A startup seeks to miniaturize particle accelerators to revolutionize semiconductor manufacturing

Inversion Semiconductor develops compact accelerators to revolutionize chipmaking.

: Inversion Semiconductor, founded in 2024 by Rohan Karthik and Daniel Vega, aims to miniaturize particle accelerators, using laser wakefield acceleration (LWFA) to transform semiconductor manufacturing. This technology could challenge current market leader ASML by creating more efficient and powerful chip production methods. LWFA enables high-energy acceleration over short distances, allowing Inversion to propose a compact device that's 1,000 times smaller than traditional accelerators and capable of delivering up to 10 kilowatts of output power. The startup faces hurdles, including the integration of new technologies with existing systems and the manufacturing demands of the semiconductor industry.

Inversion Semiconductor, founded in San Francisco by entrepreneurs Rohan Karthik and Daniel Vega in 2024, is shaking up the semiconductor industry with its groundbreaking laser wakefield acceleration (LWFA) technology. The company, operating under the wing of Y Combinator, seeks to revolutionize the established norms of chip manufacturing. LWFA involves using ultra-short laser pulses to create plasma waves that accelerate electrons with unparalleled efficiency and power, taking the form of compact particle accelerators. The startup's primary competitor, ASML, the current market dominator, uses extensive and costly extreme ultraviolet (EUV) lithography for production.

The central ambition of Inversion Semiconductor is to harness LWFA to replace traditional particle accelerators, which span kilometers, with much smaller devices capable of producing highly concentrated energy. By crafting a system that is 1,000 times smaller than conventional devices and capable of outputting up to 10 kilowatts of power, Inversion aims to promote a new era of semiconductor production efficiency. The company has caught the attention of renowned industry players like Tesla and Applied Materials, highlighting the potential real-world applications of their technology in sectors ranging from industrial imaging to semiconductor inspection.

One of Inversion's immediate technological goals is to advance its Starlight source, aiming to produce 1 kilowatt of soft X-ray light in a wavelength range from 6 to 20 nanometers. Such a development could significantly impact the accuracy and speed of semiconductor manufacturing, potentially supporting a variety of industrial applications beyond just chipmaking. Inversion is actively collaborating with notable establishments such as the Lawrence Berkeley National Laboratory and the BELLA Center, fostering research in their BELLA-LUX project that seeks to improve laser stability and enhance light generation processes.

Despite promising avenues and partnerships, the startup is not without challenges. LWFA technology requires sophisticated, petawatt-class lasers that are not only expensive and large but also difficult to operate reliably in the demanding 24/7 environment of chip fabrication. Additionally, integrating Inversion's novel light sources with existing lithography equipment poses another significant hurdle, as potentially entirely new mirrors and supporting technologies may be required for shorter wavelength utilization.

Ultimately, Inversion's overarching vision is to break through the physical limits of current chip manufacturing practices, breathing new life into Moore's Law—the observation that the number of transistors on a microchip doubles approximately every two years. Should they succeed in making this technology affordable and practical, not only would semiconductor manufacturing evolve profoundly, but a host of industries reliant on advanced light sources could also see transformation. The company's ambition to develop complete lithography systems is clear, yet achieving production-ready systems remains an optimistic endeavor with skepticism from industry analysts. However, as the founders persist, the innovation-driven race to advance chip technology continues.

Sources: TechSpot, Lawrence Berkeley National Laboratory, BELLA Center, Applied Materials, Y Combinator

startupsinnovationManufacturing