Linköping University Reveals Method to Fit Nanometre-Sized Layers of Semiconductors Together

one month ago by Alessandro Mascellino

Scientists from Linköping University and SweGaN have discovered a new method to fit together layers of semiconductors as thin as a few nanometres.

The findings resulted in the creation of a new type of transistor suitable for high-power electronic devices.

The research, which has been funded by research grants from both the Knut and Alice Wallenberg Foundation and the CoolHEMT programme, is part of the EU Horizon 2020 initiative. 

 

A vital part of the materials research. Pictured: one of the world’s most outstanding transmission electron microscopes, Arwen, at Linköping University (LiU).

A vital part of the materials research. Pictured: one of the world’s most outstanding transmission electron microscopes, Arwen, at Linköping University (LiU). Image Credit: Magnus Johansson (LiU).

 

Epitaxy in Semiconductors

In order to withstand both high temperatures and current strengths, semiconductors with particular properties need to be utilised. Gallium nitride (GaN) is one of them.

The semiconductor is currently used for efficient light-emitting diodes, but the new findings suggest it could also be useful in other applications, such as transistors.

Allowing the GaN vapour to condense onto a wafer of silicon carbide forms a thin coating. This process is known as epitaxy, and it is widely used in semiconductors as it makes it possible to determine both the crystal structure and the chemical composition of the nanometre film formed.

 

SweGaN is a company born from materials science research at Long Island University. The company specialises in tailored electronic components from gallium nitride.

The collaboration between SweGaN and Linköping University (or LiU⁠—more from which can be read here on Electronics Point) spurred the team to experiment with the combination of GaN and silicon carbide (SiC), as both materials can withstand strong electric fields.

Initial results, however, showed that the surface between the two crystalline materials fit poorly. This led to the atoms mismatching with each other and ultimately to the failure of the transistor.

 

Transmorphic Epitaxial Growth

The research team, led by LiU’s Lars Hultman and Jun Lu, then attempted to resolve the said issue by adding an even thinner layer of aluminium nitride between the two layers.

This not only led to the transistor no longer failing, but also showed that the component could cope with significantly higher field strengths than they had anticipated. Further research on this data showed that the answer to this phenomenon was to be found at the atomic level, in a couple of critical intermediate surfaces inside the components.

After analysing the phenomenon further, the scientists discovered this was due to a new epitaxial growth mechanism that they have named ‘transmorphic epitaxial growth’.

This method induces the strain between the different layers to be gradually absorbed across a couple of layers of atoms. By controlling this growth at the atomic level, it is possible to create components able to withstand high voltages up to 1,800V.

“We congratulate SweGaN as they start to market the invention,” said  Lars Hultman (pictured below).

 

Lars Hultman, Professor of Materials Science and Head of the Division of Thin Film Physics at Linköping University.

Lars Hultman, Professor of Materials Science and Head of the Division of Thin Film Physics at Linköping University. Image Credit: Linköping University.

 

“It shows efficient collaboration, and the utilisation of research results in society. Due to the close contact we have with our previous colleagues who are now working for the company, our research rapidly has an impact also outside of the academic world,” he added.

SweGaN has not yet announced a date for the commercialisation of devices based on this discovery, but the huge interest the research has shown (nearly 1,000 downloads one week after publication) may be an indicator that this will happen sooner rather than later.

Comments