LiCl@C-BMZIF Porous Composites: Synthesis, Structural Characterization, and the Effects of Carbonization Temperature and Salt Loading on Thermochemical Energy Storage

Abstract

To address the imbalance in energy supply and demand across different regions and seasons, the thermochemical conversion process was selected to efficiently utilize surplus energy. In the search for suitable novel materials, this study developed a porous matrix “in-salt” composite using a carbonized metal-organic framework as the carrier and LiCl as the primary reactant. When exposed to water vapor, the innovative material enabled both adsorption and desorption of water vapor, leading to the release and storage of thermal energy, thereby achieving effective energy storage. Using Zn(NO₃)₂·6H₂O and Co(NO₃)₂·6H₂O as metal ion sources and 2-methylimidazole as the ligand, bimetallic zeolitic imidazolate frameworks (BMZIFs) were fabricated via the liquid-phase precipitation method. The composite specimen prepared at a carbonization temperature of 1000 °C with a 20% LiCl mass concentration exhibited the most promising thermal storage performance, achieving the highest capacity, with a final water loss of 53.56% and a stable water adsorption capacity of about 0.831 g·g⁻¹.

1. Introduction

Since the industrial revolution, scientific and technological development has advanced rapidly, yet it has also brought about persistent crises that affect people’s lives. For example, in 2021 natural gas prices in Europe surged dramatically, partly due to rising demand during the winter season [1]. This highlights the urgent need to develop recyclable and renewable energy sources. A persistent challenge is the imbalance between energy supply and demand, influenced by geography, climate, and economic structure [2]. Seasonal fluctuations further intensify this issue, with demand peaking in winter and summer while output remains relatively stable.

In addition, traditional energy utilization often produces large amounts of surplus energy, particularly in the conversion of fossil fuels and electricity [3]. Recovering this wasted energy through storage technologies has therefore become a major focus. Among them, thermal energy storage (TES) is considered a promising strategy [4,5], as it enables heat to be stored during excess supply and released on demand, thus mitigating mismatches, improving efficiency, and reducing environmental impact.

TES is generally classified into sensible, latent, and thermochemical storage [6,7]. Compared with sensible and latent approaches that rely on temperature change or phase transition [3,8,9], thermochemical energy storage offers higher energy density, longer duration, and better stability through reversible reactions [10,11]. Demonstrations such as the DLR pilot system based on CaO/H₂O further highlight its practical potential [10].

Among the materials studied for TES, hygroscopic salts have received considerable attention due to their ability to chemically react with water vapor, thereby enabling reversible energy storage through hydration and dehydration cycles [12]. Their advantages include reaction simplicity, recyclability, high energy density, and tunable operating temperatures [6,13]. Various salts such as CaO, CaCl₂, SrCl₂, MgSO₄, MgCl₂, SrBr₂, and LiCl have been explored for this purpose [14]. Lithium chloride (LiCl), in particular, demonstrates strong hygroscopicity and significant heat release, making it a promising candidate for rapid energy regulation [15]. Its hydration and dehydration reactions are shown in Equations (1) and (2), where energy is released during hydration and stored as chemical energy during dehydration when external energy is supplied.

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}(\mathrm{s})+{\mathrm{H}}_2\mathrm{O}(\mathrm{g})\stackrel{-\mathsf{\Delta }\mathrm{H}}{\to }\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}·{\mathrm{H}}_2\mathrm{O}(\mathrm{s})+\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}$$

(1)

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}(\mathrm{s})+{\mathrm{H}}_2\mathrm{O}(\mathrm{s})+\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}\stackrel{+\mathsf{\Delta }\mathrm{H}}{\to }\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}(\mathrm{s})+{\mathrm{H}}_2\mathrm{O}(\mathrm{g})$$

(2)

Thus, LiCl exhibits both high hygroscopicity and significant heat release capacity. However, this same high hygroscopicity also introduces limitations in terms of stability and storage, necessitating further research into composite strategies.

To overcome such drawbacks, researchers have investigated porous carriers such as metal–organic frameworks (MOFs). MOFs are a class of porous crystalline materials formed through coordination of metal ions or clusters with organic ligands [16,17]. Among them, zeolitic imidazolate frameworks (ZIFs), which use imidazole or derivatives as linkers, display structural similarities to natural zeolites [18,19,20]. Representative structures of commonly studied ZIFs are shown in Figure 1. ZIFs exhibit high surface areas, tunable pore structures, and excellent porosity, making them attractive for applications in catalysis, gas storage, drug delivery, and energy storage [21,22]. These characteristics are particularly advantageous for TES, as they facilitate salt infiltration and water sorption [23,24,25].

Makhanya et al. further emphasized that MOFs possess exceptional porosity, thermal stability, and adsorption capacity, with surface areas sometimes exceeding 6000 m²/g [26]. However, their water adsorption capacity and charging requirements still pose significant challenges.

To address these limitations, recent studies have increasingly focused on salt@MOF composites for thermochemical energy storage. Such composites combine the high water uptake of salts with the structural stability of porous carriers, mitigating issues like swelling, deliquescence, and instability. For instance, Yang et al. reported lithium hydroxide-based covalent organic framework composites with excellent thermal conductivity and cycling performance, achieving a storage capacity of 1916.4 kJ/kg [27]. In a related study, Sun et al. developed LiCl@UiO-66 composites with a thermal storage capacity of 900 kJ/kg, offering hydrothermal stability and improved performance [28]. However, such single-metal MOF-based systems still face challenges in pore stability and limited control of salt distribution. Typically, these salt-in-matrix composites are synthesized via either a one-step or a two-step approach, with the latter (salt impregnation after MOF synthesis) providing better controllability [29,30].

In contrast, this work employs a bimetallic ZIF precursor followed by carbonization, providing a more robust porous matrix and enabling systematic investigation of carbonization temperature, salt loading, and cycling stability. This dual strategy underscores the novelty of LiCl@C-BMZIF compared to earlier studies, including Sun et al. [28].

2. Experimental

2.1. Experimental Reagents and Experimental Equipment

2.1.1. Experimental Material

The primary reagents utilized in this experiment are detailed in

Table 1. Primary experiment regents.

Chemicals	Manufacturer	CAS	Purity/%
2-Methylimidazole	Acros	693-98-1	99%
Methanol	Acros	67-56-1	98%
Cobalt(II) nitrate hexahydrate (Co(NO3)2·6H2O)	Acros	10026-22-9	98%
Zinc nitrate hexahydrate (Zn(NO3)2·6H2O)	Acros	10196-18-6	98%
Polyvinylpyrrolidone (PVP)	Sigma	9003-39-8	98%

. All materials were used without further purification.

2.1.2. Instrumentation

The specialized equipment used in this experiment included a centrifuge for precipitate separation and a carbon tube furnace for carbonization.

2.2. Synthesis of Porous Matrix “In-Salt” Composite Material

In this study, bimetallic zeolitic imidazolate frameworks (BMZIF) are synthesized containing Zn²⁺ and Co²⁺ ions, a porous carbon material C-BMZIF based on BMZIF, and subsequently impregnated with varying concentrations of LiCl to produce a Porous Matrix “In-salt” Composite Material.

2.2.1. Synthetic Method

The method chosen for the synthesis of Metal–Organic Frameworks (MOFs) has a significant impact on the performance of the resultant materials [31]. In this work, we focus on the preparation of BMZIF and its derived porous carbon (C-BMZIF), followed by LiCl impregnation to obtain the final salt@MOF composite. Several strategies for incorporating inorganic salts into MOFs have been reported, including impregnation, organic solvent-assisted, and microwave-assisted techniques. Among these, impregnation stands out as the most environmentally benign and cost-effective.

As a two-step synthesis method, integrating inorganic salt materials into pre-synthesized MOFs requires specific methodologies, and several strategies have been developed to achieve this.

The first one is the impregnation method, which involves directly soaking the MOFs in a solution containing the desired inorganic salt. Under optimal conditions, the inorganic salts permeate and are absorbed into the pore channels of the MOFs. This method’s advantages lie in its simplicity, cost-effectiveness, and the ability to achieve a relatively uniform distribution of the inorganic salts within the MOFs [32]. Next, the Organic Solvent Method employs specific organic solvents to facilitate the permeation and dispersion of inorganic salts. It can more effectively introduce inorganic salt materials into MOFs. However, the stability of these salts within the pore structure of the MOFs warrants further investigation [33]. Microwave-assisted Technique is also an emerging technology. The microwave method utilizes the high energy of microwave radiation to expedite the adsorption and permeation process of inorganic salts into MOFs [34]. The advantage of this technique is the reduced preparation time.

Nevertheless, from an environmental and techno-economic perspective, the traditional impregnation method, which only uses aqueous solutions and does not require additional energy input, stands out. Hence, this technique is more environmentally benign and cost-effective, making it an ideal synthesis means for suitable inorganic salts and MOFs.

2.2.2. Synthetic Procedure

BMZIF was synthesized according to the following procedure [35]. A solution of 3.20 g Zn(NO₃)₂·6H₂O, 0.156 g Co(NO₃)₂·6H₂O, and 2.00 g PVP in 160 mL methanol (Solution A) was prepared. Separately, 7.40 g 2-MeIm was dissolved in 160 mL of methanol (Solution B). Solution A was slowly added to Solution B and stirred at room temperature for 24 h. The precipitate was collected by centrifugation, washed several times with methanol, dried at 70 °C overnight, and ground into powder to obtain BMZIF.

The obtained BMZIF was carbonized under N₂ in a tubular furnace. Samples were heated at 5 °C·min⁻¹ from room temperature to the target temperature (600, 800, or 1000 °C) and held for 120 min, then ground to obtain C-BMZIF [36]. The composite formation process followed the scheme reported by Ahmad et al. [36].

LiCl@C-BMZIF composites were then prepared by impregnation. Typically, 2.0 g of C-BMZIF was dispersed in 100 mL H₂O, and LiCl was added at different mass ratios (5%, 10%, 20%, 30%, 50%). After impregnation, the products were dried and ground to yield a series of LiCl@C-BMZIF composites [37]. The preparation process is summarized in Figure 2.

It should be noted that precise yields of BMZIF, C-BMZIF, and LiCl@C-BMZIF could not be reported. Small but cumulative losses occurred during repeated washing, centrifugation, and grinding steps (e.g., residues adhering to mortar and pestle surfaces), making the recovered mass unsuitable for reliable yield calculation.

In this study, the synthesis variables included carbonization temperature (600, 800, 1000 °C), LiCl loading (5–50%), and cyclic stability tests, which together provided a systematic set of composites for subsequent characterization.

2.3. Characterization Methods

2.3.1. Scanning Electron Microscopy (SEM)

Utilizing SEM characterization, one can discern the microstructural morphology of composite materials prepared under varying conditions [38]. By employing X-ray sources, and optimizing parameters such as voltage 10 kV, and aperture size 30.00 µm. However, as the size of the crystals (about 100 nm) will be smaller than the volume of interaction between the electron beam and the sample (about 1 µm or more), the X-rays will “see” both topography and grain boundaries, which will add error to the numbers. Also, Li and H cannot be analyzed with EDS.

In this study, C-BMZIF rather than pristine BMZIF was selected as the host matrix for LiCl loading. The rationale is that pristine ZIFs are prone to hydrolysis and structural degradation under humid conditions, making them unsuitable for repeated hydration-dehydration cycles. After carbonization, however, BMZIF is transformed into a metal-carbon composite (C-BMZIF) with a more robust carbon skeleton that offers improved thermal and hydrothermal stability. Furthermore, while BMZIF contains mainly micropores (<2 nm) that are easily blocked by salts, carbonization generates mesopores and macropores that provide sufficient space for uniform LiCl dispersion, thereby enhancing cycling stability. SEM measurements were performed using a field-emission scanning electron microscope (Carl Zeiss, Jena, Germany), (Oxford Instruments, Abingdon, UK).

2.3.2. X-Ray Diffraction (XRD)

XRD characterization was employed to assess the structural alterations, peak positions, and relative peak intensities of the Porous Matrix “In-salt” Composite Material prepared under different conditions. Such assessments enable the determination of the material’s composition, internal crystalline structure, or morphology [39]. The X-ray diffraction analysis was carried out with a scan range from 5° to 90°. XRD patterns were obtained using a powder X-ray diffractometer (Bruker, Billerica, Massachusetts, MA, USA).

2.3.3. Fourier Transform Infrared Spectroscopy (FTIR)

When the Porous Matrix “In-salt” Composite Material is subjected to infrared radiation, specific chemical groups and bonds within its molecules, such as C-O, C-H, and C-C, will absorb radiation at distinct wavenumber (cm⁻¹), leading to vibrational motions. This results in a unique infrared wavenumber (cm⁻¹) for the composite material. By analyzing the data obtained from the Fourier Transform Infrared Spectroscopy, we can deduce the chemical composition of different composite materials [39]. FTIR spectra were recorded on a Fourier transform infrared spectrometer (PerkinElmer, Waltham, MA, USA).

2.3.4. Inductively Coupled Plasma (ICP)

Determine the proportion of elements in a sample using the inductively generated plasma state [40]. Using ICP, the sample completely dissolved in the nitric acid solution, is ionized, and the concentration ratio of each element is obtained according to the specific wavelength formed by different elements. It is worth noting that not all materials can be detected using this method, as the carbonized material cannot be completely dissolved. ICP-OES analysis was conducted using an inductively coupled plasma optical emission spectrometer (Agilent Technologies, Santa Clara, CA, USA).

2.3.5. Brunauer–Emmett–Teller (BET)

The investigation of porous materials relies on the utilization of physisorption instruments and adsorption analyzers. By leveraging the Brunauer–Emmett–Teller (BET) model along with the Barrett–Joyner–Halenda (BJH) adsorption–desorption isotherm model, one can accurately determine the material’s specific surface area and pore size distribution [41]. Moreover, by comparing the pore volume of various materials, the precise salt content in composite materials can be deduced. In light of the literature, it is hypothesized that the Porous Matrix “In-salt” Composite Material’s pore architecture comprises both micropores and mesopores. Before the experiment, the samples were degassed at 100 °C for 10 h or overnight under a high-purity N₂ (99.999%) atmosphere. Nitrogen was also used as the adsorption carrier gas during the measurements. The underlying formula for BET is described in Equation (3).

$${\displaystyle \frac{P}{V\left(P_0-P\right)}}={\displaystyle \frac{C-1}{V_{m}C}}\times {\displaystyle \frac{p}{P_0}}+{\displaystyle \frac{1}{V_{m}C}}$$

(3)

where $P$ is the equilibrium pressure of the adsorbed gas;

• $P_0$ represents the saturation vapor pressure of the adsorbing gas;
• $V$ denotes the volume of gas adsorbed at pressure P;
• $V_m$ signifies the volume of gas for monolayer adsorption;
• $C$ is a constant in the BET equation, related to the adsorption energy.

By plotting ${\displaystyle \frac{p}{P_0}}$ as a function against ${\displaystyle \frac{P}{V\left(P_0-P\right)}}$ and performing linear regression on the graphical data, Vm can be obtained, which in turn provides the specific surface area. All BET and sorption measurements were conducted using a Quantachrome Autosorb-iQ-C analyzer (Anton Paar/Quantachrome, Boynton Beach, FL, USA).

2.4. Test Methods

2.4.1. Thermogravimetric Analysis (TGA)

Thermogravimetric analysis (TGA) was performed using a Setaram Sensys Evo TG/DSC instrument (Setaram/Clamart, France) to measure sample weight changes under a high-purity N₂ (99.999%) atmosphere [42]. The system was pre-flushed with N₂ for 10 min before heating. The heating program consisted of two steps:

(1)

Temperature ramp from 25 to 250 °C at 5 °C·min⁻¹.

(2)

Isothermal step at 250 °C for 20 min.

The lower limit of 25 °C ensures that adsorbed water from ambient conditions is captured, while the upper limit of 250 °C guarantees complete removal of both physically and chemically bound water. Higher temperatures were avoided to prevent framework decomposition that would interfere with dehydration analysis. The chosen heating rate of 5 °C·min⁻¹ represents a standard compromise, providing sufficient resolution to distinguish dehydration steps while keeping the experiment time reasonable.

2.4.2. Water Adsorption Detection

Water adsorption was tested at 20 °C and 60% RH. Samples were pre-dried at 90 °C for 2 h and then weighed at 5 min intervals for 120 min. The adsorption percentage was calculated using Equation (4), and the kinetics were fitted with the Box–Lucas model (Equation (5)).

$$\mathrm{Percentage}\mathrm{Increase}={\displaystyle \frac{(w_n-w)-\left(w_0-w\right)}{w_0-w}}$$

(4)

$$y=a\left(1-e^{-bx}\right)$$

(5)

where $w_n$ is the weight after the nth interval, $w$ the Petri dish weight, $w_0$ the initial sample weight; $a$ is the equilibrium adsorption (g·g⁻¹), $b$ the rate coefficient (s⁻¹), $x$ the time (min), and $y$ the adsorption amount (g·g⁻¹).

To assess reproducibility, selected samples were subjected to replicate adsorption–desorption tests, with up to five cycles performed under identical conditions.

3. Results and Discussion

3.1. Series of Samples

Over the course of our experimentation period, a range of samples were synthesized under varying preparation conditions. These include Bimetallic Metal–Organic Frameworks (BMZIF), C-BMZIFs post different carbonization temperatures, and the Porous Matrix “In-salt” Composite Material formed by compounding the salt and 1000C-BMZIFs, specifically termed as LiCl@C-BMZIF.

A series of samples were synthesized, including BMZIF, C-BMZIF carbonized at 600, 800, and 1000 °C, and LiCl@C-BMZIF composites with LiCl loadings of 5%, 10%, 20%, 30%, and 50%.

3.2. SEM Results of Different Samples

The prepared BMZIF, C-BMZIF, and LiCl@C-BMZIF were examined by SEM. BMZIF exhibited the expected polyhedral morphology with particle sizes around 100 nm, confirming successful synthesis and consistent with literature reports. After carbonization, C-BMZIF retained the fundamental polyhedral structure, though the edges appeared more diffuse, reflecting structural changes during thermal treatment.

The SEM images of LiCl@C-BMZIF with different salt contents are presented in Figure 3, showing that the characteristic morphology of the carbonized framework was maintained after LiCl impregnation. At higher loadings, crystalline salt particles were observed on the surface, and pore spaces appeared more filled, indicating successful incorporation of LiCl.

From Figure 3, it becomes apparent that the composite material also show cases a layered construct. Further, even post salt-composition, the dodecahedral rhombic shape remains discernible, underscoring the stability of the C-BMZIF when encapsulated with salt. As the salt concentration escalates, in the imagery of the 20% LiCl@C-BMZIF, one can distinctly observe some sections of crystalline particles, suggesting the material’s surface bears salt crystals. Simultaneously, the internal pore spaces in the 20% LiCl@C-BMZIF appear more filled compared to the 10% LiCl@C-BMZIF, possibly due to the phenomenon of moisture absorption.

3.3. XRD Results of Different Samples

The synthesized BMZIF, C-BMZIF, pure LiCl, and LiCl@C-BMZIF were characterized using X-ray diffraction, with the resulting diffraction patterns depicted in Figure 4.

On the other hand, the C-BMZIF, having undergone a high-temperature annealing process leading to carbonization, evolves into an amorphous carbon structure. This transformation is indicated in Figure 4a by a broad, singular diffraction feature, giving the curve a relatively smoother appearance [43]. In Figure 4b, the PXRD pattern of pure LiCl is directly compared with that of LiCl@C-BMZIF. Finally, compared to C-BMZIF, LiCl@C-BMZIF showcases an additional prominent peak. This diffraction enhancement is attributable to the incorporation of LiCl, as illustrated in Figure 4b.

Figure 4b also highlights that pure LiCl exhibits sharp characteristic diffraction peaks at 2θ values of 30.60°, 33.30°, 35.38°, 50.69°, 60.08°, and 84.51°. In the LiCl@C-BMZIF composite, these peaks are retained, although their relative intensity is reduced due to the presence of the amorphous carbon matrix. Notably, at 33.30°, the diffraction peak of LiCl@C-BMZIF coincides with one of those from LiCl. According to literature [44], this peak is identified as a characteristic reflection of LiCl·H₂O. The emergence of this peak demonstrates the high hygroscopic nature of LiCl. Within the porous matrix, the presence of LiCl·H₂O confirms that the synthesized composite retains the excellent moisture-absorbing properties of lithium chloride, further proving the successful preparation of the Porous Matrix “In-salt” Composite Material. Hence, the final product predominantly constitutes an amorphous carbon matrix but incorporates crystalline regions introduced by the salt via the impregnation method.

3.4. FTIR Results of Different Samples

Infrared spectroscopy was employed to characterize the synthesized BMZIF, C-BMZIF at varying carbonization temperatures, and LiCl@C-BMZIF with different salt ratios.

For BMZIF, distinct absorption bands associated with organic functional groups are observable. After high-temperature annealing, the FTIR spectrum of C-BMZIF shows significant changes: the characteristic vibrations of the original framework (e.g., Zn–N and Co–O) largely disappear, while new signals corresponding to carbonized species emerge, indicating progressive transformation into an amorphous carbon structure.

Figure 5 shows the FTIR spectra of LiCl@C-BMZIF with different salt loadings (5–20%). Several absorption features can be identified: the O–H/C=O band near 3671 cm⁻¹, the C–H stretching bands around 2920 cm⁻¹ and 2850 cm⁻¹, the C≡C stretching at 2324 cm⁻¹, the C=O band at 1740 cm⁻¹, the C=C/C–N band at 1575 cm⁻¹, and the C–O band at 1064 cm⁻¹. With increasing LiCl content, the intensities of these organic functional group peaks (particularly C–H and C–O) gradually decrease, suggesting that salt impregnation partially masks or disrupts the framework vibrations.

Overall, the FTIR results demonstrate the transformation from a crystalline MOF (BMZIF) to a carbonized material (C-BMZIF), and further to a salt-impregnated composite (LiCl@C-BMZIF), while confirming that salt loading effectively modifies the organic framework signals.

3.5. ICP Results of BMZIF

Elemental analysis was performed on the samples according to the introduction in Section 2.3.4. Data were solely acquired for BMZIF due to the insolubility of post-carbonized products in nitric acid at the required pH. In this test, the 5%w nitric acid solution was prepared, and the density is 1.02563 kg/L at 20 °C. The instrumental limits of detection (LoD) for ICP-OES were 0.0001 ppm for Zn (213.857 nm) and 0.0003 ppm for Co (230.786 nm), ensuring that the detected concentrations were well above the threshold. Based on replicate measurements, the measurement uncertainty was within ±5%, consistent with typical ICP-OES quantification of trace elements.

This dataset enabled the determination of the specific molar ratios of Zn and Co ions present within BMZIF. The calculations are elucidated as per Equation (6).

$$n={\displaystyle \frac{m}{M}}$$

(6)

The data integration is shown in

Table 2. The ICP analysis of BMZIF.

Element	Concentration/mg·L−1	Molar Mass/g·mol−1	Amount of Substance/mol
Zn	0.76	65.38	0.0116
Co	0.01	58.93	1.70 × 10−4

.

According to these results, the Zn:Co molar ratio in BMZIF was approximately 69:1. Compared with the theoretical feed ratio of ~20:1 (derived from 3.20 g Zn(NO₃)₂·6H₂O and 0.156 g Co(NO₃)₂·6H₂O), a significant discrepancy is evident. Such deviations between experimental and nominal ratios have also been reported in Zn/Co-ZIF systems, where differences in the coordination kinetics of Zn²⁺ and Co²⁺ with 2-methylimidazole can result in preferential Zn incorporation and reduced Co content.

3.6. BET Results of Different Samples

The surface area of BMZIF, C-BMZIF, and LiCl@C-BMZIF was analyzed by measuring the N₂ adsorption-desorption isotherms (Figure 6). According to the IUPAC classification, both BMZIF and C-BMZIF exhibit characteristic Type IV isotherms. The hysteresis loop observed at relative pressures (${\displaystyle \frac{p}{P_0}}$) between 0.9 and 1.0 indicates the formation of mesoporous structures within the material. The near-vertical nature of the hysteresis loop suggests the assembly of a lamellar (or plate-like) structure. In contrast, upon the addition of LiCl, the composite material shows non-porous characteristics.

Furthermore, the pore size distribution of the samples is shown in Figure 7, which reveals distinct differences among the three materials. BMZIF displays a sharp distribution dominated by micropores (<2 nm), while C-BMZIF shows a broader range of mesopores (2–50 nm) and macropores (>50 nm) due to partial framework collapse during carbonization. For LiCl@C-BMZIF, the overall pore volume decreases significantly, as the introduced salt partially blocks the pore channels, thereby reducing both the microporous and mesoporous fractions.

For BMZIF, the plots exhibit distinct peaks around 1.1 and 35 nm, suggesting that the material possesses a significant number of pores at these sizes. Similarly, the C-BMZIF demonstrates pore diameters roughly at 1.4 or 30–60 nm. This indicates that both materials not only contain micropores and mesopores but also manifest macroporous structures, which are highly advantageous for subsequent moisture-absorbing salt encapsulation. At the same time, this also proves that carbonization changes the pore structure, which is beneficial to the subsequent water adsorption or salt recombination. Conversely, the composite material exhibits pore sizes around 6.25 and 27 nm. This observation underscores the successful incorporation of the salt into the C-BMZIF’s pore network, leading to a reduction in the pore dimensions of the host material.

Furthermore, BET analysis is the maximum adsorption capacity of multilayer adsorption obtained by the BET equation, so that the specific surface area of the material can be calculated [45]. In addition, DFT provides a method that can take into account the pore structure and the interaction of adsorbed molecules, resulting in a more precise pore size distribution and pore volume in the material [46]. The statistical results of different materials are shown in

Table 3. Surface area and pore volume of different samples.

	BMZIF	C-BMZIF	LiCl@C-BMZIF
Surface area with MBET/m2·g−1	1945.82	1044.74	162.83
Pore volume with DFT/cm3·g−1	1.21	1.29	0.35

.

Following carbonization, the destruction of the organic–metal framework results in a collapse of the structure, leading to an increased pore volume. Simultaneously, there is a decrease in the specific surface area. Upon introducing hygroscopic salts, the salts permeate the pore architecture, filling the voids, which subsequently results in a significant reduction in both the specific surface area and pore volume.

3.7. TGA Results of Different Samples

Thermogravimetric analysis (TGA) can be employed to investigate the dehydration performance of composite materials. The experimental temperature range was set between 25 and 250 °C, ensuring that the observed weight loss corresponds to the amount of water released. The data and calculation results are presented in

Table 4. Desorption final weight and final weight for water lost.

Sample	1000C-BMZIF	5%LiCl@C-BMZIF	10%LiCl@C-BMZIF	20%LiCl@C-BMZIF	30%LiCl@C-BMZIF	LiCl
Dehydration final weight/%	80.30	68.30	60.35	46.44	49.42	44.53
Final weight for water lost/%	19.70	31.70	39.65	53.56	50.58	55.47

.

Based on the obtained data, C-BMZIF exhibits the least weight loss. With the addition of salt to the carrier material, the resultant composites exhibit varying degrees of enhanced weight loss compared to C-BMZIF. This demonstrates the effectiveness of the composite. Meanwhile, the As the salt concentration increases, the weight loss generally exhibits an increasing trend. However, the sample with 30% LiCl@C-BMZIF displayed a poorer performance compared to its 20% sample. This can be attributed to excessive salt loading, which potentially obstructs the porous channels of the material. It suggests that salt loading of 30% surpasses the optimal capacity that this material can sustain, thereby hindering its reactivity with water vapor. Moreover, in the process of dehydration, too much salt may be precipitated in the pores, which is very unfavorable to the repeated use of materials. Concurrently, the desorption final weights for 5%LiCl@C-BMZIF, 10%LiCl@C-BMZIF, and 20%LiCl@C-BMZIF are 68.30%, 60.35%, and 46.44%, respectively. Hence, this does not exhibit a linear relationship. This discrepancy might be attributed to the fact that the actual salt content in the samples might not be exactly 5%, 10%, and 20%. Additionally, varying salt concentrations could influence the final pore structure of the formed composites. Overall, after loading salt onto 1000C-BMZIF, the 20%LiCl@C-BMZIF demonstrated the most optimal performance.

For LiCl, the weight loss observed, culminating in a final weight of 44.5275%, unfolds across three distinct phases. Initially, between 25 °C and 165 °C, there is a slow dehydration. This is followed by a second phase from 165 °C to 200 °C, where the dehydration rate noticeably accelerates. Finally, the weight remains consistent between 200 °C and 250 °C. This behavior likely points to a multi-step dehydration reaction. Given the molar mass of LiCl and water, we can calculate based on Equation (3) and get $n_{H_{2}O}:n_{LiCl}\approx 3:1$. It means that the precursor form is LiCl·3H₂O. The observed rate difference between the first and second phases lends support to prove that this is a multi-step dehydration reaction, which can be expressed as Equations (7) and (8).

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}·3{\mathrm{H}}_2\mathrm{O}+\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}\stackrel{\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}}{\to }\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}·2{\mathrm{H}}_2\mathrm{O}+{\mathrm{H}}_2\mathrm{O}$$

(7)

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}·2{\mathrm{H}}_2\mathrm{O}\stackrel{\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}}{\to }\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}+2{\mathrm{H}}_2\mathrm{O}$$

(8)

Although the carbonization temperature of the carrier material reached up to 1000 °C during synthesis, the TGA experiments were intentionally limited to 250 °C. The rationale is that both physically and chemically adsorbed water can be fully and safely removed below 250 °C, ensuring that the observed weight loss is exclusively attributed to dehydration. Extending the measurement range to higher temperatures would likely trigger decomposition or phase transitions of the MOF skeleton and carbon matrix, leading to additional weight loss unrelated to water release. Such overlap would confound the analysis and prevent a clear distinction between water desorption and framework degradation.

3.8. Water Sorption Analysis

Beyond the insights offered by TGA, water adsorption is a crucial metric in selecting the optimal sample. Both the water adsorption capacity and water adsorption rate are pivotal considerations. It is also important to compare the results between TGA and water sorption. In the subsequent sections, the hydration dynamics will be analyzed in order of carbonization temperature, the ratio of hygroscopic salts, and cycling stability.

3.8.1. The Influence of Carbonization Conditions

Under consistent preparation conditions for BMZIF, carbonization was carried out at temperatures of 600 °C, 800 °C, and 1000 °C, respectively. This resulted in three variations in C-BMZIF, designated as 600C-BMZIF, 800C-BMZIF, and 1000C-BMZIF. Water adsorption tests for these samples were conducted in a controlled environment maintained at 20 °C and 60% relative humidity. The results showed clear differences in adsorption performance among the three carbonized samples, reflecting the strong influence of carbonization temperature on the pore structure and water uptake capacity.

Furthermore, the kinetic parameters were obtained using the pseudo-first-order model fitted to the water sorption curves, and the processed data are summarized in

Table 5. Water sorption results of C-BMZIF.

	600C-BMZIF	800C-BMZIF	1000C-BMZIF
Measured Water Adsorption Capacity /g·g−1	0.15	0.12	0.24
Calculated Water Adsorption Capacity a/g·g−1	0.14	0.12	0.25
Calculated Rate of Water Adsorption b/s−1	0.05	0.05	0.07
R2	0.996	0.999	0.997

^a Fitting coefficient representing the calculated water adsorption capacity (a in y = a(1 − e^−bx)). ^b Fitting coefficient representing the calculated rate constant of water adsorption.

.

From the data presented, after carbonization, the water adsorption performance of the material has been improved to varying degrees. The 1000C-BMZIF sample exhibits superior water adsorption capability, reaching approximately 0.25 g·g⁻¹. Concurrently, it demonstrates the best adsorption rate at around 0.07 s⁻¹. This is noticeably higher than the 0.15 g·g⁻¹ and 0.05 s⁻¹ of 600C-BMZIF and the 0.14 g·g⁻¹ and 0.05 s⁻¹ of 800C-BMZIF.

This suggests that the pore structure formed post-carbonization of BMZIF at 1000 °C is more conducive to water adsorption. Primarily, the elevated temperature causes a more pronounced collapse of the originally ordered pore structure in BMZIF, favoring the formation of larger pores. While this results in a reduced specific surface area, there is a corresponding increase in total pore volume, indicating that materials with larger pore volumes exhibit enhanced water adsorption capacities. Secondly, the formation of larger pores enhances the interconnectivity within the material, facilitating more efficient water molecule transport and consequently leading to an increased water adsorption rate.

3.8.2. The Influence of Salt Content

In maintaining the synthesis conditions consistent for 1000C-BMZIF, composites with varying mass fractions of LiCl, specifically 5%, 10%, 15%, 20%, 30%, and 50%, were synthesized. These materials were correspondingly denoted as 5%LiCl@C-BMZIF, 10%LiCl@C-BMZIF, 15%LiCl@C-BMZIF, 20%LiCl@C-BMZIF, 30%LiCl@C-BMZIF, and 50%LiCl@C-BMZIF. Water adsorption experiments were subsequently conducted in a controlled environment of 20 °C and 60% relative humidity.

The results showed that the 5%LiCl@C-BMZIF sample exhibited relatively poor water adsorption performance, likely due to the insufficient salt content entering the pores, which limited the hygroscopic capacity of the material. By contrast, samples with higher LiCl loadings demonstrated markedly enhanced water uptake, confirming the positive correlation between salt content and adsorption capability.

The kinetic parameters derived from the water sorption curves are summarized in

Table 6. Water sorption results of N% LiCl@C-BMZIF and LiCl.

	5%	10%	20%	30%	50%	LiCl
Measured Water Adsorption Capacity/g·g−1	0.36	0.66	0.77	1.00	0.64	0.85
Calculated Water Adsorption Capacity a/g·g−1	0.38	0.64	0.98	1.84	1.05	4.10
Calculated Rate of Water Adsorption b/s−1	0.28	0.30	0.01	0.01	0.02	0.00
R2	0.99978	0.99976	0.99968	0.99661	0.99969	0.99985

^a Fitting coefficient representing the calculated water adsorption capacity (a in y = a(1 − e⁻ᵇˣ)). ^b Fitting coefficient representing the calculated rate constant of water adsorption.

.

Meanwhile, the 10%LiCl@C-BMZIF demonstrates an optimal water adsorption rate of about 0.29 s⁻¹, markedly above the 20%LiCl@C-BMZIF’s 0.01 s⁻¹ and the 30%LiCl@C-BMZIF’s 0.01 s⁻¹. Compared with the data for the 1000C-BMZIF, this trend hints at a decreasing water uptake rate as the salt fraction in the matrix escalates. It is postulated that the salt crystals might be clogging the carbon material’s pores and voids, thus hampering pore connectivity and inducing the agglomeration of the base material into larger particles. Both these factors augment the diffusion path and resistance for water molecules within the material, consequently leading to a reduced water adsorption rate.

According to Table 6, the 30%LiCl@C-BMZIF manifests the best water adsorption capacity, reaching roughly 1.842 g·g⁻¹. The data suggests that with increasing salt content, there is a concomitant rise in the water adsorption capacity of the composite, underscoring the salt’s role as the principal agent in water uptake.

During the experiment, the 50%LiCl@C-BMZIF exhibited deliquescence, progressively transitioning to a liquid state upon water adsorption, as visualized in Figure 8. In stark contrast, the solid-state stability of the 10%LiCl@C-BMZIF post water adsorption and post drying remained intact, as shown in Figure 9. The observed phase transition from solid to liquid in the 50%LiCl@C-BMZIF is attributed to salt precipitation from the matrix, leading to the deliquescence behavior akin to pure LiCl. It is noteworthy that the hygroscopic performance of 50% LiCl@C-BMZIF is inferior to that of pure LiCl. This suboptimal behavior can be attributed to an excessive salt loading, potentially leading to the occlusion or blockage of the porous channels. Consequently, water molecules may find it challenging to access or traverse these channels. Furthermore, when an overabundance of salt absorbs water, it tends to form a liquid solution. This phenomenon may result in the salt dissolving in the adsorbed water, further compromising the hygroscopic capabilities of the porous material.

The water adsorption performance of the 20%LiCl@C-BMZIF was relatively balanced, sitting between the capacities of the 10%LiCl@C-BMZIF and 30%LiCl@C-BMZIF. According to the sorption data, increasing salt content generally enhanced water uptake, but excessive loading (≥30%) led to pore blockage and structural stress, ultimately reducing stability. This trend is consistent with TGA results, although the data do not align perfectly due to methodological differences: TGA employs stepwise heating under an inert N₂ atmosphere, whereas water sorption is measured under steady-state humid conditions.

Reproducibility tests confirmed the reliability of these observations. The 20%LiCl@C-BMZIF consistently exhibited superior water uptake compared with the 10%LiCl@C-BMZIF across repeated cycles, with only minor variations attributable to ambient humidity fluctuations. By contrast, the 30%LiCl@C-BMZIF initially showed relatively high uptake in the first trial but suffered a marked decline in the second, accompanied by visible salt precipitation and structural consolidation; further repetitions were therefore discontinued. These results indicate that excessive salt loading compromises the cycling stability of the composite.

Overall, SEM observations, BET analysis, and sorption measurements consistently demonstrate that pristine BMZIF is microporous and unsuitable for salt loading. Carbonization produces mesopores and macropores that improve transport and uptake, while TGA and sorption both indicate enhanced performance with increasing salt content up to 20%, beyond which structural instability emerges. Consequently, 10%LiCl@C-BMZIF, 20%LiCl@C-BMZIF, and 30%LiCl@C-BMZIF were selected for the subsequent cycling stability experiments.

3.8.3. The Influence of Cycle Stability

Maintaining the experimental conditions for water adsorption, selected samples of 10%LiCl@C-BMZIF, 20%LiCl@C-BMZIF, and 30%LiCl@C-BMZIF were subjected to five repeated cycles. The cyclic results are presented in Figure 10.

The 30%LiCl@C-BMZIF exhibited higher initial uptake but suffered a pronounced decline in subsequent cycles, consistent with its inferior TGA performance and indicative of structural degradation. By contrast, the 10% and 20% loadings maintained stable adsorption capacities of ~0.62 g·g⁻¹ and ~0.83 g·g⁻¹, respectively, across five cycles, confirming that moderate salt loading (10–20%) ensures reproducibility, whereas excessive loading compromises integrity.

Compared with other salt@MOF systems, LiOH–COF derived composites achieved a high storage capacity of 1916.4 kJ·kg⁻¹ with 94.5% retention after 25 cycles, while LiCl@UiO-66 composites exhibited a capacity of 900 kJ·kg⁻¹ and maintained hydrothermal stability over 40 cycles. In contrast, the present LiCl@C-BMZIF offers simpler and more scalable synthesis, lower raw material cost, and a mesoporous carbon framework that enables higher salt loading and faster adsorption kinetics. However, its cyclic stability has so far been validated only over five cycles, which is significantly less extensive than the literature reports. Further investigation is required to establish its long-term durability.

4. Limitations and Outlook

4.1. Limitations

First, the initial synthesis of the carrier material (BMZIF) requires a protracted time span, implying inefficiencies in the material preparation phase.

Second, degradation of the material structure has been observed after carbonization. However, existing characterization techniques such as XRD, FTIR, and ICP have proven inadequate for precisely determining the molar ratios of specific metal ions, namely Zn²⁺ and Co²⁺ in C-BMZIF. This limitation results in an incomplete understanding of the stoichiometric relationships within the carrier material.

Third, narrower salt concentration gradients are needed. In addition, due to project time constraints, cycling stability tests were limited to only five cycles. Therefore, the data may not provide a comprehensive assessment of the true stability of the sample material during extended use. Furthermore, comprehensive BET results were not obtained, restricting in-depth quantitative analysis of the relationship between salt concentration and heat release.

Fourth, discrepancies were observed between TGA and water sorption results. This inconsistency primarily arises from methodological differences: TGA employs stepwise heating under an inert N₂ atmosphere, forcing the release of both physically adsorbed and chemically bound water, and may also include weight loss from residual solvents. In contrast, water sorption tests measure equilibrium uptake under ambient humidity, which may underestimate adsorption due to kinetic limitations, particularly in micropores. Moreover, structural changes induced during TGA heating could irreversibly alter the material’s adsorption sites. As a result, TGA provides information on the total water content and thermal stability, whereas water sorption reflects equilibrium performance under practical operating conditions.

Finally, from the perspective of technical economy, the costs associated with material synthesis are relatively high, suggesting that the current form of the material may not be feasible for mass production. It should also be underscored that the research presented herein remains at the laboratory scale. In practical scenarios, bulk aggregation of the material could potentially lead to deviations in performance. Consequently, practical application outcomes may diverge from controlled laboratory results.

4.2. Future Outlook

In light of the constraints delineated above, several aspects merit further investigation in future research. First, the feasibility of employing microwave-assisted synthesis techniques for the fabrication of MOFs warrants thorough exploration, as this approach may enhance synthesis efficiency and yield superior material properties.

Second, a comprehensive cycling stability analysis is essential. Future studies should place greater emphasis on extending and intensifying the cycling tests to evaluate the material’s long-term performance. Additionally, obtaining complete BET datasets would enable a more detailed examination of the correlation between salt loading, pore characteristics, and heat storage capacity.

Finally, before considering the deployment of the Porous Matrix “In-salt” Composite Material in real-world applications, pilot-scale evaluations under practical conditions are indispensable. These trials would clarify whether volumetric scaling of the material results in changes to its intrinsic properties.

5. Conclusions

In this study, a Porous Matrix “In-salt” Composite Material was synthesized using a two-step method, with an emphasis on evaluating its thermal energy storage performance. Two single-variable optimization experiments were conducted, focusing on the carbonization temperature and the ratio of hygroscopic salt to carrier material, to identify the specimen with the most outstanding thermal storage capability. The main conclusions are summarized as follows:

(1)

Bimetallic Zeolitic Imidazolate Frameworks (BMZIFs) were successfully synthesized via liquid-phase precipitation, employing salts Co(NO₃)₂·6H₂O and Zn(NO₃)₂·6H₂O in conjunction with the organic linker 2-methylimidazole. The resulting BMZIF exhibited a rhombic dodecahedron structure with an average size of ~90 nm, and the ratio of Zn²⁺ to Co²⁺ was approximately 69:1. LiCl was subsequently incorporated by the impregnation method, yielding the composite material.

(2)

Carbonization at 1000 °C produced C-BMZIF (1000C-BMZIF), which exhibited the most pronounced reactivity with water among the tested samples. Consequently, it demonstrated the optimal thermal energy storage performance, with a final weight loss of ~19.7% due to dehydration and a measured water adsorption capacity of 0.24 g·g⁻¹.

(3)

Based on the present results, the composite material with a 20 wt% LiCl loading (20%LiCl@C-BMZIF) showed the highest thermal storage capacity, with a final weight loss of 53.6%. Moreover, it retained structural stability and exhibited excellent water adsorption performance, achieving a capacity of 0.84 g·g⁻¹ and an adsorption rate of 0.01 s⁻¹.