Researchers at Samsung Electronics in Korea have successfully fine-tuned a method to build nanoscale semiconductor chips at industrial scales. The research team identified opportunities to improve the atomic layer etching (ALE) approach, which offers significant advantages over the conventional methods but was plagued by difficulties for practical use.
The arrival of the transistor began the era of electronics that has become an integral part of our daily lives. From phones to satellites, electric vehicles (EVs) to fighter jets, electronics are everywhere and are getting smaller every year. Central to this accomplishment is the reduction in the size of the transistor, which has gone from half an inch big during its invention to a few nanometers (10^7 times) smaller today.
Companies like Samsung have leveraged the power of transistors to build a wide range of devices and used their industrial prowess to produce them at scale. While doing so has helped millions of individuals access the power of electronics, the shrinking size of transistors is presenting problems at mass manufacturing scales.
ALE – Replacing the conventional approach
Chip manufacturers have conventionally used the reactive-ion etching (RIE) process for fabricating their semiconductor chips. This is a dry fabrication technique that uses a chemically reactive plasma to remove material from a substrate.
It also includes a physical step in which the substrate is bombarded with energetic ions to break surface bonds and generate patterns that serve as transistors in nanoscale devices. Although widely used, the approach has limitations such as surface damage on the substrate, plasma redeposition, and high equipment costs.
Atomic Layer Etching (ALE) has been explored as the alternative to RIE since its controlled conditions offer reduced ion-induced damage and improved selectivity. However, the approach is marred by low throughput, and the process is also quite complex to execute.
Samsung scientists solve the problem
When faced with a problem in science, researchers typically use a wide range of approaches to solve it. An approach that works in a laboratory is not necessarily the most scalable and therefore cannot be applied at an industrial scale. This is true for ALE too, which academic researchers have studied extensively in an attempt to improve processes or address its low throughput.
A research team led by Keun Hee Bai at Samsung Electronics in Korea approached ALE from an industry perspective and sought to identify what could be done to improve it and deploy it in a real-world setting.
The researchers noted that the issues raised against ALE are the series of sequential steps that need to be taken to achieve the desired result. While there are challenges to solve in these steps, Bai and team argue that these do not need to be resolved independently but as a collective challenge in ALE’s complex reaction process.
According to the researchers, careful process design and optimization as well as architecture simplification could help in solving most of these issues, something that industrial production focuses on much more than academic researchers.
Most studies have emphasized many advantages [of ALE] including low damage, high selectivity, and fine control,” said Bai in a press release. “We add to this list by reporting the advantage of a more vertical profile because ALE can create 90-degree angles with uniform widths into the layers.”
The research findings were published in the Journal of Vacuum Science & Technology A.