IonQ: Improving battery chemistry with quantum computing

The most expensive part of an electric car is usually the battery. According to a 2020 study by Oliver Wyman for the Financial Times, batteries alone account for almost 40% of the total production cost of a battery electric vehicle (BEV), which also keeps their overall costs higher: it still costs about 45% more to produce a BEV than an equivalent thermal car.

Much of this cost comes from the materials used to make the batteries. Every battery has a cathode that does most of the heavy lifting when it comes to storing and releasing energy, and that cathode is made of rare metals like lithium, nickel, and cobalt that must be mined from the earth and refined from intensively at high purity before they can be used. To pack more energy, more complex combinations incorporating even rarer metals are often used.

To help meet the growing demand for electric vehicles while remaining profitable, nearly every automaker is investing heavily in battery chemistry research, from finding cheaper, longer-lasting battery materials to sourcing and the production.

Beyond cost, there are a variety of other reasons to find better battery materials. Improving energy density – the amount of energy a fully charged battery can hold – would allow for increased range and greater flexibility in where batteries are placed in the vehicle. Improving the number of charge and discharge cycles that batteries can endure makes vehicles more durable and allows drivers to avoid the dreaded and costly “battery swap” that many early EV adopters are currently experiencing. Improvements are also possible in terms of charging speed, heat generation during discharge, reliability, flammability and safety, etc.

This intensive research has paid off. The total price of an electric vehicle battery has fallen nearly 90% since 2010, according to Bloomberg NEF. But prices still need to drop even further, to around $100 per kilowatt-hour, to become competitive with combustion car prices.

Still, finding new battery chemistry or improving the architectures we already have is hard work. You need to understand how two molecules will interact on a very deep level: if they react, how fast they react, what by-products are produced, and how the reaction could be tuned with the addition of other materials like electrolytes. The oldest way to research potential molecules is the simplest: synthesize compounds in a wet lab and see how they work.

Physically testing molecules is, of course, both slow and expensive, so for much of their research, battery chemists now turn to computers. But classical computers are not well suited to model quantum behaviors. Solving the Schrödinger equation completely even for a modestly sized molecule like lithium dioxide is impossible, so chemists resort to computational techniques that rely on lossy (and still computationally intensive) tricks like the approximation by Born-Oppenheimer. This limits the accuracy of the simulations produced, and while it helps limit wet lab work to the most promising candidates, much of the “real” understanding still occurs when these compounds are actually synthesized.

Many automakers, including Daimler, Toyota and now Hyundai, are turning to quantum computing as an accelerator for battery research. IonQ partners with Hyundai in its initiative to develop new variational clean quantum resolution (VQE) algorithms to study lithium compounds and their chemical reactions involved in battery chemistry, a key part of the vision of Hyundai’s 2025 strategy to make their organization a provider of smart mobility solutions, including selling 560,000 electric vehicles a year and a range of twelve or more battery electric vehicle models by mid-decade .

Quantum computing is an exciting catalyst for chemistry-based problems like batteries, because quantum computing was, in a sense, designed to solve chemistry problems. In his now famous 1981 talk at MIT’s 1st Conference on Physics and Computation, Richard Feynman proposed what we now call Universal Quantum Computing specifically because it would be good to solve problems of quantum physics and therefore of chemistry, by explaining that a exact the simulation of a quantum mechanical phenomenon (what these chemical interactions are) can alone be successfully performed with a quantum mechanical system. Or, to put it in its more colorful language, “nature isn’t classical, damn it! and if you want to do a simulation of nature, you better do quantum mechanics.”

To do this, we use one quantum system (our quantum bits) to simulate another (the battery chemistry) using quantum circuits, which allows us to use techniques such as a VQE algorithm to find out something about this system. In the case of VQE, we find that the system ground state energy, which is the most critical piece of information for understanding how a molecule will interact with other molecules. This technique offers more precision than a classical approximation because it can fully resolve all the atomic nuclei and electrons in the system, which might never be possible with classical computational techniques, which do not allow this direct mapping.

Teaming up with a partner like Hyundai is an ideal collaboration for IonQ. They bring deep expertise on these battery chemistry issues and their constituent molecules, and we bring our state-of-the-art hardware and our long history of innovation in quantum chemistry, including the demonstration of an end-to-end pipeline for the large molecule simulation and water simulation. Together, we believe we can continue to advance the state of the art, starting with molecules like lithium oxide – already a leap forward in quantum chemistry – and expanding to compounds and larger and more complex interactions as our techniques and hardware capability improve, potentially enabling long-sought breakthroughs such as solid-state and “post-lithium” battery technologies.

Ultimately, these tiny quantum simulations should have an impact on some of the biggest problems we face. Reducing our dependence on fossil fuels for transportation is a major part of the fight against human-induced climate change, and it cannot be done without cost-effective and efficient electric vehicle technology. Improving their range, efficiency, charging time and safety could accelerate change. At IonQ, our mission is to build the world’s best quantum computers to solve the world’s most complex problems, and it’s hard to think of a better example than helping to make electric vehicles a primary mode of transportation across the world.

Sherry J. Basler