EUV photoresist chemistry

Three companies decide whether the world's most advanced chips can be made. The light-sensitive chemistry that patterns features below five nanometres — EUV photoresist — is supplied by a market controlled to 90% by three Japanese firms, and India makes none of it. As of 2025, the country's import dependence stands at 100%.

The material sounds simple and is not. A chemically amplified EUV resist is three things working in concert: a photoacid generator (PAG) that decomposes under extreme-ultraviolet light to release catalytic acid, an acid-reactive polymer matrix that changes solubility when that acid cleaves its protecting groups, and a basic quencher that controls how far the acid diffuses. At sub-3nm nodes, the polymer must hold line-edge roughness below 3.5 nm — the jaggedness of a printed edge — or the pattern fails. JSR Corporation holds roughly 32% of global volume, TOK around 28%, delivering resist with line-edge roughness of 3.4 nm. These are qualified against specific scanners and fab processes; a formulation is not a chemical so much as a decades-long relationship between a material and a machine.

Demand is climbing steeply. In 2024, nearly 950,000 wafers at 5nm and 3nm were patterned with EUV resists, a 53% jump from 620,000 in 2023. Kioxia reported EUV adoption in advanced NAND memory approaching 40% of critical layers in 2025, projected past 70% by 2028. The market, worth USD 1.78 billion in 2025, is forecast to reach USD 4.12 billion by 2032.

India stands at the beginning. There is no known indigenous EUV photoresist programme, no production, no commercial effort. The Bharat Semiconductor Research Centre at IIT Madras was announced by the Ministry of Electronics and Information Technology to commence in 2024, but no specific EUV photoresist research has been publicly disclosed. On 9 April 2026, the government notified a Special Economic Zone for Tata Semiconductor in Dholera, Gujarat — India's first fabrication facility.

The gap persists because the supply chain rewards accumulated expertise. Concentration among Japanese, US and South Korean suppliers reflects decades of experience and sustained R&D. The barrier is visible in how rarely it is crossed: when Dongjin Semichem began supplying Samsung's 3nm lines in February 2025, it was the first new commercial entrant in over a decade.

What it would take is layered mastery: PAG chemistry, base resin synthesis, ppm-level purification, roughness-mitigation additives — and, above all, qualification against real scanners. Newer metal-oxide resists, already 28% of volumes by Q3 2024 at 3.1 nm roughness, offer one door. It is a long climb, but a mapped one.