To demonstrate initial compliance based on the materials used, you would be required to demonstrate that either the organic HAP content of each coating meets the emission limits and that you use no organic HAP containing thinners or cleaning materials, or that the total mass of organic HAP in all coatings, thinners, and cleaning materials divided by the total volume of coating solids meets the emission limits. For the latter option, you would be required to:

Option 2: Compliance based on using a capture system and addon control device

If you use a capture system and addon control device, other than a solvent recovery system for which you conduct a liquidliquid material balance, your testing and initial compliance requirements are as follows:

If you use a capture system and addon control device, other than a solvent recovery system for which you conduct liquidliquid material balances, you would determine both the efficiency of the capture system and the emission reduction efficiency of the control device. To determine the capture efficiency, you would either verify the presence of a permanent total enclosure using EPA Method 204 of 40 CFR part 51, appendix M (and all materials must be applied and dried within the enclosure); or use one of three protocols in Sec. 63.4165 to measure capture efficiency. If you have a permanent total enclosure and all materials are applied and dried within the enclosure and you route all exhaust gases from the enclosure to a control device, you would assume 100 percent capture.

To determine the emission reduction efficiency of the control device, you would conduct measurements of the inlet and outlet gas streams. The test would consist of three runs, each run lasting 1 hour, using the following EPA Methods in 40 CFR part 60, appendix A:

If you use a solvent recovery system, you could determine the overall control efficiency using a liquidliquid material balance instead of conducting an initial performance test. If you use the material balance alternative, you would be required to measure the amount of all materials used in the affected source during the compliance period and determine the total volatile matter contained in these materials. You would also measure the amount of volatile matter recovered by the solvent recovery system during the compliance period. Then you would compare the amount recovered to the amount used to determine the overall control efficiency, and apply this efficiency to the organic HAPtocoating solids ratio for the materials used. You would record the calculations and results and include them in your Notification of Compliance Status.

Operating Limits. As mentioned above, you would establish operating limits as part of the initial performance test of a capture system and control device, other than a solvent recovery system for which you conduct liquidliquid material balances. The operating limits are the minimum or maximum (as applicable) values achieved for capture systems and control devices during the most recent performance test that demonstrated compliance with the emission limits. If you operate your capture system and control device at different sets of representative operating conditions, you must establish operating limits for the parameters for each different operating condition.

[[Page 81139]]

The proposed rule specifies the parameters to monitor for the types of emission control systems commonly used in the industry. You would be required to install, calibrate, maintain, and continuously operate all monitoring equipment according to manufacturer's specifications and ensure that the continuous parameter monitoring systems (CPMS) meet the requirements in Sec. 63.4168 of the proposed rule. If you use control devices other than those identified in the proposed rule, you would submit the operating parameters to be monitored to the Administrator for approval. The authority to approve the parameters to be monitored is retained by EPA and is not delegated to States.

If you use a thermal or catalytic oxidizer, you would continuously monitor the appropriate temperature and record it at least every 15 minutes. For thermal oxidizers, the temperature monitor is placed in the firebox or in the duct immediately downstream of the firebox before any substantial heat exchange occurs. The operating limit would be the average temperature measured during the performance test, and for each consecutive 3hour period the average temperature would have to be at or above this limit. For catalytic oxidizers, temperature monitors are placed immediately before and after the catalyst bed. The operating limits would be the average temperature just before the catalyst bed and the average temperature difference across the catalyst bed during the performance test, and for each 3hour period the average temperature and the average temperature difference would have to be at or above these limits.

If you use a carbon adsorber and do not conduct liquidliquid material balances to demonstrate compliance, you would monitor the carbon bed temperature after each regeneration and the total amount of steam or nitrogen used to desorb the bed for each regeneration. The operating limits would be the carbon bed temperature (not to be exceeded) and the amount of steam or nitrogen used for desorption (to be met as a minimum).

If you use a condenser, you would monitor the outlet gas temperature to ensure that the air stream is being cooled to a low enough temperature. The operating limit would be the average condenser outlet gas temperature measured during the performance test, and for each consecutive 3hour period the average temperature would have to be at or below this limit.

For each capture system that is not a permanent total enclosure, you would establish operating limits for gas volumetric flow rate or duct static pressure for each enclosure or capture device. The operating limit would be the average volumetric flow rate or duct static pressure during the performance test, to be met as a minimum. For each capture system that is a permanent total enclosure, the operating limit would require the average facial velocity of air through all natural draft openings to be at least 200 feet per minute or the pressure drop across the enclosure to be at least 0.007 inches water.

G. What Are the Continuous Compliance Provisions?

Emission Limits. If you demonstrate compliance with the proposed emission limits based on the materials used, you would demonstrate continuous compliance if, for each monthly compliance period, the ratio of organic HAP to coating solids is less than or equal to the emission limits. You would follow the same procedures for calculating the organic HAP to coating solids ratio that you used for the initial compliance period.

For each coating operation on which you use a capture system and control device, other than solvent recovery for which you conduct a liquidliquid material balance, you would use the continuous parameter monitoring results for the month in determining the mass of organic HAP emissions. If the monitoring results indicate no deviations from the operating limits and there were no bypasses of the control device, you would assume the capture system and control device are achieving the same percent emission reduction efficiency as they did during the most recent performance test in which compliance was demonstrated. You would then apply this percent reduction to the total mass of organic HAP in materials used in controlled coating operations to determine the monthly emission rate from those operations. If there were any deviations from the operating limits during the month or any bypasses of the control device, you would account for them in the calculation of the monthly emission rate by assuming the capture system and control device were achieving zero emission reduction during the periods of deviation.

For each coating operation on which you use a solvent recovery system and conduct a liquidliquid material balance each month, you would use the liquidliquid material balance to determine control efficiency. To determine the overall control efficiency, you must measure the amount of all materials used during each month and determine the volatile matter content of these materials. You must also measure the amount of volatile matter recovered by the solvent recovery system during the month, calculate the overall control efficiency, and apply it to the total mass of organic HAP in the materials used to determine total organic HAP emissions.

Operating Limits. If you use a capture system and control device, the proposed rule would require you to achieve on a continuous basis the operating limits you establish during the performance test. If the continuous monitoring shows that the capture system and control device are operating outside the range of values established during the performance test, you have deviated from the established operating limits.

If you operate a capture system and control device that allow emissions to bypass the control device, you would have to demonstrate that HAP emissions from each emission point within the affected source are being routed to the control device by monitoring for potential bypass of the control device. You may choose from the following four monitoring procedures:

If the bypass monitoring procedures indicate that emissions are not routed to the control device, you have deviated from the emission limits.

Operations During Startup, Shutdown, and Malfunction. If you use a capture system and control device for compliance, you would be required to develop and operate according to a startup, shutdown, and malfunction plan during periods of startup, shutdown, and malfunction of the capture system and control device.

Emissions Reductions Plan for Mixing, Storage, and Waste Handling. If you use a capture system and addon control device for compliance, you would be required to develop and operate according to a plan for reducing emissions from mixing operations, storage tanks or other containers, and waste handling operations. This plan would include a description of all steps taken to minimize emissions from these sources (e.g., using closed storage containers practices to minimize emissions during filling and transfer of
[[Page 81140]]
contents from containers, using spill minimization techniques, placing solventladen cloth in closed containers immediately after use, etc.). If you do not develop a plan for reducing HAP emissions or you do not implement the plan, this would be a deviation from the work practice standard. You would have to make the emissions reductions plan available for inspection if the Administrator requests to see it. Under the option where emissions are reduced by using lowerHAP or noHAP materials, we are assuming that all the HAP in the materials entering the affected source are volatilized (emitted), unless the facility can show that a portion of the HAP released is recovered. Therefore, emissions from operations occurring within the affected source (e.g., mixing operations) are accounted for in the estimate of total materials usage at the affected source. However, when you comply by using capture systems and addon control devices, these systems and control devices may not be associated with some operations within the affected source, such as the mixing, storage, and waste handling operations. An emissions reductions plan is needed to assure that emissions are reduced from those uncontrolled operations using best available practices. When the plan is instituted as a work practice, it should provide a level of quality control and assurance.
H. What Are the Notification, Recordkeeping, and Reporting Requirements?

You are required to comply with the applicable requirements in the NESHAP General Provisions, subpart A of 40 CFR part 63, as described in the proposed rule. The General Provisions notification requirements include: initial notifications, notification of performance test if you are complying using a capture system and control device, notification of compliance status, and additional notifications required for affected sources with continuous monitoring systems. The General Provisions also require certain records and periodic reports.

Initial Notifications. If the proposed standards apply to you, you must send a notification to the EPA Regional Office in the region where your facility is located, and to your State agency, at least 1 year before the compliance date for existing sources and within 120 days after the date of initial startup for new and reconstructed sources, or 120 days after publication of the final rule, whichever is later. That report notifies us and your State agency that you have an existing facility that is subject to the proposed standards or that you have constructed a new facility. Thus, it allows you and the permitting authority to plan for compliance activities. You would also need to send a notification of planned construction or reconstruction of a source that would be subject to the proposed rule and apply for approval to construct or reconstruct.

Notification of Performance Test. If you demonstrate compliance by using a capture system and control device for which you do not conduct a liquidliquid material balance, you would conduct a performance test. The performance test would be required no later than the compliance date for an existing affected source, and no later than 180 days after startup or 180 days after publication of the final rule, whichever is later, for a new or reconstructed source. You must notify us (or the delegated State or local agency) at least 60 calendar days before the performance test is scheduled to begin, as indicated in the General Provisions for the NESHAP.

Notification of Compliance Status. Your compliance procedures would depend on which compliance option you choose. For each compliance option, you would send us a Notification of Compliance Status within 30 days after the end of the initial compliance period. In the notification, you would certify whether the affected source has complied with the proposed standards, identify the option(s) you used to demonstrate initial compliance, summarize the data and calculations supporting the compliance demonstration, and describe how you will determine continuous compliance.

If you elect to comply by using a capture system and control device for which you conduct performance tests, you must provide the results of the tests. Your notification would also include the measured range of each monitored parameter and the operating limits established during the performance test, and information showing whether the source has complied with its operating limits during the initial compliance period.

Recordkeeping Requirements. You would be required to keep records of reported information and all other information necessary to document compliance with the proposed rule for 5 years. As required under the General Provisions, records for the 2 most recent years must be kept onsite; the other 3 years' records may be kept offsite. Records pertaining to the design and operation of the control and monitoring equipment must be kept for the life of the equipment.

Depending on the compliance option that you choose, you may need to keep records of the following:

If you demonstrate compliance by using a capture system and control device, you would also need to keep records of the following:

The proposed rule would require you to collect and keep records according to certain minimum data requirements for the CPMS. Failure to collect and keep the specified minimum data would be a deviation that is separate from any emission limits, operating limits, or work practice standards.

Deviations, as determined from these records, would need to be recorded and also reported. A deviation is any instance when any requirement or obligation established by the proposed rule including, but not limited to, the emission limits, operating limits, and work practice standards, is not met.

If you use a capture system and control device to reduce HAP emissions, you would have to make your startup, shutdown, and malfunction plan available for inspection if the Administrator requests to see it. It would stay in your records for the life of the affected source or until the source is no longer subject to the proposed standards. If you revise the plan, you
[[Page 81141]]
would need to keep the previous superseded versions on record for 5 years following the revision.

Periodic Reports. Each reporting year is divided into two semiannual reporting periods. If no deviations occur during a semiannual reporting period, you would submit a semiannual report stating that the affected source has been in continuous compliance. If deviations occur, you would include them in the report as follows:

You would also have to include an explanation in each semiannual report if a change occurs that might affect the compliance status of the affected source, or you change to another option for meeting the emission limit.

Other Reports. You would be required to submit reports for periods of startup, shutdown, and malfunction of the capture system and control device. If the procedures you follow during any startup, shutdown, or malfunction are inconsistent with your plan, you would report those procedures with your semiannual reports in addition to immediate reports required by Sec. 63.10(d)(5)(ii).
III. Rationale for Selecting the Proposed Standards

A. How Did We Select the Source Category?

The surface coating of large appliances is a source category that is on the list of source categories to be regulated because it contains major sources which emit or have the potential to emit at least 10 tons of any one HAP or at least 25 tons of any combination of HAP annually. The proposed rule would control HAP emissions from both new and existing major sources. Area sources are not being regulated under this proposed rule.

The surface coating of large appliances as described in the listing includes any facility engaged in the surface coating of large appliance parts or products. We use the large appliance product lists contained in the SIC and NAICS code descriptions to describe the vast array of large appliance parts and products.

We intend the source category to include facilities for which the surface coating of large appliances is either their principal activity or an integral part of a production process that is the principal activity. Most coating operations are located at plant sites that are dedicated to these activities. However, some may be located at sites for which some other activity is principal. Collocated surface coating operations comparable to the types and sizes of the dedicated facilities, in terms of the coating operation and applicable emission control techniques, are included in the source category.

The source category does not include research or laboratory facilities or janitorial, building, and facility maintenance operations.

B. How did we select the regulated pollutants?

Organic HAP. Available emission data collected during the development of the proposed NESHAP show that the primary organic HAP emitted from the surface coating of large appliances include xylene, glycol ethers, toluene, methylene diphenyl diisocyanate, and methyl ethyl ketone. These compounds account for approximately 82 percent of this category's nationwide organic HAP emissions. However, many other organic HAP are used, or can be used, in large appliance coatings, thinners, and cleaning materials. Therefore, the proposed rule would regulate emissions of all organic HAP.

Inorganic HAP. Although most of the coatings used in this source category do not contain inorganic HAP, some special purpose coatings used by this source category do contain inorganic HAP such as chromium, cobalt, lead, and manganese. Emissions of these materials to the atmosphere are minimal because the facilities in this source category employ either water curtains or dry filters that remove overspray particles from the spray booth exhaust. At this time, it does not appear that emissions of inorganic HAP from this source category warrant Federal regulation.

C. How Did We Select the Affected Source?

In selecting the affected source(s) for emission standards, our primary goal is to ensure that MACT is applied to HAPemitting operations or activities within the source category being regulated. The affected source also serves to establish where new source MACT applies under a particular standard. Specifically, the General Provisions in subpart A of 40 CFR part 63 define the terms ``construction'' and ``reconstruction'' with reference to the term ``affected source'' and provide that new source MACT applies when construction or reconstruction of an affected source occurs. The collection of equipment and activities evaluated in determining MACT (including the MACT floor) is used in defining the affected source.

When an emission standard is based on a collection of emissions sources, or total facility emissions, we select an affected source based on that same collection of emission sources, or the total facility, as well. This approach for defining the affected source broadly is particularly appropriate for industries where a plantwide emission standard provides the opportunity and incentive for owners and operators to utilize control strategies that are more cost effective than if separate standards were established for each emission point within a facility.

Selection of Affected Source. The affected source for these proposed standards is broadly defined to include all operations associated with the coating of large appliances and the cleaning of product substrates or coating operation equipment. These operations include storage and mixing of coatings and other materials; surface preparation of the large appliances prior to coating application; coating application and flashoff, drying and curing of applied [[Page 81142]]
coatings; cleaning operations; and waste handling operations.

In selecting the affected source, we considered, for each operation, the extent to which HAPcontaining materials are used and the amount of HAP that are emitted. Cleaning and coating application, flashoff, and curing/drying operations account for the majority of HAP emissions at large appliance surface coating operations. These operations are included in the affected source.

We were not able to obtain data to adequately quantify HAP emissions from storage, mixing, and waste handling. However, solvents that are added to coatings as thinners, and other HAPcontaining additives to coatings, may be emitted during mixing and storage. The level of emissions depends on the type of mixing and the type of storage container and the work practices used at the facility. Emissions from waste handling operations depend on the type of system used to collect and transport organic HAPcontaining waste coatings, thinners, and cleaning materials in the facility. For example, solvent laden rags that are used to clean spray booths or tanks could be a source of HAP emissions. The method used to isolate and store such rags affects the level of emissions to ambient air. Mixing, storage, and waste handling operations are included in the affected source.

A broad definition of the affected source was selected to provide maximum flexibility in complying with the proposed emission limits for organic HAP. In planning its total usage of HAPcontaining materials, each facility can select among available coatings, thinners, and cleaning materials to comply with the proposed limits.

Additional information on the large appliance surface coating operations selected for regulation, and other operations, are included in the docket for the proposed standards.
D. How did we determine the basis and level of the proposed standards for existing and new sources?

The sections below present the rationale for determining the MACT floor, regulatory alternatives beyond the floor, and selection of the proposed standards for existing and new affected sources.

How did we determine the MACT floor technology? After we identify the specific source categories or subcategories of sources to regulate under section 112 of the CAA, we must develop emission standards for each category and subcategory. Section 112(d)(3) establishes a minimum baseline or ``floor'' for standards. For new sources in a category or subcategory, the standards cannot be less stringent than the emission control that is achieved in practice by the bestcontrolled similar source. The standards for existing sources can be less stringent than standards for new sources, but they cannot be less stringent than the average emission limitation achieved by the bestperforming 12 percent of existing sources (or the bestperforming five sources for categories or subcategories with fewer than 30 sources).

Within the large appliance industry, organic HAP emission control for cleaning and surface coating operations is accomplished primarily through the use of lowerHAP coatings, thinners, and cleaning materials. Addon capture and control systems for organic HAP are rarely used by the industry. While lower organic HAP materials have achieved broad use throughout the industry, each particular coating technology is not used at every facility. Rather, facilities use various combinations of lowHAP coatings, thinners, and cleaning materials. Thus, we judged the most reasonable approach to establishing a MACT floor to be the evaluation of a facility's organic HAP emissions from all coatingrelated operations. To account for differences in production levels from one facility to another, we normalized the organic HAP emission rate by the volume of coating solids used. We believe coating solids usage is an appropriate indicator of overall production levels.

We used information obtained from industry survey responses to estimate the sourcewide organic HAP emission rate from each survey respondent. We calculated total organic HAP emissions by assuming that 100 percent of the volatile components in all coatings (including adhesives), thinners, and cleaning materials (including surface preparation materials) are emitted. Major sources were identified as: those facilities that listed ``major source'' or ``synthetic minor source'' as their title V status on their questionnaire response; those facilities that reported their HAP emissions under ``maximum design capacity'' as greater than 9.1 megagrams per year (Mg/yr) (10 tpy); and other facilities that we judged to have the capacity to increase their HAP emissions to at least 9.1 Mg/yr, even though they did not identify themselves as major or synthetic minor sources. The final group of facilities were included because they reported actual HAP emissions of greater than 3 Mg (3.3 tons) during the reporting year and did not report a ``maximum design capacity.'' If these facilities operate at full capacity over multiple shifts each day, their annual emission rate may equal or exceed 9.1 Mg/yr.

The survey response information was used to determine the total volume of coating solids used by each source from all types of coatings. We included decorative, protective, and functional coatings in this total.

Using the sourcewide organic HAP emissions and the total volume of coating solids used for each survey respondent, we calculated the normalized organic HAP emissions (emission rate) in units of kilograms organic HAP per liter of coating solids used. The facilities were then ranked from the lowest emission rate to the highest, with the following exceptions. Facilities that reported the predominant use of powder coatings (greater than 90 percent of all coating solids usage) were excluded from the MACT floor calculations. While powder coating technology is a proven lowHAP coating technology, its applicability is not considered to be universal for all products manufactured within the source category. For those facilities whose products can be coated with this technology, the use of powder coatings is a very effective and efficient means of reducing HAP emissions. The degree of HAP reductions that can be achieved with the powder coating technology is close to 100 percent. However, because many large appliance parts and products cannot be satisfactorily coated with powder coating technology, we concluded that it would not be appropriate to define the MACT floors based primarily on their use. Facilities that used lesser amounts of powder coatings in combination with other lowHAP coating technologies were included in the MACT floor determination.

For some facilities, the organic HAP to coating solids ratio was very low due to the facilities' usage of unusually large quantities of lowHAP and nonHAP adhesives. The low and nonHAP adhesives usage for these facilities ranged from 40 to 84 percent of all coating solids. While many facilities in the source category use adhesives (a functional coating), their use is not as widespread compared to the decorative and protective coatings usually associated with the appearance of large appliance products. On the average, adhesive usage among all facilities in the source category database is about 4 percent of the total solids used. We concluded that because of the specific function served by adhesives, the lowHAP adhesive technology employed in the facilities described above may not be transferable to the decorative and
[[Page 81143]]
protective coatings which account for the remaining 96 percent of coating solids usage in the industry. Thus, we concluded that the facilities using atypically large quantities of these adhesives relative to decorative and protective coatings should not be included in the floor determination of existing sources or new sources.

For the existing source MACT floor, the top 12 percent of the facilities were determined based on the number of facilities in the MACT floor database (95 database facilities x 12 percent=11.4). Because the calculated value was greater than 11, we used data from 12 facilities to determine the MACT floor. The floor was calculated as the arithmetic average of the emission rates of the top 12 bestperforming representative facilities.

This process resulted in a MACT floor equal to 0.13 kg HAP/liter of coating solids (1.1 lb/gal). The survey data showed no appreciable differences between the floor facilities and the remaining facilities in the database in terms of the substrates coated, the coating technologies used, or the applicability of control measures across the various operations. Therefore, we believe the floor level of control is achievable by all existing sources.

The best performing facility in our database has an emission rate of 0.022 kg HAP/liter of coating

FOR FURTHER INFORMATION CONTACT Dr. Mohamed Serageldin, Coatings and Consumer Products Group, Emission Standards Division (MD13), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711; telephone number (919) 5412379; facsimile number (919) 5415689; electronic mail (email) address: serageldin.mohamed@epa.gov.