A 1950s Material Just Set a Modern Record For Lightning-fast Chips (sciencedaily.com)
- Reference: 0180315041
- News link: https://hardware.slashdot.org/story/25/12/06/2045205/a-1950s-material-just-set-a-modern-record-for-lightning-fast-chips
- Source link: https://www.sciencedaily.com/releases/2025/12/251204024240.htm
"The discovery also enhances the prospects for silicon-based quantum devices..."
> Scientists from the University of Warwick and the National Research Council of Canada have reported the highest "hole mobility" ever measured in a material that works within today's silicon-based semiconductor manufacturing.... The researchers created a nanometer-thin germanium epilayer on silicon that is placed under compressive strain. This engineered structure enables electric charge to move faster than in any previously known silicon-compatible material...
>
> The findings establish a promising new route for ultra-fast, low-power semiconductor components. Potential uses include quantum information systems, spin qubits, cryogenic controllers for quantum processors, AI accelerators, and energy-efficient servers designed to reduce cooling demands in data centers. This achievement also represents a significant accomplishment for Warwick's Semiconductors Research Group and highlights the UK's growing influence in advanced semiconductor materials research.
[1] https://www.sciencedaily.com/releases/2025/12/251204024240.htm
21st century material (Score:4, Informative)
Nobody was doing strained-layer epitaxy in the 1950s.
Re: (Score:3)
No but they were experimenting with germanium. My understanding is that researched stopped because while they could get better results with germanium, the problem was how to use it germanium practically as it is 1.6ppm in the Earth's crust while silicon is 282,000 ppm.
Re: (Score:3)
That's a common misconception. The real reason: it was/is a threat to germanium based lifeforms AKA Space Nazis.
I wonder... (Score:2)
TFA doesn't mention this; but if PN junctions in this new material have the same 300mV forward voltage drop that regular germanium diodes and transistors have, then that will contribute a bit to the speed. I think it should also reduce power consumption by a non-trivial amount - perhaps by a factor of two when compared with regular silicon junctions.
Operation Paperclip (Score:2)
Operation Paperclip -> Clippy -> Chippy?