National Emissions Standards for Hazardous Air Pollutants: Reinforced Plastic Composites Production

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 63

[OAR-2003-0003: FRL-7461-7]

RIN 2060-AE79

National Emissions Standards for Hazardous Air Pollutants: Reinforced Plastic Composites Production

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

This action promulgates national emissions standards for hazardous air pollutants (NESHAP) for new and existing reinforced plastic composites production facilities. The NESHAP regulate production and ancillary processes used to manufacture products with thermoset resins and gel coats. Reinforced plastic composites production facilities emit hazardous air pollutants (HAP), such as styrene, methyl methacrylate (MMA), and methylene chloride (dichloromethane). These HAP have adverse health effects including headache, fatigue, depression, irritation of skin, eyes, and mucous membranes. Methylene chloride has been classified as a probable human carcinogen. The NESHAP will implement section 112(d) of the Clean Air Act (CAA) by requiring all major sources in this category to meet HAP emissions standards reflecting the application of the maximum achievable control technology (MACT). We estimate the final NESHAP will reduce nationwide emissions of HAP from these facilities by approximately 7,682 tons per year (tpy) (43 percent).

EFFECTIVE DATE:

April 21, 2003.

ADDRESSES:

Docket.

Docket ID No. OAR-2003-0003 (formerly Docket No. A-94-52) contains supporting information used in developing the standards. The docket is available for public viewing at the Office of Air and Radiation Docket and Information Center (Air Docket) in the EPA Docket Center, EPA West, Room B108, 1301 Constitution Avenue NW., Washington, DC.

FOR FURTHER INFORMATION CONTACT:

For further information concerning applicability and rule determinations, contact the appropriate State or local agency representative. For information concerning the analyses performed in developing the NESHAP, contact Keith Barnett, U.S. EPA, Emission Standards Division, Minerals and Inorganic Chemicals Group, C504-05, Research Triangle Park, North Carolina 27711, (919) 541-5605,

barnett.keith@epa.gov.

SUPPLEMENTARY INFORMATION:

Docket.

We have established an official public docket for this action under Docket ID No. OAR-2003-0003 (formerly Docket No. A-94-52). The docket is an organized and complete file of the information considered by the EPA in the development of this rulemaking. The docket is a dynamic file because material is added throughout the rulemaking process. The docketing system is intended to allow members of the public and industries involved to readily identify and locate documents so that they can effectively participate in the rulemaking process. Along with the proposed and promulgated standards and their preambles, the contents of the docket, excluding interagency review materials, will serve as the record in the case of judicial review. (See section 307(d)(7)(A) of the CAA.) The regulatory text and other materials related to this rulemaking are available for review in the docket or copies may be mailed on request from the Air Docket by calling (202) 566-1742. A reasonable fee may be charged for copying docket materials.

Electronic Docket Access.

You may access the final rule electronically through the EPA Internet under the “

Federal Register

” listings at

http://www.epa.gov/fedrgstr/.

An electronic version of the public docket is available through EPA's electronic public docket and comment system, EPA Dockets. You may use EPA Dockets at

http://www.epa.gov/edocket/

to view public comments, access the index listing of the contents of the official public docket, and to access those documents in the public docket that are available electronically. Although not all docket materials may be available electronically, you may still access any of the publicly available docket materials through the docket facility in the above paragraph entitled “Docket.” Once in the system, select “search,” then key in the appropriate docket identification number.

Worldwide Web (WWW).

In addition to being available in the docket, an electronic copy of today's final NESHAP will also be available on the WWW through the Technology Transfer Network (TTN). Following the Administrator's signature, a copy of the NESHAP will be posted on the TTN's policy and guidance page for newly proposed or promulgated rules at

http://www.epa.gov/ttn/oarpg.

The TTN provides information and technology exchange in various areas of air pollution control. If more information regarding the TTN is needed, call the TTN HELP line at (919) 541-5384.

Regulated Entities.

Categories and entities potentially regulated by this action include:

Category

NAICS code

SIC code

Examples of regulated entities

Industry

325211

326122

325991

326191

327991

327993

2821

3084

3087

3088

3089

3281

3296

Reinforced plastic composites production facilities that manufacture intermediate and/or final products using styrene containing thermoset resins and gel coats.

332998

33312

33651

335311

335313

335312

33422

336211

336112

3431

3531

3531

3612

3613

3621

3663

3711

3711

336211

33651

33653

336399

33612

3713

3714

3714

3716

336213

336413

336214

3728

3743

3792

3999

Federal Government

Federally owned facilities that manufacture intermediate and/or final products using styrene containing thermoset resins and gel coats.

This table is not intended to be exhaustive, but rather provides a guide for readers regarding entities likely to be regulated by this action. To determine whether your facility is regulated by this action, you should examine the applicability criteria in §§ 63.5785 and 63.5787 of the final NESHAP. If you have any questions regarding the applicability of this action to a particular entity, consult the person listed in the preceding

FOR FURTHER INFORMATION CONTACT

section.

Judicial Review.

The NESHAP for Reinforced Plastic Composites Manufacturing were proposed on August 2, 2001 (66 FR 40324). This action announces EPA's final decisions on the NESHAP. Under section 307(b)(1) of the CAA, judicial review of the final NESHAP is available only by filing a petition for review in the U.S. Court of Appeals for the District of Columbia Circuit by June 20, 2003. Under section 307(d)(7)(B) of the CAA, only an objection to a rule or procedure raised with reasonable specificity during the period for public comment can be raised during judicial review. Moreover, under section 307(b)(2) of the CAA, the requirements established by the final rule may not be challenged separately in any civil or criminal proceeding brought to enforce these requirements.

Outline.

The information presented in this preamble is organized as follows:

I. Introduction

A. What is the purpose of NESHAP?

B. What is the source of authority for development of NESHAP?

C. What processes and operations are included in the Reinforced Plastic Composites Production source category?

II. Summary of the Final NESHAP

A. What source categories and subcategories are affected by the final NESHAP?

B. What are the primary sources of HAP emissions and what are the emissions?

C. What is the affected source?

D. What are the HAP emissions limits, operating limits, and other standards?

E. What are the HAP emissions factor equations in Table 1 to subpart WWWW of part 63, and how are they used in the final NESHAP?

F. When would I need to comply with the final NESHAP?

G. What are the options for demonstrating compliance?

H. What are the testing and initial compliance requirements?

I. What are the continuous compliance requirements?

J. What are the notification, reporting, and recordkeeping requirements?

III. Summary of Environmental, Energy, and Economic Impacts

A. What facilities are affected by the final NESHAP?

B. What are the air quality impacts?

C. What are the water quality impacts?

D. What are the solid and hazardous waste impacts?

E. What are the energy impacts?

F. What are the cost impacts?

G. What are the economic impacts?

IV. Summary of Changes Since Proposal

A. Above-the-Floor Capture and Control Requirements for Existing Sources

B. Replacing the Point Value Equations with HAP Emissions Factor Equations Based on the Unified Emissions Factors, and Changes to Centrifugal Casting HAP Emissions Factors

C. MACT Floors for Existing Sources

D. Cleaning

E. Compression/Injection Molding

F. Averaging Provisions

G. Pultrusion Compliance Options

H. Applicability

I. Potential Overlap with the Boat Manufacturing NESHAP (40 CFR Part 63, Subpart VVVV)

J. Determination of Resin and Gel Coat HAP Content

K. New Source MACT Floors

V. Summary of Responses to Major Comments

VI. Relationship of the Final NESHAP to Other NESHAP and the CAA Operating Permits Program

A. National Emissions Standards for Closed Vent Systems, Control Devices, Recovery Devices, and Routing to a Fuel Gas System of a Process (40 CFR Part 63, Subpart SS)

B. NESHAP for Boat Manufacturing (40 CFR Part 63, Subpart VVVV)

C. NESHAP for Plastic Parts and Products (Surface Coating)

D. Operating Permit Program

VII. Statutory and Executive Order Reviews

A. Executive Order 12866, Regulatory Planning and Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Analysis

D. Unfunded Mandates Reform Act

E. Executive Order 13132, Federalism

F. Executive Order 13175, Consultation and Coordination with Indian Tribal Governments

G. Executive Order 13045, Protection of Children from Environmental Health Risks and Safety Risks

H. Executive Order 13211, Actions that Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer and Advancement Act

J. Congressional Review Act

I. Introduction

A. What Is the Purpose of NESHAP?

The purpose of the final NESHAP is to protect the public health by reducing emissions of HAP from Reinforced Plastic Composite Manufacturing facilities.

B. What Is the Source of Authority for Development of NESHAP?

Section 112 of the CAA requires us to list categories and subcategories of major sources and area sources of HAP and to establish NESHAP for the listed source categories and subcategories. Reinforced Plastic Composites Production was included on the initial list of source categories published on July 16, 1992 (57 FR 31576). Major sources of HAP are those that have the potential to emit 10 tpy or more of any one HAP or 25 tpy or more of any combination of HAP.

The CAA requires NESHAP to reflect the maximum degree of reduction in emissions of HAP that is achievable. This level of control is commonly referred to as the MACT.

The MACT floor is the minimum control level allowed for NESHAP. This concept appears in section 112(d)(3) of the CAA. For new sources, the MACT floor cannot be less stringent than the HAP emissions control that is achieved in practice by the best-controlled similar source. The MACT standards for existing sources can be less stringent than standards for new sources, but they cannot be less stringent than the average HAP emissions limitation achieved by the best-performing 12 percent of existing sources in the category or subcategory (or the best-performing five sources for categories or subcategories with fewer than 30 sources).

In developing MACT, we also consider control options that are more stringent than the floor. We may establish standards more stringent than the floor based on the consideration of cost of achieving the HAP emissions reductions, any non-air quality health and environmental impacts, and energy requirements.

C. What Processes and Operations Are Included in the Reinforced Plastic Composites Production Source Category?

The Reinforced Plastic Composites Production source category involves the production of plastic products from cross-linking resins, usually in combination with reinforcing materials and inorganic fillers. These products may have an outer surface produced with a styrene-containing gel coat. The production of products that do not contain reinforcing materials is also included in this category, as well as the production of intermediate compounds that are later used to make the final plastic products. These non-reinforced products were included because they are produced using the same types of resins, have similar HAP emissions characteristics, and would use similar HAP emissions controls. This source category is limited to those resins and gel coats which contain styrene, either by itself or with a combination of other monomers or solvents.

There are a wide variety of operations that use styrene-containing resins to make thermoset plastics. Such manufacturing operations include manual resin application, mechanical resin application, filament application, gel coat application, compression/injection molding, resin transfer molding, centrifugal casting, continuous lamination/casting, polymer casting, pultrusion, bulk molding compound (BMC) manufacturing, and sheet molding compound (SMC) manufacturing. There are also ancillary operations such as cleaning, mixing, repair, and HAP-containing materials storage, that occur in conjunction with these manufacturing operations. Many facilities will use multiple operations in manufacturing their products.

This source category also includes some repair operations that take place at a manufacturing facility, such as repairs of parts or products that are manufactured at the same facility that must be repaired due to defects or damage that occur during manufacturing, or repairs of parts that were originally manufactured at that location and have been returned for repair due to defects in the original manufacture or damage in shipment. No other types of repair operations are included in this source category. Facilities that perform non-routine manufacture of reinforced plastic composites parts solely to replace parts of a reinforced plastic composite product that has been in use are not considered to be manufacturing facilities, and repair operations at these types of facilities are not part of this source category. See § 63.5935 of the final rule for the definition of non-routine manufacture. We believe that repair operations that are collocated with manufacturing operations that originally produce the reinforced plastic composites being repaired use the same materials as the manufacturing processes. Repair operations that are not collocated may use different materials and application techniques.

II. Summary of the Final NESHAP

A. What Source Categories and Subcategories Are Affected by the Final NESHAP?

Today's final rule applies to the Reinforced Plastic Composites Production source category. We developed subcategories based on size (

i.e.,

tpy of HAP emitted) in defining the new source MACT floors. These subcategories are sources that emit 100 tpy or more from open molding, pultrusion, centrifugal casting, continuous lamination/casting, SMC and BMC manufacturing, and mixing operations; and all other new sources. The new source MACT floors incorporate add-on controls for sources in the first subcategory, except for facilities producing large parts, and pollution prevention for other new sources.

The floors for existing sources are mainly based on pollution prevention, not add-on controls. Where floors are based mainly on pollution-prevention control techniques, we did not subcategorize by size. However, we did segregate existing sources by resin application technique, resin type, and final products, and developed separate floors for each process/product grouping.

B. What Are the Primary Sources of HAP Emissions and What Are the Emissions?

The primary source of HAP emissions from the Reinforced Plastic Composites Production source category is the evaporation of styrene and other organic liquid HAP contained in the resin during the application and/or curing of the resin. Since styrene participates in the curing reaction, not all of it is emitted. Organic HAP emissions also occur during ancillary operations such as cleaning, mixing, repair, and HAP containing materials storage. Although some gel coats or resins may contain inorganic HAP, such as lead, in resin solids or pigments, we have no data to

indicate the inorganic HAP are emitted from the production process. Therefore, only organic HAP are addressed by the final NESHAP.

Total baseline HAP emissions from the Reinforced Plastic Composites Production source category are approximately 18,000 tpy. The HAP emissions from spray lay-up and gel coating constitute approximately 52 percent and 23 percent of the total baseline HAP emissions, respectively. The remaining HAP emissions are primarily from hand lay-up/bucket and tool application, compression molding/injection molding, filament application, SMC manufacturing, and centrifugal casting.

C. What Is the Affected Source?

The affected source is the combination of all operations regulated under these standards at a reinforced plastic composites production facility. The following regulated operations are typically performed at reinforced plastic composites production facilities and are part of the affected source: open molding, closed molding, centrifugal casting, continuous lamination/casting, polymer casting, pultrusion, SMC manufacturing, equipment cleaning, mixing, BMC manufacturing, repair, and storage of HAP-containing materials. Repair operations are also included as part of the affected source if the repair is made to a part manufactured at that location.

D. What Are the HAP Emissions Limits, Operating Limits, and Other Standards?

We are promulgating the requirements of the final NESHAP in the form of HAP emissions limits (

i.e.,

HAP emissions factors, mass rate, or percent reduction), operating limits, and work practice standards. Work practice standards include design, equipment, work practices, and operational standards.

The final NESHAP contain a HAP emissions threshold that distinguishes between sources that typically can meet the HAP emissions limits using pollution prevention, and those that must use add-on controls. This threshold is called the “100 tpy threshold.” For existing sources, you determine if you are below, above, or equal to the 100 tpy threshold by summing all HAP emissions from centrifugal casting and continuous lamination/casting operations at the source. In determining HAP emissions from centrifugal casting operations, only HAP emissions from venting of the centrifugal casting mold during spinning and/or curing are considered. The HAP emissions that occur from application of resin or gel coat to an open centrifugal casting mold are considered to be open molding HAP emissions. The HAP emissions from other operations or processes are not included because the 100 tpy threshold does not apply to other operations or processes.

For new sources, you determine if you are below, above, or equal to the 100 tpy threshold by summing all HAP emissions from open molding, pultrusion, SMC and BMC manufacturing, centrifugal casting, continuous lamination/casting, and mixing operations at the source. The HAP emissions from closed molding, cleaning, repair and HAP-containing materials storage are not used in threshold determinations. In determining HAP emissions from centrifugal casting operations, only HAP emissions from venting of the centrifugal casting mold are included. The HAP emissions that occur from application of resin or gel coat to an open centrifugal casting mold are considered to be open molding HAP emissions.

The requirements for new and existing sources that are below the 100 tpy threshold are based on the MACT floor level of control. These requirements are summarized in the following table:

Table 1.—Summary for Existing Sources, and New Sources Below the 100 tpy HAP Emissions Threshold

If your operation type is . . .

And you use . . .

MACT for existing facilities and new facilities that are below the 100 tpy threshold is . . .

1. Open molding—corrosion-resistant and/or high strength (CR/HS)

a. mechanical resin application

112 lb/ton.

b. filament application

171 lb/ton.

c. manual resin application

123 lb/ton.

2. Open molding—non-CR/HS

a. mechanical resin application

87 lb/ton.

b. filament application

188 lb/ton.

c. manual resin application

87 lb/ton.

3. Open molding—tooling

a. mechanical resin application

254 lb/ton.

b. manual resin application

157 lb/ton.

4. Open molding—low-flame spread/low-smoke products

a. mechanical resin application

497 lb/ton.

b. filament application

270 lb/ton.

c. manual resin application

238 lb/ton.

5. Open molding—shrinkage controlled resin

a. mechanical resin application

354 lb/ton.

b. filament application

215 lb/ton.

c. manual resin application

180 lb/ton.

6. Open molding—gel coat

b

a. tooling gel coating

437 lb/ton.

b. white/off white pigmented gel coating

267 lb/ton.

c. all other pigmented gel coating

377 lb/ton.

d. CR/HS or high performance gel coat

605 lb/ton.

e. fire retardant gel coat

854 lb/ton.

f. clear production gel coat

522 lb/ton.

7. Centrifugal casting—CR/HS

c

N/A

25 lb/ton.

8. Centrifugal casting—non-CR/HS

c

N/A

20 lb/ton.

9. Pultrusion

d

N/A

Reduce total HAP emissions by at least 60 weight percent.

10. Continuous lamination/casting

N/A

Reduce total HAP emissions by at least 58.5 weight percent or not exceed a HAP emissions limit of 15.7 lbs of HAP per ton of neat resin plus and neat gel coat plus.

11. A closed molding operation using compression/injection molding

Uncover, unwrap or expose only one charge per mold cycle per compression/injection molding machine. For machines with multiple molds, one charge means sufficient material to fill all molds for one cycle. For machines with robotic loaders, no more than one charge may be exposed prior to the loader. For machines fed by hoppers, sufficient material may be uncovered to fill the hopper. Hoppers must be closed when not adding materials. Materials may be uncovered to feed to slitting machines. Materials must be recovered after slitting.

12. A cleaning operation

Do not use cleaning solvents that contain HAP, except that HAP containing materials may be used in closed systems, and to clean cured resin from application equipment. Application equipment includes any equipment that directly contacts resin between storage and applying resin to the mold or reinforcement.

13. A HAP-containing materials storage operation

Keep containers that store HAP-containing materials closed or covered except during the addition or removal of materials. Bulk HAP-containing materials storage tanks may be vented as necessary for safety.

14. A SMC manufacturing operation

Close or cover the resin delivery system to the doctor box on each SMC manufacturing machine. The doctor box itself may be open.

15. A SMC manufacturing operation

Use a nylon containing film or a film with an equal or lower permeability to styrene compared to a nylon containing film to enclose SMC.

16. A mixing or BMC manufacturing operation

d

Use mixer covers with no visible gaps present in the mixer covers. Gaps of up to 1 inch are permissible around mixer shafts and any required instrumentation.

17. A mixing or BMC manufacturing operation

e

Do not actively vent mixers to the atmosphere while the mixing agitator is turning.

18. A mixing or BMC manufacturing operation

e

Keep the mixer covers closed during mixing except when adding materials to the mixing vessels.

19. A new or existing pultrusion operation manufacturing parts with 1000 or more reinforcements and a cross section area of 60 square inches or more that is not subject to the 95 percent HAP emissions requirement

i. not allow vents from the building ventilation system, or local or portable fans to blow directly on or across the wet-out area(s).

ii. not permit point suction of ambient air in the wet-out area(s) unless that air is directed to a control device.

iii. use devices such as deflectors, baffles, and curtains when practical to reduce air flow velocity across the wet-out area(s).

iv. direct any compressed air exhausts away from resin and wet-out area(s).

v. convey resin collected from drip-off pans or other devices to reservoirs, tanks, or sumps via covered troughs, pipes, or other covered conveyance that shields the resin from the ambient air.

vi. cover all reservoirs, tanks, sumps, or HAP-containing materials storage vessels except when they are being charged or filled.

vii. cover or shield from ambient air resin delivery systems to the wet-out area(s) from reservoirs, tanks, or sumps where practical.

a

HAP emissions limits for open molding and centrifugal casting expressed as lb/ton are calculated using the equations shown in Table 1 to subpart WWWW of part 63. You must be at or below these values based on a 12-month rolling average.

b

These limits are for spray application of gel coat. Manual gel coat application may be included as part of spray gel coat application for compliance purposes using the same HAP emissions factor equation and HAP emissions limit.

c

Centrifugal casting operations where the resin is injected into the mold and the mold is completely closed during spinning and curing may be treated as closed molding operations.

d

Pultrusion machines that produce parts with 1000 or more reinforcements and a cross sectional area of 60 inches or more are not subject to this requirement. Their requirement is the work practice of air flow management reduction.

e

Containers of 5 gallons or less may be open when active mixing is taking place, or during periods when they are in process (

i.e.

, they are actively being used to apply resin). For polymer casting mixing operations, containers with a surface area of 500 square inches or less may be open while active mixing is taking place.

For existing sources that are equal to or above the 100 tpy HAP emissions threshold, centrifugal casting and continuous lamination/casting operations meet an above-the-floor requirement based on 95 percent control of HAP emissions.

The requirements for new sources that are equal to or above the 100 tpy HAP emissions threshold are also based on the floor level of control. The floor level of control for these sources is a 95 percent reduction of HAP emissions for open molding, pultrusion, SMC and BMC manufacturing, centrifugal casting, continuous lamination/casting, and mixing operations with one exception. For open molding and pultrusion operations at new sources that produce large parts, the floor level of control is the same as existing sources shown in the previous table. All other operations meet the requirements shown in the previous table.

In developing final requirements for reinforced plastic composites affected sources, we have provided an alternative format where possible. For example, a facility meeting a 95 percent HAP emissions reduction requirement for open molding processes can alternatively meet a HAP emissions limit. We have also provided alternatives for meeting the limits for continuous lamination/casting and SMC manufacturing operations.

E. What Are the HAP Emissions Factor Equations in Table 1 to Subpart WWWW of Part 63, and How Are They Used in the Final NESHAP?

Table 1 to subpart WWWW of part 63 presents a series of HAP emissions factor equations for open molding and centrifugal casting operations. These equations are specific to the type of resin and gel application and HAP emissions reduction technique used. These equations allow you to calculate HAP emissions factors based on HAP content and application method for each material that you use. These HAP emissions factors are then averaged and compared to limits in the final standards to determine if your open molding and centrifugal casting operations are in compliance.

The HAP emissions factor equations for open molding are identical to HAP emissions equations developed by the composites industry called the Unified Emissions Factors (UEF) as they existed at the time of final rule development. These equations can also be combined with resin and gel coat use to determine HAP emissions rates. It should be noted that although the equations are identical to the UEF at the time the rule is finalized, for purposes of compliance, only the equations actually contained in Table 1 to subpart WWWW of part 63 may be used.

F. When Would I Need To Comply With the Final NESHAP?

We are requiring that all existing sources comply by April 21, 2006. Any source that commenced construction after August 2, 2001, at a site where there were no existing reinforced plastic composite operations is a new source. New affected sources that are now in operation must be in compliance on April 21, 2003. New affected sources that startup after April 21, 2003 must comply upon startup. Existing area sources that increase their HAP emissions or their potential to emit such that they become a major source of HAP must be in compliance within 3 years of the date they become a major source. New area sources that become major sources of HAP must comply upon becoming a major source. All open molding and centrifugal casting operations that comply by meeting a specified HAP emissions limit on a 12-month rolling average will have 1 year from the compliance date to demonstrate compliance.

We are allowing new and existing facilities 3 years to comply from the time their HAP emissions reach or exceed the applicability thresholds which require the installation of add-on controls, if these HAP emissions increases occur after their initial compliance date.

In addition, we have added a one-time exemption for facilities that exceed the 100 tpy threshold due to unusual circumstances. Facilities that apply for this exemption and subsequently exceed the threshold the next year, must comply within 3 years from the time their HAP emissions first exceeded the threshold. Because this is a one-time exemption, an exceedance in any future years would result in a requirement for compliance within 3 years of the subsequent exceedance.

G. What Are the Options for Demonstrating Compliance?

Today's final NESHAP provide several options for compliance for certain operations. We are providing these options to afford industry the flexibility to decide which method is best suited for each particular situation. Operations not listed in this section have only one option for demonstrating compliance.

For open molding and centrifugal casting operations, you determine compliance with the HAP emissions limits by determining HAP emissions factors for the operations at your facility, and comparing your HAP emissions factors to the appropriate HAP emission limits for each open molding and centrifugal casting operation. To determine your HAP emissions factor you may use the HAP emissions factor equations in Table 1 to subpart WWWW, or HAP emissions factors based on facility HAP emissions testing. For open molding operations at existing and new sources, the final rule allows you to choose to comply by meeting the individual HAP emissions limits shown in Table 3 to subpart WWWW of part 63 for each operation at your affected source, or by meeting the weighted average HAP emissions limit for all open molding operations at your affected source. In addition, if you produce parts with any combination of manual resin application, mechanical resin application, filament application, or centrifugal casting operations at your affected source, you can comply using the an alternative method shown in Table 7 to subpart WWWW of part 63. You determine the highest allowable HAP resin for each individual operation from Table 3 to subpart WWWW of part 63. This same resin can then be used in all open molding and centrifugal casting operations as shown in Table 7 to subpart WWWW of part 63.

For open molding and centrifugal casting operations where the rule would require you to meet a percent reduction, you could use an add-on control device to achieve the required reduction, or you may choose to meet a HAP emissions limit that corresponds to that particular operation's percent reduction.

For continuous lamination/casting operations at existing and new sources, we are allowing you to demonstrate that each continuous casting line and each

continuous lamination line meets the appropriate standard in Table 3 to subpart WWWW of part 63, or § 63.5805(b) or (d) of the final rule. Alternatively, you can average all your continuous casting and continuous lamination lines together and demonstrate that they meet the appropriate standard. An additional alternative for sources that emit less than the 100 tpy threshold would be to capture your HAP emissions from your wet-out area in a permanent total enclosure that meets EPA's criteria, as specified in Method 204 of appendix M of 40 CFR part 51, and vent the captured wet-out HAP emissions through a closed vent system to a control device achieving 95 percent reduction of HAP emissions. Under the final rule, these alternatives can be used in combination to demonstrate compliance.

The standards for continuous lamination/casting operations are expressed as a percent reduction of HAP emissions. As an alternative, facilities can elect to meet a HAP emissions limit.

For existing and new pultrusion operations, you can capture and vent your HAP emissions to a control device that achieves the required percent reduction of HAP emissions. For all existing sources and for new sources that emit less than the 100 tpy threshold, you may use a wet-area enclosure with a resin drip collection system, direct die injection or preform injection systems that meet the criteria specified in § 63.5830 of the final rule to meet the 60 percent HAP emissions reduction requirement. For pultrusion machines that produce parts with 1000 or more reinforcements and a cross sectional area of 60 inches or more, you must implement the work practice standards in Table 4 to subpart WWWW of part 63.

For SMC manufacturing operations at new sources that exceed the 100 tpy threshold, we allow facilities to meet a 95 percent HAP emissions reduction requirement, or the HAP emissions limit specified in Table 5 to subpart WWWW of part 63.

H. What Are the Testing and Initial Compliance Requirements?

We are requiring you to conduct an initial performance test using specified EPA test methods on all affected sources which use a control device to achieve compliance. You must test at the inlet and outlet of the control device and using these results, calculate a percent reduction.

We are also requiring you to conduct a design evaluation, as specified by EPA Method 204 of appendix M of 40 CFR part 51, if you use permanent total enclosures to capture HAP emissions. If your enclosure does not meet the requirements for a permanent total enclosure, you must test the enclosure to determine the capture efficiency by EPA Methods 204B through E of appendix M of 40 CFR part 51 or an alternative method that meets the data quality objectives and lower confidence limit approaches contained in 40 CFR part 63, subpart KK. Test runs for EPA Methods 204B through E or alternative test methods must be at least 3 hours.

Prior to the initial performance test, owners and operators of affected sources would be required to install the parameter monitoring equipment to be used to demonstrate compliance with the operating limits. During the initial performance test, the owners and operators would use the parameter monitoring equipment to establish operating parameter limits.

I. What Are the Continuous Compliance Requirements?

If you use an add-on control device, we are requiring that you monitor and record the operating parameters established during the initial performance test, and calculate average operating parameter values averaged over the period of time specified in the final NESHAP to demonstrate continuous compliance with the operating limits.

If you use the HAP emissions equations in Table 1 to subpart WWWW of part 63 to demonstrate that you are maintaining a HAP emissions factor less than or equal to the appropriate HAP emissions limit listed in the final NESHAP, we are requiring that you calculate the HAP emissions factor one time if the resins or gel coats used in the operation remain the same, or if all the resins and gel coats used individually meet the applicable HAP emissions limit. You are required to calculate HAP emissions factors on a 12-month rolling average each month if the resin or gel coat varies between operations or varies over time, and not all resins or gel coats taken individually meet the required HAP emissions limit.

If you are complying with work practice standards, we are requiring that you demonstrate compliance with the work practice standards in the final NESHAP by performing the necessary work practices and by keeping a record certifying that you are in compliance with the work practices.

J. What Are the Notification, Reporting, and Recordkeeping Requirements?

We are requiring that you submit Initial Notification, Notification of Performance Tests, and Notification of Compliance Status reports by the specified dates in the final NESHAP, which may vary depending on whether the affected source is new or existing.

You are also required to submit semiannual compliance reports. If you take action that is inconsistent with your approved startup, shutdown, and malfunction (SSM) plan, then you would need to submit SSM reports within 2 days of starting such action, and within 7 days of ending such action.

We are requiring that you keep a copy of each notification and report, along with supporting documentation for 5 years. Of this time, the 2 most recent years must be on-site. You must keep records related to SSM, records of performance tests, and records for each continuous parameter monitoring system. Under the final rule, if you must comply with the work practice standards, you also need to keep records certifying that you are in compliance with the work practices for 5 years. If you use the HAP emissions factor equations to demonstrate compliance, you must keep all data, assumptions, and calculations used to determine your HAP emissions factors. For new and existing continuous lamination/casting operations, you also must keep the following records when complying with the percent reduction or pound per ton requirements: All data, assumptions, and calculations used to determine the percent reduction or pounds per ton, as applicable; a brief description of the rationale for the assignment of an equation or factor to each formula; all data, assumptions, and calculations used to derive facility-specific HAP emissions estimations and factors; identification and rationale for the worst-case scenario; and documentation that the appropriate regulatory agency has approved all HAP emissions estimation equations and factors.

III. Summary of Environmental, Energy, and Economic Impacts

A. What Facilities Are Affected by the Final NESHAP?

There are approximately 435 existing facilities manufacturing reinforced plastic composites that are major sources and subject to the final NESHAP. The rate of growth for the reinforced plastic composites industry is estimated to be 84 new facilities over the next 5 years.

B. What Are the Air Quality Impacts?

The 1997 baseline HAP emissions from the reinforced plastic composites industry are approximately 18,000 tpy. The final NESHAP will reduce HAP

from existing sources by 7,682 tpy, a reduction of 43 percent.

The final NESHAP will result in small increases in other air pollution emissions from combustion devices that will be installed in the next 5 years to comply with today's final rule. These increases result both from the combustion device directly, and from the electrical generating plants used to generate the electricity necessary to operate the add-on controls and associated air handling equipment. These emissions are estimated to be 2.3 tpy of sulfur oxides (SOx), 3.0 tpy of nitrogen oxides (NO

X

), 4.9 tpy of carbon monoxide (CO), and 0.1 tpy of particulate matter (PM) emissions.

C. What Are the Water Quality Impacts?

We estimate that the final NESHAP will have no adverse water quality impacts. We do not expect anyone to comply by using add-on control devices or process modifications that would generate wastewater.

D. What Are the Solid and Hazardous Waste Impacts?

We estimate that the final NESHAP will decrease the amount of solid waste generated by the reinforced plastic composites industry by approximately 2,650 tpy. The decrease in solid waste is directly related to switching to nonatomized resin application equipment (

i.e.,

flowcoaters and resin rollers). Switching to nonatomized resin application equipment results in a decrease in overspray because of a greater transfer efficiency of resin to the part being manufactured. A decrease in resin overspray consequently reduces the amount of waste from disposable floor coverings, cured resin waste, and personal protective equipment (PPE) for workers. Disposable floor coverings are replaced on a periodic basis to prevent resin buildup on the floor. We estimate that solid waste generation of floor coverings will decrease by approximately 620 tpy, and that cured resin solid waste will decrease by approximately 2,030 tpy.

We project that the decreased overspray from nonatomized resin application equipment will result in a decreased usage of PPE, which also consequently reduces the amount of solid waste. When using nonatomized resin application equipment, workers typically wear less PPE than when using atomized spray guns because of the reduced presence of resin aerosols and lower styrene levels in the workplace. Because we did not have information on the many different types of PPE currently used, we did not estimate this decrease in solid waste.

Some facilities that switch from atomized to nonatomized spray guns may have a small increase of hazardous waste from the used nonatomized spray gun cleaning solvents. However, most facilities would not see an increase under the final rule, and the overall impact on the industry will be small relative to the solid waste reductions. Nearly all nonatomized spray guns require resin and catalyst to be mixed inside the gun (internal-mix) and must be flushed when work is stopped for more than a few minutes. External-mix spray guns do not need to be flushed because resin is mixed with catalyst outside the gun. Facilities that switch from external-mix to nonatomized spray guns will use more solvent. Solvent usage should not change at facilities switching from internal-mix spray guns to nonatomized spray guns. The most common flushing solvents are acetone and water-based emulsifiers. Only a couple of ounces of solvent are typically needed to flush the mixing chamber and nozzle of internal-mix spray guns.

We do not have adequate data to predict the potential solvent waste impact from switching to nonatomized spray guns. The magnitude of the impact depends on the type of gun currently used (internal- or external-mix), the frequency of flushing, and the type of solvent used. However, because of the small amount of solvent used, and since most is allowed to evaporate, we believe the overall solvent waste increase will be small compared to the solid waste reductions.

E. What Are the Energy Impacts?

Energy impacts result from the final NESHAP because some facilities will be required to install add-on controls to meet certain HAP emissions limits or percent reduction requirements. We anticipate that these controls will be concentrator/oxidizer systems or thermal oxidizers. These controls increase energy requirements in two ways. First, all reinforced plastic composites facilities must ventilate work areas to maintain worker styrene exposure within acceptable limits. The ventilation systems typically exhaust air directly to the atmosphere. When an add-on control device is added to control HAP emissions, it creates an additional pressure drop for the ventilation system which, in turn, means that more electricity is required to operate system fans and to operate the control device itself. Second, fuel (usually natural gas) is required to supplement the oxidizer combustion process.

We determined that the overall energy demand for operations in the Reinforced Plastic Composites Production source category could increase by 10 million standard cubic feet per year of natural gas, and 0.6 million kilowatt hours of electricity per year as a result of the final rule. We determined this net increase based on the additional energy demand for control devices installed to meet the final standards. No information for comparison is available on the baseline energy consumption for this source category.

F. What Are the Cost Impacts?

We have estimated the industrywide capital costs for HAP emissions control equipment, including equipment such as open container covers, resin bath enclosures, capture systems, and control devices as $12.6 million for the 435 existing sources and $22.8 million for the 84 new sources. The capital costs include the costs to purchase and install the control equipment.

We have estimated the industrywide annual costs of the final rule are $21.5 million per year for the 435 existing sources and $7.7 million for the 84 new sources. Annual costs include fixed annual costs, such as reporting, recordkeeping and capital amortization, and variable annual costs such as natural gas. The estimated average cost of the final rule is $2,800 per ton of HAP emissions reductions for existing sources and $5,560 per ton of HAP emissions reductions for new sources.

G. What Are the Economic Impacts?

We conducted a detailed economic impact analysis to determine the market- and industry-level impacts associated with the final rule. We expect the aggregate price increase for reinforced plastic composites would be only 0.7 percent, or $0.03 per pound, as a result of the final rule. We project that directly affected producers would reduce total production by 1.7 percent, while producers not directly affected would increase their production by 0.7 percent. Markets for reinforced plastic composites used in corrosion-resistant products are expected to be more heavily impacted with price increases of roughly 1.6 percent and reductions in directly affected domestic production of almost 5 percent. The reason for more significant impacts in the corrosion-resistant market is that facilities in this market have higher average per-unit variable compliance costs. Corrosion-resistant product variable compliance costs are $0.13 per pound of product versus an industry average of $0.06 per pound.

In terms of industry impacts, we analyzed impacts for captive producers and merchant producers. Captive

producers make composites for use by another part of their company in a larger product. Merchant producers sell their products on the open market, either to consumers or other businesses.

In our analysis, captive producers of reinforced plastic composites are expected to fully absorb their compliance costs, which is a conservative approach. We assess impacts as if captive producers are viewed as a profit center like a merchant producer but unable to pass on costs. This is done in lieu of an analysis attempting to estimate cost-pass through for the myriad of final products that use reinforced plastics. We assume merchant producers will attempt to pass through costs to their customers.

Through the market impacts described above, the final NESHAP create both gainers and losers within the merchant segment. Some merchant facilities are projected to experience profit increases with the final rule; however, the majority that continue operating are projected to lose profits. The economic impact analysis indicates that 36 out of 301 merchant facilities (12 percent) and 89 out of 466 product lines (19 percent) at these facilities are at risk of closure because of the final NESHAP. These facilities are believed to be small businesses. Note that this number is slightly higher than the estimate of facility closure at proposal, which was 10 percent. This change is not due to any change in stringency of the final rule as applied to small businesses. It is due the reduction in stringency of the final rule for large sources. More information on the measures we have taken to minimize the small business impacts may be found in the Regulatory Flexibility Act discussion in this preamble. Furthermore, the analysis indicates that ten of the 133 captive facilities (7.5 percent) may be at risk of closure if unable to pass on costs to their customers.

Based on the market analysis, the annual social costs of the final rule are projected to be $19.9 million. The social costs are slightly less than the engineering cost estimate of $21.5 million because producers pass on a portion of these costs to consumers through price increases in an effort to reduce their regulatory burden. These costs are distributed across the many consumers and producers of reinforced plastic composites. Directly affected producers, in aggregate, are expected to lose $6.2 million annually in profits, with those not subject to the final NESHAP gaining $18 million. The consumers of reinforced plastic composites are expected to lose $31.7 million due to higher prices and lower consumption levels associated with the final NESHAP. For more information on the economic analysis, consult the final economic impacts analysis document in the docket for this project.

IV. Summary of Changes Since Proposal

A. Above-the-Floor Capture and Control Requirements for Existing Sources

In the proposed rule, existing facilities that are a small business as defined by the Small Business Administration (SBA) regulations at 13 CFR 121.201, and that emitted 250 tpy or more of HAP, or existing facilities that are not a small business and emitted 100 tpy or more of HAP, from the combination of all open molding, centrifugal casting, continuous lamination/casting, pultrusion, SMC manufacturing, mixing, and BMC manufacturing operations, were required to reduce the total HAP emissions from these operations by at least 95 percent by weight. In the final rule, this requirement now only applies to centrifugal casting and continuous lamination/casting operations, and the threshold has been changed to 100 tpy for both large and small businesses. This reduced the number of facilities we estimated would have to meet an above-the-floor requirement from 34 to 3, reduced the industry annualized costs of the final NESHAP from $26.0 million per year to $21.5 million per year, and reduced the HAP emissions reduction estimate from 14,500 to 7,700 tpy. In addition, for centrifugal casting, the percent reduction requirement only applies to HAP emissions that are vented from the closed centrifugal casting mold. It does not apply to HAP emissions that occur from other operations such as pouring or spraying resin into a centrifugal casting mold while it is open.

B. Replacing the Point Value Equations With HAP Emissions Factor Equations Based on the Unified Emissions Factors, and Changes to Centrifugal Casting HAP Emissions Factors

In the proposed rule, we used a group of equations called point value equations to determine surrogate HAP emissions factors. These factors were then used to rank existing facilities to determine existing source MACT floors for open molding operations. However, we specified that the point value equations were not considered HAP emissions factors and, therefore, should not be used for HAP emissions reporting. This resulted in the potential for facilities to have to use two different sets of equations for HAP emissions reporting and MACT compliance determinations.

In the final rule, we have eliminated the point value equations and replaced them with HAP emissions factor equations that are identical to HAP emissions factor equations that are being used in this industry for HAP emissions calculations, called the Unified Emissions Factors. Therefore, facilities now will have the same equations for MACT compliance determinations and HAP emissions calculations for HAP emissions reports.

For centrifugal casting, we have retained the HAP emissions factor equation in the proposed rule for sources that blow heated air through the mold during spinning and curing. For other centrifugal casters, we have created a new HAP emissions factor equation based on more recent information. This new HAP emissions factor significantly changes the numerical value of the floor (pounds of HAP emissions per ton of resin used) from the value in the proposed rule. However, it did not change the floor facility or the level of control a facility would need to meet the floor.

These new HAP emissions factor equations were also used to re-rank existing facilities to establish the floor level of control for existing sources. Though this change did result in different floor values in lb/ton, it did not change the level of control actually required to meet the floor. However, as discussed below, our reanalysis did result in changes to some floors for other reasons.

C. MACT Floors for Existing Sources

There are several changes to the MACT floors for existing sources, and for new sources that emit less than 100 tpy for the combination of all open molding, centrifugal casting, pultrusion, SMC and BMC manufacturing, mixing, and continuous lamination/casting operations. These changes were a result of facilities submitting additional data that indicated our original analysis of their facility HAP emissions factors were in error, or out of date.

For noncorrosion-resistant resins applied using mechanical application, the proposed rule had different floors for filled and unfilled resins. The reason for separating filled and unfilled resins was that at the time of proposal, nonatomized resin application techniques were not available for filled resins. Since proposal, filled resins now can be applied using nonatomized spray. Therefore, we now have combined the two process/product groupings into one. Also, several facilities in this process/product

grouping provided revised data. As a result, the floor level of control for noncorrosion-resistant resins using mechanical application is a HAP emissions limit of 87 lb/ton. This limit requires a resin with no more than 38.4 percent HAP applied using nonatomized mechanical resin application techniques. At proposal, facilities could use a 42.8 percent resin (filled) or a 38 percent HAP (unfilled) resin and nonatomized mechanical resin application.

For mechanical corrosion-resistant resin application, the revised floor is a HAP emissions limit of 112 lb/ton. This limit requires a resin with no more that 46.2 percent HAP and nonatomized mechanical resin application. At proposal, a resin HAP content of up to 48.3 percent was allowed if nonatomized mechanical resin application was used.

For manual application of tooling resin, the revised floor is 157 lb/ton. This allows a resin HAP content of 45.9 percent or less. At proposal, the maximum allowable HAP content was 39.9 percent.

For tooling gel coat the revised floor is 437 lb/ton. This limits gel coat HAP content to 40 percent of less. At proposal, the limit was 38 percent or less.

For SMC manufacturing, the work practices required in the proposed rule were use of nylon film, folding the edges of the film, and covering the doctor box. In the final rule, the requirements are a covered resin transport system to the doctor box and the use of nylon-containing film.

For pultrusion operations producing parts with 1000 or more reinforcements and a cross sectional area of 60 inches or more, we have changed the floor from 60 percent HAP emissions reduction to a work practice of air flow management.

In addition, we established three new floors for speciality resins and gel coats. These are shrinkage-controlled resins, fire retardant gel coats, and high performance gel coats. These speciality products were identified from comments received on the proposed rule. The new floors are shown in Table 3 to subpart WWWW of part 63.

D. Cleaning

In the proposed rule, we required that cleaning materials contain no HAP unless cleaning cured resin from application equipment. In the final rule, we have modified that requirement to allow HAP-containing cleaners to be used in closed systems such as closed tanks, and resin and gel coat delivery systems.

E. Compression/Injection Molding

In the proposed rule, we required that only one resin charge be uncovered at a time. We have clarified this requirement for the final rule to reflect that one charge may actually have to fill multiple molds. Also, we added a provision to allow the use of automated loaders and slitters. We also clarified that paste added to the mold and in-mold surface coatings are considered part of the closed molding operation.

F. Averaging Provisions

In the proposed rule, we allowed facilities to average across all open molding operations and all centrifugal casting operations. The average was based on a 12-month rolling average calculated monthly. In determining compliance, the average for each month was calculated and then the monthly averages were averaged over a 12-month period. In the final rule, the 12-month average is based on a weighted HAP emissions factor calculated from total resin and gel coat use over the 12-month period. This method will provide a more accurate value for the actual HAP emissions, in lb/ton, that the facility produced in the previous 12 months.

In the proposed rule, we did not allow pultrusion lines to average; each pultrusion machine had to meet the 60 percent reduction requirement for existing sources. In the final rule, we allow facilities to over control some lines, and under control (or leave uncontrolled) others, as long as the average reduction for all lines combined is 60 percent weighted by resin use. Also, we are allowing facilities to average the time that wet area enclosure covers are open across lines.

G. Pultrusion Compliance Options

In the proposed rule, we allowed pultrusion operations to use direct die injection as a compliance alternative to meet the 95 percent capture and control requirement. In the final rule, we are removing direct die injection as a compliance alternative because, based on industry data, it does not achieve 95 percent HAP emissions reduction. We still allow direct die injection as a compliance option to meet the 60 percent HAP emissions reduction requirement. We have also added another compliance option, preform injection, to meet a 60 percent HAP emissions reduction. We have also added another compliance option, airflow management work practices, for pultrusion machines that produce large parts as set forth in Table 4 to subpart WWWW of part 63.

H. Applicability

We made a number of changes dealing with rule applicability. First, we expanded the list of specific operations that are part of the source category, but are not subject to any control, reporting, or recordkeeping requirements. These operations include application of mold sealing and release agents, mold stripping and cleaning, repair of previously manufactured parts that is unrelated to collocated manufacturing operations, personal activities that are not part of the manufacturing operations (such as hobby shops on military bases), prepreg materials as defined in § 63.5935 of the final rule, non-gel coat surface coatings, repair or production materials that do not contain resin or gel coat, and research and development (R&D) operations as defined in section 112(c)(7) of the CAA. In addition, we exempted any facility that uses less than 1.2 tpy of resin and gel coat, and R&D facilities and operations at manufacturing facilities. The rationale for these changes is discussed in the responses to major comments section.

I. Potential Overlap With the Boat Manufacturing NESHAP (40 CFR Part 63, Subpart VVVV)

In the proposed rule, we were silent concerning situations where a facility could be subject to both the Boat Manufacturing NESHAP, 40 CFR part 63, subpart VVVV, and the Reinforced Plastic Composites NESHAP. In today's final rule, we have added § 63.5787 to clarify which subpart applies. In general, if your facility makes boat hulls and decks, or molds for boat hulls and decks, then 40 CFR part 63, subpart VVVV, applies to you. If 40 CFR part 63, subpart VVVV, does not apply to you, and you meet the applicability criteria in § 63.5785 of the final rule, then the Reinforced Plastics Composites NESHAP apply. If you are subject to 40 CFR part 63, subpart VVVV, and also make reinforced plastic composite parts that are not used in boat manufacture, then both 40 CFR part 63, subpart VVVV, and the Reinforced Plastic Composites NESHAP may apply. See § 63.5787 in the final NESHAP for more detail.

J. Determination of Resin and Gel Coat HAP Content

In the proposed rule, we stated that facilities could determine resin and gel coat HAP content using material safety data sheets (MSDS) or resin specification sheets. In the final rule, we have included § 63.5797, which describes in more detail how to determine resin and gel coat HAP

content. This new section also clarifies that only organic HAP are included in determining HAP content. The reason is that we have no data to indicate that any other HAP, such as inorganic HAP potentially present in pigments or resin solids, are emitted from the production process. We also now include a provision to account for normal manufacturing tolerances that occur in resin and gel coat manufacture.

K. New Source MACT Floors

In the proposed rule, the MACT floor for all open molding and pultrusion operations located at new sources above a 100 tpy HAP emission threshold was a 95 percent weight reduction in HAP emissions. In the final rule, we have subcategorized open molding and pultrusion operations by part size. For open molding and pultrusion operations that produce large parts the floor level of control is now the same as for existing sources. Large parts are defined in § 63.5805 (d)(2). All other new source MACT floors are unchanged.

V. Summary of Responses to Major Comments

This section presents a summary of significant public comments and responses. A summary of all the public comments that were received and our responses to those comments can be found in Docket ID No. OAR-2003-0003 (formerly Docket No. A-94-52).

Comment:

We received numerous comments on the above-the-floor requirements for existing sources. First, commenters stated that EPA had significantly underestimated the costs of add-on controls. They stated that industry estimates were, in some cases, ten times higher than our estimates. They stated that we had overestimated the HAP concentrations in the exhaust streams, underestimated the exhaust flows, and omitted costs for continuous monitors.

Second, the commenters claimed that we had not established that 95 percent capture and control was technically feasible for this diverse industry, and that only two facilities out of 433 actually had achieved the 100 percent capture that is required to meet an overall capture and control level of 95 percent. They also stated that these two facilities were atypical of the industry as a whole because they also had collocated coating operations that were also routed to the same control device. They further stated that the criteria of EPA Method 204 of appendix M of 40 CFR part 51 are not feasible for most facilities in this industry. For these reasons, the commenters recommended that the above-the-floor requirement be removed.

Response:

As a result of these comments, we reviewed the costing methodology for the above-the-floor requirements in the proposed rule and made changes to our costing methodology for add-on controls. Some of the major changes were lowering the default inlet concentration to the control device from 100 parts per million volume (ppmv) to 50 ppmv, revising the fan power equation, and using 2,000 operating hours per year, rather than 6,000 hours per year, as a default value in the absence of actual yearly operating information.

Based on these new costs, the cost per ton of HAP emissions reduction of the above-the-floor requirement significantly increased for most process/product groupings. As a result, we have removed the above-the-floor control requirements for all process/product groupings except centrifugal casting and continuous lamination/casting. It should be noted that the comments discussed above were based on open molding operations. We received no comments specifically on the above-the-floor requirements as applied to centrifugal casting and continuous lamination/casting.

Comment:

One commenter opposed allowing control requirements for new sources emitting less than 100 tpy to be the same as those for existing sources because a new site has the opportunity to design and incorporate pollution prevention and control strategies that would be cost-prohibitive for existing sources to implement. The commenter recommended that EPA consider more stringent requirements for new sources, including smaller sources, through generally available control technology or other approaches that would not be overly burdensome.

Another commenter adds that EPA's analysis indicates that the best controlled facilities have reduced HAP by only 95 percent, and 95 percent is most likely the maximum extent of historic regulatory requirements. The commenter notes that EPA looked at the experience of existing facilities to achieve greater than 95 percent control through add-on control in conjunction with pollution prevention and did not find facilities achieving greater control than that. While the assessment may be correct for what EPA looked at, the commenter states that examining past experience that lacks regulatory drivers for greater control is not the same as examining the present and future potential for control opportunities. The commenter believes that the proposal dismisses the potential for these two control techniques (add-on control and pollution prevention) to be applied to new sources.

Response:

We agree that new facilities can more easily incorporate pollution prevention and add-on controls. This is the reason we set the new source floor at 95 percent control for most new sources that emit over 100 tpy, and not at the same level as existing source floors.

Facilities that have incorporated add-on controls tend to be larger facilities. New facilities in this industry can be small operations that operate a limited number of hours and still be major sources. These small sources cannot reliably meet 95 percent capture and control given their limited operating schedules and their potential lack of on-site technical expertise. Therefore, we are not requiring a source emitting less than 100 tpy to meet the 95 percent capture and control level.

We examined whether or not we could specify some other level of control for small sources, but we could not determine what would be an appropriate level of capture and control below 95 percent. We also considered basing new source MACT floors for facilities that emit less then 100 tpy on the single best facility that incorporated pollution prevention. However, as discussed in the preamble of the proposed rule, we believed that using one facility that had the lowest HAP content resins and gel coats was unworkable, unless we could show that all new plants would build the same products as the plants that had the lowest HAP content resins and gel coats.

Given this, we had to determine a threshold value above which 95 percent capture and control is feasible for all new plants, given the diversity of this industry. We selected 100 tpy of actual HAP emissions because above this level facilities tend to operate more hours per year and are better equipped to maintain capture and control systems. Also, at the time we proposed the rule, the smallest facility in the open molding process/product grouping that was permitted at 95 percent capture and control emitted approximately 100 tpy. Therefore, we chose this number as the threshold at which 95 percent capture and control is required.

This was not the only approach we could have taken to subcategorize new sources, nor is 100 tpy the only threshold we could have chosen. For example, we could have subcategorized by annual hours of operation. However, depending on the threshold we set, this could result in large, new HAP emissions sources avoiding the 95 percent capture and control

requirements simply by building a larger facility and reducing hours of operation. By tying the requirement directly to HAP emissions, we believe we have taken the most logical approach from an environmental standpoint and an enforcement standpoint. Also, the 100 tpy threshold is a reasonable choice that means that all new large facilities in most of the process/product groupings will have to meet the most stringent HAP emissions control levels.

Comment:

We received numerous comments on the new source MACT floor for facilities with open molding, pultrusion, SMC manufacturing, mixing, and BMC manufacturing that emit 100 tpy or more of HAP from these operations. The commenters stated that the 95 percent capture and control requirements of the floor were technically infeasible and too costly. They also stated that 95 percent capture and control does not represent the best HAP emissions control approach when all environmental impacts, such as increases in emissions of criteria pollutant and greenhouse gases, are considered. The commenters note that the CAA states that the best controlled similar source must be the basis of the new source MACT floor; therefore, EPA is only authorized to apply the 95 percent capture and control requirements to facilities that are similar. The sources cited by EPA make uniformly-sized parts in high volume, employ mechanical resin application, and operate three shifts a day. However, they differ from other facilities in the industry. One of the sources is primarily a metal fabrication operation and sends significant amounts of emissions from a painting operation to the control device, making an unusually rich combustion stream. They also claimed that the facility had not been proven to meet the requirements of EPA Method 204 of appendix M of 40 CFR part 51. The other facility employs an unusual molding operation, and the ability of this facility to actually meet the 95 percent capture and control requirement is open to question. Neither of these sources are similar to any other composites open molding operation.

Response:

Our available information continues to support that the appropriate new source floor for facilities that emit 100 tpy or more of combined HAP from their open molding, pultrusion, SMC manufacturing, BMC manufacturing, mixing, centrifugal casting, continuous lamination, and continuous casting operations is 95 percent capture and control for several reasons. First, the term “best control” means best control of HAP emissions. The only other control techniques mentioned by the commenters were the pollution-prevention techniques that make up the existing source floors. The commenters claim that when other environmental impacts of add-on controls are considered, pollution-prevention control techniques are actually superior. They provided examples that showed HAP emissions reductions from pollution-prevention techniques for some facilities of up to approximately 70 percent; however, the actual HAP emissions reductions a facility will achieve based on pollution-prevention techniques will be highly site specific. Also, the highest pollution-prevention HAP emissions reduction examples assume facilities could reduce HAP emissions by enhanced process monitoring, which would reduce materials used. The HAP emissions reductions based on materials-use reductions assumes facilities are not currently using materials as efficiently as they could. There are no data to support this assumption, and the potential for HAP emissions reduction of this type could vary widely. The second example presented by one commenter assumes facilities would use nonatomized gel coat application. However, the same commenter has stated emphatically that nonatomized gel coat application cannot be used at every facility. Therefore, this example cannot be considered to fairly represent the HAP emissions reductions achievable for the industry as a whole.

Our overall estimate of the HAP emissions reduction that would occur with only pollution-prevention techniques is approximately 41 percent for open molding, compared to the significantly higher 95 percent HAP emissions reductions possible with capture and control. The CAA indicates that “best control” in the context of setting floors is the control that achieves the best HAP emissions reduction. Based on this, 95 percent capture and control represents best control for this industry.

Even if we were to consider other environmental impacts of capture and control, 95 percent control would still be considered best control. Calculations provided by one commenter indicates that a total of only 0.15 tons of criteria pollutants are generated per ton of styrene reduction; however, this number appears to be based on one of the three actual operating facilities using add-on controls shown in the commenter's example. Data from another facility using a concentrator/oxidizer system in the same report showed criteria pollutant emissions of 0.06 tons per ton of styrene emissions reduction. Our estimate at proposal was that, on average, this figure is closer to 0.04 tons of criteria pollutants per ton of HAP emissions reduction. Regardless of which number is used, the amount of HAP emissions reduction is significantly higher than any resulting criteria pollutant emissions. The commenters also cite greenhouse gas effects. They state that 30 tons of greenhouse gases are produced for every ton of styrene emissions reduction.

We reviewed the information that formed the basis of the estimate of greenhouse gas estimates. Based on our analysis, we believe that the estimate of 30 tons of greenhouse gases are produced for every ton of styrene emissions reduction is an overestimate because it is based on examples where the HAP emissions reduction varies between 77 to 84 percent. The final rule will require 95 percent HAP emissions reduction. Also, we believe the air flows used in the examples provided by the commenter are higher than will be required for new facilities. Higher air flows result in increased use of natural gas and higher greenhouse gas emissions. We believe a more accurate number would be approximately 20 tons of greenhouse gases produced for every ton of styrene emissions reduction.

Second, regardless of which number is the most accurate, any contribution of the final rule to global greenhouse gas emissions is insignificant. The total greenhouse gas emissions in the United States exceed 6 trillion tons from fossil fuel combustion alone. However, the difference between emissions of styrene from a facility controlled to the 95 percent level and one controlled using only pollution prevention is significant to the populations living near an affected facility.

The commenters also stated that the facilities that formed the basis of the new source floor are not “similar sources.” We disagree because there are actually three sources within this source category that meet the criteria to set a 95 percent capture and control floor. The commenters point out that three is a small number compared to the 433 facilities in the database at proposal. However, the CAA requires the new source floor to be based on the single best performing similar source. Therefore, only one source is sufficient to set a new source floor as long as we determine it is similar. The commenters stated that the source setting the floors operates three shifts (they shut down on weekends). However, we subcategorized new sources by annual HAP emissions. The reason was that larger sources are more likely to operate more than one shift. Also, since this floor only applies

to new sources, the facility can be designed to meet the necessary production rate with three shift operation if the operator desires to minimize control device startups and shutdowns.

The commenters stated that in two cases, the floor facilities have collocated surface coating operations. Our evaluation of these facilities was based only on the reinforced plastic composites portion of the facility. During site visits to these facilities, we determined that these facilities were required to apply 95 percent capture and control to all major processes due to State regulations. That requirement would apply regardless of whether or not the facility had collocated surface coating operations. Also, the presence of the surface coating operations does not result in a more concentrated exhaust stream compared to facilities without surface coating operations. Thus, there is no technical basis to say these facilities are not similar based on the presence of surface coating operations.

We also reviewed the commenters claim that the facilities that set the new source floor do not actually meet the requirements of EPA Method 204 of appendix M of 40 CFR part 51. Part of that claim was based on the fact that the floor facilities had doors in the PTE that were opened to move parts and materials in and out of the PTE.

One criteria of EPA Method 204 of appendix M of 40 CFR part 51 is as follows: “All access doors and windows that are not treated as natural draft openings shall be closed during routine operation of the process”. This criteria is not intended to require that these doors be closed at all times. It means that doors must be closed any time that you are not actually moving parts or equipment through them. Therefore, the fact that the floor facilities open doors to move parts in and out of the PTE does not mean they do not meet the requirements of EPA Method 204.

In addition, we reviewed the compliance determinations for two of the floor facilities. Our review did not reveal any conditions that would indicate that the requirements of EPA Method 204 of appendix M of 40 CFR part 51 are not being met.

Comment:

The commenters stated that the facilities that manufacture large parts using open molding or pultrusion are not similar to the floor facilities that are the basis of the capture and control requirements for the new source floors. They stated that the facilities used to set the 95 percent capture and control requirement only manufacture small parts and, therefore, should not be used to set a capture and control floor requirement for facilities making large parts. They also stated that achieving 100 percent capture is not feasible for large parts sources in these process groups. Though EPA had cited facilities that coated large parts in permanent total enclosures (PTE), coating operations cannot be considered similar to the manufacture of reinforced plastic composites. They suggested that any part with any dimension that exceeds 12 feet be considered a large part and be exempt from capture and control requirements.

Response:

After reviewing the comments and available data, we have determined that the facilities currently achieving 95 percent capture and control are not similar to sources producing large parts. At proposal, we noted that we had not identified any facilities in the reinforced plastic composites industry where processes producing large parts, such as storage tanks and swimming pools, have applied 100 percent efficient capture systems, but stated our belief that such PTE were technically feasible based on large PTE in other industries. We reviewed available data on the facilities in our database and found that facilities producing parts over a certain size presented different technical issues from facilities that have successfully incorporated 95 percent capture and control. As noted in the preamble to the proposed rule, one of these facilities has a PTE large enough to produce large parts. However, the air flows and HAP concentrations exiting the PTE at this facility are not the same as would be expected from a facility using a similar sized PTE to capture and control emissions from large parts production.

We also noted in the preamble to the proposed rule that surface coating operations for very large parts (as large as ocean going ships) had successfully applied PTE. However, we agree that coating operations and reinforced plastic composites operations are not similar sources. Reinforced plastic composites production typically requires more workers per part due to the necessity to both apply and roll-out the resin. Also, large parts are continuously laminated until completion rather than coated in sections.

This difference in sources, while applicable to evaluating floors based on capture and control, does not exist in the case of floors based on pollution-prevention technologies such as the use of low-HAP materials and nonatomized resin application. For that reason, we did not differentiate between large and small parts when setting floors based on pollution-prevention control techniques for either new or existing sources.

Because we determined that capture and control was not the appropriate floor for large parts manufacture, the floors for these specific operations are now the same as the floors for existing operation, which are emission limits based on the use of low-HAP materials and nonatomized resin application.

However, we do not agree with the commenter's suggested definition of large parts, because it would exempt parts from capture and control requirements where those requirements have already been demonstrated. The largest part produced at a facility where 95 percent capture and control is demonstrated has a volume of 250 cubic feet. If this part were placed in a rectangular six-sided box, the largest side of the box would be 50 square feet. Therefore we chose these criteria as the definition of a large part for open molding. For pultrusion, the largest part produced by a facility with 95 percent capture and control was 2 inches high, 10 inches wide, and had approximately 350 reinforcements. Therefore, we choose these criteria as the definition for large pultruded parts.

Comment:

Several commenters stated that capture and control requirements would make it difficult for facilities to meet Occupational Safety and Health Administration (OSHA) worker health and safety requirements. Process enclosures at current facilities are designed and operated to provide safe and efficient production of composite products. The primary purpose of enclosures in this industry is to remove contaminated air from the workplace to achieve OSHA requirements for limiting occupational exposures. Enclosures must also allow enough cool air to enter the workplace so that workers are not subject to excessive heat stress. One commenter provided a study that stated that if process enclosure exhaust flows were reduced to increase exhaust concentrations being routed to the control device, worker exposure to contaminants and heat would be increased to unacceptable levels.

Response:

The use of PTE for capture of HAP emissions should not result in increased worker exposure to contaminants or heat stress if appropriate precautions are taken. As previously noted, one solution is to design the spray enclosures based on meeting worker exposure requirements, and then enclosing the entire lamination area in a PTE. The facilities currently using PTE do not exceed OSHA exposure guidelines. Experience in the printing and publishing industry shows that use of PTE, in many cases, results

in reduced worker exposure to both contaminants and heat stress. In high heat and humidity areas, it is likely that some type of air cooling will be required during summer. However, this issue is present even without the requirement for capture and control.

Comment:

Five commenters stated that the limit of tooling gel coats (38 percent HAP) is not achievable. One commenter claimed that we set this limit based on one infrequently used product that is not representative of the industry as a whole. The commenter's products represent 70 percent of the tooling gel coat market and the maximum HAP contents range from 42 to 50 percent HAP. Their lower HAP gel coat has not gained a significant market acceptance. They have performed 2 years of research and development efforts aimed at developing a lower-HAP gel coat that would meet the requirements of the proposed rule and have been unsuccessful. They stated we had not independently tested the product on which the standard is based, so there has been no demonstration of the product's quality or suitability for broad use in the industry. The commenter also stated that setting the standard at 38 percent would have the effect of encouraging manufacturers of tooling gel coats to use para methyl styrene, which is not regulated as a HAP, as a substitute. Also, lower-HAP gel coats may be less durable than products currently on the market, which would result in reduced mold life. Therefore, more molds would have to be built to produce the same amount of product. This would result in the standard actually causing a HAP emissions increase. This commenter requested a tooling gel coat HAP limit of 52 percent HAP for clear gel coats and 49 percent for pigmented gel coats.

A second commenter asked that EPA consider tooling gel coats as speciality gel coats exempt from HAP limits similar to the speciality coating exemption contained in the Aerospace Coating MACT standards (40 CFR part 63, subpart GG). This commenter stated there is a strong possibility they will discontinue manufacturing tooling gel coats if the standard is not changed.

Another commenter stated that we must allow higher HAP limits for tooling applications in vacuum resin infusion, compression, and resin transfer molding composite tool applications, where high exotherms and heated tools are required. Durability of the mold surface is essential to the longevity of the mold. Thermal stability is a key element that requires higher-HAP content. Repeated high exotherms during the cure cycles can greatly reduce the life of low-HAP gel coats. Greater porosity found in the low-HAP materials can also create mold surface problems. Ironically, these are closed molding processes, which result in much lower HAP emissions and employee exposures than open molding processes. Closed molding facilities will not be able to offset the small amounts of high-HAP tooling gel coat used in tool production with large amounts of low-HAP general purpose open molding resins using facility averaging. The commenter recommends that the final MACT standards allow up to 48 percent HAP content for pigmented tooling gel coats.

Response:

We have received additional data since proposal. Based on these data, we increased the floor for tooling gel coats to 40 percent. We obtained very little data from industry on tooling gel coats in the original data requests and in additional efforts to obtain additional tooling gel coat data. To supplement the limited data, we looked at the tooling gel coat data used in developing the Boat Manufacturing MACT (40 CFR part 63, subpart VVVV). This is a reasonable approach because gel coat manufacturers stated that they sold the same tooling gel coats in both the reinforced plastic composites and boat manufacturing industries. The revised HAP content limit of 40 percent is the same as the Boat Manufacturing NESHAP HAP content limit for tooling gel coats.

We considered the issue raised by the commenters that a low limit in tooling gel coats would actually increase HAP emissions. While we agree that more frequent replacement of inferior molds would lead to increased HAP emissions, the factual data do not indicate that a 40 percent HAP content limit results in inferior molds. Facilities in the field (both reinforced plastic composite manufacturers and boat manufacturers) are building molds with 40 percent HAP tooling gel coat. We have no data to indicate that these facilities are producing lower quality molds than average, and none of the commenters has been able to provide objective data to substantiate that reduced mold life is inevitable with low-HAP gel coats. The information provided was based on assumed reduction in mold life. Also, the fact that one of the commenters covers 70 percent of the market is irrelevant, because MACT floors are set based on best control, not market share. In the absence of objective data that the facilities that use low-HAP tooling gel coats produce inferior molds with shorter mold lives compared to the rest of the industry, the MACT floor must be set based on the best performing facilities. In this case, that results in a floor of 40 percent HAP.

Comment:

One commenter stated that although clear cultured marble gel coats have been formulated with HAP levels as low as 40 percent, the tolerance for thermal shock and water resistance are lowered with lower-HAP levels. According to the commenter, 48 percent HAP clear coat is required for manufacturers to maintain current warranties and many have switched back to the high-HAP clear gel coats due to the poor performance of the lower-HAP clear gel coats. The commenter suggests that “clear gel coats for cultured marble” should be defined as “those used for products subject to ANSI Z124 testing” and the rule should limit the HAP content of these materials to 48 percent. A second commenter also stated that a 48 percent HAP content is necessary to meet desired gel coat performance. The commenter claims that the proposed limit of 44 percent does not take into account the entire spectrum of uses and does not satisfy the requirements of their applications.

Response:

We are bound by the statutory requirements of the CAA to set MACT floors based on the average of the best performing sources as illustrated in the available data. In the absence of specific data to support the request, we have no basis to change the floor.

In developing different process product grouping for gel coats, we did consider the different performance characteristics of different types of gel coat. These types were tooling gel coat, clear gel coat, pigmented gel coat (white/off white), pigmented gel coat (all colors except white/off white), fire retardant gel coat, and corrosion resistant/high strength gel coat. Based on information provided by industry, we determined that these different gel coat types had sufficiently different characteristics that they should be considered separately for floor determinations. However, we do not have data to demonstrate that it would be appropriate to further subcategorize clear gel coats based on each gel coat's performance characteristics.

Comment:

One commenter states that only the white/off-white and some pastels can meet a floor of 30 percent HAP because of the titanium dioxide and inert filler loading. Most solid colors require a HAP content of 38 to 40 percent. Higher performance pigmented gel coats that require high molecular weights would, therefore, need a higher monomer content to achieve workable viscosities and would probably no longer be available to the market place.

Response:

White/off-white gel coats will be defined as those containing 10

percent or more by weight titanium dioxide. As proposed, these gel coats will be subject to a HAP limit of 30 percent by weight, and all other pigmented gel coats will be subject to a HAP limit of 37 percent by weight.

At the time we developed the proposed rule, we had no data on pigmented gel coats other than white/off-white and some reds. Based on industry comments, we split pigmented gel coat into two groupings, white/off-white and other colors due to the fact that white/off-white gel coats contain titanium dioxide, which is a heavy pigment, while other colors do not. At the time we created this new grouping, we requested data from the industry concerning the HAP contents of pigmented gel coats. The industry representatives indicated that these gel coats typically have 37 percent HAP. Because non-white pigmented gel coats comprise a very small part of the total industry, we elected to accept the 37 percent number rather than attempt to gather additional data. The commenter provided no data to support their request. In the absence of new data, we have no basis to change this floor.

Comment:

Two commenters request that the category of fire retardant gel coats be exempt from HAP limits. Both commenters note that fire retardant gel coats are used in manufacturing transportation parts, building products, trains, airplane parts, and theaters. One commenter stated that these are all critical areas of applications and require various Underwriter Laboratory (UL), American Society for Testing and Materials (ASTM), and Fire Rating Certifications. It was suggested that fire retardant gel coats be defined as “those used for products for which low-flame/low-smoke resin is used.”

Response:

We have added a process/product grouping for fire retardant gel coats. These gel coats are defined as gel coats used in low-flame spread/low-smoke product applications. We have established a HAP emissions limit of 854 lb/ton which is equivalent to gel coats with a maximum HAP content of 60 percent using atomized application.

Comment:

Four commenters stated that we need to establish a separate process/product grouping for corrosion-resistant gel coats. The commenters stated that gel coats used in specific corrosion protection applications must meet the same requirements as corrosion-resistant resin. One commenter added that gel coats requiring chemical resistance to a wide range of chemicals including acids, bases, and solvents are often based on the resins similar to those that make up the structural part of the composite and provide the necessary corrosion resistance. For this reason, the commenters believe that the HAP limitation for corrosion-resistant gel coats should be 48 percent, the same as it was in the proposed rule for lamination resins used to make corrosion-resistant composites. It was suggested that “corrosion-resistant gel coats” be defined as “those used for products made with corrosion-resistant resin” and that the rule limit the HAP content of these materials to 48 percent.

Response:

We agree that there are technical limitations for corrosion-resistant applications that warrant a separate limit for corrosion-resistant gel coats, similar to the separate limits established for other specialty resins and coatings.

In the final rule, we established a separate HAP content limit of 48 percent for corrosion-resistant gel coats and defined them as “those gel coats used to manufacture products made from corrosion-resistant resin.” We believe 48 percent HAP is the appropriate number because the highest HAP content level allowed in all the corrosion-resistant resin process/product groupings is 48 percent.

Comment:

Several commenters stated that we need an additional process/product grouping for low-shrink resins. These resins have special shrinkage control properties that are unique and cannot be obtained in any other way. These resins were not identified when EPA surveyed the industry. One commenter stated that a specialty process group is needed for high molecular weight, low-shrink resins used in wind turbine blade manufacturing. The resin currently in use is 42 percent HAP unfilled. The facility would be unable to gain any relief by facility averaging because the facility predominantly uses zero-HAP epoxy resin, rather than a low-HAP production resin. Commenters requested that EPA create a subcategory for these resins with a maximum HAP level of 48 to 52 percent.

Response:

Our understanding is that these low-shrink resins are highly filled resins with special chemistry that allows them to cure at room temperature with significantly less shrinkage than a typical resin. Given the unique properties of this resin, we agree that a separate process/product grouping is appropriate. The resin manufacturer indicated that the maximum HAP content of the resin is 50 percent. Therefore, we have set HAP emissions limits for shrinkage-controlled resins that allow up to 50 percent HAP. This specialty resin costs significantly more than other resins, which provides a deterrent for facilities using the resin where its special properties are not necessary.

Comment:

One commenter believes higher HAP limits are needed for the filament application of corrosion-resistant products. The commenter claims that the rule, as proposed, will eliminate use of certain types of corrosion-resistant resins that impart required properties to certain applications. The commenter noted that the proposed limit for corrosion-resistant filament application resins was lower than for noncorrosion-resistant filament application resins. The commenter believes that the HAP emissions limit for all categories of filament application should be 178 lb/ton, and stated that this change will have insignificant impact on EPA's total HAP emissions reductions target, with the difference in HAP emissions reductions being 3 tpy.

Another commenter states that the proposed MACT of 42 percent HAP cannot be met with an isophthalic resin without some compromise to the physical properties of the cured resin. The commenter requested EPA to consider the 48 percent HAP limit found in South Coast Air Quality Management District (SCAQMD) Rule 1162.

Response:

While we acknowledge the commenters concerns, we developed the floor for this process/product grouping in the same manner as floors for other process/product groupings in open molding. We gathered data from industry and ranked the performance of the facilities in the corrosion-resistant process group and set the MACT floor based on the average of the best 12 percent, as required by law.

Though we are not changing the floor for filament application, we are retaining a provision included in the proposed rule that allows facilities to use the same resin in multiple processes. The rationale for this provision is, while our floor development ranking procedure is correct, we also realize it does not account for the fact that some facilities use multiple operations to produce components of the final product, and the resins used in the subcomponents must be compatible. This provision will allow most facilities the flexibility to use the necessary level of HAP in corrosion-resistant applications because mechanical operations have a higher-HAP content limit.

Comment:

One commenter recommended that the model point value for corrosion-resistant manual resin application be changed from 124 to 190 to reflect the use of the same

percent HAP used in mechanical resin application. The commenter notes that the facility that sets the floor using a 40 percent HAP resin is not typical of a true corrosion-resistant (CR) company because that facility uses only manual application, while true CR companies use both manual and mechanical application techniques. A second commenter requested that the MACT floor for manual corrosion-resistant resin be changed so that it is the same as the floor for mechanical corrosion-resistant resin.

Response:

As discussed in the previous response, the floor is based on the data available for this process/product grouping. However, as with filament application, the provision allowing facilities to use the same resin in multiple operations should allow enough flexibility for facilities to meet rule requirements, but still produce products with the necessary properties. Therefore, facilities that produce corrosion-resistant and noncorrosion-resistant products using both manual and mechanical resin application will be able to use the same resin in both operations.

Comment:

One commenter stated that the proposed MACT of 35.5 percent HAP for noncorrosion-resistant centrifugal casting would result in a resin too high in viscosity, which may create air release problems. The commenter states that lower molecular weight resins would cause some limitations in physical property requirements.

Response:

We received new data that changed the floor for centrifugal casting to 37.5 percent HAP. With less than 30 facilities in the process group for which we have data, the MACT floor must represent the average performance of the top five facilities. We have no data to support raising the floor any further.

Comment:

One commenter stated that they believe that new operations should be subject to new source MACT even if they are added to an existing source. The commenter understands that there are cases in which the new equipment may be incorporated within an existing manufacturing line, making it difficult to employ separate controls (

e.g.

, if all the equipment is controlled at a later end point). The commenter suggests, however, that separate and more specific provisions can be included in the rule to govern such cases.

Response:

This comment is only applicable to new source MACT for specified processes that emit over 100 tpy, because below that level, new source and existing source MACT are the same. We believe that, for this particular industry, the ability of a facility to incorporate the capture and control requirements of new source MACT for larger facilities is closely related to the structure housing the process, because the size and shape of the existing building affects the layout of the production line. Even if there are significant process changes, this by itself would not indicate that the building housing the process has been changed, thereby making retrofit of capture and control systems unfairly difficult compared to a new greenfield facility. We believe that attempting to develop a detailed set of requirements that could cover every situation would be unrealistic.

We agree that this provision may result in small facilities being able to grow significantly without becoming new sources. However, it should be noted that in the final rule, we have overridden the portion of the general provisions in 40 CFR part 63 which states that facilities that move are still considered existing. Because we believe the cost and technical feasibility of capture and control are closely related to the building housing the process, we believe that a facility that moves should be considered a new source because they can plan for capture and control prior to erecting or selecting a new building. Therefore, facilities that would be considered existing sources under the general provisions will be considered to be new sources under the final rule. Therefore, in this aspect, the final NESHAP are more stringent.

Comment:

Several commenters requested clarification in this rule on which operations at a reinforced plastics composites facility and which operations at a boat building facility will be covered by this rule and which will be covered by 40 CFR part 63, subpart VVVV (Boat Manufacturing NESHAP).

It was noted that neither the preamble nor the proposed rule explicitly states whether this rule applies to manufacturing of boats or boat components and requested that language be added to the final rule clarifying that this rule does not apply to any processes or operations subject to 40 CFR part 63, subpart VVVV. One commenter stated that boat building plants routinely produce non-boat parts and presumed that such facilities will be required to meet the composites rule when producing composite parts that are not associated with the manufacture of boats. The commenter also points out that some composite plants produce boat parts that are then used to build boats, such as when producing barge covers that are related to the manufacture of river barges.

Response:

We have added § 63.5787 to the final rule to specifically address this issue. A facility must produce boat hulls and decks, or molds for boat hulls and decks, to be covered by the Boat Manufacturing NESHAP (40 CFR part 63, subpart VVVV). If it produces reinforced plastic composites, as defined in the final rule, and is not covered by the Boat Manufacturing NESHAP, then it is covered by the Reinforced Plastic Composites NESHAP, regardless of the final use of the parts.

In the case where a facility is subject to the Boat Manufacturing NESHAP (40 CFR part 63, subpart VVVV), but the facility also makes parts that are not a component of their boats, then the non-boat parts are covered by the Reinforced Plastic Composites NESHAP. However, only resins and gel coats actually used to make parts covered by the Reinforced Plastic Composites NESHAP are considered in determining compliance.

In addition, in order to simplify compliance, we are allowing facilities that are subject to the Boat Manufacturing NESHAP (40 CFR part 63, subpart VVVV) and that also make parts subject to the Reinforced Plastic Composites NESHAP, to elect to make all their manufacturing operations subject to the Boat Manufacturing NESHAP if they can demonstrate, through the appropriate HAP emissions calculations, that this will not result in any HAP emissions increases over what would occur if they complied with the Reinforced Plastic Composites NESHAP for non-boat part production. We also clarify that HAP emissions from activities covered by the Boat Manufacturing NESHAP are not considered when calculating HAP emissions thresholds to determine the applicability of add-on controls.

Comment:

One commenter requested that the rule explain what happens in instances where the 100 tpy threshold is exceeded even by a little, temporarily. Does this require that add-on controls be installed?

Response:

It is our intent that unusual circumstances result in a facility having to add and operate add-on controls. We have included clarifying language in the final rule that allows a one-time exemption to the 95 percent capture and control requirements for facilities that were below the 100 tpy threshold and exceed the threshold due to unusual circumstances. This exemption allows facilities to average annual HAP emissions over 3 years to determine if they exceed the threshold. However, facilities are also required to document the unusual circumstances that caused the exceedance, and why they expect to remain below the threshold in the

future. If they exceed the threshold a second time, then the exemption is immediately withdrawn and they must comply with the 95 percent capture and control requirements within 3 years from when they originally exceeded the threshold.

Comment:

Two commenters requested clarification of several issues related to repair work. They are assuming the proposed rule is intended to cover manufacturing operations only. Repair processes conducted in a manufacturing facility are also covered because they are likely to use the same materials. If the processes conducted are re-manufacturing, refurbishment, repair, or maintenance, it will be considered repair for the final NESHAP. The exception would be if the repair is a part which frequently needs replacement and is made in an assembly-line type process. They also asked that since there is no de-minimums level, if any manufacturing is done, would it be covered? They noted that at some of the commenter's facilities, some minor manufacturing may occur. The repair work that may also be done at the same facility is not related to the manufacturing processes (and would be using different resin and reinforcing materials.) The commenter believes that as the rule is currently written, both the manufacturing and repair operations would be covered. The commenter does not believe that is EPA's intent and asked if we could develop language to correct that.

One commenter stated that definitions of repair and manufacturing should be added to clarify the types of repair and manufacturing covered by the rule. The preamble and rule should be consistent in stating that the facilities that only repair composites are not affected. The commenter also feels that repair operations collocated with unrelated manufacturing operations should not be covered either. In a related comment, several commenters asked that a low-use cutoff be established so that facilities that use small amounts of resin and gel coat are not subject to the rule, especially since those uses may be incidental to a completely different manufacturing operation.

Response:

The final rule has been written to make explicit what repair operations are and are not covered. Specifically, facilities at which only repair occurs are not covered by the final rule. In addition, repair of previously manufactured reinforced plastic composites unrelated to the reinforced plastic composites manufactured at the facility are also not covered by the final rule. Repair processes on parts that are manufactured at the same location are covered by the final rule. In addition, we have added a low-use cutoff exemption to the final rule. We reviewed our entire database and determined that we had no data for facilities that use less than 1.2 tpy of resin and gel coat combined. Therefore, we believe that, in the absence of any available data, facilities that use less than 1.2 tpy of resin and gel coat to produce reinforced plastic composite products or components should be exempt from the final rule.

Comment:

Many commenters requested that the rule incorporate an exemption for R&D facilities, and for R&D operations collocated with manufacturing operations. The materials used in R&D operation may be significantly different from those used in manufacturing.

Response:

We have written the final rule to exempt R&D facilities and R&D operations. The definition of R&D is the same as contained in section 112(c)(7) of the CAA.

Comment:

Several commenters stated that they believe the EPA cannot set different standards for small and large businesses based on the size of the business, rather than the size of the source. They believe that because the CAA clearly identifies “major source” by the level of HAP emissions, MACT floors must depend on the average HAP emissions reductions by the best sources without regard to cost factors of business size. They stated that this distinction was unfair because two facilities that emit the same amount of HAP would potentially have different requirements solely on the basis of their ownership. The commenter also believes that EPA did not adequately support the determination that large businesses have better access to capital than small businesses. They stated that this is not necessarily true.

Response:

Based on the revised cost analysis, we have determined that it is no longer necessary to distinguish between small and large businesses. However, we still believe the use of different thresholds in the proposed rule was appropriate because this distinction only applied to the above-the-floor regulatory option. The CAA specifically states that when we go above the floor, we must consider costs.

Comment:

One commenter states that the small business threshold of 250 tpy should apply to both existing and new sources. New capital funding to build a new facility would require due diligence on the part of the lending institution. The new facility would have to generate enough cash flow to meet the added debt load. Adding a capture and control system to the debt load would significantly reduce the cash flow available to pay back the lender's note on a new facility because the capture and control system is a non-value added asset. The lending institution would discern this and deny the loan.

Response:

For new sources, the proposed (and final standard) is the MACT floor, not an above-the-floor option. We do not have the flexibility to create small and large business distinctions when the standard is set at the MACT floor. Therefore, the final rule for new sources does not incorporate a small and large business distinction.

Comment:

Several commenters stated that a method to establish percent reduction and HAP emissions factors is needed to foster the development of new products and equipment to serve the affected industry. They recommended that EPA establish a protocol to allow the smooth introduction of equipment, products, and other technologies into the final rule.

Response:

Allowing facilities to use site-specific HAP emissions factors, and the procedure in the general provisions in 40 CFR part 63, subpart A, that allows facilities to demonstrate equivalent HAP emissions reductions, adequately address the incorporation of new HAP emissions reduction technologies. However, we have added § 63.5798 to the final rule that discusses how to obtain approval for new technologies.

Comment:

Two commenters requested that EPA change the averaging provisions to allow a facility that changes some processes to non-styrene containing resins to average these resins with the styrene-containing resins to demonstrate compliance.

Response:

We do not believe it would be appropriate to allow the use of non-styrene containing resins and gel coats to be included in the calculation of compliance. The MACT floors were developed only considering resins and gel coats that contain styrene (and other organic HAP, such as MMA) used at the facilities in our database. We did not consider non-styrene resins and gel coats used at our database facilities. Given the basis for developing the standards, it is inconsistent to allow non-styrene containing resins and gel coats to be used in the compliance calculations. Therefore, we have not added this request to the final rule.

Comment:

Numerous commenters recommended replacing the point value system with the Composite Manufacturers Association (CFA) UEF

table. The composites industry is already using these HAP emissions factors to calculate annual HAP emissions. It would simplify reporting and recordkeeping if one set of HAP emissions factor equations were used. Another commenter stated that if EPA uses the UEF, all HAP should be treated as styrene because this is how EPA developed the MACT floors. They noted that these factors are used by industry for toxic release inventory reporting and obtaining permits. According to the commenters, use of these factors for MACT will reduce the paperwork burden for small manufacturers.

Response:

We reviewed the UEF and the basis for their development. Based on this review, we believe that these equations are acceptable for estimating both HAP emissions factors for compliance purposes and HAP emissions.

As a result, in the final rule we have written the HAP emissions factor equations in Table 1 to subpart WWWW of part 63 to be identical to their equivalent UEF equations. Therefore, facilities will have one set of identical factors for both compliance and HAP emissions estimation purposes.

Because of this change, it was necessary to recalculate the floor values by recalculating HAP emissions factors using the new HAP emissions factor equations for the facilities in our database and reranking the facilities based on the new calculations. Therefore, both the numerical values of the floors (lb/ton) and the equations used to calculate compliance changed. Note the floors themselves did not change significantly because when we reranked facilities using the new HAP emissions factors, the ranking order did not change with two exceptions. In those cases, the new equations caused two facilities to switch places and changed the floor slightly. However, these changes were minor compared to the changes that resulted from other comments we received and additional data we gathered.

In addition, we have added to the final rule equations for the nonatomized gel coat application and for the mechanical atomized controlled spraying of resins. We have incorporated the latter UEF equation in the final rule so that it is applicable only where the controlled spray is achieved through automated or robotic, not manual, spraying.

Finally, we are incorporating only the UEF equations developed for styrene and not those developed for MMA. We are doing this because the data analysis forming the basis of the standards assumed all organic HAP to be styrene. This is a reasonable assumption as the amount of MMA used is a very small percentage of the total HAP monomer used.

Comment:

One commenter noted that the proposed rule does not provide for manual application of gel coats. Many gel coats are applied manually as exterior coatings when the major component part is made. The rule should require that for HAP emissions calculations from manual application, gel coat should be considered as a resin with the stated HAP content and the appropriate point value equation should be used. Companies where manual gel coat application is less than 2 percent of the total gel coat usage should be exempt from maintaining records of manual application.

Response:

We agree with the commenter that the proposed rule did not provide an equation to estimate HAP emissions from the manual application of gel coats and that the rule needs to address this. In the final rule, we have addressed this issue by allowing two options. First, the facility may elect to simply include manually-applied gel coat with spray gel coat application for compliance and HAP emissions estimation purposes. Alternatively, they can elect to treat the gel coat as spray for compliance purposes, but use the manual resin application HAP emissions factor to estimate HAP emissions.

We believe the changes discussed above are sufficient to simplify reporting and recordkeeping for manual gel coat application. Therefore, we have not added an exemption for maintaining records for manual gel coat application.

Comment:

Several commenters requested that sources be allowed to use HAP emissions factors in approved title V permits to estimate HAP emissions. It was noted that the use of such factors will reduce the administrative burden for sources and regulators and will likely improve HAP emissions estimates. One commenter suggested that such factors also be allowed to be used for compliance determinations.

Response:

We agree with this comment and believe that § 63.5798(a)(1) and (2) of the final rule already allow for the use of facility-specific HAP emissions factors. Section 63.5798(a)(1) states, in part, that “you may use any organic HAP emissions factor approved by us such as factors from the Compilation of Air Pollutant Emissions Factors, Volume I: Stationary Point and Area Sources (AP-42).” Section 63.5798(a)(1) was not intended to limit organic HAP emissions factors only to the AP-42. Paragraph (a)(2) of § 63.5789 allows the development of facility-specific organic HAP emissions factors through performance testing. If a facility has facility-specific factors that have been approved for use in title V operating permits, then those factors can be used to determine whether or not the facility is a major source under section 112 of the CAA. In addition, a facility can use facility-specific factors for comparison against applicable HAP emissions limits. We have written the language in § 63.5798 of the final rule to clarify the use of such facility-specific factors and have added the provision that such factors must be supported by test data.

Comment:

One commenter notes that the alternative point values in Table 5 to subpart WWWW of part 63 do not provide a realistic alternative to 95 percent capture and control. In an example calculation for 35 percent styrene resin in open molding, the point value calculation is equivalent to 96 percent control, which is more stringent than the add-on control requirement.

Response:

While the values may not appear realistic for some facilities, Table 5 to subpart WWWW of part 63 does present the opportunity to meet the final standards using alternative means. We believe that the values in Table 5 to subpart WWWW of part 63 provide incentive to continue to pursue lower-HAP resins and gel coats and other pollution-prevention opportunities and that even if only one facility can use the values, then their inclusion is worthwhile. For these reasons, we have retained Table 5 to subpart WWWW of part 63. However, we have made minor modifications to this table. For process/product groupings where there is an operating facility that currently meets the 95 percent control requirement, we changed the value in Table 5 to subpart WWWW of part 63 to reflect the highest actual calculated HAP emissions factors for operating facilities.

Comment:

One commenter stated that the weighted average point values should be calculated as a weighted average of resin used. The commenter pointed out that the equation in the current proposal gives equal weight to each month instead of each quantity of resin or gel coat processed. Another commenter asked for clarification on how the “weighted average floor” is calculated.

Response:

We agree with the commenter that 12-month rolling average point values should be calculated using a weighted average based on the amount of resins, rather than using an average based on monthly values, as was proposed. Therefore, the final rule incorporates the commenters' suggestion. Also, we have changed the

terminology for the averaging calculations. We now use the term “emissions factor” when discussing values calculated using actual resin and gel coat HAP contents, and “emissions limit” when discussing average values calculated from the required floor limits. This change should clarify how to calculate the weighted average floor.

Comment:

Several commenters asked that EPA include a test method to determine the effectiveness of vapor suppressants. They suggested that the CFA-developed vapor suppressant test method be used.

Response:

We agree with the commenter that the final rule should incorporate a test method applicable to vapor suppressants, which are effective at reducing HAP emissions for many resin applications. The effectiveness of vapor suppressants varies depending on the resin and the application technique used. Thus, a single effectiveness value cannot be assigned. The final rule, therefore, incorporates a test method to determine the effectiveness of vapor suppressants for facility-specific applications. This test method is being published as appendix A to subpart WWWW of 40 CFR part 63.

Comment:

One commenter states that the proposed rule is vague or silent on key issues including continuous monitoring of the preconcentrator control performance. The commenter states that the question of the practical long-term efficiency of the preconcentrator system is particularly disturbing because the proposed rule is silent on the issue of compliance assurance. Unfortunately, compliance assurance will present three problems: no available parametric measure will work to monitor absorber efficiency; continuous or semi-continuous flame ionization detectors (FID) are the only practical alternative, but are unreliable; and automated FID equipment is very expensive and prone to periods of malfunction.

The commenter also states that the only feasible available continuous emissions monitor (CEM) system that can measure styrene is an automated sampling device based on an equivalent FID sensor as described in EPA Method 25A of appendix A-7 to 40 CFR part 60 that has an annual cost of $78,200 per year. The additional cost of this necessary compliance monitoring equipment was not included in the EPA cost analysis.

Response:

We have reviewed the information on those facilities using add-on control devices with carbon adsorbers within the reinforced plastic composites industry and have found none that are using FID. These facilities are able to demonstrate compliance with 95 percent reduction. Therefore, we do not believe it is necessary to require use of FID under the final rule and have not included the cost of such devices in our cost analysis.

Comment:

One commenter notes that the requirements for sources to determine the HAP content should be the same as those in the Boat Manufacturing NESHAP (40 CFR part 63, subpart VVVV). The commenter points out that the boat rule allows sources to use information from the supplier or manufacturer and requires the use of the upper limit of a range if a range is provided and allows use of supplier information as long as a measured value does not exceed the provided value by more than 2 percentage points. The commenter notes that suppliers provide many of the same resins and gel coats to boat manufacturers and composites manufacturers.

Response:

We agree with the commenter and the final rule has been written in line with the HAP content determination provisions found in the Boat Manufacturing NESHAP, which in part allow up to a plus or minus 2 percent allowance.

Comment:

One commenter stated that the rule should allow composites manufacturers to change compliance options and should provide guidance on notification and record keeping requirements if affected sources need to switch compliance options.

Response:

We agree with the commenter and have included language in the final rule making it clear that changes in the selected compliance option are allowed.

Comment:

One commenter opposed capture and control for pultrusion sources based on worker safety. The commenter notes that the EPA analysis assumes an inlet concentration of 100 ppmv, but their measured concentrations are about 12 ppmv. At that concentration, according to the commenter, capture and control is not viable. The commenter claims that efforts to increase the inlet concentration lead to OSHA and industrial hygiene concerns and that any changes increasing the concentration to over 20 ppmv would exceed American Congress of Governmental Industrial Hygienists recommended maximums. Further, the commenter states that levels approaching 50 ppmv require installation of engineering controls (ventilation or HAP prevention) and exposure to these levels would meet with serious union objections. The commenter notes that these considerations result in higher capture and control costs. The commenter requests that health related issues be fully addressed before the proposed above-the-floor capture and control is implemented. Another commenter stated that pultrusion products requiring constant attention would have to have an enclosure large enough for the operator to be inside, and this would increase health risks due to styrene exposures.

Response:

As noted in previous responses, the above-the-floor requirement for 95 percent HAP emissions reduction is no longer required for pultrusion operations at existing sources and, therefore, capture and control is no longer an issue for existing facilities. We also note that our revised cost analysis now uses a target maximum inlet concentration of 50 ppmv rather than 100 ppmv. The 50 ppmv target is the same as the current OSHA 8-hour time weighted average limit for styrene.

We have not changed our position on capture and control for new sources, except, as discussed above, with respect to large parts made by pultrusion (and open molding) sources. New facilities can be designed with the appropriate measures in place to avoid worker exposure in excess of OSHA requirements. As previously discussed, facilities that have incorporated capture and control meet current OSHA requirements for worker safety.

Comment:

Several commenters requested that preform injection, a technique that applies resin to the reinforcements in a closed box, be an allowed control technology because it is a more viable and readily attainable control technology than either add-on control or direct-die injection. One commenter stated that preform injection should qualify for a 90 percent HAP emissions reduction, and the CFA proposed definition and requirements should be used as the criteria for preform injection. A second commenter stated that although it falls short of 95 percent reduction, reduction rates of 90 percent are attainable and an excellent trade off given the applicability, capital requirements, and operating costs associated with preform injection.

Response:

We agree with the commenters that preform injection is a viable control technology for reducing HAP emissions from pultrusion operations. Preform injection is included in the final rule as an option for meeting the 60 percent HAP emissions reduction requirement for existing pultrusion sources. However, as stated by the commenters, preform injection (and direct die injection) do

not meet the 95 percent HAP emissions reduction requirement, which is the new source MACT floor. The CAA does not allow us to be less stringent than the floor. Therefore, we cannot allow preform injection, or direct die injection, to be a compliance option to meet the 95 percent HAP emissions reduction requirement. We also included a definition for preform injection in the final rule that is based on the commenter's suggested language.

Comment:

Several commenters requested a 12-month averaging period for compliance for pultrusion. The commenters stated that pultruders should be able to use a combination of preform injection, wet area enclosures, direct die injection, and “no control” to meet the 60 percent HAP emissions reduction requirement for existing sources. The commenters pointed out that HAP emissions credits could be earned to offset the processing of products with an open bath and “no control.” According to the commenters, without averaging, facilities will be forced to discontinue manufacturing products that require constant open access (for example, certain complex profiles) or to shut down any processing line when there is an extended period of processing adjustments (which require open access to the line).

Response:

We agree that averaging will add some flexibility for you to comply with the final rule without increasing HAP emissions. Therefore, for existing sources we have included an averaging option. For purposes of averaging, we have assigned wet area enclosures a 60 percent HAP emissions reduction, and direct die injection and preform injection a 90 percent HAP emissions reduction.

Comment:

Several commenters requested that the 60 percent emission reduction requirement for existing sources, which is based in the use of a wet area enclosure, be replaced with a work practice standard of air flow management. They stated it was impossible to apply wet area enclosures to these large parts because of accessibility concerns. Large parts require almost constant access because they are extremely complex. The other control options for existing sources, preform injection and direct die injection, have also not been demonstrated on these large parts. They suggested a definition for large parts, which was parts with 1,000 or more reinforcements and at least a 60 square inch cross sectional area.

Response:

We agree that wet area enclosures, which form the basis of the existing source floor, are not feasible for large parts as defined in the comment. Therefore, we developed a separate existing source MACT floor for large pultruded parts. A review of the available data indicates air flow management (as described in more detail in Table 4 to subpart WWWW of part 63) has been used to control emissions from this process group. Therefore, the existing source MACT floor is air flow management. The final rule has been written to reflect the new floor.

Comment:

Many commenters requested that the limit on wet enclosure open times of 30 minutes per shift be changed to 90 minutes per day to allow for necessary repairs, start-ups, and shutdowns.

Response:

We evaluated the commenters' request. The facilities that actually set the floor for pultrusion are limited to 30 minutes per 8 hour shift or 45 minutes per 12 hour shift. In addition, the facility may average over all pultrusion lines. We have included averaging provisions across lines in the final rule. We have also allowed a facility to have the doors and covers open 90 minutes per day providing the machine is operated three 8-hour shifts or two 12-hour shifts.

Comment:

Three commenters claimed that the height restriction on wet area enclosures is not practical because it does not allow room above the highest part to make adjustments to the process or equipment. According to the commenters, the actual height of the wet area enclosure has no impact on HAP emissions because the puller window is the controlling factor, and styrene emissions will remain near the bath without air flow. The commenters, therefore, requested that the restriction be removed.

Response:

We have no data to suggest that limiting the height of the enclosure affects the amount of HAP emissions reduction. Therefore, we did not include the height restrictions on the wet area enclosures in the final rule.

Comment:

Commenters requested that capture and control not be required for sources engaged in SMC manufacturing. The commenter stated that EPA's proposal for control is based on one source and, according to the commenter, that source has found that they cannot operate the SMC operation and comply with EPA Method 204 of appendix M of 40 CFR part 51.

A second commenter stated that their SMC operation is permitted by Ohio EPA as a PTE with all HAP emissions vented to a thermal oxidizer. They have found it expensive to maintain the PTE and control device and may be required to install additional monitors at great expense.

Response:

For existing sources, the final rule does not require capture and control for SMC manufacturing. For new sources, however, the floor is 95 percent reduction and we do not have the flexibility to change the floor. Most of the comments raised by the commenters relate to the cost of PTE and thermal oxidizers. However, costs may not be considered in setting the floor. Additionally, the problems with compliance noted by one commenter do not, in themselves, indicate that new sources cannot be designed and operated to meet the 95 percent control requirement. For example, the facility states that they must open a large overhead door to operate their second SMC machine. In a new facility, the plant layout can be designed where large doors are not required to be continually open. Therefore, the final rule retains the requirement of 95 percent capture and control for SMC manufacturing at new sources that exceed the 100 tpy of HAP emissions threshold.

Comment:

One commenter noted that an alternative to meeting the 95 percent HAP emissions reduction requirement is provided for some operations and requests that an alternate HAP emissions limit be provided for SMC manufacturing. An alternative HAP emissions limit allows SMC manufacturers to utilize pollution-prevention efforts that have already been implemented and encourages the use of future pollution-prevention efforts.

Response:

For SMC manufacturing, we have incorporated a HAP emissions limit of 2.4 lb/ton as a compliance alternative to the 95 percent control requirement in the final rule.

Comment:

A number of commenters expressed concerns about the floor level of control for SMC manufacturing that is based on several work practices. They stated that the requirement to cover doctor boxes should be deleted because the boxes have to be open for machine operators to monitor paste levels. They also mentioned that folding the edges of the SMC had proved to create problems for some facilities that had tried the practice. Finally, they stated that the requirement to enclose the SMC in nylon film should actually say nylon-containing film.

Response:

We evaluated the basis for the MACT floor by reviewing all of the data available prior to proposal and contained in the public comment letters. On the basis of this review, we discovered that the MACT floor at proposal did not accurately reflect the actual work practices currently being used. Therefore, we changed the floor to

for both new and existing sources to cover or enclose the resin transport system up to the doctor box and use nylon-containing film to enclose the SMC. Based on the practices at the best controlled similar source, these work practices also apply to new sources that are above the 100 tpy threshold.

Comment:

Several commenters requested that the requirement for “no visible gaps in mixer covers” be revised to allow reasonable and necessary openings. In general, they stated that mixing vessels must have some opening or vents to allow air to enter or leave the vessel when materials are added or removed, or when the contained material expands or contracts due to changes in temperature. Commenters also noted necessary clearance for mixing shafts and other instrumentation are essential and suggested allowing a gap of one inch. An additional commenter stated that they have several holding tanks that are continuously agitated to prevent settling. They requested that we add clarifying language to the definition of mixers to exclude tanks that are only agitated to prevent settling.

Response:

Based on our review of the available data on the current industry control on mixing tanks, we found that the proposed rule is more stringent than the floor and that to allow some visible gaps around shafts, etc., is consistent with the data available to set the floor. Therefore, we have written the final rule to allow no more than one inch of visible gap around mixing shafts and any required instrumentation.

With regard to the request to exempt tanks that are agitated only to prevent settling, concern was that the mixing shafts required clearance. Because we have added a provision to allow up to a one inch clearance around the agitator shafts, this concern has been addressed and the suggested exemption for these specific types of mixers is not warranted. Therefore, the final rule does not include an exemption for tanks that agitate only to prevent settling.

Comment:

Several commenters requested that the rule allow active venting under three conditions: when adding filler; when using nitrogen blanketing; and prior to opening a mixer. Several commenters stated that when powders are added to mixing tanks, vent gases are directed to a dust collector to protect employees. One commenter stated that you cannot capture dust without actively venting. The commenter suggests that the proposed rule allow active venting as part of the material addition process. Two commenters actively vent covered mixers at very low flow through a dust collector. The active flow results from nitrogen flowing through the air space for safety reasons (to prevent vapor buildup). Based on stack test results, HAP emissions under these conditions were found to be very low (0.000292 lb styrene/lb styrene available). For these reasons, active venting for safety reasons, using an inert gas purge, and at low flow, should be allowed for covered mixers. Another commenter stated that some mixing operations use nitrogen blanketing for safety (to prevent formation of flammable atmospheres). These sources have an incentive to limit use of nitrogen blanketing because of cost; so, HAP emissions will be negligible. Two commenters also requested that the rule allow venting just before adding materials to clear out vapors prior to opening covers and to allow venting just after adding powders to capture residual dust in the vapor space. One commenter also asked that the term “active venting” be defined in the rule.

Response:

We believe that most HAP emissions that result from mixing operations occur when active mixing is taking place. Also, based on the data used to set the MACT floor, the facilities that responded that mixers have no active venting meant that the mixer was covered and not vented during mixing. As a result, we have written the rule requirement to read “close any mixer vents when actual mixing is occurring, except that venting is allowed during addition of materials, or as necessary prior to adding materials or opening the cover for safety.” Because we have removed the term “active venting,” no definition of this term is required.

Comment:

One commenter believes that covers should be required instead of add-on control for larger mixing operations. According to the commenter, covers can reduce HAP emissions by 84.8 percent to 96 percent. The commenter then maintains that the incremental HAP emissions reduction from oxidation cannot justify the cost and energy use of control when compared to covers. The commenter notes that there are some facilities in EPA's database that use add-on controls for mixing. However, according to the commenter, the control in all cases is incidental to the use of the add-on control for other operations in the facility. Therefore, the commenter believes that add-on control is not the best control for mixing, and the final rule should require covers instead of add-on controls for all mixing operations.

Response:

The reasons for why HAP emissions are being controlled is usually not considered in the setting of MACT standards. Further, we disagree with the commenter's characterization of the control of mixing HAP emissions as “incidental.” We do not agree that the data provided support the claim of 85 to 96 percent control using covers. Therefore, we have not written the rule as requested by the commenter. New sources that exceed the 100 tpy HAP emissions threshold will still have to cover the mixing tanks and control their HAP emissions from mixing by 95 percent, which is the new source floor level of control.

Comment:

Two commenters suggested that the definition of compression molding be changed to include a process where resin paste is added to the reinforcement at the press and to include the use of in-mold coating (IMC). According to the commenters, the resin paste process is similar to the use of SMC and BMC because there is no exposure of HAP-containing material, except where the charge is being prepared and placed in the mold. The controls for this process are the same as those available for SMC and BMC (

i.e.

limiting the quantity of exposed materials to that which is required for one press cycle). In-mold coating is a process where HAP-containing materials are mixed with catalyst and then injected into the mold cavity after the molding cycle has started. The IMC reduces the need for post-mold coating (painting) operations. The controls available for IMC are the same as those generally available for mixing operations.

Response:

We have modified the definition of closed molding to include these processes.

Comment:

Several commenters requested that the work practice standard requiring closed molding operations to uncover, unwrap, or expose only one charge per mold cycle per machine be revised so that a charge is defined as the amount of materials required to charge the mold(s) for each machine cycle. Some machines have more than one mold, and limiting the amount of material would cause a bottleneck in production capacity. One commenter added that the rule should allow multiple charges to be loaded into the hopper, provided the hopper is kept covered between loading operations and that the unlimited use of slitting machines to unwrap, cut, and prepare charges should be permitted, provided that the charges are then covered or placed in a closed container prior to use at the press.

Response:

We agree that where multiple charges are required for a single mold cycle, the rule should allow them to be prepared at the same time

and held in a closed container prior to use. Therefore, we have written the final rule to define “charge” per the commenter's suggestion and to require that multiple charges be kept covered, as for single charges, until used. We have also written the final rule to allow the use of hoppers, robotic loaders, and slitters.

Comment:

Many commenters noted in the proposed rule that polymer casting mixing operations in containers of 21 gallons or less may be open while active mixing occurs and requested that this exemption be increased. The commenters note that many are using 350 lb containers, which is equivalent to 21.6 gallons. According to the commenters, the mixing process uses an electric mixer and requires frequent manual scraping of the sides, and a requirement to cover the mixer would present a productivity disadvantage.

Response:

Changing the volume exemption from 21 to 21.6 gallons would be consistent with the intent of the proposed exemption. The surface area of exempt mixers is a more important parameter because it is directly related to the amount of HAP emissions that would occur. Therefore, we have included this exemption in the final rule but have changed the exemption parameter to 500 square inches of surface area. This change should allow the 21.6 gallon mixers, commonly used in this industry, to be exempt from the requirement to cover the mixer.

Comment:

One commenter noted that worker safety, fire prevention, and product quality requirements necessitate limited active venting of HAP-containing materials storage vessels, covered mixers, and material conveyance enclosures. Some facilities store resins in bulk tanks with passive atmospheric venting. Problems arise from resin contact with the water vapor in the atmosphere. Polymerization occurs on side walls, vents, and transfer pipes. Vents, especially conservation vents, can plug, threatening the tank's structural integrity. Nitrogen blanketing is used by some facilities to solve these problems. Nitrogen blanketing is also used to inert the head space in bulk HAP-containing materials storage tanks for fire prevention.

Another commenter requested clarifying language to allow passive vents for bulk HAP-containing materials storage tanks. The vents are small to allow for breathing of the tanks as they are filled and emptied. These vents are required under OSHA to prevent pressure build-up and to reduce the chances of explosions and major leaks or spills. The annual breathing losses from all eight of this commenter's tanks are less than 1 tpy. A third commenter suggested that the rule be changed to allow venting from HAP-containing materials storage vessels.

Response:

We did not intend to prohibit bulk HAP-containing materials storage tanks from venting to the atmosphere for safety. The final rule has been written to clarify this. However, it is not our intent to allow venting from all HAP-containing materials storage vessels because the safety concerns commenters raised are limited to bulk HAP-containing materials storage vessels. Thus, the final rule prohibits venting from HAP-containing materials storage vessels other than bulk storage tanks.

Comment:

One commenter noted that the proposed rule would require that HAP-containing materials storage containers be kept closed or covered, except when adding or removing materials. The commenter claims this provision is not workable.

Response:

We believe that covering HAP-containing materials storage containers is a simple and cost-effective way to reduce styrene evaporation. We also note that over 200 facilities that reported information on storage stated that HAP-containing materials storage containers are covered or closed. This provision has been written in the final rule.

Comment:

Several commenters requested that HAP cleaners be allowed when used in a closed system or covered tank. The reasons were that aggressive cleaners were necessary due to the presence of cured resin on some surfaces, and that it was important to use a cleaner that would not cause contamination problems. They stated that HAP emissions from these closed systems were minimal, and in many cases, the styrene used for cleaning was recycled to the process as a raw material.

Response:

The proposed rule allowed the use of HAP cleaners to remove cured resin from application equipment because of the difficulty associated with removing the cured resin. One commenter in particular identified other equipment used in the process on which cured resin may occur. We note, as the commenters have, that styrene is the main HAP used in the reinforced plastic composites industry and can be reused in the process without contaminating the end products. Therefore, we believe that the commenters' requests are reasonable and have written the final rule to expand the definition of “application equipment” and to allow the use of HAP-containing cleaners in closed systems (including covered tanks).

VI. Relationship of the Final NESHAP to Other NESHAP and the CAA Operating Permits Program

A. National Emissions Standards for Closed Vent Systems, Control Devices, Recovery Devices, and Routing to a Fuel Gas System or a Process (40 CFR Part 63, Subpart SS)

If you use an add-on control device(s) to control HAP emissions, you will need to comply with certain provisions in 40 CFR part 63, subpart SS, for add-on controls. The standards in subpart SS, cited by the final Reinforced Plastic Composites NESHAP, are applicable to most sources using an add-on control device. The final Reinforced Plastic Composites NESHAP cite these sections in subpart SS rather than repeating them in the regulatory text.

B. NESHAP for Boat Manufacturing (40 CFR Part 63, Subpart VVVV)

The final NESHAP for Boat Manufacturing were published on August 22, 2001 (66 FR 44218). There is a potential overlap between facilities that produce reinforced plastic composites if they also produce boat hulls, boat decks, or molds for boat hulls and decks. We have included provisions in the Reinforced Plastic Composites NESHAP to clarify where the Reinforced Plastic Composites NESHAP apply, and where the Boat Manufacturing NESHAP (40 CFR part 63, subpart VVVV) apply.

C. NESHAP for Plastic Parts and Products (Surface Coating)

There are currently NESHAP under development for proposal that will regulate coating of plastic parts and products. The Small Business Advocacy Review Panel, convened for the Reinforced Plastic Composites NESHAP, recommended that we consider the interaction of the Reinforced Plastic Composites NESHAP and the Plastic Parts and Products NESHAP. The Plastic Parts and Products NESHAP may potentially affect facilities that produce reinforced plastic parts and then apply a coating to the finished parts. We have coordinated with this project and have determined that there should be no overlap (

i.e.,

specific operations covered by today's final NESHAP should not also be covered in the Plastic Parts and Products NESHAP). We have not determined any requirements of today's final NESHAP that would overlap, conflict, or cause a duplication of effort.

D. Operating Permit Program

Under the operating permit program codified at 40 CFR part 70 and 40 CFR part 71, all major sources subject to standards under section 111 or 112 of the CAA must obtain an operating permit (

See

§§ 70.3(a)(1) and 71.3(a)(1)). Therefore, all major sources subject to the final NESHAP must obtain an operating permit.

Some reinforced plastic composites production facilities may be major sources based solely on their potential to emit, even though their actual HAP emissions are below the major source level. These facilities may choose to obtain a federally enforceable limit on their potential to emit so that they are no longer considered major sources subject to the final NESHAP. Sources that opt to limit their potential to emit (

e.g.,

limits on operating hours or amount of material used) are referred to by the EPA as “synthetic area” sources. To become a synthetic area source, you must contact your local permitting authority to obtain an operating permit with the appropriate operating limits. These limits must be obtained prior to the compliance date for existing sources, which is April 21, 2006. These operating limits will then be federally enforceable under 40 CFR 70.6(b).

VII. Statutory and Executive Order Reviews

A. Executive Order 12866, Regulatory Planning and Review

Under Executive Order 12866 (58 FR 51735, October 4, 1993), we must determine whether this regulatory action is “significant” and therefore subject to Office of Management and Budget (OMB) review and the requirements of the Executive Order. The Executive Order defines “significant regulatory action” as one that is likely to result in a rule that may:

(1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities;

(2) Create a serious inconsistency or otherwise interfere with an action taken or planned by another agency;

(3) Materially alter the budgetary impact of entitlements, grants, user fees, or loan programs or the rights and obligations of recipients thereof; or

(4) Raise novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

Pursuant to the terms of Executive Order 12866, OMB notified EPA at proposal that it considered this rulemaking a “significant regulatory action” within the meaning of the Executive Order. The EPA submitted the proposed rule to OMB for review. Changes made in response to suggestions or recommendations from OMB are documented and included in the public record. The OMB has informed EPA that it no longer considers this action significant. Therefore, it is not subject to further OMB review. The OMB did request a copy of the final regulation and preamble prior to publication. However, they did not request any changes in the final rule.

B. Paperwork Reduction Act

The information collection requirements in the final rule have been submitted for approval to the OMB under the Paperwork Reduction Act, 44 U.S.C. 3501

et seq.

An information collection request (ICR) document has been prepared by EPA (ICR No. 1976.01) and a copy may be obtained from Susan Auby by mail at the Office of Environmental Information, Collection Strategies Division (2822), U.S. EPA, 1200 Pennsylvania Avenue, NW., Washington, DC 20460, by e-mail at “

auby.susan@epa.gov,

” or by calling (202) 566-1672. A copy may also be downloaded from the internet at

http://www.epa.gov/icr.

The information requirements are not enforceable until OMB approves them.

The final rule contains monitoring, reporting, and recordkeeping requirements. These notices and reports are the minimum needed by us to determine if you are subject to the NESHAP and whether you are in compliance. These recordkeeping requirements are the minimum necessary to determine initial and ongoing compliance. Based on reported information, we would decide which reinforced plastic composites facilities and what records or processes should be inspected. The recordkeeping and reporting requirements are consistent with the General Provisions of 40 CFR part 63.

These recordkeeping and reporting requirements are specifically authorized by section 114 of the CAA (42 U.S.C. 7414). All information submitted to us for which a claim of confidentiality is made will be safeguarded according to our policies in 40 CFR part 2, subpart B.

We expect the final rule to affect a total of approximately 488 facilities over the first 3 years. This includes 435 existing facilities, and 53 new reinforced plastic composites facilities that will become subject to the final NESHAP during the first 3 years.

The estimated average annual burden for the first 3 years after promulgation of the final NESHAP for industry and the implementing agency is outlined below. You can find the details of this information collection in the “Standard Form 83 Supporting Statement for ICR No. 1976.01,” in Docket ID No. OAR-2003-0003 (formerly Docket No. A-94-52).

Affected entity

Total hours

Labor costs

Total annual O&M costs

Total costs

Industry

13,785

$613,623

$15,807

$629,431

Implementing agency

11,120

444,047

NA

444,047

Burden means the total time, effort, or financial resources expended by persons to generate, maintain, retain, or disclose or provide information to or for a Federal agency. This includes the time needed to review instructions; develop, acquire, install, and utilize technology and systems for the purposes of collecting, validating, and verifying information, processing and maintaining information, and disclosing and providing information; adjust the existing ways to comply with any previously applicable instructions and requirements; train personnel to be able to respond to a collection of information; search data sources; complete and review the collection of information; and transmit or otherwise disclose the information.

An agency may not conduct or sponsor, and a person is not required to respond to a collection of information unless it displays a currently valid OMB control number. Control numbers for EPA's regulations are listed in 40 CFR part 9 and 48 CFR chapter 15. When the OMB approves the information collection requirements of the final rule, the EPA will amend the table in 40 CFR

part 9 of currently approved ICR control numbers issued by OMB for various regulations.

C. Regulatory Flexibility Analysis

The EPA has prepared a Final Regulatory Flexibility Analysis (FRFA) in connection with the final rule. For purposes of assessing the impacts of today's final rule on small entities, small entity is defined as: (1) A small business ranging from 500 to 1,000 employees as defined by the Small Business Administration's size standards; (2) a small governmental jurisdiction that is a government of a city, county, town, school district or special district with a population of less than 50,000; and (3) a small organization that is any not-for-profit enterprise which is independently owned and operated and is not dominant in its field. The table below presents the size threshold for small businesses by NAICS Codes.

Category

NAICS codes

Maximum number of employees to be considered a small business

Manufacturing

335312

1000

336211

336112

33612

336213

336413

33651

325211

750

327993

332998

33312

33651

335311

335313

33422

33653

336399

All other identified NAICS Codes in this source category

500

In accordance with section 603 of the RFA, EPA prepare

This text is long and has been trimmed here. Open the source document for the complete record.