Microfabrication

Screen-printed nanostructured composites as thermal interface materials for insulated gate bipolar transistors heat dissipation applications

[+] Author Affiliations
Tien-Chan Chang, Liu Li-Yuan, Yueh-Mu Lee

Institute of Nuclear Energy Research, Atomic Energy Council, Executive Yuan, No. 1000, Wenhua Road, Jiaan Village, Longtan Township, Taoyuan County 32546, Taiwan

Yiin-Kuen Fuh, Rui-Zhong Lee

National Central University, Department of Mechanical Engineering, No. 300, Jhongda Road, Jhongli City, Taoyuan County 32001, Taiwan

J. Micro/Nanolith. MEMS MOEMS. 15(4), 044503 (Dec 20, 2016). doi:10.1117/1.JMM.15.4.044503
History: Received July 31, 2016; Accepted November 28, 2016
Text Size: A A A

Open Access Open Access

Abstract.  Thermal interface materials (TIMs) are of crucial importance in enhancing heat transfer and minimizing exceedingly high temperatures in high-density electronics. TIMs functionally aim to reduce the microscale crevices by penetrating the gap between the contacting rigid surfaces. We prepared silver nanoparticles (SNPs) and single-wall carbon nanotubes (SWCNTs)-based nanocomposites with graphite nanoplatelets (GNPs) by using a screen printing technique for conformal spreading of SNPs and SWCNTs with various weight-loading ratios on top of a layer containing the GNPs and measured its thermal conductivity and electrical conductivities in both through-plane and in-plane directions. In particular, the 10% SNPs enhanced TIMs showed highly anisotropic behavior in both electrical and thermal conductivities, viz., in-plane electrical conductivity exceeds its through-plane counterpart by three orders of magnitude, the highest in-plane electrical conductivity was 7.85 S/cm, and through-plane electrical conductivity was 0.00287  S/cm. Similarly, anisotropic behavior was found for the in-plane thermal conductivity 8.4  W/mK and through-plane thermal conductivity 0.35943  W/mK. In addition, scanning electron microscopy (SEM) was performed to reveal the typical morphology and elements’ existence of screen-printed TIMs. The proposed TIMs were put into the actual 15-kW converter to test the thermal management performance.

Figures in this Article

Many miniaturizations of electronic devices promote ubiquitous computing power. Nonetheless, severe heat and thermal management issues have presumably caused a critical bottleneck, particularly in the industry of modern chips to minimize the thermal-related failures of electronic components.1,2 Thermal interface materials (TIMs) were crucial components to necessitate efficient heat dissipation and advanced high-density electronic packaging. TIMs functionally serve as the effective thermal channel between the heat sink and hot source. TIMs have been extensively employed to substantially reduce the contact thermal resistance between cooling systems such as heat pipes and heat sinks and heat-generating electronic components in power module cooling applications and microelectronics.24 The essential prerequisite in designing useful TIMs structures is high thermal conductivity. TIMs were previously used to enhance the heat dissipation between a base-plate or heat sink and the direct bonded copper substrate under the semiconductor chips.5 There are a significant number of commercially available TIMs products in the electronic applications. In terms of adopted materials for the matrix and fillers, TIMs can be further categorized as metallic foils, sintered metallic, polymer-based such as epoxies and elastomeric pads,6,7 solders and carbon nanotube (CNT),811 phase change materials, thermal greases,12,13 and various forms of thermally conducting fillers such as graphene,1416 a few layers of graphene generated from graphite exfoliation,17,18 and graphite nanoplatelets (GNPs).1922

Concerning the heat transfer enhancement of TIMs, thermal conductivities should be divided into two major areas, i.e., in-plane and through-plane, respectively. There are very few works reported on this aspect and one recent paper investigated the highly anisotropic nature of GNP-based TIMs and demonstrated that thermal anisotropy 5 (an in-plane/through-plane thermal conductivities were 0.5 to 0.9/ to 4.5  W/mK).23 Generally speaking, through-plane thermal conductivity will be significantly lower as compared with the in-plane counterpart mainly attributed to alignment mismatch and suppression in the thermal interface layer.23 For the measurement of through-plane thermal conductivity, one fast and facile structure was proposed by laminating several layers of thermal grease on top of GNPs, resulting in a through-plane thermal conductivity 0.2  W/mK.24 Similar GNPs-epoxy thin films also reported that through-plane thermal conductivity was in the range of 0.5 to 0.9  W/mK.23 Most studies, however, reported that thermal conductivity was primarily in the in-plane direction. For example, nanostructured fillers such as carbon nanosheets and exfoliated GNPs were used and reported increased in-plane thermal conductivity values 1  W/mK.2,25,26 Furthermore, large-scale graphene was used and the reported in-plane thermal conductivity values were 1.263  W/mK.27 Extra enhancement was also pursued by introducing highly thermally conductive fillers such as metal oxide (silica and alumina) or metal microparticles (silver), into the polymer matrix,2830 with enhanced in-plane thermal conductivity values of 1 to 5  W/mK.3134 Increasing the GNPs filler loading resulting in composites with relatively high thermal conductivities (5 to 10  W/mK) had been reported.35 Previous studies of 5% loading of multilayer graphene fillers TIMs were also performed to reach the highest in-plane thermal conductivity of 9.9  W/mK.34

Several commercially available production methods were used to realize TIMs manufacture. For example, capillary flow and mechanical force (pump-out technique) induced filling techniques were reported to uniformly distribute the material between the physically contacting surfaces under pressure.2,36,37 The particular shapes or gaps can be obtained by a subtractive process of photolithographic exposure. On the other hand, additive-type patterning of screen printing technology can be massively utilized in solar cells3840 and in electronic fabrications4143 for wires or screen-printed electrodes.

In this paper, we propose a hybrid composite approach which consists of various loading ratios of nanostructured silver nanoparticles (SNPs) and single-wall carbon nanotubes (SWCNTs) embedded inside the thermal grease, realized by the screen printing technique. Experimental measurements were systematically investigated to characterize the SNPs/SWCNTs-embedded TIMs. Furthermore, high-power electronic devices with severe heat dissipation issues such as insulated gate bipolar transistors (IGBT)44,45 have been selected as the target for heat dissipation applications.

TIMs were functionally designed to overcome the surface nonflatness and surface roughness and to be capable of minimizing the air filled gap interfaces such as superimposed nano/microsurface waviness and roughness.3Figure 1(a) showed the schematic illustration of TIMs layers functionality as an efficient heat flow passage in the IGBT die direction normal to the heat sink baseplate. In this paper, a mechanical bonding and spreading process was consistently used for embedding into the thermal grease (JETART CK4000, JETART Technology Co., Ltd.) with two different nanomaterial fillers of SNPs (bundle diameter = 10 to 20 nm, SA-A00601, Top Nano Technology Co., Ltd.), and SWCNTs (bundle diameter = 1.2 to 1.5 nm, length = 2 to 5 nm, and density = 1.7 to 1.9  g/cm3 at 25°C, 07826A CTs, Aldrich Chem. Co.) and screen printed on top of the GNPs’ substrate (area=4  cm2, thickness=25  μm, T68 artificial graphite sheet, T-Global Technology Co., Ltd.). This well-developed screen printing process was a scalable and simple technique for reliably spreading and adding multiple proportion of SNPs and SWCNTs to the thermal greases, shown schematically as SNPs (1%/2%/5%/10%) [Fig. 1(b)] and SWCNTs (2.5%/5%/7%) [Fig. 1(c)]. In addition, TIMs layers were experimentally measured to exhibit a strong anisotropy in terms of thermal and electrical conductivities. This anisotropy impact of the SNPs/SWCNTs-embedded TIMs and associated performance was extensively investigated and actively applied to the thermal management of the solar/wind sustainable energy 15-KW converter systems with high-power electronics of IGBT.

Graphic Jump Location
Fig. 1
F1 :

Schematic of screen printing technique to functionally facilitate the spreading and bonding process for TIMs composites onto to aluminum sink in high-power applications. (a) Schematic showing the utilization of thin TIMs layers for interfacing and enhancing heat removal in electronic packaging of IGBT chips. There are different SNPs/SWCNTs proportions blended in the thermal greases (SNPs 1%/2%/5%/10% and SWCNTs 2.5%/5%/7%) to be screen printed on top of the substrates of GNPs in the thickness of 250  μm. Orientations of (b) uniformly distributed SNPs fillers and (c) randomly oriented aspect ratio SWCNTs fillers in thin TIMs layers are schematically shown.

First, the electrical conductivity measurements in both configurations of in-plane/through-plane of the proposed SNPs/SWCNTs-based TIMs composites (thickness 250  μm) were experimentally conducted. For the in-plane electrical conductivity measurement, the Ohm meter was used.23 For the electrical conductivity measurements of in-plane configuration, the thermal greases were initially spread and sandwiched between two copper coated glass slides at room temperature. For the through-plane electrical conductivity measurement, a two-probe electrical resistance method was implemented. The measured bulk electrical conductivity in the through-plane can be calculated from the resistance slope.23 Both the in-plane and through-plane electrical conductivity data for the various loading ratios of SNPs/SWCNTs embedded TIMs were shown in Figs. 2(a) and 2(b), respectively. For the used thermal grease, the electrical and thermal conductivities were previously investigated as 1.48  S/cm and 0.2  W/mK, respectively.24

Graphic Jump Location
Fig. 2
F2 :

Anisotropic measurements of in-plane/through-plane electrical conductivity in SNPs/SWCNTs-based TIMs layers. Measured electrical conductivities of (a) in-plane σ|| and (b), through-plane σ at different filler loading ratio SNPs/SWCNTs, and the calculated (c) anisotropy values (the ratio of σ||/σ).

In Fig. 2(a), both SNPs/SWCNTs fillers exhibited monotonically increasing trends of electrical conductivity with increasing loading ratio which, however, increased at different percentages. For example, the in-plane electrical conductivities of the SWCNTs-embedded TIMs were measured as σ||=2.52 to 3.55  S/cm, for the hybrid filler loading of the SWCNTs’ weight ratio in the range of 2.5% to 7% [Fig. 2(a)]. The enhancement of SNPs-embedded TIMs was noticeably significant such that the measured σ||=2.97 to 7.85  S/cm for the hybrid SNPs’ filler loading ratio was in the range of 1% to 10%, whereas the through-plane conductivity (σ) showed a comparatively low electrical conductivity (<0.003  S/cm), as shown in Fig. 2(b). Furthermore, the absolute values of the hybrid SNPs/SWCNTs filler loading TIMs were experimentally measured as σ=0.00255 to 0.00287  S/cm for the hybrid SNPs/SWCNTs fillers; the differences of the filler loading impact were significantly small and within 12.5% [Fig. 2(b)]. These results showed an extremely high anisotropy (more than three orders of magnitude) for the in-plane and through-plane electrical conductivities of the SNPs/SWCNTs filler loading TIMs composites.

Thus, the electrical anisotropy increases with an increasing proportion of the SNPs/SWCNTs fillers, but the contribution from SNPs/SWCNTs fillers was significantly different.

For the absolute values of the anisotropy of electrical conductivity of the SWCNTs, TIMs were calculated in the range of 988 to 1282 (SWCNTs loading ratio 2.5% to 7%), while for the SNPs fillers (filler loading ratio 1% to 10%), the anisotropy of electrical conductivity is measured in the range of 1174 to 2736 [Fig. 2(c)]. Therefore, the strongly anisotropic behavior for the SNPs/SWCNTs-embedded TIMs and the maximum electrical anisotropy for fillers weight ratio of SWCNTs 7% and SNPs 5% were measured as 1282 and 2736, respectively.

In-plane thermal conductivity (K||) of the SNPs/SWCNTs-embedded TIMs was measured by utilizing a well-developed technique as shown in Fig. 3(a). A miniature thermoelectric stage (TE stage) was employed to measure the serially mounted thermal gradient of the heat flow between the TIMs and the copper wire (0.5-mm diameter). The in-plane-oriented TIMs were cut and sampled into the lateral dimensions 2×2  cm2 and a thickness 250  μm. In addition, the nomenclature of T+ and T indicates the hot and cold sides, respectively. In Fig. 3(b), the through-plane measurement of thermal conductivity (K) was measured utilizing an LW-9389 TIMs Tester (Longwin, Taiwan) and ASTM D5470-06 standard for the steady-state heat flow measurement (two meter bars as hot and cold ends, respectively, and set the temperature in the range of 50°C to 80°C). T13 and T46 indicate the respective locations of six thermal couples (designated as T1 to T6 individually) for consistently measuring the steady-state heat flow. Figure 3(c) shows one exemplary result from the in-plane thermal conductivity experiments as the temperature gradients were flipped through the voltage reversal at the TE stage. In order to exclude any measurement errors such as black body radiation, the averaged values of oppositely polarized measurements were performed as suggested previously.26 For measuring the through-plane thermal conductivity, the thermal resistances were recorded for thicknesses of the TIMs composites in the range of 140 to 250  μm (via applied pressure 0 to 80 psi) as shown in Fig. 3(d). The general trend of the decreasing thermal resistance was obtained due to the increased applied pressure. The bulk thermal conductivity was extracted from the slope of the resistance against thickness, as indicated in the dashed lines in Fig. 3(d). As shown in Fig. 3(d), two slopes of SNPs 10%/SWCNTs 7%-embedded TIMs were measured as 0.154 and 0.426, respectively. These slopes correspond to the thermal conductivity as presented. Figure 3(e) showed the in-plane (K||) thermal conductivities of SNPs/SWCNTs-embedded TIMs and compared them with the thermal grease sample (no embedded nanostructures). Samples with SNPs 10% and SWCNTs 7%-embedded TIMs yield an increased in-plane thermal conductivity of 8.4 and 7.9  W/mK, respectively. The through-plane thermal conductivities (K) of SNPs/SWCNTs-embedded TIMs are shown in Fig. 3(f) and showed a marginal increase of through-plane thermal conductivities with increasing proportion of filler loading ratio, which functionally resembles the electrical conductivity in the through-plane [Fig. 2(b)]. Therefore, the absolute values of the anisotropy of thermal conductivity of the SWCNTs TIMs for filler loading of 2.5% to 7% were measured in the range of 25 to 29, whereas for the SNPs fillers with loading ratios of 1% to 10%, the anisotropy of thermal conductivity is measured in the range of 18 to 23 [Fig. 3(e)]. The present measurements show a relatively small anisotropy of thermal conductivity in SNPs/SWCNTs-embedded TIMs while the maximum anisotropy for fillers weight ratio of SWCNTs 7% and SNPs 10% were measured as 29 and 23, respectively. From the above measurements, two orders of magnitude difference were recorded for electrical and thermal anisotropy in the developed TIMs and the main contributing factor is primarily due to the fundamental variation of the adopted materials. For example, the electrical conductivities between the electrical conductors (silver, σ63×106  S/cm) and electrical insulators (polystyrene, PS, σ1016  S/cm) cover almost 22 orders of magnitude. As a comparison, the thermal conductors (silver 429  W/mK and thermal insulators (polystyrene, PS, 0.08  W/mK) only span 4 orders of magnitude.46,47

Graphic Jump Location
Fig. 3
F3 :

Anisotropic thermal conductivity measurements of in-plane/through-plane in SNPs/SWCNTs-embedded TIMs. Thermal conductivity measurements in the schematic setup of (a) in-plane tester by use of the comparison technique; (b) through-plane TIMs tester; (c) in-plane thermal conductivity can be measured as thermal pulses along the TIMs film (black line) and reference copper wire (redline); (d) through-plane thermal conductivity was measured as a function of thickness is recorded via thermal resistance; (e) in-plane thermal (K||); (f) through-plane (K) conductivities of TIMs-based greases of different weight ratios; and (g) anisotropy values calculated as the ratio of K||/K.

The scanning electron microscopy (SEM) images are capable of revealing the morphology and microstructure of the SNPs/SWCNTs TIMs composites. Also shown at a higher magnification, the images of the TIMs were presented with scale bars of 20, 10, 3, and 1  μm, respectively. Figure 4(a) clearly showed the existence of nano/microparticles in the size of 500 nm to 2  μm with various configurations, viz, ellipses, circle, and irregular shapes, for the sample of SNPs 10% TIMs. For the case of SWCNTs embedded TIMs [Fig. 4(b)], the image showed relatively complex nanostructures with multiple SWCNTs bridging the adjacent matrix of thermal greases. The tubular structure of the SWCNTs can be vividly identified on the higher magnification with the distribution density experimentally found as 12/μm2. Figures 4(c)4(d) showed the x-ray mapping of different elements (silver, zinc, silicon, oxygen, and carbon for SNPs TIMs and carbon, silicon, zinc, and oxygen for SWCNTs TIMs). X-ray mapping was primarily used to confirm the presence and dispersion of elements such as silver, carbon, oxygen, silicon, and zinc in the composite. According to the x-ray mapping results, the presence and dispersion of various elements of SNPs 10% TIMs sample can be quantitatively identified such as silver 60.13% (red), zinc 3.76% (blue), silicon 5.46% (green), carbon 2.76% (cyan), oxygen 5.06% (purple), and platinum-plated 22.99% [Fig. 4(c)]. Similarly for the SWCNTs TIMs sample, the elements of carbon 21.17% (cyan), silicon 16.1% (green), zinc 10.38% (blue), oxygen 13.07% (purple), and platinum-plated 36.05% [Fig. 4(d)] are identified. Previous study showed that the used thermal grease has the elements of oxygen, silicon, zinc, and minute quantity carbon,24 therefore, the SNPs/SWCNTs-embedded TIMs showed the additionally added elements of silver and carbon. Another result indicates that the SEM image of SNPs/SWCNTs composite and the SNPs/SWCNTs can be successfully screen printed on the TIMs substrate, a good indicator of our SNPs/SWCNTs filler preparation and spreading processes and that x-ray mapping furthermore confirmed the presence and dispersion of silver, carbon, silicon, zinc, and oxygen in the SNPs/SWCNTs composite such that the both in-plane/through-plane thermal conductivities can be effectively enhanced.

Graphic Jump Location
Fig. 4
F4 :

SEM morphology study of SNPs 10%/SWCNTs 7%-embedded TIMs. (a) Image of surfaces of SNPs with thermal greases at different magnifications with scale bars of 20, 10, and 3  μm. (b) SEM images with addition of SWCNTs and scale bars of 20, 10, 3, and 1  μm. Associated x-ray mapping of (c) SNPs and (d) SWCNTs TIMs composite. Inset in (c) and (d) shows mapping for all elements and individual element mapping of silver (red), zinc (blue), silicon (green), carbon (cyan), oxygen (purple), and platinum-plated for (c) SNPs TIMs; carbon (cyan), silicon (green), zinc (blue), oxygen (purple), and platinum-plated for (d) SWCNTs TIMs, respectively.

In Fig. 5(a), the thermal management of the converter system is of great importance since very high voltage/current will be switched intermittently and/or continuously and high temperature is detrimental to the service life of electronics, especially for high-power switching devices such as IGBT. In this study, a newly developed dual bidirectional IGBT-based converter in conjunction with an autonomous microgrid system is investigated with particular focus on the thermal management and performance evaluation under various operation conditions. Nanostructures of various loading ratios of SNPs and SWCNTs TIMs were applied to the 15-kW converter system of a microgrid system with IGBT module (CM100DY-24A, MITSUBISHI) and packaged by a galvanized steel plate (size 62×48×18  cm3 [Fig. 5(b)]. Due to the imperfectness of machined surface waviness, two superimposed interfaces, such as an IGBT module mounted on a heat sink, are unavoidable resulting in many micro-to-macro-scale gaps. These air gaps severely degrade the IGBT performance due to huge thermal contact resistance and accumulated heat. Therefore, in order to minimize the thermal contact resistance, various TIMs were routinely applied between the IGBT module and heat sink. As shown in Fig. 5(c), the developed TIMs composites (black square) are designed as a buffer thin layer to effectively conduct heat flow in the direction normal to the contact interface. Thermocouples were fixed at the bottom of the heat sink in transient temperature measurements on the IGBT heat sources. The red ellipse showed the three IGBTs at the normal operation position where TIMs were sandwiched between the heat generating IGBT chips and the heat sink. Figure 5(d) showed a schematic of the 15-kW converter system with an explosive view (lower figure), showing that three TIMs are sandwiched between IGBT and the aluminum heat sink.

Graphic Jump Location
Fig. 5
F5 :

(a) Schematic of a typical microgrid system, which includes sustainable resources, transformers, converter (includes IGBT), and the energy storage system. (b) Device of 15-kW converter system (inset) is initially packaged by a galvanized steel plate (size 62×48×18  cm). (c) Used TIMs (black square) and thermocouples used in measurements on the IGBT heat sources. Red ellipse showed the three IGBTs on the normal operation position and the red circle explicitly shows the temperature measurement point. (d) Schematic of the 15-kW converter with an explosive view (lower figure), showing the close-up view (upper figure) three TIMs were sandwiched between IGBT and aluminum heat sink. (e) During the converter operation, measured temperature of SNPs/SWCNTs-embedded TIMs and the filler loading ratio are as follows: SNPs 1%, 2%, 5%, 10% and SWCNTs 2.5%, 5%, 7%. As measured temperature between IGBT module and heat sink during converter discharging process.

The close-up view (upper figure) corresponds to the optical photo of Fig. 5(c). Figure 5(e) showed the measurement results of the temperature rise of SNPs and SWCNTs-embedded TIMs, during the converter operation condition of 9 kW, switching frequency 4 kHz, 220 V, and 20 A. The experimental implementation maintained the measuring time 1200  s due to the protection mechanism of a battery management system (Rich Electric Co., Ltd.) which was limited to 1200  s.

Figure 5(e) shows the experimental results of temperature measurement for SNPs 1%, 2%, 5%, 10% and SWCNTs 2.5%, 5%, 7% TIMs, respectively. At time 1200 s, the measured temperatures were recorded as 95.0°C, 93.8°C, 92.3°C, 90.9°C, 92.7°C, 92.5°C, and 91.7°C, respectively. Furthermore, Table 1 shows quantitative information of the measured values of interface temperature differences at successive times for different variants. The results indicate that the composite of SNPs 10% TIMs effectively reduced tempertures to 7.9°C as compared with only thermal grease in the 15-kW converter system.

Table Grahic Jump Location
Table 1Interface temperature differences at successive times.

The introduction of nanostructured materials such as SNPs/SWCNTs into TIMs has merit for improving the thermal conductivity. It is experimentally found that both the thermal and electrical conductivities of the proposed TIMs demonstrated a high anisotropy in the directions of in-plane and through-plane. Based on the experimental results and discussions, the following conclusions can be drawn:

  1. The best in-plane electrical conductivity is obtained at 7.85  S/cm with SNPs 10% TIMs and that of SWCNTs 7% was 3.5  S/cm. In addition, the trend in the change of in-plane electrical conductivity is found to be consistent with the literature, whereas the through-plane electrical conductivity showed no obvious impact on the loading effect of various nanostructured materials of SWCNTs and SNPs.
  2. The measurement TIMs composite samples of SNPs 10% and SWCNTs 7% showed the trend of a decrease in specific thermal resistance values with decreasing thickness (i.e., increasing pressure). The main contributing factor is attributed to the decreased contact resistance between the interfaces (with an increasing loading weight ratio). Therefore, SNPs 10% and SWCNTs 7%-embedded TIMs yielded an increased in-plane thermal conductivity of 8.4 and 7.9  W/mK, respectively.
  3. A substantial difference was observed between the through-plane and in-plane thermal and electrical conductivities of SNPs/SWCNTs-embedded TIMs. We experimentally showed the highly anisotropic transport properties in the thin TIMs layers, possibly originated from the inherently drastic difference in thermal and electrical properties (several to tens orders of magnitude difference). This anisotropy should be considered in any future development of SNPs/SWCNTs-embedded TIMs.
  4. SEM study of SNPs/SWCNTs composite morphology in thin TIMs layers was performed and samples were fabricated by screen-printed technology. For the case of SNPs, it clearly shows the existence of nano/microparticles in the size of 500 nm to 2  μm with various configurations, viz, ellipses, circle, and irregular shapes. Another case of SWCNTs showed relatively complex nanostructures with multiple SWCNTs bridging adjacent thermal greases. The tubular structure of the SWCNTs can be vividly identified on the higher magnification. Both of these cases demonstrated that the proposed TIMs were successfully fabricated with targeted elements.
  5. For the actual application of developed SNPs/SWCNTs-embedded TIMs in the high-power IGBT thermal management, the experimental results of temperature measurement were compared for SNPs 1%, 2%, 5%, 10% and SWCNTs 2.5%, 5%, 7% TIMs, respectively. At time 1200 s, the measured temperatures were 95, 93.8, 92.3, 90.9, 92.7, 92.5, and 91.7°C, respectively. The results indicated that the SNPs 10% embedded TIMs effectively reduced tempertures by 7.9°C as compared with thermal grease only in the 15-kW converter system, whereas the counterpart of SWCNTs 7% embedded TIMs comparatively reduced tempertures by 7.2°C.

This research was sponsored by the Institute of Nuclear Energy Research (INER) under Agreement No. NL1040327.

Schelling  P. K., , Shi  L., and Goodson  K. E., “Managing heat for electronics,” Mater. Today. 8, (6 ), 30 –35 (2005). 1369-7021 CrossRef
Prasher  R., “Thermal interface materials: historical perspective, status, and future directions,” Proc. IEEE. 94, , 1571 –1586 (2006). 0018-9219 CrossRef
Gwinn  J. P., and Webb  R. L., “Performance and testing of thermal interface materials,” Microelectron. J.. 34, , 215 –222 (2003). 0026-2692 CrossRef
Khuu  V.  et al., “Effects of temperature cycling and elevated temperature/humidity on the thermal performance of thermal interface materials,” IEEE Trans. Dev. Mater. Reliab.. 9, , 379 –391 (2009).CrossRef
Johnson  C. M.  et al., “Integrated high power modules,” in  7th Int. Conf. on Integrated Power Electronics Systems (CIPS) , pp. 357 –366 (2012).
Liu  J.  et al., “Recent progress of thermal interface material research—an overview,” in  14th Int. Workshop on Thermal Investigation of ICs and Systems (THERMINIC 2008) ,  Rome, Italy  (2008).
Goel  N.  et al., “Technical review of characterization methods for thermal interface materials (TIM),” in  Thermal and Thermomechanical Phenomena in Electronic Systems, (ITHERM 2008) ,  Orlando, Florida  (2008).
Shahid  P.  et al., “Factors that govern the performance of thermal interface materials,” J. Electron. Mater.. 38, (1 ), 175 –192 (2009). 0361-5235 CrossRef
Chaowasakoo  T.  et al., “Indium solder as a thermal interface material using fluxless bonding technology,” in  25th Annual IEEE Semiconductor Thermal Measurement and Management Symp. (SEMI-THERM 2009) ,  San Jose, California , pp. 180 –185 (2009).CrossRef
Goodarzi  M.  et al., “Investigation of heat transfer and pressure drop of a counter flow corrugated plate heat exchanger using MWCNT based nanofluids,” Int. Commun. Heat Mass Transf.. 66, , 172 –179 (2015). 0735-1933 CrossRef
Chen  H.  et al., “Architecting three-dimensional networks in carbon nanotube buckypapers for thermal interface materials,” J. Phys. Chem. C. 116, , 3903 –3909 (2012). 1932-7447 CrossRef
Tonapi  S.  et al., “Reliability testing of thermal greases,” in  Electronics Cooling Magazine  (2007).
Gowda  A.  et al., “Reliability testing of silicone-based thermal greases,” in  Proc. 21st IEEE SEMI-THERM Symp. , p. 64  (2005).CrossRef
Khan  M. F.  et al., “Graphene-multilayer graphene nanocomposites as highly efficient thermal interface materials,” Nano Lett.. 12, , 861 –867 (2012). 1530-6984 CrossRef
Goli  P.  et al., “Graphene-enhanced hybrid phase change materials for thermal management of Li-ion batteries,” J. Power Sources. 248, , 37 –43 (2014). 0378-7753 CrossRef
Renteria  J. D., , Nika  D. L., and Balandin  A. A., “Graphene thermal properties: applications in thermal management and energy storage,” Appl. Sci.. 4, , 525 –547 (2014).CrossRef
Dresselhaus  M. S.  et al., “Lattice vibrations in graphite and intercalation compounds of graphite,” Mater. Sci. Eng.. 31, , 141 –152 (1977). 0025-5416 CrossRef
Chung  D. D. L., “Exfoliation of graphite,” J. Mater. Sci.. 22, , 4190 –4198 (1987). 0022-2461 CrossRef
Yavari  F.  et al., “Enhanced thermal conductivity in a nanostructured phase change composite due to low concentration graphene additives,” J. Phys. Chem. C. 115, , 8753 –8758 (2011). 1932-7447 CrossRef
Raza  M. A.  et al., “Characterisation of graphite nanoplatelets and the physical properties of graphite nanoplatelet/silicone composites for thermal interface applications,” Carbon. 49, , 4269 –4279 (2011). 0008-6223 CrossRef
Raza  M. A., , Westwood  A., and Stirling  C., “Carbon black/graphite nanoplatelet/rubbery epoxy hybrid composites for thermal interface applications,” J. Mater. Sci.. 47, , 1059 –1070 (2012). 0022-2461 CrossRef
Shahil  K. M. F., and Balandin  A. A., “Graphene-multilayer graphene nanocomposites as highly efficient thermal interface materials,” Nano Lett.. 12, , 861 –867 (2012). 1530-6984 CrossRef
Xiaojuan  T.  et al., “Anisotropic thermal and electrical properties of thin thermal interface layers of graphite nanoplatelet-based composites,” Sci. Rep.. 10, , 01710  (2013). 2045-2322 CrossRef
Chang  T.-C.  et al., “Application of graphite nanoplatelet-based and nanoparticle composites to thermal interface materials,” Micro Nano Lett.. 10, , 296 –301 (2015).CrossRef
Veca  L. M.  et al., “Carbon nanosheets for polymeric nanocomposites with high thermal conductivity,” Adv. Mater.. 21, , 2088 –2092 (2009). 0935-9648 CrossRef
Xiang  J. L., and Drzal  L. T., “Thermal conductivity of exfoliated graphite nanoplatelet paper,” Carbon. 49, , 773 –778 (2011). 0008-6223 CrossRef
Liu  C., and Hu  G., “Highly efficient reduction of graphene oxide by sub/supercritical water and their application for thermal interface materials,” Appl. Therm. Eng.. 90, , 193 –198 (2015). 1359-4311 CrossRef
Xu  Y. S., and Chung  D. D. L., “Increasing the thermal conductivity of boron nitride and aluminum nitride particle epoxy-matrix composites by particle surface treatments,” Compos. Interfaces. 7, , 243 –256 (2000). 0927-6440 CrossRef
Prasher  R. S.  et al., “Thermal resistance of particle laden polymeric thermal interface materials,” ASME J. Heat Transfer. 125, , 1170 –1177 (2003).CrossRef
Ghanbarpour  M.  et al., “Thermal performance of inclined screen mesh heat pipes using silver nanofluids,” Int. Commun. Heat Mass Transfer. 67, , 14 –20 (2015).CrossRef
Prasher  R., “Thermal interface materials: historical perspective, status, and future directions,” Proc. IEEE. 94, , 1571 –1586 (2006). 0018-9219 CrossRef
Prasher  R. S.  et al., “Nano and micro technology-based next-generation package-level cooling solution,” Intel Technol. J. Electron. Package Technol. Dev.. 9, , 285 –296 (2005).CrossRef
Zhou  W. Y.  et al., “Study on insulating thermal conductive BN/HDPE composites,” Thermochim. Acta. 452, , 36 –42 (2007). 0040-6031 CrossRef
Shahil  K. M. F., and Balandin  A. A., “Thermal properties of graphene and multilayer graphene: applications in thermal interface materials,” Solid State Commun.. 152, , 1331 –1340 (2012). 0038-1098 CrossRef
Fukushima  H.  et al., “Thermal conductivity of exfoliated graphite nanocomposites,” J. Therm. Anal. Calorim.. 85, , 235 –238 (2006). 1418-2874 CrossRef
Song  W. L.  et al., “Polymer/boron nitride nanocomposite materials for superior thermal transport performance,” Angew. Chem. Int. Ed.. 51, , 6498 –6501 (2012).CrossRef
Larmagnac  A.  et al., “Stretchable electronics based on Ag-PDMS composites,” Sci. Rep.. 4, , 7254  (2014). 2045-2322 CrossRef
Green  M. A., “Ag requirements for silicon wafer-based solar cells,” Photovoltaics Res. Appl.. 19, , 911 –916 (2011).CrossRef
Yoshida  M.  et al., “Novel low-temperature-sintering type Cu-alloy pastes for silicon solar cells,” Energy Procedia.. 21, , 66 –74 (2012). 1876-6102 CrossRef
Rane  S. B.  et al., “Firing and processing effects on microstructure of fritted silver thick film electrode materials for solar cells,” Mater. Chem. Phys.. 82, , 237 –245 (2003). 0254-0584 CrossRef
Perelaer  J.  et al., “Printed electronics: the challenges involved in printing devices, interconnects, and contacts based on inorganic materials,” J. Mater. Chem.. 20, , 8446 –8453 (2010). 0959-9428 CrossRef
Sekitani  T., and Someya  T., “Ambient electronics,” Jpn. J. Appl. Phys.. 51, , 100001  (2012).CrossRef
Ahn  B. Y.  et al., “Omnidirectional printing of flexible, stretchable, and spanning silver microelectrodes,” Science. 323, , 1590 –1593 (2009). 0036-8075 CrossRef
Meysenc  L., , Jylhakallio  M., and Barbosa  P., “Power electronics cooling effectiveness versus thermal inertia,” IEEE Trans. Power Electron.. 20, , 687 –693 (2005).CrossRef
Mahajan  R., , Chiu  C. P., and Chrysler  G., “Cooling a microprocessor chip,” Proc. IEEE. 94, , 1476 –1486 (2006). 0018-9219 CrossRef
Shenogina  N.  et al., “On the lack of thermal percolation in carbon nanotube composites,” Appl. Phys. Lett.. 87, , 133106  (2005). 0003-6951 CrossRef
Itkis  M.  et al., “Thermal conductivity measurement of semitransparent single-walled carbon nanotube films by a bolometric technique,” Nano Lett.. 7, , 900 –904 (2007). 1530-6984 CrossRef
© The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

Citation

Tien-Chan Chang ; Yiin-Kuen Fuh ; Rui-Zhong Lee ; Liu Li-Yuan and Yueh-Mu Lee
"Screen-printed nanostructured composites as thermal interface materials for insulated gate bipolar transistors heat dissipation applications", J. Micro/Nanolith. MEMS MOEMS. 15(4), 044503 (Dec 20, 2016). ; http://dx.doi.org/10.1117/1.JMM.15.4.044503


Figures

Graphic Jump Location
Fig. 1
F1 :

Schematic of screen printing technique to functionally facilitate the spreading and bonding process for TIMs composites onto to aluminum sink in high-power applications. (a) Schematic showing the utilization of thin TIMs layers for interfacing and enhancing heat removal in electronic packaging of IGBT chips. There are different SNPs/SWCNTs proportions blended in the thermal greases (SNPs 1%/2%/5%/10% and SWCNTs 2.5%/5%/7%) to be screen printed on top of the substrates of GNPs in the thickness of 250  μm. Orientations of (b) uniformly distributed SNPs fillers and (c) randomly oriented aspect ratio SWCNTs fillers in thin TIMs layers are schematically shown.

Graphic Jump Location
Fig. 2
F2 :

Anisotropic measurements of in-plane/through-plane electrical conductivity in SNPs/SWCNTs-based TIMs layers. Measured electrical conductivities of (a) in-plane σ|| and (b), through-plane σ at different filler loading ratio SNPs/SWCNTs, and the calculated (c) anisotropy values (the ratio of σ||/σ).

Graphic Jump Location
Fig. 3
F3 :

Anisotropic thermal conductivity measurements of in-plane/through-plane in SNPs/SWCNTs-embedded TIMs. Thermal conductivity measurements in the schematic setup of (a) in-plane tester by use of the comparison technique; (b) through-plane TIMs tester; (c) in-plane thermal conductivity can be measured as thermal pulses along the TIMs film (black line) and reference copper wire (redline); (d) through-plane thermal conductivity was measured as a function of thickness is recorded via thermal resistance; (e) in-plane thermal (K||); (f) through-plane (K) conductivities of TIMs-based greases of different weight ratios; and (g) anisotropy values calculated as the ratio of K||/K.

Graphic Jump Location
Fig. 4
F4 :

SEM morphology study of SNPs 10%/SWCNTs 7%-embedded TIMs. (a) Image of surfaces of SNPs with thermal greases at different magnifications with scale bars of 20, 10, and 3  μm. (b) SEM images with addition of SWCNTs and scale bars of 20, 10, 3, and 1  μm. Associated x-ray mapping of (c) SNPs and (d) SWCNTs TIMs composite. Inset in (c) and (d) shows mapping for all elements and individual element mapping of silver (red), zinc (blue), silicon (green), carbon (cyan), oxygen (purple), and platinum-plated for (c) SNPs TIMs; carbon (cyan), silicon (green), zinc (blue), oxygen (purple), and platinum-plated for (d) SWCNTs TIMs, respectively.

Graphic Jump Location
Fig. 5
F5 :

(a) Schematic of a typical microgrid system, which includes sustainable resources, transformers, converter (includes IGBT), and the energy storage system. (b) Device of 15-kW converter system (inset) is initially packaged by a galvanized steel plate (size 62×48×18  cm). (c) Used TIMs (black square) and thermocouples used in measurements on the IGBT heat sources. Red ellipse showed the three IGBTs on the normal operation position and the red circle explicitly shows the temperature measurement point. (d) Schematic of the 15-kW converter with an explosive view (lower figure), showing the close-up view (upper figure) three TIMs were sandwiched between IGBT and aluminum heat sink. (e) During the converter operation, measured temperature of SNPs/SWCNTs-embedded TIMs and the filler loading ratio are as follows: SNPs 1%, 2%, 5%, 10% and SWCNTs 2.5%, 5%, 7%. As measured temperature between IGBT module and heat sink during converter discharging process.

Tables

Table Grahic Jump Location
Table 1Interface temperature differences at successive times.

References

Schelling  P. K., , Shi  L., and Goodson  K. E., “Managing heat for electronics,” Mater. Today. 8, (6 ), 30 –35 (2005). 1369-7021 CrossRef
Prasher  R., “Thermal interface materials: historical perspective, status, and future directions,” Proc. IEEE. 94, , 1571 –1586 (2006). 0018-9219 CrossRef
Gwinn  J. P., and Webb  R. L., “Performance and testing of thermal interface materials,” Microelectron. J.. 34, , 215 –222 (2003). 0026-2692 CrossRef
Khuu  V.  et al., “Effects of temperature cycling and elevated temperature/humidity on the thermal performance of thermal interface materials,” IEEE Trans. Dev. Mater. Reliab.. 9, , 379 –391 (2009).CrossRef
Johnson  C. M.  et al., “Integrated high power modules,” in  7th Int. Conf. on Integrated Power Electronics Systems (CIPS) , pp. 357 –366 (2012).
Liu  J.  et al., “Recent progress of thermal interface material research—an overview,” in  14th Int. Workshop on Thermal Investigation of ICs and Systems (THERMINIC 2008) ,  Rome, Italy  (2008).
Goel  N.  et al., “Technical review of characterization methods for thermal interface materials (TIM),” in  Thermal and Thermomechanical Phenomena in Electronic Systems, (ITHERM 2008) ,  Orlando, Florida  (2008).
Shahid  P.  et al., “Factors that govern the performance of thermal interface materials,” J. Electron. Mater.. 38, (1 ), 175 –192 (2009). 0361-5235 CrossRef
Chaowasakoo  T.  et al., “Indium solder as a thermal interface material using fluxless bonding technology,” in  25th Annual IEEE Semiconductor Thermal Measurement and Management Symp. (SEMI-THERM 2009) ,  San Jose, California , pp. 180 –185 (2009).CrossRef
Goodarzi  M.  et al., “Investigation of heat transfer and pressure drop of a counter flow corrugated plate heat exchanger using MWCNT based nanofluids,” Int. Commun. Heat Mass Transf.. 66, , 172 –179 (2015). 0735-1933 CrossRef
Chen  H.  et al., “Architecting three-dimensional networks in carbon nanotube buckypapers for thermal interface materials,” J. Phys. Chem. C. 116, , 3903 –3909 (2012). 1932-7447 CrossRef
Tonapi  S.  et al., “Reliability testing of thermal greases,” in  Electronics Cooling Magazine  (2007).
Gowda  A.  et al., “Reliability testing of silicone-based thermal greases,” in  Proc. 21st IEEE SEMI-THERM Symp. , p. 64  (2005).CrossRef
Khan  M. F.  et al., “Graphene-multilayer graphene nanocomposites as highly efficient thermal interface materials,” Nano Lett.. 12, , 861 –867 (2012). 1530-6984 CrossRef
Goli  P.  et al., “Graphene-enhanced hybrid phase change materials for thermal management of Li-ion batteries,” J. Power Sources. 248, , 37 –43 (2014). 0378-7753 CrossRef
Renteria  J. D., , Nika  D. L., and Balandin  A. A., “Graphene thermal properties: applications in thermal management and energy storage,” Appl. Sci.. 4, , 525 –547 (2014).CrossRef
Dresselhaus  M. S.  et al., “Lattice vibrations in graphite and intercalation compounds of graphite,” Mater. Sci. Eng.. 31, , 141 –152 (1977). 0025-5416 CrossRef
Chung  D. D. L., “Exfoliation of graphite,” J. Mater. Sci.. 22, , 4190 –4198 (1987). 0022-2461 CrossRef
Yavari  F.  et al., “Enhanced thermal conductivity in a nanostructured phase change composite due to low concentration graphene additives,” J. Phys. Chem. C. 115, , 8753 –8758 (2011). 1932-7447 CrossRef
Raza  M. A.  et al., “Characterisation of graphite nanoplatelets and the physical properties of graphite nanoplatelet/silicone composites for thermal interface applications,” Carbon. 49, , 4269 –4279 (2011). 0008-6223 CrossRef
Raza  M. A., , Westwood  A., and Stirling  C., “Carbon black/graphite nanoplatelet/rubbery epoxy hybrid composites for thermal interface applications,” J. Mater. Sci.. 47, , 1059 –1070 (2012). 0022-2461 CrossRef
Shahil  K. M. F., and Balandin  A. A., “Graphene-multilayer graphene nanocomposites as highly efficient thermal interface materials,” Nano Lett.. 12, , 861 –867 (2012). 1530-6984 CrossRef
Xiaojuan  T.  et al., “Anisotropic thermal and electrical properties of thin thermal interface layers of graphite nanoplatelet-based composites,” Sci. Rep.. 10, , 01710  (2013). 2045-2322 CrossRef
Chang  T.-C.  et al., “Application of graphite nanoplatelet-based and nanoparticle composites to thermal interface materials,” Micro Nano Lett.. 10, , 296 –301 (2015).CrossRef
Veca  L. M.  et al., “Carbon nanosheets for polymeric nanocomposites with high thermal conductivity,” Adv. Mater.. 21, , 2088 –2092 (2009). 0935-9648 CrossRef
Xiang  J. L., and Drzal  L. T., “Thermal conductivity of exfoliated graphite nanoplatelet paper,” Carbon. 49, , 773 –778 (2011). 0008-6223 CrossRef
Liu  C., and Hu  G., “Highly efficient reduction of graphene oxide by sub/supercritical water and their application for thermal interface materials,” Appl. Therm. Eng.. 90, , 193 –198 (2015). 1359-4311 CrossRef
Xu  Y. S., and Chung  D. D. L., “Increasing the thermal conductivity of boron nitride and aluminum nitride particle epoxy-matrix composites by particle surface treatments,” Compos. Interfaces. 7, , 243 –256 (2000). 0927-6440 CrossRef
Prasher  R. S.  et al., “Thermal resistance of particle laden polymeric thermal interface materials,” ASME J. Heat Transfer. 125, , 1170 –1177 (2003).CrossRef
Ghanbarpour  M.  et al., “Thermal performance of inclined screen mesh heat pipes using silver nanofluids,” Int. Commun. Heat Mass Transfer. 67, , 14 –20 (2015).CrossRef
Prasher  R., “Thermal interface materials: historical perspective, status, and future directions,” Proc. IEEE. 94, , 1571 –1586 (2006). 0018-9219 CrossRef
Prasher  R. S.  et al., “Nano and micro technology-based next-generation package-level cooling solution,” Intel Technol. J. Electron. Package Technol. Dev.. 9, , 285 –296 (2005).CrossRef
Zhou  W. Y.  et al., “Study on insulating thermal conductive BN/HDPE composites,” Thermochim. Acta. 452, , 36 –42 (2007). 0040-6031 CrossRef
Shahil  K. M. F., and Balandin  A. A., “Thermal properties of graphene and multilayer graphene: applications in thermal interface materials,” Solid State Commun.. 152, , 1331 –1340 (2012). 0038-1098 CrossRef
Fukushima  H.  et al., “Thermal conductivity of exfoliated graphite nanocomposites,” J. Therm. Anal. Calorim.. 85, , 235 –238 (2006). 1418-2874 CrossRef
Song  W. L.  et al., “Polymer/boron nitride nanocomposite materials for superior thermal transport performance,” Angew. Chem. Int. Ed.. 51, , 6498 –6501 (2012).CrossRef
Larmagnac  A.  et al., “Stretchable electronics based on Ag-PDMS composites,” Sci. Rep.. 4, , 7254  (2014). 2045-2322 CrossRef
Green  M. A., “Ag requirements for silicon wafer-based solar cells,” Photovoltaics Res. Appl.. 19, , 911 –916 (2011).CrossRef
Yoshida  M.  et al., “Novel low-temperature-sintering type Cu-alloy pastes for silicon solar cells,” Energy Procedia.. 21, , 66 –74 (2012). 1876-6102 CrossRef
Rane  S. B.  et al., “Firing and processing effects on microstructure of fritted silver thick film electrode materials for solar cells,” Mater. Chem. Phys.. 82, , 237 –245 (2003). 0254-0584 CrossRef
Perelaer  J.  et al., “Printed electronics: the challenges involved in printing devices, interconnects, and contacts based on inorganic materials,” J. Mater. Chem.. 20, , 8446 –8453 (2010). 0959-9428 CrossRef
Sekitani  T., and Someya  T., “Ambient electronics,” Jpn. J. Appl. Phys.. 51, , 100001  (2012).CrossRef
Ahn  B. Y.  et al., “Omnidirectional printing of flexible, stretchable, and spanning silver microelectrodes,” Science. 323, , 1590 –1593 (2009). 0036-8075 CrossRef
Meysenc  L., , Jylhakallio  M., and Barbosa  P., “Power electronics cooling effectiveness versus thermal inertia,” IEEE Trans. Power Electron.. 20, , 687 –693 (2005).CrossRef
Mahajan  R., , Chiu  C. P., and Chrysler  G., “Cooling a microprocessor chip,” Proc. IEEE. 94, , 1476 –1486 (2006). 0018-9219 CrossRef
Shenogina  N.  et al., “On the lack of thermal percolation in carbon nanotube composites,” Appl. Phys. Lett.. 87, , 133106  (2005). 0003-6951 CrossRef
Itkis  M.  et al., “Thermal conductivity measurement of semitransparent single-walled carbon nanotube films by a bolometric technique,” Nano Lett.. 7, , 900 –904 (2007). 1530-6984 CrossRef

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

Advertisement


 

  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.