manufacturing · environment · chemistry · family: making one is easy. making a million is the problem
chips builtwith forever chemicals
Semiconductor manufacturing depends on PFAS 'forever chemicals' with no viable alternatives
Problem statement
Semiconductor fabrication relies on per- and polyfluoroalkyl substances (PFAS) — persistent "forever chemicals" — in critical process steps: photolithography (photoacid generators), plasma etching (fluorinated gases), chemical-mechanical polishing (surfactants), and cleanroom equipment (fluoropolymer seals and tubing). SEMI estimates PFAS are used in >50% of critical chip fabrication steps. No viable PFAS-free alternatives match the performance of current PFAS-based materials for these applications. With incoming EPA and EU REACH regulations threatening to restrict chemicals essential to chip production, the industry faces a potential manufacturing crisis.
Why this matters
The global semiconductor industry generates >$600 billion in annual revenue and underpins the entire digital economy. PFAS contamination near semiconductor fabs has triggered >$50 billion in environmental litigation and cleanup costs across industries. The proposed EU universal PFAS restriction could directly affect chip production in European fabs. TSMC alone uses ~10 million liters of PFAS-containing chemicals annually. NSF DCL 24-043 encourages proposals on "new semiconductor manufacturing processes that reduce or preferably eliminate the use of per-and polyfluoroalkyl substances (PFAS)." A manufacturing disruption from PFAS regulation without alternatives could cascade through the entire electronics supply chain.
What’s been tried and why it hasn’t worked
Some fluorinated etching gases (SF6, NF3, C4F8) have alternatives under development, but none match the etch selectivity and profile control of current processes. EUV photolithography resists contain fluorinated polymers essential for the chemically amplified resist mechanism — alternative resist chemistries (metal-oxide, molecular resists) are 5–10 years from production readiness. Fluoropolymer seals and tubing have no replacements that resist the chemical environments in wet processing tools. Some PFAS-containing surfactants in CMP slurries have been replaced with shorter-chain alternatives, but these still contain fluorine and may fall under future regulation. The fundamental problem is that PFAS are used precisely because of the properties that make them persistent — extreme chemical stability, low surface energy, and thermal resistance.
What would unlock progress
Discovering new chemical functionalities that match C–F bond properties (stability, hydrophobicity, low surface energy) in process conditions (high temperature, plasma, aggressive solvents) without environmental persistence. This likely requires entirely new classes of materials rather than incremental modifications of existing fluorinated compounds. Alternative approaches include redesigning fabrication process steps to eliminate the need for PFAS-dependent functions entirely — for example, dry processes replacing wet chemical steps that require fluorinated surfactants.
Entry points for student teams
A student team could map the PFAS dependency tree for a specific semiconductor process step (e.g., CMP slurry formulation), identifying which PFAS properties are strictly necessary and which could be achieved through alternative chemistry. Alternatively, a team could evaluate non-fluorinated polymer seals for resistance to specific semiconductor wet-bench chemicals, generating compatibility data that the industry needs. Relevant skills: polymer chemistry, surface science, materials engineering, semiconductor process engineering.
Genes
Structural cousins: stuck the same way, in other fields
Sources
NSF DCL 24-043, "Funding Opportunities for Engineering Research in Emerging Areas of Advanced Manufacturing," NSF Directorate for Engineering, accessed 2026-02-19. go to source ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
- NSF DCL 24-043 is the primary source. SEMI White Paper on PFAS in Semiconductor Manufacturing (2023) provides industry data.
- Distinct from `environment-pfas-destruction-at-scale` — that brief covers destroying PFAS in contaminated water; this brief covers finding PFAS replacements for manufacturing.
- The `failure:ignored-context` tag applies because semiconductor process design historically treated PFAS persistence as a feature (chemical stability) rather than recognizing the environmental consequence.
- The `constraint:regulatory` tag applies because incoming regulation is the proximate driver — without EPA/EU restrictions, the industry would have no incentive to find alternatives.
- Lim, X. "Could the world go PFAS-free? Proposal to ban ‘forever chemicals’ fuels debate." Nature 620, 24–27 (2023), https://doi.org/10.1038/d41586-023-02444-5, provides accessible overview.
Reconciliation 2026-09-29: A site review read this write-up against its cited source, NSF Dear Colleague Letter 24-043, fetched in full on 2026-09-29 from https://www.nsf.gov/pubs/2024/nsf24043/nsf24043.jsp, which now redirects to https://www.nsf.gov/funding/opportunities/dcl-funding-opportunities-engineering-research-emerging-areas/nsf24-043. (1) The text quoted the letter as calling out "eliminating harmful chemicals like PFAS in semiconductor manufacturing" as an engineering frontier. Those words are not in the letter. Its one sentence on the subject, under the heading "Environmentally sustainable manufacture of semiconductors," reads "Developing new semiconductor manufacturing processes that reduce or preferably eliminate the use of per-and polyfluoroalkyl substances (PFAS)." The sentence now quotes the letter's own words. (2) Source Notes cited Lim, X., "Can the semiconductor industry eliminate 'forever chemicals'?" Nature 624, 234–237 (2023). A Crossref search on 2026-09-29 found no article with that title, and Crossref lists other articles on pages 233 to 238 of Nature volume 624. Crossref and the Nature page list one 2023 news feature on PFAS by XiaoZhi Lim, "Could the world go PFAS-free? Proposal to ban ‘forever chemicals’ fuels debate," Nature 620, 24–27 (2023). That article was read in full on 2026-09-29 from the Internet Archive copy dated 9 August 2023. It has a section on computer chips that describes PFAS in photoresists and in etching gases. The citation now carries the title, volume, pages and DOI copied from that record. This assumes the earlier citation was meant to point to that article. Confirmed against the letter and left as written: the Source line's title, publication number and directorate. NOT checked: every figure in the write-up. None of them is in the letter or in the Nature article, so they rest on the SEMI white paper, which could not be opened. The figures are PFAS use in more than 50% of critical fabrication steps, more than $600 billion in annual industry revenue, more than $50 billion in litigation and cleanup costs, about 10 million liters a year at TSMC, and 5 to 10 years for alternative resists. The SEMI white paper is cited without a link and semi.org refused automated requests. The Nature article supports the use of PFAS in photoresists and etching gases and says alternatives to these are not easy to find. The named etching gases, the statements about EUV resists, seals, tubing and polishing slurries, and the claim in the title that no viable alternatives exist are not in the letter or the Nature article, and were not checked for the same reason. All of these are left as written. Genes untouched.