SBIR Phase I: Novel Cold Cathode E-Beam Sources for Advancing Semiconductor Manufacturing

About This Grant

The broader impact/commercial impacts of this Small Business Innovation Research (SBIR) Phase I project will be in developing new solutions for semiconductor chip manufacturing for artificial intelligence (AI). Mask writers are machines that are integral to the semiconductor industry, as these create the stencils (masks) that are sandwiched together to form advanced computer chips. The number and complexity of the masks required for these chips are increasing beyond the capacity of state-of-the-art machines. The company is developing a component device that has the potential to introduce disruptive change in the fundamental design of mask writers. The project will harness this technology to tackle manufacturing bottlenecks in mask production that are caused by the accelerating demand for advanced intensive AI applications. The work will lead to a more efficient solution for producing semiconductor chips faster and with high computational power. The project has the potential to effect a financial impact of greater than $20M annually after three years, and to kickstart entry of the fundamental technology into additional market segments. This Small Business Innovation Research (SBIR) Phase I project will lead to a wafer-scale process for creating controllable cold cathode electron sources that advance integrated circuit design and manufacturing. Semiconductor chip fabrication is achieved by performing photolithography through stencil masks that are themselves created using either laser- or electron-beam mask writers. The properties of these tools dictate spatial resolution, precision, and turnaround time. The multiple electron-beam (multi-beam) writer has achieved the smallest feature sizes, thus displacing the laser writer for artificial intelligence (AI) applications. The goal of this project is to help eliminate two significant bottlenecks that hinder production and advancement in multi-beam technology: (1) Beam quality and parallelization (reliability and yield) are physically limited by the industry’s dependence on thermionic (hot) electron sources; and (2) escalating processing power for implementing design code for advanced mask sets is stressing computational resources. The proposed device will provide a means to accelerate chip production using controllable electron beams for faster and higher quality mask writing. The focus of this project on solving the first bottleneck should also help address the second bottleneck, expanding the degree of complexity of the chip sets that can be designed and manufactured for AI. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.