Chemical mechanical planarization

Every leading-edge chip passes through the same unglamorous step twenty or more times: a wafer is pressed against a spinning pad, flooded with an abrasive slurry, and polished flat to atomic-scale tolerances. This is chemical mechanical planarization. As of 2026, India imports 100% of the pads it uses and the overwhelming majority of the slurries — domestic players hold under 5% of the slurry market.

The difficulty is that CMP does something no other technique can. It achieves both global planarization across an entire 300mm wafer and local planarization within individual die features at once — a dual capability that spin-on-glass, etch-back and reflow cannot match at advanced nodes. It works by combining chemical oxidation, which softens the surface, with mechanical abrasion using slurry particles typically 50 to 250 nanometres across. Silica-based slurries dominate oxide polishing; cerium oxide slurries achieve up to ten times higher removal rates for shallow trench isolation. Each of these is a distinct formulation problem, and each must be qualified against a customer's exact process before it earns a single order.

Where India stands is early but no longer nil. In July 2026, the country's first CMP Pad Technologies Hub was inaugurated at T-Works in Hyderabad, a partnership with Japan's Toho Koki Seisakusho, and it houses India's first Toho Koki CMP groove-processing machine. Joint R&D with IITs and Indian SMEs targets commercial pad manufacturing within 18 to 24 months — roughly 2027 to 2028. On slurries, Gujarat Fluorochemicals is making a strategic entry aimed at cost-competitive formulations for legacy nodes from 180nm to 65nm. The market is real: India's CMP segment was valued at USD 56 million in 2024 and is projected to reach USD 84 million by 2032.

The gap persists because CMP is a fortress of concentration. Globally, four to five suppliers — Cabot Microelectronics, DuPont, FUJIFILM, Entegris — account for well over half of slurry revenue, protected by the sheer depth of process qualification. In India, Cabot alone holds 65% of consumption for 300mm production. Beneath the pad and the slurry sit further missing layers: CMP equipment, endpoint detection software, and post-CMP cleaning chemistries, none of which India yet makes.

What it would take is patient qualification across every dependent layer, not a single breakthrough. The Union Budget 2026-27 allocated ₹8,000 crore to the semiconductor mission — its largest single-year outlay — and ISM 2.0 names equipment and materials manufacturing explicitly. The pad hub is the first foothold on a long face.