Singapore Outlines Research Focus At SEMICON
Heterogenous integration, millimetrre wave and beyond, and wide bandgap semiconductors amongst key research pillars, says Terence Gan of Singapore's Economic Development Board
At SEMICON Southeast Asia 2021, Terence Gan, SVP, Semiconductors, Economic Development Board (EDB) of Singapore, delivered an opening address which outlined the main microelectronics research pillars for Singapore in the medium-to-long term including heterogenous integration, millimetrre wave and beyond for future communications, and wide bandgap semiconductors for next-gen power electronics.
“Our goal has been for Singapore's semiconductor industry to grow at least as fast as the global semiconductor industry. This means that as the industry doubles in size in the next decade, so too, will Singapore's semiconductor industry,” said Gan.
"Today, in Singapore, we are doing R&D on wafer to wafer and chip to wafer bonding, hybrid bonding, and chiplet packaging. But these are possible because of 25 years of learning, from when we set up the first Electronic Packaging Research Consortium in 1996. One notable achievement is fan out wafer level packaging. Around 2012, almost ten years ago, STATS-ChipPAC commercialised eWLB, a type of fan-out wafer level package.
"The technology that was ahead of its time. Today, we are doing research to extend the capabilities of fan out wafer level packaging. For example, A*STAR's Institute of Microelectronics is collaborating with the leading industry players to adapt fan out wafer level packaging for millimetre wave 5G, by putting into one package the power amplifier, MMICs, passive components and millimetre wave antenna arrays. Doing this improves performance while reducing the package area. This is especially important for mobile applications, " said Gan.
The Future of Microelectronics Initiative
How does Singapore decide what research areas to invest in?
Gan said: "The first method is to let a thousand flowers bloom. We screen research proposals via competitive grant calls and rigorous peer reviews. And we provide follow-on funding for the most promising research.
"The second method is for the big bets. Because the funding required is bigger, we need to be more deliberate in choosing the research areas. But once it is approved, it leverages the resources across the entire research ecosystem.
"And so, we have identified five microelectronics research pillars for Singapore for the medium to long term - heterogeneous integration, millimetre wave and beyond for future communications, novel architectures for edge AI, wide bandgap semiconductors for next generation power electronics and RF and piezoelectric MEMS.
"We will deploy significant capital and talent to do world-class research in these areas. For example, for the pillar on wide bandgap semiconductors, we will invest in a new lab to do R&D and pilot production for silicon carbide and gallium nitride and maybe for gallium oxide in future. We will develop our own baseline IP for epitaxy and post- epi processes. We will also develop novel devices, such as silicon carbide super junction MOSFETs and 200GHz GaN-on-SiC high electron mobility transistors and millimetre wave ICs. You can think of it as an R&D foundry for anyone with a good idea to use."
To be successful, Terence Gan said Singapore needs global partnerships and to build communities involving materials suppliers, chip designers, manufacturers, equipment OEMs, universities and research labs.
"Through R&D, we develop and retain our base of talented scientists and researchers. To make Singapore's ambition to be a world-class research and munufacturing hub a reality, we also need a strong pipeline of talent. Our goal has been for Singapore's semiconductor industry to grow at least as fast as the global semiconductor industry. This means that as the industry doubles in size in the next decade, so too, will Singapore's semiconductor industry.
"Given the increased complexity of semiconductor research and manufacturing becoming more innovation intensive, we expect the number of R&D scientists and engineers needed in Singapore to double too. Today, there are about 9000 semiconductor R&D persons in the private sector in Singapore, of which about 1,000 are PhDs. We are making plans to invest more in our schools to encourage more students to graduate with microelectronics skills. There is urgency, more than before, to invest heavily to nurture the talent we need."
He concluded: "The coming decade will be an exciting one for the semiconductor industry. The industry is expected to double in size, the demands of digitalisation will accelerate innovation and we need more of the brightest young people to join us. I hope to be able to collaborate with all of you to succeed together. "