Study on a reliable epoxy-based phase change material: facile preparation, tunable properties, and phase/microphase separation behavior

Qingsong Lian , Kai Li , Asim A. S. Sayyed , Jue Cheng * and Junying Zhang *
The Key Laboratory of Beijing City on Preparation and Processing of Novel Polymer Materials, Beijing University of Chemical Technology, Beijing 100029, China. E-mail: chengjue@mail.buct.edu.cn; zhangjy@mail.buct.edu.cn; Fax: +86 10 6442 5439; Tel: +86 10 6442 5439

Received 31st March 2017 , Accepted 3rd June 2017

First published on 5th June 2017


Room-temperature-use phase change materials (PCMs) are of vital importance in combining the sustainable development of energy with human comfort. Here, a series of novel epoxy-based polymeric solid–solid PCMs (SSPCMs) were facilely prepared by UV grafting 1-octadecanethiol (ODT) onto an allyl-based epoxy resin (DADGEBA) via thiol–ene click chemistry followed by the epoxy–amine thermal curing process. Therefore, the grafted ODT can be tightly locked in a reliable three-dimensional crosslinking network of the epoxy resin, which provides the PCMs with excellent shape-stable property. Phase change and mechanical properties characterized by DSC, DMA, hardness test, and tensile test can be easily tuned by adjusting the mass ratio of DGEBA and the ODT-grafted-DADGEBA product (D18), which was characterized by FTIR and NMR. XRD and POM analyses proved the crystallinity of EPD18-X PCMs. The structure and morphology of the EPD18-X PCMs were characterized by visual images, SEM, and POM analyses. Microphase separation was observed in all the EPD18-X PCMs, and an obvious phase separation was observed in the EPD18-25 system. However, the phase separation gradually disappeared with the increasing of the D18 content. Thermal recycling tests showed that EPD18-X PCMs can remain stable after 50 DSC thermal cycles. Due to the unique strong encapsulated epoxy curing networks, EPD18-X PCMs have excellent thermal stability with the onset degradation temperature higher than 250 °C. Tunable EPD18-X systems can be applied for room-temperature-use thermal energy storage applications such as buildings, thermoregulated fabrics, and so on.


1. Introduction

The applications of thermal energy storage (TES) technology in energy conservation and new energy fields like solar energy have gained a great deal of attention since the energy crisis of the 1970s. Phase change materials (PCMs), which can store and release quantities of energy during phase transition processes, are effective materials to resolve the TES issue and achieve the sustainable development of energy.1,2 According to the type of phase change, PCMs can be classified into three categories: solid–gas PCMs (SGPCMs), solid–liquid PCMs (SLPCMs), and solid–solid PCMs (SSPCMs). Further, most types of traditional PCMs such as paraffin, polyethylene glycol (PEG), and fatty acid are SLPCMs, which need to deal with the leakage problem during the phase change process. Hence, the preparation of form-stable PCMs (FSPCMs) has aroused great interest among researchers.

In general, FSPCMs are a kind of composite material that consist of a supporting material (maintaining the solid state of the materials) and a type of SLPCM (providing latent heat storage capacity). The main focus on these kinds of PCMs is the selection of the supporting material that can effectively maintain the solid state of the composite material to prevent leakage even when the temperature is above the phase change temperature (Tpc) of the SLPCMs. Current methods to prepare FSPCMs are microencapsulating the SLPCMs with different shells such as polystyrene,3 acrylic polymer,4,5 zirconia,6 silica,7,8 melamine–formaldehyde (MF),9 calcium carbonate,10 graphene,11–13etc. or incorporating the SLPCMs into a special structural material such as porous materials,14,15 hydrogels,16 aerogels,13,17–20 gelators,21,22 and recent metal–organic frameworks.23

However, the biggest problem that restricts the practical application of the above methods may be the reliability concern of the supporting materials, except for other problems like high cost. To be precise, owing to the fact that the intrinsic mechanism of avoiding leakage in FSPCMs is mainly physical cladding or physical absorption (through the intermolecular forces) effects, the SLPCMs will be leaked out above its Tpc once the supporting materials are destroyed under external stimulus.7

To resolve this problem, another feasible thought is the preparation of polymeric SSPCMs from its pristine SLPCMs.24 Due to the reliability of the chemical crosslinking networks, the polymeric SSPCMs can be used directly without encapsulation.25 In this field, PEG, which is a traditional type of SLPCM, may be the most successful material, and there have been countless reports about PEG-based polymeric SSPCMs.26–31 Although the PEG-based polymeric SSPCMs can be easily synthesized according to several steps of the basic chain extension reaction of isocyanates with polyols, the isocyanates used in this procedure are harmful to the environment and human beings because aromatic isocyanates can hydrolyze into highly toxic phenylamine.32 Besides this, another disadvantage that limits the application of PEG-based polymeric SSPCMs is their relatively high Tpc at around 60 °C, which is higher than the comfortable temperature (around 37 °C) that human beings can endure. Therefore, the PEG-based polymeric SSPCMs have restricted applications in thermal insulation materials such as buildings, thermoregulated fabrics, and so on.33

In order to obtain room-temperature-use polymeric SSPCMs, we are interested in finding a new type of SLPCM that not only has mild Tpc (like paraffin) but also possesses functional reactive groups (like hydroxyl groups in PEG). Besides, another proper material that can react with SLPCMs to form polymeric SSPCMs is also needed. In this paper, we designed a novel type of epoxy-based polymeric SSPCM by grafting 1-octadecanethiol (ODT) onto an allyl-based epoxy resin followed by the ring-opening reaction (ROP) of the epoxy groups. The preparation procedure of this kind of polymeric SSPCM is facile and the raw materials used are not that toxic when compared to the hypertoxic isocyanates used in the preparation of PEG-based polymeric SSPCMs.

ODT, which has been widely used for applications such as heavy metal detection,34 semiconductor nanowires,35 and etch resists36 based on its self-assembled property, has a melting point of 31 °C and was used as a room-temperature PCM for the first time such that human beings can feel comfortable. As for epoxy resin, which can form a strong three-dimensional (3-D) crosslinking network and exhibit excellent performances such as excellent mechanical properties, chemical stability, and heat resistance,37 is surely a reliable supporting material to shape the solid state of PCMs. Although some researchers have tried to blend epoxy resin with other SLPCMs followed by UV curing38 or thermal curing,39 the physical incorporation of other SLPCMs will greatly reduce the thermal mechanical properties of the epoxy resin;39 besides, the leakage problem cannot be perfectly resolved.38 Therefore, it arouses our great interest in finding a dual functional epoxy resin that can react with PCMs (ODT) without sacrificing the epoxy groups; in this case, the chemical-bonded resultant product of the ODT–epoxy compound can further go through the ROP of the epoxy groups to form a novel reliable epoxy-based polymeric SSPCM having high latent heat.

Fortunately, our group has done a great deal of work on a dual functional epoxy resin named diglycidyl ether of 4,4′-diallyl bisphenol-A (DADGEBA, see Fig. 1).40–42 The existence of double bonds makes it possible to facile graft with ODT via thiol–ene click chemistry, and then, the grafted product (named D18) can be locked in the 3-D crosslinking network of epoxy resin after the ROP of the epoxy groups.


image file: c7ta02816d-f1.tif
Fig. 1 Structures of the main chemicals used in this study.

Thiol–ene click chemistry has been an efficient and facile method in preparing materials since its discovery by Sharpless in 2001.43 However, there are no reports on the preparation of PCMs through this effective method, until recently by Kahraman.44 Therefore, thiol–ene click chemistry, despite its success in many fields according to several reviews,45–49 is overlooked in the fabrication of polymeric SSPCMs for energy conservation applications.

Herein, a brand new approach was developed for the two-stage preparation of novel epoxy-based polymeric SSPCMs via thiol–ene photo-grafted reaction followed by epoxy–amine thermal curing process. The structure of D18 was characterized by FTIR and NMR. Systematic characterizations including differential scanning calorimeter (DSC), scanning electron microscope (SEM), X-ray diffraction (XRD), polarized optical microscope (POM), and TG measurement were carried out. Besides, phase change and mechanical properties characterized by dynamic mechanical analyzer (DMA), hardness test, and tensile test of this kind of PCM can be tuned by simply changing the mass ratio of diglycidyl ether of bisphenol A (DGEBA) and D18. The phase/microphase separation of the epoxy-based polymeric SSPCMs was also discussed, which might be helpful in understanding the crystalline behavior of other liquid crystalline (LC) copolymers. These tunable and reliable epoxy-based polymeric SSPCMs (EPD18-X PCMs) may have the potential for room-temperature-use TES applications such as buildings, thermoregulated fabrics, and so on.

2. Experimental

2.1 Materials

DGEBA (CYD-128; epoxy equivalent: 194 g mol−1) was obtained from Sinopec Group. DADGEBA (epoxy equivalent: 248 g mol−1; double bond equivalent: 210 g mol−1) was synthesized according to the literature.50 Poly(propylene oxide)diamine (Jeffamine D230; amine equivalent: 61 g mol−1) was purchased from Huntsman. 2-Hydroxy-2-methylpropiophenone (1173, 99%) was obtained from Tokyo Chemical Industry. ODT (99%+) was purchased from Adamas Reagent Company.

2.2 UV grafting procedure of DADGEBA and ODT via thiol–ene click chemistry

The mixture of DADGEBA and ODT (molar ratio of double bonds to thiol groups = 1[thin space (1/6-em)]:[thin space (1/6-em)]1) with additional initiator 1173 (1% mass ratio) was charged into a 20 mL vial (using magnetic stirring). Then, the reaction mixture was stirred at 80 °C for 4 h under a 365 nm UV LED light and the irradiation intensity was 100 mW cm−2. The chemical equation can be seen in Scheme 1.
image file: c7ta02816d-s1.tif
Scheme 1 Thiol–ene click chemistry of DADGEBA and ODT.

2.3 Preparation of crosslinking D18–DGEBA polymeric SSPCMs

DGEBA and synthetic D18 were well mixed with mass ratios of 100/0, 75/25, 50/50, 25/75, and 0/100 at 60 °C (D18 is a liquid at this temperature and the state of D18 at different temperatures can be seen in the ESI Fig. S1), and then, these five mixtures were poured into D230 according to the respective stoichiometric ratios. After effective mixing of the resin and the curing agent, all the final mixtures were pre-cured at 80 °C for 2 h, cured at 100 °C for 2 h, and 125 °C for 4 h. The cured samples (also the polymeric SSPCMs) were named as EPD18-25, EPD18-50, EPD18-75, and EPD18-100 according to the D18 content, and the pure EPD18-0 system (mass ratio of DGEBA and D18 is 100/0) was called the blank sample. The preparation process of the EPD18-100 polymeric SSPCM and its structure can be seen in Scheme 2.
image file: c7ta02816d-s2.tif
Scheme 2 Preparation of crosslinking EPD18-100 polymeric SSPCM and its phase change behavior.

2.4 Measurements

FTIR tests were carried out on a Bruker Alpha FTIR machine at a resolution of 4 cm−1 in the wavenumber range of 4000–400 cm−1 using KBr pellets: all the samples were scanned 32 times.

1H NMR experiment was carried out on a Bruker Advance 400 spectrometer (400 MHz) using deuterated chloroform (CDCl3) as the solvent.

The DSC spectra were recorded on a TA Instruments Q20 equipped with an RCS 90 cooling system. All the samples were heated and cooled between −70 °C and 150 °C at a heating rate of 10 °C min−1 in a N2 atmosphere. It should be noted that each sample was quickly heated to 150 °C and kept for 3 min, and then, sharply cooled to 70 °C to eliminate the thermal history before the test.

The crystallinity of the epoxy-based polymeric SSPCMs was examined by wide-angle X-ray scattering (WAXS) measurements using a Rigaku D/Max 2500 VB2+/PC diffractometer with Cu Kα radiation.

POM (Olympus BX51 polarizing microscope) was used to further characterize the crystalline structure of the materials. All the samples were heated to 120 °C for 1 h, and then, put into a 30 °C oven for 6 h to perform the isothermal crystallization before the POM observation.

The fractured morphology (broken in liquid nitrogen) was characterized by a JEOL JSM-6700M SEM at an accelerating voltage of 5 kV, and all the surfaces of the samples were coated with gold to improve the conductivity and prevent charging.

The thermal stability of the PCMs was tested by thermogravimetric analysis (TGA) measurements, which were carried out on a TGA Q50 analyzer at a heating rate of 10 °C min−1 in a N2 atmosphere.

DMA was used to characterize the thermal mechanical properties by a TA Q800 instrument. Samples were tested with a film tensile mode at a heating rate of 5 °C min−1 from −50 to 150 °C: the oscillatory frequency was 1 Hz.

Hardness test was performed on a Fowler Shore D durometer according to ASTM D2240; the results were obtained as the average of five measurements.

Tensile test was performed at room temperature according to the Chinese National Standard GB/T 1040.2-2006.

3. Results and discussion

3.1 Synthesis of D18

Thiol–ene click chemistry was applied to graft ODT onto the backbone of DADGEBA. After the reaction of DADGEBA and ODT, we can observe that the C[double bond, length as m-dash]C stretching vibration at 1638 cm−1 and the S–H stretching vibration at 2570 cm−1 disappeared in Fig. 2, the result of which proves that the double bonds of DADGEBA as well as the thiol groups of ODT had completely reacted. Besides, the disappearance of the vinyl C–H out-of-plane deformation vibration at 995 cm−1 further proves this conclusion. It should be noted that 912 cm−1 is the combination of the characteristic peak of the epoxy group and the vinyl CH2 out-of-plane deformation vibration, and the epoxy group will not react with ODT in this reaction condition (although the thiol–epoxy reaction can occur at room temperature with a tertiary amine as the catalyst,51 the lowest reaction temperature of the thiol–epoxy without a catalyst system is 160 °C according to our previous study42). Therefore, the amount of epoxy groups still remained unchanged after the reaction of DADGEBA with ODT, and the decrease of 912 cm−1 can only be attributed to the consumption of double bonds.
image file: c7ta02816d-f2.tif
Fig. 2 FTIR spectra of D18 and the reaction mixture (DADGEBA and ODT) before reaction.

Moreover, the structure of the synthetic D18 was also characterized by 1H NMR spectrum. All the chemical shifts of the basic peaks of DADGEBA and D18 can be marked well in Fig. 3. The peaks at 5.95, 5.00, and 3.37 ppm that represent the unsaturated allyl structure of DADGEBA shifted to 1.84, 2.49, and 2.69 ppm, respectively. Besides, the peaks of the grafted ODT at 2.49, 1.56, 1.36, 1.26, and 0.89 ppm appeared on the spectrum of D18. The above result shows that the double bonds of DADGEBA had completely grafted with ODT, and D18 was successfully synthesized through the thiol–ene click chemistry, which is in consistent with the FTIR result.


image file: c7ta02816d-f3.tif
Fig. 3 1H NMR spectra of DADGEBA and D18 (CDCl3 as the solvent).

3.2 Shape-stable properties and phase separation behavior

Shape stability is of vital importance in avoiding the leakage problem in PCMs. Here, the shape-stable properties of EPD18-X PCMs were characterized by a leakage test. As shown in Fig. 4, all the samples were heated to 120 °C (which is much higher than the Tpc of EPD18-X PCMs) and maintained for 60 min, and the status of all the samples were photographed by a camera. We can observe that all the samples can retain their solid state at 120 °C without leakage. It should be noted that the practical temperature for using EPD18-X PCMs can be even as high as 250 °C, which will be discussed in Section 3.6 of this article.
image file: c7ta02816d-f4.tif
Fig. 4 Visual images of the blank sample and EPD18-X PCMs at 18 °C (room temperature) and 120 °C for 60 min (from left to right: blank sample, EPD18-25, EPD18-50, EPD18-75, and EPD18-100).

Besides, we can observe that the blank and EPD18-100 samples are quite transparent because both of them are homogeneous single phase systems. It should be noted that the transparency of the blank sample is higher than that of the EPD18-100 sample at 18 °C because the blank sample is in a completely amorphous state, like other neat epoxy curing systems,52 while the partial crystalline property affects the transparency of the EPD18-100 sample, which became even more transparent at 120 °C because the crystalline grafted ODT had turned into the amorphous state at this temperature, as shown in Scheme 2. In spite of this, it can be concluded that macroscopic phase separation did not exist in the EPD18-100 and blank samples.

As for the EPD18-X samples (except for the EPD18-100 sample), it is obvious that the visual image of the EPD18-25 sample is completely opaque, which indicates that D18 and DGEBA is incompatible and phase separation occurred at this time. With the increasing of the D18 content, the transparency of the EPD18-X sample also increases. The result shows that D18 is hard to dissolve in DGEBA, while DGEBA is relatively easy to dissolve in D18, and thus, the compatibility of DGEBA and D18 can be increased with the increasing content of D18.

To improve our understanding of the phase separation behavior of D18 and DGEBA, as well as the crystalline structure of EPD18-X PCMs, the cold-fractured interface (broken in liquid nitrogen) morphologies were characterized by SEM images. In Fig. 4, the completely transparent morphologies of the blank and EPD18-100 samples have been proven at the macroscopic level. However, the two samples exhibit quite different morphologies at the microscopic level. According to Fig. 5(a), the blank sample shows a homogeneous single phase microstructure, like other common DGEBA curing systems,53–56 as expected. As for the EPD18-100 system shown in Fig. 5(e) and (f), the white needle-like crystalline ODT that grafted on DADGEBA was well dispersed in the epoxy networks, thereby demonstrating the absence of macroscopic phase separation of the EPD18-100 sample (Fig. 4). It should be noted that the EPD18-100 sample is a liquid crystalline (LC) grafted copolymer that is similar to the LC block copolymers (BCPs), which are able to microphase separate into various types of nanostructures because of the chemically distinct features between the constituent blocks of LC BCPs.57 Although ODT was chemically bonded with DADGEBA, the huge polarity difference between ODT and DADGEBA still induced the bicontinuous phase structure like other crystalline–amorphous LC BCPs58–62 and microphase separation occurred. When it comes to the EPD18-25 system, we can observe a distinct sea-island structure, which indicates that phase separation occurred in this system. Due to the polarity difference between D18 and DGEBA and the relatively low content of D18 in this system, the dispersive spherical structure can only be attributed to the self-assembly crystalline chains of ODT in D18, and DGEBA is the continuous phase in this system. The above result confirms the occurrence of macroscopic phase separation for this system, as shown in Fig. 4. However, the SEM images of EPD18-50 and EPD18-75 samples exhibit quite different morphologies as compared to the EPD18-25 system: actually, the morphologies of the two samples are highly similar to the SEM image of the EPD18-100 system. This result suggests that phase inversion occurred, and D18 became the continuous phase in the EPD18-50 and EPD18-75 systems. It is interesting to observe that there are still a great amount of barbed structures growing on the white needle-like crystalline ODT in the EPD18-50 system; however, such barbed structures barely exist in the EPD18-75 system. Owing to the fact that the transparency of the EPD18-75 sample is better than that of the EPD18-50 sample in Fig. 4, the barbed structures could well be the result of the partial self-assembly of ODT, which caused the reduction of the transparency of the EPD18-50 sample. In conclusion, the compatibility of D18 and DGEBA can be increased with the increasing content of D18, which is consistent with the result of the macroscopic observation shown in Fig. 4. Moreover, neither macroscopic phase separation nor microphase separation was observed in the blank sample, while macroscopic phase separation occurred in the EPD18-25 system, and microphase separation existed in the EPD18-50, EPD18-75, and EPD18-100 systems. The phase/microphase separation behavior and the compatibility of DGEBA and D18 will be further discussed in Section 3.4 of this article.


image file: c7ta02816d-f5.tif
Fig. 5 SEM images of the cold-fractured interface (broken in liquid nitrogen) morphologies of (a) the blank sample, (b) EPD18-25, (c) EPD18-50, (d) EPD18-75, (e) EPD18-100, (f) and the partially magnified area of EPD18-100.

3.3 Thermal analysis

Phase change properties, which directly determine the practical application prospects, are dominant parameters for PCMs. DSC analysis is the most effective method in determining the Tpc (determines the application range) and calculating the latent heat (determines the application value)29 of the PCMs. The heating and cooling DSC curves of all the curing products are illustrated in Fig. 6(a): we can observe that there is only one melting or freezing peak in all the EPD18-X systems, and the melting point (Tm), freezing point (Tf), melting latent heat (ΔHm), and freezing latent heat (ΔHf) can be directly obtained from the DSC curves. Besides, the DSC curves of ODT and D18 can be seen in the ESI Fig. S2. As seen in Table 1, pristine ODT has the lowest Tm (31.1 °C) and Tf (25.8 °C); however, when ODT was grafted onto the double bonds of DADGEBA and formed D18, both the Tm and Tf shifted to higher temperatures of 38.6 °C and 32.5 °C, respectively. This change is possibly a consequence of the influence of the DADGEBA framework. Even so, the supercooling extent of D18 did not change a lot as compared to that of ODT. The supercooling extent is the difference between Tm and Tf: excessive supercooling extent can be an obstacle to the applications of PCMs. As for EPD18-X systems, the supercooling extent only increases a little (from 6 °C to 10 °C). This increase is not significant during this investigation and can be acceptable.
image file: c7ta02816d-f6.tif
Fig. 6 DSC analysis of the blank sample and EPD18-X PCMs: (a) melting and freezing processes; (b) characterization of glass transition temperature (Tg).
Table 1 Thermal characteristics of ODT, D18, EPD18-X PCMs, and blank samplea
System T g (°C) Melting process Freezing process
T m (°C) ΔHm (J g−1) ΔHTm (J g−1) ΔHm loss (%) T f (°C) ΔHf (J g−1) ΔHTf (J g−1) ΔHf loss (%)
a T m, Tf, ΔHm, and ΔHf were evaluated by the DSC curves; ΔHTm and ΔHTf of D18 were calculated by multiplying the weight percentage of ODT in D18 by the melting or freezing enthalpies of ODT; ΔHTm and ΔHTf of EPD18-X were calculated by multiplying the weight percentage of D18 in the D18–DGEBA–D230 curing system by the melting or freezing enthalpies of D18.
ODT 31.1 241.1 241.1 0.0 25.8 224.8 224.8 0.0
D18 38.6 104.0 139.5 25.5 32.5 96.9 130.1 25.5
EPD18-100 36.5 70.5 94.3 25.2 25.1 65.1 87.9 25.9
EPD18-75 58.3 34.0 36.5 67.5 45.9 18.7 32.8 62.8 47.8
EPD18-50 61.2 36.6 27.6 43.0 35.8 25.8 25.8 40.0 35.6
EPD18-25 70.5 36.9 14.8 20.6 28.1 27.3 14.3 19.2 25.4
Blank sample 87.9


In the meanwhile, it can be observed that all the actual latent heat (ΔHm and ΔHf) is lower than the theoretical latent heat (ΔHTm and ΔHTf). Actually, all the practical latent heat of PEG-based polymeric SSPCMs is lower than its theoretical latent heat because the crystalline regions of PEG segments are decreased owing to the restriction of the crosslinking structure.27,29,63 Just like PEG-based polymeric SSPCMs, the ΔHm loss (the specific value of (ΔHT − ΔH)/ΔHT) of the EPD18-X PCMs cannot be neglected because there are three factors reducing its practical latent heat. First, the DADGEBA backbone that ODT is grafted on plays an important role in restricting the crystallization of ODT, and the ΔHm loss for ODT is 25.5%. Second, the 3-D crosslinking network of epoxy resin further causes a 25.2% ΔHm loss for D18 in the EPD18-100 system (without DGEBA). Third, the incorporation of DGEBA into D18 significantly restricts the crystallization of ODT and the ΔHm loss reaches the maximum value at 45.9%. It should be noted that the crystallinity of the EPD18-X samples is contributed by the crystalline side chains of the grafted ODT in D18, but DGEBA networks (just like the networks of PEG-based polymeric PCMs) greatly restrict the crystalline property of the grafted ODT in D18, leading to lower crystalline regions. However, it is interesting to note that the ΔHm loss decreases with the increasing content of DGEBA. This phenomenon might be related to the compatibility of DGEBA and D18. In the discussion involving the visual images and SEM analysis of the EPD18-X system, we have proven that the compatibility of DGEBA and D18 can be increased with the increasing content of D18. In the EPD18-75 system, DGEBA can completely dissolve in D18, and thus, the dissolved DGEBA will greatly restrict the crystalline property of the grafted ODT in D18, which results in the maximum ΔHm loss value. In contrast, macroscopic separation occurred in the EPD18-25 system and the grafted ODT can even self-assemble; as a result, the crystallinity of the ODT chains in the EPD18-25 system is hard to be affected by the incorporation of DGEBA, leading to the minimum ΔHm loss value among the EPD18-25, EPD18-50, and EPD18-75 systems. As for the EPD18-50 system, the ΔHm loss value is between the EPD18-25 and EPD18-75 systems because the compatibility of D18 and DGEBA in this system is better than that in the EPD18-25 system but worse than that in the EPD18-75 system. In other words, although the ΔHm and ΔHf values still increased with the increase in the content of D18 due to more content of the ODT crystalline side chains, the ΔHm loss in the EPD18-X systems also increased with the increasing content of D18 due to better compatibility between D18 and DGEBA.

Besides, EPD18-X PCMs can display two well-separated transitions, as shown in Fig. 6(b): one from the melting of the grafted ODT and the other from the glass transition of the epoxy resin. In general, Tg is positively correlated to the crosslink density of the curing networks,64 and the long ODT side chains of D18 will decrease the crosslink density of the curing products. This is the reason why Tg decreases with the increasing content of D18 in the D18–DGEBA–D230 curing systems, as shown in Fig. 6(b). It should be noted that the glass transition cannot be found in the DSC curve of the EPD18-100 sample, which may be because the Tg of EPD18-100 is near the Tm (36.5 °C) and is covered by the melting peak.

In conclusion, it is obvious that the Tg of the EPD18-X systems can be increased (from 36 °C to 87.9 °C) with the increasing content of DGEBA, while the latent heat decreased from 70.5 to 14.8 J g−1 at the same time.

3.4 Crystalline characterization

Fig. 7 shows the XRD diagrams of EPD18-X PCMs. It is obvious that the blank sample is a typical amorphous polymer because it only demonstrates a smooth un-diffraction curve in the XRD characterization. However, all the XRD diagrams of the EPD18-X samples are a combination of a smooth amorphous area and a sharp crystalline area, which has a diffraction peak at around 21.2° and the intermolecular spacing is 4.2 Å, as calculated by the Bragg eqn (1):
 
= 2d[thin space (1/6-em)]sin[thin space (1/6-em)]θ(1)

image file: c7ta02816d-f7.tif
Fig. 7 XRD diagrams of the blank sample and EPD18-X PCMs.

Besides, we can observe that the diffraction peak at 21.2° became more and more obvious with the increase of the D18 content, that is, the crystallinity of the EPD18-X samples increases with the D18 content. This phenomenon is easy to understand because the crystallinity of the EPD18-X samples is contributed by the crystalline side chains of the grafted ODT in D18. To further prove this conclusion, the crystallinity values of the EPD18-25, EPD18-50, EPD18-75, and EPD18-100 samples calculated by the X-ray method65,66 are 28.5%, 14.6%, 11.7%, and 6.3%, respectively. These results prove that the crystallinity increases with the content of D18. It should be noted that the XRD diagram of pure D18 (see ESI Fig. S3) shows a similar diffraction peak at 21.4°.

In order to further confirm the crystallinity, POM was employed to characterize the crystalline phase structure of EPD18-X PCMs. In Fig. 8(a), there is no birefringence in the POM spectrum of the amorphous blank sample as expected. Besides, it is obvious that all the EPD18-X PCMs exhibit obvious birefringence in the POM spectra because they have different degrees of crystallinity according to the XRD result. However, the birefringence phenomenon is quite different in each EPD18-X PCM. We have pointed out that the EPD18-100 polymer is a LC grafted copolymer, which is similar to the LC BCPs discussed in Section 3.2 of this paper. Although many kinds of LC BCPs show microphase separation and exhibit obvious birefringence in their POM spectra, typical crystalline patterns are hard to be identified,67–69 which may be due to the fact that the development of a crystalline domain is suppressed by the formation of microphase separation.67 This theory is also suitable for the EPD18-100 system and D18. Although ODT is chemically bonded with DADGEBA, the huge polarity difference between the ODT side chains and DADGEBA's main backbone still induces microphase separation, which restrains the development of the crystalline domain of grafted ODT. This is the reason why the POM spectra of the EPD18-100 sample and D18 can exhibit obvious birefringence, but a typical crystalline texture cannot be identified. However, it is interesting to observe the typical crystalline texture in the EPD18-25, EPD18-50, and EPD18-75 systems, although the microphase separation still existed in these systems. This may be due to the effect of heterogeneous nucleation initiated by the incorporation of DGEBA. In spite of this, the POM morphologies of the EPD18-25, EPD18-50, and EPD18-75 systems are quite different owing to the differences of the D18 content. The macroscopic separation of the EPD18-25 sample has been observed in Fig. 4 and 5(b) because of the phase separation of DGEBA and D18, which resulted in a phenomenon in which self-assembled crystalline structures were dispersed in a large amorphous area (Fig. 8(b)). When the D18 content increased to 50%, we can observe that the crystalline domains increased and can be uniformly dispersed in the amorphous area, which is due to the increasing compatibility of D18 and DGEBA. As for the EPD18-75 system, the crystalline and amorphous domains had completely combined together according to Fig. 8(d), which is consistent with the previous result that DGEBA had completely dissolved in D18 in the EPD18-75 system. In conclusion, the POM result not only proves the crystalline property of the EPD18-X samples but also further demonstrates the phase/microphase separation behavior.


image file: c7ta02816d-f8.tif
Fig. 8 POM images of (a) the blank sample, (b) EPD18-25, (c) EPD18-50, (d) EPD18-75, (e) EPD18-100, (f) and D18 obtained at room temperature (64× magnification).

3.5 Thermal recycling properties

The DSC curves of the EPD18-X systems before and after 50 thermal cycles can be seen in the ESI Fig. S4, and the data of the ΔHm and ΔHf values as well as the ΔHm and ΔHf losses obtained from the above DSC curves are listed in Table 2. It should be noted that no significant difference was observed in the melting or freezing temperatures of the EPD18-X PCMs (Table S1). Besides, only a very slight decrease (≤2.0%) was observed in the latent heat of all the EPD18-X PCMs after 50 thermal cycles (Table 2). This result proves that this kind of novel epoxy-based polymeric SSPCM is very stable during the thermal cycling test.
Table 2 Thermal recycling properties of the blank sample and EPD18-X PCMs
System Melting latent heat (ΔHm) Freezing latent heat (ΔHf)
1 cycle (J g−1) 50 cycles (J g−1) ΔHm loss (%) 1 cycle (J g−1) 50 cycles (J g−1) ΔHf loss (%)
EPD18-100 70.5 69.7 1.1 65.1 64.2 1.4
EPD18-75 36.5 36.1 1.1 32.8 32.2 1.8
EPD18-50 27.6 27.1 1.8 25.8 25.3 1.9
EPD18-25 14.8 14.5 2.0 14.3 14.1 1.4


3.6 Thermal stability analysis

Thermal stability is of great significance for the TES applications of PCMs. In this field, the TGA test is the most commonly used method to evaluate the thermal stability of PCMs. According to the TG and DTG curves shown in Fig. 9, we can observe that there was only one thermal decomposition process for all the samples and the degradation ended at around 500 °C. The blank sample owns the best thermal stability property and the weight loss did not occur until 325 °C. As for the EPD18-X samples, the thermal stability is not as good as the blank sample. There are two reasons for this phenomenon. On the one hand, the thermal stability of DADGEBA is lower than that of DGEBA according to our previous study.42 On the other hand, the grafted ODT will decrease the thermal stability of the epoxy resin due to the relatively lower thermal stability of ODT, according to ESI Fig. S5. In spite of this, all the EPD18-X samples can remain stable until 250 °C, which is much higher than the melting point (around 35 °C) of EPD18-X PCMs.
image file: c7ta02816d-f9.tif
Fig. 9 Thermal stability analysis of the blank sample and EPD18-X PCMs: (a) TG curves, (b) DTG curves.

3.7 Mechanical properties and potential applications of the PCMs

DMA is of vital importance for measuring the viscoelastic properties by investigating the stiffness and damping characteristics of the materials. The change of storage modulus and tan[thin space (1/6-em)]δ (with the increase of temperature) of the EPD18-X PCMs are shown in Fig. 10 and tabulated in Table 3. In fact, the tan[thin space (1/6-em)]δ value can be used to characterize the Tg of materials, and the Tg value obtained from the tan[thin space (1/6-em)]δ value is usually higher than that from the DSC curve due to the frequency effect of the DMA test.70 Therefore, it is not strange that the Tg obtained by DMA is higher than that by DSC for the blank and EPD18-25 samples. However, the peak temperature of tan[thin space (1/6-em)]δ of the other three systems (namely, EPD18-50, EPD18-75, and EPD18-100) is around 40 °C, which is much lower than the Tg tested by DSC, as shown in Table 3. Besides, in Fig. 10(b), we can observe an extra small shoulder peak at 66.3 °C for the EPD18-50 system, which is more likely the Tg because 66.3 °C is higher than 61.2 °C (DSC test result). Actually, it is not well known that the tan[thin space (1/6-em)]δ of DMA can also be used to characterize the melting point of materials that have high degree of crystallinity, such as ice,71,72 isotactic polypropylene,73 and so on. Therefore, we conclude that the peak temperature of tan[thin space (1/6-em)]δ is the melting point of the ODT side chains in the EPD18-50, EPD18-75, and EPD18-100 samples, while the peak temperature of tan[thin space (1/6-em)]δ is the Tg of the EPD18-25 and blank samples. This conclusion is consistent with the SEM analysis shown in Fig. 5. In the previous study of this paper, we have known that the compatibility of D18 and DGEBA can be increased with the increasing content of D18. In the low-crystalline EPD18-25 sample, DGEBA plays the dominant role and the SEM image displays typical sea-island morphology, so the tan[thin space (1/6-em)]δ-temperature curve of the EPD18-25 sample can only show the glass transition process at around 93.7 °C. As for the EPD18-50 and EPD18-75 systems, the bicontinuous phase structure of the two samples in the SEM images is highly approximate to that of the EPD18-100 system, which shows that D18 has played the dominant role. The EPD18-50 sample can still display a small shoulder peak at 66.3 °C, which is attributed to the Tg of the sample. However, the EPD18-75 sample can only display the melting peak at 40.9 °C because the effect of D18 on DGEBA is very huge and the glass transition process cannot be observed in this system, not even for the EPD18-100 system. Besides, the storage modulus (E′) at 25 °C decreased with the increasing content of D18. However, it is interesting to note that the EPD18-25 sample possesses higher E′ than the blank sample before Tg is achieved. This may be because the self-assembly (observed in Fig. 5(b)) of a small amount of D18 side chains can form bumpy structures, which can dramatically improve the modulus of the epoxy resin, according to our previous study.74,75
image file: c7ta02816d-f10.tif
Fig. 10 DMA results of the blank sample and EPD18-X PCMs: (a) storage modulus–temperature curves; (b) tan[thin space (1/6-em)]δ-temperature curves.
Table 3 Thermal mechanical properties and mechanical properties of the blank sample and EPD18-X PCMsa
System E′ at 25 °C (MPa) Peak temperature of tan[thin space (1/6-em)]δ (°C) T g for DSC (°C) Hardness (shore D) Tensile strength (MPa) Tensile modulus (MPa)
a E′ is equal to the storage modulus tested by the DMA.
EPD18-100 353.8 46.7 56 7.66 72.17
EPD18-75 354.7 40.9 58.3 65 3.16 4.31
EPD18-50 533.5 39.5 61.2 74 6.23 29.03
EPD18-25 1188.7 93.7 70.5 85 48.53 201.30
Blank sample 854.8 102.2 87.9 90 59.08 351.91


Moreover, hardness and tensile tests were carried out to characterize the mechanical properties of the blank and EPD18-X samples. The blank sample had the highest hardness, and the hardness of the EPD18-X samples decreased with the increasing content of D18. This is easy to be understood because the structure of DGEBA is much more rigid than that of D18. As for the tensile test, the data of the EPD18-25 sample is similar to the blank sample, although the tensile strength and tensile modulus of the EPD18-25 sample are slightly lower than those of the blank sample. This result shows that DGEBA still plays an important role in the EPD18-25 system. However, the tensile strength and tensile modulus of the EPD18-50 and EPD18-75 samples are quite low, which cannot even match those of the EPD18-100 sample. This may be because DGEBA will greatly destroy the crystalline structure of D18, while the content of DGEBA is not enough to provide good mechanical properties (as the EPD18-25 system) in the EPD18-50 and EPD18-75 systems.

Therefore, EPD18-X systems not only are a novel type of reliable SSPCMs (due to their strong crosslinking networks in avoiding leakage problems) but also possess relatively good mechanical properties when compared with most types of commonly used PCMs, which makes the EPD18-X systems have tremendous potential for room-temperature-use TES applications such as buildings, thermoregulated fabrics, and so on.

4. Conclusions

In this paper, a series of novel epoxy-based polymeric SSPCMs were prepared by UV grafting ODT (to provide latent heat storage capacity) onto an allyl-based epoxy resin (DADGEBA) via thiol–ene click chemistry followed by epoxy–amine thermal curing process (to maintain the solid state of the PCMs). Further, in this case, the grafted ODT can be tightly locked in a reliable 3-D crosslinking network of the epoxy resin. Tg as well as phase change properties can be easily tuned by adjusting the mass ratio of DGEBA and the ODT-grafted-DADGEBA product (D18), which were characterized by FTIR and NMR. Tg of the EPD18-X systems can be increased (from 36 to 87.9 °C) with the increasing content of DGEBA, while the latent heat decreased from 70.5 to 14.8 J g−1 at the same time in the Tpc ranging from 18 to 37 °C, which is practical for energy storage at room temperature. XRD and POM analyses proved the crystallinity of EPD18-X PCMs. The structure and morphology of EPD18-X PCMs were characterized by visual images and SEM analysis, and an obvious phase separation was observed in the EPD18-25 system because of the incompatibility between DGEBA and D18. However, the macroscopic phase separation gradually disappeared with increasing D18 content: DGEBA can completely dissolve in D18 in the EPD18-75 system. Besides, microphase separation existed in all the EPD18-X systems because of the polarity difference between the chemically bonded ODT and DADGEBA. Moreover, thermal recycling tests suggest that the EPD18-X PCMs can remain stable even after 50 thermal cycles. Finally, the EPD18-X PCMs exhibit excellent thermal stability with the onset degradation temperature higher than 250 °C. Mechanical properties characterized by DMA, tensile test, and hardness test suggest that DGEBA still plays an important role in the EPD18-25 system due to its high storage modulus and relatively high tensile strength, tensile modulus, and hardness; while the mechanical properties of the EPD18-50 and EPD18-75 systems cannot even match the EPD18-100 system. In other words, it is interesting to observe the phase/microphase separation in the EPD18-X PCMs, which might be helpful in understanding the crystalline behavior of other LC copolymers. Besides, the tunable and reliable epoxy-based polymeric SSPCMs (EPD18-X PCMs) have the potential for room-temperature-use TES applications such as buildings, thermoregulated fabrics, and so on.

Acknowledgements

The author would like to thank the financial support from the National Natural Science Foundation of China (Project no. 21176017 and 21476013) and the National Key Research Program of China (Project no. 2016YFB0302000).

References

  1. S. Raoux, W. Wełnic and D. Ielmini, Chem. Rev., 2010, 110, 240–267 CrossRef CAS PubMed .
  2. D. Lencer, M. Salinga and M. Wuttig, Adv. Mater., 2011, 23, 2030–2058 CrossRef CAS PubMed .
  3. Y. Konuklu and H. Ö. Paksoy, Sol. Energy Mater. Sol. Cells, 2017, 159, 235–242 CrossRef CAS .
  4. Y. Ma, J. Zong, W. Li, L. Chen, X. Tang, N. Han, J. Wang and X. Zhang, Energy, 2015, 87, 86–94 CrossRef CAS .
  5. M. Graham, E. Shchukina, P. F. De Castro and D. Shchukin, J. Mater. Chem. A, 2016, 4, 16906–16912 CAS .
  6. Y. Zhang, X. Wang and D. Wu, Energy, 2016, 97, 113–126 CrossRef CAS .
  7. X. Zhang, X. Wang and D. Wu, Energy, 2016, 111, 498–512 CrossRef CAS .
  8. N. Şahan and H. Paksoy, Sol. Energy Mater. Sol. Cells, 2017, 159, 1–7 CrossRef .
  9. J. Liu, L. Chen, X. Fang and Z. Zhang, Sol. Energy Mater. Sol. Cells, 2017, 159, 159–166 CrossRef CAS .
  10. T. Wang, S. Wang, R. Luo, C. Zhu, T. Akiyama and Z. Zhang, Appl. Energy, 2016, 171, 113–119 CrossRef CAS .
  11. Y. Zhang, X. Zheng, H. Wang and Q. Du, J. Mater. Chem. A, 2014, 2, 5304–5314 CAS .
  12. Z. Chen, J. Wang, F. Yu, Z. Zhang and X. Gao, J. Mater. Chem. A, 2015, 3, 431–437 Search PubMed .
  13. S. Ye, Q. Zhang, D. Hu and J. Feng, J. Mater. Chem. A, 2015, 3, 4018–4025 CAS .
  14. Y. Li, J. Li, Y. Deng, W. Guan, X. Wang and T. Qian, Appl. Energy, 2016, 171, 37–45 CrossRef CAS .
  15. A. Sarı, Energy Convers. Manage., 2016, 117, 132–141 CrossRef .
  16. Y. Liu, Y. Yang and S. Li, J. Mater. Chem. A, 2016, 4, 18134–18143 CAS .
  17. X. Huang, W. Xia and R. Zou, J. Mater. Chem. A, 2014, 2, 19963–19968 CAS .
  18. X. Huang, Z. Liu, W. Xia, R. Zou and R. S. Han, J. Mater. Chem. A, 2014, 3, 1935–1940 Search PubMed .
  19. J. Yang, X. Li, S. Han, Y. Zhang, P. Min, N. Koratkar and Z.-Z. Yu, J. Mater. Chem. A, 2016, 4, 18067–18074 CAS .
  20. G. Li, X. Zhang, J. Wang and J. Fang, J. Mater. Chem. A, 2016, 4, 17042–17049 CAS .
  21. D. Wu, W. Wen, S. Chen and H. Zhang, J. Mater. Chem. A, 2015, 3, 2589–2600 CAS .
  22. D. Wu, B. Ni, Y. Liu, S. Chen and H. Zhang, J. Mater. Chem. A, 2015, 3, 9645–9657 CAS .
  23. Y. Luan, M. Yang, Q. Ma, Y. Qi, H. Gao, Z. Wu and G. Wang, J. Mater. Chem. A, 2016, 4, 7641–7649 CAS .
  24. J.-C. Su and P.-S. Liu, Energy Convers. Manage., 2006, 47, 3185–3191 CrossRef CAS .
  25. K. Chen, R. Liu, C. Zou, Q. Shao, Y. Lan, X. Cai and L. Zhai, Sol. Energy Mater. Sol. Cells, 2014, 130, 466–473 CrossRef CAS .
  26. W.-D. Li and E.-Y. Ding, Sol. Energy Mater. Sol. Cells, 2007, 91, 764–768 CrossRef CAS .
  27. Z. Li, W. He, J. Xu and M. Jiang, Sol. Energy Mater. Sol. Cells, 2015, 140, 193–201 CrossRef CAS .
  28. Y. Deng, J. Li, T. Qian, W. Guan, Y. Li and X. Yin, Chem. Eng. J., 2016, 295, 427–435 CrossRef CAS .
  29. X. Fu, Y. Xiao, K. Hu, J. Wang, J. Lei and C. Zhou, Chem. Eng. J., 2016, 291, 138–148 CrossRef CAS .
  30. Z. Liu, X. Fu, L. Jiang, B. Wu, J. Wang and J. Lei, Sol. Energy Mater. Sol. Cells, 2016, 147, 177–184 CrossRef CAS .
  31. W. Wang, B. Tang, B. Ju, Z. Gao, J. Xiu and S. Zhang, J. Mater. Chem. A, 2017, 5, 958–968 CAS .
  32. Q. Wu, D. Zhao, X. Jiao, Y. Zhang, K. J. Shea, X. Lu and G. Qiu, Ind. Eng. Chem. Res., 2015, 54, 8130–8136 CrossRef CAS .
  33. S. Phadungphatthanakoon, S. Poompradub and S. P. Wanichwecharungruang, ACS Appl. Mater. Interfaces, 2011, 3, 3691–3696 CAS .
  34. T. Zhang, Z. Cheng, Y. Wang, Z. Li, C. Wang, Y. Li and Y. Fang, Nano Lett., 2010, 10, 4738–4741 CrossRef CAS PubMed .
  35. S. Kim, P. D. Carpenter, R. K. Jean, H. Chen, C. Zhou, S. Ju and D. B. Janes, ACS Nano, 2012, 6, 7352–7361 CrossRef CAS PubMed .
  36. G. Lu, Y. Chen, B. Li, X. Zhou, C. Xue, J. Ma, F. Y. C. Boey and H. Zhang, J. Phys. Chem. C, 2009, 113, 4184–4187 CAS .
  37. J.-P. Pascault and R. J. Williams, Epoxy polymers: new materials and innovations, Wiley-VCH, 2010 Search PubMed .
  38. A. S. Luyt and I. Krupa, Energy Convers. Manage., 2009, 50, 57–61 CrossRef CAS .
  39. Y. Fang, H. Kang, W. Wang, H. Liu and X. Gao, Energy Convers. Manage., 2010, 51, 2757–2761 CrossRef CAS .
  40. G. Zhang, J. Cheng, L. Shi, X. Lin and J. Zhang, Thermochim. Acta, 2012, 538, 36–42 CrossRef CAS .
  41. Q. Lian, Y. Li, T. Yang, K. Li, Y. Xu, L. Liu, J. Zhao, J. Zhang and J. Cheng, J. Mater. Sci., 2016, 51, 7887–7898 CrossRef CAS .
  42. Q. Lian, Y. Li, K. Li, J. Cheng and J. Zhang, Macromolecules, 2017, 50, 803–810 CrossRef CAS .
  43. H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed., 2001, 40, 2004–2021 CrossRef CAS PubMed .
  44. E. Baştürk, D. Y. Deniz and M. V. Kahraman, Mater. Chem. Phys., 2016, 177, 521–528 CrossRef .
  45. A. B. Lowe, Polym. Chem., 2010, 1, 17–36 RSC .
  46. C. E. Hoyle and C. N. Bowman, Angew. Chem., Int. Ed., 2010, 49, 1540–1573 CrossRef CAS PubMed .
  47. M. J. Kade, D. J. Burke and C. J. Hawker, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 743–750 CrossRef CAS .
  48. A. Dondoni and A. Marra, Chem. Soc. Rev., 2012, 41, 573–586 RSC .
  49. A. B. Lowe, Polym. Chem., 2014, 5, 4820–4870 RSC .
  50. R. W. Martin and C. E. Pannell, US Pat. 2965607, 1960 .
  51. E. M. Petrie, Epoxy adhesive formulations, McGraw Hill Professional, 2005 Search PubMed .
  52. S. Chen, B. Chen, J. Fan and J. Feng, ACS Sustainable Chem. Eng., 2015, 3, 2077–2083 CrossRef CAS .
  53. E. Abraham, D. Kam, Y. Nevo, R. Slattegard, A. Rivkin, S. Lapidot and O. Shoseyov, ACS Appl. Mater. Interfaces, 2016, 8, 28086–28095 CAS .
  54. T. Li, S. He, A. Stein, L. F. Francis and F. S. Bates, Macromolecules, 2016, 49, 9507–9520 CrossRef CAS .
  55. H. Garate, M. Bianchi, L. I. Pietrasanta, S. Goyanes and N. B. D'Accorso, ACS Appl. Mater. Interfaces, 2017, 9, 930–943 CAS .
  56. T. K. L. Nguyen, S. Livi, B. G. Soares, H. Benes, J.-F. Gérard and J. Duchet-Rumeau, ACS Sustainable Chem. Eng., 2017, 5, 1153–1164 CrossRef CAS .
  57. S.-H. Huang, Y.-W. Huang, Y.-W. Chiang, T.-J. Hsiao, Y.-C. Mao, C.-H. Chiang and J.-C. Tsai, Macromolecules, 2016, 49, 9048–9059 CrossRef CAS .
  58. H. Uehara, T. Yoshida, M. Kakiage, T. Yamanobe, T. Komoto, K. Nomura, K. Nakajima and M. Matsuda, Macromolecules, 2006, 39, 3971–3974 CrossRef CAS .
  59. C. M. Bates and F. S. Bates, Macromolecules, 2017, 50, 3–22 CrossRef CAS .
  60. M. G. Jeong and K. T. Kim, Macromolecules, 2017, 50, 223–234 CrossRef CAS .
  61. Y. La, C. Park, T. J. Shin, S. H. Joo, S. Kang and K. T. Kim, Nat. Chem., 2014, 6, 534–541 CrossRef CAS PubMed .
  62. T. H. An, Y. La, A. Cho, M. G. Jeong, T. J. Shin, C. Park and K. T. Kim, ACS Nano, 2015, 9, 3084–3096 CrossRef CAS PubMed .
  63. S. Mu, J. Guo, S. Zhang, Q. An, D. Wang, Y. Liu and F. Guan, Mater. Lett., 2016, 171, 23–26 CrossRef CAS .
  64. M. Sharifi, C. Jang, C. F. Abrams and G. R. Palmese, Macromolecules, 2015, 48, 7495–7502 CrossRef CAS .
  65. B. H. Clampitt and C. E. Walker, Anal. Chem., 1965, 37, 1076–1078 CrossRef CAS .
  66. X. Shen, W. Hu and T. P. Russell, Macromolecules, 2016, 49, 4501–4509 CrossRef CAS .
  67. G.-H. Wen, B. Zhang, H.-L. Xie, X. Liu, G.-Q. Zhong, H.-L. Zhang and E.-Q. Chen, Macromolecules, 2013, 46, 5249–5259 CrossRef CAS .
  68. M. Yan, J. Tang, H.-L. Xie, B. Ni, H.-L. Zhang and E.-Q. Chen, J. Mater. Chem. C, 2015, 3, 8526–8534 RSC .
  69. P.-P. Hou, Z.-Y. Zhang, Q. Wang, M.-Y. Zhang, Z. Shen and X.-H. Fan, Macromolecules, 2016, 49, 7238–7245 CrossRef CAS .
  70. E. Zini, M. Scandola, P. Dobrzynski, J. Kasperczyk and M. Bero, Biomacromolecules, 2007, 8, 3661–3667 CrossRef CAS PubMed .
  71. Y. Vodovotz and P. Chinachoti, J. Agric. Food Chem., 1998, 46, 446–453 CrossRef CAS PubMed .
  72. M.-Y. Baik and P. Chinachoti, J. Agric. Food Chem., 2001, 49, 4031–4038 CrossRef CAS PubMed .
  73. P. Tiemblo, J. M. Gómez-Elvira, S. G. Beltrán, L. Matisova-Rychla and J. Rychly, Macromolecules, 2002, 35, 5922–5926 CrossRef CAS .
  74. T. Yang, R. Wang, X. Hou, J. Cheng and J. Zhang, Mater. Lett., 2016, 166, 150–153 CrossRef CAS .
  75. L. Ba, Q. Zou, X. Tan, J. Song, J. Cheng and J. Zhang, RSC Adv., 2016, 6, 91875–91881 RSC .

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ta02816d

This journal is © The Royal Society of Chemistry 2017