Experimental Study of LiCl/LiBr-Zeolite Composite Adsorbent for Thermochemical Heat Storage

: Adsorption-based thermochemical heat storage is a promising long-term energy storage technology that can be used for seasonal space heating, which has received signiﬁcant amount of efforts on the research and development. In this paper, the heat storage capacity of composite adsorbents made by LiCl + LiBr salt and 3A zeolite was investigated. The basic characteristics of composite material groups were experimentally tested


Introduction
Energy consumption in buildings has steadily increased in recent years. Fossil fuels are increasingly becoming unviable as an energy source due to the rapidly increasing global population and growing demands of urban construction. Instead, a range of ecologically sound renewable energy sources, such as solar, wind, and biomass energy, are increasingly being developed to supply energy demands [1][2][3][4]. This transition to renewable energy is important to slow climate change and reduce the usage of fossil fuel resources. However, at present, there is a mismatch between user demand and the supply of most renewable energy sources. For example, solar energy is an effective alternative to fossil fuels for heating buildings, which is abundant, however it is intermittent and unstable because the solar radiation varies with weather conditions and time of a day [5]. Most of the solar energy supply is in the summer, heating is most needed in the winter when solar energy is less abundant [6]. Therefore, the reasonable utilization of solar energy requires new technologies to realise short to long term heat storage.
Thermal energy storage (TES) can be used to resolve the timing mismatch between solar energy supply and energy demand [7,8]. There is increasing interest in technically advanced and economical TES systems for applications such as hot water supply and important issue to be addressed. Xu et al. [46] developed a numerical model to investigate the thermochemical reaction process in a reactor filled with composite adsorbent and calculated the heat transfer between the composite material and water vapour. Li et al. [47] numerically developed a novel multi-layered sieve reactor focusing on the interaction between the composite material and the air in an open system; in addition, they performed a parametric analysis of the inlet temperature, humidity and airflow rate. Future more, it is expected that the TCHS system will be applied to actual residential heating. A laboratory prototype of a fixed-bed open TCHS reactor developed by Zondag et al. [22] can provide a certain amount of thermal energy for home heating. The system contains 17 dm 3 of adsorbent material, which can generate 150 W of thermal energy and achieve an effective energy storage density of about 0.5 GJ/m 3 . Consider time period and material cost constraints when using energy from TCHS systems for home construction. Researchers can use the TRNSYS program to simulate and analyze transient systems. Safa et al. [48] simulated the heating performance of the heat pump system in the TRNSYS environment and found that the coefficient of performance (COP) under different heat source temperatures was 3.05-3.44 during heating. In TRNSYS-EES, Sakellari et al. [49] analyzed the performance of a heating system based on an exhaust air heat pump using simulation. And several strategies have been proposed to keep comfort within a reasonable range.
The purpose of the present study is to develop suitable LiCl/LiBr-zeolite composite adsorbents for TCHS applications and to perform test of the thermal storage performance in a lab-scale adsorption heat storage reactor. Finally, TRNSYS was used to evaluate the heating effect of this system. The studied pure zeolite and composite adsorbents were subjected to electron microscopy measurement, kinetic adsorption performance and DSC tests to investigate their pore characteristics, water adsorption, heat transfer and mass transfer properties. Based on the identified optimal LiCl/LiBr-zeolite composite adsorbent, a set of adsorption heat storage test rig was constructed. Considering the effects of both air flow rate and air humidity, the water adsorption performance of the developed composite adsorbent under different operating conditions was explored, and the energy storage density and cycle stability of the material were also obtained and analysed. Although the heat storage performance of the open system TCHS has received more attention, the performance of the combination of solar thermal collection and TCHS system heating to meet the space heating needs is still less researched. In this study, a simulation model of the dwelling was developed in Trnsys software. In the winter climate environment, the reactor is charging by solar panels during the day, and discharging by hydration reaction of the adsorbent in the reactor during the night. The heat released is used for space heating of the house to maintain the room temperature in the range of 21 ± 1 • C. When the material reaction completely stops exothermic, start gas as auxiliary energy supplement, continue to heat the house. This paper also simulates a reference system with gas as the heating energy source. The heating effect of the TCHS system with a time period of 7 months was studied. Analyze and calculate gas energy savings to assess the benefits of the system.

Preparation of Composite Adsorbent
Anhydrous LiCl and anhydrous LiBr were respectively purchased from Sinopharm Chemical Reagent Co., Ltd., (Shanghai, China). and Hefei BASF Biotechnology Co., Ltd., (Hefei, China). 3A zeolite was purchased from Clear Spring Technology Co., Ltd., (Singapore). Figure 1 shows the preparation process of the composite adsorbent. First, anhydrous LiCl and anhydrous LiBr solids with pre-defined mass ratio (1:1 to 1:5) were separately added to distilled water and continuously mixed to allow the salt powder to completely dissolve in the water to ensure uniform mixing with the 3A zeolite. To determine the optimal ratio for a composite adsorbent, five salt solution groups with various mass fractions (5 wt%, 10 wt%, 15 wt%, 20 wt% and 25 wt%) were prepared. Considering the reasonably cost effective and easy in process, this study used the water impregnation method to manufacture the composite adsorbent, which can achieve the desired effect of zeolite loading salt. In the first step, zeolite spheres of 3-5 mm size were soaked in deionised water for 12 h for pre-treatment; after soaking, the zeolite was heated and dried at 200 • C for six hours before being colded. The second step involved impregnating the previously treated dry zeolite with salt solution. 30 g of zeolite was soaked in 50 g of salt solution. To ensure that liquid solution was evenly filled in the pores of zeolite, this impregnation process was conducted with continuous stirring using a glass rod. Finally, the salt solution-carrying zeolite was dried again in an oven at 200 • C to form a dehydrated composite adsorbent. The samples were numbered Z0 (pure zeolite) and Z5, Z10, Z15, Z20 and Z25 for the samples using 5-25 wt% salt solution (e.g., Z5 corresponds to the composite adsorbent manufactured using 5 wt% salt solution). The ingredient with various ratios of the manufactured composites are showen in Table 1. Notably, it was found during the preparation that salt concentrations above 25 wt% were not easily made into composite adsorbents. The structure of zeolite particles is destroyed during the impregnation process, and some of the zeolite particles will be completely broken into powder form. Their performance was tested to be inferior to that of the intact sample set, so the maximum concentration of salt solution used in this experiment was 25 wt%.
cost effective and easy in process, this study used the water impregnation method to manufacture the composite adsorbent, which can achieve the desired effect of zeolite loading salt. In the first step, zeolite spheres of 3-5 mm size were soaked in deionised water for 12 h for pre-treatment; after soaking, the zeolite was heated and dried at 200 °C for six hours before being colded. The second step involved impregnating the previously treated dry zeolite with salt solution. 30 g of zeolite was soaked in 50 g of salt solution. To ensure that liquid solution was evenly filled in the pores of zeolite, this impregnation process was conducted with continuous stirring using a glass rod. Finally, the salt solution-carrying zeolite was dried again in an oven at 200 °C to form a dehydrated composite adsorbent. The samples were numbered Z0 (pure zeolite) and Z5, Z10, Z15, Z20 and Z25 for the samples using 5-25 wt% salt solution (e.g., Z5 corresponds to the composite adsorbent manufactured using 5 wt% salt solution). The ingredient with various ratios of the manufactured composites are showen in Table 1. Notably, it was found during the preparation that salt concentrations above 25 wt% were not easily made into composite adsorbents. The structure of zeolite particles is destroyed during the impregnation process, and some of the zeolite particles will be completely broken into powder form. Their performance was tested to be inferior to that of the intact sample set, so the maximum concentration of salt solution used in this experiment was 25 wt%.

Morphologies
The samples were imaged using a Sigma 300 model scanning electron microscope (SEM). Figure 2 shows the microscopic surface morphology of the samples at a magnification of 20,000×. As shown in the Figure 2a-f, the pure zeolite (sample Z0) exhibits clear pores, while the remaining samples demonstrate that the salt and 3A zeolite are well mixed, with salt tending to collect in the uneven areas of the zeolite surface. As the salt  The samples were imaged using a Sigma 300 model scanning electron microscope (SEM). Figure 2 shows the microscopic surface morphology of the samples at a magnification of 20,000×. As shown in the Figure 2a-f, the pure zeolite (sample Z0) exhibits clear pores, while the remaining samples demonstrate that the salt and 3A zeolite are well mixed, with salt tending to collect in the uneven areas of the zeolite surface. As the salt content of the composite increases, they gradually adhere to the surface, thereby increasing the surface roughness. The surface of the adsorbed sample displayed clear salt crystallisation, with increased crystallisation observed with increasing salt concentration. Such a phenomenon is consistent with previous descriptions of the microstructure of adsorbents loaded with salts. The outer surface of sample Z25 was almost completely covered with salt crystals, however, the channels leading from the surface of the zeolite to its core were unobstructed. More salt loading can increase the water adsorption capacity because salt embedded in the sample creates numerous micropores; this process significantly increases the surface area of the composite material and the extent of its contact area with water vapour. However, high-concentration salt solutions are also susceptible to crystallisation, which may limit the material's mass transfer capacity. It is therefore essential to identify the optimum concentration of salt solution used for manufacturing the composite adsorption material.
ing the surface roughness. The surface of the adsorbed sample displayed clear salt crystallisation, with increased crystallisation observed with increasing salt concentration. Such a phenomenon is consistent with previous descriptions of the microstructure of adsorbents loaded with salts. The outer surface of sample Z25 was almost completely covered with salt crystals, however, the channels leading from the surface of the zeolite to its core were unobstructed. More salt loading can increase the water adsorption capacity because salt embedded in the sample creates numerous micropores; this process significantly increases the surface area of the composite material and the extent of its contact area with water vapour. However, high-concentration salt solutions are also susceptible to crystallisation, which may limit the material's mass transfer capacity. It is therefore essential to identify the optimum concentration of salt solution used for manufacturing the composite adsorption material.

Pore Properties
The pore volume and pore size distribution of the materials were examined by a fully automatic surface and pore size analyzer (BET, Micromeritics APSP2460) using nitrogen adsorption-desorption method. The pore size distributions and specific surface areas of porous materials were calculated using the Barrett-Joyner-Halenda (BJH) theory and the Brunauer-Emmet-Teller (BET) theory. The porous properties of the samples, such as pore size distribution, pore volume and surface area, are listed in Table 2. Figure 3a,b show the pore volume versus different pore widths. A sufficiently large pore volume can maintain the passage of water vapor in and out of the adsorption process. Otherwise, the mass transfer of water adsorption kinetics would be limited. All samples showed a similar trend, increasing rapidly at the beginning and stabilizing at the end. They all had a peak in pore volume at about 5-7 nm pore width, which means that most of the pore sizes were distributed in this range. Moreover, the pore volume or average pore size may decrease as the concentration of the solution increases. Although a high salt concentration theoretically provides more reactants for water adsorption, it also limits water adsorption by sacrificing some mesoporous mass transport channels in the support structure. This suggests that the balance between LiCl/LiBr loading and pore volume in the composite should be carefully maintained to accomplish an optimal thermochemical thermal storage material.

Pore Properties
The pore volume and pore size distribution of the materials were examined by a fully automatic surface and pore size analyzer (BET, Micromeritics APSP2460) using nitrogen adsorption-desorption method. The pore size distributions and specific surface areas of porous materials were calculated using the Barrett-Joyner-Halenda (BJH) theory and the Brunauer-Emmet-Teller (BET) theory. The porous properties of the samples, such as pore size distribution, pore volume and surface area, are listed in Table 2. Figure 3a,b show the pore volume versus different pore widths. A sufficiently large pore volume can maintain the passage of water vapor in and out of the adsorption process. Otherwise, the mass transfer of water adsorption kinetics would be limited. All samples showed a similar trend, increasing rapidly at the beginning and stabilizing at the end. They all had a peak in pore volume at about 5-7 nm pore width, which means that most of the pore sizes were distributed in this range. Moreover, the pore volume or average pore size may decrease as the concentration of the solution increases. Although a high salt concentration theoretically provides more reactants for water adsorption, it also limits water adsorption by sacrificing some mesoporous mass transport channels in the support structure. This suggests that the balance between LiCl/LiBr loading and pore volume in the composite should be carefully maintained to accomplish an optimal thermochemical thermal storage material.

Adsorption Kinetic Performance
The adsorption rate and capacity of water vapour have significantly impact on the energy storage capacity of LiCl-LiBr-zeolite adsorption systems. The adsorption tests were conducted in a constant temperature and humidity chamber (climatic chamber) under a temperature of 25 °C and relative humidity (RH) levels of 60%, 70%, 80%, and 90% to analyse the adsorption kinetics. The mass change of the adsorbent samples over 24 h was measured by using an electronic balance. The water uptake kinetics of samples are depicted in Figure 4. When the weight difference is below 5% in two consecutive 60 min intervals, adsorption equilibrium was deemed to have been achieved.

Adsorption Kinetic Performance
The adsorption rate and capacity of water vapour have significantly impact on the energy storage capacity of LiCl-LiBr-zeolite adsorption systems. The adsorption tests were conducted in a constant temperature and humidity chamber (climatic chamber) under a temperature of 25 • C and relative humidity (RH) levels of 60%, 70%, 80%, and 90% to analyse the adsorption kinetics. The mass change of the adsorbent samples over 24 h was measured by using an electronic balance. The water uptake kinetics of samples are depicted in Figure 4. When the weight difference is below 5% in two consecutive 60 min intervals, adsorption equilibrium was deemed to have been achieved.  The pure zeolite, Z0, exhibits high adsorption rate at the early adsorption stage, reaching a maximum value of 0.18 g/g after 600 min of adsorption at 25 °C and 80% RH, as shown in Figure 4. Comparing to Z0, Z5 performs more poorly as an adsorbent with lower adsorption rate and capacity, which is due to its low salt content and partially blocked water transfer channels. Higher salt content adsorbents take longer to achieve equilibrium, suggesting that the water transfer channels were blocked more significantly The pure zeolite, Z0, exhibits high adsorption rate at the early adsorption stage, reaching a maximum value of 0.18 g/g after 600 min of adsorption at 25 • C and 80% RH, as shown in Figure 4. Comparing to Z0, Z5 performs more poorly as an adsorbent with lower adsorption rate and capacity, which is due to its low salt content and partially blocked water transfer channels. Higher salt content adsorbents take longer to achieve equilibrium, suggesting that the water transfer channels were blocked more significantly with increasing salt content; however, the adsorption rate are larger at late stage of adsorption and the final adsorption capacity is also promoted. In contrast, samples Z10, Z15 and Z25 demonstrate reasonable adsorption rates and capacities. However, the salt swells upon hydration, leading to the blockage of the zeolite pore channels, especially in Z20 as shown in BET result ( Table 2). Both the active adsorption area (specific surface area) and the mass transfer channel (pore volume) were reduced. This self-clogging effect is detrimental to the water adsorption behavior of the material. Thus, Z20 exhibited a refrained adsorption rate than other candidates until 700 min due to the corresponding significant mass transfer barrier; thereafter due to the enhanced water adsorption capacity, i.e., increased salt loading, the overall water adsorption capacity is higher than Z0 and Z5. Finally, the adsorption capacity of Z10, Z15 and Z25 are 0.22 g/g, 0.24 g/g and 0.27 g/g, respectively. Z15 always maintains a high adsorption rate and has a high adsorption capacity when the adsorption equilibrium is reached.
The RH significantly impacts the equilibrium adsorption capacity of every sample. Figure 5 shows the adsorption capacity of the samples after reaching equilibrium at 40%, 60% and 80% RH. The overall pattern indicates that high RH encourages wate vapour adsorption, thus increasing the adsorption capacity of all samples. This occurs because the increased RH elevates the water vapour pressure difference between the water vapour in the air and the equilibrium water vapour pressure of the adsorbent material, which facilitates the effective diffusion of water vapour into the adsorbent. Another significant factor is that salt deliquescence in a high-humidity environment reduces zeolite micropore blockage, which is advantageous for increasing water adsorption. Nevertheless, the swelling of salt hydration occurs at the same relative humidity conditions, which may lead to the clogging of zeolite pores. The blocking extent referring to the solution concentration was partly revealed in the BET result, which has been discussed in last section. swelling of salt hydration occurs at the same relative humidity conditions, which may lead to the clogging of zeolite pores. The blocking extent referring to the solution concentration was partly revealed in the BET result, which has been discussed in last section.

Adsorption Heat
Differential scanning calorimetry (DSC) were performed on Z0 to Z25. Using a simultaneous thermal analyser, with N2 selected as the test atmosphere. The experimental temperature conditions were set to increase from 30 °C to 300 °C, the heating rate was 5 °C/min, and the experimental time was 1 h. The heat flow was recorded throughout the process.
The DSC curves of the endothermic dehydration process of each group of samples

Adsorption Heat
Differential scanning calorimetry (DSC) were performed on Z0 to Z25. Using a simultaneous thermal analyser, with N 2 selected as the test atmosphere. The experimental temperature conditions were set to increase from 30 • C to 300 • C, the heating rate was 5 • C/min, and the experimental time was 1 h. The heat flow was recorded throughout the process.
The DSC curves of the endothermic dehydration process of each group of samples are shown in Figure 6a. The theoretical heat storage density of the composite heat storage material is obtained by integrating the curve inland in a certain range, and plotted in Figure 6b. Sample Z0 exhibites a heat storage density of 447.1 J/g; however, in all instances, the heat storage density of the composites are higher than that of the pure zeolite. This effect is due to the contribution of salt to the water absorption capacity, thereby increasing its thermal storage density. The most significant thermal storage density is recorded in Z5 (32.6% higher than Z0). Although no positive correlation was identified between the heat storage density of the composite adsorbent and the salt solution concentration, a higher solution concentration may nonetheless hinder mass transfer during water adsorption, thus resulting in a lower heat storage density. Therefore, samples Z5, Z10 and Z15 can be used as the three 'ideal' samples for the analyses in this study because their heat storage densities do not markedly differ.

Selection of the Optimal Adsorbent
Based on the analyses above, the adsorption performance of composite materials are found to be variably impacted by the salt concentration with no clear linear relationship. As confirmed by the SEM experiments, changing the salt content influenced the microscopic scale size characteristics of the composite adsorbent. The investigation of the adsorption kinetics of the samples revealed that Z15 had both a high water adsorption rate and capacity. Further DSC studies show the heat storage density of Z15 is up to 585.3 J/g, which is 30.9% higher compared to Z0. On this basis, sample Z15 was selected as the optimal composite adsorbent for lab-scale open system experiment. Figure 7a shows a schematic diagram of a reaction system. Figure 7b is the photo of the thermochemical heat storage reaction system, which comprises a pipe with a valve, a fan, a humidifier and a reactor. The reactor is placed inside the hollow cylindrical pipe that houses the entire apparatus. The rightmost side of the pipe is connected to an inclined flow booster pipe fan. In addition, the humidifier is connected to the inlet air duct. The layout of the reactor is shown in Figure 7c. The wire mesh container containing the composite adsorbent Z15 is covered with a metal disc-shaped covering. To achieve a more thorough and uniform adsorbent reaction, the reactor is fitted with a thin tube with perforations in the centre shaft. In total, 3 kg of Z15 adsorbent was placed inside the reactor. During the experiment, the pipes were covered in aluminium foil insulation foam to min-

Selection of the Optimal Adsorbent
Based on the analyses above, the adsorption performance of composite materials are found to be variably impacted by the salt concentration with no clear linear relationship. As confirmed by the SEM experiments, changing the salt content influenced the microscopic scale size characteristics of the composite adsorbent. The investigation of the adsorption kinetics of the samples revealed that Z15 had both a high water adsorption rate and capacity. Further DSC studies show the heat storage density of Z15 is up to 585.3 J/g, which is 30.9% higher compared to Z0. On this basis, sample Z15 was selected as the optimal composite adsorbent for lab-scale open system experiment. Figure 7a shows a schematic diagram of a reaction system. Figure 7b is the photo of the thermochemical heat storage reaction system, which comprises a pipe with a valve, a fan, a humidifier and a reactor. The reactor is placed inside the hollow cylindrical pipe that houses the entire apparatus. The rightmost side of the pipe is connected to an inclined flow booster pipe fan. In addition, the humidifier is connected to the inlet air duct. The layout of the reactor is shown in Figure 7c. The wire mesh container containing the composite adsorbent Z15 is covered with a metal disc-shaped covering. To achieve a more thorough and uniform adsorbent reaction, the reactor is fitted with a thin tube with perforations in the centre shaft. In total, 3 kg of Z15 adsorbent was placed inside the reactor. During the experiment, the pipes were covered in aluminium foil insulation foam to minimize heat loss to the exterior. Temperature and humidity sensor (A-Pt100) were installed at the inlet and outlet of the reactor to measure the temperature and humidity of the air. To assess the volume of air passing through the pipe, a gas flow meter was installed at the end of the pipe.

Experimental Procedures
In this study, the air flow rate and relative humidity were chosen as the main variables for investigation; thus, to examine the impacts of both on the thermal storage performance of the composite salt, orthogonal experiments were configured with airflow rates of 5 m 3 /h, 10 m 3 /h, 15 m 3 /h and 20 m 3 /h and humidity levels of 60%, 70%, 80% and 90% RH.
The 3 kg of LiCl/LiBr-zeolite composite thermal storage material was placed in the reactor, and the fan and humidifier were adjusted to ensure that the experimental conditions were within the predetermined ranges. Overall, the adsorption reaction procedure for the composite material takes 12 to 20 h to complete. In the initial stage of the reaction, the outlet temperature rises at a certain rate and then begins to fall after reaching its peak. The temperature curve subsequently tends to slowly level off and return to the inlet temperature, indicating the reaction's completion. The adsorption material in the reactor was heated in an oven following each experiment, totally dehydrated, and then cooled in preparation for the following experiment.

Energy Density Calculate
For the thermochemical heat storage system, the energy density is a crucial indicator to evaluate the performance of the composite salt heat storage, which can be calculated by the following equation. Since the humidity data measured by the temperature and humidity sensor is 0-100% RH, the water vapor partial pressure Pv in the wet air can be obtained by the following Equation (1) [50]: Ps is the saturated water vapor partial pressure [50], and the absolute humidity d of wet air can be expressed as:

Experimental Procedures
In this study, the air flow rate and relative humidity were chosen as the main variables for investigation; thus, to examine the impacts of both on the thermal storage performance of the composite salt, orthogonal experiments were configured with airflow rates of 5 m 3 /h, 10 m 3 /h, 15 m 3 /h and 20 m 3 /h and humidity levels of 60%, 70%, 80% and 90% RH.
The 3 kg of LiCl/LiBr-zeolite composite thermal storage material was placed in the reactor, and the fan and humidifier were adjusted to ensure that the experimental conditions were within the predetermined ranges. Overall, the adsorption reaction procedure for the composite material takes 12 to 20 h to complete. In the initial stage of the reaction, the outlet temperature rises at a certain rate and then begins to fall after reaching its peak. The temperature curve subsequently tends to slowly level off and return to the inlet temperature, indicating the reaction's completion. The adsorption material in the reactor was heated in an oven following each experiment, totally dehydrated, and then cooled in preparation for the following experiment.

Energy Density Calculate
For the thermochemical heat storage system, the energy density is a crucial indicator to evaluate the performance of the composite salt heat storage, which can be calculated by the following equation. Since the humidity data measured by the temperature and humidity sensor is 0-100% RH, the water vapor partial pressure P v in the wet air can be obtained by the following Equation (1) [50]: P s is the saturated water vapor partial pressure [50], and the absolute humidity d of wet air can be expressed as: The enthalpy H of wet air can be calculated from Equation (3) [39]: The energy storage density Q E can be calculated by Equation (4) [51]: ρ ƒ is the density of the dry air, q v is the volume flow rate of the gas, M is the total mass of the adsorbent material in the reactor, this experiment uses M = 3 kg.
The energy discharge efficiency of this reactor can be estimated using Equation (5) [39]: Q D is the theoretical heat storage density of the composite adsorbent, which is derived from the DSC experiment in Section 2.2.3.

Effect of Air Flow Rate
For each of the four relative humidity levels, Figure 8 illustrates the impact of various air flow rates on the exit air temperature. All the temperature differential variations in Figure 8 initially climb rapidly and then slowly decline until the temperature rise becomes close to zero. The final temperature increase reduces to zero at the point of which the adsorbent is saturated, the stored heat is completely released and the reaction is over. The rate of temperature rise increases with increasing air flow rate, with the recorded temperature rise rate at a flow rate of 5 m 3 /h noticeably lower than the other measured flow rates. Since the air flow rate affects the rate of moisture transfer and the low air flow rate limits the process of mass and heat transfer between the wet air and the adsorbent [52]. However, the effective reaction time of the material becomes longer at low flow rate conditions. The reaction time decreases consistently with increasing flow rate under conditions of 60% RH. The adsorbent still tends to have a shorter reaction time at high flow rates when the air has an RH of 70% or below, as shown in Figure 8a,b, however, as the relative humidity rises, this difference becomes less noticeable at flow rates of 10 m 3 /h, 15 m 3 /h, and 20 m 3 /h, as shown in Figure 8c,d.
ρƒ is the density of the dry air, qv is the volume flow rate of the gas, M is the total mas of the adsorbent material in the reactor, this experiment uses M = 3 kg.
The energy discharge efficiency of this reactor can be estimated using Equation (5 [39]: QD is the theoretical heat storage density of the composite adsorbent, which is de rived from the DSC experiment in Section 2.2.3.

Effect of Air Flow Rate
For each of the four relative humidity levels, Figure 8 illustrates the impact of variou air flow rates on the exit air temperature. All the temperature differential variations in Figure 8 initially climb rapidly and then slowly decline until the temperature rise become close to zero. The final temperature increase reduces to zero at the point of which th adsorbent is saturated, the stored heat is completely released and the reaction is over. Th rate of temperature rise increases with increasing air flow rate, with the recorded temper ature rise rate at a flow rate of 5 m 3 /h noticeably lower than the other measured flow rates Since the air flow rate affects the rate of moisture transfer and the low air flow rate limit the process of mass and heat transfer between the wet air and the adsorbent [52]. How ever, the effective reaction time of the material becomes longer at low flow rate conditions The reaction time decreases consistently with increasing flow rate under conditions o 60% RH. The adsorbent still tends to have a shorter reaction time at high flow rates when the air has an RH of 70% or below, as shown in Figure 8a,b, however, as the relative hu midity rises, this difference becomes less noticeable at flow rates of 10 m 3 /h, 15 m 3 /h, and 20 m 3 /h, as shown in Figure 8c

Effect of Air Relative Humidity
As the results from Figure 9, the maximum temperature difference increases as th relative humidity increases. The air temperature increased by 84.1%, 28.1%, 93.3%, an 79.8%, respectively, when the relative humidity was elevated from 60% to 90%. Similarly when more humid air flowed through the reaction pipeline, the rate of temperature ris of the air at the initial stage was higher. This behaviour was especially noticeable at a flow rate of 5 m 3 /h. It is because that the high relative humidity provides higher pressur difference to drive the adsorption reaction, which accelerates the hydration reaction of the material driving the heat generation. This causes the increase in temperatur rise value and rate.

Effect of Air Relative Humidity
As the results from Figure 9, the maximum temperature difference increases as the relative humidity increases. The air temperature increased by 84.1%, 28.1%, 93.3%, and 79.8%, respectively, when the relative humidity was elevated from 60% to 90%. Similarly, when more humid air flowed through the reaction pipeline, the rate of temperature rise of the air at the initial stage was higher. This behaviour was especially noticeable at a flow rate of 5 m 3 /h. It is because that the high relative humidity provides higher pressure difference to drive the adsorption reaction, which accelerates the hydration reaction of the material driving the heat generation. This causes the increase in temperature rise value and rate.

Effect of Air Relative Humidity
As the results from Figure 9, the maximum temperature difference increases as the relative humidity increases. The air temperature increased by 84.1%, 28.1%, 93.3%, and 79.8%, respectively, when the relative humidity was elevated from 60% to 90%. Similarly, when more humid air flowed through the reaction pipeline, the rate of temperature rise of the air at the initial stage was higher. This behaviour was especially noticeable at a flow rate of 5 m 3 /h. It is because that the high relative humidity provides higher pressure difference to drive the adsorption reaction, which accelerates the hydration reaction of the material driving the heat generation. This causes the increase in temperature rise value and rate.

Energy Storage Density
For each group of experiments, the energy storage density was calculated and plotted in Figure 10. It can be seen from the experimental results that the maximum temperature difference and the temperature rise rate have a positive correlation trend with the gas flow

Energy Storage Density
For each group of experiments, the energy storage density was calculated and plotted in Figure 10. It can be seen from the experimental results that the maximum temperature difference and the temperature rise rate have a positive correlation trend with the gas flow rate and relative humidity. However, energy storage density does not have such a law. From the formula, it shows that the energy storage density is not only related to the maximum temperature difference, but also to the effective reaction time. In addition, it cannot be ignored that in the same room temperature environment, the heat loss under different experimental conditions is different. The higher the flow rate, the more heat is lost through the pipe, metal reactor. In this experiment, the composite material has the highest heat release value of 434.4 J/g when the gas flow rate is 15 m 3 /h and the relative humidity is 70%, and its energy discharge efficiency is 74.3%.

Energy Storage Density
For each group of experiments, the energy storage density was calculated and plotted in Figure 10. It can be seen from the experimental results that the maximum temperature difference and the temperature rise rate have a positive correlation trend with the gas flow rate and relative humidity. However, energy storage density does not have such a law. From the formula, it shows that the energy storage density is not only related to the maximum temperature difference, but also to the effective reaction time. In addition, it cannot be ignored that in the same room temperature environment, the heat loss under different experimental conditions is different. The higher the flow rate, the more heat is lost through the pipe, metal reactor. In this experiment, the composite material has the highest heat release value of 434.4 J/g when the gas flow rate is 15 m 3 /h and the relative humidity is 70%, and its energy discharge efficiency is 74.3%.

Cyclic Experiments
To get the cyclic performance, the adsorbent material was subjected to ten cycles of adsorption/desorption under experimental conditions with an air flow rate of 15 m 3 /h and a relative humidity of 70%. Figure 11 shows the variation of the outlet temperature with time. It can be seen that the rate of temperature increase decreases with the number of cycles. Figure 12 shows the variation of the energy storage density for the ten cycling experiments. The average error of the cycling result was 0.41%, which proved the experimental result to be convincing. Although the adsorption heat decreased significantly after the first three trials (by 0.53%), the decrease slowed down in the subsequent studies, with a final rate of change of only 0.17%. This indicates that while certain microstructure changes do occur in the composite adsorption material during cycling, these changes become smaller and smaller as the number of cycles increases. Cycling experiments show that this material can be recycled for a long time in cross-season heat storage.
periments. The average error of the cycling result was 0.41%, which proved the experimental result to be convincing. Although the adsorption heat decreased significantly after the first three trials (by 0.53%), the decrease slowed down in the subsequent studies, with a final rate of change of only 0.17%. This indicates that while certain microstructure changes do occur in the composite adsorption material during cycling, these changes become smaller and smaller as the number of cycles increases. Cycling experiments show that this material can be recycled for a long time in cross-season heat storage.

System Description
In this study, the TRNSYS software was used to dynamically simulate the charging mode and the discharging mode. The system mainly consists of three parts. The first part is solar energy collection and energy storage; the second part is the reactor heat release as well as gas heating; the last part is the residential model as the heated unit. Figure 13 shows the charge mode and discharge mode for the entire operation. During the day, the sorbent is heated and dehydrated by a solar-triggered charging mode. All heat is stored, and there is no heat loss in this process (the heat conversion efficiency is 74.3%). The input heat from the solar panels regenerated the adsorbent in the reactor. At night, the discharge mode is triggered by blowing air at a certain flow rate and humidity into the reactor. All the heat of adsorption released by the reactor heats the air, which is mixed with ambient air and passed into the dwelling to heat the space. The specific content of the heat release analysis module is described in Section 3. The optimal heat extraction efficiency based on experimental data converts the energy stored during the day. In addition, gas is used as an auxiliary energy to supplement the insufficient power required by the house. Consid-

System Description
In this study, the TRNSYS software was used to dynamically simulate the charging mode and the discharging mode. The system mainly consists of three parts. The first part is solar energy collection and energy storage; the second part is the reactor heat release as well as gas heating; the last part is the residential model as the heated unit. Figure 13 shows the charge mode and discharge mode for the entire operation. During the day, the sorbent is heated and dehydrated by a solar-triggered charging mode. All heat is stored, and there is no heat loss in this process (the heat conversion efficiency is 74.3%). The input heat from the solar panels regenerated the adsorbent in the reactor. At night, the discharge mode is triggered by blowing air at a certain flow rate and humidity into the reactor. All the heat of adsorption released by the reactor heats the air, which is mixed with ambient air and passed into the dwelling to heat the space. The specific content of the heat release analysis module is described in Section 3. The optimal heat extraction efficiency based on experimental data converts the energy stored during the day. In addition, gas is used as an auxiliary energy to supplement the insufficient power required by the house. Considering real life, assume that a simple thermostatic controller is used to maintain the indoor temperature at 22 ± 1 • C. When the indoor average area temperature is lower than 21 • C, the fan starts to work, and the reactor provides heat; when it is higher than 23 • C, the fan stops running. In addition, when the reaction of the adsorbent in the reactor is no longer enough to supply the indoor temperature to 20 • C, no heat is provided, and the gas is started for heating work.
shows the charge mode and discharge mode for the entire operation. During the day, the sorbent is heated and dehydrated by a solar-triggered charging mode. All heat is stored, and there is no heat loss in this process (the heat conversion efficiency is 74.3%). The input heat from the solar panels regenerated the adsorbent in the reactor. At night, the discharge mode is triggered by blowing air at a certain flow rate and humidity into the reactor. All the heat of adsorption released by the reactor heats the air, which is mixed with ambient air and passed into the dwelling to heat the space. The specific content of the heat release analysis module is described in Section 3. The optimal heat extraction efficiency based on experimental data converts the energy stored during the day. In addition, gas is used as an auxiliary energy to supplement the insufficient power required by the house. Considering real life, assume that a simple thermostatic controller is used to maintain the indoor temperature at 22 ± 1 °C. When the indoor average area temperature is lower than 21 °C, the fan starts to work, and the reactor provides heat; when it is higher than 23 °C, the fan stops running. In addition, when the reaction of the adsorbent in the reactor is no longer enough to supply the indoor temperature to 20 °C, no heat is provided, and the gas is started for heating work.  To realize the residential model of the study, the 3D architectural design software SketchUp was used. A typical two-story residential model (actual size is 6150 mm × 8350 mm × 4550 mm). As shown in Figure 14, the house model is then imported into TRNSYS through the TRNSYS 3d plugin. Next, transfer the module in TRNSYS Simulation Studio to build the TCHS-building simulation platform. Figure 15 shows the dynamic simulation flow, with controllers for each zone. Table 3 gives the structural information of the building used to determine the parameters of the walls, roof, floor and windows. The TRNSYS program generally creates a flexible environment for modelling buildings, where different building descriptions can be obtained by changing design parameters. However, this study mainly focuses on the performance research of the TCHS system. It explores the degree of its effect applied to the building, so that the parameter setting of the building itself is not changed. The system lays half-roof synchronous monocrystalline silicon solar modules and uses Type103 modules to simulate the photovoltaic power generation process. The conversion efficiency of this component can reach 21.3% under standard test condition. The climatic conditions entered refer to the climatic data of a typical weather year in Newcastle, including ambient temperature, instantaneous solar radiation on the horizontal plane, wind speed, and ambient air relative humidity. In the control module of the software, temperature control, time control and heat control are adopted. At the same time, the indoor temperature is within the range of 22 ± 1 • C, and the TCHS heating working time is from 20:00 at night to 9:00 in the morning of the next day. During this time period, when the room needs heating and the energy storage of the TCHS system is not exhausted, the system heats the room. The hot air leading to the room is distributed according to the area ratio of each room. In the analysis process of building energy consumption, the Type56 analysis module is used. The terminal calls the multi-region building model TRNBUILD to simulate the thermal process of the room. By simulating, the domestic temperature, the data of photoelectric conversion and energy consumption are calculated. The above data are monitored by the Type65 online plotter and monthly statistical output by Type46. The simulation time step is 0.05 h, and the simulation period is from October to April of the following year when heating is required.
cording to the area ratio of each room. In the analysis process of building energy consumption, the Type56 analysis module is used. The terminal calls the multi-region building model TRNBUILD to simulate the thermal process of the room. By simulating, the domestic temperature, the data of photoelectric conversion and energy consumption are calculated. The above data are monitored by the Type65 online plotter and monthly statistical output by Type46. The simulation time step is 0.05 h, and the simulation period is from October to April of the following year when heating is required.

Evaluation Indicators
Compared with the energy consumption Q 1 of the traditional (gas) system, the energy consumption Q 2 achieved by the retrofitted system (TCHS and gas) is calculated according to the following Equation (6) [53].
The heating of TCHS requires raw power from fans and humidifiers. Q hum and Q fan represent the power consumption of the humidifier and fan, respectively, and Q gas is the gas consumption. Among them, 2.42 is the energy conversion factor of electricity equivalent to primary energy. For TCHS systems, the average heating situation is expressed using the coefficient of performance COP, which is defined as [54]: Among them, Q sor provides heat for the TCHS system. It can be calculated from the heat capacity obtained by air flowing through the adsorbent material.

Analysis of Residential Heating Performance Based on TCHS System
The dynamic simulation of the charging and discharging process of the established system is carried out for a period of 7 months. The results show the operating state and performance of the system under various climatic conditions. Figure 16 shows the changes in ambient temperature and room temperature during the simulation cycle. Although there is temperature control to set the temperature in the range of 21 ± 1 • C, it can be seen that the overall trend is still affected by the ambient temperature.
The dynamic simulation of the charging and discharging process of the established system is carried out for a period of 7 months. The results show the operating state and performance of the system under various climatic conditions. Figure 16 shows the changes in ambient temperature and room temperature during the simulation cycle. Although there is temperature control to set the temperature in the range of 21 ± 1°C, it can be seen that the overall trend is still affected by the ambient temperature.  Figure 17 shows the solar power generation, the heat of desorption Qdes of the adsorbent and the heat of adsorption Qsor of the adsorption system for each month in charge mode and discharge mode. As the climate temperature rises, the better the solar heat collection effect is, the more energy can be provided to the adsorption heat storage system, and the better the effect of the adsorption heat supply. The conversion efficiency between the two follows the experimental results in Section 4.3, which is close to the optimal energy conversion efficiency of 74.3%.  Figure 17 shows the solar power generation, the heat of desorption Q des of the adsorbent and the heat of adsorption Q sor of the adsorption system for each month in charge mode and discharge mode. As the climate temperature rises, the better the solar heat collection effect is, the more energy can be provided to the adsorption heat storage system, and the better the effect of the adsorption heat supply. The conversion efficiency between the two follows the experimental results in Section 4.3, which is close to the optimal energy conversion efficiency of 74.3%. To evaluate the energy consumption reduction of the heating system by the heat release of TCHS. Figure 18 plots the energy consumption Q1 of gas heating alone, and the energy consumption Q2 of the adsorption system and gas co-heating. It can be seen from the figure that the TCHS system can reduce the consumption of gas, and the total energy consumption in seven months can reduce the gas consumption by 8.8%. However, natural gas still accounts for most energy consumption, as TCHS is limited by solar energy harvesting. After the TCHS releases all the heat of adsorption, the fans and humidifiers stop working. If the ability to increase the TCHS reserve solar energy is improved, the heat To evaluate the energy consumption reduction of the heating system by the heat release of TCHS. Figure 18 plots the energy consumption Q 1 of gas heating alone, and the energy consumption Q 2 of the adsorption system and gas co-heating. It can be seen from the figure that the TCHS system can reduce the consumption of gas, and the total energy consumption in seven months can reduce the gas consumption by 8.8%. However, natural gas still accounts for most energy consumption, as TCHS is limited by solar energy harvesting. After the TCHS releases all the heat of adsorption, the fans and humidifiers stop working. If the ability to increase the TCHS reserve solar energy is improved, the heat release from TCHS can be increased. The power consumption of fans and humidifiers accounts for the central part, which is expected to save most of the gas. To evaluate the energy consumption reduction of the heating system by the heat release of TCHS. Figure 18 plots the energy consumption Q1 of gas heating alone, and the energy consumption Q2 of the adsorption system and gas co-heating. It can be seen from the figure that the TCHS system can reduce the consumption of gas, and the total energy consumption in seven months can reduce the gas consumption by 8.8%. However, natural gas still accounts for most energy consumption, as TCHS is limited by solar energy harvesting. After the TCHS releases all the heat of adsorption, the fans and humidifiers stop working. If the ability to increase the TCHS reserve solar energy is improved, the heat release from TCHS can be increased. The power consumption of fans and humidifiers accounts for the central part, which is expected to save most of the gas.
Energy consumption (kWh)   Figure 19 plots the COP values for TCHS for each month. The maximum was in December at 6.67. This is because the TCHS system relies on the hydration reaction of the adsorbent, which is affected by the climatic environment. Because the air humidity in Newcastle in December is relatively high, it can provide natural moisture for the reaction of the TCHS system, reducing the burden of electricity consumption of the humidifier. Similarly, the minimum value of 4.48 appeared in April. Although solar energy was abundant in April, the charging capacity reached the maximum ( Figure 17); but the natural air was dry, which significantly increased the power consumption of the humidifier ( Figure 18). The average monthly COP was calculated to be 5.56. If the investment in the TCHS system is increased, it is likely to replace most of the gas, and the heating demand of the house can be met with less power consumption.
Buildings 2022, 12, x FOR PEER REVIEW 20 of 23 Figure 19 plots the COP values for TCHS for each month. The maximum was in December at 6.67. This is because the TCHS system relies on the hydration reaction of the adsorbent, which is affected by the climatic environment. Because the air humidity in Newcastle in December is relatively high, it can provide natural moisture for the reaction of the TCHS system, reducing the burden of electricity consumption of the humidifier. Similarly, the minimum value of 4.48 appeared in April. Although solar energy was abundant in April, the charging capacity reached the maximum ( Figure 17); but the natural air was dry, which significantly increased the power consumption of the humidifier ( Figure  18). The average monthly COP was calculated to be 5.56. If the investment in the TCHS system is increased, it is likely to replace most of the gas, and the heating demand of the house can be met with less power consumption.

Conclusions
In this study, a LiCl/LiBr-zeolite composite adsorbent was developed. Subsequently, Figure 19. Monthly COP of TCHS system.

Conclusions
In this study, a LiCl/LiBr-zeolite composite adsorbent was developed. Subsequently, it was applied to a lab-scale reaction unit for experimental study. For the further evaluation of the practical building heating performance of TCHS system, it was applied to the domestic house. The TCHS-building model was built in the TRNSYS program. The main conclusions are as follows:

•
The different salt concentrations resulted in different adsorption capacities and thermal storage properties. Under the same experimental conditions, composite Z15 shows the best adsorption performance.

•
In the adsorption heat storage experiment using a lab-scale reactor, the temperature rise rate at the outlet of the reactor is influenced by the air flow rate and humidity. The studied composite adsorbent exhibited its highest heat density of 434.4 J/g at an air flow rate of 15 m 3 /h and an RH value of 70%, achieving an energy discharge efficiency value of 74.3%.

•
The average error of 10 cycles repeating adsorption experiment was 0.41%. The decrease in temperature lift and adsorption heat showed a slowing down trend with the number of cycling experiments, and the heat density of the last cycle test decreased by only 0.17% which indicated a good periodicity and stability of composite adsorbent.

•
The TRNSYS simulation found that the heating effectiveness of the TCHS system is dependent on the ambient temperature and humidity, and that the power consumption (i.e., fans and humidifiers) is lower in December, reaching a maximum COP of 6.67.