EPA-600/R-9 5-063
April 1995
EVALUATION OF BARRIERS TO THE USE OF RADIATION-CURED
COATINGS IN CAN MANUFACTURING
By:
Beth W. McMinn and Steven R. Church
TRC Environmental Corporation
6340 Quadrangle Drive, Suite 200
Chapel Hill, North Carolina 27514
EPA Contract No. 68-D2-0181
Work Assignment Nos. 1/005 and 1/015
EPA Project Officer: Carlos M. Nunez
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
U.S. Environmental Protection Agency-
Office of Research and Development
Washington, D.C. 20460
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
-------
ABSTRACT
In support of the Source Reduction Review Project (SRRP), maximum achievable control
technology (MACT) standards development, and the Pollution Prevention Act, FPA's Air and
Energy Engineering Research Laboratory (AEERL) is investigating the current industrial use and
barriers to the extended use of radiation-cured coatings in SRRP and MACT categories. This report
presents the results of a study to investigate and identify the technical, educational, and economic
barriers lo the use and implementation of radiation-cured coatings in can manufacturing. Some of
the important barriers were the following: (1) an applied wet film thickness of greater than 120 mg
per can of ultraviolet (UV)-curable overvarnish needed on most trial runs; (2) lower than expected
energy savings; (3) inadequate cure of overvarnish; and (4) ink "pick off during the wet-on-wet
application of the overvarnish to the inks. This report provides suggested projects that could help
overcome technical, educational, and economic barriers identified. Some of the opportunities
discussed include the following: (1) setting up a trial with a can manufacturer that is interested in
using UV-curable inks and coatings; (2) conducting research on cationic inks and coatings, which
have been billed as the next generation of UV-curable inks and coatings; and (3) working with
Radtech. the trade association representing the radiation-curable coatings industry, to develop a UV-
curable coating that could be approved by the Food and Drug Administration (FDA) for direct
contact with food.
in
-------
TABLE OF CONTENTS
Chapter Page
Abstract j_ii
List of Figures • .vrti
List of Tables ix
Conversion Factors x
Executive Summary xi.ii
1 INTRODUCTION AND PROJECT BACKGROUND 1-1
1.1 PROJECT BACKGROUND 1-1
1.2 PROJECT OBJECTIVES 1-3
1.3 INDUSTRY SEGMENT DESCRIPTION 1-4
1.4 REPORT ORGANIZATION 1-5
1.5 REFERENCES 1-6
2 CONVENTIONAL PROCESS DESCRIPTION 2-1
2.1 GENERAL 2-1
2.2 DRAW AND IRON PROCESS FOR TWO-PIECE BEER AND BEVERAGE
CANS 2-1
2.2.1 Material Inputs 2-1
2.2.2 Equipment 2-1
2.2.3 Conventional Draw and Iron Process 2-2
2.2.4 Product Outputs 2-6
2.3 DRAW-THIN-REDRAW PROCESS FOR TWO-PIECE FOOD CANS 2-7
2.3.1 Material Inputs 2-7
2.3.2 Equipment 2-7
2.3.3 Conventional Draw-Thin-Redraw Process 2-9
2.3.4 Product Outputs 2-10
2.4 THREE-PIECE PROCESS FOR FOOD CANS 2-10
2.4.1 Materiallnputs 2-10
2.4.2 Equipment 2-11
2.4.3 Conventional Three-Piece Process 2-12
2.4.4 Product Outputs 2-13
2.5 EMISSIONS AND WASTES 2-13
2.5.1 Introduction 2-13
2.5.2 Air Emissions 2-14
2.5.3 Water Releases 2-15
2.5.4 Solid Waste 2-16
2.5.5 Hazardous Waste 2-17
2.5.6 Hazardous Chemicals ..2-17
2.6 REFERENCES 2-19
v
-------
TABLE OF CONTENTS (Continued)
Chapter Page
3 DESCRIPTION OF ULTRAVIOLET-CURING TECHNOLOGY 3-1
3.1 GENERAL 3-1
3.2 PROCESS DIFFERENTIALS FOR UV-CURING 3-1
3.2.1 Material Inputs and Equipment 3-1
3.2.2 UV-Curing Process 3-2
3.2.3 Emissions and Wastes 3-7
3.2.3.1 Air Emissions . 3-7
3.2.3.2 Water Releases 3-31
3.2.3.3 Solid Waste 3-12
3.2.3.4 Hazardous Waste 3-12
3.2.3.5 Hazardous Chemicals 3-12
3.2.4 Energy 3-13
3.3 COST DIFFERENTIALS 3-15
3.3.1 Introduction 3-15
3.3.2 Material Costs 3-15
3.3.3 Operating and Maintenance Costs 3-18
3.3.4 Energy Costs 3-20
3.3.5 Total Operating Costs 3-21
3.3.6 Capital Costs 3-22
3.4 REFERENCES 3-24
4 TECHNICAL BARRIERS TO THE EXTENDED USE OF UV-CURING
TECHNOLOGY 4-1
4.1 GENERAL 4-1
4.2 PRODUCT PERFORMANCE 4-1
4.3 EQUIPMENT 4-5
4.4 HEALTH AND SAFETY 4-5
4.5 REFERENCES 4-6
5 ECONOMIC BARRIERS TO THE EXTENDED USE OF UV-CURING
TECHNOLOGY 5-1
5.1 GENERAL 5-1
5.2 CAPITAL INVESTMENT 5-1
5.3 PRICING PRESSURE 5-1
5.4 MATERIAL AND OPERATING COSTS 5-2
5.5 REFERENCES 5-3
6 EDUCATIONAL BARRIERS TO THE EXTENDED USE OF UV-CURTNG
TECHNOLOGY 6-1
vi
-------
TABLE OF CONTENTS (Continued)
Chapter Page
6.1 GENERAL 6-1
6.2 OPERATOR TRAINING 6-1
6.3 MANAGEMENT AWARENESS 6-2
6.4 REGULATORY PRESSURE 6-3
6.5 REFERENCES 6-3
7 OPPORTUNITIES TO OVERCOME IDENTIFIED BARRIERS 7-1
7.1 GENERAI 7-1
7.2 PRODUCT PERFORMANCE IMPROVEMENTS 7-1
7.3 MIGRATION OF UV-CURABLE COATINGS 7-2
7.4 FOOD AND DRUG ADMINISTRATION APPROVAL 7-2
7.5 CATIONIC COATINGS 7-3
7.6 DEVELOPMENT OF UV-CURABLE WHITE BASECOAT 7-3
7.7 USE OF UV-CURABLE COATINGS IN THREE-PIECE CAN
MANUFACTURING 7-4
7.8 REFERENCES 7-4
APPENDIX A PRELIMINARY MARKET ANALYSIS A-l
APPENDIX B SITE VISIT REPORTS B-l
v i i
-------
LIST OF FIGURES
Number Page
3-1 UV-printing Process for Tvvo-Piece Beer Cans 3-4
3-2 Tvvo-Piece Beer Can UV-Curing Process 3-5
3-3 Parabolic Reflectors 3-6
vi i i
-------
LIST OF TABLES
Number Page
2-1 Material Inputs for Conventional Draw and Iron Manufacturing Process 2-2
2-2 Equipment Used for Conventional Draw and Iron Manufacturing Process 2-3
2-3 Material Inputs for Conventional DTR Manufacturing Process 2-8
2-4 Equipment Used for Conventional DTR Pood Can Line 2-8
2-5 Material Inputs for Conventional Three-Piece Manufacturing Process 2-10
2-6 Equipment Used for Conventional Three-Piece Food Can Line 2-11
2-7 VOC Content of Waterbased Inks and Coatings 2-14
2-8 Air Emissions from Two-Piece Can Manufacturing Facilities - 1992 2-15
2-9 Water Treatment Data for Two-Piece Can Manufacturing Facilities - 1992 2-16
2-10 Hazardous Chemicals Used in Conventional Two-Piece Manufacturing Process ... 2-18
3-1 Differences in Material Inputs and Equipment 3-1
3-2 Contents of UV-Curable Ink or Coating 3-2
3-3 Contents for UV-Curable Overvarnish 3-8
3-4 Coors Test Results lor VOC Content of Coatings 3-8
3-5 VOC Emission Estimates Based on Coors Stack Testing 3-9
3-6 Coors VOC and HAP Emission Estimates for TJV and Thermal Systems 3-9
3-7 Coors Emission Reduction Estimates for Golden, CO Plant 3-10
3-8 Coors Emission Reduction Estimates for Nation 3-10
3-9 Emissions Reported to TRI by Coors Container Complex
in Golden, CO - 1992 ' 3-11
3-10 Coors Estimates for Energy Savings From UV-Curing Oven
Versus Thermal Oven 3-13
3-11 Ball Corporation - Energy Comparison of UV-Curing Versus Thermal Oven 3-14
3-12 Ball Corporation Material Cost Comparison of UV-Curable to
Waterbased Materials - Findlay, OH, 1986 - 87 3-16
3-13 Ink Prices and Consumption 3-18
3-14 Operational Efficiencies of UV-Curing System 3-18
3-15 Annual Oven Maintenance and Repair Cost Estimate 3-20
3-16 Coors Estimated Energy Cost Savings - 1993 3-21
3-17 Ball Estimated Energy Cost for Ovens 3-21
3-18 Summary of Estimated Operating Costs 3-22
3-19 UV-Curing Oven and Accessories 3-23
3-20 Thermal (Pin) Oven and Accessories 3-23
4-1 Product Standards Used by Can Manufacturers for
Coatings Evaluation 4-2
A-l Five Largest Metal Can Companies by Sales (SIC 3411) A-2
A-2 Number of Companies by Sales (SIC 3411) A-3
i x
-------
LIST OF TABLES (Continued)
Number Page
A-3 Number of Facilities and Employees A-4
A-4 Shipments (SIC 3411) A-5
A-5 Employment and Compensation (SIC 3411) A-6
A-6 Key Industry Ratios A-6
A-7 Metal Can Market Shares by End Use Segment (SIC 3411) 1990 A-7
A-8 Buyers of Beer Cans A-7
A-9 Buyers of Soft Drink Cans A-8
A-10 Metal Container Exports Compared to Total
Industry' Exports A-8
A-ll Vendors of Coating Equipment (SIC 3411) A-ll
A-12 Industry Raw Materials in 1987 A-ll
A-13 TR1 Database Emissions and Waste Streams (SIC 3411) A-13
x
-------
CONVERSION FACTORS
To Convert From
To
Multiply
LENGTH
feet (ft)
meters (m)
0.3048
meters (m)
feet (ft)
3.281
inches (in)
centimeters (cm)
2.54
MASS OR WEIGHT
ounces (oz)
kilograms (kg)
0.02835
pounds (lb)
kilograms (kg)
0.454
pounds (ib)
tons
0.0005
tons
pounds (lb)
2,000
tons
kilograms (kg)
907.2
kilograms (kg)
pounds (lb)
2.205
kilograms (kg)
tons
0.001102
VOLUME
gallons (gal)
liters (1)
3.785
gallons (gal)
cubic inches (in3)
231
gallons (gal)
cubic feet (ft5)
0.133368
gallons (gal)
fluid ounces (oz)
128
gallons (gal)
cubic meters (m5)
0.00379
milliliters (ml)
fluid ounces (oz)
0.03381
liters (1)
gallons (gal)
0.2642
cubic inches (in3)
gallons (gal)
0.004329
cubic feet (ft1)
gallons (gal)
7.48
fluid ounces (oz)
gallons (gal)
0.007813
fluid ounces (oz)
milliliters (ml)
29.57
CONCENTRATION
pounds/gallon (lb/gal)
grams/liter (g/1)
119.8
grams/liter (g/1)
pounds/gallon (lb/gal)
0.008345
DENSITY
pounds/gallon (lb/gal)
grams/milliliter (g/ml)
0.1198
grams/milliliter (g/ml)
pounds/gallon (lb/gal)
8.345
PRESSURE
pounds/inch2 (psia)
mmHg or torr (mmHg)
51.71
pounds/inch2 (psia)
atmospheres (atm)
0.0680
millimeters of mercury
pounds/inch2 (psia)
0.1934
or torr (mmHg)
(continued)
xi
-------
CONVERSION FACTORS (Continued)
To Convert From To Multiply by
TEMPERATURE
Fahrenheit (°F)
Celsius C'C)
substract 32,
then multiply by 0.5556
Celsius (°C)
Fahrenheit (CF)
multiply by 1.8,
then add 32
ENERGY
Horsepower (HP)
Kilowatts (kW)
0.747
BTU
Joules (J)
1055
Calories (cal)
BTU
0.00397
Joules (J)
BTU
0.000948
-------
EXECUTIVE SUMMARY
Section 4(b) of the Pollution Prevention Act of 1990 requires the Environmental Protection
Agency (EPA) to "review regulations of the Agency prior and subsequent to their proposal to
determine their effect on source reduction." In support of the Pollution Prevention Act, EPA
established the Source Reduction Review Project (SRRP) to focus this review on pending
regulations (and anticipated regulated industries) under the Clean Air Act (CAA), the Clean Water
Act (CWA), or the Resource Conservation and Recovery Act (RCRA). One of the goals of SRRP
tasks is to ensure that source reduction and multi-media issues are considered during the
development of upcoming air, water, and hazardous waste standards.
One important set of regulations under the CAA, and a focus of SRRP, is the standards for
maximum achievable control technology (MACT) to reduce emissions of hazardous air pollutants
(HAPs). Promulgation of these regulations began in 1992 and will continue throughout the decade
and into the next century. The MACT standards offer EPA an excellent opportunity to use SRRP
to incorporate pollution prevention measures into the upcoming standards for specific source
categories. Pollution prevention efforts may offer economic and reduced health and ecological risk
benefits to many sectors of society that are not available through traditional pollution control
methods.
In support of the SRRP Program, MACT standards development, and the Pollution
Prevention Act, EPA's Air and Energy Engineering Research Laboratory (AEERL) is investigating
pollution prevention opportunities for product and material substitutions that help industry to reduce
waste. The specific objective of this project was to investigate the current industrial use and barriers
to the extended use of waterbased and radiation-cured coatings in SRRP and MACT categories.
Metal Cans (SIC 3411), an industry facing upcoming MACT standards, was selected as an industrial
segment for study. When the MACT standards are developed, EPAfwill have a better understanding
of which coating technologies are feasible pollution prevention alternatives for the industry.
This report presents the results of a study to investigate and identify the technical,
educational, and economic barriers to the use and implementation of radiation-cured coatings within
two-piece metal can manufacturing. This project involved preparing category analyses, identifying
and classifying the use and implementation barriers, evaluating and assessing the environmental
xi.i
-------
impacts, and identifying pollution prevention and source reduction research opportunities within the
two-piece metal can industry. Information was collected for this project from a review of current
technical literature, cooperation with industry leaders and the leading trade organization, and visits
to three can manufacturing facilities. (One of the visits was to a three-piece can manufacturing
facility; however, the report focuses on two-piece manufacturing.)
This project was initially intended to study both ultraviolet (UV) radiation-cured and
waterbased screen printing inks as possible alternatives to solvent-based inks with high volatile
organic compound (VOC) emissions. During the course of this project, it became evident that the
focus should be on IJV-curable inks and coatings. The current industry standard is to use waterbased
inks and coatings that contain 6 to 15 percent volatile organic compounds (VOCs). UV-curable inks
and coatings contain less than one percent VOCs and would significantly reduce emissions from
two-piece can manufacturing operations.
Within the can manufacturing industry, there is debate over the economic and process
benefits that UV-curable inks and coatings offer. The Coors can manufacturing plant in Golden,
Colorado has been successfully using UV-curable inks and overvarnish to coat the exterior of its
cans since 1976. The UV technology has provided Coors with a number of benefits including: (1)
reduced energy costs; (2) less downtime for maintenance and repairs; (3) less floor space occupied
by the drying/curing oven; and (4) employee satisfaction with the reduced operating temperatures
and simple procedures of the UV-curing oven. Coors claims that the benefits of a UV system,
particularly the reduced energy costs, compensate for the higher material costs of UV-curable inks
and coatings.
Ball Corporation had a different experience with UV-curable inks and coatings when it
established a UV trial line at its Findlay, Ohio plant in 1986-87. The company encountered several
technological and economic barriers that prevented Ball from expanding its use of UV technology
beyond the trial stage. Some of the important barriers were the following: (1) an applied wet film
thickness of greater than 120 mg per can of UV-curable overvarnish needed on most trial runs; (2)
lower than expected energy savings; (3) inadequate cure of overvarnish; and (4) ink "pick off' during
the wet-on-wet application of the overvarnish to the inks.
xiv
-------
This report divides the barriers to implementing UV-curable inks and coatings into three
categories: technical, economic, and educational barriers. Separate chapters examine each of the
three barrier categories.
This document suggests projects that could help overcome technical, educational, and
economic barriers identified. Some of the opportunities discussed include the following:
• Setting up a trial with a can manufacturer that is interested in using UV-curable inks and
coatings. This joint venture between EPA and private industry would provide an opportunity
for suppliers of UV-curable inks and coatings, equipment vendors, and can manufacturers
to work together to overcome the technical barriers identified in the Ball trial runs.
• Conducting research on cationic inks and coatings, which have been billed as the next
generation of UV-curable inks and coatings. Cationic coatings offer promise because of their
improved abrasion resistance and their ability to dark cure (continue the curing process in
areas not exposed to UV light).
• Working with Radtech, the trade association representing the radiation-curable coatings
industry, to develop a UV-curable coating that could be approved by the Food and Drug
Administration (FDA) for direct contact with food. Because some of the acrylic compounds
in UV-curable coatings contain hazardous chemicals, they cannot be approved for direct
contact with food or beverages.
xv
-------
CHAPTER 1
INTRODUCTION AND PROJECT BACKGROUND
1.1 PROJECT BACKGROUND
Section 4(b) of the Pollution Prevention Act of 1990 requires the Environmental Protection
Agency (EPA) to "review regulations of the Agency prior and subsequent to their proposal to
determine their effect on source reduction."1 In support of the Pollution Prevention Act. EPA
established the Source Reduction Review Project (SRRP) to focus this review on pending
regulations (and anticipated regulated industries) under the Clean Air Act (CAA), the Clean Water
Act (CWA), or the Resource Conservation and Recovery Act (RCRA). One of the goals of the
SRRP is to ensure that source reduction and multi-media issues are considered during the
development of upcoming air, water, and hazardous waste standards. The following seventeen
industrial categories are affected by the SRRP:2
• Pesticide Formulating
• Pulp and Paper Production
• Pharmaceuticals Production
• Paints, Coatings, and Adhesives Manufacturing
• Printing and Publishing
• Integrated Iron and Steel Manufacturing
• Plywood/Particle Board Manufacturing
• Paint Stripper Users
• Rubber Chemicals Manufacturing
• Paper and Other Webs Coating
• Acrylic Fibers/Modacrylic Fibers
• Degreasing Operations
• Polystyrene Production
• Styrene Butadiene Latex and Rubber Production
• Reinforced Plastic Composites Production
• Machinery Manufacturing and Rebuilding
• Wood Furniture Manufacturing
One important set of regulations under the CAA, a regulation of SRRP focus, is the standards
for maximum achievable control technology (MACT) to reduce emissions of hazardous air
1-1
-------
pollutants (HAPs). Promulgation of these regulations began in 1992 and will continue throughout
the decade and into the next century. The MACT standards offer EPA an excellent opportunity to
use the SRRP to incorporate pollution prevention measures into the upcoming standards for specific
source categories. The Pollution Prevention Act of 1990 defines pollution prevention as "any
practice which reduces the amount of any hazardous substance, pollutant, or contaminant entering
the waste stream or otherwise released to the environment (including fugitive emissions) prior to
recycling, treatment, or disposal; and reduces the hazards to public health and the environment
associated with the release of such substances, pollutants, or contaminants."1 Pollution prevention
efforts may offer economic and reduced health and ecological risk benefits to many sectors of
society that are not available through traditional pollution control methods.
In support of the SRRP Program, MACT standards development, and the Pollution
Prevention Act, EPA's Air and Energy Engineering Research Laboratory (AEERL) is investigating
pollution prevention opportunities for product and material substitutions that help industry to reduce
waste. The specific objective of this project was to investigate the current industrial use and barriers
to the extended use of waterbased and radiation-curable coatings in SRRP and MACT categories.
Both radiation-curable and waterbased coatings have been demonstrated to reduce pollution in
several specific end-use categories. The three Standard Industrial Classification (SIC) categories
selected for initial investigation were Adhesive-Coated and Laminated Paper (SIC 2671 and 2672),
Metal Cans (SIC 3411), and Commercial Printing - Not Elsewhere Classified (SIC 2759). All three
of these industries face upcoming MACT standards. By initiating this report, EPA has begun a
dialogue on pollution prevention with the industries. When the MACT standards are developed,
EPA will have a better understanding of which coating technologies are feasible pollution prevention
alternatives for the three industries.
During the first task of this project, industries in 52 SIC categories were identified as having
the potential to use radiation-curable and waterbased coatings as pollution prevention alternatives.
During this phase, contacts were made with representatives from coating suppliers and trade
associations and limited literature searches were completed. From this list of 52 potential SICs, 10
were selected for further study. Preliminary market analyses were prepared for each of these ten
categories. Following the completion of the ten analyses, three categories were selected for
investigation. Conversations with resin manufacturers, coating suppliers and end users indicated that
1-2
-------
waterbased coatings were already being used extensively in the three industries, particularly in the
manufacture of metal cans. Conversely, radiation-curable coatings had made progress in each of die
three industries but were not widely used in any of them. The limited penetration of radiation-
curable coatings offered the best opportunity for research. Therefore, the focus of the project became
the use of radiation-curable coatings. The focus of this report is on barriers to the use of radiation-
curable coatings in metal can manufacturing.
1.2 PROJECT OBJECTIVES
This report presents the results of a study to investigate and identify the technical,
educational, and economic barriers to the use and implementation of radiation-curable coatings
within the metal can manufacturing industry. This project involved preparing category analyses,
identifying and classifying the use and implementation barriers, evaluating and assessing the
environmental impacts, and identifying pollution prevention and source reduction research
opportunities within the metal can manufacturing industry, in order to successfully accomplish these
objectives, information was collected from several sources including literature searches, plant visits,
pollution prevention experts, and industry and trade association personnel.
Literature searches of the EPA on-line databases, local university library databases, and
Dialog5 were conducted. The Pollution Prevention Information Clearinghouse (PPIC) and the
Pollution Prevention Information Exchange System (PIES) were also accessed. The E-Mail
capabilities of PIES were also used to communicate with other PIES users with knowledge of the
metal can manufacturing industry.
In addition to conducting literature searches, contacts were made with industry and pollution
prevention experts with the National Paint and Coatings Association (NPCA), Radtech, the Can
Manufacturers Institute (CMI), and equipment and coating manufacturing firms.
The final source of project and industry information was compiled during a total of three site
visits (see Appendix B). Together, these information gathering efforts provided the background
needed to identify the barriers and source reduction opportunities within the metal can manufacturing
industry.
1-3
-------
1.3 INDUSTRY SEGMENT DESCRIPTION
The focus of this report is the identification of barriers to the extended use of UV-curable
coatings within the two- and three-piece can manufacturing industry, represented by SIC 3411.
Although this report concentrates on the manufacture of two-piece cans, a process description is also
included for three-piece can manufacturing. The purpose for this is that many of the difficulties
encountered by two-piece can manufacturers will be similar to the problems of three-piece can
manufacturers. Likewise, research opportunities and pollution prevention techniques may be shared
among the two industry segments. Two-piece cans are comprised of two components, a can body
and an end that seals the contents inside the can. These cans are used primarily for packaging beer
and beverages although their use for food packaging is growing. Ninety percent of the 105.8 billion
two-piece cans shipped in 1992 were for the beverage can market.3 Three-piece cans have a body
and two ends that are sealed on the can. The size of this market is significantly smaller than the size
of the two-piece market. In 1992, the industry shipped 24.8 billion three-piece cans, approximately
24 percent of the number of two-piece cans. Eighty-four percent of the 1992 three-piece can
shipments served the food market, which includes cans for vegetables, fruits, pet foods, baby foods,
and several other household products.-
Another way to segment the industry is by identifying the customer served. Merchant can
manufacturers serve more than one customer, while captive plants are owned by a beer, beverage,
or food company and produce cans for internal consumption only. This report identifies the
differences in technical, economic, and educational barriers between merchant and captive can
plants.
The metal can industry was selected for investigation for three reasons. First, the industry
emits a significant amount of air pollutants each year. According to the Toxic Release Inventory
(TR1), the industry emitted over 19,000 tons of air pollutants in 1990.4 Significant amounts of these
emissions were volatile organic compounds (VOCs) and hazardous air pollutants (HAPs). Reducing
the industry's emissions would help some areas of the country reach the attainment levels for the
national ambient air quality standards (NAAQS).
The second reason for the selection of metal cans is the established, yet limited, presence of
radiation-curable coatings in the metal can market. The largest two-piece can manufacturing facility
1-4
-------
in the world, the Coors Container Complex (Coors) in Golden, Colorado, first began using
ultraviolet (UV)-curable coatings in 1975, and it has continued to use them and improve the UV-
curable coating process. Since 1975, other two-piece can manufacturers have considered using UV-
curable coatings, but none have permanently switched to a UV-curing system. There are several
barriers that have prevented two-piece manufacturers (other than Coors) from making the switch;
this report identifies those barriers. In the three-piece market, UV-curable coating systems have been
implemented by several manufacturers, but their use has been limited. This report does not cover
the barriers to UV-curable coatings in 3-piece can manufacturing.
The final reason for the selection of metal cans is the timing of the industry's MACT
standard. EPA has scheduled its promulgation of this industry's standard for the year 2,000.s This
schedule gives the agency over six years to develop a standard that will properly incorporate the
SRRP approach to regulator}' development. Using this report as a source of background information,
EPA will have the time to fully consider several pollution prevention alternatives for the metal can
industry. Based on its assessment of the alternatives, the agency will be able to develop a MACT
standard that is efficient, effective, and flexible for the metal can industry.
1.4 REPORT ORGANIZATION
This report is divided into seven chapters and one appendix. Chapter 2 describes the
conventional manufacturing processes and includes a discussion of the material inputs,
manufacturing equipment, physical processes, product outputs, and emissions and wastes. Chapter
3 includes a basic discussion of the alternative technology under investigation. This chapter
evaluates the process, cost, and emissions and wastes differentials between the conventional and
alternative processes. Chapter 4 identifies the technical barriers to the extended use of radiation-
curable coatings. It includes a description of the difficulties and available information on solutions
currently under consideration. Chapter 5 discusses economic barriers, and Chapter 6 identifies
educational barriers. Chapter 7 presents additional source reduction and pollution prevention
research opportunities. Appendix A contains a copy of the preliminary market analysis that was
developed during the early stages of this project. Appendix B contains copies of three site visit
reports that were used in the preparation of this report and are referenced throughout this report.
1-5
-------
1.5 REFERENCES
1. Pollution Prevention Act of 1990,42 U.S.C. § 13101, et seq.
2. U.S. Environmental Protection Agency. Source Reduction Review Project. Office of the
Administrator, Pollution Prevention Policy Staff, Washington, DC. EPA- 100/R-92-002.
March 1992.
3. Can Shipments Report 1992, Can Manufacturers Institute. Washington. D.C.
4. Toxic Chemical Release Inventor)' Database. U.S. Department of Health and Human
Services, National Institutes of Health. National Library of Medicine. Bethesda. MD.
Toxicology Information Program Online Services TOXNET® Files. 1990.
5. "EPA Publishes Draft Schedule for Promulgation of MACT Standards," The Air Pollution
Consultant, 3(1). pp.2.9-2.13. January/February 1993.
1-6
-------
CHAPTER 2
CONVENTIONAL PROCESS DESCRIPTION
2.1 GENERAL
This chapter describes the draw and iron (D and 1) process used to manufacture two-piece
beer and beverage cans, the draw-thin-redraw (DTR) process for two-piece food cans, and the
three-piece process for food cans. Although can plants employ variations of these processes, the
following sections describe standard applications in high-volume plants. Each process is
described separately.
2.2 DRAW AND IRON PROCESS FOR TWO-PIECE BEER AND BEVERAGE CANS
2.2.1 Material Inputs
Much of the information on the D and I process is based on a visit to the Ball can
manufacturing facility in Williamsburg, Virginia. Table 2-1 identifies the material inputs used
to manufacture two-piece beer and beverage cans and describes at what points the inputs are used
on a typical two-piece can line. Although input materials-do not vary between beer and beverage
cans, the quantities applied fluctuate, particularly for the internal coating.
2.2.2 Equipment
Table 2-2 lists the equipment used to manufacture two-piece beer and beverage cans on
a conventional line. Section 2.2.3 provides more detail on the function of each piece of
equipment.
2-1
-------
TABLE 2-1. MATERIAL INPUTS FOR CONVENTIONAL DRAW AND IRON
MANUFACTURING PROCESS
Material
Application D and I Line
Aluminum
Lubricant
Coolant
Sulfuric acid solution
Caustic solution
Deionized water
Base coat
Waterbome inks
Overvarnish
Bottom coat
Internal coat
Waxing lubricant
Cardboard or plastic
pallets
Arrives at plant in coils, processed into two-piece aluminum cans
Applied to aluminum coil in lubricator tray, prevents aluminum
from oxidizing during manufacturing process
Used in bodymaker to reduce friction
Used in washer to clean cans
Used in washer to neutralize sulfuric acid solution
Used in washer to rinse cans
Applied by base coat roller, used for background on certain labels
Applied by printer, used to print labels on cans
Applied by roll coater in printer, provides protection to label
Applied by bottom coater to bottom rim of cans, not all cans
receive bottom coating
Applied by airless spray guns, used to coat interior of cans
Applied by waxer, lubricates cans for necking
Used to stack cans for shipment or storage
2.2.3 Conventional Draw and Iron Process
Most two-piece can manufacturing plants have more than one line. The larger facilities
have three to four lines, one of which is normally dedicated to the manufacture of 16 ounce (474
ml) cans and the remaining lines to 12 ounce (355 ml) cans. The D and I process can be divided
into two phases: can bodymaking and decoration. The bodymaking section of a line includes the
first six items listed in Table 2-2: an uncoiler, lubricator, cupper, bodymaker, trimmer, and
washer/dryer. The number of cuppers, bodymakers, and trimmers depends on the design of a line
and the capability of the equipment For example, some cuppers are designed to punch six cups
per stroke while others punch 12 or 13 cups per stroke.
2-2
-------
TABLE 2-2. EQUIPMENT USED FOR CONVENTIONAL DRAW AND IRON
MANUFACTURING PROCESS
Equipment
Function
Unc oiler
Lubricator
Cupper
Body maker
Trimmer
Washer/Dryer
Basecoater
Basecoater oven
Printer
Bottom coater
Deco oven (Pin oven)
Internal coater
Internal coater oven
Waxer
Necker
Spinnecker
F1 anger
Light tester
Palletizer
Feeds coil into can line
Applies lubricant to aluminum coil
Punches cups from coil
Draws cups into cans, forms indented bottom on cans
Trims cans to desired height
Washes and dries cans before decoration
Applies basecoat to exterior surface of cans, only necessary for
certain types of cans
Cures basecoat at elevated temperatures
Applies inks to cans and overvarnish for protection
Applies coating to bottom rim of cans
Cures inks, overvarnish and bottom coat of cans at elevated
temperatures
Sprays coating on interior of can
Cures internal coating at elevated temperatures
Applies wax lubricant to neck of cans
Squeezes open end of can to desired diameter
Removes rib and smoothes neck of cans
Rolls back top edge of cans to form lip
Tests cans for leaks before packaging
Gathers and stacks cans onto pallets for shipment or storage
The bodymaking process begins after a three to five-ton coil of aluminum has been placed
on the arm of the uncoiler. The aluminum coil has a thickness of 0.0110 - 0.0114 inches
(0.0279 - 0.0290 cm) for 12 oz cans and 0.0118 - 0.0120 inches (0.0300 - 0.0305 cm) for 16 oz
cans. Most of the newer uncoilers have two arms, which allow a non-active arm to be fed into
the production line when the active arm is finished. This arrangement minimizes production
down time.
2-3
-------
The uncoiler passes the aluminum into a lubricator, consisting of a roller and tray, which
applies a synthetic, water-soluble lubricant. The roller picks up the lubricant from the tray and
applies it to the coil as it passes over the roller. The lubricant prevents the aluminum from
oxidizing during the can making process.
After passing through the lubricator, the coil moves into a cupper which punches circular
blanks of aluminum and draws them into cups approximately 3.56 inches (9.05 cm) in diameter
and 1.5 inches (3.81 cm) in height (for 12 oz. cans). Many cuppers operate at 250 strokes per
«
minute. The scrap aluminum from the coil is removed from the line after the cups have been
punched. A vacuum belt carries the cups to one of the line's bodymakers. The bodymakers use
a punch mounted on a ram to push the cups through a series of four tooling dies. This D and
I process stretches and forms the cups into cans. The wall thickness of a finished can is
approximately 0.0035 inches (0.089 mm). While the cups are being punched through the dies,
the concave bottom is formed which improves their ability to withstand the pressure generated
during later filling processes. Once the cans emerge from the bodymaker, they move to a
trimmer to be cut to their desired height. The bodymaker leaves the cans slightly thicker at their
tops because they will later be necked and flanged.
The decoration section of a D and I line includes the remaining equipment listed in
Table 2-2: a basecoater, basecoater oven, printer, bottom coater, deco oven, internal coaters,
internal coater oven, waxer, necker, spinnecker, flanger, light tester, and palletizer. Many plants
do not have a basecoater or basecoater oven because many cans do not require basecoats.
Before the decoration process can begin, the cans must be washed and rinsed to remove
lubricant, aluminum particles, and dirt A vacuum belt moves the cans from the trimmer to the
washer, which consists of four stages. The cans are rinsed with tap water, cleaned with a sulfuric
acid solution, cleaned with a caustic solution, and rinsed with deionized water. In addition to
removing foreign particles, the cleaning process etches the cans in preparation for decoration.
After being washed, the cans pass through a dryer.
Depending on the can label requirements, the decoration process may begin with the
application of a basecoat to the exterior of the cans. In the basecoater, the cans pass over a roller
that applies a white ink directly to them. The white coating serves as the base upon which other
inks will be applied. From the basecoater, the cans move along a vacuum belt to the basecoater
oven where the basecoat is cured. The basecoater oven operates at temperatures near 400°F
2-4
-------
(204°C). Inside the oven, the cans move up and down along a chain conveyor {i.e., a pin chain)
in a serpentine motion. Cans typically spend from 35 to 45 seconds inside the oven.
Once the cans leave the basecoater oven, they move down a vacuum belt to the printer.
For cans not requiring a basecoat, the printer is the first step in the decoration process. When
the cans enter the printer, they are loaded onto a mandrel wheel. The mandrel moves the cans
to the ink wheel, which applies the desired ink pattern. The ink wheel consists of a rubberized
blanket that picks up the complete color image of the label as it rotates past at least four different
ink stations. The ink stations apply the inks to the wheel through printing plates that match each
color image of the label. When the ink wheel comes into contact with the cans, the mandrel
spins the cans so that the complete image is applied. The cans are then moved to a roller where
a film of overvamish is applied to their entire exterior surface.
The application of the overvamish onto the inks is referred to as a "wet-on-wet"
application. Nearly all major two-piece manufacturers use waterbased inks and overvamish for
the decorating process. Most can manufacturers apply approximately 75 mg of overvamish to
each can.1
Larger can plants with three to four lines often have a representative of the ink supplier
on site to mix inks. The representative maintains an "ink recipe" for each of the labels that a
company runs. The ink recipe identifies the colors of the inks and the quantities to be applied
to each can. Most of the labels for beer and beverage cans require a minimum of four different
inks. Adding more than four inks to the printing process does not reduce line speeds unless the
inks are applied on top of each other to achieve various shades of color. A specialty order with
shading requirements takes more time to set up and run than a standard order using four inks.
From the printer, the cans travel along a vacuum belt to the bottom coater, which applies
a waterbased lacquer to the bottom rim of the cans. Once cured, this coating protects the bottom
of the cans and improves their mobility along the line. When the cans reach the deco oven, each
one moves onto a pin chain which lifts them up through the oven. The deco oven is similar to
the basecoater oven. It operates within the same temperature range, near 400°F (204°C), with the
same residence time of 45 seconds. The serpentine movement of the chain allows the cans to
spend more time inside the oven than they would by passing straight through it.
The cured cans exit the deco oven and move to the internal coater, which normally
consists of five to nine airless spray guns ananged in a row. The cans are turned on their sides
-------
and pass in front of one of the guns, which applies a waterbased enamel coating to their interior.
The same coating is applied to both beer and beverage cans; however, the amount of coating
varies. Beverage cans receive approximately 50 percent more coating than beer cans because of
the acidic nature of their contents. From the internal coater, the cans travel to the internal coater
(IC) oven. The IC oven is different from the basecoater and deco ovens in that the cans travel
upright through it along a conveyor belt rather than along a pin chain. The cans spend
approximately 45 seconds inside the IC oven, which operates at temperatures ranging from 320°F
to 400°F (160°C to 204°C).
Once the internal coat has been cured, the cans travel on a vacuum belt to the waxer. The
waxer prepares the cans for necking by applying a thin layer of lubricant to the outside of the
open edge of each can. The necking operation involves three steps in which the cans pass
through a necker, spinnecker and flanger. The necker squeezes the open end of each can down
to the specified diameter by creating a ridge. The spinnecker then removes the ridge and
smoothes the narrowed area near the open end of each can. Finally, the flanger rolls back the
top edge of each can to form a lip, which is later used to attach an end to the can after the filling
process has been completed.
After the cans leave the spinnecker and flanger, they pass through a light tester which
checks for leaks. If the cans are leak proof, they go to the palletizer where they are placed onto
wooden or plastic pallets. Each pallet holds from 350 to 400 cans. Once a pallet is full, it is
lowered several feet and a new pallet is stacked on top of it. The pallets are either stored in a
warehouse or shipped to the customer.
2.2.4 Product Outputs
The D and I process produces 12 and 16 ounce beer and beverage cans. The line speeds
typically range from 1,400 to 2,000 cans per minute (cpm) for beer and beverage cans with a
diameter of 2 6/16 inches (6.03 cm).2
The trend in the industry is towards "lightweighting" the aluminum can, eliminating
excess metal from the can to cut raw material costs. One can manufacturer claims that
eliminating 0.0001 inch (0.000254 cm) from the walls of its cans saves the company $1 million
a year.3 One way for two-piece manufacturers to lightweight beer and beverage cans is to reduce
2-6
-------
the diameter of their ends. Most beer and beverage cans currently have a diameter of 2 6/16
inches (6.03 cm). Professionals in the industry refer to this size as 206. The beverage industry
is moving towards a diameter of 2 4/16 inches (5.40 cm), or 204 cans. Beverage companies
favor the narrower diameters because some of the raw material savings will be passed onto them.
The beer industry, however, has not yet committed to a 204 can and will remain with the 206
diameter.2
2.3 DRAW-THIN-REDRAW PROCESS FOR TWO-PIECE FOOD CANS
2.3.1 Material Inputs
Much of the information on the DTR process is based on a visit to the Campbell Soup
plant in Maxton, North Carolina. Table 2-3 lists the major raw materials used to manufacture
two-piece food cans and describes at what points they are used on a typical DTR line. The major
difference in materials between a D and I beer/beverage can line and a DTR food can line is the
coatings. The DTR line often uses pre-coated tin-free steel, which does not require the can
manufacturer to apply any coating to it. Both sides of the steel coil have been coated by the steel
manufacturer prior to shipment to the can manufacturer; the can manufacturer then applies a
paper or film label to the exterior of the can. In the D and I process, however, the can
manufacturer applies the interior and exterior coatings; the can manufacturer prints the label
directly on the exterior surface of the can.
2.3.2 Equipment
Much of the equipment used on DTR lines is the same as used on D and I lines. The
primary equipment difference is in the can or bodymaking step. Table 2-4 lists the equipment
used to manufacture two-piece food cans on a conventional DTR line. Section 2.3.3 provides
more detail on the function of each piece of equipment.
2-7
-------
TABLE 2-3. MATERIAL INPUTS FOR CONVENTIONAL DTR
MANUFACTURING PROCESS
Material
Application On DTR Line
Pre-coated tin-free steel
Arrives at plant in coils, processed into two-piece food cans
Wax lubricant
Applied by lubricator to cups prior to cupping
Sulfuric acid solution
Used in washer to clean cans
Caustic solution
Used in washer to clean cans
Deionized water
Used in washer to rinse cans
Videojet inks
Used to mark exterior of cans for inventory purposes
Light tester
Tests cans for leaks before packaging
Cardboard or plastic pallets
Used to stack cans for shipment or storage
TABLE 2-4. EQUIPMENT USED FOR CONVENTIONAL DTR FOOD CAN LINE
Equipment
Function
Uncoiler
Feeds coil into can line
Coil lubricator
Applies lubricant to tin-free steel coil
Cupper
Punches cup from coil
Cup lubricator
Applies lubricant to cups
Can maker
Draws cups into cans
Trimmer
Trims cans to desired height
Washer/dryer
Washes and dries cans
Beader
Presses ribs into can bodies
Videojet ink applicator
Applies ink marking to cans for inventory purposes
Light tester
Tests cans for leaks before packaging
Palletizer
Gathers and stacks cans for shipment or storage
2-8
-------
2.3.3 Conventional Draw-Thin-Redraw Process
The DTR process for manufacturing two-piece food cans is similar to the D and I process.
A ten to twelve ton coil of pre-coated tin-free steel is placed on an uncoiler and unwound into
the coil lubricator. In the lubricator, the coil passes through a tray where a roller applies a thin
layer of wax to the steel. The wax reduces friction during the next process step when the coil
moves into the cupper. The cupper punches and flanges from 6 to 13 cups per stroke at
approximately 150 strokes per minute.
A second lubrication step is necessary for the cups before they can be drawn into cans.
The cups move into a chamber where a lubricator uses electrostatic attraction to apply a thin,
uniform coating of wax to all surfaces of the cup.4 The lubricator creates a wax mist inside the
chamber, and a corona grid gives a positive charge to the particles. As the cups pass through
the chamber, they are grounded (i.e., negatively charged) to attract the wax particles to their
interior and exterior surfaces. Once lubricated, the cups enter the sanitary can maker with their
open ends down. The can maker draws them to an intermediate size, normally 2.94 inches (7.47
cm) high and 3.19 inches (8.10 cm) wide, and enlarges the flange. To obtain the desired size,
the cans pass through another lubricator and can maker. The typical size of cans exiting the
second can maker is 2.56 inches (6.50 cm) wide and 3.87 inches (9.84 cm) high.4
From the can maker, the cans travel through a trimmer which cuts excess steel from their
flanges. The cans pass through the trimmer with their open ends down to ensure that steel
shavings do not contaminate the interiors.4 From the trimmer, they enter the washer/dryer, which
washes and drys the cans and prepares the surface for decoration. The next step, the beader,
forms a series of ribs in the sides of the cans. The ribs strengthen the cans, allowing them to
withstand the pressure generated during the sterilizing process. The cans then move through a
light tester which detects leaks. All leak-proof cans pass onto the palletizer where they are
stacked on pallets. After the cans have been filled and sealed in a food plant, a paper label is
attached to them.
2-9
-------
2.3.4 Product Outputs5
The DTR process typically produces ten ounce food cans. The process has not been
refined to produce larger volume food cans on a high production line. The speed of a typical
DTR line is 750 cpm.
2.4 THREE-PIECE PROCESS FOR FOOD CANS
2.4.1 Material Inputs
Much of the information on three-piece can manufacturing is based on a visit to the
Campbell Soup facility in Maxton, North Carolina. Table 2-5 lists the major raw materials used
to manufacture three-piece food cans and describes at what points they are used on a typical
three-piece line.
TABLE 2-5. MATERIAL INPUTS FOR CONVENTIONAL THREE-PIECE
MANUFACTURING PROCESS
Material
Application On Three-piece Line
Uncoated tin plate steel
Arrives at plant in coils, processed into three-piece food
cans
Applied by roll coater to steel sheets
Applied by airless spray gun to welded side seam of can
bodies
Applied by roller to pre-coated tin plate steel sheets for can
ends
Applied by compound liner to can ends before they are
attached to can bodies
Used to mark exterior of cans for inventory purposes
Tests cans for leaks before packaging
Used to stack cans for shipment or storage
Waterbased interior coating
Waterbased side seam
coating
Paraffin coating
End sealing compound
Videojet inks
Light tester
Cardboard or plastic pallets
2-10
-------
2.4.2 Equipment
Table 2-6 identifies the equipment used to manufacture three-piece food cans on a
conventional line. The process description, in Section 2.4.3, provides more detail on the function
of each piece of equipment.
TABLE 2-6. EQUIPMENT USED FOR CONVENTIONAL THREE-PIECE
FOOD CAN LINE
Equipment
Function
Uncoiler
Cutter
Sheet feeder
Roll coater
Wicket oven
Sheet stacker
Slitter
Bodymaker
Wire welder
Seam sprayer
Side seam oven
Beader/flanger
Scroll strip shearer
End press
Compound liner
End seamer
Light tester
Palletizer
Unwinds uncoated tin plate steel
Cuts coil into 4 x 4 ft (1.2 x 1.2 m) sheets
Feeds sheets into coating process
Applies waterbased interior coating to top side of sheets
Cures waterbased coating on sheets
Collects steel sheets and stacks them for transport to fabrication
section of line
Cuts steel sheets into 4 x 8 in (10.2 x 20.3 cm) blanks
Wraps blanks into cylinder shape
Welds a side seam on sheets with copper electrode
Applies waterbased coating to side seam
Cures side seam coaling
Punches series of ribs into can bodies and flanges ends
Cuts sheets into indented rectangular strips
Punches circular ends from steel strips
Applies sealing compound to circular edge of ends
Joins ends with can cylinders
Tests cans for leaks before packaging
Gathers and stacks cans for shipment or storage
2-11
-------
2.4.3 Conventional Three-Piece Process
The three-piece manufacturing processes can be divided into two operations: sheet coating
and can fabricating. The sheet coating operation consists of an uncoiler, a cutter, a sheet feeder,
a roll coater, a wicket oven, and a sheet stacker. The can fabricating operation produces cylinder
bodies and can ends. It uses a slitter, bodymaker, wire welder, seam sprayer, thermal oven,
beader/flanger, scroll strip shearer, end press, compound liner, end seamer, light tester, and
palletizer.
The sheet coating process begins with a multi-ton coil of uncoated tin plate steel. As the
coil is unwound, it is cut into sheets [often 4 x 4 ft (1.2 x 1.2m)] which are then stacked on top
of each other and placed in a sheet plate feeder. The feeder feeds the sheets to a belt which
transports them to a direct-roll coater which applies a waterbased enamel coating to the top side.
This coating will serve as the interior coating of the cans. The roll coater applies the coating by
rolling in a clockwise direction and transferring the coating from the tray below it After being
roll-coated, the sheets slide into the oven where wickets (i.e., moving grates) receive and
transport them vertically through the six-zone oven. The oven contains approximately 2,800
wickets and operates at approximately 400°F (204°C). The sheets spend approximately 15
minutes inside the oven. Upon their exit, the cured sheets are stacked and transported by loft
truck to the can fabricating operations.
The fabrication process begins with a slitter which cuts 4x8 inch (10.16 x 20.32 cm)
body blanks from the sheets. The blanks then move along a belt to the bodymaker which wraps
them around a rod to form a cylinder. With a copper electrode, a wire then welds a side seam
on the top of the cylinder where the two ends meet. An airless spray gun applies a waterbased
enamel coating to the seam of each cylinder. The cylinders exit the bodymaker in an end-to-end,
horizontal position and travel to an oven which cures the side seam spray at 400°F (204°C).
From the side seam oven, the cylinders pass through the beader/flanger where two
operations occur. First, the machine rolls a series of ribs into the cylinder bodies. The ribs
strengthen the walls, allowing the cans to withstand the pressure generated during the sterilizing
process. Second, the machine curves the rims of the cylinders to form a flange. The flange is
essential for the next step in the process where the ends are attached to the cylinder bodies.
2-12
-------
Can ends are punched and formed on a separate manufacturing line at the same time the
can cylinders are formed. A coil of pre-coated tin plate steel unwinds into a tray where it
receives a paraffin coating for lubrication. The coil travels from the tray to the scroll strip
shearer which cuts the steel into indented rectangular strips. The indented shape of the strips
minimizes the amount of scrap steel generated during the process. The strips move along a
conveyor belt to the end press which punches circular ends and removes the scrap steel from the
belt to a recycling container. The ends then travel to a compound liner where they receive a
sealing compound. After the compound liner, the cans are ready to be attached to the body
cylinders.
The two sections of the line join at the end seamer. When the body cylinders enter the
end seamer, they are turned upright and joined with a can end. The end seamer then double rolls
the flanged end of the cylinder with the can end. The first roll grips the end onto the flange, and
the second roll folds them together up toward the can body. When the cans exit the end seamer,
they pass over a light which tests them for leaks. If they pass the test, the cans move to the
palletizer which stacks them for shipping.
2.4.4 Product Outputs
There are several different sizes for three-piece food cans. The most common are 10 oz,
12 oz, and 16-19 oz. The line speeds for three-piece cans range from 350 to 800 cpm.5
2.5 EMISSIONS AND WASTES
2.5.1 Introduction
The following paragraphs describe the environmental impacts of a conventional D and I,
two-piece can manufacturing process. Because the barriers section of the report focuses on D
and I two-piece manufacturing for beer and beverage cans, this section does not include the
impacts of the DTR and three-piece processes.
2-13
-------
2.5.2 Air Emissions
The primary source of air emissions in the two-piece process are the three ovens
(basecoater oven, deco oven, and 1C oven), the printer, and the internal coater. Secondary
sources include the coating and lubricating stations. Although a percentage breakdown by
individual source is not available, it is recognized that the majority of emissions from can
manufacturing operations originate from the coatings. Most manufacturers coat the interior and
exterior of their cans with waterbased inks and coatings, which contain 10 to 15 percent volatile
organic compounds (VOCs).6 The primary VOCs in the inks and coatings are glycol ethers (in
particular butylcellosolve), n-butyl alcohol, and dimethylethanolamine.6 Table 2-7 provides a
range of the VOC contents typically found in waterbased inks and coatings. Although the
percentage of contents may vary with the type of coating, the ranges are similar for internal
coating, base coating, bottom coating, and overvarnish.6
TABLE 2-7. VOC CONTENT OF WATERBASED INKS AND COATINGS
VOC Constituent Percentage Content In Coatings Percentage Content In Inks
Glycol ethers 5-10 Up to 15
n-butyl alcohol 0-5 0-6
Dimethylethano- 0-4 0-4
lamine
Source: Refereoce 6
Table 2-8 provides stack and fugitive emissions reported to the TRI by the following two-
piece manufacturing facilities: the Ball Corporation plant in Williamsburg, Virginia; the Miller
Brewing Company plant in Reidsville, North Carolina, and the Stroh's Brewery plant in Winston-
Salem, North Carolina. American National Can Company currently owns the Stroh plant, but
at the time of the reporting, Stroh owned the facility. To provide a measure of each plant's size,
the employment range is given. Production outputs were not available from the plants due to
confidentiality.
2-14
-------
TABLE 2-8. AIR EMISSIONS FROM TWO-PIECE CAN MANUFACTURING
FACILITIES - 1992
Employment n-Butyl Alcohol Glycol Ethers
Facility Range (lbs) (lbs)
Stack Fugitive Stack Fugitive
Ball Corporation
Williamsburg, VA
100-249
120,000
120,000
190,000
190,000
Stroh Brewery Company
Winston-Salem, NC
250-499
282,000
0
415,000
0
Miller Brewing Company
Reidsville, NC
250-499
88,000
6,400
130,000
7,100
Source: References 7 and 8
The difference in the reporting methods at the plants may account for some of the
variation in the emissions quantities. The Stroh, Miller, and Ball facilities used the mass balance
approach to estimate their emissions.
The stack emissions were from the ovens and internal coaters at the facilities. The
fugitive emissions came from the basecoater (if the plants have one), the bottom coater, the
printer, the roller for the overvarnish, and the internal coater.
2.5.3 Water Releases
The only source of wastewater in a can manufacturing facility is the can washer. The
washer uses a detergent, sulfuric acid, and sometimes hydrogen fluoride to wash the cans prior
to decoration. The water leaving the washer often contains manganese (from the aluminum
cans), oils, dust, and polymers.
Plants normally run their wastewater through some type of treatment system to remove
the hazardous chemicals and solids. At least one facility uses a dissolved air flotation and
flocculation system. In this system, the water passes through a series of filters that catch
aluminum particles, oil, dust, and polymers, forming a non-hazardous filter cake. The cake,
consisting of approximately 50 percent solids, is disposed of in a sanitary landfill. The sulfuric
2-15
-------
acid and hydrogen fluoride in the water are neutralized by a caustic solution. The combination
of physical and chemical treatments removes nearly all of the hazardous chemicals from the
water.
Table 2-9 lists the amount of sulfuric acid and hydrogen fluoride treated on-site at each
of the facilities listed in Table 2-8. The reported treatment efficiencies were 100 percent at each
of the facilities, so there were no reported releases to the environment.
TABLE 2-9. WATER TREATMENT DATA FOR TWO-PIECE CAN
MANUFACTURING FACILITIES - 1992
Facility
Employment
Range
Sulfuric Acid
(lbs)
Hydrogen Fluoride
(lbs)
Ball Corporation
Williamsburg, VA
150-249
670,000
20,000
Stroh Brewery Company
Winston-Salem, NC
250-499
871,000
72,000
Miller Brewing Company
Reidsville, NC
250-499
190,000
24,000
Source: Reference 7 and 8
2.5.4 Solid Waste
The primary solid wastes generated by the two-piece process are scrap aluminum, filter
cakes from the water treatment system, rags used to clean the printer and other machines, and
spent coatings and inks. Most can manufacturing facilities recycle nearly 100 percent of their
scrap aluminum. A small amount of aluminum shavings is swept up with dirt from the floor
during normal cleaning operations. The used rags are not considered hazardous waste and are
often sent to an industrial cleaner. After their cleaning, they are returned to the plant for re-use.
The spent inks and coatings are considered solid waste even though they contain
hazardous chemicals in their virgin form. During the coating process, most of the hazardous
chemicals (e.g., glycol ethers) evaporate from the inks and coatings, leaving a residue that does
not exhibit any hazardous characteristics. Nearly all spent coatings are internal coating from the
2-16
-------
overspray of the internal coater's spray guns. These coatings are often picked up by a contractor
and shipped off-site where they are used as supplemental fuel.9
2.5.5 Hazardous Waste
The primary hazardous wastes generated by the two-piece process are solvents used to
clean the printer and other equipment, unused virgin coatings, and floor stripping. Once the
solvents are spent, they are put into drums and shipped off-site for recycling or supplemental fuel
use. Can manufacturers try to avoid generating unused virgin coating because it is an
unnecessary expense; however, they occasionally cannot use a coating within its six-month shelf
life. The coating is then disposed of in the same manner as spent solvent. However, when the
inks are spilled on the floor stripping, a mixed waste is created because the floor stripping is
considered hazardous. As a result, the combined waste must be removed and disposed of as
hazardous waste.9
2.5.6 Hazardous Chemicals
Although hazardous chemicals are prevalent throughout the two-piece manufacturing
process, a significant portion of them are recycled, treated, recovered for energy use, or emitted
into the air as described in Section 2.5.2. Table 2-10 lists the most common hazardous chemicals
found in a conventional two-piece manufacturing process. The information is based on the TRI
data reported by the three facilities identified in previous tables.
2-17
-------
TABLE 2-10. HAZARDOUS CHEMICALS USED IN CONVENTIONAL
TWO-PIECE MANUFACTURING PROCESS
Chemical Media Of Waste Source And Treatment
n-Butyl Alcohol
Air Emissions,
Solid Waste
Contained in waterbased inks and coatings.
Primarily stack emissions from three ovens
(basecoater, deco and IC ovens) or fugitive
emissions from basecoater, printer, or internal
coater. Typically no control on ovens. Small
amount disposed of off-site as nonhazardous solid
waste.
Glycol Ethers
Air Emissions,
Solid Waste
Major VOC constituent in waterbased inks. Also
present in waterbased coatings. Primarily stack
emissions from three ovens (basecoater, deco and
IC ovens) or fugitive emissions from basecoater,
printer, or internal coater. Small amount disposed
of off-site as nonhazardous solid waste.
Manganese
Solid Waste,
Water
Aluminum scrap from bodymalong process and
aluminum fines in spent lubricant and water from
washer. Aluminum scrap recycled off-site.
Aluminum fines in water removed through
filtration. Caught in filter cakes.
Hydrogen Fluoride
Water
Used as cleaning solution in washer. Neutralized
in water filtration system. Treatment reported to
be 100 percent effective.
Sulfuric Acid
Water
Used as cleaning solution in washer. Neutralized
in water filtration system. Treatment reported to
be 100 percent effective.
2-18
-------
2.6 REFERENCES
1. Telecon. Tony Grandinotti and John Burnett, Bali Corporation, Broomfield, CO, with
Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of
Ball's UV trial lines at the Findlay, OH plant. November 4, 1993.
2. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Ball Can Manufacturing Plant, Williamsburg, VA. October 21, 1993.
3. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Coors Container Complex, Golden, CO. August 16, 1993.
4. Church, Fred L. "New Draw/Thin/Redraw Process Makes a Super Can for Campbell,"
Modem Metals. 42(3), pp. 34-35. April 1986.
5. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Campbell Soup, Maxton, NC. October 27, 1993.
6. Telecon. Timothy D. Case, Ball Corporation, Metal Container Division, Broomfield, CO,
with Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion
of hazardous chemicals in can manufacturing. December 1, 1993.
7. North Carolina Manufacturing Firms: 1989-90 Directory. Prepared by North Carolina
Department of Commerce. Raleigh, NC. 1989.
8. Toxic Chemical Release Inventory 1992 Form R Submittals. Information gathered from
North Carolina Department of Environment, Health, and Natural Resources. Raleigh, NC.
November 1993.
9. Telecon. Timothy D. Case, Ball Corporation, Metal Container Division, Broomfield, CO,
with Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion
of solid wastes from can manufacturing process. December 9, 1993.
2-19
-------
CHAPTER 3
DESCRIPTION OF ULTRAVIOLET-CURING TECHNOLOGY
3.1 GENERAL
The specific focus of this chapter is ultraviolet (UV)-curable coatings. UV-curable
materials are one type of radiation-curable coatings. The use of UV-curable coatings does not
significantly alter the two-piece manufacturing process for beer and beverage cans. The
differences occur in the coating operations. Because Coors is the only two-piece can
manufacturer currently using UV-curable coatings, much of the information in this chapter was
gathered during a site visit to the Coors Container Complex in Golden, Colorado.1
3.2 PROCESS DIFFERENTIALS FOR UV-CURING
3.2.1 Material Inputs and Equipment
The primary material and equipment differences between a UV system and a conventional
thermal system are UV-curable coatings and UV-curing ovens. These differences are summarized
in Table 3-1. UV-curable inks and coatings consist of photoinitiators, organic monomers,
oligomers, pigments, fillers, compounds that affect processing, and inhibitors.2 Table 3-2 briefly
describes the function of each of the constituents. Pigments, fillers, and the miscellaneous
compounds for processing typically serve the same purpose in both UV-curable and conventional
coatings.
TABLE 3-1. DIFFERENCES IN MATERIAL INPUTS AND EQUIPMENT
Conventional Material Inputs/Equipment
UV System Material Inputs/Equipment
Waterbased inks
UV-curable, acrylic-based inks
WateTbased overvarnish
UV-curable, acrylic-based overvarnish
Thermal deco oven
UV-curing deco oven
3-1
-------
TABLE 3-2. CONTENTS OF UV-CURABLE INK OR COATING
Contents
Function
Compounds for processing
Inhibitors
Oligomers
Pigments
Fillers
Photoinitiators
Monomers
Stimulated by UV-curing lamp, release free radicals
Reduce coating's viscosity, provide final application
characteristics, do not evaporate like solids but remain part
of cured coating
Give coating its physical and performance characteristics
Add color characteristics
Increase viscosity, maintain consistency
Leveling agents, flow agents, microbiocides, etc.
Provide shelf storage stability up to 12 months at 120° F
(49°C)
Source: Reference 2
3.2.2 UV-Curing Process
Because the bodymaking section of a UV-curing line is similar to that of a conventional
thermal line, the process description begins with the decoration section of the line. A UV system
uses the same printer as a conventional system, shown in Figure 2-1. As the mandrel wheel
delivers the cans inside the printer, a rubberized blanket applies the complete color image of the
label to the exterior of the cans. The mandrel then moves the cans to a roller which applies an
overvamish to complete the "wet-on-wet" application. Coors normally applies a wet film
thickness (i.e., the weight of applied coating per can) of 0.5 to 0.8 mil (100 to 120 mg) of UV-
curable overvamish to each can.1 This coating thickness is slightly higher than the 75 mg per
can industry standard for waterbased coatings.3 Coors, however, is hopeful that they will be able
to lower their applied UV-curable coating thickness to the industry standard for waterbased
coatings.1
Another difference between the conventional and UV-curing processes is the curing of
the exterior coating. In a conventional system, waterbased inks and coatings cure inside a large
thermal deco oven that operates near 400°F (204°C). The cans spend from 3 to 45 seconds inside
the oven. In a UV system, the cans spend approximately one second inside a UV-curing oven
3-2
-------
operating at approximately 110°F (43°C). The actual curing of the UV-curable materials requires
0.7 to 3 seconds.4,16 UV-curing ovens are also smaller than conventional thermal ovens, saving
considerable floor space in a can plant. A typical UV-curing oven is 9 ft (2.74 m) long, 5 ft
(1.52 m) wide and 5 ft (1.52 m) high.4 In comparison, a thermal deco oven may be 18 ft (5.49
m) long, 5 ft (1.52 m) wide and 15 ft (4.57 m) high.4 At today's lines speeds, the ovens must
be large enough to accommodate 2,000 cans per minute (cpm). Figure 3-1 illustrates the UV-
printing process.
Coors' inks are designed to be compatible with the UV-curable overvamish. Once the
photoinitiators in the overvamish have been stimulated for curing, the inks are captured inside
it. Most of the inks that Coors uses do not have photoinitiators. This reduces the cost of the
inks and expands the range of colors available for production.
Coors uses Deco Ray 2 ovens designed by Fusion Systems of Rockville, Maryland. The
ovens have from four to eight ten-inch (25.40 cm)-long lamp modules that are positioned at a
20 degree angle to the horizontal axis. The lamp angle ensures that the cans are cured from top
to bottom. The lamps contain mercury rather than electrodes. The lamps are surrounded by
parabolic reflectors which provide a uniform intensity of light along the exterior of the cans,
enabling them to cure evenly. Figure 3-2 illustrates the curing mechanism. Figure 3-3 shows
an enlarged parabolic reflector. As the cans pass through the oven, they are positioned upright
and several inches apart (from the center of one can to the next) on a vacuum belt.1
The remainder of the decorating process in a UV system is the same as that of a
conventional thermal system. The cans receive a waterbased internal coating which is cured
inside a thermal IC oven operating at temperatures of 320°F to 400°F (160 to 204°C). They
spend 45 to 150 seconds inside the oven. After the internal coating is cured, the cans are necked
and flanged to narrow their open ends. They are then tested for leaks by a light tester and
assembled and stacked for shipment on a palletizer.
It should be noted that Coors does not use a basecoat or a bottom coat on its cans.
Hence, the application equipment for these coatings is not on Coors' lines.
Coors prints approximately 10 basic labels for its cans and 130 to 140 specialty labels that
are variations of the basic labels.1 For example, Coors prints a specialty label for the Rodeo
Showdown in Scottsdale, Arizona each year.1 The colors of the label are the same as those used
on a regular Coors label (red, gold, black and tan), but the design is different with the picture
3-3
-------
mandrel wheel blanket
segment
can feed \ wheel
transfer unit
r Ink plate
cans
pin chain
overcoat unit
overcoat application roller
Figure 3-1. UV-printlng process for two-piece beer cans.1
3-4
-------
OCCO CHAIN*
UMf * CHAM t**OCKET
CAtmiT
MPUCTOft
vwaoutoh
LOWER CHAIN
IMOCXIT
MRADIATOH
RIFUCTOA
C-SO U. V. OVEN
Figure 3-2. Two-piece beer can UV-curing process.1
3-5
-------
oN
M
u
o\
M
SU B S T R /\ T E
Figure 3-3. Parabolic reflectors.
3-6
-------
of a cowboy riding a horse on one side of the can. Over time, Coors has been able to refine the
chemistry of its inks and overvarnishes, so that the two cure properly producing the desired color
shades in a cost effective manner.
Despite Coors' success in a captive plant, there are uncertainties about the potential use
of UV-curable inks and coatings in a merchant can plant. Merchant plants may print 400 to 500
labels in a year for different beer and beverage companies.5 Most of the label specifications are
very different, requiring different designs, colors, and quantities of inks and coatings. Ball, for
example, uses approximately 2500 different colors. It is unclear whether or not a merchant
manufacturer can meet its label requirements and remain competitive using a UV-curing system.
For example, the Pepsi and Coke labels require a considerable amount of ink. Pepsi blue, an
expensive ink, is a prominent part of the Pepsi label. According to the ink prices from the Ball
trial runs, a UV-curable Pepsi blue ink costs $2.74 per lb more than a waterbased Pepsi blue.6
If 10 lbs of Pepsi blue ink are consumed per million cans, a merchant manufacturer fulfilling an
order of 100 million cans for Pepsi would spend $2,740 more on Pepsi blue using the UV-curable
alternative.
3.2.3 Emissions and Wastes
During the curing process in a UV-curing oven, the UV lamps stimulate the
photoinitiators to release free radicals that crosslink oligomers and monomers to cure the coating.
Only trace amounts of VOCs are emitted during the process. Therefore, UV curing essentially
eliminates the deco oven as a source of VOC emissions.
3.2.3.1 Air Emissions
Table 3-3 lists the contents from a Material Safety Data Sheet (MSDS) for a standard
UV-curable overvamish. Coors has conducted a series of tests to evaluate the VOC content and
resulting emissions in UV-curable and waterbased coatings. The company used testing Method
24 recommended by the American Society of Testing Methods (ASTM) to determine the VOC
content of its UV-curable and waterbased coatings.4 Method 24 is used for paints and other
surface coatings, and Coors modified it "to add UV curing prior to the gravimetric analysis in
3-7
-------
the procedure, in order to accurately reflect the UV initiated cross-linking of the inks and
overcoats."4 The results of the test are shown in Table 3-4.
TABLE 3-3. CONTENTS FOR UV-CURABLE OVERVARNISH
Contents
Percent
2-ethylhexyl acrylate
10
Ethylene glycol monopropyl ether
5
Film formers, resins, and additives
50
Reactant diluents
30
Source: Reference 7
TABLE 3-4. COORS TEST RESULTS FOR VOC CONTENT OF COATINGS
Coating
VOC Content - Tons/Billion Cans
Waterbased Coating
28.9
UV-Curable Acrylate Coating
1.68
UV-Curable Epoxy (Cationic) Coating
.22
Source: Reference 4
Coors calculated a 94 percent emissions reduction with their acrylate coating, which is
the current generation of UV-curable overvamish. The epoxy coating, commonly referred to as
a cationic coating, is a new generation of coatings that Coors has been testing on one of its lines.4
The emissions reduction potential is even greater with the cationic coating.
Coors conducted a stack test on the UV-curing oven exhaust to determine specific
emissions levels of five VOCs from its UV-curable overvamish. The company used EPA
Method 18, and chose charcoal tube absorption followed by solvent desorbtion and gas
chromatography/mass spectrometry (GC/MS) analysis for the test.4 The results of the test, shown
in Table 3-5, revealed that concentrations were below detection limits. Coors then used the
3-8
-------
results to calculate emissions for the targeted compounds on a tons per year basis. The tons per
year total indicates that the UV-curing oven emits less than 1.5 tons of VOCs per year.4
TABLE 3-5. VOC EMISSION ESTIMATES BASED ON COORS STACK TESTING
Compound
Concentration (ug/L)
Tons/Year
n-butyl alcohol
<5
<0.3
Ethoxyethanol
<5
<0.3
o-xylene
< 5
<0.3
Ethoxyethoxyethanol
< 5
<0.3
Benzophenone
< 5
<0.3
Total
< 25
< 1.5
Source: Reference 4
Coors also calculated VOC and hazardous air pollutant (HAP) emissions from UV-curable
and waterbased coatings based on the content levels provided in their material safety data sheets
(MSDS)."1 Table 3-6 lists the estimates. An estimate is provided for the bottom coat of cans
although Coors does not currently use this coating in its UV-curing process.
TABLE 3-6. COORS VOC AND HAP EMISSION ESTIMATES FOR UV AND
THERMAL SYSTEMS
Coatings
VOC Emissions
(Tons/Billion Cans)
Thermal
UV
HAP Emissions
(Tons/Billion Cans)
Thermal
UV
Overcoat
Ink
Bottom Coat
Total
26.5
0.8
1.3
28.6
1.3
0.2
0.1
1.6
13.2
0.4
0.7
14.3
0.0
0.0
0.0
0.0
Source: Reference 4
In addition to VOC and HAP estimates, Coors calculated carbon dioxide (C02) emission
estimates based on EPA conversion factors for natural gas combustion and emission factors for
3-9
-------
electrical power production.4 A conventional system uses electricity for all of the equipment on
the line except for the thermal ovens (IC, basecoater, and deco), which use natural gas.
Conversely, the deco oven on a UV system uses electricity instead of natural gas. Table 3-7
compares the emission estimates of a UV system to those of a conventional thermal system.
Coors calculated the estimates based on the four billion cans per year production volume of its
Golden, Colorado plant4 The COz emissions include those generated by the electrical power
plant for the production of electricity used by the can plant.4
TABLE 3-7. COORS EMISSION REDUCTION ESTIMATES FOR
GOLDEN, CO PLANT
Pollutant
UV System
(Tons/Year)
Thermal
(Tons/Year)
Annual Reductions
(Tons/Year)
VOCs
6.4
114.0
107.6
HAPs
0.2
57.2
57.0
C02
4,200
8,416
4,216
Source: Reference 4
Coors extrapolated the data in Table 3-8 to a national level. Table 3-8 also contains the
company's estimated emission reductions for the nation if all can manufacturing facilities
converted to a UV system. The estimates are based on the production volume of 100 billion cans
per year4
TABLE 3-8. COORS EMISSION REDUCTION ESTIMATES FOR NATION
Pollutant
UV System
(Tons/Year)
Thermal
(Tons/Year)
Annual Reductions
(Tons/Year)
VOCs
160
2,850
2,690
HAPs
5
1,430
1,425
n
o
105,000
210,400
105,400
Source: Reference 4
3-10
-------
Despite the reductions that Coors achieves with its UV-curing ovens and UV-curable
coatings, the plant emits pollutants from the other operations on its lines. Table 3-9 provides the
emissions that Coors reported to TRI for its Golden, Colorado plant in 1992. The primary source
of emissions in the plant is the internal coater and 1C oven, which are the same as used on
conventional thermal lines. The internal coater uses waterbased coatings, containing n-butyl
alcohol and glycol ethers. The quantities reported by Coors cannot be directly compared to those
reported by the facilities in Tables 2-9 and 2-10 because the production volumes are different.
Because the Coors plant is the largest can plant in the world, producing nearly four billion cans
a year, their output is significantly larger than the output of other can facilities.
TABLE 3-9. EMISSIONS REPORTED TO TRI BY COORS CONTAINER
COMPLEX IN GOLDEN, CO - 1992
VOC Emitted
Fugitive Emissions (lbs)
Stack Emissions (lbs)
n-Butyl Alcohol
111,233
124,661
Glycol Ethers
34,308
37,733
Source: Reference 8
It is important to note that the UV-curing oven has had a significant impact on emissions
reductions at Coors. The results in Table 3-7 indicate that emissions of n-butyl alcohol and
glycol ethers would have been 41 percent higher with a conventional thermal process, assuming
that these two chemicals constitute all of the VOCs eliminated by the UV-curing process.
3.23.2 Water Releases
A plant using UV-curable inks and coatings generates the same wastewater as a plant
using waterbased inks and coatings. The wastewater comes from the washer in the bodymaking
section of the line. It contains sulfuric acid from the cleaning solution, manganese from the cans,
aluminum fines, and spent oils and lubricants. Coors drains its wastewater into a tank where it
is treated with a lime slurry to neutralize the sulfuric acid. The metals settle to the bottom of
the tank, and the spent oils are skimmed from the top. The treatment efficiency for the sulfuric
acid is reported to be 100 percent.
3-11
-------
3.2.3.3 Solid Waste
The UV-curing process generates scrap aluminum, used rags from cleaning the printer and
other equipment, spent lime from the wastewater treatment tank, spent filters from the
bodymaker, spent bulbs from the UV-curing oven, and waste coatings.9 Coors gathers all of its
scrap aluminum (e.g., scrap from the cupper and cans with defective labels), crushes it into bales,
and ships it to aluminum manufacturers for recycling. The dirty rags are shipped off-site, cleaned
by an industrial cleaner, and returned to the plant. Coors returns the spent bulbs from the oven
to Fusion Systems where the mercury inside them is reclaimed. The bulbs are guaranteed by
Fusion Systems to last 5,000 hours.
During the 1970s and 80s, Coors treated its waste UV-curable inks and coatings as
hazardous waste. However, today the UV-curable materials contain only trace amounts of
solvents. The company recently conducted a series of tests, including Toxicity Characteristic
Leaching Procedure (TCLP) and EP toxicity, on its waste inks and coatings and determined that
they are not hazardous.'1 Coors considers its spent internal coating to be a nonhazardous solid
waste. A contractor picks up all of the spent coatings and incinerates them off-site.
3.2.3.4 Hazardous Waste
The primary hazardous waste generated by a UV-curing line is the materials used to clean
the printing plates. Coors has different cleaners for each line. Some of the cleaners contain
caustic compounds, while others contain propylene carbonate or propylene glycol. A contractor
transports the spent cleaners off-site where they are either recycled or incinerated.9
3.2.3.5 Hazardous Chemicals
In 1992, the Coors can manufacturing plant reported the same hazardous chemicals to TRI
that the conventional can manufacturers reported. (See Table 2-11 for a listing of the chemicals.)
Coors did not report hydrogen fluoride because it does not use the chemical in its washer. The
major difference between the UV-curing and conventional processes is the HAP emissions from
the decorating ovens. The UV-curing oven emits only trace amounts of n-butyl alcohol and
glycol ethers. A thermal deco oven emits significant quantities of the two chemicals.
3-12
-------
3.2.4 Energy
UV-curing and thermal ovens also differ in their energy consumption. A UV-curing deco
oven consumes electricity and operates at 110°F (43°C).4 A thermal deco oven consumes natural
gas while operating at 400°F (204°C). A UV-curing deco oven uses less energy than a thermal
deco oven, but proponents of the two systems disagree over how much energy is saved. Coors
estimates the energy savings of a UV-curing oven to be approximately 45 percent4 Table 3-10
provides the results of a study that Coors conducted on the energy savings with a UV-curing
oven.
TABLE 3-10. COORS ESTIMATES FOR ENERGY SAVINGS FROM UV-CURING
OVEN VERSUS THERMAL OVEN
Energy
UV-Curing Oven
(MMBTU/Billion
Cans)
Thermal Oven
(MMBTU/Billion
Cans)
Energy Savings
(MMBTU/Billion
Cans)
Natural Gas
0
15,400
15,400
Electrical
10,500
9,980
-520
Total Energy Savings
14,880
Source: Reference 4
The major energy savings is in natural gas. The UV-curing oven consumes none while
the thermal oven consumes over 15 billion BTUs per year for heating. According to Coors,
electrical energy consumption for the two ovens is similar. The UV-curing oven requires slightly
more electricity to operate its UV-curing lamps; however, the thermal oven consumes a
significant amount of electricity to operate blowers to cool the oven and a pin chain to transport
the cans.4
Ball Corporation conducted its own energy evaluation of the two technologies when it set
up a trial UV-curing line at their Findlay, Ohio plant in 1986-87.10 The plant ran several trial
runs in a full production setting over a year. Table 3-11 provides energy consumption estimates
based on the Ball trial runs. These consumption figures assume the following:10
3-13
-------
All motors were operating at 100 percent efficiency.
The UV-curing lamps were running on their high settings, operating at full power of 400
watts/in. UV-curing oven was using an increased vacuum blower size.
TABLE 3-11. BALL CORPORATION - ENERGY COMPARISON OF UV-CURING
VERSUS THERMAL OVEN
UV Oven
Thermal Oven
Vacuum belt:
(34 HP x 0.747 kw/HP = 25 kw)
6 lamps x 7.5 kw = 45 kw
4 lamps x 5.0 kw = 20 kw
Total electricity consumption = 91 kw
Natural gas consumption = 0 scfh
Pin chain, blowers =
60 HP x 0.747 kw/HP = 45 kw
Total electricity consumption = 45 kw
Natural gas consumption = 1,000 scfh
106 Btu/h
Source: Reference 10
According to Ball's comparison, the electricity consumed by the UV system is twice that
consumed by the thermal oven, which partially offsets the energy savings achieved by eliminating
the use of natural gas in the thermal oven. The Ball results, however, may overestimate
electricity consumption because the lamps were operated at full power. Coors is able to operate
its lamps at the medium setting of 300 watts/in. Ball had to operate the lamps at the higher level
because it was experiencing technical problems with the UV-curable coatings. These problems
are discussed in more detail in Chapter 4.
3-14
-------
3.3 COST DIFFERENTIALS
3.3.1 Introduction
The four types of costs associated with a can manufacturing line are material costs,
equipment costs, operating and maintenance costs, and energy costs. The following sections
compare the costs of a conventional thermal system with those of a UV system. The information
is based on data obtained from the Coors can plant in Golden, Colorado and the Ball trial line
in Findlay, Ohio (1986-87).
3.3.2 Material Costs
The primary material costs of a two-piece can line are aluminum, basecoat (if necessary),
bottom coat (if necessary), inks, overvamish, and internal coating. The costs of aluminum and
internal coating are the same for the two systems because the same materials are used. The
primary differences are the costs of the inks and overvamish used to coat the exterior of the cans.
In comparing the cost difference of a UV-curable overvamish with a waterbased overvamish, one
has to consider the percentage of solids contained in the coatings. The solids content is the
portion of the coating that remains on the substrate. A waterbased coating typically consists of
30 to 35 percent solids, 50 to 60 percent water, and 10 to 15 percent solvent.11 The water and
the solvent evaporate from the coating substrate during the drying/curing process. A UV-curable
coating consists of 99 to 100 percent solids with trace amounts of solvents and other
constituents.11 Nearly all of the coating remains on the exterior of the can during curing.
A gallon of waterbased overvamish normally costs from $5 to $7, and a gallon of UV-
curable overvamish costs from $28 to $35.1,11 However, when applied solids are considered, the
cost of a waterbased overvamish increases to approximately $20 per gallon of solids.1 Some
manufacturers use overvamish for their bottom coat, and others use a waterbased coating with
a higher concentration of solids costing from $8 to $10 per gallon.11 UV-curable bottom coats
typically cost a minimum of $25 per gallon.11 The cost of waterbased inks ranges from $4 to $7
per lb and the cost of UV-curable inks ranges from $7 to $10 per lb.12,13
3-15
-------
Coors maintains that the applied costs of UV-curable inks and overvarnish are
approximately five percent higher than the applied costs of waterbased inks and coatings.4 The
company bases its estimate on the experience it has gained in operating a UV system for nearly
20 years.
During the trial runs at the Findlay, Ohio plant, Ball Corporation performed a cost
comparison of UV-curable materials and conventional waterbased materials. Table 3-12 shows
the results of this study. There are two sets of estimates for the UV-curing line. One set is
based on actual results from the trial runs, and the other set is the projected estimates Ball
believed it could achieve with improvements to the UV-curing line. Ball never reached the
projected results because of technical difficulties that will be discussed in Chapter 4.
TABLE 3-12. BALL CORPORATION MATERIAL COST COMPARISON OF
UV-CURABLE TO WATERBASED MATERIALS -
FINDLAY, OH, 1986 - 87
Conventional
Thermal Line
Bail's UV
Trial Line
Ball's Projected
UV Line
1. Overvarnish weight (mg/can)'
90
120
100
2. Overvarnish cost ($/gallon)
5.32
25.00
25.00
3. Calculated overvarnish cost
($/l,000 cans)
0.402
0.79
0.66
4. Bottom Coat ($/gallon)
5.32
25.00
25.00
5. Calculated bottom coat cost
($/l,000 cans)
0.07
0.103
0.103
6. Calculated ink cost
($/l,000 cans)
0.424
0.573
0.573
Total calculated costs (3+5+6)
($/l,000 cans)
0.896
1.466
1.336
*Wet film application
Source: Reference 6
Ball's analysis indicates that the material costs of its trial UV-curing line were 64 percent
higher than the material costs of a conventional thermal line. The projected material costs of the
UV-curing line were 53 percent higher. Ball calculated the costs in the following manner:6
3-16
-------
Overvarnish Cost
Conventional Overvarnish
Coating weight per can = [90 mg/can x 1.10 (spoilage)]/.34 (solids) = 291.18 mg/can
Using a density of 8.5 lb/gallon, overvarnish consumption = .0755 gallon/1,000 cans
Cost = consumption x cost/gallon, .0755 x $5.32 = $.4017/1,000 cans
Where spoilage is the sum of conventional spoilage, deco scrap loss, solids loss in thermal cure,
spillage, etc.
UV-Curable Overvarnish
Coating weight per can = [120 mg/can x 1.05 (spoilage)]/.99 (solids) = 127.27 mg/can
Using a density of 8.9 lb/gallon, overvarnish consumption^ .0315 gallon/1,000 cans
Cost = consumption x cost/gallon, $.0315 gallons/1,000 cans x $25 = $.7875/1,000 cans
Where spoilage is the sum of UV spoilage, deco scrap, loss from drum transfer, spillage, etc.
Bottom Coat Cost
The same calculation method as shown above was used. Conventional and UV-curable
estimates are based on 15 mg/can and 15 percent spoilage.
Ink Cost
I
Ink prices and consumption values, based on 1988 data, are presented in Table 3-13. The
data indicate that consumption does not vary between conventional and UV-curable inks.
3-17
-------
TABLE 3-13. INK PRICES AND CONSUMPTION
Coating Cost ($/lb) Consumption (lb/million cans)
Conventional Inks
Pepsi Red 6.85 45
Pepsi Blue 6.30 10
Pepsi White 1.65 30
Classic Coke Red 6.85 80
Classic Coke White 1.65 20
Diet Coke Red 1.65 20
Diet Coke White 6.85 80
UV-Curable Inks
Coke and Pepsi Red 6.90 45
Pepsi Blue 9.04 10
Pepsi White 4.85 30
Source: Reference 6
3.3J Operating and Maintenance Costs
Proponents of UV-curing technology have identified a number of operational efficiencies
associated with the UV-curing oven. Table 3-14 lists benefits that Coors has derived from its
UV-curing lines in Golden, Colorado.
TABLE 3-14. OPERATIONAL EFFICIENCIES OF UV-CURING SYSTEM
Operational Efficiency
UV vs. Conventional System
Oven size
UV-curing oven uses 90 percent less floor space than thermal oven
Downtime
Significantly less downtime with UV-curing oven
Maintenance
78 percent less maintenance with UV-curing oven
Parts
Replacement of parts 72 percent less with UV-curing oven
Process control
UV-curing oven simpler
Source: Reference 4
3-18
-------
According to Coors, there is significantly less downtime with a UV-curing oven. This
downtime is minor when compared to other reasons such as changing the ink or a conveyor
breakdown.14 However, the high heat in the thermal ovens may contribute to other downtime
problems. The UV-curing oven operates at approximately 110°F (43°C), compared to 400°F
(204°C).4 Higher temperatures in thermal ovens tend to cause wear on the pin chain. If the chain
breaks during operation, it causes considerable delays in production.1 The parts of a UV-curing
oven are modular and can be replaced quickly by removing the used part, (e.g., a spent bulb),
and replacing it with a new part1 A UV-curing oven also has a quick start time, approximately
5 minutes; a thermal oven takes longer to heat to its operating temperature.4
Some can manufacturers have found that adhering to a strict preventive maintenance
schedule eliminates time-consuming interruptions from their thermal deco ovens. For example,
one merchant can manufacturing plant schedules 20 preventive maintenance days a year for each
of its lines.5 Employees clean the machines, change the oil, replace belts, lubricate chains, and
complete other tasks necessary to run efficient lines. The result is a clean line with few
interruptions.
During its trial runs at the Findlay, Ohio plant, Ball Corporation found that the annual
maintenance and repair costs of its UV-curing oven were less than with its thermal deco oven;
however, the savings were not nearly as large as the company expected.10 The maintenance and
repair costs that Ball developed for its two systems are presented in Table 3-15. The estimates
for the two ovens are based on line speeds of 1,200 to 1,400 cpm.10
According to the Ball estimates, the maintenance and repair costs of the UV-curing oven
are only 22 percent less than the thermal oven. It is important to note that the Ball estimates are
based on a limited number of trial runs with a UV-curing system. At the time of the trial runs,
Ball did not have years of experience operating a UV-curing oven. Therefore, these numbers
reflect the initial difference in maintenance and repair costs between the two ovens. The UV
estimates would likely decrease over time as Ball became more familiar with the operational
capabilities of the UV-curing oven.
3-19
-------
TABLE 3-15. ANNUAL OVEN MAINTENANCE AND REPAIR COST ESTIMATE
Cost ($)
Repair
Thermal
UV-Curing
Labor
Total
Pin chain replacement
Parts and Labor
Parts
NA
23,000
SO.042/1,000 cans
12,000*
11,000
NA
2,000"
15,875
14,000°
1,875
17,875
Distributed over 550 million cans
$0,032/1,000 cans
Source: Reference 10
^Assumes replacing 1,200 ft chain at S 10/ft
^Assumes replacing 200 ft chain at $ 10/ft
Includes lamps, bearings, table top belts
"Assumes 75 hours @ $25/hour
NA - Not Available
3.3.4 Energy Costs
Coors estimates that a UV-curing oven consumes about 45 percent less energy than a
thermal oven on a BTU basis.4 Section 3.2.4 contains energy consumption figures. Lower
energy consumption translates into significant energy cost savings. Table 3-16 shows the
estimated energy savings from Coors UV-curing oven. The data from the Ball trial runs in
Findlay, Ohio, presented in Table 3-17, indicate energy cost savings of only eight percent based
on electricity at $0.05/kwh and natural gas at $0.27.
3-20
-------
TABLE 3-16. COORS ESTIMATED ENERGY COST SAVINGS -1993
UV-Curing Oven
Thermal Oven
Energy ($l,000/billion cans)
($l,000/bi!lion cans)
Electrical
60
57
Natural Gas
0
50
Total
60
107
Source: Reference 4
TABLE 3-17. BALL ESTIMATED ENERGY COST FOR OVENS
Cost ($/hr)
Energy
Thermal
UV-Curing
Electrical
2.25*
4.55b
Natural Gas
2.70c
0
Total
4.95
4.55
Assuming Operation of 8,500 hours/yr
$42,075
$38,675
Source: Reference 10
'Assumes 45 kwh @ $0.05/kwh
'Assumes 91 kwh @ $0.05 kwh
'Assumes 1,000 scfh @ $2.70/1,000 scfli
3.3.5 Total Operating Costs
The following table, Table 3-18, summarizes the operating cost estimates provided in the
previous sections. According to Coors, the total operating costs of a UV-curing oven are
approximately seven percent less than those of a thermal oven.
Comparing the Coors estimates to the Ball estimates can be misleading for two reasons.
First, the years in which the data were collected are different. Ball collected its data during their
UV trial runs from 1986 to 1987. Coors collected its data during 1992 and 1993. The
technology of the two processes has improved considerably since 1988. For example, most two-
piece can manufacturers using the conventional process apply 70 mg/can of wet waterbased
overvamish rather than the 90 mg/can assumed in the Ball estimates. (See Table 3-12.) This
lowers the cost of waterbased overvamish from $.402/1,000 cans to $.313/1,000 cans.6
3-21
-------
TABLE 3-18. SUMMARY OF ESTIMATED OPERATING COSTS (in $/l,000 cans)
Thermal System
UV System
Ball Projected
Item
Coors8
Ballb
Coors*
Ball"
UV Systemb
Materials (inks and
1.025
0.896
1.076
1.466
1.336
overvarnishes)
Energy
0.107
O
1—*
00
o
o
0.06
0.07
0.07
Electricity
0.057
NA
0.06
NA
NA
Natural Gas
0.05
NA
0.00
NA
NA
Maintenance
0.130
0,042
0.04
0.032
0.032
Total
1.262
1.118
1.176
1.568
1.438
"Reference 4
•"Reference 6, 10
The energy cost for the conventional thermal line may be high. Ball included an incineration cost of $.072/1,000 cans
for the energy to run an incinerator to destroy VOC emissions from the oven. Many pin ovens operate without an
incinerator. The company also assumed high natural gas prices in the estimate.
NA - Not available
The second reason that comparisons of the Coors and Ball cost data can be misleading
is the nature of the UV-curing lines. Ball conducted its UV trial line for one year. Any can
manufacturer using a new technology is likely to operate less efficiently than a manufacturer that
has used the technology for several years. Coors has been operating its UV-curing line for nearly
20 years, during which time they have resolved many of the technical problems associated with
the process.
3.3.6 Capital Costs
The capital costs are the one-time costs of purchasing the deco oven and its accessories.
Companies depreciate these costs over several years. Table 3-19 lists the components of a
UV-curing system with a line speed of 2,200 cpm. This system does not include equipment for
bottom-coated cans. The cost of this system (without installation) is approximately $200,000.15
Table 3-20 lists the components of a thermal deco oven. The oven has two zones with
17 passes at 9 ft; each zone has a 2,500 BTU/hour maximum capacity. The system has a speed
capacity of 2,000 cpm and costs approximately $375,000. The pin chain is not included with the
3-22
-------
TABLE 3-19. UV-CURING OVEN AND ACCESSORIES
Equipment
Function
UV-curing oven
Power supply cabinet
Vacuum conveyor belt
Light shields
Side irradiators
Top irradiators
Remote pressure cooling blower
Remote exhaust blower
Major structure through which cans pass during curing
Encloses power supplies for each row of lamps
4 inch wide, stainless steel belt designed to convey
2,200 cpm with center of each can 3.5 inches apart
Supports irradiators on side of belt where cans pass
Holds UV bulbs and reflects UV light rays
Holds UV bulbs and reflects UV light rays
Cools oven, contains filters
Blows emissions up through duct work
Source: Reference 15
accessories above. It costs an additional $15,000, bringing the total purchase to approximately
$390,000*
TABLE 3-20. THERMAL (PIN) OVEN AND ACCESSORIES
Equipment Function
Pin oven Major structure through which cans pass during curing
Exhaust fan with ductwork to Ventilates emissions from oven
draw from each zone
Motor drives To move pin chain
Source: Reference 16
3-23
-------
3.4 REFERENCES
1. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Coors Container Complex, Golden, CO. October 15, 1993.
2. Cyterski, David and Peter Schessler, "Bright Future For Radiation-Curable Coatings,"
Machine Design. 60(15), pp. 66-72. June 23, 1988.
3. Telecon. Tony Grandiotti and John Burnett, Ball Corporation, Broomfield, CO, and
Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of
Ball's UV trial lines at the Findlay, OH plant. November 4, 1993.
4. Donhowe, Erik T., Coors Brewing Company, "UV Pollution Prevention Technology in
Can Manufacturing," In Proceedings: Pollution Prevention Conference on Low- and No-
VOC Coating Technologies, EPA-600/R-94-022 (NITS PB94-152246), pp. 475-487.
5. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Ball Can Manufacturing Plant, Williamsburg, VA. November 12, 1993.
6. Memorandum. J. McCarthy, Ball Corporation, Broomfield, CO, to G. Richardson, Ball
Corporation, Broomfield, CO. Cost Analysis of UV Curing. July 15, 1988.
7. Material Safety Data Sheet, PPG Industries, Inc., Pittsburgh, PA. October 14, 1987.
8. Toxic Chemical Release Inventory 1992 Form R Submittals. Received from Coors
Brewing Company, Golden, CO.
9. Telecon. Erik Donhowe, Coors Brewing Company, Golden, CO, with Steven R. Church,
TRC Environmental Coiporation, Chapel Hill, NC. Discussion of waste from UV-curable
coating process. December 9, 1993.
10. Memorandum. Burnett, J., D. Fochtman, D. Holmes, J. McCarthy, Ball Corporation,
Broomfield, CO, to T. Grandinetti, and B. Warwick, Ball Corporation. Preliminary Cost
Comparison UV-Curable Inks/Coatings vs. Conventional Thermal Cure Inks/Coatings.
December 22, 1987.
11. Telecon. Robert Zilke, AKZO Corporation, with Steven R. Church, TRC Environmental
Corporation, Chapel Hill, NC. Discussion of coating prices. December 6, 1993.
12. Telecon. Timothy D. Case, Ball Corporation, Broomfield, CO, with Steven R. Church,
TRC Environmental Corporation, Chapel Hill, NC. Discussion of waterbased inks.
December 7, 1993.
3-24
-------
13. Telecon. Robert Allara, AKZO Corporation, with Steven R. Church, TRC Environmental
Corporation, Chapel Hill, NC. Discussion of UV-curable inks. December 7, 1993.
14. Memorandum. Joette Bailey, Ball Corporation, Broomfield, CO, to Beth McMinn, TRC
Environmental Corporation, Chapel Hill, NC. Comments on Draft Evaluation of Barriers
to the Use of Radiation-cured Coatings in Can Manufacturing. March 3, 1994.
15. Memorandum. David Harbourne, Fusion Curing Systems, Rockville, MD, to Steven R.
Church, TRC Environmental Corporation, Chapel Hill, NC. October 6, 1993.
16. Oven Systems, Inc., Milwaukee, WI, Price Quote. March 5, 1993.
3-25
-------
CHAPTER 4
TECHNICAL BARRIERS TO THE EXTENDED USE OF UV-CURING
TECHNOLOGY
4.1 GENERAL
This section identifies the technical barrieis to the extended use of UV-curing technology
in two-piece can manufacturing. Subsequent chapters will discuss economic and educational
barriers. Although each barrier category is explained separately, they are inter-related. The
economic barriers, for example, often result from technical obstacles that must be overcome if
the UV technology is going to be a viable option for can manufacturers. Furthermore,
educational barriers are often perceived technical barriers that have been resolved in recent years.
There are a number of technical barriers that have prevented the more widespread use of
UV-curing technology in two-piece can manufacturing. Most of the problems involve the
chemistry of the coatings and their ability to meet the product standards desired by the can
manufacturer and the customer. Much of the information in this section is based on Ball
Corporation's experience with UV-curable coatings. Table 4-1 lists common product standards
that can manufacturers use to evaluate a coating's performance.
4.2 PRODUCT PERFORMANCE
In 1986 and 1987 Ball Corporation operated an experimental UV-curable coating line at
their Findlay, Ohio facility. During their trial runs, Ball encountered several product performance
difficulties associated with the film thickness of the UV-curable inks and overvamishes. Ball was
not able to achieve the necessary product standards from UV-curable coatings at the desired,
lower-wet film thickness of 100 mg/can. Instead, Ball had to use a higher-wet film thickness
(150 to 200 mg and higher) to meet its standards. The following paragraphs explain which
standards were not met by the UV-curable coatings at the 100 to 120 mg/can film weights.
During the trial runs, Ball tested the labels of Classic Coke, Diet Coke, and Pepsi at line speeds
of 1,200 to 1,400 cpm.1
4-1
-------
TABLE 4-1. PRODUCT STANDARDS USED BY CAN MANUFACTURERS FOR
COATINGS EVALUATION
Product Standard
Explanation
Adhesion
Spin neck
Viscosity
Lay down
Cure speed
Abrasion resistance
Cure window
Pasteurization
Coefficient of friction (CoF)
Color
Tack
How well coating adheres to surface of the cans
How well coating withstands downsizing of cans' diameter at
open end
Coating's resistance to flow
How well an ink applies to a can
Speed at which the line must run to cure the coating
How well coating withstands abrasion
Time and temperature variances in which a coating can be
cured
How well coating withstands pasteurization process
Amount of friction between cans after curing
Clarity of colors
Surface tension of coating on can
Source: Reference 1
Ink lay down. In Ball's experience, several of the UV-curable inks did not remain properly on
the substrate of the cans. At film weights below 150 mg/can, inks would "pick off (i.e., be
removed with subsequent coating applications) the cans when the overvarnish was applied.2,3 The
red inks were more prone to pick off than the other inks. Once the red pigment mixed with the
overvarnish on the rubber roller, it created a pinkish tint that was applied to subsequent cans.2
Cure speed. In several of Ball's trial runs, the cure time of the UV-curable overvarnishes was
over one second. Some cure times were as high as five and ten seconds.2 This caused mobility
problems along the line because the cans would leave the oven only partially cured. The cans
would then deposit part of their coating along rails, belts, and other parts of the line, hampering
the mobility of other cans.
4-2
-------
Abrasion resistance. Abrasion resistance is important to protect the label of the cans during the
decorating process and later during transport to the consumer outlets. Abrasion is tested by
placing a six-pack of beer or beverage cans in an abrasion machine. The machine vibrates,
simulating the motion cans might experience during transport and distribution. After vibrating
for a designated period of time, the cans are removed and their surfaces inspected for nicks,
scratches, or other coating voids. Ball found that the UV-curable overvarnish did not provide
adequate abrasion resistance during several of the trial runs.1
Pasteurization. Pasteurization occurs after the filling process when cans are heated to a specified
temperature, normally 150°F (66°C), for a specified length of time.2 Ball found that some of the
coatings developed water spots after 5 to 10 minutes of pasteurization, while coatings in other
trial runs experienced no problems.1 Pasteurization is important for some beers but not for soft
drinks.
Coefficient of friction (CoF). CoF is the amount of friction between cans after they have been
cured in the oven. Because cans left the UV-curing oven only partially cured during several of
Ball's trial runs, they would create a high degree of friction among themselves, hampering their
mobility.2
Color. Ball experienced two color difficulties. First, the overvarnish "yellowed" on some of the
trial runs, damaging the clarity of the labels. Second, when cured, some of the white inks did
not achieve the desired shade.3 White is a difficult color to cure by UV light because it reflects
rather than absorbs light.
The problems of ink pick off, poor abrasion resistance, yellowing overvarnish, and off-
shade whites are significant barriers that will have to be overcome for UV technology to succeed
in a merchant can plant The ink pick off problem occurs during the "wet-on-wet" application
of the overvarnish to the inks and may be a physical problem involving the method of coating
application. The other three problems - abrasion resistance, yellowing, and off-shade whites -
are related to coating chemistry. Although Coors does not experience difficulties with abrasion
resistance and believes that its UV-curable coatings achieve a level of abrasion resistance high
4-3
-------
enough to meet the standards of other can manufacturers, Ball has had difficulties achieving the
desired level of abrasion resistance. -
Another product performance barrier is the lack of an FDA-approved UV-curable coating
for direct contact with food, including beer and beverages. The FDA has expressed concern
about potential carcinogens in the acrylic compounds used in some UV-curable coatings." During
the trial runs, Ball claimed that components of the exterior UV-curable coatings migrated to the
interior of the cans after the cans exited the UV-curing oven. The curing process creates excited
coating particles which "jump" from the exterior of the can to the interior of the can. This
migration typically occurs only at the top pan of the can. In addition, electrical surges or
disruptions in the decorating process (e.g., coating operation) can cause fluctuations in the coating
spray pattern resulting in the application of ink and varnish to the interior of the cans.2
Coors has also studied the issue of migration and has determined that none occurs. Coors
conducted three months of analyses using Fourier Transform Infrared (FTTR) technology and gas
chromatography/infrared detection (GC/IRD) instruments and found no volatile components from
exterior inks or coatings on the internal coating of its cans at detection levels between 10 and 100
parts per billion.5 Coors concluded that neither photoinitiators nor acrylates migrate to the
internal coating of its cans during production. The tests were conducted before the cans reached
the IC oven.5 With waterborne coatings, migration may not be a problem because the migrated
elements from the overcoat cure from the heat of the IC oven. However, tests have shown a loss
of coating weight in the internal coating ovens which may indicate that waterbased external
coatings have a migration problem as well.6
The lack of FDA approval for UV-curable coatings presents another hurdle for the
technology. Many can manufacturers acknowledge that emissions from the internal coating
process in their plants are greater than emissions from the external coating process. Some claim
that the ratio of emissions from internal to external coatings is three to one in a typical two-piece
can plant.7 As a result, manufacturers view improvements to waterborne coatings as the best way
to reduce emissions. Many do not want to invest in UV-curing equipment and make the
necessary process changes if the change will not significantly reduce emissions.
4-4
-------
4.3 EQUIPMENT
Curing equipment difficulties also impact the desired product performance levels. During
many of Ball's trial runs, the UY-curing oven did not adequately cure the coatings.2 The oven,
which is designed to cure cans in less than a second, required more than one second and, in some
instances, up to 10 seconds for curing.2 Ball tried to correct the problem by operating the lamps
at their high setting, which significantly reduced the energy savings that the company hoped to
achieve with the UV system. Ball used Electrode Arc Lamps UV XL, which have a medium and
a high setting, and are guaranteed for 1,000 hours.2 The company claims that at today's cure
speeds in excess of 2,000 cpm, it would need a significantly larger UV-curing oven than the one
used during the trial runs to adequately cure its cans.
The problems Ball encountered with the UV-curing oven are inter-related with cure speed
and CoF difficulties. The UV-curing oven could not adequately cure the UV-curable overvarnish
in less than a second, hence cans left the oven partially cured. Ball claims that the UV-curable
coatings often needed a "thermal bump" from the heat of the IC oven to completely cure.2,3 One
explanation for this is that the film weights of the coatings (often 150 to 200 mg/can) were too
thick to achieve proper cure.
4.4 HEALTH AND SAFETY
Another issue that frequently arises when can manufacturers consider the UV-curing
process is the disposal of waste inks and coatings. During their trial runs, Ball treated waste UV-
curable inks and coatings as hazardous. The hazardous classification is a regulatory and financial
burden because of the disposal requirements and liability provisions under the Resource
Conservation and Recovery Act (RCRA). In the 1970s and early 1980s, Coors treated its UV-
curable waste inks and coatings as hazardous because they had a higher solvent content.
However, the company recently performed TCLP and EP toxicity analyses on its waste inks and
coatings and determined that they are not hazardous. Therefore, Coors now disposes of waste
inks and coatings as non-hazardous waste.8
During their trial runs, Ball identified the odor of UV-curable inks and coatings as a
worker-safety problem. The odor seemed most prevalent around the printer.2 Coors has not
4-5
-------
reported a similar problem. Taking extra precautions to cover unused coatings and inks may help
reduce odors. However, Ball's experience indicates that most of the odor were generated by the
open application of varnish on the decorator.
Through recent advances in monomer chemistry, the health and safety hazards associated
with radiation-curable coatings have decreased significantly. The monomers have a high
molecular weight which reduces the volatility and removes almost any danger associated with
vapor inhalation. There continues to be concerns with skin becoming sensitized when in direct
contact with the radiation-curable coatings, and workers are still required to wear appropriate
personal protective equipment when handling these materials.9
4.5 REFERENCES
1. Memorandum. Joette Bailey, Ball Corporation, Broomfield, CO, to Steven R. Church,
TRC Environmental Corporation, Chapel Hill, NC. August 24, 1993.
2. Telecon. Tony Grandiotti and John Burnett, Ball Corporation, Broomfield, CO, with
Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of can
coatings. November 4, 1993.
3. Memorandum. Joette Bailey, Ball Corporation, Broomfield, CO, to Steven R. Church,
TRC Environmental Corporation, Chapel Hill, NC. November 2, 1993.
4. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
visit - Coors Container Complex, Golden, CO. August 16, 1993.
5. Crabtree, Terry A. UV Curing of Two-piece Cans: An Update. Fusion UV Curing
Systems. Rockville, MD. 1989.
6. Memorandum. Joette Bailey, Ball Coiporation, Broomfield, CO to Beth McMinn, TRC
Environmental Corporation, Chapel Hill, NC. Comments on Draft Evaluation of Barriers
to the Use of Radiation-Cured Coatings in Can Manufacturing. March 3, 1994.
7. Geer, Robert. American National Can Company. Discussion on UV-curable coatings at
quarterly meeting of environmental managers at Can Manufactures Institute. September
23, 1993.
8. Telecon. Erik Donhowe, Coors Brewing Company, Golden, CO, with Steven R. Church,
TRC Environmental Corporation, Chapel Hill, NC. Discussion of waste from UV-curable
coating process. December 9, 1993.
4-6
-------
9. Memorandum. Ross, Alexander, Rad Tech International North America, Falls Church,
VA to Beth McMinn, TRC Environmental Corporation, Chapel Hill, NC. Comments on
Draft Evaluation of Barriers to the Use of Radiation-Cured Coatings in Can
Manufacturing. March 4, 1994.
4-7
-------
CHAPTER 5
ECONOMIC BARRIERS TO THE EXTENDED USE OF UV-CURING
TECHNOLOGY
5.1 GENERAL
This section identifies the economic barriers to the extended use of UV-curing technology
in two-piece can manufacturing. Although each barriers category is explained separately, they
are inter-related. The economic barriers, for example, often result from technical obstacles that
must be overcome if the UV technology is going to be a viable option for can manufacturers.
5.2 CAPITAL INVESTMENT
The most common deterrent that can manufacturers face when considering a UV-curing
system is the capital and material costs. With their plants operating at low profit margins and
high volume, manufacturers view a change in technology as an expensive proposition. A plant
is unlikely to consider purchasing a UV-curing oven for $195,000 to $200,000 unless it needs
to replace the oven it is currently using.1 However, a plant may consider a UV-curing oven when
adding additional capacity because the thermal system is a larger capital investment ($390,000).2
5.3 PRICING PRESSURE
As previously mentioned, the can manufacturing business is very competitive because it
operates on low margins and high volume. Companies have to produce a significant quantity of
cans to be profitable. The industry is constantly pressured to reduce costs to prevent further
deterioration of margins and to minimize further capitalization in a highly capitalized industry.3
For a company to increase its profitability, it must increase the volume of cans that it produces.4
Plants operating 24 hours per day depend on faster line speeds to increase volume. Therefore,
line speed is a critical element that can manufacturers consider when evaluating new
technologies.5 Profitable companies limit the amount of downtime on their lines. Unexpected
interruptions to a line can be expensive in terms of lost production and material waste.
5-1
-------
In order to more fully understand the pricing pressures experienced by the can
manufacturing industry, it is important to understand the industry organization. There are two
types of manufacturers of two-piece beer and beverage cans: merchant and captive. Merchant
manufacturers serve more than one customer. American National Can Company is the largest
merchant manufacturer in the United States.6 It has several plants located around the United
States to serve the needs of beer and beverage companies in different regions. Captive
manufacturers are owned by a beer or beverage company and produce cans solely for the
company who owns them. The Coors Container Complex in Golden, Colorado, the largest can
plant in the world, is a captive manufacturer that has, at some point, supplied cans to each of
Coors' three breweries located in Elkton, Virginia; Memphis, Tennessee; and Golden, Colorado.
Beer and beverage companies own can manufacturing facilities to increase their leverage
with merchant suppliers. For example, a beverage company will designate its captive facility to
supply a certain percentage of the company's demand for cans. Beer and beverage companies
prefer to keep their can facilities operating at 100 percent capacity, which is most profitable. The
remainder of the company's demand will be served by merchant suppliers, who are forced to
absorb the fluctuations in demand for the beverage company's product. In 1992, merchant
manufacturers supplied 80 percent of the beer and beverage companies' demand for cans, while
captive manufacturers supplied the remaining 20 percent.4
5.4 MATERIAL AND OPERATING COSTS
Although the price of a UV-curing oven compares favorably to that of a thermal oven,
the higher UV-coating costs are often enough to deter can manufacturers from considering the
UV-curing alternative.2 Section 3.3.2 contains cost estimates for UV-curable and waterbome
materials. Coors' experience with UV-curable inks and coatings indicates that they are
approximately five percent more expensive than waterbome coatings on an applied basis.7 For
Coors, the energy savings from their UV-curing oven has compensated for the higher price of
the UV-curable materials.
Ball Corporation had a different experience with UV-curable coatings during its trial runs
at the Findlay, Ohio plant during 1986 and 1987. Table 3-12 listed Ball's material costs and
Table 3-18 summarized their operating costs. During their trial runs, Ball's goal was to reduce
5-2
-------
the wet film thickness of its UV-curable overvamish to 100 mg per can.8 This goal, however,
was never achieved. In the majority of their trial runs, Ball had to use a minimum wet film
thickness of 200 mg per can to achieve the desired product qualities after cure.2 Because of
increased material consumption and, therefore, increased costs, the UV-curing alternative was not
economically feasible for Ball. Even using the projected consumption for the lower film
thickness, Ball's energy savings, approximately eight percent, did not make up the difference in
the higher material costs of the UV system. Ball's projected UV-curable material costs were 49
percent higher than its waterbome materials.
Availability of UV-curable materials and equipment is another cost barrier to more
widespread use of UV systems. Coatings and ink suppliers consider UV-curable products a
"specialty" line with a limited market9 Coors believes that if other can manufacturers used UV-
curable inks and coatings, competition would increase among vendors and material costs would
fall. The company estimates that a 90 percent penetration of UV-curing technology into the can
manufacturing market would reduce the cost of UV-curable inks and overvamish by 10 percent.10
Several manufacturers of coatings and equipment claim to include UV-curable products as part
of their product line. At an industry trade show in 1992, nine of the 180 participants (five
percent) offered UV-curing systems." The best known vendor of UV-curing systems for two-
piece cans is Fusion UV Curing Systems of Rockville, Maryland, which supplies UV-curing
ovens to Coors. AKZO Coatings and PPG Industries supply the UV-curable overvamish to
Coors, and Martinez Ink Company supplies the UV-curable inks.9
5.5 REFERENCES
1. Memorandum. David Harbourne, Fusion Curing Systems, Rockville, MD to Steven R.
Church, TRC Environmental Corporation, Chapel Hill, NC. October 6, 1993.
2. Telecon. Tony Grandiotti and John Burnett, Ball Corporation, Broomfield, CO, and
Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of
Ball's UV trial lines at the Findlay, OH plant. November 4, 1993.
3. Memorandum. Joette Bailey, Ball Corporation, Broomfield, CO to Beth McMinn, TRC
Environmental Corporation, Chapel Hill, NC. Comments on Draft Evaluation of Barriers
to the Use of Radiation-cured Coatings in Can Manufacturing. March 3, 1994.
5-3
-------
4. Telecon. Robert Graham, Miller Brewing Company, Reidsville, NC, with Steven R.
Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of the economics
of can manufacturing. August 5, 1993.
5. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
visit - Ball Can Manufacturing Plant, Williamsburg, VA. October 21, 1993.
6. Gale Research, Inc. Ward's Business Directory of U.S. Private and Public Companies,
Volume 5. Detroit, MI. 1992.
7. Donhowe, Erik T., Coors Brewing Company, "UV Pollution Prevention Technology in
Can Manufacturing," In Proceedings: Pollution Prevention Conference on Low- and No-
VOC Coating Technologies, EPA-600/R-94-022 (NTIS PB94-152246), pp. 475-487.
8. Memorandum. J. McCarthy, Ball Corporation, Broomfield, CO to G. Richardson, Ball
Corporation, Broomfield, CO. Cost Analysis of UV Curing. July 15, 1988.
9. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
visit - Coors Container Complex, Golden, CO. August 16, 1993.
10. "UV-Curable Coatings For Aluminum Can Production," Volume II, Coors Brewing
Company, Golden, CO. April 30, 1993
11. Editors, "Where to Buy Canmaking Equipment, Materials, and Services," Modern Metals.
48(5):44DD-UU, June 1992.
5-4
-------
CHAPTER 6
EDUCATIONAL BARRIERS TO THE EXTENDED USE OF UV-CURING
TECHNOLOGY
6.1 GENERAL
It is generally well-known that educated employees (both management and production
personnel) are more productive, efficient, resourceful, and economically aware than are
employees who are less informed. In addition to providing specific instructions for use of
chemicals or operation of equipment, facility training programs can promote health and safety
procedures, waste reduction methods, and ingrain pollution prevention attitudes in the everyday
activities of plant personnel.
This chapter identifies some of the educational barriers to the extended use of UV-curing
technology in two-piece can manufacturing. The spread of UV-curing technology within the two-
piece can manufacturing industry has been hampered by negative perceptions, lack of
information, and education. When can manufacturers first considered the UV-curing technology
during the 1970s, they discovered several flaws. Since then, many of the difficulties have been
overcome, yet the negative perceptions linger within the industry.
6.2 OPERATOR TRAINING
One of the most pressing educational barriers involves operator perception of the word
"radiation." Although radiation-curing (e.g., UV-curing) equipment manufacturers have designed
their ovens to harness the radiant energy used to cure the coatings, many people are still hesitant
about using such equipment. Facilities and equipment manufacturers must recognize the
importance of proper training and radiation safety.
Another perceived problem is worker sensitization to acrylate compounds in the inks and
coatings. Although skin sensitization was an issue in the 1970s, coating formulations have been
improved to reduce sensitization. Coors has also taken extra precautions to improve their
workers' understanding of UV-curable inks and coatings. The company color-coded all of its
UV-curable materials to facilitate chemical recognition.1 The company also increased the
6-1
-------
frequency with which workers applied barrier creams for dermal protection and required all
workers to wear gloves compatible with UV-curable materials.1 Like workers in other can plants,
Coors employees are receptive to improved hygiene because they are in the food packaging
business.1 Coors has demonstrated that sensitization is no longer a problem if the proper
precautions are taken. However, the perception lingers among some companies that worker
sensitization is a problem with UV-curable coatings.
6.3 MANAGEMENT AWARENESS
As previously mentioned, past difficulties with UV-curable materials still influence current
industry perceptions regarding UV-curing technologies. In order to further the use and
implementation of UV-curing technologies, it is important that industrial managers are aware of
the current state of the technology. One example of a previous difficulty that has since been
overcome involves oxygen inhibition. In the past, some acrylic compounds in UV-curable
coatings lost their reactivity when exposed to air. Molecular oxygen would react with free
radicals in the coating, forming a peroxide.2 The reactions would deplete the number of free
radicals available for polymerization during the curing process.2 With earlier generations of UV-
curable coatings, measures had to be taken to eliminate the coatings' exposure to oxygen before
curing. Even though Coors still uses acrylic-based coatings, it has resolved the oxygen inhibition
problem through improved formulations from its coating suppliers.1 However, it is still the
perception of the industry that oxygen inhibition remains a problem.
Another example of the need for management awareness is in the area of cost Many
industry personnel believe that UV systems are not cost effective. Although the capital expense
for a UV-curing oven is substantial, it is no more than the investment in a new thermal line.
Studies also indicate UV-curing oven maintenance and energy costs to be less than those for
thermal systems. One recognized cost increase is the expense of coating raw materials. This
cost, however, may be offset by savings in emissions fees resulting from fewer releases of VOCs
and HAPs.
6-2
-------
6.4 REGULATORY PRESSURE
A significant barrier to the expanded use of UV-cuiable coatings is the current regulatory
environment. Regulators should be made aware of the effect that current requirements have on
expanding technology. Most components of UV-curable materials fall under the jurisdiction of
the Toxic Substance Control Act (TSCA), which requires manufacturers of new chemical
substances to submit Pre-Manufacturing Notices (PMN) 90 days before commercial production
of the substance is to begin. In the 1980s, the EPA used Section 5(e) of TSCA to ban or place
limits on many acrylic-based compounds to be used in UV-curable materials.2 As a result,
research into many new UV-curable products stopped.3 Since then, the pace of research into new
chemical compounds for UV-curable coatings has resumed. However, low-molecular-weight
acrylic compounds, which constitute many of the new chemistries for UV-curable products, are
subject to Significant New Use Restrictions, which require additional paperwork and safety
precautions for researchers.3
6.5 REFERENCES
1. Memorandum. Steven R. Church, TRC Environmental Corporation, to Carlos Nunez,
U.S. Environmental Protection Agency. Site Visit - Coors Container Complex, Golden,
CO. August 16, 1993.
2. Walata, S.A. Ill and C.R. Newman. Radiation-Curable Coatings. EPA-600/2-91-035.
(NTIS PB91-219550). Control Technology Center, Research Triangle Park, NC. July
1991.
3. Mullin, Rick. "Spotlight On Radiation Curing," Chemical Week. 151(5), pp. 22-26.
August 5, 1992.
6-3
-------
CHAPTER 7
OPPORTUNITIES TO OVERCOME IDENTIFIED BARRIERS
7.1 GENERAL
UV-curing technology has the potential to be used more widely in two-piece can
manufacturing. The technology has worked successfully in a captive can plant. The Coors Brewing
Company has competed in a cost effective manner with other can manufacturers while using the
technology on its high volume lines at the Coors Container Complex in Golden, Colorado. Coors
has resolved many of the technical problems that UV-curable coatings had when can manufacturers
first considered the technology in the 1970s. A remaining question is whether or not the technology
can succeed in a merchant can plant where the label requirements are more varied. The following
paragraphs offer research opportunities to pursue in evaluating the uncertainties surrounding UV
technology in two-piece can manufacturing. Opportunities are classified into the following areas:
product performance, coating migration, FDA approval, cationic coatings, white basecoats, and
three-piece can manufacturing.
7.2 PRODUCT PERFORMANCE IMPROVEMENTS
The trial runs conducted at Ball's Findlay, Ohio plant indicated that UV-curable coatings had
several technical shortcomings that would have to be overcome if the technology were to succeed
in a merchant can plant. The following were primary problems:
• Wet film thickness - Film weights of 100 to 120 mg/can that meet product standards need
to be achieved to make UV-curable coatings economical for manufacturers.
• Ink lay down - Inks must avoid "pick off" during the application of ovcrvarnish.
• Colors - Desired shades of white must be attained and yellowing of ovcrvarnish eliminated.
• Abrasion resistance - Cured overvarnish must be resistant enough to withstand normal
abrasion from the manufacturing process and during transport.
7-1
-------
• Partial cure - Cans leaving the oven must be completely cured to improve mobility on UV-
curing lines and prevent coating migration.
One opportunity to investigate these difficulties would be if can manufacturers (merchant and
captive), UV-curable coating manufacturers, and UV-curing equipment vendors were to set up a trial
line. An existing line in a can plant could be converted to a UV system if an outside party were
willing to co-fund and participate in the conversion.
7.3 MIGRATION OF UV-CURABLE COATINGS
One concern of the industry is that photoinitiators and acrylatc compounds from UV-curable
coatings migrate to the interior surface of cans before and during the curing of the internal coating.
Ball witnessed this occurrence during its trial runs and stated that this phenomenon is not uncommon
on a can line.1 Can manufacturers express concern over the possibility of UV-curable chemicals
migrating because they could affect the flavoring of the product and they have not been approved
by the FDA for direct contact with food. Coors has conducted a study indicating that no migration
occurs at detection levels between 10 and 100 parts per billion. A third party could conduct a study
of the issue. Such a study would contribute to the industry's understanding of UV-curable coatings.
7.4 FOOD AND DRUG ADMINISTRATION APPROVAL
Can manufacturers are in the food packaging business. The coatings they use on the interior
of their cans come into direct contact with a food product, and, therefore, must meet FDA standards.
Developing a UV-curable coating that satisfies FDA requirements for direct contact with food would
remove a significant barrier to more extensive use of UV-curing technology. Can manufacturers
prefer to use coatings of the same family. An FDA-approved UV-curable coating would eliminate
two major sources of emissions in a can plant, the internal coater and IC oven, while providing
incentive for manufacturers to convert their entire line to a UV system.
7-2
-------
7.5 CATIONIC COATINGS
Cationie coatings arc a new generation of UV-curable coatings that offer promise to
overcoming some of the technical problems with existing acrylic based coatings. The curing process
for these coatings involves irradiating onium salts and other light-activated compounds with UV
light of the proper wavelength.2 Once the curing is initiated, it continues after the exposure to IJV
light ends. This "dark curing" allows for a strong polymerization to take place, which improves the
coating's abrasion resistance.2 Coors is testing a cationic-based overvarnish on its 16 ounce can line. 3
This line can accommodate the slower cure rate of the cationie coating because it only produces 650
cans per minute.3 The company is working with AKZO Coatings to reduce the cure time of the
cationie coatings to less than 0.7 seconds.3
Cationie chemistry is an important area of research for UV-curable coating technology.
Besides the improved abrasion resistance and "dark curing," cationie coatings offer the benefit of
"shadow curing."2 Shadow curing occurs on areas of a substrate that were never directly exposed
to the UV light. The polymerization process is initiated in other areas of the substrate and passes
into the unexposed areas.2 This technology expands the potential applications of UV-curable
coatings in two-piece can manufacturing.
7.6 DEVELOPMENT OF UV-CURABLE WHITE BASECOAT
The largest buyer of beer cans in the United States is Anheuser-Busch Corporation. In 1991,
the company bought 44 percent of all manufactured beer cans.4 Most merchant manufacturers are
interested in selling cans to Anheuser-Busch because of its size. The company uses a white basecoat
on several of its labels. Critics of UV-curable technology claim that a commercially viable UV-
curable white basecoat does not exist. White reflects light, making it difficult to cure. Proponents
of the technology claim that UV-curable white basecoats not only exist but work.3 However, they
have not been used on a high volume line. A third party could assist the industry by working with
manufacturers to develop and test a UV-curable white basecoat.
7-3
-------
7.7
USE OF UV-CURABLE COATINGS IN THREE-PIECE CAN MANUFACTURING
IJV-curable coating systems have found limited use in three-piece can manufacturing. In the
United States, there are approximately 10 to 12 commercial lines using a UV-curable system to coat
can ends and 3 to 4 commercial lines using UV-curable overvamish on three-piece can bodies (e.g,
for juice and aerosol cans.)5 The UV-curing process is essentially the same as a conventional
thermal process for three-piece manufacturing except that a UV system applies UV-curable
overvamish with photoinitiators to uncoated tin-free steel and cures it in a UV-curing oven rather
than a thermal oven. A thermal oven is still used to cure the waterborne coating used for the interior
surface of the cans. Three-piece facilities using a UV system were not visited during the course of
the project. However, the three-piece can manufacturing industry offers a growing market segment
for UV-curable coatings, and would be worth further investigation.
7.8 REFERENCES
1. Telecon. Tony Grandiotti and John Burnett, Ball Corporation, Broomfield, CO. with Steven
R. Church, TRC Environmental Corporation, Chapel Hill, NC. Discussion of can coatings.
November 4, 1993.
2. Koleski, Joseph V. Cationic Radiation Curing. Federation Series on Coatings
Technologies. Federation of Societies for Coatings Technologies. June 1991.
3. Memorandum. Steven R. Church, TRC Environmental Corporation, Chapel Hill, NC, to
Carlos Nunez, U.S. Environmental Protection Agency, Research Triangle Park, NC. Site
Visit - Coors Container Complex, Golden, CO. August 16, 1993.
4. Darnay, Arsen J. and Marlita A. Reddy, The Market Share Reporter 1992, second edition,
Gale Research Inc., Detroit, MI, pg. 227.
5. Telecon. Robert Zilke, AKZO Corporation, and Steven R. Church, TRC Environmental
Corporation, Chapel Hill, NC. Discussion of uses of UV-curing technology in three-piece
can manufacturing. November 17,1993.
7-4
-------
APPENDIX A
PRELIMINARY MARKET ANALYSIS
A-l
-------
METAL CANS (SIC 3411)
1. INDUSTRY DESCRIPTION
Metal cans are used primarily for the packaging and distribution of food, beer and soft
drinks. The industry is a mature industry that has become concentrated among a few national
companies. During the 1980's, the industry experienced strong growth as food and beverage
sales climbed. Metal cans enjoyed a resurgence in popularity as consumers became concerned
about recycling. Since the late 1980's, the market has become saturated, and the competition is
fierce among the national producers as they try to consolidate their market shares.
The market's sales are concentrated among three companies. Each of these companies
reported sales over $2 billion in 1992, and together they comprised more than 73 percent of the
market (see Table A-1).1 The remainder of the market is spread among smaller companies with
a national presence or among regional companies that serve a particular niche. None of the other
companies reported sales greater than $400 million (see Table A-2).
TABLE A-l. FIVE LARGEST METAL CAN COMPANIES BY SALES (SIC 3411)
Company
Sales ($, millions)
1. American National Can Company
4,500
2. Crown Cork and Seal, Inc.
3,807
3. Ball Corporation
2,267
4. Inter-American Packaging, Inc. (U.S. Can Company)
400
5. Reynolds Metals Company (Can Division)
390
Total Sales Of Top Five
11,364
Total Industry Sales
14,392
Percentage Of Total Industry Sales
78.96
Source: Ward's Business Directory of U.S. Private and Public Companies 1993, Volume 5
A-2
-------
TABLE A-2. NUMBER OF COMPANIES BY SALES (SIC 3411)
Sales ($, millions)
Number of Companies
> 1,000
3
1,000- 500
0
500- 100
10
100-50
9
1 -50
36
< 1
3
Total Number of Companies
61
Source: Ward's Business Directory of U.S. Private and Public Companies 1993, Volume 5
There are two types of metal cans: three piece cans and two piece cans. Three piece cans
are made from a rectangular sheet and two circular ends. Two piece cans are made from a
rectangular or circular sheet and one end. An end is attached after the can is filled. In the
United States, most companies manufacture two piece cans for beverage containers.2
The most common method for producing a two piece can is the draw-and-iion method.
Aluminum or tin-plated steel is fed to a press in coil form. The press makes cups out of the
plate and then feeds them into other presses where they are pushed through a series of
progressively smaller die rings. The rings thin and lengthen the cups' walls. After the last ring,
the cups reach a bottom punch which creates the desired profile. Circular knives then trim the
cups to their desired height.2
Once the cans are formed, they are washed and dried in preparation for coating and
printing. The exterior of each can is coated with a white primer before the inks are applied. The
cans also receive a protective interior coating, which is normally a waterborne coating.
The ends of the cans are produced by feeding a metal sheet into presses where the ends
are cut and shaped. The ends are then passed between a wheel and segmented rails. The rails
roll the edge of the ends. Sealing material is then applied between the end profile and the curl.
When the can body and the curl are attached, the two pieces are compressed and ironed.2
A-3
-------
2. INDUSTRY ECONOMICS
Although sales in the industry are concentrated among a handful of national companies,
there are 369 can manufacturing facilities spread throughout the country (see Table A-3). Most
employees work at a medium-sized facility with 100 to 499 employees.3 Because the
manufacturing process is highly automated and the end product is small, a few hundred
employees can produce a substantial number of cans.
TABLE A-3. NUMBER OF FACILITIES AND EMPLOYEES
Range of Employees
Number of Facilities
Number of Employees
1 - 49
164
2,600
50-99
53
3,900
100 - 499
146
28,900
500 - 999
5
4,000
1,000 - 2,499
1
W*
Total
369
39,400
W = Information withheld by Census Bureau
Source: 1987 Census of Manufactures
During recent years, the cost of producing metal cans steadily increased (see Table A-4).
From 1987 to 1991, the cost of materials rose by nearly 25 percent.5,4 To counter the increased
material costs, the industry held down manufacturing costs (i.e., labor) as the value added by
manufactures decreased by nearly 7 percent during the same period. The reduction in
manufacturing costs, however, was not enough to keep pace with the increased material costs.
Over the same period, the value of the industry's shipments rose by 13 percent, below the 25
percent increase for material costs. To remain competitive, can companies were forced to reduce
their profit margins.
A-4
-------
TABLE A-4. SHIPMENTS (SIC 3411)
Year
Cost Of Materials
(5, Millions)
Value Added By Manufacture
($, Millions)
Value of Shipments
($, Millions)
1987
7,194.6
3.816.0
11,013.6
1988
7.492.2
3,920.3
11,407.1
1989
7,985.9
3,418.5
11,389.3
1990
8,676.4
3,668.4
12,342.4
1991
8.977.9
3,557.3
12,449.6
Source: Census of Manufactures 1987 and Annual Surveys of Mamifcicturing 1988 - 1991
In their efforts to remain competitive in an increasingly saturated market, metal can
manufacturers reduced employment. From 1987 to 1991, the total number of employees and the
number of production workers in the industry decreased by 12 and 10 percent, respectively (see
Table A-5)M
Compared to the workers in other industries, production workers in the metal cans
industry are productive. According to Manufacturing USA, in 1988 the average production
worker in the metal cans industry produced 1.6 times more shipments than the average worker
the other manufacturing industries.5 This is not surprising because the manufacture of metal cans
is a capital-intensive, highly automated process. The workers are well compensated for their
productivity. Wages for production workers in the metal cans industry were 1.54 times higher
than the national average for other manufacturing industries in 1988 according to Manufacturing
USA.5 The average salary for a metal can production worker grew by nearly 11 percent
(unadjusted) from 1987 to 1991 (see Table A-5).3,4
A-5
-------
TABLE A-5. EMPLOYMENT AND COMPENSATION (SIC 3411)
Year
All Employees
Production Worker
Number
(Thousands)
Payroll
(S, Millions)
Average
$ Salary
Per
Employee
(B/A)
Number
(Thousands)
Wages
($, Millions)
Average
S Salary
Per
Production
Worker
(E/D)
A
B
C
D
E
F
1987
39.4
1325.4
33,640
32.7
1,058.0
32355
1988
39.0
1,361.5
34,910
32.9
1,081.9
32,884
1989
36.9
1,342.1
36371
31.1
1,082.0
34,791
1990
35.9
1,319.4
36,752
30.5
1,077.2
35,318
1991
34.6
1,315.0
38.006
29.3
1,048.8
35,795
Source: Census of Manufacturts 1987 and Annual Surveys of Manufactures 1988 - 1991
The increased cost of producing cans has hurt the industry's profitability in recent years.
The profit earned on each dollar of sales in the industry decreased by 36 percent from 1989 to
1992 (see the return on sales ratio in Table A-6).6 The industry's return on assets experienced
an even steeper decline (42 percent). Despite the reduction in profits, the industry was able to
maintain its liquidity position near 2.0; however, in 1992 its liquidity slipped to a level that
would be considered unsafe by some financial analysts (see current ratio - the ration of current
assets to current liabilities, in Table A-6).
TABLE A-6. KEY INDUSTRY RATIOS
Year
Return On Sales
(Percentage)
Return On Assets
(Percentage)
Current Ratio
Assets To Sales
(Percentage)
1989
3.5
6.2
1.8
54.9
1990
2.7
5.0
1.8
52.8
1991
3.3
4.1
2.0
64.0
1992
2.6
3.6
1.7
57.6
Source: Duns Analytical Services. Industry Norms & Key Business Ratios (1989 - 1992)
A-6
-------
The profitability of the metal cans market depends upon the growth of the industry's
major market segments - soft drinks, beer and food (see Table A-7).7 Consumer demand for
these products increased steadily during the 1980's and has remained stable since then.
TABLE A-7. METAL CAN MARKET SHARES BY END USE SEGMENT (SIC 3411)
1990
End Use
Number Of Units (Billions)
Market Share Percentage
Food
53.26
42.2
Beer
39.25
31.1
Soft Drinks
29.66
23.5
General Packaging
4.04
3.2
Total
126.21
100
Source: Market Share Reporter. 1992
Within the beer and soft drink markets, Anheuser-Busch and Coke are the two largest
buyers of cans (see Tables A-8 and A-9).7 None of the other brewing companies significantly
rivals Anheuser-Busch's consumption of cans in the beer market. Pepsi is the only company to
nearly matches Coke's consumption in the soft drink market.
TABLE A-8. BUYERS OF BEER CANS
Company
Market Share Percentage
Anheuser-Busch
44
Miller
22
Coors
13
Stroh
11
Other
10
Source: Market Share Reporter. 1992
A-7
-------
TABLE A-9. BUYERS OF SOFT DRINK CANS
Company
Market Share Percentage
CCE (Coke)
20
COBO (Pepsi)
17
Consolidated Pkg. Group - Pepsi
10
Beverage Association
7
Other
46
Source: Market Share Reporter. 1992
The companies listed in Tables A-8 and A-9 not only buy cans from other companies but
also manufacture their own cans. At this time, it is uncertain what percentage of cans is
manufactured by metal can companies and what percentage is manufactured by beverage
companies.
Foreign markets are important to the industry. Eight percent of its shipments and
employment are tied to exports (see Table A-10).8 The largest producer of cans in the United
States, American National Can Company, is owned by a French company, Pechiney Incorporated,
which has a strong presence in the European market and other markets around the world.
TABLE A-10. METAL CONTAINER EXPORTS COMPARED TO TOTAL
INDUSTRY EXPORTS
Value of Exports as a Percentage of Total Industry Shipments
8
Export Manufacturing Employment as a Percentage of Total Manufacturing
Employment for Metal Container Industry
8
Source: National Trade Data Bank - The Export Connection 1987
Because of the highly automated manufacturing process and the dominance of a few
national companies, it is difficult for a new company to enter the metal cans market and compete
on the national level. According to Manufacturing USA, in 1988 metal can manufacturers held
investments in plant and equipment equal to 3.5 times the average manufacturing plant. The
level of new capital investment fluctuated for the metal cans industry from 1987 to 1991 due to
A-8
-------
decreased profitability.5 However, the level of investment remained between 3 and 5 percent of
the cost of materials and 7 and 11 percent of the value added by manufacture over that period.3,4
A-9
-------
3. PROCESS
3.1 Process Flow Description
The most common method to coat the exterior of two piece cans is reverse roll coating
of solvent-based materials.9 A roller applies a coating to the cans by spinning in the opposite
direction from which they are moving. Roller coating machines usually have one roller that runs
partially immersed in the coating and passes the coating to another parallel roller above it. The
second roller then applies the coating to the cans. A white coating is often applied first to the
cans. The coating is then cured at temperatures between 170° and 200° C (325° and 400° F).y
Once the white coating has been applied, inks are transferred to the cans as they rotate on a
mandrel. A protective varnish is then roll coated over the inks. To cure the coating, the cans
are passed through a single or multipass oven at temperatures between 180° and 200° C (350° and
400° F).10
After curing, the can interiors are spray coated. The most common method of spray-
coating two piece cans is air atomized sprays. Using compressed air, a gun forms the coating
into tiny droplets and propels them onto the interior surface of the cans. The bottom end of the
exterior of the cans may also be spray or roll coated.10 The coating process normally occurs in
a well ventilated enclosure that protects the surface of the cans from dirt.
3.2 Equipment
There are several pieces of equipment to coat cans. Common coating systems are
composed of coating/printing rollers, curing ovens, and ink monitor controls. A number of
American and foreign companies supply coating equipment. Table A-11 lists some of the
companies that supply a comprehensive line of coating and curing equipment.11
A-10
-------
TABLE A-ll. VENDORS OF COATING EQUIPMENT (SIC 3411)
Equipment Vendor
Bartell Machinery Systems Corporation
Rome, New York
Flynn Burner Corporation
New Rochelle, New York
LTG Technologies
Spartanburg, South Carolina
Moco Thermal Industries
Romulus, Michigan
Thermal Innovation Corporation
Manasquan, New Jersey
Source: Modern Metals, lime 1992
3.3 Raw Materials
The basic raw materials used in manufacturing metal cans include metals (steel and
aluminum), liquid coatings and inks. Table A-12 provides the quantities and the costs of the
industry's consumption of these raw materials in 1987.3 (These are the most recent consumption
data available from the Census Bureau.)
TABLE A-12, INDUSTRY RAW MATERIALS IN 1987
Material
Quantity
Delivered Cost ($, Millions)
Carbon steel (tons, thousands)
2,695.1
2,148.1
Aluminum (lbs., million)
2,170.4
2,400.6
Paints, varnishes, lacquers,
shellacs, japans, and enamels
(1,000 gallons)
31,284.2
199.4
Source: 1987 Census of Manufactures
A-ll
-------
4. EMISSIONS
The can coating process generates air emissions and liquid and solid waste streams. Air
emissions are influenced by the type of coating, the coated area, the thickness of the coat, and
the efficiency of the application.10 The primary sources of emissions are the coating area and the
curing ovens. Fugitive emissions also result when coatings are mixed and loaded into the
application device, during transport of coated parts from the spray booth to the oven, and during
post curing. The rate at which solvent vapors are emitted depends on the speed of the line, the
size of the cans and the type of coating used.1"
The liquid and solid wastes in the metal cans industry are similar to those in other
industries that use coatings. Most liquid waste streams are caused by spent cleaning solvents and
contaminated coating materials. Solid wastes include solvent-soaked rags used for cleaning
equipment and discarded packaging materials. The rags are likely to be classified as hazardous
wastes although some facilities allow their solvent-soaked rags to dry by evaporation before
discarding them with landfill wastes.
Table A-13 contains data from the Toxic Release Inventory (TRI).12 The data are based
on the information provided by 966 facilities. The number of facilities reporting to the TRI
database is significandy higher than the total number of can manufacturing facilities (369)
reported in the 1987 Census of Manufactures. The discrepancy could be the result of two
different definitions of a can manufacturing facility or the result of facilities reporting under
secondary or tertiary SICs. The 1987 Census lists facilities whose primary purpose is the
manufacture of cans. Although the definition of a can manufacturing facility for the TRI
database is not known at this time, it could be any facility that has the capacity to manufacture
cans even though this may not be the facility's primary activity.
A-12
-------
TABLE A-13. TRI DATABASE EMISSIONS AND WASTE STREAMS (SIC 3411)
Waste Type
1991 Releases (lbs)
Air Emissions
36,072,387
Waste Water
34,467
Solid Wastes
266
Source: Toxic Chemical Release Inventory Database. National Library of Medicine, U.S. Department of Health and Human Services
A-13
-------
5. CONCLUSION
The metal cans industry is a capital-intensive industry with highly automated manufacturing
processes. In recent years, the industry became very competitive as costs increased and profitability
decreased. To remain competitive in the market, companies constantly look for new ways to
improve the productivity of their manufacturing processes or reduce costs. For this reason, they are
receptive to new technologies and processes. Radiation-cured coatings have already found limited
use within the industry. One of the major buyers of cans, the Coors Brewing Company, has been
using U V-curcd inks and coatings on the exterior of its cans since the late 1980s. Other companies
are interested in following the Coors example. SIC 3411 has strong potential to provide valuable
research opportunities on radiation-cured coatings. The industry cannot be considered a good
industry for researching the future use of waterborne coatings because they are already used
extensively. Industry contacts indicated that nearly all beverage and food can interior and exterior
coating is done with waterborne coatings applications.
A-14
-------
6. REFERENCES
1. Gail Research Inc. Ward's Business Directory of U.S. Private and Public Companies.
Volume 5. 1993.
2. Editors, "How To Specify Metal Cans," Packaging, 35(9): 87-89, July 1990.
3. United States Department of Commerce, Bureau of the Census. Census of Manufactures,
1987. GPO 1988.
4. United States Department of Commerce, Bureau of the Census. Annual Surveys of
Manufacturing, 1988-1991. GPO.
5. Editorial Code and Data, Inc., Manufacturing USA: Industry Analyses, Statistics, and
Leading Companies, Gail Research Inc., 1992.
6. Duns Analytical Services, Industry Norms & Key Business Ratios, 1989-1992.
7. Damay, Arsen J. and Marlita A. Reddy, Market Share Reporter, second edition, Gale
Research Inc., Detroit, MI, pg.227, 1992.
8. National Trade Databank ¦ The Export Connection, 1987.
9. U.S. Environmental Protection Agency, Compilation of Air Pollutant Emission Factors.
AP-42, Volume I, 4th edition (GPO 055-000-00251-7) with Supplements. Office of Air
Quality Planning and Standards, Research Triangle Park, NC, September 1985.
10. Buonicore, Anthony J., and W.T. Davis, eds. Air Pollution Engineering Manual; Van
Nostrand Reinhold, 1992.
11. Editors, "Where to Buy Canmaking Equipment, Materials, and Services," Modem Metals,
48(5) :44DD-44UU, June 1992.
12. United States Department of Health and Human Services, National Library of Medicine,
Bethesda, MD. Toxic Chemicals Release Inventory Database, 1991. TOXNET.
A-15
-------
APPENDIX B
SITE VISIT REPORTS
Firm
Coors
Ball
Campbell
Lpcaljcp
Golden, CO
Williamsburg, VA
Maxton, NC
Dais
08/16/93
10/21/93
10/27/93
Page
B-2
B-19
B-27
B-l
-------
TRC
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hilt, NC 27514
•a {919) 968-9900 Fax (919) 968-7557
Environmental Solutions through Technology
November 4, 1993
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory
MD-61
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Prime Contract 68D20181
Coors Trip Report
TRC Environmental Reference Number 1645005
Dear Carlos:
Attached is the trip report from our visit to the Coors Container Complex on August
16. I apologize for the delay in sending it to you; however, we were waiting for the
emissions data from Coors. Please let me know if you have any questions or comments on
the report.
Sincerely,
A
Steven R. Church
Environmental Scientist
Offices in California, Colorado, Connecticut, Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texas,
Washington, Washington, D C., and Puerto Rico A TRC Compony
Prir.fc tn Recyc'cd Paper
B-2
-------
Date:
October 15, 1993
Subject:
Site Visit - Coors Container Complex
Manufacturer of Two-piece Aluminum Cans
EPA Contract 1-68-D2-0181, Work Assignment Number 1/005
TRC Reference Number 1645005
From:
Steven R. Church
TRC Environmental Company
To:
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory (MD-61)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
I. Purpose
As part of its emphasis on pollution prevention, the U.S. Environmental Protection
Agency (EPA) is identifying the barriers to the extended use of radiation-cured and water-based
coatings in Source Reduction Review Project (SRRP) categories and Maximum Achievable
Control Technology (MACT) categories. TRC Environmental Corporation (TRC) is supporting
EPA in this effort by evaluating the current use of these coatings in the metal can manufacturing
industry under Work Assignment Number 1/005, EPA Contract Number 1-68-D2-0181.
The primary source of emissions in metal can manufacturing plants is volatile organic
compounds (VOCs) used in the coatings of the cans. Coatings cured by ultraviolet (UV) light
(i.e., UV coatings) are considered a pollution prevention alternative for the industry because they
consist of nearly 100 percent solids which remain on the substrate during the curing process.
Few, if any, solvents are emitted by UV coatings.
The Coors can manufacturing plant in Golden, Colorado was selected for a site visit
because it has been coating the exterior of its two-piece aluminum cans with UV coatings since
1976. The purpose of the visit to Coors was to review their coating process and to discuss with
the staff the barriers that the company overcame in successfully implementing its UV system.
The visit also provided valuable information on the technological progress that has been made
with UV coatings in metal can manufacturing.
Coors recently won the National Industrial Competitiveness Through Efficiency, Energy,
Environment, and Economics (NICE3) grant to advance the development and implementation of
UV coatings for cans. The U.S. Department of Energy (DOE) and EPA jointly award the grant
each year to an organization that is planning to use a new process and/or equipment to reduce
the generation of pollution in manufacturing.
B-3
-------
This trip report includes four sections. Section II identifies the location of the Coors
facility. Section III presents the individuals who participated in the site visit, and Section IV
includes the technical information compiled during the site visit
n. Place and Date
Coors Container Complex
South Service Road
Golden, CO 80401
(303)277-5067
August 16, 1993
HI. Attendees
Coors Brewing Company
Erik T. Donhowe, Manager of Environmental and Safety Services
David H. Johnson, Senior Environmental Control Project Manager
Jack S. Kowal, Jr., Principal Chemical Project Manager
Acurex Environmental Corporation
Mitchell R. Wool, Regional Program Manager
TRC Environmental Corporation
Beth W. McMinn, Environmental Engineer
Steven R. Church, Environmental Scientist
IV. Discussion
The discussion began with TRC describing the twofold purpose of its visit: to learn more
about the Coors can manufacturing process, and in particular the application of UV-cured
coatings to 2-piece aluminum cans, and to identify the barriers that Coors had to overcome in
implementing its UV system. The following areas were discussed with the Coors personnel:
• Market Profile
• Manufacturing Supplies
Manufacturing Process Profile
• Environmental Impacts
• Implementing a UV System: the Coors Experience
• Barriers to the Extended Use of UV Coatings
B-4
-------
Each topic is discussed in detail below.
A. Coors History and Market Profile
The Coors Brewing Company has three major facilities located in Golden, Colorado;
Memphis, Tennessee; and the Shenandoah Valley, Virginia. Coors is the third largest producer
of beer in the United States. The Coors can manufacturing plant in Golden is the largest 2-piece
aluminum can plant in the world, and it produces approximately 4.2 billion cans a year. (The
plant is the only can manufacturing facility owned by Coors.) Coors employs 6,500 people at
the Golden location, 1,200 of whom work in the can manufacturing plant.
B. Manufacturing Supplies
The major raw materials used in manufacturing cans at the Coors plant are aluminum,
inks and overvarnish. The plant consumes 65,000 tons of aluminum, 85 tons of ink, and 490 tons
of overvarnish each year. The company has approximately 10 standard design labels that it uses
for its cans. The inks used on a given line vary with the type of can being produced. For
example, cans produced for the original Coors beer use four different inks: red, gold, black and
yellow. All of the cans receive a clear overvarnish.
C. Manufacturing Process Profile
The Coors plant has 5 lines for manufacturing cans, 4 of which were functional at the
time of the visit. One of the lines manufactures 16 ounce (.4736 liter) cans and the remaining
4 lines manufacture 12 ounce (.3552 liter) cans. The 16 ounce (.4736 liter) line produces from
650 to 1,900 cans per minute while the 12 ounce lines produce from 2,800 to 4,000 cans per
minute. The process on each line can be divided into two phases: can bodymaking and
decoration. The bodymaking section of a typical line includes 4 to 6 cuppers, 16 to 22
bodymakers, 1 can washer, and 1 can dryer. The decoration section of a line includes 1 to 3
printers, 2 to 3 UV ovens, 8 to 15 internal coating machines, 1 internal coating oven, 1 to 3
necker/flangers, and 1 to 3 leak testers.
The manufacturing process begins with large aluminum coils. A tag on each coil
identifies the supplier and the Coors' date of receipt. At each line a coil is placed on an uncoiler.
As the coil is unwound, the sheet of aluminum passes through a tray containing petroleum
lubricant, fatty acids, and surfactants. The lubricant prevents the aluminum from oxidizing during
the manufacturing process. Once the aluminum is lubricated, it passes into a cupper that punches
blank disks from the sheet and draws them into cups approximately 4 inches (10.16 cm) in
diameter and 1.5 inches (3.81 cm) high.
The cups next move along a vacuum conveyor belt to the bodymaker. In the bodymaker,
the cups pass through a series of 3 dies that strike them into a more elongated shape. During
this draw and iron (D and I) process, the walls of the cans are drawn to a thickness of 0.0035
inches (0.089 mm). The cans are also punched to form a concave bottom. The concave bottom
B-5
-------
improves the cans' ability to withstand the pressure generated during the filling process. Once
the can bodies have been formed, the walls are trimmed to within .002 to .003 inches (0.051 -
0.076 mm) of the desired can height. The cans emerging from the bodymaker are slightly thicker
near their tops because they will be necked and flanged after the interior and exterior coating
processes.
Before the coating processes can begin, the cans must be washed and rinsed. In the
washers, the cans pass through four stages. In the first two stages, the washer applies a sulfuric
acid/surfactant chemical solution to clean the contamination off the cans. In the last two stages
of the washer, the cans are rinsed with deionized water, which removes any remaining residues.
Finally, the cans are dried to prepare them for the decorating phase of the process.
To print images on its cans, Coors uses essentially the same type of printer as those used
in thermal systems. (See Figure 1.) The cans enter the printer from the top where they are
loaded onto a mandrel wheel. The mandrel moves the cans to the ink wheel, which applies the
inks to the aluminum cans. The ink wheel holds a rubberized blanket that picks up the complete
color image of the cans as it rotates by four different ink stations. The ink stations apply the
inks to the wheel through printing plates that match each color image of the cans. When the ink
wheel comes into contact with the cans, the mandrel spins the cans so that the complete image
is applied. The cans are then immediately moved to a roller where 0.5 - 0.8 mil (100 to 120 mg)
of UV-curable overvarnish is applied to each can. Printing the images and applying the
overvarnish to each can takes less than a second.
It is important to note that the overvarnish is applied to the inks before curing takes place.
This "wet-on-wet" application is possible because the overvarnish is the only component of the
UV system that requires curing. The inks are designed to be compatible with the overvarnish.
Once the photoinitiators in the overvarnish have been stimulated for curing, the inks are captured
inside it. Most of the inks that Coors uses do not have photoinitiators. This reduces the cost of
the inks and expands the range of colors available for production.
After the printer, the cans are transferred by chain to a vacuum belt. The vacuum belt
carries the cans upright through the UV oven where each can is cured in less than 0.7 seconds.
(See Figure 2.) The Deco Ray 2 UV oven used at Coors was designed by Fusion Systems of
Rockville, Maryland. The oven uses 10 inch (25.40 cm)-long lamp modules that are positioned
at a 20 degree angle. The angle guarantees that the cans are cured from top to bottom. The
lamps contain mercury bulbs that are sunounded by parabolic reflectors. The design of the
parabolic reflectors provides a uniform intensity of light along the substrate of the cans, enabling
them to cure evenly. As the cans pass through the oven, they are three inches apart (from the
center of one can to the next). The oven cures approximately 1,700 cans per minute, and there
is normally I oven per printer on a line.
B-6
-------
mandrel wheel
can feed
blanket
segment
Ink plate
transfer unit
L overcoat unit
o/c application roller
VULCAN PRINTER
-------
FIGURE 2
oeco chain-
U»H» CHAM MNOCKET
CAMNiT
MUXCTOfl
nuADurofK
LOWER CHAIN
SPROCKET
IRRADIATOR
REFLECTOR
C-M U.V. OVEN
B-8
-------
Once the exterior coating of the cans has been cured, they are transported by vacuum belt
to the interna] coater. At this point, the cans are turned on their sides. Each can arrives at an
airless spray gun station that applies a water-based epoxy coating throughout its interior. When
the spraying occurs, the can is spun to uniformly apply the coating. The cans are then moved
horizontally through the internal coating (1C) oven, which operates at temperatures ranging from
320°F to 400°F (160°C to 204°C). The heat cures the internal coating as water and solvent
evaporate from the substrate of the cans.
The Coors staff estimates that 85 percent of the overvamish is cured by UV light The
remaining 15 percent is cured by the heat from the IC oven. During the final curing process in
the IC oven, the exterior coating loses approximately 8 percent of its weight. The Coors staff
believes that this can be partially attributed to the small amount of solvent, water, and other trace
constituents in the UV overvamish. Coors plans to use some of the money from the NICE3 grant
to identify the contents and to verify the quantities of the coating loss.
From the IC oven, the cans are transported to a necker/flanger machine that forms the
neck on the open end of the cans. The cans then travel to a spin necker that narrows the necks
to a diameter of 2.25 inches (5.72 cm). The ends, which will be sealed with a double seam after
the cans are filled, are manufactured in another building located near the Coors Container
Complex. The can ends are coated on both sides prior to filling with a water-based epoxy
coating.
Once the cans leave the spin necker, they are tested for leaks in a light tester. If the cans
are leak proof, they go to a palletizer where they are put onto a pallet Some of the cans go
directly to semi-trailer cell bins and are shipped directly to the brewery for filling. During the
winter months when beer consumption is down, Coors will build an inventory of 200 to 300
million cans. That inventory disappears during the summer months when beer consumption
increases faster than the facility can form and coat cans. Building up an inventory allows the
can manufacturing plant to operate at a steady production rate throughout the year.
Coors operates its can lines 24 hours a day with a 3-shift rotation. The plant closes for
2 to 3 days a year. Each line receives roughly 36 hours of preventive maintenance on a quarterly
basis. Production workers perform the maintenance during their regular shift hours.
B-9
-------
D. Environmental Impacts
The data in the appendix are a summary of the Form R releases that the Coors Container
Complex reported to the EPA in accordance with Section 313 of the Superfund Amendments and
Reauthorization Act (SARA) of 1986 for years 1990 to 1992. The chemical releases from the
container complex are part of the 313 data that Coors reports for the Coors Brewing Company
(CBC) as a whole. The CBC (referred to as "Valley" in the appendix) includes the brewery, the
container complex and several other buildings. The container complex typically accounts for 90
percent or more of the CBC's releases.
For each chemical, the releases are organized into four pathways: fugitive emissions, stack
emissions, water discharges and off-site transfers. The off-site transfers refer to drums of waste
shipped to another location for disposal. Chemicals that are listed as "Not Reported" for a given
year were emitted in quantities below the Form R reporting threshold.
E. Implementing a UV System: the Coors Experience
The implementation of the Clean Air Act (1970) and the energy crisis of 1973 forced the
upper management at Coors to consider alternatives for its coating operations. Coors believed
that the UV system would provide substantial energy savings by eliminating the need for natural
gas in thermal ovens and by reducing the overall amount of energy consumed in the coating
process. Coors believed that the reduction in VOC emissions associated with a UV system would
make compliance with the CAA regulations significantly less burdensome than it would have
been with a thermal system.
Another important benefit that Coors anticipated from the new system was the smaller
space required for UV ovens. Because the can manufacturing industry operates on low profit
margins, plants continually have to increase their output to be profitable. Coors foresaw that
thermal ovens would become too large to accommodate the increased line speeds. UV ovens are
significantly smaller than thermal ovens because they cure cans in less than 1 second. Thermal
ovens, on the other hand, cure cans by baking them for over 30 seconds. This requires a long
chain passing through a larger oven.
Because of the perceived benefits of UV coatings, Coors switched from a thermal coating
system to a UV system in 1975. The conversion took 3.5 months, from mid September through
December of that year. There were several motivating factors behind Coors' switch to a UV
system. The most important of which was an internal mandate to reduce VOC emissions from
the company's coating operations to zero. The president of the Coors Container Division, Robert
Momin, was personally committed to achieving that target.
After full-scale production began with UV coatings in January 1976, the company needed
another 18 months to overcome the problems associated with implementing the new system.
Nearly all of the problems involved the chemistry of the new coatings rather than the mechanics
B-10
-------
of the equipment. The chemistry problems tended to manifest themselves in the color and print
quality of the cans. The production workers had difficulty exactly matching the UV inks with
the colors of the original Coors beer can (i.e., red, yellow and black). Black was particularly
difficult to color match because of its ability to absorb significant amounts of light. Furthermore,
many of the cans produced in 1976 and early 1977 did not meet Coors' internal standards of
image clarity. The production staff worked through these problems on a trial-and-error basis.
The Coors staff considers printing to be an art. As a result, the production workers had to make
many small adjustments in mixing the inks to produce coated cans that met the company's
standards.
It should be noted that Coors did not overcoat its cans at the time of the conversion.
Instead a high gloss UV ink system was utilized. The company did not begin using a UV
overvarnish until the early 1980s.
An important part of the conversion process was training the production workers to use
the new UV inks. Early in the process, skin sensitivity was a problem because of the
transparency and low tack of the UV materials. A small number of production workers became
sensitized to the new materials because they could not tell when the materials were on their skin.
The old solvent-based materials were sticky; therefore, workers knew immediately when they
needed to wash off the materials. While implementing the UV system, Coors enhanced the
hygiene training it gave to its workers. Workers were told to apply barrier creams more
frequently to protect against skin sensitivity. The company also color coded the containers of
UV materials and equipment to make them easier to recognize.
Because Coors produces its cans for beer, a food product, Coors workers were cognizant
of the importance of hygiene and Food and Drug Administration (FDA) requirements before the
conversion took place. Using the new UV materials was different for them, but it was not an
overwhelming challenge. They knew how to handle materials cleanly. To make sure that its
workers were not exposed to undue hazards with the new materials, the company conducted
several carcinogenic studies.
F. Barriers to the Extended Use of UV Coatings
Through continuous improvements to the UV system over the past 17 years, Coors has
been able to overcome the initial difficulties it had in implementing the system. However, there
are some lingering barriers that may prevent other 2-piece can manufacturing facilities from
adopting UV technologies. The most obvious barrier is the cost of UV coatings. An overvarnish
for a thermal system typically costs $20/gallon of applied solids. The cost for a comparable UV
overvarnish ranges from $28 to $35/galIon of applied solids. There have been several analyses
indicating that the production efficiencies and energy savings from a UV system more than
compensate for the initial high price of the UV coatings. Nevertheless, many managers in the
industry do not consider the UV alternative because of the higher price of the coatings.
B-ll
-------
Because Coors is the only user of UV coatings in the 2-piece can market, the resin
manufacturers and coating suppliers have not invested a substantial amount of money into
developing UV products. Most of these companies consider the UV market to be a specialty
market that receives limited attention and resources. The lack of a large, competitive end market
for UV coatings has helped keep their prices well above those for commodity coatings. It has
also slowed progress to resolve some of the technical issues. For example, there has been debate
within the industry over the feasibility of using a UV white basecoat on a high volume 2-piece
can line. The largest user of white basecoat does not use UV technology; therefore, there is little
incentive for resin manufacturers and coating suppliers to develop the technology.
Another barrier to more widespread use of UV coatings is the perception that they have
low abrasion resistance. Coors uses an acrylic-based coating on all of its 12 ounce (.3552 litre)
can lines. Although the company claims that its cans meet the industry standards for abrasion
resistance, acrylic based UV coatings have a reputation of low resistance. Coors uses cardboard
packaging on its six packs rather than plastic rings because of the potential abrasion caused by
the rings.
Coors is in the process of developing a cationic-based UV overvarnish. The cationic
coating offers abrasion resistance superior to the acrylic-based coating, but it cures differently
than acrylates do. The cationic coatings are designed to begin curing when exposed to UV light
and to continue curing after the exposure ends. This "dark curing" allows for a strong
polimerization process to take place. Coors is testing a cationic-based overvarnish on its 16
ounce (.4736 liter) can line. This line can accommodate the slower cure rate of the cationic
coating because it only produces 650 cans per minute. The company is working with AKZO
resins to reduce the cure time of the cationic coating to less than .7 seconds. It hopes to
implement the cationic coating on all of its 12 ounce (.3552 liter) can lines.
Cationic chemistry is an important area of research for UV coating technology. Besides
the improved abrasion resistance and "dark curing", cationic coatings offer the benefit of "shadow
curing." Shadow curing occurs in areas of a substrate that were never directly exposed to the
UV light. The polymerization process is initiated in other areas of the substrate and passes onto
the unexposed areas. This technology expands the potential applications of UV coatings in 2-
piece can manufacturing.
B-12
-------
APPENDIX
B-13
-------
SARA SECTION 313 FORM R SUMMARY
VALLEY AND CONTAINER EMISSIONS
1990
1991
1992
Chemical/
Total
Container
Total
Container
Total
Container
Release Pathway
Reported for
contribution to
Reported for
contribution to
Reported for
contribution to
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total(lbs)
Ammonia
Fugitive
8,300
563
11,000
25.5
13,000
182
Stack
600
0
0
0
0
0
Water
63,000
0
34,000
0
36,000
0
Off-site
19,000
0
920
1,282(3)
1,100
1,100
I a-Butyl Alcohol
120,000
Fugitive
125,637
140,000
137,750
110,000
111,233
Stack
140,000
139,981
150,000
154,145
120,000
124,661
Water
0
0
0
0
0
0
Off-site
33
4,467
1,300
5,333(3)
2,400
34
Chlorine
Fugitive
11-499
LITTLE
11-499
0
0
Not reported
Stack
0
LITTLE
0
0
0
Water
0
0
0
0
0
Off-site
0
0
0
0
0
Chromium
compounds
Fugitive
1
Not reported
14
Not reported
0
Not reported
Stack
0
2
4
Water
82
99 '
11-499
Off-site
710
1,840
24,310
1
Coors 313 Summary
- The S.M. Stoller Corporation
-------
1990
1991 |
1992
Chemical/
Total
Container
Total
Container
Total
Container
Release Pathway
Reported for
contribution to
Reported for
contribution to
Reported for
contribution to
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total (lbs) ]
Valley (lbs)
Valley total(lbs)
Copper
compounds
Fugitive
2
0
Not
Not reported
Stack
0
0
reported
Water
850
0
Off-site
21,300
7
Glycol ethers
Fugitive
47,000
45,964
67,000
65,490
35,000
34,308
Stack
50,000
50,404
47,000
47,343
38,000
37,733
Water
0
0
o
0
0
0
Off-site
73
7,283
420
1,674(3>
1,800
3,933(3)
Hydrogen
fluoride
Fugitive
11-499
LITTLE
0
0
Not reported
Stack
11-499
LITTLE
0
0
Water
0
0
0
0
Off-site
1
0
41
0
Lead compounds
Fugitive
Slack
0
Not reported
Not
Not reported
Water
0
reported
•
Off-site
110
440
Coors 313 Summary
- The S.M. Stollcr Corporation
-------
1990
1991
1992
Chemical/
Total
Container
Total
Container
Total
Container
Release Pathway
Reported for
contribution to
Reported for
contribution to
Reported for
contribution to
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total(lbs)
Manganese
compounds
Fugitive
34
0
1
0
0
0
Stack
0
0
0
0
0
0
Water
4
4
4
0
0
0
Off-site
80
300,748
379,050
274,451
222,150
28
Methanol
Fugitive
12,000
Not reported
Not
Not reported
Stack
0
reported
Water
0
Off-site
0
Methyl ethyl
ketone
Fugitive
9,600
8
Not
Not reported
Stack
0
o
reported
Water
0
0
Off-site
660
0
Nickel
compounds
Fugitive
19
Not reported
7
Not reported
0
Not reported
Stack
0
1
2
Water
740
97
16
Off-site
850
1,240
30,350
Coors 313 Summary -The S.M. Stoller Corporation
-------
1990
1991
1992
Chemical/
Total
Container
Total
Container
Total
Container
Release Pathway
Reported for
contribution to
Reported for
contribution to
Reported for
contribution to
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total (lbs)
Valley (lbs)
Valley total(lbs)
Phosphoric acid
Fugitive
Stack
0
0
0
Not reported
0
Not reported
Water
0
0
0
0
Off-site
0
0
0
0
280
0
230
11
Sulfuric acid
Fugitive
0
0
0
0
0
0
Stack
0
0
0
0
0
0
Water
0
0
0
0
0
0
Off-site
200
0
213
0
28
0
1,1,1-
Trichloroethane
Fugitive
130,000
106,808
Not
Not reported
Stack
0
0
reported
Water
0
0
Off-site
4,500
8,825(2)
Trichlorotri-
fluoromcthane
Fugitive
Not
25,000
23,200
Not reported
Stack
reported
0
0
Water
0
0
-
Off-site
0
0
Coors 313 Summary -The S.M. Stoller Corporation
-------
C^The total fugitive emissions from the Yalley appear to be less than the contribution from container. The Valley fugitive
emissions were based on a material balance of chemical used minus stack, water, and off-site transfers. The total off-site transfer
was more than reported by container by about 2,000 lb, and therefore the fugitive emissions were lower by about 2,000 lb.
Emissions are reported to two significant figures, so fugitive emissions of 123,600 lb were reported as 120,000.
^Some or all of container off-site transfers were to recyclers and were therefore not reportable for 1990.
^Container contribution includes transfers to Investment Recovery or off-site.
Coors 313 Summary -The S.M. Stoller Corporation
-------
TRC
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hill, NC 275U
*(919) 968-9900 Fax (919) 968-7557
Environmental Solutions through Technology
November 24, 1993
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory
MD-61
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Prime Contract 68D20181
Ball Trip Report
TRC Environmental Reference Number 1645005
Dear Carlos:
Attached is the trip report from our visit to the Ball Can Manufacturing Plant in
Williamsburg, Virginia. Ball has declared the entire report to be nonconfidential. Please let
me know if you have any questions or comments on the report.
Sincerely,
Steven R. Church
Environmental Scientist
Offices in California, Colorado, Connecticut, Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texas,
Washington, Washington, D.C., and Puerto Rico A TRC Ccmpony
Priaiec oo Recyded Paper B" 1 9
-------
Date:
November 12, 1993
To:
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory (MD-61)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
From:
Steven R. Church
TRC Environmental Company
Subject:
Site Visit - Ball Metal Container Plant
Manufacturer of Two-piece Aluminum Cans
EPA Contract 68-D2-0181, Work Assignment Number 1/005
TRC Reference Number 1645005
I. Purpose
As part of its emphasis on pollution prevention, the U.S. Environmental Protection
Agency (EPA) is identifying the barriers to the extended use of radiation-cured and waterborne
coatings in Source Reduction Review Project (SRRP) categories and Maximum Achievable
Control Technology (MACT) categories. TRC Environmental Corporation (TRC) is supporting
EPA in this effort by evaluating the current use of these coatings in the metal can manufacturing
industry under Work Assignment Number 1/005, EPA Contract Number 68-D2-0181.
The primary source of air emissions in metal can manufacturing plants is volatile organic
compounds (VOCs) used in the coatings of the cans. Coatings cured by ultraviolet (UV) light
(i.e., UV coatings) are considered a pollution prevention alternative for the industry because they
consist of nearly 100 percent solids which remain on the substrate during the curing process.
Few, if any, solvents are emitted by UV coatings.
The trip to the Ball can manufacturing plant in Williamsburg, Virginia was arranged with
the help of the Can Manufactures Institute (CMI). The purpose of the visit was to review the
coating process of a two-piece manufacturing facility that uses waterborne coatings, the industry-
standard for interior and exterior coatings. The trip also provided an opportunity to see a
merchant can manufacturing plant. The Ball plant sells cans to a variety of breweries and
beverage manufacturers in the Eastern United States. The demands on merchant can
manufacturing plants are different from those on captive can manufacturers, plants that serve only
one end user.
This trip report includes four sections. Section D identifies the location of the Ball
facility. Section III presents the individuals who participated in the site visit, and Section IV
includes the technical information compiled during the site visit.
B-20
-------
II. Place and Date
Ball Metal Container Plant
8935 Pocahontas Trail
Williamsburg, VA 23185-6249
(804)887-2061
October 21, 1993
HI. Attendees
Ball Corporation. Metal Container Division
Lou Dunn, Production Manager, Metal Beverage Container Group
Timothy D. Case, Environmental Engineer, Metal Container Operations
TRC Environmental Corporation
Beth W. McMinn, Environmental Engineer
Steven R. Church, Environmental Scientist
IV. Discussion
The visit to the plant included an opening conference, during which TRC viewed a video
of Ball's two-piece manufacturing process. The video was followed by a question and answer
session and eventually by a tour of the production lines. During the visit, the following topics
were discussed:
• Company Profile
• Manufacturing Supplies
• Manufacturing Process Profile
• Environmental Impacts
• Waste Minimization
Each topic is discussed in detail below.
A. Profile of Ball Metal Container Division
The Ball Corporation began manufacturing two-piece metal cans in 1968 when it acquired
Jeffco Manufacturing Company in Golden, Colorado. Jeffco had been producing beer cans and
ends for Adolph Coors Company since 1962. During the 1970s and 1980s, Ball steadily
B- 21
-------
expanded its operations to become the industry's third largest producer of metal cans.1 In 1992,
Ball's seven U.S. can plants produced over 12 billion cans.2 Production at the Williamsburg
plant, Ball's second plant, began in 1972. The plant employs 230 workers, 34 of whom are
salaried employees. The Metal Container Operations Division of Ball employs 3,500 people
nationwide, and the company as a whole employs 14,200 people. The Williamsburg plant
operates 24 hours a day, 7 days a week. There are two 12-hour shifts each day. The employees
work on a 4-day-on, 4-day-off rotation. The plant closes for 5 days a year.
B. Manufacturing Supplies
The major raw materials used in manufacturing cans at the Ball plant are aluminum,
basecoat, inks, overvarnish, and internal coating. The aluminum arrives at the plant in five ton
coils. Each coil lasts approximately 4 hours during a production run. Because Ball is a merchant
supplier of cans, the plant will run oyer 300 different labels in a given year. A large supply of
inks are stored, maintained, and mixed on site by a representative of Acme Inks, the major ink
supplier for Ball. Glidden, Valspar, PPG, and BASF supply overvarnish and interior coatings
with VOC contents of 10 to 15 percent.
C. Manufacturing Process Profile
The Ball plant has four lines for manufacturing cans. Three of the lines are dedicated to
the manufacture of 12 ounce (355 ml) cans, and one line manufactures both 12 (355 ml) and 16
ounce (474 ml) cans. The plant currently produces 206 cans. This number refers to the diameter
of the open end of the cans. A 206 can has a 2 and 6/32 inch (5.56 cm) diameter. During the
two weeks following our visit, Ball was planning to re-tool two of its lines for production of 204
beverage cans. The trend in the industry is toward narrower can ends because they save the can
manufacturers millions of dollars in decreased raw material costs. The beverage companies have
accepted the narrower ends because some of the raw material savings are passed onto them. The
beer industry, however, has not yet committed to a 204 can and will remain with the 206
diameter.
The lines at the Ball plant can be divided into two phases: can bodymaking and
decoration. The bodymaking sections of the line include lubricators, cuppers, bodymakers,
trimmers, and washers. The number of machines varies with each line. For example, the plant
has two 12-out and two 6-out cuppers that serve the three 12 ounce lines. There is also one 13-
out cupper serves the 12/16 ounce line. (A 6-out cupper punches 6 cups per stroke, a 12-out
cupper 12 cups per stroke, and a 13-out cupper 13 cups per stroke.) The decoration sections of
the lines include basecoaters, basecoat ovens, printers, bottom coaters, deco ovens, internal
1 Ward's Business Directory of U.S. Private and Public Companies 1993, Volume 5
2 Ball Corporation, Metal Container Division Brochure
B-22
-------
coaters, internal coater ovens, waxers, neckers, spinneckers and flangers, light testers, and
palletizers. One of the lines does not have a basecoater or basecoater oven.
The manufacturing process begins at the uncoiler. A five-ton aluminum coil is placed on
each of the uncoiler's two arms. When the coil is finished on one arm, the other arm is flipped
around and fed into the production line. This arrangement minimizes down time. From the
uncoiler, the sheet of aluminum passes through a lubricator that applies a synthetic, water-soluble
lubricant to it. The sheet then passes into a cupper which punches circular blanks of aluminum
and draws them into cups approximately 3.56 inches (9.05 cm) in diameter and 1.5 inches (3.81
cm) in height. The cupper operates at 250 strokes per minute.
From the cupper, the cans travel on a conveyer belt to one of the line's seven
bodymakers. The bodymakers use a punch mounted on a ram to push the cups through a series
of four tooling dies. This draw and iron (D and I) process stretches and forms the cups into
cans. While the cups are being punched through the dies, the concave bottom is formed which
improves their ability to withstand the pressure generated during the filling process. Once the
cans emerge from the bodymaker, they move to a trimmer to be cut to their desired height.
Before the decoration process can begin, the cans must be washed and rinsed. The cans
pass from the trimmer along a vacuum belt up to the washer, which is located on the second
floor of the plant. The washer consists of four stages in which the cans are rinsed with tap
water, cleaned with a sulfuric acid solution, cleaned with a caustic solution, and rinsed with
deionized water. The cleaning process removes oil, dirt, and metallic fines from the cans' surface
and etches them in preparation for decoration.
The decoration process at the Ball plant often begins with the application of a basecoat
to the exterior of the cans. In the basecoater, the cans pass over a roller that applies a white ink
directly to them. The white ink serves as the base upon which other inks will be applied. From
the basecoater, the cans move along a vacuum belt up to the basecoat oven where the white ink
is cured. The basecoater oven operates at temperatures near 400° F (204° C). Inside the oven,
the cans move up and down along a pin chain that is shaped in a continuous "S" formation. The
cans spend from 35 to 45 seconds inside the oven.
Once the cans leave the basecoat oven, they move down a vacuum belt to the printer.
For those cans not requiring a basecoat, the printer is the first step in the decoration process.
When the cans enter the printer, they are loaded onto a mandrel wheel. The mandrel moves the
cans to the ink wheel, which applies the desired ink pattern. The ink wheel consists of a
rubberized blanket that picks up the complete color image of the label as it rotates past at least
four different ink stations. The ink stations apply the inks to the wheel through printing plates
that match each color image of the label. When the ink wheel comes into contact with the cans,
the mandrel spins the cans so that the complete image is applied. The cans are then moved to
a roller where a thin film of overvarnish is applied to their entire exterior surface. The
application of the overvarnish onto the inks is referred to as a "wet-on-wet" application.
B-2 3
-------
The inks for the printer are mixed by the representative from Acme Inks in a room
adjacent to the production floor. He maintains an "ink recipe" for each of the labels that Ball
runs. The ink recipe identifies the colors of the inks and the quantities to be applied to each can.
Although most of the labels that Ball prints require four different colors, the plant occasionally
prints labels with six colors. The addition of two inks to the printing process does not reduce
line speeds unless the inks are applied on top of each other to achieve various shades of color.
A specialty order with shading requirements takes more time to set up and to run than a standard
order. The ink representative must spend time mixing various quantities of the inks to obtain the
correct shades. The production personnel then have to complete trial runs to ensure that the ink
mixtures cure to the desired shades. Once the line has been set up correctly, the production time
will be longer for the specialty order because the inks will be applied on top of each other.
From the printer, the cans travel along a vacuum belt up to the second floor. Before they
reach the deco oven, the cans pass through the bottom coater which applies a waterbome lacquer
to the bottom rim of the cans. This coating not only protects the cans but improves their
mobility. The cans then move to a pin chain which takes them through the deco oven. The deco
oven is similar to the basecoater oven. It operates within the same temperature range, near 400°F
(204° C), as the basecoater oven, and the cans spend approximately 45 seconds inside. Once the
cans exit the deco oven, they are cured and ready to move down to the internal coater on the first
floor.
The internal coater consists of seven airless spray guns arranged in a row. The cans will
pass in front of one of the seven guns, which applies a waterborne enamel coating to their
interior. The same coating is applied to both beer and beverage cans; however, the amount of
coating varies. Beverage cans receive approximately 50 percent more coating than beer cans
because of the acidic nature of their contents. From the internal coater, the cans travel upstairs
along a vacuum belt to the internal coater (IC) oven. The IC oven is different from the basecoat
oven and the deco oven in that the cans travel upright along a conveyor belt through it rather
than along a pin chain. The cans spend approximately 45 seconds inside the IC oven.
Once the internal coat has been cured, the cans travel on a vacuum belt to the waxer. The
waxer prepares the cans for necking by applying a thin layer of lubricant to the outside of the
open edge of each can. The necking operation involves three steps in which the cans pass
through a necker, spinnecker and flanger. The necker squeezes the open end of each can down
to the specified diameter (e.g. 204) by creating a ridge. The spinnecker then removes the ridge
and smoothes the narrowed area near the open end of each can. Finally, the flanger rolls back
the top edge of each can to form a lip, which is later used to attach an end to the can after the
filling process has been completed.
After the cans leave the spinnecker and flanger, they pass through a light tester which
checks for leaks. If the cans are leak proof, they go to the palletizer where they are placed onto
cardboard or plastic pallets. Each pallet accepts 389 cans. Once a pallet is full, it is lowered a
few inches and a new pallet is stacked on top of it. The process is repeated until a stack of 21
layers has been formed. The stacks are either stored in a Ball warehouse or shipped to the customer.
B-24
-------
D. Environmental Impacts
The major emission points in the can making process are the three ovens (basecoat oven,
deco oven, and IC oven) and the internal coater. Although the coatings are waterborne, they
contain 10 to 15 percent solvents. The hazardous constituents include n-butyl alcohol in the
coatings and glycol ethers in the inks and coatings. These pollutants are released into the air
when the coatings are baked in the ovens. None of the ovens has a control device for capturing
or destroying the emissions.
The internal coater has an estimated transfer efficiency of greater than 95 percent. Ball
has a receptacle on the internal coater which captures the overspray (i.e., the coating that misses
the cans) and mixes it with waste basecoat and overvarnish. The waste coatings are blended
together with used oils and lubricants from the bodymaking process. A contractor, Heritage
Environmental Services, picks up the mixture, treats it, and uses it as fuel in cement kilns at an
offsite location.
Wastewater containing sulfuric acid from the can washer is treated in a dissolved air
flotation and flocculation system located on the first floor of the plant. Following the treatment,
the flocculent is pushed through a series of filters that catch aluminum fines, oil, dust, and
polymers, forming non-hazardous "filter cake." The cake, which is 50 percent solids, is disposed
of in a landfill.
The hazardous waste generated by the plant consists of contaminated solvent used to clean
the printing presses. The solvents are kept in quart-sized safety cans. The small size of these
containers limits the amount of solvents that workers may contact during cleaning operations.
Furthermore, Ball has managed to reduce the amount of solvents in the cleaning solutions by
using a higher concentration of water. Once the solvents become spent, they are put into drums.
The drums are picked up by Heritage Environmental Services and disposed of offsite. The rags
used to apply the solvents to the printing presses are not considered hazardous and are cleaned
offsite by an industrial cleaner. They can then be re-used.
E. Waste Minimization At Ball
Ball has been able to minimize the amount of waste generated in its plant by adhering to
a strict preventive maintenance schedule. The plant schedules twenty maintenance days a year
for each of the four lines. To maintain the lines, employees clean the machines, change the oil,
replace the belts, lubricate chains, and complete other tasks necessary to run efficient lines. The
schedule is essential to preventing the problems traditionally associated with thermal coating
systems. For example, pin chains in thermal deco ovens can break easily because of constant
wear and exposure to high temperatures (around 400° F, 204° C). When a pin chain in a deco
oven breaks, it can shut down a line for hours because the oven has to be turned off and cooled
before a production operator can enter the oven to fix the chain. This is not only expensive to
a company because of lost production time, it also creates a significant amount of off-quality
cans, which have to be removed from the oven and other parts of the line before production can
B- 25
-------
resume. Ball has avoided this problem by requiring its workers lubricate the chain everyday and
replace it on a regular schedule.
The success of the plant's preventive maintenance program can be attributed to the
company's commitment to total quality management (TQM). Although the Williamsburg plant
has not fully implemented a TQM program, it has required its employees to complete a 32-hour
course on quality management. During the course, employees learn the importance of preventive
maintenance as a way to reduce waste and save money. Employees also learn to use and
interpret results from a computerized statistical process control program. Ball employees
randomly sample cans from the lines and test column strength, can mobility, coating adherence,
and wall/cylinder thickness. Test data is then entered into a computer. Maintaining the data base
not only improves the quality of Ball's cans, but it reduces waste because Ball can detect a
production problem as soon as it develops. The problem can then be resolved immediately rather
than going undetected until the batch has been completed.
The focus of Ball's TQM program has been to empower employees to work together as
a team. Employees have learned the importance of each function on a line. When a printer
breaks down, fixing it becomes the responsibility of not only the printer operator but other
production workers on the line. This approach toward problem solving encourages employees
to consider the productivity of the line as a whole rather than to focus solely on the maintenance
of one production station. Ball also encourages production workers to accompany managers on
visits to customer facilities. These visits allow production associates to hear firsthand a
customer's compliments or complaints about a particular batch from the plant. The workers gain
a better understanding of the importance of product quality.
The end result of the TQM program is a plant in which employees take pride. The
production workers sense that they share in the plant's success. Their commitment to the plant
was reflected in the cleanliness of the lines. Workers seem to care about maintaining clean lines
and about minimizing waste. Ball actively participates in EPA's 33/50 program. The
Williamsburg plant was able to reduce its use of 33/50 chemicals by 100 percent two years ahead
of the program's schedule.
The can making industry is very competitive. Each region of the country has several
merchant suppliers who compete to meet the needs of beer and beverage companies. Some beer
and beverage companies also have their own can manufacturing facilities to satisfy a portion of
their demand. The profitability of the Williamsburg plant depends on its production volume.
Because the profit margin is small on every order that the plant runs, it must produce a large
number of cans each year to be profitable. Therefore, line speed, the number of cans that a line
produces in a given amount of time, is essential to the plant's success. Ball constantly searches
for ways to increase its line speeds whether it be reducing the down time between label changes
or implementing an effective TQM program.
B-26
-------
TRC
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hill. NC 27514
¦a (919) 968-9900 Fax (919) 968-7557
Environmental Solutions through Technology
December 14, 1993
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
EPA Prime Contract 68D20181
Campbell Soup Company Trip Report
TRC Environmental Reference Number 1645005
Dear Carlos:
Attached is the trip report from our visit to the Campbell Soup can plant on October
27. It contains no confidential information about Campbell's or its manufacturing processes.
Please let me know if you have any questions or comments on the report.
Sincerely,
Steven R. Church
Environmental Scientist
Enclosure
Offices in California, Colorado, Connecticut, Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texas,
Washington, Washington, D.C., and Puerto Rico A TRC Company
Pfi.Tlad on ftacyded Poper B~27
-------
Date:
November 18, 1993
Subject:
Site Visit - Campbell Soup Container Plant
Manufacturer of Two-piece and Three-piece Food Cans
EPA Contract 68-D2-0181, Work Assignment Number 1/005
TRC Reference Number 1645005
From:
Steven R. Church
TRC Environmental Company
To:
Carlos Nunez
Organics Control Branch
Air and Energy Engineering Research Laboratory (MD-61)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1. Purpose
As part of its emphasis on pollution prevention, the U.S. Environmental Protection
Agency (EPA) is identifying the barriers to the extended use of radiation-cured and waterborne
coatings in Source Reduction Review Project (SRRP) categories and Maximum Achievable
Control Technology (MACT) categories. TRC Environmental Corporation (TRC) is supporting
EPA in this effort by evaluating the current use of these coatings in the metal can manufacturing
industry under Work Assignment Number 1/005, EPA Contract Number 68-D2-0181.
The primary source of air emissions in metal can manufacturing plants is volatile organic
compounds (VOCs) used in the coatings of the cans. Coatings cured by ultraviolet (UV) light
(i.e., UV coatings) are considered a pollution prevention alternative for the industry because they
consist of nearly 100 percent solids which remain on the substrate during the curing process.
Few, if any, solvents are emitted by UV coatings.
Campbell Soup Company volunteered to host the visit to its plant in Maxton, North
Carolina. The purpose of the visit was to review the coating processes for two-piece and three-
piece food cans. TRC had already viewed the manufacturing process for two-piece beverage cans
during visits to two other can facilities. The Campbell's visit offered TRC the opportunity to see
the draw-thin-redraw (DTR) process for two-piece food cans and the welding process for three-
piece cans. Both of the Campbell's processes use conventional waterborne coatings.
This trip report includes four sections. Section II identifies the location of the Campbell's
facility. Section III presents the individuals who participated in the site visit, and Section IV
includes the technical information compiled during the site visit.
B-28
-------
II. Place and Date
Campbell Soup Facility
Route 2, Highway 71 North
Maxton, NC 28364
(919)844-5631
October 27, 1993
HI. Attendees
Camobell Soup Company
Robert C. Locke, Manager - Environmental Services
Thomas M. Braydich, Manager - Engineering and Power
Environmental Protection Agency - Air and Energy Engineering Research Laboratory
Carlos M. Nunez, Chemical Engineer, Organics Control Branch
TRC Environmental Corporation
Beth W. McMinn, Environmental Engineer
Steven R. Church, Environmental Scientist
IV. Discussion
The visit to the plant included an opening session, during which EPA discussed the goals
of the visit and the project, TRC representatives reviewed health and safety requirements and the
handling of confidential business information, and the Campbell representatives answered
questions about the plant's 2-piece and 3-piece can manufacturing processes and their
environmental impacts. The session was followed by a tour of the production lines.
Unfortunately, due to the company policy that prohibits visitors and personnel with beards from
entering the manufacturing facility, the EPA representative was unable to tour the production
area. During the visit, the following topics were discussed:
• Company Profile
Manufacturing Supplies
Manufacturing Process Profile - 2-piece Cans and 3-piece Cans
~ Environmental Impacts
Each topic is discussed in detail below.
B-29
-------
A. Profile of Campbell Soup Company
Campbell Soup Company processes and packages food products. The company is best
known for its soups; however, it produces a number of other well known food products, including
Pepperidge Farm Cookies and Godiva candies, at other facilities. Campbell's has owned the
facility in Maxton, North Carolina since 1978. It opened a regional distribution warehouse there
in 1979. In 1982 Campbell's attached a complete food production facility, which included can
manufacturing and food (i.e., soup, pasta, and bean products) processing operations, to the
warehouse. The facility currently manufactures 2-piece and 3-piece food cans. All of the cans
are used to package products manufactured at the plant.
The Maxton facility is one of four canned soup production facilities that Campbell owns
in the United States. The facilities compete with each other, as well as with other soup
companies, for regional markets. The Maxton facility primarily serves the Southeast and
Midatlantic regions of the country. It employs 1.123 workers: 670 regular hourly, 292 temporary,
62 weekly salary, and 99 monthly workers. Campbell's operates its can lines twenty-four hours
a day, 240 days a year. During soup season (from October - March), the lines run six to seven
days a week and five days a week during the rest of the year.
B. Manufacturing Supplies
The major raw materials used in manufacturing cans at the Campbell plant are uncoated
tinplate steel (for 3-piece cans), pre-coated tin-free steel (for 2-piece cans), waterborne interior
coatings, waterborne exterior coatings, Videojet inks, parafin, and end sealing compounds. Table
1 lists the manufacturing supplies that the Maxton facility uses.
C. Manufacturing Process Profile
The Maxton facility produces 2-piece and 3-piece cans of the following sizes: 211 x 400
(10 oz), 300 x 407 (14-16 oz), and 303 x 500 (16-19 oz). Of the eight can manufacturing lines
at the facility, six produce 3-piece cans and two produce 2-piece cans. Of the two 2-piece lines,
one uses draw and iron (D & I) manufacturing process and the other a draw-thin-redraw process.
Table 2 lists the production rates of the various lines. At the time of the visit, Campbell's was
in the process of constructing a D & I, 2-piece line. During the tour of the facility, TRC saw
a 3-piece line and DTR 2-piece line. Because the 2 and 3-piece manufacturing processes are
significantly different, they are described separately below.
Two-piece Manufacturing Process
The DTR process for manufacturing 2-piece cans is similar to the D & 1 process. The
DTR line consists of one coil lubricator, one cupper, one cup lubricator, two redraw presses, one
tester, three trimmers, one beader, and one palletizer. Because no coatings are applied to the
cans other than lubricants, the DTR line does not have an oven.
B-30
-------
TABLE 1. Campbell Soup Manufacturing Supplies - Maxton
Raw Material
Trade Name
Hazardous Ingredients
(Weight Percentages)
Supplier
Application
Ecoliner 5004AL
Ecoliner 3610BL
2-Butanone, Methyl
Ethyl Ketone
T 680 Thinner
Zep Extra
Videojet 16-8600
Videojet 16-8605
Bisphenol-A resin (< 25%)
Phenolic resin*
n-Butanol (< 6%)
Diethylene glycol
monobutyl (< 4%)
2-Butoxyethanol (3%)
Dimethyl ethanolamine (2%)
Bisphenol-A resin (23%)
n-Butanol (8%)
2-Butoxyethanol (3%)
Diethylene glycol monobutyl
ether (1%)
Dimethyl ethanolamine (2%)
Melamine resin (>2%)
Formaldehyde (<0.01%)
Proprietary
Proprietary
Ethylene glycol monobutyl
ether (5-10%)
Sodium hydroxide (<5%)
Nonylphenoxpoly ethanol
(<5%)
None
None
Valspar Corp. Interior lining
Hi-Tek
Polymers, Inc.
Prillman Co.
Zep
Manufacturing
Co.
Videojet
Systems Int.
Videojet
Systems Int.
Interior lining
Mobil Oil Corp. Cleaner
Ink and varnish
remover
Cleaner
Ink jet printing
Ink jet printing
* Weight percentage not reported.
B- 31
-------
TABLE 2. Production Rates Of Lines
Line
Type Of Can
Production Rate (cans per minute)
300 x 407 line
211 x 400 line
10 oz, 3-piece
14-16 oz, 3-piece
430 - 600
500
380
400
750
2,500
303 x 500 line
16 - 19 oz, 3-piece
Draw and iron line
303 x 404 line
Draw-thin-redraw line
16 oz, 3-piece
10 oz, 2-piece
10 oz, 2-piece
The process begins with an 11 ton coil of tin-free steel that has been pre-coated on both
sides. The coil unwinds into a tray where a roller applies a thin layer of wax to the steel. It then
passes into the cupper which punches and flanges six cups per stroke at 150 strokes per minute.
Before the cups can be drawn into cans, they must be lubricated. They move into a
chamber where a lubricator uses an electrostatic attraction to apply a thin, uniform coating of wax
to all surfaces of the cup.1 The lubricator creates a wax mist inside the chamber, and a corona
grid gives a positive charge to the particles. As the cups pass through the chamber, they are
grounded to attract the wax particles to their interior and exterior surfaces. Once lubricated, the
cups pass to the sanitary can maker where they enter with their open ends down. The can maker
draws them to their intermediate size of 2.94 inches (7.47 cm) high and 3.19 inches (8.10 cm)
wide and enlarges the flange. To obtain their desired size, the cans pass through another
lubricator and can maker where they are drawn to 3.875 inches (9.84 cm) high and 2.56 inches
(6.50 cm) wide.1
Upon exiting the second can maker, the cans travel through a trimmer which cuts excess
steel from their flanges. The cans pass through the trimmer with their open ends down to ensure
that steel shavings do not contaminate their interiors.1 From the trimmer, they enter the beader
which forms a series of ribs in their sides. The ribs strengthen the cans, allowing them to
withstand the pressure generated during the sterilizing process. The cans then move through a
light tester which detects leaks. All leak-proof cans pass onto the palletizer where they are
stacked on pallets.
Three-piece Manufacturing Process
The 3-piece can manufacturing process can be divided into two operations: sheet coating
and can fabricating. The sheet coating operation consists of a sheet feeder, a roll coater, a wicket
1 Church, Fred L.; "New Draw/Thin/Redraw Process Makes a Super Can for Campbell,"
Modern Metals, April 1986.
B- 32
-------
oven, and a sheet stacker. The can fabricating operation produces cylinder bodies and can ends.
It uses a slitter, a bodymaker, a wire welder, a seam sprayer, a thermal oven, a flanger/beader,
a scroll strip shearer, an end former, a compound liner, an end seamer, a light tester, and a
palletizer.
The sheet coating process begins with a 4 ft (1.2 m)-wide, 11 ton coil of uncoated tin
plate steel. As the coil is unwound, it is cut into 4 ft x 4 ft (1.2m x 1.2m) sheets which are then
stacked on top of each other and placed in a sheet plate feeder. From the feeder, the sheets
travel along a belt to a direct roll coater which applies a waterbome enamel coating to the top
side. This coating will serve as the interior coating of the cans. The roll coater applies the
coating by rolling in a clockwise direction and transfeiring the coating from the tray below it.
After being roll-coated, the sheets slide into the oven where wickets receive and transport them
vertically through the six oven zones. The oven contains approximately 2,800 wickets and
operates at approximately 400° F (204° C). The sheets spend approximately 15 minutes inside
it. Upon their exit, the cured sheets are stacked and transported by truck to the can fabricating
operations.
The fabrication process begins with a slitter which cuts 4x8 inch (10.16 x 20.32 cm)
body blanks from the sheets. The blanks then move along a belt to the bodymaker which wraps
them around a rod to form a cylinder. With a copper electrode, a wire then welds a side seam
on the top of the cylinder where the two ends meet. An airless spray gun then applies a
waterbome enamel coating to the seam of each cylinder. The cylinders exit the bodymaker in
an end-to-end, horizontal position and travel to an oven which cures the side seam spray at
approximately 400°F (204°C).
From the side seam oven, the cylinders pass through the beader/flanger where two
operations occur simultaneously. First, the machine rolls a series of ribs into the cylinder bodies.
The ribs strengthen the walls, allowing the cans to withstand the pressure generated during the
sterilizing process. Second, the machine curves the rims of the cylinders to form a flange. The
flange is essential for the next step in the process where the ends are attached to the cylinder
bodies.
Can ends are punched and formed on a separate manufacturing line at the same time the
can cylinders are formed. While the can bodies are taking shape on one section of the line,
another section forms the ends. A coil of pre-coated tin plate steel unwinds into a tray where
it receives a parafin coating for lubrication. The coil travels from the tray to the scroll strip
shearer which cuts the steel into indented rectangular strips. The indented shape of the strips
minimizes the amount of scrap steel generated during the process. The strips move along a
conveyor belt to the end press which punches circular ends and removes the scrap steel from the
belt to a recycling container. The ends then travel to a compound liner where they receive a
sealing compound. After the compound liner, the cans are ready to be attached to the body
cylinders.
B- 33
-------
The two sections of the line join at the end seamer. When the body cylinders enter the
end seamer, they are turned upright and joined with a can end. The end seamer then double rolls
the flanged end of the cylinder with the end itself. The first roll grips the end onto the flange,
and the second roll folds them together up toward the can body. When the cans exit the end
seamer, they pass over a light which tests them for leaks. If they pass the test, the cans move
to the palletizer which stacks them for shipping.
D. Environmental Impacts
The major emission points in the can-making operations at Maxton are the roll coater, the
oven, and the compound liner on the three-piece lines. Each of these stations has a duct above
it which vents fumes to a thermal oxidizer for destruction. The oxidizer is designed to destroy
volatile organic compounds (VOCs) with 95 percent efficiency by raising the temperature of its
gas stream to 1,400°F (760°C) for a minimum of 0.5 seconds.2 For 1990 the VOC sources
emitted approximately 334 tons of VOCs to the thermal oxidizer. With a destruction and
removal efficiency (DRE) of 93 percent, the thermal oxidizer emitted approximately 22.7 tons
of non-methane hyTdocarbons. Table 3 contains the emissions that Campbell's reported to the
Toxic Release Inventory in 1992.
The Maxton facility has a general water permit for boiler blow down and non-contact
cooling water discharges. The company uses a spray irrigation system for process waste water
generated by soup manufacturing and container operations.
2 Pacific Environmental Services, Inc., Site Specific Test Protocol For Air Pollution Testing;
Campbell Soup; Maxton, North Carolina, August 2, 1990.
B-34
-------
TABLE 3. Emissions Reported To Toxic Release Inventory For 1992 - Maxton Facility
Chemical
Source
Pounds Reported
(Kgm)
F orm/T reatment
Methyl ethyl
ketone
Cleaner in sheet
basecoat operation
14 (6.3)
1,931 (869)
11,990 (5,396)
Fugitive emissions
On-site energy recovery
Collected as fumes at
source and discharged to
thermal oxidizer
Phosphoric acid
Can washer
10,041 (4,518)
Mixed with process waste
water for spray irrigation
Cyclohexane
Cleaner in compound
liner operation
550 (248)
123,333 (55,500)
460,617 (207,278)
Fugitive emissions
On-site energy recovery
Collected as fumes at
source and discharged to
thermal oxidizer
N-butyl alcohol
Interior and Exterior
Coatings
350 (158)
20,238 (9,107)
113,622 (51,130)
Fugitive emissions
On-site energy recovery
Collected as fumes at
source and discharged to
thermal oxidizer
B-35
-------
TECHNICAL REPORT DATA , —,M - (l , -- x
(Please read Iniiructtons on the reverse before compleiii || | |||| || |||||| || 11| || 111II1 III
1 REPORT NO. 2.
EPA-600/R-95-063
3. f iii mi iiiiiiiimil iii inniii
F395-21581C
4. TITLE AND SUBTITLE
Evaluation of Barriers to the Use of Radiation-cured
Coatings in Can Manufacturing
5. REPORT DATE
April 1995
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Beth W. McMinn and Steven R. Church
8. PERFORMING ORGANIZATION REPORT NO.
CH-91-21
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TRC Environmental Corporation
6340 Quadrangle Drive, Suite 200
Chapel Hill, North Carolina 27514
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D2-0181, Tasks 1/005 and
1/015
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 3-12/93
14 SPONSORING AGENCY CODE
EPA/600/13
is.supplementary notes ^EERL project officer is Carlos M. Nunez, Mail Drop 61, 919/541-
1156. $
16. ABSTRACT-fj-j-ie report gives results of a study to investigate and identify the technical,
educational, and economic barriers to the use and implementation of radiation-cured
coatings in can manufacturing. The study is part of an EPA investigation of current
industrial use and barriers to the extended use of radiation-cured coatings in Source
Reduction Review Project (SRRP) and maximum achievable control technology (MA-
CT) standards development categories.) Among the important barriers were: (1) an
applied wet film thickness of > ^O^mg^er can of ultraviolet (UV)-curable overvarnisb
needed on most trial runs ;_(2 flower than expected energy savings; (3) inadequate cure
of overvarnish; and^(4)"ihk "pick off" during the wet-on-wet application of the over-
varnish to the inksXThe report suggests projects that could be,of help in overcoming
technical, educational, and economic barriers identified.^Among the opportunities
discussed were: (1) setting up a trial with a can manufacturer who is interested in
using UV-curable inks and coatings; (2) conducting research on cationic inks and
coatings, which have been billed as the next generation of UV-curable inks and coat-
ings; and (3) working with Radtech, the trade association representing the radiation-
curable coatings industry, to develop a UV-curable coating that could be approved by
the Food and Drug Administration for direct contact with food.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDEO TERMS
c. cosati Field/Group
Pollution Inks
Cans Varnishes
Manufacturing Cations
Curing
Coatings
Radiation
Pollution Control
Stationary Sources
Can Manufacturing
Radiation-cured Coatings
13B 14E
13 D
05C 07D
13 H
11C
14G
19. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report}
Unclassified
21 NO. OF PAGES
132
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73) £6
------- |