Once considered quality issues, substrate defects now allow precise control of semiconductor crystal growth

A team led by researchers at Rensselaer Polytechnic Institute (RPI) has made a breakthrough in semiconductor development that could change the way we produce computer chips, optoelectronics and quantum computing devices.

The team, which also includes researchers from the National High Magnetic Field Laboratory, Florida State University and SUNY Buffalo, published their findings last month Nature. Their work deepens the understanding of remote epitaxy, a manufacturing technique in which high-quality semiconducting films are grown on one substrate and then transferred to another substrate.

Remote epitaxy works by placing a thin buffer layer between a substrate and a growing crystal film. The atomic structure of the substrate guides the crystal’s growth through the buffer, but the buffer prevents permanent binding, meaning the completed crystal layer can be peeled off and moved elsewhere.

Until now, researchers believed that the remote epitaxy technique only worked with buffer layers less than one nanometer thick, and that a thicker layer would almost completely overshadow the substrate’s subtle electrostatic forces that drive crystal film growth.

But the recently graduated RPI materials science Ph.D. Student Ru Jia and her lab mates grew crystals through carbon buffer layers up to seven nanometers thick – a 600% improvement – ​​and found that they were still well aligned with the underlying substrate.

“This work shows that remote epitaxy can be mediated by substrate defects such as dislocation,” said Jian Shi, Ph.D. a professor of materials science and engineering at RPI, Jia’s Ph.D. advisor and a senior author on the paper. “In practice, this broadens material choices, improves process windows, and supports scalable membrane release and wafer recycling strategies in real devices.”

Using a zinc oxide/gallium nitride model system in collaboration with Yan RPI professor of materials science Yunfeng Shi, Ph.D., and his team validated the findings with calculations showing that defects called dislocations can mediate remote epitaxy over long distances.

“This work would not have been possible without close collaboration between experts in materials growth and characterization, advanced characterization and atomistic-scale computer simulations,” said Jian Shi.

The researchers tested their approach with multiple crystal/substrate combinations, demonstrating the universality of their findings. As a proof of concept, they built working photodetectors by transferring perovskite crystal films onto flexible substrates, demonstrating the practical feasibility of the technique.

The work suggests that substrate defects, traditionally viewed as quality control issues, can be deliberately designed to remotely control epitaxy processes. For example, manufacturers could use the technique to program functional ‘islands’ or epilayers at specific locations on their crystal films – a level of precision needed to fabricate quantum devices that require precise control of crystal growth.

“The paper provides a mechanism – long-range defect-assisted electrostatic interactions – that engineers can deliberately use to achieve nucleation and alignment in crystal films,” said Jian Shi.