Thermal characterization of Al2O3 and ZnO reinforced silicone rubber as thermal pads for heat dissipation purposes
Introduction
Recent advancement in electronics technology has resulted in the miniaturization of transistors, allowing more transistors to be crammed and integrated into a single device, resulting in a higher performance device [1]. Nevertheless, integration and cramming of transistors has resulted in the escalation of power dissipation as well as an increase in heat flux at the devices. It is well known that the reliability of devices is exponentially dependant on the operating temperature of the junction, whereby a small difference in operating temperatures (in the order of 10–15 °C) can result in a two times reduction in the lifespan of a device [2]. Therefore, it is essentially crucial for the heat generated from the devices to be dissipated as quickly and effectively as possible, to maintain the operating temperatures of the device at a desired level [3], [4].
Among the various methods used to dissipate heat from the devices includes the attachment of a high thermal conductivity and low coefficient of thermal expansion (CTE) heat sink or heat spreader on the devices [5], [6], [7]. However, without good thermal contacts, the performance of a high thermal conductivity heat sink to dissipate heat is limited, due to interfacial thermal resistance arising from non-surface flatness and surface roughness of both the devices and heat sink. Non-surface flatness, are commonly observed in the form of convex, concave and wavy surfaces, resulting in as much as 99% of the interfaces being separated by air gaps [8]. Interstitial air gaps trapped due to improper mating of the surfaces significantly reduces the capability to dissipate heat, due to the low thermal conductivity value of air (kair = 0.026 Wm−1 K−1). One method that is commonly used to reduce the thermal contact resistance between the two surfaces is to include an additional material, commonly referred as thermal interface materials (TIM), to provide an effective heat path, as shown in Fig. 1 [9], [10], [11], [12], [13], [14].
TIMs are typically made up of polymer or silicone matrix reinforced with highly thermally conductive but electrical insulating fillers such as aluminum nitride, boron nitride, alumina or silicon carbide [15], [16], [17]. An ideal TIM should not only have high thermal conductivity but must also have low coefficient of thermal expansion. Besides that the material must be easily deformed by small contact pressure to contact all the uneven areas of the mating surfaces [15].
TIMs can be categorized into elastomeric thermal pads, thermal greases, solders and phase change materials [18], [19]. Of all the classes of TIMs mentioned, elastomeric thermal pads are popular for cooling of low power devices, such as chip sets and mobile processors [2]. Elastomeric thermal pads, typically 200–1000 μm thick, consists of elastomer filled with either ceramic or metal fillers. The advantages of elastomeric thermal pads is that they are easy to handle, in addition to being compressible to 25% of their total thickness, enabling the pads to absorb tolerance variances in assemblies [2].
In this study, elastomeric thermal pads were developed from silicone rubber filled with alumina (Al2O3) or zinc oxide (ZnO) fillers at various loadings up to 10 vol.%. Filler loading in the present study has been limited to 10 vol.% to avoid the hardening of pads, which could consequently result in an increase in contact resistance. The effect of Al2O3 or ZnO fillers at various filler loadings on the thermal conductivities and coefficient of thermal expansion (CTE) of the silicone rubber were studied. Experimental data obtained was fitted into a model equation, namely Maxwell–Eucken, Bruggeman, Cheng–Vochan and Agari for thermal conductivity and rules of mixture for CTE, with the values obtained, analyzed and compared. The developed thermal pads were also investigated for thermal stability using the thermal gravimetry analysis (TGA).
Section snippets
Materials
Silicone rubber used in this study, is a siloxane based polymer manufactured by Shin–Etsu silicones, while the curing agent used was 2,5-bis(tert-butyl peroxy)-2,5-dimethylhexane, also from Shin–Etsu silicones. All the above chemicals are used as received. The fillers used in this study are aluminium oxide (Al2O3) 99.9% and zinc oxide (ZnO) 99.7% from Aldrich, with an average particle size of 10 and 1 μm, respectively. Silicone rubber was chosen as the matrix for the thermal pads due to the
Morphology observations
The state of filler distribution is important, as under the percolation theory; filler units need to touch one another to form a continuous heat conduction path [27], [28]. SEM micrographs of the Al2O3 (Fig. 3) and ZnO fillers (Fig. 4), showed that Al2O3 fillers have a flat platelet shape, while ZnO fillers are rhombohedral in shape. Fig. 5(a) is a SEM micrograph showing the cross-section of the Al2O3 filled thermal pads while Fig. 5(b) is that of ZnO filled thermal pads. Both the samples are
Conclusion
Elastomeric thermal pads were successfully prepared from silicone rubber and thermal conductive Al2O3 or ZnO fillers. The addition of either Al2O3 or ZnO fillers into the silicone rubber increases both its thermal stability and thermal conductivity but reduces its CTE. Relationship between fillers and the silicone rubber matrix obtained from the Agari model, showed that Al2O3 filled thermal pads have a higher likelihood for the formation of conductive chains, due to its larger particle size, as
Acknowledgements
The authors thankfully acknowledge the research grant awarded by Intel Technologies Sdn. Bhd. One of the authors, Mr. Sim Lim Chong a recipient of Intel fellowship grant, thankfully acknowledges Intel Technologies Sdn. Bhd. for the award.
References (35)
- et al.
Thermal characterization of an epoxy-based underfill material for flip chip packaging
Thermochim. Acta
(2000) - et al.
Prediction of thermal contact resistance between polished surfaces
Int. J. Heat Mass Transfer.
(1998) - et al.
Carbon black dispersions as thermal pastes that surpass solder in providing high thermal contact conductance
Carbon
(2003) DSC and DMTA studies of a thermal interface material for packaging high speed microprocessors
Thermochim. Acta
(2002)- et al.
Performance and testing of thermal interface materials
Microelectron. J.
(2003) - et al.
Thermally conducting aluminum nitride polymer-matrix composites
Compos. Part A
(2001) - et al.
Thermal Conductivity of polystyrene-aluminum nitride composites
Compos. Part A
(2002) - et al.
Development of new class of electronic packaging materials based on ternary systems of benzoxazine, epoxy, and phenolic resins
Polymers
(2000) Hardness measurement of silicone rubber and polyurethane rubber cured by ionizing measurement
Radiat. Phys. Chem.
(1997)- et al.
Preparation of high strength and optically transparent silicone rubber
Eur. Polym. J.
(1998)
Damage effects and mechanism of proton irradiation on methyl silicone rubber
Mater. Chem. Phys.
Rheological, extractive and thermal studies of the room temperature vulcanized polydimethylsiloxanes
Polymer
Thermal conductivity of polystyrene-aluminum nitride composite
Compos. Part A
Thermal performance challenges from silicone to systems
Intel Technol. J.
Lithium doped polyethylene-glycol based thermal interface pastes for high contact conductance
J. Electron. Packag.
Sodium silicate based thermal interface material for high thermal contact conductance
J. Electron. Packag.
Applications of phase-change materials in Pentium® III and Pentium® III Xeon™ Processor Cartridges
Cited by (462)
Preparation of continuous PBO fiber-filled silicone rubber with high thermal conductivity through simple Wrapping
2024, Composites CommunicationsOptimization of freeze granulation and sintering behavior of MgO granules for thermal interface materials
2024, Ceramics InternationalThermal resistance of Open-Cell metal foam with thermal interface materials (TIM)
2023, Applied Thermal Engineering