Perfluorooctane sulfonic acid, its salts and perfluorooctane sulfonyl fluoride

While alternatives to PFOS are available for some applications, this is not always the case in developing countries where existing alternatives may need to be phased in. For some applications like photo imaging, semi-conductor or aviation hydraulic fluids, technically feasible alternatives to PFOS are not available to date.

Note that following information is extracted from the risk management evaluation document(UNEP-POPS-POPRC.4-15-Add.6)

A. Uses for which at present, according to responses received, no technically feasible alternatives are available

In addition to the below, China noted that there is continuing use of PFOS-related substances in the petroleum industry and in nano-material processing for which at present they do not have alternatives available.

Photo imaging

According to the photographic industry, chemicals or classes of chemicals that may be considered alternatives to PFOS or PFOS-related substances on an industry-wide basis (or even a company-wide basis) are reported as not currently being available. Successful alternatives to PFOS materials have included non-perfluorinated chemicals such as hydrocarbon surfactants, chemicals with short perfluorinated chains (C3 - C4), silicones, telomers. In very few cases, it has been possible to reformulate coatings so that they are inherently less sensitive to static build-up.

According to the industry, the imaging products/applications where there are currently no identified alternatives to PFOS-related substances and which represent critical uses are as follows:

  • surfactants for mixtures used in coatings applied to films, papers, and printing plates; PFOS chemicals are critical for creating coatings of high complexity in a highly consistent manner, thus avoiding the creation of large amounts of waste due to irregularities in coating thickness;
  • electrostatic charge control agents for mixtures used in coatings applied to films, papers, and printing plates. The antistatic properties of PFOS materials also provide important safety features by controlling the build-up and discharge of static electricity, thus preventing injuries to employees and users, damage to operating equipment and products, and fire and explosion hazards;
  • friction control and dirt repellent agents for mixtures used in coatings applied to films, papers, and printing plates; and

adhesion control agents for mixtures used in coatings. Adhesion control is a property imparted to film coatings as a result of the use of PFOS materials as coating aids.

Estimates of releases from the photo imaging industry, as developed by UK DEFRA, are 1.02 kilogram into waste water and 0.051 into air from manufacturing uses in the EU. The industry estimates a total of less than 2 kilogram worldwide, by extrapolation.

Most consumer and professional imaging papers do not contain PFOS-related substances. For papers that do contain the substances, the coatings contain concentrations in the range of 0.1-0.8 µg/cm2. Most of this material will not be on the surface of the coating as the PFOS-related substance is contained within a matrix and is bound to coating matrices.

The cost, so far, for replacement of PFOS materials is estimated to be in the range of € 20-40 M for the full range of imaging products. These costs are based on the estimated cost of achieving the current reduction of 83% in the use of PFOS-related substances. The cost to be incurred from further work on replacements (for the remaining 17%) is expected to be significantly higher than the above figure as the replacement work is increasingly more difficult.

Based on the previous cost estimates of US$20-40 M for reduction that took place between 2000 and 2004, i.e., a reduction of roughly 15 tonnes, the average cost is US$2 M per tonne. Further reductions are estimated to cost more than twice as much, up to US$5 M per tonne. The cost of substituting the remaining 10 tonnes would be US$50 M. Since only 2 kilogram is estimated to be released into the environment the cost of reducing the release to zero, using these estimates would be US$25 M per kilogram. This calculation indicates the level of magnitude of the costs of reducing the release.

Photoresist and Semi-conductor

According to the semi-conductor industry, the operation of PFOS based photo acid generators (PAGs) is critical to the semiconductor industry in the photolithography process. ESIA, JSIA, SIA and SEMI indicate that there are currently no substitutes known that give the same level of critical functionality to cause effective, efficient transformation in leading edge photoresists and which can be used in volume manufacturing.

For anti-reflective coatings used in combination with photoresists, ESIA indicates that there is also no alternative available that fulfils the specific technical requirements necessary (ESIA, 2003). The industry is also evaluating one additional specialized application for which PFOS use may have no current substitute -- use in liquid etchant in the photo mask rendering process.

The semiconductor industry indicates that the industry and its suppliers continue to search for alternatives for these critical uses. The nature of semiconductor production is such that if alternatives to PFOS are eventually identified at the fundamental research stage, critical adjustment to the chemistry of inputs such as PFOS use in the photolithography process will trigger far-reaching adjustments throughout the manufacturing process and supply chain to ensure that the chemical processes throughout the production process remain aligned. Thus, the semiconductor industry believes it could take an additional ten years to design, operationalize and integrate the new technology, once it has been identified, into the semiconductor manufacturing process. According to the industry, the delay is a necessary function of the semiconductor technology development cycle: technological innovations generally require 10 years of further development before they can be reflected in high volume manufacturing (ESIA, JSIA, SIA, SEMI 2007).

It should also be noted that during the chemical formulation of photolithography products, worker exposure potential is very low because the process occurs under highly automated, largely closed system conditions. The same process for electronics fabrication is similarly automated, with a low volume of PFOS used, and use of protective equipment.

Chemical isolation is also an intrinsic part of quality control procedures.

Environmental release potentials are deemed to be low. Due to the low vapour pressure of PFOS, and the nature of the process, no emissions to air are expected. However, waste products, including 93% of the resist formulation (PAGs and surfactants) are incinerated. Releases to water are also considered to be negligible. Furthermore, there is no residual PFOS compound present in manufactured microprocessors and therefore no consumer exposure or concern about releases from electronic waste disposal or recycling.

PFOS releases from photolithography uses are small compared with PFOS use in other industry sectors. In 2002 for the whole of Europe, an estimated 43 kilogram of PFOS were released in the effluent from photolithography uses, on the order of only 0.45 percent of all PFOS releases at that time in Europe. Mass balance data for Europe in 2004 indicates an estimated 54 kilogram of these releases. It has been estimated that a similarly small proportion of releases in the United States and Japan is attributable to the photolithography uses, based on recent past use patterns.

It is difficult to quantify the costs that will ultimately be involved in replacing PFOS use in the photolithography industry with alternative substances, given that such alternatives are not currently available. The requirements for innovation and the limits of technical feasibility are the main factors that currently limit access to alternatives. If those hurdles can eventually be overcome, however, there will be substantial costs associated with the transition to the use of alternative substances in the photolithography process. For example, there are likely to be extensive introduction costs associated with bringing a new system into high volume production, including re-qualification costs and possible loss of revenues associated with much lower yield as new systems are brought on line. Many resists are specifically tailored to one individual company’s process, which means that a valid replacement for one cannot necessarily be applied industry-wide. Given those uncertainties, the estimate below, derived for this evaluation, is only an indication of the order of magnitude of the costs involved.

Replacing existing resists systems would require extensive R&D followed by a time-consuming manufacturing process re-qualification. The development cost of one completely new photoresists system for the industry has been estimated at US$192M for 193nm resist, US$287M for 157nm, and US$218M for EUV resist. The cost for 157nm resist development is the highest, because it has more novel requirements than either 193nm or EUV resists.

Development costs of a new photoresist system would be US$700M. Assuming that variable costs are the same as in the present system, it takes 5 years to develop the new system and the time span for the analysis is 25 years. This would imply that the reduction in release of PFOS related substances is equal to 20 years of releases (50 kilogram per year), i.e. a total of 1000 kilogram. Costs would be US$0.7M per kilogram PFOS. This calculation indicates the level of magnitude of the costs of reducing the release. By comparison, the semiconductor industry had global sales of $248 billion in 2006.

The semiconductor industry recently signed an agreement to curtail the use of PFOS-based chemicals at the global level. Under the agreement, members of the World Semiconductor Council, which comprises the trade associations representing the microchip industries of most of the world’s leading semiconductor-producing countries (including SIA, ESIA and trade associations in Asia), and SEMI have committed to the following actions: (i) ending non-critical uses for PFOS by specific dates; (ii) working to identify substitutes for PFOS in critical uses for which no other materials are presently available; (iii) destroying solvent wastes from critical uses; and (iv) taking other steps to mitigate the potential environmental impacts of PFOS use in these critical applications.

Photo masks in the Semiconductor and Liquid Crystal Display (LCD) Industries

Photo masks are an essential part of the photolithography process of semiconductor and LCD production. Photo mask production is mainly outsourced from semiconductor or LCD producers to other companies.

Three major photo mask producers in Japan report that a wet process is used in the production of most photo masks. PFOS and PFOS-related substances are contained in etchants for semiconductor and Thin Film Transistor (TFT) panels, because these products require very fine patterning. In the case of photo masks for semiconductors, a dry process is also used for some specific cases. All TFT photo masks are produced using a wet process because of their large size.

The total amount of PFOS (including PFOS moiety in PFOS-related substance) use for this purpose in Japan is estimated at approximately 70 kilogram per year. It is estimated that Japanese companies play a major role in photo mask production, and have more than a 70% share of the worldwide market. Thus it is estimated that total use of PFOS and PFOS-related substances for this use in the world is approximately 100 kilogram.

Because of strong acid of etchants, non-fluoro surfactant is not stable in etchants, thus it is not applicable for this process. Furthermore, other fluoro-surfactants such as shorter chain PFAS are not suitable because their ability to lower surface tension is not sufficient.

A dry etching process is applied to high-end ultra-fine patterns of semiconductor photo masks. However, the yield and productivity of the dry etching process is much (15 to 20 times) lower than the wet process. Furthermore, the dry process is not useable for LCD panels because of their large size (more than 1m by 1m). 

Aviation hydraulic fluids

According to information received from one of the major producers of hydraulic fluids, there are no alternatives to the PFOS substances currently being used in aircraft systems and there is no known alternative chemistry that will provide adequate protection to aircraft.

The process of qualifying a new fluid for use in commercial aircraft has historically taken about 10 years from concept to actual commercial manufacture. There are no current alternatives to PFOS substances currently being used in aircraft systems and there is no information on costs or environmental/human health attributes of alternatives.

Certain Medical devices

The medical device industry has been using many raw materials containing PFOS for a very long time. For example, PFOS is used as an effective dispersant when contrast agents are incorporated into an ethylene tetrafluoroethylene copolymer (ETFE) layer. PFOS plays an essential role in radio-opaque ETFE production, allowing the achievement of the levels of accuracy and precision required in medical devices (e.g., radioopaque catheters, such as catheters for angiography and indwelling needle catheters).

Since about 2000, when the effects of PFOS on the environment were identified as a problem, radio-opaque ETFE manufacturers have been working with chemical material suppliers to find alternatives.

The 2006 OECD survey identified use of perfluorobutane sulfonate (PFBS) as surfactant in coating products. In some cases, this substance can be used as a dispersant for inorganic contrast agent when it is mixed into ETFE. For many other medical devices, alternatives that allow the achievement of the same standard, however, remain to be found. It is expected that, due to its unique properties, PFOS will continue to be used for a variety of medical devices.

B. Uses for which alternative substances or technologies may be available but would need to be phased in.

Metal plating

PFOS-related substances are used in the following main applications:

  • decorative chromium plating; and
  • hard chromium plating.

Other important uses are: pre-treatment agent for plastic plating, PTFE powder plating treatment agent, pre-treatment agent for printed circuit board plating; chromic acid anodizing; nickel cadmium, or lead plating; alkaline zinc plating; stainless steel electric grinding agent; and chemical abrasive agent for copper alloy.

The 2006 OECD survey identified use of perfluorobutane sulfonate (C4 PFAS) as mist suppressants. Other information indicates that there are currently no known effective alternative chemical mist suppressants to PFOS-related substance for these applications (Japan, 2007; US, 2007).

However, information received from a number of industry and regulatory authorities indicates that the substitution of Cr (VI) or hexavalent chromium with the less hazardous Cr (III) in decorative plating applications would eliminate the need for the use of PFOS-related substances in this application. Such substitution has potentially significant cost savings and health and safety and environmental benefits for the metal plating sector.

The higher costs of using Cr (III) are more than offset by the savings from reduced waste treatment costs, reduced air monitoring costs, record keeping, and the reduced reject rate. The major benefit, however, relates to the significantly reduced risk of employee ill health induced by working with hexavalent chromium. The progress of substitution is different due to the quality requirements of the different markets e.g. in Japan only 40-50 of about 1000 companies have changed their process. In such cases, PFOS mist control agents are still needed to protect workers’ health

For hard chromium plating, information received indicates that the direct substitution of Cr (VI) with Cr (III) is not currently a viable option. While industry has indicated that substitute processes for Cr (VI) hard chrome plating have been developed for certain small applications, currently there are no technologies that are available on a large, commercial scale to replace the majority of Cr (VI) plating applications. In Japan alternatives for uses other than hard chromium plating are not yet identified partially because of the requirements for high reliability e.g. for automobile pumping parts.

The cost of improved ventilation with extraction, which is the recommended substitute for PFOS-based mist suppressants, has been calculated to be €3400 per year in each production unit where the investment period is 15 years (RPA 2004). Assuming a few hundred units in the EU the total cost would be one or two million euros. In Japan it has been estimated that the cost would be US$40 000 for each 1000 litre bath (Japan, 2007).

The anticipated costs of the proposed Canadian regulations (See Section 1.5) by firm size are US$0.65 M for 34 small firms, US$2.6 M for 52 medium firms and US$0.68 M for 14 large firms. The total estimated compliance costs for Canadian facilities using PFOS fume suppressants to comply with the proposed regulations is approximately US$3.9 M (discounted at 5.5% over 25 years). This would result in a reduction in PFOS releases of approximately 86 tonnes over the 2013 to 2032 period (Canada, 2006)). Based on these Canadian calculations, the cost of reduction is US$46 per kilogram of PFOS reduced.

Fire fighting foam

A number of alternatives to the use of PFOS-based fluorosurfactants in fire fighting foams are now available/under development. These alternatives include: non-PFOS-based fluoro-surfactants; silicone based surfactants; hydrocarbon based surfactants; fluorine-free fire fighting foams; and other developing fire fighting foam technologies that avoid the use of fluorine. The efficacy of alternatives would need to be considered.

Fluorine-free foams are approximately 5-10% more expensive than the fluorosurfactant-based foams (including those PFOS-based foams marketed previously). The manufacturers, however, indicate that prices for fluorine-free foams would reduce if the market size increased. It is, therefore, assumed that prices are broadly comparable.

As the transition from PFOS-based products has already taken place for most uses in many countries, there are only limited developmental or operational costs associated with the substitution of PFOS-based foams by foam manufacturers or users. The main costs for phasing out PFOS-based foams are related to managing stockpiles and waste containing such foams.

With regard to the toxicological and ecotoxicological suitability of non-PFOS based fluorosurfactants, data are limited. Whether telomers represent a significant concern for human health and the environment is under review elsewhere and conclusions are awaited.

With regard to fluorine-free foams, current information indicates that compared to PFOS based foams, they do not persist or bio-accumulate in the environment (due to the absence of fluorine). With regard to acute toxicity, fluorine- free foams appear to have a slightly lower acute toxicity, although the information provided to date is not conclusive.

For Canada, it is estimated that the proposed regulations would reduce the release of PFOS based AFFF into the environment in the order of 2.83 tonnes over the 2008 to 2032 period. The present value of the disposal and replacement costs experienced by airports, military facilities and refineries would be in the order of approximately US$0.64 M (in 2006 $) discounted at 5.5% over the 25-year time period (Canada, 2006). Based on these Canadian calculations the cost of reduction is US$226 per kilogram of PFOS reduced.

For the EU, costs of replacement and destruction of foam have been estimated at €6000 per tonne. The stocks in the EU are 122 tonnes (RPA 2004). Based on the RPA's calculations, the cost of reduction is €6 per kilogram of PFOS reduced. Once the foam has been renewed, the cost of destruction may be as low as €1 per kilogram.

In Japan, it has been estimated that 86 tonnes of PFOS equivalent exist in AFFF products on the market. Based on this information, the estimated total amount of PFOS in the market is less than 200 tonnes in fire fighting foam concentrate. The market has stockpiled some 21,000 tons of PFOS fire fighting foam concentrate, and some 11,400 tons of fire fighting foam contains PFOS itself, the rest 9,600 tones contains PFOS derivatives. The majority of market stock is fire fighting foam for water-immiscible liquids such as oil, naphtha and hydrocarbon-fuels, and non-PFOS alternatives are already marketed for this use. It is estimated that replacement will take about15 years based on present production capacity. On the other hand, some 2,000 tones of market stock, fire fighting foam for water miscible liquid such as alcohols glycols and acetone is more indispensable before long for biological fuels (bio-ethanol etc.). The foam for water miscible liquid is required to fulfil government standards and a non-PFOS alternative is not yet developed due to technical difficulties and technical feasibility. It is estimated that the alternative development will take several years and that replacement will take also about 15 years. Furthermore, fire fighting foam containing PFOS is also stored at airports (Japan, 2007).

The SNUR regulations in the U.S. restrict only new manufacture or importation of PFOS chemicals and PFOS- containing products. The U.S. regulations do not impose any restriction on the use of existing stocks of PFOS-based AFFF manufactured or imported into the U.S. prior to the effective date of the regulations, and no mandatory phase out of those existing stocks is either in place or contemplated.

Electric and electronic parts

PFOS is widely used in the production of electric and electronic parts. Major uses are as sealing agents and adhesives. For these uses, alternatives are available or are under development, and PFOS will be replaced relatively quickly.However, several uses have been identified for which alternatives will not soon be available. One such use is in the intermediate transfer belt of colour copiers/multi-function printers.

The intermediate transfer belt is an essential part of colour printers and colour copying machines. According to information provided to the Japanese government, the largest manufacturer (which supplies more than 60% of polyimide intermediate transfer belts) uses PFOS to ensure the required properties. Intermediate transfer belts of this manufacturer contain up to 100 ppm of PFOS. The part is used by 12 colour copier/multi-function printer manufacturers that dominate the global market; it is also supplied worldwide as a spare part. The properties of the intermediate transfer belt determine the design of the copier/multi–function printer. Due to the long life of copiers/multi-function printers, if supplies of this part are stopped, millions of copiers/multi-function printers might be discarded before the end of their product life, leading to unnecessary potential environmental damage.

Similar to intermediate transfer belts, PFA rollers and belts in fixing units contain PFOS for the same reason. The largest manufacturer of these units has reported that PFOS in the amount of 8×10-4 ppm is contained in an additive used in producing the units, and that additive is used in the amount of 3μg/cm2. The production volume is 300,000 units per month, and annual consumption of PFOS is less than 3 g.

Furthermore, PFOS is used in various kinds of additives, such as grease additives for mechanical slides and micrometers, as a component of an etchant used in the plating process to produce electronic devices, as well as a wider range of other uses in electric and electronic industries. However, due to the very low concentrations involved, as mentioned above, and the complex supply chain, use in this area was only recognized very recently, and thus further study is needed.

It is not clear what the impact of using alternatives to PFOS would be with regards to product performance.

Use of PFOS derivative in production of ant baits for control of leaf-cutting ants

Sulfluramid (1-octanesulphonamide-N-ethyl-1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluoro; CAS: 4151-50-2), is manufactured using a PFOS-related derivative (perfluorooctyl sulfonyl fluoride (PFOSF), CAS No 307-35-7).

Sulfluramid is the active ingredient in the manufacture of ant baits in ready-to-use formulations and is known to degrade to PFOS. It is estimated that the production of sulfluramid in Brazil is about 30 tonnes per annum. Sulfluramid is used at a 0.3% concentration, resulting in a production of around 10,000 tonnes of ant baits/year. In 2006, around 400 tonnes of ant baits (sulfluramid 0.3%) were exported to 13 countries in South and Central America. Sulfluramid cannot be manufactured without the use of PFOS-related derivatives. Since 10 % of sulfluramid is degraded to PFOS its use represents a direct release of PFOS to the environment.

Several mechanical, cultural, biological and chemical methods, including different formulations, have been studied for controlling leaf-cutting ants. Granulated baits represent the most widely used method for leaf-cutting ant control, consisting of a mixture of an attractant (usually orange pulp and vegetable oil) and an active ingredient (insecticide), presented in the form of pellets. This method features some significant advantages over other methods. It is a low-cost method, delivering high efficiency with reduced health hazards to humans and the environment during application and being specific to the pest target. Its formulation is developed with low concentrations of active ingredients, and its localized application does not require application equipment. The utilization of ready-to-use formulations should reduce or impede releases to humans but the release of 30 tonnes of sulfluramid annually to the environment will eventually result in a significant part of it being degraded to PFOS.

Currently, the active ingredients used in ant baits are: sulfluramid, fipronil and chlorpyrifos. Fipronil and chlorpyrifos are more acutely toxic to mammals, water organisms, fish and bees than sulfluramid. Comparative studies demonstrate low efficiency of ant baits with chlorpyrifos and fipronil. According to the Brazilian Annex F information, sulfluramid cannot presently be efficiently replaced in Brazil by other registered products commercialized for the same purpose. In the EU, PFOS-related substances are not used in the manufacture of pesticides (RPA 2004). Ant baits containingS-methoprene and pyriproxifen are registered in New Zealand for the control of exotic ants by aerial and ground applications (ERMANZ, 2007).

Limited information provided on associated worker exposure to PFOS from the manufacture of sulfuramid baits indicates low exposure to the workers. There was no information on the exposure of the local community and environment from the use of sulfluramid baits.

A. Uses for which alternatives are available in developed countries

For the following uses which have been used historically in the US, Canada and the EU, alternatives are available and in use: fire fighting foams; carpets; leather/apparel; textiles/upholstery; paper and packaging; coatings and coating additives; industrial and household cleaning products; and pesticides and insecticides.

It is presently unclear whether these uses of PFOS-related substances still occur in some parts of the world. However, in China PFOS and/or PFOS-related substances are still used in clothing manufacturing and for surface coating.

For further information, please refer to 

  • UNEP/POPS/POPRC.5/10/Add.1 – General guidance on considerations related to alternatives and substitutes for listed persistent organic pollutants and candidate chemicals
  • UNEP-POPS-POPRC.8-INF-17-Rev.1 - Technical paper on the identification and assessment of alternatives to the use of perfluorooctane sulfonic acid in open applications
  • UNEP-POPS-POPRC.12-INF-15-Rev.1 - Consolidated guidance on alternatives to perfluorooctane sulfonic acid and its related chemicals
  • Risk profile ArChEnFrRuSp (PDF)
  • Risk management evaluation (RME) ArChEnFrRuSp (PDF), addendum to the RME ArChEnFrRuSp (PDF)