Highly-oriented graphite/polyimide–carbon nanotube supported composite phase change materials with high thermal conductivity and photothermal conversion performance

Abstract

Phase change materials have significant application prospects in thermal energy storage and management. However, challenges such as low thermal conductivity, liquid leakage and solid rigidity have hindered their practical applications. In this study, a dual encapsulation strategy was adopted, using a highly-oriented graphite framework (HOGF) as the large framework and a polyimide/carbon nanotube (PI/CNT) aerogel as the small framework to construct an oriented carbon skeleton with high thermal conductivity. Subsequently, it was impregnated with n-octadecane (OD), and composite phase change materials (OHPC-x) with bidirectional high thermal conductivity, heat storage and high photothermal performance were successfully prepared. The increase of OD endows OHPC with excellent heat storage capacity, and the enthalpy value of OHPC-2 can reach 164.46 J g⁻¹. In addition, the lamellar structure of the HOGF provides phonon transmission channels, endowing the OHPC composite with a relatively high in-plane thermal conductivity (5.8913 W m⁻¹ K⁻¹). CNTs, as thermally conductive fillers and light collectors, can not only expand the heat transfer area but also reduce thermal resistance. Their addition enabled OHPC to achieve an enhanced axial thermal conductivity (2.2934 W m⁻¹ K⁻¹) and a high photothermal conversion rate (86.9%). The developed composite material has achieved a perfect combination of multiple functions and holds great application potential in the efficient utilization of solar energy, building thermal management, and the protection of electronic equipment.

---

New concepts

In thermal energy storage and management systems, phase change materials (PCMs) have attracted much attention due to their outstanding energy storage density and economic feasibility. However, challenges such as low thermal conductivity, liquid leakage and solid rigidity have hindered their practical applications. This study proposes a dual encapsulation strategy using a porous carbon skeleton to address the limitations of PCM leakage and low thermal conductivity. We designed an oriented carbon skeleton with high thermal conductivity using expanded graphite paper (HOGF) as the large skeleton and a polyimide/carbon nanotube (PI/CNTs) aerogel as the small skeleton. Subsequently, the composite phase change material (OHPC-x) with high thermal conductivity, heat storage and high photothermal performance was prepared by octadecane (OD) impregnation. This work offers a promising and feasible approach for the production of high-performance materials for energy storage and thermal management applications. The prepared composite materials have potential applications in the efficient utilization of solar energy, building thermal management, and the protection of electronic devices.

1. Introduction

With the exponential growth of global energy demand, the development of efficient thermal energy storage (TES) technology has become an inevitable path to achieving the sustainable development goals. Among diverse thermal management solutions, phase change materials (PCMs) are widely recognized as an innovative approach to resolve spatiotemporal mismatches in thermal energy supply and demand.^1–5 Nevertheless, solid–liquid PCMs often face inherent limitations, such as low thermal conductivity and leakage risks.^6–8 Composite PCMs (CPCMs) designed using the synergistic coupling effect between the support matrix and highly thermally conductive fillers exhibit enhanced shape stability and superior heat transfer characteristics.^9–12 This technology has great potential in key areas such as building energy efficiency and thermal management of power battery systems.

The fabrication system of form-stable CPCMs primarily encompasses micro/nano-encapsulation of PCMs within core–shell architectures,^13–15 physical immobilization of PCMs via confinement effects in porous media,^16–18 and construction of polymer–PCM crosslinked networks through intermolecular interaction modulation.^19–21 Compared to the interfacial incompatibility challenges inherent in microencapsulation techniques and the kinetic hysteresis limitations of chemical crosslinking approaches, porous substrate materials have emerged as pivotal functional carriers in CPCM design due to their dual advantages of high specific surface area with tunable porosity distribution and intrinsic high thermal conductivity.^22,23 Recent research has systematically explored diverse porous substrates, including metal–organic frameworks (MOFs),²⁴ metallic foams,²⁵ porous carbons,^26–28 boron nitride,²⁹ and MXene aerogels.^30,31 Notably, highly aligned carbon-based porous materials (e.g., graphene aerogels,^32,33 carbonized wood,³⁴ and highly-oriented graphite^35,36) have demonstrated revolutionary breakthroughs owing to their anisotropic thermal transport characteristics and enhanced interfacial π–π interactions.^37,38

The oriented carbon-based materials are structurally and regularly arranged in a certain direction, forming phonon or electron transport paths, which improves the thermal conduction efficiency of CPCMs.³⁹ At present, some 1D/2D fillers with oriented high thermal conductivity have been prepared into 3D aerogels with oriented structures through external factors such as directional freezing or pressure, which are used to encapsulate PCMs to enhance their thermal conductivity.⁴⁰ Hu et al. fabricated high-thermal-conductivity CPCMs with oriented carbon fibers (CF) through single-screw extrusion.⁴¹ The axial thermal conductivity of a CPCM with oriented CF was significantly increased to 6.31 W m⁻¹ K⁻¹. In addition, it also demonstrated excellent photothermal conversion and thermal management capabilities. Gao et al. constructed a graphene structure with high vertical orientation by using unidirectional freezing and high-temperature carbonization technology.⁴² The interface thermal resistance was significantly reduced, and it demonstrated an excellent thermal conductivity of up to 4.85 W m⁻¹ K⁻¹ at only 4.26 vol% graphene oxide (GO), thereby achieving rapid heat transfer and storage of PCMs.

Nevertheless, 1D and 2D thermally conductive fillers require other “adhesives” to assist in building the 3D framework, and it is inevitable that there will be situations of poor bonding and discontinuous structure. The presence of “adhesives” can also cause high interfacial thermal resistance between fillers in the framework, making it difficult to significantly increase the overall thermal conductivity.^43–45 The emergence of high thermal conductivity carbon materials with 3D large frames has addressed this problem. They not only have a complete structure to support the entire aerogel but also provide a complete transmission channel for heat conduction, significantly improving the thermal conductivity. Graphite paper (GP) is a multi-layer stacked carbon material, which is formed by pressing and orienting expanded graphite particles. It features readily available raw materials and is suitable for large-scale production. Its sp² hybrid carbon atoms form a honeycomb lattice, making the in-plane thermal conductivity exceed 100 W m⁻¹ K⁻¹. Zhang's team developed a directional construction technology of highly-oriented graphite (HOG) skeletons based on a room temperature acid expansion process, which enables the controlled preparation of highly oriented expanded graphite networks in a one-step process, providing an innovative path for the development of functional carbon-based materials.⁴⁶ The obtained HOG thermally conductive filler, when compounded with polydimethylsiloxane (PDMS), exhibits an excellent in-plane thermal conductivity of 35.4 W m⁻¹ K⁻¹. Moreover, by regulating the oxidation degree of the precursor, three-dimensional anisotropic thermal conduction channels can be precisely constructed. This 3D carbon skeleton has great application potential in the preparation of oriented high thermal conductivity CPCMs.

However, although HOGF has the advantage of directional structure, it is limited by the geometric limitations of two-dimensional layer stacking, making it hard to achieve seamless encapsulation of PCMs under high-temperature conditions.³⁵ In this regard, a dual constraint mechanism can be established by constructing a cross-scale secondary interpenetrating network. This strategy can not only enhance the PCM fixation capacity through the topological locking effect, but also achieve multi-level strengthening of the heat conduction path, thereby breaking through the performance ceiling of traditional single-phase skeleton materials. Polyimide (PI) aerogels, due to their unique imide ring and imide heterocyclic structures, possess outstanding mechanical strength, low dielectric constants, excellent chemical resistance and temperature resistance.^47–49 PI aerogels feature a highly interconnected three-dimensional nanopore structure, with pore diameters typically at the nanoscale (<100 nm). Therefore, filling PI as a secondary network between HOGF layers can effectively limit the leakage of PCM molecules. However, PI aerogels tend to shrink and their overall structure is easily destroyed during the molding process, and it is generally necessary to add enhanced nano-fillers in the preparation process.⁵⁰ Carbon nanotubes (CNTs), as highly efficient thermally conductive fillers and light capture agents, when filled between HOGF layers, can not only increase the thermal conduction pathways but also enhance the photothermal conversion rate.

To address the aforementioned challenges, this study proposes a dual-stage encapsulation strategy: utilizing a highly-oriented graphite framework (HOGF) as a macroscopic framework while constructing microscale secondary networks via polyimide (PI) and carbon nanotubes (CNTs) for synergetic encapsulation of the high-latent-heat phase change material n-octadecane (OD). The incorporation of HOGF endows the CPCMs with anisotropic high thermal conductivity (5.8913 W m⁻¹ K⁻¹ in-plane), while the PI/CNTs architecture optimizes pore structure homogeneity, enhancing PCMs’ loading capacity. The synthesized OD/HOGF/PI/CNTs composite material (OHPC) exhibits excellent enthalpy (ΔH_m = 164.46 J g⁻¹), thermal cycling stability (structural stability after 100 cycles), and extremely high photothermal conversion capacity (η = 86.9%). The prepared OHPC is applied to the thermal management of buildings, taking advantage of its high photothermal conversion performance and passive cooling capacity to meet the all-weather demands. This work offers a promising and feasible approach for the large-scale production of high-performance materials for energy storage and thermal management applications.

2. Materials and methods

2.1. Materials

Commercially available graphite papers (thickness: 1–3 mm, density: 0.9–1.1 g cm⁻³) were purchased from Guangshengjia New Material LLC, China. N-Octadecane (OD) was purchased from Shanghai Yuanye Biotechnology Co., Ltd. Multi-walled carbon nanotubes (CNTs, 99%) were purchased from Shenzhen Suiheng Technology Co., Ltd. N,N-Dimethylacetamide (DMAc, 99%) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Homophthalic anhydride (PMDA, 99%) was purchased from Shanghai Maclin Biochemical Technology Co., Ltd. 4,4′-Diaminodiphenyl ether (ODA, 98%) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.

2.2. Preparation of HOGF

Initially, concentrated sulfuric acid (95.5–96.5%) and nitric acid (68%) were mixed at a 1 : 1 volume ratio and subjected to intense agitation in an ice bath for 15 minutes. For HOGF production, graphite sheets with varying thicknesses (1–3 mm) underwent room-temperature immersion in the mixed acid solution for 60 minutes to facilitate the expansion process. The resulting HOGF was then thoroughly rinsed with deionized water until neutral pH was achieved. Subsequent drying at 100 °C for 24 hours yielded a HOGF with tunable density and dimensions through precise optimization of preparation parameters (Fig. 1a).

Fig. 1 (a) Schematic diagram of the increase in the interlayer distance of the HOGF caused by expansion in acidic solution and (b) schematic diagram of the preparation process; (c) molecular synthesis reaction of polyimide.

2.3. Preparation of PAA/CNTs

A solution was prepared by dissolving 2 g of ODA in 24 g of DMAc under continuous magnetic stirring. Subsequently, 2.21 g of PMDA was gradually introduced into the mixture, which was maintained in an ice-water bath for 2 hours to ensure homogeneous polymerization. The reaction system was then supplemented with 1.01 g of triethylamine (TEA) and subjected to mechanical agitation for 5 hours, yielding a poly(amic acid) (PAA) precursor solution. The reaction process is shown in Fig. 1c. Freeze-drying of this solution produced water-soluble PAA solids. For poly(amic acid) salt preparation, 3 g of PAA and 0.84 g of TEA were dissolved in 200 g of deionized water through vigorous stirring, forming an aqueous PAA salt solution. CNT composites were fabricated by adding 0.5, 1, or 2 wt% CNTs into 5 g aliquots of the PAA solution, followed by magnetic stirring until uniform dispersion was achieved (Table S1).

2.4. Preparation of OD/HOGF/PI/CNTs

As shown in Fig. 1b, the HOGF was immersed in the PAA/CNTs solution and vacuum-stabilized for 1 hour to ensure that the solution completely penetrated the interlayer space of the HOGF. The obtained mixture was freeze-dried to obtain the HOGF/PAA/CNTs aerogel. This aerogel was then transferred to a tubular furnace and subjected to amidation by heating to 200 °C for 1 hour and holding at 300 °C for 2 hours under the protection of argon gas, resulting in the HOGF/PI/CNTs aerogel. Finally, molten n-octadecane was added to the HOGF/PI/CNTs aerogel and vacuum-impregnated at 80 °C for 8 hours. After removing the excess OD, the OD/HOGF/PI/CNTs-x composite material was obtained (denoted as OHPC-x, where x represents the mass fraction of CNTs).

2.5. Characterization

The micro-morphology of the aerogels and composites was characterized using a scanning electron microscope (SEM, Hitachi SU8000, Japan). The samples were pre-dried before preparation, and then they were torn or cut with tweezers or a knife and pasted on a conductive adhesive. Afterwards, the surface was treated with platinum spray. Finally, the samples were fed into the instrument and observed at an operating voltage and current of 5 kV and 10 μA. Diffraction analysis of PAS, PI, C-PI composite aerogels and CPCMs was carried out using an X-ray diffractometer (XRD, Bruker D8 ADVANCE), and their diffraction patterns were analyzed to obtain relevant information on the materials. The bulk materials were directly placed on the sample stage and flattened, the powder samples were spread on the sample stage and flattened, and the working voltage and current were set at 40 kV and 40 mA with a sweep speed of 5° min⁻¹ for the tests. The surface functional groups and chemical structures of the samples were characterized using a Fourier transform infrared spectrometer (FT-IR, Nicolet 6700). A small amount of sample and potassium bromide (at a ratio of about 1 : 100) was ground thoroughly with a mortar and pestle, and then the sample was pressed using tablet pressing equipment. The samples were then placed in the instrument and observed in the wavelength range of 4000–400 cm⁻¹. A Raman spectroscopy tester (Finder Vista) was used to analyze the degree of graphitization of the biocarbon materials. The mass loss of the material during the heating process was analyzed using a synchronous thermal analyzer (TGA, Netzsch STA449F, Germany). About 10 mg of the sample was weighed and placed in the instrument. The thermal stability of the sample was measured under a nitrogen atmosphere in the temperature range of 30–800 °C at a heating rate of 10 °C min⁻¹. The heat storage capacity of the samples was tested using a differential scanning calorimeter (METTLER TOLEDO DSC3, Switzerland). 5–10 mg of the sample was weighed and placed in a crucible and the mass was recorded. The phase transition parameters of the samples were characterized from 10 °C to 90 °C at a heating–cooling rate of 10 °C min⁻¹ under a N₂ atmosphere. In order to accurately assess the thermal diffusivity of the samples, measurements of the thermal diffusion coefficients were carried out at room temperature using a laser thermal conductivity meter (NETZSCH LFA 467). The absorbance of the samples was determined using a Shimadzu UV-Vis-NIR spectrometer (UV-2550). The broad spectral absorption strength of the samples was analyzed using a barium sulfate white plate as a baseline and a scanning wavelength of 200–800 nm. The photothermal conversion efficiencies of various CPCMs were evaluated using a laboratory self-assembly setup. To simulate solar irradiation, a xenon lamp (HSX-F300, China) with a constant power of 300 W was placed 30 cm above the sample. The thermal radiation density of the xenon lamp was measured using an optical power meter (THORLABS, Germany), which was maintained at 100 mW cm⁻². An infrared thermometer (Optris CT, Germany) was used to capture the surface temperature of the samples in the photothermal conversion experiments. The monolayer composites and the heat-insulating-thermal-conducting integrated composites were placed on a heating table at 55 °C, and an infrared thermal imager (FLUKE Ti480 PRO) was used to photograph and read the temperature distributions and variations in the samples. We also recorded infrared thermal images of the samples' infrared stealth on a human hand or a cell phone.

3. Results and discussion

3.1. Microstructure and characterization of HOGF, HPC and OHPC

Different from the common aerogel packaging, here, HOGF was used as the large framework and the PI/CNTs aerogel as the small framework to synthesize a 3D packaging network with high thermal conductivity. This unique design wraps the OD molecules like a net, effectively preventing their leakage. The oriented graphite layer not only serves as a large framework to improve the mechanical stability of traditional aerogels, but also provides high overall thermal conductivity, endowing it with excellent thermal management capabilities. Among them, the small cross-linked network formed by the filled PI/CNTs aerogel further mitigates the leakage problem of OD molecules at high temperatures, and the filling of CNTs also increases the thermal conductivity in the vertical direction of OHPC, making its application more diversified.

Fig. 2 shows the structural characteristics of the HOGF before and after acid treatment modification. Through the cross-scale analysis in Fig. 2a–c, it can be seen that the material thickness has increased from the initial 3.11 mm to 13.88 mm, achieving a significant volume expansion rate of 446%, but its lateral dimensions have not undergone lateral deformation. This type of directional expansion behavior, combined with the porous structure and layered arrangement, enables it to maintain the advantage of low mass density (0.28 g cm⁻³) while achieving anisotropic structural regulation. The thickness variation of graphite paper in the acid mixture is shown in Fig. S1. With the increase of time, the graphite paper becomes thicker and thicker. However, after 60 minutes, the excessive expansion of the graphite paper leads to structural delamination. Therefore, the expansion time needs to be controlled. Fig. 2d–f show the microscopic morphology differences between graphite paper and the HOGF under different expansion degrees. The original graphite paper presents a dense stacked morphology, while the HOGF shows a loose layered structure, with the interlayer spacing showing a gradient expansion feature with the increase of the expansion rate. Whether viewed from a macro perspective (Fig. 2c) or a micro perspective (Fig. 2d–f), the orientation structure of the graphite layers is clearly observable. As the expansion rate increases, the degree of orientation decreases, which may be attributed to deformation of the graphite layers caused by expansion forces. Notably, even under expansion stress, the graphite layers maintain structural integrity. Compared to porous systems, the layered structure of the HOGF provides an ideal pathway for the uniform penetration of PI/CNTs, thereby facilitating the construction of a stable structure. Fig. 2g–i show the TEM images of the graphite sheet in the HOGF. The (002) crystal plane spacing of the HOGF is 0.406 nm, which is significantly larger than that of natural graphite at 0.335 nm. The inherent properties of graphite crystals can be clearly identified through electron diffraction combined with hexagonal crystal system matching analysis. Further XRD tests indicated that the crystal defects produced by acid treatment were effectively retained (Fig. 2j). It is worth noting that after acid–base activation treatment, the intensity of the characteristic peaks on the (002) crystal plane was significantly enhanced. This structural optimization might be due to the effective removal of interlayer impurities and the restoration of lattice distortion. The Raman analysis results in Fig. 2k further confirm this situation. There is a distinct D peak on the graphite paper, which disappears in the HOGF. Based on this, it can be inferred that low-temperature acid treatment can wash away impurities on the surface of the graphite layer without affecting the crystallinity and damaging the overall microstructure of the graphite layer.⁴⁶

Fig. 2 The thickness of the GP (a) before and (b) after expansion; (c) comparison of GP before and after expansion; (d)–(f) the morphology of the fracture surface of the HOGF with different expansion rates; (g) TEM image and (h) HR-TEM image of the HOGF, where the lattice plane corresponds to the (002) plane; (i) the corresponding diffraction spectrum of the graphite sheets; (j) XRD patterns and (k) Raman results of GP and the HOGF.

The morphology of the HOGF/PI/CNTs composite aerogel is shown in Fig. 3a. It can be clearly seen that there are irregular porous aerogels mixed between the expanded graphite sheets. Among them, the laminates of HOGF provide phonon transport channels, enhancing the planar thermal conductivity. The mixed aerogel channels help improve the vertical thermal conductivity and mechanical properties of the material, while at the same time offering the possibility for subsequent impregnation and coating of phase change materials. Fig. 3b–d show the scanning electron microscopy images of OHPC-x, from which it can be clearly observed that OD is completely impregnated and uniformly dispersed within the skeleton of the aerogel. This result indicates that there is excellent compatibility between OD molecules and the aerogel, which not only enhances the overall stability of the CPCMs, but also greatly improves its heat storage and release efficiency. In addition, due to the pore space filled in the aerogel, the capillary effect and hydrogen bonding effect ensure that the composite phase change material will not leak easily even in the molten state, thus guaranteeing the safety and reliability of the material during use. Fig. 3e and f show the XRD test of the CPCM. The PAS composite CNT aerogel has diffraction peaks at around 25° and 43°, corresponding to the (002) and (100) crystal planes of CNTs (Fig. S2). With the increase of CNT content, the diffraction peaks slightly shift to the left. This might be due to the introduction of CNTs causing changes in the lattice spacing, which in turn affects the position of the diffraction peaks. After imination, the PI composite CNT aerogel still showed diffraction peaks at around 25° and 43°, indicating that the imination process did not change the main crystal structure of the material.⁵¹

Fig. 3 (a) SEM image of the HOGF/PI/CNTs-1 aerogel; (b)–(d) SEM images of OHPC-x; (e) and (f) XRD patterns of the composite aerogel before and after imination; (g)–(i) FT-IR spectra of the composite aerogel before and after imination and OHPC.

The FT-IR test results are shown in Fig. 3g–i and Fig. S3. The peaks near 3726 and 3433 cm⁻¹ in the infrared spectra of PAS series aerogels are related to the stretching vibration of hydroxyl groups (–OH), the peak at 1633 cm⁻¹ corresponds to the stretching vibration of carbonyl groups (C=O) in amide bonds, and the peaks at 1497 and 1386 cm⁻¹ are related to the bending vibration of methylene groups (–CH₂). The peak near 1038 cm⁻¹ might result from the stretching vibration of the C–O bond, while the peaks at 705, 567 and 490 cm⁻¹ correspond to the vibrations of the carbon–carbon single bond (C–C).⁵² After imination, the characteristic peaks of the PI/CNTs aerogel showed obvious changes. This is mainly due to the transformation of amide bonds during the imination process, which leads to a change in the vibration mode of the relevant functional groups. The infrared spectra of the OHPC series composite phase change materials show peaks similar to those of pure OD, indicating that OD maintains its original functional group structure in the composite materials. However, due to the interaction between OD and the HOGF/PI/CNTs aerogel, some characteristic peaks have undergone subtle shifts in position or intensity changes.

3.2. Thermal properties of OHPC

The phase transition thermophysical properties of complex aerogel/OD composites were investigated by DSC. From the obtained DSC thermograms, it can be seen that the pure OD and its composites exhibit single-peak absorptive and exothermic behaviors during the heating and cooling processes, respectively. The DSC curve in Fig. 4 clearly presents the phase transition peak shapes during heating and cooling, corresponding, respectively, to the solid–liquid phase transition and liquid–solid phase transition processes of the samples. By comparing it with the DSC curve of pure OD, it can be seen that all CPCM samples exhibit similar peak shape characteristics on the heating and cooling curves, which further confirms that the aerogel and OD have achieved effective physical recombination.⁵³Fig. 4c and d show the data of the loading rate and enthalpy value of the phase change material obtained by the DSC test. Among them, the enthalpy value of the OD/HOGF composite phase change material encapsulated only by the HOGF large frame is only 74.16 J g⁻¹, and the loading rate is only 35.6%. This indicates that the HOGF alone cannot encapsulate OD, and leakage will occur during heating. Therefore, it is necessary to fill secondary frames in it. The enthalpy value of the OHPC-2 sample was up to 164.46 J g⁻¹, showing a loading rate as high as 78.94% (Table S2). The significant improvement of this value indicates that more phase change materials are effectively encapsulated within the aerogel. The high loading rate not only increases the energy storage density of the material, but also enhances its phase transformation performance, enabling a more superior performance of the material in thermal energy storage and release. Furthermore, with the increase of CNT content, both the melting enthalpy and crystallization enthalpy of the OHPC series phase change materials show a significant upward trend. This might be because CNTs, as an efficient thermally conductive material, enhance the heat transfer efficiency of the composite material and promote its phase transition process. In addition, the introduction of CNTs improves the dispersion and compatibility of OD in the aerogel and promotes the ordered arrangement and crystallization growth of OD molecules, thereby increasing the enthalpy of melting and crystallization. This indicates that, during the heat absorption and release processes, the OHPC series phase change materials have a higher energy storage and release capacity, and can effectively absorb and release a large amount of latent heat, thus showing broad application prospects in the field of thermal energy management.

Fig. 4 For OD and OHPC, (a) DSC heating curves, (b) DSC cooling curves, (c) enthalpy values, and (d) loading rates; (e) DSC diagram of 100 heating and cooling cycles of OHPC-1; and (f) FT-IR spectra before and after cycling.

To delve deeper into the thermal reliability and stability of the CPCMs, DSC was used to investigate the composite during a heating/cooling cycle test within the temperature range of 0–50 °C. As depicted in Fig. 4e, even after undergoing up to 100 heating and cooling cycles, the melting and crystallization curves of the composite remained stable, showing no substantial changes. Fig. 4f shows the FT-IR spectra of OHPC-1 before and after cycling. Neither the peak shape nor the peak value has changed significantly, indicating that the structure of OHPC-1 has not changed before and after cycling. The experimental data conclusively demonstrate that OHPC-1 exhibits exceptional thermal cycling stability, thereby providing robust empirical support for its sustained operational integrity and reliability in practical implementation scenarios (Table 1).

Table 1 Phase transition parameters of OD and OHPC-x

Samples	Melting process		Cooling process
	T m (°C)	ΔHm (J g^−1)	T C (°C)	ΔHC (J g^−1)
OD	28.06	208.34	27.48	213.98
OD/HOGF	27.56	74.16	28.68	73.11
OHPC-0.5	27.22	121.49	28.94	121.15
OHPC-1	27.56	143.60	29.09	141.89
OHPC-2	27.19	164.46	29.00	166.15

---

Fig. 5a shows the melting temperatures of pure OD and OHPC. The melting temperature of pure OD is 28.06 °C. When it is compounded with the aerogel, the melting temperature decreases slightly. This change might be due to the interaction between OD and the aerogel matrix, which leads to partial limitation of the OD molecular chain and thereby affects its melting behavior. With the addition of CNTs, the melting temperature of the OHPC-x series shows a trend of rising first and then falling. The melting temperature of OHPC-0.5 is 27.22 °C, which is similar to that of pure OD. This might be because a small amount of CNTs has a relatively small influence on the melting behavior of OD. However, with the further increase of CNT content, the melting temperatures of OHPC-1 and OHPC-2 were 27.46 °C and 27.19 °C, respectively. The addition of CNTs can change the microstructure and thermal conductivity of the composite materials. However, excessive addition of CNTs can lead to agglomeration or uneven distribution, thereby having a significant impact on the melting behavior of OD.

Fig. 5 (a) Melting points of OD and CPCMs; (b) leakage test images of OD and CPCMs; (c) TG and (d) DTG of OHPC.

The samples were subjected to heating tests to evaluate their shape stability, as shown in Fig. 5b. At room temperature, all samples maintained a stable form. However, as the temperature of the heating plate gradually rises, the state changes of the samples show significant differences. Pure OD undergoes a significant phase transformation from solid to liquid after being heated to its melting point. This process is accompanied by the dissociation of OD molecular chains and the enhancement of fluidity, causing the sample to gradually lose its original solid shape and eventually melt completely. In contrast, the composite phase change material with the aerogel as the matrix demonstrated excellent shape stability during the heating process. Even when heated to a temperature far above its melting point, it can still maintain its initial form without obvious shape changes or leakage. This result indicates that the aerogel matrix not only provides a stable supporting structure for OD, but also effectively restricts the fluidity of OD molecules through its unique pore structure, thereby achieving shape stability at high temperatures.

The thermal stability and degradation process of the material were analyzed through thermogravimetric testing (Fig. 5c and d). The OHPC series materials all exhibit the characteristic of one-time thermal degradation. As a linear polymer, OD is prone to chain decomposition at high temperatures and almost completely decomposes at high temperatures, showing relatively poor thermal stability. However, when OD is compounded with the HOGF/PI/CNTs aerogel, the thermal stability of the composite material has been improved to a certain extent. CNTs, as a kind of nanomaterial with high thermal stability, can maintain the structural integrity at high temperatures, thereby inhibiting the thermal degradation of OD to a certain extent. When the mass ratio of carbon nanotubes is 0.5 wt%, the mass retention rate of the composite material is approximately 45%. However, when the CNT content was further increased to 1 wt% and 2 wt%, the mass retention rate decreased to about 42%. This might be due to the poor dispersion of CNTs within the composite material or the enhanced interaction with other components, which affects the overall structural stability of the composite material, resulting in a slight decrease in the thermal stability of the composite material.

3.3. Thermal conductivity of OHPC

In addition to leakage issues, thermal conductivity efficiency is another critical factor influencing the energy storage efficiency and temperature control performance of PCMs. This study selected the HOGF as the primary thermal conductivity framework, primarily due to its light weight, low cost, ease of mass production, and highly ordered layered structure, making it a highly promising thermal conductivity material. The introduction of PI/CNTs as the secondary network not only enhances the loading capacity of the PCMs but also has a certain promotional effect on its thermal conductivity. The in-plane and out-of-plane thermal conductivities of OHPC-x were experimentally measured, with the results shown in Fig. 6a, b and Table S3. The analysis shows that as the CNT loading increases, both in-plane and out-of-plane thermal conductivities exhibit an upward trend. When the CNT loading reaches 2%, the in-plane thermal conductivity reaches 5.8913 W m⁻¹ K⁻¹, and the out-of-plane thermal conductivity is 2.2934 W m⁻¹ K⁻¹. Further research revealed that the content of CNTs has a relatively minor effect on the in-plane thermal conductivity of OHPC, while it has a more significant impact on the out-of-plane thermal conductivity. This phenomenon can be attributed to the highly oriented graphite layer structure, which acts as a “high-speed channel” for phonon propagation, enabling rapid heat transfer and effectively enhancing the in-plane thermal conductivity of the OHPC composite material (Fig. S4). However, in the direction perpendicular to the HOGF layers, due to the absence of a continuous layer network, the phonon transmission channels are blocked and phonon scattering occurs between the graphite layers, resulting in a relatively low thermal conductivity. However, the addition of CNT fillers has significantly improved this. As thermally conductive fillers, CNTs can reduce the interfacial thermal resistance between PI and PCMs, expand the heat exchange area, and thus enhance the heat transfer efficiency. Secondly, it can enhance the mutual adsorption between the matrix material and PCMs. This is conducive to reducing phonon scattering, thereby improving the thermal conductivity. Therefore, although the out-of-plane thermal conductivity is lower than the in-plane thermal conductivity, it is still higher than that of most thermally conductive materials. This kind of material with high thermal conductivity and bidirectional thermal conductivity is of great significance for the rapid heat exchange of electronic devices.

Fig. 6 (a) In-plane and (b) through-plane thermal conductivity of OHPC; (c) finite element analysis of heat transfer in the (i) horizontal and (ii) vertical directions.

Finite element simulation was carried out using COMSOL Multiphysics® simulation software to simulate the heat flow and temperature distribution of OHPC composites during the heating process in two different directions. Transient finite element simulation was used to simulate heat flow and temperature distribution. The outer boundary was set to be thermally insulated, with an initial temperature of 20 °C and a bottom surface temperature of 70 °C. As shown in Fig. 6c, in the direction perpendicular to the plane (model (i)), heat diffusion is slow due to excessive phonon scattering within the composite material and the lack of effective heat conduction pathways. However, in the direction within the plane of the OHPC (model (ii)), heat diffusion from the bottom to the top of the sample is significantly faster. It only takes about 1 minute for heat to be conducted to the upper surface of the material, thus proving the excellent heat conduction ability of the HOGF.

3.4. Photothermal conversion of OHPC

Light absorption properties are some of the important factors affecting photothermal conversion. The absorption spectra of the substrate materials at various wavelengths of the sun were measured using a UV-visible near-infrared spectrophotometer (Fig. 7a). Pure OD showed limited light absorption behavior, whereas OHPC-x with the addition of the HOGF and CNTs exhibited light absorption values greater than 50% in the 200–800 nm band. The photothermal conversion efficiency and thermal energy storage capacity of the composite phase change materials were tested using a xenon lamp with an intensity of 100 mW cm⁻² to simulate sunlight in order to evaluate the photothermal conversion behavior of the composites (Fig. 7b). The transient temperature changes of the samples during solar irradiation were recorded using an infrared pyrometer recorder with thermocouples, and the recorded temperature–time curves are shown in Fig. 7c. It can be seen that the heating rates and maximum temperatures of all OHPC-x are higher than those of pure OD, which suggests that they have a stronger light and heat absorption capacity. Pure OD can only melt slowly into the liquid state under sunlight irradiation due to its white appearance, poor light-absorbing ability and low phase transition temperature. The OHPC-x series samples have excellent photothermal conversion ability, and under simulated sunlight irradiation, the temperatures of the composites are first rapidly elevated followed by a temperature plateau around 30–35 °C, which is consistent with the T_m and T_c of OD. This is consistent with the T_m and T_c of OD, which indicates that the temperature evolution of OHPC-x composites is delayed by the absorption of thermal energy and release of latent heat generated by the melting and crystallization of OD. CPCMs undergo the solid–liquid phase transition process, and the composites can efficiently store the thermal energy generated from the photothermal conversion in the form of latent heat in their internal storage, which provides a strong guarantee for the efficient utilization and long-term storage of thermal energy. After the phase transition is completed, the temperature of OHPC-x rises again and finally reaches the constant temperatures of 58 °C, 56.5 °C and 54.9 °C, respectively. Under the simulated condition of sunlight being turned off, the temperature of the sample dropped significantly due to the loss of energy supply. When the temperature drops to the phase transition temperature, the energy stored in the composite material begins to be released, which enables the OHPC-x series samples to maintain a temperature of around 30 °C for approximately 200 seconds, demonstrating excellent energy storage and release performance. In addition, the temperature of OHPC-2 containing 2 wt% CNTs was slightly higher than that of the other samples, which was attributed to the fact that CNTs are an efficient light trapping agent with good light absorption ability.

Fig. 7 (a) UV-Vis-NIR spectra of OHPC-x; (b) photothermal test simulations; (c) photothermal conversion curves and (d) photothermal conversion efficiencies of CPCMs, (e) temperature–time curves and (f) Fourier transform infrared spectra of OHPC-2 after 20 photothermal cycles.

Fig. 7d shows the photothermal conversion efficiency (η) of all samples:

Here, A is the irradiated area of the sample, m is the mass of the sample, ΔH is the melting enthalpy, P is the simulated sunlight intensity, and Δt is the phase transition time. The photothermal conversion efficiencies of the OHPC-x series samples were calculated to be 81.2%, 85.7% and 86.9%, respectively (Table S4). Therefore, 2 wt% of CNTs is the optimal filling amount, and its photothermal conversion efficiency is better than that of samples with other ratios. The above findings indicate that the photothermal conversion efficiency of OHPC-2 is significantly improved and the heat energy transfer is more effective, which can convert light energy into heat energy more rapidly and efficiently, providing strong support for the efficient utilization of heat energy. Compared with the work published in the past five years, the OHPC-2 sample prepared in this study has significant comprehensive advantages, with higher heat storage density and photothermal conversion efficiency than those previously reported in the literature, showing higher latent heat capacity and excellent photothermal conversion efficiency.

The photothermal cycling stability of OHPC-2 composites was evaluated by subjecting the composites to 20 consecutive simulated solar irradiation–room temperature cooling cycles. Fig. 7e shows the photothermal cycling curves of OHPC-2, and the temperature profiles exhibit a high degree of consistency, which indicates that it has excellent photothermal cycling stability performance. In addition, by performing FT-IR tests on the samples before and after photothermal cycling, the test results showed that the chemical structure of the samples remained unchanged, and the main characteristic peaks also remained unchanged, which further confirmed its excellent photothermal cycling stability (Fig. 7f). Therefore, the OHPC-2 sample prepared in this study has excellent cycling stability and photothermal conversion performance. Compared with the research results published in the past five years, the OHPC prepared in this study has significant comprehensive advantages. Its heat storage density, thermal conductivity and photothermal conversion rate are all higher than those reported in most previous studies. Particularly, OHPC-2 has a high in-plane thermal conductivity and excellent photothermal conversion efficiency, which proves that it has great potential in thermal management (Table 2).

Table 2 Comparison of the thermal performance of OHPC-2 in the present work with other CPCMs in the literature

CPCM	Thermal conductivity (W m^−1 K^−1)	η (%)	ΔHm (J g^−1)	T m (°C)	Ref.
C/RGO@CoNC/PW	0.96	95.1	209.2	42	32
PI/MXene aerogel/PEG	—	87.8	177.1	62.5	48
PI/PPy-CNTs@PEG	0.72	92.9	148.2	64.3	54
PI-BN/PEG	5.34	—	125.2	29	55
C-PI-GNP/PEG	7.032	40.65	136.4	67.1	56
PI/XG/TiO2/PEG	0.82	94.23	160.38	61.24	57
C-PI/KNF@PEG/CoFe2O4@PPy	0.406	84.8	150.1	55.2	58
PI/PR/PEG	0.3469	82.5	174.4	63.3	59
NF/UMCC-HA-ODA60%	1.094	—	110.41	27.89	60
MXene/ADP-modified DWs/OD	—	65.5	191.6	33.4	61
OHPC-2	5.8913 (in-plane orientation)	86.9	164.46	27.19	This work
	2.2934 (axial orientation)

---

3.5. Applications of OHPC

Experiments simulating outdoor natural sunlight exposure were conducted on the developed composite materials to study their solar–thermal energy conversion and storage performance in practical applications. As shown in Fig. 8a, a wooden house model is adopted, and the roof is covered with the OHPC-x composite material. Then, the house model was exposed to 100 mW cm⁻² of simulated sunlight for 20 minutes, and then the xenon lamp was turned off for natural cooling. It can be observed from the temperature–time evolution curve in Fig. 8b that after 30 minutes of simulated sunlight exposure, the surface temperature of the OHPC-2 composite material was the highest among the three samples, at 58 °C. This can be attributed to its outstanding light absorption capacity and thermal energy storage capacity. Furthermore, in the absence of sunlight, OHPC-0.5 exhibits a faster solar–thermal energy storage/release process due to the less encapsulated phase change material and lower enthalpy value. In contrast, OHPC-2 has a higher OD loading and enthalpy value, remaining at around 33 °C after 20 minutes without sunlight. It can be seen from the infrared image in Fig. 8c that under simulated sunlight irradiation, the surface temperature changes of OHPC-2 and OHPC-1 are faster than those of OHPC-0.5. There is no doubt that the presence of a high content of CNTs enables OHPC to have a higher light absorption capacity, thus showing a faster temperature evolution under solar illumination. Moreover, the presence of PCMs delays its heat release, allowing for better thermal management of objects. These results further confirm the outstanding solar–thermal energy conversion/storage capabilities of OHPC-x composites in the collection and utilization of solar thermal energy.

Fig. 8 (a) Light simulation of a house; (b) temperature–time profiles of the composite material in the presence and absence of light; (c) infrared thermography images of a model of a wooden house covered with OHPC-x in the presence and absence of light; temperature-transformed infrared imaging of OHPC-x (d) along the axial direction and (e) along the planes on a hot stage at 60 °C.

A series of OHPC-x samples of the same thickness and height were placed on a heating plate at 60 °C, and their temperatures were monitored in real time using an infrared thermal imager to obtain the thermal response data. As shown in Fig. 8d, the OHPC-x samples are placed parallel on the hot stage, and the heat is transferred upward along their axial direction. At the beginning, the temperature of all samples was close to room temperature, approximately 28 °C. With the increase of heating time, different samples exhibited different heating rates and stable temperatures. When heated for 120 seconds, the temperature of the OHPC-0.5 composite material increased slowly, reaching only 33.51 °C. This might be due to the relatively low content of CNTs in OHPC-0.5, which has a relatively low thermal conductivity and slow heat transfer. In contrast, the composites with a high content of CNTs showed a faster temperature increase rate. The surface temperatures of OHPC-1 and OHPC-2 reached 43.11 °C and 42.94 °C, respectively. This indicates that the addition of CNTs significantly enhances the axial thermal conductivity of the material. With the extension of heating time, the surface temperature of all samples continued to increase and gradually stabilized. At 240 seconds, all the composite materials reached a relatively high temperature. Fig. 8e shows the infrared imaging of OHPC-x placed vertically on a hot plate, with heat conducted along the HOGF plane. When heated for 30 seconds, the fastest increase was observed in the temperature of OHPC-2. It can be seen that the temperature changes of all samples are faster than the axial conduction. Within just 90 seconds, the inner surface temperature has approached the temperature of the hot plate, which also corresponds to the previous thermal conductivity. This indicates that when conducting along the HOGF plane, CNTs have a relatively small influence on its thermal conductivity. The above results indicate that the introduction of CNTs can significantly enhance the axial thermal conductivity of the composite phase change material. The addition of the HOGF endows OHPC with high thermal conductivity in the planar direction, enabling the material to respond more quickly to external heat sources. This phenomenon is of great significance for improving the application performance of composite phase change materials in the fields of thermal energy storage and temperature regulation.

4. Conclusions

In this study, a dual-encapsulation framework was designed by ingeniously filling PI/CNTs between HOGF sheets, and oriented CPCMs with high thermal conductivity and high photothermal conversion were successfully fabricated through freeze-drying and vacuum impregnation. The OD loading of OHPC-2 is as high as 78.94%, and its enthalpy value reaches 164.46 J g⁻¹, which has great application potential in the fields of energy storage and conversion. Furthermore, the presence of the HOGF endows OHPC with a relatively high in-plane thermal conductivity. The in-plane thermal conductivity of OHPC-2 is as high as 5.8913 W m⁻¹ K⁻¹. The incorporation of CNTs enhances the axial thermal conductivity and photothermal conversion efficiency of the composite material. The axial thermal conductivity of OHPC-2 can reach 2.2934 W m⁻¹ K⁻¹, and the photothermal conversion efficiency can also reach 86.9%. When OHPC was vertically placed on a hot plate at 60 °C, all samples could transfer heat to the surface within 90 seconds, indicating that OHPC has excellent heat transfer and dissipation capabilities and holds potential application prospects in the thermal management of electronic devices. In addition, the prepared OHPC also has excellent thermal and photothermal cycling capabilities, maintaining a high enthalpy and structure even after 100 consecutive heating and cooling cycles and 20 photothermal cycles. This work overcomes the limitations of traditional PCMs and provides new ideas for solar energy collection and thermal management technologies.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability

The data supporting the results of this study have been included as part of the supplementary information (SI) and additional data are available from the corresponding author upon reasonable request. Supplementary information (the expansion process of graphite paper, comparative infrared images of various materials, and some tabular data) is available. See DOI: https://doi.org/10.1039/d5nh00543d.

Acknowledgements

This work was financially supported by the Fundamental Research Funds for the Central Universities (No. FRF-KST-25-001) and the Natural Science Foundation of Beijing Municipality (No. L253029).

References

1. L. Lai, P. Liu, C. Zhang, D. Zhou, Q. Li and S. Fan, Nanoscale Horiz., 2025, 10, 1446–1452.
2. Y. Meng, Y. Liu, Z. Wan, Y. Huan, Q. Guo, D. Fan, X. Zhou, J. Liu, Y. Cao, X. Cao, Z. Gu, T. Qian and C. Yan, Chem. Eng. J., 2023, 453, 139967.
3. Y. Tian, R. Yang, H. Pan, N. Zheng and X. Huang, Nano Energy, 2025, 133, 110440.
4. R. Li, X. Sun, R. Zhang, S. Wang, Q. Liu, P. Bai, Y. Yuan and Y. Li, Chem. Eng. J., 2025, 504, 158964.
5. J. Tong, Z. Liu, Y. Liu, X. Huang and G. Wang, Adv. Funct. Mater., 2025, 35, 2411744.
6. P. Guo, W. Gu, J. Lu, K. Zhang, F. Liu, H. Liu, N. Sheng and C. Zhu, Chem. Eng. J., 2025, 504, 158832.
7. H. B. Ghadim, A. Godin, A. Veillere, M. Duquesne and D. Haillot, Renewable Sustainable Energy Rev., 2025, 208, 115039.
8. Q. Du, M. Yang, H. Sun, M. Li, J. Zhao, D. Li and D. Sun, Adv. Funct. Mater., 2025, 35, 2500131.
9. J. Yang, S. Wang, J. Zhao, S. Wang, S. Wang, Y. Peng, Q. Huang and C. Liu, Chem. Eng. J., 2025, 511, 162039.
10. S. Song, R. Wang, L. Lv, W. Zhu, R. Feng and L. Dong, Composites, Part B, 2025, 305, 112721.
11. D. Banswar, J. K. Anand, S. A. Bukhari, S. Singh, R. Prajesh, H. Kumar, S. K. Makineni and A. Goswami, Nanoscale Horiz., 2025, 10, 549–560.
12. Y. Cao, P. Lian, Y. Chen, L. Zhang and X. Sheng, Sol. Energy Mater. Sol. Cells, 2024, 268, 112747.
13. X. Zhang, Y. Yang, C. Zhong, D. Xiang, H. Sun, D. Li, G. Yan and Y. Wu, Small, 2025, 21, 202500683.
14. Q. Zhang, Y. Li, C. Liu, X. Wu, X. Zhang, J. Song, Y. Mao and K. Yuan, J. Mater. Chem. A, 2024, 12, 32526–32547.
15. J. Shen, Y. Ma, F. Zhou, X. Sheng and Y. Chen, J. Mater. Sci. Technol., 2024, 191, 199–208.
16. L. Hu, L. Zhang, W. Cui, Q. An, T. Ma, Q. Wang and L. Mai, J. Mater. Sci. Technol., 2025, 210, 204–226.
17. H. Li, C. Hu, Y. He, Z. Sun, Z. Yin and D. Tang, Matter, 2022, 5, 3225–3259.
18. Q. Liang, H. Zhang, Y. Li, X. Zhang and D. Pan, J. Energy Storage, 2024, 83, 110621.
19. J. Koo, J. Hyeong, S. Kim, M. Rim, C. Sung, S. Seo, S.-E. Kim, S. Kim, D.-Y. Kim and K.-U. Jeong, Chem. Eng. J., 2025, 510, 161680.
20. Y. Cao, Y. Meng, Y. Jiang, S. Qian, D. Fan, X. Zhou, Y. Qian, S. Lin, T. Qian and Q. Pan, Chem. Eng. J., 2022, 433, 134549.
21. Y. Huan, M. Gu, Y. Ni, H. Xue, H. Zhu, Y. Zhu, Q. Guo, D. Fan, X. Zhou, J. Liu, Y. Cao, Y. Lu, C. Yan and T. Qian, Chem. Eng. J., 2023, 468, 143742.
22. R. Yang, N. Zheng, Z. Yu, F. Zhang, H. Gai, J. Chen and X. Huang, Appl. Therm. Eng., 2023, 230, 120808.
23. Y. Li, X. Huang, J. Lv, F. Wang, S. Jiang and G. Wang, Composites, Part B, 2022, 234, 109735.
24. P. Liu, M. Huang, X. Chen, Y. Gao, Y. Li, C. Dong and G. Wang, Interdiscip. Mater., 2023, 2, 423–433.
25. M. Aramesh and B. Shabani, Renewable Sustainable Energy Rev., 2022, 155, 111919.
26. S. Wang, Q. Huang, Z. Sun, Y. Wang, T. Liang, B. Wang, C. Fan and C. Liu, Carbon, 2024, 226, 119174.
27. Q. Liang, D. Pan and X. Zhang, Chem. Eng. J., 2023, 453, 139441.
28. Y. Tian, N. Zheng, Z. Tao, J. Tong, T. Yuan and X. Huang, Carbon Neutraliz., 2025, 4, e70040.
29. T. L. Wong, C. Vallés, A. Nasser and C. Abeykoon, Renewable Sustainable Energy Rev., 2023, 187, 113730.
30. Y. Gao, X. Chen, X. Jin, C. Zhang, X. Zhang, X. Liu, Y. Li, Y. Li, J. Lin, H. Gao and G. Wang, eScience, 2024, 4, 100292.
31. Y. Ma, M. Zou, W. Chen, W. Luo, X. Hu, S. Xiao, L. Luo, X. Jiang and Q. Li, Appl. Energy, 2023, 349, 121658.
32. T. Shi, X. Gao, H. Liu and X. Wang, Nano-Micro Lett., 2025, 17, 236.
33. Y. Wu, C. An, Y. Guo, Y. Zong, N. Jiang, Q. Zheng and Z.-Z. Yu, Nano-Micro Lett., 2024, 16, 118.
34. X. Shi, Y. Meng, R. Bi, Z. Wan, Y. Zhu and O. J. Rojas, Composites, Part B, 2022, 245, 110231.
35. G. Zhou, L. Li, S.-Y. Lee, F. Zhang, J. Xie, B. Ye, W. Geng, K. Xiao, J.-H. Lee, S.-J. Park, Z. Yang, C. Huang and Y. Zhang, Compos. Sci. Technol., 2023, 243, 110256.
36. T. Hong, J. Xue, H. Li, Y. Wen, X. Gao, D. Liu and Q. Zheng, Chem. Eng. J., 2025, 519, 164747.
37. Z. Tu, Z. Xie, Q. Zhou, J. He, Y. Dong, C. Huang, F. Zhang, Q. Zheng and Y. Zhang, Composites, Part A, 2025, 199, 109178.
38. F. Zhang, L. Guo, Y. Shi, Z. Jin, Y. Cheng, Z. Zhang, C. Li, Y. Zhang, C. H. Wang, W. Feng and Q. Zheng, Chem. Eng. J., 2023, 452, 139664.
39. C. Lucchesi, R. Vaillon and P.-O. Chapuis, Nanoscale Horiz., 2021, 6, 201–208.
40. W. Zheng, Z. Liu, G. Xi, T. Liu, D. Wang, L. Wang and W. Liao, Nanoscale Horiz., 2025, 10, 1054–1076.
41. J. Hu, Z. Yao, A. Chen, C. Xiao, G. Zhang and X. Yang, Appl. Therm. Eng., 2025, 265, 125566.
42. J. Gao, B. Zhou, C. Liu, C. He, Y. Feng and C. Liu, Chem. Eng. J., 2023, 475, 146087.
43. Y. Li, N. Zheng, Y. Ren, X. Yang, H. Pan, Z. Chai, L. Xu and X. Huang, Ceram. Int., 2024, 50, 18923–18931.
44. N. Zheng, H. Pan, Z. Chai, Z. Liu, F. Gao, G. Wang and X. Huang, ChemSusChem, 2024, 17, e202301971.
45. J. Li, X. Hu, C. Zhang, W. Luo and X. Jiang, Renewable Energy, 2021, 178, 118–127.
46. F. Zhang, D. Ren, Y. Zhang, L. Huang, Y. Sun, W. Wang, Q. Zhang, W. Feng and Q. Zheng, Chem. Eng. J., 2022, 431, 134102.
47. L. Zhang, X. Li, E. Ding, Z. Guo, C. Luo, H. Zhang and J. Yu, Compos. Sci. Technol., 2024, 251, 110588.
48. Z. Zheng, H. Liu, D. Wu and X. Wang, Chem. Eng. J., 2022, 440, 135862.
49. J. Jing, H. Liu and X. Wang, Adv. Funct. Mater., 2024, 34, 2309269.
50. X. Zhang, X. Zhao, T. Xue, F. Yang, W. Fan and T. Liu, Chem. Eng. J., 2020, 385, 123963.
51. H. Zheng, X. Lu and K. He, J. Energy Chem., 2022, 68, 454–493.
52. T. Shi, Z. Zheng, H. Liu, D. Wu and X. Wang, Compos. Sci. Technol., 2022, 217, 109127.
53. G. Zhu, M. Zou, W. Luo, Y. Huang, W. Chen, X. Hu, X. Jiang and Q. Li, Chem. Eng. J., 2024, 488, 150930.
54. T. Shi, J. Jing, Z. Qian, G. Wu, G. Tian, H. Liu and X. Wang, Adv. Sci., 2025, 12, 2411758.
55. Q. Zhang, T. Xue, J. Tian, Y. Yang, W. Fan and T. Liu, Compos. Sci. Technol., 2022, 226, 109541.
56. M. Su, G. Han, J. Gao, Y. Feng, C. He, J. Ma, C. Liu and C. Shen, Chem. Eng. J., 2022, 427, 131665.
57. Z. Sun, H. Zhang, Q. Zhang, R. Jing, B. Wu, F. Xu, L. Sun, Y. Xia, F. Rosei, H. Peng and X. Lin, J. Mater. Chem. A, 2022, 10, 13556–13569.
58. T. Shi, H. Liu and X. Wang, ACS Appl. Mater. Interfaces, 2024, 16, 10180–10195.
59. Z. Zheng, T. Shi, H. Liu, D. Wu and X. Wang, Appl. Therm. Eng., 2022, 207, 118173.
60. H. Jiang, J. Hu, P. Yang, Y. Qian, T. Ma, R. Wei, Y. Zhang, W. Sun and X. Guo, J. Energy Storage, 2025, 131, 117589.
61. H. Yue, Y. Ou, J. Wang, H. Wang, Z. Du, X. Du and X. Cheng, Energy, 2024, 286, 129441.

---