Program Manager: Michael Filler
It may be possible to reinvent how microelectronics are made using a two step process:
These nanomodular electronics could enable a "fab in a box" and make fabricating microelectronics as straightforward as printing this document.
You can find a detailed program roadmap here.
The goal of the nanomodular electronics program is to build a new process for manufacturing microelectronics that are customizable down to the transistor, use a plethora of materials, and can be made on-demand without etching silicon or requiring a massive, expensive facility. By creating tiny, modular components separately from laying them down and wiring them together, we could make custom microelectronics as easily as printing this document.
The process of using light and chemicals to construct microelectronics from a single piece of silicon has both enabled Moore’s Law and created many unfortunate downstream effects: industry consolidation in volatile parts of the world, fragile supply chains, high overhead for making custom circuits, and large barriers for innovating on the process because everything is so tightly coupled. Incremental improvements to the current system are possible, but won’t get around the fundamental limitations of the underlying process paradigm – the planar process.
Price per mole for bulk products
Even the transistors at the core of microelectronics are still shockingly expensive compared to other manufactured goods. At roughly $10 billion per kilogram, transistors cost thousands to millions of times more than even drugs, which are created in bulk chemical processes that scale with volume. If you could create transistors and other circuit components in the same way, making microelectronics could be as ubiquitous as writing software. However, changing how we make transistors requires an entirely new process for creating microelectronics.
Developments in nanotechnology, colloidal chemistry, precision additive manufacturing, and computer vision suggest that this new process is possible!
Creating nanomodular electronics needs a research program that coordinates several parallel component-focused projects towards a single goal: a student competition to explore the possibility space of nanomodular electronics and uncover compelling uses.
A student competition is a powerful “forcing function” for creating a general purpose technology. It ensures the system will be flexible and consistent enough to accommodate unexpected ideas and non-expert users. Both synthetic biology and VLSI electronics became widespread because immature tools were put in the hands of creative and energetic students.
If the history of general purpose technologies is a guide, nanomodular electronics could have impact far beyond their obvious near term applications like tamper-resistant electronics with unique identifiers, physically implemented neural networks, and myriad low-volume applications that depend on custom circuits.