Thermally Robust SiC Materials for Next-Generation Power Electronics
13-12-2022 | By Liam Critchley
Materials that are thermally robust and exhibit high thermal conductivity and stability are crucial in power electronics and electronic systems where a lot of heat is generated as a by-product. The inability to dissipate this heat can cause device failure, and if the materials used to build power electronic components are not thermally robust, then the components are going to break down over time.
Thermally robust materials can play both a thermal management or an active role within electronic and photonic devices. Silicon carbide (SiC) is a material that is now starting to be used in next-generation advanced devices, including power electronics, optoelectronics, and quantum computing, but its thermal properties are highly dependent on the crystal phase of the SiC.
Nevertheless, it’s thought SiC has the potential to displace silicon-based architectures or at least be used alongside them because it facilitates fast-switching speeds, low losses, and high blocking voltage. Moreover, power electronic devices often overheat due to localised heat flux, so the thermal management of devices and having materials that can better facilitate thermal dissipation are going to be key for next-generation technologies—as new devices are getting more powerful and are therefore producing more heat in the process.
The Silicon Carbide Materials of Interest
There are several different crystalline structures of SiC, all of which have different properties. When it comes to thermal conductivity, the most studied crystal structures, which are 6H-SiC and 4H-SiC, have lower thermal conductivity values of metals such as copper and silver—which are typically 320-350 W m-1 K-1 for the SiC compared to 400-430 W m-1 K-1 for the metals. While there are other thermal management materials out there that are better than these crystalline phases, the 3C-SiC phase shows promise—with much better theoretical electronic and thermal conductivities—but this crystal structure is much less-studied than the other SiC materials.
One of the challenges with developing 3C-SiC materials for electronic applications is the literature values, and measured values for thermal conductivity don’t add up and are not only much lower than the theoretical values but also lower than 6H-SiC materials. This was cleared up recently, and the lower-than-expected experimental values have been attributed to the presence of boron defects within the material. The presence of these defects leads to an exceptionally strong boron defect-phonon scattering. It’s been found that only 0.1% boron in the material lattice is enough to decrease the thermal conductivity of the SiC by a factor of 2.
Despite not living up to its theoretical values, there is a lot of potential for realising high thermally conducting 3C-SiC materials as both active materials and as thermal management materials for cooling semiconductor devices, so more research is now going into these SiC crystal structures.
Rare Silicon Carbide Moissanite Mineral Stone
Thermal Robustness Found in Wafer-Scale 3C-SiC Crystals
Researchers have now looked into high-quality 3C-SiC crystals (without the boron defects) that were produced using chemical vapor deposition (CVD) methods and grown on the (111) plane of a silicon substrate. This latest study aimed to investigate the real thermal conductivity properties sans defects to determine the true thermal conductivity of wafer-scale cubic 3C-SiC materials and to see how inaccurate the current literature values are.
The researchers used a wide range of methods to probe the thermal properties of 3C-SiC. This included measuring the in-plane and cross-plane thermal conductivity with beam-offset time-domain thermoreflectance (BO-TDTR), as well as performing structural analyses with Raman spectroscopy, X-ray diffraction (XRD), high-resolution scanning transmission electron microscopy (HR-STEM), electron backscatter diffraction (EBSD), and second ion mass spectroscopy (SIMS) to understand the relationship between the microstructure, chemical composition, and thermal conductivity of 3C-SiC samples.
The study results showed an agreement with the theoretical thermal conductivity values of a pure crystal and not with the previously reported defect-present analyses. The analyses showed that the 3C-SiC has a thermal conductivity of over 500 W m-1 K-1 and is the second most conductive material among larger crystals, only surpassed by diamond. This value is 50% higher than commercially available 6H-SiC materials and more than 50% higher than the previously measured values for 3C-SiC. 3C-SiC thin films were also shown to have a record-high in-plane and cross-plane thermal conductivity, which was higher than diamond films of the same thickness.
The SiC material was also grown epitaxially on silicon and showed a thermal boundary conductance that is among the highest for semiconductor interfaces and about ten times larger than diamond-based interfaces. Out of all the SiC material phases, 3C-SiC has also been shown to have the thermal conductivity properties and the highest channel mobility (better electronic properties).
As expected, it was found that the high thermal conductivity was a direct result of the crystal being of high purity and quality, so avoids the strong defect-phonon scatterings. Because the thermal conductivity is higher than the more structurally complex 6H-SiC crystal, the research also showed that the structural complexity and thermal conductivity are inversely related.
The study’s findings have provided fundamental insights into the phonon transport mechanisms within SiC and have shown that 3C-SiC could be an excellent wide-bandgap semiconductor. Because high thermal conductivity has been observed in bulk crystals and thin films, and a high thermal boundary conductance is observed at interfaces, there is a lot of potential for use in active components and thermal management substrates in the next generation of next-generation power electronic and optoelectronic devices.
There are a number of materials out there with high thermal conductivities, but most have one or more challenges when it comes to designing a scalable approach for device manufacturing. So, a scalable and facile manufacturing route can be achieved for thermally conductive 3C-SiC that is defect-free. In that case, there’s a lot of potential for them to be integrated within existing silicon architectures for the next generation of power electronics and optoelectronics.
Reference:
