Luisier wins SNSF Advanced Grant to develop s


image: Example of a modern semiconductor chip
see After

Credit: @Mathieu Luisier

The Swiss National Science Foundation has awarded Professor Mathieu Luisier more than two million Swiss francs over a five-year period to develop a computer-based simulator that will simplify the design of electronic components and speed up their manufacturing process.

Moore’s scaling law, which observes that the sizes of transistors shrink at an exponential rate, has seen the dimensions of these components, which form the heart of semiconductor chips, approach the atomic limit. Quantum mechanical effects dominate here and the development and production of reliable devices at this scale is complicated – Intel, for example, had to delay the introduction of its 7-nanometer chip by four years as it went back to the drawing board. drawing on the design of transistors.

While complicated, it’s also the way to go: smaller transistors mean more can be fitted to a piece of silicon, allowing more powerful and complex components to be built. Luisier envisions, for example, the co-integration of light-emitting/detecting modules that enable high-speed intra-chip communication, thermoelectric generators that help recycle heat dissipated by microprocessors, and non-volatile memories that will enable energy-efficient data storage.

These innovations and new functionalities will however only be possible if new materials such as, for example, two-dimensional materials, inorganic compounds such as BaTiO3 and complex oxides can be combined with the silicon that forms the basis of these chips. Besides the integration of new materials, all these features involve electrical, optical and thermal effects. They lead to important interactions between electrons, phonons and photons which could, in turn, have a negative impact on the performance of electronic devices and should therefore be designed and controlled from the initial design phase.

Moore’s law has held up so far, says Luisier, due to the continued adaptation of transistor manufacturing recipes and the gradual introduction of technology boosters such as stubs or 3D FinFETs. These advances, while driven by the intuition of visionary researchers, were underpinned by conventional technology computer-aided design (TCAD) tools that were used to validate new ideas.

Device engineers have realized, however, that such design approaches cannot handle the atomic scale due to the strong influence of quantum mechanical effects. The combination of multiple functionalities, the mixture of heterogeneous materials and the close interaction between electrical, thermal and optical phenomena further complicate the situation. The design of next-generation chips, says Luisier, will depend on the availability of advanced modeling tools that can be used during the design process to accurately predict the characteristics of multi-functional, multi-material nano-devices. Although there are more advanced modeling tools, including one, OMEN, developed by Luisier himself, their practical use is limited by various considerations. In some cases, they can only be used in tiny structures of less than 1,000 atoms, use empirical parameters as inputs, or ignore certain particle-particle interactions, for example. Researchers need a new class of device simulators with improved modeling capabilities.

This is where the SNSF-funded Quantum Transport Simulations at the Exascale and Beyond project comes in. Its ultimate goal is to go “beyond the state of the art” with the development and release of open source, versatile and portable software. , scalable and advanced device simulator, “QuaTrEx”. The tool will showcase unique capabilities and set new standards in terms of simulation capabilities, code implementation, and device applications.

“By considering appropriate physics, offering a wide palette of device geometries, leveraging all types of computing resources, and being released as open source software, the proposed QuaTrEx has the potential to accompany the semiconductor industry for at least 20 years, save it from design problems that could result from inaccurate predictions provided by standard device simulators, and pave the way for semiconductor chips. next-generation drivers,” said Luisier.

The SNSF funding will support two post-docs and three doctoral students for the duration of the project.

The SNSF The Advanced Grants were introduced in 2021 to provide researchers from Swiss institutions with a substitute for the Advanced Grants of the European Research Council, for which they are currently not eligible: Switzerland is considered as a non-associated third country by the program of EU Horizon Europe research. Of 232 Advanced Grant applications submitted to the SNSF last year, 24 were eligible for funding.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.


About Author

Comments are closed.