PFAS in semiconductor manufacturing: Policy, current data and a look ahead

PFAS (per- and polyfluoroalkyl substances), often referred as “forever chemicals”, are synthetic chemicals widely used for their repellent properties to water, oil, and heat and chemical stability. Several PFAS compounds are extremely persistent in the environment because they degrade slowly under natural conditions, and some can accumulate in the environment and the human body over time. Scientific studies have shown that exposure to certain PFAS could be linked to harmful health effects in both humans and animals.

Policy

In April 2024, the United States Environmental Protection Agency (EPA) finalized the first ever legally enforceable National Primary Drinking Water Regulation (NPDWR) for PFAS. This landmark regulation has drawn attention to the sources of PFAS entering the public drinking water supplies. The regulation establishes Maximum Contaminant Levels (MCLs) for selected PFAS, with particularly stringent limits and health‑based goals of zero detection for PFOA and PFOS. Monitoring requirements start in 2027, with compliance required by 2029, with potential extensions for certain compounds by 2031. Compliance with the MCLs will be determined by using annual average concentrations measured at dedicated sampling points.

1. PFOA (perfluorooctanoic acid): MCL of 4.0 parts per trillion (ppt) with Maximum Contaminant Level Goal (MCLG) of zero.
2. PFOS (perfluorooctane sulfonic acid): MCL of 4.0 ppt with MCLG of zero.
3. PFNA (perfluorononanoic acid): MCL & MCLG of 10 ppt.
4. PFHxS (perfluorohexane sulfonic acid): MCL & MCLG of 10 ppt.
5. HFPO-DA (hexafluoropropylene oxide dimer acid, "GenX"): MCL & MCLG of 10 ppt
6. Mixtures containing two or more of PFHxS, PFNA, HFPO-DA, and PFBS: regulated using a hazard‑index approach.

Estimates for the cost of complying with EPA’s PFAS drinking water rule vary substantially. EPA projects annual compliance of approximately $1.5 billion, while AWWA-sponsored study estimates annual costs of approximately $3.48 billion per year ^[7,11]. The AWWA study further estimates national capital improvement cost to range from $37.11 billion to $48.25 billion ^[11]. These projected costs substantially exceed the currently available federal funding in the range of 9-10 billion dollars for PFAS and emerging contaminant made available through the Infrastructure Investment and Jobs Act and related federal programs. While the initial funding allocations are encouraging, implementations tend to be dynamic in nature due to market forces, and additional funding mechanism are needed to bridge the gap between estimated national compliance costs and currently dedicated federal support.

Data

PFAS compounds represent a very large class of chemicals with different uses, transport fate, and persistence. Therefore, their dominant sources can vary by compound, location, and test data availability. Industrial discharges are considered to be one of the significant contributor of PFAS to wastewater, particularly from sectors that utilize fluorinated chemicals in their manufacturing processes. Estimates indicate that 12-61% of PFOA and PFOS in wastewater can be attributed to domestic sources^[1]. Since most existing public wastewater treatment plants are currently not equipped to remove or destroy PFAS, the PFAS containing treated effluent discharged from these plants ends up in surface water bodies, and thus our ecosystem.

The recent expansion of domestic semiconductor manufacturing, driven by the CHIPS and Science Act (2022), has amplified concerns about potential PFAS contributions from semiconductor manufacturing. Semiconductor manufacturing is a water intensive operation and generates a significant amount of wastewater.

Data on PFAS in semiconductor manufacturing process is not commonly available in the public domain. The 2023 survey conducted by Semiconductor Industry Association (SIA) on PFAS in semiconductor wastewater remains the most exhaustive study with publicly available results ^[2]. So far, three (3) process areas in the chip manufacturing chain have been identified as a source of PFAS: 1) Photolithography 2) PFAS containing wet chemicals 3) Assembly, test, packaging and substrate operations. In addition to these three processes, plasma etch and certain polymeric materials are also considered potential sources for PFAS. The results from SIA survey, which covered 26 semiconductor fabs operated by 7 companies across the US, Europe, and Asia, found a substantial variation in PFAS release from fab to fab. Within this dataset, three key factors of variation were noted:

1. Type of product manufactured at the facility (for example: 200 mm versus 300 mm lines or silicon versus silicon carbide-based wafers).
2. Process chemistries and manufacturing technology utilized.
3. In-house effluent management practices including total volume of effluent generated, how waste streams are collected and segregated, and the treatment systems in place prior to discharge.

Across the surveyed semiconductor facilities, daily wastewater discharge volume ranged from roughly 200 cubic meters per day (m³/day) about 53,000 gallons per day (gpd) to 28,000 m3/day (about 7.4 million gallons per day), indicating large differences in the range among fabs. The concentration of target PFAS compounds in these effluent streams, measured by methods such as EPA 537.1 (modified), EPA 1533, ASTM 7979-20 spanned from non-detect levels to 2,824 nanograms per liter (ng/L), with a study average of 840 ng/L. On a mass basis, the associated PFAS load varied between 0.002 gram per day (g/day) and 13 g/day, with an average of approximately 4g/day of target PFAS per facility. Extrapolated over a year, this equates to an annual average PFAS discharge of about 3.2 pounds per facility, highlighting the potential cumulative impact of the semiconductor wastewater on downstream PFAS loading ^[2]. When compared with municipal wastewater systems, this discharge intensity appears particularly significant. Li et al. reported an average total PFAS concentration of about 98 ng/L in treated effluent from 38 municipal wastewater treatment plants in the United States, with a median treatment capacity of approximately 30 million gallons per day, illustrating typical PFAS levels for mid‑sized public treatment facilities ^[3]. On average, therefore, semiconductor effluent in this dataset is around 8 to 9 times more concentrated in PFAS than treated municipal effluent, positioning semiconductor fabs as relatively significant point sources even though their total discharge volumes are often lower than those of municipal plants. Even if subject to dilution and some attenuation downstream, sustained PFAS discharges of several grams per day per facility can materially influence PFAS loading to receiving waters and, by extension, potential impacts on downstream drinking water sources and aquatic environments.

The look ahead

Domestic semiconductor manufacturing fabs are critical for national security, economic resilience, and technological leadership, as they safeguard a stable supply chain for advanced chips essential to defense systems, critical infrastructure, and emerging technologies such as artificial intelligence (AI) and high-performance computing. At the same time, PFAS chemistry remains deeply embedded in the semiconductor value chain. The current technical and performance constraints makes complete elimination of PFAS‑containing materials from semiconductor manufacturing technically challenging in the near term. The semiconductor industry has taken some steps towards better sustainability, including phasing out perfluorooctanesulfonic acid (PFOS) from the photoacid generators (PAGs) used in photolithography ^[5]. However, fluorinated polymers, perfluorocarbons (PFCs), hydrofluorocarbons (HFCs), and fluorinated heat transfer fluids continue to be crucial for process control and efficiency in the semiconductor chip manufacturing ^[4]. The limitations of analytical techniques add further challenges and complexity to this issue. EPA standard test methods mentioned above are aimed at detecting “target” PFAS compounds and do not fully capture the diversity of PFAS used or generated. More advanced methods such as high-resolution mass spectrometry (HRMS) combined with liquid chromatography (LC) have shown existence of various short-chain and ultra-short-chain PFAS compounds which are not captured by the existing methods ^[6]. Thus, the current monitoring framework may likely underestimate or not reflect the total PFAS loads associated with industrial point sources including the semiconductor manufacturing sector.

The current PFAS regulations highlight a gap between industrial wastewater discharge standards and drinking water limits. The EPA's Effluent Guidelines Program Plan 15 (2023) deferred immediate revisions to the discharge standards for the Electrical and Electronic Components category (40 CFR Part 469), pending further data on PFAS contributions, while public water systems face substantial investments, in both monitoring and compliance, to meet the EPA’s NPDWR MCLs, for which a meaningful share could be traced back to upstream industrial sources.

EPA’s included a key data-gathering step as the agency explicitly launched a nationwide publicly owned treatment works (POTW) Influent Study to monitor industrial PFAS loads in municipal influent. If this study reveals that industrial loads are a primary driver of POTW non-compliance with the new drinking water MCLs, then it could prompt new federal pre-treatment requirements ^[8]. Furthermore, the pressure is not solely domestic in the United States. Globally, the European Union's proposal to restrict approximately 10,000 PFAS compounds under REACH regulations may impact the global semiconductor supply chain, potentially forcing operational changes even in U.S. facilities that rely on international equipment and chemical suppliers ^[9].

Given the funding gaps and significant pressure on public systems to comply with PFAS regulations, the semiconductor industry, proactively could pursue two complementary path for shared goal: advancing on-site PFAS destruction technologies and supporting shared financing models (i.e. public-private partnership) with public entities in their jurisdiction.

Technologically, fabs must pivot from PFAS separation methods to on-site PFAS destruction technology. While treatment technologies such as granular activated carbon (GAC), reverse osmosis (RO), and Ion Exchange are effective at capturing long-chain PFAS compounds, these separation technologies merely transfer PFAS to a solid or concentrated waste stream, often incinerated or landfilled. The scientific frontier is shifting toward destruction technologies capable of mineralizing PFAS on-site. Emerging solutions, such as electrochemical oxidation, have demonstrated significant destruction efficiency in concentration waste streams for both long- and short-chain compounds, which should be further piloted and implemented in large scale commercial applications ^[10].

On the policy initiatives targeted at fund raising, manufacturers could champion Collaborative Water Security Funds. Under this model, semiconductor firms, many of whom are beneficiaries of the CHIPS Act, would directly contribute capital to local municipal utilities for implementation of advanced PFAS treatment infrastructure.

Looking ahead, continued data collection, treatment technology innovation, integrated policy, and shared financial responsibility will be essential to reducing PFAS impacts while supporting the continued growth of the semiconductor industry and bolstering the resilience of municipal infrastructure as communities work together to meet public health goals.

References

[1] Omobayo A. Salawu, Ziwei Han, Naomi L. Senehi, Shannon L. Roback, & Adeyemi S. Adeleye. (2025). Determination of household wastewater PFAS composition and concentrations via neighborhood sampling. Journal of Hazardous Materials, 498, 139834.

[2] 2023 Semiconductor PFAS Consortium Survey Results: PFAS in Semiconductor Fabrication Facility Wastewater. Semiconductor PFAS Consortium (January 2025).

[3] Li, Y., Liu, Y., Loos, M., Ruan, T., & Ebinghaus, R. (2025). Occurrence and fate of per- and polyfluoroalkyl substances (PFAS) in municipal wastewater treatment plants: A comprehensive review and meta-analysis. Environmental Science & Technology, 59(6), 3542–3557.

[4] Semiconductor PFAS Consortium. (2023). PFOS and PFOA Conversion to Short-Chain PFAS-Containing Materials.

[5] Semiconductor PFAS Consortium. (2023). PFAS-Containing Photo-Acid Generators Used in Semiconductor Manufacturing.

[6] ACS Measures. (2025). Practical Guidance on Selecting Analytical Methods for PFAS in Semiconductor Wastewater.

[7] Fact Sheet: Benefits and Costs of Reducing PFAS in Drinking Water. (EPA, April 2024)

[8] U.S. Environmental Protection Agency. Effluent Guidelines Program Plan 15 (EPA, 2023).

[9] European Chemicals Agency (ECHA). Annex XV Restriction Report – Per- and polyfluoroalkyl substances (PFASs) (ECHA, 2023)

[10] Axine Water Technologies. ElectraCLEAR™ PFAS Destruction in Semiconductor Wastewater (Axine, 2024)

[11] Final Technical Memorandum Updating National Cost Estimate for PFAS standards using UCMR-5 (American Water Works Association, July 2024)