The Mexichem Conjecture: Regulating HFCs under CAA ... or TSCA?

Since 1994, EPA has regulated hydrofluorocarbons (HFCs), which are potent greenhouse gases, through several programs under Title VI of the Clean Air Act (CAA) governing stratospheric ozone protection. In a recent decision, Mexichem Fluor, Inc. v. EPA, the U.S. Court of Appeals for the D.C. Circuit (D.C. Circuit) vacated and remanded in part a 2015 EPA rule regulating the use of certain HFCs in particular end uses. Two petitions for rehearing and rehearing en banc are now pending in the case, and eleven states and a group of administrative law professors recently filed amicus briefs urging the court to grant rehearing. Assuming, however, that the opinion remains unchanged, this decision significantly constrains EPA’s ability to regulate HFCs through this type of rulemaking and narrows EPA’s options for controlling HFCs under its existing CAA authority. This article provides an overview of the decision and discusses a potential alternative pathway identified by the D.C. Circuit for EPA to regulate HFCs: through the Toxic Substances Control Act (TSCA). It appears that TSCA is a limited tool for this purpose.


One of the programs EPA has used to regulate HFCs is the Significant New Alternatives Policy (SNAP) program under Section 612 of the CAA, which requires EPA to evaluate substitutes for ozone-depleting substances (ODS) and replace such substances “to the maximum extent practicable” with “chemicals, product substitutes, or alternative manufacturing processes that reduce overall risks to human health and the environment.” EPA reviews substitutes within a comparative risk framework, evaluating the human health and environmental impacts—including the global warming potential—of the substitute against the current substance in use. To implement the replacement of ODSs with “safer” substitutes, EPA is required under CAA Section 612(c) to publish a list of both approved and prohibited alternatives. Manufacturers may not replace an ODS with a prohibited substitute substance.

The 2015 SNAP Rule

In the 2015 SNAP rule at issue in Mexichem (the Rule), EPA changed the status of certain HFCs or HFC blends from “acceptable” to “unacceptable,” “acceptable, subject to use conditions,” or “acceptable, subject to narrowed use limits." Thus, the Rule resulted in a ban on certain HFCs or HFC blends in various end-uses (e.g., the use of HFC-134a in new light-duty motor vehicle air conditioning systems beginning in Model Year 2021) in the aerosols, foam blowing, refrigeration, and air conditioning sectors. 

EPA stated in the Rule that the SNAP program:

[D]oes not provide a static list of alternatives but instead evolves the list as the EPA makes decisions informed by our overall understanding of the environmental and human health impacts as well as our current knowledge about available substitutes. In the more than twenty years since the initial SNAP rule was promulgated, EPA has modified the SNAP lists many times, most often by expanding the list of acceptable substitutes, but in some cases by prohibiting the use of substitutes previously listed as acceptable. 

EPA went on to note that, as the lists have expanded, it has not performed a systematic review of chemicals previously approved to determine if less environmentally adverse alternatives have since become available, but that EPA was beginning this process with the Rule. As directed by the Obama Administration’s 2013 Climate Action Plan, EPA explained that in “our first effort to take a broader look at the SNAP lists, we have focused on those listed substitutes that have a high global warming potential relative to other alternatives in specific end-uses” as part of a full comparative risk analysis performed on those substances. Under the Rule, certain HFCs that had been previously approved as substitutes for ODSs would now be prohibited or restricted in specified end-uses, as EPA determined that lower global warming potential alternatives were available that posed less environmental impact.

Court Opinion

In a two-to-one opinion, the D.C. Circuit vacated the Rule “to the extent the Rule requires manufacturers to replace HFCs with a substitute substance.” It remanded to EPA the issue of whether the Agency “has the authority to conclude that a manufacturer’s past decision to replace an ozone-depleting substance with HFCs is no longer lawful.” The majority’s opinion turned on the concept of “replacement,” holding that once a manufacturer had initially replaced an ODS with an alternative, EPA could not subsequently order the replacement of a non-ODS alternative with a substitute that contributed less to climate change. For example, under the majority’s reasoning, if EPA had approved at some point in the past an HFC with high global warming potential as a “safer” alternative to an ozone-depleting substance, like a chlorofluorocarbon (CFC) or an hydrochlorofluorocarbon (HCFC), EPA could not subsequently prohibit the use of that HFC, even if a lower global warming potential substitute had been developed that posed less risk to the environment, because the original ozone-depleting substance had already been “replaced.” The majority held that CAA Section 612 “and other statutes” do not give EPA authority to order the replacement of substances that are not ozone-depleting, like HFCs, due to their impact on climate.

The dissent strongly departed from the majority’s interpretation of the word “replacement,” arguing that replacing an ODS is not necessarily a one-time event, and that such an interpretation “makes a mockery of the statutory purpose, because a product manufacturer could ‘replace’ a class I substance with a substitute before the EPA has a chance to evaluate it completely, and if the agency later determines that a different substitute ‘reduce[s] overall risks to human health and the environment’ ..., the agency would be powerless to tell that product manufacturer that it could no longer use the more risky substitute.” Such an interpretation, the dissent concluded, “undermines Congress’s intent to ‘reduce overall risks to human health and the environment’” to the maximum extent practicable, as required by the CAA. 

Notwithstanding its conclusion flowing from its interpretation of “replacement,” the majority found that EPA still has “several statutory authorities” to regulate HFCs. The majority recognized EPA’s authority to move HFCs from the safe list to the prohibited list of substances, assuming all statutory criteria are satisfied and found that EPA had not acted in an arbitrary or capricious manner in changing the listing of the HFCs covered by the Rule from “acceptable” to “unacceptable.” The majority further held that EPA may continue to prohibit any manufacturer still using an ODS from replacing it with an HFC because the manufacturer has yet to “replace” the ozone-depleting substance with a substitute. 

The majority went on to note that “EPA possesses other statutory authorities, including the Toxic Substances Control Act, to directly regulate non-ozone-depleting substances that are causing harm to the environment.” Their opinion also listed several other programs under different parts of the CAA as authorities that EPA could potentially use to regulate HFCs, including the National Ambient Air Quality Standards program, the Hazardous Air Pollutants program, the Prevention of Significant Deterioration program, and CAA Section 202, which governs emission standards for new motor vehicles or new motor vehicle engines. “Our decision today,” the Court wrote, “does not in any way cabin those expansive EPA authorities.”

However, none of these programs under the CAA appears well suited for HFC controls, either because of the structure of the program (i.e., it focuses on stationary sources and may not be readily adaptable to substances used in products), or because of the nature of the substances at issue (i.e., HFCs are not toxic air pollutants under the CAA). Accordingly, it may be useful to look at the extent to which, as a practical matter, TSCA could serve as another source of statutory authority for EPA to regulate HFCs, as suggested by the majority.

Regulatory Options for HFCs under TSCA

Threshold Questions

A threshold question is whether potential regulation of HFCs under TSCA for their global warming potential would be precluded by HFC regulations under the CAA. Section 9(b) of TSCA directs EPA to use other statutes that it administers, such as the CAA, rather than TSCA if those other statutes could eliminate or sufficiently reduce to a sufficient extent a risk to health or the environment associated with a chemical substance or mixture, such as HFCs, unless EPA determines in its discretion that it is in the public interest to use TSCA to protect against that risk. The D.C. Circuit’s decision arguably could contribute to a determination either that the CAA cannot sufficiently reduce the global warming potential risk of HFCs, or that use of TSCA is in the public interest. If EPA were to make that determination, it would have to consider “all relevant aspects of the risk” and compare the estimated costs and efficiencies of action under TSCA as compared with an action under another EPA-administered statute.

A second threshold issue is whether EPA has authority to regulate the global warming potential of HFCs under TSCA. Various provisions of TSCA authorize EPA to regulate certain risks of “injury to health or the environment.” While TSCA does not define “health,” Section 3(6) defines “environment” to include “water, air, and land and the interrelationship which exists among and between water, air, and land and all living things.”

EPA has not previously taken action under TSCA based on global warming potential and its impact on climate change as a risk to health or the environment. If it were to do so, it would arguably be expanding its understanding of the scope of TSCA. 

Thus, EPA would have to decide that TSCA authorizes it to regulate the global warming potential of HFCs. EPA has already done so under the CAA. In 2007, the Supreme Court held that HFCs are air pollutants under the CAA and that the CAA gives EPA authority to regulate greenhouse gases, including HFCs, in new motor vehicles. Subsequently, EPA made an “endangerment” finding under the CAA with respect to HFCs in combination with certain other greenhouse gases and their impact on health:

Pursuant to CAA section 202(a), the Administrator finds that greenhouse gases in the atmosphere may reasonably be anticipated both to endanger public health and to endanger public welfare. Specifically, the Administrator is defining the “air pollution” referred to in CAA section 202(a) to be the mix of six long-lived and directly-emitted greenhouse gases: carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF6).

Section 5: PMNs and Significant New Use Rules

Under TSCA Section 5, no one may manufacture (a term that includes import) a “new” chemical substance (one not on the TSCA Inventory) unless the substance is covered by an exemption or the person first submits a premanufacture notice (PMN) for the substance to EPA and EPA reviews the PMN. Based on its review, EPA may issue an order banning or restricting the manufacturer’s use of that substance if it finds that the substance “may present an unreasonable risk to health or the environment” or a related finding.

This provision is only relevant to HFCs that are not already on the TSCA Inventory. All of the HFCs ever approved by EPA under the SNAP program are on the Inventory.

Also under Section 5, no one may manufacture or process a chemical substance for “significant new use,” as determined by EPA through a significant new use rule (SNUR), unless such person submits and EPA reviews a significant new use notice (SNUN, the same as a PMN). As with a PMN, EPA’s review of a SNUN may lead to an order imposing a ban or a variety of restrictions use if EPA finds that the use “may present an unreasonable risk” or a related finding. SNURs are available for “existing” chemical substances (those on the Inventory), such as those approved under the SNAP Program.

Section 5(a) sets a low bar for SNUR rulemaking. Unlike under some other provisions of TSCA, EPA need not make any determination about the unreasonableness of a risk to health or the environment in order to promulgate a SNUR. Instead, it need only find that a new use “may reasonably be expected to have health or environmental importance” and consider “all relevant factors,” including:

  • The projected volume of manufacturing and processing of a chemical substance;
  • The extent to which a use changes the type or form of exposure of human beings or the environment to a chemical substance;
  • The extent to which a use increases the magnitude and duration of exposure of human beings or the environment to a chemical substance; and
  • The reasonably anticipated manner and methods of manufacturing, processing, distribution in commerce, and disposal of a chemical substance.

In recent years, EPA has promulgated SNURs for four substitute chemicals under the SNAP program:

  • HFO-1234yf / 1-Propene, 2,3,3,3-tetrafluoro- (CAS No. 754-12-1);
  • HFO-1336mzz(Z) / 2-Butene, 1,1,1,4,4,4-hexafluoro-, (2Z)- (CAS No. 692-49-9);
  • HFE-7300 / Pentane, 1,1,1,2,2,3,4,5,5,5-decafluoro-3-methoxy-4-(trifluoromethyl)- (CAS No. 132182-92-4); and
  • HFE-347pcf2 / Ethane, 1,1,2,2-tetrafluoro-1-(2,2,2-trifluoroethoxy)- (CAS No. 406-78-0).

EPA did not cite global warming potential or climate change in connection with any of these SNURs. Instead, it cited anticipated adverse environmental or health effects from the particular uses. These examples show that EPA at least is willing to consider SNURs as a tool to review or manage SNAP chemicals. Based on the factors under Section 5(a) and EPA’s past practices, it might take the position that the global warming potential of HFCs means that their use “may reasonably be expected to have health or environmental importance,” thus justifying promulgation of a SNUR.

SNURs are a limited tool, however. They are only available for “new” uses, not “ongoing” uses. All of the HFCs previously regulated by EPA under the SNAP program have “ongoing” uses, generally making SNURs unavailable for those uses. If ongoing uses were to be discontinued, EPA could promulgate a SNUR for those uses since at that point they would no longer be ongoing and could be considered to be “new.”

SNURs would be more useful to EPA where new HFCs are introduced. If the HFCs are “new” chemical substances, not on the TSCA Inventory, and not exempt, they would need to undergo PMN review. Once EPA completes its evaluation of a PMN, it could adopt a SNUR for the PMN substance since at that point all of its uses would be “new.”

Section 6: Risk Management Regulation

 The other regulatory tool potentially relevant under TSCA is the risk management provision, Section 6. Section 6(a) offers EPA a wide variety of risk management options, ranging from labeling requirements to a ban on all uses of a chemical substance. However, the Section 6 process takes years to complete.

With the exception of certain persistent, bioaccumulative, and toxic chemicals, EPA may only promulgate a risk management rule under Section 6(a) after it has conducted a risk evaluation for a chemical substance and determined that the substance presents an unreasonable risk to health or the environment, without regard to cost or other non-risk factors.

There are three ways for EPA to begin a risk evaluation under Section 6. One is through prioritization. EPA would have to designate an HFC as a high priority for a risk evaluation, a process that takes nine to twelve months. That process is likely to be preceded by an as-yet undefined period of pre-prioritization activity as EPA considers potential candidates for prioritization.

EPA could also decide to conduct a risk evaluation for a chemical substance by accepting a manufacturer’s request that it do so, or by designating the substance as one of ten initial chemicals to receive a risk evaluation.

Once EPA begins a risk evaluation, it has a statutory deadline of three years (extendable by six months) to complete the evaluation. EPA has some discretion about which conditions of use to evaluate, so the scope document for the risk evaluation, due six months after initiation of the risk evaluation, would have to specify conditions of use relevant to the global warming potential of the HFC. In the final risk evaluation, EPA would have to determine whether or not any of the evaluated conditions of use of the HFC presents an unreasonable risk of injury to health or the environment without regard to costs or other non-risk factors.

If EPA determines that at least one condition of use presents an unreasonable risk, it must proceed to adopt risk management measures through rulemaking. A proposed rule is due one year after publication of the risk evaluation, with the final rule due two years after publication, possibly subject to an extension of up to two years.

EPA did designate a SNAP program chemical, methylene chloride, as one of the initial ten chemicals to receive a risk evaluation, and the scope document did address the chemical’s global warming potential to some degree. However, the scope document focused on potential hazards to human and ecological receptors from direct exposure. 

In some cases, a risk management rule under Section 6 must consider “whether technically and economically feasible alternatives that benefit health or the environment, compared to the use so proposed to be prohibited or restricted, will be reasonably available as a substitute when the proposed prohibition or other restriction takes effect.” EPA published a proposed Section 6(a) rule for a few uses of methylene chloride, based on an earlier risk assessment of those uses. The preamble noted that the chemical has “a global warming potential chemical with a value of 8.7 GWP, or approximately 8.7 times more heat absorptive than carbon dioxide.” It also noted its status under the SNAP program and that it is not an ODS. The proposed rule was based only on the human health hazards of direct exposure to methylene chloride, however, not on its global warming potential. Nevertheless, the discussion in the preamble of alternatives to methylene chloride for the restricted applications did address the global warming potential of the alternatives.

The methylene chloride example shows that EPA is aware of the global warming potential of the chemicals it will regulate under Section 6 and their alternatives. EPA has yet to take the next step of making that potential a basis for either a risk evaluation or a risk management rule.


If the D.C. Circuit denies further review of the decision in Mexichem, it will significantly limit EPA’s ability to regulate the use of HFCs on the basis of their global warming potential under the existing CAA.

The Court suggested TSCA as a possible alternative route to regulation. We have shown that there are several threshold questions that must be addressed before reaching the conclusion that TSCA is available to regulate HFCs on the basis of global warming potential. 

Assuming that it is available, the fastest route to regulation under TSCA, significant new use rules under Section 5, are generally not available to EPA for existing HFCs. New HFCs, those not on the TSCA Inventory, may be subject to restriction by order once they undergo PMN review and through SNURs following the completion of that review. 

The alternative of risk management rulemaking under Section 6 is a lengthy process. EPA has put a toe in the water, so to speak, by considering the greenhouse gas potential of a chemical substance being regulated under Section 6 and its alternatives, but it has yet to commit itself to include the impact on climate change as a basis for decision making under Section 6.