Pete Singer, Editor-in-Chief
PerFluoroalkyl and Polyfluoro Alkyl Substances (PFAS) are vital to the semiconductor industry due to their unique and unparalleled properties. They are used throughout the entire semiconductor manufacturing supply chain.
Although some PFAS pose a risk to health and the environment, they have come under increased scrutiny by regulators and government bodies around the world, unfortunately, in a very politicized way. Already, some states in the U.S. have enacted legislation that would ban PFAS (Minnesota, home to 3M, has passed the most restrictive legislation to date, banning the sale of some PFAS containing products by 2025, enforcing the need to report all PFAS containing products in 2026, and banning the sale of all bar essential PFAS containing products by 2032), with legislation also being proposed at the Federal level and in Europe.
Although well intentioned, these measures – which many observers see as poorly thought out and incredibly over-reaching – could have a dramatic and negative impact on the semiconductor industry in the very near future.
“The intent might be quite good,” notes Chris Jones, Environmental Solutions Business Development Manager at Edwards Vacuum. “Some of the PFAS compounds are unpleasant and getting them out of any supply chain where their management is neither well-regulated or controlled is going to be really important moving forward.” The problem is that the regulations as they are being written are overly broad. “Trying to list 10,000 or 12,000 compounds as being of interest is not a particularly great way of going about environmental regulations,” Jones said.
David Isaacs, Vice President, Government Affairs at the Semiconductor Industry Association, said, “policymakers should proceed carefully in placing restrictions on the uses of fluorinated chemicals. Given the critical role of semiconductors in our economy and national security, it is important to avoid policies that unduly restrict current semiconductor operations or future innovation. At a minimum, broad restrictions on PFAS as a class should provide critical use exemptions for the uses of fluorinated chemicals in the semiconductor industry … and provide sufficient time for the industry and its suppliers to identify and qualify potential substitutes.”
To address these concerns, SEMI has formed a PFAS Working Group, which notes in it’s mission statement that “many jurisdictions are aiming to restrict the use of PFAS and to require the reporting of PFAS imports without a complete understanding of how dependent common technologies are on PFAS substances, how PFAS fill a very unique ( and currently essential) technical niche, and how pervasively PFAS are used in semiconductor production, semiconductor manufacturing equipment, the full semiconductor industry supply chain, and in technology in general.” In June 2023, The PFAS Working Group published an excellent “PFAS Explainer” series that will be updated frequently.
PFAS are also a focus at this year’s SEMICON West. At Wednesday’s Sustainability & EHS Forum, Kevin Fay, Executive Director of SPAN, said at the state level, the trend has been toward a risk-based approach that emphasizes targeted definitions, prioritization of high-risk compounds, and a uniform federal approach. At the federal level, the EPA is currently pursuing points laid out in the PFAS Strategic Roadmap, including a reporting program expected this year. Congress is also working on a solution with the recent draft of Senate EPW legislation and the formation of the PFAS Task Force.
The Semiconductor Industry Association (SIA) has formed the Semiconductor PFAS Consortium, which is producing a series of educational white papers. The first paper, “The Impact of a Potential PFAS Restriction on the Semiconductor Sector,” provides an overview of the uses of PFAS in various applications in the semiconductor industry and the timeline needed to develop, qualify, and implement alternatives. The report notes:
• Some materials are unique and have such specific technical requirements that it is incredibly challenging to find a viable alternative. No known alternatives exist for many of the industry’s uses of fluorocarbon-containing materials, and in many instances, a successful invention of an alternative substance needs to occur before subsequent steps of qualification can be undertaken.
• The organization of collaborative research and industry alignment is sometimes needed when making significant changes. For instance, the industry’s voluntary commitments to a perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) phase-out, and reductions in perfluorocarbon (PFC) emission were only effective due to the alignment and efforts of the industry as a whole.
• The timeline needed to develop, qualify, and implement alternatives falls into the following four broad categories:
- 3 to 4 years: If an existing non-PFAS alternative is available, does not require infrastructure alterations, and can be demonstrated to provide adequate performance for a specific application, then it typically takes 3 to 4 years to conduct the necessary manufacturing trials and implement the successful alternative into high volume manufacturing (HVM).
- From 3 to more than 10 years: In some applications, an existing non-PFAS alternative may be viable but requires tooling and/or process or facility changes before it can be successfully introduced into high-volume manufacturing. In these cases, it may take 3 to more than 10 years to introduce changes to the semiconductor manufacturing and related equipment (SMRE) and/or processes, perform qualification testing, and implement the non-PFAS alternative into HVM.
- From 5 to more than 25 years and successful invention is required: For some applications, it is not currently possible to demonstrate that a non-PFAS alternative can fulfill the application-specific performance requirements. In these cases, it may be necessary to invent and synthesize new chemicals and/or develop alternative approaches to device fabrication that provides the necessary electrical and computational performance.
- No alternative is achievable: In some cases, it may ultimately be found that a non-PFAS alternative is not capable of providing the required chemical function. If a non-PFAS alternative chemical cannot be invented, then the integrated circuit device structure may need to be abandoned in favor of an alternative device structure that may or may not provide equivalent performance. In some cases, the fundamental aspects of chemistry and physics prevent the use of PFAS-free alternatives.
“If you have to get rid of some PFAS, then you may end up having to move away from fluorine-based chemistry and if you’re moving away from fluorine-based chemistry, then you may have to move away from silicon-based chemistry,” Jones said. “It’s very difficult to say which of these is going to be possible, whether in the near term, medium term or long term and some of them may not be possible at all.”