graphene (GN) composite material

Applied Thermal Engineering 140 (2018) 13–22

Contents lists available at ScienceDirect

Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Improving thermal management of electronic apparatus with paraffin (PA)/ expanded graphite (EG)/graphene (GN) composite material

T

⁎

Tao Xua, Yantong Lib, Jiayu Chenb, , Huijun Wua, Xiaoqing Zhoua, Zhengguo Zhangc a

Academy of Building Energy Efficiency, School of Civil Engineering, Guangzhou University, Guangzhou 510006, China Department of Architecture and Civil Engineering, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong c Key Laboratory of Enhanced Heat Transfer and Energy Conservation, The Ministry of Education, School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China b

H I GH L IG H T S

PA-EG-GA composite PCM. • AThenovel new material has better thermal properties. • Testsproposed examined the heat transfer properties of the new material. • An experiment simulative electronic chips. • The new materialforhas better thermal management capacity. •

A R T I C LE I N FO

A B S T R A C T

Keywords: Phase change material Heat storage Thermal management Paraffin-EG-graphene Graphene nanoplatelets

This study introduced a novel PA-EG-GN composite phase change material (PCM) with better thermos-physical properties to help dissipate heat in electronic apparatus. Both the X-ray diffractometer and Fourier transformation infrared spectra patterns show that the composite PCM is the pure physical combination of no chemical interactions. Raman spectroscopy results suggest that the structural symmetry of the GN decreases with vibration increases. Results from Scanning Electron Microscopy show that GN can improve the compatibility with the mixture of PA and EG. Differential scanning calorimetry curves indicate that the composite PA-EG-GN has a lower latent heat than that of pure PA and PA-EG. The weight-loss ratios of the PA-EG and PA-EG-GN are roughly equivalent to the mass ratio of PA in the composite PCMs. The thermal conductivity of the PA-EG-GN is evidently higher than PA-EG and shows a strong linear relationship with the compress density. To verify these conclusions, an experiment was conducted to compare the thermal management capabilities of the PA-EG composites and the PA-EG-GN composites with several simulative chips. Both the surface peak temperature and the apparent heat transfer coefficient of chips were measured. The final results confirmed that the PA-EG-GN has a better performance than regular PA-EG composites.

1. Introduction In the informational era, there are two emerging trends in electronics development. One is faster computational performance and the other is higher chip integration level [1–3]. Both trends lead to higher heat flux that generated by the electronics, which is the major challenge for the electronics industry [4]. Traditional cooling methods, such as natural and single-phase forced convection, are not efficient in the next generation of electronic chips due to their high heat fluxes [5]. Thus, researchers proposed many new technologies for the electronic apparatus cooling, such as heat pipes [6], energy selective electron devices

⁎

Corresponding author. E-mail address: [email protected] (J. Chen).

https://doi.org/10.1016/j.applthermaleng.2018.05.060 Received 15 March 2018; Received in revised form 20 April 2018; Accepted 14 May 2018 1359-4311/ © 2018 Elsevier Ltd. All rights reserved.

[7,8], thermoelectric cooler [9,10], and etc. Apart from aforementioned technologies, phase change materials (PCM) attract increasing attention, given its simplicity, high reliability, and low power consumption [11,12]. Besides, the large latent heat of PCM can also be utilized to maintain the temperature of the electronic apparatus within a stable range, which ensures a safer work environment [13]. Therefore, PCM for active and passive electronic cooling apparatus has been intensively investigated in recent years. For example, Wang et al. [14] discussed the influence of properties of paraffin and porosity of metal on heat dissipation and optimized the performance of phase change thermal control apparatus. Kandasamy et al.

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

[15] analyzed the influences of component direction, input power, and cool-heat cycle time on the thermal management of electronic apparatus. Baby et al. [16] reported that utilizing fins in the heat sinks that encapsulated with PCM could drastically enhance the performance of electronic devices. Gharbi et al. [17] found that the alliance of the PCM and well-space fins could effectively control the temperature of the electronic devices in an appropriate range. Wu et al. [18] concluded that phase change material board performs better than natural cold air cooling for electronics. Jaworski et al. [19] designed a new heat spreader filled with PCM which could improve the cooling effect on the electronic devices. Samimi et al. [20] simulated the thermal management performance of the battery cell using the carbon fiber-PCM composites. They reported that the utilization of the composite PCM with higher mass fraction carbon fibers contributed to an even temperature distribution within the battery cell. Although PCM has been proven effective in controlling the temperature of electronic apparatus, the low thermal conductivity of PCM is still the largest barrier to its wider application. To overcome this limitation, researchers have developed many methods to create the composite PCM with high thermal conductivity, such as embedding dispersing metallic particles into pure PCMs [21–23], blending PCMs with nanoparticles [24,25], and adding carbon materials into pure PCMs [26,27]. Among these approaches, using carbon materials cannot only effectively reduce the weight and cost of the energy storage systems, but also be naturally compatible with PCMs [28]. EG is one of the most popular carbon materials, which is intensively utilized to improve the thermal conductivity of PCMs, such as LiNO3-KCl-NaNO3 [29], LiNO3-KCl [30], stearic acid [31], polyethylene glycol [32], and palmitic-stearic acid eutectic salts [33]. Previous studies [28,34,35] show that adding the EG can effectively enhance the thermal conductivity of PA. Another typical carbon material is GN, which is also widely applied to enhance the thermal conductivity of PCMs, such as palmitic-stearic acid [36], palmitic acid-polypyrrole [37], and lauric acid [38]. However, few studies reported about integrating both GN and PA-EG to further enhance the thermal conductivity of the PCMs. In the meantime, the thermal conductivity of PA-EG could be improved with the increase of EG, but the latent heat of PA-EG also would decline dramatically. Thus, only little nanoscale GN with superior thermal conductivity easily absorbed into the holes of EG could enhance greatly the thermal conductivity of PA in the holes of EG and decrease the storage capability decline rate. In this paper, the novel composite PA-EG-GN was proposed and prepared to improve the thermal conductivity of the PCMs. To examine the thermal-physical properties of the new composite, the pure PA, PAEG, and PA-EG-GN were compared and analyzed. The X-ray Diffraction (XRD) was used to characterize the crystalline phases of the PCMs. The Fourier Transform Infrared Spectroscopy (FTIR) spectra of the PCMs was recorded by the KBr disk method. The Raman spectrometer was used to observe the Raman spectroscopy patterns of EG and GN. Atomic Force Microscopy (AFM) was used to observe the shape of GN. The Scanning Electron Microscope (SEM) was used to observe the microstructures of the PCMs. The thermogravimetric analysis (TGA) was conducted with a thermal analyzer. The Differential Scanning Calorimetry (DSC) was used to measure the phase change temperatures and latent heat of the PCMs. The thermal constant analyzer was used to measure the thermal conductivity of the PCMs. Also, a validation experiment on simulative electronic chips was conducted to compare the thermal performance using the PA-EG-GN with that using the PA-EG.

Table 1 Physical properties of GN. Parameters

Value

Diameter Thickness Carbo Content Density Thermal Conductivity Specific Surface Area

5–15 μm 1–5 nm > 99 wt% 0.5 g/cm3 ∼5000 W/(m·K) 100 m2/g

from 48 °C to 50 °C) was chosen as the base organic PCM. EG was selected as the inorganic supporting material for the paraffin phase change material. EG can be produced through expansion treatment of the raw expandable graphite (average particle size: 500 μm, expansion ratio: 300 ml/g, from Qingdao Hengsheng Graphite Co., Ltd) in a microwave oven (Midea Inc, China) with a power of 800 W for 30–40 s. The EG was used to absorb a mass of the mixture of the PA and GN. Table 1 shows the physical properties of the graphene nanoplatelets (Zhuhai Lingxi New Material Technology Co., Ltd, China). The material composition process includes three steps. Firstly, the PA was melted by water bath heating at 80 °C. Then, the GN was put into the liquid paraffin with a roll mixer to ensure mixing uniformity. Finally, the EG was dissolved in the liquid. After natural cooling, the new PA-EG-GN composite material can be acquired. Based on the conclusion of a previous study [35], the maximum amount of paraffin (melting point from 48 °C to 50 °C) can be absorbed by EG is about 85.6 wt%. Thus, the PA-EG composites with a fixed EG mass fraction of 80% were chosen in this study to prevent the paraffin leakage. To investigate the relationship between the mass fraction of GN and the thermal-physical properties of the composite PCM, a series of the composite PCM samples with different GN mass fractions were prepared. Table 2 shows the mass fractions of composite samples.

2.2. Characterization Crystalline phases of EG, GN, PA, and PA-EG-GN composite PCM were characterized by an X-ray diffractometer (XRD, D8 Advance, Bruker, German) with Cu-Kα irradiation (k = 1.5406 Å) accelerating voltage and 40 mA currents. Fourier transformation infrared (FT-IR) spectra of EG, GN, PA and PA-GN-EG composite PCM were recorded between 400 and 4000 cm−1 on a spectrometer (Tensor 27, Bruke, Germany) using the KBr disk method. Raman spectroscopy patterns of EG and GN were obtained by using the Raman spectrometer (LabRAM Aramis, HJY Inc., France). The excitation line at 632.8 nm was emitted by the AR ion laser. The laser power and scanning time of this ion laser were respectively 20 mW and 20, and then when it normally works. The shape of GN, especially the thickness, was observed by using the Atomic Force Microscopy (Veeco Multimode, America). The microstructures of EG, GN, PA-EG, and PA-EG-GN were observed using a scanning electron microscope (SEM) (S3700N, Hitachi Lnc., JNP). The phase change temperatures and latent heat of PA, PAEG and PA-EG-GN were measured by a differential scanning calorimeter (DSC 214 Polyma, NETZSCH, Germany) under a nitrogen atmosphere from 10 to 80 °C. The sample mass was controlled in the range of 5–10 mg and the heating/cooling rate was set at 2 °C/min. Furthermore, the sample of PA-EG-GN was placed in a high and low temperature heat

2. Methodology and experiment design

Table 2 Mass fractions of composite samples.

2.1. Preparation of the composite PCM The proposed composite PCM composes of base organic PCM and inorganic supporting materials. PA (Shanghai Huayong paraffin Co., Ltd, China) with appropriate phase change temperature (melting point 14

Sample Name

Mass Fraction of PA-EG Composite

Mass Fraction of GN

PA-EG PA-EG-GN

PA: EG = 80%: 20% PA: EG = 80%: 20%

PA-EG:GN = 1: 0.0% PA-EG:GN = 1: 5.0%

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

1 mm

a

4 mm

38 mm

34 mm

44 mm 48 mm Fig. 1. Round plates with the same thickness.

b

chamber (SG-80L-CC-2, SANWOOD, China) to perform heating–cooling cycles, and then was measured by DSC under the same conditions. In order to investigate the thermal stability of PA, PA-EG, and PAEG-GN, the thermogravimetric analysis (TGA) used a thermal analyzer (TG 209F3, NETZSCH, Germany) in the range of 30–400 °C at a heating rate of 10 °C/min in a nitrogen atmosphere. The thermal conductivities of the prepared round plates with different compress densities were measured by a hot disk thermal constant analyzer (TPS2500, Hot Disk Inc., Sweden). The analyzer adopts a transient plane source method. The probe was placed between two round plates with the same thickness as shown in Fig. 1, and then heat source began to work for a preset scanning time with constant power. The round plates of PA-EG and PAEG-GN have respectively 60 g (PA = 48 g and EG = 12 g) and 63 g (PA = 48 g, EG = 12 g, and GN = 3 g).

48 mm

2.3. Assessment of the thermal management performance Fig. 3. Schematic diagram of the heat storage unit.

To assess the thermal management performance of the proposed composite PCM, an experiment was designed to investigate the thermalphysical properties of the materials during the process of thermal energy storage. The experiment aims to reveal the impact of adding GN into the composite on its heat transfer behavior. Fig. 2 shows the schematic diagram of the experimental system. The system includes an incubator, a heat storage unit, a simulative chip, a voltage output operation box, and a temperature monitoring device. The temperature of the incubator (LRH-150, Guangdong Xinteng Co., Ltd, China) has a built-in close-loop control system so that the temperature will be automatically adjusted to the destined temperature with a tolerance of ± 1 °C. During the heat storage process, the

incubator was set to 21 °C. To mimic the real thermal management performance of the composites PCM block (44 × 34 × 44 mm) for electronic chips, the samples were encapsulated in an alloy aluminum hollow radiator (48 × 38 × 48 mm), forming a heat storage unit as shown in Fig. 3. A simulative electronic chip (16 × 16 × 2 mm) was agglutinated to the heat storage unit by the thermal silicone grease with a thermal conductivity of 4.95 W/(m·K) and a thickness of 0.5 mm. In order to ensure the same thickness, the same weight of the thermal silicone grease was put in the center of the simulative chip and squeezed by the alloy

Fig. 2. Schematic diagram of the experimental system. 15

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

d

d Transmittance (%)

Intensity (a.u.)

c

c

b

b a

a 10

20

30

40

50

500

60

1000

1500

2000

2500

3000

3500

4000

-1

Wavenumber (cm )

2 (°)

Fig. 5. FTIR spectra of (a) EG, (b) GN, (c) PA and (d) PA-EG-GN.

Fig. 4. XRD patterns of (a) EG, (b) GN, (c) PA and (d) PA-EG-GN.

aluminum hollow radiator being confined to one location when the test species were changed. The heating power of the simulative electronic chip (Guangzhou Zhize Electronic Technology CO., Ltd, China) can be adjusted through applying various working voltages with an output operation box. In the experiment, the simulative chip was heated for 4000 s and then cooled for another 4000 s. The voltage variations that generated by the thermocouples were automatically recorded by a data logger (Agilent 34970A, USA). The data logger is able to associate and convert the collected voltage signals to digital temperatures with a margin of errors within ± 1 °C. A K-type thermocouple was placed at the top center of the simulative chip.

Intensity(a.u.)

G

D GN

EG

3. Results and discussion 3.1. XRD analysis

1000

1200

1400

1600

Raman shift / cm

Fig. 4 shows the characteristic XRD patterns of EG, GN, PA and PAEG-GN composites. The peak located at 26.4° in EG is attributed to its regular crystal structure. The peak and XRD pattern of EG is similar to GN. The XRD curve of PA contains two main peaks located at 21.5° and 23.9°. Also, the XRD pattern of PA-EG-GN composite PCM consists of EG, GN, and PA, indicating that the crystal structures of EG, GN, and PA remain unchanged. Thus, the PA-EG-GN composite PCM is a physical integration of PA, EG, and GN.

1800

2000

-1

Fig. 6. Raman spectra of EG and GN.

(1332 cm−1), appears. The G-peak, corresponding to the E2g phonon vibration of the center, belongs to an instinct Raman pattern of the EG. The D-peak, denoting the A1g phonon mode of the K-point which locates at the Brillouin region of the EG, is the Roman feature of the disorder induction of the carbon material [39]. Therefore, the sp2 hybridized carbon atoms of the EG arrange very tightly and neatly. The G-peak of the GN is weaker and broader than that of the EG, but the D-peak intensity of the GN is stronger than that of the EG. It means that the structural symmetry of the GN decreases and vibration pattern increases. Part of sp2 hybrid carbon atoms in its structure has converted into the sp3 hybrid carbon atoms, resulting in the destruction of the C]C double bonds in the EG. Besides, the intensity ratio between the D-peak and G-peak (ID/IG) also reflects the number ratio between sp3 and sp2 carbon atoms. The intensity ratio of the EG (ID/IG = 1.28) is higher than that of the GN (ID/IG = 0.052), which indicates that graphitic layer fracture and the area of this layer evidently decreases. The average size of the sp2 hybrid carbon plane in the GN is larger than that in the EG, and the disorder degree of the GN increases. Fig. 7 shows the AFM image of GN. It can be seen that the GN nanosheet evenly distributes on the basement of the mica sheet in a state of the small stack. The uniform color of the sample is clearly distinguished with the basement color. Stankovich et al. [40,41] had proved from both the theoretical and experimental perspectives that, the thickness of the GN with a single layer which is prepared by the Liquid Phase Chemical Reduction method was approximately 1.1 nm.

3.2. FT-IR spectra analysis Fig. 5 displays the FTIR spectra of the EG, GN, PA and PA-EG-GN composites. The peaks (at 3457 and 1638 cm−1) of GN are attributed to the eOH stretching vibrations and C]O stretching vibrations, respectively. The curve of EG is similar to the curve of GN. The peaks (at 2918 and 2849 cm−1) of PA are assigned to the stretching vibration of CH3 and CH2, respectively. Besides, the asymmetric stretching vibrations of CH2 band is in accordance with the strong absorption peak located at 1465 cm−1 and the rocking vibration of CH2 led to a strong absorption peak, which located at 719 cm−1. On the other hand, the absorption peaks on the PA and PA-EG-GN composites curve contain almost all the characteristic absorption peaks of EG, GN, and PA without showing any new absorption peaks. As a result, it could be concluded that there was no chemical reaction among EG, GN, and PA. 3.3. Raman spectra analysis and Atomic force microscopy Fig. 6 shows the Raman Spectra of EG and GN. Two diffraction peaks, the sharp and strong G-peak (1576 cm−1) and the weak D-peak 16

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

Fig. 7. AFM images of GN.

According to the AFM results in this study, the average thickness of the GN sample is nearly 1.1 nm. This indicates that the layer of the prepared GN sample is mainly single, and some two-layers or three-layers structure probably appear in the sample.

This indicates that GN has the larger surface area which leads to improved compatibility for PA and EG mixtures.

3.4. SEM analysis

The phase change characteristics of the composite PCM samples, including the phase change temperature and phase change latent heat, were inspected by DSC. The DSC curves of paraffin and composite PCMs are presented in Fig. 9. The corresponding test values are listed in Table 3. The pure PA and PA-EG composite PCM were introduced as the references to investigate the impacts of adding GN on the thermal properties of the material. All composite PCMs have the similar DSC curve shapes with that of the pure paraffin, which suggests the paraffin is the critical ingredient to store the latent thermal energy during the phase change process. As shown in Table 3, although the phase change

3.5. DSC analysis

EG has large volume worm-like microstructure with rich pores, which are suitable to be filled with various materials [42,43]. Fig. 8 displays the Scanning Electron Microscopy (SEM) images of the EG, GN, PA-EG, and PA-EG-GN. It can be seen that the EG consists of flattened irregular honeycomb network made of elementary graphite sheets in Fig. 8a. Fig. 8b shows GN has a laminate structure. As shown in Fig. 8c–d, it can be clearly seen that PA and GN had been easily absorbed into the porous network of EG. By comparing Fig. 8d with Fig. 8c, it is found that the pores of EG were filled up because of GN.

Fig. 8. SEM images of the EG, GN and composite PCM: (a) EG, (b) GN, (c) PA-EG, (d) PA-EG-GN. 17

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

0.8

GN is respectively 47.3 °C and 49.8 °C higher than the pure PA. The reason is that the PA is coated inside the micropores of the EG, and the surface tension and capillary force are required to be overcome due to the volatilization of the gases produced by the decomposition of the PA. Thus, the starting point of weightlessness will rise for PA-EG-GN and higher than PA-EG. Such phenomenon is caused by the huge specific surface area and scale effect (as shown in Fig. 8d), which can well fill the micropores of the EG. Therefore, the surface tension and capillary force are enhanced. Secondly, the decomposition temperature of the PA, PA-EG, and PAEG-GN are respectively 219 °C, 305.5 °C, and 300.9 °C, which means that adding EG improves the decomposition temperature and reduces the maximum weight-loss ratio. This result is also caused by the surface tension and capillary force are required to be overcome due to the volatilization of the gases produced by the decomposition of the PA. As the temperature increases, the mass of the gases will increase, and the pressure in the micropores will increase. The increase of the pressure will restrain the thermal decomposition of the PA, which will delay the decomposition reaction to the higher temperature. Meanwhile, the decomposition temperature of PA-EG-GN is lower than that of PA-EG. This results from the good thermal conductivity of GN, which can accelerate the breakdown of the molecular chain. Thirdly, the maximum weight-loss ratio of the PA, PA-EG, and PAEG-GN are respectively 14.94%, 10.47%, and 8.82%, which means that the maximum weight-loss ratio of the PA is the highest, and that of PAEG-GN is the lowest. The main reason is that the micropores of the EG slow down the volatilization of the gases produced by the decomposition of the PA. Meanwhile, the GN can effectively reduce the scale of the micropores, which increases the surface tension and capillary force of the EG. Therefore, the volatilization of the gases can be slowed down greatly. Finally, the spanning range of the PA, PA-EG, and PA-EG-GN are respectively 176.0 °C, 184.7 °C and 188.8 °C, which means that the maximum spanning a range of the PA-EG-GN is the highest, and that of PA is the lowest. The main reason is also due to the characteristics of the GN and micropores of EG.

Heat Flow (mW/mg)

Exo 0.4

0.0

PA PA-EG PA-EG-GN

-0.4

-0.8 10

20

30

40

50

60

70

80

Temperature (ºC) Fig. 9. DSC curves of paraffin and composite PCMs.

temperatures are approximately same, the phase peak temperatures of all composite PCMs drop slightly. The temperature deviations are also displayed in Fig. 9, which suggests that the endothermic and exothermic peaks of all composite PCMs are narrower than that of the paraffin. This phenomenon can be attributed to the porous EG network and high thermal conductivity of GN. EG can provide heat conduction path in the paraffin and GN can improve the heating and cooling rates, which can accelerate the phase transition speed of PCM. Since the phase transition processes of EG and GN does not occur in the process, the latent heat decreases with an increase of the mass fraction of EG and GN. The formula for calculating phase change latent heat is expressed as Eq. (1):

ΔHPEG = (1−wEG %−wGN %)ΔHPA

(1)

where ΔHPEG and ΔHPA stand for the calculated values of composite PCMs and the latent heat of PA respectively, WEG% and WGN% are the mass fraction of EG and GN. Besides, the experimental phase change latent heats of the composites agree with its calculated values with relative errors less than 5%.

3.7. Thermal conductivity analysis For PCM, thermal conductivity is a key parameter because it reflects heat transfer rate. The composite PCMs were put into steel shell mould to make the cylinder composite PCMs by tablet press (DY-40, Tianjin Keqi Co. Ltd, China). The compress densities of the cylinder composite PCMs samples could be calculated using the following formula:

3.6. Thermal stability analysis In order to investigate the thermal stabilities of the PCMs, TG (Thermogravimetric Analysis) and DTG (Derivative Thermogravimetric Analysis) were adopted to investigate the mass loss of the PCMs under different temperatures. Fig. 10 displays the mass loss of PA, PA-EG, and PA-EG-GN. It can be seen from Fig. 10a that the weight-loss ratios of PA-EG and PA-EG-GN are approximately 80.6% and 75.1%, which are roughly equivalent to the mass ratio of PA in the composite PCMs. This is due to the fact that PA can be decomposed more easily than EG and GN. Also, the mass ratio of PA in PA-EG is higher than that in PA-EGGN. Table 4 summarizes the thermal stability characteristics presented in Fig. 10. With Fig. 10 and Table 4, we have following observations. Firstly, it can be seen from this table that the starting point of PA-EG and PA-EG-

ρ=

4m πd 2h

(2)

where m was the weight of the cylinder samples which could be measured using an analytical balance with a precision of 0.001 g, d and h were the diameters and heights of the cylinder samples respectively which could be measured using a vernier caliper with a precision of 0.02 mm. The thermal conductivities of the cylinder samples were measured using the previously described method in this paper. Fig. 11 shows the thermal conductivity comparison between PA-EG and PA-EG-

Table 3 Phase change properties of the paraffin and composites. Type

PA PA-EG PA-EG-GN

Temperature (°C)

Phase change latent heat

Phase peak temperature

Phase temperature

Calculated Value (J/g)

Test Value (J/g)

Relative Error (%)

Heat

Cool

Heat

Cool

Heat

Cool

Heat

Cool

Heat

Cool

53.54 53.17 53.07

52.37 51.48 51.54

46.72 46.18 46.16

53.78 53.97 53.98

– 165.6 157.3

– 166.6 158.2

207.0 162.4 159.1

208.2 164.9 158.5

– 1.92 1.14

– 1.02 0.19

18

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

Fig. 10. TGA curves of paraffin and composite PCMs: (a) TG and (b) DTG.

GN. It can be seen that all of the thermal conductivities increase linearly with the rising compress density due to the reduction of loose space within the composite PCMs and the expansion of the area of contact with all particles [44]. The relationship can be express by the following equation: For PA-EG:

y = 0.0135x −1.8256(R2 = 0.9888)

(3)

For PA-EG-GN:

y = 0.0155x −2.4392(R2 = 0.9897)

(4)

where y is the thermal conductivity, x is the compress density. The linear equations between the thermal conductivity and the compress density of EG-based composite PCMs have been studied in previous works [28,45,46]. The increasing trend could be attributed to the reduction of micropore volume of the porous expanded graphite and the enlargement of the contact area. It can be noticed that the thermal conductivity of PA-EG-GN at the same compress density is higher than that of PA-EG, which could be attributed to the higher heat conduction and special structure of GN. GN has extremely high thermal conductivity with super thin 2-D nanolayer structure and special wrinkle surface, which can easily form thermal conductivity bridge between PA and 3-D microstructure EG. The schematic of GN acting as a bridge between PA and EG as shown in Fig. 12. In addition, the existence of GN can enlarge the contact area and decrease interface thermal resistance between PA and EG.

Fig. 11. Thermal conductivity comparison between PA-EG and PA-EG-GN.

3.8. Thermal management performance of the electronic component To investigate the thermal management ability of electronic component with the composite PCM, heating power was provided in the following experiment. The PA-EG and PA-EG-GN composite PCMs with the same higher compression density of 760 kg/m3 were selected for the ease of comparison in order to prevent PA squeezed from the micro hole of EG. The surface temperature variations of the simulative chips with PA-EG and PA-EG-GN composite PCMs are plotted in Fig. 13. The figure illustrates how the temperature reaches equilibrium after the heat was transferred to PCM and then dissipated into the air by the

Fig. 12. Schematic of GN acting as a bridge between PA and EG.

Table 4 Thermal stability characteristics of PA, PA-EG, and PA-EG-GN. Material

Starting point of weightlessness (°C)

Terminal point of weightlessness (°C)

Spanning range (°C)

Maximum weight-loss ratio (%/min)

Decomposition temperature (°C)

PA PA-EG PA-EG-GN

102.3 149.6 152.1

279.1 334.3 340.9

176.0 184.7 188.8

14.94 10.47 8.82

219.9 305.5 300.9

19

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

60

40

55

36 34 32 PA-EG PA-EG-GN

30 28

50

Temperature(0C)

Temperature (0C)

38

26

45 40 35

PA-EG PA-EG-GN

30

24 25

22

20

20 0

1000

2000

3000

0

4000

1000

2000

Time (s)

Time (s)

(a) 5W

(b) 10W

3000

4000

80

Temperature ( C)

90

60

0

Temperature (oC)

100 70

PA-EG PA-EG-GN

50 40

80 70 60

PA-EG PA-EG-GN

50 40

30

30 20

0

1000

2000

3000

20

4000

0

1000

Time (s)

3000

4000

(d) 20W

120

140

100

120

Temperature (0C)

Temperatue (0C)

(c) 15W

80

PA-EG PA-EG-GN

60

2000

Time (s)

40

100

PA-EG PA-EG-GN

80 60 40

20

20 0

1000

2000

3000

4000

Time (s)

0

1000

2000

3000

4000

Time (s)

(e) 25 W

(f) 30W

Fig. 13. Thermal management performance comparison between using PA-EG and PA-EG-GN with different heat power.

heat sink. It can also be observed from the figure that the temperature development in Fig. 13a and b are distinctive from others. When the power is low, the chip temperature is not able to reach the phase change temperature of PCM and more heat will be released into the air. This explains why Fig. 13a and b have no buffer/stair through temperature development. All figures show that PA-EG-GN composite PCMs have better conductivity, as its heat dissipation speed is higher. In Fig. 13c–f, the curves exhibit a three-step behavior, which is similar to the temperature variation in different types of PCM latent thermal energy storage (LTES) systems [43]. In the first step, the

surface temperature of the simulative chip rises sharply without phase transition of PCMs. In the second step, PCMs gradually change phases and absorb a large amount of heat. Finally, after the completion of the phase change, the temperature becomes stable. It also can be seen that the equilibrium temperature of the composite with GN is lower than that without GN. This is due to the thermal conductivity improvement by adding GN into PCMs. In addition, PA-EG-GN composite PCMs reach the third stage earlier than PA-GE composite PCMs, because they contain fewer PCMs. One interesting case is Fig. 13c, the power is large enough to exhibit the three-stage temperature development but too low 20

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

the mass ratio of PA in the composite PCMs. Both the starting point of PA-EG and PA-EG-GN is higher than the pure PA, and the spanning range of the PA-EG-GN is higher than that of the pure PA and PA-EG. The thermal conductivity analysis results indicate that the thermal conductivity of the PA-EG-GN is evidently higher than PA-EG. The strongly linear relationship between the thermal conductivity with the compress density is presented. An experiment was also conducted to compare the performance of thermal management in simulative electronic chips with the PA-EG and that with the PA-EG-GN. Both the surface peak temperature and the apparent heat transfer coefficient were measured for the chips. The results suggest that the PA-EG-GN has a higher thermal conductivity and performs better than the PA-EG. When the compress density increases, the thermal conductivity increases. The relationships between them show a strong linear correlation. The experiment results suggest that the surface peak temperature of the simulative chip with PA-EGGN was lower than that of PA-EG. Besides, the apparent heat transfer coefficient of the cooling system with PA-EG-GN was also higher than that with PA-EG. Both conclusions proof the thermal conductivity of the composite PA-EG-GN is higher than the PA-EG for the thermal management in the electronic apparatus. Therefore, these excellent thermal properties of the proposed composites offer a better thermal management capacity for electronic apparatus.

Fig. 14. Surface peak temperature comparison between using PA-EG and PAEG-GN.

to show obvious improvements in temperature reduction and heat dissipation. Since the composite with higher conductivity can absorb and dissipate the heat generated by the simulative chip at a higher speed, the variation trend of the temperature of the simulative chip with GN is larger than that without GN during the phase change process. It can also be observed from Fig. 13 that the peak temperature of the simulative chip with GN is lower than that of the simulative chip without GN. Based on the heat conduction-dominated mechanism, the thermal energy storage process with higher conductivity coefficient can reach thermal equilibrium faster and obtain lower peak temperature. The comparison of the surface peak temperature between using PAEG and PA-EG-GN is shown in Fig. 14. It can be seen that the temperature increases when the heating power increases, and they show a strong linear relationship. This relationship can be presented as: For PA-EG:

T = 19.206 + 3.8031P (R2 = 0.9995)

Acknowledgment This work was supported by Guangzhou Science and Technology Program (201704030137), the Research Project of Guangdong Province (2017A050506058), and the Shenzhen Science and Technology Funding Programs (JCYJ20150318154726296). References [1] Z. Ling, Z. Zhang, G. Shi, X. Fang, L. Wang, X. Gao, Y. Fang, T. Xu, S. Wang, X. Liu, Review on thermal management systems using phase change materials for electronic components, Li-ion batteries and photovoltaic modules, Renew. Sustain. Energy Rev. 31 (2014) 427–438. [2] J. Yang, H. Wu, M. Wang, Y. Liang, Prediction and optimization of radiative thermal properties of nano TiO2 assembled fibrous insulations, Int. J. Heat Mass Transf. 117 (2018) 729–739. [3] Y. Liang, H. Wu, G. Huang, J. Yang, H. Wang, Thermal performance and service life of vacuum insulation panels with aerogel composite cores, Energy Build. 154 (2017) 606–617. [4] P. Naphon, D. Thongkum, P. Assadamongkol, Heat pipe efficiency enhancement with refrigerant–nanoparticles mixtures, Energy Convers. Manage. 50 (2009) 772–776. [5] Y. Wang, X. Gao, P. Chen, Z. Huang, T. Xu, Y. Fang, Z. Zhang, Preparation and thermal performance of paraffin/Nano-SiO2 nanocomposite for passive thermal protection of electronic devices, Appl. Therm. Eng. 96 (2016) 699–707. [6] H. Peng, J. Li, X. Ling, Study on heat transfer performance of an aluminum flat plate heat pipe with fins in vapor chamber, Energy Convers. Manage. 74 (2013) 44–50. [7] J. He, X. Wang, H. Liang, Optimum performance analysis of an energy selective electron refrigerator affected by heat leaks, Phys. Scr. 80 (2009) 035701. [8] G. Su, Y. Pan, Y. Zhang, T.-M. Shih, J. Chen, An electronic cooling device with multiple energy selective tunnels, Energy 113 (2016) 723–727. [9] S. Manikandan, S.C. Kaushik, Energy and exergy analysis of an annular thermoelectric cooler, Energy Convers. Manage. 106 (2015) 804–814. [10] Y. Cai, D. Liu, F.-Y. Zhao, J.-F. Tang, Performance analysis and assessment of thermoelectric micro cooler for electronic devices, Energy Convers. Manage. 124 (2016) 203–211. [11] Y. Li, G. Huang, H. Wu, T. Xu, Feasibility study of a PCM storage tank integrated heating system for outdoor swimming pools during the winter season, Appl. Therm. Eng. 134 (2018) 490–500. [12] Y. Li, G. Huang, T. Xu, X. Liu, H. Wu, Optimal design of PCM thermal storage tank and its application for winter available open-air swimming pool, Appl. Energy 209 (2018) 224–235. [13] Y. Li, Y. Du, T. Xu, H. Wu, X. Zhou, Z. Ling, Z. Zhang, Optimization of thermal management system for Li-ion batteries using phase change material, Appl. Therm. Eng. 131 (2018) 766–778. [14] J. Wang, Z. Qu, W. Li, W. Tao, T. Lu, Experimental research on the encapsulation of the radiator using the metal foam with phase change material, J. Eng. Thermophys. 32 (2011) 295–298. [15] R. Kandasamy, X.-Q. Wang, A.S. Mujumdar, Application of phase change materials in thermal management of electronics, Appl. Therm. Eng. 27 (2007) 2822–2832. [16] R. Baby, C. Balaji, Experimental investigations on phase change material based

(5)

For PA-EG-GN:

T = 19.317 + 3.5914P (R2 = 0.9995)

(6)

where T is the surface peak temperature; P is the heating power. The surface peak temperature of the simulative chip with GN under different heating power is lower than that without GN. The slope of the sample with GN is lower than that without GN. As a result, the temperature control capacity of the sample with GN is superior to that without GN. 4. Conclusions This research proposed to utilize a novel PA-EG-GN composite PCM with better thermal management for the electronic apparatus. The pure PA, PA-EG, and PA-EG-GN materials were compared to examine their thermal-physical properties. XRD patterns of the PA-EG-GN indicate that the crystal structures of EG, GN, and PA remain unchanged, and therefore the composite PCM is a physical combination of PA, EG, and GN. FTIR spectra suggest that no chemical reaction among EG, GN, and PA occurs in the composite. Raman spectroscopy patterns indicate that the structural symmetry of the GN decreases, and vibration pattern increases. AFM image of the GN shows that the layer of the prepared GN sample is mainly single. SEM analysis indicates that GN can improve the compatibility between PA and EG. DSC analysis suggests that the composite PA-EG-GN has a lower latent heat than that of the pure PA and PA-EG. In addition, thermal stability analysis concludes that weight-loss ratios of the PA-EG and PA-EG-GN are roughly equivalent to 21

Applied Thermal Engineering 140 (2018) 13–22

T. Xu et al.

[17]

[18]

[19] [20]

[21] [22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

finned heat sinks for electronic equipment cooling, Int. J. Heat Mass Transf. 55 (2012) 1642–1649. S. Gharbi, S. Harmand, S.B. Jabrallah, Experimental comparison between different configurations of PCM based heat sinks for cooling electronic components, Appl. Therm. Eng. 87 (2015) 454–462. W. Wu, G. Zhang, X. Ke, X. Yang, Z. Wang, C. Liu, Preparation and thermal conductivity enhancement of composite phase change materials for electronic thermal management, Energy Convers. Manage. 101 (2015) 278–284. M. Jaworski, Thermal performance of heat spreader for electronics cooling with incorporated phase change material, Appl. Therm. Eng. 35 (2012) 212–219. F. Samimi, A. Babapoor, M. Azizi, G. Karimi, Thermal management analysis of a Liion battery cell using phase change material loaded with carbon fibers, Energy 96 (2016) 355–371. Y. Tian, C.Y. Zhao, A numerical investigation of heat transfer in phase change materials (PCMs) embedded in porous metals, Energy 36 (2011) 5539–5546. Y. Li, J. Li, Y. Deng, W. Guan, X. Wang, T. Qian, Preparation of paraffin/porous TiO2 foams with enhanced thermal conductivity as PCM, by covering the TiO2 surface with a carbon layer, Appl. Energy 171 (2016) 37–45. P. Chen, X. Gao, Y. Wang, T. Xu, Y. Fang, Z. Zhang, Metal foam embedded in SEBS/ paraffin/HDPE form-stable PCMs for thermal energy storage, Sol. Energy Mater. Sol. Cells 149 (2016) 60–65. S. Wu, D. Zhu, X. Zhang, J. Huang, Preparation and melting/freezing characteristics of Cu/paraffin nanofluid as phase-change material (PCM), Energy Fuels 24 (2010) 1894–1898. P.D. Myers, T.E. Alam, R. Kamal, D.Y. Goswami, E. Stefanakos, Nitrate salts doped with CuO nanoparticles for thermal energy storage with improved heat transfer, Appl. Energy 165 (2016) 225–233. R.D. Weinstein, T.C. Kopec, A.S. Fleischer, E. D’Addio, C.A. Bessel, The experimental exploration of embedding phase change materials with graphite nanofibers for the thermal management of electronics, J. Heat Transfer 130 (2008) 042405. Y. Hamada, W. Ohtsu, J. Fukai, Thermal response in thermal energy storage material around heat transfer tubes: effect of additives on heat transfer rates, Sol. Energy 75 (2003) 317–328. Z. Zhang, N. Zhang, J. Peng, X. Fang, X. Gao, Y. Fang, Preparation and thermal energy storage properties of paraffin/expanded graphite composite phase change material, Appl. Energy 91 (2012) 426–431. T. Xu, Y. Li, J. Chen, J. Liu, Preparation and thermal energy storage properties of LiNO3-KCl-NaNO3/expanded graphite composite phase change material, Sol. Energy Mater. Sol. Cells 169 (2017) 215–221. Z. Huang, Z. Luo, X. Gao, X. Fang, Y. Fang, Z. Zhang, Investigations on the thermal stability, long-term reliability of LiNO3/KCl – expanded graphite composite as industrial waste heat storage material and its corrosion properties with metals, Appl. Energy 188 (2017) 521–528. G. Fang, H. Li, Z. Chen, X. Liu, Preparation and characterization of stearic acid/ expanded graphite composites as thermal energy storage materials, Energy 35 (2010) 4622–4626. W. Wang, X. Yang, Y. Fang, J. Ding, J. Yan, Preparation and thermal properties of

[33]

[34]

[35] [36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

22

polyethylene glycol/expanded graphite blends for energy storage, Appl. Energy 86 (2009) 1479–1483. N. Zhang, Y. Yuan, Y. Du, X. Cao, Y. Yuan, Preparation and properties of palmiticstearic acid eutectic mixture/expanded graphite composite as phase change material for energy storage, Energy 78 (2014) 950–956. C. Chen, H. Zhang, X. Gao, T. Xu, Y. Fang, Z. Zhang, Numerical and experimental investigation on latent thermal energy storage system with spiral coil tube and paraffin/expanded graphite composite PCM, Energy Convers. Manage. 126 (2016) 889–897. Z. Zhang, X. Fang, Study on paraffin/expanded graphite composite phase change thermal energy storage material, Energy Convers. Manage. 47 (2006) 303–310. Y. Yuan, N. Zhang, T. Li, X. Cao, W. Long, Thermal performance enhancement of palmitic-stearic acid by adding graphene nanoplatelets and expanded graphite for thermal energy storage: A comparative study, Energy 97 (2016) 488–497. M. Silakhori, H. Fauzi, M.R. Mahmoudian, H.S.C. Metselaar, T.M.I. Mahlia, H.M. Khanlou, Preparation and thermal properties of form-stable phase change materials composed of palmitic acid/polypyrrole/graphene nanoplatelets, Energy Build. 99 (2015) 189–195. S. Harish, D. Orejon, Y. Takata, M. Kohno, Thermal conductivity enhancement of lauric acid phase change nanocomposite with graphene nanoplatelets, Appl. Therm. Eng. 80 (2015) 205–211. A.C. Ferrari, J. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. Novoselov, S. Roth, Raman spectrum of graphene and graphene layers, Physical review letters 97 (2006) 187401. S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T. Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide, Carbon 45 (2007) 1558–1565. I. Jung, M. Pelton, R. Piner, D.A. Dikin, S. Stankovich, S. Watcharotone, M. Hausner, R.S. Ruoff, Simple approach for high-contrast optical imaging and characterization of graphene-based sheets, Nano Lett. 7 (2007) 3569–3575. Y. Zhong, S. Li, X. Wei, Z. Liu, Q. Guo, J. Shi, L. Liu, Heat transfer enhancement of paraffin wax using compressed expanded natural graphite for thermal energy storage, Carbon 48 (2010) 300–304. S. Kim, L.T. Drzal, High latent heat storage and high thermal conductive phase change materials using exfoliated graphite nanoplatelets, Sol. Energy Mater. Sol. Cells 93 (2009) 136–142. Z. Huang, X. Gao, T. Xu, Y. Fang, Z. Zhang, Thermal property measurement and heat storage analysis of LiNO3/KCl–expanded graphite composite phase change material, Appl. Energy 115 (2014) 265–271. Z. Ling, J. Chen, T. Xu, X. Fang, X. Gao, Z. Zhang, Thermal conductivity of an organic phase change material/expanded graphite composite across the phase change temperature range and a novel thermal conductivity model, Energy Convers. Manage. 102 (2015) 202–208. T. Xu, Y. Li, J. Chen, J. Liu, Preparation and thermal energy storage properties of LiNO3-KCl-NaNO3/expanded graphite composite phase change material, Sol. Energy Mater. Sol. Cells 169 (2017) 215–221.