United States
            Environmental
        Office of Air Quality
        Planning and Standards
EPA-453/R-01-010
October 2001
http://www.epa.gov/ttn/ua
SEPA
National Emission Standards
for Hazardous Air Pollutants
(NESHAP) for Source
Category: Metal Furniture
Surface Coating --
Background Information for
Proposed Standards

-------
                                           EPA-453/R-01-010
         National Emission Standards for
      Hazardous Air Pollutants (NESHAP)
               for Source Category:
       Metal Furniture Surface Coating -
Background Information for Proposed Standards
                Emission Standards Division
             U.S. Environmental Protection Agency
           Office of Air Quality Planning and Standards
          Research Triangle Park, North Carolina 27711
                     October 2001

-------
                                     DISCLAIMER

       This report has been reviewed by the U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Emission  Standards Division and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or recommendation for use.

-------
                     ENVIRONMENTAL PROTECTION AGENCY

NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR POLLUTANTS (NESHAP)
FOR SOURCE CATEGORY: METAL FURNITURE SURFACE COATING - BACKGROUND
INFORMATION FOR PROPOSED STANDARDS

1.      The standards regulate organic hazardous air pollutant (HAP) emissions from the surface
       coating of metal furniture. Only those metal furniture surface coating operations that are part of
       major sources under section 112(d) of the Clean Air Act (CAA) will be regulated.

2.      For additional information contact:

       Dr. Mohamed Serageldin, Ph.D.
       Coatings and Consumer Products Group
       U.S. Environmental Protection Agency (C539-03)
       Research Triangle Park, NC 27711
       Telephone:  (919) 541-2379
       E-MAIL: serageldin.mohamed@epamail.epa.gov

3.      Paper copies of this document may be obtained from:

       U.S. Environmental Protection Agency Library (C267-01)
       Research Triangle Park, NC 27711
       Telephone:  (919) 541-2777

       National Technical Information Service (NTIS)
       5285 Port Royal Road
       Springfield, VA 22161
       Telephone:  (703) 487-4650

4.      Electronic copies of this document may be obtained from the EPA Technology Transfer
       Network (TTN) over the internet by going to the following address:

       http ://www. epa.gov/ttn/uatw/coat/mfurn/  (Select met_furn.html)

-------
               METAL FURNITURE SURFACE COATING NESHAP
                 BACKGROUND INFORMATION DOCUMENT
                           TABLE OF CONTENTS
1.0 SUMMARY	1-1
      1.1 INTRODUCTION	1-1
      1.2 PROPOSED STANDARDS FOR AFFECTED SOURCES  	1-2
      1.3 ENVIRONMENTAL IMPACTS	1-3
      1.4 COST IMPACTS  	1-3
      1.5 REFERENCES  	1-4

2.0 INTRODUCTION	2-1
      2.1 OVERVIEW 	2-1
      2.2 PROJECT HISTORY 	2-1
           2.2.1 Regulatory Background 	2-1
           2.2.2 Data Gathering	2-4
      2.3 REFERENCES  	2-5

3.0 INDUSTRY CHARACTERIZATION	3-1
      3.1 INTRODUCTION	3-1
      3.2 SOURCE CATEGORY DESCRIPTION	3-1
      3.3 PROCESS DESCRIPTION AND EMISSION POINTS	3-2
           3.3.1 Raw Material Preparation	3-2
           3.3.2 Cleaning Operations	3-4
           3.3.3 Coating Application Systems 	3-6
           3.3.4 Assembly	3-9
      3.4 REFERENCES  	3-12

4.0 EMISSION CONTROL TECHNIQUES	4-1
      4.1 POLLUTION PREVENTION MEASURES 	4-1
           4.1.1 Powder Coatings	4-1
           4.1.2 Waterbased Coatings	4-3
           4.1.3 Solventbased. Higher Coating Solids Coatings  	4-3
           4.1.4 Work Practice Procedures	4-4
           4.1.5 Equipment Substitutions	4-6
      4.2 POLLUTANT ABATEMENT AND RECOVERY DEVICES	4-7
           4.2.1 Recovery of Coating Overspray	4-7
           4.2.2 Add-on Control Devices 	4-7

-------
            4.2.3 Other Applicable Add-on Control Technologies	4-8
      4.3 REFERENCES 	4-9

5.0 NATIONWIDE BASELINE CHARACTERISTICS AND MODEL PLANTS 	5-1
      5.1 INTRODUCTION	5-1
      5.2 DATA SOURCES  	5-2
            5.2.1 Economic Census	5-3
            5.2.2 American Business Index (ABI) Database  	5-3
            5.2.3 Toxics Release Inventory (TRD Database	5-4
            5.2.4 Industry Questionnaires	5-4
      5.3 MODEL PLANT DEVELOPMENT	5-5
            5.3.1 Selection of Model Plants	5-5
            5.3.2 Nationwide Number of Facilities	5-8
            5.3.3 Model Unit Operations	5-10
            5.3.4 Selection of Model Plant Parameters  	5-11
      5.4 NOTES AND REFERENCES	5-19

6.0 REGULATORY APPROACH	6-1
      6.1 INTRODUCTION	6-1
      6.2  METHODOLOGY FOR DETERMINING THE MACT FLOOR 	6-2
            6.2.1 Description of MACT Floor Format	6-2
            6.2.2 Definition of "Average"	6-5
            6.2.3 Meaning  of "Best Performing" and "Best Controlled"	6-6
      6.3 COLLECTION AND ANALYSIS OF DATA  	6-6
            6.3.1 Site Visits	6-6
            6.3.2 Industry Questionnaires  	6-6
            6.3.3 Data Analysis to Determine the MACT Floor	6-7
      6.4 REGULATORY ALTERNATIVES MORE STRINGENT THAN THE MACT
            FLOOR	6-14
            6.4.1 Organic HAP Emission Control Technologies (by coating type)	6-15
            6.4.2 Emission Reduction of Add-on Capture and Control Systems	6-19
            6.4.3 Beyond-the-floor Regulatory Alternative	6-20
            6.4.4 Costs of Beyond-the-floor Regulatory Alternative	6-22
            6.4.5 Conclusions	6-27
      6.5 NOTES AND REFERENCES	6-27

7.0 ENVIRONMENTAL AND ENERGY IMPACTS	7-1
      7.1 INTRODUCTION	7-1
      7.2 DETERMINATION OF THE NUMBER OF MAJOR SOURCE FACILITIES ... 7-2
            7.2.1 Existing Major Sources  	7-3
            7.2.2 New Major Sources  	7-3

-------
      7.3 ENVIRONMENTAL IMPACTS FOR EXISTING MAJOR SOURCES	7-5
            7.3.1 Organic HAP Emission Reduction 	7-5
            7.3.2 VOC Emission Reduction	7-8
            7.3.3 Secondary Impacts 	7-12
      7.4 ENVIRONMENTAL IMPACTS FOR NEW MAJOR SOURCES	7-12

            7.4.1 Estimated Nationwide Baseline Organic HAP and VOC Emissions for New
                  Sources	7-14
            7.4.2 Organic HAP Emission Reduction 	7-14
            7.4.3 VOC Emission Reduction	7-18
      7.5 NOTES AND REFERENCES	7-18

8.0 COST IMPACTS  	8-1
      8.1 INTRODUCTION	8-1
      8.2 METHODOLOGY FOR ESTIMATING EXISTING SOURCE COSTS	8-2
            8.2.1 Determination of How Existing Sources Will Comply  	8-2
            8.2.2 Cost Methodology for Existing Sources 	8-2
      8.3 EXISTING SOURCE COST IMPACTS 	8-5
            8.3.1 Coating Operations 	8-8
            8.3.2 Cleaning Operations	8-10
            8.3.3 Monitoring. Recordkeeping. and Reporting	8-11
      8.4 METHODOLOGY FOR ESTIMATING NEW SOURCE COSTS	8-11
            8.4.1 Number of New Major Sources	8-11
            8.4.2 Determination of How New Sources Will Comply 	8-14
            8.4.3 Cost Methodology for New Sources  	8-14
      8.5 NEW SOURCE COST IMPACTS 	8-15
            8.5.1 Coating Operations 	8-19
            8.5.2 Cleaning Operations	8-19
            8.5.3 Monitoring. Recordkeeping. and Reporting	8-19
      8.6 NOTES AND REFERENCES	8-20

9.0 ECONOMIC IMPACT AND SMALL BUSINESS ANALYSIS 	9-1
      9.1 INTRODUCTION	9-1
      9.2 ECONOMIC IMPACTS	9-1
            9.2.1  Market Impacts	9-3
            9.2.2  Social Costs and Their Distribution	9-5
      9.3 SMALL BUSINESS IMPACTS 	9-9
      9.4 REFERENCES 	9-13

                             LIST OF APPENDICES
                                       111

-------
Appendix A   Evolution of the Background Information Document  	  A-l

Appendix B   Participants in the Data Collection Effort	B-l

Appendix C   Standard Industrial Classification (SIC) Code Data Summaries	  C-l

Appendix D   Estimated Emission Reduction and Cost of Powder Coating	  D-l

Appendix E   Detailed Cost Calculations for Permanent Total Enclosures and  Oxidizers	E-l

Appendix F    Calculation Methodology for the Affected-source-wide Emission Rate Used to
              Determine the Metal Furniture NESHAP MACT Floor	F-l

Appendix G   Summary Data for EPA Sampled Companies Operating Metal Furniture Manufacturing
              Facilities	  G-1
                                    LIST OF TABLES

Table 5-1      Summary of Model Plant Parameters Based on Questionnaire Response Informatidn-13

Table 5-2      Summary of Cleaning and Coating Application Unit Operations Material Usage and
              Emissions Data for Facilities in the Small Model Plant Designation	5-14

Table 5-3      Summary of Cleaning and Coating Application Unit Operations Material Usage and
              Emissions Data for Facilities in the Medium Model Plant Designation	5-15

Table 5-4      Summary of Cleaning and Coating Application Unit Operations Material Usage and
              Emissions Data for Facilities in the Large Model Plant Designation	5-16

Table 5-5      Summary of Number of Employees, Operating Schedules, and Number of Coating
              Lines for Facilities in the Small Model Plant Designation	5-17

Table 5-6      Summary of Number of Employees, Operating Schedules, and Number of Coating
              Lines for Facilities in the Medium Model Plant Designation	5-18

Table 5-7      Summary of Number of Employees, Operating Schedules, and Number of Coating
              Lines for Facilities in the Large Model Plant Designation	5-18

Table 6-1      Products Coated and FLAP Emission Control Technology Used by Facilities Included in
              the MACT Floor Analysis  	6-3
                                            IV

-------
Table 6-2     Facility Cleaning and Coating Application Operations Organic HAP Emissions
              Normalized by Coating Solids Usage for the MACT Floor Determination	6-9

Table 6-3     Organic FLAP Emission Rates Achievable Through the Use of Capture and Control
              Systems for Existing and New Metal Furniture Surface Coating Model Plants  . . . 6-21

Table 6-4     Estimated Model Plant Cost per Megagram of Organic FLAP Emission Reduction for
              Existing Metal Furniture Surface Coating Facilities for Installation of Emission Capture
              and Control Systems  	6-25

Table 6-5     Estimated Model Plant Cost per Megagram of Organic FLAP Emission Reduction for
              New Metal Furniture Surface Coating Facilities for Installation of Emission Capture and
              Control Systems	6-26
Table 7-1
Model Plant Parameters By Unit Operation  	7-4
Table 7-2     Estimated Annual Number of New Major Source Facilities for the Metal Furniture
              Surface Coating Source Category	7-6

Table 7-3     Estimated Number of New Major Source Facilities for Five Years After Promulgation
              for the Metal Furniture Surface Coating Source Category	7-7

Table 7-4     Nationwide Baseline Organic FLAP Emission Estimates for Existing Major Source
              Metal Furniture Surface Coating Facilities	7-9

Table 7-5     Nationwide Organic FLAP Emission Estimates for Existing Major Source Metal
              Furniture Surface Coating Facilities at the MACT Floor Level of Control  	7-10

Table 7-6     Nationwide Baseline VOC Emission Estimates for Existing Major Source Metal
              Furniture Surface Coating Facilities	7-11

Table 7-7     Nationwide VOC Emission Estimates  for Existing Major Source Metal Furniture
              Surface Coating Facilities at the MACT Floor Level of Control	7-13

Table 7-8     Nationwide Baseline Organic FLAP Emission Estimates for New Major Source Metal
              Furniture Surface Coating Facilities	7-15

Table 7-9     Nationwide Baseline VOC Emission Estimates for New Major Source Metal Furniture
              Surface Coating Facilities	7-16

Table 7-10    Nationwide Organic FLAP and VOC Emissions for the 20 New Sources After the

-------
             Five-Year Period After Promulgation of a Rule
. 7-17
Table 8-1     Model Plant Parameters by Unit Operation	8-3

Table 8-2     Estimated Capital Costs and Annual Costs (excluding MR&R costs) for All Facilities
             Converting to All Lower Organic HAP Content Coatings and Organic HAP-free
             Cleaning Materials-Metal Furniture Surface Coating Source
             Category	8-6

Table 8-3     Nationwide Monitoring, Recordkeeping and Reporting Costs for Existing
             Sources Summary-Metal Furniture Surface Coating Source Category	8-7
                                           VI

-------
Table 8-4

Table 8-5


Table 8-6


Table 8-7
Table 8-8


Table 8-9


Table 9-1


Table 9-2

Table 9-3


Table 9-4
Nationwide Annual Cost Summary-Metal Furniture Surface Coating Source Categofy-9

Estimated Annual Number of New Major Source Facilities for the Metal Furniture
Surface Coating Source Category	8-13

Estimated Number of New Major Source Facilities for Five Years After Promulgation
for the Metal Furniture Surface Coating Source Category	8-14

Estimated Capital Costs and Annual Costs (excluding MR&R costs) for New Facilities
Using Lower Organic HAP Content Coatings and Organic HAP-free Cleaning
Materials - Metal Furniture Surface Coating Source Category  	8-16

Nationwide Monitoring, Recordkeeping, and Reporting Costs for New Sources -
Metal Furniture Surface Coating Source  Category 	8-17

Nationwide Annual Cost Summary for New Sources - Metal Furniture Surface
Coating Source Category	8-18
Effect of Compliance Costs on Metal Furniture Producers by Industry
Segment: 1997	
9-4
Market Impacts on Metal Furniture Producers by Industry Segment: 1997  	9-5

Economic Welfare Impacts of Metal Furniture MACT on Producers, Consumers, and
Society  	9-7

Summary Statistics for SBREFA Screening Analysis on Metal Furniture Sample:
MACT Floor	9-12
                                   LIST OF FIGURES
Figure 3-1

Figure 3-2

Figure 3-3

Figure 3-4

Figure 3-5
Example Flow Diagram of Unit Operations .
UOS for Cleaning Operation System Prior to Coating	3-5

Number of Coating Lines By Coating Type and Application Method	3-7

UOS for Powder Coating Application System (with recycle)	3-10

UOS for Powder Coating Application System (without powder recycle)	3-11
                                           vu

-------
Figure 5-1     Total Coating Solids Usage by Facility for Liquid and Powder Coatings  	5-7




Figure 9-1     Full-Cost Pass-Through of Regulatory Costs  	9-8




Figure 9-2     Partial-Cost Pass-Through of Regulatory Costs 	9-8
                                            vui

-------
                                      1.0 SUMMARY







1.1 INTRODUCTION



       This background information document (BID) supports proposal of the national emission



standards for hazardous air pollutants (NESHAP) for limiting emissions of organic hazardous air



pollutant (HAP) emissions from the metal furniture surface coating source category. The standards are



being developed under the authority of section 112(d) of the Clean Air Act as amended in 1990



(CAA).l



       This document is divided into nine chapters providing a combination of background information



and EPA rationale for decisions made in the standards development process.  Chapter 2 presents an



overview of the NESHAP regulatory process and briefly describes the history of this project.



Chapters 3 through 5 provide background information including:  an industry description in Chapter 3,



the emission control techniques available to this industry in Chapter 4, and nationwide baseline



characteristics and model plants representing the metal furniture surface coating industry in Chapter 5.



Chapter 6 describes how we determined the maximum achievable control technology (MACT)



"floors", and an evaluation of the control alternatives beyond the floor.  Chapters 7 and 8 present the



predicted HAP emission reduction and cost impacts associated with the proposed standards,



respectively. Chapter 9 presents the results of the economic analysis for the proposed standards.



Relevant background material has been repeated in several of these chapters. While this leads to a



certain amount of repetitiveness in this document, the intent was to allow the reader to focus on a



specific topic without necessarily having to read one or more previous chapters to understand the



context in which the relevant material was developed. The repetitive material has been kept to a



minimum, and references to the chapters that contain more detailed information have been provided
                                             1-1

-------
throughout the text for the reader who requires a more in-depth understanding of the background



material.



        The appendices to this document provide additional background information.  Supporting



information and more detailed descriptions for the technical and rationale chapters of this document are



provided in the items referenced in each chapter and located in the project docket.



        The term "coating application" as used in this BID refers to the application of protective



coatings, adhesives, and other types of coatings. Protective coatings mean either protective or



decorative coatings, and generally refer to the paint applied to the metal furniture parts, components, or



completed assemblies.  In this BID, the term "coatings" refers to all coatings and adhesives used in the



metal furniture manufacturing process, unless otherwise limited.







1.2 PROPOSED STANDARDS FOR AFFECTED  SOURCES



        The proposed standard for new sources is an affected-source-wide organic FLAP emission limit



of 0.094 kg organic FEAP/liter coating solids (nonvolatiles) used (0.78 Ib/gal). For existing sources, the



emission limit is 0.12 kg organic FIAP/liter coating solids used (1.0 Ib/gal). The term "coating solids



used" refers to the volume of coating solids, or nonvolatiles, contained in the total amount of coatings



(including adhesives) used. It is not related to the transfer efficiency or the amount of coating solids



actually applied (deposited) on the surfaces being coated.  These limits take into account emissions



from all unit operations that may emit organic FIAP from the metal furniture manufacturing operations



associated with surface coating (i.e., the affected source).  This collection of operations includes all of



the following:



•       Surface preparation of the metal furniture prior to coating application



•       Preparation  of a coating for application (e.g., mixing in additives, dissolving resins)



•       Application of a coating to metal furniture



•       Flashoff, drying, and curing following coating application



•       Cleaning of equipment used in the coating application operation



•       Storage of coatings, additives, and cleaning materials



•       Conveyance of coatings, additives, and cleaning materials from storage  areas to mixing areas or




                                               1-2

-------
       to coating application areas, either manually or by automated means
•      Handling and conveyance of waste materials generated by the surface coating operation.
       This approach is consistent with the general industry trend of lowering emissions by reducing
the mass of pollutants in coatings and cleaning materials rather than by the use of add-on control
devices. The performance-based nature of these emission limits allows the metal furniture surface
coating industry flexibility in choosing between many available control methods (including but not limited
to coating reformulation, conversion to powder coating, solvent elimination, work practices, and add-
on control devices) to achieve compliance.

1.3 ENVIRONMENTAL IMPACTS
       As stated above, there are a variety of compliance methods  available to and in use by the
industry to meet the MACT floor level of control for organic HAP emissions.  Various combinations of
the available control methods may be utilized to achieve the MACT floor level of control.
Environmental impacts for new and existing sources were estimated  assuming that all sources will
convert from existing liquid coatings and organic HAP cleaning materials to lower organic HAP content
liquid coatings and organic  HAP-free cleaning materials such that the organic HAP emission rate for the
affected source is equal to the proposed emission limit for existing and new sources.  Detailed analyses
of the environmental impacts are discussed in Chapter 7 of this document. The nationwide organic
HAP emissions for existing sources in the fifth year after promulgation of standards implementing the
MACT floor level of control were estimated to be 6,400 Mg/yr. This represents an organic HAP
emission reduction of 13,900 Mg/yr (15,300 tons/yr) from existing sources.  The estimated organic
HAP emission reduction for the 20 new sources anticipated to be in operation in the fifth year after
promulgation of standards implementing the MACT floor level of control was estimated to be 465
Mg/yr (511 tons/yr).

1.4 COST IMP ACTS
       Cost estimates for implementing control methods to comply with the proposed emission limits
were based on applying the same compliance methodology presented above for the environmental
                                             1-3

-------
impacts.  Estimates of nationwide capital and annual costs are detailed in Chapter 8.  Capital costs

result from purchasing equipment necessary to implement the specific control methods of each option.

For new and existing sources, no capital costs would be incurred because the conversion from higher

organic HAP content coatings and cleaning materials to lower organic HAP content coatings and

organic HAP-free cleaning materials would not require the purchase of new equipment. Annual cost

impacts for new and existing sources reflect the increased cost for coatings, cleaning materials, and the

cost of implementing monitoring, reporting, and recordkeeping (MR&R) requirements.  Nationwide

annual costs for existing sources, including MR&R costs, were estimated to be $14.8 million in the fifth

year after promulgation of the standards.  Nationwide annual costs for new sources, including MR&R

costs, were estimated to be $0.6 million in the fifth year after promulgation of the standards.


1.5 REFERENCES
1.      United States Congress. Clean Air Act as amended 1990. 42 U.S.C. 7401, et seq.
       Washington, D.C. Government Printing Office.  November 1990.
                                             1-4

-------
                                   2.0 INTRODUCTION







2.1 OVERVIEW



       The Clean Air Act as amended in 1990 (CAA) requires that the emission standards for new



sources be no less stringent than the emission control achieved in practice by the best controlled similar



source. For existing sources, the emission control can be less stringent than the emission control for



new sources, but it must be no less stringent than the average emission limitation achieved by the best



performing 12 percent of existing sources (for which the EPA has emissions information). In categories



or subcategories with fewer than 30 sources, emission control for existing sources must be no less



stringent than the average emission limitation  achieved by the best performing 5 sources.  The



NESHAP are commonly known as maximum achievable control technology (MACT) standards.



       The purpose of this document is to summarize the background information gathered  during the



development of the metal furniture surface coating industry NESHAP.







2.2 PROJECT HISTORY



2.2.1  Regulatory Background



       Federal regulations that apply to metal furniture surface coating include a New Source



Performance  Standard (NSPS) under 40 CFR Part 60, Subpart EE, "Standards of Performance for



Surface Coating of Metal Furniture," which is applicable to each metal furniture surface coating



operation in which organic coatings are applied. For the purposes of subpart EE, a surface  coating



operation may be a prime coat or topcoat operations, and includes the coating application station,



flashoff area, and curing oven.  The metal furniture surface coating NSPS regulates emissions of volatile




                                             2-1

-------
organic compounds (VOC) and limits these emissions to 0.90 kilogram of VOC per liter of coating
solids (nonvolatiles) applied.  The NSPS was proposed on November 28, 1980, and promulgated on
October 29, 1982.  All metal furniture surface coating operations that were modified or began
construction or reconstruction after November 28, 1980, must be in compliance with the NSPS.
       In addition to the NSPS, the EPA also published a Control Techniques Guideline (CTG)
document 1 that covers metal furniture surface coating operations. The CTG was intended as guidance
for States in the development of State Implementation Plans (SIP). The CTG defined reasonably
available control technology (RACT) for metal furniture surface coating operations as 0.36 kilograms of
organic solvent emitted per liter of coating (minus water and 'exempt' solvents). This limit is based on
the use of low organic solvent coatings or waterbased coatings, and is approximately equivalent (on the
basis of coating solids applied) to the use of an add-on control device that collects or destroys about 80
percent of the solvent from a high organic solvent coating. 2
       Most States that have emission limitations specific to metal furniture surface coating follow the
CTG guidance. As of 1997, thirty states have limits substantially the same as the CTG, some with
different limits for individual coating types or curing methods (e.g., specialty coatings, air-dried coatings,
baked coatings). Three States have limits less stringent than the CTG, and one State is more stringent.
One State has an emission limit in units not directly convertible to those of the CTG. The remaining 15
States have no VOC limits specific to metal furniture surface coating operations.3
       None of the Federal or State regulatory efforts is specifically directed toward HAP; however,
most HAP of concern in the metal furniture surface coating industry are VOC and the same methods
used to limit VOC  emissions are also applicable to HAP emissions.  The primary use of HAP is as a
solvent in the coatings applied to metal furniture. The specific HAP used in the metal furniture surface
coating industry and the  sources of HAP emissions are described in Chapter 3 of this document.
       The MACT standard development for the metal furniture surface coating industry began in
April 1997 with a Coating Regulations Workshop for representatives of the EPA and interested
stakeholders and continues as a coordinated effort to promote consistency and joint resolution of issues
common across nine surface coating source categories.  The workshop covered eight categories: fabric
printing, coating, and dyeing; large appliances; metal can; metal coil; metal furniture; miscellaneous metal
                                             2-2

-------
parts; plastic parts; and wood building products.  The automobile and light duty truck project was



started subsequently.



       The first phase was one in which the EPA gathered readily available information about the



industry with the help of representatives from the regulated industry, State and local air pollution control



agencies, small business assistance providers, and environmental groups.  The goals of the first phase



were to either fully or partially:



•      Understand the coating process



•      Identify typical emission points and the relative emissions from each industry



•      Identify the range(s) of emission reduction techniques and their effectiveness



•      Make an initial determination on the scope of each source category



•      Determine the relationship and overlaps of the source categories



•      Locate as many facilities as possible, particularly major sources



•      Identify and involve representatives for each industry segment



•      Complete informational site visits



•      Identify issues and data needs and develop a plan for addressing them



•      Develop questionnaire(s) for additional data gathering and



•      Document results of the first phase of regulatory development for each category.



       The industry members that participated in the stakeholder process included members of the



American Furniture Manufacturers Association, Business and Institutional Furniture Manufacturer's



Association, National Paint and Coatings Association, representatives of individual companies in the



regulated industry, and representatives of companies that supply coatings to the industry. States that



participated in the process included Florida, Illinois, and Pennsylvania. In addition, data were obtained



from several other states including Alabama, California, Georgia, Indiana, Kansas, and Tennessee. The



U.S. EPA was represented by EPA Region 5, the EPA Office of Air Quality Planning and Standards



(OAQPS), the EPA Office of Enforcement and Compliance Assurance (OECA), the EPA Office of



Pollution Prevention and Toxic Substances (OPPTS), and an EPA Small Business Ombudsman.  A list



of participants in the surface coating of metal furniture rule development effort is presented in Appendix



B of this document.




                                                2O
                                               -j

-------
       The first phase of the MACT standard development concluded with the drafting of a
preliminary industry characterization (PIC) document for the metal furniture surface coating industry. 4
The information summarized in the PIC document can be used by States that may have to make case-
by-case MACT determinations under Section 112(g) or 112(j) of the CAA. The initial phase of the
regulatory development focused primarily on familiarizing the project team with metal furniture surface
coating operations, identifying facilities that make up the industry, and investigating the emission control
technologies in use by facilities in the industry.
2.2.2 Data Gathering
       Information presented in this document was collected from a variety of sources. Data
collection began with a review of information collected by the EPA during development of the NSPS.
A total of five stakeholder meetings were held for the purpose of information exchange and the
identification of potential data sources. (The participants  are listed in Appendix B of this document.)
Information was also collected during site visits to nine metal furniture surface coating facilities that
operate metal furniture coating operations with a wide variety of production rates, coating types, and
product types. On August 20, 1997, a telephone conference meeting was held with the regulatory
subgroup, which is made up of EPA and State representatives. 5
       In June 1997 and June 1998 industry questionnaires were developed for gathering information
for the development of the metal furniture surface coating industry MACT standard. The questionnaires
were sent to 39 companies with metal furniture surface coating operations identified through literature
sources and stakeholder contacts. Responses were received from a total of 85 individual facilities. Of
these 85 facilities, 59 were determined to be metal furniture surface coating facilities, 49 of which were
major or synthetic minor sources of HAP. The coating information obtained from the questionnaire
responses included approximately 680 coatings representing over 9 million liters  of usage.
                                              2-4

-------
2.3  REFERENCES


1.      Control of Volatile Organic Emissions from Existing Stationary Sources Volume HI: Surface
       Coating of Metal Furniture. EPA-450/2-77-032. U.S. Environmental Protection Agency,
       Office of Air and Waste Management, Research Triangle Park, North Carolina. December
       1977.

2.      Note 1. p. iv.

3.      Enflex Database of Federal and State Statutes.  West Group, Eagan, MN. 1997.

4.      Preliminary  Industry Characterization:  Surface Coating of Metal Furniture. U.S. Environmental
       Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park,
       North Carolina. September 1998.

5.      Memorandum submitted by Dr. Mohamed Serageldin, EPA:ESD, to the Metal Furniture
       Project Docket No. A-97-40.  Integrated Rule Development (P-MACT/P-BAC Phase),
       Regulatory Subgroup Teleconference.  August 20, 1997.
                                           2-5

-------
                           3.0  INDUSTRY CHARACTERIZATION

3.1 INTRODUCTION
       This chapter provides a general description of the metal furniture industry, the source category,
and the production process.  Discussions of emission sources from each unit operation, the number of
potentially affected sources, and national baseline emissions are also included.
       As discussed more fully in Chapter 5, it was estimated that there are 3,002 facilities that
produce metal furniture parts or products. These facilities are located throughout the U.S., with the
highest concentration of facilities in California, Michigan, New York, Florida, and Illinois. 1  Of these
facilities, it was estimated that 655 are major sources of hazardous air pollutants (HAP). The remaining
2,347 facilities are minor (area) sources of HAP emissions, with 1,435 of these area sources located in
urban areas.  Baseline (before additional control) organic HAP emissions were estimated to be 20,300
Mg/yr (22,300 tons/yr) from the major sources.

3.2 SOURCE CATEGORY DESCRIPTION
       For the purpose of developing national emission standards for hazardous air pollutants
(NESHAP), the EPA initially defined the metal furniture surface coating source category as "any facility
engaged in the surface coating and manufacture of metal furniture parts or products.'2  This description
was meant to identify what may be included in the metal furniture source category and did not represent
a complete delineation of all possible emission sources within the source category. Therefore, using
definitions from the new source performance standards and control  techniques guidelines^ as well as
various state regulations, the source category definition has been clarified as encompassing facilities that
                                             3-1

-------
apply coatings in the manufacture of metal furniture or component parts of metal furniture. Metal



furniture means furniture or components of furniture constructed either entirely or partially from metal.



Metal furniture includes, but is not limited to, components of the following types of products as well as



the products themselves: household, office, institutional, laboratory, hospital, public building, restaurant,



barber and beauty shop, and dental furniture. Metal furniture also includes office and store fixtures,



partitions, shelving, lockers, lamps and lighting fixtures, and wastebaskets.



        The corresponding Standard Industrial Classification (SIC) codes and North American



Industry Classification System (NAICS) codes for these products may be divided into two groups.



The first group are those codes that deal almost exclusively with metal furniture products and are shown



in Appendix C, Table C-l. The second group, shown in Table C-2, are those codes related to metal



furniture but that only partially encompass metal furniture products.  Table C-3 lists all of these SIC



codes and their corresponding NAICS  codes.







3.3 PROCESS DESCRIPTION AND EMISSION POINTS



        The metal furniture industry typically utilizes liquid coatings with a wide range of coating solids



content, as well as powder coatings.  Typical organic HAP reported in liquid coatings include, but are



not limited to, methyl ethyl ketone, toluene, xylene, and methyl isobutyl ketone.



        A simplified process flow diagram of the metal furniture manufacturing process is provided in



Figure 3-1 in which the different unit operations are shown.  The metal furniture manufacturing process



may be  divided into five main unit operations: (1) raw material preparation,  (2) cleaning operations, (3)



coating  application systems, (4) adhesive application operations, and (5) assembly. Each of these unit



operations is described briefly in the following sections.



3.3.1  Raw Material Preparation



        Raw materials generally consists of steel rods, tubes, or coiled steel  sheets. The material is cut



to size and processed through various stamping, forming, bending, and welding steps. At this point in



the process, the metal furniture unit may be completely assembled, as in the case of
                                              3-2

-------
               EP-1
                                                                            EP-4
                                                                              4
                                                                              I
                                                                            Oxidizer
                                                                             CD-2
        EP-2   EP-3    F-l
          444
                                                                           Dry Filters
                                                                             CD-I
                                                                    EP-8    F-2
                                                                      4      4
                                                                        EP-10
                                                                      Wastewater
                                                                      Treatment
                                                                        ES-17
                                                                    EP-7
                                                                      4
Machine Shop
Operations
(stamping, bending,
forming welding etc )
ES-1 through ES-6




	 »•




FP Q
Parts Cleaning

ES-7, ES-8



T7 -J














• ! i
• ! i
: ! {

Spray Coating
ES-9
EP-5
4
i
Dip Coating
ES-10, ES-11








T7 1 o oh off
ES-12
EP-6
.
i
Flashoff
ES-13


— ^-




       Packaging
      and Shipping
Touch-up/Repair
   Operations
    ES-16
  Final
Assembly
  ES-15
Drying
Ovens
ES-14
ES = Emission Source
EP = Emission Point
CD = Control Device
F = Fugitive Emissions
               Air Emissions

               Product Flow
                    Figure 3-1
   Example Flow Diagram of Unit Operations
                                                           3-:

-------
an outdoor patio chair, and ready for surface finishing.  However, other items, particularly office
furniture such as filing cabinets, will require assembly after surface coating.
3.3.2 Cleaning Operations
       Before a metal furniture part or component can be coated, its surface must be thoroughly
cleaned.  The cleaning unit operation system (UOS)5 shown in Figure 3-2 provides a representation of
a typical metal furniture cleaning operation and the steps where air emissions may occur. This operation
consists of the following basic processing stages: 1) alkaline or acid cleaning, 2) water rinse, 3)
phosphate treatment (typically iron phosphate), 4) water rinse, and 5) pretreatment and/or water rinse.
The last stage in that operation involves drying the parts in an oven.
       In the alkaline or acid cleaning stage, metal parts are sprayed with, or immersed in, a cleaning
bath to dissolve and remove oil, grease, and dirt. This bath, which can be alkaline or acidic, typically
includes one or more other ingredients such as surfactants or corrosion inhibitors. Generally, acid-
based solutions are preferred for removing corrosion and scale from metal pieces. However, because
alkaline formulations are generally somewhat milder, they are recommended for certain metal substrates
when the corrosivity of acid solutions is a concern.
       The cleaning stage is followed by a phosphate treatment stage. The purpose of this treatment is
to provide corrosion resistance to the surface of the metal part. The final pretreatment stage, if utilized,
may be a rust inhibitor or adhesion promoter.
       Following each treatment stage, the substrate is typically sent through several rinse stages in
series, which are schematically represented by one rectangle in Figure 3.2.  A counterflow rinsing
system is commonly utilized.  A counterflow rinsing system is a sequence of rinse steps in which
replenished rinse water moves in the opposite direction of the substrate flow. The parts being  cleaned
progress from dirtier to cleaner rinse water. The system maximizes water use by  adding fresh water
only at the final rinse stage in the sequence. Thus, the part is exposed to the cleanest rinse water just
before proceeding to the next treatment stage.
       In general, the chemicals used contain little organic HAP or volatile organic compound (VOC)
materials and, therefore, this type of cleaning operation generates negligible emissions.
                                              3-4

-------
                      To Atmosphere
                           A
               To Atmosphere
                    i
                                 To Atmosphere
                                      i
Uncleaned
Partg	
                                                                                                          System Boundary
                         Alkaline
                         Cleaning
Rinsing
Phosphating
Pretreatment
 or Rinsing
                    Waste    Alkaline      Waste  Water   Waste Phosphate   Waste Water     Waste  Water
                             Cleaner
                     Figure 3-2. UOS for Cleaning Operation System Prior to Coating
                                                                     Cleaned
                                                                     Parts
            physically connected unit operations
            air emissions
                                                         3-5

-------
       An alternate cleaning method uses an enclosed shot-blasting operation as the means of cleaning
prior to coating.  The operation uses steel shot (fine particles) to abrasively remove dirt and grease, as
well as to smooth rough edges and welds.  The operation can also be used to remove cured coatings
when parts require rework.  Although the steel shot is recycled back to the enclosure containing the
parts,  a small amount of particulate matter emissions is generated by this operation. However, the
cleaning operation does not involve any liquid chemicals, and no wastewater discharge is produced.6
       While the two cleaning operations discussed above result in minimal organic HAP and VOC
emissions, more significant emissions may occur from other cleaning operations including spray gun
cleaning, paint line flushing, rework operations, and touchup cleaning at final assembly.
3.3.3  Coating Application Systems
       Surface coating is accomplished by means of applying a coating to the metal part, then curing or
drying the coating. The coating itself may be in the form of a liquid or powder, and may be applied by
means of spray or dip application operations. Nearly all sprayable coatings are electrostatically
applied, as well as many dip coatings.  The presence of the electrostatic field creates an electrical
attraction between the paint, which is positively charged, and the grounded metal part and enhances the
amount of coating deposited on the part.  The distribution of coating line types as reported in the 1997
and 1998 industry questionnaire responses is shown in Figure 3-3.
       Sprayable liquid coatings are applied in a booth by manual or automatic means.  In some
instances, productivity is maximized by using automatic application followed by manual touchup.
Typically, overspray is  collected on dry filters within the booth.  Waterwash booths are less commonly
used in the metal furniture industry.  Alternatively, the overspray can be collected on a series  of baffles
installed prior to the dry filters or waterwash and collected for reuse.  Both air emissions and waste
(including spent dry filters) generated by the coating application operation are substantially reduced
through the use of this recycling method.7
                                              3-6

-------
0
0
.a

E
     30
     25
     20
     15
     10
                 I
I
/7
I
I
/I       I
/I       I
            Liquid Spray                   Adhesive Spray or Roll                 Electrocoat Dip

                             Powder Spray                     Autophoretic Dip                    Electrolytic Dip



                                                        Line Type
                         Figure 3-3. Number of Coating lines by Coating Type and Application Method
                                                          3-7

-------
       Dip coating is another available coating application method, typically used on parts that do not
require a high-quality appearance, such as interior components of a filing cabinet. The parts to be
coated are manually or automatically dipped into a tank containing the coating.  The parts are then
raised from the tank and any excess coating is allowed to drain, achieving very high coating transfer
efficiencies. Typical systems have some means of recirculation of the tank contents, filters to remove
paint sediment and solid contaminants,  and means for controlling viscosity of the fluid. Because of the
large surface area of liquid coating exposed, solvent losses occur from the tank.  To maintain the
desired coating viscosity in the tank, these losses are compensated by adding thinner (water or solvent,
depending on the coating used).
       Flow coaters were designed to overcome some of the problems associated with conventional
dip coaters.  The coating is applied to the parts at low pressure as they pass under a series of nozzles.
Typical flow coater tanks are enclosed and are smaller than the equivalent dip coating tank.  As a
result, less coating is used and less solvent is evaporated than in dip tank operations.  This modification
results in an increase in production rate and more rapid coating color changes.
       Adhesives are used primarily to attach seat cushions to the seat bottom or frame, attach cloth to
seat cushions, and attach decorative laminate to wood or metal substrates for desk tops and table tops.
The adhesive is typically spray applied to both the substrate and laminate, then the two parts are
assembled. Spray application is used when parts with a large surface area are to be  coated, such as a
desk top, and the viscosity of the adhesive is low enough to pass through a spray nozzle. Roll
application is used for high viscosity adhesives and for small surface areas.  In most instances, the
adhesive is activated by pressure, not heat.
       Electrocoating is a specialized  form of dip coating where opposite electric charges are applied
to the coating and the part.  The coating is deposited on the part by means of electrical attraction, which
produces a more uniform coating on the part than traditional dip application.  Autophoretic coating is a
dip application method where a chemical  reaction deposits the coating on the surface of the part.
Emissions of organic HAPs are considerably less than a comparable liquid spray coating process.
VOC emissions from the autophoretic process are negligible.8
                                              3-i

-------
       Powder coatings are applied almost exclusively by means of electrostatic spray in the metal
furniture industry. The electrostatic spray gun directs the flow of powder to the product. If a powder
recovery system is used, the oversprayed powder is recovered and recycled (see Figures 3-4 and 3-
5).  Powder coatings may also be applied using a dip application operation.  The part to be coated is
first heated to a temperature above the powder's melting point. The hot part is then immersed in a
fluidized bed of the powder, melting the powder in contact with it and forming a continuous coating on
the part.
       Each of the liquid and powder coatings described above is heat dried or cured after application,
with the exception of adhesives which are activated by pressure.  For liquid spray and dip coating
operation, the coated parts are typically first slowly moved through a flashoff area after the coating
application operation, which allows solvents in the coating to evaporate slowly and avoids bubbling of
the coating while it is curing in the oven.  The amount of organic FtAP and VOC emissions from the
flashoff area depends on the type of coating used, line speed, and the distance between the application
area and the bake oven.  For liquid spray applications, it is estimated that 65-80 percent of the volatiles
are emitted during the application and flashoff operations, and the remaining 20-35 percent from the
drying/curing operation. 9 However, the amount of evaporation is dependent on the type of coating and
the residence time of the part in the zone. After application of the powder coating, the metal part is
conveyed to an oven and heated to cure the powder.  This curing process melts the powder, forming a
continuous coating on the metal part.  Depending on the powder coating used, the finish may be smooth
or textured. Following the curing step, the final unit is assembled (if necessary) and packaged for
shipment.
3.3.4  Assembly
       Many metal furniture items require final assembly operations after coating. The majority of
these operations are mechanical in nature, such as assembly of drawers into file cabinets and attachment
of handles and decorative trim, and involve no emissions.
                                            3-9

-------
Uncoated Parts
Powder




t





To Atmosphere
A
;
|
Powder
Booth
Recycle


\
Wa

ste




To Atmosphere
A
i
Curing
Oven
i
Na
Ga
i
tural
s
System Be
,
Coated
Parts


/

physically connected unit operations
 air emissions
Figure 3-4. UOS for Powder Coating Application System
             (with recycle)
                                          3-10

-------
                    To Atmosphere
To Atmosphere
Powder
Uncoated Parts

i A
j System






Powder
Booth



\




Curing
Oven
i

i
Coated
Parts


Boundary
/

connected unit operations
 air emissions
                                   Waste
           Natural
           Gas
                                       Figure 3-5. UOS for Powder Coating Application System
                                                     (without powder recycle)
                                        3-11

-------
3.4 REFERENCES


1.      Preliminary Industry Characterization:  Surface Coating of Metal Furniture. U.S. Environmental
       Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park,
       North Carolina. September 1998.

2.      Documentation for Developing the Initial Source Category List. Final Report.  United States
       Environmental Protection Agency, Office of Air Quality Planning and Standards. EPA 450/3-
       91-030. July 1992.

3.      Standards of Performance for Surface Coating of Metal Furniture.  40 CFR 60, Subpart EE.

4.      Control of Volatile Organic Emissions from Existing Stationary Sources Volume HI: Surface
       Coating of Metal Furniture. EPA-450/2-77-032. U.S. Environmental Protection Agency,
       Office of Air and Waste Management, Research Triangle Park, North Carolina. December
       1977.

5.      "Standardized Accounting for a Formal Environmental Management and Auditing System,"
       Chapter 20 in Waste Minimization Through Process Design, A.P. Rossiter, ed., McGraw-Hill,
       Inc.  1995.

6.      Memorandum from Serageldin, M., EPA:ESD, to Yow, T., U.S. Furniture Industries. Report
       of the May 14,  1997,  site visit to the High Point, North Carolina, facility.

7.      Memorandum from Colyer, R., EPA:ESD, to Grady, K., HON Industries. November 3,
       1998. Report of the August 7, 1997, site visit to the Cedartown, Georgia, facility.

8.      Memorandum from Herring, L., EPA:ESD, to Mika, M.E., Steelcase, Inc. March 11, 1998,
       Report of the June 30 and July 1, 1998, site visits to the Grand Rapids, Michigan facility.

9.      Reference 4.
                                           5-12

-------
                         4.0 EMISSION CONTROL TECHNIQUES

       This chapter details techniques that are currently utilized by the metal furniture surface coating
industry to control organic hazardous air pollutant (HAP) and volatile organic compound (VOC)
emissions. Control techniques include pollution prevention measures such as coating substitution or
reformulation from conventional solventbased coatings, solvent substitution, and the use of add-on
control devices such as oxidizers, absorbers, and biorectors.

4.1 POLLUTION PREVENTION MEASURES
       Pollution prevention measures including lower organic HAP content coatings, work practice
procedures, and equipment modifications may be used to decrease organic HAP emissions from
coating application operations. Lower organic HAP coatings, such as waterbased and higher solids
content coatings, as well as powder coatings, may be used to reduce organic HAP emissions by
reducing or eliminating the organic solvent present in the coating. Work practice procedures and
equipment modifications may also result in pollution prevention when they reduce organic HAP
emissions at the source.
4.1.1  Powder Coatings
       Powder coatings have minimal organic HAP and VOC emissions (cure volatiles), generally
result in a smaller waste stream, and have higher durability as compared to traditional liquid coatings. 1
Because powder coatings are applied as dry particles, no solvent-based volatiles are released during
the application operation, and cure volatile emissions from the curing operation, if any, are generally
much less than the volatile emissions from liquid coating systems. Typically, powder overspray is
                                            4-1

-------
recycled and reused rather than discarded as waste (see Figure 3-4). Transfer efficiency for powder
without a recovery system is estimated to be approximately 60 percent, but can be greater than 99
percent with recovery.2
       Two types of powder coating resin materials exist: thermosetting and thermoplastic.
Thermosetting powders harden during heating inside a bake oven as a result of cross-linking or
polymerizing of the resin.  Common thermosetting resin types include epoxies, polyesters, hybrids,
polyurethanes, and acrylics.3 Thermoplastic powders soften with the application of heat and resolidify
during cooling, but continue to have the same chemical composition. Typical thermoplastic resins
include polyethylene, polypropylene, nylon, polyvinyl chloride, thermoplastic polyamides, and
thermoplastic polyesters.  The general metal finishing industry accounts for approximately 53 percent of
thermoset powder sales.4
       Powder coating application systems used in the metal furniture industry generally consist of a
powder delivery system, electrostatic spray gun system, and a spray booth. A powder recovery
system may also be included. Powder delivery systems utilize pneumatic pumps to transport the
powder to the spray gun.  Since powder coatings contain no solvents, organic HAP and VOC
emissions are eliminated during coating preparation and application  as compared to conventional liquid
coating systems.
        Some organic HAP and VOC emissions may be released after powder coating application
during the curing process (cure volatiles). Depending on the specific resin type and additives used in
the powder formulation, cure volatiles may be produced by two mechanisms.  First, organic
components in the formulation may be volatilized when the powder is subjected to heat without
undergoing a chemical reaction. The second mechanism is a chemical reaction between the additives in
the powder when exposed to the heat of curing that creates organic compounds, and then these organic
compounds are volatilized. The amount of cure volatiles released is dependent on many factors
including resin type, cure time, and cure temperature.
        Emissions may occur from the curing of powder coatings at temperatures greater than 160°C
(320°F).5 Two to six mass percent of urethane polyester powder coatings may be emitted as volatile
compounds in the curing step.6 Urethane polyester powders represent the powder type with the
                                             4-2

-------
greatest potential for volatile emissions due to the use of isocyanate curing agents which are blocked
with caprolactam. The unblocking reaction occurs when heat is applied in the curing oven, and
caprolactum (which is not a HAP) is released. The typical powder type used in the metal furniture
industry appears to be modified epoxy-based powders, which do not use isocyanate curing agents or
caprolactam blockers. Consequently, volatile emissions from these powder coatings are expected to
be considerably less than the urethane polyester powders.  However, application of powder coatings
may result in the release of paniculate matter emissions into the surrounding atmosphere, unless these
emissions are controlled.
       The use of powder coatings appears to be increasing. Numerous metal furniture manufacturing
facilities have converted existing liquid coating lines to powder. Powder coatings had an estimated
overall growth rate in North America of 12 percent between 1992 and 1996.7
4.1.2 Waterbased Coatings
       Waterbased coatings have recently gained acceptance as  an automotive topcoat due to their
lower VOC content levels and improved appearance compared to higher coating solids, solventbased
coatings.  This successful commercialization in the automotive industry is expected to lead to the
increased use of waterbased coatings in other industries. The use of waterbased coatings is limited in
the metal finishing industry because waterbased coatings tend to corrode mild steel and some stainless
steels.  The metal ions released from the corrosive attack can contaminate the coating and upset its
chemistry. 8
       Waterbased coatings reduce organic HAP and VOC emissions due to the reduction of organic
HAP and VOC contained in the coating as compared to conventional solventbased coatings.  They
may contain up to 80 percent water and the remaining 20 percent consists of solids, and may also
include organic HAP or VOC materials. Emission reductions may be realized during coating
preparation, application, and curing due to the overall reduction of organic HAP and VOC materials in
the coating formulation.  Some waterbased coatings may be recovered and reused, thereby decreasing
organic HAP and VOC material usage. 9
4.1.3 Solventbased. Higher Coating Solids Coatings
                                             4-:

-------
       Conventional solventbased coatings contain 25-60 percent coating solids by volume.  Higher



coating solids coatings contain greater than 60 percent coating solids by volume, and use coating resins



with highly reactive sites to help in coating polymerization. 10 Because less solvent is used with higher



coating solids coatings, surface preparation is more critical as compared to conventional solventbased



coatings. There is less solvent in the coating to self-clean the substrate surface. The surface finish



achieved in the metal furniture industry with higher coating solids coatings is similar to the surface finish



achieved with conventional solventbased coatings.



        Organic HAP and VOC emissions are reduced through the use of higher coating solids



coatings because they contain less solvent per unit volume of solids than conventional solventbased



coatings. Thus, a lesser amount of organic HAP and VOC emissions are released during coating



preparation, application, and curing.  While higher coating solids coatings typically utilize conventional



spray equipment, additional organic HAP and VOC emission reduction may be achieved during coating



application due to the reduction in number of spray applications necessary to achieve a given dried film



thickness on the substrate. Also, higher coating solids coatings generally achieve a higher transfer



efficiency as compared to conventional solventbased coatings.  These factors may lead to lower overall



coating usage as compared to conventional solventbased coatings. The reduction of organic HAP and



VOC material coupled with reduction in overall coating usage may lead to emission reductions of up to



50 percent, as compared to conventional solventbased coatings. 11



4.1.4 Work Practice Procedures



       It is estimated that 25 to 50 percent of all waste in furniture coating operations can be attributed



to poor operation and maintenance. 12  Coating waste is generated during coating material preparation,



coating application, and equipment cleaning. If coating waste is reduced, overall organic HAP and



VOC emissions from coating operations will be reduced because less organic HAP and VOC coating



material will be needed for production.  Coating waste may be reduced by effectively controlling



material preparation, maximizing the amount of coating transferred to the part through the use of more



efficient application methods and proper form (spray technique), and using proper equipment



maintenance procedures.  Six operational factors that may impact emissions are viscosity of the coating
                                             4-4

-------
material, air and fluid pressure, shape and size of the spray pattern, proper positioning of the
workpiece, operator training, and equipment maintenance. 13
       By increasing the transfer efficiency, or percentage of coating applied to the part, less coating is
needed to produce a given number of parts. This reduction in overspray and therefore, coating usage,
leads to a reduction in organic HAP and VOC emissions.
       The viscosity of the as-purchased (as-supplied) coating is an important parameter which affects
the shelf life of a coating, whereas the viscosity of the as-applied coating affects its properties after
application.  There are two methods to control viscosity: thinning of the coating with a solvent or
heating the coating. Typically, the less viscous the coating material the easier the atomization and thus,
the easier the coating application. Heating the coating material may lead to lower organic HAP and
VOC emissions, as opposed to thinning the material with a solvent, while still achieving comparable
atomization results. Air and fluid pressure may also be controlled to provide optimum atomization
results while reducing overspray.
       Operator training plays an essential role in efficient material usage and reduction of finish
defects.  Operators should be trained on the proper distance from gun tip to workpiece, position of the
gun tip, and spray gun triggering. Depending on the type of spray gun used, the gun tip should be held
approximately 20 to 30 centimeters (8 to 12 inches) from the product.  If the distance from the gun tip
to the product is too great, a decreased transfer efficiency may result because the spray pattern will be
too large, resulting in a greater amount of overspray. Running of the coating occurs when too much
coating is applied to a small surface area of the part resulting in increased rejects. This often occurs
when the spray gun is too close to the substrate. The spray gun should be held perpendicular to the
workpiece to reduce uneven coating coverage. 14  It should be triggered after the stroke is started and
released before the end of the stroke to reduce material usage and finish defects. Operator training
should be repeated periodically to reinforce proper spray coating techniques.
       Proper maintenance of equipment will also decrease material usage and defects in finished
products.  To minimize rejects and reworks due to defects in finished products from contamination
occurring at the spray booth, the floor of the spraybooth should be periodically  cleaned. Lighting
conditions should be adequate to allow the painter to better view the workpiece, thereby minimizing
                                              4-5

-------
defects from incomplete coating coverage. Turbulent air in the spraybooth should be avoided, as finish
defects may be caused when dry overspray is carried on to previously coated parts. Old peelcoat on
the walls and ceiling of the spraybooth should be removed when layers of dry overspray accumulate,
which can land on moving parts. 15
       Spray guns should be kept clean and lubricated according to manufacturer's recommendations
to ensure proper operation. If the spray gun is cleaned in solvent,  only the tip of the gun should be fully
immersed to avoid scale build-up in the gun. The gun spray pattern should be checked periodically for
wear or clogging to ensure maximum coating transfer efficiency. 16
4.1.5  Equipment Substitution
       The use of the most effective application equipment may reduce emissions of organic HAP and
VOC. Conventional systems utilize higher atomizing air pressure with typical transfer efficiencies of 25
to 40 percent.  More modern technologies, such electrostatic and high volume/low pressure (HVLP)
spray equipment, can achieve much higher transfer efficiencies. HVLP systems have improved nozzles
which provide better air and fluid flow, which allow for more gentle atomization of the air stream.
These nozzles or atomizers shape the air/spray pattern and guide the charged coating particles to the
product being coated. The electrostatic attraction of the charged  particles pulls them onto the part's
surface. Transfer efficiencies of up to 90 percent may be achieved depending on the product shape,
size, and substrate. 17 This increase in transfer efficiency translates to a decrease  in usage of materials
containing organic HAP and VOC.
       Another spray coating application technology which can reduce emissions measurably utilizes
supercritical fluid (SCF), especially carbon dioxide (CO2), in place of organic solvents to apply
coatings to metal substrates. In conventional coating formulations, solvents are used, among other
things, to reduce the viscosity of the coating low enough to allow  atomization to occur in the spray
process.  However in the SCF  process,  CO2 replaces a portion of the organic solvents and is dissolved
in the coating material to produce decompressive atomization. 18  Unlike the solvents it replaces, CO2 is
not an organic HAP or  VOC. Using CO2 as a coating solvent not only reduces the amount of organic
HAP and VOC emissions but also reduces the amount of CO2 gas that is emitted from coating
operations. One kilogram (2.2 Ibs) of organic solvent emitted to the air may eventually produces 2.0 to
                                             4-6

-------
3.0 kg (4.4 to 6.6 Ibs) of new CO2 as it is oxidized naturally in the environment, whereas with
supercritical CO2, the solvent is replaced by 1.0 kg (2.2 Ibs) or less of by-product carbon dioxide. 19
Volatile emissions in commercial applications of the SCF spray process using a variety of resin types
have been reduced from 50 to 89 percent.20 However, CO2 does not work for a few resin systems.

4.2 POLLUTANT ABATEMENT AND RECOVERY DEVICES
       In addition to pollution prevention measures, organic HAP and VOC emissions from coating
application operations can be reduced by recovering and reusing overspray or the use of add-on
control devices.
4.2.1  Recovery of Coating Overspray
       Spray booths are typically equipped with dry filters or waterwash to control overspray.  A less
common alternative is to modify the back of the spray booth with a series of baffles that run the height
of the spray booth and are several inches wide. These baffles overlap each other, forcing the
overspray-laden air to change direction several times. The overspray droplets carried in the air are
collected on the baffles. As the coating builds up on the baffles, it drips into collection troughs under
the baffles and can be collected for reuse. This reduces overall emissions because instead of the
overspray becoming waste, it is collected and reused, thereby reducing the overall amount of new
organic HAP and VOC material used in the coating application operation.
4.2.2  Add-on Control Devices
       Organic HAP and VOC emissions from coating application and curing operations can be
reduced through the use of add-on control devices. While add-on control devices are available to the
industry, the EPA is aware of only a few cases where add-on control technologies are utilized in the
metal furniture surface coating industry. Technologies applicable to the control of organic HAP and
VOC emissions include oxidation, absorption, adsorption, and bioreactors (biofilters).
       4.2.2.1 Thermal oxidation  Organic HAP, VOC, CO, and condensable organic particulate
matter emissions in an air stream may be destroyed by exposure to an oxidizing atmosphere at high
temperatures. Oxidizers may be of thermal  or catalytic design and may use primary or secondary heat
recovery to reduce energy consumption. Catalytic oxidizers  employ a catalyst to aid in the oxidation
                                             4-7

-------
reaction, which helps lower the required combustion temperature relative to that achieved in thermal
oxidizers.  Both types of oxidizers generally utilize either regenerative or recuperative techniques to
preheat inlet gas in order to decrease energy costs associated with high oxidation temperatures.
       In general, thermal oxidizers may achieve destruction efficiencies of greater than 95 percent as
applied to coating application operations with high and constant concentrations of organic emissions.21
Primary heat recovery ranges from approximately 55 to 95 percent.22,23
       4.2.2.2 Absorption.  The process of absorption consists of contacting a gas stream with a
liquid so that one or more of the components of the exhaust stream will dissolve in the liquid.  Water is
the most common absorbent, but organic solvents may also be used. Removal efficiency can be
enhanced by the addition of reactive chemical additives to the absorbent to increase solubility of the
absorbed pollutant or change the equilibrium. Some particulate matter may also be removed by the
liquid, although excessive particulate matter can lead to plugging.
       4.2.2.3 Adsorption.  The unbalanced molecular forces on the surface of solids attract and
retain gases and particulate matter that come in contact with the solid. This phenomenon is known as
adsorption.  Several materials are widely used as the adsorbent,  such as activated carbon, organic resin
polymer, and inorganic materials.24 Each has substantial surface area per unit of volume. Adsorption
has been used for coating application  operation exhaust streams  at ambient temperature to
approximately 38°C  (100°F)25
       Carbon adsorption removal efficiency is dependent upon several factors, including the flow rate
of the inlet air stream, the inlet concentration of the pollutant,  the chemical and physical characteristics
of the pollutant, and the bed design. Existing systems have generally been designed for efficiencies
between 90 to 95 percent, although efficiencies of up to 99 percent can be achieved in some cases.26
4.2.3 Other Applicable Add-on Control Technologies
       This section describes several add-on control technologies which are not currently utilized by
the metal furniture surface coating industry.  However, they are applicable control technologies for
organic HAP emissions from coatings.
       4.2.3.1 Biodegradation.  Low concentrations of organic materials in exhaust streams can be
removed through the use of biodegradation.  A biodegradation system first involves dissolving the
                                             4-8

-------
organic materials in a liquid phase. Microorganisms then metabolize the organic materials, aiding in their
biodegradation.  The organic material is oxidized at close to room temperature and breaks down into
carbon dioxide, water, and other byproducts.27

       4.2.3.2  Condensation.  Organics can be removed from gas streams by cooling the gas to a
temperature less than the dew point of the organics. The gas may be cooled with indirect or direct heat
exchangers.  The typical coolant is cold water. For low concentration streams (less than about 1
percent or 10,000 ppmv), refrigerant coolants are required. Some paniculate matter in the gas stream
may also be removed, generating a condensate sludge.
       4.2.3.3  UV Oxidation.  Oxidants such as ozone and peroxide mixed with organics in an air
stream are irradiated with ultraviolet (UV) light to produce highly reactive hydroxy and oxygen radicals.
These radicals then react with the organics in the air stream, converting them into carbon dioxide and
water. The chemistry of this process is similar to that by which sunlight degrades organics in the
atmosphere.
       UV/ozone oxidation technology has been successfully demonstrated for control of coating
application operation emissions.28 This technology can achieve VOC destruction efficiencies of
greater than 95 percent.


4.3 REFERENCES

1.      Ouellette, J. "A Major Player." Chemical Marketing Reporter. October 10, 1994; Volume 24,
       number 15, pp. SR9-SR10.

2.      Baker, N.C. "Paint Shops Take to Powder." Environment Today. September 1994. Volume 5,
       Number 9, pp.  1,  18.

3.      Anderson, B., Liberto, N. and Oliver, L. "Converting to Powder Coating - Low Production
       Volume." In: Conference Proceedings Powder Coating '91. Sponsored by The Powder
       Coating Institute.  October 7-9, 1997. p. 123-153.

4.      "POWDER COATINGS: A Pollution Prevention Alternative To Conventional Shop-Applied
       Coatings."  The Powder Coating Institute, Alexandria, Virginia. 1993.
                                            4-9

-------
5.      Chattanooga-Hamilton County (Tennessee) Air Pollution Control Bureau engineering report for
       Village Smith, L.P. June 15, 1997.

6.      Powder Coatings Group Technology Brief #8 - VOC's in Powder Coatings.  Morton
       International, Reading, Pennsylvania. October 1991.

7.      Reference 1.

8.      Bankert, P. "Water-Borne Paint Circulation."Metal Finishing. July 1997.

9.      Pollution Prevention in Metal Painting and Coating Operations: A Manual for Pollution
       Prevention Technical Assistance Providers. Chapter 6, "Alternatives to Solvent-Borne
       Coatings."  The Northeast Waste Management Officials'Association. Boston, MA.  April
       1998.

10.    Randall, P.M. "Pollution Prevention Methods in the Surface Coating Industry." Journal of
       Hazardous Materials. January 1992. Volume 29, Number 2, pp. 275-295.

11.    Reference 9.

12.    North Carolina Department of Environment, Health, and Natural Resources, North Carolina
       Division of Pollution Prevention and Environmental Assistance. Waste Reduction Fact Sheet.
       "Incentives and Techniques for Pollution Prevention in Furniture Coating Operations."
       Presented at The Furniture Industry and the Environment Conference, Hickory, NC.
       November 18, 1992.

13.    Pollution Prevention in Metal Painting and Coating Operations: A Manual for Pollution
       Prevention Technical Assistance Providers. Chapter 4, "Overview of Pollution Prevention in
       Coating Application Processes." The Northeast Waste Management Officials' Association.
       Boston, MA.  April 1998.

14.    Managing and Recycling Solvents in the Furniture Industry.  Pollution Prevention Program-
       North Carolina Department of Environment, Health, and Natural Resources.  May 1986.

15.    Joseph. R "Environmental Coating Problems." Metal Finishing. July 1996.

16.    Reference 12.

17.    Adams, J., "Reducing Spray Booth VOC's."  Industrial Finishing. March 1990.
                                           4-10

-------
18.    Busby, D.C. and Woods, RJ. "Cost-Effective Coating with Supercritical Fluid Technology."
       Materials Performance. November 1995.

19.    Pollution Prevention in the Paints and Coatings Industry. U.S. Environmental Protection
       Agency, Research Triangle Park, North Carolina. Publication No. EPA-625/R-96/003.
       September 1996.

20.    Reference 18.

21.    Lukey, M.E. and Reed, G.T., "Add-On VOC Control Options for the Surface Coating
       Industry." For Presentation at the Air & Waste Management Association's 87th Annual
       Meeting & Exhibition, June 19-24, 1994, Cincinnati, Ohio.

22.    Hussey, Frank, "Emissions Reduction Options for Painting Operations." Industrial Paint and
       Powder. March 1995. pp. 12-16.

23.    Reference 21.

24.    Perry, R.H. and Green, D.W., eds. Perry's Chemical Engineers Handbook: Sixth Edition
       (New York:  McGraw-Hill, 1984).

25.    Reference 22.

26.    Reference 21.

27.    United States Environmental Protection Agency.  1997 Air Biotreatment Meeting—US. EPA,
       NC (a government/industry/academia partnership), June 3, 1997, Research Triangle Park,
       North Carolina. URL: http://www.epa.gov/ttnatw01/bio/bio_rsch.html

28.    Ayer, J. "Controlling VOC and Air Toxic Emissions from Aircraft Refinishing Facilities - A
       New Approach." For Presentation at the Air & Waste Management Association's 90th Annual
       Meeting & Exhibition, June 8-13, 1997, Toronto, Ontario, Canada.
                                          4-11

-------
         5.0  NATIONWIDE BASELINE CHARACTERISTICS AND MODEL PLANTS







       The purpose of this chapter is to present the nationwide baseline characteristics of the metal



furniture surface coating industry and the methodology used to estimate each characteristic.  This



chapter also presents the methodology used to characterize model plants to represent the industry.  The



different types of model plants (based on size) were first determined, and then the values of model plant



parameters that affect the level of emissions were calculated. The baseline characteristics, along with



the model plants, provide a reference point against which impacts of regulatory alternatives being



considered are compared in Chapters 7 through 9.







5.1 INTRODUCTION



       Because of the large number of facilities in the metal furniture source category, a plant-specific



estimation of impacts was  not feasible.  Therefore, a model plant approach was selected to estimate the



impacts.  A model plant does not represent any single actual facility, but rather it represents a range of



facilities with similar features that may be impacted by the standards.  Each model plant is characterized



in terms of facility size and other parameters that affect estimation of emissions, control costs, and



secondary environmental impacts. This approach works well even when there are a large number of



facilities involved, as in the metal furniture surface coating source category. It is also an efficient



approach for estimating plant-level and nationwide impacts of control options when reliable data from



all potentially impacted facilities in a source category are not available or are difficult to obtain, which is



the case for the metal furniture source category. Thus, the use of model plants provides a reasonable



estimate of plant-level and nationwide impacts of control  options that are representative of the source




                                              5-1

-------
category without having to simulate the effects of applying control options on all potentially impacted
facilities in the source category.  The control options are similar across the model plants and are
technically feasible for all sizes of facilities, including small businesses.  The model plant approach will,
therefore, provide impact estimates that are representative of the source category. The model plants
developed for this source category incorporate the baseline characteristics presented here.

5.2 DATA SOURCES
        The primary data source used to estimate the total number of metal furniture facilities was the
1997 U.S. Census Bureau's Economic Census, Manufacturing Industry Series 1 because it provided the
most comprehensive determination of the number of facilities by Standard Industrial Classification (SIC)
and North American Industry Classification System (NAICS) codes.  To determine the percentage of
facilities located in urban and rural areas,  the American Business Index (ABI) database2 was utilized
because this is the only comprehensive listing of facility names and addresses by SIC code that was
found (at the time this analysis was performed, no databases were found that linked NAICS codes with
facilities names and addresses).  Then,  the EPA's Toxics Release Inventory (TRI) database3 was used
to determine the split between major and area sources.4  The TRI database was the only database
found that provided speciated emissions data organized by facility name and also included addresses
and SIC codes (the TRI database did not include NAICS codes). The primary data source for
estimating emissions was the database created from the responses to the industry questionnaires
conducted in June 1997 and June 1998. The only other readily available and comprehensive database
of HAP emissions information by SIC  or NAICS code was contained in the TRI database. The
questionnaire database was selected  over the TRI database because the questionnaire responses
contained detailed information by unit operation and represented the most accurate detailed information
available.  The following sections discuss each data source in more detail.
5.2.1 Economic Census
                                             5-2

-------
       The Economic Census provides a variety of information arranged by NAICS code.  The



census data were used to determine the total number of facilities for each of the NAICS codes relevant



to the metal furniture industry.



       Fifteen Standard Industrial Classification (SIC) codes that include metal furniture parts or



products were identified. Six of these codes deal almost exclusively with metal furniture products (see



Appendix C, Table C-l). The other nine codes deal with a mixture of metal furniture products, as well



as products for numerous other industries (see Appendix C, Table C-2). While the first six SIC codes



constitute the majority of the industry, all of the 15 SIC codes will probably contain facilities affected by



the rule. Consequently, the baseline emissions and economic estimates  must take into account these



other nine relevant SIC codes, even though they contain some facilities  that do not produce metal



furniture. Once the NAICS codes were made publicly available, the NAICS codes corresponding to



each of these 15 SIC codes were determined. The NAICS codes were then used to obtain the



Economic Census data. Appendix C, Table C-3 shows the relationship between SIC codes used to



obtain the Economic Census data and the NAICS codes, published after 1997.  Section 5.3.2 explains



how the Economic Census data were used to estimate the number of metal furniture facilities.



5.2.2 American Business Index (ABI) Database



       The ABI database can be searched using numerous criteria, such as facility name, city where



the facility is located, sales volume, and SIC code (NAICS codes were not listed in the ABI database).



We chose SIC code for consistency with the U.S. Census Bureau data. Up to four SIC codes can be



listed for each facility in the ABI database.  Only the primary SIC code was used for this search in



order to avoid double counting of facilities.  For example, many facilities were listed with both SIC



code 3645 (residential electric lighting fixtures) and SIC code 3646 (commercial lighting fixtures). If



the search was not limited to the primary SIC code, then these facilities would have been counted under



each of the  SIC codes. This would have led to inflated estimates of the  total number of facilities.



       The ABI search yielded a list of facilities for each SIC code, along with the facility's address.



From this information, the location of each facility and whether this location was in an urban or rural



area was determined.



5.2.3 Toxics Release Inventory (TRD Database




                                            5-3

-------
       The TRI database was searched using the SIC codes in Appendix C, Tables C-l and C-2 as
the basis of the search (the TRI database does not list NAICS codes). This search yielded site-specific
TRI information for each facility, including speciated emissions data.  Using these emissions data, each
facility's total HAP emissions were determined by summing the information tabulated under the point
and fugitive (not captured) emission values for each HAP in the speciated emissions data.
       The TRI database may include multiple SIC codes for each facility. It was not possible to limit
the search to only the primary SIC code, as with the ABI database. Therefore, to avoid double
counting, the TRI listings for each SIC code were cross-referenced, and duplicate entries were
removed.
5.2.4 Industry Questionnaires
       Questionnaires were sent to a total of 39 companies, including both the lune 1997 and lune
1998 questionnaires. Responses were received from 85 individual facilities.  Of these 85 facilities, 59
were determined to be in the metal furniture source category. The industry questionnaire response
database contains the information provided by these 59 facilities.  The database was further refined by
separating the area sources from the major sources.  In order to be classified as an area source, not
only did the facility have to have HAP emissions below the 9.1/22.7 megagrams per year (10/25 ton
per year) major source threshold, but also be technologically limited from exceeding the threshold.
Technologically limited means that the facility does not have the capacity to emit HAP at a level equal to
or greater than the major source HAP threshold from the existing collocated  operations that are under
common control. For example, a facility with total HAP emissions of 1 Mg/yr that applies only powder
coatings and maintains no liquid coating application operations or other major emitting collocated
operations would be judged to be technologically limited from exceeding the major source threshold.  A
number of the facilities in the database that were judged to be area sources of HAP used powder
coatings exclusively. This analysis showed that 49 facilities in the questionnaire database were major  or
synthetic minor. 5
       Of the 49 major or synthetic minor facilities left in the database, 22 provided complete
information on their cleaning and coating operations such that total organic HAP emissions and total
                                             5-4

-------
coating solids usage could be calculated.  The information provided by these 22 facilities was used to



estimate baseline emissions.







5.3 MODEL PLANT DEVELOPMENT



5.3.1  Selection of Model Plants



       The affected source for estimating the impacts of previous metal furniture surface coating



standards6 was limited to the coating application, flashoff, and curing operations. Therefore, the model



plants developed to aid in the estimation of the impacts of these previous rules were limited to these



same operations. The EPA considered a broader affected source in this rulemaking, so the model



plants represent the combination of all unit operations (see section 5.3.3 for a description of the unit



operations) associated with coating application and cleaning operations.  The basic approach was to



develop a small number of model plants that reflect the combination of unit operations found at typical



facilities, rather than numerous model plants that include only a single unit operation (such as a series of



model plants for cleaning operations and a series of model plants for coating application and curing



operations). This provided the flexibility to evaluate regulatory alternatives that allow compliance to be



determined across all coating application, cleaning, and related operations at a facility.



       The most logical parameter on which to distinguish model plants was size.  However, there are



many ways to measure size. Annual sales was considered to be a determining factor of size, but it was



rejected because one facility could produce a high volume of low-priced products,  while another



produced a low volume of high-priced products. The overall annual sales of these two facilities may be



similar, but other representative parameters or characteristics would be very different.  The number of



employees at a facility was also considered to be a determining factor of size but was rejected because



it may not take into account the level of automation.



       Surface area coated provides the best indicator of size for the purpose of estimating emissions



because it is directly related to the amount of coating used. However, available  data on surface area



coated on a facility basis were limited.  A parameter for which data were available that serves as a



surrogate for amount of surface area coated was the volume of coating solids (nonvolatiles) used. In
                                             5-5

-------
general, the dry film coating thickness is relatively uniform across metal furniture product types such that



the volume coating solids used is an adequate indicator of size for the model plants.



        To span the range of types and sizes of facilities in the source category, three model plants were



developed. These model plants were distinguished by size as measured by the total volume coating



solids used.  The three model  plants are referred to as small, medium, and large.  Figure 5-1 presents



the coating solids usage for the facilities in the industry questionnaire database, ranked from lowest to



highest usage.  The facilities fell into two general groups-facilities with coating solids usage above



100,000 liters per year and those below.  Based on the knowledge of the industry gained primarily



through site visits, these two groups did not appear to adequately describe the range of facility sizes



observed. Therefore, three groups were developed.



        The first was for small facilities similar to ones observed during the site visits. From information



obtained during the site visits, the coating solids usage for small facilities (i.e., primarily privately held



companies consisting of a single manufacturing location) could be very low, in some cases no more than



about 1,000 liters/yr. However, the coating solids usage is highly dependent on the type of coating



used.  For example, a lacquer,  which is a solution of high molecular weight polymers, will need more



solvent to dissolve the polymer. Hence, it will contain much less coating solids than a two-reactant



coating  system that polymerizes after application. The type of product produced and production



volume are two other parameters that will also affect the coating solids usage. Based on these factors,



an upper limit for coating solids usage of 40,000  liters/yr for this type of small facility was believed to be



reasonable and was used to define the upper limit of the small model plant.  This is also shown in the



large cluster of facilities in Figure 5-1 below the 40,000 liter/yr level.
                                              5-6

-------
                                       Total Solids Usage

                                    Liquid and Powder Coatings
(U  T3
«*   W

!  I
P
"o
C/)
500
400
300
200
100
                                                                                       n
        CM
        O
        CO
            LU
                         CO
                         CD
                         9
                         LLJ
                        CD
                        O
                        LU
CO

CO

9
LLJ
CD
O
                                 LLJ
    06
    O
CL
9
oo
9
                 CO
                 o
                                                  LLJ
                         CO
                         O
                                                      LLJ
9
oo
9
    gCD
    9
LLJ   LLJ
             9
                                          CO   CO
X
O
06
                                              Facility ID
            Figure 5-1. Total Coating Solids Usage by Facility for Liquid and Powder Coatings
                                             5-7

-------
       The next model plant to be defined was the large model plant. This model plant is



characterized by the larger, corporate-owned facilities that produce large quantities of standard metal



furniture products, typically for home and office use.  Based on site visits and the information in Figure



5-1, this model plant was best defined as using 100,000 liters/yr or more of coating solids.



       For the remaining facilities, those using between 40,000 and 100,000 liters/yr of coating solids,



no distinguishing requirements in terms of coatings used or products produced were observed to



warrant further division.  Thus, this group was selected as representative of the medium model plant.



5.3.2 Nationwide Number of Facilities



       The total nationwide number of facilities corresponding to each  model plant size was estimated



using the U.S. Census Bureau's Economic Census data.7 The facilities  in the Toxic Release Inventory



System (TRIS) databases were then used to determine the overall percentage of major and minor



(area) sources of organic FLAP emissions in the metal furniture manufacturing industry. The percentage



of major sources from the TRIS database was applied to the total number of sources in the census data



to give a nationwide estimate of 655  major sources.9 Then, the nationwide number of facilities that fell



into the small, medium, and large model plant categories was determined based on the corresponding



size distribution of facilities in the industry questionnaire responses. The small model plant group



accounted for 45 percent of the facilities, while the medium and large model plant groups accounted for



32 and 23 percent, respectively.  Using these percentages, the estimated nationwide number of major



sources by model plant size was 295  small, 209 medium, and 151 large facilities.



       The Economic Census data were used as the primary source of information for the number of



metal furniture facilities in the United States. A search was performed for each of the SIC codes listed



in Appendix C, Tables C-l and C-2  (when this search was performed, NAICS codes were not



included in the Census data).  While this search could have been limited by using number of employees



(for example, excluding facilities with less than 5 employees on the assumption that they would not be



major sources), the decision was made to include all facilities, regardless of size, because the Economic



Census data do not contain information that can be used to estimate emissions.  Instead, the distinction



between major and area sources was made using the TRI data as described below.
                                             5-i

-------
       Three of the SIC codes in Appendix C, Table C-2 were not considered after reviewing the



Economic Census data.  Two of these SIC codes, 3499 and 3999, represent general categories of



facilities that cannot be classified under more specific codes.  Many of the non metal furniture products



under these SIC codes would likely be regulated under the miscellaneous metal parts and products



source category. It is expected that there will only be a few metal furniture manufacturers that could not



be classified under the remaining codes identified in Appendix C, Table C-l and C-2.  Because of the



large number of facilities listed under these two SIC codes and that many of the facilities would clearly



be outside of the scope of the metal furniture source category, including them in the baseline number of



facilities would bias the estimate of nationwide impacts.  Consequently, excluding these two codes will



not have a significant effect on the estimate of the nationwide number of facilities.  For the third SIC



code, 7641, census data were not available on facilities by primary SIC  code. Appendix C, Table C-4



lists the estimated  number of facilities for the remaining SIC codes.



       A major drawback of using the Economic Census data is that it provided no information on the



level of emissions. Such information was needed to determine the number of major sources (the



number of sources that will be affected by the proposed rule). The TRI  database was used to



determine emissions because it was the most readily available source of speciated HAP emissions data



on a facility basis.  The TRI database does have limitations (e.g., not all section 112(b) HAP are



included under TRI), but it was the best source of speciated emissions that was readily available.



       The TRI database was searched using each of the metal furniture SIC codes (TRI data by



NAICS codes were not available). For each facility returned under these searches, the speciated air



emissions data were obtained.  The TRI "point" and "fugitive" emissions  for each HAP were summed



and then the result was compared to the 9.1/22.7 megagrams per year (10/25 tons per year) major



source threshold.  If the HAP emissions were above the threshold, the facility was considered to be a



major source. The percentage of TRI-reporting facilities that were  major sources was then calculated,



as well as the percentage that were area sources. Applying these values to the number of facilities



obtained from the Economic Census data, an estimate was made of the total number of major and area



sources nationwide.  These values are presented in Appendix C, Table C-5.
                                             5-9

-------
       To determine the number of area sources located in urban areas, the ABI facility lists by SIC



code were utilized (census data could not be used because it does not list individual facility names or



addresses). For each SIC code, 10 percent of the facilities were randomly selected and a



determination was made as to whether they were located in an urban area. 10 A sample of 10 percent



was chosen because available resources were insufficient to check each of the listed facilities. For each



SIC code, the percentage of facilities located in urban areas was calculated, then applied to the total



number of facilities from the Economic Census data. Appendix C, Table C-5 presents the estimates of



the total nationwide number of major sources, area sources, and area sources in urban areas.



5.3.3 Model Unit Operations



       A unit operation is an industrial operation classified according to its function in the



manufacturing process. For the purposes of the model plant and impacts analyses, the following unit



operations were considered:



•      Cleaning



•      Coating application and curing



•      Mixing and storage



•      Handling  and conveyance of waste materials



       These unit operations cover all areas of a facility that organic HAP emission control methods



affect.  Thus, by adequately describing each of these unit operations and defining their input and output



parameters on a model plant basis, an estimate can be made of the impacts the control methods will



have on the model plants.



       5.3.3.1 Cleaning Unit Operations. Cleaning unit operations encompass all production-related



cleaning activities within the model plant.  The production-related cleaning activities include cleaning of



the item being produced (including raw materials and component parts before assembly or subassembly



operations), as well as cleaning of equipment (such as spray guns, spray booths, roll coaters,  and



mixing and storage tanks).  Janitorial cleaning is excluded.



       5.3.3.2 Coating Application and Curing Unit Operations.  The coating application and curing



unit operation system includes coating application, flashoff, and drying or curing. The coating applied



may be liquid or solid (powder) and the term coating includes adhesives. It may be applied in a booth




                                             5-10

-------
or other enclosure by spray, dip, brush, roll, or any other means of transferring the coating to the



substrate.  For determining the emissions, the system of unit operations includes the application unit and



flashoff area (which is the period between coating application and the curing or drying step). It also



includes the drying or curing unit operation, whether accomplished by air drying or in an oven.



However, when any add-on control system is used the coating emissions that are captured and



destroyed will need to be accounted for and subtracted from the total emissions from the coating.



       5.3.3.3 Mixing and Storage Unit Operations. These unit operations encompass all mixing and



storage operations that involve organic HAP-containing materials, such as coatings,  solvents used for



thinning or cleaning, and other cleaning materials.  Conveying of coating and cleaning materials from



storage areas to mixing areas  or to the coating application areas are also included and represent



subcategories of these unit operations.



       5.3.3.4 Handling. Conveying, and Treating of Waste Materials.  This unit operation is



comprised of all equipment used to handle and treat organic HAP-containing waste  materials (such as



waste paint and solvents) produced by the metal furniture coating and cleaning unit operations.



5.3.4  Selection of Model Plant Parameters



       The model plant parameters are the values that will be used to estimate the impacts on a model



plant level. These parameters, shown in Table 5-1, describe the raw material usage and operational



parameters of each of the three model plants.



       The industry questionnaire response database was the primary source of data for the model



plant parameters.  Since model plant size was based on total volume of coating solids used, those



facilities that  did not provide complete information on coating solids content of their coatings were not



used to determine model plant parameters. Of the 49 facilities in the database, 22 provided complete



coating solids information and adequate information to calculate an emission rate. The 22 facilities were



divided into three groups, each one containing the facilities that fell into the range of coating solids



corresponding to small, medium, and large model plants (see  Tables 5-2 through 5-7).



       For each model plant  parameter (e.g., cleaning material usage), the arithmetic average of the



values from each group of facilities was used.  Where the parameter described discrete items, such as



coating application lines, the average value was rounded to the next highest integer.  Rounding of the




                                             5-11

-------
average model plant parameters was done in this way in order to provide a more conservative estimate



of the impacts. For example, an average value of 1.25 coating application lines per model plant would



be rounded to two coating lines. The resulting cost impacts would then reflect the cost of applying



control options to two coating application lines, rather than just one.
                                             5-12

-------
                        Table 5-1.  Summary of Model Plant Parameters
                         Based on Questionnaire Response Information"
Parameter
Small Model Plant
<40,000 liters/yr
Medium Model Plant
40,000 - 99,999
liters/yr
Large Model Plant
>99,999 liters/yr
Cleaning Unit Operations
Cleaning Material Usage (L/yr)
3,000
1,500
90,000
Coating Application Unit Operations
Liquid Coating Usage (L/yr)
Powder Coating Usage (L/yr)
Powder Coating Usage" (kg/yr)
Coating Solidsb From Liquid
Coatings (L/yr)
Coating Solids From Powder
Coatings (L/yr)
Total Coating Solids (L/yr)
Number of Liquid Coating
Lines
Number of Powder Coating
Lines
66,000
950
1,300
21,000
950
22,000
2
1
160,000
3,600
5,100
50,000
3,600
54,000
2
1
440,000
11,000
16,000
240,000
11,000
250,000
4
1
a An average powder coating density of 1.41 kg/liter was used to convert from liters to kilograms.
b Nonvolatiles (film formers).
                                            5-13

-------
            Table 5-2. Summary of Cleaning and Coating Application Unit Operations Material Usage and Emissions Data for
                                          Facilities in the Small Model Plant Designation3





Facility ID
MFA-08-CP
MFD-01
MFF-01
MFE-06-I
MFE-06-F
MFE-06B
MFE-04
MFA-08-TX
MFB-02
MFE-03-B
Average


Cleaning
Material
Usage
(L/yr)
6,057
0
0
0
2,214
0
0
7,589
13,948
0
2,981


Liquid
Coating
Usage
(L/yr)
73,080
8,417
93,992
114,335
64,988
97,663
37,654
50,972
27,339
94,504
66,294


Powder
Coating
Usage
(L/yr)
0
9,450
0
0
0
0
0
0
0
0
945

Coating
Solids from
Liquid
Coatings
(L/yr)
37,892
3,114
33,014
27,669
17,656
12,319
12,025
28,706
8,750
24,766
20,591

Coating
Solids from
Powder
Coatings
(L/yr)
0
9,450
0
0
0
0
0
0
0
0
945


Total
Coating
Solids Usage
(L/yr)
37,892
12,564
33,014
27,669
17,656
12,319
12,025
28,706
8,750
24,766
21,536



Total HAP
Emissions
(kg/yr)
4,186
1,481
5,481
4,910
11,202
13,297
1,771
7,771
3,857
21,061
7,502



Total VOC
Emissions
(kg/yr)
12,131
3,241
5,769
4,910
11,202
65,943
4,142
18,377
12,453
26,240
16,441
Normalized
Facility
Emissions
(kg HAP/L
coating
solids)
0.110
0.118
0.166
0.177
0.634
1.079
0.147
0.271
0.441
0.850
0.399
' Source:  1997 and 1998 industry questionnaire responses.
                                                             5-14

-------
            Table 5-3.  Summary of Cleaning and Coating Application Unit Operations Material Usage and Emissions Data for
                                         Facilities in the Medium Model Plant Designation3





Facility ID
MFF-03-C
MFE-06-K
MFA-08-CF
MFB-03
MFE-03-A
MFE-06-J
MFE-06-G
Average


Cleaning
Material
Usage
(L/yr)
988
0
6,664
2,991
0
0
0
1,520


Liquid
Coating
Usage
(L/yr)
96,142
206,006
99,849
74,959
188,879
148,162
333,754
163,964


Powder
Coating
Usage
(L/yr)
0
0
0
0
8,635
0
16,399
3,576

Coating
Solids from
Liquid
Coatings
(L/yr)
65,338
63,862
57,640
45,717
32,097
41,041
45,113
50,115

Coating
Solids from
Powder
Coatings
(L/yr)
0
0
0
0
8,635
0
16,247
3,555


Total
Coating
Solids Usage
(L/yr)
65,338
63,862
57,640
45,717
40,732
41,041
61,360
53,670



Total HAP
Emissions
(kg/yr)
6,154
6,300
13,910
22,880
22,362
24,713
41,046
19,624



Total VOC
Emissions
(kg/yr)
26,887
6,300
23,580
26,268
51,890
24,713
176,540
48,025
Normalized
Facility
Emissions
(kg HAP/L
coating
solids)
0.094
0.099
0.241
0.500
0.549
0.602
0.669
0.393
a Source: 1997 and 1998 industry questionnaire responses.
                                                             5-15

-------
            Table 5-4. Summary of Cleaning and Coating Application Unit Operations Material Usage and Emissions Data for
                                          Facilities in the Large Model Plant Designation3





Facility ID
MFA-07-J
MFA-08-CX
MFF-03-A
MFF-04
MFA-07-HAZ
Average


Cleaning
Material
Usage
(L/yr)
0
55,304
14,780
91,631
286,983
89,740


Liquid
Coating
Usage
(L/yr)
629,321
662,726
188,149
611,459
128,681
444,067


Powder
Coating
Usage
(L/yr)
0
0
0
0
53,085
10,617

Coating
Solids from
Liquid
Coatings
(L/yr)
305,693
396,862
127,866
258,522
89,666
235,722

Coating
Solids from
Powder
Coatings
(L/yr)
0
0
0
0
53,085
10,617


Total
Coating
Solids Usage
(L/yr)
305,693
396,862
127,866
258,522
142,751
246,339



Total HAP
Emissions
(kg/yr)
39,476
68,901
52,448
118,705
182,651
92,436



Total VOC
Emissions
(kg/yr)
132,013
165,614
52,448
179,684
220,185
149,989
Normalized
Facility
Emissions
(kg HAP/L
coating
solids)
0.129
0.174
0.410
0.459
1.280
0.490
a Source: 1997 and 1998 industry questionnaire responses.
                                                             5-16

-------
  Table 5-5.  Summary of Number of Employees, Operating Schedules, and Number of Coating Lines
                       for Facilities in the Small Model Plant Designation3'1'
Facility ID
MFC-02
MFE-04
MFB-02
MFA-11B
MFE-06B
MFD-01
MFA-11A
MFE-03B
MFE-06F
MFA-08-TX
MFF-05A
MFE-06I
MFF-05B
MFA-08-CP
MFF-01
MFE-06D
Total Number
of Employees
130
130
475
100
75
227
527
343
103
91
498
96
433
720
240
314
Operating Schedule
Hours/Day



18


16



24

24



Days/Week



5.5


5



6

6



Number of Coating
Lines
Liquid
1
2
1

1
1

2
1
2
3
1
2
1
1
1
Powder
1
1



1




1

2
2


a Source: 1997 and 1998 industry questionnaire responses.
b Where no entry is made in this table, the information was not supplied by the facility in their
questionnaire response.
                                            5-17

-------
  Table 5-6.  Summary of Number of Employees, Operating Schedules, and Number of Coating Lines
                      for Facilities in the Medium Model Plant Designation3'1'
Facility ID
MFA-08-CF
MFA-08-GA
MFB-03
MFE-03A
MFE-06G
MFE-06J
MFE-06K
MFF-03C
Total Number
of Employees
582
300
203
650
900
171
270
116
Operating Schedule
Hours/Day








Days/Week








Number of Coating
Lines
Liquid
2
2
1
4
2
3
1
1
Powder



1
1



a Source: 1997 and 1998 industry questionnaire responses.
b Where no entry is made in this table, the information was not supplied by the facility in their
questionnaire response.
  Table 5-7.  Summary of Number of Employees, Operating Schedules, and Number of Coating Lines
                       for Facilities in the Large Model Plant Designation3'11
Facility ID
MFF-03A
MFA-07-HAZ
MFF-04
MFA-07-J
MFA-08-CX
Total Number
of Employees
265
285
490
620
659
Operating Schedule
Hours/Day



24

Days/Week



6

Number of Coating
Lines
Liquid
1
2
5
4
4
Powder

1



a Source: 1997 and 1998 industry questionnaire responses.
                                            5-18

-------
b Where no entry is made in this table, the information was not supplied by the facility in their
questionnaire response.

 5.4 NOTES AND REFERENCES

1.      U. S.  Department of Commerce, Bureau of the Census.  1997 Economic Census,
       Manufacturing:  Industry Series (Various Reports).  Washington, DC. U.S. Government
       Printing Office.

2.      American Business Index Database, 1996/1997.  First Edition. American Business Information
       Incorporated. Omaha, Nebraska.

3.      U.S.  Environmental Protection Agency. Toxic Release Inventory System.
       http://www.epa.gov/enviro/html/tris/tris_queryJava.html.  Accessed June 1997.

4.      Section 112(a)(l) of the CAA defines a major source as a stationary source that emits or has
       the potential to emit 9.1 Mg/yr (10 tons/yr) of any HAP or 22.7 Mg/yr (25 tons/yr) of any
       combination of HAP. An area source is any stationary source that is not a major source.

5.      A synthetic minor source is a source that would be a major source if uncontrolled, but limits its
       emissions below the major source thresholds with controls that must remain in place under an
       enforceable commitment.

6.      Surface Coating of Metal Furniture - Background Information for Proposed Standards. Draft.
       EPA-450/3-80-007a.  U.S.  Environmental Protection Agency.  Research Triangle Park, North
       Carolina.  September 1980.

7.      Reference 1.

8.      References.

9.      For a more detailed explanation of this procedure, see the following: Memorandum from
       Hendricks, D., EC/R Incorporated to Serageldin, M., EPA:ESD. July 10, 2000. Revised
       August 28, 2001. Nationwide Baseline Characteristics of the Metal Furniture Industry.

10.    The Office of Management and Budget's  (OMB) definition of Metropolitan Statistical Area
       (MSA) was used to delineate urban areas. OMB defines an MSA  as 1) one city with 50,000
       or more inhabitants, or 2) a Census Bureau-defined urbanized area (of at least 50,000
       inhabitants) and a total metropolitan population of at least 100,000  (75,000 in New England).
       For more information, see the Census Bureau's Internet site at
       http://www.census.gov/population/www/estimates/aboutmetro.html.

                                            5-19

-------
                               6.0 REGULATORY APPROACH







6.1 INTRODUCTION



       This chapter presents the methodology used to determine maximum achievable control



technology (MACT) floors for existing and new major sources 1 of hazardous air pollutant (HAP)



emissions in the metal furniture surface coating source category. Based on this methodology, the



MACT floor for existing major sources would be 0.12 kg organic HAP/liter coating solids



(nonvolatiles) used (1.0 Ib/gal), and for new and reconstructed major sources the MACT floor would



be 0.094 kg organic HAP/liter coating solids used (0.78 Ib/gal).  Regulatory alternatives more stringent



than the MACT floor level of control for existing sources are discussed in Section 6.4, but none of the



regulatory alternatives was determined to be feasible.



       The metal furniture surface coating source category encompasses facilities that apply coatings in



the manufacture of metal furniture or component parts of metal furniture. Metal furniture means



furniture or components of furniture constructed either entirely or partially from metal. Metal furniture



includes, but is not limited to,  components of the following types of products as well as the products



themselves:  household, office, institutional, laboratory, hospital, public building, restaurant, barber and



beauty shop, and dental furniture. Metal furniture also includes office and store fixtures, partitions,



shelves, lockers, lamps and lighting fixtures, and wastebaskets.



       The corresponding Standard Industrial Classification (SIC) codes and North American



Industry  Classification System (NAICS) codes for these products have been identified. The SIC and



NAICS codes were divided into two groups: those that are comprised almost exclusively of metal
                                              6-1

-------
furniture products (Appendix C, Table C-l), and those that are comprised of metal furniture products
as well as other products (Appendix C, Table C-2).  Appendix C, Table C-3 lists all the SIC codes
from Tables C-l and C-2 along with their corresponding NAICS codes.2

6.2 METHODOLOGY FOR DETERMINING THE MACT FLOOR
       For standards established under section 112(d) of the Clean Air Act as amended in 1990
(CAA), the minimum level of control required by the statute is commonly referred to as the "MACT
floor."  For new sources, emission standards "shall not be less stringent than the emission control that is
achieved in practice by the best controlled similar source."  For existing sources, the emissions
standards must be at least as stringent as either "the average emission limitation achieved by the best
performing 12 percent of the existing sources," or "the average emission limitation achieved by the best
performing five sources" for categories or subcategories with less than 30 sources. As explained in the
following sections, the average of the best performing 12 percent of sources was used in this analysis.
6.2.1  Description of MACT Floor Format
       The format selected for the MACT floor analysis (and the proposed standard) is the affected-
source-wide organic HAP emissions normalized by the volume of coating solids used (referred to as the
emission rate). The emission rate of a source calculated  on this basis takes into account emissions from
all operations that may emit organic HAP from the metal furniture operations associated with  coating
application and cleaning (i.e., the affected source).  This collection of operations includes cleaning,
coating application and curing (including adhesives), mixing and storage, and handling and conveyance
of waste materials. The emissions are normalized by the  volume of coating solids used within the
boundary of the affected source. Thus, the units of the emission rate are kilograms organic HAP
emitted per liter coating solids used. Facilities utilize a variety of emission control technologies, and
these technologies are reflected in a MACT floor analysis based on an affected-source-wide  emission
rate. As  shown in Table 6-1, the 22 facilities for which the emission rate could be calculated (see
Section 6.3.3  for further discussion of these 22 facilities)  used a number of different emission control
                                              6-2

-------
technologies to reduce organic HAP emissions.  These technologies included waterbased coatings, high

solids coatings, powder coatings, and add-on control devices.

           Table 6-1. Products Coated and HAP Emission Control Technology Used By
                         Facilities Included in the MACT Floor Analysis3
Facility ID
MFF-03-C
MFE-06-K
MFA-08-CP
MFD-01
MFA-07-J
MFE-04
MFF-01
MFA-08-CX
MFE-06-I
MFA-08-CF
MFA-08-TX
MFE-06-J
MFF-03-A
MFB-02
Products Coated
Lockers, racks, storage cabinets
Bedframes, bed rails, rollaway
beds, day beds
Chairs
Office furniture
Office furniture
Office and restaurant equipment,
copier stands
Framing and struts
Office furniture
Bed frames, bed rails, trundle beds,
springs
Computer office furniture
Office furniture
Bedframes, bed rails, trundle beds
Lockers, shelving, shop furniture
Residential and commercial lighting
fixtures
HAP Emission Control Technology
High solids coatings
Waterbased coatings
High solids coatings
Powder coatings
Waterbased coatings
High solids coatings
High solids coatings
Collect and reuse overspray
Automatic painting system
Waterbased coatings
Waterbased coatings
High solids coatings
Carbon adsorber/oxidizer system
Waterbased coatings
High solids coatings
High solids coatings
Waterbased coatings
High solids coatings
Collect and reuse overspray
Waterbased coatings
Powder coatings
Non-HAP cleaners
                                            6-3

-------
          Table 6-1.  Products Coated and HAP Emission Control Technology Used By
                    Facilities Included in the MACT Floor Analysis3 (cont.)
Facility ID
MFF-04
MFB-03
MFE-03-A
MFE-06-F
MFE-06-G
MFE-03-B
MFE-06B
MFA-07-HAZ
Products Coated
Lockers, storage shelves, racks
Commercial and industrial lighting
fixtures
Mechanisms for recliners and
sleepers, springs, bedframes,
institutional beds
Office furniture components
Mechanisms for recliners, rockers,
gliders, and sleepers
Mechanisms for recliners, rockers,
sleepers; baby crib spring units
Sleeper sofa mechanisms
Office furniture
HAP Emission Control Technology
High solids coatings
Waterbased coatings
High solids coatings
Waterbased coatings
Powder coatings
Improved cleaning before coating allowed
use of coatings with lower solvent content
Waterbased coatings
Powder coatings
Waterbased coatings
Dip coating
High solids coatings
Powder coatings
Source:  1997 and 1998 industry questionnaire responses.
                                          6-4

-------
       The general industry trend observed through site visits, industry questionnaires, and literature
searches is to reduce emissions by reducing the mass of pollutants in coating and cleaning materials
rather than by the use of add-on control devices.  Therefore, the MACT floor for the standards was
based on an emission rate in units of kg HAP/L coating solids, which is a production-based parameter
that can be used to compare effectiveness of various pollution prevention and other control
technologies.  The use of such an emission unit allows an affected source in the metal furniture surface
coating industry flexibility in choosing any reasonable means (including but not limited to coating
reformulation, conversion to powder coating, solvent elimination, work practices, and capture systems
and add-on control devices) to meet the MACT floor level of control. The selected format encourages
emission reduction by reformulation but also allows the industry the flexibility to utilize add-on control
devices if desired.
       Normalizing the organic HAP emissions was necessary to compare emissions from facilities of
all sizes, as well as facilities using different coating technologies. Normalizing by the amount of surface
area coated was the preferred method because it is the one factor that is consistent across all facilities.
However, insufficient surface area data were available for all facilities in the MACT floor analysis. As a
substitute for surface area, the volume coating solids used was selected as the normalizing factor. The
volume coating solids used is an adequate measure of surface area coated since the average dry film
thickness of coatings on most metal furniture products is generally consistent.
6.2.2 Definition of "Average"
       As discussed above, the minimum level of control  defined under section 112(d) of the CAA is
commonly referred to as the MACT floor.  The term "average" is not defined in the CAA. In a Federal
Register notice published on June 6, 1994 (59 FR 29196), the EPA announced its conclusion that
Congress intended "average," as used in section 112(d)(3), to be the mean, median, mode, or some
other measure of central tendency.  The EPA concluded that it retains substantial discretion, within the
statutory framework, to set MACT floors at appropriate levels, and that it construes the word
"average" (as used in section 112(d)(3)) to authorize the EPA to use any reasonable method, in a
                                              5-5

-------
particular factual context, of determining the central tendency of a data set.  As discussed in Section
6.3.3, the arithmetic mean was used as the average in this MACT floor analysis.
6.2.3 Meaning of "Best Performing" and "Best Controlled"
       For the MACT floor analysis, performance was evaluated in terms of an affected-source-wide
estimated emission rate of mass of organic HAP emitted per volume of coating solids used.  The "best
performing" metal furniture manufacturing facilities were judged to be those with lower emission rates
estimated on this basis.
       Section 112(d)(3) of the CAA requires that the basis of the MACT floor for new sources be
"the emission control achieved in practice by the best controlled similar source."  The facility with the
lowest estimated affected-source-wide emission rate was considered to be the best controlled source
for this analysis.

6.3 COLLECTION AND ANALYSIS OF DATA
6.3.1 Site Visits
       Site visits were conducted at nine separate facilities (comprising eight companies) that apply
coatings to a variety of relevant products including stadium seating, residential furniture, office furniture,
and recliner mechanisms. These facilities ranged from a small plant with less than 100 employees to a
major manufacturing facility comprised of multiple buildings employing over 1,000 people.
       The purpose of the site visits was to obtain information on facility operations, with particular
emphasis on cleaning operations and coating application and curing systems.
6.3.2 Industry Questionnaires
       Eight companies were selected3 to receive the initial metal furniture industry  questionnaire in
June 1997,  in an effort to obtain a broad representation of the metal furniture surface coating industry.
The initial questionnaire requested information about the general facility, unit operations (including
description, flow diagrams, coating specifications, type of parts and substrate material coated, and
waste handling procedures), control measures and applicable regulations, and collocated sources.
                                              6-6

-------
       Various methods were used to select the recipients of the initial questionnaire, with the desired
result being a representative cross-section of the industry. Four companies under SIC code 2522
(office furniture, except wood) received the initial questionnaire as a result of their position as leading
manufacturers of office furniture. Through discussions with the National Association of Store Fixture
Manufacturers (NASFM), two store fixture manufacturing companies (SIC code 2542) were identified
to receive the initial questionnaire.  A product search was performed on the Dental Manufacturers of
America (DMA) website4 for manufacturers of dental and laboratory furniture. One dental chair
manufacturer (SIC code 3843) and one laboratory furniture manufacturer (SIC code 3821) were
chosen from the compiled list to receive the initial questionnaire.
       Thirty-three companies received a second, more comprehensive industry questionnaire that was
sent in June 1998.  The following metal furniture industry segments were surveyed: household, office,
and public building furniture; store fixtures, partitions, and shelves; residential and commercial lighting
fixtures; laboratory and dental furniture; furniture repair; metal furniture parts and hardware; and
miscellaneous metal furniture products. The June 1998 questionnaire generated responses from 75
facilities.
6.3.3 Data Analysis to Determine the MACT Floor
       For both the June 1997 and June 1998 questionnaires, responses were received from a total of
85 facilities.  Fifty-nine of these facilities were determined to be metal furniture surface coating facilities.
For each of these facilities, the questionnaire responses were used to determine the potential to emit
HAP and current permit restrictions.  Where the potential to emit HAP was above the major source
thresholds and there were no reported permit restrictions limiting the emissions below the threshold, the
facility was identified for the purpose of this analysis as a major source. Those facilities with potential
HAP emissions above the major  source threshold, but which also reported permit restrictions limiting
HAP emissions below the major source threshold, were identified as synthetic minor6 sources in this
analysis.  Those facilities with a potential to emit HAP below the major source threshold were identified
as area sources.  In some instances where data on potential to emit were not available, the
determination of major or area source status was made on the basis of technological limitations.  For
                                               5-7

-------
example, a facility that reported actual HAP emissions well below the 10 or 25 tons per year major
source threshold, used all powder coatings, and did not report any liquid coating capability was judged
technologically limited from emitting HAP at major source levels.  Such a facility was considered to be
an area source. A total of 49 facilities were assumed to be non-area sources, and data from these
facilities were used to develop a database of coating and cleaning material information (questionnaire
response database).
       6.3.3.1 General Data Set Used.  The purpose of the data collection effort was to obtain
representative data to characterize the metal furniture surface coating industry, and culminated in a
comprehensive database of facility characteristics, material usage, and HAP emissions. The information
contained in this database was used to calculate the facility emission rates used in the MACT floor
analysis.  Data gaps, resulting primarily from  incomplete questionnaire responses, limited the number of
facilities for which an emission rate could be  calculated. Typically, these data gaps consisted of material
usage, HAP content, or coating solids content.  As a result of these data gaps, facility emission rates
could be calculated for only 22 of the 49 facilities in the database.
       6.3.3.2. General Procedure for Calculation of the MACT Floor. The calculation of the
affected-source-wide organic HAP emissions for each facility was accomplished by assuming that 100
percent of the organic HAP components in all cleaning materials (including surface preparation),
thinners, and coatings (including adhesives) are emitted. 7  For the one facility in the MACT floor data
set that used an add-on control device, the reported capture and control  efficiencies were used to
determine actual emissions.  These emission values were then normalized for each facility by the volume
of coating solids used. Because the format of the MACT floor was an affected-source-wide emission
rate based on the materials used, emissions from each individual emission point were not calculated.
       The facilities were then ranked from the lowest emission rate to the highest (see
Table 6-2). To determine the "average emission limitation achieved by the best performing 12 percent
of existing sources"  as the MACT floor is defined by section 112 of the CAA, an arithmetic mean was
used in this analysis.

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                     Coating Solids Usage for the MACT Floor Determination
Number
Facility ID
Product Description
Facility Status for
FLAP Emissions
Total HAP
Emissions
(kg), (1)
Total Coating Solids
Volume
(L), (1)
Normalized Facility
Emission Rate
(kgHAP/L
coating solids)
Major and synthetic minor facilities that reported all information necessary to calculate the normalized facility emission rate
1
2
3
4
5
6
VIACT Floor =
7
8
9
10
11
12
13
MFF-03-C
MFE-06-K
MFA-08-CP
MFD-01
MFA-07-J
MFE-04
0.116
MFF-01
MFA-08-CX
MFE-06-I
MFA-08-CF
MFA-08-TX
MFF-03-A
MFB-02
Storage cabinets, lockers, and racks
Bedframes, bed rails, rollaway beds, and day
beds
Metal office furniture, chairs
Modular furniture, bookcases, chairs, tables,
desks, partitions, file cabinets, shelving
counters, racks, and lockers
Metal office furniture
Metal furniture parts and hardware, copier
stands, office equipment, and other misc.
metal products
(AVERAGE OF TOP SIX FACILITIES.)
Bolted framing/strut
Metal office furniture
Metal bed frames, bed rails, trundle beds and
top springs
Computer office furniture
Metal office furniture, desks, cabinets,
storage cabinets, movable walls
Fabricated metal products, lockers, shelving,
and shop furniture
Residential and commercial lighting fixtures
Major
Major
Major
Major
Major
Major
6,154
6,300
4,186
1,481
39,476
1,771
65,338
63,862
37,892
12,564
305,693
12,025
0.094
0.099
0.110
0.118
0.129
0.147

Synthetic minor
Major
Synthetic minor
Major
Synthetic minor
Major
Major
5,481
69,030
4,910
13,909
7,771
52,448
3,857
33,014
396,862
27,669
57,640
28,706
127,866
8,750
0.166
0.174
0.177
0.241
0.271
0.410
0.441
                                            6-9

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                Coating Solids Usage for the MACT Floor Determination (continued)
1 14
MFF-04
Lockers, shelving, racks and other
miscellaeous metal furniture
Major
118,705
258,522
0.459 1
Number
15
16
17
18
19
20
21
22

Facility ID
MFB-03
MFE-03-A
MFE-06-J
MFE-06-F
MFE-06-G
MFE-03-B
MFE-06B
MFA-07-HAZ

Product Descriotion
Commercial, industrial lighting fixtures
Recliner mechanisms, bed frames and rails,
spring units
Bed frames, bed rails, trundles, and mirror
supports
Office furniture components
Sofa sleeper beds and recliner mechanisms
Motion mechanisms, sleepers, baby crib
parts, and RV steps
Sleeper sofa mechanisms
Metal office furniture

Facility Status for
FLAP Emissions
Major
Major
Major
Synthetic Minor
Major
Major
Major
Major

Total HAP
Emissions
flceY m
22,880
22,362
24,713
11,202
41,046
21,061
13,297
182,651

Total Coating Solids
Volume
(LA m
45,717
40,732
41,041
17,656
61,360
24,766
12,319
142,751

Normalized Facility
Emission Rate
(kgHAP/L
coatins solids")
0.500
0.549
0.602
0.634
0.669
0.850
1.079
1.280

Major and synthetic minor facilities that reported incomplete information and for which the emission rate could not be calculated
23
24
25
26
MFE-06D (6)
MFC-02
MFA-08-GA
MFA-11A
Metal bed frames
Dental chairs/stools
Metal office furniture
Metal office furniture, wall panels
Synthetic Minor
Insufficient
Synthetic minor
Major
Insufficient
data to
calculate
87
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient data to
calculate
Insufficient data to
67,984
Insufficient data to
calculate




                                            6-10

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                Coating Solids Usage for the MACT Floor Determination (continued)
127


MFA-11B


Metal office furniture, wall panels


Major


Insufficient
data to
calculate
Insufficient data to
calculate




Number
28
29
30
31
32
33
34
35
36
Facility TD
MFE-06E
MFA-08-FP
MFA-09
MFE-02
MFE-06A
MFE-06C
MFA-08-SP
MFA-08-KP
MFA-08-DP
Product Description
Metal furniture diecasting
Metal office furniture
Metal chairs and tables
Bedding and furniture springs
Metal bedding components
Recliner and swivel chair mechanisms
Metal office furniture
Metal office furniture, movable office panels
and partitions
Metal office furniture
Facility Status for
HAP Emissions
Major
Major
Major
Insufficient
information to
determine
Major
Major
Major
Major
Major
Total HAP
Emissions
flcgim
510
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
calculate
12,242
13,738
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
calculate
Total Coating Solids
Volume
CTACn
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Normalized Facility
Emission Rate
(kg HAP/L coating
solids")









                                            6-11

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                Coating Solids Usage for the MACT Floor Determination (continued)
137


MFA-07G


Metal file cabinets, laterals, bookcases, and
chairs

Major


Insufficient
data to
calculate
Insufficient data to
calculate

1


                                            6-12

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                Coating Solids Usage for the MACT Floor Determination (continued)
Number
38
39
40
41
42
43
44
45
46
47
Facility ID
MFA-07-OS
MFA-08-AL
MFA-08-CA
MFA-10
MFB-04-LDL
MFD-04
MFF-03B
MFF-06-A
MFF-06-B
MFF-06-C
Product Description
Metal office furniture
Metal office furniture
Metal office furniture, desks, tables, file
cabinets, office panels/dividers, chairs and
systems furniture
Household furniture
Commercial and residential lighting fixtures
Metal office furniture
Rack storage units, shelving units, mezzanine
Metal store shelves
Metal store shelves
Metal store shelves
Facility Status for
FLAP Emissions
Major
Major
Major
Major
Insufficient
information to
determine
Insufficient
information to
determine
Major
Insufficient
information to
determine
Insufficient
information to
determine
Insufficient
information to
Hptprminp
Total HAP
Emissions
(ke\ m
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
calculate
6,273
4,019
Insufficient
data to
calculate
Insufficient
data to
calculate
Insufficient
data to
r.nlr.nlntp
Total Coating Solids
Volume
(LA m
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data to
calculate
Normalized Facility
Emission Rate
(kg FLAP/L coating
solids)










                                            6-13

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Normalized by
                Coating Solids Usage for the MACT Floor Determination (continued)
Number
48
49
Facility ID
MFF-05A
MFF-05B
Product Description
Custom metal merchandizing systems
Custom metal merchandizing systems
Facility Status for HAP
Emissions
Insufficient information
to determine
Major
Total HAP
Emissions fkgX (I)
Insufficient data to
calculate
Insufficient data to
calculate
Total Coating
Solids Volume
(Lim
26,652
25,000
Normalized Facility
Emission Rate
(kg HAP/L coating solids)


Area source facilities not included in the MACT floor analysis
50
51
52
53
54
55
56
57
58
59
MFB-04-COC (2)
MFA-04 (2)
MFA-5 (2), (5)
MFA-07-ALL (2)
MFB-04-LW
MFB-04-HI (2)
MFA-07-GEN (2)
MFA-07-WIN (2)
MFC-05 (2), (3)
MFC-06 (2), (4)
Commercial and residential lighting fixtures
Store fixture hardware
Metal dormitory furniture
Metal office chairs
Indoor lighting
Indoor/outdoor lighting fixtures
Metal office chairs
Metal office chairs
Dental chairs/units
Dental laboratory furniture
Area
Area
Area
Area
Area
Area
Area
Area
Area
Area
0
0
117
0
23
0
0
0
Insufficient data to
calculate
Insufficient data to
calculate
Insufficient data
to calculate
119
37,070
5,499
Insufficient data
to calculate
Insufficient data
to calculate
36,473
15,825
2,980
2,693

0.000
0.003
0.000


0.000
0.000


sTOTES:
1) Facility did not provide organic HAP content or coating solids content information in their questionnaire response.
2) Emissions from cure volatiles were not considered.
'3) HAP materials used in surface preparation - No quantity of cleaner given.
4) Xylene used in some cleaning applications - No quantity of cleaner given.
5) HAP emissions are from cleaning/surface preparation.
6) Emission rate for this facility cannot be calculated due to possible collocation and data discrepancy issues.
                                            6-14

-------
Table 6-2. Facility Cleaning and Coating Application Operations Organic HAP Emissions Norm
                Coating Solids Usage for the MACT Floor Determination (continued)
                          6-15

-------
       The mode and median were also considered to calculate the average emission limitation.
However, both of these indicators of central tendency are more relevant in other situations. The mode
concept was not appropriate for this data set because each value in the data set is unique; thus, there
was no emission rate that appeared frequently.  The median concept is often used when selecting
between control technologies rather than for determining an average emission rate.  Such an analysis
allows the MACT floor level of control to correspond directly to a control technology. Since this
MACT floor analysis did not involve selection of a particular control technology, using the median of
the data set was not considered an appropriate means of determining average emission limitation. Also,
there was no indication that viable control technologies would be excluded by choosing one calculation
methodology over the other. Thus, there was no compelling reason to use the mode or the median, and
the arithmetic mean was chosen as being the most representative methodology for determining the
average of the data set.
       The MACT floor for existing sources was thus determined by the arithmetic mean of the
affected-source-wide organic HAP emission rates of the top 12 percent of these facilities, which were
the top six facilities (12 percent of 49) shown in Table 6-2.8 This mean value, which is the existing
source MACT floor, is 0.12 kg organic HAP/liter coating solids used (1.0 Ib/gal). The
MACT floor for new and reconstructed sources, based on the best performing source in Table 6-2, is
0.094 kg organic HAP/liter coating solids used (0.78 Ib/gal).

6.4 REGULATORY ALTERNATIVES MORE STRINGENT THAN THE MACT FLOOR
       Based on information reported in industry questionnaire responses, observations made during
site visits, and information obtained through literature and database searches, several organic HAP
emission control technologies in use by surface coating industries were identified. This section presents
these technologies and evaluates whether each is technically feasible for implementation by the metal
furniture  surface coating industry. For those technologies determined to be technically feasible, further
analysis was performed to determine if they can effectively reduce organic HAP emissions to a level
                                            6-16

-------
below that represented by the MACT floor technology (low organic HAP content coatings) and to



determine the cost of such reduction.
                                           6-17

-------
6.4.1 Organic HAP Emission Control Technologies (by coating type)
       6.4.1.1 Powder Coatings for Thermal/IR Cure.  Powder coatings cured by thermal means
(convection heating) or infrared (IR) radiation (or a combination of both) have minimal organic HAP
and volatile organic compound (VOC) emissions (cure volatiles), generally result in a smaller waste
stream, and have higher durability as compared to traditional liquid coatings.9 Because these powder
coatings are normally applied as dry particles, no solvent-based volatiles are released during the
application operation, and cure volatile emissions from the curing operation, if any, are generally much
less than the volatile emissions from liquid coating systems.     Powder coating application systems
used in the metal furniture industry generally consist of a powder delivery system, electrostatic spray
gun system, and a spray booth. A powder recovery system may also be included.  Since powder
coatings applied in the metal furniture surface coating industry contain no solvents, organic HAP and
VOC emissions from organic solvents are eliminated during coating preparation as compared to
conventional liquid coatings.  The use of powder coatings in the metal furniture surface coating industry
appears to be increasing. Numerous metal furniture manufacturing facilities have converted existing
liquid coating lines to powder or have added powder coating lines.
       Powder coating application operations are best suited for long production runs of consistently
sized parts without color changes.  Whenever there are deviations from this "ideal," powder coating can
become a less desirable alternative to conventional liquid coatings.  For example, small production runs
with multiple color changes would require one of three means of operation. The first is to shut down
the powder coating line and perform  a complete cleaning of the spray booth and the powder handling
and application equipment. This can be a time consuming procedure  and may not be feasible in a high
production environment. Alternatively, the powder application line could be equipped with multiple
spray booths so that one can be cleaned while the others  are in production use.  This is technically
feasible as observed during a site visit to one high production facility, but the  equipment cost increases
rapidly as the number of booths increase. A third means of operation is to not recycle the powder.
However, this results in increased costs for raw materials and increases the amount of waste produced.
                                             6-18

-------
       While the color selection of powder coatings has increased significantly, it is still not possible to
produce the varied surface finishes and colors available from liquid coatings or to apply the coatings at
low dried film thicknesses achievable with liquid coatings.  Specialty finishes such as antique and
crackle, as well as the palette of designer colors offered by some metal furniture manufacturers, may
not be adequately duplicated by powder coatings.  Some metal furniture manufacturers specialize in
products with these unusual finishes.  Requiring them to use only powder coatings could eliminate the
market niche they supply. However, new powder technologies are being developed to address the
limitations that prevented some metal finishers from adopting powder coating systems. These
advancements will help reduce the time it takes to change colors under certain conditions 10 and reduce
significantly the average film thickness below the present achievable film build of approximately 2
mils. 11
       Even though there are several drawbacks to using these powder coatings, they can be
effectively used for many metal furniture coating situations.  However, they are not currently
demonstrated as a viable control option for all metal furniture products. Therefore, powder coatings
are not a technically feasible emission control option and are not evaluated further as a beyond-the-
floor option. For information purposes,  costs and emission reductions for this technology were
estimated and are presented in Appendix D.
       6.4.1.2 Powder Coatings for UV Cure.  Ultraviolet (UV) curable coatings are used for heat
sensitive substrates as they allow much lower curing temperatures (<120°C) than thermal/IR curable
coatings which may require curing temperatures of up to 220°C. These UV cured powder coatings,
formulated with chemical photoinitiators sensitive to UV light, offer the same quality advantages as
thermal/IR cured powder coatings. Upon expose to UV light, these photoinitiators form free radicals
that trigger cross-linking (curing) of the resin. 12
       In order to achieve complete curing of UV coatings, the entire coating must be exposed to the
UV light source. For metal furniture coating applications, this presents two important problems.  First,
the vast majority of metal furniture coatings are pigmented.  The pigment acts to block the UV light, and
this effect intensifies with the dry film thickness of the coating. 13 The shape of the metal  furniture
                                             6-19

-------
components also presents curing problems. Metal furniture products typically have many bends or are



box-shaped, thus "shading" some coated areas from exposure to the UV light source.  For these



reasons, UV coatings have found limited acceptance in the metal furniture surface coating industry and



are not evaluated further as a beyond-the-floor option.



       6.4.1.3 Low Organic HAP Content Liquid Coatings. A number of liquid coating technologies



have been identified through data gathering efforts that contain either very low amounts of organic HAP



or contain no organic HAP. These coating technologies fall into two general categories. The first and



most common are conventional coatings formulated with solvents that are not organic HAP (but may be



VOC), waterbased coatings, and higher coating solids content coatings. Because these coatings do not



constitute a different emission control technology than that used by the six facilities in the MACT floor



analysis, they were not considered to be a more stringent regulatory alternative. The second category



of lower organic HAP content coatings is nonconventional liquid coatings, including liquid formulations



of UV curable coatings and autophoretic coatings.  These coatings have the potential to reduce organic



HAP emissions beyond that achievable by conventional low organic HAP content coatings, so the



technical feasibility of this group of coatings was evaluated further.



       6.4.1.3.1  UV coatings.  UV curable liquid coating formulations have been used for several



decades on parts made of wood, composite, and metal. However, they are not being used in the metal



furniture industry and the same conclusions reached in the discussion of UV curable powder coatings



apply here.  Therefore, these coatings were not evaluated further as the basis for a beyond-the-floor



option.



       6.4.1.3.2  Autophoretic coatings.  The autophoretic coating process consists of a series of dip



tanks in which the parts to be coated are immersed. This process cleans the parts, then deposits the



coating solids on the surface of the parts via a chemical reaction. The coating solids are then heat



cured.  The only reported use of autophoretic coatings for metal furniture applications was a black



coating, which effectively limits its use to parts hidden from view.  Because of the limited potential use



of autophoretic coatings, they were not evaluated further as the basis for a beyond-the-floor option.
                                             6-20

-------
       6.4.1.4 Add-on Capture and Control Systems. Organic HAP emissions from coating



application and curing operations can be reduced by capturing and directing them to an add-on control



device. While add-on control devices are technically feasible options for reducing organic HAP



emissions in the metal furniture surface coating industry, information was obtained for only two cases



where they are utilized in this industry. In one case, a thermal oxidizer is used, and a thermal oxidizer



preceded by a carbon concentrator is used in the other. Other add-on controls applicable to the



control of organic HAP emissions include carbon adsorption, absorption, and bioreactors (biofilters).



       Capture systems in use by the metal furniture industry are typically limited to the spray booth in



which the coatings are applied.  While there were no reported uses of permanent total enclosures on



the coating application, flash-off, and curing operations (i.e., the coating operation) in our data gathering



efforts, such enclosures are used in other coating industries. No technical reasons have been reported



that would preclude the use of permanent total enclosures by the metal furniture surface coating



industry.



       Any add-on control device that will remove or destroy organic HAP emissions from an exhaust



stream is technically feasible for emission control of metal furniture surface coating operations.



However, the performance of some control devices is  affected by variations in the exhaust flow rate



and pollutant concentration in the exhaust stream. In the metal furniture surface coating industry,



facilities often have a number of spray booths that may or may not be operational at any one time.



Also, the actual application of coatings is not continuous. These factors lead to highly variable flow



rates and pollutant concentrations. Because of these factors, only thermal oxidizers were considered in



this analysis because they have more tolerance to handle such variation. While other control devices



may be less expensive on both a capital and annual cost basis, they would not be as likely as thermal



oxidizers, given these conditions, to be able to meet a level  of control more stringent than the MACT



floor level of control on a continuous basis. In order to achieve the level of emission reduction



necessary to be considered a more stringent regulatory alternative, complete capture of emissions from



the coating operation would be  necessary (see Section 6.4.2).  Because there were  no technical



reasons why permanent total enclosures could not be used on metal furniture surface coating
                                             6-21

-------
operations, permanent total enclosures in conjunction with thermal oxidizers were considered as a
potential regulatory alternative to achieve organic HAP emission reductions more stringent than the
MACT floor.
        6.4.1.5 Organic HAP-free Cleaning Materials. There are two basic types of items cleaned in
metal furniture surface coating cleaning operations. The first is cleaning of metal furniture parts and
assemblies prior to coating.  These cleaning operations typically involve non-HAP acid and caustic
solutions.  Because cleaning prior to coating usually does not result in HAP emissions, there was no
need to perform a beyond-the-floor analysis.
        The second type of item cleaned is equipment used in the coating application operation.  These
items generally consist of spray guns and paint distribution lines. A number of cleaning materials may be
used to clean these items, many of which are non-HAP materials. Based on information obtained from
site visits, industry questionnaire responses, and literature searches, non-HAP cleaning materials are
available and in use by the industry and the general industry trend is to increase usage of non-HAP
cleaning materials. Because it is expected that new sources will be more likely to use available emission
reduction technology, the cost analysis for new sources included the use of all non-HAP cleaning
solvents. This was the basis for determining the emission reduction and for the cost analysis of the
beyond-the-floor option.
6.4.2 Emission Reduction of Add-on Capture and Control Systems
        Model plants were developed (see Chapter 5) as a tool to estimate the impacts the standards
will have on the metal furniture surface coating industry.  A model plant does not represent any single
actual facility, but rather it represents a range of facilities with  similar characteristics that may be
impacted by the standards. Each model plant is characterized in terms of facility size and other
parameters that affect estimates of emissions, control costs, and secondary environmental impacts. The
model plant approach was used to determine whether the technically feasible organic HAP emission
control technology (add-on capture and control systems) can achieve an emission rate less than that
represented by the existing source MACT floor technology.
                                             6-22

-------
       Three model plants, distinguished by size as measured by the total volume coating solids
(nonvolatiles) used, were developed. Coating data from the industry questionnaire response database
were sorted from lowest to highest total volume coating solids used by each facility.  The volume
coating solids used ranged from a low of about 8,700 liters/yr to a high of nearly 400,000 liters/yr, and
fell into three general ranges.  These ranges were less than 40,000 liters/yr; 40,000 to 99,999 liters/yr;
and greater than 99,999 liters/yr. The model plant sizes of small, medium, and large, respectively, were
based on these ranges.  The facilities that provided complete responses to the industry questionnaires
were divided into three groups based on correspondence to the model plant sizes. Within each of these
groups, the average coating and cleaning material usage, coating solids usage, and HAP emissions were
calculated.  These values were then used to estimate the emission reduction achievable by the add-on
capture and control systems.
       A facility could choose to capture and control any  portion of the emissions from their coating
operations. 14  Table 6-3 presents the emission rates achievable by capturing and controlling certain
amounts of the emissions from coating operations at each model plant, in conjunction with converting to
all non-HAP cleaning materials. Only by capturing and controlling all coating emissions (assuming 100
percent capture of emissions from all coating operations (coating lines) and 98 percent control) can an
emission rate be achieved that represents a regulatory alternative more stringent than the existing source
MACT floor.
6.4.3  Beyond-the-floor Regulatory Alternative
       The existing and new source MACT floors,  as well as the proposed rule, are expressed in
terms of the total organic HAP emissions from the affected source normalized by the total coating solids
used.  While it may at first appear that any emission rate below the existing source MACT floor could
be considered a more stringent regulatory alternative, the emission control technology used to further
reduce emissions below the MACT floor must also be considered. The six facilities on which the
existing and new source MACT floors are based use low organic HAP content coatings. Five facilities
used low organic HAP  content coatings (either higher coating solids content or waterbased coatings)
exclusively.  The remaining facility used a combination of low organic HAP content coatings and
                                             6-23

-------
powder coatings.  The emission rates achieved for this approach ranged from 0.094 to 0.147 kg
HAP/L coating solids (the facility that used both liquid and powder coatings had an emission rate of
0.118 kg HAP/L coating solids).
       Each emission control technology may be assumed to represent a range of possible emission
rates depending on the specific coatings used, the emission capacity of a coating technology, or the
pollutant removal efficiency of an add-on capture and control unit.  Because
                                            6-24

-------
                  Table 6-3.  Organic HAP Emission Rates Achievable Through the Use of Capture and Control Systems for
                                     Existing and New Metal Furniture Surface Coating Model Plants3
Model
Plant






Small
Medium
Large
(A)
Total
Coating
Solids
Usage
(L/yr)




22,000
54,000
250,000
(B)
Total
HAP
Emissions
At MACT
Floor
Level of
Control
(kg/yr)




7,500
19,600
92,400
Amount of Organic HAP Emissions Captured and Controlled

25 Percent


(C)
HAP
Emissions
After Capture
and Control15
(kg/yr)

5,700
14,800
69,800
(D)
Emission
Ratec
(kg HAP/L
coating
solids)

0.259
0.274
0.279
50 Percent


(E)
HAP
Emissions
After
Capture and
Controld
(kg/yr)
3,800
10,000
47,100
(F)
Emission
Rate6
(kg HAP/L
coating
solids)

0.173
0.185
0.188
75 Percent


(G)
HAP
Emissions
After
Capture and
Control
(kg/yr)
2,000
5,200
24,500
(H)
Emission
Rateg
(kg HAP/L
coating
solids)

0.091
0.096
0.098
100 Percent


(I)
HAP
Emissions
After
Capture and
Control11
(kg/yr)
200
400
1,800
(J)
Emission
Rate1
(kg HAP/L
coating
solids)

0.0091
0.0074
0.0072
a Assumes that existing and new model plants would have the same coating solids usage and organic HAP emissions in the absence of a standard.
b C = (B x 25/100) x (100 - 98)/100 + B*(100 - 25)/100
c D = C/A
d E = (B x 50/100) x (100 - 98)/100 + B*(100 - 50)/100
e F = E/A
f G = (B x 75/100) x (100 - 98)/100 + B*(100 - 75)/100
g H = G/A
111 = (B x 100/100) x (100 - 98)/100 + B*(100 - 1000)/100
1 J = I/A
                                                                  6-25

-------
the beyond-the-floor emission limit cannot be set arbitrarily, an emission rate more stringent than the



MACT floor emission rate was established by evaluating the technically feasible emission control



technology (i.e., add-on capture and control systems). The analysis provided emission rates for each of



the three model plants.  While the emission rate was derived by specifically applying add-on capture



and control systems, it would not be necessary to use only this technology in actual practice. Any



emission control technology that could achieve the emission rates in Table 6-3,  Column J, would be



acceptable. For example, a facility could use powder coatings or organic HAP-free liquid coatings to



achieve these lower emission rates.



6.4.4 Costs of Beyond-the-floor Regulatory Alternative



       6.4.4.1 Basis of Cost Estimates.  The regulatory alternative cost analysis estimates the



additional costs that a source would have to incur above a set baseline to implement the emission



controls necessary to achieve the additional emission reduction beyond the MACT floor level of



control. The baseline in this analysis is the costs a facility would have incurred to achieve the MACT



floor level of control. It was assumed that this baseline is the use of conventional liquid coatings (i.e., a



combination of some or all of the following: low organic HAP content solventbased coatings, higher



coating solids content solventbased coatings, and waterbased coatings). The costs presented here for



the beyond-the-floor regulatory alternative represents the cost of adding a capture and control system



to further reduce emissions from the coating operation as well as the cost of organic HAP-free cleaning



solvents. Emission capture and control system costs were based on the installation of a permanent total



enclosure achieving 100 percent capture and a thermal oxidizer (TO) achieving 98 percent control



operating on all of the coating application lines at each model plant.  Details of the cost analysis  are



presented in Appendix E.



       6.4.4.1.1  Thermal oxidizers. Insufficient information was available from the questionnaire



responses to determine the likely flowrate to the thermal oxidizer. Hence, a flowrate of 200,000



standard cubic feet per minute (scfm) was assumed to be reasonable 15 for the large model plant if all



four coating lines  were to be totally enclosed and the thermal oxidizer cost was based on this value.
                                             6-26

-------
For small and medium model plants, a flowrate of 100,000 scfm to the thermal oxidizer was assumed
because emissions from two coating lines would be controlled from each of these model plants.
       A regenerative thermal oxidizer (RTO) was chosen for costing purposes because information
provided in the EPA OAQPS Control Cost Manual indicates that for flow rates of 50,000 scfm and
greater, an RTO should be used. 16 The COST-AIR Control Cost Spreadsheets 17 for RTOs, and the
Control Cost Manual were used to estimate add-on control costs.  The facility parameter inputs to the
spreadsheets were based primarily on data from questionnaire responses. Inputs for other parameters
such as labor rates and cost of electricity and natural gas, were based on a variety of standard
sources. 18,19,20,21 Annual costs included the annualized costs of purchased equipment, assuming an
interest rate of 7 percent and equipment life of 10 years for the RTO and 30 years for the permanent
total enclosure.
       6.4.4.1.2 Permanent total enclosures.  The cost of permanent total enclosures was estimated
using the Control Cost Manual22 in conjunction with the EPA's CO$T-AIR Control Cost spreadsheets
and spreadsheets obtained from literature sources.23 The cost associated with permanent total
enclosure installations varies with the scope of the project. The construction costs of a permanent total
enclosure is dependant upon how much construction is needed to place walls or ceilings, type of doors
used, the amount of duct work that has to be modified to meet the EPA Method 204 criteria, how
much air conditioning is needed (if any), and the degree to which modifications to the make-up air
system are required.24  The cost of the permanent total enclosure also included information from case
studies and cost factors presented in the literature and metal furniture model plant data.25,26,27  These
cost factors were related to total enclosure room volume and the volumetric air flow rate of the room
exhaust (see Appendix E).
       An average enclosure volume for a single coating line was determined from the questionnaire
response information and estimated the cost of a permanent total enclosure that would enclose this
volume. This estimate included the cost of installing spot air conditioning (localized air conditioning
where needed for operator comfort, rather than air conditioning the entire enclosure). The cost of the
air conditioning was based on the volumetric  exhaust flowrates for each coating line and cost factors
                                            6-27

-------
presented in the literature.28 This single coating line total enclosure cost was then multiplied by the
number of coating lines in a model plant to determine the model plant cost.29  Li addition to the
recovery of the capital costs of the permanent total enclosure and air conditioning, the annual cost of
electricity for the air conditioning was included in the overall annual cost of the permanent total
enclosure.
        The size of the total enclosure used in this cost analysis was based on enclosing the entire
coating application line (application booth, flashoff area, and drying/curing oven). In actual practice,
such an extensive total enclosure may not be necessary to meet the emission limit. Depending on the
solvents used in a particular coating, the majority of evaporative emissions may occur at the application
booth or in the drying oven, resulting in a smaller portion of the evaporative emissions occurring in the
flashoff area. Sufficient emission reduction may be achieved simply be constructing a total enclosure
around the application booth and routing those emissions along with the curing oven exhaust to the add-
on control device. The extent of the total enclosure needed is a case-by-case decision that can be
made only after evaluation of the particular circumstances of each facility.  In this analysis, the most
conservative assumption was used where the entire coating line would be enclosed.
        6.4.4.1.3 Organic HAP-free cleaning materials.  To determine the cost of converting to
organic HAP-free cleaning materials, the average organic HAP-containing cleaning material usage was
first determined for each model  plant, based on the industry questionnaire responses.  One prevalent
organic HAP-containing cleaning material (xylene) and one prevalent organic HAP-free cleaning
material (isopropyl alcohol) were also selected from the questionnaire responses. The costs of these
cleaning materials were $0.40 per liter of xylene and $0.80 per liter of isopropyl alcohol.30  The cost
difference between organic HAP-containing cleaning material and the organic HAP-free cleaning
material was then calculated for each model plant based on cleaning material usage.  It was assumed
that cleaning material usage would remain constant.
        6.4.4.2 Cost of Emission Capture and Control Systems. Tables 6-4 and 6-5 present the
costs, additional organic HAP emission reduction, and cost per megagram of additional emission
reduction for existing and new model plants, respectively, for the regulatory alternative of installing
                                             6-28

-------
emission capture and control systems on all coating application lines at each model plant.  For existing
model plants, the cost per megagram of additional emission reduction ranged from $71,000 to
$436,000 ($65,000 to $396,000 per ton). On a nationwide basis, the cost was
                                            6-29

-------
                          Table 6-4. Estimated Model Plant Cost per Megagram of Organic HAP Emission Reduction for Existing
                             Metal Furniture Surface Coating Facilities for Installation of Emission Capture and Control Systems
Model
Plant








Small
Medium
Large
Total
(A)
Nationwide
Number of
Facilities in
Each
Model
Plant Sizea



314
197
144
^\.
(B)
Model
Plant
Coating
Solids
Usageb
(L/yr)



22,000
54,000
250,000
""""--^
(C)
Model
Plant
Annual
Costs0
(1998$)




1,064,000
1,064,000
2,003,000
^\
(D)
Nationwide
Annual
Costs0
(1998
MM$)




334
210
288
832
(E)
Model
Plant
Capital
Costs0
($)




3,512,000
3,512,000
6,344,000
^-\
(F)
Baseline
Level of
Controld
(kg HAP/L
coating
solids used)



0.12
0.12
0.12
^^\
(G)
Regulatory
Alternative
Level of Control6
(kg HAP/L
coating solids
used)



0.0091
0.0074
0.0072
^^^^
(H)
Additional
Model
Plant
Organic
HAP
Emission
Reductionf
(Mg/yr)

2.44
6.08
28.2
^\.
(I)
Additional
Nationwide
Organic
HAP
Emission
Reduction8
(Mg/yr)


766
1,198
4,061
6,025
(J)
Model Plant
Annual Cost
per Mg of
Additional
Organic
HAP
Emission
Reduction11
($/Mg)
436,000
175,000
71,000
^^^
(K)
Nationwide
Annual Cost
per Mg of
Additional
Organic
HAP
Emission
Reduction1
($/Mg)
436,000
175,000
71,000
138,000
a Source: Memorandum from Hendricks, D., EC/R Inc., to Serageldin, M., EPS:ESD.  August 28, 2001.  Nationwide Baseline Characteristics of the Metal Furniture Industry.
bSource: Memorandum from Hendricks, D., EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September  14, 2001. Model Plants for the Metal Furniture Surface Coating Source
Category.
0 Annual and capital costs presented are the additional costs incurred beyond the baseline.
 D = (AxC)/106
d The baseline is the MACT floor level of control.  For details on the MACT floor, see: Memorandum from Hendricks, D., and Holmes, K., EC/R Inc., to Serageldin, M.,
EPA:ESD:CCPG.  September 19, 2001. Recommended MACT Floors for Existing and New Major Sources for the Metal Furniture Surface Coating Source Category.
e From Table 2, Column J.
fH=(F-G)xB/1,000
gI = AxH
hJ = C/H
'K = (D* 106)/I
                                                                          6-30

-------
                           Table 6-5. Estimated Model Plant Cost per Megagram of Organic HAP Emission Reduction for New
                             Metal Furniture Surface Coating Facilities for Installation of Emission Capture and Control Systems
Model
Plant








Small
Medium
Large
Total
(A)
Nationwide
Number of
Facilities in
Each
Model
Plant Sizea



10
5
5
^\.
(B)
Model
Plant
Coating
Solids
Usageb
(L/yr)



22,000
54,000
250,000
""""--^
(C)
Model
Plant
Annual
Costs0
(1998$)




1,064,000
1,064,000
2,003,000
^\
(D)
Nationwide
Annual
Costs0
(1998
MM$)




11
5
10
26
(E)
Model
Plant
Capital
Costs0
($)




3,512,000
3,512,000
6,344,000
^-\
(F)
Baseline
Level of
Controld
(kg HAP/L
coating
solids used)



0.094
0.094
0.094
^^\
(G)
Regulatory
Alternative
Level of Control6
(kg HAP/L
coating solids
used)



0.0091
0.0074
0.0072
^^^^
(H)
Additional
Model
Plant
Organic
HAP
Emission
Reductionf
(Mg/yr)

1.87
4.68
21.7
^\.
(I)
Additional
Nationwide
Organic
HAP
Emission
Reduction8
(Mg/yr)


18.7
23.4
108.5
151
(J)
Model Plant
Annual Cost
per Mg of
Additional
Organic
HAP
Emission
Reduction11
($/Mg)
569,000
227,000
92,000
^^^
(K)
Nationwide
Annual Cost
per Mg of
Additional
Organic
HAP
Emission
Reduction1
($/Mg)
569,000
227,000
92,000
172,000
a Fifth year after promulgation.  Source: Memorandum from Hendricks, D., and Homes, K., EC/RInc., to Serageldin, M., EPS:ESD. October 1, 2001. New Source MACT Cost
Impacts for the Metal Furniture Surface Coating Source Category.
bSource: Memorandum from Hendricks, D., EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September 14, 2001. Model Plants for the Metal Furniture Surface Coating Source
Category.
0 Annual and capital costs presented are the additional costs incurred beyond the baseline.
 D = (AxC)/106
d The baseline is the MACT floor level of control. For details on the MACT floor, see: Memorandum from Hendricks, D., and Holmes, K., EC/R Inc., to Serageldin, M.,
EPA:ESD:CCPG.  September 19, 2001. Recommended MACT Floors for Existing and New Major Sources for the Metal Furniture Surface Coating Source Category.
e From Table 2, Column J.
fH=(F-G)xB/1,000
gI = AxH
hJ = C/H
'K = (D* 106)/I
                                                                          6-31

-------
$138,000 per megagram of additional emission reduction ($125,000 per ton). For new model plants,

these costs ranged from $92,000 to $569,000 per megagram of additional organic HAP emission

reduction  ($84,000 to $517,000 per ton), and $172,000 per megagram ($156,000 per ton) on a

nationwide basis.

6.4.5 Conclusions

       While the emission capture and control regulatory alternative has been found to be technically

feasible for the metal furniture surface coating industry, the estimated cost per megagram of additional

organic HAP emission reduction above the baseline is greatly disproportional to the additional emission

reduction that would be achieved. This is true whether the analysis is on a model plant or nationwide

basis.


6.5 NOTES AND REFERENCES

1.      Section 112(a)(l) of the CAA defines a major source as a source that emits or has the potential
       to  emit 9.1 Mg/yr (10 tons/yr) of any HAP or 22.7 Mg/yr (25 tons/yr) of any combination of
       HAP.  A synthetic minor source is a source that has taken federally enforceable permit
       restrictions to limit HAP emissions below major source thresholds.

2.      Detailed information  concerning the conversion from SIC to NAICS codes can be obtained
       from the U.S. Census Bureau. See the U.S. Census Bureau's Internet site at
       http://www.census.gov/epcd/www/naics.html.

3.      At the time the questionnaire recipients were selected, data were available only by SIC code.
       No facility listings by NAICS code were available.

4.      See http://www.dmanews.org

5.      10 tons/year of any one HAP or 25 tons/year of any combination of HAP.

6.      A  synthetic minor source is a source that has taken federally enforceable permit restrictions
       such that their potential to emit HAP does not exceed the 10 tpy/25tpy major source threshold.
       Without the permit restrictions, a synthetic minor source would be a major source.
                                            6-32

-------
7.      Some of the data used for these calculations were obtained from material safety data sheets
       (MSDS). Where the MSDS provided a range rather than a single value, we used the midpoint
       of the range in the calculations.

8.      The average of the top 12 percent of sources in the database was used, not the average of the
       top five sources, because there are more than 30 sources in the source category, even though
       complete emissions data were available for less than 30 sources.

9.      Ouellette, J. "A Major Player." Chemical Marketing Reporter. October 10, 1994; Volume 24,
       number 15, pp. SR9-SR10.

10.    Guskov, S.  "Equipment Techniques for Fast Color Change." Proceedings of Powder Coating
       2000.  September 26-28, 2000. Indianapolis, Indiana, pp.  77-86.

11.    Heflin, D. "Price or Cost That is the Question." Proceedings of Powder Coating 2000.
       September 26-28, 2000.  Indianapolis, Indiana, pp. 39-46.

12.    "Manufacturing Process Advantages of UV Curable Coatings," available at
       http://www.sabreen.com/uv_curable_coatings.html.

13.    Note  12.

14.    In general terms, the coating operation consists of coating application (the spray booth),
       flashoff, and drying. One or more of these parts of the coating operation could be enclosed by
       a permanent total enclosure. For example, a facility could enclose the spray booth and flashoff,
       but not the drying oven. This would result in two emissions streams from the coating operation:
       one from the enclosure and one from the drying oven.  One or both of these emission streams
       could be vented to an add-on control device.

15.    In fact, a flow rate of 200,000 scfm may be high.  The volume of a total enclosure for a large
       model plant was estimated to be about 180,000 ft3, which would result in a complete turnover
       of the enclosure air about every minute. However, the exact size of the enclosure for any given
       facility is difficult to determine and the 200,000 scfm flow rate was used to be sure that the
       estimated costs were high enough to represent most situations that may be encountered by
       industry.

16.    U.S. Environmental Protection Agency. OAQPS Control Cost Manual, 5th Edition. EPA-
       453/B-96001. December  1995. pp. 3-43 to 3-47.

17.    U.S. Environmental Protection Agency. CO$T-AIR Control Cost Spreadsheets. Internet
       address:  http://www.epa.gov/ttn/catc/products.htmWcccinfo.  "Total Annual Cost Spreadsheet

                                           6-33

-------
       Program - Thermal Incinerators (Total flowrate > 50,000 scfm)."  Accessed on March 13,
       2000.

18.    U.S. Government, Bureau of Labor Statistics.  Table 11, Private Industry: Goods-producing
       and Service-producing Industries. Website address:
       http://stat.bls.gov/news.release/ecec.ttl.htm. Accessed on December 17, 1999.

19.    U.S. Energy Information Administration., Department of Energy, Washington, D.C. World
       wide web homepage.  Table 9.11, Natural Gas Prices. Website address:
       http://www.eia.doe.gov.emeu/mer/contents.htm. Accessed on March 4, 1999.

20.    U.S. Energy Information Administration., Department of Energy, Washington, D.C. World
       wide web homepage.  Table 9.9, Retail Prices of Electricity Sold by Electric Utilities.  Website
       address: http://www.eia.doe.gov.emeu/mer/contents.htm. Accessed on March 4, 1999.

21.    U.S. Office of Management and Budget. OMB Circular A-94, "Discount rates to be used in
       evaluating time-distributed costs and benefits." Revised October 29, 1992. Website address:
       http://www.whitehouse.gov/WH/EOP/OMB/html/circular.html. Accessed on March 4, 1999.

22.    Note 16.  pp. 3-20 to 3-64.

23.    Lukey, M. "Designing Effective and Safe Permanent Total Enclosures," Air and Waste
       Management Association. June 1993.

24.    Lukey, M. "Permanent Total Enclosures needed in Response to Subpart KK and Changes in
       Test Procedures." Paper No. 97-TA4B.05. Presented at Air and Waste Management
       Association 1997 Annual Meeting, Toronto, Ontario, Canada. June 8-13, 1997.  p. 4.

25.    Note 16.

26.    Lukey, M. "Designing Effective and Safe Permanent Total Enclosures." Paper No. 93-TA-
       33.05. Presented at Air and Waste Management Association 1993 Annual Meeting, Denver,
       Colorado. June 13-18, 1993.

27.    Lukey, M. "Five Design Options for Permanent Total Enclosures."  Paper No. VTP-69.
       Presented at Air and Waste Management Association Specialty Conference "Emerging
       Solutions to VOC and Air Toxics Control." February 26-28, 1997.  San Diego, California.

28.    Note 24.
                                           6-34

-------
29.    The small and medium model plants each have two coating lines.  The large model plant has
       four coating lines.

30.    Cleaning material costs were derived from the following sources:
       Chemical Marketing Reporter, Schnell Publishing Company,  http://www.chemexpo.com
       accessed in August 1999; SouthChem, Durham, NC. July 1999; Worth Chemical, Durham,
       NC.  August 1999.
                     7.0  ENVIRONMENTAL AND ENERGY IMPACTS


7.1 INTRODUCTION

       The purpose of this  chapter is to present the estimated environmental and energy impacts

related to implementing the maximum achievable control technology (MACT) floor level of control for

existing metal furniture surface coating facilities. The impact estimates for both new and existing sources

were based on conversion to lower organic hazardous air pollutant (HAP) content coatings (including

adhesives) and organic HAP-free cleaning materials.

       Existing sources are not expected to achieve compliance with the existing source MACT floor

level of control until the beginning of the fourth year after promulgation of a rule.  Therefore, there

would be no environmental  or energy impacts the first three years after promulgation. During each of

the fourth and fifth years after promulgation, the nationwide organic HAP emission reduction for existing

sources was estimated to be 13,900 Mg/yr (15,300 tons/yr), and the VOC emission reduction was

estimated to be 21,700 Mg/yr (23,900 tons/yr).  This represents a reduction from the baseline organic

HAP emissions of approximately 70 percent, and a reduction from the baseline VOC emissions of
                                           7-1

-------
approximately 60 percent.  There were no energy impacts or other secondary environmental impacts
associated with the conversion to reformulated coatings and cleaning materials.
       The impacts for new sources were also based on utilization of lower HAP content liquid
coatings to achieve the new source MACT floor level of control. For the 5-year period following
promulgation of the rule, it was estimated that 20 new sources will be constructed.  The organic HAP
emission reduction for these new sources was estimated to be 465 Mg/yr (511 tons/yr) in the fifth year
after promulgation of a rule implementing the MACT floor level of control. The VOC emission
reduction in the fifth year was estimated to be 380 Mg/yr (418 tons/yr).  This represents a reduction
from the  baseline organic HAP emissions of approximately 73 percent. And a reduction of 35 percent
from the  VOC baseline.  There were no energy impacts or other secondary environmental impacts
associated with the conversion to reformulated coatings and cleaning materials.
       The metal furniture surface coating source category encompasses facilities that apply coatings in
the manufacture of metal furniture or component parts of metal furniture.  Metal furniture means
furniture  or components of furniture constructed either entirely or partially from metal.  Metal furniture
includes,  but is not limited to, components of the following types of products as well as the products
themselves: household, office, institutional, laboratory, hospital, public building, restaurant, barber and
beauty shop, and dental furniture. Metal furniture also includes office and store fixtures, partitions,
shelves, lockers, lamps and lighting fixtures, and wastebaskets.
       The corresponding Standard Industrial Classification (SIC) codes and North American
Industry Classification System (NAICS) codes for these products were identified to aid in the
estimation of impacts. These SIC and NAICS codes were divided into two groups: those that are
comprised almost exclusively of metal furniture products, and those that are related to metal furniture
but only partially encompass metal furniture products.  Appendix C, Table C-l lists the product groups
and manufacturing SIC codes that are almost exclusively metal furniture.  The SIC codes related to
metal furniture and their associated relevant products are listed in Appendix C, Table C-2. Appendix
C, Table  C-3  lists all the SIC codes from Tables C-l and  C-2 along with their corresponding NAICS
codes. 1
                                             7-2

-------
7.2 DETERMINATION OF THE NUMBER OF MAJOR SOURCE FACILITIES
       As described in Chapter 5, model plants were developed to aid in the estimation of the impacts
that meeting the MACT floor level of control (0.12 kg organic HAP/liter coating solids (1.0 Ib/gal) for
existing sources, 0.094 kg organic HAP/liter coating solids (0.78 Ib/gal) for new sources) would have
on metal furniture surface coating facilities. The range of the volume of coating solids (nonvolatiles)
used by the facilities that responded to the 1997 and 1998 questionnaires was used to set the size of the
three model plants. In addition, distinctive parameters were developed for each model plant, including
average coating usage. The parameters that describe each of these model plant sizes are shown in
Table 7-1. The model plant parameters in conjunction with other industry questionnaire response data
were used to estimate the environmental impacts.  The use of model plants provides a reasonable
estimate of plant-level and nationwide impacts of control options that are representative of the source
category without having to simulate the effects of applying control options at all potentially impacted
facilities in this source category.
7.2.1.  Existing Major Sources
       The total nationwide number of existing metal furniture surface coating facilities was estimated
using the U.S. Census Bureau's Economic Census.2 The metal furniture surface coating facilities in the
Toxic Release Inventory System (TRIS) databases were then used to estimate the percentage of major
sources of HAP emissions in each SIC code. Applying the percentage of major sources from the TRIS
database to the total number of sources in the Economic Census data gave a nationwide estimate of
655 major sources.
       The nationwide number of facilities that fall into the small, medium, and large model plant
categories was then determined based on the corresponding size distribution  of facilities in the industry
questionnaire responses.  The small model plant group accounted for 45 percent of the facilities, while
the medium and large model plant groups accounted for 32 and 23 percent,  respectively. Using these
percentages, the 655 estimated nationwide number of existing major source facilities subject to the
proposed rule, broken down by model plant size, would be 295 small facilities, 209 medium facilities,
and 151 large facilities.
                                             7-:

-------
7.2.2.  New Major Sources



       Information obtained from an industry trade group, questionnaire responses, and industry



representatives, were used to estimate the anticipated number of new major sources in the metal



furniture surface coating industry.  The industry trade group provided an estimate of the percent



 increase in the sales volume measured in dollars for the United States office furniture market.4  This



information indicated a 3 to 5 percent sales volume increase in terms of current dollars, which included



the effect of price increases and inflation. This rate  of increase was assumed to
                                              7-4

-------
                     Table 7-1. Model Plant Parameters By Unit Operation3
Parameter
Small Model Plant
<40,000 liters/yr
Medium Model
Plant
40,000 - 99,999
liters/yr
Large Model Plant
>99,999 liters/yr
Cleaning Unit Operations
Cleaning Material Usage (L/yr)
3,000
1,500
90,000
Coating Application Unit Operations
Liquid Coating Usage (L/yr)
Powder Coating Usage (L/yr)
Powder Coating Usageb
(kg/yr)
Coating Solids0 From Liquid
Coatings (L/yr)
Coating Solids From Powder
Coatings (L/yr)
Total Coating Solids (L/yr)
Number of Liquid Coating
Lines
Number of Powder Coating
Lines
66,000
950
1,300
21,000
950
22,000
2
1
160,000
3,600
5,100
50,000
3,600
54,000
2
1
440,000
11,000
16,000
240,000
11,000
250,000
4
1
a Source: 1997 and 1998 industry questionnaire responses.
b An average powder coating density of 1.41 kg/liter was used to convert from liters to kilograms.
c Nonvolatiles (film formers).
                                            7-5

-------
remain consistent over the 5-year period. After further discussions with the trade group, the actual
sales volume increase anticipated for the year 2001, excluding price increases and inflation, was
determined to be a 1 to 3 percent increase. 5  This change in sales volume was assumed to be directly
related to the number of pieces produced rather than to the same number of more expensively priced
pieces, indicating a direct relationship between sales volume and production.  Further, it was assumed
that production is directly related to coating solids usage.  Coating solids usage was taken to be a
reliable indicator of overall production level.
       The annual model plant coating solids usage values were multiplied by 2 percent, which was the
midpoint of the estimated sales volume increase obtained from the industry trade group.6 The resulting
values were the estimated increase in coating solids usage due to the growth in the metal furniture
industry. Based on conversations with industry representatives, excess capacity currently exists in the
industry, but it is very difficult to quantify because of changing product types and market demands.
Additional capacity can also be added at existing facilities. This combination of existing capacity and
new capacity at existing facilities was estimated to be large enough to absorb 75 percent of the coating
solids usage increase before any new facilities would be required.7
       Hence, the predicted increase in nationwide annual coating solids usage that will not be
absorbed by existing sources was determined to be 35,000 liters for small facilities, 53,000 liters for
medium facilities, and 180,000 liters for large facilities (see Table 7-2). Comparing these increases to
the individual model plant coating solids usage values, four new facilities would be required each year to
handle the increased production (two small, one medium, and one large)8 (see Table 7-2), or 20 new
major source facilities during the  5-year period after promulgation of the  rule (see Table 7-3).

7.3 ENVIRONMENTAL IMPACTS FOR EXISTING MAJOR SOURCES
7.3.1  Organic HAP Emission Reduction
       To estimate the overall organic HAP emission reduction, the organic HAP emissions from
major sources was first estimated  assuming that the MACT floor level of control is implemented. The
22 facilities used to estimate the baseline emissions (see Chapter 5) were used
                                              7-6

-------
 Table 7-2. Estimated Annual Number of New Major Source Facilities for the Metal Furniture Surface
                                      Coating Source Category

Model
Plant





Small
Mediu
m
Large
A
Nationwide
Number of
Major
Source
Existing
Facilities"

314
197

144
B
Total
Coating
Solids
Usage per
Model
Plantb
(liters/yr)
22,000
54,000

250,000
C
Nationwide
Coating
Solids
Usage0
(liters/yr)


6,908,000
10,638,000

36,000,000
D
Annual
Increase
in
Coating
Solids
Usaged
(%)
2
2

2
E
Coating
Solids Usage
Absorbed By
Existing
Capacity6
(%)

75
75

75
Total
F
Annual
Coating Solids
Not Absorbed
by Existing
Sources
Capacityf
(liters/yr)
35,000
53,000

180,000
268,000
G
Equivalent
Annual
Number of
New
Sources8


2
1

1
4
a Total number of facilities nationwide determined from U.S. Census data. Percent of these that were major sources
determined from toxic release inventory system (TRIS) data. Breakdown by model plant size determined by industry
questionnaire response data.
b Average values from industry questionnaire response data. See Chapter 5.
cAxB
d Based on industry publication and conversations with industry representatives.
e Inferred from conversations with industry representatives.
fCx(D/100)x(l-E)
g F/B, rounded up to the next highest integer.
                                                 7-7

-------
Table 7-3. Estimated Number of New Major Source Facilities for Five Years After Promulgation for
                     the Metal Furniture Surface Coating Source Category
Year
Yearl
Year 2
Year3
Year 4
YearS
Total
Small Model
Plant
2
2
2
2
2
10
Medium Model
Plant
1
1
1
1
1
5
Large Model
Plant
1
1
1
1
1
5
Overall
Increase
4
4
4
4
4
20

-------
in this analysis.  These facilities were divided into three groups corresponding to the model plant sizes,
and the organic HAP emissions at the MACT floor level of control were calculated for each facility by
multiplying the MACT floor emission rate (0.12 kg organic HAP/L coating solids used) by the coating
solids usage of the facility. For each of the facilities in a model plant size group, the organic HAP
emissions were summed, then scaled up to nationwide levels based on the total number of facilities
nationwide corresponding to each model plant size, as was done in Chapter 5 to estimate the baseline
emissions.  The overall nationwide organic HAP emissions at the MACT floor level of control were
then determined by summing the scaled up values for each model plant size group.  Finally, the
estimated organic HAP emissions at the MACT floor level of control were subtracted from the
nationwide baseline organic HAP emissions to determine the nationwide organic HAP emission
reduction. The nationwide baseline organic HAP emissions  are presented in Table 7-4. The
nationwide organic HAP emissions after implementing the MACT floor level of control, as presented in
Table 7-5, were estimated to be 6,400 Mg/yr. This represents a reduction of 13,900 Mg/yr (15,300
tons/yr), or approximately 70 percent, from the baseline organic HAP emissions.
7.3.2  VOC Emission Reduction
       The emission reduction for the purposes of this environmental impact analysis was estimated
using the same general procedure that was used in the previous section for estimating HAP reduction.
First, we determined the nationwide VOC baseline emissions by calculating the VOC emission rate in
terms of kg VOC/L coating solids for each of the 22 industry questionnaire response facilities.  These
values were scaled by the nationwide number of facilities corresponding to each model plant size and
summed to determine nationwide baseline VOC emissions. The nationwide baseline VOC emissions
for existing sources are presented in Table 7-6.
       We then calculated the arithmetic average VOC emission rate for those facilities already
meeting the MACT floor level of control for organic HAP. This average VOC emission rate was 0.26
kg VOC/L coating solids and was considered a cut-off (or maximum) VOC emission rate. Because
these facilities used lower organic HAP content coatings to achieve the MACT
                                            7-9

-------
                  Table 7-4. Nationwide Baseline Organic HAP Emission Estimates for Existing
                            Major Source Metal Furniture Surface Coating Facilities
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
(A)
Reported Organic HAP
Emissions3
(kg/yr)
4,186
5,481
7,771
4,910
21,061
11,202
1,481
13,297
1,771
3,857
6,154
6,300
41,046
13,909
22,880
24,713
22,362
68,901
39,476
118,705
182,651
52,448
(B)
Corresponding Model Plant
Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
(C)
Nationwide Baseline
Organic HAP Emissions
(kg/yr)
2,210,000C
4,100,000d
14,000,000e
20,300,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 295/10.
d This value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 209/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 151/5.
                                                   7-10

-------
7-11

-------
             Table 7-5. Nationwide Organic HAP Emission Estimates for Major Source Metal Furniture
                          Surface Coating Facilities at the MACT Floor Level of Control
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
(A)
Calculated Organic HAP
Emissions at 0.12 kg HAP/L
coating solidsa
(kg/yr)
4,405
3,838
3,337
3,217
2,879
2,053
1,461
1,432
1,398
1,017
7,596
7,425
7,134
6,701
5,315
4,771
4,735
46,139
35,540
30,055
16,596
14,866
(B)
Corresponding Model Plant
Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
(C)
Nationwide Organic HAP
Emissions at 0. 12 kg
HAP/L coating solids
(kg/yr)
739,000C
l,304,000d
4,325,000e
6,368,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 295/10.
d This value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 209/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 151/5.
                                                   7-12

-------
                       Table 7-6.  Nationwide Baseline VOC Emission Estimates for Existing
                            Major Source Metal Furniture Surface Coating Facilities
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
Reported VOC Emissions*
(kg/yr)
12,129
5,769
18,377
4,910
26,240
11,202
3,241
65,943
4,142
12,453
26,887
6,300
176,540
23,580
26,268
24,713
51,889
104,400
132,013
179,684
220,185
52,448
Corresponding Model
Plant Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
Nationwide Baseline VOC
Emissions
(kg/yr)
4,850,000C
10,037,000d
20,800,000e
35,687,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 295/10.
d This value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 209/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 151/5.
                                                   7-13

-------
floor level of control, it was assumed that their VOC emission rate would be indicative of that achieved
by facilities complying with the MACT floor level of control.
       The average VOC emission rate was then multiplied by the coating solids usage for each of the
22 facilities to obtain the estimated VOC emissions for each facility after achieving the MACT floor
level of control for organic HAP emissions. Similar to the procedure used to estimate the organic HAP
emission reduction, the 22 facilities were grouped by model plant size, the VOC emissions were
summed for each group, then the emissions were scaled to nationwide levels.  The nationwide VOC
emissions after implementing the MACT floor level of control for organic HAP emissions,  as presented
in Table 7-7, were estimated to be 14,000 Mg/yr (15,400 tons/yr).  This represents a reduction of
21,700 Mg/yr (23,900 tons/yr), or approximately 60 percent, from the baseline VOC emissions.
7.3.3  Secondary Impacts
       Since it was assumed that no add-on control devices would be used to meet the MACT floor
level of control for existing sources, there would be no change in emissions of non-HAP pollutants
(other than VOC). No information has been obtained to indicate that there would be any change in the
amount of waste produced by coating or cleaning operations after conversion to lower organic HAP
content coating and organic HAP-free cleaning materials. In addition, no information was obtained that
indicated there would be a change in energy consumption or wastewater generation as a result of the
conversion.

7.4 ENVIRONMENTAL IMPACTS FOR NEW MAJOR SOURCES
       For new sources, the MACT floor level of control was determined to be an emission rate of
0.094 kg organic HAP/L coating solids used (see Chapter 6). The impacts were based on new
sources using low organic HAP content coatings (including adhesives) and organic HAP-free cleaning
materials.
       For the five-year period following promulgation of the rule, it was estimated that 20 new
sources will be constructed.  This growth was estimated to occur evenly, with four new sources each
                                           7-14

-------
year. Due to the increasing number of new sources, the impacts increase each year as well. For the



fifth year after promulgation of a rule, the estimated organic HAP emission reduction for
                                            7-15

-------
             Table 7-7. Nationwide VOC Emission Estimates for Existing Major Source Metal Furniture
                          Surface Coating Facilities at the MACT Floor Level of Control
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
Calculated VOC Emissions at
0.12 kg HAP/L coating solidsa
(kg/yr)
9,665
8,421
7,322
7,058
6,317
4,504
3,205
3,142
3,067
2,232
16,666
16,290
15,652
14,703
11,661
10,469
10,390
101,231
77,976
65,643
36,413
32,616
Corresponding Model Plant
Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
Nationwide VOC
Emissions at 0. 12 kg
HAP/L coating solids
(kg/yr)
1,621,000°
2,861,000d
9,488,000e
13,970,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 295/10.
d This value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 209/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 151/5.
                                                   7-16

-------
the 20 new sources was estimated to be 465 Mg/yr (511 tons/yr), which represents a 73 percent
reduction from the baseline organic HAP emissions. The VOC emission reduction for the fifth year was
estimated to be 380 Mg/yr (418 tons/yr).  This represents a 35 percent reduction from the baseline
VOC emissions.
7.4.1. Estimated Nationwide Baseline Organic HAP and VOC Emissions for New Sources
       Nationwide baseline organic HAP and VOC emissions for new sources were estimated using
the same procedure described in Section 7.3.1.  However, the emission for the 22 questionnaire
response facilities were scaled by the total number of new sources in the fifth year after promulgation of
a rule corresponding to each model plant size, rather than scaling by the number of existing sources as
was done in Section 7.3.1. As presented in Tables 7-8 and 7-9, respectively, the nationwide baseline
organic HAP emissions for new major sources were estimated to be 635 Mg/yr (698 tons/yr), and the
baseline VOC  emissions were estimated to be 1,090 Mg/yr (1,200 tons/yr).
7.4.2. Organic HAP Emission Reduction
       The organic HAP emission reduction was estimated using the lowest emission rate of the 22
facilities for which emission rates could be calculated. This facility was used as the indicator of the
emission rate for new facilities because it was used to establish the new source MACT floor and is the
only questionnaire response facility achieving the new source MACT floor emission rate.  Therefore, it
was assumed that this facility was the best available indicator of what new sources would achieve.
Although this facility fell into the medium model plant size classification, it was assumed that this
emission rate would apply equally to the small  and large model plant size classifications because the
same emission control technology (low organic HAP content coatings and organic HAP-free cleaning
materials) is expected to be used regardless of facility size.
       The organic HAP emission reduction increases each year because four new sources  are
expected to come on line each year. The values for the fifth year after promulgation are shown in Table
7-10. The fifth year after promulgation was chosen for consistency with the  cost impacts (Chapter 8),
which are also presented on a fifth-year basis.  As presented in Table 7-10, the organic HAP emissions
for the 20 new sources in the fifth year after promulgation of a rule were
                                            7-17

-------
                    Table 7-8. Nationwide Baseline Organic HAP Emission Estimates for New
                            Major Source Metal Furniture Surface Coating Facilities
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
(A)
Reported Organic HAP
Emissions*
(kg/yr)
4,186
5,481
7,771
4,910
21,061
11,202
1,481
13,297
1,771
3,857
6,154
6,300
41,046
13,909
22,880
24,713
22,362
68,901
39,476
118,705
182,651
52,448
(B)
Corresponding Model Plant
Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
(C)
Nationwide Baseline
Organic HAP Emissions
(kg/yr)
75,000C
98,000d
462,000e
635,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 10/10.
dThis value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 5/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 5/5.
                                                   7-18

-------
7-19

-------
                        Table 7-9. Nationwide Baseline VOC Emission Estimates for New
                            Major Source Metal Furniture Surface Coating Facilities
Facility ID
MFA-08-CP
MFF-01
MFA-08-TX
MFE-06-I
MFE-03-B
MFE-06-F
MFD-01
MFE-06B
MFE-04
MFB-02
MFF-03-C
MFE-06-K
MFE-06-G
MFA-08-CF
MFB-03
MFE-06-J
MFE-03-A
MFA-08-CX
MFA-07-J
MFF-04
MFA-07-HAZ
MFF-03-A
Reported VOC Emissions3
(kg/yr)
12,129
5,769
18,377
4,910
26,240
11,202
3,241
65,943
4,142
12,453
26,887
6,300
176,540
23,580
26,268
24,713
51,889
104,400
132,013
179,684
220,185
52,448
Corresponding Model
Plant Sizeb
Small
Small
Small
Small
Small
Small
Small
Small
Small
Small
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Large
Large
Large
Large
Large
TOTAL
Nationwide Baseline VOC
Emissions
(kg/yr)
164,000C
240,000d
689,000e
1,093,000
a Source: Industry questionnaire responses.
b Small model plant defined as less than 40,000 liters/yr of coating solids usage.
 Medium model plant defined as 40,000 to 99,999 liters/yr of coating solids usage.
 Large model plant defined as greater than 99,999 liters/yr of coating solids usage.
0 This value equals the sum of the values in Column A for the small facilities scaled up by a factor of 10/10.
dThis value equals the sum of the values in Column A for the medium facilities scaled up by a factor of 5/7.
e This value equals the sum of the values in Column A for the large facilities scaled up by a factor of 5/5.
                                                   7-20

-------
                     Table 7-10. Nationwide Organic HAP and VOC Emissions for the 20 New Sources After the
                                         Five-Year Period After Promulgation of a Rule
Model Plant


Small
Medium
Large
Total
A
Number of New
Sources After
Five- Year Period
After
Promulgation11
10
5
5
^^^^
B
Total Coating
Solids Usage per
Model Plantb
(L/yr)
22,000
54,000
250,000
^^^^
C
Organic HAP
Emission Rate
After Control0
(kg/L coating
solids)
0.094
0.094
0.094
^^^^
D
VOC Emission
Rate After
Controld
(kg/L coating
solids)
0.41
0.41
0.41
^^^^
E
Nationwide
Organic HAP
Emissions After
Control6
(Mg/yr)
21
25
120
170
F
Nationwide VOC
Emissions After
Controf
(Mg/yr)
90
110
510
710
a From Table 3.
b Average values from industry questionnaire response data.
c This value represents the new source MACT floor as determined in the following:  Memorandum from Hendricks, D., EC/R, to Serageldin,
M., EPA:ESD. September 14, 2001. Model Plants for the Metal Furniture Surface Coating Source Category.
d This VOC emission rate is from the facility used to establish the new source MACT floor.
eE = (AxBxC)/1000
fF = (AxBxD)/1000
                                                           7-21

-------
estimated to be 170 Mg organic HAP/yr.  This represents a reduction of 465 Mg/yr (73 percent) from

the baseline value shown in Table 7-8.

7.4.3.  VOC Emission Reduction.

       The VOC emission reduction for new sources was estimated using the same procedure as that

for organic HAP emissions. The VOC emission rate was calculated for the facility used to determine

the new source MACT floor (that is, the facility with the lowest organic HAP emission rate out of the

22 facilities for which the emission rate could be calculated). This emission rate was 0.41 kg VOC/L

coating solids. This emission rate was assumed to be equally applicable to all model plant size

classifications because there is no indication that the coatings used by a facility are necessarily affected

by the size of the facility.

       Table 7-10 presents the VOC emissions for the 20 new sources expected to be in operation in

the fifth year after promulgation of a rule.  The VOC emissions in the fifth year were estimated to be

710 Mg/yr. This represents a reduction of 380 Mg/yr (35 percent) from the baseline value shown in

Table 7-9.


 7.6 NOTES AND REFERENCES

1.      Detailed information concerning the conversion from SIC to NAICS codes can be obtained
       from the U.S. Census Bureau. See the U.S. Census Bureau's Internet site at
       http://www.census.gov/epcd/www/naics.html.

2.      U.S. Department of Commerce, Bureau of the Census. 1997 Economic Census,
       Manufacturing: Industry Series (Various Reports). Washington, DC. U.S. Government
       Printing Office.

3.      U.S. Environmental Protection Agency.  Toxic Release Inventory System. Internet Address:
       http://www.epa.gov/enviro/html/tris/tris_queryjava.html. Accessed  in June 1997.

4.      BIFMA International, Statistical Overview (Updated 12/16/99), Obtained from BIFMA
       website http://bifma.org/statover.html on February 4, 2000.

5.      Electronic Mail.  Miller,  B., BIFMA to Holmes, K., EC/R, Incorporated, "Estimated New
       Sources." March 24, 2000.
                                            7-22

-------
6.     Note 4.

7.     To a large extent, this assumption was based on the definition of affected source in the
       proposed regulation.  Because the affected source for all practical purposes encompasses all
       coating and cleaning related activities, it would be possible for many existing sources to add
       production capacity and, thus, new source requirements would not apply.

8.     Example calculation for small facilities:
       22,000 L coating solids/yr = annual coating solids usage for a small model plant.
       314 = nationwide number of existing facilities corresponding to the small model plant   size.
       0.02 = 2 percent increase in coating solids usage.
       (1-0.75) = amount of coating solids usage not absorbed by existing capacity.

       (22,000 L coating solids/yr) x (314 facilities) x (0.02) x (1-0.75) = 35,000 L coating  solids/yr.
       (35,000 L coating solids/yr) / (22,000 L coating solids/yr) = 1.6, which was rounded to two
       new facilities per year.
                                              7-23

-------
                                    8.0 COST IMP ACTS







8.1 INTRODUCTION



       The purpose of this chapter is to present the methodology used to estimate the cost impact of



implementing the existing source maximum achievable control technology (MACT) floor level of control



for the metal furniture surface coating source category. Costs were developed on a model plant basis



and were then scaled to nationwide costs.



       The costs presented here cover the first 5 years after promulgation of the standards. Because



existing sources have 3 years to achieve compliance with the emission limitations, the cost schedule will



build up during the first 3 years, reaching a maximum value in the fourth year.  The costs then decrease



slightly in the fifth year to a value that should reflect the projected cost of compliance from that point on.



Nationwide annual costs for existing sources, including monitoring, recordkeeping, and reporting



(MR&R) costs, were estimated to be approximately $14.8 million in the fifth year after promulgation of



the standards. There were no capital costs for existing sources.



       New sources must come into compliance upon startup.  Consequently, new sources will be



affected by all compliance costs, including monitoring, recordkeeping, and reporting (MR&R) costs,



beginning in the first year of their startup. Nationwide annual costs, including monitoring,



recordkeeping, and reporting  (MR&R) costs, were estimated to be approximately $0.6 million in the



fifth year after promulgation of the standards.  There were no capital costs for new sources.
                                             J-l

-------
8.2 METHODOLOGY FOR ESTIMATING EXISTING SOURCE COSTS



8.2.1  Determination of How Existing Sources Will Comply



       There are a variety of compliance methods available to and in use by the industry to meet the



MACT floor level of control for organic HAP emissions.  These include the use of lower organic HAP



content liquid coatings, powder coatings, lower organic HAP content cleaning materials, lower organic



HAP content adhesives, and add-on capture and control systems.  Various combinations of the



available compliance methods may be utilized to achieve the existing source MACT floor level of



control. Information obtained from the industry questionnaire responses, industry  site visits, trade



groups, and industry representatives was analyzed to determine which compliance methods would most



likely be used by existing sources and, therefore, which compliance methods to use in this cost analysis.



       The cost analysis was based on existing sources using lower organic HAP  content liquid



coatings (including adhesives), cleaning materials, and thinning solvents to meet the proposed emission



limit.  Add-on control devices and conversion of a liquid coating operation to powder coating were also



considered as a compliance option. While two facilities are known to use add-on control devices,



there is no indication that existing sources will use add-on control devices to any significant extent in the



future because of the availability of reformulated coating and cleaning materials. While conversion costs



on a per liquid coating operation basis were available, the portion of the industry that would convert to



powder coating could not be determined. Hence, these costs could not be scaled  up to nationwide



levels.  Thus, conversion to lower organic HAP content coatings and thinners and  organic HAP-free



cleaning materials was chosen as the basis for estimating the cost impacts for existing sources.



8.2.2  Cost Methodology for Existing Sources



       Model plants were developed (see Chapter 5) to aid in the estimation of the impacts the



standards will have on the metal furniture surface coating industry.  Three model plants (small, medium,



and large) were developed.  The parameters that describe each of these model plant sizes are



presented in Table 8-1.
                                             5-2

-------
                     Table 8-1. Model Plant Parameters By Unit Operation3
Parameter
Small Model Plant
<40,000 liters/yr
Medium Model
Plant
40,000 - 99,999
liters/yr
Large Model Plant
>99,999 liters/yr
Cleaning Unit Operations
Cleaning Material Usage (L/yr)
3,000
1,500
90,000
Coating Application Unit Operations
Liquid Coating Usage (L/yr)
Powder Coating Usage (L/yr)
Powder Coating Usageb
(kg/yr)
Coating Solids0 From Liquid
Coatings (L/yr)
Coating Solids From Powder
Coatings (L/yr)
Total Coating Solids (L/yr)
Number of Liquid Coating
Lines
Number of Powder Coating
Lines
66,000
950
1,300
21,000
950
22,000
2
1
160,000
3,600
5,100
50,000
3,600
54,000
2
1
440,000
11,000
16,000
240,000
11,000
250,000
4
1
a Source: 1997 and 1998 industry questionnaire responses.
b An average powder coating density of 1.41 kg/liter was used to convert from liters to kilograms.
c Nonvolatiles (film formers).

-------
       Estimated cost impacts were developed for each of the three model plants. To determine
nationwide cost impacts, the model plant costs were multiplied by the estimated nationwide number of
affected sources corresponding to each model plant size. The total nationwide number of facilities was
estimated using the U.S. Census Bureau's Economic Census. 1 The metal furniture surface coating
facilities in the Toxic Release Inventory System (TRIS) database2 were then used to determine the
overall percentage of major and area sources of HAP emissions in the industry. Applying the
percentage of major sources from the TRIS database to the total number of sources in the Economic
Census database gave a nationwide estimate of 655 major sources (see Chapter 5).
       Information provided in response to industry questionnaires was used to determine the
percentage of existing major source facilities that already meet the emission limit for this rule (9 percent
or 59 facilities), and the percentage of existing sources that are synthetic minors (12 percent or 79
facilities). Existing sources that have emission rates equal to or lower than the emission limit would not
have to implement controls, but would incur MR&R costs, and synthetic minor sources (based on their
potential (capacity) to emit) would not have to implement controls because of the proposed rule. The
number of existing sources used for estimating the implementation costs  for meeting the emission limit
was reduced to account for both of these types of sources, leaving a total of 517 existing sources.
However, for annual MR&R costs, sources already meeting the emission limit will have the same
MR&R requirements as those sources that are not presently meeting the emission limit, unless they are
permitted with a federally enforceable emission limit designating the facility as a synthetic minor source.
Therefore, for determining nationwide MR&R costs, the 59 existing sources already meeting the
emission limit were included for a total of 576 existing sources.
       The nationwide number of facilities that fall into the small, medium, and large model plant
categories was then determined based on the corresponding size distribution of facilities in the industry
questionnaire responses.  The small model plant group accounted for 45 percent of the facilities, while
the medium and large model plant groups accounted for 32 and 23 percent, respectively.  Using these
percentages, the 517 estimated nationwide number of major source facilities subject to the proposed
                                              8-4

-------
rule, broken down by model plant size, is 233 small facilities, 165 medium facilities, and 119 large



facilities.



        There are three types of control costs that may be incurred by a facility in the course of



complying with the standards: capital, direct, and indirect.  Capital costs represent the one-time



purchase of equipment. Because the compliance option expected to be used by most facilities to



comply with the standards utilizes reformulated raw materials rather than a different coating technology



or add-on control devices, we assumed that no capital costs would be incurred. That is, all existing



equipment related to the coating application unit operation (spray guns, spray booths, dip tanks, and



storage and mixing systems) was assumed to be compatible with lower organic HAP content coatings.3



Direct costs are incurred on a continuing basis for materials consumed in the manufacturing process,



primarily coatings and solvents.  Utilities are also included in the direct costs, but are expected to be



unchanged since there will be no change in equipment. Indirect costs include overhead, taxes,



insurance, and administrative costs, as well as capital recovery costs. Since no capital costs were



projected, it was assumed that overhead, taxes, insurance, and administrative costs would not change



as a result of converting to lower organic HAP content coatings. For this cost analysis, therefore, only



direct costs associated with raw material usage were developed.



        In addition to direct costs, affected facilities will incur MR&R costs. Annual MR&R costs



were based on the OMB 83-1 Supporting Statement, "Information Collection Request for the Metal



Furniture Surface Coating Operations Source Category."4







8.3 EXISTING SOURCE COST IMPACTS



        Table 8-2 presents the nationwide capital and annual (direct and indirect) cost impacts



associated with conversion to lower organic HAP content coatings and thinners and organic HAP-free



cleaning materials. The costs presented in Table 8-2 represent the costs that would be incurred starting



in the fourth year after promulgation of the standards. There would be no such cost in the first three



years after promulgation because facilities would not have to be in compliance with the standards until

-------
after the third year.  Table 8-3 presents the nationwide MR&R costs for existing sources for years 1



through 5 after promulgation of the standards.
                                              8-6

-------
   Table 8-2. Estimated Capital Costs and Annual Costs (excluding MR&R costs) for All Facilities Converting to All Lower Organic HAP Content Coatings
                             and Organic HAP-free Cleaning Materials — Metal Furniture Surface Coating Source Categorya
Unit Operation
Cost per Model Plant
A
Capital Costs
B
Indirect Costsb
C
Direct Costs0
D
Number of
Facilities
Requiring
Control
Nationwide Costs
E
Capital Costs
(AxD)
F
Indirect Costsb
(BxD)
G
Direct Costs0
(CxD)
SMALL MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$1,200
233
233
$0
$0
$0
$0
$0
$279,600
MEDIUM MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$600
165
165
$0
$0
$0
$0
$0
$99,000
LARGE MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$36,000
119
119
TOTAL NATIONWIDE COSTS
$0
$0
$0
$0
$0
$0
$0
$4,284,000
$4,662,600
a The costs shown in this table represent the costs incurred during each of years 4 and 5 after promulgation of the standards.  No costs associated with
 coating, cleaning, or adhesives will be incurred during the first 3 years after promulgation. However, we recognize that there will be costs associated
 with qualifying new cleaners and coatings for product use, but those costs are highly source specific and cannot be quantified here.
b Indirect costs include capital recovery, overhead, taxes, insurance, and administrative costs.
0 Direct costs consists of the operating costs (including utilities), but exclude monitoring,  recordkeeping, and reporting costs.

-------
     Table 8-3. Nationwide Monitoring, Recordkeeping and Reporting Costs for Existing Sources
                         Metal Furniture Surface Coating Source Category
A
Number of Existing Major Facilities
B
MR&R Costs per Facility3
C
Nationwide MR&R Costs
(AxB)
Year One
576
$1,671
Cumulative
$962,000
$962,000
Year Two
576
$335
Cumulative
$193,000
$1,155,000
Year Three
576
$13,124
Cumulative
$7,559,000
$8,714,000
Year Four
576
$17,701
Cumulative
$10,196,000
$18,910,000
Year Five
576
$17,555
Cumulative
$10,112,000
$29,022,000
a OMB 83-1 Supporting Statement, ICRfor the Metal Furniture Surface Coating Operations Source
Category, September 24, 2001. Represents the annual average cost per facility.

-------
Table 8-4 summarizes the total annual costs (including MR&R costs) on a nationwide basis for the 5-
year period after promulgation of the standards.  Nationwide annual costs, including MR&R costs,
were estimated to be approximately $14.8 million in the fifth year after promulgation.
8.3.1 Coating Operations
       The change in direct costs associated with converting from higher organic HAP content liquid
coatings to lower organic HAP content liquid coatings is related to two factors. The first is the cost per
unit volume of the coatings. From the information collected through initial data gathering efforts,
industry questionnaire responses, and industry trade group representatives, there was no indication that
a liquid coating with an organic HAP content at or below the MACT floor level of control will cost any
more or less than a liquid coating with a higher organic HAP content. The cost of a coating will also
vary with the quantity purchased.  Usually, the cost goes down as the volume purchased increases. For
this analysis, then, it was assumed that there will be no change in the unit volume cost of these liquid
coatings, resulting in no change in the annual costs associated with conversion to lower organic HAP
content liquid coatings. However, certain costs will be incurred qualifying new liquid coatings for
production use, but such costs are highly source  specific and cannot be quantified here.
       The second factor affecting the direct costs for liquid coating operations is the volume of lower
organic HAP content liquid coatings needed to replace an equivalent amount of higher organic HAP
content liquid coatings. Based on information  provided in the industry questionnaire responses, many of
the lower organic HAP content liquid coatings also had a higher coating solids content.  This would
indicate that overall liquid coating usage may actually decrease as a result of conversion to lower
organic HAP content liquid coatings. However, the coating solids content values are variable and are
not consistent for all liquid coatings. Due to this  variability, whether a reduction in usage would be
realized throughout the industry could not be determined. Therefore, it was assumed there would be no
decrease  in the volume of liquid coating usage in order not to overestimate any possible cost savings.
                                              8-9

-------
                        Table 8-4. Nationwide Annual Cost Summary
                       Metal Furniture Surface Coating Source Category
Year After
Promulgation


1
2
3
4
5
Total
A
Coating and Cleaning
Annual Costs"

$0
$0
$0
$4,662,600
$4,662,600

B
Monitoring,
Recordkeeping, and
Reporting Annual Costsb
$962,000
$193,000
$7,559,000
$10,196,000
$10,112,000

C
Total Annual Cost
(A + B)

$962,000
$193,000
$7,559,000
$14,858,600
$14,774,600
$38,347,200
a From Table 8-2.
b From Table 8-3.
                                          8-10

-------
8.3.2 Cleaning Operations



       For costing purposes, one organic HAP cleaning material (xylene) and one organic HAP-free



cleaning material (isopropyl alcohol) were selected from the questionnaire responses based on



prevalent use by the industry. Xylene represented approximately 34 percent (1,203,000 liters xylene of



3,492,000 liters total cleaning materials) of the reported organic HAP cleaning material usage, and



isopropyl alcohol was the most widely reported non-HAP cleaning material in the questionnaire



responses.  These choices were made solely to illustrate the possible cost differential between the



cleaning materials,  as isopropyl alcohol is approximately twice the cost of xylene, and is not meant to



imply that isopropyl alcohol is an appropriate replacement in all circumstances. This illustrates one of



the more conservative choices for a non-HAP alternative cleaning material on the basis of cost, and



should adequately account for the maximum cost that may be incurred to make such a change in



cleaning materials.  However, many types of solvent blends which contain low amounts of organic HAP



exist. These may often cost less than the non-HAP materials.



       To determine the cost of converting to organic HAP-free cleaning materials, the average



organic HAP-containing cleaning material usage was first determined for each model plant, based on



the industry questionnaire responses.  The costs of these cleaning materials were $0.40 per liter of



xylene and $0.80 per liter of isopropyl alcohol. 5  The cost difference between organic HAP-containing



cleaning material and the organic HAP-free cleaning material was then calculated for each model plant.



       There are no capital costs associated with converting to organic HAP-free cleaning materials.



The nationwide annual costs were $279,600, $99,000, and  $4,284,000 for small, medium, and large



model plants, respectively (see Table 8-2, Column G).  It should also be noted that there will be costs



associated with qualifying any new cleaning material for production use. These costs, however, are



highly source-specific and cannot be quantified here.  For example, if a facility switches to a less volatile



cleaner, the facility's cleaning emissions will likely be reduced, as well as its usage of cleaning solvents.



Significant annual cost savings have been reported in such cases. 6
                                             8-11

-------
8.3.3 Monitoring. Recordkeeping. and Reporting
       Each affected source will incur costs for implementing the MR&R requirements of the proposed
rule. These costs result primarily from the labor necessary to implement and maintain a system for
obtaining information (organic HAP content, coating solids content, density, etc.) on materials used,
tracking material usage, performing compliance calculations, and generating reports.  The annual
MR&R costs presented in Table 8-3 were based on the OMB 83-1 Supporting Statement "Information
Collection Request for the Metal Furniture Surface Coating Operations Source Category."?
       In years 1  through 3 it was assumed that existing sources will gradually implement only the
MR&R activities necessary to prepare to meet the requirements of the proposed rule. Thus, these
costs are relatively low compared to subsequent years. During year 1, only larger existing sources are
expected to begin  basic activities related to establishing monitoring and recordkeeping systems.
However, also occurring in year 1 all existing sources would read the rule to determine whether it
applies to them. This one time cost is reflected in the higher MR&R costs for year 1 over year 2.
Then, in years 2 and 3, all existing sources are expected to begin setting up monitoring and
recordkeeping systems, with an increase in activities in year 3. All existing sources in year 4 will incur
the full costs associated with implementing controls to achieve the emission limits, as well as MR&R
activities. The MR&R costs decrease slightly in year 5 because certain activities (e.g., initial
compliance status  notification) are only performed in year four.

8.4 METHODOLOGY FOR ESTIMATING NEW SOURCE COSTS
8.4.1 Number of New Major Sources
       Information obtained from an industry trade group, questionnaire responses, and industry
representatives was used to estimate the anticipated number of new major sources.  The industry trade
group provided an estimate of the percent increase in the sales volume measured in dollars for the
United States office furniture market. 8 They provided information which indicated a 3 to 5 percent
sales volume increase in terms of current dollars, which included the effect of price increases and
inflation, as well as increases in sales volume. It was assumed that this rate of increase would be
                                             8-12

-------
consistent over the 5-year period. After further discussions with the trade group, the actual sales
volume increase anticipated for the year 2001, excluding price increases and inflation, was determined
to be a 1 to 3 percent increase.9 This change in sales volume was assumed to be directly related to the
number of pieces produced rather than to the same number of more expensively priced pieces.
Therefore, it was assumed that a direct relationship exists between sales volume and production.  It was
further assumed that production is directly related to coating solids (nonvolatiles) usage.  Coating solids
usage is an accurate indicator of overall production level because the dry film thickness is generally
consistent throughout the industry.
       As described in Section 8.2.2 for  existing sources, model plants were also used to aid in the
estimation of the impacts the proposed rule would have on new metal furniture surface coating facilities.
As shown in Table 8-1, the annual coating solids usage for small, medium, and large model plants is
22,000, 54,000, and 250,000 liters, respectively. When multiplied by the estimated nationwide
number of existing major source facilities  corresponding to  each model plant size, the nationwide annual
coating solids usage was calculated to be  6,490,000 liters for small facilities, 11,290,000 liters for
medium facilities, and 37,750,000 liters for large facilities (Table 8-5, Column C).
       The nationwide coating solids usage values were multiplied by 2 percent, which was the
midpoint of the estimated sales volume increase obtained from the industry trade group.  The resulting
values were the estimated increase in coating solids usage due to the growth in the metal  furniture
industry. Based  on conversations with industry representatives, excess capacity currently exists in the
industry, but it is very difficult to quantify because of changing product types and market demands.
Additional capacity can also be added at existing sources and it was estimated that this combination of
existing and new capacity would absorb 75 percent of the coating solids usage increase before any new
facilities would be required. 10
       Hence, the predicted increase in nationwide annual  coating solids usage that will not be
absorbed by existing sources was determined to be 32,450  liters for small facilities,  56,450 liters for
medium facilities, and 188,800 liters for large facilities (Table 8-5, Column F). Comparing these
                                              8-13

-------
increases to the individual model plant coating solids usage values, four new facilities would be required
each year to handle the increased production (two small, one medium, and one
                                             8-14

-------
       Table 8-5.  Estimated Annual Number of New Major Source Facilities for the Metal Furniture Surface Coating Source Category

Model
Plant
Small
Medium
Large
A
Nationwide
Number of
Major Source
Existing
Facilities a
295
209
151
B
Total Coating
Solids Usage per
Model Plantb
(liters/yr)
22,000
54,000
250,000
C
Nationwide
Coating Solids
Usage0
(liters/yr)
6,490,000
11,290,000
37,750,000
D
Annual
Increase in
Coating Solids
Usaged
(%)
2
2
2
E
Coating Solids
Usage Absorbed
By Existing
Capacity6
(%)
75
75
75
Total
F
Annual Coating Solids
Not Absorbed by
Existing Sources
Capacityf
(liters/yr)
32,450
56,450
188,800
277,700
G
Equivalent
Annual Number
of New Sources8
2
1
1
4
a Total number of facilities nationwide determined from U.S. Census data. Percent of these that were major sources determined from toxic release inventory
system (TRIS) data. Breakdown by model plant size determined by industry questionnaire response data.
b Average values from industry questionnaire response data.
cAxB
d Based on industry publication and conversations with industry representatives.
e Inferred from conversations with industry representatives.
fCx(D/100)x(l-E/100)
gF/B
                                                                    8-15

-------
large)! 1 (see Table 8-5), or 20 new major source facilities during the 5-year period after promulgation

of the rule (see Table 8-6).
 Table 8-6. Estimated Number of New Major Source Facilities for Five Years After Promulgation for
                      the Metal Furniture Surface Coating Source Category
Year
Yearl
Year 2
Year3
Year 4
YearS
Total
Small Model
Plant
2
2
2
2
2
10
Medium Model
Plant
1
1
1
1
1
5
Large Model
Plant
1
1
1
1
1
5
Overall
Increase
4
4
4
4
4
20
8.4.2 Determination of How New Sources Will Comply

       As described in Section 8.2.1, there are a variety of compliance methods available to and in

use by the industry to meet the MACT floor level of control for organic HAP emissions. Because the

emission limit for new sources in the proposed rule is not significantly more stringent than the emission

limit for existing sources, it was assumed that the same emission control technology could be used by

new sources to meet the emission limit. That is, new sources are expected to reduce emissions using

lower organic HAP content liquid coatings (including adhesives), cleaning materials, and thinning

solvents.  In addition, it was assumed that the coatings used will have a higher coating solids content,

which was based on information derived from the industry questionnaire response database.

8.4.3 Cost Methodology for New Sources

       As discussed in the existing source cost analysis in Section 8.3, model plants were used to

facilitate the estimation of impacts of the proposed rule on the metal furniture surface coating industry.
                                            8-16

-------
These model plants were also used to estimate the new source cost impacts, using the number of new



facilities corresponding to each model plant size to scale up individual model plant costs to nationwide



levels.



       The new source cost analysis reflects the costs that are a direct result of having to achieve the



new source MACT floor level of control.  This requires determining the costs that would have been



incurred in the absence of this requirement.  It was assumed that a new source would use conventional



liquid coatings and organic HAP thinners and cleaning solvents if it did not have to comply with an



emission limit or rule. Thus, the incremental cost of achieving the new source MACT floor level of



control would be the difference in cost between these materials and  the cost of lower organic HAP



content coatings and thinners and organic HAP-free cleaning materials.



       In addition to direct costs, new affected sources will incur MR&R costs.  Annual MR&R costs



were based on the OMB 83-1 Supporting Statement, "Information Collection Request for the Metal



Furniture Surface Coating Operations Source Category."12







8.5 NEW SOURCE COST IMPACTS



       Table 8-7 presents the nationwide capital and annual (direct and indirect) cost impacts



associated with the use of lower organic HAP content coatings and thinners and organic HAP-free



cleaning materials. The costs presented in Table 8-7 represent the costs that would be incurred by the



four new sources in the first year after promulgation of the standards because new sources would have



to comply upon startup. In the second year after promulgation, an additional four new sources would



come on line, and annual costs would double from the that shown in Table 8-7. Costs would increase



similarly in each of the five years after promulgation as four new sources come on line each year. Table



8-8 presents the nationwide MR&R costs for new sources for years 1 through 5 after promulgation of



the standards.  Table 8-9 summarizes the total annual costs (including MR&R costs) on a nationwide



basis for the 5-year period. Nationwide annual costs, including MR&R costs, were estimated to be



$0.6 million in the fifth year after promulgation.
                                            8-17

-------
       Table 8-7. Estimated Capital Costs and Annual Costs (excluding MR&R costs) for New Facilities Using Lower Organic HAP Content Coatings
                             and Organic HAP-free Cleaning Materials — Metal Furniture Surface Coating Source Categorya
Unit Operation
Cost per Model Plant
A
Capital Costs
B
Indirect Costsb
C
Direct Costs0
D
Annual Number
of Facilities
Requiring
Control
Nationwide Costs
E
Capital Costs
(AxD)
F
Indirect Costsb
(BxD)
G
Direct Costs0
(CxD)
SMALL MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$1,200
2
2
$0
$0
$0
$0
$0
$2,400
MEDIUM MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$600
1
1
$0
$0
$0
$0
$0
$600
LARGE MODEL PLANT
Coating
Cleaning
$0
$0
$0
$0
$0
$36,000
1
1
TOTAL NATIONWIDE COSTS
$0
$0
$0
$0
$0
$0
$0
$36,000
$39,000
a The cost shown in this table represent the first year after promulgation.  The number of new sources coming on line in each subsequent year is the same as shown in
this table. Thus, the total costs in the second year after promulgation would be twice that shown in this table, cost in the third year would be three times that shown in
this table, etc. See Column A in Table 8-9.
b Indirect costs include capital recovery, overhead, taxes, insurance, and administrative costs.
0 Direct costs consists of the operating costs (including utilities), but exclude monitoring, recordkeeping, and reporting costs.
                                                                     8-18

-------
      Table 8-8. Nationwide Monitoring, Recordkeeping, and Reporting Costs for New Sources
                         Metal Furniture Surface Coating Source Category
A
Number of New Major Sources
B
MR&R Costs per Facility3
C
Nationwide MR&R Costs
(AxB)
Year One
4
$33,750
Cumulative
$135,000
$135,000
Year Two
8
$26,750
Cumulative
$214,000
$349,000
Year Three
12
$24,400
Cumulative
$293,000
$642,000
Year Four
16
$21,060
Cumulative
$337,000
$979,000
Year Five
20
$20,350
Cumulative
$407,000
$1,386,000
a OMB 83-1 Supporting Statement, ICRfor the Metal Furniture Surface Coating Operations Source
Category, September 24, 2001.  Represents the annual average cost per facility.
                                             8-19

-------
                 Table 8-9. Nationwide Annual Cost Summary for New Sources
                       Metal Furniture Surface Coating Source Category
Year After
Promulgation
1
2
3
4
5
Total
A
Coating and Cleaning
Annual Costs"
$39,000
$78,000
$117,000
$156,000
$195,000

B
Monitoring,
Recordkeeping, and
Reporting Annual Costsb
$135,000
$214,000
$293,000
$337,000
$407,000

C
Total Annual Cost
(A + B)
$174,000
$292,000
$410,000
$493,000
$602,000
$1,971,000
a Calculations for the first year are shown in Table 3.  Because we estimated that the same number of
new sources will come on line each year, these costs are additive in each subsequent year.
b From Table 8-8.
                                           8-20

-------
8.5.1 Coating Operations.



       Similar to the cost analysis for existing sources, it was assumed that there will be no change in



the unit volume cost of using lower organic HAP content coatings as compared to higher organic HAP



content coatings (see Section 8.3.1).  Thus, no change in the annual costs



associated with conversion to lower organic HAP content coatings is expected. However,  certain



costs will be incurred qualifying new liquid coatings for production use, but such costs are highly source



specific and cannot be quantified here.



8.5.2 Cleaning Operations.



       The new source cost analysis for cleaning materials followed the procedure detailed in Section



8.3.2 for existing sources. There were no capital costs associated with converting to organic HAP-free



cleaning materials.  The annual costs for the four new facilities coming  on line each year after



promulgation were estimated to be $2,400, $600, and $36,000 for small, medium, and large model



plants, respectively (see Table 8-7, Column G).  It should also be noted that there will be costs



associated with qualifying any new cleaning material for production use.  These costs, however, are



highly source-specific and cannot be quantified here. For example, if a facility switches to a less volatile



cleaner, the facility's cleaning emissions will likely be reduced, as well  as its usage of cleaning solvents.



Significant annual cost savings have been reported in such cases. 13



8.5.3 Monitoring. Recordkeeping. and Reporting.



       Each new affected facility will incur costs for implementing the MR&R requirements of the



standard. These costs result primarily from the labor necessary to implement and maintain a system for



obtaining information (organic HAP content, coating solids content, density, etc.) on materials used,



tracking material usage, performing compliance calculations, and generating reports. The annual



MR&R costs presented in Table 8-8 were based on the OMB 83-1 Supporting Statement "Information



Collection Request for the Metal Furniture Surface Coating Operations Source Category."14
                                             8-21

-------
8.6 NOTES AND REFERENCES

1.      U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census,
       Manufacturing:  Industry Series (Various Reports).  Washington, DC. U.S. Government
       Printing Office.

2.      U.S. Environmental Protection Agency. Toxic Release Inventory System. Internet Address:
       http://www.epa.gov/enviro/html/tris/tris_queryjava.html. Accessed in June 1997.

3.      Based on the industry questionnaire responses, the facilities that spray apply liquid coatings all
       used the same general types of application equipment (nearly all facilities reported some form of
       electrostatic spray application equipment and a few reported that the use of high volume low
       pressure (HVLP) spraying equipment). From this, the conclusion was made that low organic
       HAP content coatings reflect a change in the choice of solvent used in the coating formulation,
       not a significant departure in the nature of the coating itself. The solvent, whether HAP or non-
       HAP, is chosen for compatibility with the resin system and to impart characteristics, such as
       viscosity, that allow the coating to be applied as desired. The choice of solvent, thus, typically
       does not affect the application method.

4.      OMB 83-1 Supporting Statement,  "ICR for the Metal Furniture Surface Coating Operations
       Source Category."  ICR Number 1952-01. U.S. Environmental Protection Agency, Research
       Triangle Park, North Carolina. September 24, 2001.

5.      Cleaning material costs were derived from the following sources:
       Chemical Marketing Reporter, Schnell Publishing Company. Web site address:
       http://www.chemexpo.com accessed in August 1999.
       SouthChem, Durham, NC. July 1999.
       Worth Chemical, Durham, NC. August 1999.

6.      Alternative Control Techniques Document - Industrial Cleaning Solvents. EPA-453/R-94-
       015. U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
       Research Triangle Park, North Carolina.  February 1994.  pp. 5-8 and 5-9.

7.      Note 4.

8.      BIFMA International,  Statistical Overview (Updated 12/16/99), Obtained from BIFMA
       website http://bifma.org/statover.html on February 4, 2000.

9.      Electronic Mail.  Miller, B., BIFMA to Holmes, K., EC/R, Incorporated, "Estimated New
       Sources." March 24, 2000.
                                            8-22

-------
10.    To a large extent, this assumption was based on the definition of affected source in the
       proposed regulation. Because the affected source for all practical purposes encompasses all
       coating and cleaning related operations, it would be possible for many existing sources to add
       production capacity and, thus, new source requirements would not apply.

11.    Example calculation for small  facilities:
       22,000 L coating solids/yr = annual coating solids usage for a small model plant.
       295 = nationwide number of existing facilities corresponding to the small model plant size.
       0.02 = 2 percent increase in coating solids usage.
       (1-0.75) = amount of coating solids usage not absorbed by existing capacity.

       (22,000 L coating solids/yr) x (295 facilities) x (0.02) x (1-0.75) = 32,450 L coating solids/yr.
       (32,450 L coating solids/yr) / (22,000 L coating solids/yr) =1.5, which was rounded to 2 new
       facilities per year.

12.    Note 4.

13.    Note 6.

14.    Note 4.

-------
              9.0 ECONOMIC IMPACT AND SMALL BUSINESS ANALYSIS







9.1 INTRODUCTION



       This chapter evaluates the economic impacts of pollution control requirements on metal furniture



surface coating operations.  These requirements are designed to reduce emissions of hazardous air



pollutants (HAP) into the atmosphere.  The Clean Air Act's purpose is to protect and enhance the



quality of the nation's air resources (Section 101(b)).  Section 112 of the Clean Air Act Amendments



of 1990 establishes the authority to set national emission standards for HAP. The emissions of HAP



from metal furniture manufacturing originates from the cleaning and coating of these products.



       To reduce emissions of HAP, the EPA establishes maximum achievable control technology



(MACT) standards. The term "MACT floor" refers to the minimum control technology on which



MACT standards can be based. For existing major sources, the MACT floor is the average emissions



limitation achieved by the best performing 12 percent of sources (if there are 30 or more sources in the



category or subcategory). The MACT can be more stringent than the floor, considering costs, non-air



quality health and environmental impacts, and energy requirements. The estimated costs for individual



plants to comply with the MACT standards are inputs into the economic impact analysis presented in



this report.







9.2 ECONOMIC IMPACTS



       The MACT standards for metal furniture surface  coating facilities require these producers to



reduce the level of HAP in their coatings and solvents to meet the levels specified by the floor. The
                                            9-1

-------
costs of meeting the MACT standards will vary across facilities depending upon their physical



characteristics and current usage of coatings and solvents.  These regulatory costs will have financial



implications for the affected producers, and broader implications as these effects are transmitted



through market relationships to other producers and consumers. These potential economic impacts are



the subject of this section.



       Inputs to the economic analysis include:



1.      Baseline characterization of metal furniture industry



2.      Baseline market data as projected from industry and secondary sources



3.      Compliance cost estimates for industry segments (through model  plants) to meet the MACT



       floor standards.



The EPA has estimated the nationwide compliance costs of this regulation on existing sources to be



$14.77 million in the fifth year after promulgation.



       Metal furniture production is an assembly-line process in which components are cut,



assembled, and coated. The common structural materials used in production are steel and aluminum;



however, there has been a recent trend toward the use of plastics for certain components. Production



of metal furniture involves coating operations that emit HAP through use of coatings containing organic



solvents.  Coatings are applied to the metal surfaces to protect them from wear and corrosion.  The



coatings possess varying characteristics which make them suitable for different applications.



       Households, businesses, and institutions purchase and use metal furniture and related products.



The Standard Industrial Classification (SIC) codes of the industries that manufacture the various



products covered under this source category are provided in Appendix C, Table.  For the purposes on



this analysis, the metal furniture industry segments are defined as:



1.      Metal furniture classified by SIC codes 2514, 2522, and 2531 and include household metal



       furniture, office metal furniture, and public building metal furniture



2.      Metal fixtures classified by SIC 2542, 3645, 3646, and 2599 and includes cabinets,



       counters, display cases, residential lighting fixtures, commercial and industrial lighting fixtures,



       and institutional lighting fixtures
                                             9-2

-------
3.     Fabricated metal products covered by SIC codes 3429, 3469, and 3495 and includes



       furniture hardware, wastebaskets, stamped metal, and furniture springs



4.     Dental and laboratory metal furniture and apparatus covered by SIC codes 3821 and



       3843 and include dental cabinets and chairs; and laboratory furniture, benches, tables, and



       cabinets.



Appendix C, Table C-3 lists the corresponding North American Industry Classification System



(NAICS) codes.



       The following subsections address the economic impacts of the regulation on the individual



industry segments and the product markets served by those facilities within each segment.



9.2.1  Market Impacts



       In conducting an economic impact analysis, the EPA typically models the responses by



producers and markets to the imposition of the proposed regulation. The alternatives available to



producers in response to the regulation and the context of these choices are important in determining



the economic and financial impacts. Economic theory predicts that producers will take actions to



minimize their share of the regulatory costs.  Producers decide whether to continue production and, if



so, determine the optimal level consistent with market signals. These choices and market feedback



allow them to pass costs forward to the consumers of their end-products or services and/or to pass



costs backward to the suppliers of production inputs.



       Table 9-1 presents total annual compliance costs as a share of the value of shipments for the



major industry segments affected by this regulation.  These estimates are also provided for each SIC



code within the metal furniture industry segment.



       Table 9-1 shows that compliance costs are an extremely small share of the value of shipments.



Within the metal furniture industry segment,  costs range from 0.02 to 0.07 percent of the value of



shipments, indicating that the costs of meeting this regulation are not deemed significant.  If the metal



furniture producers were to partially or fully absorb the costs of complying with this rule, market prices



would either increase by less than shown in Table 9-1 or not at all. Because of the product diversity



within these SIC codes, the government and industry
                                             9-:

-------
   Table 9-1. Effect of Compliance Costs on Metal Furniture Producers by Industry Segment: 1997

Industry Segment


Metal Furniture
Household (SIC 25 14)
Office (SIC 2522)
Institutional (SIC 2531)
Metal Fixtures (SICs 2542, 3645, 3646,
2599)
Fabricated Metal Products (SICs 3429,
3469, 3495)
Dental and Laboratory (SICs 3821,
3843)
Total, all industry segments
Value of
Shipments
($106/yr)a

$11,791
$2,275
$8,001
$1,515
$10,334

$5,150

$4,686

$31,961
Total
Compliance
Costs
($106/yr)
$4.4
$1.7
$1.9
$0.9
$7.5

$1.8

$1.1

$14.8
Cost Shareb
(%)


0.04
0.07
0.02
0.06
0.07

0.04

0.02

0.05
aTotal compliance cost are representative of the expected costs faced by affected facilities within the
listed SIC codes.
bRelative cost shares computed as the total compliance costs divided by the value of shipments.
data do not provide the requisite production and/or price data upon which to base the economic

modeling. In lieu of these data, the EPA has employed a 1997 baseline characterization for each

industry segment where price is normalized to $1 so that the "value of shipments" proxies the

production quantity. The cost shares across the industry segments are then used as the "shifters" of the

market supply curve in a partial equilibrium model.

       Based on the EPA's partial equilibrium modeling, as shown in Table 9-2, the projected change

in market price and output is minimal as a result of the proposed MACT standard on existing sources.

The market price and output impacts are less than 0.1 percent across all industry segments. The metal

household furniture and the metal fixtures industry segments are projected to incur the largest impacts of

0.04 percent.
                                             9-4

-------
        Table 9-2. Market Impacts on Metal Furniture Producers by Industry Segment: 1997
Industry Segment
Metal Furniture
Household (SIC 25 14)
Office (SIC 2522)
Institutional (SIC 2531)
Metal Fixtures (SICs 2542, 3645, 3646,
2599)
Fabricated Metal Products (SICs 3429,
3469, 3495)
Dental and Laboratory (SICs 3821,
3843)
Total, all industry segments
Cost Share of
Sales
(%)
0.04
0.07
0.02
0.06
0.07
0.04
0.02
0.05
Market Impacts3 (%)
Price
0.02
0.04
0.01
0.03
0.04
0.02
0.01
0.02
Output
-0.02
-0.04
-0.01
-0.03
-0.04
-0.02
-0.01
-0.02
3 Percent change in market price and output result from the EPA's partial equilibrium model with unitary
market supply and demand elasticities. As a result, the predicted percent change for price and output
will be the same.
9.2.2  Social Costs and Their Distribution

       The value of a regulatory action is traditionally measured by the change in economic welfare

that it generates.  Welfare impacts, or the social costs required to achieve the environmental

improvements, stem from the regulation's effect on market outcomes and will extend to the many

consumers and producers of metal furniture and related products. For this analysis, based on applied

welfare economics principles, social costs are measured as the sum of the regulation induced changes in

consumer and producer welfare (otherwise known as 'surplus').  Consumers experience reductions in

their surplus because of increased market prices and reduced levels of consumption. Producers may

experience either increases or decreases in their surplus (i.e., profits) as a result of increased market
                                             9-5

-------
prices and changes in production levels and compliance costs. However, it is important to emphasize
that these surplus measures do not include benefits that occur outside the market, that is, the value of
reduced levels of air pollution with the regulation.
        The national estimate of compliance costs is often used as an approximation of the social cost of
the rule.  Under the MACT floor, the engineering analysis estimated annual costs for existing facilities to
be $14.77 million.  However, this estimate does not account for behavioral responses by producers or
consumers to the imposition of the regulation (e.g., shifting costs to other economic agents, closing
product lines  or facilities).  Accounting for these responses results in a social cost estimate that differs
from the engineering estimate and provides insights on how the regulatory burden is distributed across
society (i.e., the many consumers and producers of metal furniture and related products). The
economic welfare impacts of the regulation on producers and consumers can be considered under three
different scenarios:
1.       Full-cost absorption by  producers
2.       Full-cost pass-through to consumers
3.       Partial-cost pass-through to consumers.
Full-cost absorption lacks any accounting for behavioral responses to regulation, and  in this scenario
producers bear the full compliance costs  of the regulation. The other scenarios account for behavioral
responses to regulation both by  consumers and producers.  Full-cost pass-through refers to a situation
where producers are able to pass the social  costs of the regulation fully onto consumers. Alternatively,
partial-cost pass-through refers to a situation  where regulatory costs are borne both by consumers and
producers.
        9.2.2.1   Full-Cost Absorption. Under full-cost absorption, producers have no behavioral
response to the implementation of a regulation. The full regulatory compliance costs  are incurred by
affected facilities, whose owners experience a loss in profits equal to that amount, i.e., $14.77 million.
Since output is unchanged, market prices remain the same under  the full-cost absorption scenario and
consumers continue to demand the same  quantity.  As shown in Table 9-3, the welfare change is
                                              9-6

-------
composed entirely by a loss in producer surplus with no change (by assumption) in consumer surplus in



this case.
                                             9-7

-------
   Table 9-3. Economic Welfare Impacts of Metal Furniture MACT on Producers, Consumers, and
                                           Society
Stakeholders
Producers
Consumers
Society
Welfare Change
Full-Cost
Absorption
- $14.77 million
$0
- $14.77 million
Partial-Cost
Pass-Through (Fig. 9-
2)
- $7.38 million
- $7.38 million
- $14.77 million
Full-Cost
Pass-Through (Fig. 9-1)
0
- $14.77 million
- $14.77 million
Note: Totals may not add due to rounding.

       9.2.2.2 Full-Cost Pass-Through. Under full-cost pass-through, producers can pass the entire

burden of the regulation onto consumers of metal furniture and related products. In Figure 9-1, the

demand of consumers is represented by the downwards-sloping curve D and the original supply curve

of producers is represented by S0.  Implementing the regulation results in a shift in the supply curve from

S0 to Si. This leads to an increase in the market price from P0 to P! to incorporate the compliance

costs. This rise in price leads consumers to purchase a smaller quantity, Qb as can be seen by

examining the market demand curve (the new equilibrium point c). As shown in Figure 9-1, the loss in

consumer surplus here is the area P0acPb which is less than the full compliance costs, i.e., area P0abPb

because consumers reduce their consumption from Q0 to Qb  Thus, as shown in Table 9-3, the welfare

change is composed entirely by a loss in consumer surplus of $14.77 million with no change in producer

surplus.
       9.2.2.3 Partial-Cost Pass-Through. The economic welfare effects of a partial cost pass

through can be examined by referring to Figure 9-2.  In this case, both consumers and producers

experience a change in welfare. Once again market demand is represented by a standard downward-

sloping curve.  The supply curve is represented as an upward-sloping curve; equilibrium is determined

by the intersection.  The effect of the regulation is to shift the supply curve from S0 to Sb  This will lead

to a change in both consumer and producer surplus. The loss
                                             9-8

-------
   Incru
                                                                 S,  With k'f>rul,-=tifyi
                                                        Unil C«t lucre
                                                                 S_:  Without
                                        ^        l~J&               Output

                 Figure 9-1.  Full-Cost Pass-Through of Regulatory Costs
in consumer surplus is represented by the area P0bcP!. This loss in surplus occurs because consumers
           P1
   Price f
  I ncreaset
                                                                Sr With Regulation
SQ:  Without Regulation
              Figure 9-2.  Partial-Cost Pass-Through of Regulatory Costs
                                            9-9

-------
face a higher price for metal furniture and related products and as a response, they purchase a smaller



quantity. The net change in producer surplus is equal to the area abde (loss) - P0dcP! (gain due to a



transfer from consumers).  Combining the losses in surplus leads to the social costs of the regulation,



which is equal to the area abce.  This is less than the full compliance costs represented by area abfe in



Figure 9-2.  Thus, as shown in Table 9-3, the welfare change here is $14.77 million and is composed of



a change in both consumer surplus ($7.38 million) and producer surplus ($7.38 million).



       9.2.2.4  Summary. As summarized in Table 9-3, the economic welfare impacts for producers,



consumers, and society as a whole vary across the three scenarios considered.  The largest economic



impact would occur if producers made no behavioral change in response the regulation and were to



fully absorb the compliance costs of $14.77 million.  Consumers would bear no costs; therefore, the



total welfare change of society would be equal to the change in welfare experienced by  producers.



Under partial-cost pass-through, both producers and consumers experience a welfare change.



However, in this case, the sum of the loss in welfare is less than the full compliance costs. In full-cost



pass-through, the reduction in welfare consumers would incur would also be less than the total



estimated compliance costs of $14.77 million.



       Regardless of whether the costs of regulating the metal furniture manufacturing industry were



fully absorbed by producers or fully passed on to consumers, the per unit costs are negligible. As a



result, the effect of this regulation on the price of metal furniture and related products is not



distinguishable from random price fluctuations (or 'noise'). Therefore, the trivial magnitude of these



relative costs indicate negligible  distributional effects of this regulation across society.







9.3 SMALL BUSINESS IMPACTS



       This regulatory action will potentially affect the economic welfare of owners of metal furniture



surface coating facilities. The  ownership of these facilities ultimately falls on private individuals who may



be owner/operators that directly conduct the business of the firm (i.e., "mom and pop shops" or



partnerships) or, more commonly, investors or stockholders that employ others to conduct the business



of the firm on their behalf (i.e., privately-held or publicly-traded corporations).  The individuals or
                                             9-10

-------
agents that manage these facilities have the capacity to conduct business transactions and make



business decisions that affect the facility. The legal and financial responsibility for compliance with a



regulatory action ultimately rests with these agents; however, the owners must bear the financial



consequences of the decisions.  While environmental regulations can affect all businesses, small



businesses may have special problems in complying with such regulations.



       The Regulatory Flexibility Act (RFA) of 1980 requires that special consideration be given to



small entities affected by federal regulation. The RFA was amended in 1996 by the Small Business



Regulatory Enforcement Fairness Act (SBREFA) to strengthen the RFA's analytical and procedural



requirements. Under SBREFA, the EPA implements the RFA as written with a regulatory flexibility



analysis required only for rules that will have a significant impact on a substantial number of small



entities.  This section examines the metal furniture surface coating industry and provides a preliminary



screening analysis to determine whether this rule is likely to impose a significant impact on a substantial



number of the small entities (SISNOSE) within this industry. The screening analysis employed here is a



"sales test," which computes the annualized compliance costs as a share of sales for each company.



       Based on facility responses to the industry questionnaires, the EPA identified the ultimate parent



company and obtained their sales and employment data from either their questionnaire response or one



of the following secondary sources:



1.      Dun and Bradstreet Market Identifiers (Dun & Bradstreet, 1999)



2.      Hoover's Company Profiles (Hoover's Inc., 1999)



3.      Company Websites.



The facilities that received the questionnaires represent a sample of the total number of facilities included



in this source category (estimated at 655 major sources nationwide). Appendix G provides a listing of



the 24 companies that own and operate the 62 potentially affected facilities that responded to these



questionnaires.



       The Small Business Administration (SB A) defines a small business in terms of the sales or



employment of the owning entity.  These thresholds vary by industry and are evaluated based on the



industry classification  (SIC/NAICS code) of the impacted facility. Responses to the industry
                                             9-11

-------
questionnaires indicated multiple SIC/NAICS codes with a small business definition ranging from 100

to 1,000 employees or less than $5 million in annual sales. The EPA developed a company's size

standard based on the reported industry classification for these facilities. In cases where companies

own facilities with multiple classifications, the primary SIC/NAICS code and associated SBA definition

was used. Based on the EPA's database, 10 companies were identified as small (42 percent) and the

remaining 14 being large (58 percent) (See Appendix G for detailed listing).

       To assess the potential impact of this rule on these small businesses, the EPA calculated the

share of annual compliance cost relative to baseline sales for each company (i.e., employed the "sales

test"). When a company owns more than one facility, the costs for each facility are summed to develop

the numerator of the test ratio, or cost-to-sales ratio (CSR). Annual compliance costs are defined in

this analysis as the engineering estimate of regulatory costs imposed on these companies; thus, they do

not reflect the changes in production expected to occur in response to imposition of these costs and the

resulting market adjustments.  Table 9-4 reports total annual compliance costs, the number of

companies impacted at the one percent and three percent levels, and summary statistics for the

cost-to-sales ratios for small and large companies.

       Although small businesses represent 42 percent of the companies sampled within this source

category, Table 9-4 shows that their aggregate compliance costs represents only 14 percent, or

$176,000, of the industry sample's total of $1.3 million. The  annual compliance costs for small

businesses range from zero to 0.7 percent of their sales with 30 percent of the small businesses (i.e.,  3

out of 10) not incurring any regulatory costs. The vast majority of small companies with sales data have

CSRs below 0.5 percent. The mean (median) compliance cost-to-sales ratio is 0.15 (0.10)  percent for

the identified small businesses and 0.01 (0.01) percent for the large businesses. These results are

expected to be "representative" of the distributional impacts across companies by size and, of course,

depends upon the sample's representativeness of the total population of potentially affected facilities.
 Table 9-4.  Summary Statistics for SBREFA Screening Analysis on Metal Furniture Sample: MACT
                                            Floor
                                             9-12

-------

Total Number of Companies
Total Annual Compliance Costs
;$103/yr)
\verage TAG per company ($103/yr)

Companies with Sales Data
Not Impacted, i.e., = 0%
Impacted at >0 to 1%
Impacted at > 1 to 3%
Impacted at >3%
Small
10
$176
$17.6
Number
10
O
7
0
0
Share
100%
30%
70%
0%
0%
Large
14
$1,117
$79.8
Number
14
2
12
0
0
Share
100%
14%
86%
0%
0%
All Companies
24
$1,293
$53.9
Number
24
5
19
0
0
Share
100%
21%
79%
0%
0%
Z!ost-to-Sales Ratios
Average
Median
Minimum
Maximum




0.15%
0.10%
0.00%
0.70%




0.01%
0.01%
0.00%
0.10%




0.06%
0.01%
0.00%
0.70%
       The U.S. Census Bureau (1998) reports the after-tax return to sales for corporations in the
Furniture and Fixtures industry grouping at 4.5 percent for 1997. Corporations with less than $25
million in assets within this grouping experienced higher return to sales of 5.1 percent during this time
period. Reviewing the range of costs to be borne by small businesses in light of the 4.5 to 5.1 percent
profit margins typical of this industry, the EPA has determined the costs are typically small  and, overall,
do not constitute a significant impact on a substantial number.
       Because of the small questionnaire sample, the EPA conducted a supplemental SBREFA
screening analysis using the Toxics Release Inventory (TRI)  database that was employed by the
engineering analysis to estimate the number of major source facilities within this source category.  Based
on the TRI sample of facilities, the EPA identified the owning entities and obtained sales and
                                             9-13

-------
employment data where available.  A total of 28 small companies were identified from this sample of 57

companies that owned 70 major source facilities.  Lacking compliance estimates specific to these

facilities, the potential impacts were analyzed using the following costing scenarios:

1.      Minimal impact = $17,600 per major source, which reflects the average cost per small

       business from Table 9-4; and

2.      Maximum impact = $53,900 per major source, which reflects the costs for a large model plant.

       The minimal impact scenarios is likely to be more representative of the cost impacts for small

businesses because they are likely to own facilities represented by the small model plant.  Alternatively,

the maximum impact scenario is a worst-case costing scenario since most small businesses are not likely

to  own facilities represented by the large model plant.

       The supplemental screening analysis provided the following small business impacts for each cost

scenario:

1.      Minimal impact had an average CSR of 0.15% (median of 0.09%) with range of 0.04 to

       1.04%.

2.      Maximum impact had an average CSR of 0.45% (median of 0.27%) with range of 0.13 to

       3.15%.

The minimal impact scenario provides results comparable to those summarized in Table 9-4. Although

the maximum impact scenario is a worst-case scenario, we observe only 2 of the 28 small companies

(7  percent) with CSRs greater than 1 percent, and only 1 small company (3.2 percent) with a CSR > 3

percent. Therefore, the EPA believes that the supplemental analysis confirms the negligible impacts

observed from the initial SBREFA screening analysis based on the industry questionnaire.


9.4 REFERENCES
1.      Dun & Bradstreet.  1999. Dun's Market Identifiers [computer file].  New York, NY: Dialog
       Corporation.

2.      Hoover's Incorporated.  1999.  Hoover's Company Profiles [computer file]. Austin, TX:
       Hoover's Incorporated. 
-------
3.      Securities Exchange Commission. 1998. 10-K Reports, 

4.      U.S. Environmental Protection Agency.  1998. Preliminary Industry Characterization: Surface
       Coating of Metal Furniture. >.

5.      U.S. Department of Commerce, Bureau of the Census. 1993, 1994, 1995, and 1996. Annual
       Survey of Manufactures.

6.      U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Metal Household Furniture Manufacturing.

7.      U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Office Furniture (Except Wood) Manufacturing.

8.      U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Institutional Furniture Manufacturing.

9.      U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Showcase, Partitions, Shelving, and Locker Manufacturing.

10.    U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Commercial, Industrial, and Institutional Electric Lighting
       Fixture Manufacturing.

11.    U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Residential Electric Lighting Fixture Manufacturing.

12.    U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Dental Equipment and Supplies Manufacturing.

13.    U.S. Department of Commerce, Bureau of the Census.  1997 Economic Census:
       Manufacturing Industry Series for Laboratory Apparatus and Furniture Manufacturing.

14.    U.S. Department of Commerce, Bureau of the Census.  1999.  Concentration Ratios in
       Manufacturing (for the year 1992).
                                          9-15

-------
                   APPENDIX A




EVOLUTION OF THE BACKGROUND INFORMATION DOCUMENT

-------
INTRODUCTION

       The objective of the metal furniture integrated rule development project is to develop a
technical basis for supporting the proposed NESHAP for the metal furniture source category.  This
BID represents our current state of knowledge on the metal furniture source category.

       To accomplish this objective, technical data were acquired on the following aspects of the metal
furniture source category (1) representative processes and operations, (2) product characteristics, (3)
HAP emission points, including magnitude and composition of HAP emissions, and (4) the types and
costs of control options applicable to identified HAP emission points in this source category. The
primary sources of technical data included (1) technical references and literature, (2) State and local
regulatory agencies, (3) site visits,  (4) contact with representatives of the metal furniture industry and
trade associations, and (5) distribution of a section 114 questionnaire to metal furniture companies,
including summarization and analysis of the data collected in this effort.

       A chronological history of the development and evolution of significant events relating to the
emergence of the BID are presented in Table A-l.
                                             A-l

-------
                         TABLE A-l.  EVOLUTION OF THE BID
  Date
      Company, consultant, or agency
               and location
          Nature of action
04/08/97
04/09/97
05/11/97
05/14/97
05/14/97
05/28/97
06/11/97
U.S. Environmental Protection Agency, State
and Local agencies, and Industry
Durham, NC
U.S. Environmental Protection Agency, State
and Local agencies, and Industry
Durham, NC
U.S. Environmental Protection Agency and
Persons Interested in the Surface Coating of
Metal Furniture Rule Development
Durham, NC

Metal Creations
High Point, NC

U. S. Furniture Industries
High Point, NC

U.S. Environmental Protection Agency and
Persons Interested in the Surface Coating of
Metal Furniture Rule Development
Research Triangle Park, NC

Nevin Laboratories, Incorporated, Chicago,
IL
Steelcase, Incorporated, Grand Rapids, MI
Kimball, Incorporated, Jasper, IN
HON Industries, Muscatine, IA
Pelton & Crane Company, Charlotte, NC
Allsteel, Incorporated, Milan, TN
Darling Store Fixtures, Paragould, AR
Lozier Corporation, Omaha, NE
U.S. Coating Workshop with EPA,
State and Local agencies, and Industry
to familiarize them with the regulatory
process.

Coating Regulations Workshop Metal
Furniture/Large Appliance Breakout
Session. Discussion of the rule
development process and an informal
question and answer section. Also,
introduction of key persons in the rule
development process.

Draft example of completed
questionnaire response for review and
comment.
Site visit to High Point facility.


Site visit to High Point facility.


First Roundtable Meeting (P-
MACT/P-BAC Phase) with
EPA/Industry/States Working Team.


Distribution of section 114
questionnaire.
                                          A-2

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
06/3 0/97   The HON Company
           A Division of HON Industries
           Cedartown, GA

06/30/97   Steel case Incorporated
           Kentwood, MI

07/01/97   Steel case Incorporated
           Grand Rapids, MI

07/02/97   American Seating Company
           Grand Rapids, MI

07/07/97   Darling Store Fixtures
           Paragould, AR
07/09/97   Steelcase, Incorporated
           Grand Rapids, MI


07/09/97   Kimball International
           Jasper, IN
07/10/97   HON Industries, Incorporated
           Muscatine, IA
07/10/97   Nevin Laboratories, Incorporated
           Chicago, IL
                                     Response to June 1997 section 114
                                     questionnaire. Response for
                                     Cedartown plant.

                                     Site visit to Corporate Development
                                     Center.

                                     Site visit to Desk Plant and File Plant.
                                     Site visit to Grand Rapids facility.


                                     Response to June 1997 section 114
                                     questionnaire. Response for
                                     Paragould, AR and Corning, AR
                                     facilities.

                                     Response to June 1997 section 114
                                     questionnaire. Response contained in
                                     Confidential Business Information File.

                                     Response to June 1997 section 114
                                     questionnaire. Response for (Artec
                                     Panel) Plant,  Jasper, IN and Harpers,
                                     Post Falls, ID.

                                     Response to June 1997 section 114
                                     questionnaire. Response for Oak
                                     Steel metal case goods facility and
                                     Geneva chair plant.

                                     Response to June 1997 section 114
                                     questionnaire.
                                          A-3

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
      Company, consultant, or agency
               and location
          Nature of action
07/11/97
07/17/97
07/21/97
07/21/97



07/22/97


07/29/97


07/30/97


07/31/97
08/04/97
Siemens Medical System, Incorporated
Pelton and Crane Group
Charlotte, NC

Lozier Corporation
Omaha, NE
All steel, Incorporated
Milan, TN
Stanley Environmental, Incorporated
Coralville, LA
Husted, Husted and Associates, Incorporated
High Point, NC

Lozier Corporation
Omaha, NE

Steelcase, Incorporated
Grand Rapids, MI

U.S. Environmental Protection Agency and
Persons Interested in the Surface Coating of
Metal Furniture Rule Development
Research Triangle Park, NC

U.S. Environmental Protection Agency and
Regulatory Subgroup
Research Triangle Park, NC
Response to June 1997 section 114
questionnaire.


Response to June 1997 section 114
questionnaire. Response for Omaha,
Nebraska-North Plant, Omaha,
Nebraska-West Plant and Scottsboro,
Alabama Plant.

Response to June 1997 section 114
questionnaire. Response for Milan,
TN, Tupelo Systems, and Jackson
Seating facilities.

Response to request for additional
information for HON Industries section
114 questionnaire.

Transmittal of site visit questionnaire
for Metal Creations.

Facsimile transmitting paintiine
coverage data.

Transmittal of Material Safety Data
Sheets (MSDS's).

Second Roundtable Meeting (P-
MACT/P-BAC Phase) with
EPA/Industry/States Working Team.


Metal Furniture Integrated Rule
Development (P-MACT/P-BAC
Phase), Regulatory Subgroup
Teleconference.
                                          A-4

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
08/07/97   HON Industries
           Cedartown, GA

08/22/97   Steelcase, Incorporated
           Grand Rapids, MI
09/02/97   Steelcase, Incorporated
           Grand Rapids, MI

03/19/98   U.S. Environmental Protection Agency and
           Persons Interested in the Surface Coating of
           Metal Furniture Rule Development
           Research Triangle Park, NC

04/15/98   Royal Development
           High Point, NC

04/16/98   Charleston Forge
           Boone, NC

04/16/98   Johnson Casualties
           North Wilkesboro, NC

05/11/98   U.S. Environmental Protection Agency and
           Persons Interested in the Surface Coating of
           Metal Furniture Rule Development
           Durham, NC

05/15/98   BIFMA International
           Grand Rapids, MI
                                     Site visit to Cedartown Plant.
                                     Transmittal of corrected TRIS data
                                     and response to information request of
                                     July 21, 1997, based on the June 1997
                                     section 114.

                                     Transmittal of revised pages lib and
                                     12b to the section 114 submittal.

                                     Third Roundtable Meeting (P-
                                     MACT/P-BAC Phase) with
                                     EPA/Industry/States Working Team.
                                     Site visit to High Point facility.


                                     Site visit to Boone facility.


                                     Site visit to North Wilkesboro facility.


                                     Posting of DRAFT Example of
                                     completed questionnaire response.
                                     Posted on the Metal Furniture website.
                                     Electronic mail - Comments on the
                                     draft information collection request,
                                     including attachment of an alternative
                                     form set.
                                          A-5

-------
                    TABLE A-l.  EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
06/03/98   Accuride International, Sante Fe Springs, CA
           Identical letters sent to:
           Atlas Spring Manufacturing Corporation,
           Gardena, CA
           Hickory Springs Manufacturing Co., Hickory,
           NC
           National Metal Industries, West Springfield,
           MA
           Rabun Metal Products Incorporated, Tiger,
           GA
           Stylelander Metal Stampings, Inc., Verona,
           MS
           United Receptacle, Pottsville, PA

06/03/98   B Line Systems,  Incorporated, Highland, OH
           Identical letters sent to:
           Framecrafters, Chicago, IL
           Penco Products Incorporated, Oaks, PA
           Republic Storage Systems, Canton, OH
           Sunlight Casual Furniture, Paragould, AR

06/03/98   A-Dec, Incorporated, Newburg, OR
           Identical letters sent to:
           Den-Tal-Ez Manufacturing, Bay Minette, AL
           Medical Lab Automation, Incorporated,
           Pleasantville, NY
           Sheldon Lab Systems, Crystal Springs, MS

06/03/98   Davies Office Refurbishing, Incorporated,
           Albany, NY
           Identical letters sent to:
           Furniture Medic International, Memphis, TN
           Office Repair and Services, San Francisco,
           CA
           Professional Refinishing, Los Angeles, CA
                                     Distribution of section 114 industry
                                     questionnaire for Metal Furniture Parts
                                     and Hardware Manufacturing
                                     Companies.
                                     Distribution of section 114 industry
                                     questionnaire for Miscellaneous Metal
                                     Furniture Products Manufacturing
                                     Companies.
                                     Distribution of section 114 industry
                                     questionnaire for Laboratory and
                                     Dental Furniture Manufacturers.
                                     Distribution of section 114 industry
                                     questionnaire for Furniture Repair
                                     Operation Companies.
                                           A-6

-------
                    TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
06/03/98   A&J Manufacturing Company, Tustin, CA
           Identical letters sent to:
           American Desk Manufacturing, Temple, TX
           Cramer, Incorporated, Kansas City, KS
           Crown Metal Manufacturing, Elmhurst, IL
           Dehler Manufacturing, Chicago, IL
           Edsal Manufacturing, Chicago, IL
           Virco Manufacturing, Torrance, CA
           Steelcase, Incorporated, Grand Rapids, MI
           HON Industries, Muscatine, IA

06/03/98   Venture Lighting International, Solon, OH
           Identical letters sent to:
           Lightolier, Incorporated, Fall River, MA
           Mid-West Chandelier Company, Kansas
           City, KS
           Lithonia Lighting Company, Conyers, GA

06/11/98   Sheldon Laboratory Systems
           Crystal Springs, MS
06/18/98   Medical Laboratory Automation, Inc.
           Pleasantville, NY
06/24/98   Leggett & Platt Incorporated
           Carthage, MO
07/01/98   Venture Lighting
           Solon, OH
                                      Distribution of section 114 industry
                                      questionnaire for Household, Office,
                                      and Public Building Furniture and
                                      Store Fixtures, Partitions and Shelving
                                      Companies.
                                      Distribution of section 114 industry
                                      questionnaire for Residential and
                                      Commercial Lighting Fixture
                                      Companies.
                                      Transmittal of letter explaining Sheldon
                                      Laboratory Systems' status in regard
                                      to the June 1998, section 114
                                      questionnaire.

                                      Transmittal of letter regarding status of
                                      the June 1998,  section 114
                                      questionnaire response.

                                      Transmittal of letter stating that the
                                      Leggett & Platt Stylelander facility in
                                      Verona, Mississippi is no longer used
                                      for manufacturing.

                                      Transmittal of letter stating Venture
                                      Lighting's status in regard to the June
                                      1998 section 114 questionnaire
                                      request.
                                           A-7

-------
                    TABLE A-l.  EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
07/16/98   Accuride International Inc.
           Santa Fe Springs, CA


07/21/98   A&J Manufacturing Company
           Brea, CA


07/28/98   U.S. Environmental Protection Agency and
           Recipients of the Surface Coating of Metal
           Furniture Questionnaire
           Durham, NC

07/29/98   Leggett & Platt Incorporated
           Carthage, MO


08/17/98   Davies Office Refurbishing
           Albany, NY

08/17/98   Adec
           Newberg, OR

08/18/98   U.S. Environmental Protection Agency and
           Persons Interested in the Surface Coating of
           Metal Furniture Rule Development
           Durham, NC

08/19/98   Republic Storage Systems Company
           Canton, OH

08/20/98   EST Division of Leggett Partners, L.P.
           Leggett & Platt Incorporated
           Carthage, MO
                                     Transmittal of letter stating Accuride's
                                     status in regard to the June 1998
                                     section 114 questionnaire request.

                                     Transmittal of letter stating that all
                                     surface coating operations are
                                     procured from outside suppliers.

                                     Distribution of clarifications to the June
                                     1998 questionnaire.
                                     Transmittal of memorandum detailing
                                     metal furniture facilities completing
                                     June 1998 section 114 response.

                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of letter concerning section
                                     114 questionnaire request.

                                     Distribution of coating calculation sheet
                                     for use in completing the industry
                                     questionnaire.


                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire response for the Grafton,
                                     WI facility.
                                           A-8

-------
                    TABLE A-l.  EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
08/20/98   Leggett & Platt Incorporated
           High Point, NC


08/20/98   Leggett & Platt Incorporated
           Carthage, MO


08/20/98   Leggett & Platt Incorporated
           Carthage, MO


08/20/98   Leggett & Platt Incorporated
           Carthage, MO


08/20/98   Hickory Springs Manufacturing Company
           Hickory, NC
08/20/98   Virco Manufacturing Corporation
           Torrance, CA

08/21/98   Steelcase Incorporated
           Grand Rapids, MI
08/21/98   The HON Company
           Muscatine, IA

08/21/98   Arco Bell Corporation
           Temple, TX
           (Originally sent under the previous company
           name of American Desk)
                                     Transmittal of completed section 114
                                     questionnaire response for High Point
                                     Sleeper.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Simpsonville, KY facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Linwood Branch facility.

                                     Transmittal of completed section 114
                                     questionnaire response for Duro Metal
                                     Manufacturing facility in Dallas, TX.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Hickory, NC Metal Plant and the Fort
                                     Smith, AR Metal Plant.

                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire response for the Chair I,
                                     Revest-Dallas, and Revest-Atlanta
                                     facilities.

                                     Electronic mail transmittal of section
                                     114 questionnaire status.

                                     Distribution of section 114 industry
                                     questionnaire for Household, Office,
                                     and Public Building Furniture and
                                     Store Fixtures, Partitions and Shelving
                                     Companies.
                                          A-9

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
08/24/98   National Metal Industries
           West Springfield, MA

08/24/98   B-Line Systems
           Highland, IL
08/24/98   Lightolier
           Fall River, MA
08/24/98   Mid-West Chandelier Company
           Kansas City, KS

08/24/98   Crown Metal Manufacturing Company
           Elmhurst, IL

08/24/98   Lithonia Lighting
           Conyers, GA
08/24/98   Collier-Keyworth, Incorporated
           Leggett and Platt, Incorporated
           Liberty, NC

08/24/98   The HON Company
           Muscatine, IA
08/24/98   The HON Company
           Muscatine, IA
                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Highland Plant.

                                     Transmittal of completed section 114
                                     questionnaire response for the Fall
                                     River facility.

                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire responses for Lithonia
                                     Electronic  Systems Group, Lithonia
                                     Lighting-Conyers, Lithonia Lighting-
                                     Lithonia West, Lithonia Lighting-Hi
                                     Tek Division, Lithonia Lighting-
                                     Cochran, and Lithonia Down Lighting.

                                     Transmittal of completed section 114
                                     questionnaire response.


                                     Transmittal of completed section 114
                                     questionnaire response for the Geneva
                                     Plant.

                                     Transmittal of completed section 114
                                     questionnaire response for the Oak
                                     Steel Plant.
                                          A-10

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
08/24/98   HON Industries, Allsteel
           Jackson, TN


08/24/98   The HON Company
           South Gate, CA


08/24/98   The HON Company
           Winnsboro, SC


08/24/98   The HON Company, Allsteel
           West Hazelton, PA


08/24/98   Leggett & Platt, Incorporated
           Winchester, KY


08/24/98   Leggett & Platt, Incorporated
           A Division of Dresher Incorporated
           York, PA

08/24/98   Leggett & Platt, Incorporated
           Whittier, CA
08/25/98   Penco Products, Incorporated
           Oaks, PA
08/25/98   Atlas Spring Manufacturing Corporation
           Gardenia, CA
                                     Transmittal of completed section 114
                                     questionnaire response for the Jackson
                                     facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the South
                                     Gate facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Winnsboro Plant.

                                     Transmittal of completed section 114
                                     questionnaire response for the West
                                     Hazleton facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Winchester, KY facility.

                                     Transmittal of completed section 114
                                     questionnaire response for Harris Hub
                                     facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Whittier, CA facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Newtown Square, PA facility.

                                     Transmittal of completed section 114
                                     questionnaire response.
                                         A-ll

-------
                   TABLE A-l. EVOLUTION OF THE BID (continued)
  Date
Company, consultant, or agency
         and location
Nature of action
08/25/98   Dehler Manufacturing Company,
           Incorporated
           Chicago, IL

  08/98    U.S. Environmental Protection Agency and
           Persons Interested in the Surface Coating of
           Metal Furniture Rule Development
           Durham, NC

09/02/98   Leggett & Platt, Incorporated
           Tupelo, MS
09/04/98   The HON Company
           Cedartown, GA
09/11/98   Steelcase Incorporated
           Grand Rapids, MI
09/24/98   Professional Refinishing Organization
           Newport Beach, CA

09/25/98   Steelcase Incorporated
           Grand Rapids, MI
  09/98    U.S. Environmental Protection Agency and
           Persons Interested in the Surface Coating of
           Metal Furniture Rule Development
           Durham, NC
                                     Transmittal of completed section 114
                                     questionnaire response.


                                     Distribution of the Draft Preliminary
                                     Industry Characterization: Surface
                                     Coating of Metal Furniture.


                                     Transmittal of completed section 114
                                     questionnaire response for the Super
                                     Sagless facility in Tupelo, MS.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Cedartown, GA facility.

                                     Transmittal of completed section 114
                                     questionnaire response for the
                                     Computer Furniture Plant, File Plant,
                                     Desk Plant, Tustin Plant, and Athens
                                     Plant facilities.

                                     Transmittal of completed section 114
                                     questionnaire response.

                                     Transmittal of completed section 114
                                     questionnaire response for the Panel
                                     Plant, Context Plant and Systems I
                                     Plant facilities.

                                     Distribution of the Preliminary Industry
                                     Characterization: Surface Coating of
                                     Metal Furniture.
                                          A-12

-------
                    TABLE A-l. EVOLUTION OF THE BID (continued)
   ^             Company, consultant or agency                  -T      .,   .
   Date                      ,,    •                             Nature of action
                          and location

 10/06/98  Dental EZ                                 Transmittal of completed section 114
           Bay Mnette, AL                           questionnaire response.

 07/13/99  U.S. Environmental Protection Agency and     Fourth Pre-MACT
           Persons Interested in the Surface Coating of   (EPA/Industry/States) Working Team
           Metal Furniture Rule Development            Meeting
           Research Triangle Park, NC

 07/16/99  Hickory Springs Manufacturing Company      Transmittal of letter regarding floor
	Hickory, NC	calculation.	
                                         A-13

-------
               APPENDIX B




PARTICIPANTS IN THE DATA COLLECTION EFFORT

-------
                                     TABLE B-l. LIST OF STAKEHOLDERS
Name
Company
Mailing Address
Telephone/Fax Number
e-mail Address
                                              EPA Representatives
Mohamed
Serageldin
Karen Borel
Kathy Davey
Bob Rose
Scott Throwe
Eric Wilkinson
U.S. EPA
U.S. EPA
U.S. EPA
U.S. EPA
U.S. EPA
U.S. EPA
OAQPS/ESD/CCPG (C539-03)
Research Triangle Park, NC 2771 1
Air Permits Branch
61 Forsyth Street
Atlanta, GA 30303
OPPTS-OPPT PPD
Mail Code 7409
401 M Street, S.W.
Washington, DC 20460
OSDBU
Mail Code 1230-C
401 M Street, S.W.
Washington, D.C. 20460
OECA
Mail Code 2223A
401 M Street, S.W.
Washington, D.C. 20460
OPPTS/ PPD
Mail Code 7409
401 M Street, S.W.
Washington, D.C. 20460
(919) 541-2379
fax-(919) 541-5689
(404) 562-4300
fax-(404) 562-9019
(202) 260-2290
fax-(202) 260-0 178
(202) 564-9744
fax-(202) 565-2078
(202) 564-7013
fax-(202) 564-0050
(202) 260-3575
fax-(202) 260-0 178
serageldin.mohamed@epa.gov
b orel . karen@epa. gov
davey.kathy@epa.gov
rose.bob@epa.gov
throwe. scott@epa.gov
wilkinson.eric@epa.gov
                                                  Consultants
                                                    B-l

-------
TABLE B-l.  LIST OF STAKEHOLDERS
Name
David Hendricks
Karen Holmes
Company
EC/R Incorporated
EC/R Incorporated
Mailing Address
2327 Englert Drive
Suite 100
Durham, NC 27713
2327 Englert Drive
Suite 100
Durham, NC 27713
Telephone/Fax Number
(919) 484-0222
ext. 335
fax-(9 19) 484-0 122
(919) 484-0222
ext. 310
fax-(9 19) 484-0 122
e-mail Address
hendricks.david@ecrweb.com
holmes.karen@ecrweb.com
        State Representatives
Ken Barrett
Dan Belik
Bob Colby
Stan Cowen
Somnath Dasgupta
Jorge Deguzman
Cindy Eisfelder
Jon Heinrich
Alabama DEM
Bay Area AQMD
Chattanooga/ Hamilton
County Air Pollution
Control Bureau
Ventura County
APCD
Iowa Waste Reduction
Center
Sacramento
Metropolitan APCD
Michigan DEQ
Wisconsin DNR
Air Division
P.O. Box 301463
Montgomery, AL 36130-1463
939 Ellis Street
San Francisco, C A 94109
3511 Rossville Boulevard
Chattanooga, TN 37407-2495
669 County Square Drive
Ventura, CA 93003




(334) 271-7870
fax-(334) 279-3044
(415)749-4786
fax-(415)928-0338
(423) 867-4321
fax-(423)867-4348
(805) 645-1408
fax-(805) 645-1444
fax-(3 19) 268-3733
fax-(916) 386-7040
fax-(5 17) 24 1-7440
fax-(608) 267-0560








              B-2

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Robert Hodanbosi
Susan Hoyle
Lee Huo
Dick Johnson
Jimmy Johnson
Martha Lee
Fred Lettice
Christy Myers
Hank Naour
Company
Ohio EPA
Pennsylvania Bureau
of Air Quality
San Joaquin Valley
Unified APCD
Placer County APCD
Georgia Department of
Natural Resources
Sacramento
Metropolitan APCD
South Coast AQMD
Alabama DEM
Illinois EPA
Mailing Address

400 Market Street
12th Floor
Harrisburg, PA 17105-8468
1999 Tuolumne
Suite 200
Fresno, CA 93721
1 1 464 B Avenue
Dewitt Center
Auburn, CA 95603
Air Protection Branch
4244 International Parkway
Suite 120
Atlanta, GA 30354
8411 Jackson Road
Sacramento, CA 95826
2 1865 East Copley Drive
Diamand Bar, CA 91765
Air Division
P.O. Box 301463
Montgomery, AL 36130-1463
Bureau of Air
P.O Box 19506
Springfield, IL 62794-9506
Telephone/Fax Number
fax-(614) 644-3681
(717) 787-9257
fax-(717) 772-2303
(209) 497-1075
fax-(209) 233-0140
(916) 889-7130
fax-(916) 889-7107
(404) 363-7127
fax-(404) 363-7100
(916) 386-6660
fax-(916) 386-6674
(909) 396-2576
fax-(909) 396-2608
(334)271-7861
fax-(334) 279-3044
(217)785-1716
fax-(217) 524-5023
e-mail Address

shoyle@state.pa.us





hank.naour@epa.gov. state.il .us

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Todd Nishikawa
Venkata
Panchakarla
John Patten
John Ramsey
Frank
St. Clair
Doug Wagner
Richard Wales
Company
Placer County APCD
Florida Department of
Environmental
Protection
Tennessee Department
of Environmental
Conservation
Kansas Department of
Health and
Environment
Mojave Desert
AQMD
Indiana Department of
Environmental
Management
Mojave Desert -
Antelope Valley
APCD
Mailing Address

Mail Station #5500
2600 Blair Stone Road
Tallahassee, FL 32399
Division of Air Pollution Control
L&C Annex, Ninth Floor
401 Church Street
Nashville, TN 37243-1531
Forbes Field, Building 740
Topeka, KS 66620
15428 Civic Drive
Suite 200
Victorville, CA 92392-2383
Office of Air Management
100 North Senate
P.O. Box 6015
Indianapolis, IN 46206-6015
15428 Civic Drive
Suite 200
Victorville, CA 92392-2383
Telephone/Fax Number
fax-(530) 889-7107
(850)488-0114
fax-(850) 922-6979
(615) 532-0554
fax-(615) 532-0614
(913)296-1593
fax-(913)296-1545
(760)245-1661x6101
fax-(760) 245-2022
(317)232-0286
fax-(3 17) 232-6749
(760) 245-1661
fax-(760) 245-2699
e-mail Address

panchakarla_v@dep. state.fl .us





       Industry Representatives
Thomas Ashley
Charleston Forge
251 Industrial Park Drive
Boone,NC 28607
(704) 264-0100
fax-(704) 264-5901

               B-4

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Quentin Baker
Clyde Blaco
Kevin Booth
Steve Byrne
Carlos Casillas
Andy Counts
Jennifer Depolo
Mick Durham
William English
Robert Eshbach
Steve Foster
Company
Royal Development
Company
NASFM
Penco Products
Cytec
Leggett & Platt
American Furniture
Manufacturers
Association
Leggett & Platt
Stanley Environmental
PPG Industries
Republic Storage
Systems
Johnson Casuals
Mailing Address
325 Kettering Road
High Point, NC 27263
3595 Sheridan Street
Suite 200
Hollywood, FL 33021
4080 West Farm Road
West Jordan, Utah 84088
1300 Mt. Kemble Avenue
Morristown, NJ 07960
P.O. Box 4956
Whittier, CA 90602
P.O. Box HP-7
High Point, NC 27261
P.O. Box 140
Linwood, NC 27299
225 Iowa Avenue
Muscatine, IA 52761
One PPG Place
Pittsburgh, PA 15272
1038 Beldon Avenue N.E.
Canton, OH 44705

Telephone/Fax Number
fax-(336) 889-6736
(954) 893-7300 ext. 27
fax-(954) 893-7500
(801)280-1541
fax-(80 1)280-3450
(973) 425-8406
fax-(973) 425-01 85
(562) 945-2641
fax-(562) 945-3 190
(910) 884-5000
fax-(910) 884-5303
(336) 956-5000
fax-(336) 956-5013
(319)264-6342
fax-(3 19) 264-665 8
(412)434-3198
fax-(412) 434-3705
(330) 454-5800
fax-(330) 454-7772
fax-(336) 667-0998
e-mail Address
royal@northstate.com

kevin.booth@pencoproducts. com
steve_by rne@gm . cy tec . com

acounts@ng.infi.net
jdepololegg@aol.com
durhammick@stanleygroup.com

beshbach@republicstorage. com

             B-5

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Ken Gabele
Charlie Gardner
Walt Hammond
Madelyn Harding
Mary Husted
Michael Jonas
Dennis Kane
Glen Kedzie
Terry Knight
Albert Kula
Sidney Lefkovitz
Company
The Sherwin-Williams
Company
Leggett & Platt
Thomasville Furniture
Industries
The Sherwin-Williams
Company
Husted & Associates
Lozier Corporation
Leggett & Platt
National Paint and
Coatings Association
B-Line Systems
Leggett & Platt
Mid-West Chandelier
Mailing Address
101 Prospect Avenue, N.W.
Cleveland, OH 44115-1075
2017 South Green Street
Tupelo, MS 38802
P.O. Box 339
Thomasville, NC 27361-0339
101 Prospect Avenue, N.W.
Cleveland, OH 44115-1075
P.O. Box 5256
High Point, NC 27262
6336 Pershing Drive
Omaha, NE 68110
915 Woodland View Drive
York, PA 17402
1500 Rhode Island Avenue, NW
Washington, DC 20005
509 W. Monroe
Highland, IL 62249
P.O. Box 7327
High Point, NC 27264
P.O. Box 15097
Kansas City, KS 661 15
Telephone/Fax Number
(216)566-3316
fax-(216) 556-2920
(662) 791-7136
fax-(662)791-7187
(910) 476-2263
fax-(91 0)472-4080
(216) 566-2630
fax-(216) 556-2730
(910) 869-3097
fax-(91 0)869-3031
(402) 457-8497
fax-(402) 457-8554
(717) 843-6288
fax-(717) 843-6185
(202) 462-6272
fax-(202) 328-0688
(618)654-2184
fax-(6 18) 654-2 184
(336) 889-4998
fax-(336) 889-5066
(913)281-1100
fax-(9 13) 28 1-1967
e-mail Address
klgabele@sherwin.com


mkharding@sherwin.com

mj onas@compuserve.com

gkedzie@paint. org
tknight@cooperbline.com


             B-C

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Scott Lesnet
Charles Lindsey
Diane Luo
Bob Maindelle
Archie Martz
Jeffery Masi
Richard Mathis
Dave Mazzocco
Michael McMullen
David McNeil
Company
HON Industries
Leggett & Platt
Duro Metal
Manufacturing
Pelton & Crane
WilsonArt
International
Lilly Industries
Allsteel, Inc.
Metal Creations
PPG Industries
American Seating
Company
Hickory Springs
Mailing Address
SM4 Technical Center
505 Ford Avenue
Muscatine, IA 52761
P.O. Box 170520
Dallas, TX 75217
11 727 Fruehauf Drive
Charlotte, NC 28241

P.O. Box 2358
High Point, NC 27261
71 Denton Fly Road
Milan, TN
P.O.Box 1104
High Point, NC 27261
4325 Rosanna Drive
Allison Park, PA 15101
401 American Seating Center
Grand Rapids, MI 49504
P.O. Box 128
Hickory, NC 28603
Telephone/Fax Number
(319)262-7865
fax-(3 19) 262-7899
(214)391-3181
fax-(214) 391-7629
(704) 587-7294
fax-(704) 587-7214
fax-(254) 207-2948
(910) 802-4326
(910) 889-2157
fax-(910) 889-6007
(901)686-4116
fax-(901) 686-4120
(910) 889-2083
fax-(910) 885-2442
(412) 492-5476
fax-(4 12) 492-53 77
(616) 732-6650
fax-(616) 732-6401
(828) 328-2201
fax-(828) 324-47 15
e-mail Address
lesnet@honcompany . com
charles.lindsey@gte.net





mazzocco@ppg.com
mcmullen@amscco. com
davemcn@twavenet
             B-7

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Mary Ellen Mika
Brad Miller
Chuck Millisor
Bob Nelson
Robert Nevin
Loc Nguyen
Mary Ellen Roddy
Rhonda Ross
Larry Runyan
Stan Schmitt
Company
Steelcase, Inc.
BIFMA International
Leggett & Platt
National Paint and
Coatings Association
Nevin Laboratories
Davies Office
Refurbishing
National Paint and
Coatings Association
Warner, Norcross &
Judd (for BIFMA)
American Furniture
Manufacturers
Association
Kimball, Inc.
Mailing Address
P.O. Box 1967
Mail Code: PS
Grand Rapids, MI 49501
2680 Horizon Drive, SE
Suite A-l
Grand Rapids, MI 49546
P.O.Box 1109
Liberty, NC 27298
1500 Rhode Island Avenue, NW
Washington, DC 20005

40 Loudonville Road
Albany, NY 12204
1500 Rhode Island Avenue, NW
Washington, DC 20005
2000 Town Center
Southfield, MI 48075
P.O. Box HP-7
High Point, NC 27261
1155 West 12th Avenue
Jasper, IN 47549
Telephone/Fax Number
(616) 246-9787
fax-(6 16) 246-91 91
(616) 285-3963
fax-(6 16) 285-3765
(336) 622-0120
fax-(336) 622-1050
(202) 462-6272
fax-(202) 462-8549
fax-(773) 624-7337
(518)449-2040
fax-(5 18) 449-403 6
(202) 462-6272
fax-(202) 462-8549
(284) 784-5088
fax-(284) 784-3250
(910) 884-5000
fax-(910) 884-5303
(812)634-3274
fax-(8 12) 634-3250
e-mail Address
mmika@steelcase.com
bmiller@bifma.com
cmillisor@mindspring.com
bnelson@paint.org


mroddy @paint. org
rross@wnj.com
lfrun@aoi.com
staschm@kimball . com
             B-8

-------
TABLE B-l. LIST OF STAKEHOLDERS
Name
Jim Sell
Jo Spiceland
Andy Sticker
Sherry Stookey
Ron Tucker
Wayne Vangsness
Robert Walp
Ronald Westgate
Bob Wood
Lynn Zimmerman
Bernard Zysman
Company
National Paint and
Coatings Association
Charleston Forge
Darling Store Fixtures
Lilly Industries
Lilly Industries
National Metal
Industries
Penco Products
Lightolier
Lexington Furniture
Industries
Steelcase, Inc.
Occidental Chemical
Mailing Address
1500 Rhode Island Avenue, NW
Washington, DC 20005
251 Industrial Park Dr.
Boone, NC 28607

P.O. Box 2358
High Point, NC 27261
2137BrevardRoad
High Point, NC 27261
203 Circuit Avenue
West Springfield, MA 01089
99 Brower Avenue
Oaks, PA 19456
63 1 Airport Road
Fall River, MA 02720
P.O. Box 1008
Boone, NC 27293
P.O. Box 1967
Mail Code: PS
Grand Rapids, MI 49501
P.O. Box 344
Niagara Falls, NY 14302
Telephone/Fax Number
(202) 462-6272
fax-(202) 462-8549
(828) 264-4901
fax-(828) 264-5901
fax-(870-23 9-6429
(336) 802-43305
fax-(336) 889-6007
(910) 802-4337
fax-(910) 889-6007
(413)785-5861
fax-(413) 737-2309
(610)666-0500
fax-(610) 650-5257
(508) 646-3341
fax-(508) 674-4710
(910)249-5316
fax-(910) 249-5588
(616)475-2183
fax-(6 16) 246-91 91
(716) 278-7894
fax-(7 16) 278-7297
e-mail Address
jsell@paint.org
kspiceland@yahoo. com

stookey s@lillyindustries. com


bwalp@pencoproducts. com
rwestgate@gentyte.com
bwood@infoave.net
Izimmerl @steel case.com
bernie zysman@oxy.com
             B-9

-------
B-10

-------
TABLE B-2. METAL FURNITURE SITE VISIT FACILITIES
COMPANY VISITED
American Seating Company
Grand Rapids, Michigan
Charleston Forge
Boone, North Carolina
HON Industries
Cedartown, Georgia
Johnston Casuals
North Wilkesboro, North Carolina
Metal Creations
High Point, North Carolina
Steelcase, Incorporated
Grand Rapids, Michigan
(Two Facilities)
Royal Development
High Point, North Carolina
U.S. Furniture
High Point, North Carolina
PRODUCTS PRODUCED
Stadium Seating and Public Transportation Seating
Residential Furniture
Office Furniture
Residential Furniture
Residential Furniture
Office Furniture
Recliner Mechanisms
Residential Furniture
                     B-ll

-------
TABLE B-3. METAL FURNITURE INDUSTRY QUESTIONNAIRE RECIPIENTS
Facility Name
Lozier Corporation - Scottsboro Plant
Steelcase - Athens Plant
Dental Ez Group
Darling Store Fixtures - Paragould Plant
Darling Store Fixtures - Corning Plant
Hickory Springs Manufacturing Compant
The HON Company - South Gate
Professional Refinishing Organization
Atlas Spring Manufacturing Organization
Virco Manufacturing Corporation
Leggett & Platt Incorporated
Steelcase - Tustin Facility
Revest, Incorporated
Lithonia Down Lighting
The HON Company - Cedartown Facility
The HON Company - Oak Steel Facility
Harpers, Incorporated
B-Line Systems - Highland Plant
Artec Manufacturing
Mid-West Chandelier Company
Leggett & Platt, Incorporated
Leggett & Platt, Incorporated
Lightolier - Fall River Facility
National Metal Industries
Steelcase - Systems I Plant
Steelcase - Computer Furniture Plant
City
Scottsboro
Athens
Bay Minette
Paragould
Corning
Fort Smith
South Gate
Los Angeles
Gardena
Torrance
Whittier
Tustin
Lithia Springs
Vermillion
Cedartown
Muscatine
Post Falls
Highland
Jasper
Kansas City
Simpsonville
Winchester
Fall River
West Springfield
Grand Rapids
Kentwood
State
AL
AL
AL
AR
AR
AR
CA
CA
CA
CA
CA
CA
GA
GA
GA
IA
ID
IL
IN
KS
KY
KY
MA
MA
MI
MI
                           B-12

-------
TABLE B-3. METAL FURNITURE INDUSTRY QUESTIONNAIRE RECIPIENTS (cont.)
Facility Name
Steelcase - Desk Plant
Steelcase - Panel Plant
Steelcase - File Plant
Steelcase - Chair I Plant
Steelcase - Context Plant
Super Sagless
PENCO Products
Metal Creations
Leggett & Platt - Linwood Branch Facility
Collier-Keyworth, Incorporated
High Point Sleeper
Hickory Springs Manufacturing Company
Lozier Corporation - Omaha North Plant
Lozier Corporation - Omaha West Plant
Davies Office Refurbishing, Incorporated
Republic Storage Systems Company, Incorporated
PENCO Products
Harris Hub
HON/ALLSTEEL
Duro Metal Manufacturing
Revest, Incorporated
PENCO Products - Salt Lake Plant
EST Division of Leggett Partners
City
Grand Rapids
Kentwood
Grand Rapids
Grand Rapids
Kentwood
Tupelo
Vicksburg
High Point
Linwood
Liberty
High Point
Hickory
Omaha
Omaha
Albany
Canton
Oaks
York
West Hazelton
Dallas
Farmers Branch
West Jordan
Grafton
State
MI
MI
MI
MI
MI
MS
MS
NC
NC
NC
NC
NC
NE
NE
NY
OH
PA
PA
PA
TX
TX
VT
WI
                              B-13

-------
TABLE B-4. SUMMARY OF DATA CONTRIBUTED TO THE EPA FROM STATES
State/Local Agency
Alabama DEM
Bay Area AQMD (California)
South Coast AQMD (California)
Ventura County APCD
(California)
California Air Resources Board
Illinois EPA
Indiana
Michigan
Missouri DNR
Ohio EPA
Tennessee Metropolitan
Government of Nashville and
Davidson Counties
Chattanooga-Hamilton County
APCB (Tennessee)
Texas
Wisconsin DNR
Data Contributed
Listing of metal furniture manufacturing facilities
Emission inventory listing and Regulation 8, Rule 14: Surface
Coating of Large Appliances and Metal Furniture
AQMD Rule 1 107-Coating of Metal Parts and Products, and
AQMD BACT for metal furniture
Facility permits
ARE Database of surface coating facilities
Title V permit applications for three facilities; Initial CAAPP permits
for three facilities, facility list of metal furniture manufacturers
Airs Facility Subsystem Quick Look Report, Facility emissions data
by SCC code, and Voluntarily reported data for the 189 HAPs
Seven Title V permit applications and multiple operating permit
applications
Facility operating permits, emissions inventories, Title V permit
applications
STARDUST Database, Ohio BAT Clearinghouse Data, and Title
V permit applications for three facilities
Construction permit, Title V permit for one facility, VOC Report,
and Construction and Operating Permit for one facility
Engineering reports for two facilities, Material Safety Data Sheets
on powder coating
Chapter 115 surface coating rules and definitions, database and
mailing list for fabricated metal products
Listing of Title V and synthetic minor facilities
                           B-14

-------
                        APPENDIX C

     STANDARD INDUSTRIAL CLASSIFICATION (SIC) CODE AND
NORTH AMERICAN INDUSTRY CLASSIFICATION SYSTEM (NAICS) CODE
                     DATA SUMMARIES

-------
                              TABLE C-l. SIC CODES
                   (All products listed for each code are metal furniture)
SIC Code
           Description
            Typical Products
  2514
Metal Household Furniture
Bookcases, Chairs, Tables, Swings,
Kitchen Cabinets, Medical Cabinets, Camp
Furniture, Frames for Boxsprings, Cribs,
Cots, Garden Furniture, Serving Carts
  2522
Office Furniture, Except Wood
Bookcases, Chairs, Tables, Desks, File
Cabinets, Wall Cases, Partitions, Modular
Furniture, Benches
  2531
Public Building and Related Furniture
Benches, Portable Bleacher Seating,
Stadium Seating, Theater Seating, School
Furniture, Church Furniture
  2542      Office and Store Fixtures, Partitions,
             Shelving, and Lockers, Except
             Wood
                                   Cabinets, Counters, Display Cases, Display
                                   Fixtures, Bar Fixtures, Shelving,
                                   Showcases, Sorting Racks, Lunchroom
                                   Fixtures
  3645
Residential Electric Lighting Fixtures
Chandeliers, Floor Lamps, Lamps, Wall
Lamps, Desk Lamps, Lamp Shades
(metal), Table Lamps
  3646      Commercial, Industrial, and
             Institutional Electric Lighting Fixtures
                                   Chandeliers (commercial), Desk Lamps
                                         C-l

-------
                                  TABLE C-2.  SIC CODES
                             (Only products listed are metal furniture3)
SIC Code
2599
3429
3469
3495
3499
3821
3843
3999
7641
Description
Furniture and Fixtures, Not
Elsewhere Classified
Hardware, Not Elsewhere Classified
Metal Stampings, Not Elsewhere
Classified
Wire Springs
Fabricated Metal Products, Not
Elsewhere Classified
Laboratory Apparatus and Furniture
Dental Equipment and Supplies
Manufacturing Industries, Not
Elsewhere Classified
Reupholstery and Furniture Repair
Typical Products
Hospital Beds, Bowling Center Furniture,
Cafeteria Furniture, Factory Furniture, Ship
Furniture
Furniture Hardware, Convertible Bed
Mechanisms
Wastebaskets, Stamped Metal
Furniture Springs, Spring Units for Seats
Metal Chair Frames, Metal Furniture Parts
Laboratory Furniture, Benches, Tables,
Cabinets
Dental Cabinets, Dentists' Chairs
Beauty Shop and Barber Shop Furniture
Furniture Repair/Refinishing, Antique
Repair Restoration
a These SIC code descriptions contain many other products that are outside the scope of the metal
furniture source category and are not listed here. This table only includes the products that are
considered to be within the scope of the metal furniture source category.
                                             C-2

-------
TABLE C-3.  SIC CODES AND CORRESPONDING NAICS CODES
Category
Metal Household
Furniture
Office Furniture,
Except Wood
Public Building and
Related Furniture
Office and Store
Fixtures, Partitions,
Shelving, and Lockers,
Except Wood
Furniture and Fixtures,
Not Elsewhere
Classified
Hardware, Not
Elsewhere Classified
Metal Stampings, Not
Elsewhere Classified
(Except Kitchen
Utensils, Pots and Pans
for Cooking and Coins)
Wire Springs
Fabricated Metal
Products, Not
Elsewhere Classified
Residential Electric
Lighting Fixtures
Commercial, Industrial,
and Institutional Electric
Lighting Fixtures
Laboratory Apparatus
and Furniture
1987 SIC
Code
2514
2522
2531
2542
2599
3429
3469
3495
3499
3645
3646
3821
Equivalent 1997
NAICS Code
337124
337214
337127*
337215
(b)
332510°
332116d
3326126

335121
335122
339111
Equivalent 1997 NAICS Category
Metal Household Furniture
Manufacturing
Nonwood Office Furniture
Manufacturing
Institutional Furniture Manufacturing
Showcase, Partition, Shelving, and
Locker Manufacturing
Institutional Furniture Manufacturing
Hardware Manufacturing
Metal Stamping
Wire Spring Manufacturing
Showcase, Partition, Shelving, and
Locker Manufacturing
Residential Electric Lighting Fixture
Manufacturing
Commercial, Industrial, and
Institutional Electric Lighting Fixture
Manufacturing
Laboratory Apparatus and Furniture
Manufacturing
                       C-3

-------
       TABLE C-3. SIC CODES AND CORRESPONDING NAICS CODES (continued)
Category
Dental Equipment and
Supplies
Manufacturing
Industries, Not
Elsewhere Classified
Reupholstery and
Furniture Repair
1987 SIC
Code
3843
3999
7641
Equivalent 1997
NAICS Code
339114


Equivalent 1997 NAICS Category
Dental Equipment and Supplies
Manufacturing
Institutional Furniture Manufacturing
Reupholstery and Furniture Repair
a Includes 3371271, 3371274.
b Includes 3391137 and 3371277/A.
c Only includes 3325101.
d Only includes 3321165.
e Only includes 3326124.
                                         C-4

-------
TABLE C-4. ESTIMATED NATIONWIDE NUMBER OF METAL FURNITURE FACILITIES
                       BYNAICS AND SIC CODE
NAICS Code
337124
337214
337127
337215
335121
335122
3391 13 and
337127
332510
332116
332612
339111
339114
SIC Code
2514
2522
2531
2542
3645
3646
2599
3429
3469
3495
3821
3843
Description
Metal Household Furniture
Office Furniture, Except Wood
Public Building Furniture
Office and Store Fixtures, Partitions,
Shelving, and Lockers, Except Wood
Residential Lighting Fixtures
Commercial and Industrial Lighting
Fixtures
Furniture and Fixtures, not elsewhere
classified
Hardware, not elsewhere classified
Metal Stampings, not elsewhere
classified
Wire Springs
Laboratory Apparatus and Furniture
Dental Equipment and Supplies
Total
Number of
Facilities
163
194
150
466
245
211
303
59
298
180
384
349
3,002
                               C-5

-------
           Table C-5. Estimated Nationwide Number of Major and Area Sources in the
                                  Metal Furniture Industry
NAICS
Code
337124
337214
337127
337215
335121
335122
339113
and
337127
332510
332116
332612
339111
339114
Total
SIC
Code
2514
2522
2531
2542
3645
3646
2599
3429
3469
3495
3821
3843

Total
Number
of
Facilities
163
194
150
466
245
211
303
59
298
180
384
349
3,002
Percent
Major3
44
43
25
38
23
16
23
18
15
14
5
8

Nationwide Number of Facilities
Major
Sources
72
83
38
177
56
33
69
11
44
26
19
27
655
Area
Sources'3
91
111
112
289
189
178
234
48
254
154
365
322
2,347
Percent Area
Sources in
Urban Areas
51
44
22
38
85
62
56
61
58
60
76
80

Area
Sources in
Urban
Areas
46
49
25
110
161
110
131
29
147
92
277
258
1,435
a From TRI data.
b May include synthetic minor sources.
                                          C-6

-------
                      APPENDIX D




ESTIMATED EMISSION REDUCTION AND COST OF POWDER COATING

-------
                    Estimated Emission Reduction and Cost of Powder Coating

       Although not considered a technically feasible beyond-the-floor regulatory option for the entire
metal furniture source category, there may be some sources that would chose to use powder coatings
to reduce organic HAP emissions. Therefore, this appendix presents estimated organic HAP emission
reductions achievable beyond the MACT floor level of control and the estimated cost to achieve these
reductions.

A. Emission Reduction of Powder Coating

       We observed through site visits and questionnaire responses that many metal furniture surface
coating facilities use powder coatings for only a portion of their coating needs.  Thus, we determined
what portion of each model plant's production would have to be converted to powder coating, in
conjunction with conversion to all non-HAP cleaning materials, to achieve an emission rate less than the
existing source MACT floor.  As shown in Table D-l,  each model plant would have to convert all
liquid coating usage to powder coating to achieve an emission rate that represented a level of control
more stringent than that achieved by the MACT floor technology.

       At 75 percent conversion to powder coatings, the emission rate for each of the model plants
was approximately 0.09 kg HAP/L coating solids, which is at the low end of the emission rate range
represented by low organic HAP content coatings. Each model plant would have to convert all liquid
coating usage to powder coating to achieve a greater emission reduction than the existing source
MACT floor. Therefore, the emission reduction and costs presented in this appendix represent that
associated with complete conversion to powder coatings. We did not consider any level of conversion
between 75 and 100 percent because the available cost data were not sufficiently refined to allow such
an incremental analysis.

B. Powder Coating Cost Estimate

       The capital cost of the powder coating line was based on the conversion of a liquid coating line
to powder.  By using only the conversion cost, we effectively accounted for the cost that would have
been incurred for the liquid coating line. We also subtracted the cost of liquid coatings that would have
been used by the facility, then added back the cost of an equivalent amount of powder coatings.

       Cost information was obtained for the operation of a powder coating line from one metal
furniture facility,1 which was used to estimate the annual cost associated with powder coating for large
model plants.  Additional capital and annual cost information was obtained from a published case
       1 Confidential Business Information provided by a metal furniture manufacturer. July 1999.

                                            D-l

-------
study.2  Since the coating solids usage for this case study facility was between that for the small and
medium model plants, this information was used to estimate the capital costs for both of these model
plant sizes.

        For liquid coatings, a single average cost value was used which encompasses solventbased,
waterbased, higher solids content, lower solids content, and a range of organic HAP content coatings.
A single value was used for liquid coatings because of the wide range of coatings that are available.
Since it was not possible to determine what mix of coatings may be used by any particular facility, an
average liquid coating cost was determined to be the most accurate representation of this cost. Liquid
coating costs obtained through published literature3 were converted from cost per gallon to cost per
liter coating solids. These values were then averaged to obtain the liquid coating cost used for this
analysis ($12.05/L coating solids). This average value was then scaled to 1998 dollars using the
Chemical Engineering Plant Cost Index.4

        Similarly, the cost of powder coatings vary according to the supplier, volume purchased, and
resin system, to name just a few factors. The metal furniture industry provided a range of costs for
powder coatings, varying from about $7/kg to $26/kg. Again, since it was not possible to determine
the mix of coatings used by a facility, we chose the midpoint of the range, $17/kg, as the most accurate
representation of this cost.

        For this analysis, it was assumed that the total coating solids used by a facility that uses powder
coating application operations would decrease as compared to the coating solids used with liquid
coating operations due to the ability to recycle the powder. Based on a published case study,5 the
amount of powder solids used was 31 percent less than the equivalent coating solids from liquid
coatings.

        Table D-2 presents the costs, additional organic HAP emission reduction, and cost per
megagram of additional emission reduction for existing model plants for the regulatory alternative of
conversion to powder coating.  Table D-3 presents this same information for new model plants.  For
existing model plants,  the cost per megagram of additional emission reduction ranged from $51,000 to
       2 Pollution Prevention in Metal Painting and Coating Operations:  A Manual for Pollution
Prevention Technical Assistance Providers.  The Northeast Waste Management Officials' Association.
Boston, MA.  1998. p. 78.

       3 Bocchi, Gregory; Products Finishing. "Powder Coating Advantages," The Powder Coating
Institute; June 1997.

       4 Chemical Engineering. Chemical Engineering Plant Cost Index,  June 1998.

       5 Note 2.

                                             D-2

-------
$70,000 ($46,000 to $64,000 per ton). For new model plants, these costs ranged from $65,000 to
$89,000 per megagram of additional organic HAP emission reduction ($59,000 to $81,000 per ton).
                                           D-3

-------
              Table D-l.  Organic HAP Emission Rates Estimated to be Achievable By Conversion to Thermal/IR Curable
                         Powder Coating for Existing and New Metal Furniture Surface Coating Model Plants3






Model
Plant
Small
Mediu
m
Large




(A)
Total
Coating
Solids
Usage
(L/yr)
22,000
54,000

250,00
0



(B)
Total
Organic
HAP
Emission
s (kg/yr)
7,500
19,600

92,400

Amount of Liquid Coating Usage Converted to Powder Coatingb: c
25 Percent
(C)
HAP
Emissions
After
Conversion
(kg/yr)
5,600
14,700

69,300

(D)
Emission
Rate6
(kg
HAP/L
coating
solids)
0.255
0.272

0.277

50 Percent
(E)
HAP
Emissions
After
Conversio
nf
(kg/yr)
3,800
9,800

46,200

(F)
Emission
Rateg
(kg
HAP/L
coating
solids)
0.173
0.181

0.185

75 Percent
(G)
HAP
Emissions
After
Conversio
nh
(kg/yr)
1,900
4,900

23,100


(H)
Emission
Rate1
(kg HAP/L
coating
solids)
0.086
0.091

0.092

100 Percent
(I)
HAP
Emissions
After
Conversion
j
(kg/yr)
0
0

0

(J)
Emission
Ratek
(kg
HAP/L
coating
solids)
0
0

0

a Assumes that existing and new model plants would have the same coating solids usage and organic HAP emissions in the absence of a
standard.
b HAP emissions after conversion to powder coatings assumes that there are no cure volatile emissions from the powder coatings.
c Emission rate after conversion assumes that the coating solids usage will not change.
dC = Bx(100-25)/100
e D = C/A
fE = Bx(100-50)/100
« F = E/A
hG = Bx(100-75)/100
; H = G/A
jI = Bx(100-100)/100
                                                          D-4

-------
J =
                                                  D-5

-------
                Table D-2. Estimated Model Plant Cost per Megagram of Organic HAP Emission Reduction for Existing
                            Metal Furniture Surface Coating Facilities for Conversion to Powder Coating
Model
Plant







Small
Medium
Large
(A)
Model
Plant
Coating
Solids
Usage3
(L/yr)


22,000
54,000
250,000
(B)
Model Plant
Annual
Costsb
(1998 $)




184,000
328,000
1,674,000
(C)
Model Plant
Capital Costs'3
($)





550,000
550,000
3,350,000
(D)
Baseline Level
of Control0
(kgHAP/L
coating solids
used)



0.12
0.12
0.12
(E)
Level of Control
After Implementing
Powder Coatings
(kg FLAP/L coating
solids used)



0
0
0
(F)
Additional
Model Plant
Organic FLAP
Emission
Reductiond
(Mg/yr)


2.64
6.48
30.0
(G)
Model Plant
Annual Cost per
Mgof
Additional
Organic FLAP
Emission
Reduction6
($/Mg)
70,000
51,000
56,000
aSource: Memorandum from Hendricks, D., EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September 14, 2001.  Model Plants for the
Metal Furniture Surface Coating Source Category.
b Annual and capital costs presented are the additional costs incurred beyond the baseline.
c The baseline is the MACT floor level of control. For details on the MACT floor, see: Memorandum from Hendricks, D., and Holmes, K.,
EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September 19, 2001.  Recommended MACT Floors for Existing and New Major Sources
for the Metal Furniture Surface Coating Source Category.
dF = (D-E)xA/l,000
eG = B/F
                                                          D-6

-------
                 Table D-3.  Estimated Model Plant Cost per Megagram of Organic HAP Emission Reduction for New
                            Metal Furniture Surface Coating Facilities for Conversion to Powder Coating
Model
Plant







Small
Medium
Large
(A)
Model
Plant
Coating
Solids
Usage3
(L/yr)


22,000
54,000
250,000
(B)
Model Plant
Annual
Costsb
(1998 $)




184,000
328,000
1,674,000
(C)
Model Plant
Capital Costs'3
($)





550,000
550,000
3,350,000
(D)
Baseline Level
of Control0
(kgHAP/L
coating solids
used)



0.094
0.094
0.094
(E)
Level of Control
After Implementing
Powder Coatings
(kg FLAP/L coating
solids used)



0
0
0
(F)
Additional
Model Plant
Organic FIAP
Emission
Reductiond
(Mg/yr)


2.07
5.08
23.5
(G)
Model Plant
Annual Cost per
Mgof
Additional
Organic FLAP
Emission
Reduction6
($/Mg)
89,000
65,000
71,000
aSource: Memorandum from Hendricks, D., EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September 14, 2001. Model Plants for the
Metal Furniture Surface Coating Source Category.
b Annual and capital costs presented are the additional costs incurred beyond the baseline.
c The baseline is the MACT floor level of control. For details on the MACT floor, see: Memorandum from Hendricks, D., and Holmes, K.,
EC/R Inc., to Serageldin, M., EPA:ESD:CCPG. September 19, 2001.  Recommended MACT Floors for Existing and New Major Sources
for the Metal Furniture Surface Coating Source Category.
dF = (D-E)xA/l,000
eG = B/F
                                                          D-7

-------
              APPENDIX E

     DETAILED COST CALCULATIONS FOR
PERMANENT TOTAL ENCLOSURES AND OXIDIZERS

-------
Regenerative Thermal Oxidizer Cost Calculations
   1. WASTE GAS HEAT CONTENT CALCULATED BASED ON UNCONTROLLED EMISSIONS AND EXHAUST FLOWRATE
   2. PTE BASED ON LUKEY, SPOT A/C, ASSUMED DUCTWORK COSTS INCLUDED IN TOTAL CAPITAL COST
   SINCE UNITS ARE FIELD ERECTED.
   TOTAL ANNUAL COST SPREADSHEET PROGRAM-REGENERATIVE THERMAL OXIDIZERS
                 FLOW <500,000 SCFM
   COST BASE DATE:  December 1988 [1]
   VAPCCI (1998): [2]
                              INPUT PARAMETERS
   MODEL PLANT
   — Gas flowrate (scfm):
   — Reference temperature (oF):
   — Inlet gas temperature (oF):
   — Inlet gas density (Ib/scf):
   — Primary heat recovery (fraction):
   -- Waste gas heat content (BTU/scf):
   -- Waste gas heat content (BTU/lb):
   -- Gas heat capacity (BTU/lb-oF):
   — Combustion temperature (oF):
   — Heat loss (fraction):
   — Exit temperature (oF):
   -- Fuel heat of combustion (BTU/lb):
   -- Fuel density (Ib/ft3):
Large
                  Medium
                                       Small
200000
77
100
0.0739
0.95
0.030
0.41
0.255
1600
0.01
175
21502
0.0408
100000
77
100
0.0739
0.95
0.030
0.41
0.255
1600
0.01
175
21502
0.0408
100000
77
100
0.0739
0.95
0.030
0.41
0.255
1600
0.01
175
21502
0.0408
                              DESIGN PARAMETERS
   Auxiliary Fuel Requirement (Ib/min):
                                 (scfm):
   Total Gas Flowrate (scfm):
            15.558
             381.3
            200381
    7.779
    190.7
  100191
    7.779
    190.7
  100191
                 TOTAL CAPITAL COST ($) [3]
                 (Cost correlations range: 5000 to 500,000 scfm)
                 @ 85 % heat recovery—base:
                  '  '     '  —escalated:
                 @ 95 % heat recovery—base:
                  '  '     '  —escalated:
   PERMANENT TOTAL ENCLOSURE
   TOTAL CAP COST($)
                 0
                 0
         5,149,540
         6,127,576
           216,000
         6,343,576
       0
       0
2,853,170
3,395,063
  117,300
3,512,363
       0
       0
2,853,170
3,395,063
  117,300
3,512,363
                                                         E-l

-------
Regenerative Thermal Oxidizer Cost Calculations
                                ANNUAL COST INPUTS
 Operating factor (hr/yr):
 Operating labor rate ($/hr):
 Maintenance labor rate ($/hr):
 Operating labor factor (hr/sh):
 Maintenance labor factor (hr/wk):
 Electricity price ($/kwh):
 Natural gas price ($/mscf):
 Annual interest rate (fraction):
 Oxidizer  control system life (years):
 Oxidizer  capital recovery factor:
 Permanent total enclosure control system life (years):
 Permanent total enclosure capital recovery factor:
 Taxes, insurance, admin, factor:
 Pressure drop (in. w.c.):
6600
37.61
41.37
1
1
0.05
3.10
0.07
10
0.1424
30
0.0806
0.04
20.0
6600
37.61
41.37
1
1
0.05
3.10
0.07
10
0.1424
30
0.0806
0.04
20.0
                                        6600
                                       37.61
                                       41.37
                                            1
                                            1
                                        0.05
                                        3.10
                                        0.07
                                          10
                                      0.1424
                                          30
                                      0.0806
                                        0.04
                                        20.0
                                    ANNUAL COSTS
     Item
                                                               Cost ($/yr)
            Cost ($/yr)
           Cost ($/yr)
 Operating labor
 Supervisory labor
 Maintenance labor
 Maintenance materials
 Natural gas
 Electricity
 Overhead
 Taxes, insurance, administrative
 Oxidizer capital recovery
 Permanent total enclosure capital recovery
 Permanent total enclosure related electricity cost
15,515
2,327
2,151
2,151
467,966
232,617
13,287
253,743
872,429
17,407
87,232
15,515
2,327
2,151
2,151
233,983
116,309
13,287
140,495
483,381
9,453
43,616
                                       15,515
                                       2,327
                                       2,151
                                       2,151
                                     233,983
                                     116,309
                                       13,287
                                     140,495
                                     483,381
                                       9,453
                                       43,616
 Total Annual Cost
                                Indirect Cost
                                Direct Cost
$1,966,826
$1,156,865
  $809,960
$1,062,667
  $646,615
  $416,052
$1,062,667
  $646,615
  $416,052
 [1] Base total capital investment reflects this date.
 [2] VAPCCI = Vatavuk Air Pollution Control Cost Index (for regenerative
 thermal oxidizers) corresponding to year and quarter shown. Base
 total capital investment has been escalated to this date via VAPCCI and
 control equipment vendor data. Available at:  http://www.epa.gov/ttn/catc/products.htmljifcccinfo
 [3] Source: Vatavuk, William M. ESTIMATING COSTS OF AIR POLLUTION
 CONTROL. Boca Raton, FL: Lewis  Publishers, 1990.

 Assumptions:
 1) Monitoring and recordkeeping costs are not included
 2) Permanent total  enclosure (PTE) costs estimated based on case studies by M. Lukey, PES, and engineering judgement.
 3) Permanent total enclosure costs assume  engineering =  10% PTE cost; spot air conditioning,  30 year life.
 4) Because regenerative thermal incinerators are field erected, it is assumed that ductwork costs are included
 in the Total Capital Cost estimate.
 5) Electricity cost $0.0451/kwh, natural gas cost $3.099/mscf, both based on information from Energy Information
 Administration for 1998.
 6) Operator labor rate = 17.91/hr*1.67=$29.91/hr, maintenance labor rate  = l.l*operator rate =$32.90/hr.
 Both were based on Bureau of Labor Statistics data for March 1999.

 Revised on 11/9/00, 9/26/01. 10/5/01
                                                              E-2

-------
Regenerative Thermal Oxidizer Cost Calculations
          Calculate waste gas heat contents for each of the model plants
          Calculation based on average uncontrolled emission rate for  each model plant in kg HAP/I solids (See MF-MACTFLOORREV16)
          Assumes HAP is xylene with heat of combustion (Btu/lb)  =                                           17559
          Annual Operating hours=                          6600
Model Plant Exhaust Flowrate Unctrl Emissions Unctrl Emissions Gas stream ht of combustion Corresponding Cone.
(scfm) (tpy) (Ib/min) (Btu/scf) (ppm)
Small
Medium
.arge
100,000
100,000
200,000
4.844101295
11.711492595
67.557824125
0.0245
0.0591
0.3412
0.0043
0.0104
0.0300
1
2
6
          Gas stream heat of combustion (Btu/scf) = [unctrl. emiss. (lb/min)]*[xylene ht. of combustion (Btu/lb)]/exhaust flowrate (scfm)

          Concentration (ppm)  ={[unctrl. emissions (lb/min)]*[397 ft3 xylene/lb mole xylene]/[106.16 Ib xylene/lb mole xylene]}*{[1000000/exhaust(scfm)]}
          Small, Medium, and  Large model plants - assumes that all emissions/lines vented to control device
                                                                          E-3

-------
Permanent Total Enclosure Cost Calculations
        AWMA-based  capital A/C cost, 30 yr life
        Base PTE Cost               $50,000
        Base Room Volume (KFT3)
                                                          PERMANENT TOTAL ENCLOSURE (PTE) CAPITAL COSTS
Spot A/C Factor    0.00125
                                                  270


Model Plant
>mall
Medium
.arge


Room Vol.
(KFT3)
180
180
270

Calc. PTE
Cost
($)
$33,333
$33,333
$50,000
(A)
PTE Cost
to Use [1]
($)
$33,000
$33,000
$50,000
(B)
Engineering
Cost
($)
$3,300
$3,300
$5,000


Exhaust
(scfm)
100,000
100,000
200,000
A/C
Capacity
Needed
(tons)
125
125
250

Calc. A/C
Capital Cost
($)
$150,000
$150,000
$300,000
AWMA-based
A/C Capital
Cost
($)
$81,031
$81,031
$160,831
(C)
A/C Capital
Cost to Use
[1]
($)
$81,000
$81,000
$161,000
A/C
Electrical
Use
(kW)
147
147
293

Total PTE Capital
Cost (A+B+C)
($)
$117,300
$117,300
$216,000
        [1] - "to use" refers to the values used to determine the total capital cost and reflect the significant figures in the calculated values.
        Assumptions:
        - Room Volume based on information obtained from industry surveys and scaled by model plant coating lines
        - Base PTE Cost based on case studies  by M. Lukey, PES, and engineering judgement
        - PTE costs of model plants based on estimated size of the  enclosure, and engineering judgement
        - Engineering cost estimated as  10% of PTE cost
        - A/C calculations assume spot air conditioning is installed
        - A/C capacity based on cost factors presented by M. Lukey, PES, as 25 tons/20,000 scfm
        -A/C cost based on  cost factors  presented by M. Lukey, PES, as $30,000 per 25 tons
        -AWMA A/C cost estimated using formulas in AWMA Lukey/EPA PTE costing spreadsheet
        -Electricity required for calculated A/C capacity calculated using equation presented in "Mechanical Engineering Reference
           Manual", M. Lindeburg, 8th Edition.  1990.  Page 7-28.
        Small - represents 2 coating lines at 180 kft3
        Medium - represents 2 coating lines at 180 kft3
        Large - represents 4 coating lines at 270 kft3
                                                                                     E-4

-------
Permanent Total Enclosure Cost Calculations
                                               MAKEUP AIR FAN COST

           From AWMA spreadsheet, the makeup air fan cost was $5,733 for an air flow rate of 26,200 scfrn. This cost
           was scaled by the ratio of the calculated makeup ar flow rate from "Makeup Ar Flow/rate"
           and the AWMA ar flow rate (87,333/26,200).
                                      Calc'd Makeup Air Flow from
            Model Plant Total Exhaust    "Makeup Air Flowrate"    Scaled Makeup Fan Cost
                            (scfrn)                (scfrn)                      ($)
           Lage                200,000                       174,667                 38,220
           Medium              100,000                       87,333                 19,110
           Small                100,000                       87,333                 19,110
                                                    E-5

-------
Permanent Total Enclosure Cost Calculations
                                                            SPOT AIR CONDITIONING COST
           Will assume spot air conditioning is needed.
           Spot air conditioning refers to the use of small air conditioning units placed where needed, rather than central air conditioning
           for the entire enclosure.
           Note:  AWMA example assumed that no A/C was needed. However, formulas in the spreadsheet were used to calculate the
           materials and installation costs for both total and spot A/C
           Formulas for spot vs total A/C apply different multipliers to (scfm), so assume scfm to be entered  is TOTAL EXHAUST
           and not just the amt. of exhaust cooled by spot A/C
           Spreadsheet formula spot A/C:
                        Materials ($)    (987 + (0.693*scfm))
                        Installation ($)   (244 + (0.105*scfm))

           Model Plant   Total Exhaust   Spot A/C material Cost  Spot A/C Installation Cost   Spot A/C Total Cost              Spot A/C capacity
                             (scfm)               ($)                      ($)                    ($)                      (tons)
           Large                200,000               139,587                   21,244             160,831                  250
           Medium              100,000                70,287                   10,744              81,031                  125
           Small                100,000                70,287                   10,744              81,031                  125
           (Spot A/C capacity from ptemr&r.wk4)
                                                                   E-C

-------
Permanent Total Enclosure Cost Calculations
                                                   MAKEUP AIR FLOW/RATE
           The electricity associated with operation of the makeup air fan was calculatedtaking the following factors into account:
           1. The electricity usage is a function of makeup air flowrate.
           2. Because we do not have makeup air flowrates and they cannot easily be calculated, they were estimated for
           electricity usage purposes by applying the ratio of makeup airflow to total exhaust flow from the example in
           the AWMA spreadsheet.
           3. All other makeup air fan electricity usage related parameters from AWMA spreadsheet were used (press, drop, etc.)
           Ratio of makeup airflow to total exhaust from AWMA spreadsheet:  (26,200/30,000) =               0.87

           Model Plant Total Exhaust  Calc'd Makeup Air flowrate
                           (scfm)              (scfm)
           Large               200,000                 174,667
           Medium            100,000                  87,333
           Small              100,000                  87,333
                                                             E-7

-------
Permanent Total Enclosure Cost Calculations
                            ANNUAL COSTS ASSOCIATED WITH INSTALLATION AND OPERATION OF
                                           PERMANENT TOTAL ENCLOSURE (PTE)
            MODEL PLANT

            TOTAL CAPITAL INVEST.
                                 ANNUAL COST INPUTS
            Operating hours per year                 6600
            Electricity price ($/kwh):                  0.05
            Annual interest rate (fraction):             0.07
            PTE system life (years):                    30
            Capital recovery factor:                 0.0806
Small      Medium   Large

  $117,300   $117,300   $216,000
                                     ANNUAL COSTS (1998$)
                Item
            Electricity
            Capital recovery
            Total Annual Cost
 Cost($/yr)

    43,616
     9,453
43,616
 9,453
87,232
17,407
    53,069     53,069    104,639
            Assumptions:

            1) Base PTE Cost based on case studies by M. Lukey, PES, and engineering judgement
            2) PTE costs of model plants based on estimated size of the enclosure, and engineering judgement
            3) Engineering cost estimated as 10% of PTE cost
            4) AC calculations assumespot air conditioning is installed
            5) AC capacity based on cost factors presented by M. Lukey, PES, as 25 tons/20,000 scfm
            6) AC cost based on equations in AWMA Lukey/EPA paper and assoc. spreadsheet
            7) Electricity required for calculated AC capacity calculated  using equation  presented  in
               "Mechanical Engineering Reference Manual", M. Lindeburg, 8th Edition. 1990. Page 7-28.
            8) Capital recovery based  on a 30 year equipment life based on AWMA Lukey/EPA paper
            9) Electricity cost $0.0451/kwh, based on info, from Energy  Information Administration for 1998

-------
                           APPENDIX F

CALCULATION METHODOLOGY FOR THE AFFECTED-SOURCE-WIDE EMISSION RATE
USED TO DETERMINE THE METAL FURNITURE NESHAP MACT FLOOR IN CHAPTER 6

-------
Calculation Methodology for the Affected-source-wide Emission Rate

       This example is based on the materials used by a hypothetical facility as presented in the
spreadsheet below. Refer to this spreadsheet for the explanations that follow.

A. HAP Emissions from Each Material

HAP emissions were calculated by multiplying material usage (in liters) by the density (in kg/liter) to
obtain the mass of the material used. This value was then multiplied by the HAP content (as a decimal)
to obtain the mass of HAP in the material used. In the example spreadsheet for a hypothetical facility
(below) for Material #1, the emissions of 2-butoxyethanol is (51,200 L) x (1.11 kg/L) x (12/100) =
6,820 kg.  This procedure was repeated for each HAP component of each material.

B. Coating Solids Volume

Coating solids volume was calculated by multiplying the coating usage  (in liters) by the coating solids
content in percent by volume (as a decimal).  In the attached example for Material #1, the coating
solids volume is (51,200 L) x (32/100) = 16,384 L. This procedure was repeated for each coating
material. Note that Material #5 is a thinning solvent and contains no coating solids.

C. Total HAP Emissions

Total HAP emissions were calculated as the sum of the HAP emissions from each material component.
In the attached example, total HAP emissions are the sum of Column E, which is 17,346 kg.

D. Total Coating Solids Volume

Total coating solids volume was calculated as the sum of the coating solids volume from each coating
material. In the attached example, total coating solids volume is the sum of Column G, which is 25,513.

E. Normalized Facility Emissions

Normalized facility emissions were calculated as the total HAP emissions divided by the total coating
solids volume.  In the attached example, the normalized facility emissions are
(17,346 kg HAP)/(25,513 L coating solids) = 0.68 kg HAP/L coating solids.
                                            F-l

-------
                                 Example Spreadsheet for the Affected-source-wide Emission Rate Calculation Methodology
XYZ Company
Facility ID Material
MFX-01 1

2


3
4

5
Usage
ID (L)
(A)
51,200.00
51,200.00
19,235.00
19,235.00
19,235.00
2,341.00
9,658.00
9,658.00
7,642.00
Material
Density (Kg/L) HAP Component
(B) (C)
1.11
1.11
1.03
1.03
1.03
0.99
0.96
0.96
0.87
2-Butoxyethanol
Formaldehyde
2-Butoxyethanol
Formaldehyde
Ethylbenzene
2-Butoxyethanol
Xylene
Naphtha (see note c)
2-Butoxyethanol
HAP
Content HAP Emissions Coating Solids Coating Solids
(Mass %) (Kg) Content (Vol %) Volume (L)
(D) (E) (see note a) (F) (G) (see note b)
12.00
0.10
6.00
0.02
0.02
5.00
27.00
0.05
100.00
6819.84
56.83
1188.72
4.16
4.16
115.88
2503.35
4.64
6648.54
32.00

28.20


21.30
33.20

0.00
16384.00

5424.27


498.63
3206.46

0.00
                                                                                            Total HAP
                                                                                          Emissions (Kg)
                                                                                         (H) (see note d)
                                                                                            17346.13
  Total Coating
Solids Volume (L)
 (I) (see note e)
Normalized Facility
    Emissions
(Kg HAP/L Coating
      Solids)
  (J) (see note f)
                                                                                                                                 25513.36
                                                                                                                                                        0.68
(a) HAP Emissions (E) = (A)*(B)*((D)/100)
(b) Coating Solids Volume (G) = (A)*(F/100)
(c) Solvent blend for Naphtha was assigned 1% HAP by mass.  HAP content values were taken from information provided by the Chemical Manufacturer's
Association Solvent Council, and were used only when the solvent blend HAP content was reported to be zero.
(d) Total HAP Emissions (H) = Sum of Column  (E)
(e) Total Coating Solids Volume (I) = Sum of Column (G)
(f) Normalized Facility Emissions (J) = (H)/(I)
                                                                                F-2

-------
                       APPENDIX G

SUMMARY OF DATA FOR EPA SAMPLED COMPANIES OPERATING METAL
           FURNITURE MANUFACTURING FACILITIES

-------
Appendix G. Summary Data for EPA Sampled Companies Operating Metal Furniture Manufacturing Facilities
No. of Facilities
Comnanv Name
Arrowhead Holdings Corporation
Atlas Springs Manufacturing Corporation
B-Line Systems
Crown Metal Manufacturing Company
Davies Office Refurbishing, Inc.
Dehler Manufacturing Company, Inc.
Den-Tal-Ez, Inc.
Genlyte Group Incorporated
Hickory Springs Manufacturing Company
HON Industries
Kimball International
Leggett & Platt Incorporated
Lozier Corporation
L.A. Darling Company, Inc.
Metal Creations
Mid- West Chandelier Company
National Service Industries, Inc.
Nevin Laboratories, Inc.
Professional Refinishing Organization
Republic Storage Systems, Inc.
Siemens Medical System, Inc.
Standex International Corporation
Steelcase Incorporated
Virco Manufacturing Corporation

Sales f$106)
$165.50
$9.40
$223.50
$13.00
$2.50
$10.00
$23.10
$664.10
$23.10
$1,696.40
$1,107.00
$3,370.40
$281.10
$300.00
$37.00
$17.80
$2,031.30
NA
$2.20
$52.00
$11,144.00
$616.20
$2,742.50
$273.60

Emnlovment
1,990
140
1,400
125
200
120
327
3,490
295
9,824
9,556
27,000
2,400
3,000
NA
NA
16,700
NA
58
450
57,950
5,500
16,400
2,373

Total
3
1
1
1
1
1
1
1
2
7
2
10
3
2
1
1
7
1
1
1
1
1
11
1
62
Major
Source
3
1
1
0
1
0
1
1
2
4
2
10
3
2
1
1
1
0
1
1
1
1
11
1
49
Small Business
No
Yes
No
Yes
Yes
Yes
Yes
No
Yes
No
No
No
No
No
Yes
Yes
No
No
Yes
Yes
No
No
No
No
10
                                                       G-l

-------
                               TECHNICAL REPORT DATA
                           (Please read Instructions on reverse before completing)
 1. REPORT NO.
   EPA-453/R-01-010
                                                               3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
 National Emission Standards for Hazardous Air Pollutants
 (NESHAP) for Source Category:  Metal Furniture Surface
 Coating - Background Information for Proposed Standards
5. REPORT DATE
 October 2001
6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
8. PERFORMING ORGANIZATION
  REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
   U.S. Environmental Protection Agency
   Office of Air Quality Planning and Standards
   Coatings and Consumer Products Group
   Research Triangle Park, NC  27711
                                                               10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
   68D01055
 12. SPONSORING AGENCY NAME AND ADDRESS
   U.S. Environmental Protection Agency
   Office of Air Quality Planning and Standards
   Coatings and Consumer Products Group
   Research Triangle Park, NC  27711
13. TYPE OF REPORT AND PERIOD
COVERED
14. SPONSORING AGENCY CODE
   EPA/200/04
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
 A draft rule for the regulation of hazardous air pollutants (HAP) from metal furniture coating
 operations is being proposed under the authority of Section 112(d) of the Clean Air Act. This
 document contains comments the background information used to develop the draft rule.
 17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                              b. IDENTIFIERS/OPEN ENDED TERMS
                                                                               c. Field/Group
   Air Pollution
   Hazardous Air Pollutants
   Surface Coating
   Metal Furniture
 18. DISTRIBUTION STATEMENT
   Release Unlimited
                                              19. SECURITY CLASS (Report)
                                                Unclassified
               21. NO. OF PAGES
                     201
                                              20. SECURITY CLASS (Page)
                                                Unclassified
                                                                               22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION IS OBSOLETE

-------