Amir Arbabi and Peng Bai Receive DARPA Young Faculty Awards

Content

This story was first published by the UMass News Office.

Two faculty members in the College of Engineering have received Young Faculty Awards from the Defense Advanced Research Projects Agency (DARPA). Each will receive $500,000 over two years, with a potential for an additional third year.

Amir Arbabi, associate professor of electrical and computer engineering, is developing new technology that enables 3D integration of photonic integrated circuits.

“Most of the semiconductor chips that you might hear about are electronic, but there are also photonic chips that work at optical frequencies,” he explains. “Instead of electricity going around in the plane of the chip, what you have is light that is propagated and processed, and data is encoded in that.”

He explains that different types of functions of optical chips – generation, detection and encoding of data – come from different materials. “You cannot have a single chip that optimally does all of those functions. So, for example, light generated in one chip is transferred to another one for encoding. And then for detection, it is sent to another chip. You need to have links between these chips.”

Current approaches to establishing these connections consume significant chip real estate on a millimeters-by-millimeters sized chip, provide limited optical bandwidth, and are prone to high losses and misalignment errors.

Arbabi says his method aims to address these limitations and increase the misalignment tolerance, meaning that connected chips can still have high performance even if the alignment isn’t perfect. The resulting technology will have applications for quantum optics, optical AI accelerators and optical fiber communications, “which is the backbone of the internet,” he adds.

Peng Bai, associate professor of chemical engineering, is the principal investigator on a project to help discover new nanoporous materials that can selectively separate mixed compounds. To do so, he is building a machine learning framework that can both rapidly identify the optimal material structure for a given application and predict if such a structure is feasible to create in reality.

He compares it to a highly sophisticated sponge. “If you put a sponge in oily water, both oil and water will be absorbed, but depending on the application, what we want is for the sponge — the nanopores — to only absorb one of the components.” As such, nanoporous materials have been investigated for a wide range of applications such as water harvesting and separation of critical elements.

What the “sponge” material will absorb is highly dictated by its structure and shape. With this in mind, Bai and his team will encode materials by viewing them as 3D analogues of images for their machine learning framework.

The Materials Genome Initiative has led to the proliferation of large materials databases. However, whether these materials can feasibly be synthesized is hard to rationalize. “They all look reasonable even to a seasoned synthetic chemist — the computationally predicted materials have the same atomic arrangement and bonding geometry exactly like known materials — but a machine learning model was able to tell 99% of the computationally predicted materials apart from the real materials,” says Bai.

He also sees this project as a way to explore the integration of machine learning tools with existing bodies of knowledge. “From a research point of view, a very exciting aspect is how we can couple what we know from physics-based thermodynamic knowledge with data-based machine learning tools,” he says.

More information about the Young Faculty Awards can be found on DARPA’s website.