New research, led by Queen Mary University of London, demonstrates that a double-layer electrode design, guided by fundamental science through operando imaging, shows remarkable improvements in the cyclic stability and fast-charging performance of automotive batteries, with strong potential to reduce costs by 20–30%.

The research, published today in Nature Nanotechnology, was led by Dr. Xuekun Lu, Senior Lecturer in Green Energy at Queen Mary University of London.

In the study, the researchers introduce an evidence-guided double-layer design for silicon-based composite electrodes to tackle key challenges in the Si-based electrode— a breakthrough with strong potential for next-generation high-performance batteries.

The evolution of automotive batteries has been driven by ever-increasing demand for driving range and charging speed since EVs took off 15 years ago. Silicon electrodes can provide 10 times higher theoretical capacity and faster charging, but their large-scale deployment is held back by substantial volume changes of up to 300% during charge/discharge cycles. This means they degrade quickly and don’t last long.

Assisted by multiscale multimodal operando imaging techniques, this research reveals unprecedented insights into the electro-chemo-mechanical processes of the graphite/silicon composite electrodes. Guided by these improved mechanistic understandings, a novel double-layer architecture is proposed, which addresses key challenges in material design, exhibiting significantly higher capacity and lower degradation compared to conventional formulations.

Dr. Xuekun Lu, who led the study, said, “In this study, for the first time, we visualize the interplay between microstructural design and electro-chemo-mechanical performance across length scales—from single particle to full electrode—by integrating multimodal operando imaging techniques.

“This study opens new avenues for innovating 3D composite electrode architectures, pushing the boundaries of energy density, cycle life, and charging speed in automotive batteries, and thereby accelerating large-scale EV adoption.”

Professor David Greenwood, CEO of the WMG High Value Manufacturing Catapult Center commented, “High silicon anodes are an important technology pathway for high energy density batteries in applications like automotive. This study offers a much deeper understanding of the way in which their microstructure affects their performance and degradation, and will provide a basis for better battery design in the future.”

A research team led by Profs. Wang Mingtai and Chen Chong from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences has developed an antimony trisulfide (Sb₂S₃) solar cell with a record conversion efficiency of 8.21%. This achievement marks the highest performance ever reported for this type of solar cell.

The study was published in Advanced Energy Materials.

Sb₂S₃ has drawn increasing attention as a promising light-absorbing material due to its abundance, non-toxicity, and favorable optoelectronic properties. However, devices fabricated via solution methods typically suffer from high defect densities and interface mismatches, which limit carrier transport and restrict photovoltaic conversion efficiencies to around 6–7%.

To overcome these challenges, the researchers proposed a full-dimensional defect passivation approach using the permeation effect of degradable phenethylammonium iodide (PEAI) in amorphous Sb₂S₃ films.

PEAI pretreatment of amorphous Sb₂S₃ films enables [hk1]-oriented crystallization, full-dimensional defect passivation (bulk and interfaces), and dual-interface energy-level reconstruction via Cd-I and Sb-I bonding. The PEAI reduces CdS surface energy and preferentially adsorbs on Sb₂S₃ (211) planes, promoting [hk1] orientation and enhancing carrier transport.

Furthermore, the penetrated PEAI increases the carrier lifetime by a factor of 3.7, verifying effective defect suppression.

As a result, the researchers successfully fabricated an Sb₂S₃ bulk heterojunction solar cell with a conversion efficiency of 8.21%, the highest reported to date.

This work sets a new performance benchmark for Sb₂S₃ solar cells and provides valuable insights for the design of next-generation, high-efficiency thin-film solar cells.

Lauren Goode: Well, they’re all interested in growing more. Who among us, Mike? But the hyperscalers refers to this class of major tech companies or cloud service providers. So Meta, Amazon, Microsoft, Google, they’re all in that category.

Molly Taft: Yeah, and I think it’s important to remember that these companies have so much money and they have an ability to raise capital like nobody’s business. So they’re able to do some really crazy stuff to build quick and to build-out really, really big. And they’re getting pretty creative, because their goals right now are to build these things quickly and get them up and running so they can basically use this physical infrastructure to compete with each other.

Lauren Goode: I think that’s right, Molly. I think there’s a lot of frenemy building happening right now, and I would just love to be a part of their group chats when all of these announcements are being made.

Michael Calore: Yeah, and speaking of frenemies, the other sphere of influence that these companies are operating in is the political sphere. Obviously, in order to build a giant data center somewhere, you need to have the political will to do it, which means you need buy-in from the local residents, the local government, the state, the country. So what’s happening in the political sphere with folks who want to build more data centers and people who oppose it, regulation? How is that playing out?

Molly Taft: That’s a great question, and I think if you look at the national conversation, it’s quite different than what’s happening on the local level. You have Washington, you obviously have an administration that is very friendly to the idea of an American AI empire. Importantly for the energy conversation, the way that the Trump administration has approached this support has been through support of fossil fuels. They would really like for all data centers to be powered with oil and gas, a little bit of nuclear and coal. And this works out great for those industries as well. If you’re going to have this massive expansion of power demand, it’s really cool to be in the middle of that and be the one that everyone wants to turn to for energy resourcing. And then on the other side, there has been this influx of local opposition to these data centers for a variety of reasons, be it the water use, be it fears about rising electricity rates, be it noise, and some of the really big struggles have catapulted this issue to national conversation. I’m thinking about xAI in Memphis. When Elon Musk wanted to get xAI up and running, he installed a bunch of unpermitted gas turbines in order to get xAI working that he installed in a majority Black community in Memphis that already had severe issues with air pollution and asthma. And those folks made themselves known. Earlier this year, there was an attempt in DC to impose a moratorium on any state regulation around AI at all. It was an incredibly broad inclusion in the Big Beautiful Bill that ultimately didn’t succeed. But one of the people who opposed it publicly was Marjorie Taylor Greene, who actually mentioned data centers in her opposition, and she compared AI to Skynet, the fictional AI from the Terminator movie franchise. So, this is getting some strange bedfellows in league with each other, I think this kind of contrast between what the administration is trying to push forward and some very powerful energy companies that stand to gain from it, versus some truly grassroots local movements and people concerned about the impacts of what these things are going to do in their communities.