Microencapsulated phase change material through cellulose nanofibrils stabilized Pickering emulsion templating

Phase change materials (PCMs) possess remarkable capability to store and release substantial amounts of energy during the processes of melting and crystallization across a wide temperature range, thus holding great promise in applications related to temperature regulation and thermal energy storage. Herein, to effectively address PCM leakage and enhance thermal conduction, PCM microcapsules with melamine–formaldehyde resin (MF) shell were prepared using in situ polymerization of Pickering emulsions stabilized by cellulose nanofibrils (CNFs). CNFs were selected as the stabilizers for the Pickering emulsions and as reinforcing nanofillers for the MF shell, owing to their excellent emulsifying capability, high mechanical strength, and sustainable nature. Paraffin wax (PW) was utilized as the PCM material. The resulting PCM microcapsules with MF resin shells and PW core had a diameter ranging from 2 to 4 µm. Results showed that microcapsule with the core–shell ratio of 2 (Micro-2.0) exhibited the highest latent heat of crystallization and latent heat of fusion, measuring approximately 128.40 J/g and 120.23 J/g, respectively. The encapsulation efficiency of Micro-2.0 was determined to be approximately 79.84%.


Introduction
Energy consumption is expected to increase significantly in the coming decades, because of population and economic growth [1][2][3]. To date, most energy is still generated by traditional fossil fuels [4,5]. Therefore, improving energy efficiency and transitioning to green energy sources have become crucial [6][7][8]. As the most abundant energy source available on the earth, solar energy is a widespread green thermal energy resource and is considered to be the most promising solution to the growing energy crisis [9,10]. Additionally, it is worth noting that waste heat is a major source of recoverable energy loss. Thermal energy storage is an effective solution to make full use of waste heat and solar thermal energy, offering great possibilities for reducing fossil fuels consumptions, and alleviating energy crisis [11,12]. Among current thermal energy storage technologies, latent heat energy storage that can provide high-energy densities per unit mass/volume at nearly constant temperatures by changing the phase of a material, usually a phase change material (PCM), has been recognized as one of most promising approaches [13,14]. Of note, PCM are materials that can absorb or release a significant amount of heat during the phase transition while maintaining a constant temperature [15,16]. The heat absorbed or released during this phase transition is known as latent heat [17,18]. PCMs are generally classified into three main categories including organic (e.g., paraffin wax (PW), fatty acids, and esters), inorganic (e.g., salt hydrates, metals, and metal alloys), and eutectic PCMs (e.g., sodium acetate trihydrate and potassium nitrate, or organic compounds like fatty acids combined with alcohols) [19,20]. Almost these PCMs, organic phase change materials, such as PW and fatty acids, are 1 3 149 Page 2 of 9 attractive for latent heat energy storage due to their stable chemical properties, minimal undercooling, wide range of phase transition temperatures, and high heat enthalpy [21,22]. However, the wide application of organic PCMs still face significant challenges such as volume change during phase transition, leakage, and low thermal conductivity [23,24]. One solution to these issues is microencapsulation, which involves enclosing the PCM in microcontainers or capsules [25]. Microencapsulation can effectively prevent leakage during phase transition and increase the specific surface area between PCMs and the matrix, thereby improving thermal conductivity [26,27].
Of note, it is important to ensure that the microcapsules remain stable and uniform over time during the microencapsulation process. To protect phase change materials, polymer shells can be used as encapsulating materials, such as melamine-formaldehyde resin (MF) [28], polyurethane (PU) [29], polystyrene (PS) [30,31], acrylic resin [32], and poly(methyl methacrylate) (PMMA) [33]. These shells can provide a barrier between the PCM and the surrounding environment, protecting it from external factors that may cause it to degrade or lose its thermal properties [19].
Pickering emulsions, a type of emulsion stabilized by solid particles instead of conventional surfactants, can also be used as a promising platform for incorporating and stabilizing PCMs [34,35]. The solid particles adsorb at the oil-water interface, forming a physical barrier that prevents the droplets from coalescing [36,37]. This results in stable emulsions with a long shelf life and increased resistance to shear and coalescence [38]. The use of Pickering emulsions as a platform for PCM stabilization has several advantages over other methods, including improved stability, reduced agglomeration, and increased surface area for heat transfer [39,40]. Moreover, the physical barrier formed by the interfacial solid particles can effectively prevent PCM leakage [41,42]. The solid particles used to stabilize Pickering emulsions can be a wide range of materials including silica, metal oxides, clays, and cellulose nanomaterials [43,44].
Among them, cellulose nanomaterials, for instance, cellulose nanofibers (CNFs), derived from cellulose, have attracted much attention for a wide range of applications including but not limited to those related to Pickering emulsions because of their high specific surface area, easy modification, and good biocompatibility, just to name a few [45][46][47]. In our previous work, FeCl 3 -catalyzed formic acid hydrolysis combined with high-pressure homogenization to produce CNFs from industrial kraft pulp was reported [48]. The surface properties of CNFs, especially the surface wettability, can be easily tuned by changing the hydrolysis time. In addition, the obtained CNFs were able to successfully stabilize Pickering emulsions, and the emulsions showed good stability at various ionic strengths from 0 to 1 M NaCl and high temperature of 80 °C.
Herein, as shown in Fig. 1, various PW/MF microcapsules with MF polymer as the shell and PW as the core were fabricated using CNFs-stabilized Pickering emulsions as the template. The chemical components, morphology, thermal storage performance, and thermal reliability of PW/MF microcapsules were comprehensively studied.

Materials
CNFs suspension with a concentration of around 1 wt% was prepared via FeCl 3 -catalyzed formic acid hydrolysis in combination with high-pressure homogenization method according to our previous work [48]. Sodium chloride (NaCl) was purchased from Aladdin Reagent (Shanghai) Co., Ltd. Paraffin wax (Pathological grade, the melting point is 60 ~ 62 °C) and citric acid (CA) were obtained from Maclin Biochemical (Shanghai) Co., Ltd. Formaldehyde solution was provided by Yantai Sanhe Chemical Reagent Factory Co., Ltd. Melamine and Triethanolamine were received from Tianjin Damao Chemical Reagent Factory. All chemicals were used as received without further purification.

Preparation of CNCs stabilized Pickering emulsions
To prepare CNFs stabilized Pickering emulsions, 0.5 wt% NaCl (0.08 g) was added to the 10 mg/mL CNFs (8 mL) in order to decrease the repulsion between CNFs. Paraffin (2 g) was added into 8 mL of the CNFs suspension and heated to 80 °C until the paraffin completely melted. The molten mixtures were emulsified by a tip sonicator (FB505, Fisher Scientific, USA) for 10 min (3 s on/2 s off) under the power lever of 60%. After ultrasonication, 10 g of deionized water was added to dilute the emulsion and CA solution (0.5 mol/L) was used to adjust the pH of the emulsion to about 5.

Preparation of the PCM microcapsules with CNFs and MF hybrid shell
To synthetize MF prepolymer solution, formaldehyde and melamine were sequentially added in a mass ratio of 1:8 into the flask containing 10 mL of the deionized water in order. The mixtures were stirred at 70 °C while adjusting the pH to around 9 with triethanolamine until a clear solution was obtained. After cooling to ambient temperature, the obtained solution was added to the as-prepared CNFs-stabilized PCM Pickering emulsions (10 mL) drop by drop and stirred at 80 °C with a constant stirring speed of 250 rpm. The initial core (paraffin) to shell (MF) ratio varied from 0.5, 1.0, 1.5 to 2.0. The mixtures of the pH were then adjusted to 5 with citric acid solution. After 3 h, the reaction was stopped, and the PCM microcapsules with CNF/MF shell were collected by suction filtering. The wet filter cake was thoroughly washed with 30 wt% ethanol aqueous solution (40 ~ 50 °C) and then dried in the oven for 24 h. The obtained PCM microcapsules were labeled as Micro-0.5, Micro-1.0, Micro-1.5 and Micro-2.0, respectively.

Characterization
The morphology of the Pickering emulsions was observed using an optical microscope (Leica, DM4000 Germany). A MasterSizer ® 3000 granulometer (Malvern ® , UK) was used to examine the droplet size distribution.
The morphologies of the microcapsule samples were observed by using a scanning electron microscope (JEOL JSM-IT300LV, Japan) at an accelerating voltage of 10.0 kV. A small amount of sample was sprinkled onto conductive tape and sputter coated with gold for 3 ~ 4 times.
The Zeta potential of CNFs suspension was measured using a microscopic electrophoresis device (BTG-Mutek Szp06, Germany). Prior to analysis, the CNFs suspension with a concentration of 0.1 wt% was subjected to ultrasonication at 45 kHz for 10 min. The testing of the CNFs solution was conducted at a pH of 6 and a temperature of 25 °C. To ensure accuracy, all measurements were performed in triplicate, with each measurement consisting of a minimum of 10 runs.
Fourier transform infrared (FTIR) spectra of the samples were recorded using a FTIR-650 spectrometer (Tianjin Gang Dong Sci. and Tech. Development Co., Ltd., China) over the region of 4000 to 400 cm −1 with a resolution of 4 cm −1 . The powder samples were blended with KBr and compressed into a thin pellet for FTIR examination.
The thermal stabilities of PW, MF shell, and PW/MF microcapsules with CNFs was characterized by a TA Instruments-Waters LLC (USA) at a heating rate of 20 °C/min ranged from room temperature to 600 °C. Three to 5 mg of samples were treated with nitrogen (N 2 ) atmosphere at a 50 mL/min flow rate.
The thermal properties of PW, MF shell, and PW/MF microcapsules with CNFs were examined by using a differential scanning calorimetry (Netzsch, DSC214 Polyma, Germany), and measured at a heating-cooling rate of ± 10 °C min −1 in the temperature range of − 10 ~ 80 °C under the N 2 gas with a flow rate of 50 mL/min. The encapsulation efficiency (η) of PCM microcapsules was calculated according to the following Eq. (1) [49]: where the ΔH m0 and ΔH c0 are the melting and solidifying enthalpies of PW, respectively; ΔH m and ΔH c represent the enthalpies of the PCM microcapsules during melting and solidifying, respectively.

Results and discussion
In this work, the CNFs employed to stabilize PCM Pickering emulsions were prepared by FeCl 3 -catalyzed formic acid hydrolysis in combination with high-pressure homogenization method. The resultant CNFs with an average diameter of around 10 nm (Fig. 2a) have a negative zeta potential value of − 20 mV (Fig. 2b). Such nanofibers showed excellent Pickering emulsifying ability because they could be partially wetted by both oil and water phases. Of note, NaCl was added to the CNFs dispersion to lower the electrostatic repulsion between CNFs, thus facilitating the formation of stable Pickering emulsion. Figure 2c shows the optical images of PW Pickering emulsions. Although the emulsion droplets showed slight aggregation, no coalescence was observed. More importantly, even after 3 months of storage at room temperature, no obvious phase separation was observed, indicating the high stability of the CNFs stabilized PW Pickering emulsions. The average diameter of the Pickering emulsions was around 4 µm. The PCM microcapsules with different core-shell ratios were fabricated via in situ polymerization of MF precursor with CNFs-stabilized PW Pickering emulsions as templates. The formula of PCM microcapsules was listed in Table S1. The size of PCM microcapsules was mainly related to the size of Pickering emulsion droplets, while the core-shell ratio has little effect on it. From Fig. 3, it can be seen that the size of the microcapsules with different core-shell ratios is about 2 ~ 4 µm. PCM microcapsules demonstrated nearly spherical shape with dense MF shell. Compared with other samples, some small indentations were observed in Micro-2.0 ( Fig. 3d and h), which may be caused by the volume change of PCM during solidification. CNFs can be clearly observed on the surface of the synthesized microcapsules with MF shells (Fig. 3). It can be found that the microcapsules did not break even under high-vacuum conditions of the SEM, indicating the high mechanical properties of the shells. According to studies previously reported, MF resin incorporated with CNF was expected to have higher mechanical properties than that of pure MF resin. Thus, the microcapsules synthesized in this work are believed to a promising candidate for encapsulating PMs.
The FTIR spectra of the CNFs, PW, MF, and PCM microcapsules are shown in Fig. 4  of ester groups, which is because the fact that FA could react with cellulose through esterification [45,53]. MF displays vibration at 3410, 1300-1500, 996, and 810 cm −1 , which are attributed to the N-H stretching vibrations of a secondary amine, the methylene C-H bending vibration, the C-H out of plane deformations, and bending vibration of triazine ring, respectively [42]. PW shows typical FTIR spectra of alkanes, such as the symmetrical stretching vibration of -CH, -CH 3 , and -CH 2 at 2955, 2915, and 2847 cm −1 , deformation vibration of -CH 2 and -CH 3 at 1461 cm −1 , and rocking vibration of -CH 2 at 719 cm −1 [54]. The PCM microcapsules display the characteristic vibration of PCM, CNFs, and MF, and no new vibration appears, suggesting that there is no chemical reaction between the MF shell and the PCM core materials. The FTIR spectra of the microcapsules contained all the characteristic absorption peaks of the MF shell and the PW core without any other obvious absorption peaks, indicating that the PW had been successfully microencapsulated into the MF shell, and no chemical reaction had occurred between the PW and the MF resin. It is worth note that there is no absorption peak at 1719 cm −1 for the microcapsules, which is due to the hydrolysis of the ester groups from the CNFs due to the polymerization under alkaline conditions. DSC analyses were conducted to evaluate the phase change properties of the PCM microcapsules including melting and solidifying temperatures, and latent heat storage capacities. The DSC scans of bulk PW, MF shell, and PCM microcapsules are shown in Fig. 5a. During both heating and cooling scans, bulk PW exhibits two distinct phase-change peaks. The first peak, occurring within a lower temperature range, signifies the solid-solid transition from an ordered phase to a more disordered rotator phase. The second peak, observed at higher temperature ranges, corresponds to the solid-liquid melting process [55]. Moreover, no peak was observed in the DSC curve of MF shell, suggesting that all the latent heat of PW/MF microcapsules was from the PW core. The thermal properties of PW and PCM microcapsules are summarized in Table 1. The melting and solidifying temperatures of PW are approximately in the range of 46-72 °C and 63-34 °C, respectively. The latent heat of melting and freezing of bulk PW was measured as 156.54 and 154.87 J/g, respectively. Thermal properties such as encapsulation efficiency and phase transition enthalpy significantly affect the performance and energy storage capacity of PCM microcapsules. The encapsulation efficiency and phase enthalpy of PCM microcapsules gradually increased when the core-shell ratio increased (Table 1). When the core-shell ratio is 2, the crystallization and melting latent heat values of Micro-2.0 are the largest, about 128.40 J/g and 120.23 J/g, respectively. The encapsulation efficiency of Micro-2.0 is about 79.84%. The thermal properties of microcapsules containing PCM were examined using TGA. Figure 5b illustrates the weight loss and derivative weight loss curves for PCM, MF shell, and PCM microcapsules. PW demonstrates a characteristic weight loss between 180 and 230 °C due to evaporation. The TGA plot reveals that CNFs exhibits a high pyrolysis temperature at approximately 360 °C. Both PW/MF microcapsules exhibit a similar three-step degradation process. In the first step, the thermal degradation of PW primarily contributes to the weight loss of PW/MF microcapsules. The second step occurs between 200 and 340 °C, where the weight loss of the microcapsules aligns precisely with the PW evaporation in this temperature range. Leakage of PW from the microcapsules leads to a rapid increase in internal pressure, causing the fragile MF shell to crack. The third step corresponds to the pyrolysis of the MF shell. Compared to pure PW, the thermal degradation process of microencapsulated PW is significantly delayed, indicating that the MF shell impedes PW volatilization at high temperatures. This phenomenon can be explained by the effective bonding or chemical linking of appropriate CNFs with MF prepolymers, resulting in composite MF shells with enhanced strength against cracking, destruction, and improved thermal stability [56]. Moreover, increasing the core-shell ratio further enhances thermal stability.

Conclusions
In summary, PW was successfully microencapsulated using the CNF-stabilized Pickering emulsions as the template. CNFs were employed as Pickering emulsifiers due to their remarkable emulsifying ability, surface functionality, renewability, and biodegradability. The MF shell was formed through the in situ polymerization of MF precursors within the Pickering emulsion resulting in PCM microcapsules with strengthened MF shells reinforced by CNFs. Microcapsules with different core/shell ratios were successfully prepared. Among these microcapsules, Micro-2.0 exhibited the highest latent heat of crystallization and latent heat of fusion, measuring approximately 128.40 J/g and 120.23 J/g, respectively. The encapsulation efficiency of Micro-2.0 was determined to be approximately 79.84%. This work provides a simple yet effective strategy for PCM microencapsulation and could give inspirations for the design of novel thermal energy storage systems.