Advanced
A Review on Thermal Conductivity of Polymer Composites Using Carbon-Based Fillers : Carbon Nanotubes and Carbon Fibers
A Review on Thermal Conductivity of Polymer Composites Using Carbon-Based Fillers : Carbon Nanotubes and Carbon Fibers
Carbon letters. 2010. Dec, 11(4): 347-356
Copyright ©2010, Korean Carbon Society
  • Received : August 08, 2011
  • Accepted : November 11, 2011
  • Published : December 30, 2010
Download
PDF
e-PUB
PubReader
PPT
Export by style
Share
Article
Author
Metrics
Cited by
TagCloud
About the Authors
Jinho Hong,
Dong Wha Park
Sang Eun Shim
seshim@inha.ac.kr
Abstract
Recently, the use of thermal conductive polymeric composites is growing up, where the polymers filled with the thermally conductive fillers effectively dissipate heat generated from electronic components. Therefore, the management of heat is directly related to the lifetime of electronic devices. For the purpose of the improvement of thermal conductivity of composites, fillers with excellent thermally conductive behavior are commonly used. Thermally conductive particles filled polymer composites have advantages due to their easy processibility, low cost, and durability to the corrosion. Especially, carbon-based 1-dimensional nanomaterials such as carbon nanotube (CNT) and carbon nanofiber (CNF) have gained much attention for their excellent thermal conductivity, corrosion resistance and low thermal expansion coefficient than the metals.This paper aims to review the research trends in the improvement of thermal conductivity of the carbon-based materials filled polymer composites.
Keywords
1. Introduction
Heat release from the electronic devices by excellent thermal conduction is involved in the use of heat sink to dissipate heat from an electronic package, the operation of heat exchanger, the melting of ice on the runway, the cooking and the numerous industrial processes. For effective thermal conduction, thermal interface materials or thermally conductive composites must have excellent thermal conductivity across which heat transfer occurs with the low contact resistance. Without good thermal contacts, the use of a thermal conductive material is just cost-consuming [1] . Most polymer resins are known to be thermal insulators. In insulators, heat transport is governed by the motion of free particles, called “phonon”. For insulating materials, heat transfer by the mobile electrons is ignored which dominates the thermal conductivity of metals [2] . Thermally conductive polymer composites offer the new, easy and economic way to improve the heat dissipation from the heat source. Thermally conductive polymer composites have much potential than the traditional thermally conductive metals in terms of the good resistance to corrosion, light and processibility for their purposes [3] .
Beside thermally conductive polymeric solid composites,thermal interface materials (TIM) as the excellent thermal conducting materials play very important roles in electronics [4] . Today, as the electronic products are chasing light, pocketsized and large scale integration, thermal problems occurring inside of products bring about more troubles [5] . Thermal management of these products is very important for design, packaging and housing of goods. Generally, commercialized thermal conductive pads or films are using the metal powder or ceramic particles in the polymeric matrix for using their high thermal conductive properties [1] . High performance materials need the superior thermal conducting behavior for their high exothermic character.
For improving the thermal conductivity of a polymeric material, polymer matrices are usually filled with the thermally conductive fillers like metal powders, ceramic powders and carbon-based materials. General thermal conductivity of polymer resins is in the range from 0.1 to 0.3W/m·K. Some common fillers for enhancing thermal conductivity are 234W/m· K for aluminum, 400W/m ·K for copper and 600W/m ·K for graphite. In Table 1 , the thermal conductivities of conventional thermally conductive fillers are listed [1] . For using the polymer composites as a thermal interface material, the thermal conductivity of the fabricated composites should have approximately 1 to 30W/m·K [6 , 7] .
Many researchers have used the metal oxides, metals and ceramics such as alumina (Al 2 O 3 ), diamond, silicon carbide (SiC), and zinc oxide (ZnO) as the thermal dissipative filler of the composite materials [1] . Goh et al ., prepared the Al 2 O 3 and ZnO reinforced silicone rubber where about 70%of the thermal conductivity enhancement was achieved at 10 vol% of the filler loading [8] . Kou et al . explored the effect of the particle size of alumina on the thermal
Thermal conductivity of conventional thermal conductive fillers and composites[1]
Lager Image
Thermal conductivity of conventional thermal conductive fillers and composites [1]
conductivity of silicone rubber. Thermal conductivity of the prepared composite was improved with alumina particles with a smaller diameter at the same volume fraction [9] . Kim et al . reported the linear increase of the thermal conductivity of water and ethylene glycol containing the alumina, zinc oxide, and titanium dioxide particles. [10] . The improvement of thermal conductivity has been explored for various systems including boron nitride (BN) in polyethylene matrix, ZnO in silicone rubber, metal oxide fillers in ethylene vinyl acetate (EVA) for the solar cell, modified ZnO in EVA, and ZnO in ethylene glycol [11 - 15] .
Carbon-based fillers are known as very ideal thermal conductive fillers for polymer composites because they have a higher thermal conductivity, excellent corrosion resistance, and a lower thermal expansion coefficient than metals. When the matrix and filler are both carbon materials, it is called carboncarbon composite. The carbon-carbon composite shows very high thermal conductivity and it is most suitable to use as heat sink. But the carbon-carbon composite is very expensive, so the application of carbon-carbon composites to industrial purpose is confronted with difficulties. For this reason, many researchers are working on the development of cheap polymer composites using carbon-based materials as the filler.
Graphite, carbon fiber, diamond and carbon nanotube are frequently explored as the thermal conductive fillers in polymer-carbon composites. After the discovery of carbon fiber (CNF), most of the works are focused on the use of CNF as thermal, electrical, and mechanical reinforcing filler to improve the polymer characteristics [16 , 17] . Recently, an article shows the 150% of thermal conductivity enhancement with 1 wt% of CNF in polymer matrix [16] . But the
Lager Image
3-D schematic demonstration of nanocomposite systems: (a) with spherical fillers (such as Al2O3 and SiO2) (b) with rodlike filler (such as clay or carbon nanotubes) [18].
polymer-CNF composite still lacks the models for predicting the effective thermal conductivity for their unusual structure.
A significant amount of work has been conducted by using the 1-dimensional carbon fillers like carbon nanotube (CNT) and carbon nanofiber (CNF). These fillers enhance not only the thermal conductivity, but also the electrical, and mechanical properties. 3-Dimensional microstructure of the nanofiller loaded polymer composites is shown in Fig. 1 [18] . When the nanofillers are dispersed in the soft polymeric matrices, mechanical properties of the composites can be enhanced by a reinforcing effect from transferring the loads. When the polymer composites filled with spherical filler (aspect ratio of the particle is equal to unity), the modulus of the composite depends on the modulus, density, size, shape, volume ratio, and number of the incorporated particles. The 1-dimensional filler may connect more polymer chains and afford more effective load transfer, leading to an improvement of mechanical properties. Upon adding a critical volume fraction of 1-dimensional filler, the filler stretches and perturbs the polymer matrix, generating elongated domains that are reinforced by these fillers [19] . 1-Dimensional fillers can easily transfer the loads or reduce it than spherical fillers. Here, the uniform dispersion of CNT or CNF in the polymer matrix is the main concern. Welldispersed polymer composites filled with CNT or CNF can be achieved by using the fine twin-screw extrusion, surfactant, oxidation of fillers and incorporation of functional groups on the surface of the fillers.
2. Carbon Nanotube (CNT) Incorporated Polymer Composites
- 2.1. Thermal conductivity of CNT-polymer composites
CNT incorporated polymer composites with enhanced properties have been widely explored in the materials science and engineering, especially in the reinforcing and conducting behavior of the composites. Recently, ultra-high thermal conductive nature of CNT has gained much attention to use CNT as thermally conductive filler. With the expensive manufacturing cost and limited production, initially, the use of CNT as the thermal conductive filler was rare. But the reduced unit cost for manufacturing and the mass production of CNT with the technological development lead the massive use of CNT as the thermal conductive filler. Because CNT exhibits the extremely high thermal conductivity and mechanical property, a great attention has been paid to the application of CNT in electronic devices [20 - 23] . As the electronic devices need the multifunction and high performance in a small size, the heat dissipation from the products has become more important. For example, a small difference in operating temperature in the order of 10 ~ 15℃ can result in a two-fold reduction in the life time of the electronic devices [24] . CNT has the one-dimensional unique structure which enables the composites filled with CNT to have low percolation threshold concentrations [25 , 26] . By using this characteristics, CNT incorporated composites show the thermal conductivity enhancement at low concentrations [27 , 28] .
Theoretically and experimentally, the thermal conductivity of CNT is about 3,300 and 6,600W/m·K for multiwall and singlewall CNT, respectively. One-dimensional structural characteristics of CNT may promote the thermal conduction behavior of the polymer-CNT composites with a low concentration of CNT incorporation [29 , 30] . Theoretically, six-fold of thermal conductivity enhancement of composite can be achieved by 0.1 wt% of CNT incorporation to the polymer matrix. But experimental results show that it is impossible with current techniques [31 - 34] .
Lager Image
Carbon nanotube dispersion state of (a) raw and (b) masterbatched multiwall carbon nanotube (MWCNT)/poly(dimethylsiloxane) (PDMS) composites at 0.1 phr MWCNT [37].
- 2.2. Models for predicting thermal conductivity of CNTpolymer composites
Since CNT has a very large aspect ratio and anomalous thermal conductivity, the prediction of the thermal conductivity of CNT incorporated polymer composites from the existing theoretical model is not possible [16] . Nan et al . have derived a theoretical model for predicting the thermal conductivity of 1-dimensional CNT incorporated composites [35 , 36] . This equation model is useful at low volume fraction (f < 0.02) of CNT incorporation. And the thermal conductivity of CNT incorporated composites was differed by varying the aspect ratio of CNT. The equation is expressed as follows:
Lager Image
where K c , K e , K m , f, p, and d are thermal conductivity of CNT, effective thermal conductivity of the composite, thermal conductivity of matrix, volume fraction of nanotube, aspect ratio of CNT, and diameter of CNT, respectively. And the Kapiza radius (a K ) is defined by
Lager Image
For nanotube composites, a K is 16~40 nm when K m is 0.2~0.5W/m·K and R K is 8×10 -8 m·K/W. By using equation (1), Hong et al . predicted the thermal conductivity of multiwall CNT incorporated poly(dimethylsiloxane) (PDMS) nanocomposites [37] . In their work, the thermal conductivity of the composites was increased with the amount of CNT and the higher thermal conductivity was achieved for the highly dispersed CNT/PDMS composite prepared by a masterbatch technique. Fig. 2 shows the dispersibility of CNT in the PDMS matrix. At the same concentration of CNT, excellent dispersion of CNT was achieved for the masterbatched CNT/PDMS composite, corresponding improved thermal conductivity. In CNT incorporated polymer composite systems, the fine dispersion of CNT is very important in the
Lager Image
Thermal conductivity of raw MWCNT and silica-coated MWCNT incorporated PDMS nanocomposites [38].
heat dissipation. CNT easily forms the self-aggregated clusters in the polymer matrix and the clusters disturb the heat conduction within the polymer matrix. Also the surface resistance of the polymer-CNT is very important to heat penetration. Hong et al . explored the effect of surface resistance between the CNT/PDMS composites [38] . It is mentioned that CNT and PDMS has poor compatibility. In order to reduce the surface resistance, they coated the MWCNT by silica layer by sol-gel method using 2-aminoethyl 3-aminopropyltrimethoxysilane (AEAPS) coupling agent as an activator [39] . Because the phonon wave can easily penetrate the interface between PDMS-MWCNT by silica layer with enhanced wettability, prepared PDMSMWCNT composite showed enhanced thermal conductivity at the same concentration of MWCNT incorporation ( Fig. 3 ).
- 2.3. Experimental review on thermal conductivity of CNT-polymer composites
Bryning et al . prepared the purified singlewall CNT (SWCNT) dispersed epoxy composite using the N-Ndimethylformamide (DMF) and surfactant [28] . They showed the thermal conductivity enhancement of DMF-processed composite at SWCNT volume fraction between 0.001 to 0.005 and surfactant-processed composite at SWCNT volume fraction up to 0.1. The DMF-processed composite showed the excellent thermal conduction behavior at the lower concentration of SWCNT compared to the surfactant-stabilized composite ( Fig. 4 ). Gojny et al . explored the influence of the type of carbon nanotube (carbon black (CB), singlewall CNT, doublewall CNT, and multiwall CNT) and the relevance of
Lager Image
Thermal conductivity of SWCNT dispersed epoxy nanocomposites using DMF and surfactant as dispersant [28].
Lager Image
TEM-images of the nanofillers: (a) carbon black (b) singlewall CNT (c) doublewall CNT (d) amino-functionalized doublewall CNT (e) multiwall CNT and (f) amino-functionalized multiwall CNT [31].
surface-functionalization (-NH 2 ) in epoxy matrices [31] . The transmission electron microscopy (TEM) images of the fillers used in the work are shown in Fig. 5 . Because the interfacial adhesion between the CNT and epoxy matrix influences to the phonon conduction, the experimental results showed the composites containing amino-functionalized CNT had lower thermal conductivity than pristine CNT incorporated composites due to reduced interfacial adhesion ( Fig. 6 ). Huang
Lager Image
Thermal conductivity vs. filler contents with different fillers [31].
Lager Image
Thermal conductivity as a function of weight fraction of CNT [40]. 706: low thermal conductive silicone rubber (0.23W/m·K) and S106: thermally conductive silicone elastomer (0.56W/m·K)
et al . fabricated the aligned CNT composite film by in-situ injection molding method in the silicone rubber (S160) [40] . As shown in Fig. 7 , the aligned CNT incorporated composite shows remarkably improved thermal conductivity than isotropically distributed CNT composites.
Song et al . treated MWCNT with different ball-milling time and the average length and max length of MWCNT were decreased with the ball-milling time [41] . Thermal conductivity of the fabricated composite was highest at the average length of MWCNT close to the mean curvature radius of MWCNT (600 nm, about 9 hrs of ball-milling) ( Fig. 8 ). They referred that this effect was the length efficiency. The longer MWCNT with the inappropriate length efficiency caused the self-entanglement of MWCNT and the too short MWCNT also reduced the thermal
Lager Image
Thermal conductivity enhancement with different ballmilling times at 0.64 wt% CNT [41].
Thermal conductivity of 2 wt% carbon nanotube (CNT) incorporated poly(dimethylsiloxane) (PDMS) with different conditions[44]
Lager Image
Thermal conductivity of 2 wt% carbon nanotube (CNT) incorporated poly(dimethylsiloxane) (PDMS) with different conditions [44]
conductivity from decreased aspect ratio.
Liu et al . [42] discussed the influence of the purification of CNT on the thermal conductivity of CNT incorporated nanocomposites. Because the pristine CNT contains impurities such as amorphous carbon and metallic catalyst, the purification of CNT is needed to remove these impurities. Also the purification of CNT has following advantages: (1) introduction of the functional group which promotes the wettability and adhesion with host matrix on CNT, (2) prevention of self-agglomeration of CNT originated from strong van der Waals force. Liu et al . prepared the chemically treated CNT using nitric acid and explored the effect of purification of CNT on thermal conductivity [43] . Table 2 shows the thermal conductivity of purified and raw CNT incorporated PDMS rubber. After 30 min of the chemical treatment, the thermal conductivity of the composites was increased from 0.110 to 0.192 W/m·K at 2 wt% of chemically treated CNT incorporation. It is noted that the thermal conductivity of raw CNT incorporated PDMS composites was 0.185W/m·K. But after 90 min of acid treatment, the thermal conductivity of CNT-PDMS composite was decreased. Chemical functionalization of CNT significantly reduces CNT-matrix thermal boundary
Lager Image
NIR spectroscopy of four SWCNT fractions; (a) raw (b) acid treated (c) low speed centrifuged and (d) high speed centrifuged CNTs and thermal conductivity enhancement as a function of the purity of CNT [44].
resistance and decreases intrinsic CNT conductivity by defect generated during the acid treatment. Yu et al . purified SWCNT by nitric acid and fabricated the SWCNT-epoxy composites [44] . The relative purity was examined using near infrared spectroscopy (NIR) ( Fig. 9 ). Thermal conductivity of the composites was increased with the increased relative purity. This positive effect is ascribed from the fact that (1) introduced carboxylic acid during nitric acid treatment aids the dispersion of SWCNT in the polymer matrix, (2) carboxylic acid group on the surface of SWCNT forms the crosslink with the polymer chain and it reduces the phonon scattering, and (3) the higher purity of SWCNT, the higher effective concentration. Hong et al . purified MWCNT and SWCNT with H 2 O 2 aqueous solution followed by HCl aqueous solution [45] . Fabricated poly(methyl methacrylate) (PMMA)-pristine CNT composites exhibited the thermal conductivity up to 3.44W/m·K for 4.0 wt% of MWCNT content. But in SWCNT incorporated composites, the decrease of thermal conductivity was observed due to the poor dispersion of pristine SWCNT at high loading. Fig. 10 (a) and (b) show the thermal conductivity of PMMA-unpurified CNT composites and -purified CNT composites, respectively. After the incorporation of purified CNT, functional groups were generated on the surface of CNT and they inhibited the formation of percolation chains among SWCNT bundles. Finally, purified SWCNT incorporated composites have good dispersibility of the filler and the thermal conductivity of the composites was increased. Severe acid treatment of MWCNT destructs the graphite structure of MWCNT, leaving the amorphous carbon remaining on MWCNT. As shown in Fig. 10 (b), the destructed structure of MWCNT dominates the decrease in the thermal conductivity of purified MWCNT incorporated composites.
Lager Image
Thermal conductivity of (a) raw-CNT and (b) purified CNT incorporated PMMA-based composites [45].
3. Carbon Nanofiber (CNF)/Graphite Incorporated Polymer Composites
- 3.1. Thermal conductivity of CNF-polymer composites
In recent years, nanomaterials are increasingly being used in polymer resins to enhance a variety of properties without sacrificing any properties of the polymer. For example, CNT and carbon nanofiber (CNF) have very high thermal and electrical conductivities with the unusual 1-dimensional structures. Especially, CNF reinforced thermoplastic laminates have been used in aerospace sector due to their excellent heat stability, mechanical property, and low flammability [46] . In principle, proper alignment of the nanofillers can give the desired directional properties, but it is not always easy to obtain such alignment with the normal polymer processing methods of extrusion and injection molding. CNF and CNT have same 1-dimensional structures and both are fabricated by the catalytic decomposition of aliphatic hydrocarbons. But they are different in the diameter (CNF: ~100 nm, CNT: 1~10 nm) and their thermal conduction properties. Recently, CNF has become readily available in
Lager Image
Schematic of the process used to produce (a) PANbased and (b) pitch-based CNF [47].
large quantities and at a price that is significantly lower than that of CNT. So, the use of CNF as the filler in the polymer matrices has much potential in industrial applications.
In 1960, the high performance carbon fiber was firstly commercialized by Union Carbide. Most of the CNFs are produced by converting a carbonaceous precursor into fiber form, then crosslinked. After the carbonization at a temperature from 1200 to 3000℃ in an inert condition, final carbon fiber products are fabricated [47] . Most of the carbon fibers can be classified into two main base materials, namely, poly(acrylonitrile), PAN, precursor carbon fiber and pitch precursor carbon fiber. The schematic process for producing the PAN-base and pitch-base fibers is shown in Fig. 11 (a) and (b). When preparing the PAN-base fiber, the rate of fiber formation is controlled by the solution concentration, the concentration of coagulation bath, the temperature of the bath, the speed of wind up, and so on. Like PAN-base fiber, the pitch-base fiber formation can be controlled by adjusting parameters.
Carbon fiber offers the highest tensile strength and modulus out of all reinforcing fillers. In Table 3 , typical properties of different two carbon fibers are compared. Carbon fibers do not show the stress failure, stress corrosion or stress rupture at room temperatures, as other fibers do(polymer or glass fiber) [48] . Because PAN-based CNF consists of particle-like structure and small crystals in comparison with pitch-based CNF, PAN-based CNF has
Axial tensile properties of carbon fibers with the precursor[49]
Lager Image
Axial tensile properties of carbon fibers with the precursor [49]
Lager Image
SEM pictures of (a) PAN-based and (b) pitch-based CNF. Terminations (-1) and (-2) refer low and high resolutionrespectively [48].
higher tensile and compressive strength than pitch-based CNF. Due to these characters of two types of CNF, PANbased CNF generally shows lower thermal conductivity than that of pitch-based CNF. The scanning electron microphotographs of the cross-section of PAN-based CNF and pitch-based CNF are represented in Fig. 12 .
The thermal conductivity of CNF is not exactly explored, but it is thought to be very high. The thermal conductivity value of CNF might be in the range of the thermal conductivity of the graphite (600W/m·K) and the corresponding value of MWCNT (3,300W/m·K). Similarly to CNT, the thermal conductivity of the polymer composite can be significantly increased by the addition of a small concentration of CNF with their unusual 1-dimensional structure without the adverse effect of the polymer matrix. Many researchers have investigated the effect of CNF on the thermal conductivity of CNF filled polymer composites. But the interface resistance between the polymer and CNF has not been established yet. So, the enhancement of the thermal conductivity by the addition of CNF is much lower than our expectation [49] . It is for this reason that the thermal conductivity of particulate composites is not sensitive to the thermal conductivity value of the filler, provided that the ratio of the filler to matrix thermal conductivity is large [50] . To increase the interfacial interaction between the CNF and the polymer matrix, acid treatment of the fibers has been attempted [51] .
- 3.2. Experimental review on thermal conductivity of CNFpolymer composites
Kuriger et al . explored the thermal properties of polypropylene (PP) composites reinforced with vapor-grown carbon fiber (VCGF) [52] . VGCF, different from PAN-based or pitch-based CNF, has the physical properties like single crystal graphite and a lamellar structures. The thermal conductivity of the fabricated PP-VCGF composite showed a good thermal diffusivity with a longitudinal direction with increasing fiber volume fraction. Due to the 1-dimensional structure of CNF, however, lower thermal conductivity enhancement was observed in a transverse direction of the composite ( Fig. 13 ). Biercuk et al . prepared epoxy-VCGF composites [17] where the thermal conductivity of the epoxy-VCGF composite was compared with epoxy-SWCNT composite. The thermal conductivity enhancement of the epoxy-VCGF composite was lower than that of epoxy-SWCNT composite due to the smaller aspect ratio of VCGF than SWCNT because VCGF had higher diameter than SWCNT and tended to break to length below 100 ㎛. In spite of the lower enhancing characters on thermal conductivity of VCGF than SWCNT, VCGF incorporated composite showed about 100% enhancement in thermal conductivity at 2 wt% loading. Ghose et al . used a polyimide resin as a matrix and CNF [53] . The composite was prepared in the form of the aligned and randomly oriented CNF composites and via two mixing processes, namely, melt and solution mixing. Better
Lager Image
Thermal diffusivity results in the longitudinal and transverse directions for PP composites at various CNF volume fractions[52].
Thermal conductivity of polyimide/carbon nanofiber composites[54]melt: melt mixingsol: solution mixinga: CNF was aligned in polyimide matrixb: CNF was not aligned in polyimide matrix
Lager Image
Thermal conductivity of polyimide/carbon nanofiber composites [54] melt: melt mixing sol: solution mixing a: CNF was aligned in polyimide matrix b: CNF was not aligned in polyimide matrix
Lager Image
(a) Electrophoresis process for deposition of CNT on CNF and (b) CNF with MWCNT deposited by electrophoresis [55].
orientation of CNF in the polyimide matrix was achieved via melt mixing and the thermal conductivity of the composite was higher with the melt mixing process. But when CNF was not aligned, the better dispersion of CNF in polymer matrix
Lager Image
Thermal conductivity of pristine and MWCNT-grafted PAN-based (T1000GB) and pitch-based (K13D) carbon fibers[56].
was observed in the process via solution mixing. Table 4 shows the thermal conductivity of the polyimide-CNF composites. The thermal conductivity of the composite was significantly increased with the alignment of CNF.
Recently, many researchers made an effort on hybrid filler systems using CNT and CNF together. Kim et al . used MWCNT as filler in the conventional carbon fiber/phenolic resin composites for thermal dissipation [54] . The addition of 7 wt% MWCNT resulted in about 57% thermal conductivity enhancement without orientation of fibers. MWCNT with a high aspect ratio formed a bridge between CNF, thus providing a continuous way for heat transfer which resulted in the highly enhanced thermal conductivity. Bekyarova et al . fabricated the CNT-coated CNF via electrophoretic deposition [55] . Fig. 14 shows the experimental scheme for electrophoresis deposition and the morphology of CNT-coated CNF. Then the hybrid CNT-CNF filler was incorporated into epoxy resin. When CNT was coated on the surface of CNF, CNT-CNF incorporated composites had the increased electrical conductivity and enhanced mechanical properties. And it was expected that the thermal conductivity of CNT-CNF incorporated composites would be increased. Naito et al . coated the PAN-based and pitch-based CNFs by CNT using thermal chemical vapor deposition (CVD) method [56] . CNT-coated CNF filler showed an excellent thermal conductivity after the growth of CNT on CNF and it seems to have great potential in the application on the thermally conductive polymer composites ( Fig. 15 ).
4. Summary
Herein, the thermal conductivity of 1-dimensional carbonbased filler incorporated polymeric composites is reviewed.1-Dimensional carbon based materials such as carbon nanotubes and carbon fibers are frequently mixed with polymer resins in efforts to increase the thermal conductivity of the polymer composites. For their unique structure and excellent thermal transportation properties, the thermal conductivity of the composite can be significantly increased with the small concentration of the fillers. Commonly, the thermal conductivity of 1-dimensional filler incorporated composites was found to be much higher in the longitudinal direction due to axial alignment of nanotubes and fibers. And nanotubes and fibers can easily generate the clusters in a polymer matrix with the curvature characters of the fiber. Also, the thermal resistance between the polymer and fibrous filler prevents the penetration of the phonons through the interface. So the dispersion of the nanotubes/fibers in the matrix and thermally connectable techniques are very important for enhancing thermal conductivity of the composites.
By using some important models for predicting the thermal conductivity of composites, CNT and CNF filled nanocomposites show the very high thermal conductivity enhancement than any other inorganic fillers. However, in reality, the increment of a thermal conductivity is not high enough than our expectation. In order to predict the thermal conductivity of 1-dimensional fillers incorporated composites, the degree of the dispersion of nanotubes/fibers and the thermal resistance between the polymer and filler must be taken into account.
Thermal management of the materials becomes more important with the development of the electronic devices, aerospace technology, power generation, thermo coatings, and so on. Therefore, the pace to exploit these materials must be expedited and the exploration to find the potential uses must be continued such as for LED application.
Acknowledgements
This work was supported by the Regional Innovation Center for Environmental Technology of Thermal Plasma (ETTP) at Inha University designated by MKE.
References
Chung D. D. L 2001 Appl. Therm. Eng. 21 1593 -
Tritt T. M 2004 “Thermal Conductivity: Theory Properties and Applications” Springer Science New York
Sanada K , Tada Y , Shindo Y 2009 Compos. Part A-Appl. S. 40 724 -
Prasher R 2006 Proceedings of the IEEE 94 1571 -
Gwinn J. P , Webb R. L 2003 Microelectron. J. 34 215 -
Finan J. M 1999 Proceedings of Society of Plastic Engineers' Annual Technical Conference 1547 -
Heiser J. A , King J. A 2004 Polym. Composite 25 186 -
Sim L. C , Ramanan S. R , Ismail H , Seetharamu K. N , Goh T. J 2005 Thermochim. Acta 430 155 -
Zhou W , Qi S , Tu C , Zhao H , Wang C , Kou J 2007 J. Appl. Polym. Sci 104 1312 -
Kim S. H , Choi S. R , Kim D. J 2007 Heat Trans.-T. ASME 129 298 -
Zhou W , Qi S , An Q , Zhao H , Liu N 2007 Mater. Res. Bull. 42 1863 -
Mu Q , Feng S , Diao G 2007 Polym. Compos. 28 125 -
Lee B , Liu J. Z , Sun B , Shen C. Y , Dai G. C 2008 Exp. Polym. Lett. 2 357 -
Lee B , Dai G 2008 J. Mater. Sci. 44 4848 -
Yu W , Xie H , Chen L , Li Y 2009 Thermochim. Acta. 491 92 -
Choi S. U. S , Zhang Z. G , Yu W , Lockwood F. E , Grulke E. A 2001 Appl. Phys. Lett. 79 2252 -
Biercuk M. J , Llaguno M. C , Radosavljevic M , Hyun J. K , Johnson A. T , Fischer J. E 2002 Appl. Phys. Lett. 80 2767 -
Xu B , Fu Y. Q , Ahmad M , Luo J. K , Huang W. M , Kraft A , Reuben R , Pei Y. T , Chend Z. G , Th J , De Hossond M 2010 J. Mater. Chem. 20 3442 -
Buxton G. A , Balazs A. C 2004 Mol. Simulat. 30 249 -
Kuchibhatla S. V. N. T , Karakoti A. S , Bera D , Seal S 2007 Prog. Mater. Sci. 52 699 -
Xie X. L , Mai Y. X , Zhou S. P 2005 Mater. Sci. Eng. R. 49 89 -
Ramasubramaniam R , Chen J , Liu H 2003 Appl. Phys. Lett. 83 2928 -
Parekh B. B , Fanchini G , Eda G , Chhowalla M 2007 Appl. Phys. Lett. 90 121913 -
Viswanath R , Wakharkar V , Watwe A , Lebonheur V 2000 Intel Technol. J. 4 1 -
Bonnet P , Sireude D , Garnier B , Chauvet O 2007 Appl. Phys. Lett. 91 201910 -
Lu C , Mai Y. W 2008 J. Mater. Sci. 43 6012 -
Xu Y , Leong C. K , Chung D. D. L 2007 J. Electro. Mater. 36 1181 -
Bryning M. B , Milkie D. E , Islam M. F , Kikkawa M , Yodh A. G 2005 Appl. Phys. Lett. 87 161909 -
Berber S , Kwon Y. K , Tamánek D 2000 Phys. Rev. Lett. 84 4613 -
Yang D. J , Wang S. G , Zhang Q , Sellin P. J , Chen G 2004 Phys. Lett. A 329 207 -
Gojny F. H , Wichmann M. H. G , Fiedler B , Kinloch I. A , Bauhofer W , Windle A. H , Schulte K 2006 Polymer 47 2036 -
Kashiwagi T , Grulke E , Hildong J , Groth K , Harris R , Butler K 2004 Polymer 45 4227 -
Xia H , Song M 2005 Soft Matter. 1 386 -
Xu Y , Ray G 2006 Compos. Part A-Appl. S. 37 114 -
Nan C. W , Shi Z , Lin Y 2003 Chem. Phys. Lett 375 666 -
Nan C. W , Liu G , Lin Y , Li M 2004 Appl. Phys. Lett. 85 3549 -
Hong J , Lee J , Hong C. K , Shim S. E 2010 Current Appl. Phys. 10 359 -
Hong J , Lee J , Hong C. K , Shim S. E 2010 J. Therm. Anal. Calorim. 101 297 -
Kim M , Hong J , Lee J , Hong C. K , Shim S. E 2008 J. Colloid Interf. Sci. 322 321 -
Huang H , Liu C , Wu Y , Fan S 2005 Adv. Mater. 17 1652 -
Song P. C , Liu C. H , Fan S. S 2006 Appl. Phys. Lett. 88 153111 -
Kim M , Hong C. K , Choe S , Shim S. E 2007 J. Polym. Sci. Part A: Polym. Chem. 45 4413 -
Liu C. H , Fan S. S 2005 Appl. Phys. Lett 86 123106 -
Yu A , Itkis M. E , Bekyarova E , Haddon R. C 2006 Appl. Phys. Lett. 89 133102 -
Hong W. T , Tai N. H 2008 Diam. Relat. Mater 17 1577 -
Shen Z , Bateman S , Wu D. Y , McMahon P , Dell'Olio M , Gotama J 2009 Compos. Sci. Technol. 69 239 -
Edie D. D 1998 Carbon 36 345 -
Chand S 2000 J. Mater. Sci. 35 1303 -
Nan C. W , Birringer R , Clarke D. R , Gleiter H 1997 J. Appl. Phys. 81 6692 -
Rajaiah J , Andrews G , Ruckenstein E , Gupta R. K 1993 Chem. Eng. Sci. 47 3863 -
Lee G. W , Lee J. I , Lee S. S , Park M , Kim J 2005 J. Mater. Sci. 40 1259 -
Kuriger R. J , Alam M. K 2002 Exp. Heat Trans. 15 19 -
Ghose S , Working D. C , Connell J. W , Smith Jr J. G , Watson K. A , Delozier D. M , Sun Y. P , Lin Y 2006 High Perform. Polym. 18 961 -
Kim Y. A , Kamio S , Tajiri T , Hayashi T , Song S. M , Endo M , Terrones M , Dresselhaus M. S 2007 Appl. Phys. Lett. 90 093125 -
Bekyarova E , Thostenson E. T , Yu A , Kim H , Gao J , Tang J , Hahn H. T , Chou T. W , Itkis M. E , Haddon R. C 2007 Langmuir 23 3970 -
Naito K , Yang J. M , Xu Y , Kagawa Y 2010 Carbon 48 1849 -