Cost-effective composite substRatE for Advanced GaN applicaTIons : DeVEloPment and evaluatIon – CREATIVEPI

Wide bandgap technologies, and more particularly the GaN technology, introduce new possibilities in the landscape of power electronics and microwave through intrinsic qualities superior to silicon and III-V conventional technologies. The rapid evolution of this technology over the last decade is due to the considerable progress made on the growth techniques, improvements in materials, design of new architectures and components. GaN technology has undeniable advantages for high power applications, high frequency, but also for low noise linear and nonlinear applications. Besides the indisputable success of III-N optoelectronics, the transfer of the GaN power electronics technology from research laboratories to commercial foundries has already begun (eg. Cree MOCVD-grown GaN/SiC), but the GaN market will only reveal its true potential once production costs will be reduced when the technology will be perfectly reliable for high power and high temperature applications. GaN industry will become a major player in electronics by offering new solutions, and by transforming traditional architectures.
The high cost and low availability of bulk GaN substrates have driven nitride-chip manufacturers to use other platforms for growth, such as silicon carbide (SiC) and sapphire. These two alternatives have provided the basis for the production of countless devices, but in every case intermediate layers have had to be inserted into the structure to combat significant lattice and thermal mismatches. Most GaN-based electronic and optoelectronic devices are today industrially grown on these substrate materials but, they are proven to be costly and not available in diameters larger than 6-inch. The composite substrate (i.e. deposition of a buffer layer on a low-cost large-area handle wafer) has been identified as one of the solutions to remove these technological barriers by providing an increase in useable area (less edge exclusion) while significantly decreasing the processing cost per wafer and opening full access to automated SMIF wafer handling equipment. It presents, moreover, a unique opportunity for the integration of advanced devices with standard CMOS electronics.
The CREATIVEPI project aims to develop and validate a cost-effective high-performance 2-inch (6-inch during dissemination) thin-film composite substrate intended to bridge the gap between low-cost single-crystal silicon and high-performance, high-cost single-crystal compound semiconductor materials currently used as the starting substrates for III-N devices. This program will demonstrate that the new AlN-CSOS (Compound semiconductor on silicon) composite substrates facilitate the growth of high quality III-nitride device structures and are less costly than the current platforms. The present collaborative research project involves a consortium of three leading Laboratories in the domain of Material Science and high-frequency high-power devices, l’Institut des Matériaux Jean Rouxel de Nantes (IMN), l’Institut d'Electronique, de Microélectronique et de Nanotechnologie de Lille (IEMN) and le Centre de Recherches sur l’Hétéro Epitaxie et ses Applications (CRHEA). It also involves multidisciplinary approaches by including a wide range of technical domains.

In the frame of the project, demonstrators, in this case HEMT devices and power LEDs, will be realized on the Si/AlN 2-inch substrates to allow comparison with the state-of-the-art and enable to characterize the complete technology chain. Inventions to be patented will be methodically selected in order to avoid any dissemination of knowledge or IP infringement. The project results will be preferentially exploited through the creation of a spin-off company dedicated to the production of AlN-CSOS substrates. Nevertheless, technology transfer to a SME (eg Soitec, IQE, etc.) or a group (e.g. Saint-Gobain, Sumitomo, etc.) will also be taken into consideration.