Accelerated Materials Design & Discovery (AMDD) Explained

Toyota Research Institute
Apr 9 · 5 min read

How TRI is making the future of electric mobility a reality

By: Brian Storey, Director of AMDD, Toyota Research Institute

The Accelerated Materials Design and Discovery (AMDD) group at Toyota Research Institute (TRI) is dedicated to making electrified mobility a reality. We are a group of researchers working closely with universities to help make electric vehicles (EVs) efficient, affordable, reliable, and truly emissions free. At Toyota, we think of EVs as one solution in our diverse portfolio of sustainable vehicles that includes battery electric, plug-in hybrid, and fuel cell powertrains.

Toyota is embracing multiple forms of electrification, from Plug-In Hybrid’s to Hydrogen Fuel Cells

We are passionate about EVs because of their environmental benefits, which many people first think of as reducing carbon dioxide (CO2) emissions. While CO2 is an important consideration, EVs can also play a critical role in reducing the pollutants which are responsible for millions of premature deaths globally every year. We are also living in an exciting time of rapid change in the auto industry. Customer demand for EVs is getting stronger everyday and with further improvements we expect that the EV will soon represent a better option for many customers.

“LF-Z Electrified” a BEV Concept Car Symbolizing the Next Generation of Lexus

We are not alone and automakers are ramping up to collectively spend tens of billions of dollars on EVs over the next few years. So what will our small group at TRI contribute to this transition? And more importantly, what does the AMDD group actually do? To start to answer these questions, let’s consider our name.

A is for accelerated. The development and evolution of new energy technologies is inherently slow, often taking decades for noticeable change. Batteries are a perfect example. In the early 1900s, a large fraction of cars were battery powered. While the modern lithium-ion battery found in today’s car is much more advanced than those from over a century ago, the pace of change in batteries seems unremarkable when compared to other technologies. For example, the Wright brothers’ first successful flight was in 1903, and now we are sending spaceships carrying robot explorers to other planets to look for evidence of life. Increased adoption of EVs requires accelerating battery innovation timelines so they are closer to what we see in many other fields.

M is for materials. While the challenges for driving EV adoption involve technological, social and economic factors, it is only through developing new materials that we can make leaps in performance and system longevity. We have a long way to go before reaching the theoretical limits of energy density and efficiency for batteries or fuel cells. Finding better materials can get us closer to these performance goals. Further, many materials that are vital to current EVs — platinum for the fuel cell and cobalt for the battery — have a negative environmental impact due to their mining and/or scarcity. We need sustainable replacements. Just as energy is an inherently slowly evolving field, so is new materials development. The need for acceleration is doubly urgent.

D is for design. Materials design means that we would like to intentionally control the structure of the material, from the atomic to macroscopic scale, to provide targeted properties and performance. Complex technologies like airplanes, are fully designed in simulation before ever taking flight. The role of simulation in designing new materials does not quite work the same way. Even though compute power has increased by a factor of 10,000,000 since the 1970s, this growth has not led to a dramatic increase in the number of new technologically useful materials. Simulation still plays an important role in materials design, but it is apparently not sufficient; we also need large sources of experimental data to help us create fundamentally new design principles.

D is (also) for discovery. While related to design, discovery is subtly different. Discovery implies a leap beyond design rules and moving toward new theories, insights and understanding. Traditionally, discovery has resulted from expert insight with a touch of serendipity. There are many examples where an expert scientist trying to do one thing, actually found something unexpected but even more important. Louis Pasteur is famously quoted, “chance favors only the prepared mind”. In recent years, discovery has been enhanced by both computational and high-throughput experimental screening of many possibilities. These methods have been successful in conducting broad explorations, but are more limited in making discoveries. The possible combinations of new materials is literally infinite and our data is so complex that deriving insights is impossible for the human mind alone. We need new approaches that can better “prepare the mind” to capitalize on large data sources, connect different disciplines and make breakthrough discoveries.

You might have noticed that I still really haven’t answered the question — what does AMDD actually do? How are we going to solve the problems I’ve outlined above?

AMDD team and university collaborators at our annual research meeting in 2019

To get the real answers to those questions, you will have to follow us. On this Medium page, our team will discuss their work and how we are solving these grand challenges by bringing automation and artificial intelligence (AI) to the design and discovery process. We will speculate on how we can use the power of the computer and the expert scientist working together to make breakthroughs. We will demonstrate how we are using robots to conduct faster, more reliable experiments. We will examine how to bridge the enormous gap between computer designed materials and ones we can actually make. We will explain why we are embracing open source software and open data within the research community. In the coming months, we’ll share the excitement of our projects, our people, and our approach.

The Battery Informatics Lab at SLAC National Accelerator Lab is supported by AMDD. (Credit Jacqueline Orrell/SLAC National Accelerator Laboratory)

New materials are critical to making EVs the best choice for the customer and the environment. We won’t rest until we can bring the same qualities of affordability, dependability, and safety found in a 22 year-old Camry to electric vehicles.

Interested in joining AMDD or TRI? Please visit the careers section of our website.

Toyota Research Institute

Amplifying Human Ability with Human-Centric AI, Robotics, Energy & Materials, and Machine Learning