PRESENTATION FOR THE SOCIETY OF MANUFACTURING ENGINEERS

U.S. ENVIRONMENTAL PROTECTION AGENCY

OFFICE OF TOXIC SUBSTANCES (TS-779)

This paper represents the work and views of the author. It does not necessarily represent the policy or position of the U.S. Environmental Protection Agency.

Today I would like to speak to you about issues concerning the usage and handling of radiation curable coatings. I am a Regulatory Impact Analyst with the Office of Toxic 8~ibstances at the U.S. Environmental Protection Agency (EPA). The Office administers the Toxic Substances Control Act, commonly referred to as TSCA. The Act was passed by Congress in 1976, and gives the Agency broad regulatory authority over new and existing chemicals. After giving a brief description of some of the major sections of TSCA₁ I will focus on recent OTS actions taken on chemicals designed for use in radiation-curable coatings. I will devote most of the time to those actions taken on new chemical substances under section 5 of the law. Specifically, I will be addressing the Agency's concerns over monomers and oligomers which contain the unreacted acrylate functionality.

The purpose of my talk is to enable you, as users of radiation-curable coatings, to understand EPA'S concerns and the reasons for taking certain regulatory actions.

The Toxic Substances Control Act was passed by Congress in 1976. The basic purpose of TSCA is to identify unreasonable risks to human health or the environment. Congress gave EPA broad authority under various sections of the Act to achieve this goal. The most important sections of the Act are sections 4, 5, 6, and 8.

Under section 4 of TSCA, the Agency can require manufacturers and processors of raw materials to test a particular chemical for health effects, such as teratogenicity (causes birth defects) or cancer.

Under section 5 of TSCA, companies are required to submit pre-manufacture notices (PMNs) for new chemicals to the EPA prior to the commencement of commercial manufacture; the Agency reviews the submissions for potential health and environmental effects.

Under section 6, the Agency can regulate existing chemical substances, i.e., those chemicals contained in the TSCA Inventory. Under section 6, the Agency has a host of options available to it, ranging from labeling requirements to an outright ban on the manufacture and use of specific chemicals.

Some of the more notable activities under Section 6 have concerned asbestos and polychlorinated biphenyls (PCBs). In the coatings area, OTS has investigated a group of four ethylene glycol ether solvents for possible regulatory action. These chemicals have been shown to cause birth defects, liver and kidney effects, and reproductive effects in laboratory animals. Section 8 of TSCA deals with data gathering authority. Under this section, EPA can collect data on production, use, and exposure to chemical substances. Under 8(b), OTS set up the TSCA Inventory, a compilation of approximately 60,000 chemicals which represents those chemicals that were commercially available between 1975 and 1978. The Inventory also includes those chemicals which have passed through the new chemicals review

process under section 5 since 1979. Also, there is a provision under section 8 that requires firma to submit test data to the Agency when such data shows a chemical to cause harmful effects (e.g. causes cancer in laboratory animals).

I would like to spend some time now explaining in some detail section 5 of TSCA, which gives the Agency authority to review and regulate new chemical substances. It is section 5 that has been used recently to regulate chemicals used in radiation-curable coatings.

Under section 5, the Agency reviews new chemical substances for health and environmental effects prior to the commencement of commercial manufacture. Chemical manufacturers are required to submit a pre-manufacture notice (PMN) at least 90 days prior to commencement of manufacture (or importation) of a new chemical substance. A new chemical substance is defined as one not contained on the TSCA Inventory, which I already described.

Between 1979 and July, 1986 the Agency had received over 7,000 PMNs. Notices of commencement of manufacture, indicating the start of commercial production, had been received for approximately 3,000 PMNs, or 44% of the total. PMN submissions cover the entire range of the chemical industry, including solvents, surfactants, resins, lubricants, dyes, photographic chemicals, and various performance additives. Applications range

from enhanced oil recovery to structural components for the space shuttle.

Under section 5 (as well as sections 4 and 6), decisions center around the concept of unreasonable risk. Congress, when enacting TSCA, did not define "unreasonable risk". Congress did, however, give some guidance as to what constitutes "unreasonable risk". Congress stated that the determination of unreasonable risk is a judgement which involves balancing the severity of harm and the probability that the harm will occur against the effects of a proposed regulatory action on the availability of the benefits of the chemical. This determination must also take into account the availability of substitutes for the chemical and other adverse effects the proposed regulatory action may have on society. Thus, section 5 involves balancing the risks and benefits associated with the introduction of a new chemical substance.

Making a determination of unreasonable risk involves consideration of various factors, including:

This work is carried out by teams of EPA staff, which include program managers, technical managers, engineers, chemists, toxicologists, economists/regulatory impact analysts, and other disciplines.

The Agency has many options available to it when reviewing PMNS. If there is no hazard and/or exposure identified, the chemical may be dropped, which means that the manufacturer is free to produce and sell the chemical. If the Agency needs more time to review the case (beyond the original 90 days) the review period can be extended for another 90 days If the Agency believes that there is enough data to characterize the hazard, the Agency may ban its use and production under section 5(f). This is a rare occurrence. If the Agency has no concerns for the intended use as specified by the manufacturer, but has concerns for another use which may result in a changed exposure scenario, it may promulgate a Significant New Use Rule (SNUR) outlining what uses would be considered new. For example, the use of an "industrial solvent" in consumer products might constitute a significant new use. Companies wanting to use the solvent in consumer products would be required to submit a SNUR notice to the Agency. A SNUR notice Would then be reviewed in a manner similar to that for a PMN.

Section 5(e) of TSCA is used to regulate new chemicals pending the development of data. Under 5(e) the Agency can limit production, distribution, and use of a new chemical in various ways. One approach is to issue a 5(e) order that specifies certain industrial hygiene practices, labeling requirements, marketing limitations, and methods of disposal. Another option is to set a production volume trigger, which specifies that once the aggregate production volume of a chemical reaches a certain threshold, the manufacturer must deliver the results of certain tests (i.e. health effects or environmental effects) to the Agency. Under the triggered testing approach, it is likely that engineering and/or personal protective equipment controls would be required pending results of the testing.

With this background information about TSCA in mind, let's turn now to the subject of radiation-curable coatings.

As I mentioned before, EPA has the authority, under section 5 of TSCA, to review new chemical substances prior to manufacture. In recent years, the Agency has received an increasing number of chemical substances intended to be used a" components of radiation-curable inks, coatings, or adhesives. We define radiation-curable coatings as those coatings designed tQ be cured (hardened) via exposure to UV-light or electron beam (EB) energy sources.

Radiation-curable coatings normally contain four distinct components: photoinitiators, monomers, oligomers, and additives. The photoinitiator is responsible for initiating the free radical polymerization of the coating. Benzophenone is a common initiator. Photoinitiators are not employed in electron beam-cured coatings₁ because the amount of energy generated by the electron beam is sufficient to initiate polymerization without the use of a photoinitiator.

Monomers are added to the coating to reduce viscosity, effect cross-linking, or impart other desirable properties such as hardness or flexibility. These monomers are normally mono- or multifunctional acrylate esters, such as hydroxyethyl acrylate or trimethyolpropane triacrylate. The mono-functional monomers primarily effect viscosity reduction, while the multifunctional monomers effect cross-linking.

The oligomers, or pre-polymers, serve as the backbone, defining hardness, solvent resistance, etc. Common oligomers are acrylated epoxies, polyurethanes, or polyesters. For example it is not uncommon to "cap" a polyurethane molecule with an acrylate ester such as hydroxyethyl acrylate. The acrylate ester is attached to the polyurethane because it provides a good cross-linking site.

Finally, additives include such Substances as pigments for coloring or surfactants (surface active agents) that affect flow properties.

The Agency, under section 8(e) of TSCA, has received the results of several animal studies on acrylates. These studies indicate that these chemicals cause cancer in laboratory animals. The mechanism by which these chemicals cause cancer is not known. One common feature of these chemicals is the presence of a highly reactive double bond. This toxicological information has led toxicologists to come to suspect any new acrylate or methacrylate-containing molecule as a potential carcinogen. This includes monomers and oligomers, which together comprise more than 90% by weight of any typical coating. At the present time, the Agency is most concerned about chemicals having a molecular weight below 1000, which is related to the potential for dermal (skin) absorption.

I should point out that solvent based acrylic coatings, which many of you may be familiar with, involve a different type of chemistry than radiation-curable coatings. In solvent based acrylic coatings, the double bond on the acrylates has been reacted away. In radiation-curing, the double bonds on the acrylates are meant to be available, because of the different curing process involved.

There are many people who argue that exposure to acrylates is self-regulating, because the chemicals are known. to be severe irritants. Thus, there is a strong incentive, to avoid exposure. At the same time, we have seen chemicals submitted as PMNs that are designed to be less irritating. When a chemical is non-, or less-irritating, it may be possible that people will not take appropriate safeguards to avoid exposure, because they think it's less toxic. There is, however, a difference between acute toxicity (e.g. irritation) and chronic toxicity (e.g. whether a chemical causes cancer). Coating users often say that solvents are chosen on the basis of smell, when the more pertinent question is the relative toxicity.

Radiation-curable coatings present one example of how EPA must balance the risks and the benefits associated with the introduction of new chemical substances, and involves the central concept of unreasonable risk. The Agency is generally aware of the benefits associated with the use of radiation-curable coatings, including energy savings, the reduction or elimination of solvent emissions, apace savings, and increased output. At the same time, there is a belief that uncontrolled use of these monomers and oligomers may present an unreasonable risk to human health. This finding is based on a comparison of new chemicals to known carcinogens, and exposure estimates arrived at by EPA engineers.

One may ask then why the Agency doesn't simply require manufacturers to test the new substances to see if they cause cancer or not. The problem is that most new chemicals do not have projected production volumes that would justify expenditures of up to ~l million for a 2-year animal feeding study and a three year delay in the introduction of the new chemical substance.. Short-term screening tests, costing less than $10,000, are not considered valid for this class of chemicals. Some work is currently being performed by RPA and industrial toxicologists aimed toward the development of a valid screening teat for this class of chemicals. The results of this work, however, may not be available for a few years.

The Office of Toxic Substances has not banned the use of any components of radiation-curable coatings. There is concern, however, over the unregulated use of potential carcinogens. Therefore, the Agency has issued a number of 5(e) orders. Generally, these orders require;

1) wearing of protective equipment - e.g., gloves, goggles, respirators.

2) recordkeeping - e.g., customers lists, names of employees exposed to chemicals.

3) limiting sales to processor/users until a significant new; use rule (SNUR) is developed.

4) attaching warning labels to products stating that the materials are potential carcinogens, Letters to customers, and material safety data sheets (MSDs') are also required.

5) Specifying disposal methods in order to avoid potential exposure of the general public to the chemicals through drinking water.

From a business standpoint, there are problems with these orders, especially with respect to the marketing limitations and labeling requirements. Let me first talk about ma limitations. In practical terms, limiting sales to processors/users means that the PNN submitter can only sell to companies who both formulate end cure the coating. This means in most cases that the PMN submitter cannot sell to formulators until a significant new use rule (SNUR) is promulgated. Some Submitters claim that formulators represent up to 90~ of their sales; and that the wait (18 months - 2 years) for a SNUR is too long. One of the arguments here is that the PMN submitter is unable to realize the full return on the investment in research and development in a timely manner. Once the SNUR is promulgated, anyone can produce the chemical as long as they comply with the provisions of the SNUR. By the time the SNUR is issued, the original submitter may have lost valuable lead time, other firm's research efforts may have caught up to the submitter's in the interval between the PMN submission and the SNUR. The promulgation of a SNUR takes longer than a 5(e) because a SNUR involves formal rule-making, while the 5 (e) order is an administrative order. Formal rule-making includes review by the rest of the Agency and the Office of Management and Budget (OMB); the final SNUR must be signed by the Administrator of the EPA. The issuance of a 5(e) Order involves a more limited review within the Agency. The Assistant Administrator for Pesticides and Toxic Substances has the authority to sign the 5(e) Order on behalf of the Agency.

There are legal and administrative aspects of the problem. In essence, the 5(e) order only applies to manufacturers end their direct customers. It cannot, generally, be extended to third-tier parties - i.e. customers of customers. SNURs are written because they apply to all parties who manufacture, process, or use the chemical subject to the SNUR. Recent $NU~ have been drawn along the lines of the 5(e) Order: thus new uses would be uses other than in industrial inks, coatings, or adhesives; use without protective equipment; release to water, etc. Firms wanting to use the chemical for a new use have to submit SNUR notices, which are reviewed like PMNs.

In an attempt to shorten these time delays, a few alternatives to the present process have been considered and pursued during the past two years. These alternatives include multi-party 5(e) orders, a listing SNUR for those chemicals awaiting the development of SNURs, and a generic SNUR for future acrylate PMNs. In multi-party 5(e) orders, manufacturers, formulators, distributors, and some users all sign the 5(e) order. In effect, the various parties constitute co-submitters of the PMN substance. The practical result of this action is to enable the manufacturer to sell the new chemical substance to a wider customer base than under the normal 5(e) orders.

The listing SNUR approach represents an attempt on the part of EPA to group together chemicals that have been subject to 5(e) orders, but have not been subject to SNURs. Under normal circumstances, SNURs are developed on a chemical by chemical basis. The listing SNUR lists various standard requirements for chemicals (e.g., labeling, personal protective equipment); as new chemicals are added to the list, they would be subject to specific requirements. The effect of the listing SNUR would be to cut down on the time between the issuance of the 5(e) order and the promulgation of the SNUR. The listing SNUR approach was published in the Federal Register in April, 198a. The EPA is currently reviewing public comments.

The generic SNUR for acrylates and methacrylates would apply to PMN chemicals submitted subsequent to Promulgation of the rule. Under the generic SNUR, Which is currently being reviewed within EPA, chemicals meeting certain criteria (e.g. molecular weight below 1000 and containing the unreacted acrylate/methacrylate moiety) would automatically become subject to the SNUR. The provisions of the SNUR may be the same as those contained in the 5(e) orders (i.e., labeling, protective equipment, recordkeeping). The practical effect of the generic SNUR might be to obviate the need for 5(e) orders and individual SNURs, thus facilitating the introduction of new chemicals into the marketplace.

The labeling requirements and warning letters to customers are also bothersome to some submitters. The complaints center around the argument that many existing chemicals that are just as dangerous as the new chemicals are not subject to the same requirements. Thus, new chemicals are made to appear more dangerous than existing chemicals, because the existing chemicals are not labeled as potential carcinogens. The Agency is also concerned with this issue. It is pert of a much larger issue of new vs. existing chemicals regulation. The Agency is currently investigating various strategies to deal with whole categories of chemicals, including acrylates and ethylene glycol ethers. Under such an approach, the EPA might employ sections 4,5,6 and 8 to deal with large numbers of new and existing chemicals belonging to the same chemical family.

As far as acrylates are concerned, EPA is not contenting itself with only pursuing regulations. Strategies are being developed that will pursue fundamental toxicological questions related to the carcinogenic potential of this class of chemicals.

EPA has been engaged in a dialogue with industry in a few areas. The Basic Acrylic Monomers Manufacturers (BAMM), made up of Dow, Badische, Celanese, Union Carbide and Rohm & Haas, has been formed to deal with various issues related to acrylates. The group has met with EPA on several occasions to discuss toxicological and commercial issues.

In 1985, the RADCURE division of AFP/SME came and presented a half-day seminar on radiation curing technology to about 75 EPA staff engineers, economists, chemists, and program managers.

The EPA is also considering the funding of a study of the potential impact of radiation curing on the furniture industry. The study, which would take about two years to complete, would deal with such issues as solvent reduction, process economics, and finishing properties (compared to conventional solvent-based coatings). This kind of study could help EPA to quantify the impact of radiation curing on an important segment of the coating industry. The EPA would welcome data from industry quantifying the benefits of radiation curing as well as examples of specific solvent-based systems replaced by radiation-curable formulations. 

From a user's standpoint, what does all this mean? I think it means that faced with complying with the provision of a 5(e) Order or a SNUR, it Should not be automatically assumed that It would be safer to employ an existing chemical not subject to these provisions. Until more is known about the mechanism of cancer, care should be taken to avoid exposure to radiation-curable coatings. Work practices should be designed to avoid contact if possible. If not, protective equipment such as gloves, goggles, and respirators should be routinely employed. As customers of formulators and/or manufacturers, you have the right to ask your supplier for guidance as ti the proper equipment to use. In addition, the suppliers of protective equipment can provide you with information. Coating lines should be designed to eliminate or severely reduce potential exposure to mists. Any unreacted materials should be disposed of properly (e.g. incineration) to protect the general public from exposure to these chemicals through their drinking water. After a1l, the real point here is to protect workers and the public against exposure to potential carcinogen, not to sidestep problems by denying they exist because there isn't a specific regulation.