Accelerator miniaturization proposals promise 1000 times reduction, potentially boosting semiconductor manufacturing speed by 15 times - Inversion Semiconductor puts forth the concept of compact particle accelerators using petawatt lasers
In a world where making tiny, reliable, and repeatable light sources for chips is a massive headache, a San Francisco startup, Inversion Semiconductor, is shaking things up with their ballsy plan. This Y Combinator-backed bunch aspires to revolutionize the industry by creating a microscopic yet mighty powerful light source based on a compact particle accelerator. They claim this tech would outshine ASML's existing technology by a staggering 33 times, potentially paving the way for producing even tinier chip features.
At the core of Inversion's ingenious contraption is a pocket-sized particle accelerator, 1,000 times smaller than conventional ones, yet capable of delivering a whopping 10 kW of output power. Though petite, it leverages the 'Laser Wakefield Acceleration' (LWFA) method to zoom ahead in the chipmaking race. According to the plan, this powerful light source could either turbocharge chip production 15 times or simultaneously power multiple chipmaking tools, slashing costs significantly.
Despite boasting these titanic claims, Inversion Semiconductor is not looking to rest on its laurels. The company intends to take things a step further by building complete lithography tools to go head-to-head with ASML. This vision, though intriguing, faces some humongous challenges. For instance, Their specific type of accelerator requires a hefty dose of pricey, power-guzzling petawatt-class lasers. And unless they collaborate with ASML (or maybe some other manufacturers), they'd need to develop their own lithography systems and establish a whole new ecosystem—a time-consuming and costly undertaking.
Given that the company lacks experience in manufacturing high-volume, 24/7 devices, their aspirations are nothing short of audacious, and there's no guarantee they'll meet their paper promises.
The Science Behind Inversion
The dynamic duo behind Inversion Semiconductor, Rohan Karthik (CEO) and Daniel Vega (CTO), with their master's degrees in mechanical engineering and applied physics, aim to create a tabletop particle accelerator that's 10 times more powerful than ASML's future plans. This small wonder could produce lasers with wavelengths between 20nm and 6.7nm, including 13.5nm light used by ASML today in both Low-NA EUV and High-NA EUV lithography.
Lasers with sub-10nm wavelengths, called soft X-rays, are not yet being used in chip production due to their high absorption rates by most materials. However, these sub-10nm wavelengths might prove to be a fascinating field for future research.
Inversion's ambitions transcend just a powerful light source. They have dreams of dominating the lithography sector entirely. Using particle accelerators as light sources for lithography tools is an oft-discussed concept in the industry. Inversion seeks to use what they call 'tabletop particle accelerators' that can accelerate electrons to extraordinary energies over centimeters, rather than kilometers, like those seen in luminous giants such as the Large Hadron Collider used by CERN.
Hoisting Sails on the Wakefield Waves
Immersion Semiconductor intends to use accelerators that rely on the LWFA technique, which is significantly distinct from methods employed by ASML and CERN. LWFA harnesses powerful, eye-blink-short (femtosecond-scale) laser pulses interacting with plasma to create potent electric fields, push electrons aside, and generate plasma waves, or ‘wakefields.’ Electrons become trapped and rapidly accelerated within these waves, gaining substantial energy in a short distance as they scamper back to their original position. The plasma wave belts accelerate electrons in fields that are 100 - 1000 times stronger than conventional accelerators, according to the Imperial College London.
These accelerated electrons can then be utilized for various practical applications, including compact X-ray sources and semiconductor lithography, among others. Unlike traditional EUV sources, LWFA generates radiation that is coherent, monochromatic, and precisely tuneable, enabling wavelengths shorter than 13.5 nm, which could be critical for next-gen lithography systems.
The LWFA mechanism propels electrons to energies reaching multiple giga-electron volts (GeV) over extremely brief distances, miniaturizing high-energy electron acceleration systems significantly, from vast installations down to tabletop-sized devices. This solar-powered innovation might spark further creativity in the semiconductor sphere.
Immersion Semiconductor's Short-term Goals
Inversion Semiconductor's accomplishments to date include setting up a small laser lab within the Y Combinator offices to develop new laser stabilization techniques and constructing initial LWFA prototypes capable of emitting short-wavelength radiation. They have also collaborated with the Lawrence Berkeley National Laboratory and the BELLA Center on the BELLA-LUX project, focusing on refining laser stability and enhancing the generation of suitable light for semiconductor use.
The company's immediate goal is to develop Starlight: a high-power, tunable light source capable of yielding 1 kW of soft X-ray light within the 20 nm to 6 nm wavelength range. If successful, this device could find applications in industry X-ray imaging and advanced semiconductor mask inspection. Tesla and Applied Materials are among those expressing interest in these avant-garde leaps.
Parallel to this, the company is working on advanced mirror systems to reflect and focus the generated EUV light, required for precise wafer patterning. The upcoming lithography system based on this technology, designated LITH-0, will be fueled by Starlight, with the objective of demonstrating practical silicon wafer patterning capabilities. However, nobody knows when Inversion Semiconductor's LITH-0 will see the light of day, so to speak.
The Nasty Details
On paper, Inversion Lithography's plans seem solid, and LWFA seems like a nearly perfect approach for generating EUV radiation or light. However, things get mighty convoluted in reality.
First off, an LWFA accelerator chamber may be small, but it calls for mind-boggling petawatt-class, ultrafast laser systems that are complex, large, and fancy. Keeping these lasers at the ready for non-stop fab work is a new feat nobody has tried. It's also unclear whether Inversion Semiconductor's setup can fire these lasers at an acceptable repetition rate per second.
Second, the experts at the John Adams Institute for Accelerator Science at Imperial College London have acknowledged that LWFA produces electron beams with significant energy spread and divergence beyond 1 GeV. For lithography, the generated light must be incredibly stable in wavelength, direction, and coherence to achieve precise and repeatable patterning. Instability means poor resolution, leading to performance variability and yield loss.
Lastly, while today's LWFA-based tools featuring a 13.5nm laser source can utilize mirrors and optics developed for ASML's Low-NA and High-NA EUV tools, they inch closer to shorter wavelengths, their team would need to innovate and create new mirrors and optics. This would be a hassle, especially if Inversion Lithography actually decides to develop their own lithography systems—which would mean building a whole new industrial ecosystem.
A more realistic approach might be making their LWFA-based source compatible with existing tools from ASML. However, integrating an LWFA light source with existing EUV lithography scanners would be a complex endeavor, with numerous challenges such as developing new beam shaping, focusing, and metrology systems to name just a few. While ASML has tackled all the challenges related to its Cymer light sources, one can only wonder whether they are interested in playing nice with third-party tools. As for other makers of litho machines—Canon and Nikon—they haven't managed to create EUV (or beyond EUV) scanners, so the odds of them joining the dance aren't particularly high. And don't forget, we're talking about a powerhouse light source here, so the sector would also need new resists, pellicles, and other consumables to make everything work swimmingly.
Perhaps the most significant hurdle for Inversion Lithography is that they have no experience producing constantly serviced, mass-produced tools engineered for high-octane, around-the-clock fabs and compatible with other equipment in the building.
Science and technology are central to Inversion Semiconductor's ambitious plans. They aim to create a powerful light source based on a compact particle accelerator, harnessing the 'Laser Wakefield Acceleration' (LWFA) method. This technology, if successful, has the potential to outshine existing ASML technology by 33 times, potentially leading to even tinier chip features. Moreover, Inversion seeks to develop complete lithography tools, venturing beyond just a powerful light source and aspiring to dominate the lithography sector.
