Advanced materials, essential for everyday items like batteries, construction materials, and pharmaceuticals, have traditionally been developed through lengthy trial-and-error processes. However, advancements in artificial intelligence (AI) are revolutionizing this field. Polaron, a spin-out from Imperial College London, leverages AI in a novel way by analyzing microscopy images to extract valuable microstructural data. This approach allows for the generation of 3D reconstructions of materials, optimizing their manufacturing processes and enabling the design of new materials with tailored properties, such as faster-charging batteries with longer lifespans.

Polaron's AI-driven methodology enhances materials characterization by transforming raw microscopy data into quantifiable insights, significantly reducing time and labor costs. For example, one client saved approximately 1,000 engineer hours through Polaron’s automated image analysis. The technology also facilitates design optimization, allowing manufacturers to explore various production methods on a microstructural level to improve performance while adhering to practical manufacturing constraints.

While Polaron focuses on battery technologies, its applications extend to various industries, including construction materials, electronics, and pharmaceuticals, where microstructure influences performance. Traditional sectors, where imaging the material structure can lead to enhancements in product quality, stand to benefit greatly.

Looking ahead, the integration of AI will enable more intelligent, data-driven processes that harmonize discovery, design, and manufacturing. By reducing the need for extensive experiments and fostering continuous optimization, AI holds the promise of transforming materials science into a more efficient and innovative field.

Hydrogen peroxide plays a critical role in disinfectants, medical sterilization, environmental cleanup, and manufacturing. However, its traditional production methods are energy-intensive and mainly reliant on large-scale industrial processes. In response, researchers have developed a groundbreaking computational framework aimed at creating cleaner alternatives.

This new framework enables the identification of effective catalysts for generating hydrogen peroxide directly from water and electricity through an electrochemical process dubbed the two-electron water oxidation reaction. Lead researcher Hao Li explains the challenges in catalyst design due to the diverse forms they can take—metal alloys, metal oxides, and single-atom materials—each presenting unique atomic structures that complicate performance comparison.

To overcome these barriers, Li and his team introduced a weighted atom-centered symmetry function to describe catalytic active sites at the atomic level, combining geometric and chemical information for a unified analysis. This innovation, paired with machine learning and reaction modeling, allows the team to predict the performance of various materials effectively.

Their model successfully identified lithium scandium oxide (LiScO₂) as a promising catalyst, achieving approximately 90% efficiency in hydrogen peroxide production while maintaining stability over nearly a week of continuous operation. Significantly, this framework minimizes trial-and-error in catalyst development, making the process more systematic and efficient.

Moreover, implemented in the Digital Catalysis Platform, the framework enhances the predictive capability for reaction properties, paving the way for broader applications beyond hydrogen peroxide production. Researchers anticipate that this method could be vital for designing catalysts for other crucial electrochemical reactions, ultimately contributing to more sustainable chemical production and energy technologies.

The U.S. Department of Energy (DOE) is spearheading the Genesis Mission, a national initiative that integrates artificial intelligence (AI) and advanced computing to expedite American scientific discovery. Savannah River National Laboratory (SRNL) plays a pivotal role alongside 16 other national laboratories in this transformative effort. DOE Under Secretary for Science Darío Gil emphasized the mission's ambitions during a House Science Committee hearing, underscoring a collective dedication to enhancing scientific inquiry through the application of AI.

The Genesis Mission seeks to double the productivity of U.S. science and engineering within the next decade, addressing pressing global challenges in energy, scientific advancement, and national security. Leveraging AI, SRNL has initiated projects targeting the cleanup of legacy waste at sites like Savannah River Site (SRS). By using extensive subsurface characterization data from federal sites, SRNL could potentially save billions while accelerating environmental remediation and optimizing domestic energy production.

Furthermore, SRNL’s Advanced Manufacturing Collaborative facility is positioned to deploy state-of-the-art AI technologies to enhance the production efficiency of defense materials. With decades of data from previous nuclear cleanup operations, SRNL is well-equipped to innovate in fields such as novel materials development, isotope recovery, and subsurface science.

The Genesis Mission will establish Model Teams across national labs and industry partners to develop AI-driven approaches to a range of objectives, including:

• Enhancing power system operations using AI and autonomous technologies fueled by grid data.
• Securing critical mineral supply chains through advanced discovery and processing techniques.
• Shortening design timelines for magnetic confinement fusion studies.
• Promoting next-generation nuclear fission innovations.

These collaborative endeavors promise to catalyze groundbreaking discoveries, advance America’s energy leadership, and bolster national security.

HotHouse Therapeutics has launched with £2.9 million in pre-seed funding to innovate an AI-driven plant bioengineering platform aimed at revolutionizing drug discovery and production. This Norwich-based startup secured investments from SynBioVen, Start Codon, UKI2S, Twin Path Ventures, Wren Capital, and various angel investors, along with a grant from Innovate UK.

The startup’s platform tackles a significant challenge in the vaccine industry: the reliance on the unsustainable bark of a rare Chilean tree for the critical adjuvant QS-21. This fragile supply chain is unable to keep pace with global demand, particularly impacting low- and middle-income countries.

HotHouse Therapeutics proposes to solve this issue by bioengineering plant systems that can be cultivated efficiently anywhere. Their AI-powered design tools work in tandem with a transient plant expression system to transform plant leaves into chemical factories capable of producing natural and novel molecules in a matter of days. This approach promises a scalable, carbon-positive production method suitable for global health initiatives that seek reliable supply and affordability.

The BotanAI design engine and BotanBIO production system streamline the generation of high-purity compounds, with a current focus on enhancing next-generation saponin-based vaccine adjuvants. Additionally, the company plans to extend its reach into traditional drug discovery.

With the potential to unlock over 20 million novel molecules through AI-driven pathway prediction, HotHouse’s sustainable production methods support vertical farming and long-term environmental goals. The firm anticipates collaborations with global health organizations and biopharma companies to expedite access to its innovative adjuvant offerings.

David Sheppard, CEO, encapsulates the mission: “Unlock powerful chemistry, make it sustainable, and deliver it to those who can reach patients worldwide.”