News: Microelectronics
17 September 2020
SMI develops low-temp, high-rate homogeneous epi of 4H-SiC
Funded by the United States Air Force Research Lab (AFRL) in a Phase I Small Business Technology Transfer Research (STTR) project ‘Tooling and Processing for Low Temperature Homogeneous Epitaxy of 4H-SiC Using Novel Precursors’, Structured Materials Industries Inc (SMI) of Piscataway, NJ, USA – which provides chemical vapor deposition (CVD) systems, components, materials and process development services – has developed low-temperature high-rate CVD growth for 4H-SiC (silicon carbide) epilayers.
The effort is to develop a CVD chemistry/process that supports a high growth rate at lower-than-standard temperatures while maintaining quality by effectively eliminating the silicon droplet and other defects. The result will be to enable the growth of thick 4H-SiC epilayers for power devices at relatively high rates and with greater process tool resilience.
In the STTR project, SMI worked with a team of academic institutions, including the University of South Carolina (USC), which has researched chloride- and fluoride-enhanced CVD growth. The program also includes researchers at Morgan State University and a researcher at the Mississippi State University (MSU) who are SiC growth and device experts. The SMI-led team offers expertise in enhanced process SiC material growth and device fabrication.
Thicker SiC films are desired for producing high-voltage devices, necessitating the development of processes with increased growth rates preferably at lower temperatures. Lower temperatures decrease wear on the reactor and decrease thermal cycle time – a more efficient growth rate also decreases process cycle time.
SiC can operate with voltage drops of ~1.2kV per 10μm of thickness. Increasing precursor concentrations to achieve higher growth rates generally leads to the formation of homogeneous nucleation of silicon particles or droplets in the gas phase. Some of these droplets land on the substrate and make the epitaxial layer defective and hence useless for devices. While hardware design and process steps mitigate this problem, it can still be improved on. One solution is to develop an epitaxy growth process using a lower growth temperature of <1500˚C (reducing gas-phase pre-reactions) but generally with a penalty of reduced growth rates.
Another issue is the degradation of graphite parts used in the growth reactor at higher temperature by reactant/process gases. The degradation of the parts continuously changes the thermal profile; leading to process instability and thus resulting in reduced production yield. Hence, a process with a modest reduction in growth temperature while achieving high growth rates will have a significant impact on tool life and throughput. SMI technology development within this STTR project will develop a low-temperature process using novel precursors and other enhancements.
“Now [that] we have an opportunity to create the tooling and processes that increases production yield, reduce cost and improve the quality of the SiC epilayers layers - all are critical for advancing the SiC power device manufacturing - we look forward to scaling the process and demonstrating devices, growth swiftly and efficiently at low temperatures,” says principal investigator and SMI research scientist Dr Arul C. Arjunan.
“Routinely and robustly producing thicker device-quality defect-free SiC at high growth rates and at lower temperatures will be a great benefit to the rapidly growing SiC power device market for decades to come,” believes SMI’s president Dr Gary S. Tompa.
“The Air Force funding of this work recognizes the importance of addressing production costs and capabilities in the SiC supply chain and that an advanced process and the associated hardware to achieve this goal are needed sooner than later,” says an STTR collaborator at USC. “Reducing SiC growth temperature while maintaining high quality at higher rates - and thus lowering costs – is the power device manufacturing dream,” adds the STTR collaborator at MSU.
