Advanced Composites and Hybrid Materials

, Volume 2, Issue 3, pp 471–480 | Cite as

Latent heat and thermal conductivity enhancements in polyethylene glycol/polyethylene glycol-grafted graphene oxide composites

  • Junyang Tu
  • Hairong Li
  • Jingjing Zhang
  • Die Hu
  • Ziqing Cai
  • Xianze Yin
  • Lijie Dong
  • Leping Huang
  • Chuanxi XiongEmail author
  • Ming JiangEmail author
Original Research


Although high-thermal conductivity fillers can enhance the thermal conductivity of organic phase change materials (PCMs), it is still difficult to simultaneously prevent the loss in latent heat. The realization of enhanced thermal conductivity without adversely affecting the latent heat of polymeric PCMs remains challenging. We report experimental results demonstrating that polyethylene glycol (PEG)-based PCM composites exhibit large latent heats beyond the expected values and significantly improved thermal conductivities. Latent heat and thermal conductivity enhancements are achieved by optimizing the interaction between the filler and PCM in the PEG/graphene oxide (GO) composites. PEG-grafted GO (PEG-g-GO) is synthesized and introduced into the PEG to regulate the interaction between the PEG base and GO. The result can be largely attributed to the enhanced GO dispersion and heightened mobility of the PEG chains. The grafted PEG in the PEG-g-GO acts as a plasticizer, leading to a considerable effect on the crystallization kinetics of the PEG in the composite PCMs, and thus the improved crystallizability.

Graphical abstract

Latent heat and thermal conductivity enhancements are achieved by optimizing the interaction between the filler and PCM in the PEG/GO composites.


Phase change material Nanocomposite Latent heat Thermal conductivity Polyethylene glycol Graphene oxide 

1 Introduction

Phase change materials (PCMs) can be used for thermal energy storage because of their ability of storing and releasing latent heat by transforming one physical state (e.g., crystalline state) to another (e.g., amorphous state). Unfortunately, their applications in thermal energy storage are greatly confined because of the low thermal conductivities and the leakage of most of the PCMs when they undergo solid-liquid phase change [1, 2, 3, 4, 5]. To this end, considerable efforts have been devoted to solve these problems by adding thermally conductive fillers or supporting materials such as clay [6], silica [7], diatomite [8], active carbon [9], CNTs [10], expanded graphite [11, 12], and graphene [13, 14, 15]. However, the incorporation of the fillers into PCMs to enhance the thermal conductivity and/or keep the same form in solid state even when PCMs change phase from solid to liquid state may normally result in an undesirable significant decrease of the phase change enthalpies. This is expected given that some of the PCM weight is replaced by the fillers that cannot undergo phase change.

Yavari et al. [16] reported a 2.5 times increase in thermal conductivity (0.91 W/mK) and a loss of 15.4% heat of fusion for the graphene flakes/stearyl alcohol nanocomposite at 4 wt% graphene loading. Silakhori et al. [17] found that the thermal conductivity of palmitic acid/polypyrrole/graphene nanoplatelet composite increased by 38.7% and 34.3% in comparison with those of PA and PA/PPy and the latent heat dropped by 27.4% with 1.6 wt% graphene (151 J/g). Mehrali et al. [18] obtained an enormous increase of about 460% in thermal conductivity after impregnation with a very high mass fraction (55.19 wt%) of GO in paraffin, but there was also a significant 55.2% reduction in melting latent heat. Rufuss et al. [19] prepared paraffin/GO composites by impregnating low mass fraction (0.3%) of GO nanoparticles in paraffin. The study showed an improved thermal conductivity over that of virgin paraffin giving a 101% increase in solid state, but a 36.5% deterioration in latent heat compared to virgin paraffin. Yang et al. [20] prepared a gel PCM by impregnating of polyethylene glycol (PEG) into the cellulose/graphene aerogels exhibiting a high thermal conductivity of 1.35 W/mK and yet a latent heat loss of 12.7% with 5.3 wt% of graphene.

Although high-thermal conductivity fillers can enhance the thermal conductivity of organic PCMs, it is still difficult to simultaneously prevent the loss in latent heat. When mixed with PCMs, those fillers actually significantly reduce the energy storage density of PCMs in most cases. Recently, several cases of breakthrough have been reported. Amin et al. [21] showed that graphene nanoplatelets enhanced both the latent heat and thermal conductivity of beeswax. The melting latent heat increased by 22.5% compared to pure beeswax. The latent heat increase can be explained by Brownian motion and particle clustering mechanisms. Kim and Drzal [22] reported the latent heat of paraffin/xGnP composite PCMs did not decrease as up to 5 wt% exfoliated graphite nanoplatelet loading to paraffin because of good dispersion of exfoliated graphite nanoplatelet in paraffin with high surface area.

However, the enhanced thermal conductivity and non-decrescent latent heat have been simultaneously observed only for the composite PCMs based on low-molecular-weight organics, such as beeswax and paraffin. Unfortunately, most of the low-molecular-weight PCMs tend to induce explosion danger due to their higher vapor pressures and lower flash points compared with polymeric PCMs [15, 23, 24]. In addition, they are less reliable in term of their lower cycling stability during service in comparison with that of polymeric PCMs. The realization of enhanced thermal conductivity without adversely affecting the latent heat of polymeric PCMs remains challenging. On account of their relatively complex crystallization processes and highly imperfect crystals, the improvement mechanism of the thermal energy storage property of polymeric PCMs may be substantially different with those of low-molecular-weight organic PCMs. Theoretically, the crystallinities of PCM polymers could be optimized further through improved nucleation by using fillers as nucleating agents. However, a large amount of fillers usually utilized with semi-crystalline PCM polymers for good shape stability and high thermal conductivity actually could reduce the mobility of the chains and thus affect the kinetics of crystallization. As a result, both the crystallinity and the phase change enthalpy of the composites are found to decrease significantly. As understood from the literature survey, no case of breakthrough in polymer composite PCMs with both enhanced thermal conductivity and non-decrescent latent heat has been reported so far.

In this paper, we report experimental results demonstrating that PEG-based composite PCM exhibits significantly improved thermal conductivity and large latent heat beyond the expected value. The realization of enhanced thermal conductivity without adversely affecting the latent heat is achieved by optimizing the crystallinity of the PEG in PEG/graphene oxide (GO) composites. PEG-grafted GO (PEG-g-GO) is synthesized and introduced into the PEG to regulate the interaction between the PEG base and GO. The crystalline behaviors of the composite PCMs are verified by X-ray diffractometer (XRD) and polarized optical microscopy (POM), and the structures and thermophysical properties of the composite PCMs are analyzed by differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), and scanning electronic microscopy (SEM).

2 Experimental

2.1 Materials

Natural flake graphite (NG) with high purity of over 99.8% and an average particle size of 325 mesh was provided by XFNANO Materials Tech Co., Ltd. Sodium nitrate (NaNO3), hydrogen peroxide (H2O2), hydrochloric acid (HCl), PEG (Mn = 4000), and dimethyl formamide (DMF) were purchased from Sinopharm Chemical Reagent Co., Ltd. Concentrated sulfuric acid (H2SO4) was purchased from YATAI Chemical Reagent Co., Ltd. Potassium permanganate (KMnO4) was purchased from Wuhan Zhongtian Chemical Co., Ltd.

2.2 Fabrication of GO

GO is synthesized from NG powder by modified Hummers method with decreased oxidation time to avoid excessive oxidization and produce more hydroxyl groups. In this method, a solution of H2SO4 (85 ml) is added to a mixture of NG (2.5 g) and NaNO3 (1.875 g) and subsequently cooled below 20 °C. KMnO4 (12.5 g) is slowly added into the above mixture in 10 min with the aid of stirring. The reaction is kept stirring below 20 °C in ice bath for 30 min. Then, the mixture is heated to 35 °C and maintained for 2 h. Subsequently, distilled water (125 ml) is slowly added, resulting in a rapid rise in temperature to 90 °C. Then, 30 wt% of H2O2 (10 ml) is gradually added within 1 h. The color of the mixture changes from black to bright yellow. The suspension, while still warm, is filtered. Then, the GO is repeatedly washing using an about 5% solution of HCl for three times to remove metal ions. It is then washed continually with deionized water until the pH is nearly neutral. Finally, it is centrifuged to remove graphite which has not been oxidized and then dried in vacuum at 60 °C to constant weight.

2.3 Preparation of PEG-g-GO

PEG-g-GO is prepared by modifying the method originally proposed by Xu et al. [25]. The as-prepared GO is modified by an excess amount of TDI. In this synthesis procedure, GO (2 g) is loaded into a 100-mL three-necked flask equipped with a magnetic stirring bar, and 100 mL DMF dehydrated is then added under nitrogen to create a homogenous suspension. TDI (2 g) is next added and the mixture is stirred under nitrogen at 80 °C for 24 h. After that, the excessive PEG (5 g) is added into the slurry reaction mixture under nitrogen atmosphere and then reacted for 24 h at 80 °C. The suspension is poured into acetone (50 mL) to coagulate the product, followed by centrifugation and washing several times to remove the unreacted PEG. Finally, the resulting product is dried at 100 °C under vacuum for 48 h before use.

2.4 Fabrication of composite PCMs

PEG/GO and PEG/PEG-g-GO composite PCMs are prepared by a solution blending method. GO or PEG-g-GO nanosheets are first dispersed in DMF with the aid of sonication to form a homogeneous suspension. Then, PEG is added to the above suspension under stirring at 100 °C within 4 h. Finally, after being casted, the dispersion is first dried at 70 °C for 6 h, and subsequently peeled off the flat glass dish for further drying in vacuum at 100 °C for 48 h to constant weight.

2.5 Characterization

Transmission electron microscopy (TEM, Joel JEM-2001F) images are obtained by placing a few drops of the dispersion on a copper grid, and evaporating them at room temperature prior to observation. Infrared measurements are taken on a Fourier transform infrared (FTIR) spectrometer (Thermo Nicolet Nexus, America) in the range of 4000 to 400 cm−1 using KBr pellets. The X-ray measurements are carried out on a Rigaku D/Max IIIA X-ray generator with Cu Kα radiation at a wavelength of 1.54 Å. Specimens are scanned from 5° to 80° with a scan speed of 0.02°/min. POM observation is conducted on an Olympus BX51-P polarizing optical microscope equipped with a LTS 350 hot stage. DSC results are obtained using a DSC (PE Co., USA) at a heating and cooling rate of 5 °C min−1 at a temperature range from 20 to 80 °C under nitrogen atmosphere. To confirm the reproducibility of the results, at least each sample are measured for two DSC traces. TGA is carried out on a TG/DTA 220U (Seiko Instrument Co. Ltd., Tokyo Japan) with the Exstar 6000 Station. The samples are scanned from 20 to 800 °C with the heating rate of 10 °C/min and nitrogen gas purging. The thermal conductivities of the various samples are measured at 28 °C by the guarded heat flow test method (TA Instruments DTC 300, USA). The samples for thermal conductivity measurement are compact round disks of 50 mm in diameter and about 2 mm in thickness. The thermal conductivity testing is conducted in a direction perpendicular to the plane of the disks. SEM measurements are performed with a Hitachi S-4800 field emission electron microscope. The cryogenic fracture surfaces of the hot pressed samples are contrasted with gold before SEM observation.

3 Results and discussion

3.1 Morphology of the GO and PEG-g-GO flakes

A SEM image and two TEM images of the prepared GO platelets are shown in Fig. 1. The lateral dimensions of the GO flakes can reach several micrometers. The graphene flakes have relatively wrinkled surface texture, which could play a beneficial role in enhancing the interlocking of the flakes with each other and enabling interfacial connection with the PEG.
Fig. 1

SEM (a) and TEM (b, c) images of the GO flakes. Scale bar: a 1 μm. b 0.5 μm. c 100 nm

3.2 Chemical properties of the PEG-g-GO

Chemical characterization of the PEG-g-GO is carried out using FTIR spectroscopy and the result is shown in Fig. 2. In the spectrum of GO, the broad band at 3386 cm−1 and 1618 cm−1 are attributed to the stretching and flexural vibrations of O-H groups. The peak at 1719 cm−1 responds to the stretching vibrations of C=O of carbonyl. In comparison with those of GO and TDI, the FTIR spectrum of GO-TDI displays pronounced additional absorbance at around 1705 cm−1, 1651 cm−1, and 1534 cm−1 corresponding to the absorption peaks of amide and urea carbonyl group stretching vibrations as well as amide II band. The presence of strong peak at 2270 cm−1 for the GO-TDI spectrum confirms that the TDI molecules have reacted at one end with the PEGs and the as-synthesized GO-TDI has preserved unreacted active isocyanate groups at the other end. The strong absorption peak at 2270 cm−1 disappears in PEG-g-GO spectrum and a weak peak at 2887 cm−1 is ascribed to the stretching vibration of -CH- in the grafted PEG, suggesting the successful grafting of the PEG chains to the GO.
Fig. 2

FTIR spectra of GO, TDI, GO-TDI, and PEG-g-GO

3.3 Crystalline properties of PEG/PEG-g-GO composite PCMs

To reveal the crystalline properties of the composite PCMs, the XRD patterns of the GO, PEG, PEG-g-GO, PEG/GO, and PEG/PEG-g-GO are shown in Fig. 3. The typical diffraction peak of GO is observed at about 10.5°, which indicates that most of the GO are efficiently exfoliated. The XRD pattern of the PEG is characterized by two strong 2θ peaks at 19.1° and 23.6°, which is assigned to the typical plane of (120) and (112) of PEG. For PEG-g-GO, its WAXD peak of (001) crystal plane locates at 2 theta = 6.7°, which indicates the PEG grafting enlarges the interlayer space of the GO. Meanwhile, it is easily observed that the PEG-g-GO has a weak and broad diffraction peak at 2 theta = 23°, indicating an amorphous structure. The PEG/GO composites exhibit very similar information to the pristine PEG in the XRD patterns, which implies that they have similar crystal structure. However, an additional peak at 7.4° from the GO is also observed. On the other hand, the PEG-g-GO/GO also presents similar scattering patterns to those of the PEG or PEG/GO. Compared with the PEG-g-GO, the diffraction peak of GO of the PEG/PEG-g-GO changed to 2θ = 5.4°, and compounding of PEG-g-GO with PEG results in broadening GO diffraction peak and decreasing its intensity. These indicate that the GO layer spacing further increases and PEG-g-GO is dispersed in PCMs more evenly relative to the GO.
Fig. 3

XRD traces of the PEG-g-GO (a), PEG/GO (b) and PEG/PEG-g-GO (c)

The crystalline property of the composite PCMs is further investigated by POM and the POM images shown in Fig. 4 are taken below the crystallization temperature of the pure PEG and composite PCMs. In the images of the composite PCMs, the darker regions marked by arrows correspond to the GO or PEG-g-GO, whereas the lighter regions indicate the PEG. There is a large amount of excess PEG as shown in the POM images of the PEG/GO and PEG/PEG-g-GO and the saturated GO or PEG-g-GO sheets disperse in the excess PEG. Figure 4b reveals that GO aggregates are formed as a result of interparticle attraction due to hydrogen bonding and van der Waals forces. The PEG chains are strained as the GO particles migrate towards each other during crystallization, and thus, the crystal growth is inhibited. In case of the PEG/PEG-g-GO, a superior dispersion of the GO is confirmed owing to the plasticized effect of the grafted PEG. The addition of the PEG-g-GO does not weaken the crystallizability of the PEG, which is attributed to increased PEG chain mobility and nucleation effect of the finely dispersed grafted GO particles.
Fig. 4

POM images taken below the crystallization temperature of pure PEG (a), 5 wt.% GO/PEG composite (b), and 5 wt.% PEG-g-GO/PEG (c). Scale bar 100 μm

3.4 Thermal storage performance and thermal stability of the PEG/PEG-g-GO composite PCMs

The latent heats and phase change temperatures of the PEG/PEG-g-GO and PEG/GO samples are measured using DSC shown in Fig. 5 and the detailed results of phase transition properties are listed in Table 1. Figure 5 shows the melting and freezing DSC curves of the pristine PEG, PEG/PEG-g-GO, and PEG/GO composite PCMs. The detailed results of DSC thermal analysis of the peak melting/crystallization temperature (Tm/Tc), the measured melting/crystallization enthalpy (ΔHm/ΔHc), and the expected melting/crystallization enthalpy (ΔHm*/ΔHc*) obtained are presented in Table 1 and Fig. 5c. The expected melting/crystallization enthalpies of the PEG/PEG-g-GO and PEG/GO composites are calculated based on the impregnated contents of the PEG into the PEG-g-GO or GO and the fact that the GO or PEG-g-GO cannot contribute to the latent heat. The PEG-g-GO cannot contribute to the phase change latent heat due to the largely confined chain mobility of the grafted PEG by the GO sheets. The expected enthalpy value (ΔH*) can be calculated from the following relation:
$$ {\varDelta H}^{\ast }={\varDelta H}_{\mathrm{PEG}}\left[1-\left(1+{R}_{\mathrm{G}}\right){\omega}_{\mathrm{G}\mathrm{O}}\right] $$
where ΔHPEG represents the measured melting/crystallization enthalpy of the neat PEG, RG is the PEG grafting ratio or the mass fraction of PEG in the PEG-g-GO, and ωGO is the mass fraction of GO in the PEG/PEG-g-GO.
Fig. 5

DSC traces of the PEG/PEG-g-GO (a) and PEG/GO (b) with various PEG contents. Comparison between the measured and expected values of latent heat for PEG/PEG-g-GO, PEG/GO, and PEG (c)

Table 1

Phase transition temperatures and phase transition enthalpies of the PEG/PEG-g-GO and PEG/GO


Melting process

Crystallization process

Tm (on) °C

Tm (peak) °C

ΔHm (J/g)

ΔHm* (J/g)

Tc (on) °C

Tc (peak) °C

ΔHc (J/g)

ΔHc* (J/g)








− 169.92









− 168.88

− 160.36








− 154.93

− 150.80








− 138.93

− 141.25








− 159.55

− 160.62








− 136.86

− 152.16








− 124.98

− 143.71

The latent heat capacities of the composite PCMs containing the GO are obviously lower than that of the pristine PEG due to the incorporation of the GO that does not undergo phase change. However, in this study, the DSC results indicate that the mixture rule is typically invalid for this system. Most of the measured latent heats of the PEG/GO composites are evidently lower than their expected latent heats. This is partially expected given that some of the PCM weight is replaced by the GO that does not undergo phase change. Also, the phase transition of the PEG could be inhibited to a large extent due to the confined chain motion of the PEG induced by agglomeration of the GO, as discussed before. It is noteworthy that the melting enthalpy of the 95 wt% PEG/GO composite is slightly higher than the expected value. However, considering the typical error range of DSC measurement (i.e., 5%), the enhancement in latent heat of the nanocomposite is uncertain. Moreover, the PEG crystallization rate estimated based on the difference Tc(on) - Tc(peak) falls normally due to the confined PEG chain mobility, which indicates that the PEG crystallization is inhibited in the PEG/GO system. This should also occur because the crystal growth is retarded by the largely decreased interparticle space at higher GO contents [26] because a more serious aggregation may occur. Thus, the obviously reduced latent heat of phase change is detected by DSC (Table 1 and Fig. 5c).

On account of the aforementioned issues of the PEG/GO, PEG-g-GO is synthesized and utilized for crystallinity enhancement of the PEG base. By adding a proper loading of the PEG-g-GO, relatively high latent heats are attained. Notably, when the GO content in the composites is lower than 10 wt%, the melting enthalpies are significantly higher than the expected values. Introducing the PEG-grafted GO readily improved the dispersion of GO in the matrix, and thus remarkably elevated the PEG chain mobility in comparison to the ungrafted GO. Also, the grafted PEG chains, act as the plasticizer, inhibit the interaction between GO and PEG and heighten the mobility of the free PEG chains [27], leading to the decrease in the crystallization activation energy of the PEG/PEG-g-GO composites different from that of PEG/GO composites. The PEG crystallization rate raises as expected due to the enhanced PEG chain mobility. As shown in Table 1, the melting temperatures of these samples fall off with the increase of PEG-g-GO content, whereas those of the PEG/GO rise, which demonstrates that the addition of PEG-g-GO promotes the motion of the PEG chains. Consequently, the relative latent heats increase markedly. Remarkably, the melting enthalpy of the PEG/PEG-g-GO is much higher than expected value when the content of GO declined to 10 wt%. Moreover, when the content of GO decreases below 5 wt%, the latent heat of the composite is even higher than that of the pristine PEG (Fig. 5c). In view of the typical error range of DSC measurement (i.e., 5%), despite the enhancement in latent heat of the composite when compared with that of the neat PEG may be misleading, but this possibility of the enhancement cannot be excluded considering the enhanced crystallizability induced by increased PEG chain mobility.

The TG curves of the PEG/GO, GO, and pure PEG are shown in Fig. 6. Primary mass loss of GO occurs in the temperature range of 150~200 °C, which corresponds to the thermal decomposition of oxygen-containing functional groups on the GO surfaces. At the temperature above 200 °C, the GO shows a slow weight loss as a function of temperature due to release of oxygen away from GO leaving [28]. The degradation of the pristine PEG starts around 330 °C and ends at 440 °C. As can be seen from the curves, all the PEG/GO samples have two thermal degradation stages. The first decomposition stage of the PEG/GO takes place at the range of 200–350 °C, corresponding to the thermal decomposition of oxygen-containing functional groups on the GO surfaces. The major weight loss up to 80% in the range from 350 to 450 °C mainly originates from the pyrolysis of the blended PEG. Hydrogen-bonding interaction between the GO and PEG makes these composite PCMs better thermally stable over the pure PEG on account of the higher decomposition temperatures of the oxygen-containing functional groups on GO surfaces.
Fig. 6

TG curves of the PEG/GO (a) and PEG/PEG-g-GO (b) composites containing various GO loadings

The TG curves of the PEG-g-GO and PEG/PEG-g-GO are shown in Fig. 6b. The PEG-g-GO samples have three degradation stages. The loss below 150 °C should correspond to the evaporation of absorbed and interlayer water of the PEG-g-GO. The weight loss of the PEG-g-GO in the range from 150 to 380 °C would be due to release of oxygen away from GO leaving. The weight loss of the PEG-g-GO due to release of oxygen away from GO is significantly lower than that of GO, which is attributed to the conversion of oxygen-containing functional groups into relatively stable urethane groups (-NHCOO-). At the temperature above 380 °C, the major weight loss up to 80% mainly originates from the pyrolysis of the grafted PEG chains. The residual weight of PEG-g-GO is lower than that of GO, which is due to the complete removal of the grafted PEG at high temperatures. Hence, the grafting ratio of PEG-g-GO is calculated to be 12.5 wt%, taking into account the residue at 800 °C. The PEG/PEG-g-GO PCMs manifest similar decomposition stages with a relatively low weight loss rate peak temperature compared to that of the pure PEG. It indicates that the PEG-g-GO does not significantly affect the thermal stability of the PEG. This is attributed to the weaker interaction between the PEG-g-GO and the PEG base induced by the plasticized effect of the grafted PEG.

3.5 Thermal conductivities of the PEG/PEG-g-GO composite PCMs

Quantitative values of the thermal conductivities (κ) of the PEG/GO and PEG/PEG-g-GO composites are shown as functions of the GO weight fraction in the composites in Fig. 7. As expected, the thermal conductivities of both the composite PCMs tend to raise as the GO content increases. The thermal conductivities of the PEG/GO and PEG/PEG-g-GO at 28 °C reach 2.0 W/mK and 2.35 W/mK at 20 wt.% GO content, respectively. This value of κ is about 7.6 and 9.1 times higher than that of pristine PEG (0.263 W/mK). This increase can be attributed to the high thermal conductivities of the interconnected networks of the thermally conductive fillers that provide a path of lower resistance for phonons to travel [29, 30, 31, 32]. The composites with the PEG-g-GO fillers exhibit higher thermal conductivity enhancement as compared to the composite containing the GO fillers. This can be attributed to the large interfacial contact area as a result of the good compatibility between the PEG-g-GO and PEG as well as the sufficient dispersion of the PEG-g-GO particles. Hence, the heat transfer process is more effective in the PEG/PEG-g-GO than those in the ungrafted GO doped counterparts. These results suggest that the PEG-g-GO is a promising filler for the thermal conductivity enhancement of the PEG-based PCMs. It is feasible to add a small amount of PEG-g-GO to attain a higher energy storage performance, as well as to improve the thermal conductivity of the composite PCMs.
Fig. 7

Plots of thermal conductivity versus GO content of the PEG/GO and PEG/PEG-g-GO composite PCMs

3.6 Microstructure of the PEG/PEG-g-GO composite PCMs

The SEM images of freeze-fracture surfaces of the PEG/GO and PEG/PEG-g-GO composites, as well as the pure PEG, are shown in Fig. 8. Both the composites appear much rougher than that of the pure PEG. The image confirms the good affinity between the PEG and PEG-g-GO as well as sufficient dispersion of the PEG-g-GO in the PEG. It appears that the dispersion and distribution of the GO in the PEG are not uniform and ordered as the PEG-g-GO because some aggregations occur. An increase in thickness of the PEG-g-GO sheets is observed as intuitively expected due to the complete covering of the sheets by the PEG. There is a major difference in filler orientation between the PEG/GO and PEG/PEG-g-GO. That is, the PEG-g-GO particles are more oriented due to the plasticized effect of the grafted PEG in comparison with the GO. After the PEG-g-GO dispersion is cast, the PEG-g-GO particles with relatively high mobility tend to preferentially align parallel to the plane of the casted film. A more uniform dispersion of the PEG-g-GO without agglomeration in the mutually miscible PEG leads to the formation of a more efficient thermally conductive network, when compared with the PEG/GO.
Fig. 8

SEM images of freeze-fracture surfaces of the pure PEG (a) and PEG/GO (b) and PEG/PEG-g-GO (c) composite PCMs with 5 wt.% GO. Scale bar 2 μm

4 Conclusions

We investigated the thermal storage and conductive performances of PEG-based composite PCMs. PEG-g-GO is introduced into PEG to regulate the interaction between the PEG and GO. It is found that the composite PCM with the PEG-g-GO filler reveals a high thermal storage performance beyond expected value, ΔH = 174.91 J/g, while simultaneously providing a higher thermal conductivity when compared with the pure PEG. Namely, the realization of enhanced thermal conductivity without adversely affecting the latent heat is achieved by optimizing the interaction between the filler and PCM in the PEG-based composites. The result can be largely attributed to the enhanced GO dispersion and heightened mobility of the PEG chains. The grafted PEG in the PEG-g-GO acts as a plasticizer, leading to a considerable effect on the crystallization kinetics of the PEG in the composite PCMs, and thus the improved crystallizability. This strategy allows one to substantially enhance both the thermal storage and conductive performances based on polymeric PCMs. As a result, the PEG/PEG-g-GO composite PCMs can be considered as promising PCM candidates with large latent heats and high thermal conductivities for highly efficient thermal energy storage applications.


Funding information

We gratefully acknowledge the support by the National Natural Science Foundation of China (Nos. 51503158, 51803155).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflicts of interest.


  1. 1.
    Li Y, Xu G, Guo Y, Ma T, Zhong X, Zhang Q, Gu J (2018) Fabrication, proposed model and simulation predictions on thermally conductive hybrid cyanate ester composites with boron nitride fillers. Compos Part A-Appl S 107:570–578CrossRefGoogle Scholar
  2. 2.
    Feng CP, Bai L, Bao RY, Liu ZY, Yang MB, Chen J, Yang W (2018) Electrically insulating POE/BN elastomeric composites with high through-plane thermal conductivity fabricated by two-roll milling and hot compression. Adv Compos Hybrid Mater 1:160–167CrossRefGoogle Scholar
  3. 3.
    Sharma SD, Sagara K (2005) Latent heat storage materials and systems: a review. Int J Green Energy 2:1–56CrossRefGoogle Scholar
  4. 4.
    Tian Y, Zhao CY (2013) A review of solar collectors and thermal energy storage in solar thermal applications. Appl Energ 104:538–553CrossRefGoogle Scholar
  5. 5.
    Ruan K, Guo Y, Tang Y, Zhang Y, Zhang J, He M, Kong J, Gu J (2018) Improved thermal conductivities in polystyrene nanocomposites by incorporating thermal reduced graphene oxide via electrospinning-hot press technique. Compos Commun 10:68–72CrossRefGoogle Scholar
  6. 6.
    Sarier N, Onder E, Ozay S, Ozkilic Y (2011) Preparation of phase change material-montmorillonite composites suitable for thermal energy storage. Thermochim Acta 524:39–46CrossRefGoogle Scholar
  7. 7.
    Wang Y, Xia TD, Zheng H, Feng HX (2011) Stearic acid/silica fume composite as form-stable phase change material for thermal energy storage. Energ Buildings 43:2365–2370CrossRefGoogle Scholar
  8. 8.
    Fu X, Liu Z, Wu B, Wang J, Lei J (2016) Preparation and thermal properties of stearic acid/diatomite composites as form-stable phase change materials for thermal energy storage via direct impregnation method. J Therm Anal Calorim 123:1173–1181CrossRefGoogle Scholar
  9. 9.
    Feng L, Zheng J, Yang H, Guo Y, Li W, Li X (2011) Preparation and characterization of polyethylene glycol/active carbon composites as shape-stabilized phase change materials. Sol Energ Mat Sol Cells 95:644–650CrossRefGoogle Scholar
  10. 10.
    Karaipekli A, Biçer A, Sarıcd A, Tyagi VV (2017) Thermal characteristics of expanded perlite/paraffin composite phase change material with enhanced thermal conductivity using carbon nanotubes. Energ Convers Manage 134:373–381CrossRefGoogle Scholar
  11. 11.
    Yuan K, Zhou Y, Sun W, Fang X, Zhang Z (2018) A polymer-coated calcium chloride hexahydrate/expanded graphite composite phase change material with enhanced thermal reliability and good applicability. Compos Sci Technol 156:78–86CrossRefGoogle Scholar
  12. 12.
    Zhang Z, Zhang N, Peng J, Fang X, Gao X, Fang Y (2012) Preparation and thermal energy storage properties of paraffin/expanded graphite composite phase change material. Appl Energ 91:426–431CrossRefGoogle Scholar
  13. 13.
    Cao R, Liu H, Chen S, Pei D, Miao J, Zhang X (2017) Fabrication and properties of graphene oxide-grafted-poly(hexadecyl acrylate) as a solid-solid phase change material. Compos Sci Technol 149:262–268CrossRefGoogle Scholar
  14. 14.
    Yang J, Qi G, Liu Y, Bao R, Liu Z, Yang W, Xie B, Yang M (2016) Hybrid graphene aerogels/phase change material composites: thermal conductivity, shape-stabilization and light-to-thermal energy storage. Carbon 100:693–702CrossRefGoogle Scholar
  15. 15.
    Li H, Jiang M, Li Q, Li D, Chen Z, Hu W, Huang J, Xu X, Dong L, Xie H, Xiong C (2013) Aqueous preparation of polyethylene glycol/sulfonated graphene phase change composite with enhanced thermal performance. Energ Convers Manage 75:482–487CrossRefGoogle Scholar
  16. 16.
    Yavari F, Fard HR, Pashayi K, Rafiee MA, Zamiri A, Yu ZZ, Ozisik R, Borca-Tasciuc T, Koratkar N (2011) Enhanced thermal conductivity in a nanostructured phase change composite due to low concentration graphene additives. J Phys Chem C 115:8753–8758CrossRefGoogle Scholar
  17. 17.
    Silakhori M, Fauzi H, Mahmoudian MR, Metselaar HSC, Mahlia TMI, Khanlou HM (2015) Preparation and thermal properties of form-stable phase change materials composed of palmitic acid/polypyrrole/graphene nanoplatelets. Energ Buildings 99:189–195CrossRefGoogle Scholar
  18. 18.
    Mehrali M, Latibari ST, Mehrali M, Metselaar H, Silakhori M (2013) Shape-stabilized phase change materials with high thermal conductivity based on paraffin/graphene oxide composite. Energ Convers Manage 67:275–282CrossRefGoogle Scholar
  19. 19.
    Rufuss DDW, Suganthi L, Iniyan S, Davies PA (2018) Effects of nanoparticle-enhanced phase change material (NPCM) on solar still productivity. J Clean Prod 192:9–29CrossRefGoogle Scholar
  20. 20.
    Yang J, Zhang E, Li X, Zhang Y, Qu J, Yu Z (2016) Cellulose/graphene aerogel supported phase change composites with high thermal conductivity and good shape stability for thermal energy storage. Carbon 98:50–57CrossRefGoogle Scholar
  21. 21.
    Amin M, Putra N, Kosasih EA, Prawiro E, Luanto RA, Mahlia TMI (2017) Thermal properties of beeswax/graphene phase change material as energy storage for building applications. Appl Therm Eng 112:273–280CrossRefGoogle Scholar
  22. 22.
    Kim S, Drzal LT (2009) High latent heat storage and high thermal conductive phase change materials using exfoliated graphite nanoplatelets. Sol Energ Mat Sol Cells 93:136–142CrossRefGoogle Scholar
  23. 23.
    Li H, Jiang M, Li Q, Li D, Huang J, Hu W, Dong L, Xie H, Xiong C (2014) Facile preparation and thermal performances of hexadecanol/crosslinked polystyrene core/shell nanocapsules as phase change material. Polym Compos 35:2154–2158CrossRefGoogle Scholar
  24. 24.
    Zhang J, Li H, Tu J, Shi R, Luo Z, Xiong C, Jiang M (2018) Shape stability of polyethylene glycol/acetylene black phase change composites for latent heat storage. Adv Mater Sci Eng 2018:3954163Google Scholar
  25. 25.
    Xu C, Wu X, Zhu J, Wang X (2008) Synthesis of amphiphilic graphite oxide. Carbon 46(2):386–389Google Scholar
  26. 26.
    Lee S, Hong J, Jang J (2013) The effect of graphene nanofiller on the crystallization behavior and mechanical properties of poly(vinyl alcohol). Polym Int 62:901–908CrossRefGoogle Scholar
  27. 27.
    Xu J, Zhang Z, Xu H, Chen J, Ran R, Li Z (2015) Highly enhanced crystallization kinetics of poly(L-lactic acid) by poly(ethylene glycol) grafted graphene oxide simultaneously as heterogeneous nucleation agent and chain mobility promoter. Macromolecules 48:4891–4900CrossRefGoogle Scholar
  28. 28.
    Li H, Chen Z, Liu L, Chen J, Jiang M, Xiong C (2015) Poly(vinyl pyrrolidone)-coated graphene/poly(vinylidene fluoride) composite films with high dielectric permittivity and low loss. Compos Sci Technol 121:49–55CrossRefGoogle Scholar
  29. 29.
    Yang X, Liang C, Ma T, Guo Y, Kong J, Gu J, Chen M, Zhu J (2018) A review on thermally conductive polymeric composites: classification, measurement, model and equations, mechanism and fabrication methods. Adv Compos Hybrid Mater 1:207–230CrossRefGoogle Scholar
  30. 30.
    Liu Z, Shen D, Yu J, Dai W, Li C, Du S, Jiang N, Li H, Lin C (2016) Exceptionally high thermal and electrical conductivity of three-dimensional graphene-foam-based polymer composites. RSC Adv 6:22364–22369CrossRefGoogle Scholar
  31. 31.
    Qin M, Xu Y, Cao R, Feng W, Chen L (2018) Efficiently controlling the 3D thermal conductivity of a polymer nanocomposite via a hyperelastic double-continuous network of graphene and sponge. Adv Funct Mater 28:1805053CrossRefGoogle Scholar
  32. 32.
    Mehra N, Mu L, Ji T, Yang X, Kong J, Gu J, Zhu J (2018) Thermal transport in polymeric materials and across composite interfaces. Appl Mater Today 12:92–130CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Materials Science and Engineering, State Key Laboratory for New Textile Materials & Advanced Processing TechnologyWuhan Textile UniversityWuhanPeople’s Republic of China
  2. 2.Mechanical Metrology DivisionHubei Institute of Measurement and Testing TechnologyWuhanPeople’s Republic of China
  3. 3.School of Materials Science and EngineeringWuhan University of TechnologyWuhanPeople’s Republic of China

Personalised recommendations