D

J. Chem. Chem. Eng. 6 (2012) 515-519

DAVID

PUBLISHING

The Development of Thermal Conductive Polymer Composites for Heat Sink Yeo-Seong Yoon, Mee-Hye Oh*, Ah-Yeong Kim and Namil Kim Environmental Materials and Component R&D Center, Korea Automotive Technology Institute, Cheonan, Chungnam 330-912, South Korea

Received: March 27, 2012 / Accepted: April 25, 2012 / Published: June 25, 2012. Abstract: The thermal and mechanical properties of the polyamide 6/boron nitride and polyphenylene sulfide/graphite composites have been investigated as a function of composition and size of fillers. The addition of highly thermal conductive h-BN and graphite gives rise to large increase (about 2 times) of thermal conductivity of individual polymer. In PPS/graphite system, the higher conductivity value was obtained when smaller graphites were added. Meanwhile, the tensile and flexural strength are reduced upon increasing filler loading. Key words: Polymer composites, thermal conductivity, PA6 (polyamide 6), PPS (polyphenylene sulfide), h-BN (boron nitride), graphite.

1. Introduction As the electronic devices become denser and powerful, the heat evolved during operation becomes severe. The heat generated may cause degradation and subsequently reduce the reliability and performance of devices. Therefore, the management of excess heat becomes an important issue. In order to dissipate and spread the waste heat a heat sink retaining high thermal conductivity is commonly used. Even though the metals are the promising materials, in the electronic devices where weight is an important factor they are not suitable. There have been many efforts to replace the metallic materials, especially focusing on developing the polymer composites with light weight, ease of processing, flexibility in design, and high thermal conductivity in the range of 1 to 30 W/mK [1, 2]. The thermal conductivity of thermoplastics that are intrinsically not thermally conductive can be improved remarkably by adding highly conductive inorganic *

Corresponding author: Mee-Hye Oh, Dr., Ph.D., research fields: polymer composites, electrochemistry. E-mail: [email protected].

fillers such as graphite, carbon black, carbon fibers, ceramic, or metal particles [3-6]. The conductive polymer composites and their potential applications are dependent on the dimension (size and shape) of fillers, the amount of filler loading, and the degree of dispersion. Higher filler loading and close packing are typical requirements to achieve the appropriate level of thermal conductivity. However, there is no distinct improvement when the disparity in thermal conductivity between polymers and fillers is more than 100 times [5]. The high content of inorganic fillers, however, may reduce the process ability of the composites. Therefore, a composite with high thermal conductivity at low filler loading is highly desired. The effect of inorganic fillers on mechanical properties should also be taken into account. Several techniques have been proposed to measure the thermal conductivity of polymer composites [7, 8]. Steady state method is one of mainly used techniques. It measures the temperature difference across the specimens in response to the external heating source. The conductivity value may be either absolute or

516

The De evelopment of o Thermal Co onductive Po olymer Composites for He eat Sink

relative wiith respect to the refference mateerial maintained in series or o parallel to the sam mple. Regarding non-steady n staate methods, hot h wire/or plate, p temperature wave, and laser flash are the typpical examples [9]. In presennt paper, thhe thermal and mechannical properties of o polyamidee 6 (PA6)/booron nitride and polyphenyleene sulfide (P PPS)/graphite composites have h been measurred for the potential p appllication as a heat sink. Regarrding polymeer matrix, thhe PA6 and PPS were selecteed because off high thermall stability.

2. Experim ment 2.1 Materialls and Methods PA6 annd glass fiber-reinforcced PPS-G GF30 (polyphenylene sulfide) supplied s by Hyosung H Co Ltd. Asia plasticss were used as a a polymer matrix. Theyy are semicrystalline polymerss, revealing a melting poinnt at around 220 and 280 oC, C respectivelly. Two diffeerent inorganic fillers, f hexaggonal boron nitride (h--BN, Momentive Performancee Materials Innc.) and grapphite (KS 150 and KS 500, TIMCAL T Ltdd.) were addeed to PA6 and PPS-GF30, P respectively, to adjust the mechanical and thermal properties. p Inn order to rem move the residuall moisture, both b polymerr and filler were w o dried in a vaacuum at 60 C for 1 dayy prior to mixxing. The amount of inorganic fillers was varied v from 0w wt% to 40wt% with w respect to the polym mer content. The mixtures weere homogeniized in a twin screw extruuder (Model STS S 32, Han koook E. M. Ltdd.) equipped with w side feeder. The specim men for tenssile and flexxural strength tesst was preppared by staandard injecction molding (M Model LGH 200 N, LG Cable Ltd.). Since S latent heat is evolved during d the exxtrusion proccess, temperature control insside barrel is of particcular importance.

Co. LTD.). The particle sizee and size diistribution off ng LS™ 2000 the inorganic fiillers were analyzed usin Series (Beckmann Coulter, IN NC.). The sizee distributionn wass obtained byy monitoring tthe diffusion of laser lightt indu uced by parrticles. The hh-BN and grraphite weree disp persed in DI (de ionized)) water first and a a couplee drop ps of disperssant (Coulter Anionic Dispersant Typee IIA A) were addeed. The analyysis was con nducted afterr ultrrasonificationn treatment foor 10 min. Th he change off mecchanical propperties including tensile (A ASTM D638)) and d flexural streength (ASTM M D790-3 po oint bending)) wass investigatedd using a uuniversal testting machinee (SF FM-10, Uniited Calibraation Co.). Test wass perfformed at ambient temperature and a relativee hum midity of 60% %. The shappe and size of o h-BN andd grap phite were fuurther analyzeed and characcterized usingg SEM M (scanning ellectron microsccopy, JSM-670 01, JEOL). Thermal T connductivity oof the com mposite wass meaasured at diifferent com mpositions an nd presentedd usin ng a therm mal conductivvity meter (Unitherm™ ™ Model 2022, Anter A Corp) (Fig. 1). It follows thee AST TM E1530 guarded g heatt flow meterr method. Att steaady state, thee Fourier heaat flow equaation may bee desccribed as: ⁄ (1)) whiich indicates that the therm mal resistancce (Rs) of thee specimen is dettermined from upper (Tu) and lowerr platte surface tem mperature (T Tm), heat flux x through thee testt sample (Q)), and total interface reesistance btnn sam mple and surfface plates (R Rint). From th he resistancee valu ue obtained from f Eq (1), the thermal conductivityy () of the samplle having a coonstant thick kness (d) mayy ⁄ . be evaluated e from m

2.2 Characteerization Followingg the ASTM M D 792, thhe SG (speccific gravity) of polymer p and its composites was measuured using electroonic densimetter (MD-300 S, ALFA Miirage

Fig.. 1

Scheme off thermal cond ductivity testing machine.

The Development of Thermal Conductive Polymer Composites for Heat Sink

3. Results and Discussion Fig. 2 shows the particle size and size distribution of h-BN and graphite, i.e., KS150 and KS500. The h-BN exhibits a broad size distribution in a range of 2~80 m with the average particle size of 15 m (Fig. 2a). Similarly, KS150 and KS500 also reveal a broad distribution, but the size of KS500 particles (average

517

80/20 PPS-GF30/graphite composition, the lower strength value is obtained with larger graphite, i.e., 85.4 MPa for PPS-GF30/KS150 and 75.3 MPa for PPS-GF30/KS500. The flexural strength of PA6/h-BN is also slightly decreased upon increasing fillers content. In the case of PPS-GF30/graphite system, the

size ~ 300 m) is almost over three times that of KS150 particles (average size ~ 100 m). As shown in SEM photomicrographs ( 300), the aggregation of plane micro-particles (plate-like shape) is clearly discernible in both h-BN and graphites. The thermal conductivity of the composites is varied depending on the shape and the extent of inter-connectivity of inorganic fillers (network density) inside a polymer matrix. The fillers with platelet shape are advantageous over the spherical or cylindrical one in that they readily overlap each other with large contact area and moreover facilitate high filler loading. The weight of heat sink is an important factor that determines the energy efficiency and performance of the devices. Fig. 3 depicts the effect of filler loading on the SG (specific gravity) of the composites. Specific gravity of PA6/h-BN composites, as denoted by “◇”, is increased as the h-BN loading is increased from 30 wt% to 40 wt% (Fig. 3). It is associated with relatively high density of h-BN (SG~2.1) compared to PA6 (1.13-1.15). The SG of PPS-GF30/KS150 composite (“△”) is also increased from 1.78 wt% at 20 wt% to 1.81 wt% at 30 wt% upon increasing filler loading. Considering the high specific gravity of metallic materials such as aluminum (~ 2.7) and copper (~ 8.9), it is inferred that the weight of heat sink is reduced more than 40%-50% by simply replacing the materials. The mechanical properties of PA6- and PPS-based composites have been measured at different polymer/filler composition. As shown in Fig. 4, the tensile strength decreased as the mount of fillers increased. Upon increasing fillers content, they tend to aggregate, which in turn reduce the tensile strength. At

Fig. 2 Size distribution and microscopy images of (a) h-BN, (b) KS150, and (c) KS500 graphite fillers.

518

The Development of Thermal Conductive Polymer Composites for Heat Sink

particle size has little influence on the flexural strength, exhibiting almost the same value of about 110 MPa at 80/20 composition (Fig. 5). The thermal conductivity of pure PA6 and PPS is increased significantly by the incorporation of highly thermal conductive fillers, where the inter-connected

inorganic fillers may provide the pathway for the heat flow. However, the maximum loading of inorganic fillers is restricted because of the limited processing capability and expensive prices. As shown in Fig. 6, the thermal conductivity of neat PA6 and PPS is increased more than 100% upon adding inorganic fillers. A conductivity of 0.52 and 0.75 W/mK were achieved at 70/30 PA6/h-BN and 80/20 PPSGF30/KS150 composition. These values are further increased upon increasing the fillers content, i.e., 0.66 W/mK at 60/40 PA6/h-BN and 1.25 W/mK at 70/30

Fig. 3 The effect of filler loading on the specific gravity of PA6/h-BN (◇) and PPS-GF30/KS150 ( ) composites.

Fig. 4 The variation of tensile strength of PA6/h-BN (A: 70/30, B: 60/40), PPS-GF30/KS150 (C: 80/20, D: 70/30), and PPS-GF30/KS500 (E: 80/20) composites.

Fig. 6 The variation of thermal conductivities of (a) PA6/h-BN and (b) PPS-GF30/KS150 composites as a function of composition and temperature. Table 1 The effect of inorganic fillers on thermal conductivity of 70/30 PPS/graphite composites (W/mK). Components Fig. 5 The variation of flexural strength of PA/h-BN (A: 70/30, B: 60/40), PPS-GF30/KS150 (C: 80/20) and PPS-GF30/KS500 (D: 80/20) composites.

PPS-GF30/KS150 PPS/KS150 PPS-GF30/KS500

30 1.252 0.826 0.774

Temperature (oC) 60 90 1.354 1.366 0.833 0.812 0.783 0.775

110 1.362 0.806 0.773

The Development of Thermal Conductive Polymer Composites for Heat Sink

PPS-GF30/KS150. The improved thermal conductivity may be attained due to highly contact inorganic fillers and therefore formation of filler network. When the samples were heated, the thermal conductivity of PA6/h-BN and PPS-GF30/graphite composites show an opposite behavior. The former reveals a slightly lower value, while the latter shows higher thermal conductivity. In general, the thermal conductivity of composites increases upon heating. The slight reduction in PA6/h-BN composite may be attributable to the increase of distance between adjacent conductive fillers or the presence of voids at high temperature. The higher thermal conductivity in PPS-based composite compared to PA6-based one indicates that the graphites are dispersed better in a polymer matrix. As shown in Table 1, the PPS-based composite displays a better thermal conductivity when glass fiber (GF30) is reinforced. At 70/30 PPS/KS150, the conductivity is improved more than 50% in the presence of glass fiber, i.e., 1.252 W/mK for PPS-GF30/KS150 and 0.826 W/mK for raw PPS/KS150. Although the glass fibers are not thermally conductive, they may provide the conductive pathway by inducing aggregation of graphites on the surface. The use of finer particles leads to a higher thermal conductivity because of larger surface area, 1.252 W/mK for PPS-GF30/KS150 and 0.774 W/mK for raw PPS-GF/KS500. The effect of glass fiber and fillers size on thermal conductivity is summarized in Table 1.

loading should be optimized depending on its applications. Thermal conductivity of composites was improved when the polymer was reinforced with glass fiber presumably due to the formation of denser networks. The elucidation of exact mechanism is still under investigation.

Acknowledgments This technology development was supported by Parts. Materials Technology Development Project of Ministry of Knowledge Economy, Republic of Korea.

References [1]

[2]

[3]

[4] [5]

[6] [7]

4. Conclusions The effect of h-BN and graphite fillers on thermal conductivity and mechanical properties of polymer composites has been investigated. The addition of inorganic fillers improved the thermal conductivity more than 50%, but reduced the mechanical properties of polymer composites. Therefore, the amount of filler

519

[8]

[9]

Zhou, W. Y.; Qi, S. H.; Li, H. D.; Shao, S. Y. Study on Insulating Thermal Conductive BN/HDPE Composites. Thermochim. Acta. 2007, 452, 36-42. Nayak, K. C.; Saha, S. K.; Srinivasan, K.; Dutta, P. A Numerical Model for Heat Sinks with Phase Change Materialsand Thermal Conductivity. Inter. J. Heat and Mass Transfer 2006, 49, 1833-1844. Xu, Y. S.; Chung, D. D. L. Increasing the Thermal Conductivity of Boron Nitride and Aluminum Nitride Particle Epoxy-Matrix Composites by Particle Surface Treatments. Compos. Interfaces 2000, 7, 243-256. Chung, D. D. L. Materials for Thermal Conduction. Appl. Thermal Eng. 2001, 21, 1593-1605. Lee, G. -W.; Park, M.; Kim, J. K.; Lee. J. I.; Yoon. H. K. Enhanced Thermal Conductivity of Polymer Composites Filled with Hybrid Filler. Composites A 2006, 37, 727-734. Tu, H. M.; Ye, L. Thermal Conductive PS/Graphite Composites. Polym. Adv. Technol. 2009, 20, 21-27. Tritt, T. M.; Weston, D. Measurement Techniques and Considerations for Determining Thermal Conductivity of Bulk Materials. Inthermal Conductivity Theory, Properties, and Applications. Tritt, T. M. Ed.; Kluwer Academic/Plenum Publishers: New York, 2004; pp 187-203. Rides, M.; Morikawa, J.; Halldahl, L.; Hay, B.; Lobo, H.; Dawson, A.; et al. Intercomparison of Thermal Conductivity and Thermal Diffusivity Methods for Plastics. Polym. Test. 2009, 28, 480-489. Han, Z. D.; Fina, A. Thermal Conductivity of Carbon Nanotubes and Their Polymernanocomposites: A Review. Prog. Polym. Sci. 2011, 36, 914-944.