Keynotes

The presence of non-idealities and defects in VLSI technologies is an immutable reality, which, however, is often overlooked by device engineers, researchers, technologists, and designers alike. I will argue that VLSI reliability, in fact, “makes or breaks” any new technology. Superficial device reliability optimization is possible using phenomenological observations only, but I will argue that solid physical foundations and a thorough understanding of the underlying degradation mechanisms are essential both for truly dependable lifetime projections and for novel device pathfinding. Based on the detailed investigation of gate oxide defects, our “defect engineering” approach enables, among other things, oxidation techniques for advanced VLSI architectures. In deeply scaled devices, degradation mechanisms can be decomposed down to individual defects, with each defect measured separately. Such knowledge then allows us, e.g., to model the degradation statistics of deeply-scaled devices and to predict the likelihood of their failure. Robust degradation models enable us to project wider safe operating areas, which in turn allow us to design better-performing circuitry at a given technology node and thus limit costs. Moreover, the ubiquitous presence of defects can be, in fact, embraced, and the in-depth knowledge of defect properties can be used to our advantage to design new devices and applications, ranging from tamper-aware chip ageing monitors to physically unclonable functions and reservoir computing to identify users.

Modern (asymmetric) encryptions are based on unproven mathematical assumptions such as the complexity of the factorization problem. Quantum computers will be able to efficiently break these encryptions, making internet connections insecure. One of the replacement candidates is quantum key distribution (QKD), where the information is encoded in single photon states of light. Due to the fundamental laws of quantum physics, any eavesdropping attempt would be immediately revealed.

For QKD to work in the field, one needs to generate single photons and transmit them over long distances. In this talk, I will present our efforts in realizing a compact room temperature single photon source based on a color center in the 2D material hexagonal boron nitride. The performance is sufficient to outperform laser-based QKD protocols. The emitter is directly coupled to a photonic integrated circuit that routes the single photons to different experiments. The payload is currently being integrated on a 3U CubeSat and will be launched in 2025 as part of the QUICK3 mission. A constellation of small satellites equipped with the technology developed in QUICK3 could serve as a backbone for a satellite-based quantum internet. Moreover, 2D materials can also be used to build radiation-tolerant and ultra-low weight electronics or opto-electronics and have been qualified by us for use on satellites.

Performance of 2D material-based devices for space applications. (a) Simulation of the radiation environment in LEO at 500 km altitude. Licensed under CC BY 4.0 (Ref. [1]) (b) Quantum emitter array in hexagonal boron nitride (hBN) for satellite-based quantum communication. Reprinted (adapted) with permission (Ref. [2]). Copyright 2024, American Chemical Society. (c) Atomically-thin field-effect transistor (FET) for light-weight satellite electronics. Licensed under CC BY 4.0 (Ref. [1]) (d) Performance of the hBN emitter for space-to-ground scenarios in daylight conditions (red dots). Licensed under CC BY 4.0 (Ref. [3]) (e) Performance of the 2D-FET before and after space qualification. Licensed under CC BY 4.0 (Ref. [1]).

[1] Nat. Commun. 10, 1202 (2019)
[2] ACS Nano 18, 5270-5281 (2024)
[3] APL Quantum 1, 016113 (2024)