mathrm{n}}^{}-{\dot{m}}_{\mathrm{o}\mathrm{u}\mathrm{t}}^{}=0,

(11){\left(\stackrel{}{\dot{m}}X\right)}_{\mathrm{i}\mathrm{n}}-{\left(\stackrel{}{\dot{m}}X\right)}_{\mathrm{o}\mathrm{u}\mathrm{t}}=0.

The circulation ratio (CR) decides both the amount of consumption of high-grade electricity and the ratio of solution mass flowrate (m₅) from the absorber to the generator to the total refrigerant mass flowrate (m₁) of 1 kg/s, which depends on the equilibrium mass fraction of [mmim][DMP] within the absorber and the generator as

(12)\mathrm{C}\mathrm{R}={\dot{m}}_5/{\dot{m}}_1=X_{\mathrm{g}}/(X_{\mathrm{g}}-X_{\mathrm{a}})\mathrm{.}

The overall and unit wise energy balance expression of ARS are given in Eqs. (13) and (14), respectively. The energy balance expression comprises the heat dissipation through the condenser and the absorber and the low-grade heat gain within the generator and cooling production in the evaporator via refrigerant evaporation at a low temperature. The overall energy balance across ARS is

(13)Q_{\mathrm{e}}+Q_{\mathrm{g}}-Q_{\mathrm{c}}-Q_{\mathrm{a}}=0.

The unit wise general energy balance expression contains the inlet and outlet enthalpy (h_{\mathrm{i}\mathrm{n}} and h_{\mathrm{o}\mathrm{u}\mathrm{t}} of each stream with heat gain and dissipation ({\dot{m}\dot{Q}}_{\mathrm{i}\mathrm{n}}^{} and {\dot{Q}}_{\mathrm{o}\mathrm{u}\mathrm{t}}^{}).

(14){\left(\dot{m}h\right)}_{\mathrm{i}\mathrm{n}}-{{\left(\dot{m}h\right)}_{\mathrm{o}\mathrm{u}\mathrm{t}}}_{}+{\dot{Q}}_{\mathrm{i}\mathrm{n}}^{}+{\dot{m}\dot{Q}}_{\mathrm{o}\mathrm{u}\mathrm{t}}^{}=0.

The heat duty of the evaporator (Q_e), the generator (Q_g), the condenser (Q_c), the absorber (Q_a), and the heat recovery exchangers (Q_SHE) are

(15)Q_{\mathrm{e}}=m_4{h}_4-m_3{h}_3=m_1(h_4-h_3),

(16)Q_{\mathrm{g}}=m_1{h}_1-m_7{h}_7+m_8{h}_8,

(17)Q_{\mathrm{c}}=m_1{h}_1-m_2{h}_2=m_1(h_1-h_2),

(18)Q_{\mathrm{a}}=m_{10}{h}_{10}+m_4{h}_4-m_5{h}_5,

(19a)Q_{\mathrm{S}\mathrm{H}\mathrm{E}}=m_8(h_8-h_9),

(19b)Q_{\mathrm{S}\mathrm{H}\mathrm{E}}=m_7(h_7-h_6)\mathrm{.}

Finally, the energy and exergy-based performances for the proposed H₂O-[mimm][DMP] mixture is estimated by simulations and by solution of mass and energy models, the evaluation of state properties of each stream, and the calculation of the heat load across each unit of the ARS. The COP and ECOP expression are respectively

(20)\mathrm{C}\mathrm{O}\mathrm{P}=Q_{\mathrm{e}}/Q_{\mathrm{g}},

(21)\mathrm{E}\mathrm{C}\mathrm{O}\mathrm{P}=\frac{Q_{\mathrm{e}}(T_{\mathrm{r}\mathrm{e}\mathrm{f}}/T_{\mathrm{e}}-1)}{Q_{\mathrm{g}}(1-T_{\mathrm{r}\mathrm{e}\mathrm{f}}/T_{\mathrm{g}})}\mathrm{.}

3 Result and discussion

The estimation of the mixture thermodynamics property like P-T-X and h-T-X is performed by using NRTL based excess Gibbs free energy. Besides, the resulted activity coefficient models are presented in detail. Moreover, the comprehensive performance analysis and the optimization single effect of the ARS based on energy and exergy performances are explored and optimized for utilization of low-grade heat more effectively for various configurations (parallel flow T_a = T_c, series flow from absorber to condenser, series flow from the condenser to the absorber) of T_a and T_c.

3.1 Model validation and comparison

As exhibited in Fig. 3, primarily the simulation outcomes of the single effect absorption cycle using H₂O-[mimm][DMP] were compared with the COP data published by Dong et al. [37]. This confirms the developed thermodynamics NRTL models and agrees well with the simulation methodology. Further, the comparison of the proposed H₂O-[mimm][DMP] mixture with the conventional working pairs H₂O-LiBr [45], NH₃-H₂O [45], and the two ILs, i.e., H₂O-[emim][BF₄] [46], H₂O-[emim] [DMP] [47] are illustrated in Table 2 for a T_g, T_a, T_c, and T_e of 100°C, 30°C, 40°C, and 10°C. These working pairs can be arranged in a declining order of COP as H₂O-[mimm][DMP] (COP= 0.834)>H₂O-LiBr (0.833)>H₂O-[emim][DMP] (0.822)>NH₃-H₂O (0.646)>H₂O-[emim][BF₄] (0.525). These COP comparisons of working pairs confirm that the best performance for the investigated H₂O-[mimm][DMP] mixture is with a maximum COP of 0.834 and that the proposed IL based absorbent can be used as an alternative to replace the typical absorbent LiBr. As demonstrated in Table 3, detailed state properties like P, T, X_IL, h, and m of the single effect ARS as shown in Fig. 1 are simulated using H₂O-[mimm][DMP] at a T_g, T_a, T_c and T_e of 100°C, 30°C, 40°C, and 10°C and a constant refrigerant flowrate of 1 kg/s. The tabulated data demonstrate an elevated CR of 6.44 for absorbent [mimm][DMP], which is 1.5 times the conventional absorbent LiBr. The result in the heat load of each components of the ARS (the generator, the condenser, the absorber, the evaporator, and the recovery heat exchanger) is presented in Fig. 4 based on the refrigerant flowrate of 1 kg/s. This generates a maximum cooling capacity of 2351.7 kW and an internal heat recovery of 732.17 kW.

3.2 Thermodynamics equilibrium properties of H₂O-[mmim][DMP] mixture

As displayed in Figs. 5, 6, and 7, two separate computer programs were developed in SCILAB to compute excess Gibbs free energy (g^E) and subsequently equilibrium thermodynamic properties (h-T-X and P-T-X) of a proposed binary mixture of H₂O-[mmim][DMP] by using the NRTL model. Variation of excess g^E with a mass fraction of [mmim][DMP] (X_IL) for temperatures range of 20°C– 140°C is displayed in Fig. 5. This covers the operating equilibrium condition of ARS that arises within the absorber and the generator. These g^E plots reveal that the investigated binary mixture is endothermic wherein the interactions of unlike molecules (in between IL and H₂O) are weaker in relation to the interaction between molecules IL-IL and H₂O–H₂O. Similar equilibrium enthalpy plot and the vapor pressure in terms of T and X_IL are shown in Figs. 6 and 7 respectively. These plots are helpful in choosing the operating pressure, the composition, and the enthalpy of the ARS based on input temperatures of T_e, T_a, T_g and T_c.

3.3 Performance assessment and optimization

3.3.1 Effect of T_g

The variation of COP, ECOP, CR, and mass fraction of [mmim][DMP] with the rise in the generator temperature for various evaporation temperatures of ARS are displayed in Figs. 8, 9, and 10 respectively at a constant T_a = T_c of 30°C. For each evaporation temperature, both the COP and ECOP start to increase, achieve maxima, and then start to decline with the rise in T_g. As a case for T_e of 5°C, the ARS becomes feasible to provide the cooling effect at a minimum generator temperature of 59°C at a COP of 0.157 and an ECOP of 0.11. Then, the cycle achieves a maximum COP of 0.858 at a T_g of 83°C and a maximum ECOP of 0.502 at a T_g of 63°C. After attaining the maxima, the COP and ECOP start to drop with the rise in T_g to more than 83°C and 63°C respectively. This shows that the rate of decline in the ECOP is sharper than that in the energy-based COP.

For a constant T_g, only the energy-based COP of the ARS improves with the upswing in T_e from 5°C to 15°C. Like the constant T_g of 80°C, the high evaporation of the temperature of 15°C shows a higher COP of 0.891 and a lower ECOP of 0.199 in comparison to the evaporation temperature of 5°C at a lower COP of 0.857 and a higher ECOP of 0.396. This enhanced COP performance at a higher T_e is expected due to the rise in refrigerant vapor absorption rate into the incoming refrigerant solution (high mass fraction of IL). Such a high T_e value increases the absorption pressure within the absorber, which enhances the low-pressure water vapor absorption capacity and the equilibrium water concentration into the binary mixture. The high T_e of 15°C has a high saturation pressure of 1.706 kPa which is 1.95 times more than the T_e of 5°C. These effect of absorption performance and mass fraction of ionic liquid [mmim][DMP] into the generator and the absorber is displayed in Fig. 11. For a constant T_e, the mass fraction difference of [mmim][DMP] in the generator and the absorber (X_g–X_a) is found to be in expanding direction with the increase in T_g and respective equilibrium X_g. The mixture flowrate and respective circulation ratio increases at a low T_g due to the low composition difference (X_g–X_a) to fulfill the designed refrigeration capacity at a refrigerant flowrate of 1 kg/s, as viewed in Figs. 10 and 11.

3.3.2 Effect of flow direction with the variation of T_a and T_c

The effect of the temperature of the condenser and the absorber to dissipate heat on the atmosphere is very important in deciding the performance of the ARS. These temperatures depend on the selection of the flow direction of cooling water which is needed to investigate and optimize the performance of the ARS. Here onwards, the effect of the flow of cooling water was discussed separately like the parallel flow with T_a = T_c, the series flow from T_a to T_c (the absorber to the condenser), and the reverse direction series flow from T_c to T_a (the condenser to the absorber).

1) Parallel flow at T_a = T_c

Figure 12 shows the impact of T_a = T_c on COP, ECOP, CR, and X_IL, at a constant T_e of 5°C. As displayed Figs. 12(a) and 12(b), both the COP and ECOP reduce with the increase at T_a = T_c from 30°C to 50°C. Besides, the maxima of COP and ECOP is diminished with the rise in temperature T_a = T_c. In another word, a lower value of T_a = T_c is more beneficial to utilization of low grade heat more efficiently and to give improved COP and ECOP. This enhanced performance is due to the better absorption capacity at a lowered condensation pressure. Therefore, both T_a and T_c determine the circulation ratio of the mixture based on the difference in the mass fraction of the IL within the absorber and the generator (X_g–X_a) as presented in Figs. 12(c) and 12(d), from which, it is observed that the mass fraction difference (X_g–X_a) expands with the rise in T_g. However, at a constant T_g, these differences increase for a low value of T_a = T_c, which also displays and confirms the required minimum generator temperature for the ARS for which the ARS becomes feasible to generate the cooling effect. It can be seen that the minimum generation temperature rises with the rise in T_a = T_c, like the highest at T_a = T_c of 50°C, and the lowest at T_a = T_c of 30°C. This is due to the improved refrigerant vapor absorption capacity in the absorber at a lowered absorption temperature which increases the mass fraction difference (X_g– X_a). As an example, at a T_g of 120°C, X_g–X_a is 0.1475 (X_a = 0.8243) at a T_a = T_c of 30°C, which is 7.97 times more than T_a = T_c of 50°C with an X_a of 0.916. At T_a = T_c of 50°C, the ARS become feasible to produce a cooling effect at a T_g of 110°C while it reduces to 71°C with a drop at T_a = T_c = 35°C. This clearly indicates that a low T_a and T_c favor the efficient use of low-grade energy for cooling purpose.

The optimum T_g to attain maximum performances of ARS (COP and ECOP) declines with the drop at T_a = T_c from 50°C to 30°C. As a case, the ARS in parallel flow mode at T_a = T_c of 30°C demonstrates a maximum COP of 0.86 at a T_g of 83°C and a maximum ECOP of 0.50 at a T_g of 63°C. However, with the rise at T_a = T_c to 50°C, the maximum COP and ECOP of the ARS is reduced to 0.78 at a T_g of 150°C and 0.21 at a T_g of 123°C respectively. In the case of CR, a higher T_g reduces the feed (refrigerant weak solution) solution flowrate (and CR) from the absorber to the generator due to enlarged IL mass fraction difference (X_g–X_a) and feed flowrate and CR become nearly constant at a higher T_g.

2) Series flow

From the previous condition of T_a = T_c, it can be noticed that the combination of T_a and T_c determines the performance of ARS for a given T_e and a supplied T_g. Accordingly, the performance variation with the series flow of cooling water from the absorber to the condenser and back from the condenser to the absorber are examined in Figs. 13 and 14 respectively. From Fig. 13, it is clearly observed that both performances decrease with the rise in T_c from 30°C to 40°C due to the rise in the pressure of the condenser. In this way, the equilibrium mass fraction of IL becomes increasingly more with the rise in the temperature of the generator. It can be seen that the maximum COP and ECOP of the ARS is found to be better for a lowered T_c of 30°C. Further, a T_c of 30°C requires a reduced generator temperature. As shown in Fig. 13(b), an optimum T_g to have the maximum COP of 0.83 and the maximum ECOP of 0.38 is 95°C and 76°C respectively at a T_c of 40°C. However, these optimum T_g at a T_c of 30°C is decreased to 83°C and 63°C to attain the maximum COP and ECOP, respectively. This optimum T_g increases to 76°C to obtain the maximum ECOP of 0.39 as T_c rises to 40°C. The effect of the rise in T_c from 30°C to 40°C on ECOP is very small for a T_g greater than 120°C. In this case, the ECOP can be considered in the similar range for the selected T_c.

Likewise, the COP and ECOP variation with the absorber temperature T_a from 30°C, 35°C, and 40°C is presented in Fig. 14 at a constant T_c of 30°C. Like the T_c effect, the COP and ECOP of the ARS are degraded with the acceleration in T_a. For a generator temperature of more than 120°C, the effect of T_a on the COP and ECOP is very minimal. This indicates that the variation of T_c affects both performances on large in comparison to the variation of T_a. The optimum T_g to give maximum performances is found to be in falling order with the rise in T_a from 30°C to 40°C.

3.3.3 Effect of T_c

Figure 15 shows the effect of condenser temperature (from 25°C to 50°C) on the COP and X_g with consideration of generator temperatures of 80°C and 90°C at a T_e of 5°C, 10°C, and 15°C. As illustrated in Fig. 15(a), the low value of T_c reveals a better COP for both a generator temperature of 80°C and 90°C. Like the T_e of 10°C, the T_c below 41°C gives a better COP for a generator temperature of 80°C than that of 90°C. Afterward, the COP trends to become the opposite wherein a generator temperature of 90°C shows a better COP than that of 80°C. Such a transition of T_c is found in increasing order with the rise in T_e from 5°C to 15°C. Like the T_e of 5°C, the transition condensation temperature is 32°C which increases to 50°C at a higher T_e of 15°C. This implies that for a higher value of condenser temperature, a high evaporation temperature for the ARS is more useful for the efficient utilization of low-grade heat. Figure 15(b) displays the X_g variation plot with the rise in T_c for a generator temperature of 80°C and 90°C. This shows that the reduction in mass fraction of IL (X_g) within the generator with the rise in T_c from 25°C to 30°C as the saturation pressure of condensation within the generator rises with the increase in T_c. For a constant T_c, X_g is greater for the T_g of 90°C than that of 80°C due to the constant saturation pressure within the high-pressure side of the ARS.

3.3.4 Effect of T_a

The effect of T_a on COP and X_a is discussed individually as shown in Figs. 16(a) and 16(b) at a T_e of 5°C, 10°C, and 15°C and two low generator temperatures of 80°C and 90°C. Like the previous subsection, the COP of the absorption system for cooling is found to be better for a low absorption temperature. This cooling performance is found to be reduced with an increase of absorption temperature from 30°C to 40°C as revealed in Fig. 16(a). The rate at which the COP performance decline with the rise in T_a is faster for a low T_e of 5°C than for a higher T_e of 15°C. The IL mass fraction in the absorber is lower at a higher T_e due to the excess H₂O vapor absorption rate at an elevated saturation pressure as seen in Fig. 16(b). At a certain T_a, the COP at a T_g of 90°C becomes higher than that at a T_g of 80°C and such transition in absorption temperature increases with the rise in T_e from 5°C to 15°C. As a case, a T_e of 5°C has the same COP for both T_g (80°C and 90°C) at a T_a of 31° and this increases to 41°C for a higher T_e of 15°C. Such transition in a T_a of above 31°C and 41°C at a high T_g of 90°C reveals a better performance for a T_e of 5°C and 15°C, respectively. This clearly demonstrates that the selection of T_a and T_c with consideration of the IL composition plays a significantly important role in optimizing the performance of the absorption cooling system.

3.3.5 Optimum performance with the utilization of low-grade heat

Efficient utilization and conversion of low-grade heat into cooling based on accessible sink temperature in terms of T_a and T_c is very vital to optimization. Accordingly, in this paper, the T_g of the ARS for the various operating condition is optimized to obtain maximum performances as reported in Table 4. The optimized T_g to generate the maximum COP and maximum ECOP consists of the variation of T_e, T_a, and T_c. In addition, the utilization of low-grade heat below 100°C at a T_g of 70°C, 80°C, and 90°C for each T_e, T_a, and T_c combination are simulated and compared with the optimized T_g.

Based on the variation of T_e from 5°C to 15°C for a constant T_a = T_c of 30°C, an optimum T_g to obtain the maximum COP is noticed to be declined from 83°C to 61°C with the rise in the maximum COP from 0.858 to 0.9. An optimum T_g to achieve the maximum ECOP is reduced from 63°C to 49°C with the rise in T_e from 5°C to 15°C but maximum ECOP values are reduced from 0.5 to 0.39, which clearly indicates that an optimum T_g obtained via the maximum ECOP is lower in comparison to the energy-based COP. Besides, a high T_e utilizes the low grade energy more efficiently to obtain a better COP and ECOP.

Similar simulation to optimize the ARS at T_e = 5°C is extended at a higher T_a = T_c of 40°Cand 50°C wherein the optimum T_g is elevated to 111°C and 153°C with a maximum COP of 0.82 and 0.787 respectively. The temperature difference between the optimized T_g by considering the maximum COP is reduced by 42°C and 70°C with the decline in T_a = T_c by 50°C to 40°C (a drop of 10°C) and 50°C to 30°C (a drop of 20°C) respectively. However, for the same T_a = T_c reduction, the maximum COP of the ARS is enhanced by 0.071 (as T_a = T_c is reduced from 50°C to 30°C) and 0.033 (as T_a = T_c is reduced from 50°C to 40°C). Further, it confirms that the ARS is unviable to produce the cooling effect at a T_e of 5°C to utilize low grade temperature heat below 100°C and 90°C at T_a = T_c of 50°C and 40°C respectively.

Further tabulated data includes the optimization of COP and ECOP for the series flow of cooling water to dissipate heat at different configurations in T_a and T_c with a constant T_e of 5°C. For a constant T_a of 30°C and with the rise in T_c from 30°C to 50°C, the maximum COP is reduced from 0.858 to 0.813 with the escalation of an optimized T_g from 83°C to 108°C. However, at a constant T_c of 30°C and a rise in T_a from 30°C to 50°C, the maximum COP is declined from 0.858 to 0.83 with the rise in an optimized T_g from 83°C to 124°C.

Moreover, the effect of various low generator temperatures of 70°C, 80°C, and 90°C are mentioned in Table 4. At T_g = 90°C, the COP is reduced from 0.857 to 0.785 with the rise in T_c from 30°C to 50°C (at T_a = 30°C) besides the fact that the variation of T_a from 30°C to 50°C (with constant T_c = 30°C) reduces the COP from 0.857 to 0.736. This indicates that at 90°C, the series flow from the absorber to the condenser gives an enhanced performance of the ARS. But based on the maximum COP, the series flow configuration from the condenser to the absorber is the best option in comparison to the absorber to the condenser. This further indicates that a high temperature difference from the optimum T_g declines the ARS performances due to the more supplied heat loss. Such analysis is helpful for the selection of low-grade source temperature of the ARS. Further this analysis indicates that the reduced temperature and pressure of condensation produces better outcomes than the refrigerant vapor absorption rate with a reduced absorption temperature. In summary, it can be concluded that a lower value of T_a and T_c is more beneficial to utilization of low-grade energy more efficiently to produce efficient cooling at a lower evaporation temperature of the ARS.

4 Conclusions

In this paper, the single effect absorption refrigeration system using H₂O-[mmim][DMP] as working pair was proposed, and the performance based on energy and exergy were investigated as a potential alternative to conventional H₂O-LiBr. It can be concluded that, primarily, the thermodynamic equilibrium properties of the proposed H₂O-[mmim][DMP] mixture using the NRTL activity coefficient model are obtained to generate equilibrium Dühring’s (P-T-X_IL) and enthalpy plots (h-T-X_IL) for a [mmim][DMP] mass fraction from 0.1 to 0.9. These equilibrium charts are useful to assess the operating constraints of the ARS based on various temperatures within the generator, the absorber, and the condenser.

The validation of energy based COP of the proposed H₂O-[mmim][DMP] mixture powered ARS is observed to be better with a COP of 0.834 than NH₃-H₂O, H₂O-LiBr, and ILs, i.e., H₂O-[emim][DMP], H₂O-[emim][BF₄] (COP= 0.525) for T_e, T_a, T_c and T_g = 10°C, 30°C, 40°C and 100°C. This COP comparison indicates that the ionic liquid [mmim][DMP] based absorbent could become an alternative to the current commercial absorbent LiBr.

Overall, the effect of energy and exergy-based performance along operating temperatures (T_e, T_g, T_a, and T_c) were studied and optimized for efficient use of low temperature thermal energy. The simulation results of H₂O-[mmim][DMP] mixture show that COP is directly proportional to T_e and T_g and inversely proportional to sink source temperature used as an input to the condenser and the absorber. A T_e of 15°C shows a maximum COP of 0.9 for an optimized T_g of 61°C and for a lower T_a and T_c of 30°C.

For each T_e, an optimum T_g is found to be more for energy-based COP than exergy based ECOP.

An optimum T_g to attain the maximum COP is declined from 83°C to 61°C with the rise in the evaporation temperature from 5°C to 15°C for parallel flow configuration at T_a = T_c = 30°C.

Further cooling water flow (series and parallel) to dissipate heat from the ARS were studied and optimized for the temperature difference between T_a and T_c. It is found that the impact of the reduced temperature within the condenser produces better outcomes than refrigerant vapor absorption with reduced sink temperature. Finally, the presented simulation study indicates that overall, lower values of T_a and T_c are more effective for efficient utilization of low-grade heat toward generation of cooling effect.

References

[1] Fernandes M S, Brites G J V N, Costa J J, Review and future trends of solar adsorption refrigeration systems. Renewable & Sustainable Energy Reviews, 2014, 39: 102–123

[2] Sleiti A K. Tidal power technology review with potential applications in Gulf Stream. Renewable & Sustainable Energy Reviews, 2017, 69: 435–441

[3] Sleiti A K, Al-Ammari W A, Al-Khawaja M. Review of innovative approaches of thermo-mechanical refrigeration systems using low grade heat. International Journal of Energy Research, 2020, 44(13): 9808–9838

[4] Takalkar G. Simulation and experimental study of heat based refrigeration cycles. Dissertation for the Doctoral Degree. India: Institute of Chemical Technology, 2013

[5] Sleiti A K, Al-ammari W A, Al-khawaja M. A novel solar integrated distillation and cooling system – design and analysis. Solar Energy, 2020, 206: 68–83

[6] Sarbu I, Sebarchievici C. Review of solar refrigeration and cooling systems. Energy and Buildings, 2013, 67: 286–297

[7] Shublaq M, Sleiti A K. Experimental analysis of water evaporation losses in cooling towers using filters. Applied Thermal Engineering, 2020, 175: 115418

[8] Sun Y, Di G, Wang J, Gaseous solubility and thermodynamic performance of absorption system using R1234yf/IL working pairs. Applied Thermal Engineering, 2020, 172: 115161

[9] Perez-Astudillo D, Bachour D. DNI, GHI and DHI ground measurements in Doha, Qatar. Energy Procedia, 2014, 49: 2398–2404

[10] Bachour D, Perez-Astudillo D. Ground-measurement GHI map for Qatar. Energy Procedia, 2014, 49: 2297–2302

[11] Papadopoulos A I, Kyriakides A S, Seferlis P, Absorption refrigeration processes with organic working fluid mixtures—a review. Renewable & Sustainable Energy Reviews, 2019, 109: 239–270

[12] Boman D B, Hoysall D C, Staedter M A, A method for comparison of absorption heat pump working pairs. International Journal of Refrigeration, 2017, 77: 149–175

[13] Shiflett M B, Yokozeki A. Solubility and diffusivity of hydrofluorocarbons in room-temperature ionic liquids. AIChE Journal, 2006, 52(3): 1205–1219

[14] Yokozeki A, Shiflett M B. Ammonia solubilities in room temperatures ionic liquids. Industrial & Engineering Chemistry Research, 2007, 46: 1605–1610

[15] Shiflett M B, Yokozcki A. Absorption cycle utilizing ionic liquids and water as working fluids. US patent: US-8715521-BZ, 2005

[16] Mehari A, Xu Z Y, Wang R Z. Thermal energy storage using absorption cycle and system: a comprehensive review. Energy Conversion and Management, 2020, 206: 112482

[17] Parab P, Takalkar G, Bhagwat S. Vapour liquid equilibrium of Potassium formate–water: measurements and correlation by e-NRTL model. Indian Chemical Engineer, 2019, 61(4): 361–373

[18] Wu W, You T, Leung M. Screening of novel water/ionic liquid working fluids for absorption thermal energy storage in cooling systems. International Journal of Energy Research, 2019, 44(12): 9367–9381

[19] Takalkar G D, Bhosale R R, Mali N A, Thermodynamic analysis of EMISE–water as a working pair for absorption refrigeration system. Applied Thermal Engineering, 2019, 148: 787–795

[20] Liu X, Bai L, Liu S, Vapor-liquid equilibrium of R1234yf/[HMIM][Tf₂N] and R1234ze(E)/[HMIM][Tf₂N] working pairs for the absorption refrigeration cycle. Journal of Chemical & Engineering Data, 2016, 61(11): 3952–3957

[21] Wu W, Zhang H, You T, Thermodynamic investigation and comparison of absorption cycles using hydrofluoroolefins and ionic liquid. Industrial & Engineering Chemistry Research, 2017, 56(35): 9906–9916

[22] Kim S, Kohl P A. Theoretical and experimental investigation of an absorption refrigeration system using R134/[bmim][PF₆] working fluid. Industrial & Engineering Chemistry Research, 2013, 52(37): 13459–13465

[23] Shiflett M B, Yokozeki A. Solubility and diffusivity of hydrofluorocarbons in room-temperature ionic liquids. AIChE Journal, 2006, 52(3): 1205–1219

[24] Wang M, Infante Ferreira C A. Absorption heat pump cycles with NH₃ – ionic liquid working pairs. Applied Energy, 2017, 204: 819–830

[25] Chen W, Bai Y. Thermal performance of an absorption-refrigeration system with [emim]Cu₂Cl₅/NH₃ as working fluid. Energy, 2016, 112: 332–341

[26] Takalkar G D, Bhosale R R, Mali N A, Energetic and exergetic performance of NH₃-H₂O-based absorption refrigeration cycle: effect of operating factor. International Journal of Exergy, 2020, 31(4): 352

[27] Zhang B, Chen W, Sun Q, Numerical evaluation of thermal performances of diffusion–absorption refrigeration using 1,3-dimethylimidazolylium dimethylphosphate/methanol/helium as working fluid. Energy Conversion and Management, 2017, 152: 201–213

[28] He Z, Zhao Z, Zhang X, Thermodynamic properties of new heat pump working pairs: 1,3-dimethylimidazolium dimethylphosphate and water, ethanol and methanol. Fluid Phase Equilibria, 2010, 298(1): 83–91

[29] Chen W, Liang S. Thermodynamic analysis of absorption heat transformers using [mmim]DMP/H₂O and [mmim]DMP/CH₃OH as working fluids. Applied Thermal Engineering, 2016, 99: 846–856

[30] Zheng D, Dong L, Huang W, A review of imidazolium ionic liquids research and development towards working pair of absorption cycle. Renewable & Sustainable Energy Reviews, 2014, 37: 47–68

[31] Popp S, Bösmann A, Wölfel R, Screening of ionic liquid/H₂O working pairs for application in low temperature driven sorption heat pump systems. ACS Sustainable Chemistry & Engineering, 2015, 3(4): 750–757

[32] Wang M, Becker T M, Infante Ferreira C A. Assessment of vapor–liquid equilibrium models for ionic liquid based working pairs in absorption cycles. International Journal of Refrigeration, 2018, 87: 10–25

[33] Kim S, Kohl P A. Analysis of [hmim][PF₆] and [hmim][Tf₂N] ionic liquids as absorbents for an absorption refrigeration system. International Journal of Refrigeration, 2014, 48: 105–113

[34] Sujatha I, Venkatarathnam G. Performance of a vapour absorption heat transformer operating with ionic liquids and ammonia. Energy, 2017, 141: 924–936

[35] Chen W, Xu C, Wu H, Energy and exergy analyses of a novel hybrid system consisting of a phosphoric acid fuel cell and a triple-effect compression–absorption refrigerator with [mmim]DMP/CH₃OH as working fluid. Energy, 2020, 195: 116951

[36] Wu W, Leung M, Ding Z, Comparative analysis of conventional and low-GWP refrigerants with ionic liquid used for compression-assisted absorption cooling cycles. Applied Thermal Engineering, 2020, 172: 115145

[37] Dong L, Zheng D, Nie N, Performance prediction of absorption refrigeration cycle based on the measurements of vapor pressure and heat capacity of H₂O+[DMIM]DMP system. Applied Energy, 2012, 98: 326–332

[38] Anand S, Gupta A, Tyagi S K, An absorption chiller system using lithium bromide and water as working fluids: exergy analysis. ASHRAE Transactions, 2014, 120: 226–239

[39] Kim Y J, Gonzalez M. Exergy analysis of an ionic-liquid absorption refrigeration system utilizing waste-heat from datacenters. International Journal of Refrigeration, 2014, 48: 26–37

[40] Ayou D S, Currás M R, Salavera D, Performance analysis of absorption heat transformer cycles using ionic liquids based on imidazolium cation as absorbents with 2,2,2-trifluoroethanol as refrigerant. Energy Conversion and Management, 2014, 84: 512–523

[41] Swarnkar S K, Srinivasa Murthy S, Performance of a vapour absorption refrigeration system operating with ionic liquid-ammonia combination with water as cosolvent. Applied Thermal Engineering, 2014, 72(2): 250–257

[42] Wagner W, Pruß A. The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use. Journal of Physical and Chemical Reference Data, 2002, 31(2): 387

[43] Abumandour E S, Mutelet F, Alonso D. Performance of an absorption heat transformer using new working binary systems composed of {ionic liquid and water}. Applied Thermal Engineering, 2016, 94: 579–589

[44] Kamali M, Parham K, Assadi M. Performance analysis of a single stage absorption heat transformer-based desalination system employing a new working pair of (EMIM) (DMP)/H₂O. International Journal of Energy Research, 2018, 42(15): 4790–4804

[45] Yokozeki A. Theoretical performances of various refrigerant-absorbent pairs in a vapor-absorption refrigeration cycle by the use of equations of state. Applied Energy, 2005, 80(4): 383–399

[46] Shiflett M B, Yokozeki A. Absorption cycle utilizing ionic liquids and water as working fluids. US Patent Application 20070144186, 2006

[47] Takalkar G. Thermodynamic properties and performance evaluation of [EMIM] [DMP]-H₂O working pair for absorption cooling cycle. International Journal of Energy Research, 2019,44(15): 12269–12283

RIGHTS & PERMISSIONS

Higher Education Press