Scholarly article on topic 'Cellulose nanofibers enable paraffin encapsulation and the formation of stable thermal regulation nanocomposites'

Cellulose nanofibers enable paraffin encapsulation and the formation of stable thermal regulation nanocomposites Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Nano Energy
OECD Field of science
Keywords
{Nanocellulose / "Phase change materials" / Encapsulation / "Thermal regulation" / Biocomposites}

Abstract of research paper on Materials engineering, author of scientific article — Yuanyuan Li, Shun Yu, Pan Chen, Ramiro Rojas, Alireza Hajian, et al.

Abstract Non-leaking, green materials with high content of phase change materials (PCM) can conserve solar energy and contribute to a sustainable society. Here, paraffin was encapsulated by nanocellulose (CNF) through a pickering emulsion method, while simultaneously forming a composite material. The thermodynamic drive for phase separation was confirmed by molecular modeling. Particle formation was characterized by dynamic light scattering and they were processed into stable PCM/CNF composites in the form of PCM paper structures with favorable mechanical properties. The PCM composite was lightweight and showed a solid content of paraffin of more than 72wt%. Morphology was characterized using FE-SEM. The thermal regulation function of the PCM composite was demonstrated in the form of a model roof under simulated sunlight. No obvious leakage was observed during heating/cooling cycles, as supported by DSC and SAXS data. The PCM composite can be extended to panels used in energy-efficient smart buildings with thermal regulation integrated in load-bearing structures.

Academic research paper on topic "Cellulose nanofibers enable paraffin encapsulation and the formation of stable thermal regulation nanocomposites"

Contents lists available at ScienceDirect

Nano Energy

journal homepage: www.elsevier.com/locate/nanoen

Full paper

Cellulose nanofibers enable paraffin encapsulation and the formation of stable thermal regulation nanocomposites

CrossMark

Yuanyuan Li*, Shun Yu, Pan Chen, Ramiro Rojas, Alireza Hajian, Lars Berglund*

Wallenberg Wood Science Center, Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, SE-100 44 Stockholm, Sweden

ARTICLE INFO

Keywords:

Nanocellulose

Phase change materials

Encapsulation

Thermal regulation

Biocomposites

ABSTRACT

Non-leaking, green materials with high content of phase change materials (PCM) can conserve solar energy and contribute to a sustainable society. Here, paraffin was encapsulated by nanocellulose (CNF) through a pickering emulsion method, while simultaneously forming a composite material. The thermodynamic drive for phase separation was confirmed by molecular modeling. Particle formation was characterized by dynamic light scattering and they were processed into stable PCM/CNF composites in the form of PCM paper structures with favorable mechanical properties. The PCM composite was lightweight and showed a solid content of paraffin of more than 72 wt%. Morphology was characterized using FE-SEM. The thermal regulation function of the PCM composite was demonstrated in the form of a model roof under simulated sunlight. No obvious leakage was observed during heating/cooling cycles, as supported by DSC and SAXS data. The PCM composite can be extended to panels used in energy-efficient smart buildings with thermal regulation integrated in load-bearing structures.

1. Introduction

Around two thirds of the greenhouse gas emissions in the world are related to energy production and use [1]. In the building sector, requirements for electric light, air conditioning, refrigeration, water heating, etc. accounts for as much as 30-40% of the total energy consumption [2,3]. It is therefore a priority to reduce the consumption in the building sector through cleaner energy with increased harvesting and application efficiency. Solar energy is attractive, because the estimated potential of solar energy is in the range of 1575- 49,837 EJ/year, which is roughly 3-100 times the primary energy consumption worldwide during 2008 [4]. Additionally, solar energy is clean, free, and inexhaustible [5,6]. Nanomaterials science and engineering can extend energy-saving technologies by proposing new building materials, where structural and energy-saving characteristics are integrated [7].

Paraffin-based phase change materials (PCM) can store and release large amounts of energy, including solar energy, as latent heat. However, paraffin leakage and low thermal conductivity restrict their applications [8]. Typically, encapsulation technologies are applied to address the leakage problem by enclosing the paraffin in a stable "container" [9,10]. Such encapsulation can be obtained by physical methods such as spray drying, air suspension coating, solvent evaporation, or shear force emulsification. Also chemical methods have been

used; interfacial, in-situ, or emulsion polymerization, sol-gel methods, or their combinations [11]. Various polymeric shell materials have been investigated. This includes petroleum-based polymethylmetha-crylate (PMMA), polystyrene (PS), urea formaldehyde (UF), polyurea, and melamine formaldehyde (MF) [12]. However, more environmentally friendly chemicals or encapsulation processes are desirable.

Nanocellulose derived from renewable resources, such as wood, is highly interesting for combination with phase change constituents. It is the load-bearing component in plant cell walls [13], and can be disintegrated from chemical wood pulp fibers in the form of colloidal nanofibrils in water suspension [14]. The industrial importance will increase, since cellulose nanofibrils (CNF) are now commercially available. Filtration and drying can be used to prepare nanopaper with a modulus as high as 16 GPa and a tensile strength of 330 MPa [15]. Even aerogels from CNF have good mechanical properties [16], and can provide a skeletal reinforcement structure in materials containing functional particles [17,18].

Cellulose exhibits an amphiphilic character and, combined with its nanoparticle nature, can be a surfactant and/or stabilizer at the water-oil interface in pickering emulsions [19,20]. Nanocellulose has been used to prepare stable oil-in-water emulsions, where oil was encapsulated in a nanocellulose shell [21,22]. Capron et al. produced pickering emulsions of hexadecane in water stabilized with bacterial cellulose nanocrystals (CNC) [20]. High stability was observed without any

* Corresponding authors. E-mail addresses: yua@kth.se (Y. Li), blund@kth.se (L. Berglund).

http://dx.doi.org/10.1016/j.nanoen.2017.03.010

Received 23 December 2016; Received in revised form 20 February 2017; Accepted 6 March 2017 Available online 08 March 2017

2211-2855/ © 2017 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/BY-NC-ND/4.0/).

droplet size change after centrifugation. Inspired by the primary plant cell wall, Svagan et al. reported nanocellulose-based liquid-core capsules of high mechanical stability [21]. The excellent mechanical performance was attributed to factors such as chemical crosslinking of cellulose nanofibrils (CNF) and CNC, local organisation, and the high aspect ratio of CNF/CNC. CNF was also reported to form a network at the oil/water interface, and this prevented coalescence of oil droplets [23]. Nanocellulose films and/or coatings have also been studied for their excellent barrier properties, a feature of great interest in phase change material encapsulation [24,25].

In this present study, we attempt a water-based process to prepare a shape-stabilized PCM composite, where the PCM particles are formed during the process. The process is designed to exercise nanostructural control. The specific nanomaterial structure in the form of a continuous CNF nanofiber network, is addressing the problem of leakage. Furthermore, the CNF imparts load-bearing properties to a stimuli-responsive phase change material. Specifically, CNF is intended to encapsulate paraffin for shape-stable and load-bearing thermal regulation materials for energy saving. A simple CNF-water-paraffin system is the starting point. Ultra-sonication is applied to form a paraffin-in-water emulsion. CNF diffuses and is stabilized at the waterparaffin interfaces, forming a CNF shell around paraffin droplets. Excess CNF forms a three-dimensional (3D) network and prevents the coalescence of paraffin capsules. Fig. 1 shows a sketch of shape-stabilized thermo-regulated nanopaper. CNF-shell paraffin-core capsules are embedded in a three-dimensional nanocellulose network. Thermal regulation is achieved by heat absorption or heat release via the phase change (melting/solidification) of paraffin. During use, paraffin works as a thermal buffer to regulate the temperature in the range around its melting point. The present nanocomposite concept could for example be used to cover inner walls or ceilings to reduce the need for air-conditioning, or be used to regulate the temperature in greenhouses.

2. Material and methods

2.1. CNF preparation

CNF was prepared according to previously reported methods [26]. 156 mg 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO, Sigma-Aldrich), 1028 mg sodium bromide (NaBr, Sigma-Aldrich) and 10 g never-dried commercial softwood pulp from sulfite process (13.8% hemicelluloses and 0.7% lignin, Nordic Paper, Sweden) were mixed together in a three-necked round-bottom flask with pulp concentration of 1.0 wt%. Thereafter, 30 mL of a 14% NaClO solution (VWR

Release Absorb

Low temperature High temperature

Fig. 1. Representation of the three-dimensional (3D) structure of form-stable thermal regulation biocomposites containing paraffin-core/CNF-shell capsules.

International) was added dropwise under stirring. The pH of the mixture was maintained at 10.5 by adding 1 mol/L NaOH during the process. The reaction lasted for 1 h and the pulp was washed with distilled water. Further oxidization treatment of the pulp was conducted with 1% w/v NaClO2 in acetate buffer at pH 4.8 for 48 h. After that, the pulp was washed again with distilled water. After the oxidation pre-treatment, the fibers were thoroughly washed with distilled water and disintegrated using a Microfluidizer (M-110EH, Microfluidics Ind., USA) by passing one time through a pair of 400 and 200 ^m microchannels, followed by a pair of 200 and 100 ^m microchannels to obtain a CNF viscous solution of around 0.7 wt%.

2.2. Capsules preparation

CNF was adjusted to a concentration of 0.2 wt% using distilled water. Next, paraffin (RT25HC, obtained from Rubitherm® Technologies GmbH) with 10 times the weight of CNF was added into CNF solution. Ultra-sonication was applied at 85% amplitude for 5 min in order to form the emulsion. After sitting for 24 h, the excess paraffin that phase-separated to the top of the solution was removed.

2.3. PCM Gel and PCM paper preparation

PCM gel was prepared by calcium ion-assisted gelation of the emulsion in a polytetrafluoroethylene mold. Briefly, 200 g of paraffin-water emulsion with CNF concentration of about 0.2 wt% was poured into a 90 mm-diameter polytetrafluoroethylene mold. The emulsion was left undisturbed for 2 h at 4 °C for pre-gelation. Afterwards, 20 mL of 1 wt% CaCl2 solution was added to the emulsion. After sitting for 10 min, a piece of strong gel was obtained. PCM paper was obtained by carefully drying the gel at room temperature.

2.4. Molecular modeling

The adsorption of cellulose on the oil/water interface was studied using molecular dynamics (MD) simulation on the canonical NPT ensemble (constant number of particles, pressure and temperature). The designed simulation box had the initial dimensions of 4.14x20.0x3.97 nm in x, y, z direction, respectively. As shown in Fig. S1, the oil/water phase is parallel to the xz plane. With the application of periodic boundary conditions in all directions, the interface is infinite. In the oil phase, linear alkane chains with 17 carbon units were used, which is identical to the performance in experiment. The thickness of the oil phase is about 8 nm, and that of water phase is about 12 nm.

Based on X-ray and neutron-refined crystal data [27], a single cellulose fibril containing 18 chains with DP =8 was built with exposed hydrophilic (110, 1-10, 010) and hydrophobic (200) surfaces. Along the chain axis, the fibril was covalently bonded to its own periodic image to mimic the infinite chain. The fibril was placed in the water phase with its center of mass distant to the oil/water interface by about 5 nm. The system contained 256 alkane chains, 18 cellulose chains and 5927 water molecules, resulting in a total of 24,149 atoms.

MD simulations were performed with GROMACS version 4.6.4, together with GROMOS 54A7 force field applied for alkane, single point charge (SPC) model for water, and GROMOS 56Acarbo force field with one optimized Lennard-Jones (LJ) parameter for cellulose [2831]. Table S1 shows non-bonded parameters of the GROMOS 56Acarbo force field employed for molecular dynamics simulation of cellulose, alkane, as well as SPC water model. The bonded parameters can be found in the reference [29]. The detailed structure of cellulose, water and alkane are shown in Fig. S2. Energy minimization by steepest descent and the conjugate method was firstly performed before heating up from 0 to 320 K in 10 ns. Subsequently, the production calculation continued for 290 ns. Bond lengths were constrained using LINCS algorithm, thus allowed the use of standard leap-frog algorithm with a

time of 2 fs to solve the equation of motions [32]. Cut-off distance for short-range Coulomb interaction and Lennard-Jones non-bonded interactions was set to 0.9 nm. Long-range dispersion was corrected for energy and pressure. Particle mesh Ewald summation was used for the long-range Coulomb Berendsen algorithm [33]. Temperature was controlled by the velocity-rescaling algorithm of Bussi et al. at 320 K and pressure was regulated at 1 bar using Berendsen barostat in a semi-isotropic manner [34,35]. MD frames were saved every 20 ps.

2.5. Thermal regulating model test

The PCM paper was fixed on the model house roof shown in Fig. 6a. A temperature sensor was secured under each roof to test the variation during simulated solar radiation. An AM 1.5G solar simulator with intensity of 1000 W/m2 was applied to introduce the solar radiation, similar to a high intensity of sun at around noon. Temperature as a function of time was recorded to show the thermal regulating property.

2.6. Characterization

AFM images were captured in the tapping mode with a multimode Nanoscope Ilia Atomic Force Microscope (Bruker Corp.) The samples were prepared by casting 0.1 g/L CNF solution on a silica wafer. SEM images were taken with a Field-Emission Scanning Electron Microscope (FE-SEM, Hitachi S-4800, Japan) operating at an acceleration voltage of 1 kV. The cross-section of the samples was prepared by fracturing the freeze-dried samples after cooling with liquid nitrogen. Pt/Pd coating was sputtered on the sample before characterization. The Zetasizer ZEN3600 (Malvern Instruments Ltd., U.K.) was used for the dynamic light scattering (DLS) measurements to determine the capsules size distribution. One milliliter of diluted emulsion was placed in disposable polystyrene cuvettes. The measurements were performed at room temperature and each of the curves is the average of

16 measurements. Tensile testing of the paper was performed on an Instron 5944 equipped with a 500 N load cell. All samples were kept at 23 °C under 50% humidity for at least 24 h prior to measurement under the same conditions. Tensile test was performed with 10%/min strain rate and 25 mm of span. The specimens were cut into a rectangular strip with a dimension of 5 mmx50 mm. Thermogravimetric analysis (TGA) was performed to measure the composition of the PCM paper at a heating rate of 10 K/min from room temperature to 700 °C under N2 flow of 50 mL/min. Differential scanning calorimetry (DSC) was performed with a heating rate of 10 K/ min in the temperature range of -10-60 °C under a nitrogen atmosphere. The samples were sealed in a 100 ^L aluminum crucible. To investigate the cycling performance, the data was recorded from the second scan. Small-angle X-ray scattering (SAXS) measurements were carried out at MAX II storage ring, MAX IV laboratory. The X-ray wavelength was 0.91 nm with sample-to-detector distance of 1885 ± 1 mm. Pilatus 1 M was used as the detector with a pixel size of 172x172 ^m2. The 1D scattering profile was extracted from the 2D scattering pattern with proper background subtraction and by using DPDAK software [36]. Beam damage to the sample was carefully checked by continuously exposing the sample to X-ray beam at a single spot and monitor the scattering intensity change as a function of time. The combined SAXS and thermal treatment experiment were carried out by using a heat/water cooling stage at the following selected temperature (in °C) in a step-wise sequence: 15, 18, 20, 22, 23, 24, 25, 26, 27, 28, 30, 32, 35, 37, 40, 35, 30, 28, 27, 26, 25, 23. At each temperature, 6 scattering patterns were collected with 10 s exposure time.

3. Results and discussion

Starting with the nanocellulose solution, the encapsulation of paraffin was completed by forming stable oil-in-water emulsion

Fig. 2. a) AFM image showing the morphology of the neat CNF network. b) Low magnification SEM image of paraffin-containing capsules. c) SEM image with the CNF network structure apparent at the capsule surface. d) Size distribution of capsules obtained by DLS, the inset is the optical image of the fresh emulsion (1) and the emulsion after 6 months (2). e) Photograph of capsule-containing gel based on 0.3 wt% CNF suspension and f) rheological properties showing shear-thinning.

Fig. 3. Molecular modeling and sketch illustrating the formation of paraffin capsules: (a) Snapshots of CNFs immersed in water and oil/water interface. Changes over simulation time of (b) distance between cellulose surface and oil/water interface; (c) system enthalpy and (d) number of water molecules in the solvation shell ( < 1.0 nm) of cellulose. (e) Graphic representation of the process of capsule formation as interpreted from molecular modeling simulations.

through ultra-sonication. The CNFs (Fig. 2a) were obtained by disintegrating wood pulp fibers by TEMPO-mediated oxidation followed by high-pressure homogenization [37]. The nanofiber diameter was around 4 nm according to AFM height measurement (Fig. S3) with average CNF length in the order of one micrometer. The morphology of the capsules is exhibited in Fig. 2b, with particle size ranging from hundred nanometers to several micrometers. Parts of capsules in the SEM images appeared collapsed. This is due to the pressure-change during sputter-coating and SEM operation, leading to paraffin evaporation. A densely packed nanocellulose layer was part of the shell,

preventing paraffin leakage. Dynamic light scattering (DLS) was utilized to characterize particle size distribution and resulted in diameter estimates of less than 8 pm, in agreement with SEM images. DLS experiments resulted in two main peaks representing capsule size of about 800 nm and 5.5 pm respectively. The inset image in Fig. 2c shows emulsions freshly prepared (1) and an emulsion after 6 months (2), with no obvious phase separation or other change and therefore highly stable. An advantage of the preparation method is that a gel could easily be obtained only by increasing the CNF concentration. Fig. 2e is the gel prepared from a CNF suspension of 0.3 wt%. The gel

Fig. 4. Photograph of a) PCM gel showing the mechanical flexibility and b) PCM paper obtained by drying the PCM gel. c) Low and d) high magnification SEM images of the PCM paper cross section, the capsule particles are apparent. Surface SEM images e) and f) of PCM paper reveal the rough surface due to the capsules. High magnification SEM g) of the capsules show a dense and "smooth" surface at this scale.

Fig. 5. a) TGA (N2) and b) DSC curves of CNF paper, PCM paper and pure PCM. c) Optical image of PCM paper during the heat/cool cycles between 20 °C (left) and 50 °C (right) after 10 cycles. d) the stress-stain curve of the PCM paper.

shows apparent shear-thinning (Fig. 2f), which is of interest in 3D printing.

To investigate the mechanism of paraffin encapsulation, molecular dynamics (MD) simulation was applied. The cellulose crystal is amphiphilic due to edges with polar hydroxyl groups and hydrophobic faces dominated by non-polar methylene units [38]. This amphiphilic property may contribute to the CNF stabilization at the interface between paraffin and water phases. The diffusion of cellulose from bulk water to the oil/water interface was studied by MD simulation as shown in Fig. 3a. In the first 200 ns of the simulation, CNF kept hovering in the water phase. In the last 100 ns, CNF started to diffuse into the oil/water interface region and stabilize there with its hydro-phobic surface orienting to the interface. Fig. 3b shows the distance between the cellulose surface and the interface as a function of the simulation time. After 200 ns, CNF stabilized the oil/water interface. A video showing CNF diffusion and stabilization at the oil/water interface is supplied in the supporting materials. The diffusion is both enthalpy-driven and entropy-driven process. Cellulose aggregation in water is a spontaneous process partially due to the decreased system enthalpy ascribed to the stronger hydrogen bonds between water molecules than cellulose-water ones [39,40]. This mechanism can explain the CNF diffusion to the oil/water interface. As shown in Fig. 3c, the total enthalpy of the system tends to decrease during the adsorption process. The diffusion is also an entropically driven process. Water molecules near cellulose were confined and possessed less entropy than bulk water. The water molecules near paraffin/water interface also possessed lower entropy than those in bulk due to the polar-apolar interaction. After the CNF diffused to the oil/water interface from water phase, both the water molecules solvating CNF and those near the interface are expelled from the surface to the bulk, leading to a net increase in the solvent entropy (Fig. 3d) [41]. CNF can maintain the

minimum system enthalpy and maximum system entropy by stabilizing at the interface. Fig. 3e represents the formation of the stable paraffin-core capsules in water according to information gathered from molecular modeling: Ultra-sonication allowed paraffin to be dispersed in the continuous water phase in the form of droplets. CNFs in the water phase then diffused to and were stabilized at the paraffin-water interface, leading to the formation of paraffin-core, CNF-shell capsules. Extra CNFs in the system presumably formed a network to prevent coalescence of the capsules.

An advantage of this preparation method is that, the capsules formation and dispersion in the high-strength CNF nanofiber matrix were completed at the same time. Thus, the emulsion can be directly transferred to nanostructured materials in different forms, such as gels, paper structures, and thicker fiber boards. Fig. 4a shows the free standing PCM gel prepared by casting the emulsion in a Teflon mold. The gel is soft and flexible. By gradually drying the gel, a PCM film or paper structure as shown in Fig. 4b is obtained. The PCM paper is white and opaque due to the light-scattering of paraffin particles. As a reference, pure nanocellulose paper is clear and transparent (Fig. S4). The PCM paper structures is composed of capsules immobilized in the 3D nanofiber network formed from the CNFs in the emulsion. Fig. 4c is the cross-sectional SEM image of the PCM paper, with a thickness of about 300 pm. The high magnification image shows that the paper is composed of capsules, as revealed by the hollow spheres in Fig. 4d. The diameters of the hollow spheres range from hundred nanometers to micrometers, which is in agreement with the capsule sizes obtained by DLS. Surface characterization was done to further understand the paper structure. Fig. 4e reveals the rough surface of the PCM paper. The high magnification image shows that the paper surface roughness is caused by the presence of capsules. The capsules are well preserved even under the low pressure during SEM characterization. Focusing on

Time (min)

1,6 1,8 2,0 2,2 2,4 2,6 2,8 3,0 3,2 3,4 0 10 20 30 40 50 60 110 120 130 140

q (nm"I) Measurement sequence

Fig. 6. a) Illustration of the model house thermal regulating test. b) Temperature inside "house" vs. time plots of the "sun" irradiation onto bare model roof, roof with cover of NFC paper, and cover with PCM paper respectively, as well as c) cover with stored PCM paper and PCM paper after water absorption. d) 1D SAXS intensity profiles of PCM paper during a heating and cooling cycle. e) Diffraction peaks of PCM crystalline structure at step-wise increasing temperature, showing the evolution of PCM crystalline structure. f) The peak intensity (red) and position (blue) of diffraction peaks of PCM crystalline structure with corresponding temperature.

the capsule surface (Fig. 4g), a dense layer was observed, preventing leakage.

Thermogravimetric analysis (TGA) was used to characterize the composition and thermal stability of the PCM paper. The results showed a composition of 72.2 wt% paraffin and 27.8 wt% of CNF (Fig. 5a). The thermal behavior of PCM paper is similar to pure nanocellulose paper with one more weight loss region at around 350450 °C. The degradation of paraffin is shifted to higher temperatures, as it is encapsulated in CNF. The PCM paper weight loss of about 10 wt % from room temperature to 120 °C is due to evaporation of residual moisture. CNF starts to degrade at T > 200 °C, resulting in weight loss and char formation from 200 °C to 350 °C. Further increase in temperature degraded the CNF completely. In PCM paper, the encapsulated paraffin contributed to the weight loss from 350 to 450 °C.

In DSC calorimetry, the paraffin demonstrated melting in the range 21-31 °C, with the main melting point at around 29 °C (Fig. 5b). The PCM paper gave similar exothermic peak and melting point. An extra peak appeared at around 20 °C, maybe due to the presence of water in the composite. The total enthalpy of the composite calculated in the

temperature range of 15-35 °C is 139 J/g, while for pure paraffin is 183 J/g. This high enthalpy is ascribed to the high paraffin content. Cycling performance was investigated by heating the PCM paper to 50 °C and then cooling down to 20 °C repeatedly. The PCM paper was milky and opaque at 20 °C as shown in Fig. 5c, left. After heating to 50 °C, the paper became clear and transparent due to the melting of PCM (Fig. 5c, right). No obvious leakage was detected during the heating/cooling cycles. Fig. 5c is the image of PCM paper after 10 heating/cooling cycles. DSC was also utilized to monitor the leakage of PCM paper between -10 and 60 °C (Fig. S5). After 100 cycles, the paper still shows repeatable phase change behavior, confirming no leakage.

For application purposes, load-bearing PCM composite materials are preferred. The mechanical performance of the PCM paper was characterized by uniaxial tensile testing. The paper showed a tensile strength of 30 MPa and strain to failure of 12%, while pure PCM was too weak for tensile testing. The good mechanical performance was due to the nanocellulose network with good compatibility and interaction with the PCM/CNF capsules.

Thermal regulation is a desirable function of PCM composites. We tested the thermal regulation performance of PCM paper on a model house as shown in Fig. 6a. A piece of PCM paper with a weight of 2 g was fixed on the model house roof with a surface area of 25 cm2. The house was placed under a sun simulator with an intensity of 1000 W/ m2, which translates to intense solar radiation at noon. A digital thermometer probe was fixed to the inner roof. The PCM paper showed excellent thermal regulation performance as illustrated in Fig. 6b. With PCM paper, 23 min was required to reach 35 °C, much longer than the bare roof or roof with NFC paper. This was due to thermal absorption by the paraffin (melting) and a contribution from evaporation of free water. The second cycling showed decreased thermal regulation capacity, shown in Fig. 6b. It took 7 min to reach 35 °C, much less time compared with the first exposure to the simulated sun radiation. This is mainly due to the free water in the PCM paper present during the first cycle. Water has very high evaporation enthalpy; 2.3 MJ kg-1. Nevertheless, the paper still showed thermal regulation function and this could be improved by using thicker PCM fiber boards rather than thin paper structures. An advantage of PCM-nanocellulose materials is that the paper can be designed to absorb and retain a large amount of water. This could enable the paper to use rain or high humidity during evenings, for thermal regulation during following hot days. Fig. 6c shows thermal regulation of PCM paper during wetting and drying cycles. The PCM paper can absorb about 30 wt% its own weight of water. After water absorption, the PCM paper showed improved thermal regulation with 17 min to reach 35 °C. The PCM gel was also characterized and showed high thermal regulation performance due to the large amount of water in its matrix. Accordingly, the temperature at the ceiling for a PCM gel remained at room temperature for more than 50 min due to the evaporation of water.

SAXS measurements were performed to study the structural changes of the PCM paper during heating and cooling cycles. The measurements were carried out in such manner that 6 SAXS patterns (10 s/exposure) were collected at various temperatures controlled by a heating/cooling stage unit. Selected full-scale 1D SAXS profiles are shown in Fig. 6d and reflected the scattering of the interfacial structures present on the capsules. At different temperature, the intensity slope in the low q region follows the trend as I(q) x q~4, which obeys the Porod law [42]. The exponent was -4, indicating smooth interfaces, presumably as those observed with the SEM image of the capsule surfaces. As the exponent remained at -4 during thermal cycling, no percolated PCM networks were formed. The SAXS data show that the PCM particle phase was unchanged in terms of distribution (paraffin stayed in the particle, also in molten state). The evolution of the diffraction peaks of the PCM crystalline structure is shown in Fig. 6e at increasing temperatures. The peak intensity monotonically decreases upon increasing temperature and disappears at 32 °C and beyond. During the cooling process, the PCM crystalline peak reappeared at 25 °C. The intensity (red) and position (blue) of the PCM peak is plotted in Fig. 6f and related to the corresponding temperature. PCM melting starts at around 22 °C, as noted by decreased peak intensity. Concurrently, the peak position shifted toward lower scattering vector, q value, indicating that the lattice constant of the crystalline phase expanded. Moreover, the melting rate abruptly increased at around 27 °C, with rapid decrease in peak intensity. This is in good agreement with the DSC curves obtained for the PCM paper. Furthermore, the peak position displayed great fluctuation while the peak intensity steadily decreased. The minimum peak position from these experiments is 2.54 nm-1, which corresponds to 2.47 nm as the maximum lattice constant in real space. This may correspond to the largest lattice constant at which the PCM can still hold a crystalline phase. The observed fluctuation of the peak position may be explained with the contributions of the liquefied phase shortening the residual crystalline phase to obtain a decreased lattice constant.

4. Conclusions

For the purpose of thermal storage materials, paraffin-core, CNF-shell capsules were successfully fabricated using an emulsion preparation method with CNF as the stabilizer. The preparation procedure has the advantage that capsules are formed during the process, and integrates capsule stabilization and dispersion in the high-strength CNF nanofiber matrix. Potentially, the materials could be processed in scalable paper-like routes. Molecular dynamics simulation provided thermodynamic data for the cellulose diffusion mechanism, away from the water phase, to the oil/water interface. The PCM composite showed high enthalpy and excellent thermal regulation performance with no leakage during heating/cooling cycles. For the purpose of temperature regulation, an advantage of this PCM paper composite is the high water absorption/retention, which may lead to outdoor applications. This PCM paper is a promising lightweight, structural and functional candidate for smart building applications.

Acknowledgments

We acknowledge funding from the Knut and Alice Wallenberg foundation through the Wallenberg Wood Science Center at KTH Royal Institute of Technology. MAX-IV Laboratory is acknowledged for providing beamtime and beamline staff at beamline I911- SAXS at MAX-II storage ring. Xuan Yang and Yingxin Liu are acknowledged for the help for tensile test and rheology test. Mingjing Lin is acknowledged for the help in the scheme drawing.

Appendix A. Supplementary material

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.nanoen.2017.03.010.

References

[1] International Energy Agency, CO2 Emissions From Fuel Combustion Highlights 2015 (2015). (http://www.iea.org/publications/freepublications/publication/ CO2EmissionsFromFuelCombustionHighlights2015.pdf) (accessed 21.03.16).

[2] P. Nejat, F. Jomehzadeh, M.M. Taheri, M. Gohari, M.Z.A. Majid, Renew. Sustain. Energy Rev. 43 (2015) 843-862.

[3] A. Rotzetter, C. Schumacher, S. Bubenhofer, R. Grass, L. Gerber, M. Zeltner, W. Stark, Adv. Mater. 24 (2012) 5352-5356.

[4] M. Aman, K. Solangi, M. Hossain, A. Badarudin, G. Jasmon, H. Mokhlis, A. Bakar, S. Kazi, Renew. Sustain. Energy Rev. 41 (2015) 1190-1204.

[5] N.S. Lewis, Science 315 (2007) 798-801.

[6] Z. He, H. Wu, Y. Cao, Adv. Mater. 26 (2014) 1006-1024.

[7] Y. Zhang, X. Zheng, H. Wang, Q. Du, J. Mater. Chem. A 2 (2014) 5304-5314.

[8] T. Khadiran, M.Z. Hussein, Z. Zainal, R. Rusli, Sol. Energy Mater. Sol. Cells 143 (2015) 78-98.

[9] C. Liu, Z. Rao, J. Zhao, Y. Huo, Y. Li, Nano Energy 13 (2015) 814-826.

[10] J. Wang, M. Yang, Y. Lu, Z. Jin, L. Tan, H. Gao, S. Fan, W. Dong, G. Wang, Nano Energy 19 (2016) 78-87.

[11] D.C. Hyun, N.S. Levinson, U. Jeong, Y. Xia, Angew. Chem. Int. Ed. 53 (2014) 3780-3795.

[12] W. Su, J. Darkwa, G. Kokogiannakis, Renew. Sustain. Energy Rev. 48 (2015) 373-391.

[13] J. Pérez, J. Munoz-Dorado, T. de la Rubia, J. Martinez, Int. Microbiol. 5 (2002) 53-63.

[14] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Chem. Soc. Rev. 40 (2011) 3941-3994.

[15] H. Sehaqui, N.E. Mushi, S. Morimune, M. Salajkova, T. Nishino, L.A. Berglund, ACS Appl. Mater. Interfaces 4 (2012) 1043-1049.

[16] M. Pääkkö, J. Vapaavuori, R. Silvennoinen, H. Kosonen, M. Ankerfors,

T. Lindström, L.A. Berglund, O. Ikkala, Soft Matter 4 (2008) 2492-2499.

[17] R.T. Olsson, M. Samir, G. Salazar-Alvarez, L. Belova, V. Strom, L.A. Berglund, O. Ikkala, J. Nogues, U.W. Gedde, Nat. Nanotechnol. 5 (2010) 584-588.

[18] Y.Y. Li, H.L. Zhu, H.B. Gu, H.Q. Dai, Z.Q. Fang, N.J. Weadock, Z.H. Guo, L.B. Hu, J. Mater. Chem. A 1 (2013) 15278-15283.

[19] Y. Li, H. Zhu, F. Shen, J. Wan, S. Lacey, Z. Fang, H. Dai, L. Hu, Nano Energy 13 (2015) 346-354.

[20] I. Kalashnikova, H. Bizot, B. Cathala, I. Capron, Langmuir 27 (2011) 7471-7479.

[21] A.J. Svagan, A. Musyanovych, M. Kappl, M. Bernhardt, G. Glasser, C. Wohnhaas, L.A. Berglund, J. Risbo, K. Landfester, Biomacromolecules 15 (2014) 1852-1859.

[22] I. Capron, B. Cathala, Biomacromolecules 14 (2013) 291-296.

[23] T. Winuprasith, M. Suphantharika, Food Hydrocolloid 32 (2013) 383-394.

[24] C. Aulin, M. Gällstedt, T. Lindström, Cellulose 17 (2010) 559-574.

[25] P.A. Larsson, L.A. Berglund, L. Wagberg, Biomacromolecules 15 (2014) 2218-2223.

[26] Y. Kobayashi, T. Saito, A. Isogai, Angew. Chem. Int. Ed. 53 (2014) 10394-10397.

[27] Y. Nishiyama, P. Langan, H. Chanzy, J. Am. Chem. Soc. 124 (2002) 9074-9082.

[28] N. Schmid, A.P. Eichenberger, A. Choutko, S. Riniker, M. Winger, A.E. Mark, W.F. van Gunsteren, Eur. Biophys. J. 40 (2011) 843-856.

[29] H.S. Hansen, P.H. Hünenberger, J. Comput. Chem. 32 (2011) 998-1032.

[30] P. Chen, Y. Nishiyama, K. Mazeau, Cellulose 21 (2014) 2207-2217.

[31] B. Hess, C. Kutzner, D. Van Der Spoel, E. Lindahl, J. Chem. Theory Comput. 4

(2008) 435-447.

[32] B. Hess, H. Bekker, H.J. Berendsen, J.G. Fraaije, J. Comput. Chem. 18 (1997) 1463-1472.

[33] U. Essmann, L. Perera, M.L. Berkowitz, T. Darden, H. Lee, L.G. Pedersen, J. Chem. Phys. 103 (1995) 8577-8593.

[34] G. Bussi, D. Donadio, M. Parrinello, J. Chem. Phys. 126 (2007) 014101.

[35] H.J. Berendsen, Jv Postma, W.F. van Gunsteren, A. DiNola, J. Haak, J. Chem. Phys. 81 (1984) 3684-3690.

[36] G. Benecke, W. Wagermaier, C. Li, M. Schwartzkopf, G. Flucke, R. Hoerth, I. Zizak, M. Burghammer, E. Metwalli, P. Müller-Buschbaum, J. Appl. Crystallogr. 47 (2014) 1797-1803.

[37] H. Fukuzumi, T. Saito, T. Wata, Y. Kumamoto, A. Isogai, Biomacromolecules 10

(2009) 162-165.

[38] S. Besombes, K. Mazeau, Plant Physiol. Biochem. 43 (2005) 299-308.

[39] X. Zhao, Y. Chen, X. Jiang, Y. Shang, L. Zhang, Q. Gong, H. Zhang, Z. Wang, X. Zhou, J. Therm. Anal. Calorim. 111 (2013) 891-896.

[40] J. Taylor, Trans. Faraday Soc. 53 (1957) 1198-1203.

[41] P. Chen, Y. Nishiyama, J. Wohlert, et al., Translational entropy and dispersion energy jointly drive the dsorption of urea to cellulose[J], J. Phys. Chem. B (2017). http://dx.doi.org/10.1021/acs.jpcb.6b11914.

[42] W. Ruland, J. Appl. Crystallogr. 4 (1971) 70-73.

Dr. Yuanyuan Li received her Ph.D. degree in Nanjing Forestry University in 2014. During her Ph.D., she worked as an exchange student at the Department of Materials Science and Engineering, University of Maryland, focusing on nanocellulose based multifunctional composites and flexible electronics. Currently, she is a postdoc at KTH Royal Institute of Technology working on biomaterials for functional structural materials.

Shun Yu received his PhD degree in surface physics in Royal Institute of Technology (KTH), Stockholm, Sweden in 2012. Then, he held a Knut and Alice Wallenberg Postdoc fellowship of "MAX IV synchrotron radiation program" and worked at Deutsches Elektronen-Synchrotron(DESY), Hamburg, Germany. Now, he works as a researcher at the department of Fiber and Polymer Technology, KTH. His research focuses on understanding the structural formation and evolution of soft matters and organic/inorganic hybrid systems by using advanced synchrotron radiation based X-ray techniques.

Dr. Pan Chen got PhD degree in polymer science from university of Grenoble Alpes, CERMAV-CNRS, France in June 2013, under the supervision of Dr. Jean-Luc Putaux, Dr. Karim Mazeau and Dr. Yoshiharu Nishiyama. Since April 2014, he joined the department of progress engineering of RWTH Aachen university as a scientific staff working on the dissolution mechanism of crystalline cellulose for two years. Currently, he is a postdoc associate in WWSC, KTH and focuses on development and understanding of bio-based nanocomposites materials using the combination of experimental and molecular modeling techniques.

Ramiro Rojas is currently a researcher at the Wallenberg Wood Science Center at KTH Royal Institute ofTechnology working on the general topic of "wood nanotechnology". He received his Ph.D. in Chemistry from Rutgers University in New Jersey, USA and held a Swedish Institute postdoctoral fellowship at the Angstrom Laboratory in Uppsala University, Sweden. His research focuses on new materials technologies with an emphasis on method design, processing, and the study of structure-property relationships of natural and synthetic polymers.

Alireza Hajian received his M.S. degree from KTH Royal Institute of Technology in macromolecular materials. He is currently a Ph.D. candidate in the Wallenberg Wood Science Center at KTH under supervision of Prof. Lars Berglund. His present research is focused on cellulose composites.

Lars Berglund is a professor at KTH Royal Institute of Technology with a strong interest in bio-based materials and nanocomposites. He is the director of Wallenberg Wood Science Center on New materials from trees. He has published more than 200 journal papers, and examined more than 20 PhD-students. He is a member of the Royal Swedish Academy of Engineering Science.