United States
Environmental Protection
Agency
Office of Pollution
Prevention and Toxics
Washington, DC 20460
EPA-745-R-98-016
December 1998
1996 Toxic Release Inventory
Data Quality Report
-------
-------
TABLE OF CONTENTS
1.0
2.0
3.0
EXECUTIVE SUMMARY i
INTRODUCTION 1-1
1.1 EPA's Overall Quality Assurance Program 1-2
1.2 Site Survey Objectives 1-2
1.3 Prior Site Survey Efforts 1-3
APPROACH 2-1
2.1 Survey Instrument 2-1
2.2 Sample Selection 2-3
2.3 Site Surveyor Selection and Training 2-5
2.4 Arranging Site Visits 2-5
2.5 Conducting Site Visits 2-6
2.5.1 Data Collection 2-7
2.5.2 Threshold Determinations 2-7
2.5.3 Release and Other Waste Management Estimates 2-8
2.6 Data Management/Data Quality Assurance 2-9
2.6.1 Quality Review of Survey Instrument and Data Entry 2-9
2.6.2 Data Weighting 2-9
2.6.3 Limitations of the Analysis 2-10
2.7 Data Analysis and Reporting 2-11
DESCRIPTION OF INDUSTRIAL PROCESSES SURVEYED 3-1
3.1 Iron and Steel Manufacturing and Processing 3-2
3.1.1 Steelmaking Using the Basic Oxygen Furnace 3-2
3.1.1.1 Cokemaking 3-2
3.1.1.2 Ironmaking 3-3
3.1.1.3 Sintering 3-6
3.1.1.4 Steelmaking 3-6
3.1.2 Steelmaking Using the Electric Arc Furnace (EAF) 3-7
3.1.3 Forming and Finishing Operations 3-8
3.1.3.1 Forming 3-8
3.1.3.2 Finishing 3-9
3.2 Primary Aluminum Processing 3-10
3.3 Primary Zinc Processing 3-13
3.4 Semiconductor Manufacturing 3-16
3.4.1 Photolithography 3-17
3.4.2 Thin Films 3-19
3.4.3 Etching 3-20
3.4.4 Cleaning 3-21
3.4.5 Doping 3-22
3.4.6 Chemical Mechanical Planarization 3-23
3.5 Printed Wiring Board Manufacturing 3-23
3.5.1 Board Preparation 3-24
3.5.2 Electroless Plating 3-25
3.5.3 Imaging 3-26
3.5.4 Electroplating 3-27
-------
TABLE OF CONTENTS (Continued)
4.0
5.0
3.5.5 Soldering Coating ,. 3-29
3.5.6 Electrical and Mechanical Testing 3-29
3.5.7 Printed Wiring Board Assembly and Soldering 3-29
3.6 Motor Vehicle Equipment Manufacturing 3-30
3.6.1 Foundry Operations 3-30
3.6.2 Metal Fabricating 3-33
3.6.3 Metal Finishing/Electroplating 3-34
3.7 Motor Vehicle Painting/Finishing 3-36
THRESHOLD DETERMINATIONS 4-1
4.1 Approaches Used for Determining Thresholds 4-1
4.2 Frequency of Incorrect Threshold Determinations 4-2
4.3 Reasons for Making Incorrect Threshold Determinations 4-9
4.3.1 Reasons Why Facilities Failed to Submit EPCRA Section 313
Reports for Chemicals That Exceeded Thresholds 4-10
4.3.2 Reasons Why Facilities Submitted EPCRA Section 313 Reports
for Chemicals That Did Not Exceed Thresholds 4-11
4.3.3 Chemical Activity Classification 4-13
4.3.4 Impact of Not Calculating Thresholds 4-15
4.3.5 Other Factors Correlating with the Frequency of Incorrect
Threshold Determinations 4-15
4.4 Lessons Learned 4-16
SOURCES AND TYPES OF RELEASE AND OTHER WASTE MANAGEMENT
ACTIVITIES 5-1
5.1 Observed Release and Other Waste Management Activities 5-2
5.2 Incorrectly Reported Release and Other Waste Management Activity
Types 5-13
5.2.1 Air Releases 5-16
5.2.2 Off-Site Transfers for Recycling and Disposal, and On-Site
Recycling 5-16
5.2.3 On-Site Treatment and Land Disposal 5-17
5.2.4 Water Discharges 5-18
5.3 Overlooked Release and Other Waste Management Activities 5-18
5.3.1 Container Residue 5-21
5.3.2 Process Areas and Discharge Streams 5-22
5.3.3 Other Treatment Areas 5-23
5.3.4 Combustion 5-23
5.4 Calculation Methodologies 5-23
5.4.1 Air Releases 5-24
5.4.2 Water Discharges 5-27
5.4.3 Off-Site Transfers for Release and Other Waste Management
Activities 5-27
5.4.4 Correct Methodology Usage 5-28
5.4.5 Emission Factors 5-29
5.5 On-Site Recycling, Treatment, and Energy Recovery 5-30
-------
TABLE OF CONTENTS (Continued)
6.0
7.0
8.0
RELEASE AND OTHER WASTE MANAGEMENT ACTIVITIES 6-1
6.1 Estimates of On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R 6-2
6.1.1 Comparison of the Facility Estimates to the Surveyor
Estimates 6-2
6.1.2 Comparison of the Facilities Surveyed to the National TRI
Database 6-13
6.2 Estimates of Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R 6-16
6.2.1 Comparison of the Facility Estimates to the Surveyor
Estimates 6-16
6.2.2 Comparison of the Facilities Surveyed to the National TRI
Database 6-25
6.3 Estimates of On-Site Waste Management Quantities as Reported in
Sections 7, 8.2, 8.4, and 8.6 of the Form R 6-28
6.3.1 Comparison of the Facility Estimates to the Surveyor
Estimates 6-28
6.4 Production Ratio/Activity Index 6-32
6.5 Source Reduction Activities 6-36
6.5.1 Source Reduction Reporting 6-37
6.5.2 Errors Made When Claiming Source Reduction 6-38
6.5.3 Overall Accuracy of Source Reduction Data 6-40
PREPARATION OF THE FORM R 7-1
7.1 Facility Personnel and References 7-1
7.2 Amount Of Time Needed To Prepare Form Rs 7-5
7.3 Use of the Hotline 7-8
7.4 Comments on the Form R Instructions and Guidance Manuals 7-10
7.5 Comments on the Automated Form R (APR) 7-12
7.6 Comments on Use of the Form A 7-15
RECOMMENDATIONS 8-1
8.1 Additional Guidance Concerning Form R Instruction and
Documentation 8-1
8.2 Additional Guidance Concerning Threshold Determinations 8-3
8.3 Additional Guidance Concerning Release Estimates 8-3
9.0
REFERENCES 9-1
-------
LIST OF TABLES
Table 3-1 Process Materials Inputs/Waste Outputs - Zinc 3-15
Table 3-2 Chemicals Used in Lamination, Drilling, and Cleaning 3-25
Table 3-3 Materials Used in Copper and Tin-Lead Electro- and
Electroless Plating Processes 3-26
Table 3-4 Chemicals Used in Photolithography for Printed Wiring Boards 3-27
Table 3-5 Materials Used During Etching 3-27
Table 4-1 Accuracy of Threshold Determinations for Reporting Year 1996, by
SIC Code 4-6
Table 4-2 Estimated Impact of Incorrect Threshold Determinations on
the Number of EPCRA Section 313 Reports Submitted to EPA 4-7
Table 4-3 Accuracy of Threshold Determinations by Reporting Year 4-7
Table 4-4 Reasons Why Facilities Failed to Submit EPCRA Section 313
Reports for Chemicals That Exceeded Thresholds 4-11
Table 4-5 Reasons Why Facilities Submitted EPCRA Section 313
Reports for Chemicals That Did Not Exceed Thresholds 4-12
Table 4-6 Frequency with Which Facilities Calculated Thresholds for EPCRA
Section 313 Chemicals, by SIC Code 4-16
Table 5-1 Distribution of Sources and Types of Release and Other Waste Management
Activities 5-3
Table 5-2 Misclassified and Overlooked Types of Release and Other
Waste Management Activities 5-14
Table 5-3 Misclassified and Overlooked Sources of Release and
Other Waste Management Activities 5-19
Table 5-4 Distribution of Calculation Methodologies 5-25
Table 5-5 Frequency the Best Methodology was Used by Facilities to Estimate
Release and Otherwise Waste Managed Quantities 5-29
Table 5-6 Types of Emission Factors Used for Fugitive and Stack Air Releases 5-30
Table 5-7 Facilities Incorrectly Reporting the Quantity Sent to
Treatment Rather than Actually Treated 5-31
-------
LIST OF TABLES (Continued)
Table 5-8 Observed On-Site Recycling Activities 5-32
Table 5-9 Chemicals For Which On-Site Recycling Was Claimed
(SIC Codes 33, 36, and 37 Combined) 5-32
Table 6-la Summary of SIC Code 33 On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R 6-4
Table 6-lb Summary of SIC Code 36 On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R 6-5
Table 6-lc Summary of SIC Code 37 On-Site Release and Other Waste Management
Quantities as Reported in Sections 5 of the Form R 6-5
Table 6-2 Comparison of Differences Between Facility Estimates and
Site Surveyor Estimates Across the SIC Codes 6-12
Table 6-3a Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 33 6-14
Table 6-3b Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 36 6-15
Table 6-3c Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 37 6-15
Table 6-4a Summary of SIC Code 33 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R 6-18
Table 6-4b Summary of SIC Code 36 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R 6-19
Table 6-4c Summary of SIC Code 37 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R 6-19
Table 6-5 Comparison of Differences Between Facility Estimates and
Site Surveyor Estimates Across the SIC Codes 6-22
Table 6-6 Percent Difference of Facility and Site Surveyor Estimated Total Release and
Other Waste Management Quantities as Reported in Sections 5 and 6 of the
Form R for Reporting Years 1996, 1995, 1994,1988, and 1987 6-25
-------
LIST OF TABLES (Continued)
Table 6-7a Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R by the Facilities Surveyed
to the TRI Database, SIC Code 33 6-26
Table 6-7b Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 36 6-27
Table 6-7c Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 37 , 6-27
Table 6-8a On-Site Waste Management for SIC Code 33 6-30
Table 6-8b On-Site Waste Management for SIC Code 36 6-30
Table 6-8c On-Site Waste Management for SIC Code 37 6-30
Table 6-9 Estimation Method Used by Facilities and Surveyors to Calculate
Production Ratio/Activity Index 6-33
Table 6-10 Percent of Time Surveyor Agreed with Facility Basis of
Production Ratio Estimate 6-35
Table 6-11 Frequency With Which Facilities Claimed Source Reduction Activities .... 6-37
Table 6-12 Source Reduction Activities Claimed by the Surveyed Facilities 6-39
Table 6-13 Errors in Source Reduction Activity Classifications 6-40
Table 7-1 Number of Employees at Visited Facilities 7-2
Table 7-2 Types of Personnel Completing the Form Rs 7-3
Table 7-3 Common References Used to Complete the Form Rs 7-4
Table 7-4 Number of Hours Required to Complete All Form Rs at Surveyed Facilities . 7-6
Table 7-5 Average Number of Hours Needed to Complete Each Form R 7-6
Table 7-6 Comments on the Form R Chemical Specific Instructions 7-10
Table 8-1 Recommendations for Avoiding Errors in Threshold Determinations 8-4
Table 8-2 Recommendations for Avoiding Errors in Identifying Release and
Other Waste Management Activity Types and Sources 8-5
-------
LIST OF FIGURES
Figure 2-1
Figure 3-1
Figure 3-2
Figure 3-3
Figure 3-4
Figure 3-5
Figure 3-6
Figure 4-1
Figure 4-2
Figure 4-3
Figure 5-la
Figure 5-lb
Figure 5-lc
Figure 5-Id
Figure 5-le
Figure 5-If
Figure 5-lg
Figure 5-2
Figure 5-3
Figure 6-la
Approach Used to Perform the RY 1996 EPCRA Section 313
Site Survey Program
Iron and Steel Manufacturing Process Overview Using a EOF
Iron and Steel Manufacturing Cokemaking and Iron Making
Bayer Process (Alumina Refining)
Overall Process Flow Diagram - Semiconductor Manufacturing . .
General Foundry Flow Diagram
Car Painting Process
Approaches Used by Facilities to Make Threshold Determinations
Accuracy of Threshold Determinations by SIC Code
. 2-2
. 3-4
. 3-5
3-11
3-18
3-32
3-38
. 4-4
. 4-8
Comparison of Chemical Activity Classifications
Made by Facilities to Those Made by Reviewers
Distribution of Release and Other Waste Management Activity Types
(RY 1996)
Distribution of Sources for Fugitive Releases
Distribution of Sources for Stack Releases (RY 1996)
Distribution of Sources for Surface Water Discharges (RY 1996)
Distribution of Sources for Land On-Site (RY 1996)
Distribution of Sources for Discharges to POTWs (RY 1996) ...
Distribution of Sources from Off-Site Transfers for Release and Other
Waste Management Activities
Misclassified or Overlooked Types of Release and Other Waste
Management Activities
Misclassified and Overlooked Sources of Release and Other Waste
Management Activities
4-14
. 5-6
. 5-7
. 5-8
. 5-9
5-10
5-11
5-12
5-15
5-20
Comparison of Estimates of Total Release and Other Waste Management
Quantities as Reported in Sections 5 and 6 of the Form R in SIC Code 33
.. 6-6
-------
LIST OF FIGURES (Continued)
Figure 6-lb Comparison of Estimates of Total Release and Other Waste Management
Quantities as Reported in Sections 5 and 6 of the Form R in SIC Code 36 ... 6-7
Figure 6-lc Comparison of Estimates of Total Release and Other Waste Management
Quantities as Reported in Sections 5 and 6 of the Form R in SIC Code 37 ... 6-8
Figure 6-2a Frequency of Agreement Between Facility and Surveyor Estimates by
Release and Other Waste Management Activity Type for SIC Code 33 6-9
Figure 6-2b Frequency of Agreement Between Facility and Surveyor Estimates by
Release and Other Waste Management ActivityType for SIC Code 36 6-10
Figure 6-2c Frequency of Agreement Between Facility and Surveyor Estimates by
Release and Other Waste Management Activity Type for SIC Code 37 6-11
Figure 6-3 Comparison of Facility and Site Surveyor Estimates of Total Release
and Other Waste Management Quantities as Reported in Sections 5 and 6
of the Form R 6-23
Figure 6-4 Estimation Method Used by Facilities and Surveyors to Calculate PR/AI ... 6-34
Figure 7-1 Time Needed to Complete All Form Rs at Survey Facilities (SIC Codes 33,
36, and 37) 7-7
Figure 7-2 Percent of Facilities Calling the Hotline by Industry 7-9
Figure 7-3 Percent of Facilities Using the Automated Form R 7-13
-------
EXECUTIVE SUMMARY
As part of a continuing effort to assess and improve the quality of the data
contained in the Toxic Release Inventory (TRI) database, the U.S. Environmental Protection
Agency (EPA) conducted TRI data quality site surveys for reporting year (RY) 1996. The goals
of these site surveys were to:
• Provide a quantitative assessment of the accuracy of TRI data;
• Identify ways to improve the TRI data collection process; and
• Identify where further guidance on the completion of the TRI report forms
(EPCRA Section 313 reports) would be beneficial. :
This summary presents the significant findings and conclusions from the site surveys conducted.
This report focuses on surveys completed for RY 1996. Previous reports have
presented findings from site surveys for RYs 1987, 1988, 1994, and 1995. For RY 1996, site
surveys were completed at 60 facilities:
• 27 facilities in SIC Code 33, the primary metals industry;
• 14 facilities in SIC Code 36, the electronic and other electrical equipment
industry; and
• 19 facilities in SIC Code 37, the transportation equipment industry,
Accuracy of TRI Data
Evaluation of the methodologies used by facilities and accuracy of these methods
in completing threshold determinations and release and other waste management estimates
provide a basis to assess the accuracy of the TRI data. The following information presents the
approaches used by facilities and the accuracy of those approaches for completing threshold
determinations and release and other waste management estimates.
-------
EXECUTIVE SUMMARY
100%
SIC Code 37
Purchase or
Inventory Records Process Recipes
Assumed
Threshold Monitoring Data
Exceeded Production Data
Approach Used to Determine Thresholds
tafotmttloa for this /fcure Is discussed In Section 4.2.
Approaches Used by Facilities to Make Threshold Determinations
Facilities primarily use purchase or inventory records, or process recipes to make
threshold determinations.
Facilities in the electronic and other electrical equipment industry (SIC Code 36)
and primary metals industry (SIC Code 33) use production data more frequently
than those facilities in the transportation equipment industry (SIC Code 37).
Facilities in the primary metals industry (SIC Code 33) and the electronic and
other electrical equipment industry (SIC Code 36) are more likely to assume
thresholds are exceeded.
n
-------
EXECUTIVE SUMMARY
SIC Code 36
0%
Facility Did Not Submft
an EPCRA Section 313
Report fora Chemical
that Old Not Exceed a
Threshold
SIC Code 33
Facility Submitted
an EPCRA Section 313
Report fora Chemical
thatExceeded a
Threshold
Facility Submitted
an EPCRA Section 313
Report fora Chemical
that Did Not Exceed a
Threshold
Outcome
FacilrtyDidNotSubmit
an EPCRA Section 313
Report for a Chemical
that Exceeded a
Threshold
Note: Tne first two outcomes represent cases in which facilities correctly determ'ne thresholds, while the last two outcomes represent
cases in which facilities incorrectly deterrrine thresholds.
Data for this figure can be found in Table 4-1.
Accuracy of Threshold Determinations
Considering all EPCRA Section 313 chemicals used on-site, facilities determine
thresholds correctly 95% of the time.
Considering only EPCRA Section 313 chemicals that actually exceeded
thresholds, facilities accurately identified the threshold exceedences 88% of the
time.
in
-------
EXECUTIVE SUMMARY
Type
Datt tor this Hgtm can ba found on Table 5-1.
SIC Coda 37
SIC Code 36
SIC Code 33
Distribution of Release and Other Waste Management Activity Types
Fugitive and stack releases and off-site transfers for further waste management
activities were observed at most facilities in all industry sectors.
No facilities in any industry sector visited had on-site underground injection.
Most facilities in the electronic and other electrical equipment industry (SIC Code
36) had releases to Publicly Owned Treatment Works (POTW)s.
IV
-------
EXECUTIVE SUMMARY
Type
Data for this figum can be found In Table 5-2.
Misclassified and Overlooked Types of Release and Other Waste management Activities
Facilities often correctly identified release and other waste management activities
that were occurring, but reported them to the wrong type (particularly between
stack and fugitive releases and between various off-site transfers for further waste
management activities).
On-site treatment was misclassified or overlooked frequently in the primary
metals industry (SIC Code 33) because facilities were confused as to whether air
pollution control devices collecting metal should be reported as on-site treatment.
Recycling, both on and off site, was frequently misclassified due to confusion
over the definitions of recycling and reuse.
Off-site disposal was frequently overlooked at facilities using EPCRA Section
313 Chemicals (typically metals) that were present in dust collected in baghouses,
electrostatic precipitators, and rotoclones.
-------
EXECUTIVE SUMMARY
On-Slte Releases and Transfers to POTWs
SIC Code 33
(FacJities-27,
Form Rs=74)
SIC Code 36
(FacJities=14,
Fonn Rs=48)
SIC Code 37
(Facili(ics=19,
Form Rsa64)
On-Slte Waste Management Quantities
SIC Code 33
(Facil«les=27,
FormRs-74)
Transfers Off-Site for Further Waste Management (excluding POTWs)
SIC Code 36
(Facilitte=14,
Form Rs=48)
SIC Code 37
(Facilitles=19,
Form Rs=64)
| D Facility H Site Surveyor |
Facility and site surveyor estimates agreed for the most part for the transportation
equipment industry (SIC Code 37).
Facility and site surveyor release and other waste management estimates for the
primary metals industry (SIC Code 33) and the electronic and other electrical
equipment industry (SIC Code 36) differed significantly, primarily due to the
difference in recycling estimates. This difference is attributed to industry
confusion over definitions of recycling and reuse.
Facilities in the primary metals industry (SIC Code 33) processed millions of tons
of metals. Thus, the releases, transfers, and other waste management quantities
from this industry are much higher than the releases and transfers from the other
industries visited.
VI
-------
EXECUTIVE SUMMARY
TRT Data Collection Process
The following section provides a summary of findings on the mechanics of the TRI data
collection process. Time required to prepare EPCRA Section 313 reports, feedback on usage of
the automated Form R, and usage of the EPCRA Hotline are presented.
100%
SIC Code 36
21-40
Time Estimate (Hours) "'"'uu
May not add up to 100% because not all facilities reported the time estimate
Time Needed to Complete all Form Rs for Facilities Visited in SIC Codes 33,36, and 37
The average number of hours to complete each Form R across all three SIC Codes
visited is 18 hours, less than the 43 hours per Form R estimated by EPA.
It takes more time per Form R for a facility to complete one Form R, compared to
multiple Form Rs. Facilities filing only one Form R took an average of 22 hours
to complete it, while facilities filing more than one Form R took an average of 16
hours per Form R.
Vll
-------
EXECUTIVE SUMMARY
Did not Use APR
37%
Used APR (found it helpful)
42%
Used APR (found it not
helpful)
21%
Percent of Facilities Using the Automated Form R
Of the facilities using the Automated Form R (AFR), approximately two-thirds
found it helpful.
Use of the AFR sho'uld eliminate data entry errors and thereby increase the
accuracy of data in the TRI database. However, site surveyors found several
instances where data listed hi the TRI database did not match the AFR submitted
by the facility. The nature of these errors should be fixed within the AFR system
and corrected.
Facilities commented that they would use the AFR more often if certain printing
problems, computer compatibility issues, and cross-section linkage problems were
resolved.
vm
-------
EXECUTIVE SUMMARY
SC Code 33
SIC Code 36
SIC Code 37
Percent of Facilities Calling the Hotline by Industry
Facilities in the transportation equipment industry (SIC Code 37) called the
hotline 68% of the time, whereas the facilities in SIC Code 36 and SIC Code 33
called the hotline 57% and 48%, respectively. Facilities in the transportation
equipment industry also made the least number of incorrect threshold
determinations and release and other waste management estimates out of the three
industries surveyed. While direct linkage is not certain, this finding suggests that
contact with the EPCRA Hotline improves reporting.
IX
-------
EXECUTIVE SUMMARY
Further TRI Guidance
The following section discusses areas where additional instructions or guidance would improve
EPCRA Section 313 reporting and increase the accuracy of the TRI database. The areas
highlighted in this summary were commonly found in all of the industries surveyed.
Recommendations for Improving Accuracy in
in EPCRA Section 313 Reporting
Area for Improvement
Recommendation for Improvement in
Future TRI Reporting Years
Overlooked container residue
Instructions and guidance materials should
emphasize that even a "RCRA empty"
container is expected to contain a residual
(possibly up to two inches) and that it must
be considered for TRI reporting. Also, note
that on-site container rinsing and disposal of
the rinsate will result in a water discharge.
Overlooked acid aerosol manufacturing
Instructions and guidance materials should
indicate that if H2SO4 or HC1 is used
anywhere in the plant as an aerosol,
regardless of whether the process is enclosed
or not, their usage should be applied to the
threshold determination and release and other
waste management calculations.
Incorrectly reporting disposition for off-site
transfers
Instructions and guidance materials should
emphasize that facilities should attempt to
determine the type of receiving facility that is
accepting the transfers and exactly how the
material sent is being managed (or directly
reused) by the receiving facility.
Definitions of recycling versus reuse
Provide a definition of recycling and include
examples of streams that can be considered
as being recycled in Sections 7 and 8 of the
TRI reporting instructions. A good example
would be handling of used metals or metal
compounds. Specifically, a discussion of
what waste management activity must be
applied to a used metal for it to be considered
recycled versus reused, would be helpful.
-------
EXECUTIVE SUMMARY
Area for Improvement
Recommendation for Improvement in
Future TRI Reporting Years
Definitions of source reduction
Consider shortening the list of codes for
source reduction and providing definitions
for each code.
Clarification of on-site treatment of waste
streams containing metals
Provide clarification of on-site treatment
definitions pertaining to waste streams
containing metals. Guidance is needed to
instruct how to report collection of metals in
an air pollution control device, whether a
treatment efficiency or collection efficiency
should be reported in Section 7A, and what to
put in Section 8 of the Form R in this
scenario.
Clarification of the treatment definitions in
Sections 7 and 8 of the Form R for organic
and inorganic chemicals. Section 7A applies
to the waste stream containing the toxic
chemical while Section 8 applies to the toxic
chemical itself. These differences have
caused confusion when reporting. Confusion
occurs when chemicals go through a
treatment system but are not destroyed.
Facilities need direct guidance in the TRI
reporting instructions to claim zero
efficiency, and then what to put in Section 8
(zero or NA).
Evaluation of common references used and the percentage of facilities that attended EPA training
workshops assists in characterizing the usage of existing support materials for TRI reporting.
The following illustrates the level of usage of support materials.
XI
-------
EXECUTIVE SUMMARY
Common References Used to Compile Form Rs for RY1996
Most facilities surveyed use the Form R instructions and material safety data
sheets as their main references in compiling Form Rs.
Over half of the facilities visited in the transportation equipment industry (SIC
Code 37) attended EPA-sponsored training workshops. It is noteworthy that
threshold determinations and release and other waste management estimates in
this sector were the most accurate of the industry sectors surveyed.
Many facility contacts and a trade association representative in the primary metals
industry (SIC Code 33) requested that EPA sponsor a training workshop
specifically for their industry. In addition, facility contacts expressed an interest
in a TRI guidance document specific to foundry operations, and another document
specific to reporting for metals. Such efforts would improve accuracy,
particularly if focused on metal versus metal compounds, de minimus
applicability, the definitions of recycling versus reuse, and the reporting of metals
collection/treatment.
xn
-------
1.0
INTRODUCTION
Section 313 of the Emergency Planning and Community Right-to-Know Act
(EPCRA) requires the U.S. Environmental Protection Agency (EPA) to collect information from
manufacturers, processors, and users of listed toxic chemicals on release and other waste
management activities of those chemicals to the environment. To collect such information,
EPCRA requires that EPA implements a yearly reporting requirement from such facilities.
EPCRA Section 313 reports, referred to as Form R/A reports, are due to EPA by July 1 each year
to fulfill the reporting requirement for the previous calendar year. The reporting requirement was
first implemented for the 1987 calendar year. The data from the Form R chemical reports are
compiled in EPA's Toxics Release Inventory (TRI) database for future analysis, distribution, and
evaluation. TRI data, as mandated by EPCRA, are made available to the public. The
information collected under EPCRA Section 313 is useful in helping inform the general public
and the communities surrounding affected facilities of release and other waste management
activities of toxic chemicals, assisting the Agency to focus their research into the effects and
control of toxic substances, and aiding in the development of regulations, guidelines, and
standards.
The study discussed in this report reviewed data from the 1996 reporting year (RY
1996) to provide a quantitative assessment of the accuracy of the data collected, to identify areas
in the TRI data collection process that could be improved, and to disseminate further guidance on
the completion of EPCRA Section 313 reports.
The study was conducted by performing on-site surveys of TRI information at 60
randomly chosen facilities:
• 27 facilities in SIC Code 33, primary metals industry;
• 14 facilities in SIC Code 36, electronic and other electrical equipment
industry; and
• 19 facilities in SIC Code 37, transportation equipment industry.
This report presents the data gathered for RY 1996 and compares it to data from similar studies
completed for previous reporting years, where appropriate.
1-1
-------
1.1
This data quality report is organized as follows:
• Section 1 - Introduction, including site survey objectives;
• Section 2 - Sample selection and approach;
• Section 3 - Description of industrial processes surveyed;
• Section 4 - Threshold determinations made by the facilities;
• Section 5 - Sources and types of release and other waste
management activities;
• Section 6 - Quantity of release and other waste management activities;
• Section 7 - Preparation of the Form R; and
• Section 8 - Conclusions and recommendations.
EPA's Overall Quality Assurance Program
Because of the wide audience and many intended uses of the TRI data, EPA
designed and implemented a program to assess the quality of the data collected under EPCRA
Section 313 and to identify areas where better guidance materials would be useful for improving
the accuracy of future reported data. The site surveys described in this report are a component of
EPA's overall quality assurance program.
1.2
Site Survey Objectives
EPA's site surveys were designed to provide a quantitative assessment of the
accuracy of the data submitted for a calendar year by identifying the frequency and the
magnitude of errors in the Form R data and the reasons these errors occurred. EPA believed that
on-site review of industrial processes, pollution control technologies, and documentation
supporting the Form R reports would reveal errors in the database not obvious from review of a
facility's Form R submissions. Expected error types included overlooked chemicals, incorrectly
included chemicals, and errors in the release and other waste management quantity estimate
calculations. The goal of the surveys was to obtain information that could be used to improve
1-2
-------
the Form R reporting instructions and definitions, and associated guidance materials, and thus
improve the quality of data in the TRI database in future years.
Anyone who uses the results of the site survey program, as well as the TRI
database itself, should be aware of a basic limitation of the EPCRA reporting process. Under
EPCRA, facilities are not required to perform any additional monitoring or measurement of the
quantity of toxic chemicals released to the environment to calculate Form R release and other
waste management quantity estimates. Therefore, the methods selected by facilities to estimate
release and other waste management quantities depend on the nature of the best readily available
data. The accuracy of release and other waste management quantity information reported to
EPCRA therefore depends on the accuracy of the best readily available data and any subsequent
estimation methodologies.
At facilities where supporting data were available, site surveyors carefully
examined the facility's estimation calculations and data sources and then recalculated the
estimates. In many instances, the site surveyors were able to identify data sources overlooked by
facility personnel. These new data were then used to recalculate release and other waste
management quantity estimates during the site visits. However, site surveyors did not conduct
any monitoring or measurements during the site visits. Site surveyors also assessed the quality
of the estimation methods by recalculating release and other waste management quantities using
alternative approaches where more accurate estimation methods were appropriate or where
available data warranted the use of alternative approaches.
1.3
Prior Site Survey Efforts
EPA has conducted five sets of quality assurance site surveys since the first
submittal of Form Rs from industry. The RY 1987 site surveys covered all SIC Codes affected
by the EPCRA Section 313 requirements. The RY 1988 site surveys covered SIC Codes 28,291,
and 34 - 38. The RY 1994 site surveys covered SIC Codes 25, 281, 285, and 30; RY 1995 site
surveys covered SIC Codes 26 and 286.
The results of the site surveys help EPA identify ways additional guidance can be
structured to improve the overall quality of the data generated under EPCRA reporting.
1-3
-------
-------
2.0
APPROACH
The RY 1996 site survey program was conducted using the following steps:
(1) Revising the survey instrument;
(2) Selecting facilities to be visite4 (Sample selection);
(3) Training site surveyors (Training);
(4) Arranging site visits;
(5) Performing site visits (Site visit methodology);
(6) Data management and data quality assurance; and
(7) Data analysis and reporting.
This approach is presented schematically in Figure 2-1, and each of these steps is discussed in
the following subsections. The approach was originally established for the RY 1987 and RY
1988 site survey programs to ensure consistency in conducting site surveys and accuracy of the
results. It was improved for each of the RY 1994, RY 1995, and RY 1996 site survey programs
based on experience and lessons learned from the previous programs.
2.1
Survey Instrument
The survey instrument, shown in Appendix A, was designed to standardize and
facilitate the review of threshold determinations, and the calculations used to assess release and
other waste management activities at facilities. The engineers and scientists who performed the
site surveys used the survey instrument as a detailed checklist to ensure that all pertinent items
were reviewed. The survey instrument also provided a consistent format for recording both the
data collected during site surveys and the errors made by facility personnel on their Form R
reports.
2-1
-------
Revise the Survey Instrument
Define the Goals
for the Sample
Selection Process
Select a Random
Group of Facilities
for the Sample
Obtain Facility's Voluntary
Participation and Arrange
Site Visits
Perform the Site Visits
and Complete the Survey
Instrument for Each Site
Develop a Computerized
Data Management System
Internally Check the
Consistency of the Completed
Survey Instruments
Revise the Training Manual
Train Site Surveyors
Prepare for Each Site Visit
Hold an Informal Wrap-up
Meeting to Discuss
Qualitative Items
Enter and Verify the
Data in the Data
Management System
Summarize the Results
of the Wrap-up Meeting
in a Memorandum
Generalize the Results
Using Facility Weights
REPORT RESULTS
Figure 2-1. Approach Used to Perform the RY 1996 EPCRA Section 313
Site Survey Program
2-2
-------
In addition to its primary focus on chemical-specific information, the survey
instrument contains questions regarding the usefulness of the reporting instructions, EPCRA
Section 313 hotline, and the other published guidance materials. Each section of the survey
instrument focuses on identifying specific types of errors made by facility personnel on their
Form R submittals.
The RY 1996 data quality site surveyors used a revised version of the survey
instrument from previous site survey programs. Most of the questions remained the same, but
some additional questions concerning available documentation, possible sources for threshold
determinations, source reduction activities, pollution prevention technology, and use of the Form
A were added to clarify information received during the site visits and to assess the usefulness of
the new guidance and materials that are now available. The increments for amount of time
needed to complete all Form Rs at the facility were adjusted slightly in the RY 1996 survey
instrument to obtain a more precise estimate of time needed. The formafwas also revised to
make the survey instrument easier for the site surveyors to use.
2.2
Sample Selection
For RY 1996, a total of approximately 64,000 Form R reports and 7,000 alternate
threshold certification statements, referred to as Form A reports, covering all SIC Codes required
to report toxic chemicals, were submitted to EPA by approximately 22,000 facilities and entered
into the TRI database. At the time the site surveys for RY 1996 were conducted, the following
number of Form Rs and Form As had been submitted and incorporated in the TRI database for
the selected SIC Codes:
SIC Code
33
36
37
Number of Facilities
1,902
1,233
1,248
Number of Form Rs
6,138
3,031
4,117
Number of Form As
465
90
214
2-3
-------
The primary objective of the sample selection approach was to obtain a random
group of facilities from the key industry groups within the specific SIC Codes being surveyed to
scale up appropriately the results to reflect the reporting of the entire SIC Code group. The RY
1996 site visit program targeted 60 completed site visits at facilities in SIC Codes 33, 36, and 37.
These SIC Codes represent the primary metal industries, electronic and other electrical
equipment and components, and transportation equipment, respectively. The SIC Codes
ultimately visited were further defined as the three-digit SIC Codes 331, 332, 333, 334, 335, 367,
369,371, and 372. The other three-digit codes within SIC Codes 33, 36, and 37 were not
targeted by the survey so as to increase the confidence level of the sample by decreasing the
target population size. The three-digit codes selected have the most Form Rs and facilities
within the targeted two-digit SIC Code.
The number of facilities visited in each three-digit SIC Code was based on a
statistical weighting technique, calculated as the product of the total number of targeted facilities
(60) and the percentage of facilities in each three-digit SIC Code (with respect to the total
number of facilities in all the three-digit SIC Codes listed above). The table below shows the
number of facilities visited within each three-digit SIC Code.
SIC SIC
Code. Description
331 Steel works, blast furnaces, and rolling
and finishing mills
332 Iron and steel foundries
333 Primary smelting, refining of
nonferrous metals
334 Secondary smelting, refining of
nonferrous metals
335 Rolling, drawing, extruding of
nonferrous metals
367 Electronic components and accessories
369 Miscellaneous electrical machinery,
equipment, and supplies
371 Motor vehicles and equipment
372 Aircraft and parts
Number of
Facilities
Visited
8
7
1
11
3
15
4
2-4
-------
Only facilities with 15 or fewer Form Rs were considered candidates for site visits
due to time and budget considerations. Site visits to facilities with more than 15 Form Rs would
have taken considerable time, and would have limited the total number of facilities that could be
visited. Since the same facility personnel may complete multiple reports at a given facility,
visiting more facilities gives a better indication of the range of reporting practices.
2.3
Site Surveyor Selection and Training
The engineers and scientists who staffed the site survey program were selected on
the basis of their experience in performing environmental audits of industrial processing
facilities, and thus were required to have a thorough understanding of chemistry, technical
calculations, multimedia environmental concerns, and pollution control technologies.
A surveyor training program was developed to ensure consistency and high
quality work among all site surveyors. The training program consisted of three steps:
2.4
1) Compiling a comprehensive training manual, including copies of EPA
guidance documents and other references;
2) Holding training sessions to familiarize project personnel with program
requirements; and
3) Conducting standardized reviews of the completed survey instruments
with feedback to the site surveyor to maintain a consistent approach
among all surveyors.
Arranging Site Visits
The goal in arranging site visits was to provide each facility in the sample with an
equal opportunity to participate in the site visit program. Participation was voluntary; the
facilities were not legally required to participate. A key factor encouraging voluntary
participation was the assurance that the facilities would remain anonymous. Names, location,
and all other facility identification data are shielded from EPA. Upon facility request, the
contractors performing the site visits signed a written confidentiality agreement.
2-5
-------
As a first step, introduction letters were sent to each facility's technical contact,
and where appropriate, to each facility's senior management official (a sample of a letter is
provided in Appendix B). This letter explained the purpose of the quality assessment program
and the anticipated burden and benefits to the facility. They also provided assurance to the
facility that all facility-specific data would be treated as confidential. EPA's contractors then
called the technical contacts at the facilities to solicit their participation, to arrange a date for the
site visit, and to review a preliminary agenda for the site visit.
2.5
Conducting Site Visits
The goal of the site visit was to collect all the information needed to complete the
survey instrument accurately, while minimizing burdens on facility staff. On-site survey
activities included tours of the facilities, which focused on material storage areas, industrial
processing operations, and pollution control equipment; careful review of all readily available
documentation, which could include MSDSs, production data, monitoring data, purchasing
records, and facility spreadsheets or computer software with this information; and interviews
with appropriate facility employees regarding documentation materials. Site surveyors did not
perform any monitoring or measurements during the site visits.
The site visits were designed to determine:
• Overlooked chemicals;
• Release and other waste management activities;
• Errors in the Form R reports submitted to EPA;
• Whether more accurate release estimation methods could have been used,
based on information available to the facilities; and
• Whether further EPA guidance is needed on certain issues for that
particular industry.
Release and other waste management estimates were recalculated or recreated by site surveyors
from available documentation during the visit. Site surveyors recorded these results on the
survey instrument and reviewed the results with facility personnel before leaving the site. Site
2-6
-------
surveyors held a wrap-up meeting at the facility with the person who filled out the Form R
reports at the end of the visit to discuss the facility contact's issues or questions and to review the
conclusions and recommendations of the site surveyor. Site surveyors may have called the
facility contact after the on-site visit occurred to discuss any remaining issues that required
clarification or additional research.
2.5.1
Data Collection
Site surveyors reviewed 191 Form R chemical reports and 605 additional
chemicals with amounts used or activities that did not meet the reporting criteria at the 60
facilities visited for RY 1996. Site surveyors reviewed threshold determinations, and release and
other waste management estimates separately to identify the frequency, magnitude, and sources
of errors in these areas. Site surveyors followed the steps described in the Form R reporting
instructions for completing threshold determinations and release and other waste management
estimates. The Form R reporting instructions state that facilities must first assess which
chemicals are manufactured, processed, or otherwise used in excess of appropriate thresholds.
Facilities must then estimate and report all release and other waste management quantities of
listed chemicals exceeding thresholds.
2.5.2
Threshold Determinations
Facilities may make the following types of errors in determining which chemicals
at their site meet a EPCRA Section 313 thresholds:
• Overlooking a chemical;
• Incorrectly calculating a threshold amount;
• Incorrectly applying de minimus exclusions;
• Incorrectly applying an exemption; and
• Misclassifying a chemical activity.
To identify errors in threshold determinations, site surveyors looked for problems
in a facility's documentation. On the plant tour, site surveyors looked for evidence of chemicals
that were reported but should not have been reported, and for evidence of chemicals that were not
2-7
-------
reported but should have been reported. The site surveyors reviewed each facility's
documentation to track the decision process used to determine whether a chemical should have'
been reported. Furthermore, site surveyors used all available documentation to recalculate
threshold estimates for reported chemicals and for chemicals present but not reported to verify
the accuracy of facility calculations.
2.5.3
Release and Other Waste Management Estimates
Facilities most commonly make the following types of errors in calculating
release and other waste management estimates for EPCRA Section 313 chemicals:
• Overlooking a chemical;
• Overlooking a source of data;
• Incorrectly calculating a release or other waste management quantity; and
• Incorrectly interpreting the reporting instructions.
During the site survey program, site surveyors used a two-part approach to identify facility
errors in estimating release and other waste management activities. First, site surveyors always
recalculated release and other waste management quantities using the same technical approach
used by the facility. Second, whenever the site surveyor's experience and training indicated that
a calculation approach different than the one used by the facility was appropriate, the surveyor
attempted to obtain the data needed to calculate release and other waste management quantities
using the more appropriate approach. In many such instances, data were not readily available
during the site visit to recalculate these amounts using the alternative approach. In the cases
where site surveyors were able to recalculate release and other waste management amounts using
alternative approaches, they were able to assess the reasonableness of the estimation techniques
used by facility personnel.
The surveyors quantified all numerical differences between the facility's estimates
and the recalculated values, even in instances where surveyors identified only small differences.
As discussed later, these numerical differences were used to assess quantitatively the accuracy of
the total aggregate release and other waste management quantities contained in the TRI database.
2-8
-------
2.6
Data Management/Data Quality Assurance
Many steps were taken to ensure the data quality of the surveyor's estimates and
to verify the data in the survey database. This section outlines the procedures used to review the
survey instruments after they were completed by the site surveyor, the database system, and the
data entry into the master database; describes the verification procedures for the data entered into
the database; presents the data weighting used to apply the results to the entire population of
facilities for each SIC Code surveyed; and discusses potential sources of error in the site survey
program.
2.6.1
Quality Review of Survey Instrument and Data Entry
All survey instruments were reviewed twice by a set of reviewers to ensure site
survey calculations and methodologies used were correct and consistent for all site surveys. All
data were double entered into the database, compared to each other, and then verified with the
actual survey if an inconsistency was found. Project staff also reviewed the database entries for
internal consistency and completeness by comparing responses to various questions as
appropriate.
2.6,2
Data Weighting
To allow EPA to compare the site survey program results to the TRI database for
the SIC Codes surveyed, weighting factors were applied to the site visit data. These factors or
"weights" represent the number of facilities in the TRI database represented by each of the
surveyed facilities. Each facility was randomly selected and the number of facilities chosen from
each three-digit SIC Code was proportional to the percentage of facilities in any three-digit SIC
Code; therefore, each facility was weighted equally.
The sample facilities weights in each SIC Code group are summed to represent
the total population of facilities included in the TRI database for that SIC Code group. A total
population of 1,492 facilities for SIC Code 33 (covering SIC Codes 331, 332, 333, 334, and
335), 781 facilities for SIC Code 36 (covering SIC Codes 367 and 369), and 1,034 facilities for
SIC Code 37 (covering SIC Codes 371 and 372) represents the TRI data for RY 1996.
2-9
-------
2.6.3 Limitations of the Analysis
The design and implementation of the survey may have introduced unavoidable
inaccuracies in the study results. The three primary sources of error are:
• Sample selection bias;
• Survey implementation; and
• Data reduction and analysis.
The relatively small number of facilities sampled introduced a sample selection
bias. The smaller the number of facilities sampled, the greater the likelihood that these facilities
do not accurately represent the universe of reporting facilities. For the selected sample size of 60
facilities in SIC Codes 33, 36, and 37, the 90% confidence interval is plus or minus 11%. That
is to say, if 50% of the visited facilities reported accurate data, a 90% probability exists that
between 39% and 61% of the facilities in the national database reported accurate data.
Another possible source of error concerns the fact that approximately 15 different
surveyors performed the survey. This source of inaccuracy was controlled to the extent possible
by the use of a carefully designed survey instrument and extensive quality assurance provisions.
Nevertheless, it is possible that different surveyors made different judgments in the course of the
site surveys.
Finally, certain assumptions were made to simplify data analysis. The key
assumption was that the facilities and Form Rs examined in the site visits accurately represent all
facilities in their SIC Code group in terms of the accuracy of the data submitted. Aside from
possible errors introduced by the relatively small size of the sample, the sampled facilities may
not fully represent their SIC Code group because the sampled facilities excluded any facility with
more than 15 Form Rs for budgetary reasons. In addition, we have observed that, in general,
facilities that volunteer to participate in the program are smaller in size and have less throughput
than facilities that decline to participate. To the extent that facilities submitting more than 15
Form Rs or larger facilities with more throughput report more (or less) accurate data than the
sampled facilities, the facilities sampled may not fully reflect the universe of facilities in the
database.
2-10
-------
2.7
Data Analysis and Reporting
Once the results of the site surveys were loaded into a database and the database
was validated through the quality assurance process described above, the data were evaluated to
discern trends in the quality of data in the TRI forms. The results of that analysis are presented
in the following sections.
2-11
-------
-------
3.0
DESCRIPTION OF INDUSTRIAL PROCESSES SURVEYED
This section describes the major industrial processes seen on the site visits in SIC
Codes 33, 36, and 37. The manufacturing processes described include:
• Iron and Steel Manufacturing at facilities in SIC Code 33 (Section 3.1);
• Primary aluminum processing at facilities in SIC Code 33 (Section 3.2);
• Primary zinc processing at facilities in SIC Code 33 (Section 3.3);
• Semiconductor manufacturing at facilities in SIC Code 36 (Section 3.4);
• Printed board manufacturing at facilities in SIC Code 36 (Section 3.5);
• Motor vehicle equipment manufacturing at facilities in SIC Code 37
(Section 3.6); and
• Motor vehicle painting/finishing at facilities in SIC Code 37 (Section 3.7).
The section is designed to provide a general understanding of the industries
surveyed, and to provide an overview of the inter-relationship between the industrial process,
pollutant outputs, and pollution prevention opportunities within those industries.
This section specifically contains a description of commonly used production
processes, associated raw materials, the byproducts produced, and the materials either recycled or
transferred off-site for further waste management. This discussion, coupled with schematic
drawings of the identified processes, provides a concise description of where wastes may be
produced in the process. This section also describes the potential fate (air, water, land) of these
waste products.
3-1
-------
3.1
Iron and Steel Manufacturing and Processing
Steel is an alloy of iron usually containing less than one percent carbon. Steel
production occurs in several sequential steps as shown in Figure 3-1. The two types of
steelmaking technology in use today are the basic oxygen furnace (EOF) and the electric arc
furnace (EAF). Although these two technologies use different input materials, the output for
both furnace types is molten steel which is subsequently formed into steel mill products. The
EOF input materials are molten iron, scrap, and oxygen. In the EAF, electricity and scrap are the
input materials used. BOFs are typically used for high tonnage production of carbon steels,
while EAFs are used to produce carbon steels and low tonnage alloy and specialty steels. The
processes leading up to steelmaking in a EOF are very different than the steps preceding
steelmaking in an EAF; the steps after each of these processes used to turn molten steel into steel
mill products are the same.
3.1.1
Steelmaking Using the Basic Oxygen Furnace
The process of making steel in a EOF is preceded by cokemaking and ironmaking
operations. In cokemaking, coke is produced from coal. In ironmaking, molten iron is produced
from iron ore and coke. Each of these processes and the subsequent steelmaking process in the
EOF are described below. Figure 3-1 shows the process overview of steelmaking using a EOF.
3.1.1.1
Cokemaking
Coal processing in the iron and steel industry typically involves producing coke,
coke gas, and by-product chemicals from compounds released from the coal during the
cokemaking process as shown in Figure 3-2. Coke is carbon-rich and is used as a carbon source
and fuel to heat and melt iron ore in ironmaking. In cokemaking, pulverized coal is heated at
high temperatures in the absence of air in batteries of ovens. Volatile byproducts are collected
and processed for other uses. The solid carbon remaining in the oven (coke) is then processed
for ironmaking. The necessary heat for coke distillation is supplied by external combustion of
fuels (e.g., recovered coke oven gas, blast furnace gas) through flues located between ovens. At
the end of the heating cycle, the coke is pushed from the oven into a rail quench car. The quench
3-2
-------
car takes it to the quench tower, where the hot coke is cooled with a water spray. The coke is
then screened and sent to the blast furnace or to storage. ,
In the by-products recovery process, volatile components of the coke oven gas
stream are recovered including the coke oven gas itself (which is used as a fuel for the coke
oven), naphthalene, ammonium compounds, crude light oils, sulfur compounds, and coke breeze
(coke fines). During the coke quenching, handling, and screening operation, coke breeze is
produced. Typically, the coke breeze is reused in other manufacturing processes on-site (e:g.,
sintering) or sold off-site as a by-product.
3.1.1.2
Ironmaking
In the blast furnace, molten iron is produced, as shown in Figure 3-2; Iron ore,
coke, and limestone are fed into the top of the blast furnace. Heated air is forced,into the bottom
of the furnace. The carbon monoxide from the burning of the coke reduces iron ore to iron. The
acid part of the ores reacts with the limestone to create a slag which is drawn periodically from
the furnace. This slag contains unwanted impurities in the ore. When the furnace is tapped, iron
is removed through one set of runners and molten slag via another. The molten iron is tapped .
into refractory-lined cars for transport to the steelmaking furnaces. Residuals from the process
are mainly sulfur dioxide or hydrogen sulfide, which are driven off from the hot slag. The slag is
the largest by-product generated from the ironmaking process and is reused extensively in the
construction industry. Blast furnace flue gas is cleaned and used to generate steam to preheat the
air coming into the furnace, or it may be used to supply heat to other plant processes. The
cleaning of the gas may generate air pollution control dust in removing coarse particulates
(which may be reused in the sintering plant or landfilled), and water treatment plant sludge in
removing fine particulates by venturi scrubbers.
3-3
-------
UJ
a:
o
1
o
QL
UJ
t
1
< ^:
t
QO
1. m::>
* i.Q
0 CCO
^ uj.o:
M
JL tD
©@l_^
pl
r
/T-
^
H,
o
Dl,
3 fc=
A A
CJ)
CONTINUOUS CASTING
O
O
CO
5
CD
UJ
PQ
«
on
.3
5
<5
K)
2
PL,
ISA
a
I
i
%
%
•a
a
C3
c
2
i
3-4
-------
CO
3
Q
LU
•y.
t£
ID
U.
CO
LU
z
u_
111
O
CO
LU
Z
LL
0
5
co
O
-J
CO
LU
Z
LL
LU
2
O
LU
•5
CO
LU
• 3»
c^
J
. .g
k j
O
o
L i
LU
Q:
o
k t
LU
1
LU
^
~*
a:
DC
I
03
•e
a
a
-------
3.1.1.3
Sintering
Sintering is the process that agglomerates fines (including iron ore fines, pollution
control dusts, coke breeze, water treatment plant sludge, and flux) into a porous mass for
charging to the blast furnace. Through sintering operations, a mill can recycle iron-rich material,
such as mill scale and processed slag. Not all mills have sintering capabilities. The input
materials are mixed together, placed on a slow-moving grates and ignited. Windboxes under the
grate draw air through the materials to deepen the combustion throughout the traveling length of
the grate. The coke breeze provides the carbon source for sustaining the controlled combustion.
In the process, the fine materials are fused into the sinter agglomerates, which can be
reintroduced into the blast furnace along with ore. Air pollution control equipment removes the
particulate matter generated during the thermal fusing process. If electrostatic precipitators or
baghouses are used as the air pollution control equipment, the dry particulates captured are
typically recycled as sinter feedstock, or are landfilled as solid waste.
3.1.1.4
Steelmaking
In a batch mode, molten iron from the blast furnace, flux, alloy materials, and
i
scrap are placed in the EOF, melted and refined by injecting high-purity oxygen. A chemical
reaction occurs, where the oxygen reacts with carbon and silicon generating the heat necessary to
melt the scrap and oxidize impurities. Slag is produced from impurities removed by the
combination of the fluxes with the injected oxygen. Various alloys are added to produce
different grades of steel. The molten steel is typically cast into slabs, beams or billets.
The waste products from the basic oxygen steelmaking process include slag,
carbon monoxide, and oxides of iron emitted as dust. Also, when the hot iron is poured into the
furnace, iron oxide fumes are released and some of the carbon in the iron is precipitated as
graphite (kish). The EOF slag can be processed to recover the high metallic portions for use in
sintering or blast furnaces, but its applications as saleable construction materials are more limited
than the blast furnace slag.
3-6
-------
Basic oxygen furnaces are equipped with air pollution control systems for
containing, cooling, and cleaning the volumes of hot gases and sub-micron fumes that are
released during the process. Water is used to quench or cool the gases and fumes to temperatures
at which they can be effectively treated by the gas cleaning equipment. The resulting waste
streams from the pollution control processes include air pollution control dust and water .
treatment plant sludge. The principal pollutants removed from the off-gas are total suspended
solids and metals (primarily zinc, and some lead).
3.1.2
Steelmaking Using the Electric Arc Furnace (EAF)
In the steelmaking process that uses an electric arc furnace (EAF), the primary
raw material is scrap metal, which is melted and refined using electric energy. During melting,
oxidation of phosphorous, silicon, manganese, carbon, and other materials occurs and a slag
containing some of these oxidation products forms on top of the molten metal. Oxygen is used
to decarburize the molten steel and to provide thermal energy. Since scrap metal is used instead
of molten iron, there are no cokemaking or ironmaking operations associated with steel
production that uses an EAF.
The process produces metal dusts, slag, and gaseous products. Particulate matter
and gases evolve together during the steelmaking process and are conveyed into a gas cleaning
system. These emissions are cleaned using a wet or dry system. The particulate matter that is
removed as emissions in the dry system is referred to as EAF dust, or EAF sludge if it is from a
wet system. The composition of EAF dust can vary greatly depending on the scrap composition
and furnace additives. The primary component is iron or iron oxides, and it may also contain
flux (lime and/or fluorspar), zinc, chromium and nickel oxides (when stainless steel is being
produced) and other metals associated with the scrap. Oils are burned off "charges" of oil-
bearing scrap in the furnace. Minor amounts of nitrogen oxides and ozone are generated during
the melting process. The furnace is extensively cooled by water which is recycled through
cooling towers. - . ,
3-7
-------
3.1.3
Forming and Finishing Operations
Whether the molten steel is produced using a EOF or an EAF, to convert it into a
product, it must be solidified into a shape suitable and finished.
3.1.3.1
Forming
The traditional forming method, called ingot teeming, has been to pour the metal
into ingot molds, allowing the steel to cool and solidify. This method has been largely replaced
by continuous casting. The continuous casting process bypasses several steps of the
conventional ingot teeming process by casting steel directly into semifinished shapes. Molten
steel is poured into a reservoir from which it is released into the molds of the casting machine.
The metal is cooled as it descends through the molds, and before emerging, a hardened outer
shell is formed. As the semifinished shapes proceed on the runout table, the center also
solidifies, allowing the cast shape to be cut into lengths.
Process contact water cools the continuously cast steel and is collected in settling
basins along with oil, grease, and mill scale generated in the casting process. The scale settles
out and is removed and recycled for sintering operations, if the mill has a sinter plant. Waste
treatment plant sludge is also generated.
The steel is further processed to produce slabs, strips, bars, or plates through
various forming operations. The most common hot forming operation is hot rolling, where
heated steel is passed between two rolls revolving in opposite directions. The final shape and
characteristics of a hot formed piece depend on the rolling temperature, the roll profile, and the
cooling process after rolling. Wastes generated from hot rolling include waste treatment plant
sludge and scale.
In subsequent cold forming, the cross-sectional area of unheated steel is
progressively reduced in thickness as the steel passes through a series of rolling stands.
Generally, wires, tubes, sheet and strip steel products are produced by cold rolling operations.
Cold forming is used to obtain improved mechanical properties, better machinabiliry, special size
accuracy, and the production of thinner gages than hot rolling can accomplish economically.
3-8
-------
During cold rolling, the steel becomes hard and brittle. To make the steel more ductile, it is
heated in an annealing furnace.
Process contact water is used as a coolant for rolling mills to keep the surface of
the steel clean between roller passes. Cold rolling operations also produce a waste treatment
plant sludge, primarily due to the lubricants applied during rolling. Grindings from resurfacing
of thp worn rolls and disposal of used rolls can be a significant contributor to the plant's
wastestream.
3.1.3.2
Finishing
One of the most important aspects of a finished product is the surface quality. To
prevent Corrosion, a protective coating may be applied to the steel product. Prior to coating, the
surface of the steel must be cleaned so the coating will adhere to the steel. Mill scale, rust,
oxides, oil, grease, and soil are chemically removed from the surface of steel using solvent
cleaners, pressurized water or air blasting, cleaning with abrasives, alkaline agents or acid
pickling. In the pickling process, the steel surface is chemically cleaned of scale, rust, and other
materials. Inorganic acids such as hydrochloric or sulfuric acid are most commonly used for
pickling. Stainless steels are pickled with hydrochloric, nitric, and hydrofluoric acids.
Steel generally passes from the pickling bath through a series of rinses. Alkaline
cleaners may also be used to remove mineral oils and animal fats and oils from the steel surface
prior to cold rolling. Common alkaline cleaning agents include: caustic soda, soda ash, alkaline
silicates, and phosphates.
Steel products are often given a coating to inhibit oxidation and extend the life of
the product. Coated products can also be painted to further inhibit corrosion. Common coating
processes include: galvanizing (zinc coating), tin coating, chromium coating, aluminizing, and
terne coating (lead and tin). Metallic coating application processes include hot dipping, metal
spraying, metal cladding (to produce bi-metal products), and electroplating. These coating
processes contain many EPCRA Section 313 chemicals (mainly metals) that end up in the
facility's wastestream.
3-9
-------
3.2
Primary Aluminum Processing
Primary aluminum producers generally employ a three step process to produce
aluminum ingots. First, alumina is extracted from bauxite ore using the Bayer process (Figure 3-
3). In the Bayer process, finely crushed bauxite is mixed with an aqueous sodium hydroxide
(caustic soda) solution to form a slurry. The slurry is then reacted at a high temperature under
steam pressure in a vessel known as a digester, and creates a mixture of dissolved aluminum
oxides and bauxite residues. During the reaction a majority of the impurities such as silicon,
iron, titanium, and calcium oxides drop to the bottom of the digester and form a sludge. The
remaining sodium aluminate solution is then flash cooled by evaporation and sent for
clarification. During clarification, agents such as starch are added to help any fine impurities that
remain in the slurry, such as sand, to drop out, further purifying the sodium aluminate solution.
The solution is then fed into a precipitation tank to be crystallized. In the precipitation tank the
sodium aluminate solution is allowed to cool with the addition of a small amount of aluminum
hydroxide "seed." The seed stimulates the precipitation of solid crystals of aluminum hydroxide.
The aluminum hydroxide crystals settle to the tank bottom, and are removed. The
crystals are then washed to remove any caustic soda residues, vacuum dewatered, and sent on for
calcination. In the calciners (a type of rotating kiln) the aluminum hydroxide is roasted for
further dewatering.
In the second step, the aluminum oxide (alumina) produced during the Bayer
process is reduced to make pure molten aluminum. Alumina is dissolved in molten cryolite, and
the alumina is separated into aluminum and oxygen by electric current. The electrolytic
reduction process begins with the placement of the alumina into electrolytic cells, or "pots,"
filled with molten cyrolite. Within each pot a positive electric current is passed through the
cryolite by means of a carbon anode submerged in the liquid cryolite. The oxygen atoms,
3-10
-------
Fuel Condensate
1
Condensate
Impurity
Removal
Evaporation
T
Impurities Condensate
Heating
Steam
Cooling
Water
Fuel
Figure 3-3. Bayer Process (Alumina Refining)
3-11
-------
separated from aluminum oxide, carry a negative electrical charge and are attracted to the carbon
anodes. The carbon and the oxygen combine immediately to form carbon dioxide and carbon
monoxide. These gases bubble free of the melt. The nearly pure aluminum collects at the
bottom of the pot, is siphoned off, placed into crucibles, and then transferred to melting/holding
furnaces.
The third step consists of either mixing the molten aluminum with other metals to
form alloys of specific characteristics, or casting the aluminum into ingots for transport to
fabricating shops. Casting involves pouring inolten aluminum into molds and cooling it with
water. At some plants, the molten aluminum may be batch treated in furnaces to remove oxide,
gaseous impurities, and active metals such as sodium and magnesium before casting. Some
plants add a flux of chloride and fluoride salts and then bubble chlorine gas, usually mixed with
an inert gas, through the molten mixture. Chloride reacts with the impurities to form HC1, and
metal chloride emissions. A dross forms to float on the molten aluminum and is removed before
casting.
Two types of anodes may be used during the reduction process: an anode paste or
a pre-baked anode. Because the carbon is consumed during the refining process (about one-half
pound of carbon is consumed for every pound of aluminum produced), if anode paste (Soderberg
anode) is used, it needs to be continuously fed through an opening in the steel shell of the pot.
The drawback to pre-baked anodes is that they require that a pre-baked anode fabricating plant be
located nearby or on-site. Most aluminum reduction plants include their own facilities to
manufacture anode paste and/or pre-baked anode blocks.
One waste material produced during the primary production of aluminum are
fluoride compounds. Fluoride compounds are principally produced during the reduction process.
One reason that pre-baked anodes are favored is that the closure of the pots during smelting
facilitates the capture of fluoride emissions, though many modern smelters employ other
methods to capture and recycle fluorides and other emissions.
Primary aluminum processing activities result in air emissions, process
wastewaters, and other solid-phase wastes. Large amounts of particulates are generated during
the calcining of hydrated aluminum oxide, but the economic value of this dust for reuse in the
3-12
-------
process is such that extensive controls are used to reduce emissions to relatively small quantities.
Small amounts of particulates are emitted from the bauxite grinding and materials handling
processes. Emissions from aluminum reduction processes are primarily gaseous hydrogen
fluoride and particulate fluorides, alumina* carbon monoxide, volatile organics, and sulfur
dioxide from the reduction 'cells; and fluorides, vaporized organics and sulfur dioxide from the
anode baking furnaces; A variety of" control devices such as wet scrubbers are used to abate
emissions from reduction cells and anode baking furnaces.
Wastewaters generated from primary aluminum processing are produced during
clarification and precipitation though much of this water is fed back into the process to be reused.
Solid-phase wastes are generated at two stages in the primary aluminum process;
red mud produced during bauxite refining, and spent potliners from the reduction process. Red
mud normally contains significant amounts of iron, aluminum, silicon, calcium, and sodium.
The types and concentrations of minerals present in the mud depends on the composition of the
ore and the operating conditions in the digesters. The carbon potliners used to hold the
alumina/cryolite solution during electrolytic aluminum reduction process eventually crack and
need to be removed and replaced. These potliners contain cyanide complexes which form during
the aluminum reduction process.
3,3
Primary Zinc Processing
The primary production of zinc begins with the reduction of zinc concentrates to
metal (the zinc concentration process consists of separating the ore, which may be as little as two
percent zinc, from waste rock by crushing and flotation). Zinc reduction is accomplished in one
of two ways: either pyrometallurgically by distillation (retorting in a furnace) or
hydrometallurgically by elecfrowinning. Hydrometallurgical zinc refining is discussed in this
section, as this was the process seen on the site visits.
Four stages are generally used in hydrometallurgic zinc refining: calcining,
leaching, purification, and electrowinning. Calcining, or roasting, is performed to eliminate
sulfur and form leachable zinc oxide. Roasting is a high-temperature process that converts zinc
sulfide concentrate to an impure zinc oxide. Roaster types include multiple-hearth, suspension,
3-13
-------
or fluidized-bed. The gas zinc oxide stream is directed to the baghouse (filter) area where the
zinc oxide is captured in baghouse dust. All of the calcining processes generate sulfur dioxide,
which is controlled and converted to sulfuric acid as a marketable process by-product.
Electrolytic processing of desulfurized calcine consists of three basic steps;
leaching, purification, and electrolysis. Leaching refers to the dissolving of the captured calcine
in a solution of sulfuric acid to form a zinc sulfate solution. After leaching, the solution is
purified in two or more stages by adding zinc dust, which precipitates the copper and cadmium
which are then filtered out.
Zinc electrowinning takes place in an electrolytic cell and involves running an
electric current from a lead-silver alloy anode through the aqueous zinc compound solution. This
process charges the dissolved zinc ion and deposits it onto an aluminum cathode (a plate with an
opposite charge) which is immersed in the solution. Periodically, the zinc-coated cathodes are
removed and rinsed, and the zinc mechanically stripped from the aluminum plates. The zinc
concentrate is then melted and cast into ingots.
Electrolytic zinc smelters contain as many as several hundred cells. A portion of
the electrical energy is converted into heat, which increases the temperature of the electrolyte.
During electrowinning a portion of the electrolyte passes through cooling towers to decrease its
temperature and to evaporate the water it collects during the process.
The sulfur dioxide generated during the primary zinc refining process is further
reacted with oxygen and water to produce sulfuric acid. The wastes from the acid plant may
contain sulfur, cadmium, and lead.
During the electrolytic refining of zinc, solid materials in the electrolytic solution
that are not captured previously during purification may precipitate out in the electrolytic cell.
i
When the cells undergo their periodic shutdown to recover zinc, this precipitated waste (known
as anode slimes/sludges) is collected during cell cleaning.
Primary zinc processing activities generate air emissions, process wastes, and
other solid-phase wastes. The material inputs and waste outputs resulting from primary zinc
3-14
-------
processing are presented by media in Table 3-1. Air emissions are generated during roasting.
Sulfur dioxide emissions from the roasting processes at zinc processing facilities are generally
recovered at on-site sulfuric acid plants. Much of the particulate matter emitted from primary
zinc facilities is also attributable to roasters.
Table 3-1
Process Materials Inputs/Waste Outputs - Zinc
.-;..'• Process
Zinc Calcining
Zinc Leaching
Zinc Purification
Zinc Electrowinning
Material Input
Zinc ore, coke
Zinc calcine, sulfuric
acid, limestone, spent
electrolyte
Zinc-acid solution, zinc
dust
Zinc in a sulfuric
acid/aqueous solution,
lead-silver alloy anodes,
aluminum cathodes,
barium carbonate, or
strontium, colloidal
additives
Air Emissions
Sulfur dioxide,
particulate matter
containing zinc
and lead
Process Wastes
Wastewaters
containing
sulfuric acid
Wastewaters
containing
sulfuric acid,
iron
Dilute sulfuric
acid
Other Wastes
Acid plant
blowdown slurry
Copper cake,
cadmium
Electrolytic cell
slimes/sludges
Though the amount and composition of particulate varies with operating
parameters, the particulate is likely to contain zinc and lead.
Wastewaters may be generated during the leaching, purification, and
electrowinning stages of primary zinc processing when electrolyte and acid solutions become too
contaminated to be reused again. This wastewater is treated before discharge.
Solid wastes are generated at various stages in primary zinc processing.
Blowdown slurry generated during the operation of sulfuric acid plants is generally transferred
off site. The solid copper cake generated during purification is generally sent off site to recover
the copper.
3-15
-------
3.4
Semiconductor Manufacturing
A semiconductor is a material that has an electrical conductivity between that of a
conductor and an insulator; its electrical characteristics can be manipulated to behave like either
depending on how it is processed. Silicon has traditionally been the substrate used to
manufacture semiconductors; recently other materials such as gallium arsenide (GaAs) and
indium phosphide (InP) have been used as a substrate material.
The semiconductor manufacturing process is continually evolving. The variety of
distinct processing steps involved results in a range of processes that may occur at a single plant.
Process designs are not uniform from plant to plant. An average semiconductor manufacturing
process consists of hundreds of process steps, of which a significant percentage may be potential
sources of EPCRA Section 313 chemicals. Many of the manufacturing steps are repeated several
times during the production process. For these reasons, this overview will discuss general
manufacturing steps and does not attempt to describe a specific type of plant.
A clean environment is essential to the manufacture of semiconductors. Thus
cleaning operations precede and follow many of the manufacturing process steps. Wet
processing, during which semiconductor devices are repeatedly immersed in or sprayed with
solutions is commonly used to minimize the risk of contamination. Wet processes are the
primary source of EPCRA Section 313 chemicals found in semiconductor manufacturing.
The primary component of a semiconductor is the semiconductor wafer, or chip.
The manufacture of a semiconductor chip is essentially a six-step process (see Figure 3-4) with
the following steps:
Photolithography;
Thin Films;
Etching;
Cleaning;
Doping; and
Chemical Mechanical Planarization.
The following sections discuss each of the processing steps identified above. It
should be noted that many of the chemicals used in semiconductor manufacturing are used in
3-16
-------
more than one process, and some of the chemicals are used as a raw material as well as generated
indirectly through the use of other chemicals. For example, hydrochloric acid is used in wet
etching processes and is also generated in small quantities during dry etching, where chlorine
plasma reacts with a hydrogen carrier to produce hydrochloric acid. Also, regardless of which
process a chemical is used in, a general rule is that acid/base streams are treated in an on-site
wastewater treatment plant and waste solvents are typically collected and sent off site for waste
management (such as recycling or energy recovery).
3.4.1
Photolithography
Photolithography is used in semiconductor manufacturing to form surface patterns
on the wafer. These patterns will in turn allow various materials to be deposited on or removed
from selected, precise locations. In this process a viscous, solvent-based, light-sensitive
photoresist is applied to the wafer on a spin track. On the spin track, a fixed amount of
photoresist is metered onto the wafer, which is then spun at high speed on a rotating element to
uniformly coat the wafer surface.
After a "soft bake" to remove most of the carrier solvent, a pattern is introduced
into the resist by exposing predefined areas of the wafer with specific wavelengths of light,
lasers, electron beams, or other means. A template mask, which is a glass plate containing an
image of the desired circuit, may be used to introduce the pattern. Depending on the photoresist
system, a developer solution is applied to dissolve some of the photoresist, yielding a stencil for
further processing. Materials may be added or removed from the unmasked areas, giving a
printed circuit. The number of photolithography steps required depends on the type of integrated
circuit.
After the subsequent processing steps, residual photoresist is removed by using
wet stripping (solvent or acid) or plasma gas stripping.
3-17
-------
Chemical
Mechanical
Planarization
Figure 3-4. Overall Process Flow Diagram - Semiconductor Manufacturing
3-18
-------
The most common potential sources of EPCRA Section 313 chemicals from
photolithography are photoresist solvents, strippers, and developers. The primary release source
is tool exhaust from the photoresist applicators, developing, and stripping stations; spent solvent,
developer, and stripping solutions collected and sent either to a POTW or for off-site treatment,
disposal, or energy recovery; and container residue. Typical EPCRA Section 313 chemicals
include n-methyl-2-pyrrolidone, xylene (mixed isomers), certain glycol ethers, ammonia, and
methyl ethyl ketone. Ammonium compounds are commonly used as developers.
Release and other waste management activity types include fugitive and point
source emissions to the air from wet chemical and solvent stations and control equipment
exhaust, and off-site transfers to POTWs, treatment facilities, and waste management facilities.
Vapors from organic solvents found in the photoresist and photoresist strippers are typically sent
to an on-site treatment unit, usually some type of concentrator (such as a carbon absorber)
followed by thermal destruction.
3.4.2
Thin Films
In thin film deposition, layers of single crystal silicon, polysilicon, silicon nitride,
silicon dioxide, and other materials are deposited on the wafer to provide desirable properties on
portions of the device or to serve as masks. Deposition of these films is frequently performed in
a chemical vapor deposition (CVD) reactor or a high-temperature tube furnace using
silicon-containing gases as reactants. The deposition rate can be further enhanced by striking a
plasma to overcome kinetic barriers. Selected impurity compounds or dopants may be used in
the deposition process to alter the electrical characteristics of the deposited film or layer.
Sometimes chlorine gas, hydrochloric acid (acid aerosols), or 1,1,1-trichloroethane is used during
oxidation to modify the oxide characteristics.
Typical EPCRA Section 313 chemicals used in these deposition processes are
chlorine, anhydrous ammonia, hydrochloric acid (acid aerosols) 1,1,1-trichloroethane, and
1,2-dichloroethylene.
To interconnect electrical devices on an integrated circuit and to provide for
external connections, metallic layers are deposited onto the wafer by evaporation, sputtering, or
3-19
-------
CVD. Evaporation consists of vaporizing a metal under a vacuum at a very high temperature.
I
Sputtering processes (also called physical vapor deposition or PVD) involve bombarding
metallic targets with a plasma gas, which displaces ions from the target and deposits them on the
wafer. CVD of metal is similar to the other deposition processes described above except that the
reactive gas is a metal-containing vapor. Devices may have a single layer or multiple layers of
metal.
Typical EPCRA Section 313 chemicals used in metallization processes are
ammonia and compounds of chromium, copper, and nickel.
The most common potential source of EPCRA Section 313 chemicals from
application of thin films is ammonia gas used as a nitrogen source in silicon nitride deposition,
organics used as chlorine sources and as a cleaner for deposition furnaces, and metals deposited
to interconnect electrical devices. The primary point of release is the thin film deposition tool
(furnace or oxidation chamber). Exhaust from these tools is typically routed to a scrubber which
vents to the atmosphere and also result in wastewater generation. It should be noted that metals
are not typically used hi amounts that would exceed the reporting threshold. Release and other
waste management activity types from thin film deposition processes include point source air,
on-site treatment, and off-site disposal.
3.4.3
Etching
Etching is used to chemically remove specific unwanted areas of silicon substrate
or deposited film so that an underlying material may be exposed or another material may be
deposited in the etched material's place. Etch processes usually occur after a photoresist pattern
has been applied, so that the etching is accomplished in specific areas.
Etching may be performed using either solutions of acids, bases, or oxidizers (wet
etch), or by using various gases (usually halogenated) excited by striking a plasma (dry etch). In
either case, the fluoride ion or radical is almost always introduced if the substrate or film to be
etched contains silicon.
3-20
-------
Typical examples of chemicals used in etch processes include 1,1-dichloro-l-
fluoroethane, chlorine, ethylene glycol, hydrochloric acid (acid aerosols), hydrogen fluoride,
nitric acid, phosphoric acid, and sulfuric acid (acid aerosols).
The most common potential source of EPCRA Section 313 chemicals from
etching is the etchant material itself, usually inorganic acids used in wet etching processes, and
certain halogenated organics used in dry etching processes. Acid gasses from the etching tools
are usually vented through a scrubber, which in turn sends spent scrubber water to an on-site
treatment plant for neutralization prior to discharge to a receiving stream or POTW. Release and
other waste management activity types include stack air, fugitive air, container residue, and
discharge to a POTW.
3.4.4
Cleaning
Cleaning of wafers is required to prepare them for each chemical and physical
process to ensure that contaminants on the wafer surfaces do not affect the final integrated
circuit's electrical performance. Before, and sometimes after, wafers are subjected to any
specialized manufacturing processes, they are typically immersed in or sprayed with various
aqueous and/or organic solutions. In some cases they are mechanically scrubbed in some manner
to remove films, residues, bacteria, or other particles. Fog chambers may also be used for wafer
cleaning.
Typical sources of EPCRA Section 313 chemicals from cleaning operations
would be cleaning station exhaust vents, waste solvents, and container residue (from "empty"
containers of the cleaning solution). Release and other waste management activity types from
this process include stack and fugitive air emissions; off-site transfer of the spent cleaning
solutions; wastewater discharges (either direct or to a POTW); and transfers of "empty" shipping
containers to off-site locations.
3-21
-------
3.4.5
Doping
Doping is a process in which specific atoms of impurities are introduced into the
silicon substrate to alter the electrical properties of the substrate by acting as charge carriers.
Doping is typically accomplished through ion implantation or diffusion processes.
Ion implantation is the most common method used to introduce impurity atoms
into the substrate. The dopant atoms are first ionized with a medium-to-high-current filament,
then accelerated toward the wafer surface with large magnetic and electrical fields. Because of
the high kinetic energy of the ions during bombardment, damage to the substrate's crystalline
structure occurs. To restore the substrate's structure to a satisfactory level, slow heating or
"annealing" of the amorphous material in various gaseous atmospheres is subsequently
performed.
Diffusion is a high-temperature process also used to introduce a controlled
amount of a dopant gas into the silicon substrate. The process occurs in a specially designed tube
furnace where dopants may be introduced in one of two primary ways: dopant gases may be
introduced into the furnace that will diffuse into the exposed areas of the substrate (gaseous
diffusion), or dopant atoms may diffuse into the substrate from a previously deposited dopant
oxide layer in the areas where the two are in contact (nongaseous diffusion).
Typical EPCRA Section 313 chemicals used in doping processes are boron
trifluoride and compounds of antimony, barium, and nickel.
The most common potential source of EPCRA Section 313 chemicals from
doping are the dopants themselves, as well as certain organic compounds that may be used as
furnace cleaning gases or chlorine sources. The physical release points are tool and control
device exhaust vents, spent cleaning solutions, and solid or hazardous waste generated as part of
the process.
Organic chemicals may be emitted from furnace exhaust and may also be
collected and sent off site for energy recovery or disposal.
3-22
-------
3.4.6
Chemical Mechanical Planarization
Chemical Mechanical Planarization (CMP) is used in semiconductor
manufacturing to remove the top layer of material from the wafer in a controlled manner, leaving
a smooth and flat surface for further processing. This technology is applied in two ways. The
first is to selectively remove the top part of a nonconducting layer or film to reduce the
topography on the wafer (also called planarization). The second type of CMP is removal of
excess material from the surface of conducting layers (metals).
The only notable source of EPCRA Section 313 chemicals from CMP is the
planarization process which typically contains ferric nitrate (Fe(NO3)3), a nitrate compound.
Spent slurries containing nitrates are typically sent off site to a POTW.
3.5
Printed Wiring Board Manufacturing
Printed wiring boards (PWBs) are the physical structures on which electronic
components such as semiconductors and capacitors are mounted. PWBs are subdivided into
single-sided, double-sided, multilayer, and flexible boards. Multilayer boards are manufactured
similarly to single and double-sided boards, except that conducting circuits are etched on both
the external and internal layers. Multilayer boards allow for increased complexity and density.
PWBs are produced using three methods: additive, subtractive, or semi-additive technology. The
subtractive process accounts for a significant majority, perhaps 80%, of PWB manufacturing.
The conventional subtractive manufacturing process begins with a board,
consisting of epoxy resin and fiberglass, onto which patterns are imaged. In most operations,
conducting material, usually copper, is bonded onto the substrate surface to form copper-clad
laminate. After drilling holes through the laminate and making those holes conductive,
unwanted copper is etched off, leaving copper patterns. The patterns on the board form the
electric circuits that conduct electricity. Multilayer boards typically use metals such as platinum,
palladium, and copper to form electric circuits. Specialized PWBs may use nickel, silver, or
gold.
3-23
-------
Additive technology is used less often than subtractive technology because it is a
more difficult and costly production process. This capital-intensive technology is used primarily
for small interconnection components used in multi-chip devices. The production process begins
with a base plate upon which a dielectric material is deposited. An interconnecting layer of
copper is plated onto the dielectric layer which connects the layers of dielectric material and
copper. Copper posts are plated-up and another layer of dielectric material is deposited exposing
the posts. The next interconnect layer is plated and makes contact with the posts. Layers of
dielectric material, copper, and copper posts are added to complete the interconnect. A
lithographic process, similar to the one used in semiconductor manufacturing, diminishes the
spaces and widths of the PWB.
This section provides a simplified discussion of the steps commonly performed
during conventional subtractive manufacturing. These steps are outlined below:
3.5.1
Board preparation;
Application of conductive coatings (plating);
Soldering;
Fabrication; and
Assembly.
Board Preparation
Board preparation begins with a lamination process. Two-side etched copper
dielectric boards (consisting usually of fiberglass and epoxy resin) are separated by an insulating
layer and laminated or bonded together, usually by heat and pressure. Photographic tools are
j
used to transfer the circuit pattern to the PWB, and computer control programs are used to
control the drilling, routing, and testing equipment. Preparing the copper-clad board involves
drilling holes to establish an electrical path between the layers and to mount components. The
boards are then mechanically cleaned to remove drilling wastes (i.e., fine particulate
contaminants, such as copper). Vapor degreasing, abrasive cleaning, chemical cleaning with
alkaline solutions, acid dips, and water rinses are techniques used to clean the boards and prepare
them for the next process, electroless plating. Table 3-2 shows materials used during lamination,
drilling, and cleaning processes.
3-24
-------
Table 3-2
Chemicals Used in Lamination, Drilling, and Cleaning
Lamination
Drilling:
Cleaning
Epoxies
Sulfuric acid
Potassium permanganate
Ammonium bifluoride
Oxygen
Fluorocarbon gas
Acetone
1,1,1 -Trichloroethane
Silican (and other abrasives)
Sulfuric acid
Aqueous ammonia
Hydrochloric acid
Source: Based on EPA DfE 1993: Industry Profile and Description of Chemical Use for the Printed Wiring Board Industry: Preliminary Draft.
Wastes generated include: airborne particulates, acid fumes, and organic vapors
from cleaning, surface preparation, and drilling; spent acid and alkaline solutions; spent
developing solutions, spent etchings, and waste rinsewaters in the wastewater; and scrap board
materials and sludges from wastewater treatment. Drilling and routing dust (copper, aluminum,
and gold) are collected and recycled.
3.5.2
Electroless Plating
The first process in this step is to prepare the surfaces of the drilled holes. The
holes are prepared by an etchback process to remove smeared epoxy resin and other
contaminants using one of the following: sulfuric or hydrochloric acid; potassium permanganate;
or carbon tetrafluoride, oxygen and nitrogen. The holes are then coated with a material such as
copper or graphite carbon, by a chemical process called electroless plating.
Electroless plating coats a uniform conducting layer of copper or other material on
the entire surface including the barrels of the holes of the prepared board without outside power
sources. This coating of copper is not thick enough to carry an electrical current, but provides a
base upon which additional copper can be deposited electrolytically. Chemical deposition is the
technique used to coat the board. After the electroless plating, the boards are dried to prevent the
board from oxidation (e.g., rusting). The board may also be cleaned to prepare for a following
electroplating processing. Table 3-3 shows a list of material used. Wastes generated include:
3-25
-------
spent electroless copper baths; spent catalyst solutions; spent acid solutions; waste rinsewaters;
and sludges from wastewater treatment.
Table 3-3
Materials Used in Copper and Tin-Lead Electro- and
Electroless Plating Processes
Type of Plating
Copper
Tin-Lead
Electroplating Chemicals
Copper pyrosphate
Orthophosphate
Pyrophosphate
Nitrates
Ammonia
Acid copper
Copper sulfate
SulfUric acid
Tin-Lead
Fluoroboric acid
Boric Acid
Peptone
Electroless Chemicals
Hydrochloric acid
Palladium chloride
Stannous chloride
Metallic tin pellets
Sodium hydroxide
Copper sulfate
Formaldehyde
Tin chloride
Sodium hypophosphite
Sodium citrate
Source: Based on EPA DfE 1993: Industry Profile and Description of Chemical Use for the Printed Wiring Board Industry: Preliminary Draft.
3.5.3
Imaging
During imaging, circuit patterns are transferred onto the boards through
photolithography or a stencil printing process. Photoresist (i.e., a light sensitive chemical) is
applied to the board in areas where the circuit pattern will not be set. The board is exposed to a
,i
radiation source and developed to remove the unwanted areas of the resist layer. Stencil printing
uses a printing process, such as silk screening, to apply a protective film that forms the circuit
pattern.
After photolithography, the boards are subjected to a light etching process,
typically using etchants containing ammonia, to remove rust inhibitor or other metals (usually
copper). After the stencil printing process, the protective film is dried, and the exposed copper is
etched away. SulfUric acid and hydrogen peroxide are common etchants used during imaging.
After plating or etching, the photoresist is removed with a photoresist stripper.
3-26
-------
Tables 3-4 and 3-5 present a list of materials used during photolithography and
etching process. Wastes generated include organic vapors and acid fumes, spent developing
solutions, spent resist material, spent etchant, spent acid solutions, and sludges from wastewater
treatment.
Table 3-4
Chemicals Used in Photolithography for Printed Wiring Boards
Resists
Phptopolymef Developers
l^otopiiriymer Strippers
Mylar
Vinyl
Photoresists
Isopropyl alcohol
Potassium bicarbonate
Sodium bicarbonate
1,1,1 -Trichloroethane
Amines
Glycol ethers
Sodium hydroxide
Potassium hydroxide
Methylene chloride
Source: Based on EPA Dffi 1993: Industry Profile and Description of Chemical Use for the Printed Wiring Board Industry: Preliminary Draft.
Table 3-5
Materials Used During Etching
Ammonia
Ammonium chloride
Ammonium persulfate
Ammonium sulfate
Boric acid
Carbon tetrafluoride
Chlorine
Cupric chloride
Hydrochloric acid
Hydrofluoric acid
Hydrogen peroxide
Lead
Nickel
Nickel chloride
Nickel sulfamate
Nitrate
Nitric acid
Nitrogen
Orthophosphate
Oxygen
Peptone
Permanganates
Sodium citrate
Sodium hydroxide
Stannous chloride
Sulfuric acid
Tin
Source: Based on EPA DfE 1993: Industry Profile and Description of Chemical Use for the Printed Wiring Board Industry: Preliminary Draft.
3.5.4
Electroplating
Electroplating is a process in which a metal is deposited on a substrate through
electrochemical reactions. Electroplating is required to build up the thickness and strength of the
conducting layers to provide reliable electrical conductivity between inner layers or from one
side of the PWB to the other. Electroplating can also protect against corrosion, wear, or erosion.
3-27
-------
This process involves immersing the article to be coated/plated into a bath containing acids,
bases, or salts.
The electroplating process for PWBs usually begins with the copper laminate
which is coated with a plating resist (photolithography), by stenciling, leaving the area exposed
to form the circuit pattern. The resist prevents the conductive material from adhering to other
areas of the board and forms the circuit pattern.
The PWB plating process typically uses copper and tin-lead as plating materials,
although silver, nickel, or gold can be used. Copper in a plating bath solution is deposited to a
sufficient thickness, and a solvent or aqueous solution is applied to remove the plating resist.
The copper coating forms interconnections between the layers and provides electrical contact for
electronic parts mounted or assembled on the PWB surface. PWB manufacturers then typically
electroplate a tin or tin-lead solder on the board to protect the circuit pattern during the following
etching or stroppihg prdcesses. An acid etch solution (ammonia^ peroxide solutions, sodium
persulfate, cupric chloride, or ferric chloride) removes the exposed copper foil, leaving the
thicker copper plating to form the circuit pattern. Ammonia and cupric chloride are the primary
etchants used by PWB manufacturers. Fluoroboric acid is used in the tin-lead plating process to
keep the metals dissolved in the solution and to ensure a consistent deposition of the tin-lead
alloy onto the circuit board.
After the plating bath, the board is rinsed with water,, scrubbed, and then dried to
remove the copper, spray etch solutions, arid other materials. Rinsing ends the chemical
reactions during plating and prevents contamination or dragout from being released in the next
bath or rinse water (dragout is the plating solution that sticks to parts after taken out of the
plating bath). Dragout can occur in any bath step, not just in one plating bath. The tin-lead layer
is generally removed and the panel is electrically tested for discontinuities in the electrical
pathway and shorts. Table 3-3 presents a list of materials used during the electroplating process.
Wastes generated during plating include: spent acid solutions, waste rinsewaters,
spent developing solutions, spent etchant, and spent plating baths in the wastewater; organic
vapors from spent developing solution and spent resist removal solution; and acid and ammonia
fumes.
3-28
-------
3.5.5
Soldering Coating
Solder coating is used to add solder to PWB copper components before
component assembly. All methods of solder coating involve dipping the panel into molten
solder. The solder, an alloy consisting of 60 percent tin and 40 percent lead, coats the pads and
holes not covered by solder mask. The excess solder is removed with a blast of hot oil or hot air.
However, the hot oil or hot air does not remove the solder that has formed a chemical
(intermetallic) bond with the copper. The most common process is hot-air leveling.
3.5.6
Electrical and Mechanical Testing
A cross section is cut from a sample panel from each lot using a grinding process
called routing, and the plated holes are examined with a photomicrograph. Individual circuit
boards are cut out of panels that pass quality control. Routing generates dust which may contain
copper, lead, or other metals plated to the panel, but the dust is recycled. Electrical tests,
dimensional and visual inspections, and quality audits are performed to ensure compliance with
customer requirements. Finally, the finished PWBs are packaged, labeled, and shipped to the
customer.
3,5.7
Printed Wiring Board Assembly and Soldering
After the PWBs are manufactured, the electrical components are attached during
assembly. Adhesives are applied to the boards, and then the components are attached and
soldered to the boards. Components are attached to the PWB by a process called soldering.
There are several different kinds of soldering processes, including wave, dip, and drag. A type
of chemical known as "flux" is used before soldering to facilitate the production of the solder
connection. Not only does flux clean the surface and remove oxidized material, it prevents
oxidation from occurring during the solder process. After the solder has been applied, flux
residue may be removed from the board, and the board may be cleaned and dried.
3-29
-------
The wastes generated during assembly include: solder dross, post-solder scrap
boards, filters, gloves, rags, and spent gaseous or semi-gaseous solvents from cleaning processes.
The wastes that may be generated during soldering, flux application, and cleaning include:
organic vapors, copper, lead, spent solvents, and spent deionized water into the wastewater;
solder dross; and wastewater treatment sludge. Solder dross is primarily oxidized solder skin
that forms on any molten solder exposed to oxygen and can be recycled off site.
3.6
Motor Vehicle Equipment Manufacturing
Motor vehicle parts and accessories include both finished and semi-finished
components. Approximately 8,000 to 10,000 different parts are ultimately assembled into
approximately 100 major motor vehicle components, including suspension systems,
transmissions, and radiators. These parts are eventually transported to an automotive
manufacturing plant for assembly.
The manufacturing process used to produce the thousands of discrete parts and
accessories vary depending on the end product and materials used. Different processes are
employed for the production of metal components versus the production of plastic components.
Most processes, however, typically include casting, forging, molding, extrusion, stamping, and
welding.
3.6.1
Foundry Operations
Foundries, whether they are integrated with automotive assembly facilities or
independent shops, cast metal products which play a key role in the production of motor vehicles
and motor vehicle equipment. Iron and steel are currently the major metal components of an
automobile with increasing use of aluminum and other metals. The following discussion focuses
on iron foundries and associated production processes.
3-30
-------
The main steps in producing cast iron motor vehicle products are as follows (see
Figure 3-5):
Pattern design and production;
Sand formulation;
Mold and core production;
Metal heating and alloying;
Metal molding;
Mold shakeout;
Production finishing and heat treating; and
Inspection.
The process begins with the mixing of moist silica sand with clay and water to
produce the "green sand," which forms the basis of the mold. Other additives, including organics
such as seacoal or oat hulls, may be added to the green sand to help prevent casting defects. The
core is then created using molded sand and often includes binders, such as resins, phenol, and/or
formaldehyde. The core is the internal section of a casting used to produce the open areas
needed inside such items as an engine or a drive train. After the core has been molded, it is
baked to ensure its shape, and then combined with the rest of the casting mold in preparation for
casting. At the same time the core is being created, iron is being melted. The iron charge,
whether it be scrap or new iron, is combined with coal (as a fuel) and other additives such as
calcium carbide and magnesium, and fed into a furnace, which removes sulfur (usually an
electric arc, an electric induction, or a cupola furnace).
Calcium carbide may be added for certain kinds of iron casting, and magnesium is
added to produce a more ductile iron. Once the iron reaches the appropriate temperature, it is
poured into the prepared mold. The mold then proceeds through the cooling tunnel and is placed
on a grid to undergo a process called "shakeout." During shakeout the grid vibrates, shaking
loose the mold and core sand from the casting. The mold and core are then separated from the
product which is ready for finishing.
The finishing process is made up of many different steps depending upon the final
product. The surface may be smoothed using an oxygen torch to remove any metal snags or
chips, it may be blast-cleaned to remove any remaining sand, or it may be pickled using acids to
achieve the correct surface. If necessary, the item may be welded to ensure the tightness of any
3-31
-------
CD
CD
rl
c
Q> O
.£ a, S
"*•* "7T "i"1
2 -e o
-^
LU
O.
CO
z
CD
d
5I
Q? ro
P
S a.
•i!
.s
CD
X
1
CL
6
a.
2
^
ra
m
•
.
S
I1
CD
B
>
i
fi.
M
C
CO
C
1
CL
D
C
1
3
O
i.
)
5.
n
t. c
§5
LL
z
a:
i
0.
CO
=
i
CD
X
CO
z
LL
CO CL
C
C
co
CL
I
n°
if
CO
~
(0
a.
g
^
ca
CO
c
_N
O
C
CO
_c
•h
co
s
«
I
Q
I
"3
fe
S
O
I
OX)
E
3-32
-------
seams or seals. After finishing, the item undergoes a final heat treatment to ensure it has the
proper metallurgical properties. The item is then ready for inspection. Inspection may take place
in any number of ways, be it visually, by x- or gamma ray, ultrasonic, or magnetic particle. Once
an item passes inspection, it is ready to be shipped to the assembly area.
Iron foundries create a number of air emissions, process wastewaters, and solid
wastes. Gas and particulate emissions occur throughout the casting process. Dust created during
sand preparation, molding, and shakeout is prevalent. Gases containing lead and cadmium and
other particulate matter and sulfur dioxide are also created during foundry operation, especially
during the melting of the iron.
The wastewaters generated during foundry operations are primarily from slag
quenching operations (water is sprayed on the slag to both cool it as well as pelletize it) and by
the wet scrubbers employed as air pollution control devices connected to furnaces and sand and
shakeout operations. Due to the presence of cadmium and lead in iron, these metals may both be
present in wastewaters.
3.6.2
Metal Fabricating
Another major process in the manufacturing of automotive parts is metal
fabrication. Metal fabrication involves the shaping of metal components. Many automotive
parts, including fenders, hubcaps, and body parts are manufactured in metal fabricating shops.
Typical large-scale production of these items starts with molten metal (ferrous or nonferrous)
containing the correct metallurgical properties. Once the metal has been produced, it is cast into
a shape that can enter the rolling process. Shearing and forming operations are then performed to
cut materials into a desired shape and size and bend or form materials into specified shapes.
Shearing (or cutting) operations include punching, piercing, blanking, cutoff,
parting, shearing, and trimming. Basically, these are operations that produce holes or openings,
or that produce blanks or parts. Forming operations shape parts by forcing them into a specified
configuration, and include bending, forming, extruding, drawing, rolling, spinning, coining, and
forging.
3-33
-------
Once shearing and forming activities are complete, the material is machined. This
entails shaping or forming a workpiece by removing material from pieces of raw stock with
machine tools. The principal processes involved in machining are hole-making, milling, turning,
shaping/planing, broaching, sawing, and grinding.
Each of the metal shaping processes can result in wastes containing EPCRA
Section 313 chemicals. In general, there are two categories of waste generated in metal shaping
operations: scrap metal and metalworking fluids/oils.
Scrap metal may consist of metal removed from the original piece (e.g., steel or
aluminum). Quite often, scrap is reintroduced into the process as feedstock.
In general, metalworking fluids can be petroleum-based, oil-water emulsions, or
synthetic emulsions that are applied to either the tool or the metal being tooled to facilitate the
shaping operation. Metalworking fluids typically become contaminated and spent with extended
use and reuse. When disposed, these fluids may contain contaminants, including chlorine, sulfur
an phosphorous compounds, phenols, cresols, and alkalines, as well as metals. Air emissions
may result through volatilization during storage, fugitive losses during use, and direct ventilation
of fumes.
3.6.3
Metal Finishing/Electroplating
The final process in manufacturing motor vehicle parts is metal finishing and
electroplating. Numerous methods are used to finish metal products. However, prior to applying
j
the finishing application, the surface must be prepared. One of the most important aspects of a
finished product is the surface cleanliness and quality. Without a properly cleaned surface, even
the most expensive coatings will fail to adhere or prevent corrosion. The steel motor vehicle
parts are generally cleaned with nitric, hydrochloric, and hydrofluoric acid, as well as alkaline
cleaning agents.
Metal finishing and electroplating activities are performed on a number of metals
and serve a variety of purposes; the primary purpose being protection against corrosion. This is
particularly important to the automotive industry because of the harsh weather and road
3-34
-------
conditions to which automobiles may be subject. Metal finishing and electroplating can also be
performed for decorative purposes. These plating processes involve immersing the article to be
coated/plated into a bath consisting of acids, bases, salts, etc.
The metals used in electroplating operations (both common and precious metal
plating) include cadmium, lead, chromium, copper, nickel, zinc, gold, and silver. Cyanides are
also used extensively in electroplating solutions and in some stripping and cleaning solutions.
Electroless plating is the chemical deposition of a metal coating onto a metal
object, by immersion of the object in an appropriate plating solution. In electroless nickel
plating, the source of nickel is a salt, and a reducer is used to reduce the nickel to its base state.
A complexing agent is used to hold the metal ion in the solution. Immersion plating produces a
metal deposit by chemical displacement. Immersion plating baths are usually formulations of
metal salts, alkalies, and complexing agents (typically cyanide or ammonia).
Etching is the process used to produce specific design configurations or surface
appearances on parts by controlled dissolution with chemical reagents or etchants. Etching
solutions are commonly made up of strong acids or bases with spent etchants containing high
concentrations of spent metal. The solutions include ferric chloride, nitric acid, ammonium
persulfate, chromic acid, cupric chloride, and hydrochloric acid.
Anodizing uses the piece to be coated, generally with an aluminum surface, as an
anode in an electrolytic cell. Anodizing provides aluminum parts with a hard abrasion- and
corrosion-resistant film. This coating is porous, allowing it to be dyed or to absorb lubricants.
This method is used both in decorative applications, including automotive trim and bumper
systems, and in engineering applications such as aircraft landing gear struts. Anodizing is
usually performed using either sulfuric or chromic acid often followed by a hot water bath,
though nickel acetate or sodium potassium dichromate seal may also be used.
Surface preparation operations generate wastes contaminated with solvents and/or
metals depending on the type of cleaning operation. Concentrated solvent-bearing wastes and
releases may arise from degreasing operations. Degreasing operations may result in solvent-
bearing wastewaters, air emissions, and materials in solid form.
3-35
-------
Surface finishing and related washing operations account for a large volume of
wastes associated with automotive metal finishing. Metal plating and related waste account for
the largest volumes of metal (e.g., cadmium, chromium, copper, lead, mercury, and nickel) and
cyanide-bearing wastes.
Electroplating operations can result in solid and liquid wastestreams that contain
EPCRA Section 313 chemicals. Liquid wastes result from workpiece rinses and process cleanup
waters. Most surface finishing (and many surface preparation) operations result in liquid
wastestreams. Centralized wastewater treatment systems are common, and can result in solid-
phase wastewater treatment sludges. In addition to these wastes, spent process solutions and
quench baths are discarded periodically when the concentrations of contaminants inhibit proper
function of the solution or bath. When discarded, process baths usually consist of solid- and
liquid-phase wastes that may contain high concentrations of acids, bases, and cyanide wastes.
3.7
Motor Vehicle Painting/Finishing
Automotive finishing is a multi-step process subdivided into four categories: 1)
anti-corrosion operations, consisting of cleaning applications, a phosphate bath, and a chromic
acid bath; 2) priming operations, consisting of an electrodeposition primer bath, an anti-chip
application, and aprimer-surfacer application; 3) joint sealant application; and 4) finishing
operations, consisting of a color coat application, a clear coat application, and any painting
necessary for two-tone color or touch-up applications. The stages of the automotive finishing
process are illustrated in Figure 3-6
After the automobile body has been assembled, anti-corrosion operations prepare
the body for the painting/finishing process. Initially, the body is sprayed with or immersed in a
I
cleaning agent, typically consisting of detergents, to remove residual oils and dirt. The body is
then dipped into a phosphate bath, typically zinc phosphate, to prevent corrosion. The phosphate
I
process also improves the adhesion of the primer to the metal. The body is then rinsed with
chromic acid, further enhancing the anti-corrosion properties of the zinc phosphate coating. The
anti-corrosion operations conclude with another series of rinsing steps.
3-36
-------
Priming operations further prepare the body for finishing by applying various
layers of coatings designed to protect the metal surface from corrosion and assure good adhesion
of subsequent coatings. Prior to the application of these primer coats, however, plastic parts to
be painted and finished with the body are installed.
A primer coating is applied to the body using an electrodeposition method,
creating a strong bond between the coating and the body to provide a more durable coating. In
electrodeposition, a negatively-charged auto body is immersed in a positively-charged bath of
primer. The coating particles, insoluble in the liquid and positively-charged, migrate toward the
body and are, in effect, "plated" onto the body surface.
Although the primer bath is mostly water-based with only small amounts of
organic solvent (less than five percent to ten percent), fugitive emissions consisting of volatile
organic compounds (VOCs) can occur. However, the amount of these emissions is quite small.
In addition to solvents and pigments, the electrodeposition bath contains small amounts of lead.
Prior to baking, excess primer is removed through several rinsing stages. The
rinsing operations use various systems to recover excess electrodeposited primer. Once the body
is thoroughly rinsed, it is baked. VOC emissions resulting from the baking stage are generally
incinerated.
Next, the body is further water-proofed by sealing spot-welded joints of the body.
Water-proofing is accomplished through the application of a paste or putty-like substance. This
sealant usually consists of polyvinyl chloride and small amounts of solvents. The body is again
baked to ensure that the sealant adheres thoroughly to the spot-welded areas.
After water-proofing, the automobile body proceeds to the anti-chip booth. Here,
a substance usually consisting of a urethane or an epoxy ester resin, in conjunction with solvents,
is applied locally to certain areas along the base of the body, such as the rocker panel or the front
of the car. This anti-chip substance protects the lower portions of the automobile body from
small objects, such as rocks, which can fly up and damage automotive finishes.
3-37
-------
Body Shop
Seal Deck
Cleaning
Operation
Zinc
Phosphate
Bath
Primer Electro
Deposition
Install
Plastic Parts
Chromic
Acid Dip
fc
Anti-Chip
Booth
^
Primer -
Surface Water -
Wash Booth
I
1
r
Clear Coat
Booth
i
t
4
Main Color
Booth
^
^
Wet Sand
Deck
4t
\
Finesse
Operations
Deadener
Trim Shop
Repairs and
Two-Tone
Finishing
Assembly
Final Repairs
Figure 3-6. Car Painting Process
3-38
-------
The primer-surfacer coating, unlike the initial electrodeposition primer coating, is
applied by spray application in a water-wash spray booth. The primer-surfacer consists primarily
of pigments, polyester or epoxy ester resins, and solvents. Due to the composition of this
coating, the primer-surfacer creates a durable finish which can be-sanded. The pigments used in
this finish provide additional color layers in case the primary color coating is damaged. A
continuous stream of air, usually from ceiling to floor, is used to transport airborne particulates
and solvents from the primer-surfacer overspray. The air passes through a water curtain which
captures a portion of the airborne solvents for reuse or treatment at a waste water facility. Efforts
have been made at certain facilities to recycle this air to reduce VOC emissions.
After the primer-surfacer coating is baked, the body is then, sanded •,: if necessary, *
to remove any dirt or coating flaws. This is accomplished using a dry sanding technique.
Particulate matter is generated during this process.
The next step of the finishing process is the application of the primary color
coating. This is accomplished in a manner similar to the application of primer-surfacer. One
difference between these two steps is the amount of pigments and solvents used in the ;
application process. VOC emissions from primary color coating operations can be double that
released from primer-surfacer operations. In addition to the pigments and solvents, aluminum or
mica flakes can be added to the primary color coating to create a finish with unique reflective
qualities. Instead of baking, the primary color coat is allowed to "flash off," in other words, the
solvent evaporates without the application of heat. This evaporation contributes to significant air
emissions.
After the primary color coating is allowed to air-dry briefly, the final coating, a
clear coat, is applied. The clear coat adds luster and durability to the automotive finish. This
coating generally consists of a modified acrylic or a urethane and is baked on. Following the
baking of the clear coat, the body is inspected for imperfections in the finish. Operators finesse
minor flaws through light sanding and polishing and without any repainting.
Once the clear coat is baked, a coating known as deadener is applied to certain
areas of the automobile underbody. Deadener, generally a solvent-based resin of tar-like
consistency, is applied to areas such as the inside of wheel wells to reduce noise. In addition,
3-39
-------
anti-corrosion wax is applied to other areas, such as the inside of doors, to further seal the
automobile body and prevent moisture damage. This wax contains aluminum flake pigment and
is applied using a spray wand.
The finished vehicle is then inspected to ensure that no damage has occurred as a
result of the final assembly stages. If there is major damage, the entire body part is replaced.
However, if the damage is minor, such as a scratch, paint is taken to the end of the line and
applied using a hand-operated spray gun.
Many of the wastes generated during automotive production are the result of
painting and finishing operations. These operations result in air emissions as well as the
generation of solid and liquid wastes.
Air emissions, primarily VOCs, result from the painting and finishing application
processes (paint storage, mixing, applications, and drying) as well as cleaning operations. These
emissions are composed mainly of organic solvents containing EPCRA Section 313 chemicals
which are used as carriers for the paint. Solvents are also used during cleanup processes to clean
spray equipment between color changes, and to clean portions of the spray booth. Solvents are
often composed of a mixture of dimethyl-benzene, acetone, 4-methyl-2-pentanone, butyl ester
acetic acid, light aromatic solvent naphtha, ethyl benzene, hydrotreated heavy naphtha, 2-
butanone, toluene, and 1-butanol.
Various solid and liquid wastes may be generated throughout painting operations.
These wastes generally contain the solvents with EPCRA Section 313 chemicals listed above.
Solid and liquid wastes may also contain metals from paint pigments and organic solvents.
3-40
-------
4.0 THRESHOLD DETERMINATIONS
This section reviews how accurately facilities determined whether EPCRA Section 313
chemicals exceeded TRI reporting thresholds and the extent to which incorrect threshold
determinations ultimately affect the quality of the TRI database. Threshold determinations are a
critical element in determining whether facilities must report the release and other waste
management quantities of EPCRA Section 313 chemicals. More specifically, errors in threshold
determinations can cause facilities to fail to submit Form Rs for chemicals that meet the
reporting criteria, which may cause the TRI database to understate the magnitude of release and
other waste management quantities. This section considers the following topics when evaluating
how threshold determinations affect the quality of TRI data:
• Approaches commonly used to calculate thresholds (Section 4.1)
• Frequency of incorrect threshold determinations (Section 4.2)
• Reasons for making incorrect threshold determinations (Section 4.3)
The section concludes by reviewing key findings and offering several recommendations
for improving future reporting practices.
This section does not differentiate facilities that submitted Form R reports from those
facilities that submitted Form A reports (i.e., alternate certification statements). Section 7 of this
report provides specific details on the frequency with which facilities choose to submit Form As.
4.1 Approaches Used for Determining Thresholds
This subsection summarizes the approaches that the surveyed facilities used to calculate
thresholds and correlates selected approaches with the accuracy of threshold determinations. For
every EPCRA Section 313 report reviewed during the site visits, surveyors documented the
approach used to determine thresholds using the following categories:
• Purchasing data and inventory records;
• Emission factors;
4-1
-------
• Mass balance;
• Assumed a threshold was exceeded;
• Process recipes;
• Monitoring data;
• Production data; and/or
Other.
During each site visit, surveyors reviewed information in the facility's supporting
documentation to identify which approach, or which combination of approaches, was used to
calculate thresholds. In cases "where a facility's documentation was incomplete, site surveyors
identified an approach that could reasonably have been used to estimate thresholds. Figure 4-1
summarizes the approaches that facilities in SIC Codes 33, 36, and 37 most commonly used to
estimate thresholds for the chemicals that were reported to the TRI database. The figure
indicates some notable trends:
For all industries considered, facilities most often used purchasing data, inventory
records, and process recipes to calculate thresholds for chemicals during RY
1996. Site surveys conducted following RYs 1987, 1988, 1994, and 1995 also
found facilities frequently used these same approaches to calculate thresholds.
The site surveyors generally agreed that facilities' use of purchasing data,
inventory data, and process recipes was an appropriate approach for estimating
thresholds, particularly for raw materials.
Facilities in the primary metals industry (SIC Code 33) and the electronic and
other electrical equipment industry (SIC Code 36) were twice as likely to assume
that a threshold was exceeded than facilities in the transportation equipment
industry (SIC Code 37). Previous analyses of site visit data found facilities that
assume thresholds are exceeded are more likely to make incorrect threshold
determinations than facilities that calculate actual quantities of chemicals
manufactured, processed, and otherwise used. Sections 4.2 through 4.4 revisit the
effect of assuming thresholds are exceeded on the quality of the TRI database.
4.2 Frequency of Incorrect Threshold Determinations
The following analyses indicate the frequency and type of incorrect threshold
determinations made by facilities and project how erroneous determinations ultimately affect the
i
quality of the TRI database. During site visits, surveyors used information provided by facility
contacts to calculate thresholds for all EPCRA Section 313 chemicals that were manufactured,
4-2
-------
processed, or otherwise used. Based on these calculations and the required reporting threshold
quantities, surveyors then listed the chemicals for which the facilities should have submitted
EPCRA Section 313 reports. Errors in threshold determinations were identified by comparing
the list of chemicals that the surveyor determined had exceeded thresholds to the list of
chemicals for which facilities actually submitted EPCRA Section 31.3 reports. These
comparisons yielded four possible outcomes:
• The facility submitted an EPCRA Section 313 report (i.e., either a Form R or a
Form A) for a chemical that exceeded a threshold;
• The facility did not submit an EPCRA Section 313 report for a chemical that did
not exceed a threshold;
• The facility submitted an EPCRA Section 313 report for a chemical that did not
exceed a threshold; or
• The facility did not submit an EPCRA Section 313 report for a chemical that
exceeded a threshold.
In the first two outcomes, the facilities correctly determined thresholds. The third
outcome represents an incorrect threshold determination that causes facilities to unnecessarily
submit Form Rs or Form As. The fourth outcome represents an incorrect threshold
determination, and therefore causes facilities to fail to meet their statutory requirementpf
reporting environmental release and other waste management quantities to the TRI database.
Important "right-to-know" information is then unavailable to the public. V . ,
4-3
-------
o'
I
Q
O
1
o>
Q
2
o
JS
H
a
•S
S
o
-------
Using the four possible outcomes, Tables 4-1, 4-2, and 4-3 and Figure 4-2 summarize the
frequency that facilities made correct and incorrect threshold determinations during RY 1996.
The tables and figure suggest several important observations:
Over all SIC Codes considered, facilities determined thresholds correctly for over
90% of the EPCRA Section 313 chemicals used at their respective plants.
According to a trend analysis of threshold determinations (see Table 4-3), the site
survey project has found that the percentage of making correct threshold
determinations has remained relatively constant over RYs 1987, 1988, 1994,
1995, and 1996.
Facilities visited in the primary metals industry (SIC 33) had the highest
frequency (8.3%) of incorrect threshold determinations, and most of the incorrect
determinations were made for metals and metal compounds. The incorrect
determinations were nearly evenly split between failing to submit EPCRA Section
313 reports for chemicals that exceeded thresholds (4.3% of errors) and
submitting EPCRA Section 313 reports for chemicals that did not exceed
thresholds (4.0% of errors). The impact of incorrect threshold determinations on
the quality of the entire TRI database can be estimated by scaling the results of the
current site visits up to the 6,603 EPCRA Section 313 reports that were submitted
for RY 1996 for all facilities in SIC 33.' For example, the site surveyors
concluded that 11 EPCRA Section 313 reports out of the 75 reports that were
submitted by the selected facilities were actually for chemicals that did not exceed
thresholds (see Table 4-1). This fact suggests that 970 of the 6,603 EPCRA
Section 313 reports filed by facilities in SIC 33 for reporting year 1996 also were
filed for chemicals that did not exceed thresholds. Similarly, the site surveyors
concluded that the selected facilities should have submitted 12 EPCRA Section
313 reports in addition to the 75 that were reviewM, which suggests that facilities
in SIC 33 should have submitted an additional 1,100 reports to TRI (see Table 4-
2).
1 These extrapolations assume the reporting practices of the facilities visited are representative of the reporting
practices of the industry as a whole.
4-5
-------
(I)
T3
O
u
•\
1
1-K
!M
I
>x
bJD
O
3 ^
C3 O
S
P
2
"o
H
«M
o
2
i
o
•a .a
o "•
0 |
U o
s§
eviewed, by
WC Code Is
||
'i "S §
1 §
IfJ
o ?
yt -Si
% 1
^
|S
**^ ^*
•»•*
o S
^D ^J
B fe
3 5j.
a
co *P ^^
|l 1
o e B
U B..B-
i|l
^Q ^-^.
HI
rj pS3 3
Be«
f) W
0> C
a b
C3
^j C
^^ B
53 -c
i
o
S
e
'is
B
1
<1>
O
2
j§
g
J3
H
^
in
r-
•g
[w
"S
2
S
•a
•s
CO
(11
lifted by facilitif
a
•2
CO
O
&1
CO
T— 4
m
a
•§
CO
W
•s
1
c
"3
o
H
oo
S
« "0 -S
Hi
ail
.S
*3 a
II
a
|2-|
-S
4-6
-------
WJ HH
C O
O -M
TT E
JJ ^
•° ^
CS £
H t
is
- s
TS S
O -D
« 1
§S
i-e
r
O
^ ,
o -a ««
*- -2 b
*«^
o
O
S;
CO
o
VO
VO"
O
co
CN
co"
equip
e
VO
CO
I
3,
S1
I
•e
fa
&
I
%
I
It
II
«!
g.s
gj
a a
qj -^-
I*
«s
2 o
•S S
c .:
ox
S3
•P^
•*j
O
cs
Tf «H
« S
3 fe
HO
Q
2
*O
H
4tH
O
I
CJ
e
o
••a
a
<3
*j
Q
•a
"3
to
-------
O
U
U
HH
C/5
en
a
o
.S3
2
S
-------
• Facilities visited in the electronics equipment industry (SIC Code 36) made
incorrect threshold determinations less frequently than facilities in the primary
metals industry (SIC Code 33); however, every incorrect threshold determination
observed for facilities manufacturing electronics equipment resulted in the
facilities failing to submit EPCRA Section 313 reports for chemicals that actually
exceeded thresholds. Most of the incorrect threshold determinations were made
for inorganic acids. Using the same extrapolation scheme presented in the
previous bulleted item, the frequency of incorrect threshold determinations
suggest that, on average, facilities in SIC Code 36 should have filed 530 EPCRA
Section 313 reports in reporting year 1996, beyond the 3,121 reports actually filed
(see Table 4-2). ,
• Facilities visited in the transportation equipment industry (SIC Code 37) made the
fewest incorrect threshold determination errors of the three industries considered
in the latest site survey program. The incorrect threshold determinations for this
industry included metals, inorganic acids, and organic compounds. Of the 75
EPCRA Section 313 reports that site surveyors reviewed for facilities in this
industry, seven were filed for chemicals that did not exceed thresholds. Further,
site surveyors concluded that facilities did not submit EPCRA Section 313 reports
for an additional four chemicals. Extrapolating the frequency of incorrect
threshold determinations to the entire industry, the site survey results suggest that
400 of the 4,331 EPCRA Section 313 reports filed by facilities in SIC Code 37 for
reporting year 1996 were for chemicals that did not exceed thresholds, and that an
additional 230 reports should have been filed by facilities in this industry (see
Table 4-2).
In summary, site surveyors found that in the industries surveyed, 95% of threshold
determinations are made correctly, considering chemicals that both exceed and do not exceed
thresholds. When only chemicals that actually exceed thresholds are considered, surveyors
found that they were correctly reported as exceeding thresholds 88% of the time. Consistent with
findings from previous years, the industry that had the fewest errors (SIC Code 37) was also the
industry that was least likely to assume that thresholds were exceeded.
4.3 Reasons for Making Incorrect Threshold Determinations
This section summarizes why facilities made incorrect threshold determinations. During
the site visits, surveyors classified the reasons for making each incorrect threshold determination
into several general categories. The site surveyors identified the reasons for erroneous threshold
calculations from discussions with facility personnel and from information in supporting
documentation kept at the facility. The following discussion first explains why facilities failed to
4-9
-------
submit EPCRA Section 313 reports for chemicals that exceeded thresholds, then explains why
facilities submitted reports for chemicals that did not exceed thresholds.
4.3.1 Reasons Why Facilities Failed to Submit EPCRA Section 313 Reports for Chemicals
That Exceeded Thresholds
Table 4-4 summarizes the reasons why facilities failed to submit EPCRA Section 313
reports for chemicals that exceeded thresholds during reporting year 1996. As shown in the
table, the most common reason why facilities did not identify chemicals used above threshold
levels was because facilities simply overlooked the use of EPCRA Section 313 chemicals. Only
two of the facilities visited made calculation errors that led to the erroneous conclusion that a
chemical was not used at reportable levels. None of the facilities visited failed to submit reports
due to misclassifying a chemical activity between "manufacture," "process," or "otherwise use."
Other reasons for making incorrect threshold determinations all represent cases where facilities
misinterpreted reporting exemptions or some other aspect of the reporting instructions. The
following lists provide further insight into why facilities made incorrect threshold
determinations, and indicate specific examples of errors documented during site visits:
Overlooking a chemical activity:
• To determine which chemicals exceed reporting thresholds, representatives from a
steel mill calculated annual usage of metal alloys that are added to an electric arc
furnace. The site surveyor noted that scrap metal used at the facility probably
included other metal alloys that were not accounted for by the facility's
calculations. Analytical data for baghouse dusts and hazardous waste manifests
suggested that the steel mill "processed" two additional metal compounds at
levels exceeding 25,000 pounds.
i
I
• A manufacturer of transportation equipment was aware that the painting
operations at the facility used glycol ethers but assumed that the usage could not
possibly have exceeded threshold quantities. Review of purchasing data indicated
that total annual usage was significantly greater than threshold amounts.
4-10
-------
Table 4-4
Reasons Why Facilities Failed to Submit EPCRA Section 313
Reports for Chemicals That Exceeded Thresholds
Reason for not submitting a report
for a chemical that exceeded a
threshold
Chemical was overlooked
Chemical activity was misclassified
Threshold quantity was miscalculated
Other
Total number of errors
Number of chemicals that were not reported for the listed
reason, by SIC Code
(percent of total shown in parentheses)
SIC Code 33
10
(83%)
0
(0%)
0
(0%)
2
(17%)
12
SIC Code 36
2
(29%)
0
(0%)
2
(29%)
3
(42%)
1
SIC Code 37
3
(75%)
0
(0%)
0
(0%)
1
(25%)
4
Misinterpreting reporting instructions:
• A semiconductor manufacturer used hydrochloric acid and sulfuric acid in a series
of enclosed spray-cleaning operations. The facility thought that the "acid
aerosols" activity qualifier for these chemicals applied only to "acid aerosols"
found in ambient air, and not enclosed equipment. The site surveyor noted that
any hydrochloric or sulfuric acid in aerosol form—even aerosols present in
enclosed spray-cleaning operations—should be counted towards thresholds.
• A metals processing facility was aware that it "processed" over 25,000 pounds of
copper, but thought the copper wire was used only as an article. Noting that the
facility's processes cause releases of some copper and change the original size of
the wire, the site surveyor concluded that the article exemption did not apply.
4.3.2 Reasons Why Facilities Submitted EPCRA Section 313 Reports for Chemicals That
Did Not Exceed Thresholds
Table 4-5 summarizes why facilities submitted EPCRA Section 313 reports for chemicals
that did not exceed thresholds during reporting year 1996. Clearly, there were many different
reasons why facilities made incorrect threshold determinations, with no single reason dominating
the site survey results. (The "other" category included several different reasons for making
incorrect threshold determinations.) The following list provides further insight into why
4-11
-------
facilities made incorrect threshold determinations and describes specific instances when site
surveyors concluded that facilities submitted an EPCRA Section 313 report for a chemical that
did not exceed a threshold:
A metals processing facility calculated a 35,000-pound threshold for aluminum,
but did not account for the "fume or dust" activity qualifier. The site surveyor
noted that the facility actually "processed" less than 2,000 pounds of aluminum
"fume or dust."
A facility submitted reports for all chemicals that were "manufactured,"
"processed," or "otherwise used" in quantities greater than 10,000 pounds per
year, including for N,N-dimethylformamide, of which the facility "processed"
approximately 15,000 pounds. The site surveyor noted that the 10,000 pound
threshold applies only to chemicals that are "otherwise used" and that a 25,000
pound threshold applies to chemicals that are "manufactured" or "processed." As
a result, the surveyor concluded that the facility should not have reported for
N,N-dimethylformamide.
Table 4-5
Reasons Why Facilities Submitted EPCRA Section 313
Reports for Chemicals That Did Not Exceed Thresholds
Reason for submitting a report for a
chemical that did not exceed a threshold
Decided to report, even though aware that the
chemical usage was below threshold levels
Chemical activity misclassified
Miscalculated the threshold quantity
Chemical was delisted or modified
Other
Total number of errors
Number of chemicals that were reported for the listed
reason, by SIC Code
(percent of total shown in parentheses)
SIC Code 33
2
(18%)
0
(0%)
1
(9%)
2
(18%)
6
(55%)
11
SIC Code 36
NA
NA
NA
NA
NA
0
SIC Code 37
2
(29%)
3
(43%)
0
(0%)
0
(0%)
2
(29%)
1
NA: Not applicable. None of the facilities in SIC 36 that were visited submitted an EPCRA Section 313 report for a chemical that did not
exceed a threshold.
4-12
-------
A smelting facility kept a detailed materials inventory which indicated that the
facility "processed" over 25,000 pounds of chromium and manganese. The site
surveyor noticed that these metals were trace impurities in the facility's processes
and probably were never present at concentrations exceeding de minimis levels.
Based on a detailed review of Material Safety Data Sheets (MSDS)s, laboratory
analytical data, and usage records, the site surveyor concluded that the de minimis
exemption applied to the metals and that the facility should not have filed the
corresponding EPCRA Section 313 reports.
A facility submitted a Form R for phosphoric acid in every reporting year since
1987. In RY 1996, however, the facility noted that usage of phosphoric acid was
below the corresponding thresholds. Fearing that not submitting a Form R for a
chemical that was previously reported might somehow trigger an audit or
enforcement response, the facility reported for phosphoric acid anyway.
As noted previously, Section 4.4 lists several recommendations to help facilities avoid
making similar errors in future reporting years when determining which chemicals exceed
thresholds.
4.3.3 Chemical Activity Classification
Because appropriate TRI reporting thresholds (e.g., 10,000 or 25,000 pounds) depend on
how facilities use EPCRA Section 313 chemicals, it is important that facilities correctly classify
chemical activities as either "manufacture," "process," or "otherwise use." Table 4-5 indicates
that facilities in SIC Code 37 incorrectly submitted EPCRA Section 313 reports for three
chemicals that exceeded thresholds as a result of misclassifying the chemical activities. To
evaluate how accurately facilities classify chemicals, site surveyors documented activities for all
EPCRA Section 313 chemicals based on information provided by facility contacts and on
observations made during facility tours. Figure 4-3, which compares chemical activity
classifications made by facilities to those made by site surveyors for RY 1996, suggests that the
site surveyors generally agreed with the chemical activity classifications that facilities indicated
on their EPCRA Section 313 reports. Site surveyors did not notice consistent reasons why
facilities misclassified chemical usage.
4-13
-------
a
U u
j» <
Is
U
So
o fe
100
90
80
70
60
50
40
30
20
10
0
Manufacture
Process
SIC Code 33
Otherwise Use
Manufacture
Process
SIC Code 36
Otherwise Use
100
Manufacture
Process
SIC Code 37
Otherwise Use
D Reviewer
• Facility
Figure 4-3. Comparison of Chemical Activity Classifications
Made by Facilities to Those Made by Reviewers
4-14
-------
4.3.4 Impact of Not Calculating Thresholds
An important factor to consider in the accuracy of threshold determinations is whether
facilities actually calculated threshold levels for EPCRA Section 313 chemicals or whether they
just assumed that thresholds were or were not exceeded. At each facility visited, site surveyors
used feedback from facility contacts and data in supporting documentation to determine which
method was adopted to make threshold determinations. For EPCRA Section 313 chemicals
found to exceed reporting thresholds, Table 4-6 summarizes the frequency with which facilities
in the selected industries actually calculated annual usages. Not surprisingly, facilities in the
industries that calculated thresholds least often (SIC Code 33 and SIC Code 36) made more
errors when determining thresholds than facilities in the industry that calculated thresholds more
frequently (SIC Code 37). This observation, which is consistent with findings from site visits for
reporting year 1994 and 19951, suggests that errors in threshold determinations may be
significantly reduced if facilities actually calculate annual usages for EPCRA Section 313
chemicals, as opposed to assuming that chemicals are below or above reporting thresholds.
Section 4.4 further discusses this issue.
4.3.5 Other Factors Correlating with the Frequency of Incorrect Threshold
Determinations
As noted on the Survey Instrument (see Appendix A), site surveyors collected additional
data from facilities on factors that are suspected to affect the quality of TRI reporting. These
additional data include, but are not limited to, the amount of time facilities spend to prepare
EPCRA Section 313 reports, the frequency with which facilities call the EPCRA hotline, and the
title or position of the persons who completed a facility's EPCRA Section 313 reports.
Comparing the frequency of making incorrect threshold determinations with these additional data
revealed two interesting trends:
4-15
-------
Table 4-6
Frequency with Which Facilities Calculated Thresholds for EPCRA
Section 313 Chemicals, by SIC Code
SIC Code
33
36
37
Percent of Chemicals Reviewed by Site Surveyors
Calculated Thresholds
for Which Facilities
58.6%
81.3%
93.7%
Note: Data based on (1) chemicals that facilities submitted Form Rs or Form As and (2) chemicals that were "incprrectly not reported" (see
Section 4,2 for definition).
• Industries in which "facility environmental staff" prepared EPCRA Section 313
reports tended to make the least amount of incorrect threshold determinations.
More specifically, 68% of facilities in SIC Code 37 had "facility environmental
staff prepare EPCRA Section 313 reports, compared to 57 percent of facilities in
SIC Code 36 and just 33% of facilities in SIC Code 33. (See Section 7 for further
information on these distributions.)
• Industries in which the EPCRA Section 313 report preparers call the EPCRA
hotline most frequently were found to make the least number of incorrect
threshold determinations. More specifically, 68% of the facilities visited in SIC
Code 37 called the EPCRA hotline, while 57% of the facilities visited in SIC
Code 36 and only 48% of facilities visited in SIC Code 33 called the EPCRA
hotline. (Again, see Section 7 for further information on these distributions.)
Although the previous associations between the type of staff completing EPCRA Section
313 reports and the frequency-with which staff ^consult the EPCRA hotline appear to correlate
with the accuracy of threshold determinations, further study is needed to determine whether this
trend is statistically significant and applicable to other industries.
4.4 Lessons Learned
In summary, site surveyors found that facilities in the transportation equipment industry
(SIC Code 37) determined thresholds more accurately than facilities in the primary metals and
electronic equipment industries (SIC Codes 33 and 36). Although many factors likely contribute
to this trend, the survey results suggest that facilities that calculate thresholds for EPCRA Section
313 chemicals make fewer incorrect threshold determinations than facilities that assume
thresholds are exceeded.
4-16
-------
Overall, facilities correctly calculated thresholds for 95% of the EPCRA Section 313
chemicals used at the selected industries. However, the frequency of incorrect threshold
determinations suggests that the TRI database might not account for a significant quantity of
chemicals used at reportable levels. More specifically, the site survey results suggest that for RY
1996, facilities correctly reported for 88% of the chemicals that actually exceeded thresholds.
Although the nature and extent of threshold determinations varies from one industry to
the next, some general lessons can be learned from the mistakes identified by the site surveyors.
Section 8 further discusses this issue.
4-17
-------
-------
s.o SOURCES AND TYPES OF RELEASE AND OTHER WASTE MANAGEMENT
ACTIVITIES
This section provides an overview of the sources of on-site releases and both on-site and
off-site waste management activities. It also discusses corresponding release and other waste
management activity types as they pertain to the RY 1996 Form R. Errors and oversights in
identifying potential sources of chemical usage can result in significant errors when estimating
the quantity released and otherwise managed as waste. Similarly, misidentifying the type of
release or other waste management activity can result in data being misclassified in the TRI
database. This section analyzes the frequency and types of errors facilities have made in
identifying the sources and types of release and other waste management activities. The analysis
can help identify the reasons for certain systematic errors made by the regulated community, in
general, and by the specific industries visited, in particular. Once the reasons are identified, EPA
can take appropriate action to help facilities reduce the frequency of errors, and thereby increase
the accuracy of the estimates.
This section considers the following topics when evaluating how release and other waste
management estimates affect the quality of TRI data:
• Distribution of release and other waste management activity sources and release
and other waste management activity types within each SIC Code (Section 5.1);
• Incorrectly reported release and other waste management activity types (Section
5.2);
• Overlooked release and other waste management activities (Section 5.3);
• Calculation methodologies (Section 5.4); and
• On-site waste management activities (Section 5.5).
Trends and corresponding qualitative discussions regarding observations made during the
site visits are presented as applicable, and issues that are specific to individual industries or unit
operations are discussed whenever possible. The information combined with the quantitative
data presented in Section 6 will help determine the primary sources of error in data entered in the
TRI database.
5-1
-------
For the purposes of this report, "sources" mean the streams or unit operations that
generate the potential release or other waste management activity (such as process vents,
container residue, or spills) and "types" mean the ultimate disposition of the release or other
waste management activity corresponding to elements in Sections 5 through 7 of the RY 1996
Form R (such as releases to fugitive air, releases to stack air, discharges to a publicly owned
treatment works (POTW), releases to land, and transfers to off-site disposal). In most cases, this
section presents data both in a tabular form for quantitative analysis and in a graphical format for
qualitative trend analyses. Data are presented for RY 1996 for each of the three major SIC
Codes visited (33, 36, and 37). A trend analysis between these SIC Codes has been conducted
and a general comparison to the findings from previous survey efforts is made when applicable
(see 1994 and 1995 Toxic Chemical Release Inventory Data Quality Report, EPA 745-R-98-002,
for details on results of surveys in SIC Codes 25, 26, 281, 285, 286, and 30).
5.1 Observed Release and Other Waste Management Activities
II
Table 5-1 presents the distribution of sources and the corresponding types of release or
other waste management activity that was observed during the site visits for SIC Codes 33, 36,
and 37. One facility may have multiple sources for a given type. Therefore, a "total" row is
I
included to show the percent of facilities that had at least one source for the given type. Note
j
that, for on-site energy recovery, on-site treatment, and on-site recycling, data were not available
for distribution from specific sources. Figure 5-la presents the "totals" by type of release and
other waste management activity and Figures 5-lb through 5-lg present the data graphically by
source for each release type and management activity. No transfers to underground injection
were reported or observed; therefore, no corresponding figure is presented.
Site-surveyors identified fugitive air releases at most facilities in these SIC Codes (59%
to 74%). However, the percent of facilities with fugitive air releases was less than the percent
observed in site visits conducted in RYs 1994 and 1995, when nearly all facilities had fugitive air
releases. (Releases of 67% in SIC Code 30 and 80% to 100% in SIC Codes 286, 26, 25, 281, and
285.) A lower percentage was expected for the RY 1996 study, because fewer volatile organic
chemicals and significantly more inorganic chemicals are processed and used in SIC Codes 33,
5-2
-------
1
OS
%
OS
£
o
-t)
*N a
ft^S ^
1 I
-d
S3
cs
I
o
a
I
I
£
8 8
:2'|
4> «-J
2 1
'•*•* *^.
2 &
jjjj OS
if
o »
•s s
1*
o
PH
s'
i
0
Jj
JB •
Is
•S
u
53
v«
•o
oa
«
fO
•e
o
y
S3
«
g
o
en
L
^"H
if
g ._§
rl
8 ^^
t
«
00
ro
S
0?
^^
1
oo
^^
^
T-H
S?
Tj"
i>
CA
OO
12
§
06
XO
p^»
<^
Storage tank/stock pile losses
-•5
*
^
f-H
xP
0^
^^
T— 4
1
0
T3
Housekeeping practices/waste
•-§
wS
^1
2
0s-
Y"H
T— I
Accidental spills/releases
xP
0s-
ro
O
0
xP
0s-
t-;
ro
a
§
Volatilization from treatment £
-x=
o
o
0
S?
' *""J
^
s?
p
^
•o
eS
f^
ON
"S
9
e
cu
a
O
"£
5
HN
S
S
TOTAL Reporting Release fi
xP
OO
VO
0?
r-
oo
0s
CO
Os
a
Volatilization from process an
o
xP
in
^
T— 1
x=
O
O
I
xP
00
S
=£
*— 1
^
^
Storage tank/stock pile losses
SN
f)
•0
xP
0s
cn
2;
xP
0s-
O
C5
§"
§
"o
Housekeeping practices/waste
o
o
xP
0s-
•^
0?
O
O
Accidental spills/releases
00
12
V
XP
0s-
«
00
CS
0^
^f«
>>
1
S
Volatilization from treatment i
^o
iri
i
o
S?
o
o
Process discharge streams
S?
o
o
g
o
CN
*o
00
i—)
,
§
06
VC
l>
oo
J-x
^f
en
\
3
e
t/3
4)
e
0
OQ
J
ts
S
e
TOTAL Reporting Release £
o
o
xP
0s-
o
o
xP
0s-
r--
ro
1
S
*o
Housekeeping practices/waste
~a
ca
•
§
0?
ro
wS
xO
0s-
T— H
^
0\
^^
o
Accidental spills/releases
a
cd
0
o
S
g
o
0s?
O
o
0s
^-
^
g
a
| Waste treatment discharge sta
5-3
-------
Pi
6.
i
rn
CO
u
o
3
S3
O
T!
OH
I
&
1
§
ffi
CO
g
C3
^a
tal spi
dental spills/r
C
g
1
CO
6
O.
O
P-I
5-4
-------
V
•er--
3
u
co
CO
CO
O
O
ao
t^
r-
a
o
CO
IT)
a>
1
H
tm
I
g
I
£
P
•8
i/3
t0
s
I
o =a
5-5
-------
1
0)
8P
d
03
CA
I
•d
a
et
a
•I
£
99
JaqjQ pue saseaiajj
5-6
-------
A6J3U3 JO 'BuipAoaa 'saSpnjg juawjeaji
enpisay jaujejuoo
siueajis s&iegosia }uauj}Baji ajseM
seajv luaujjeaJi uioij uo!}ez!|i)E|OA
seajv ssaoojd uioj^ uo»ezui}B|OA
-------
spnpojd-Ag uoijsnqujoo
ABjaug jo 'BuipAoay 'saBpnis juawjBajj.
anpjsay jaujBjuoo
SLUE8JJS 86jBl)OS|a )U3UJ)B9JX 3)SB/\A
SB9JV }UaiUJB8JX LUOJJ UO!}BZ|||)E|OA
sassa|ay/S||ids isjuapjoov
dn-usaio ajSBM/saojpeJci 6u|daa>)asnoH
sasscri and >|oo)g/>)UBx aBBJojg
SB3JV SSaOOJjJ OIOJJ UO!IBZ!|I}B|OA
o
V)
ON
C\
?— i
I
0)
CQ
.2
a
§
o
a
.S3
Q
OJD
.1
saojnos m|M sajjnpEj .jo juaojaj
5-8
-------
spnpojd-^a uoiisnqujoo
spnpoJcj-Ag AjaAOoay
ABJ8U3 jo 'BuipAoay 'ss8pn|S juauijeaji
sasson and M30JS/MUB1 aBsjois
SUJESJJS 96jEqosiQ juaaneaji ajseM
SE9JV JUaLUJBSJl OIOJJ UO!}BZ!|l}e|OA
i
in
dn-UB8|0 ajsBA/usaoipBJd 6ui|asnoH
SBSJV sssoojd LUOJJ UOBBZIIUBIOA
-------
spnpoJd-A'g uoijSnqaioo
spnpojd-A'g AJSAOoay
A6J8U3 jo 'BujpAoay 'ssBpnis }U8W)B8Ji
SLUE3JJS BBjElJOSia )U8U1}B9J1 8)Se/\A
SB8JV }U8UJte8JJ. LUOJJ UOI)BZI|l)E|O/\
S8SSOT 8|]d >|00}S/>|UB1 86BJOJS
OT
C/l
i
o
a
03
I
I
o
02
H
o
•-C
.c
SB8JV SSSOOJd LUOJJ UOI)BZj|j)B|O/\
-------
sionpaid-Ag uoijsnqiuoo
sjonpoJd-Ag /OsAooay
'BuipAoay 'saBpnis }uaui)Baii
SLUBSJJS eBjeqosia }U8imesJi SISBM
SB8JV lUSUflESJl 0104 UOI)BZ!|!}B|OA
o
w
dn-ueao 8}SB/wsao!pBJd 6u|daa>|asnoH
S83JV SSSOOJd LUOJJ UO!}BZI|l}e|OA
o
PH
a
•§•
o
CO
«M
p
I
1
B..
£
5-11
-------
spnpoJd-Ag uo|)snqiuoo
spnpojd-Ag AjaAooay
A&iaug jo 'BuipAoay 'saBpnig luaiujeaii
snpjsaa JSUIBJUOO
jjouny j9)8Miuiojg
SBSJV juaaijeaji uiojj uo|}ez||!)e|OA
sasea|9y/s|||ds leiuapioov
S
CO
sa6uey/s9A|BA/sduind
seajv ssaooJd uioj; uo!}ezj|i}E|o/\
SUJEOJJg 86JEHOSIQ SS30OJd
saojnog IOI/M saumoej jo ;ueojad
>
'•§
s
on
C5
fl
03
£
SM
M
6
•a
s
I
.0
fl
C3
H
s
s
o
a
|
•M
I
DJO
M)
£
5-12
-------
36, and 37 than in the previously studied SIC Codes. A similar trend is seen when comparing
stack air releases.
Site surveyors observed at least one type of off-site transfer for release and other waste
management activities at nearly all facilities (78%, 86%, and 100% in SIC Codes 33, 36, and 37,
respectively). In RYs 1994 and 1995, site surveyors observed a significantly smaller percent of
facilities with off-site transfers for release and other waste management activities (90% in SIC
Code 25 and 55% to 70% for all other SIC Codes). The number of off-site transfers for release
and other waste management activities increased for the RY 1996 study, because facilities in SIC
Codes 33, 36, and 37 generally collect a large quantity of "scrap" or "unusable" material
containing metals and send it off site for waste management activities.
Many facilities also discharged EPCRA Section 313 chemicals to water, both indirectly
discharged to a POTW and directly discharged to surface water. The percent of discharges to
water is slightly higher than that observed during site visits to facilities in RY 1994 and RY
1995. Underground injection was never observed during site visits to facilities, in contrast to
what was seen in RY 1994 and RY 1995.
5.2 Incorrectly Reported Release and Other Waste Management Activity Types
This section presents the release and other waste management activity types that facilities
misclassified and overlooked. An analysis of these results can help identify the sections of the
RY 1996 Form R that cause confusion for the regulated community. A comparison with site
surveys conducted in individual SIC Codes can help identify areas of confusion that are specific
to various industries. These data alone cannot be used to quantitatively assess the accuracy of
data in the TRI database because the magnitude of errors in estimates is not considered. Section
6 presents a detailed quantitative analysis. Table 5-2 and Figure 5-2 presents the percent of
misclassified or overlooked release and other waste management activities.
5-13
-------
•a &
>—' M
i—i W
ra
A
§
4)
£
o
"S
^
i
§!
•§
1
o
1
Pi
t~
«
U
U
r/i
\o
U
U
3
«
,
To Off-Site Energy Recover
s
oo
o\
v->
To Off-Site Recycle
o
o
oo
o!
Os
CS
1 To Off-Site Treatment
5-14
-------
po>|OO|jaAO JO p9|j!SSE|DS|ifj spodey jo tuaojad
5-15
-------
A comparison of the release and other waste management activities reported by facilities
with those identified by site surveyors showed that a large number of facilities reported the
wrong release or other waste management activity type. This section discusses those types that
were incorrectly reported and presents a qualitative discussion regarding the corresponding error.
5.2.1 Air Releases
A small number of facilities misclassified air emissions or overlooked them entirely.
Although facilities often had difficulty in quantifying their releases, most recognized they existed
and made attempts to calculate release estimates from most sources. The facilities' most
common source of error was their failure to identify fugitive emissions of metals or metal
i
compounds when processes were conducted at extremely high temperatures. Facilities also
commonly failed to report EPCRA Section 313 chemicals in particulate matter that escaped from
• • „ i
baghouses or other dust collection systems. The number of facilities in SIC Codes 33, 36, and 37
that misclassified or overlooked air emissions is lower than that observed in RYs 1994 and 1995.
A primary reason is that many facilities visited in RYs 1994 and 1995 misclassified releases to
general room air. Facilities in SIC Codes 33, 36, and 37 were likely to have process air releases
to large, open areas rather than in an enclosed room. Therefore, the potential for this mistake did
not occur. Another common error hi RYs 1994 and 1995 was overlooking stack releases from
.! ' I
storage tanks. However, this error was not detected in facilities in SIC Codes 33, 36, and 37 as
i
these facilities were less likely to have storage tanks.
5.2.2 Off-Site Transfers for Recycling and Disposal, and On-Site Recycling
Many facilities hi RY 1996 misclassified or overlooked off-site transfers for release and
i
other waste management activities. In particular, transfers off-site for disposal and off-site
transfers for recycling were often misreported, in addition to on-site recycling.
A primary reason for misclassified release and other waste management activities is
confusion over "direct reuse" (which is not reportable) and "recycling" (which is reportable).
For example, facilities often reported large quantities of off-specification products containing
:|
metals that were directly reused in secondary smelting operations (without further waste
5-16
-------
management) as sent off-site for recycling. Other facilities reported scrap, slag, and dust even
though it was directly reused in the production of asphalt as either off-site or on-site recycling.
Similarly, facilities misclassified on-site recycling when various process streams were directly
reused.
A frequently observed error for transfers off-site for disposal was overlooking large ,
quantities of EPCRA Section 313 chemicals (typically metals) that were present in dust collected
in baghouses, electrostatic precipitators, and rotoclones. This dust is often disposed to landfills.
Finally, some facilities reported transfers off-site for recycling without knowledge of how the
waste was actually handled. In these instances, most facility contacts could not provide a basis
for claiming recycling and indicated that the material may actually be disposed.
Many of these facilities recognized that they may have been misreporting and expressed a
desire for clarification on the reuse/recycling issue in general and its applicability, specifically to
typical operations in SIC Codes 33, 36, 37. A comparison to previous data shows that many
facilities in RYs 1994 and 1995 also frequently misclassified off-site transfers for release and
other waste management activities. However, those facilities did not typically have "recycling"
vs. "direct reuse" concerns.
5.2.3 On-Site Treatment and Land Disposal
On-site treatment and on-site land disposal were overlooked at a large number of
facilities (not misclassified), in particular in the primary metals industry (SIC Code 33). Most
facilities failed to consider the removal of dust from an air stream as "treatment or removal"
when it applies to metals and metal compounds. Some facilities felt that because the metal was
not destroyed, it should not be reported in Section 7A as being treated. [The 1996 TRI
instructions say that the waste treatment efficiency reported must represent physical removal of
the parent metal from the waste stream, p.41] Other facilities entirely overlooked the dust being
treated and its subsequent disposal to land. This situation was observed more frequently in
facilities that employed large dust collection systems and when metals were present in the
process.
5-17
-------
5.2.4 Water Discharges
Discharges containing EPCRA Section 313 chemicals from facilities in SIC Codes 33,
36, and 37 to POTWs and receiving streams were less prevalent than observed at site visits for
RYs 1994 and 1995. However, there were some instances where discharges were overlooked.
5.3 Overlooked Release and Other Waste Management Activities
Section 5.2 discussed the release and other waste management activity types that were
either misclassified or overlooked. This section identifies the process or unit operation sources
that were overlooked. An analysis of this information may be used to identify specific unit
operations or processes that are problematic for EPCRA Section 313 reporting. Additional
guidance and a focussed effort to analyze the fate of EPCRA Section 313 chemicals from these
,i
sources will increase the accuracy of data in the TRI database. Again, these data do not reflect a
quantitative measurement of the estimates associated with release and other waste management
activities, but a quantitative analysis of the estimates is presented in Section 6.
In general, fewer facilities in SIC Codes 33, 36, and 37 completely overlooked release
and other waste management activity sources compared to those visited in RYs 1994 and 1995
(although the magnitude of the associated errors may be greater).
As shown on Table 5-3 and Figure 5-3, several facilities overlooked some sources of
release and other Waste management activities entirely. In some cases, the result was an
underestimation of the overall quantity of the toxic chemical managed as waste by the facility.
However, in cases where a mass balance was used as the method to determine the quantity of the
•|
toxic chemical managed as waste, the facility may have included the quantity that was
overlooked in another source. For example, a facility may have overlooked a 1% stack air
release from a dust collection system that is 99% efficient. However, after conducting a material
balance and analyzing the total throughput, the facility may have assumed this quantity was
released from process areas as fugitive emissions. In this case, the stack release to air would
have been under reported, while the fugitive air emissions would have been over reported.
5-18
-------
>—N
•9
|
IE
"8
i
s
f
o
&
o
**
w
*
-
«
S
VO
4)
e
53
«
*«
5
w
en
•
1
CO
vq
So
S
*:
*
1
1
u
ff-i
0
0
o
Storage tank/stock pile losst
a<
0
Q
O
o.
1
s
Housekeeping practices/wa
IN
o
o
o
Pumps/valves/flanges
5
£)
VO
oo
vd
CO
s
«
Volatilization from process
o
o
o
o
o
o
Process vents
3
"Sources of "other" include:
5-19
-------
spnpoJd-Ag uojjsnquioo
sjonpoJd-A'g AjaAOoay
X6J9U3 JO 'BujpAoay 'sa6pn|g tusiujeaji
siueaiis aBjeyosjo JUSIUIBSJX a;seM
seajv juauujeajj. LUOJJ UOIIBZHUBIOA
siusA ssaoojd
seajv ssaoojd UIDJJ uonezi|i)E|OA
o
.OT
sasson and >|OO}SAUEI aBeJOts
anpisay jau]E}uoo
p»H°°|J»AO puc pouissBiosiw spodoy jo juaojoj
5-20
-------
This error was observed on a site-specific basis, typically when a mass balance was
used for facility-wide estimates. Many facilities overlooked sources entirely and did not account
for the associated EPCRA Section 313 chemicals when reporting for any types of release and
other waste management activities on the Form R. The sources most often overlooked were
container residue (typically as liquid residue in "empty" drums), stack emissions of particulate
matter, transfers and disposal of collected particulate matter, and transfers or management of off-
specification product.
5.3.1 Container Residue
hi RY 1996, the largest source of overlooked release and other waste management
activities (considering frequency, not overall quantity) was from container residue, as was the
case in RYs 1994 and 1995; however, both the frequency and magnitude were considerably less
in SIC Codes 33, 36, and 37. A main reason for the decrease in errors is that fewer facilities in
these SIC Codes purchase or use drums of organic liquids, which decreases the opportunity to
overlook container residue. Although the EPCRA Section 313 instructions specify that container
residue should be considered as a release or other waste management quantity, facilities assumed
that all used drums, totes, or small containers were completely empty and the subsequent transfer
of the empty containers off site for disposal (disposal on site was rarely observed in SIC Codes
33, 36, and 37) did not result in any release or other waste management activities of EPCRA
Section 313 chemicals. Many facilities did not consider the potential for reportable quantities of
residual chemicals in these containers. Other facilities considered this potential release or other
waste management quantity but felt it was negligible (and did not report it) if drums were
shipped as "empty", as defined by federal and/or state shipping regulations.
In practice, liquids are often removed from drums by gravity draining or by pumping.
Neither of these methods removes all material from the drum and an appreciable quantity may
remain. Hazardous Materials Transportation Act (HMTA), Hazardous Materials Transportation
Uniform Safety Act (HMTUSA), and Resource Conservation Recovery Act (RCRA) regulations
require special handling precautions when transporting drums containing hazardous materials
(drums are often defined as "RCRA empty" for shipping purposes if they contain less than one
inch of a liquid substance). Therefore, facilities often remove the materials in the containers to
levels which are below state or federal regulations, but they do not completely empty them. It
5-21
-------
should be noted that some facilities (less than was observed in RYs 1994 and 1995) sent
hundreds of "empty" drums off-site that potentially contained some residual EPCRA Section 313
i
chemical. This transfer results in a significant quantity released or otherwise managed as waste
mat was overlooked.
•
•
Some facilities also overlooked release and other waste management quantities due to
residual powdered EPCRA Section 313 chemicals in empty bags. This quantity was
I
significantly less than observed in previous reporting years, primarily because pigments in the
form of solids are commonly used in painting operations (SIC Code 285, RY94), while polymer
beads were the only solid material that result in container residue observed at more than one site
in SIC Codes 33, 36, and 37. Note that most facilities that add powdered metal alloys to molten
i
processes do so by dumping the alloy, including its container, to the kettle. Therefore, in this
instance there is no container residue.
Most of the liquid release and other waste management quantities from overlooked
container residue should have been reported as off-site transfers for disposal. However, some
i
should have been reported to off-site recycling, off-site treatment, or off-site energy recovery.
i i
Other overlooked liquid discharges should have been reported as discharged to either POTWs or
to surface water streams because the drums were rinsed on site and the rinsewater was collected
i
and sent to the local POTW or receiving stream. Most overlooked solid releases from bag
residue should have been reported as being disposed to on-site landfills or to off-site disposal.
5.3.2 Process Areas and Discharge Streams
The next most overlooked sources were volatilization from process areas and process
discharge streams. Many facilities in the primary metals industry (SIC Code 33) overlooked
significant quantities of metals that volatilize during smelting and foundry operations. Metals
and metal compounds often volatilize or become entrained as fumes or dust during these
operations. The concentration of these metals in the air stream is often small (in the parts per
million range); however, the throughput is so high that a significant quantity may be overlooked.
!, , , II
Facilities often failed to apply metals with the "fume or dust" qualifier to the manufacturing
threshold. Similar oversights were observed, although less frequently, in SIC Codes 36 and 37.
5-22
-------
Overlooking EPGRA Section 313 chemicals in air streams also resulted in
underestimating or overlooking the quantity from other sources, such as dust that is ultimately
released or disposed in stormwater, other process discharge streams, and waste treatment
discharge streams.
5.3.3 Other Treatment Areas
Facilities rarely overlooked volatilization from treatment areas, most likely because
most treatment chemicals are either non-volatile or are completely destroyed during the treatment
process. This observation is consistent with RYs 1994 and 1995 surveys. Similarly, release and
other waste management quantities from storage tank losses; releases from process vents, pumps,
valves, and flanges; and housekeeping wastes were rarely overlooked.
5.3.4 Combustion
Site surveyors did not identify any overlooked release or other waste management
activities from combustion by-products at facilities hi SIC Codes 33, 36, or 37. In contrast to the
facilities visited for RYs 1994 and 1995 survey programs, very few of the facilities visited for the
RY 1996 survey used on-site boilers, industrial furnaces, or incinerators.
5.4 Calculation Methodologies
EPA requires facilities to designate one of four calculation methodology categories
used for each release or other waste management activity estimate (monitoring data, mass
balances, emission factors, and engineering judgment or calculations). Table 5-4 presents the
distribution of calculation methodologies that were used to determine estimates for each release
or other waste management activity type. An analysis of the methodologies used and how
frequently facilities used the best available methodology provides insight on the accuracy of
some estimates and on the reason for some errors.
It was observed during the review of facility notes that facilities often used multiple
methods or reported a method that was inconsistent with the method actually used. Therefore,
the data reported in Table 5-4 represents the site surveyor's opinion as to the primary method
5-23
-------
actually used by the facility, not necessarily the method reported on the facility's Form R (and
entered in the TRI database). Using the surveyor's opinion allows for a better analysis of data
accuracy when compared to the actual methods used. Additionally, a significant number of
facilities used hazardous waste manifests to calculate estimates of off-site transfers. Site
surveyors noted these occurrences, when applicable. Their frequency of use is presented along
with the four EPA-accepted methods. It should be noted that few or no release or other waste
management activity quantities were reported for several release or other waste management
activity types. In these circumstances the table shows zero percent.
5.4.1
Air Releases
For RY 1996, nearly all facilities reported at least one fugitive release. As in RYs
1994 and 1995, site surveyors observed that fugitive releases were typically the most difficult for
facilities to estimate. Engineering calculations were the predominant method used by most
facilities. Site surveyors observed that many facilities actually used one or more of the methods
;. , i
to estimate fugitive emissions, and then applied engineering judgment to total the emissions from
all sources. They used engineering judgment for partitioning releases between stack and fugitive
if monitoring data were not available. Mass balances, monitoring data, and emission factors are
presented in Table 5-4 only when they were the predominant method used.
It was uncommon for facilities to have access to monitoring data for fugitive releases.
However, facilities did use it when available (typically in the form of periodic leak tests). Only
a few facilities used emission factors, a contrast to survey results from RYs 1994 and 1995 when
•,i i
several facilities used emission factors. Many facility contacts inquired whether emission factors
i
that were relevant to their processes existed (for fugitive or stack emissions). They were not
aware of EPA-published factors or any relevant trade association factors and very few had
,i
conducted testing to develop facility-specific emission factors. The type of emission factors
used and a subsequent discussion is presented below.
5-24
-------
«
o>
o\
CS
ts
cs
o
o
g
i
o
10
oo
ts
3
55
£
o
£
g
•5
I
o\
o
c>
ts
ts
o
CO
i
s
o
i
i
en
00
•a
o
1
o
Oi
1
i
VO
co
vd
i
rn
c
-------
CD
I
§
3
O)
I
is
"o
1
I
§
•jig
o
fa*
«*•!
o
•*•»
s
u
a>
P«
-*
!
. i
|j
S
1
*
c
ro
1
U
U
55
S
1
HH
C/3
fl
f3
0)
"8-
o
u
03
1
"3
•8
JS
•*^
i
i
i
u
u
.»
i
p
Energy Recc
S
'dp
o
0
s
o
o
o
i
I
"ea
u
M
I
Recycling
-•§
bo
S?
o
o
o
o
o
0
•o
60
S
1
12
documen
J
"
S
}
1
bo
.g
1
C
p
fi
3
CL|
fi
8
c
o
;|
s
o
:s, facility or te
i
f
CO
.Ir
1
!?
es include: off
facility not
2
is
1
lethodologies ;
c
S
.^
P
5-26
-------
Due to a lack of monitoring data and relevant emission factors, facilities used mass
balances to determine fugitive releases from at least one process line or unit operation when a
material balance around the entire facility resulted in a chemical quantity that was unaccounted
for.
Most facilities also reported a stack release. Although facilities had difficulty in
estimating these releases, they typically indicated less difficulty in identifying and quantifying
these releases than observed with fugitives. Engineering calculations and mass balances were the
most often used methods. However, surveyors also observed the use of emission factors and
monitoring data (actual releases from stack tests). As with fugitive emissions, few facilities were
aware of published emission factors that applied to their processes. However, some had
conducted stack testing and used the appropriate monitoring data to develop site-specific factors
accordingly.
5.4.2 Water Discharges
Many facilities reported discharges to POTWs and/or surface water. Using
monitoring data was the primary method to calculate POTW and surface water discharges.
Typically, discharges were monitored for compliance with various local, state, or other federal
regulations, resulting in an accurate estimate. If monitoring data were not available, facilities
typically used a mass balance around processes involving contact water to determine the quantity
of EPCRA Section 313 chemical that could not be accounted for. Then, engineering judgment
(usually based on knowledge of chemical volatility and solubility) was used to estimate a
partition factor between releases of the unaccounted quantity that would be lost to fugitive air
versus the quantity discharged to water.
5.4.3 Off-Site Transfers for Release and Other Waste Management Activities
Table 5-4 shows that most facilities used monitoring data and/or hazardous waste
manifests to estimate off-site transfers for release or other waste management. Monitoring data
came from two main sources: (1) periodic facility sampling of the process waste streams that
were collected prior to shipment, and (2) sampling conducted by the receiving facility.
Documentation for this data was typically more prevalent and more complete than methods used
5-27
-------
to estimate release and other waste management activities to most other sources. However, in
contrast to data observed in RYs 1994 and 1995, test data often provided only the concentration
of EPCRA Section 313 chemicals. Facilities often made mistakes in calculating the throughput
of material sent off site. Also, many sources were overlooked and some transfers were reported
when they were actually directly reused (and not reportable). Therefore, the accuracy of the
overall off-site transfer estimates is questionable.
5.4.4 Correct Methodology Usage
Figure 5-5 presents the frequency with which the site surveyors felt the method used
by the facility would result in the most accurate estimate, based on information and data
available to the surveyor at the time of the site visit. It does not present the frequency that the
facilities correctly calculated the quantity of release or other waste management activity. This
figure shows concurrence with the selected method in most cases.
As observed during surveys from RYs 1994 and 1995, it should be noted that during
many visits the surveyor identified another, more accurate method that could have been used to
, t ;, i
estimate release and other waste management quantities, if a particular variable had been tracked
for RY 1996. In many cases, the facility contact indicated that it would have been fairly easy for
the facility to implement the suggestion and that they planned to take the surveyors' advice for
j
subsequent years. However, it was not always possible to recreate the required variable.
Another limitation to this analysis is the fact that surveyors often identified a more accurate
method that could be used based on data the facility claimed to have, but the facilities were
unable to gather the information immediately for use by the site surveyor.
5-28
-------
Table 5-5
Frequency the Best Methodology was Used by Facilities to Estimate
Release and Otherwise Waste Managed Quantities
SIC Code
33
36
37
Frequency
85.9%
99.3%
87.7%
5.4.5
Emission Factors
Chemical-specific emission factors were sometimes used to estimate fugitive and
stack releases. EPA instructed site surveyors to determine the type of chemical-specific emission
factors used, when applicable. The potential types were designated as facility-derived, EPA-
approved or published, trade association-derived, and other. Table 5-6 presents the percentage of
use for each type of chemical-specific emission factor. «•-...
These factors were typically employed to estimate fugitive releases of volatile
chemicals from process areas or piping (leaks from pumps, valves, flanges, etc.) or to estimate
stack releases from storage tanks and stack releases from gasses generated by unit operations that
were channeled through air pollution control devices (typically baghouses). Non-chemical-
specific factors in trade association guidance or derived by the facility were treated as
engineering calculations. '
5-29
-------
Table 5-6
Types of Emission Factors Used for Fugitive and Stack Air Releases
Release Type
Fugitive
Stack
Release Source
Facility derived
EPA derived
Trade Association derived
Other
Facility derived
EPA derived
Trade association derived
Other
Percent (by report)
SIC Code 33
83.3%
0.0%
16.7%
0.0%
100.0%
0.0%
0.0%
0.0%
SIC Code 36
100.0%
0.0%
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
SIC Code 37
25.0%
25.0%
25.0%
25.0%
75.0%
0.0%
25.0%
0.0%
5.5
On-Site Recycling, Treatment, and Energy Recovery
in • ' i
With the exception of gases routed through dust collection systems, the RY 1996
survey finds that EPCRA Section 313 chemicals were rarely managed on site (recycling,
treatment, or energy recovery). Table 5-2 and Figure 5-2 show that some facilities incorrectly
identified these waste management activities. Additionally, EPA recognized the potential
confusion in reporting requirements for on-site treatment in Section 7A and Section 8.6 of the
EPCRA Section 313 report. Therefore, site surveyors specifically determined whether the
quantities reported were sent to treatment versus actually treated. Facilities typically correctly
I
identified on-site treatment activities when they existed (with the exception of significant
confusion regarding whether to report dust collection systems, particularly for metals entrained
in the dust). Only a few facilities incorrectly reported, as shown in Table 5-7.
5-30
-------
Table 5-7
Facilities Incorrectly Reporting the Quantity Sent to
Treatment Rather than Actually Treated
SIC Code
33
36
37
' Percent Incorrectly Reported
7.4%
14.3%
5.3%
It should be noted that the quantitative values presented in Figure 5-2 and Table 5-2 regarding on
and off-site recycling may not be accurate because most facilities were confused by the definition
of "recycling". EPA recognized that this potential might exist and instructed site surveyors only
to analyze release and other waste management activities to recycling activities if the facility
reported them. Therefore, site surveyors only recorded on-site recycling as incorrect if such
activities were claimed but did not exist.
Site surveyors discussed on-site management issues and acquired feedback from
facility contacts. The primary concern raised was that definitions of the terms "recycling",
"direct reuse", and "waste management" are generally unclear. Additionally, facilities felt that
these terms were particularly confusing when applied to large quantities of off-specification
material that was "recycled", either on site or off site. Tens of thousands of pounds of metals
could be involved in these processes.
Facilities also expressed confusion about how to report on-site treatment of metals or
metal compounds. Many realized that metals cannot be treated for destruction; however, they
can be removed from a process waste stream. Facilities questioned whether this removal should
be considered "treatment" in Section 7A, and whether the removal efficiency (opposed to the
destruction efficiency) should be reported. [Facilities should report the removal efficiency of
metals in the waste stream.]
Very few facilities were confused or concerned regarding when to report EPCRA
Sections 313 chemicals sent to treatment versus those chemicals sent to energy recovery, perhaps
because relatively few EPCRA Section 313 chemicals in these SIC Codes were incinerated.
5-31
-------
On-site recycling was rarely claimed at these facilities. Typically recycl
included off-specification product, process solvents, or waste dust collected
Tables 5-8 and 5-9 summarize data that were collected for on-site recycling
during site visits. Table 5-9 presents the frequency that each EPCRA Section
chemical category was recycled, as reported by these facilities.
Table 5-8
Observed On-site Recycling Activities
e streams
in a baghouse.
that was observed
313 chemical or
# Of Facilities
Reporting
1
2
1
1
1
Type of Recycling Claimed
Other
Other
Metals Recovery, Electrolytic, and
Ion Exchange
Other (Process Discharge Stream)
Spent Process Solvent
Description of Recycling Stream
By-product
Polymer remelt or "reshred"
Spent metal plating bath
Product grinding-back to process stream
Solvents/organics recovery batch still
SIC Code
33
33
36
37
37
Table 5-9
Chemicals For Which On-Site Recycling Was Claimed
(SIC Codes 33,36, and 37 Combined)
EPCRA Section 313 Chemical or Chemical Category
Copper or copper compounds
Lead compounds
Antimony compounds
Hydrogen fluoride
Methyl ethyl ketone)
Phenol
Toluene
Xylene (mixed isomers)
Number of Facilities Reporting
2
2
1
1
1
1
1
1
5-32
-------
6.0
RELEASE AND OTHER WASTE MANAGEMENT ACTIVITIES
Release and other waste management estimates are the most highly scrutinized and
publicized data in the TRI program. Thus, comparing the facility estimates to the surveyor
estimates gives an indication of how accurately the facilities in the three SIC Codes have
reported. This section discusses release and other waste management estimates made by
facilities and site surveyors. Major differences in release and other waste management estimates
between the facilities and site surveyors are noted, and the reasons for the differences are
explained. The following topics are discussed in each subsection: ,
• On-site release and other waste management estimates, as reported in Section 5 of
the Form R (Section 6.1)
• Off-site transfers for release and other waste management quantities, as reported
in Section 6 of the Form R (Section 6.2) „ ' .
• On-site release and other waste management activities as reported in Section 7 of
the Form R (Section 6.3)
• Production ratio/activity index (Section 6.4)
• Source reduction (Section 6.5)
A discussion of the methodology used by the site surveyors to gather the data necessary to
estimate the release and other waste management quantities is contained in Section 2. A
discussion of the specific techniques used by the facilities and by the site surveyors when
estimating release and other waste management quantities is presented in Section 5.
Facilities are required to report estimates of release and other waste management by
chemical for each release and other waste management activity type. On-site release and other
waste management activities (as reported in Section 5 and Section 8.1 of the Form R) must be
apportioned among the following five categories: -
Fugitive or non-point air emissions;
Stack or point air emissions;
Discharges to receiving streams or water bodies;
Underground injection on site; or
Releases to land on site.
6-1
-------
Off-site transfers for release and other waste management activities (as reported in Section 6 and
Sections 8.3, 8.5, and 8.7 of the Form R) are categorized according to how the waste is managed:
j
— Discharges to POTWs;
— Off-site transfer for disposal;
— Off-site transfer for treatment;
— Off-site transfer for recycling; and
— Off-site transfer for energy recovery.
Facilities also report on their on-site waste management activities in Sections 7, 8.2, 8.4, and 8.6
' i r. i
of the Form R according to these type categories:
6.1
— On-site treatment;
— On-site recycling; and
— On-site energy recovery.
i
Estimates of On-Site Release and Other Waste Management Quantities as
Reported in Section 5 of the Form R
Section 6.1.1 compares the estimates of on-site release and other waste management
!
quantities between the facilities and surveyors. Section 6.1.2 compares the scaled up facility
estimates to the TRI database.
6.1.1 Comparison of the Facility Estimates to the Surveyor Estimates
'i
To assess the accuracy of the estimates reported by the facilities, the facility estimates
for each medium were compared to those calculated by the site surveyors. First, the chemical-
specific estimates were summed at the facility level for each release and transfer medium. Next,
the estimates were totaled by type for facilities in the SIC Codes 33, 36, and 37. These totals for
, j
each of the SIC Codes were compared to evaluate overall accuracy within and among the
'!
industries. Because the chemical-specific estimates are combined by type, the accuracy of site-
specific estimates for each chemical at each facility is not evaluated in this report. Such
information was provided to the facility at the time of the site visit.
6-2
-------
Tables 6-la through 6-lc show the percent difference between the facility estimates
and the site surveyor estimates for each release and transfer medium in Section 5 of the Form R
for SIC Codes 33, 36, and 37. The percent difference is calculated as:
Percent Difference = (Fa - SS)/(SS) x 100
where: Fa = Facility Estimate
SS = Site Surveyor Estimate
The site surveyor estimates were used as the basis for comparison as they are a more accurate
representation of "true value" than the facility estimates. Negative percent difference values
indicate that, overall, the facilities surveyed underestimated the release or other waste
management activity, while positive values indicate an overestimate. These differences are
depicted graphically in Figures 6-la through 6-lc. Note that none of the surveyed facilities had
transfers to underground injection.
Figures 6-2a through 6-2c illustrate the percentage of facility estimates which were
greater than, equal to within 5%, or less than the surveyor estimates for SIC Codes 33, 36, and
37, respectively. The number of facilities with each type is given in parentheses below each
release and other waste management activity type.
Release and Other Waste Management Estimates in the Primary Metals Industry, SIC
Code 33
For the primary metals industry, SIC Code 33, 27 facilities were surveyed and
estimates for 74 EPCRA Section 313 chemicals were reviewed. Comparing the facility estimates
to the site surveyor estimates (Table 6-la), the largest discrepancies in total pounds are for on
site land disposal (480%). For on-site land disposal, the large difference can be attributed to one
facility. This facility completed the wrong column on the Form R and should have reported the
chemical as sent off-site for disposal. The closest agreement is for discharges to receiving
streams or other bodies of water at 0.5% (based on five facilities). The total percent difference
for all on-site release and other waste management activities in this SIC Code is 0.22%.
6-3
-------
Release and Other Waste Management Estimates in the Electronic and Other
Electrical Equipment Industry, SIC Code 36
In the electronic and other electrical equipment industry, SIC Code 36, a total of 14
facilities were surveyed. Site surveyors reviewed and estimated release and other waste
management activities for 48 EPCRA Section 313 reports. The range of the percent differences
in the estimates in SIC Code 33, is 1.5 to 53.1% (see Table 6-lb). The total percent difference
for all on-site release and other waste management activities in the SIC Code is -24.5%.
Release and Other Waste Management Estimates in the Transportation Equipment
Industry. STC Code 37
A total of 19 facilities were surveyed in the transportation equipment industry, SIC
,l
Code 37, and 64 EPCRA Section 313 reports reviewed. Total estimates for each type by facility
are within 10% of the site surveyor estimates, with the exception of on-site land disposal which
' i
differed by 212% (see Table 6-lc). Note, however, that total releases of this type were only 255
pounds at the facilities visited.
Table 6-la
!|
Summary of SIC Code 33 On-site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form Ra
Type6
Fugitive air
Stack air
Receiving stream
Land on site
Total
Release and Other Waste
Management Quantities
Reported by the Facilities
(million pounds)
0.112
0.784
4.55E-03
0.029
0.930
Release and Other Waste
Management Quantities
Estimated by the
Surveyors
(million pounds)
0.121
0.797
4.53E-03
0.005
0.928
Percent Difference0
-7.75%
-1.63%
0.541%
480%
0.22%
T>!o underground injection was reported.
"Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
6-4
-------
Table 6-lb
Summary of SIC Code 36 On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form Ra
Type"
Fugitive air
Stack air
Receiving stream
Land on site
Total
Release and Other Waste
Management Quantities
Reported by the Faculties
(million pounds)
0.027
0.045
6.36E-04
1.35E-03
0.074
Release and Other Waste
Management Quantities
Estimated by the
Surveyors
(million pounds)
0.022
0.074
6.26E-04
8.80E-04
0.098
Percent Difference0
22.9%
-39.4%
1.52%
53.1%
-24.5%
T^umber of facilities = 14; number of EPCRA Section 313 reports represented = 48.
T^o underground injection was reported.
"Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
Table 6-lc
Summary of SIC Code 37 On-Site Release and Other Waste Management
Quantities as Reported in Sections 5 of the Form Ra
Type"
Fugitive air
Stack air
Receiving stream
Land on site
Total
Release and Other Waste
Management Quantities
Reported by the Faculties
(million pounds)
0.268
1.08
0.0
7.96E-04
1.35
Release and Other Waste
Management Quantities
Estimated by the
, purveyors
(million pounds)
0.297
1.11
0.0
2.55E-04
1.41
Percent Difference"
-9.96%
-2.78%
0%
212%
-4.26%
dumber of facilities = 19; number of EPCRA Section 313 reports represented = 64.
Tfo underground injection was reported.
"Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
6-5
-------
.
OT
a
E
.s e
u
r
I
9
a
O>
•4->
1
(U
a u
o s
4) 0>
e§ ^
*"§
o «
•s
03
1S
_; »a
B. A
U
cs
vo
S8|)j)ueno )uouia6euey\|
ajSBM jaqK) PUB esBejay ^o sa}eiu!}sa
.a
I
(TJ
1
6-6
-------
g
•a
s
I
I
o>
I
« o
-------
r
CO
CO
.£ E
> m
11
a
•M
S
s
V
§£
3-8
« o
^ u
o a
•2 s
0H ^
H
H vi
CtH <»
o g
«> .2
S ^
•-S d
05 •••
^
<6
1
-------
saieunjsa joAa/uns am uem ssa-| jo 'o; jenbg '
jaieejo SGM a;eiui;s3 ajaqM saifiipej ^o juaojaj
6-9
-------
s8}eui{)S3 joAe/uns SIB ueqi ssei jo 'oj |enb^ '
jejeejg SBM ejEUiijsg ajeqM saijnpBj jo ;ueojad
6-10
-------
in
S
^ fa
a s
co «S
^ i
U C
w &
-g ea
II
a
« O
«"g
aj a
fa es
ta «
, ce
o>
a
saieuiijsg joAa/uns aqt ueqj. ssai jo *
jajeajo SBM a)euij)S3 ajai]M sanjijoej jo tuaoiad
6-11
-------
Summary of On-Site Release and Other Waste Management Estimates in all Three
SIC Codes
|
Table 6-2 summarizes the differences in on-site release and other waste management
quantities for all three SIC Codes. The high percent differences for on-site disposal to land are
due to errors from a few facilities, as explained above. The overall magnitude of these errors is
small, and does not contribute significantly to the total release and other waste management
quantities.
Table 6-2
Comparison of Differences Between Facility Estimates and
Site Surveyor Estimates Across the SIC Codes
Medium
Fugitive air
Stack air
Receiving stream
Land on site
Total
Percent Difference for
SIC Code 33
-7.75%
-1.63%
0.54%
432%
0.22%
Percent Difference for
SIC Code 36
22.9%
-39.4%
1.52%
53.1%
-24.5%
Percent Difference for
SIC Code 37
-9.96%
-2.78%
0.0%
212%
-4.26%
Note: Percent Difference - (Fa - SSy(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
6-12
-------
6.1.2 Comparison of the Facilities Surveyed to the National TRI Database
Estimates made by surveyed facilities were compared to national estimates in the TRI
database to determine how closely the release and other waste management quantities reported
by the surveyed population matched the national population. To make this comparison, the
estimates of the surveyed facilities were scaled-up to place them on the .same basis as the national
estimates. The scale-up factor used is the ratio of the number of Form Rs reported by the
surveyed facilities to the number of Form Rs submitted to the TRI database by all the facilities in
the SIC Codes visited. Only release and other waste management quantities from facilities with
15 or fewer Form Rs were included in the national estimates, since this was also a selection
criteria when identifying facilities to visit (see discussion in Section 2). Tables 6-3a through 6-
3c show this comparison and the percent differences for SIC Codes 33, 36, and 37, respectively.
A comparison of transfers to underground injection has not been made since none of the facilities
surveyed used underground injection systems. The percent difference is calculated as:
Percent Difference = (Fa - TRI)/(TRI) x 100
where: Fa = Scaled Facility Estimate
TRI = Total Releases and Transfers Reported in TRI database
The TRI database values have been used as the basis for comparison as these data are being used
nationwide.
6-13
-------
Table 6-3a
Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 33
Type"
Fugitive Air
Stack Air
Receiving Stream
On Site Disposal
Total
Scaled Release and Other
Waste Management
Quantities Reported by
Surveyed Facilities
(million pounds)
7.22
50.6
0.294
1.89
60
Total Reported
Nationwide (SIC Codes
331,332,333,334,335)
(million pounds)
39.9
108
32.3
215
395
Percent Difference"
Facility: TRI
-81.9%
-53.2%
-99.1%
-99.1%
-84.8%
>Io underground injection was reported.
Tcrccnt difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TRI=Total Release Estimate as Reported in the TRI
database.
Overall, there are substantial differences between the scaled-up release and other
waste management quantities from the surveyed facilities compared to national estimates in the
TRI database. We believe this finding relates to the voluntary nature of the program. Anecdotal
'.:'", • ' -I
evidence suggests that smaller companies and facilities have been more willing to participate in
the survey program, and larger companies and facilities have a greater tendency to decline to
participate. Further examination of the TRI database records of facilities that declined to
participate in the site survey program for RY 1996 indicates they had approximately five times
the amount of release and other waste management quantities per Form R than those that
volunteered to participate in the survey. Thus, the difference in the release and other waste
management estimates between the surveyed facilities and the national estimates reflect that
surveyed facilities tended to have lower throughput, on average, than the industry as a whole.
6-14
-------
Table 6-3b
Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 36
Type'
Fugitive air
Stack air
Receiving stream
On Site disposal
Total
Scaled Release and Other
Waste Management
Quantities Reported by
Surveyed Facilities
(million pounds)
1.73
2.90
0.041
0.087
4.76
Total Reported
Nationwide (SIC Codes
367 and 369)
(million pounds)
3.22
7.91
1.39
0.299
12.8
Percent Difference11
Facility: TRI
-46.2%
-63.4%
-97.1%
-70.9%
-62.8%
>Jo underground injection was reported.
"Percent difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TRI=Total Release Estimate as Reported in the TRI
database.
Table 6-3c
Comparison of Scaled On-Site Release and Other Waste Management
Quantities as Reported in Section 5 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 37
Type*
Fugitive air
Stack air
Receiving stream
On site disposal
Total
Scaled Release and Otber
Waste Management
Quantities Reported by
Surveyed Facilities
(million pounds)
17.3
69.6
0.0
0.051
87.0
Total Reported
Nationwide (SIC Codes
371 and 372)
(million pounds)
16.2
66.6
0.275
0.69
83.8
Percent Difference6
Facility: TRI
6.55%
4.53%
r-100%
-92.6%
3.82%
^No underground injection was reported.
T>ercent difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TRI=Total Release Estimate as Reported in the TRI
database.
6-15
-------
6.2 Estimates of Off-Site Release and Other Waste Management Quantities as
Reported in Section 6 of the Form R
Section 6.2.1 compares the estimates of off-site release and other waste management
quantities between the facilities and surveyors. Section 6.2.2 compares the scaled up facility
estimates to the TRI database.
-
6.2.1 Comparison of the Facility Estimates to the Surveyor Estimates
To assess the accuracy of the estimates reported by the facilities, the facility
1
estimates for each medium were compared to those calculated by the site surveyors. This was
done in the same manner that the on-site releases and waste management quantities were
tabulated.
!
Tables 6-4a through 6-4c show the percent difference between the facility estimates
and the site surveyor estimates for each off-site release and transfer medium in Section 6 of the
Form R for SIC Codes 33, 36, and 37. The percent difference is calculated as:
Percent Difference = (Fa - SS)/(SS) x 100
where: Fa = Facility Estimate
SS = Site Surveyor Estimate
The site surveyor estimates were used as the basis for comparison as they are a more accurate
representation of "true value" than the facility estimates. Negative percent difference values
indicate that, overall, the facilities surveyed underestimated the release or other waste
managemenf activity, while positive values indicate an overestimate. These differences are
depicted graphically in Figures 6-la through 6-lc. Note that none of the surveyed facilities had
transfers to underground injection.
Figures 6-2a through 6-2c illustrate the percentage of facility estimates which were
greater than, equal to within 5%, or less than the surveyor estimates for SIC Codes 33, 36, and
37, respectively. The number of facilities with each type is given in parentheses below each
release and other waste management activity type.
6-16
-------
Release and Other Waste Management Estimates in the Primary Metals Industry, STC
Code 33
For the primary metals industry, SIC Code 33, 27 facilities were surveyed and
estimates for 74 EPCRA Section 313 chemicals were reviewed. Comparing the facility estimates
to the site surveyor estimates (Table 6-4a), the largest discrepancies in total pounds are for
transfers off site disposal (-91.4%), transfers off site for treatment (73.4%), and discharges to
POTWs (50%). The total percent difference for all off-site release and other waste management
activities in this SIC Code is -37.3%.
In the case of off-site disposal, estimates were reported by 12 facilities from all five of
the three-digit SIC Code 33 facilities surveyed; nine of these facilities underestimated the
quantity sent off site for further waste management. Two facilities account for much of the
difference. One incorrectly thought copper qualified for the article exemption; this facility
disposed of large amounts of copper at an off-site landfill, negating the exemption. The second
miscalculated the percent of a metal sent to an off-site landfill. Because the total amount of
waste landfilled was high in both cases, the amount underestimated was large. An additional
seven should have reported greater release quantities to this type, which results in a large
discrepancy between the totals for the facility and surveyor estimates.
In the case of chemicals sent off site for treatment, one facility greatly overestimated
its transfers for two reasons: the facility incorrectly assumed the metals in waste transferred off
site were treated instead of being disposed, and they miscalculated the amount of nitric acid
produced, thus overestimating the quantity treated. For discharges to POTWs, one facility
accounts for the discrepancy by reporting releases (Code A, 1-10 pounds) where they should not
have. The chemical was zinc which in this case would not be released as a dust or fume to the
POTW. Note that although this contributes a 50% error in the quantity sent to POTWs, the size
of the error is orders of magnitude smaller than that of other types.
Figure 6-la shows the impact of the estimating differences on the amounts of
chemicals released and otherwise managed as waste. The errors in the estimates of off-site
transfers are the most significant. The 4 million pound difference hi off-site recycling estimates
between the facilities and surveyors accounts for 83% of the difference hi the total off-site
6-17
-------
release and other waste management quantities in SIC Code 33. The
to one facility overlooking 4 million pounds of metal in slag that was sent to
for further use in road maintenance.
As presented in Figure 6-2a, the majority of the air releases arid
difference can be attributed
an off-site recycler
off-site transfers for
disposal were underestimated. Transfers for off-site treatment and off-site energy recovery
tended to be overestimated.
Table 6-4a
Summary of SIC Code 33 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form Ra
Type"
POTW
Off-site disposal
Off-site treatment
Off-site energy recovery
Off-site recycling
Total
Release and Other Waste
Management Quantities
Reported by the Facilities
(million pounds)
1.65E-05
0.086
0.261
0.030
7.77
8.15
Release and Other Waste
Management Quantities
Estimated by the
Surveyors ;
(million pounds)
1.10E-05
1.005
0.150
0.022
11.8
13.0
Percent Difference"
50%
-91.4%
73.4%
37.5%
-34.1%
-37.3%
"Number of facilities - 27; number of EPCRA Section 313 reports represented = 74.
Tfo underground Injection was reported.
"iPcTccnt Difference - (Fa - SSy(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
6-18
-------
Table 6-4b
Summary of SIC Code 36 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form Ra
Type"
POTW
Off-site disposal
Off-site treatment
Off-site energy recovery
Off-site recycling
Total
Release and Other Waste
Management Quantities
Reported by the Facilities
(million pounds)
0.243
0.121
0.020
0.211
1.00
1.60
Release and Other Waste
Management Quantities
Estimated by the
Surveyors
(million pounds)
0.246
0.182
0.026
0.215
1.46
2.13
Percent Difference*
-1.23%
-33.3%
-21.7%
-2.16%
-31.5%
-24.9%
Mumber of facilities = 14; number of EPCRA Section 313 reports represented = 48.
T4o underground injection was reported.
°Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
Table 6-4c
Summary of SIC Code 37 Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form Ra
' :::::^Type"-.\(: ':
POTW
Off-site disposal
Off-site treatment
Off-site energy recovery
Off-site recycling
Total
Release and Other Waste
Management Quantities
Reported by the Facilities
:• (million pounds) '
0.045
0.136
0.006
0.346
1.16
1.69
Release and Other Waste
Management Quantities
Estimated by the
• - " •.-Surveyors'." • V
(million pounds)
0.045
0.146
0.007
0.403
0.942
1.54
Percent Difference6
0%
-6.57%
-10.82%
-14.1%
22.9%
9.74%
lumber of facilities = 19; number of EPCRA Section 313 reports represented = 64.
T$O underground injection was reported.
"Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Note: Due to rounding, calculated values may not yield exact numbers.
6-19
-------
Release and Other Waste Management Estimates in the Electronic and Other
Electrical Equipment Industry, SIC Code 36
In the electronic and other electrical equipment industry, SIC Code 36, a total of 14
facilities were surveyed. Site surveyors reviewed and estimated release and other waste
management quantities for 48 EPCRA Section 313 reports. The total percent difference for all
! I,'!1 1
off-site release and other waste management activities in this SIC Code is -24.9% (see Table 6-
4b).
As with the primary metals industry, errors in the off-site transfers have the greatest
i
consequence on the total estimate of release and other waste management quantities (see Figure
6-lb). Two facilities underestimated off-site recycling by a significant amount. One facility
failed to report for copper (overlooking the chemical entirely), accounting for 25,000 pounds of
i
the difference. The other facility assumed all metals were directly reused when taken off site
when some were managed and then recycled. Similar to the primary metals industry, the
460,000 pound difference in off-site recycling between the facility and surveyors estimates
accounts for 87% of the difference in the total off-site release and other waste management
quantities in SIC Code 36.
" i
Release and Other Waste Management Estimates in the Transportation Equipment
Industry, STC Code 37
A total of 19 facilities were surveyed in the transportation equipment industry, SIC
Code 37, and 64 EPCRA Section 313 reports reviewed. Total estimates for each type by facility
are within 25% of the site surveyor estimates (see Table 6-4c). Of the three SIC Codes surveyed,
• . j
tills SIC Code shows the best overall agreement with a total percent difference of off-site releases
and waste management quantities of 9.7%. The most significant impact in terms of the error in
the overall amount of release and other waste management activities is again due to the
inaccuracies in estimating the off-site transfers to recycling (see Figure 6-lc). One facility
i i' f
significantly overestimated off-site recycling of metals due to the fact that most of this metal was
directly reused.
6-20
-------
Summary of Off-Site Release and Other Waste Management Estimates in all Three
STC Codes
Table 6-5 summarizes the differences in off-site release estimates and waste
management quantities for all three SIC Codes. Figure 6-3 presents the sum of the total (on-site
and off-site) release and waste management quantities for each SIC Code graphically.
Overall, off-site transfers to recycling, disposal, and treatment were the most
problematic to estimate for all the SIC Codes. Because these transfer types account for a large
portion of the total quantity of release and other waste management activities, improving on
these estimates would improve the accuracy of the total estimates reported.
The main reason for the difficulty facilities had estimating off-site transfers to
recycling relates to differentiating between "reuse" and "recycle". Facilities felt that definitions
for these were not clearly stated in the reporting instructions in general, nor do they address
specific concerns that are unique to these SIC Codes. In particular, these definitions were
perceived to be unclear as they apply to metals and metal compounds present in scrap, off-
specification product, dust, slag, and other spent process streams that are subsequently used by
other facilities.
Some facilities reported confusion based on the EPA issue paper, Clarification and
Guidance for the Metal Fabrication Industry. January 1990. This guidance was issued prior to
inclusion of Section 8 of the Form R and states that amounts sent off site for recycling should not
be reported, which is incorrect under current reporting requirements. The document also
references EPA's Toxic Chemical Release Inventory Questions and Answers: 1990 Update,
which also contains some outdated information. Efforts are currently underway to revise the
metal fabrication and electroplating guidance documents, and updated guidance should be
available within the next year.
6-21
-------
Table 6-5
Comparison of Differences Between Facility Estimates and
Site Surveyor Estimates Across the SIC Codes
Medium
POTW
Off-site disposal
Off-site treatment
Off-site energy recovery
Off-site recycling
Total
Percent Difference for
SIC Code 33
50.0%
-91.4%
73.4%
37.5%
-34.1%
-37.3%
Percent Difference for
SIC Code 36
-1.23%
-33.3%
-21.7%
-2.16%
-31.5%
-24.9%
Percent Difference for
SIC Code 37
0%
-6.57%
-10.8%
-14.1%
22.9%
9.74%
Note: Percent Difference » (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
6-22
-------
(•sqi ^0 suoiiMUi)
y uuoj ai|; jo 9 pue s suoi)oes u; payoday
tuauiaBeuen e;se/\A J3MIO
fl^ 4)
"S Z*
-S o
O Vfi
i
'-a
o
II
fl
OS
ea
i
s
O'
6D
3
| s
II
•d
a
6-23
-------
A secondary reason for inaccuracies in the estimates of off-site
facilities mistakenly assuming metals sent off site in waste that are treated rather
This assumption causes some treatment estimates to be overestimated, and
disposal to be underestimated. Another discrepancy is that facilities did not
of waste removed from the site and incorrectly assumed recycling. Unless
supporting documentation of these waste management practices, the site
these transfers to be sent for disposal. Again, disposal would be underestimated
transfers for recycling would be overestimated when this occurred.
transfers is due to
than disposed.
the corresponding
always know the fate
facilities had
considered
and off-site
surveyors <
to
As an indication of how industries surveyed this year compare
previous years, the overall percent differences are presented in Table 6-6.
differences account for all on-site and off-site release and other waste
This comparison suggests that industries in SIC Codes 33, 36, and 37 have
reporting. On closer evaluation, industries in SIC Code 33 account for much
difference, and SIC Code 36 to a lesser extent. Industries in SIC Code 37
comparable to the facilities in the SIC Codes from previous years.
(Th
management quantities.)
less accurate
of the 28%
at 3% difference are
Industry confusion over the definitions of recycling and reuse is
the 28% difference in facility and surveyor estimates in SIC Code 33. The
difference in estimates is due to the large amount of throughput, use, and
industry. In addition to providing guidance on recycling and reuse, industry
appears warranted.
6-24
those surveyed in
.ese percent
the main reason for
magnitude of the
of metals in this
-specific guidance
reuse
-------
Table 6-6
Percent Difference of Facility and Site Surveyor
Estimated Total Release and Other Waste Management Quantities as
Reported in Sections 5 and 6 of the Form R for Reporting Years 1996,1995,
1994,1988, and 1987
Reporting Year
1996
1995
1994
1988
1987
SIC Codes Surveyed
331, 332, 333, 334, 335, 367, 369, 371, 372
26, 286
25,281,285,30
28, 291, 34 through 38
20 through 39
Percent Difference
-28%
-1.2%
-6.7%
1.1%
-2.2%
6.2.2 Comparison of the Facilities Surveyed to the National TRI Database
Estimates made by surveyed facilities were compared to national estimates in the TRI
database to determine how closely the release and other waste management quantities reported
by the surveyed population matched the national population. To make this comparison, the
estimates of the surveyed facilities were scaled-up to place them on the same basis as the national
estimates. The scale-up factor used is the ratio of the number of Form Rs reported by the
surveyed facilities to the number of Form Rs submitted to the TRI database by all the facilities in
the SIC Codes visited. Only release and other waste management quantities from facilities with
15 or fewer Form Rs were included in the national estimates, since this was also a selection
criteria when identifying facilities to visit (see discussion in Section 2). Tables 6-7a through
6-7c show this comparison and the percent differences for SIC Codes 33, 36, and 37,
respectively.
Percent Difference = (Fa - TRI)/(TRI) x 100
where: Fa = Scaled Facility Estimate
TRI = Total Release and Other Waste Management Quantities
Reported in TRI database
The TRI database values have been used as the basis for comparison as these data are being used
nationwide.
6-25
-------
Table 6-7a
Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R by the Facilities Surveyed
to the TRI Database, SIC Code 33
Type'
POTW
Off-Site Transfers
Total
Scaled Release and Other
Waste Management
Quantities Reported by
Surveyed Facilities
(million pounds)
0.001
526
526
Total Reported
Nationwide (SIC Codes
331,332,333,334,335)
(million pounds)
5.49
1,067
1,072
Percent Difference11
Facility: TRI
-100%
-50.7%
-50.9%
s'o underground injection was reported.
TPcreent difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TRI=Total Release Estimate as Reported in the TRI
database.
Overall, there are substantial differences between the scaled-up release and other
waste management quantities from the surveyed facilities compared to national estimates in the
TRI database. We believe this finding relates to the voluntary nature of the program. Anecdotal
evidence suggests that smaller companies and facilities have been more willing to participate in
,j
the survey program, and larger companies and facilities have a greater tendency to decline to
1 " " , ;i
" "' , : !l
participate. Further examination of the TRI database records of facilities that declined to
: , . , , : j
participate in the site survey program for RY 1996 indicates they had approximately five times
the amount of release and other waste management quantities per Form R than those that
volunteered to participate in the survey. Thus, the difference in the release and other waste
• • ', . • ' ' !l
management estimates between the surveyed facilities and the national estimates reflect that
i| :
surveyed facilities tended to have lower throughput, on average, than the industry as a whole.
6-26
-------
Table 6-7b
Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 36
Type*
POTW
Off-site transfers
Total
Scaled Release and Other
Waste Management
Quantities Reported by
Surveyed Facilities
(million pounds)
15.7
87.3
103
Total Reported
Nationwide (SIC Codes
36Yand369)
(million pounds)
12.0
324
336
Percent Difference6
Facility: TRI
30.1%
-73.1%
-69.3%
Mo underground injection was reported.
""Percent difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TKKTotal Release Estimate as Reported in the TRI
database.
Table 6-7c
Comparison of Scaled Off-Site Release and Other Waste Management
Quantities as Reported in Section 6 of the Form R for the Facilities Surveyed
to the TRI Database, SIC Code 37
Type'
POTW
Off-site transfers
Total
Scaled Release and Other
Waste Management
., Quantities Reported by
Surveyed Facilities
(million pounds)
2.91
106
109
Total Reported
Nationwide (SIC Codes
371 and 372)
(million pounds)
6.99
193
200
; s
Percent Difference6 -
Facility: TRI
-58.4%
-45.1%,
-45.5%
Tercent difference = (Fa-TRI)/(TRI) x 100, where Fa=Scaled Facility Estimate and TRI=Total Release Estimate as Reported in the TRI
database.
6-27
-------
6.3 Estimates of On-Site Waste Management Quantities as Reported in Sections 7,
8.2, 8.4, and 8.6 of the Form R
ij
. i
In addition to evaluating the accuracy of the on-site releases and off-site transfers to
disposal, treatment, energy recovery, and recycling. Site surveyors also reviewed facility
estimates for EPCRA Section 313 chemicals in on-site waste management activities.
11
Comparisons of the percent differences are made between the facility and the surveyor estimates
for each of the three SIC Codes and each of the three waste management types: treatment, energy
recovery, and recycling.
I
To calculate the percent difference, the amounts of EPCRA Section 313 chemicals in
waste managed on site estimated by the facilities and the site surveyors were summed to
determine totals for each waste management type at each facility. Totals for each facility were
then summed for all facilities in each SIC Code. These totals are presented in Tables 6-8a
through 6-8c. The percent difference is calculated as:
1 ', i
percent difference = (Fa - SS)/(SS) x 100
where: Fa = facility estimate
SS = site surveyor estimate
The site surveyor estimates were used as the basis for comparison as they are a more accurate
', • I •
representation of "true value" than the facility estimates.
i
i
i
6.3.1 Comparison of the Facility Estimates to the Surveyor Estimates
In general, most facilities in SIC codes 33 and 36 expressed considerable confusion
over reporting for on-site waste management activities. This resulted in significant quantitative
errors on a facility basis. However, the total amount of EPCRA Section 313 chemicals managed
I
as waste on site is relatively small compared to quantities released and transferred off site for
further waste management. Therefore, these errors do not significantly affect the overall
i
estimates in the TRI database. Facilities in SIC code 37 also expressed confusion, but to a much
lesser degree.
6-28
-------
Primary Metals Industries, SIC Code 33
Of the 27 facilities surveyed, only one reported a quantity greater than zero for on-site
treatment (some correctly reported a quantity of zero). Site surveyors found this facility
significantly underestimated the quantity treated (destroyed) and that three other facilities
overlooked significant quantities. Additionally, as discussed in Section 4, several facilities failed
to report on-site treatment or removal of metals in Section 7A of the Form R due to confusion
regarding whether removal of metals from a process wastestream should be considered. Most of
these facilities also expressed confusion regarding the percent efficiency that should be entered in
these situations; destruction efficiency (0%) or removal efficiency (typically 99%). [The 1996
TRI instructions say that the waste treatment efficiency reported must represent physical removal
of the parent metal from the waste stream, p. 41. It should be noted, however, that Section 8.6 of
the Form R asks for the amount destroyed in on-site treatment. Therefore, the correct amount
for metals treatment in this Section 8.6 is zero.]
Only one facility reported a quantity sent to energy recovery. This was expected
because facilities in the primary metals industry rarely employ energy recovery systems for
EPCRA Section 313 chemicals, partly because most recycle or reuse activities relate to metals
and metal compounds, which do not have a heat content high enough to sustain combustion. In
this particular case, the site surveyor determined that the EPCRA Section 313 chemicals were
directly reused, not sent to energy recovery.
Four facilities reported on-site recycling, two of which account for the majority of the
total amount reported. One of these reported on-site recycling for a direct reuse activity and the
second considerably overestimated the quantity recycled. A fifth facility overlooked a recycling
activity entirely.
6-29
-------
Table 6-8a
On-Site Waste Management for SIC Code 33a
Type
On-site treatment
On-site energy recovery
On-site recycling
Total
Number of Facilities
with On-Site Waste
Management
4
0
4
Amount
Reported by the
Facility
(million pounds)
0.002
0.950
21.2
22.2
Amount
Estimated by the
Surveyor
(million pounds)
0.040
0
6.12
6.16
Percent
Differenceb
-94.8%
—
246%
260%
lumber of facilities « 27, number of EPCRA Section 313 reports represented = 74.
TPereent Difference — (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
Table 6-8b
On-Site Waste Management for SIC Code 36a
Type
On-site treatment
On-site energy recovery
On-site recycling
Total
Number of Facilities
with On-Site Waste
Management
7
2
2
Amount
Reported by the
Facility
(million pounds)
0.575
0.486
12.7
13.7
Amount
Estimated by the
Surveyor
(million pounds)
1.40
0.468
0.065
1.94
Percent
Difference"
-59%
3.81%
19,416%
610%
lumber of facilities - 14, number of EPCRA Section 313 reports represented = 48.
T>erecnt Difference - (Fa - SSy(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
I " ' : j
Table 6-8c
On-Site Waste Management for SIC Code 37a
Type6
On-site treatment
On-site recycling
Total
Number of Facilities
with On-Site Waste
Management
3
2
Amount
Reported by the
Facility
(million pounds)
0.296
0.016
0.311
Amount
Estimated by the
Surveyor
(million pounds)
0.302
0.016
0.318
Percent
Difference*
-2.13%
0%
-2.02%
"Number of facilities ~ 19, number of EPCRA Section 313 reports represented = 64.
wo on-site energy recovery was reported.
"Percent Difference = (Fa - SS)/(SS) x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
6-30
-------
Electronic and Other Electrical Equipment Industry, SIC Code 36
Fourteen facilities from this SIC Code were surveyed. Five of these reported a
quantity greater than zero for on-site treatment. Site surveyors determined that four of these
underestimated the quantity treated and identified two additional facilities that employed some
type of system that resulted in the destruction of an EPCRA Section 313 chemical. The primary
reason for the quantitative errors was not confusion about how to report. Rather, site surveyors
identified a number of chemicals that were overlooked entirely. Facilities in the electronic and
other electrical equipment industry expressed confusion regarding on-site treatment of metals,
similar to the confusion in the primary metals industry. However, the "treatment or removal" of
metals (dust removal systems in particular) was less prevalent. Therefore, the percent difference
is lower. These facilities were more likely to treat mineral acids (via neutralization) or organic
chemicals.
Two of the 14 facilities reported on-site energy recovery. Site surveyors did not
identify additional energy recovery operations and quantitative estimates were in close
agreement. On-site energy recovery is not an area of confusion for SIC code 36.
Three facilities reported on-site recycling activities. These activities resulted in
approximately 12.7 million pounds reported in the TRI database. The site surveyors concluded
that two of these facilities correctly reported (both the recycling activity and the quantity).
However, the third facility incorrectly reported over 12.6 million pounds for recycling that were
actually directly reused. As with SIC code 33, this indicates that clarification of the terms
"recycle" versus "reuse" will greatly increase the accuracy of the TRI database. It also shows
that due to extremely high throughputs, an error by one facility may significantly affect the total
estimates for the entire SIC code.
Transportation Equipment Industry, SIC Code 37
Results indicate that on-site waste management is rarely conducted in this industry.
Also, when employed, facility estimates generally agreed with surveyor estimates, indicating that
both the processes used and the chemicals managed cause less confusion than observed in SIC
Codes 33 and 36.
6-31
-------
Three of the 19 facilities surveyed reported a quantity greater than zero to on-site
treatment. Site surveyors agreed within two percent of the estimated value and identified an
additional facility that employed on-site treatment activities, but the quantity destroyed was less
than 0.5 pounds.
ij
On-site energy recovery was not reported or observed at these facilities.
I
j
" ; j
On-site recycling was reported, and observed at two facilities. Site surveyors agreed
with facility estimates in both cases.
6.4
Production Ratio/Activity Index
The production ratio/activity index is a chemical-specific measure of the changes in
business activity between subsequent reporting years. The production ratio/activity index can be
1 • '!
determined using the following methods:
• TCM - the ratio of the amount of the chemical manufactured in the current
reporting year to the previous reporting year;
i' :• j
• TCPV - the ratio of production volume in the current reporting year to the
previous reporting year;
• TCU - an activity index of the amount of the toxic chemical used in the current
reporting year to the previous reporting year;
'•' !|, • !|
• HR - an activity index of the amount of operating hours for an activity in the
current reporting year to the previous reporting year;
• WT - an activity index or production ratio based on a weighted average of data
from several processes; and
i
• OTH - any other estimation method.
The site surveyors reviewed the method used by each facility for each Form R, and
determined whether it was the most appropriate method to use based on the facility's available
data. Figure 6-4 and Table 6-9 present by SIC code the distribution of the use of each method as
reported by the facilities and by the site surveyors. The site surveyors recommended changes to
the reported method for 25 of 157 (16%) Form Rs. The predominant method used by the
6-32
-------
facilities and the surveyors for each SIC Code is TCPV followed by TCU. The RY 1996 result is
consistent with the results of the data quality surveys previously conducted by EPA.
Table 6-9
Estimation Method Used by Facilities and Surveyors to Calculate
Production Ratio/Activity Index
Method of
Estimate
TCM
TCPV
TCU
HR
WT
OTH
Percent of Form Rs Reviewed Using Each Method of Estimate
SIC Code 33
Facilities
0
59
22
0
0
19
Surveyors
0
78
17
0
2
3
SIC Code 36
Facilities
0
68
20
0
0
12
Surveyors
0
78
22
0
0
0
SIC Code 37
Facilities
0
79
7
7
0
7
Surveyors
0
83
7
10
0
0
TCM - the ratio of the amount of the chemical manufactured in the current reporting year to the previous reporting year.
TCPV - the ratio of production volume in the current reporting year to the previous reporting year.
TCU - an activity index of the amount of the toxic chemical used in the current reporting year to the previous reporting year.
HR - an activity index of the amount of operating hours for an activity in the current reporting year to the previous reporting year.
WT - an activity index or production ratio based on a weighted average of data from several processes.
OTH - any other estimation method.
Table 6-10 shows the frequency which the surveyors agreed with the facility's choice
of method. It also provides explanations for the Form Rs where the surveyors disagreed with the
facility's choice of method, and shows that the surveyors disagreed most often with the "other"
basis of estimate. For only eight Form Rs (from four facilities), the surveyors thought that a
defined method not used by the facility was more appropriate and/or accurate for determining the
production ratio/activity index from the data available at the facility.
6-33
-------
Ik
TCFV TO)
HR
TCM WT
OTH
SIC Code 33
I
~
fi
II
on.
100.0%
90.0%
80.0%
70.0%
60.0%
50.0% 4-
40.0% - -
30.0% - -
20.0% - -
10.0% - -
0.0%
TCFV TO) HR TCM
SIC Coda 36
WT OTH
TCFV TCU
HR TCM
SIC Code 37
WT
OTH
D Facilities
I Surveyors
Data for this figure can be found on Table 6-6.
Figure 6-4. Estimation Method Used by Facilities and Surveyors to Calculate PR/AI
6-34
-------
Table 6-10
Percent of Time Surveyor Agreed with Facility Basis of
Production Ratio Estimate
SIC Code
33
36
37
Facility Basis of
Estimate
TCPV
TCU
OTH
TCPV
TCU
OTH
TCPV
TCU
HR
OTH
Percent of Time
! Surveyor
Agreed with Basis
97
77
18
100
100
0
100
25
75
0
"•-..'•
Changes Made by Surveyor
1 TCPV changed to WT
3 TCU changed to TCPV
2 OTH changed to TCPV because the facilities did not
know how to calculate this value
5 OTH changed to TCPV because the facilities reported
chemical release ratio of 1996 to 1995
1 OTH changed to TCPV because the facility reported
the production ratio from 1995 to 1996
Not applicable
Not applicable
4 OTH changed to TCPV because the facilities reported
a ratio of projected sales for 1997 to 1996 sales
1 OTH changed to TCPV because the facility reported a
ratio of sales for 1996 to sales for 1995
Not applicable
3 TCU changed to HR
1 HR changed to TCPV
2 OTH changed to TCPV because the facilities reported
a ratio of sales for 1996 to sales for 1995
1 OTH changed to TCU because the facility did not
know how to calculate this value
1 OTH changed to TCPV because the facility did not
know how to calculate this value
TCM - the ratio of the amount of the chemical manufactured in the current reporting year to the previous reporting year.
TCPV - the ratio of production volume in the current reporting year to the previous reporting year.
TCU - an activity index of the amount of the toxic chemical used in the current reporting year to the previous reporting year.
HR - an activity index of the amount of operating hours for an activity in the current reporting year to the previous reporting year.
WT - an activity index or production ratio based on a weighted average of data from several processes.
OTH - any other estimation method.
6-35
-------
II
Most (14 of 19) of the "other" responses were actually errors on the part of the
'!
facility, or their lack of knowledge of what this ratio was supposed to represent. Three of the
"other" responses were based on the ratio of the facility's sales from 1996 to 1995. In each case
these facilities had data available to calculate an activity index (the TCPV method), which the
: . • . I
surveyors thought was more appropriate. Surveyors recommended changing all of the "other"
responses except two. These two Form Rs (from the same facility) also used the ratio of the
facility's sales from 1996 to 1995, but this ratio approximated the production volume for both
chemicals.
A common error was that facilities used an activity index that was not specific to the
processes involving the EPCRA 313 chemical. This was often seen when facilities based the
ratio on total revenue, sometimes even including revenue from foreign sources.
Because many of the reporting errors for the production ratio/activity index were due
to facilities not understanding the value or by calculating it using sales data, EPA can improve
the accuracy of these values by preparing TRI reporting instructions that more clearly explain the
'
ratio, including a numerical equation example, and by emphasizing that production-based data
measures are preferred over sales data when available.
6.5
Source Reduction Activities
The following discussion reviews how accurately facilities report source reduction
activities on Form Rs. Starting in RY 1991, EPA requested that facilities include on their Form
Rs information describing source reduction activities implemented to reduce the quantity of
EPCRA Section 313 chemicals in waste. This information provides the users of the data insight
i
into the types and frequency of source reduction activities by industrial facilities. To assess the
i
accuracy of source reduction entries in the TRI database, analyses in this section address the
i
following three questions:
• Are the source reduction activities that facilities indicate on Form Rs legitimate?
!, i
• Why do facilities make errors when claiming source reduction?
' . i
• Do all facilities report source reduction activities on Form Rs?
6-36
-------
Section 6.5.1 describes the source reduction activities reported for SIC Codes 33,36,
and 37. Section 6.5.2 presents the errors found by the surveyors and answers the preceding
questions. Section 6.5.3 summarizes the findings and recommends ways to improve the accuracy
of source reduction activity reporting. It should be noted that this section focuses only on source
reduction activities that facilities indicate on Form Rs. Form As do not contain source reduction
information.
6.5.1 Source Reduction Reporting
Table 6-11 summarizes how often source reduction activities were reported for
EPCRA Section 313 chemicals. The data indicate that more source reduction was reported in the
electronic and other electrical equipment industry (SIC Code 36) than the transportation
equipment industry (SIC Code 37) or the primary metals industry (SIC Code 33). Compared to
similar data from previous years, the RY 1996 results for the electronic and other electrical
component industry was the largest value of reported source reduction on Form Rs surveyed
(39%) of any SIC code previously included in an EPA data quality survey. The results for source
reduction reporting in SIC Codes 33 and 37 (13% and 28%, respectively) were similar to the
values reported for other SIC codes hi other years.
Table 6-11
Frequency With Which Facilities Claimed Source Reduction Activities
SIC Code
33
36
37
Number of
Facilities
6
8
4
Percent of
FacOities
Visited
22
57
21
Percent of Form Rs
Submitted"
13
39
28
Total Number of Source
Reduction Activities
Claimed by Facilities
8
26
22
Table 6-12 shows the source reduction activities most commonly reported for each
SIC code. A variety of responses were received for each SIC code, although raw material
substitutions and process modifications account for most of the source reduction activities
reported for each SIC code.
6-37
-------
6.5.2 Errors Made When Claiming Source Reduction
i
I
i
i • •
To identify errors commonly made by facilities and reasons why facilities made these
: '' .•':.; I ' -.
errors, site surveyors determined whether facilities indicated source reduction activities that were
consistent with definitions of source reduction presented in the EPCRA Section 313 reporting
instructions. In cases where facilities did not claim source reduction activities, site surveyors
attempted to determine whether facilities overlooked source reduction activities. The rate of
>!"!'„ •:, , ' I
occurrence of errors in reporting source reduction activities and of not reporting source reduction
••• . . . . -'. ' ' • ' I ; •"
activities is shown in Table 6-13. Only a few errors were identified in reporting source reduction
activities. The percentage of errors found was lower than those found in previous EPA data
quality surveys. This EPA data quality survey is the first that attempted to identify overlooked
source reduction activities. Surveyors did find several overlooked source reduction activities.
Site surveyors disagreed with the source reduction activities reported for only five
Form Rs (from two facilities). Three Form Rs, from one SIC Code 369 facility, reported a
"change in operating practice" (Code W19) for three metal compound categories because they
started re-melting scrap metal. This process is not source reduction because the facility is still
i
processing the same amounts of metals, but now are just receiving some of them from a different
source. Another Form R, from an SIC Code 371 facility, reported "other changes in inventory
control" (Code W29) for dichloromethane because they improved their drum reconditioning
i
activities. This same facility also reported "other spill and leak protection" (Code W39) because
they added a vapor collection system above a process area to capture methanol fumes. This
facility may be reducing the amount of dichloromethane and methanol in their waste, but not
because of source reduction activities. The errors made by both of these facilities resulted from
their not understanding exactly what activities constitute source reduction.
6-38
-------
Table 6-12
Source Reduction Activities Claimed by the Surveyed Facilities
SIC
Code
33
36
37
Source
Reduction
Code
W14
W52
W13
W41
W55
W82
W58
W42
W52
W19
W13
W41
W73
W78
W82
W42
W72
W73
W74
W19
W49
W52
W58
Description
Changed production schedule to minimize equipment changeovers
Modified manufacturing equipment and layout
Added a recordkeeping system for chemical additions to a bath
Increased purity of raw materials
Changed from small volume containers to bulk containers
Changed composition of raw materials
Other process modifications
Substituted raw materials
Modified equipment, layout, or piping
Other changes in operating practices
Improved maintenance scheduling, recordkeeping, or procedures
Increased purity of raw materials
Substituted coating materials used
Other surface preparation and finishing modifications
Modified design or composition of product
Reduced concentrations of reportable chemicals in raw materials
Modified spray systems
Substituted coating materials used
Improved application technique
Use fewer storage tanks and transfers for fewer emissions
Raw material modifications
Modified equipment layout or piping
Other process modifications
Percent of Form Rs
That Correctly
Used This Code
25.0
25.0
12.5
12.5
12.5
12.5
26.1
21.6
21.6
8.7
4.4
4.4
4.4
4.4
4.4
35.0
15.0
15.0
15.0
5.0
5.0
5.0
5.0
6-39
-------
Table 6-13
j
Errors in Source Reduction Activity Classifications
SIC
Code
33
36
37
Number of Errors
in Source
Reduction Activity
Claims
0
3
2
Estimated Percent of
Source Reduction
Activities that are
Claimed In error*
0
12
9
Approximate
Number of Source
Reduction Activities
not Reported by the
Selected Facilities
7
0
4
Estimated Percent of
Source Reduction
Activities that are not
Reported"
11
0
7
Perccnts in this column were calculated using the weighting factors discussed in Section 2.6.
The source reduction activities that the surveyors identified as not having been
reported are summarized as follows:
• One SIC Code 33 facility had an ongoing effort to increase aluminum yield, and
could have reported administrative source reduction activities (Code W13) for two
chemicals;
i
ii
i
• Two SIC Code 33 facilities modified their equipment (Code W52) and made other
changes (Code W58), affecting several chemicals at each site;
i
• One SIC Code 37 facility could have reported source reduction for one chemical
through use of a higher purity raw material (Code W41); and
• Three SIC Code 37 facilities replaced or closed a particular process and stopped
using a particular chemical altogether, or eliminated a waste stream to a particular
media (Codes W52, W58, or W61).
i
6.5.3 Overall Accuracy of Source Reduction Data
Site surveyors found that some facilities in the selected industries misinterpreted
definitions of source reduction and should not have claimed all the source reduction activities
that they did for RY 1996. Observations made by site surveyors suggest that some facilities did
i j
not claim legitimate source reduction activities on their Form Rs, but the current site survey data
are insufficient for evaluating how often it occurs. The source reduction data in the TRI database
6-40
-------
may not be completely accurate, however it does indicate that pollution prevention efforts are
being considered by industries, and that the reporting of these activities is increasing.
Because the reporting errors of source reduction activities seem to be due to facilities
misinterpreting definitions, EPA can help improve the accuracy of source reduction data by
preparing TRI reporting instructions and guidance manuals that clearly define which activities
are and are not considered to be source reduction.
6-41
-------
-------
7.0
PREPARATION OF THE FORM R
Site surveyors interviewed facility personnel during each site visit to obtain general
information regarding completion of the EPCRA Section 313 reports (Form Rs or Form As) and
to identify trends among the surveyed facilities. The information obtained during these
interviews included quantitative information such as facility size (the number of employees at the
facility), estimated time to complete the Form Rs, the types of personnel primarily responsible
for preparing the Form Rs, and the types of references used by these personnel. In addition, the
surveyors obtained qualitative feedback on the Form R Instructions, the Automated Form R
(APR), the TRI Hotline, use of the Form A, and suggestions for additional guidance that EPA
should develop to assist facilities in estimating release and other waste management quantities
and in preparing thePorm Rs. Each of these topics is discussed in a subsection as follows:
• Section 7.1- Facility Personnel and References;
• Section 7.2 - Amount of Time Needed to Prepare Form R Reports;
• Section 7.3 - Use of the Hotline;
• Section 7.4 - Comments on the Form R Instructions;
• Section 7.5 - Comments on the Automated Form R; and
• Section 7.6 - Comments on Use of the Form A.
7.1
Facility Personnel and References
Table 7-1 identifies the percentage of surveyed facilities by size (based on number of
employees) for each SIC Code. The table indicates that most of the primary metals facilities
(SIC Code 33) had fewer than 500 employees (an average of 190), while the electronic and other
electrical equipment (SIC Code 36) and transportation equipment (SIC Code 37) facilities were
evenly split between facilities with 50 to 499 employees and facilities with more than 500
employees.
7-1
-------
Table 7-1
I
Number of Employees at Visited Facilities
Employee
Range
10-49 employees
50-499 employees
>500 employees
Number of Sites
Average Employees
Percentage of Visited Facilities with a Given Number of Employees
SIC Code 33
22
74
4
27
190
SIC Code 36
0
50
50
14
690
SIC Code 37
11
47
42
19
710
JiLl
Each facility was asked to identify the type of personnel responsible for completing
the Form Rs from among the following choices:
• Facility Environmental - A full-time, on-site employee whose primary
responsibility is dealing with environmental issues.
;:' ' : -is " . ' I
:.i, •,. . > I
• Corporate Environmental - A person with environmentally-related responsibilities
for more than one individual facility and may or may not be physically located at
the visited facility.
• Facility Staff- An on-site employee whose responsibilities extend beyond the
environmental area.
• Consultant/Contractor - Personnel contracted outside the company to prepare the
facility's Form Rs.
• Safety Personnel - Similar employee to Facility Environmental but includes safety
issues. This person may have responsibilities dealing with Environmental Health
and Safety issues.
• Other - Anyone who completed the Form R that does not belong to one of the
previously described staff types.
Table 7-2 lists the types of personnel responsible for preparing the Form Rs for each
SIC Code. Both facility staff and facility environmental staff were common responses for SIC
Code 33, while facility environmental staff (alone) was the most common response for SIC
Codes 36 and 37. Consultants/contractors and facility staff were typically reported by smaller
7-2
-------
facilities (based on number of employees) across each SIC Code, while facility environmental
staff was typically reported by larger facilities.
Table 7-2
Types of Personnel Completing the Form Rs
Staff Type
Facility environmental
Corporate environmental
Facility staff
Consultant/contractor
Safety personnel
Other
Percentage of Facilities Using Each Staff Type to Prepare The Form Rs"
SIC Code 33
33
19
37
15
4
7
SIC Code 36
57
7
14
29
14
0
SIC Code 37
68
16
0
16
5
5
Totals may equal more than 100 percent due to facility personnel identifying themselves as more than one staff type,
Table 7-3 identifies the reference materials facilities most commonly used to prepare
their Form Rs. All but two of the facilities visited used the TRI Reporting Form R instructions
for RY 1996 and MSDSs to prepare their Form Rs. Many of the facility contacts had attended
EPA-sponsored training workshops. The number of respondents that attended EPA-sponsored
training workshops was much greater than reported in EPA's previous surveys. Facility use of
other references is similar among the three SIC Codes. One difference between the SIC Codes is
that SIC Code 36 commonly reported using EPA's Estimating Releases and Waste Treatment
Efficiencies for TRI (Green Book) and Industry Trade Association Materials, while SIC Code 37
commonly reported the use of EPA's Compilation of Air Pollutant Emission Factors Document
(AP-42), use of privately sponsored seminar materials, and other resources.
7-3
-------
Table 7-3
•I • • i
Common References Used to Complete the Form Rs
Reference
TRJ reporting Form R instructions
Material safety data sheets
Estimating Releases and Waste Treatment
Efficiencies for TRI (Green Book)
EPCRA Section 313 Release Reporting
Guidance, Estimating Chemical Releases
Compilation of Air Pollution Emission
Factors, AP-42
Industry trade association materials
Privately sponsored seminar materials
EPA-sponsored training workshop
Computer programs
Other references
Percentage of Facilities Using a Particular Reference*
SIC Code 33
93
93
22
7
19
19
7
33
15
22
SIC Code 36
100
100
29
7
14
29
7
29
21
7
SIC Code 37
100
100
16
16
42
16
37
53
37
32
"otals may equal more than 100% as facilities often used more than one reference.
1" !
Eight facilities reported the use of other references that were not specifically
identified in Table 7-3. These sources included supplier data, manufacturer's control
efficiencies, hazardous waste manifests, non-AP-42 emission factors, Occupational Safety and
Health Administration (OSHA) air sampling data, internal guidance from a corporate office,
corporate seminars, EPA's TRI Question and Answer Document, and various software packages.
Some of the software packages included MSDS tracking software, purchasing tracking software,
chemical tracking software, internal spreadsheets, and named software packages such as OSHA-
.I:''.' ' , I
Soft, Wixel, HMMIS System, Reg Master, and Environmental Management Information System.
7-4
-------
7.2
Amount Of Time Needed To Prepare Form Rs
Table 7-4 and Figure 7-1 show the number of hours required to collect the necessary
data and complete all the Form Rs as reported by the facilities surveyed in RY 1996. Surveyors
requested that the facility select one of five time period ranges (shown in Table 7-4). The time
range reported by each facility was then divided by the total number of Form Rs filed by each
facility to estimate the time required per Form R as shown in Table 7-5.
The estimates in Table 7-5 were calculated using the midpoint of each time range.
Facilities that reported spending more than 100 hours provided an actual hour estimate which
was used in the calculations. Therefore, these values can be compared to one another to identify
differences between the SIC Codes, but are not necessarily an accurate estimate of the time
required to complete each Form R. For example, if the upper end of each time range had been
used to perform these calculations rather than the midpoint, the average time per Form R (across
all facilities) is 24 hours rather than 18, as shown in Table 7-5. Table 7-5 discusses one
particular data outlier: one SIC Code 371 facility reported a time of 200 hours per Form R which
was more than double the next highest value in the database. Several estimates, as noted on
Table 7-5, include and exclude this value. This SIC Code 371 facility reported this value
because they believe they need to have someone key-enter MSDS information on a weekly basis
to track their usage of EPCRA Section 313 chemicals.
For SIC Codes 33 and 36, Table 7-5 shows is that it takes more time per Form R for a
facility to complete one Form R, compared to multiple Form Rs. These calculations were
repeated (data not shown) and the same conclusions apply to facilities that complete "one or
two" Form Rs, compared to more than two Form Rs. While this conclusion is reasonable, it does
not appear to apply to SIC Code 37. Of the four SIC Code 37 facilities that filed only one Form
R, all made one particular product. The remaining SIC Code 37 facilities made one or more
products or were assembly lines handling hundreds of parts. The time each facility reported for
completing Form R reports was similar. Thus, simple facilities making a limited number of
products and those facilities making the same product(s) year after year may realize the same
time efficiency in EPCRA Section 313 reporting as facilities submitting multiple Form Rs.
7-5
-------
Table 7-4
i
Number of Hours Required to Complete All Form Rs at Surveyed Facilities
Time Range Estimate
s8 hours
9-20 hours
21-40 hours
41-100 hours
>100 hours
Unknown
Percentage of Facilities
SIC Code 33
26
18
30
18
4
4
SIC Code 36
14
36
21
29
0
0
SIC Codes 37
26
21
11
26
16
0
Table 7-5
:• • i i . •
Average Number of Hours Needed to Complete Each Form R
SIC Code
33, 36, 37
33
36
37
33, 36, 37
33, 36, 37
33, 36, 37
33
33
36
36
37
37
37
Data Subset
All data
All data
All data
All data
Facilities filing only 1 Form R
Facilities riling more than 1 Form R (with 1 outlier)
Facilities filing more than 1 Form R (without 1 outlier)
Facilities filing only 1 Form R
Facilities filing more Than 1 Form R
Facilities filing only 1 Form R
Facilities filing more than 1 Form R
Facilities filing only 1 Form R
Facilities filing more than 1 Form R (with 1 outlier)
Facilities filing more than 1 Form R (without 1 outlier)
Time Estimate
(Hours)
18
16
16
22
22
16
11
26
9
23
10
9
25
12
7-6
-------
I
03
TS
a
A
I
O
U
u
'3
s
s
0>
u
3
1! "8
jz;
i
H
go
a
7-7
-------
As indicated above, the average time needed per Form R is a function of the number
of hours in the time range checked, and whether the midpoint or maximum of the range is used.
However, the various time estimates listed in Table 7-5 represent an average range, and are
i
significantly lower than the estimated average burden of 43 hours per Form R as listed in the
,|
Toxic Chemical Release Inventory Reporting Form R and Instructions for RY 1996.
7.3
Use of the Hotline
Of 60 facilities visited, 57% reported calling the EPCRA hotline for assistance in
completing the Form Rs (although not necessarily about RY 1996). Figure 7-2 shows the
percentage of visited facilities that called the hotline for the SIC Codes included in this analysis.
Figure 7-2 indicates that personnel at about half of the primary metals facilities (SIC Code 33)
called the hotline, and a greater percentage of the electronic and other electrical equipment (SIC
Code 36) and transportation equipment facilities (SIC Code 37) called the hotline.
Most of the respondents in RY 1996 (85%) indicated that the hotline was helpful.
However, two facilities stated that they had difficulty getting through to speak to an operator, and
two other facilities stated that they received different answers from different operators. These
I
same two complaints about the hotline were received during the analyses that EPA conducted for
RY 1994 and RY 1995, in about the same frequency. One other facility commented that the
hotline response was riot helpful in RY 1995, but was helpful in RY 1996.
7-8
-------
-------
7.4
Comments on the Form R Instructions and Guidance Manuals
Surveyors asked facility personnel for feedback on the Form R instructions, and for
,i
requests they may have for any additional guidance materials. Table 7-6 shows the number of
respondents who identified a particular portion of the Form R instructions as unclear, for each
SIC Code. The comments received were similar for each SIC Code.
Table 7-6
Comments on the Form R Chemical Specific Instructions
Subject Area
Toxic chemical identity
Mixture component identity
Activities and uses of the toxic chemical
Releases to the environment on site
Transfers in waste to off-site locations
On-sitc waste treatment methods and
efficiency and on-site energy recovery and
recycling methods
Source reduction and recycling activities
Number of Res]
Sub
SIC Code 33
0
0
2
2
2
1
0
ipndents Stating That a Particular
ect Area Was Unclear
SIC Code 36
0
0
1
0
0
0
1
SIC Code 37
0
1
3
1
1
3
1
General comments received about unclear portions of the Form
summarized below, with the number of facilities making each comment shown
The list includes comments received from facilities in each SIC Code, and
be SIC-code specific. After the general comments are three separate lists,
Code, of additional guidance materials that have been requested or comments
instructions that are SIC-code specific.
R. instructions are
in parentheses.
were not interpreted to
for each SIC
on the Form R
one
7-10
-------
Unclear Areas of the Form R Instructions
• Definitions of manufacturing, processing, and otherwise use (4 facilities).
• The de minimis exemption (2 facilities).
• The article exemption (1 facility).
• The exemption for vehicles used on site (1 facility).
• The definitions of recycle versus reuse (5 facilities).
• The definition of metals versus metal compounds, e.g., nickel versus a nickel
alloy (3 facilities).
• How to report metals entering POTWs (2 facilities).
• How to determine a facility's latitude and longitude (1 facility).
• The definition of an aerosol (1 facility).
STC Code 33 - Primary Metals (RY 1996)
• Two facilities requested a guidance document that is industry specific (one was a
steel manufacturer and the other was a foundry), and a third facility simply
requested that more examples be presented.
• One facility suggested that EPA opinions be published as part of the guidance.
• One facility requested more information on which chemicals are in the glycol
ethers category.
SIC Code 36 - Electronic and Other Electrical Equipment (RY 1996)
• One facility requested that EPA suggest methods (e.g., spreadsheet formats) for
tracking EPCRA Section 313 chemicals used on site.
• One facility requested more guidance on estimating stormwater releases.
• One facility requested more guidance on estimating releases of acid aerosols (e.g.,
for hydrochloric acid).
• One facility requested that EPA release its instructions and guidance manuals in a
more timely manner.
7-11
-------
SIC Code 37 - Transportation Equipment (RY 1996)
I
• One facility suggested that the instructions contain less technical jargon; another
facility requested "anything" to make the process simpler.
!j
• One facility suggested that the instructions contain a list of references. Another
facility requested that EPA explain where to find the answers to frequently asked
questions. Another facility asked for a list of web sites (e.g., both EPA and
environmental group web sites) that provide TRI information.
• One facility requested that more examples be included over a broader range of
operations.
, , ij
• One facility requested additional pollution prevention guidance.
I
• One facility requested a consistent method for estimating metal releases in
stormwater.
ii , .
• One facility requested that welding emission factors be published, with
apportionments between different media categorized.
• One facility requested more guidance on emissions calculations.
• One facility requested, for EPA computer programs, that EPA write more easily
understandable explanations of the program assumptions. A specific example was
cited, the WIND program for estimating stockpile fugitive emissions.
7.5
Comments on the Automated Form R (AFR)
Sixty-three percent of the facilities surveyed for RY 1996 used the AFR to prepare
their Form Rs. Two-thirds of the facilities that used the AFR stated that it was helpful, while
one-third stated that it was not helpful. This information is shown in Figure 7-3.
•i
The types of feedback received on the AFR are provided below. Every comment
shown below was provided by more than one facility. Some of the comments, while described
separately below (the way they were reported), may actually be different ways of describing the
same problem.
7-12
-------
t
-------
Positive AFR Comments
:i
I
• Several facilities said the AFR is easy to use and saves typing time from year to
year. Common data (such as facility name) are maintained from year to year and
only the release information needs to be updated.
• Two commenters stated that the validation queries help reduce reporting errors,
particularly in Section 8.
'.; • , I ,
• Several commenters stated that the AFR software seems to be getting better with
each release version.
Negative AFR Comments and AFR Suggestions
• Six facilities reported receiving diskettes for RY 1996 that were not readable, and
therefore were not useable. One facility said they did not receive the AFR early
enough in the year to use it; and that the AFR should be sent out earlier in the
year.
' '': , jj
• Several facilities said they had previously used a Windows version of the
software, but they were mailed a DOS version this year. Most facilities did not
warit the DOS version and called the hotline to obtain a Windows version instead.
• Two facilities reported having problems installing the Windows version of the
software for RY 1996; one of these facilities said it had worked better for RY
1997. One facility reported problems installing a Windows NT version of the
software. One facility said that no information was available indicating that the
AFR could not be loaded onto a computer on a network.
•; ;' ' i
• Two facilities said that the Windows version of the software did not let them carry
over data from the previous year (the opposite of the first positive statement made
about the AFR).
I
• Five facilities reported that the software contains bugs but they did not elaborate.
Many of the other bullets in this section probably explain some of these bugs.
One facility said that they had technical problems with Sections 3.0 and 6.2 of the
software. Another facility said that the AFR had switched some release totals
around between reporting categories.
I
ii
• Three facilities stated that the AFR would not let them report metals entering a
POTW as a release for disposal; the software wanted to classify it as treatment.
Two facilities said that the software would not let them enter NA in every place
where they believed that NA was an acceptable response.
1
•!
• • j
One facility had a hard time entering extra off-site transfers. The software said
releases to the environment and did not distinguish on site from off site releases.
7-14
-------
This site ended up double-counting these releases. Once submitted, the facility
could not access the database to correct the mistake.
One facility received a Notice of Significant Error because data that they put into
the AFR were not transferred onto the printed reports created by the AFR.
One facility suggested that more specific error analyses be included. This facility
had left out an EPA identification number for a hazardous waste disposal site, but
the error message they received simply said an error occurred in the previous
section.
One facility said that the AFR User Support Hotline voicemail is always full of
messages, and is unable to receive any more messages.
Several facilities had trouble printing their reports once they were finished, and
some could not print the reports at all. One facility stated that 1) each AFR page
was printed onto two separate pieces of paper and 2) some of the printed text was
garbled. One facility said that it could not print the AFR Users Guide.
One facility suggested that the printer selection section needs to be updated. New
printers come out every year and the AFR printer selection list is not keeping up.
One facility suggested that the program should have an easier way to exit, or
abort, the program.
7.6
Comments on Use of the Form A
Five facilities surveyed for RY 1996 filed at least one Form A report. The surveyors
agreed that three of these facilities had used the Form A correctly, but two of the facilities had
used the Form A incorrectly. At least eight other facilities could have filed at least one Form A
but did not. Three of these facilities did not know about the Form A. Five of these facilities
knew about Form As, but chose instead to file Form Rs because it was more convenient. These
facilities said that they had to do the same threshold, release, and other waste management
calculations regardless of the form they filed, and recording their estimates on the Form R, a
form they were already familiar with, would take less time by comparison than learning the
requirements of the new Form A.
7-15
-------
The Form A requirements have been discussed in the reporting instructions for the
••I
past two years. Some facilities have correctly learned these requirements and used the Form As.
"' ' ''
The Form A requirements are also being emphasized in the new industry-specific guidance
manuals that EPA is now preparing.
7-16
-------
8.0
RECOMMENDATIONS
This section presents several recommendations for the EPCRA Section 313 program
based on the results and conclusions of the RY 1996 data quality assessment based on site
surveys. Improvements in reporting guidance and in the reporting instructions, as well as
facilities' experience in completing Form R reports for the previous reporting years will continue
to improve the data quality in the TRI database. Recommendations for continued improvement
of the TRI database are listed in the following subsections.
8.1
Additional Guidance Concerning Form R Instruction and Documentation
General recommendations noted by site surveyors for all SIC Codes include a section
introducing and explaining the Question/Answer document and guidance documents currently
available from EPA in the front of the TRI instructions. Many facilities and trade associations
did not read the entire TRI instruction booklet and, therefore, were not aware these documents
existed. Many facilities also requested a TRI guidance document specific to their industry.
Facilities frequently requested a guidance document on foundry operations as well as one on
metal and metal compounds reporting. Site contacts responsible for filling out EPCRA Section
313 reports mentioned that a readily available list of TRI Internet sites would also be helpful.
Facilities across all SIC Codes visited expressed concerns about the Automated Form
R (APR). Three of the facilities reported that the original diskettes they submitted to EPA were
deemed "unreadable". Therefore, EPA issued a Notice of Technical Error. EPA and the facility
eventually resolved the issue and in each case the facility re-sent the appropriate forms. None of
these facilities received a confirmation letter from EPA. When site surveyors arrived on site with
copies of information extracted from the TRI database, discrepancies existed between the TRI
information and the information the facilities sent to EPA. This problem indicates a potential
systematic error when corrected AFRs are sent to EPA.
Specific comments from facilities in each of the SIC Codes visited are as follows:
8-1
-------
SIC Code 33 - Primary Metals Industry
• Better definitions are needed in order to distinguish between recycling and reuse.
• More examples and applicability guidelines are needed for the de minimis and
article exemptions.
' , • •, . i
1 i/ ....•'' ' . r
• Better definitions are needed of metals versus metal compounds, and EPA
guidance should be available on which to report if both the metal and metal
compounds exceed the threshold.
• Clearer instructions are needed in reporting metals in Section 7 and 8.6 of the
Form R (on-site treatment). One issue in question is whether an air pollution
control device removing dust containing the metal should be included as treatment
(and if so, what efficiency to report since the metal is not destroyed but is
removed from the gas stream). Specific examples should be given in this scenario
when the toxic chemical has a "dust or fume" qualifier and/or when the metal dust
collected is actually the product to be sold. Another issue related to on-site
treatment is whether to report "0" or "N/A" in Section 8 when an on-site treatment
unit is reported in Section 7.
'',''" . • " I
SIC Code 36 - Electronic and Other Electrical Equipment
Clear guidance is needed on whether HC1 and H2SO4 acid aerosols should be
reported as being treated (Section 7 and/or Section 8) if the chemical was simply
removed from an air stream and incorporated into an aqueous stream. Facilities
felt the current guidance in the reporting instructions was unclear.
Clear guidance is needed on whether a chemical that is destroyed when it is used
to treat other chemicals is considered to be treated itself. Facilities were unsure if,
or how, Sections 7 and 8.6 should be completed in this situation.
\ . ' " I
More information is requested on determining production ratio, specifically for
those EPCRA Section 313 Chemicals produced as by-products or where the
production ratio is determined by something other than the annual production
ratio of the final product.
8-2
-------
SIC Code 37 - Transportation Equipment
8.2
• Better documentation is requested when EPA delists a chemical, but includes it in
a chemical category. A few facilities were tracking specific diisocyanate or glycol
ether chemicals that were once on the EPCRA Section 313 list, but are now
included in the diisocyanates or glycol ethers category instead of being listed
separately. Facilities suggested leaving these chemicals in the EPCRA Section
313 with a note to include them in the chemical category instead.
• Better definitions are needed to distinguish between the minor category
classifications (manufacturing aid vs. processing aid).
Additional Guidance Concerning Threshold Determinations
Although the nature and extent of threshold determinations varies from one
industry to the next, some general lessons can be learned from the mistakes identified by the site
surveyors. Table 8-1 lists common errors made by facilities when determining thresholds and
offers several recommendations to avoid making such errors in the future. These
recommendations may also be useful to EPA when developing future releases of TRI reporting
instructions.
8.3
Additional Guidance Concerning Release Estimates
Table 8-2 lists common errors made by facilities in all SIC Codes surveyed when
estimating release and other quantities managed as waste, and offers several recommendations to
avoid making such errors in the future.
8-3
-------
Table 8-1
Recommendations for Avoiding Errors in Threshold Determinations
Error Observed in
Determining Thresholds
Recommendation for Avoiding Error in Future TRI
Reporting Years
Facility did not document results of
threshold determinations.
Reporting instructions should emphasize that
documentation requirements apply to both threshold
determinations and release and other waste
management estimates.
Facility assumed EPCRA Section
313 chemicals exceeded thresholds,
rather than calculating annual usages
and comparing these amounts to
reporting thresholds.
Facilities should be informed that assuming thresholds
are exceeded, rather than calculating annual usages for
EPCRA Section 313 chemicals, is a common source of
errors in TRI reporting. Reporting instructions should
encourage facilities not to assume thresholds are
exceeded, even for chemicals used in very large or very
small quantities.
Facility overlooked EPCRA Section
313 chemicals that were purchased in
mixtures.
Facilities should carefully review the most recent
MSDS for every mixture brought on site to identify all
EPCRA Section 313 chemicals used during a reporting
year.
Facility considered only raw
materials used for production and
overlooked chemicals used for other
purposes.
Facilities should take a systematic approach to identify
all chemicals and mixtures used in production and non-
production capacities, including catalysts, underground
injection well treatment chemicals, wastewater
treatment chemicals, and the like.
Facility environmental staff was
unaware that certain EPCRA Section
313 chemicals were used at the plant.
Facilities should implement measures, such as chemical
usage logs or hazardous chemical inventories, to ensure
that environmental staff are aware of all EPCRA
Section 313 chemicals used in industrial applications.
Facility did not account for EPA's
most recent threshold determination
guidance.
EPA should enhance outreach efforts to ensure that all
facilities are aware of revised reporting guidelines well
in advance of submission deadlines.
8-4
fc,l
-------
Table 8-2
Recommendations for Avoiding Errors in Identifying Release and
Other Waste Management Activity Types and Sources
Observed Error
Recommendation for Avoiding Error in Future
I TORI Reporting Years
Overlooked stack emissions from
storage tanks, or reporting these
emissions as fugitives.
Instructions should emphasize this potential release
source and briefly discuss the definition of loading,
working, and breathing losses from tanks (and the
methodology to calculate them).
Overlooked container residue
Instructioiis should emphasize that even a "RCRA
empty" drum is expected to contain a residual
(possibly up to two inches) and that it must be
considered for TRI reporting. Also, note that on-site
drum rinsing and disposal of the rinsate will result in
a discharge to water.
Overlooked acid aerosols
manufacturing
Instructions should indicate that if H2SO4 or HC1 is
used anywhere in the plant as an aerosol, regardless
of whether the process is enclosed or not, their usage
should be japplied to the threshold determination and
release and other waste management calculations.
Incorrectly reporting disposition for
off-site transfers
Instructions should emphasize that facilities should
attempt toidetermine the type of receiving facility that
is accepting the transfers and exactly how the material
sent is being managed (or directly reused) by the
receiving facility.
Questions on on-site recycling
Provide a definition of recycling and include
examples of streams that can be considered as being
recycled in Sections 7 and 8. An example would be
used metals or metal compounds. Specifically, what
waste management activity must be applied to a used
metal for ijt to be considered recycled versus reused.
Definitions of source reduction
Consider shortening the list of codes for source
reduction and providing definitions for each code.
Questions of on-site treatment of
waste stream containing metals
Provide clarification of on-site treatment definitions
pertainingjto waste streams containing metals.
Facilities completing Sections 7a and 8.6 of the Form
R for metals are confused as to when treatment refers
to collection versus actual destruction of a metal.
8-5
-------
Table 8-2 (Continued)
Recommendations for Avoiding Errors in Identifying Release and
Other Waste Management Activity Types and Sources
Observed Error
Recommendation for Avoiding Error in Future
TRI Reporting Years
Confusion on energy recovery
Few facilities marked metals as going to energy
recovery, either on-or off-site. The TRI Reporting
Instructions should explicitly state that metals do not
have a BTU value high enough for energy recovery,
and should be marked as going to disposal or
recycling.
Reporting releases as zero versus a
range code representing a small
amount
For air and water releases, standard guidance is
needed on when it is reasonable to claim NA or zero
versus "guessing a small amount" - range code A or
B. (e.g. metal processed in amounts up to 10 pounds a
year through furnaces, reactors, etc. Is it reasonable
to claim zero or NA?) Standard guidance is needed
for consistent release and other waste management
estimates.
Questions on Section 8 amounts.
Facilities would like a simple formula for releases in
each block of Section 8. (e.g., Section 8.1 = 5.1 + 5.2
+ 5.3 + 5.4 + 5.5 + 6.2 (disposal only)). This will cut
down on errors and double counting.
Clarification of the treatment
definitions in Sections 7 and 8 of the
Form R for organic and inorganic
chemicals.
The definitions in the two sections are currently
different, which can cause problems when reporting.
Confusion occurs when: 1) chemicals go through a
treatment system but are not destroyed. Facilities
need direct guidance to claim zero efficiency, and
then what to put in Section 8 (zero or NA); 2)
facilities may report the amount sent to treatment
versus the amount treated. Current guidance is
confusing because facilities are supposed to report the
amount sent to energy recovery and the amount sent
to recycling, but not the amount sent to treatment
(they should correctly report the amount treated
instead).
Clarification on how to calculate
production ratio for "processed" and
"otherwise used" chemicals.
Facilities often used sales receipts or quantities
released from year to year rather than an activity
index that relates directly to the chemicals used.
8-6
-------
9.0 REFERENCES
1) 1994 and 1995 Toxic Release Inventory Report, USEPA, March 1998
9-1
-------
Appendix A
-------
SURVEY INSTRUMENT
-------
Facility ID: |_|_|_|-L|_|_|-|_|_|J
1998
(TRI REPORTING YEAR 1996)
TRI DATA QUALITY
SURVEY INSTRUMENT
-------
Date of Visit:
Facility Name:,
Facility Address:
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
FACILITY FACT SHEET
through.
City:
State:
Zip Code:
Mailing Address (if different from street address):
Telephone:.
Facility Contact:
Site Surveyors:
Fax:
Pre-visit Telephone Contact:
Reviewers:
A-l
-------
Facility ID:
PRE-VISIT
TELEPHONE CONTACT
TECHNICAL REVIEW
\. How many Form R (reporting year 1996) chemical reports were submitted for this facility?
2. How many 313 chemicals were identified by this facility, but not reported, for reporting year 1996?
3. Did the facility submit any revised Form R chemical reports for reporting year 1996?
YES....HH NO...D (SkiptoQ.5)
4, List the chemicals which had revised chemical reports.
Did the facility submit any withdrawal requests to EPA for the reporting year 1996?
YES....D NO...D (SkiptoQ.7)
List the chemicals which had withdrawal requests.
,:: Approved Q
Approved LH
Approved HH
Approved LJ
How many full-time equivalent employees did the facility have in 1996?
NOTE: If there were less than 10 fall-time equivalent employees in 1996, do not v sit this facility.
Terminate discussion with facility at this time.
A-2
Denied CH
Denied U
Denied LJ
Denied LH
Milt , ,;.,!
-------
Facility ID: |_|_|_|-|_|_|_|-|_|_|
Briefly describe the industrial processes performed at this facility in 1996.
9.
10.
Has the facility's process operations significantly changed since 1996 (including equipment, chemicals,
feedstock, etc.)?
YES D NO....D (SkiptoQ.ll)
Briefly describe any process changes.
11. Has the facility implemented any new treatment, disposal, energy recovery, recycling or source reduction
activities since 1996?
YES D NO D (Skip to Q. 13)
12. Briefly describe any new treatment, disposal, energy recovery, recycling or source reduction activities.
LOGISTICS
13. Will the facility be operating under typical conditions at the time of the visit?
YES....D NO D
A-3
-------
14.
16.
Facility ID: I
..." . ii
1 !
What personal protective equipment will be needed to participate in a facility tour?
Hard Hat D
Safety Boots
Safety Glasses
Respirator
Other:
n
n
15. How long is a typical tour? If unknown, how many square feet does the facility occupy?
_
(Consider this information when planning the type and duration of tour that would be most useful).
Hotel recommendation:
17. Directions to facility:
18.
19.1
19.2
20.
Time to meet:
Is a confidentiality agreement required to be completed for this facility?
YES...CH NO....D (Skip to Q. 20)
Has a confidentiality agreement been completed?
YES...D NO....D
' ' ' I
Will the person who completed the Form R and all supporting materials be available during the site visit?
YES...D NO....D Alternate Contact:
A-4
-------
Facility ID: |_|_|_|-
21. Describe the type and quantity of supporting material available for the Form R calculations.
A-5
-------
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
SECTION 1.0
REPORT PREPARATION
i!
1.1 Who prepared the release estimates in the facility's Form R chemical reports? (Check all that apply)
Facility Environmental Staff C]
1
Corporate Environmental Staff CU
Facility Staff D
Consultant/Contractor Q
Safety Department Staff D
Other, specify
1.2 Check all EPA documents and other references used to estimate releases and control efficiencies.
NONE D
TRI Reporting Form R and Instructions, 1996 Version
(EPA 745-K-95-051) D
Estimating Releases and Waste Treatment Efficiencies for the TRI ("Green Book"]
EPA 560/4-88-002
Title III Section 313 Release Reporting Guidance
EPA/560-4-88-004 a through q, Estimating Chemical Releases D
Compilation of Air Pollution Emission Factors, AP-42
Industry Trade Association Materials/Seminars EH
Privately Sponsored Seminar Materials
EPA-Sponsored Training Workshops
MSDSs
A-6
-------
Computer Programs (list)_
Facility ID: |_|_|_|-|_|_|_|-]_|_|_|
D
Other
D
1.3
What is your estimate of .the time needed to fulfill the reporting requirements of Section 313 for 1996?
Please include familiarization with the regulation and reporting instructions, completion and internal review
of the reporting forms, and documentation of all information in your reports. (This is the total time for all
FormJRs.)
8 Hours
9 - 20 Hours
21-40 Hours
41 -100 Hours D
> 100 Hours D
If > 100 Hours, please fill in number of hours
1.4 Did you find the 1996 Form R reporting instructions useful?
YES D NO....D
1.5 Did you feel any section of the instructions provided with the Form R were unclear?
YES....D NO.....D (Go toQ. 1.6) NA....D (Skip to Q. 1.6)
1.5.1 Check the appropriate section below and briefly explain the difficulty encountered.
Facility Reporting Determination CH
Part I. Facility Identification Information Cj
A-7
-------
Facility ID: ___________
Part II. Chemical Specific Information (Circle number of all that apply)
D
1.6
1.6.1
1.7
1.8
1.9
1.10
1. Toxic Chemical Identity
2. Mixture Component Identity
3. Activities and Uses of the Toxic Chemical '
4. Maximum Amount On-Site
5. Releases to the Environment On-Site ,
6. Transfers in Waste to Off-Site Locations
, 7. On-Site Waste Treatment Methods and Efficiency and On-Site Energy
• Recovery and Recycling Methods
8. Source Reduction and Recycling Activities
Did you call the Emergency Planning and Community Right-to-Know Hotline?
YES....D NO....D (SkiptoQ.1.7)
Die! you find the operator's response helpful?
YES....D NO....D
If no, explain
a
Have you ever received any assistance from EPA Regional or headquarters staff to prepare the
Form R reports?
YES....D
NO....D
Has EPA or your state ever contacted you with questions about any of the reported estimates
(excluding computer generated notices)?
YES....L
NO....D
Has the facility received any Notices of Significant Error, Notices of Noncompliance, or Notices
of Technical Error from EPA or the state for any 1996 reports?
YES....I
NO....L
Does the facility use any computer software to track toxic chemicals brought on site, used, or
identified in MSDSs?
YES....D
If yes, identify:
NO....D
A-8
-------
1.11
1.11.1
1.11.2
Facility ID: i_|_j_|-|_|_|_
Did you use the Automated Form R (AFR) electronic reporting to submit your Form Rs?
YES....D NO....D (Skip t<> Q.l.12)
Did you feel the AFR helped to reduce any errors on the Form R?
YES....D NO....D
Describe any comments on the use of the AFR.
1.12
Are there additional guidance manuals that EPA should develop to provide more clarification on
Form R reporting?
YES....D
NO....D
1.13
If a Form R was completed and the total annual reportable amount was less than 500 pounds, why
did the facility not complete the short form for the alternate threshold reporting?
A-9
-------
2.1
Facility ID:
SECTION 2.0
INTRODUCTION AND FACILITY TOUR
(313 CHEMICALS PRESENT ON-SITE)
List all chemicals reported on the facility's Form R Chemical Reports.
Chemical Name
CAS#
Not a Section 313 Chemical
n
n
n
n
n
n
n
n
n
n
n
n
n
A-10
-------
2.2
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
List all Section 313 chemicals not reported on the facility's Form R chemical reports, but
documented by the facility.
NONE....L
Chemical Name
Not a Section
313 Chemical
n
n
n
n .
n
2.3
List ALL other Section 313 chemicals not reported or documented, but identified by the surveyor
during the site visit.
NONE....D
Chemical Name
CAS#
A-ll
-------
Facility ID
2.4
List all mixtures identified during the facility tour which may contain Section 313 chemicals.
(a)
Mixture Name
"i
- ...
, ' !
:i . „
(b)
Identify Section
313 Chemical
Present
(c)
Concentration of
Chemical1
(d)
Amount of Mixture
Used in 19962
(e)
Amount of Section
313 Chemical
Used2
'if concentration of chemical is below de minimis (0.1 wt.% for carcinogens, 1.0 wt.% for all others), do not include
mixture in threshold determination.
2 Complete columns d and e during threshold determination.
A-12
-------
Facility ID:
Process Diagram(s):
(identify release points and chemicals)
A-13
-------
Facility ID |_|_|_|-L|_|_|-|_|_|_|
Treatment Unit, Disposal, Energy Recovery, Recycling or Source Reduction Operations):
(identify release points and chemicals)
A-14
-------
Facility ID:
Facility Tour Notes:
A-15
-------
Facility ID: "|_l-l-|-|_l-l_|-|-l-l_l
Facility Tour Notes (Cont'd):
A-16
-------
Chemical Name:"
CAS # |_| jj
Facility ID:
SECTION 3.0
REVIEW OF THRESHOLD DETERMINATION
3.1 How is this chemical employed at the facility? (Check all that apply)
Facility Reviewer
3.1.1, I—I LJ Manufacture
a. Produced at the facility
b. Imported by the facility
c. For on-site use/processing
d. For sale/distribution
e. By-product
f. Impurity1 (% = )
Facility Reviewer
n
n
n
n
n
n
n
n
3.1.2 n
n
Process (incorporative activity)
a. Chemical reactant (raw materials,
intermediates, etc.)
b. Formulation component
c. Article component
d. Repackaging
n
n
n
n
n
n
n
'If impurity is present below de minimis concentrations (0.1% for carcinogens, 1% for others), it is exempt from
reporting.
A-17
-------
Chemical Name:
Facility ID: '_!__- _U_-__U
Facility Reviewer
3.1.3 f~1 CD Otherwise Use (nonincorporative activity)
a. Chemical processing aid (added to reaction
mixture)
b. Manufacturing aid (process lubricants,
coolants, etc.)
3,1.4 D
3.2
3.2.1
c. Ancillary use (cleaners, degreasers, lubricants)
EH Exempt Uses
a. Used in laboratory activities
b. Structural component
c. Routine janitorial/facility grounds maintenance
d. Personal employee use
e. Motor vehicle maintenance
f. Intake water component
g. Contained in an article
Was the chemical reported by the facility?
YES....n (GotoQ.3.3) NO....CH
If no, why did the facility decide this chemical was not reportable?
Facility Reviewer
D D
n n
n n
n n
n n
n n
n
n
a
b
c
d.
Below threshold
Exempt
Overlooked chemical
Other (specify)
n
;i
n
.. n
i n
i
A-18
-------
Chemical Name:
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
3.3
3.4
3.4.1
3.5
Are all uses of the chemical exempt from reporting according to the surveyor or are all uses of the
chemical a non-aerosol form of sulfuric acid or hydrochloric acid?
YES....LJ (GotoQ.3.10) NO....LJ (Continue)
Does documentation which supports the threshold determination exist?
(Documentation is defined as any type of data available at the facility in any form which can
be used to recalculate the estimate)
YES....D NO....D
If no, why not?
a. Documentation cannot be located I I
b. Documentation was not retained by facility I I
c. Facility unaware that documentation is required I I
d. Facility overlooked the chemical (Skip to Q.3.6) I I
e. Other (specify) I I
What was the basis of estimate used by the facility for the amount manufactured, processed, or
otherwise used in 1996? Check all that apply.
a. Purchase/inventory records I I
b. Emission Factors I I
c. Mass balance CU
d. Assumed threshold exceeded (no calculations completed) I I
e. Process recipes/MSDS I I
f. Monitoring data CU
g. Production data f~l
h. Other (specify) CU
A-19
-------
Chemical Name:
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
3.6
3.7
3,8
How much chemical did the facility manufacture, process, or otherwise use in 1996?2
Facility
Reviewer
a. Manufactured
b. Processed
c. Otherwise used
.Ibs
.Ibs
Ibs
.Ibs
.Ibs
Ibs
d. Facility did not estimate these quantities LJ
Was the reviewer's estimate of the amount of chemical manufactured, processed, or otherwise used
recalculated using available documentation or recreated using other facility data?
j
a. Recalculated, with no error LJ
II
b. Recalculated, within a factor of 2 , LJ
i>' i
c. Recalculated, within a factor of 10 LJ
i
d. Recalculated, greater than a factor of 10 LJ
! '
e. Recreated, with no error LJ
f. Recreated, within a factor of 2 LH
g. Recreated, within a factor of 10 CD
h. Recreated, greater than a factor of 10 LJ
i. Facility did not estimate these quantities LJ
i
Was a threshold exceeded for this chemical in 1996?
YES....LJ (This chemical should have been reported. Continue)
j
NO LJ (This chemical should not have been reported. Skip to Q 3.10)
. II
3Record calculations and assumptions for the threshold determination on the worksheet in Section 6.0.
A-20
-------
Chemical Name:
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
3.10
3.11
This chemical was:
a. Correctly reported (Go to Section 4.0) D
b. Correctly not reported (Skip to next chemical) d
c. Incorrectly reported (Go to Q.3.11) U
d. Incorrectly not reported (Go to Q.3.12)U
Why was this chemical incorrectly reported?
a. Facility reported, although amount used was below threshold I—I
b. Facility incorrectly assumed threshold was exceeded I—1
c. Chemical activity was misclassified I—I
d. Threshold quantity was miscalculated I—I
e. Chemical was exempt I—'
f. Chemical has been delisted/modified I—I
g. Other (specify) . >—'
(Skip to next chemical)
A-21
-------
Chemical Name:
Facility ID:
3.12
Why was this chemical incorrectly not reported?
a. Chemical activity was overlooked
/
b. Chemical activity was misclassified
c.
d.
Threshold quantity was miscalculated
Other (specify)
3.13
I" (Continue to Section 4.0)
I
If the facility completed a short form for this chemical, are the releases less than 500 pounds?
. I
YES....!!] (Skip to the next chemical and document the release calculations)
i
NO D (Go to Section 4.0)
A-22
-------
Facility ID: |_|_|_|-|_|_|_|-|-l-l-l
SECTION 4.0
REVIEW OF RELEASE TYPES
A-23
-------
I
t
j:
H-
II
II
i
l
.a
1
te
O
P*
c
0 «
no .1
•11
!j.j
Ml
.||
I*
a
a
•.c
•s
b
r
i
i
i
i
i
•^J
U
1 '
O "^
2
0
3
S
0
•0
L>
-4
H
d
Pumps/valves/flanges
pq
rrt
Volatilization from process
U
Volatilization from treatme
Q
CO
Combustion by-products
^
I
!5
III
5«-a
P3
-------
T3
§
CO
JD
"s
CD
Releases are only fugitiv
en
CS
CD
S
"Q
,J "5
are not released to a stac'
Releases are to a stack ai
_o
CD
•g
C
b/l
as fugitives.
Chemical is a volatile or;
.is
C3
C3
T3
•£
O
§*
i
en
C3
O JD
1± 22
•a <3 £P « -S 8
CD K^ g O 2 -^ >-.
.E3 s > c rt *S 5
1 9 8 53* j| § 8
S ^ "- "S c « 2
g C3 C3 g 5 CD "-1
c S ^^-=3 -2 o>>
g 2 2^:^ u -0 go
CD en ent-iti -s >%cD
* 'S s 'S o '•? ^ § §
1 lil-l-'l-.is
^llilililii-3
ISlS|l|«|iiS
S £f£ c^tnpi §Di §03,
W>
• r-c
S
CD
(-^
c-3
CD
> a
c3 .S
X3 ^
system.
Toxic chemical does not
value high enough to sus
•H
§
o
3
r^
O
combustion.
Site visit concluded that
released to this medium.
Other
CD
CL,
CD
J8 — •
"CD o
IH .y
-o S
.CD g
Facility correctly identifi
or facility overlooked ch
*c3
W
O
fr
Sent to treatment, not dii
p ^
(— ( cd
'O (D
a
op Jj
13 O
^— ' c!
Overlooked release type
residue not considered)
Releases are to disposal
00
u
e
I
ffl
S
8
§
^ DC
en
a
-------
k
Y
§ ;
a
•J3
o c
0
^ .->
^
.
•
S'
-< i
i
t
11 ^
)
>
3
•>
:
;
J
3
5
i*
^
1
!»*
3 .
5 c
i
-i c
•4 t
' •
.
3
) t
i :
•4 /
a
1
3
7
3
2
3
3
;
3
J
^
2
o
J
3
3
3
V,
•i r
i
4
•4 K
C
C
•
c
e
- j
i
"* *
3
y
t
j
>
3
5-
3 •
3 «•
3
3
H
5 -
3
3
3
4
r f
H ^
1
r V
-1 ^
3
3
5
0
-4
4
3
5
5
)
3
o
•4
J
- 1
3
3
J
3 ;
3
•4
1
•*
-1
.2
e
es
8
a
03
O
bo
t3 O
rt bO t .
« I |
C ^n >,
•S w O
' w
« «
O
CO
O
• -
§!§§
a
II II
oo
y identified release
1
but facility incom
$
n
s
unclear or incomp
CA
e
but documentatiol
1
ii
•o
-2
^3
B
I
imentation cannot
3
8
Q
II
tained by facility
mentation required
SI
imentation was no
ity unaware that d
II
n n
•3
O
fli
ity overlooked ch<
1
ft
II
1
o
§
ft O & r
II II II
V
(
- f!
i
T
3
y
3 :
3 r
3
J
3
3
1-
3 ;
-4 r
j;
s ^
3 1
1 1
|j
3
?.-
-<
i,
•4 f
1
r >-
3
J
o
3
_t
2
|
3
3°
|
I 1%
\ II II
C—
a
ific
1
J
•a
£
!-H
u
•a
$
r3
^
ft
o
o
o<
CO
•a
o
S
u
43
O
4->
O
13
I
&
T3
J3
|M
15
3
CXi
<
&<
W
CO
O
C3
0
S2
0
I
>S
1
CO
CTJ
-------
(X,
Jl
H3
§ a
5 o
£ v •*
f**
» »0
04- ,
|i
" is
",£
i
°s
1. Did the facility identify a release type on the Form R?
2. Enter surveyor's release types.
|| 3. Did the facility correctly identify the release type?
>,
•S3
P
E
1
i
4. If Q.3 is NO, identify the reason that the release type v
identified, otherwise enter NA.
§?
•E-1
fe
co
3
1 5. Is documentation on the facility's release estimate avai
•' -.'-
*
0)
I
O
8
P.,
i
o
,|
ex
&
": '
-
\ 6. If monitoring data were used, is it available for review
I 7. Were chemical-specific EPA emission factors used?
„
(3
'«
«J
o
1 10. Was on-site treatment of this chemical included in re
-
a
a
•o
*3
c
1
i
1 1. Were treatment efficiencies reported consistent with
vendor specs, or EPA-published efficiencies? (2)
CA
&
0
*-»
CH
§
«
S
112. Does the facility have information available to estima
chemical released during 1996?
cp
V. 3
3 "
1:1
-------
%
sa
o
„ i4 'SB
if is
OH 2,°°°
.-§ &¥
£f g 03 WO
O nEj ^3
is
PL,
1 Chemical Name
i
i
V.
C
II 1. Did the facility identify a release type on the Form R?
II 2. Enter surveyor's release types.
|| 3. Did the facility correctly identify the release type?
13
<3
8
.3
»4. If Q.3 is NO, identify the reason that the release type was
identified, otherwise enter NA.
£
>
a
. j
-------
9
|
[2
-35
,g
"o
sr
-.
8
^^
S
>•
o
O
fe
fy an on-site treatment, enerj
Did the facility identi
*— i ^^
S
ification of on-site methods.
Enter surveyor's idenl
01
^"^-
• — •
o-
•a
O
rtly identify the on-site meth
Did the facility correc
CO
s
^
;z;
(_i
1
'S
1
0
•8
§
g
i
o
H
O
.S
CO
§
the reason that the method \
c
o
•a
1"
CO
1— 1
^
60
3
•0
>*
1
•3
o
.s
o
"8
.g
1
CO
•S
•*->
8
^
"c
o
1
to
"S
-3
©
od
005
For on-site treatment
itment? (4)
.
c
t
OJ
S
o
•o
recove
.s
.S
o
o
(3
3
cr
o
•S
t!
o
in §8.4B, did the facility rep
For on-site recycling
vo" S
-
_
1
£3
1
'•*
1
^
1
"a
g
1
terize the recycling stream (i
If appropriate, charac
r-^
- .~. '
~t ' s
i ,
•-
recycling unit: (6)
Describe the type of
06
method
on
itify th
II II
bo S
.S ^
'S o c5
•S. -5 e *-
>
"o
".
5 a >• C
? es "S o
.fSl I
S .H
c «3 « e
o D ea u
co 0 ^ co co"O
ycling
o
I
&
fr,
0
ton
0)
en
O
•s
V
ft
o
co r^
U e2 H
co co O
ft
S£
o
o
CO
•o
^
o
o
1
o
ssi
II II II
treatment
'co
o
II
1
o
t \
energy rei
'S3
O
II
T3
O
•3
£
jf!
P CO
£?
'to ~
^'§
O [in
II II
M
CO
C
O
e
i
i off-site landfill,
§
0
1
CO
(D
!
1
n
1
!
ake place in a
covery does not t
energy re
ifi 'cp
II
OS1
iting value high
recovery system
DCS not have a he;
&
§
s
d
•a
jg
tiemical di
Toxic cl
n
pa
o
combustion
not recycled
to sustain
tiemical is
method or facility
identified on-site
cal.
|l
f-iitl
O " i s O U
II II II
o\
cN
1
-------
Facility ID: |_|_|_|-|_|_|_|-|_|_|_|
SECTION 5.0
REVIEW OF RELEASE ESTIMATES
A-30
Oli;l!«[ ,,| Ill" , ',:, , ,„ ,1 illii
-------
•• '#'
' |i
ti
§ 8
•S S560"
fl -
D
r-
* « 5
S ^S *o*
$ CO
>w
•^E
co" "
••^"'
4) ^H
,gj-«>a
B
s ^
' . ,*; —
J!
1' """=-
2 ,:±
"3 *
"3 *
6 1
u
PQ
U
PQ
^^«
U
PQ
<
O
PQ
<
0
PQ
^~
£
1
1
1. Enter facility's release
^
1
O
1
S
s
CO
' use to e
S
1
o
2. What method(s) did th
1
1
1
£
1
CD
.2
•S
CO
^»
'o
_g ^
s i
O H
13. Based on data availabl
to determine a release esti
. '^'
;-•
•
'~
I
,co
:-l
s
.^M
'-EG
'.&
0
1
01
1
3
O
0
"8
3
JS
ch could
•f 5^
fg
14. What is a better metho
accurate release estimate1?
U
PQ
U
PQ
^
^«
U
PQ
<
U
PQ
<
U
PQ
1
o
43
H
1
03
•
CO
1
"a
1
*s
Release estimate was included b
K
II
f>
z
,0
121
1
*§
J
5
ave been, do not continue with
S
1
Cfl
1
•S
V
s
J
1
i
B
§
0
^,
p
1
o
CD
00
C
1
3
8
S
e
"3
O
1
"3
0
_g
's
o.g
^ ^
-o
1
*rt
°
•°
!
(U
J£
s
s
3
O
o
p
B
CXc
^*s CO
o"
O
ut enter facih'ty release (i.e, N3
j=
J3
§
§
cn
= Monitoring data or direct mea
2 3
o —
1 S!
*4-< b
s
1
o S «
= Mass balance calculations
\
= EPA chemical-specific emissi
= Engineering calculations ("mi
(this includes factors that are no
= Engineering judgement ("gue
= Hazardous waste manifests
- U Hr,
S U W O
^^^
G'
o
ffi
o
1
O
II
O
C/l
= Facility did not estimate releai
<^
Z
"c3
!
u
— i
II
CO
_Q
0\
^
i — i
i — i
II
CO
_2
i
cr\
>n
II
•4->
Release estimate was nc
3
3
0
J3
V3
included on Form R but
o
Tt P
a o.
0 J=
•a (-
I
i
o
B
o
•o
have been, skip to Ques
Facility overlooked this
skip to Question 4
Facility does not have a
release to this medium, i
u u
i— < r-J
<
PQ
U
&
Jz;
g
§
continue with this medii
II
<;
z
-------
!|
5 2§
SQ £r
Si a
oe
U
If 3
C/2 U
O
b
P1^
O
Is
PM
r
-1 —
1
i
z —
1 -5T
§ 00
6 d
U
pq
<£
U
PQ
<
U
PQ
^
U
PQ
U
pq
<
^-*
\^s
1?
.a
_e
^j*
estimate <
o
TO
"a
jX)
1. Enter faci
f?
c-
o
1
1
«
.3
to
use to e
£•
1
o
•5
12
CO
X3
1
as
£
ts
"8
1
^
3
CJ
he most ao
+3
'•3
CO
o ^
^O .§
*2 ^
data avai
a release
3. Based on
to determine
"&
$
P
en
B
a
o
£— <
GO
oo
S3
a
fo
»
OS
§
d to calculi
U
pq
<;
u
PQ
<
U
PQ
U
PQ
*'
•3
o
1
(D
ore accura
El
t»
'§
§
lease estii
-> 0
"f 00
•u a
a --1
5! 1 i
i 11
^_^
s
s
c
>]
"c
O
T:
•*-
—
-------
a
1
«j '»?
S SG
Q e<|
QO S
W30 g
||
*s o
^ o
^ aa
* S
^S
VI ^
fl\
_^ w
c -^
.2 33
^ i
£0
li
t"
§
83
O
1
uJS
u ?«
PS
&
G
§
O1
&:
la
» «
**" CC
tfi q>
S ^
o1'
"S
"3 M
tfS
.|*
; -;.;!,-
rhpmiral Niinie
^AftJl.l 1: 1. 1.1.1 l.l-l 1 : -.'.-..' -
^»^ :, l_l_l_l — .— I I— 1 — 1 1 — 1 , i
1. Enter facility's estimate from §8, Column B, on the
Pnrm R (Enter NA iffacilitv did not estimate)
^— .
»— t
*s».
%
•s
1 f
•I
I
- 1
i 1
1 *
< o
3. Calculate the quantity released or transferred using
flip mprhnH in fnntnntp. Ci).
4. Are the facility's estimate (Q. 1 ) and the quantity
rplpacpH nr transfe.rrp.d from O.3 the same? (3)
5. If Q.4 is NO, provide notes or an explanation
detailing any differences in the calculation of Section
8 Hata.
«
S
•s
£
•M
o
a
S o ns
>
-------
Facility ID:
Section 5.4 Review of Form R §8 Data
(On-Site Treatment, Energy Recovery, or Recycling)
For the on-site treatment or energy recovery method(s) identified in Section 4.2c (Question 2), complete the following table.
Only recreate on-site recycling estimates that were provided by the facility. Do not estimate on-site recycling releases NOT
identified by the facility.
Chemical Name
CAS#LL_ L-LJ-L
1. Enter the facility's estimate of quantity from §8, Column
B, of the Form R. (1)
2. What method did the facility use to estimate the amount
treated, sent to energy recovery, or recovered from
recycling? (2)
3, Based on data available to the facility, is this the most
accurate method to estimate the amount treated, sent to
energy recovery, or recovered from recycling? (3)
IF Q.3 IS YES, SKIP TO QUESTION 6
4. What is a better method which could be used to calculate
a more accurate; estimate? (2)
5, Enter the reviewer's estimate using a more accurate
method. (4)
6. Enter the reviewer's estimate using the same, method. (4)
On-Site~Treatment
(§7Aor8.6B)
On-Site Energy
Recovery
(§7Bor8.2B)
On-Site Recycling
(I)
(2)
Estimate was not included on Form R but should have been, skip to Question 4.
Facility overlooked this chemical, skip to Question 4.
Facility does not have this on-site method, do not continue with this medium.
* !l
Monitoring data or direct measurements
Mass balance calculations
EPA chemical-specific emission factors
Engineering calculations (this includes use of factors which are not chemical-specific)
Engineering judgement
Hazardous waste manifests
Other
(3)
(4)
Nl
N2
NA
M
C
E
OC
OJ
OH
O
NA = Facility did not estimate quantities for this on-site method
Document the method used by the facility and/or alternate methods used in Section 6.0
Y = Yes
N = No
NA = Facility does not have this on-site method
Document calculations in Section 6.0.
NA = Facility did not estimate release.
A-34
-------
Facility ID:
Section 5.5 Review of Form R §8 Data
(Production Ratio/Activity Index and Source Reduction Activities)
/,£ - ^irf.>?J-
Oiemidal Name" * - . * '<
Production Ratio/Activity Index
I. Enter the facility's estimate from §8.9 of
the Form R. (Enter NO if facility did not
estimate)
2. Enter facility's basis of estimate (1).
3a. Is mis estimate based on a variable that
most directly affects the quantities of the
toxic chemical generated as "waste"
quantities? (2)
3b. If Q.3a is NO, enter surveyor's choice
for alternate basis.
3c. If Q.3a is NO, enter surveyor's
production ratio estimate.
Source Reduction Activities! <,f •„', ~' *
4a. Enter the source reduction activity
codes from Section 8.10 of the Form R.
4b. Provide a text description of the source
reduction activity.
4c. Is this activity "source reduction" (i.e.,
not recycling, treatment, energy recovery,
or disposal) (2)7
^ . "^ >*' " ^ -^
/> ;5ii
3a.
3b.
3c.
^.,, iSitivj^Sj^-^IR
4a-l.
4b-l.
4c-l.
''^.%£j$i&
ion Ratio/^ctivi%^lL|i
fee Reduction AetJyiftfs
te-l.
4b-2.
4c-2.
W;^MiliW:;£^
4a-3.
4b-3,
4c-3.
Possible source reduction activities noted by the surveyor:
(1)
TCM
TCPV =
TCU
HR
WT
OTH
NA
Ratio of amount of the toxic chemical manufactured in 1996 to 1995
Ratio of production volume in 1996 to 1995
An activity index of the amount of toxic chemical used in 1996 to 1995
An activity index of the amount of operating hours for an activity in 1996 to 1995
An activity index or production ratio based on a weighted average of data from several processes
Other:
The manufacture or use of the chemical began in 1996.
(2)
Y
N
Yes
No
SECTION 6.0
CALCULATION WORKSHEETS
A-35
-------
Facility ID
THRESHOLD DETERMINATION WORKSHEET
Chemical Name:
Description of Use
TOTALS
Amount
Manufactured
Amount Processed
Amount Otherwise
Used
Calculations:
A-36
-------
Facility ID:
MAXIMUM AMOUNT ONSITE WORKSHEET
Chemical Namc:-
INSTRUCTIQNS;
Calculate the. maximum amount of the chi
reporting year. Keep in mind the following:
All storage a
The amount of eh
I at any time; and
The amount of chemical in each waste streai
Storage Areas:
Total:
Chemical in Use:
Total:
Chemical in Waste Streams:
Total:
Total On-Site:
A-37
-------
Facility ID:
RELEASE ESTIMATE WORKSHEET
Chemical Name:
CAS#
Release Type.
SI Page #
Question #
INSTRUCTIONS: Record all calculations for release estimates below, in the appropriate sections. Be sure
to identify if calculations use the same method as the facility or a more accurate method.
A-38
-------
Appendix B
-------
September 21, 1998
Mr,, Environmental Contact
Company Name
Address
City, ST 00000
Dear Mr. Environmental Contact:
j!
!
Your facility has been randomly selected by the U.S. Environmental Protection Agency
(EPA) to participate in a data quality survey of facilities that submitted 1996 reporting year
Form Rs for the Toxic Chemical Release Inventory (TRI) under Section 313 of the
Emergency Planning and Community Right-to-Know Act (EPCRA). The purpose of the
survey is to evaluate the quality of the data submitted on the Form R chemical reports and
to provide feedback to EPA that can be used to improve the reporting instructions,
guidance, or the reporting form. Eastern Research Group, Inc. (ERG) is working with
EPA to conduct survey visits to approximately 60 facilities randomly selected from the TRI
database in the following SIC Codes: 33, 36 and 37.
'ul ' ||
I
Participation in this program is voluntary. However, we believe your participation will
provide a valuable evaluation of the threshold calculations and release and transfer
estimates for your facility, which will assist you in future Form R reporting.
ERG technical staff members will visit your facility at a date between May and July 1998.
Most visits will last 1 day, depending on the number of chemical reports you submitted and
the number of Section 313 chemicals located on site. We will spend the majority of that
time reviewing the methodology and data that your facility used to make threshold
calculations and release and transfer estimates for the 1996 reporting year. During the site
visit, we will need your assistance to tour the plant in enough depth to allow us to
understand your manufacturing processes and to identify potential release points. We will
make every effort to minimize disruptions to your schedule while we are on site.
We will contact you by telephone within the next week to arrange a date for the site visit. I
would like to emphasize that the name and location of your facility will not be released to
EPA in order to ensure confidentiality of your facility's information. All data collected on
this project will be tabulated and reported to EPA in summary form only. At the project
conclusion, facility identification information will be destroyed.
CEB3W627-05.U
-------
Mr. Environmental Contact
September 21, 1998
Page 2
The EPA Project Manager for this program is Velu Senthil of the Office of Pollution
Prevention and Toxics (OPPT). Enclosed is a letter from him requesting your
participation.
Thank you for your cooperation. If you have any questions, please feel free to call me.
Sincerely,
Program Manager
Enclosure
CEB3\0627-03.ij
-------
•'It.
-------
|