ASBESTOS EXPOSURE ASSESSMENT
                 Prepared by

               ICF Incorporated
                Prepared for

                 Dr. Kin Wong
           Chemical Engineering Branch
      Office of Pesticides and Toxic Substances
       U.S. Environmental Protection Agency
             Washington, D.C. 20460
                March 21, 1988
              REVISED REPORT

-------
                              TABLE OF CONTENTS


                                                                          Page

LIST OF TABLES	     viii

EXECUTIVE SUMMARY ,	      xii

  I.   INTRODUCTION 	        1

      A.   Approach 	     •   i

          1.  Availability of Occupational Exposure Data 	:. .        3

             a.   Asbestos Exposure Survey 	        3
             b.   OSHA Compliance Inspections 	        8
             c.   NIOSH and Other Studies 	       10
             d.   Summary 	       11

          2.  Analysis of Occupational Exposure Data 	       13

             a.   Current Exposure Levels 	       13
             b.   Populations Exposed/Duration and Frequency of
                   Exposure 	       15
             c.   Proj ected Exposure Levels 	       16

      B.   Conversion Factor for Asbestos Measurement 	      '17

      C.   Report Format 	       21

 II.   OCCUPATIONAL EXPOSURE 	       23

      A.   Mining and Milling 	       24

          1.  Process Description 	       24

             a.   Conventional "Dry" Processing	       24
             b.   Unconventional "Wet"  Processing 	       26
             c.   Exposure Controls	       26

          2.  Manufacturers and Production 	       28

          3.  Current Exposures	       28

             a.   KCAC Incorporated 	       31
             b.   Calaveras Asbestos,  Ltd	       35
             c.   Vermont Asbestos Group 	       42

          4.  Summary 	       45
                                   - ii -

-------
                        TABLE OF CONTENTS
                           (Continued)
B.  Product Manufacture

    1.  Paper Products .
       a.  Product Descriptions 	
           (1)  Millboard 	
           (2)  Pipeline Wrap 	
           (3)  Beater-Add Gaskets 	
           (4)  High-Grade Electrical Paper
           (5)  Specialty Papers 	
       b.  Process Descriptions 	
           (1)  Primary Manufacture 	
           (2)  Secondary Manufacture 	
       c.  Production and Employment 	
       d.  Exposure Profile 	
    2.  Asbestos-Cement Pipe
       a.  Product Description
       b.  Process Description
       c.  Exposure Profile ...
    3.  Asbestos-Cement Sheet
       a.  Product Descriptions
       b.  Process Descriptions
       c.  Exposure Profile ....
    4.  Friction Products
       a.  Product Descriptions  	
           (1)  Drum Brake-Linings  	
           (2)  Disc Brake Pads  (Light and Medium)
           (3)  Disc Brake Pads  (Heavy)  	
           (4)  Brake Blocks 	
           (5)  Clutch Facings 	
           (6)  Automatic Transmission Components
           (7)  Friction Materials  	
       b.  Process Descriptions	
           (1)  Primary Manufacture 	
           (2)  Secondary Manufacture 	
       c.  Production and Employment 	
       d.  Exposure Profile 	
           (1)  Primary Manufacture 	
           (2)  Secondary Manufacture 	
                             - iii -

-------
                        TABLE OF CONTENTS
                           (Continued)
                                                                    Page

    5.   Textiles	      84

       a.   Product Descriptions 	      84
       b.   Process Descriptions 	      86
           (1)   Primary Manufacture 	      86
           (2)   Secondary Manufacture 	      88
       c.   Exposure Profile 	'..,.      88

    6.   Sheet Gaskets and Packing 	      91

       a.   Product Descriptions 	      91
       b.   Process Descriptions 	      92
           (1)   Primary Manufacture 	      92
           (2)   Secondary Manufacture 	      92
       c.   Exposure Profile 	      93
           (1)   Primary Manufacture 	      93
           (2)   Secondary Manufacture 	      95

    7.   Roof Coatings,  Non-Roofing Coatings, Missile Liner
         and Sealant Tape 	      95

       a.   Product Descriptions 	      95
       b.   Process Descriptions 	      98
       c.   Exposure Profile	      98

    8.   Asbestos-Reinforced Plastics 	     101

       a.   Product Description 	     101
       b.   Process Descriptions 	     102
           (1)   Primary Manufacture 	     102
           (2)   Secondary Manufacture 	     102
       c.   Exposure Profile 	     103

    9.   Miscellaneous Products 	 	     106

       a.   Production Data 	     106
       b.   Product Descriptions 	     107
           (1)   Filler Acetylene Cylinders 	     107
           (2)   Battery Separators 	     117
           (3)   Arc Chutes .	     109

C.  Chlorine Manufacture (Asbestos Diaphragm Cells) 	     109

    1.   Process Description 	     109

    2.   Manufacturers Using Asbestos Diaphragms 	     Ill
                              -  iv  -

-------
                              QF CONTENTS
                           (Continued)
    3.   Exposure Profile 	

    4.   Frequency and Duration of Exposure ..

D.  Brake Repair Service 	

    1.   Exposure Setting/Process Description
       a.   Compressed Air/Solvent Mist . .
       b.   Brush 	
       c.   Water Spray/Rag 	
       d.   Brake Washer 	
       e.   Vacuum Unit Without Enclosure
       f.   Vacuum Unit With Enclosure ...
    2.   Current Exposures 	

    3.   Populations Exposed 	

       a.  Duration of Exposure for One Brake Job  .
       b.  Full-Time Equivalent Populations 	
           (1)  Drum Brake Linings for Automobiles
           (2)  Disc Brake Linings for Automobiles
           (3)  Drum Brake Linings for Trucks 	
           (4)  Disc Brake Pads for Trucks 	
    4.   Frequency and Duration of Exposure ...

    5.   Summary 	

E.  Construction Industry Exposure 	

    1.   Exposure Settings and Operations 	

    2.   Pre-0.2 f/cc PEL Exposures 	

    3.   Projected Post-0.2 f/cc PEL Exposures

    4.   Populations Exposed	

    5.   Frequency and Duration of Exposure ..

    6.   Summary 	
                             .  - v -

-------
                                    OF CONTENTS
                                  (Continued)

                                                                          Page


 III.  AIR RELEASES  ."	     163

      A.  Milling and Primary Manufacturing Emissions  	     163

          1.  Methodology  	     163

             a.  Operating Schedule 	     165
             b.  Quantity of Asbestos Collected by the
                   Control Device  	     165
             c.  Collection Efficiency of the Control Device  	     166

          2.  Emission Estimates  	     169

      B.  Secondary Manufacturing Emissions 	     176

          1.  Methodology  	     176

          2.  Emission Estimates	     177

      C.  Mining and Trade Use Emissions 	     181

          1.  Methodology  	     185

          2.  Emission Estimates	     188

             a.  Mining 	     189
             b.  Brake Repair 	     191
             c.  Construction 	     200

      D.  Emissions from Asbestos-Containing Waste Piles  	     203

          1.  Regulations Affecting Management of Asbestos Wastes ....     203

             a.  Mining and Milling Wastes 	     203
             b.  Manufacturing and Fabricating Wastes 	     205
             c.  Installation Wastes 	     206
             d.  Demolition and Renovation Wastes	     207

          2.  Emission Estimates from Mining/Milling Waste Piles 	     208

             a.  Potential Emission Points During Waste
                   Handling Operations 	     209
             b.  Methodology 	     212
             c.  Emission Estimates 	     213

REFERENCES 	     217
                                    - vi  -

-------
                                    QF CONTENTS
                                 (Continued)
APPENDIX A.  OCCUPATIONAL EXPOSURE PROFILES AND AIR RELEASES FOR
            "PRODUCTS NO LONGER PRODUCED OR USED IN THE U.S	
            1.   Occupational Exposure ..

                a.  Product Manufacture
                    (1)  Paper Products 	
                    (2)  Corrugated Asbestos Cement Sheets
                    (3)  Vinyl-Asbestos Floor Tile 	
                b.   Construction Industry Exposure 	

                    (1)  Exposure Settings and Operations  	
                    (2)  Current Pre-0.2 f/cc PEL Exposures  ...
                    (3)  Projected Post-0.2 f/cc PEL Exposures
                    (4)  Population Exposed 	,
                    (5)  Frequency and Duration of Exposure  ..,
            2.  Air Releases
                a.  Primary and Secondary Manufacturing Sources
                b.  Construction Sources 	
                                   - vii -

-------
                                LIST OF TABLES
                                                                          Page


Table ES-1 -- Summary of Occupational Exposures and Air Releases 	      xv

Table 1    - - Summary of ICF Exposure Survey Results 	       4

Table 2    -- Availability of Occupational Exposure Data for
               Primary Processors 	•..       5

Table 3    -- Availability of Occupational Exposure Data for
               Secondary Processors 	       7

Table 4    - - Annual Domestic Production of Asbestos Fiber for 1985 ..      29

Table 5    -- Asbestos Exposure Profile for the Mining and Milling
               Operations at KCAC Incorporated 	      32

Table 6    -- Asbestos Exposure Profile for the Mining and Milling
               Operations at Calaveras Asbestos, Ltd	      37

Table 7    -- Asbestos Exposure Profile for the Mining and Milling
               Operations at the Vermont Asbestos Group 	      44

Table 8    -- Production and Employment for Primary Manufacture
               of Paper Products 	      54

Table 9    - - Exposure Profile for Paper Products 	      56

Table 10   -- Exposure Profile for A/C Pipe -- Primary Manufacture  ...      63

Table 11   - - Exposure Profile for A/C Sheet - - Primary Manufacture ..      68

Table 12   - - Production and Employment for Primary Manufacture of
               Friction Products 	      76

Table 13   -- Exposure Profile for Friction Products --
               Primary Manufacture 	      79

Table 14   -- Exposure Profile for Friction Products --
               Secondary Manufacture and Rebuilding 	      82

Table 15   - - Exposure Profile for Textiles 	      89

Table 16   •- Exposure Profile for Sheet Gaskets and Packings 	      94

Table 17   - - Production and Employment for Coating Type Products ....      97

Table 18   - - Exposure Profile for Coatings -- Primary Manufacture  ...      99

Table 19   -- Exposure Profile for Asbestos-Reinforced Plastics 	     104


                                   - viii  -

-------
                               LIST OF TABLES
                                 (Continued)
Table 20   - -  Production Data for Miscellaneous Products 	

Table 21   --  Producers of Chlorine Using Asbestos Diaphragms ....

Table 22   - -  Chlorine Production/Asbestos Fiber Consumption ......

Table 23   --  'Exposure Profile for Asbestos Diaphragm Cells 	

Table 24   --  Asbestos Exposure During Brake Servicing,
               by Control Method 	

Table 25   --  Summary of Asbestos Exposure by Control Method,
               With Calculated Means 	

Table 26   •-  Facilities Where Brake Repair is Performed
               (1984 Data) 	

Table 27   --  Area Asbestos Concentration During Brake Servicing .

Table 28   --  Automobile Drum Brake Shoe Repair by Facility Type
               and Estimated Full-Time Equivalent Workers Exposed
               to Asbestos 	

Table 29   -•  Automobile Disc Brake Pad Repair by Facility Type
               and Estimated Full-Time Equivalent Workers Exposed
               to Asbestos 	

Table 30   --  Truck Drum Brake Lining Repair by Facility Type
               and Estimated Full-Time Equivalent Workers Exposed
               to Asbestos 	

Table 31   --  Truck Disc Brake Pad Repair by Facility Type and
               Estimated Full-Time Equivalent Workers Exposed to
               Asbestos 	

Table 32   --  Pre-0.2 f/cc PEL Exposures to Asbestos Products in
               the Construction Industry	

Table 33   --  Projected Exposures to Asbestos Products in the
               Construction Industry 	

Table 34   - -  FTE Populations in the Construction Industry	

Table 35   --  Summary of Occupational Exposure to Asbestos in
               the Construction Industry 	

Table 36   --  Asbestos Emissions from Milling and Primary
               Manufacturing Sources 	
                                    -  ix  -

-------
                                LIST OF TABLES
                                 (Continued)
                                                                          Page
Table 37   -- 1981 Primary Manufacturer Asbestos Consumption
               and Waste Generation	     173

Table 38   -- Asbestos Emissions from Secondary Manufacturing
               Sources 	, .  .     178

Table 39   -- 1981 Secondary Manufacturer Asbestos Mixture
               Consumption and Waste Generation 	     179

Table 40   -- Asbestos Emissions from Mining Sources 	     190

Table 41   -- Distribution of Brake Repair Work by Types of
               Facilities 	     192

Table 42   - - Asbestos Emissions from Brake Repair 	     195

Table 43   -- Asbestos Emission Estimates in g/yr for Each Region --
               Installation of Drum Brake Shoes on Cars 	     196

Table 44   -- Asbestos Emission Estimates in g/yr for Each Region --
               Installation of Drum Brake Shoes on Trucks 	     197

Table 45   -- Asbestos Emission Estimates in g/yr for Each Region --
               Installation of Disc Brake Pads on Cars 	     198

Table 46   -- Asbestos Emission Estimates in g/yr for Each Region --
               Installation of Disc Brake Pads on Trucks 	     199

Table 47   -- Asbestos Emissions from Construction Activities 	     202

Table 48   - - Summary of Current Waste Generation from Milling
               Operations 	     210

Table 49   -- Parameters for Estimating Emissions Resulting from
               Wind Erosion	     214

Table 50   -- Emission Estimates from Milling Waste Piles :	     215

Table A-l  -- Exposure Profiles for Paper Products No Longer
               Manufactured in the U. S	     237

Table A-2  -• Exposure Profile for Corrugated Asbestos Cement
               Sheet - - Primary Manufacture 	     242

Table A-3  -- Exposure Profile for V/A Floor Tile -- Primary
               Manufacture 	     246
                                    - x -

-------
                               LIST OF TABLES
                                 (Continued)
Table A-4  --  Exposure to Asbestos Floor Products in the Construction
               Industry Under Pre-0.2 f/cc PEL 	
Table A-5  --  Exposure to Asbestos Floor Products in the Construction
               Industry Under Post-0.2 f/cc PEL 	
Table A-6  -- Asbestos Emissions from Primary Manufacturing Sources
               for Products No Longer Manufactured in the U.S	
Table A-7  -- Asbestos Emissions from Secondary Manufacturing Sources
               for Products No Longer Manufactured in the U.S	
Table A-8  -- Asbestos Emissions from Construction Activities Using
               Products No Longer Used in the U.S	
                                    -  xi -

-------
EXECUTIVE SUMMARY




    EPA is assessing the risks associated with the production and use of




asbestos and of products made from asbestos.   As background for EPA's January




29, 1986 proposal to ban and phase down asbestos production and usage,




occupational exposure levels and air releases for primary and secondary




processing of asbestos from 1981 8(a) submissions and exposure levels for




construction from the late 1970's and early 1980's were used.  Recently OSHA




has lowered its permissible exposure limit (PEL) for asbestos to 0.2 f/cc from




the previous 2 f/cc.  Because of OSHA's new PEL, additional controls and




respirators are likely to be used to reduce current exposure levels to the new




PEL or lower.  The purpose of this study is to update occupational exposure




data and air emissions,  used as inputs to determine ambient exposures, to




better represent current industry practices.




    Available occupational exposure and air emission data from NIOSH,




academic, and industry studies were supplemented by OSHA Compliance data and




the IGF Exposure Survey.  The ICF Exposure Survey, which covered both




occupational exposures and air releases,  was sent to all miners/millers of




asbestos, primary and secondary manufacturers of asbestos products, and




several relevant industry groups.  The overall positive response rate of this




voluntary survey was 14 percent (refer to Chapter I, Section A.I for a more




thorough discussion of the survey).  OSHA Compliance data were supplied to us




by the OSHA Office of Management Data Systems for the SICs corresponding to




manufacturing, construction, and automotive servicing.




    Due to the limited availability of exposure data on any one product, we




estimated exposure levels for product categories.  However, these product




category exposures are applied to the worker populations, and population




distributions, for each individual product, thus allowing for some




distribution of risk.  This analysis, however, assumes that job category




                                   - xii -

-------
exposures for all products in a product category are  identical.  The product

categories are:

           Paper products;
           Friction products;
           Asbestos cement products;
           Asbestos-reinforced plastics;
           Coatings;
           Packings and gaskets;
           Textiles; and
           Miscellaneous uses.

This analysis does not cover products no longer produced in the U.S. or

imported into the U.S. such as commercial paper, corrugated paper, rollboard,

flooring felt, roofing felt (imported only), corrugated A/C sheet  (imported

only), and vinyl asbestos floor tile.  Occupational exposure  levels and

population factors for products no longer produced or used in the  U.S. are

presented in Appendix A for use in sensitivity analysis.

    Current exposure levels associated with each job category or task are

based on historical data.  Both geometric and arithmetic means of  the raw data

are presented throughout the text of this report (and in Appendix  A).  The

geometric mean represents a typical exposure level for a worker performing a

particular job, assuming that the observations follow a log normal

distribution which is common for exposure data.  The arithmetic mean, which

represents the total worker exposure when multiplied by the exposed

population, is used in the health benefits model to assess the consequences of

exposure.

    Total 1985 worker populations for primary and secondary product

manufacturing for each product are calculated by summing up the populations

for each producer gathered during the ICF Market Survey.  These populations

are distributed into the various job categories identified by the  monitoring

results using the population distributions obtained from 1981 TSCA Section
                                   -  xiii  -

-------
8(a) submittals.  Exposed populations for mining and milling were obtained




through telephone contacts with the respective company representatives.




    Since installation, repair, and removal jobs are intermittent, populations




for brake repair and construction have been calculated as full-time




equivalents (FTEs).   The FTE population is the number of workers working 250




days/year and 8 hours/day at installing, repairing, and removing the total'




quantity of an asbestos product manufactured or imported each year (from the




ICF Market Survey).   Short-term exposures, which represent the exposure during




the period of time in which the actual task is performed, are applied to this




population.




    We used a simplified and conservative approach for projecting exposure




levels under the 0.2 f/cc PEL.  This approach assumes that for those




operations where 8-hour TWA exposures are currently below 0.2 f/cc, work




practices will remain unchanged.  However, for those operations where the




8-hour TWA exposures are currently above 0.2 f/cc, work practices will be




changed either with the addition of engineering controls or respirators to




reduce the exposures to 0.2 f/cc.




    Table ES-1 summarizes the occupational exposure results.  Throughout the




asbestos manufacturing industry, exposure levels are a function of specific




job and product type.  Activities with relatively high time-weighted average




exposures are fiber receiving and storage, fiber introduction and mixing, and




some finishing operations.  Product characteristics also affect exposure




levels.  In a product such as a coating, the asbestos is encapsulated thus




reducing its potential for release.




    Air releases are estimated for each mining/milling and product




manufacturing facility using site specific data and engineering estimates of




baghouse collection efficiencies.  Air releases from brake servicing and




construction are calculated as annual industry emissions due to the lack of




                                   - xiv -

-------
                                           Table ES-1.   Sumnary of Occupational Exposures and Air Releases
Occupational Exposure
Exposure Level
(f/cc)'
Industry Sector Geometric Mean
Mining
Milling
Milling Waste Piles
Primary Product Manufacture
Paper Products
A/C Pipe
A/C Sheet
Friction Product*
Textiles
Gaskets end Packings
Coatings
Reinforced Plastics
Secondary Product Manufacture
Paper Products
Friction Products8
Textile*
Gaskets and Packings
Reinforced Plastic*
Chlorine Manufacture (Asbestos Diaphragms)
Brake Repair
Construction
0.02*
0.05£
N/A

0.04
0.08
o.ie
0.11
0.18
0.03
0.03
0.06

0.02
0.01
0.1S
0.08
0.07
0.02
0.09h
0.10
Arithmetic Mea
0.03f
0.06*
N/A

O.OS
0.10
0.18
0.1S
0.18
0.08
0.11
0.06

0.02
0.05
0.16
0.11
0.09
0.04
0,15h
0.13
Air Releases
Population 1
n Exposed
44
111
N/A

299
286
23
2,603
78
168
1,449
138

1,877
7,045
208
903
456
650
114,234
2,032
Duration of Exposure Total Emissions Duration of Emissions
(hours/yeer)c (kg/yr)d (hours/year)
900-1.600
1,490-1,920
N/A

1,630
2,000
2,000
2,000
1,920
2,000
1,720
2,000

2,000
2,000
2,000
2,000
2,000
1,760
2,000
2,000
11 (14)d
4.495
9

957
552
79
3,536
152
303
80
17

1,405
18
8
9
2
0.08
17 (23 )d
318 (422)d
900-1,600
8,760
8,760

8,760
8,760
6,760
8,760
8,760
8,760
8,760
8,760

8,760
8,760
8.760
8,760
8,760
8,760
8,760
8.760
"Geometric and arithmetic means of available raw monitoring data projected under the 0.2 f/cc PEL.   Exposure levels are 8-hour TKAs except for brake
repair and construction for which short-term exposures are used.  Exposure levels are weighted arithmetically by the number of workers exposed to each
level in the Industry sector to determine the "average" values presented in this table.

Population figures are the summation of the number of exposed workers at each facility, except for brake repair and construction for which full-time
equivalent populations are presented.

CA default of 2,000 hours/year Is used when no other data are available.  (The 8-hour TWA has been calculated such that It accounts for short dally
exposures, thereby making the effective duration of exposure 8 hours/day.)

-------
                                                               Table ES-1  (Continued)


dTotal emissions for mining, brake repair, and construction are based on the occupational exposure levels associated with each task;  therefore,  two
values are presented -- one using the geometric mean exposure and the other using the arithmetic mean exposure.   The total emissions  using the
arithmetic mean exposure are presented in parentheses.

8The duration of emissions Is 8,760 hours/year In all cases,  except for mining,  because emissions are modeled as continuous releases.  The duration of
mining emissions Is the actual time spent on mining activities.

fAssumes use of respirators specified by the mining/milling companies.

8Includes both secondary processing and brake rebuilding.

Geometric and arithmetic mean exposure levels for brake service using various non-engineering and engineering controls were weighted arithmetically by
the estimated fraction of facilities using each control; the numbers given are the overall weighted geometric and arithmetic means, respectively.

4The population exposed during brake repair Includes not only the mechanics performing the brake Jobs, but also all other full-time mechanics
(full-time equivalents) in the brake servicing facilities.  All full-time mechanics are Included because monitoring data show similar levels for area
exposure* as were observed for breathing zone exposures for workers performing brake Jobs.

-------
site specific information in these ubiquitous industries.  Tables ES-1




summarizes the air release results.




    This report is divided into three major chapters:  Introduction,




Occupational Exposure and Air Releases.  Chapter 1 presents our approach for




gathering and analyzing occupational exposure and air release data, and our




review of conversion factors for asbestos measurement.  Chapter II presents




our estimates of occupational exposures during mining/milling, asbestos




product manufacture, chlorine manufacture using asbestos diaphragm cells,




brake repair, and construction.  Each subsection includes a brief discussion




of the products included in each product category, the processes used to




manufacture the products or a description of the operations involved, current




and projected geometric mean exposure levels, populations exposed, and




duration and frequency of exposure.




    Chapter III presents our estimates of air releases from milling, primary




manufacturing, and secondary manufacturing sources from stacks; and from




mining, brake repair, and construction as area releases.  Each subsection




presents our methodology for estimating air emissions and our emission




estimates.




    Appendix A presents geometric and arithmetic mean exposure levels,




population factors, and air emissions for production and use of asbestos




products no longer produced in or imported into the U.S.  These data are




reported for use in sensitivity analysis to estimate likely occupational




exposures and air releases should these products be produced or used in the




U.S. in the future.
                                   - xvii -

-------
I.  INTRODUCTION




    EPA is assessing the risks associated with the production and use of




asbestos and of products, made from asbestos.  As background for EPA's January




29, 1986 proposal to ban and phase down asbestos production and usage, occupa-




tional exposure levels and air releases for primary and secondary processing




of asbestos from 1981 8(a) submissions and exposure levels for construction




from the late 1970's and early 1980's were used.  Recently the Occupational




Safety and Health Administration (OSHA) has lowered its permissible exposure




limit (PEL) for asbestos to 0.2 f/cc from the previous 2 f/cc (OSHA 1986a).




Because of OSHA's new PEL, additional controls and respirators are likely to




be used to reduce current exposure levels to the new PEL or lower.  The




purpose of this study is to update occupational exposure data and air




emissions, used as inputs to determine ambient exposures, to better represent




current industry practices.




    A.  Approach




    Our approach was to identify and gather all available data from the




literature on occupational exposures and air emissions.  We performed an




on-line literature search using the NIOSH, Pollution Abstracts, Enviroline,




and NTIS databases to identify NIOSH, academic, and industry studies on




asbestos exposures.  In addition, we performed a search of all OSHA and




National Institute of Safety and Health (NIOSH) publications using the




NIOSH/OSHA publications index and the OSHA Technical Data Center's on-line




system.  NIOSH reports include "Hazard Evaluation and Technical Assistance




Reports (HHEs)," "Industry Wide Study Reports (IWs)," "Control Technology




Reports (CTs or Walk-Through Survey Reports)," and "Contractor Reports."  We




also gathered the Toxic Substances Control Act (TSCA) Section 8(a) data and




other literature referenced in the original "Exposure Assessment for




Asbestos," OSHA's background documents and docket on the new asbestos




                                    - 1 -

-------
standard, and submittals related to occupational exposure and air releases to




EPA's docket.  We also accessed the OSHA Computerized Information (OCIS) data




base developed in Salt Lake City under the SIC Industrial Hygiene (I.H.)




Information File; however, the data available on this system is so general




that it could not be accurately used to estimate exposures for specific




products.  NIOSH is currently performing several site visits to better




characterize exposures from brake repair; three new brake repair reports have




been received from this effort to date.




    Mining and milling exposure data are collected annually by the Mine Safet;




and Health Administration (MSHA).   We received copies of all the data




available from MSHA for the asbestos mining/milling facilities.  MSHA




regulates mining and milling operations and sets an exposure limit of 2 f/cc;




OSHA's new 0.2 f/cc PEL does not affect the miners and millers of asbestos.




    We searched the literature for data on air releases, but found that the




best base of data for estimating air releases was the OAQPS draft background




document entitled "National Emission Standards for Asbestos -- Background




Information for Proposed Standards" and the Section 114 Letters.  The draft




background document supplied baghouse efficiencies and stack information for




typical plants producing each primary product; these model plants include the




fiber consumption and the number of stacks with the corresponding flow rate




through each stack.




    The available literature data are limited; therefore, we supplemented the




data by gathering OSHA Compliance  data for asbestos (OSHA 1987) and




distributing a survey to all affected industry members identified in the ICF




Market Survey (1986-1987).  The ICF Exposure Survey (1986-1987) was sent to




all miners/millers of asbestos, primary and secondary manufacturers of




asbestos products, and several relevant industry groups.  OSHA Compliance dat,




(OSHA 1987) were supplied to us by the OSHA Office of Management Data Systems




                                    - 2 -

-------
for the Standard Industrial Classifications (SICs) corresponding to




manufacturing, construction, and automotive servicing.




        1.  Availability of Occupational Exposure Data




        This section summarizes the results of ICF's voluntary asbestos




exposure survey and the availability of occupational exposure data from the




ICF Exposure Survey; OSHA Compliance data; and NIOSH, industry, and academic




studies.




           a.  Asbestos Exposure Survey




           Table 1 presents the results of the ICF Exposure Survey (1986-1987)




for mining/milling, primary processing, chlorine manufacture, and secondary




processing.  The final column labeled "industry groups" was an effort to




obtain exposure data for downstream uses.




    One of the 3 asbestos mining/milling companies responded to the survey,




for a positive response rate of 33 percent.  This company, however, did not




provide monitoring data.




    The survey was sent to the 135 active primary processors identified by the




ICF Market Survey (1986-1987).  Many companies process asbestos in multiple




locations; the total number of plants represented by the primary processors is




183.  Of these 135 companies, 31 responded to the survey in some way, for a




total response rate of 23 percent.  However, only 15 of the facilities (or 11




percent) completed the survey to some degree.  (Not every positive responder




provided exposure data.)  Of the remaining respondents, 10 have ceased pro-




cessing asbestos and 6 refused to respond to the survey.  Table 2 presents a




distribution of the positive responses by product category.  Both an




indication of the number of plants and the number of companies are provided




for each product category.




    The survey was sent to 17 chlorine manufactures, some of which may not be




using asbestos diaphragm cells.  Of these 17 companies, 10 respondents




                                    - 3 -

-------
               Table 1.   Summary of ICF Exposure Survey Results
                      Mining/   Primary      Chlorine    Secondary   Industry
                      Milling  Processing  Manufacture3  Processing   Groups
Number of Companies
That Received
Surveys

Number of Plants
Using Asbestos
Operated By These
Companies

Total Number of
Responses
135
183
 31
17'
 56'
 10
 87
101
 21
 11
N/A
Survey Completed
Ceased Production
Refusal
Percent Response
Rate
Survey Completed
Ceased Production
Refusal
1
-
-
33%

33%
-
•
15
10
6
23%

11%
7.5%
4.5%
10
-
-
59%

59%
-
•
7
13
1
24%

8%
15%
1%
-
-
-
0%

_
-
-
N/A - Not applicable.

aUse of asbestos diaphragms for chlorine manufacture.

"These numbers of companies and plants may include facilities that do not use
asbestos diaphragms.
                                    - 4 -

-------
    Table 2.  Availability of Occupational Exposure Data for Primary Processors
 Product Category
1CF Exposure
   Survey
  (Number
of Plants)a
OSHA Compliance
  Inspections
   (Number of
    Plants)b
 NIOSH    Industry
Studies    Group,
(Number  Company,  or
  of       Academic
Plants)    Studies
Friction Materials
Sheet Gaskets
and Packings

Textiles

Roof Coatings and
Non-Roofing
Coatings
                (3 inspections)

     5                13
(4 companies)  (19 inspections)
                (6 inspections)

     1                 1

     8                25
(6 companies)  (33 inspections)
                     0

                     5
              0

              0
Total
Paper Products
A/C Pipe
A/C Sheet
2
1
0
3
2
2
0
0
0
0
1
0
5
4
2
                                          21
  2

 38
Asbestos -
Reinforced
Plastics
Total

0


18
(15 companies)
0 0


51 9
(67 inspections)
0


1

0


79

aThe number of plants equals the number of companies responding to the survey
unless indicated otherwise.  Many companies have multiple locations.  (Note:  Not
all survey responses were fully completed.)

"The number of inspections equals the number of plants unless specified otherwise.
Several facilities were inspected multiple times from 1979 to the present.
                                     - 5 -

-------
completed the survey, for a response rate of 59 percent.  Fifteen facilities

are represented by the 10 respondents; however, exposure data are limited for

a few of the facilities.

    The survey was also sent to the 87 active secondary processors identified

by the ICF Market Survey (1986-1987).  Many companies process asbestos in

multiple locations; the total number of plants represented by the secondary

processors is 101.  Of these 87 companies, 21 responded to the survey in some

way, for a total response rate of 24 percent.  However, only 7 of the

facilities (or 8 percent) completed the survey to some degree.  (Not every

positive responder provided exposure data.)  Of the remaining respondents, 13

have ceased processing asbestos and one refused to respond to the survey.

Table 3 presents a distribution of the positive responses by product category

Both an indication of the number of plants and the number of companies are

provided for each product category.

    The exposure survey was sent to 11 industry groups in an effort to gather

exposure data for downstream uses.  The following groups were sent surveys,

but none of them supplied any monitoring data:

        c  American Federation of State, County, and Municipal
           Employees, Washington, D.C.;

        •  Asbestos Information Association, Arlington, VA;

        •  Asphalt Recycling and Reclaiming Association, Annapolis,
           MD;

        •  Building and Construction Trades Department (AFL-CIO),
           Washington, D.C.;

        •  Edison Electric Institute, Washington, D.C.;

        •  Electrical/Electronics Insulation Conference, Washington,
           D.C.;

        •  Friction Materials Standards Institute, Paramus, NJ;

        •  Gasket Fabricators Association, Philadelphia, PA;
                                    - 6 -

-------
   Table 3.   Availability of Occupational Exposure Data for Secondary Processors
Product Category
                a
ICF Exposure
   Survey
  (Number
of Plants)b
OSHA Compliance
  Inspections
   (Number of
    Plants)0
 NIOSH    Industry
Studies    Group,
(Number  Company,  or
  of       Academic
Plants)    Studies     Total
Paper Products
Friction Materials
Asbestos -
Reinforced
Plastics
Missile Liner
Sheet Gaskets
1
2
0


1
4
0
2
0


0
8
1
1
2


0
0
0
0
0


0
0
2
5
2


1
12
and Packings
Textiles
  TOTAL
(3 companies)

     0


     8
(7 companies)
       11
                        25
aThe are fewer product categories for secondary processing because products such as
A/C pipe, roof coatings, and non-roof coatings do not require secondary processing.
In addition, there are currently no secondary processors of A/C sheet.
     number of plants equals the number of companies responding to the survey
unless indicated otherwise.  Many companies have multiple locations.  (Note:  Not
all survey responses were fully completed.)

cThe number of inspections equals the number of plants unless specified otherwise.
Several facilities were inspected multiple times from 1979 to the present.
                                     - 7 -

-------
        •  International Association of Heat and Frost Insulators and
           Asbestos Workers,  Washington, D.C.;

        •  International Union,  United Automobile,  Aerospace,  and
           Agricultural Implement Workers of America, Detroit, MI;  and

        •  Thermal Insulation Manufacturers Association,  Mt. Kisco,
           NY.

    In a further effort to obtain downstream exposure data, we called:

(1) Electrical/Electronics Insulation Conference, Washington,  D.C.;  (2) Unit

Automobile Workers, Detroit,  MI; (3) Friction Materials Standards Institute,

Paramus,  NJ;  (4) National Roofing Contractors Association, Chicago,  IL; and

(5) Edison Electric Institute, Washington, D.C.  Still, we were unable to

obtain any exposure data.

           b.  OSHA Compliance Inspections

           We reviewed a printout of OSHA Compliance Inspection summaries

dated 1979 through 1986 for asbestos (OSHA 1987) to extract summaries that

applied to the scope of this exposure assessment.  The printout was provided

by OSHA's Office of Management Data Systems for asbestos mining  (SIC  1499),

construction (SICs 1520-1799), manufacturing (SICs 2210-3999), and auto  repai

 (SICs  4171,  4231,  5511,  5541, and 7510-7549).

     No OSHA  compliance data  were found for asbestos  mining and milling.   This

 is reasonable  since  the Mine Safety and Health Administration (MSHA)  regulate

 mining operations.

     The  OSHA compliance summaries  for manufacturing totalled 920.  We

 cross-referenced the companies  (and plant locations) inspected with  the  1981

 Section 8(a) list of primary and secondary processors of asbestos  and with th

 most recent  list of processors  from the ICF Market Survey (1986-1987).

 Summaries for those companies and their respective plants that could be

 identified on any one of the asbestos processor lists were extracted.  Six

 additional inspections for companies that appeared to be asbestos processors


                                     - 8  -

-------
but could not be found on any of the lists were also extracted.  (For an




exposure assessment, it is appropriate to use data for facilities that have




since ceased production because one cannot assume that facilities ceased




production due to their exposure levels being out of the normal range.  It is




reasonable to assume that facilities ceasing production used similar process




technologies and engineering controls as are used by current manufacturers and




had exposure levels that are, therefore, representative of those experienced




by the remaining producers.)  The company (and plant) inspections selected as




appropriate for the exposure assessment were further divided by primary and




secondary processing and by product category, using the Section 8(a) and ICF




Market Survey (1986-1987) data.  Tables 2 and 3 present the distribution of




OSHA inspections by product category for primary and secondary processing,




respectively.  Both an indication of the number of plants and the number of




inspections are provided for each product category; several plants were




inspected multiple times.



    The remaining OSHA inspections for manufacturing appear to include




downstream users of asbestos products, and facilities that either do not use




asbestos but had concerned employees or have asbestos in either their




buildings or machinery (e.g., publishers).  However, 12 of these inspections




appeared to be brake rebuilders based on either the company names or the job




categories.  Words such as rebuilders, debonder, exchange, tear down,




stripper, and remanufacture in reference to brakes were used to select




appropriate companies for the brake rebuilding exposure assessment.




    We were not quite so successful in our review of the 1096 OSHA  inspections




for construction.  The difficulties in this review were the vague nature of




the SIC categories and the nondescript company names.  We found it  difficult




to confidently say an inspection referred to a specific product category.




Also,  in the vast majority of the inspections, the job titles  seemed to refer




                                    - 9  -

-------
to asbestos abatement rather than construction activities (e.g., insulation




remover, laborer scraper, asbestos remover), to activities where the worker is




indirectly exposed (e.g., boiler maker, electrician, HVAC technician,




plumber),  or to applications using asbestos products no longer used in the




U.S. (e.g., insulator).  IGF identified only 4 inspections that looked at all




applicable.  After speaking with the OSHA regional offices to obtain further




information on the activities at the time of the inspections, this number was




reduced to only one usable inspection for A/C shingle tear-off.  The OSHA




regional offices keep the original inspection files on hand for 3-4 years at




which time they are sent to the archives.  Of the two that were more recent




than this, only the one had to do with one of the products we are currently




investigating.



    The automotive repair printout yielded 82 inspections for automotive




repair, servicing, and garages; motor vehicle dealers; and terminal




maintenance facilities.  (OSHA inspection data were also obtained for service




stations,  but no relevant personal sampling results were identified).  The




asbestos concentration was non-detectable for 86 out of 109 personal samples.




The 8-hour TWA for the remaining 21 personal samples ranged from 0.005 to




0.940 f/cc.  These data are not included in our exposure estimate because of




the uncertainties associated with them.  It was not possible to determine




whether or not the monitoring data were taken during brake repair operations;




data for job titles such as "worker" or "laborer" were presented with no




indication of the activity being performed.  Even the mechanics may not have




been doing brake work at the time of. the inspection.




           c.  NIOSH and Other Studies




           In addition  to the data from the ICF Exposure Survey (1986-1987)




and the OSHA inspections (OSHA 1987), we collected NIOSH reports, contractor




reports, trade association studies, and company studies.  The availability of




                                    -  10 -

-------
these studies by product category for primary and secondary processing is




presented in Tables 2 and 3, respectively.  We also have 3 NIOSH studies for




construction.  For brake repair, there are 7 NIOSH studies, and for brake




rebuilding, there are 2 NIOSH studies.  We have one NIOSH report and one trade




association report for chlorine manufacture using asbestos diaphragms.  And




finally for mining and milling, MSHA supplied extensive monitoring data for




each company, and one company provided some of its own monitoring results.



           d.  SiifimiflT-y




           A good quantity of exposure data is available for mining/milling,




primary manufacture of friction materials and coatings, and secondary




manufacture of sheet gaskets.  Limited data are available for primary




manufacture of paper products, A/C pipe and sheet, packings and gaskets, and




textiles; secondary manufacture of paper products, friction materials




(including brake rebuilding), asbestos-reinforced plastics, coatings, and




textiles; construction using A/C products and roofing felts; and brake repair.




No exposure data are available for primary manufacture of asbestos-reinforced




plastics; exposure estimates for this product category, therefore, rely




heavily upon OSHA's recent assessment for the new PEL.




    There are uncertainties surrounding each major type of exposure data:




data supplied by companies or industry groups such as those obtained in the




ICF Exposure Survey, OSHA compliance data, and NIOSH or academic studies.  It




is believed that there are biases associated with each type of exposure data.




For example, OSHA compliance data is thought to represent high exposures




because OSHA is likely to focus its attention on facilities for which it




receives employee complaints or which have historically been out of




compliance.  However, OSHA also performs random inspections at facilities




which have a greater likelihood of compliance than those for which complaints




are received.  Conversely, company supplied data is thought to be biased




                                    -  11  -

-------
towards facilities using best available control technologies and, therefore,

having low exposure levels.  The basis for this belief is the unlikelihood

that facilities with high exposure levels would respond to a voluntary survey

and expose themselves as being out of compliance.   Finally,  NIOSH or academic

studies would likely focus on facilities or operations having high exposures

or on specific control technologies which would likely provide low exposure

levels (e.g., evaluation of vacuum systems to control exposures during brake

repair).  To determine the significance of these potential biases in the data,

we evaluated the available exposure data for primary manufacture of friction

products.  Primary manufacture of friction products was chosen for this

evaluation because of the wealth of available data from each type of source.

For two job categories, fiber introduction/mixing and finishing, the current

exposure data (pre-0.2 f/cc PEL) were segregated by the source type (i.e.,  ICF

Exposure Survey, OSHA compliance data, or NIOSH/academic studies), and

separate geometric means were calculated yielding the following  results:

        •  Fiber Introduction/Mixing

           --   ICF Exposure Survey:     0.329  f/cc
           --   OSHA compliance data:    0.566  f/cc
           --   NIOSH  studies:           0.085  f/cc
                Combination of sources:  0.251  f/cc

         •  Finishing

            --   ICF Exposure Survey:     0.149  f/cc
            --   OSHA compliance data:    0.424  f/cc
            --   NIOSH  studies:           0.193  f/cc
                Combination of  sources:  0.174  f/cc

The results  indicate  that there  is somewhat of a  bias in the data.   In

general,  OSHA coppliance data  yield the highest geometric mean exposures,

while ICF Exposure  Survey and  NIOSH data  yield comparable lower geometric  mean

 exposures.   In the  case of the NIOSH data for  fiber introduction/mixing,  the

 geometric mean exposure is significantly  lower than for the OSHA and ICF

 Exposure Survey data because the bulk of  the NIOSH data focused on evaluating

                                     -  12  -

-------
the effectiveness of bag opening equipment to control airborne fibers.  The

mix of the various data types for most product categories, however, minimizes

the effects of the biases, as indicated by the overall geometric mean levels

presented for friction products using a combination of sources.

        2.  Analysis of Occupational Exposure Data

           a.  Current Exposure Levels

           For product manufacturing, a large quantity of exposure monitoring

data has been obtained.  Due to the limited availability of exposure data on

any one product, we were only able to estimate exposure levels for product

categories.  However, these product category exposures are applied to the

worker populations for each individual product, thus allowing for some

distribution of risk by product.  This analysis, however, assumes that job

category exposures from all products in a product category are identical.  The

product categories are:

           Paper products;
           Asbestos cement products;
           Friction products;
           Textiles;
           Packings and gaskets;
           Coatings;
           Asbestos-reinforced plastics; and
           Miscellaneous uses.

This  analysis does not cover products no longer produced  in the U.S.  or

imported  into the U.S. such as commercial paper, corrugated paper, rollboard,

flooring  felt, roofing felt  (imported only), corrugated A/C sheet  (imported

only), and vinyl asbestos floor tile.  Occupational  exposure  levels  and

population factors  for products no  longer produced or used in the U.S. are

presented in Appendix A for use in  sensitivity analysis.

    Current  exposure levels  associated with each job category or task are

based on  historical  data.  Both geometric and  arithmetic  means of  the raw data

are presented throughout  the  text of this report  (and in  Appendix A).


                                    - 13 -

-------
    The traditional summary statistic used in industrial hygiene studies is




the geometric mean of the measured exposures.  It appears that this value is




used because:  (1) the distribution of exposures to toxic substances has been




found in other studies to follow a log normal distribution; and (2) the




geometric mean of a log normal distribution is the mode (most common single




value) of that distribution.  The geometric mean is, therefore, the single




level of exposure that workers are more likely to be exposed to than any other




single level of exposure.




    The geometric mean of exposure values would be the appropriate summary




statistic to use if our exposure concern centered on the level of exposure




typically encountered by a representative worker.  In the case of asbestos,




however, the health benefits model used to assess the consequences of exposure




is a linear, no-threshold, dose-response model.  Using a model of this type,




the total dose of asbestos fiber delivered to all workers determines the total




number of anticipated cases of lung cancer, mesothelioma, and cancers of the




gastrointestinal tract (Augustyniak 1987).




    If the concern is with total dose, rather than with "typical" levels of




exposure, the useful summary statistic becomes the arithmetic mean of the




distribution of exposures rather than the geometric mean of those exposures.




The arithmetic mean of the distribution is the summary statistic which, when




multiplied by the number of exposed workers, yields the total amount of worker




exposure to asbestos (Augustyniak 1987).




    The  important summary statistic for the  sampled population becomes that




summary  statistic which  is an unbiased estimator of the population mean.  The




arithmetic mean of the sampled values is  the unbiased estimator of the




population mean and, therefore, is used in the health benefits model to assess




the consequences of exposure  (Augustyniak 1987).
                                    - 14 -

-------
           b.  Populations Exposed/Duration and Frequency of Exposure




           Total 1985 worker populations for primary and secondary product




manufacturing for each product are calculated by summing up the populations




for each producer gathered during the ICF Market Survey (1986-1987).  These




populations are distributed into the various job categories identified by the




monitoring results using the population distributions obtained from the 1981




TSCA Section 8(a) submittals.  When total populations are not available due to




refusal to respond to the ICF Market Survey (1986-1987), we applied a ratio of




number of workers to asbestos fiber consumption based on data received from




responding companies manufacturing the same or similar products.  (Exposed




populations and duration and frequency of exposure for mining and milling were




obtained through telephone contacts with the respective company representa-




tives.)  All exposure levels for mining, milling, and product manufacture are




converted to 8-hour time weighted averages (TWAs) such that the appropriate




duration of exposure for all job categories is 8 hours/day even if a worker is




actually only exposed for a short period of time.  Eight-hour TWAs are




calculated from short-term exposure levels assuming "zero" exposure to




asbestos during the time the worker is not involved in asbestos operations and




taking a weighted average over the 8-hour work day.




    Since installation, repair, and removal jobs are intermittent, populations




for brake repair and construction have been calculated as full-time




equivalents (FTEs).  The FTE population is the number of workers working 250




days/year and 8 hours/day at installing, repairing, and removing the total




quantity of an asbestos product manufactured or imported each year (from the




ICF Market Survey (1986-1987)).  In reality, a much larger population would be




exposed for a significantly shorter period of time.  FTEs are calculated based




on the average time of a brake job or construction work crew productivity.




Short-term exposures, which represent the exposure during the period of time




                                    -  15 -

-------
in which the actual task is performed, are applied to this population.




Short-tern exposure levels are converted to 8-hour TWAs  (assuming no exposure




to asbestos during the remaining portion of the work day) to project exposures




under the new PEL.  Projections are made using the methodology described



below.




           c. Proj ected Exposure Levels




           Due to the lack of data on engineering controls corresponding to




the current exposure levels,  a comparative control evaluation to decide which




controls are effective in achieving exposures of less than 0.2 f/cc and what




exposure levels will actually be reached using these controls is not possible,



Therefore, we used a simplified and conservative approach for projecting




exposure levels under the 0.2 f/cc PEL.  This approach assumes that for those



operations where 8-hour TWA exposures are currently below 0.2 f/cc, work




practices will remain unchanged.  However, for those operations where the




8-hour TWA exposures are currently above 0.2 f/cc, work practices will be




changed either with the addition of engineering controls or respirators to




reduce the exposures to 0.2 f/cc.   It is further assumed that facilities




unable to meet the new PEL would likely cease production or use of asbestos




products.  Once the raw data  have been manipulated by this methodology, new




geometric and arithmetic mean exposures are calculated;  these are the



projected exposures.




    In projecting exposures under the new asbestos standard, OSHA (1986b)




generally used a methodology  which assumed that current exposure: levels




greater than or equal to 0.2  f/cc would be reduced to 0.15 f/cc,  while levels




less than 0.2 f/cc would be reduced to 80 percent of their initial values (in




certain cases,  OSHA used a slightly different methodology).   We chose not to




adopt OSHA's approach because of the lack of evidence to support these




assumptions,  and the unlikelihood that facilities with exposures already belov




                                    -  16  -

-------
0.2 f/cc would attempt to reduce exposures further with the installation of




expensive engineering controls or the implementation of a mandatory respirator




program.  Bragg (1986) states that under the new asbestos standard, industry




will achieve levels of 0.05 f/cc because (1) producers typically aim for




exposures below the PEL to insure compliance and (2) exposures of 0.5 f/cc (or




one-fourth of the PEL) were achievable when the PEL was 2 f/cc.  We chose not




to use Bragg's assumption because it becomes more difficult to reduce




exposures below the PEL by such a large margin as the PEL is reduced.  We




agree with OSHA and Bragg that producers and users are likely to achieve




exposures below the asbestos standard, and we allow for this by not adjusting




data currently below the 0.2 f/cc.  However, for facilities with exposure




levels greater than 0.2 f/cc, the level of reduction possible is not obvious.




Our methodology, therefore, provides conservative estimates of exposure, with




no attempt to "guess" at industry's ability to control exposures below the new




asbestos standard due to the lack of data to support such judgments.




    B.  Conversion Factor for Asbestos Measurement




    In analyzing air releases and ambient exposures, it is necessary to




convert back and forth between f/cc and ng/cc units to allow for dispersion




modeling (requiring mass inputs) and health benefits modeling (requiring fiber




count inputs).  This section discusses the phase contrast microscopy (PCM) and




the transmission electron microscopy (TEH) analysis techniques, and the factor




which has been derived to convert measurements from one technique to the




other.



    In the PCM technique, the analyst counts the number of fibers (all fibers,




whether or not they are asbestos) larger than 5 urn in length and with an




aspect ratio (length to diameter) of at least 3 to 1.  The results are




typically reported as optical fibers/cc.  The smallest visible fiber diameter




is approximately 0.2-0.5 urn (Ayer et al. 1965, Chatfield 1983).  The choice of




                                    -  17 -

-------
fiber sizes to be counted was designed to permit optimum reproducibility of

results by different analysts.  The PCM method does not differentiate between

asbestos and other fibers; but in most asbestos workplaces, where the only

significant fibrous contaminant is asbestos, this limitation is not very

serious.  The size distribution of asbestos fibers, and hence the percentage

of fibers fulfilling the method's counting requirements, is not the same in

all industrial processes or for all types of asbestos.  Thus a measurement in

a workplace with a high proportion of small fibers (i.e., fibers with sizes

under the resolution limit of the PCM technique) could be improperly judged to

pose less of a hazard than a measurement in an environment with fewer but

larger fibers (i.e., those which the method would be able to count).

    The measurement of asbestos contamination in the ambient air cannot be

adequately performed with PCM.  The principal reason for this is that the

proportion of airborne fibers which are actually asbestos in ambient air is

far smaller than that which is found even in "clean" asbestos workplaces, so

that "although it is found that an optical fibre count made on an

environmental sample usually yields a definite value, this value is totally

unrelated to the presence or absence of any asbestos" (Chatfield 1983).  The

proportion of total fibers which are asbestos in ambient air has been

estimated to be less than 10 percent (Spumy and Stober 1978).  This

difficulty is compounded by the fact that ambient air tends to have a smaller

proportion of the larger, PCM-countable fibers than workplace air (NRC 1984),*

a fact which has also made investigators wary of trying to evaluate the

presence of asbestos in ambient air by a method which only detects the larger

fibers.
     * Although there is wide variability, Chatfield (1983) indicates, as an
example, that only 0.7 percent of the total number of fibers in ambient air
exceeded 5 urn in length (i.e., are countable by PCM).

                                    -  18 -

-------
    As a result of these difficulties with the PCM method, ambient levels of

asbestos have been primarily characterized by transmission electron microscopy

(TEM) (Chatfield 1983).  With proper attachments,* a transmission electron

microscope can detect all asbestos fibers, of any size, present in air.

Furthermore, since the dimensions of each fiber can be measured, and the

density of each asbestos type is known, the results can be calculated and

reported in mass units (e.g., nanograms or micrograms per m^) or as particle

counts with quite detailed size distributions.  Mass concentration is the most

commonly reported parameter.

    As discussed above, the PCM method and the TEM method have radically

different resolution characteristics.  In effect, if the two methods were to

count the same asbestos cloud simultaneously, they would come up with very

different numbers.  It is necessary, therefore, to accurately define a

conversion factor to be able to extrapolate concentrations based on optical

fibers/cc to comparable relationships expressed in the typical TEM units of ug

of asbestos per cubic meter of air (ug/m^) (which are in turn calculated from

TEM counts).

    Investigators have attempted to empirically determine such conversion

factors by making parallel measurements with the two methods while adjusting

the TEM technique so that only fibers of the size which would be resolved and

counted by the phase contrast microscope get counted (and their weight

estimated) by the electron microscope analyst.  A few such studies have

reported a wide range of conversion factors, and there are at least two

important reasons for the variation in reported values.

    The most important reason is quite simple; there is in fact no single

conversion factor, but one for each specific set of conditions.  In the words
     * Selected Area Electron Diffraction (SAED) and Energy Dispersive X-Ray
Analyzer (EDXA).

                                    -  19 -

-------
of a recent important review of this issue by the National Research Council




(MC 1984), "these conversion factors usually cannot ... be applied to samples




obtained under a different set of conditions."  The conversion factor under




one set of conditions will almost inevitably be different from that obtained




under other conditions (e.g., different process, different source of raw




asbestos) because the size distribution of the airborne asbestos fibers will




be different.




    The second important reason for the variability in the conversion factor




is that different methods have been used in different studies to prepare




samples for the transmission electron microscope, and these methods can yield




quite different conversion factors even when other conditions are constant




(Chatfield 1983),  Different methods, for example, can result in different




proportions of fibers being lost in sample preparation before counting, or in




significant alterations in the size distribution of fibers from that which




actually exists in the sampled atmosphere (Chatfield 1983).




    Despite these uncertainties, there is a measure of scientific consensus




about the appropriate range of values to use as a conversion factor.  The most




widely quoted conversion factor of 30 ug/nr per optical fiber/cc has been




reported by Nicholson (EPA 1986g) as the geometric mean (with a very wide




variance) of six studies; the range of conversion factors was 5 to 150.  The




National Research Council (NRC 1984) in a recent review cited an equivalent




factor of 30 ug/m-* per optical fiber/cc, based on ratios developed in four




studies, three of which were also included in Nicholson's analysis (EPA




1986g).  Another recent review by Conmins (1985) estimates a similar range foi




the conversion factor of 2.5 to 50 ug/m^ per optical fiber/cc, citing eight




references which almost certainly go back to the same original data cited by




the other reviews.
                                    - 20 -

-------
    The range of values reported for the conversion factor (ug/m3 per optical




fiber/cc) fundamentally reflect the true variability of the conditions being




measured and, secondarily, the inherent inaccuracy of the different




measurement methods;  The bottom line is that, in effect, there is a "true"




and different factor for each conversion from optical fiber to gravimetric




units, depending on the fiber size distribution of the asbestos cloud in a'




particular situation.  There is currently no body of data available, however,




to allow investigators to make elegant estimates of conversion factors under




different circumstances, such as different asbestos processes.




    Any single value for a given conversion factor is, therefore, a rough




approximation at best.  In this context, we conclude that it is reasonable to




use the reported range of 2.5 to 150 ug/m3 per optical fiber/cc for the needed




conversion factor.  An approximate midpoint of 30 ug/m3 per optical fiber/cc




has recently become popularized and is a reasonable value to use; this




conversion is equivalent to 0.03 ng/fiber or 30 fibers/ng.




    C.  Report Format




    This report is divided into two maj or chapters:  Occupational Exposure and




Air Releases.  Chapter II presents our estimates of occupational exposures




during mining/milling, asbestos product manufacture, chlorine manufacture




using asbestos diaphragm cells, brake repair, and construction.  Each




subsection includes a brief discussion of the products included in each




product category, the processes used to manufacture the products or a




description of .the operations, current and projected geometric and arithmetic




mean exposure levels, populations exposed, and duration and frequency of




exposure.




    Chapter III presents our estimates of air releases from milling, primary




manufacturing, and secondary manufacturing sources from stacks; and from




mining, brake repair, and construction as area releases.  Each subsection




                                    -  21 -

-------
presents our methodology for estimating air emissions and our emission




estimates.




    Appendix A presents geometric and arithmetic mean exposure levels,




population factors,  and air emissions for production and use of asbestos




products no longer produced in or imported into the U.S.  These data are




reported for use in a sensitivity analysis to estimate likely occupational




exposures and air releases should these products be produced or used in the




U.S. in the future.
                                    - 22 -

-------
II.  OCCUPATIONAL EXPOSURE




    This chapter presents our estimates of occupational exposures during




mining/milling, asbestos product manufacture, chlorine manufacture using




asbestos diaphragm cells, brake repair, and construction.  Each subsection




includes a brief discussion of the products included in each product category,




the processes used to manufacture the products or a description of the




operations, current and projected exposure levels, populations exposed, and




duration and frequency of exposure.




    Due to the lack of data on engineering controls corresponding to the




current exposure levels,  a comparative control evaluation to decide which




controls are effective in achieving exposures of less than 0.2 f/cc and what




exposure levels will actually be reached using these controls is not possible.




Therefore, we used a simplified and conservative approach for projecting




exposure levels under the 0.2 f/cc PEL.  This approach assumes that for those




operations where 8-hour TWA exposures are currently above 0.2 f/cc, work




practices will be changed either with the addition of engineering controls or




respirators to reduce the exposures to 0.2 f/cc.  Once the raw data have been




manipulated by this methodology, new geometric and arithmetic mean exposures




are calculated; these are the projected exposures.  Where available,




information on current use of respirators is provided.




    The occupational exposure and population data bases from which exposure




estimates for jobs and job categories are derived is limited, often dated, and




contains data gaps and other uncertainties that require the use of




assumptions.  Several judgments have been made in the absence of actual data




and these are stated throughout the discussion.  Therefore, the results of




this analysis represent estimates given the available data and should be used




cautiously.  The exposure estimates, exposed populations, and frequencies and




durations of exposure are by no means absolute values.




                                    -  23  -

-------
    A,  Mining and MillinE




        1,  Process Description




        Asbestos fiber is currently mined in the continental United States by




either of two methods.  Depending on the type of deposits, asbestos is mined




by open-pit methods using either conventional processes or a hydraulic




processing technique.




    Conventional processing methods include blasting, crushing, grinding, and




air classifying the asbestos ore.  This mechanism of ore processing is termed




the "dry" method.  Alternately, asbestos ore can be mined by the wet process




involving wet screening and grinding.  The following sections describe the




mining, crushing and drying, and milling of asbestos ore for each of the two




methods of ore beneficiation.  Further details of the processes specific to




each miner/miller are discussed in the exposure sections below.




           a.  Conventional "Dry" Processing




           In the "dry" process, asbestos fiber containing ore is blasted or




drilled from an open-pit bench mine and shovel-loaded into trucks for




transport to the primary and secondary crushing areas at the top of the pit.




The ore is initially crushed at the primary jaw-crusher to reduce the ore




size.  The ore is further reduced in size (approximately 3/8-inch in diameter;




at the secondary cone-crusher.




    After crushing, the ore is transported by a belt conveyor to the drying




area  (rotary and tower kilns).  After drying, the ore is transported by




conveyor  to the dry ore storage bin.  The dried ore is then ready for the




milling operation.




    Milling of asbestos fiber-containing ore commences with transport of the




dry ore by conveyor or bottom-dump truck to the milling area.  Milling




consists  of removing  the imbedded asbestos fiber from the ore by a repeated




series of crushing, fiberizing, screening, aspirating, and grading operations




                                    - 24 -

-------
    The ore is introduced into a vibrating screen apparatus where the short




fiber is released from the granular material.  The low-density fibers tend to




rise to the top of the screen bed, while the granular material stays on the



screen surface (Kirk-Othmer 1984).




    The resultant mixture of rock fragments and asbestos fiber is then passed




over inclined shaking screens to separate the various sizes of material.  The




smaller particles of rock and the shorter asbestos fibers pass through the




screen openings and are carried away for further treatment.  The coarser rock




fragments and the longer asbestos fibers remain on the screen.  The shaking




action of the screen causes the fiber to rise to the top where it is lifted




off by air suction as it nears the end of the screen.  The fiber is aspirated




through large air ducts and transported to the collectors (Hwang 1981).




    The material remaining on the screen, which consists mainly of unfreed




asbestos in rock fragments, passes to fiberizers that release the rest of the




fiber.




    The coarse material that has fallen through the initial screening is




rescreened and refiberized using finer mesh screens to recover medium and




shorter length fibers.




    Finally all of the fibers are cleaned and separated into various




commercial grades and bagged in woven polylaminate or paper bags for shipment.




Packing of finished fibers is usually done by loose- or pressure-packing




machines ("baggers") that consolidate the fibrous material into the bag (NIOSH




1982a).




    In pressure packing the fiber is transported from a bin by screw conveyors




to pressure packers preceded by precompressors and pre-weighers (all




operations are automatically controlled).  Loose-packing requires only ambient




pressure and no compression of the fibers.
                                    -  25 -

-------
           b.  Unconventional "Wet" Processing




           The recovery of milled asbestos fiber from ore is usually quite lo



(about five percent).  The New Idria mining district, however, contains a




unique asbestos deposit that is an agglomerated ore consisting of more than 5




percent asbestos with moisture content of up to 20 percent.  The deposit is




made up of soft, friable sheets and clumps of asbestos fiber that consist



mainly of very short chrysotile fibers.




    The asbestos ore is mined by a conventional open pit method.  After




scraping off 10 to 20 feet of overburden, bulldozers and scrapers are used to




remove the ore from the deposit (Myers 1986a).   The ore is loaded into bottom




dump trailers though a 3/8-inch screen and hauled about 60 miles to the mill




site, where it is stockpiled.  Asbestos ore stored at the mill stockpile is



periodically watered down to prevent fibers from becoming airborne.




    Ore from the stockpile at the mill is slurried, crushed, sized, screened,




dewatered, pelletized,  and dried.  A portion of the product is sold in pellet



form and is shipped by bulk or in bags.   The remaining pellets are further




processed through a hammer mill to produce open fibers that are packaged and



bagged in the conventional manner (NIOSH 1982a).




           c-  Exposure Controls




           Typically, the emissions from asbestos mining and milling opera-



tions are kept to a minimum by the use of fairly simple process controls.  Ai




releases of asbestos fibers encountered during mining activities are




controlled by wetting the ore and fiber deposits.   Due to the outdoor nature




of mining, it is difficult to use any type of process controls or filtering




systems other than personal air-filtering devices (Myers 1986a, Toney 1986) t



control airborne asbestos levels.




    Protective factors  represent the "minimum anticipated workplace level of




respiratory protection that would be provided by a properly functioning




                                    - 26  -

-------
respirator or class of respirators to a percentage of properly fitted and




trained users" (Myers n.d.).  There are possible complications associated with




using protective factors.  The protective factors are for the ideal scenario




and do not take into account elements that would reduce the protective ability




of the various respirators and masks.  The protective devices must be worn




properly and cleaned regularly to insure that they are working correctly.




They must fit well and movement must be kept to a minimum (Rosenthal and Paull




1985).  The Chemical Engineering Branch of the Office of Toxic Substances at




EPA recommends a protection factor of 10X for any air-purifying half-mask




respirator including disposable equipped with any type of particulate filter




except single use, any air-purifying full facepiece respirator equipped with




any type of particulate fitter, or any supplied air respirator equipped with a




half-mask and operated in demand (negative pressure) mode (Myers n.d.).  A




protection factor of 25X is recommended for any powered air-purifying




respirator equipped with a hood or helmet and any type of particulate filter,




or any supplied air respirator equipped with a hood or helmet and operated in




a continuous flow mode (Myers n.d.).  These recommendations are based on




examination of literature and various experts in government and industry.




These protection factors are conservative estimates of actual protection which




is difficult to predict.  The actual protection factors are not less than the




recommended factors, but may be greater than* the recommended factors for these




types of respirators (Myers n.d.).




    Milling activities are such that airborne asbestos levels can in some




cases be controlled.  Many processing areas are enclosed and many operations




are automated.  These measures significantly reduce the exposure level that




workers would encounter, but cannot be used in all process areas.  Local




exhaust ventilation and air filtering systems (e.g., baghouses) reduce




airborne fiber levels in the workplace and help to minimize exposure during




                                    - 27 -

-------
milling activities.  The baghouse is also an efficient means of recycling




fiber to the milling operation, thereby increasing the efficiency of the




process and reducing the amount of waste that must ultimately be disposed.




During some milling activities, airborne fiber levels can be reduced by




wetting the fiber, but this technique is limited to those processing




techniques where moisture is permissible (e.g., in "wet" processing prior to




the drying operation and in conventional processing prior to kiln drying and




screening).




    The most efficient method of controlling airborne fiber levels, as in th«




case of mining, is by the use of respirators.  Other systems, if they are




employed, are an additional method of controlling fiber levels, but do not




insure that workplace fiber levels meet the required MSHA standard (2 f/cc)




for mining and milling activities.




        2.  Manufacturers and Production




        The three domestic producers of asbestos fiber are located in




California (2 companies) and Vermont (1 company).  The mines, their location,




and the annual production of asbestos fiber are listed in Table 4.  The volui




of domestic asbestos fiber production has decreased steadily over the years.




Current production levels are about one-third of 1980 levels, but are consi-




dered to be at a static level; production should remain at 1986 levels for




1987 and beyond until new regulations or use restrictions are proposed and




implemented.




        3.  Current Exposures




        The current standard for occupational exposure during mining and




milling activities is 2 fibers per cubic centimeter (f/cc) as set by the




Department of Labor, Mine Safety and Health Administration (MSHA).  The




standard (0.2 f/cc) that has been set by the Occupational Safety and Health




Administration (OSHA) does not apply to mining and milling activities because




                                    - 28  -

-------
Table 4.  Annual Domestic Production of Asbestos Fiber for 1985
                                          Annual Production
        Mine and Location                   Volume (tons)
     Calaveras Asbestos, Ltd.                 34,000
     Copperopolis Mine
     Calaveras County, CA

     KCAC Incorporated                        20,000
     Joe 5 Pit
     San Benito County, CA

     Vermont Asbestos Group                    8,070
     Lowell Mine
     Orleans County, VT                       •	

                    Total                     62,070
     Sources:  Myers 1985, Kenmer and Hall 1986,
               Phelps 1987.
                             - 29 -

-------
OSHA has no jurisdiction in these areas.  The exposure data presented herein




allows for comparison with exposures from other activities over which OSHA



does have jurisdiction.




    MSHA publishes personal exposure data for various mining and milling




activities, including asbestos.  Additional exposure data for one of the




asbestos mining and milling facilities  (Calaveras Asbestos Ltd..) have been




provided by a contact at that facility  (Toney 1987).  Data from this source




are also analyzed.




    The MSHA personal exposure data for each mining and/or milling site are




broken down by labor category, and the  asbestos exposure concentrations are




presented  in f/cc.  All exposures of interest are shift weighted averages




 (8-hour time-weighted average (TWA)).



    The procedure used to determine the presence of asbestos fibers is a




modification of a standard asbestos fiber counting method (NIOSH Standard




#582).  The modified procedure initially employs phase contrast microscopy




 (400-450X  magnification) to verify the  presence of fibers.  If two or more




fibers are detected per cubic centimeter, the sample is stained with a




solution to determine if the fibers initially detected are asbestos fibers.




This technique, known as staining dispersion, tests positive in the presence




of asbestos.  If the results are positive, the sample is examined under the




electron microscope to determine the aspect ratio of the fibers.  The



analytical and sampling procedures are  not 100 percent reliable; and,




therefore, an error factor is taken into account in determining whether to




issue a citation for exposure samples that are shown to contain asbestos




fibers in  concentrations greater than the exposure limit.  All MSHA personal




exposure data presented for the asbestos mining and milling facilities are foi




filter-type samples (Autio 1986).
                                    - 30 -

-------
    To assess the magnitude of possible exposure at each site, an




understanding of the processing techniques employed is necessary.  Therefore,




a brief discussion of each mining and milling facility, its practices, and its




engineering controls (if available) is presented below.  These discussions are




accompanied by a detailed description of the job categories and exposure data




for each site.  The exposure data are first presented as airborne breathing




zone samples for each site.  The exposure data are then adjusted using the




respirator protection factor for those job categories for which respiratory




equipment is designated as mandatory by each facility; this adjusted exposure




is the level to which the worker is actually exposed.




           a.  KCAC Incorporated




           The mine operations used by KCAC Inc. are conventional open-pit




stripping methods.  There is no drilling or blasting; and as  the ore contains




up to 20 percent moisture, emissions are minimal (Myers 1986a).  The stripping




and scraping operations are only undertaken about every three years (other




mining operations are performed annually, seven months per year), and the ore




is stockpiled at the mine for subsequent screening and hauling to the mill.




The work areas at the mine are wetted as necessary with water trucks during




stripping, screening, and truck loading to keep asbestos fibers  from becoming




airborne.  The mine operates eight hours/day and four days/week  for the dry




months of April through October (Myers 1986a).




    KCAC Incorporated employs three people at their Joe 5 Pit mining facility.




The three employees at the mine site can be designated as two production




workers and one supervisor (Myers 1986b).  The tasks that are undertaken at




the mine fall into six MSHA Job Code categories according to  the personal




exposure data.  The job categories and exposure levels for the Joe 5 Pit area




are given in Table 5.  When multiple exposure samples are available, geometric




and arithmetic means have been calculated.




                                    - 31 -

-------
                             Table S.  Asbestos Exposure Profile for the Mining and Milling Operations at KCAC Incorporated

Job Category (MBHA Job Code)
Mining
Scraper/Loader Operator (682)
Front-End Loader Operator (762)
Bulldozer Operator (368)
Dry Screening Plant Operator (488)
Scalper/Screener Operator (388)
Dryer Operator (379)
Number of
Workers"

N/A
H/A
N/A
N/A
N/A
H/A
Current Exposure Level
Without Respiratory Protection
(f/cc)b
Geometric Mean

0.
0.
0.
0.
0.
Oj

39
26
45
80
62
63

(2)
(2)
(3)
(1)
(1)
(1)
Arithmetic Mean

0.44
0.34
0.59
0.80
0.62
Ml
Current Exposure Level
With Respiratory Protection
(f/cc)d
Geometric Mean

0.039
0.026
0.045
0.080
0.062
0.063
Arithmetic Mean Frequency

0.044
0.034
0.059
0.080
0.062
0.043

8 hr/d.
8 hr/d.
8 hr/d,
8 hr/d.
8 hr/d,
8 hr/d,

4
4
4
4
4
4
and Duration

d/wk,
d/Mk,
d/wk.
d/wk.
d/wk.
d/wk,

7 mo/yr
7 mo/yr
7 mo/yr
7 mo/yr
7 mo/yr
7 mo/yr
    Total Number of Workers Involved
    in Mining Operations

  Weighted Average Exposure Level
  for Mining Operations0
           0.33
0.57
0.053
    Total Number of Workers Involved
    in Milling Operations

  Weighted Average Exposure Level
  for Milling Operations*
30
           0.41
                               0.43
                                                0.041
0.057
Milling.
Front -End Loader Operator (782)
Bulldoter Operator (368)
Slurry, Mixing, Pumping Operator (379)
Concentrator Operator (679)
Ball, Rod, or Pebble Mill Operator (179)
Palletizing Operations Worker (779)
Dryer Operator (379)
Bagging or Packaging Operations Worker (879)
Forkllft Operator (389)
Maintenance (618)

2
1
2
4
N/A
2
2
4
4
_2_

0
0
0
0
0
0
0
0
0
p

.21
.86
.39
.20
.50
.51
.49
.36
.29
• 9?

(3)
(1)
(1)
(1)
(1)
(2)
(3)
(10)
(3)
(1)

0.22
0.86
0.39
0.20
0.30
0.51
0.49
0.51
0.36
JLS

0.021
0.086
0.039
0.020
0.050
0.051
0.049
0.036
0.029
0.052

0.022
0.086
0.039
0.020
0.050
0.051
0.049
0.051
0.036
0.052 1

hr/d,
hr/d,
hr/d.
hr/d,
hr/d.
hr/d,
hr/d,
hr/d,
hr/d,
J hr/d.

3
3
3
5
3
5
3
3
5
3

d/wk.
d/wk,
d/wk,
d/wk.
d/wk,
d/wk.
d/wk.
d/wk,
d/wk,
d/wk,

12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
12 mo/yr
                                                                  0.043
N/A - Not available.

"Myers 1986b.

bMSHA 1986.  Exposure levels are geometric and arithmetic mean,  8-hour TWA*.   The number of data points is given in parentheses.

cAssumes all employees spend equal time performing each job and are,  therefore,  all exposed to the same average exposure level.

-------
                                                                   Table 5 (Continued)


 Assumes all employees use full-face mask respirators with a protective factor of 10X (Myers  1986b, Myers  n.d.).

aThe average Is calculated by weighting exposure values by number of employees in each  category.   In  categories  for  which no data on nunfeer of employees
are available, exposure values are excluded from the average exposure.

-------
    HSHA designates six job categories for the mining operations of KCAC

Incorporated,  while KCAC registers only 3 employees for that site.  It is,

therefore,  obvious that some of the jobs are only operational for a portion of

the shift.   For purposes of this analysis, we assume that the 3 employees are

involved in each of the jobs for equal amounts of time during each shift.

    After screening at the mine site, the asbestos containing ore (3/8-inch

pieces and less) is transported about 60 miles to the King City Mill in

Monterey County.  Prior to milling, the ore is stockpiled about 1/4-mile from

the mill and is kept wet with water trucks and sprinkler systems as necessary

to reduce emissions (Myers 1986a).

    The milling process utilized by KCAC is the "wet" method and involves

slurry ing the ore  (about 90 percent asbestos) with water and pumping it to the

mill  (Myers 1985).  The ore beneficiation system is wet until the final drying

and packaging operations, which are all equipped with baghouse-type dust

collection systems.  Collected dust is either directly sent to a product bin

or slurried with water and returned to the process.  The King City Mill is

operational eight hours/day and five days/week for the full year.

    The KCAC King City Mill has 43 employees involved in various activities.

However, not all of these employees are involved in the production of

asbestos.  The breakdown of asbestos production workers by job category is

approximately as follows (Myers 1986b):

         9 - Maintenance employees
         4 - Forklift operators
         4 - Concentrator operators
         4 - Bagging and packaging (includes RG-244 and wet-end)
         2 - Slurry/pumping/mixing operators
         2 - Dryer operators
         2 - Pelletizing operators
         2 - Front-end loader operators
         1 - Bulldozer operator

        30   Production Workers and Supervisors
                                    -  34 -

-------
The exposure data from MSHA does not cover all of the categories identified by




KCAC, but weighted average exposure levels for the available categories based




on the number of exposed workers are presented in Table 5.   All of the data




presented in Table 5 are personal exposure data,  eight-hour time weighted




averages.




    It should be noted that 80 percent of all personnel at both KCAC




facilities (mine and mill) wear full-face respirators depending on the




particular activities in which they are involved (Myers 1986b).  Respirators




are required in certain areas and while performing certain activities (e.g.,




packaging and loading operations), but are not mandatory in all areas (Myers




1987).  It is assumed, however, that all workers are wearing respirators at




all times until more exact information becomes available from KCAC Inc.  Using




this assumption, actual exposure accounting for use of respiratory protection




would be 10X (Myers n.d.) less than that indicated by the MSHA personal




exposure data.




           b.  Calaveras Asbestos. Ltd.




           The mining and milling operations of Calaveras Asbestos, Ltd. are




located at the same facility in Copperopolis, California.  The operations are




seasonal (10 months/year); the mine and mill are closed in January and




February.  Due to the seasonal nature of the operation at Calaveras Asbestos'




facility, the number of production workers varies.  During active periods, the




mine is in operation for eight hours/day and five days/week, while the mill is




in operation 24 hours/day and five days/week (Toney 1986) . •




    The mining operation at Copperopolis is a conventional open-pit bench




mine.  Ore is blasted or drilled and then loaded into trucks and hauled to the




primary and secondary crusher area at the top of the pit.  After the ore has




been crushed (jaw, cone, and impact crushers), it is transported by conveyor




to the drying area.  After drying, the ore is transported, again by conveyor,




                                    -  35 -

-------
 to  the dry ore  storage bin for holding until needed for the milling  operation:

 (NIOSH 1982a).

     Calaveras Asbestos, Ltd. produces  asbestos grades from their milling

 operations that can be classified as Group 4 (1/8-inch long) and Group  6  (3/8-

 inch long) fibers.   To produce these fibers,  the ore is beneficiated by

 crushing, screening,  and  fiberizing the various sizes of asbestos-containing

 ore. Finally,  the  asbestos fibers that reach the top of each shaking screen

 are vacuumed off, sent to the collector,  separated by fiber size, cleaned, and

 bagged  in woven polylaminate or paper  100-pound bags for shipment.   Bagging or

 packaging is performed by a hydraulic  pressure packer (NIOSH 1982a).

     The  average number of workers at Calaveras Asbestos is 135 or more  (Toney

 1986).   Not  all of  these  workers,  however,  are involved in the production of

 asbestos fiber  from ore.   Depending on the time of year and whether  the mill

 is  operating at full or partial capacity,  the actual number of workers  exposed

 to  asbestos  in  the  mining and milling  operations is about 80 (Toney  1987).*

     A breakdown of  production workers  by job category is presented in Table 6.

 The number of workers and exposure data by job category are provided for

 mining and milling  activities, as  well as  exposure reduction achieved by  the

 use of air-purifying equipment.   Both  MSHA and Calaveras (Toney 1987)

 monitoring data are used  to estimate geometric and arithmetic mean exposures

 by  job category.  The Calaveras Asbestos samples are taken as part of an

 on-going monitoring program (Toney 1987).**
       It should be noted that conventional mining and milling operations are
more labor intensive than "wet" processing operations due to considerations
involving the nature of the ore recovery and beneficiation methods and the
relative age and inefficiency of the equipment used in the conventional
operations.

     ** Mining exposure data provided by Calaveras Asbestos, Ltd., are from
samples taken in the spring months when atmospheric conditions cause the
exposure levels to be less than in summer months.  The exposure values can be
expected to be 50-75 percent higher in the summer months than the actual
values for mining activities obtained in April (Toney 1987).
                                    -  36 -

-------
                          Table 6.   Asbestos Exposure Profile for the Mining and Milling Operations  at Calaveras  Asbestos,  Ltd.

Job Category (MSBA Job Code)
Mining
Powdergang Worker* (607)
Churn Drill Operator* (434)
Front-End Loader (762)
Quarry Truck Driver (376)
Bulldozer Operator (368)
Dunp Operator (622)
Hater Truck/Blade Operator (479)
Supervisor (649)
Greaser/Oiler (618)
Number of
Workers"

3
2
2
11
2
1
1
2
_i
Current Exposure Level
Without Respiratory Protection
(f/cc)B
Geometric Mean

0.08
0.12
0.07
0.02
0.11
0.11
0.10
N/A
0.18

(1)
(2)
(1)
(1)
(1)
(1)
(2)

(2)
Arithmetic Mean

0.08
0.13
0.07
0.02
0.11
0.11
0.11
N/A
0.13
Current Exposure Level
With Respiratory Protection
(f/cc)d
Geometric Mean

0.008
0.012
0.007
0.002
0.011
0.011
0.010
N/A
0.012
Arithmetic Mean Frequency

0.008
0.013
0.007
0.002
0.011
0.011
0.011
N/A
0.013

6
8
8
8
8
8
8
8
8

hr/d.
hr/d.
hr/d,
hr/d.
hr/d,
hr/d,
hr/d.
hr/d,
hr/d,

5
3
3
3
5
5
5
5
5
and Duration

d/wk,
d/«*,
d/wk,
d/wk.
d/wk,
d/wk.
d/wk,
d/wk,
d/wk,

10/mo/yr
10/Do/yr
10/mo/yr
10/mo/yr
10/mo/yr
10/mo/yr
10/mo/yr
10/mo/yr
10/mo/yr
    Total Number of Workers Involved in
    Mining Operations

  Weighted Average Exposure Level
  for Mining Operations0
                                                    25
           0.06°
0.06*
                                                0.006
    Total Number of Worker* Involved
    In Milling Activities

  Weighted Average Exposure Level for Milling
  Operation*6
49
           0.39
                               0.49
                                                0.034
                                                                  0.006
Hilling
Front End Loader Operator (782)
Gathering Arm Loader Operator (043)
Mill Operators (179)
Crusher Operators (079)
Sizing and Hashing Operation* Workers (388)
Dryer Operator* (379)
Bagging or Packaging Operation* Worker* (879)
Forklift Operators (389)
Janitor* (413)
Laboratory Technician* (314)

1
1
3
4
3
2
19
7
3
6

0.34
0.30
0.41
0.48
0.69
2.14
0.26
0.14
0.57
o,??

(2)
(1)
(1)
(5)
r (5)
f (1)
(14)
(3)
(5)
(2)

0.33
0.30
0.41
0.73
0.70
2.14£
0.40
0.24
0.65
0.25

0
0
0
0
0
0
0
0
0
0

.034
.030
.041
.048
.069
.086
.026
.014
.057
^025

0
0
0
0
0
0
0
0
0
o

.033
.030
.041
.073
.070
.086
.040
.024
.063
,025

hr/d,
hr/d,
hr/d,
hr/d.
hr/d,
hr/d.
hr/d.
hr/d,
hr/d,
hr/d,

3 d/wk,
3 d/wk,
5 d/wk,
5 d/wk,
3 d/wk.
5 d/wk,
3 d/wk.
3 d/wk,
3 d/wk,
5 d/wk,

10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
10 mo/yr
                                                                  0.044
N/A - Not available.

*Toney 1987.

TCHA 1986, Toney 1987.  Exposure level* are geometric and arithmetic mean,  6-hour  TWA*.  MSHA job code* are assigned to the job titles used in the
Calaveras exposure data summary.  The number of data points is given in parentheses.

-------
                                                                               Table 6  (Continued)


            cTh» average le calculated by weighting exposure values by number of employees  in  each Job category.  Categories  for which no data on number of employees
            ot exposure levels ate available are excluded from the exposure analysis.

             Assumes all mine employees use a half-face mask with a protective factor of  10X and all mill employees use half-face masks  (protection  factor of  10X)
            except dryer operator* who use battery powered helmets (protection factor of  25X)  (Toney  1987, Myers n.d.).

            elt should be noted that the average value for exposure during  mining operations is low for  the Calaveras site.   This la due in part to  the large  number
            of truck drivers responsible for ore-hauling at the facility.   There la  a high  demand for ore-hauling because ot  the low asbestos content of the ore, and
            the truck driver exposure is comparatively low.

             MSHA data for this job category indicated an 8-hour TWA exposure of 0.04 f/cc.  This data point has not been included  in this analysis  because Calaveras
            Asbestos, Ltd. (Toney 1987) Indicated that exposure for these workers is usually high.  Calaveras requires a battery powered helmet for  workers in this
            category due to the high concentrations of asbestos that they encounter.
00

I

-------
    The 25 workers involved in the mining of asbestos-containing ore can be

divided approximately as follows (Toney 1987):

        11 - Quarry Truck Drivers
         3 - Powder Gang (1 powderman, 2 helpers)
         2 - Front-End Loader Operators
         2 - Supervisors
         2 - Bulldozer Operators
         2 - Drill Operators
         1 - Greaser/Oiler
         1 - Dump Operator
         1 - Hydration Plant (Water Truck/Blade) Operator
        25   Production Workers and Supervisors

In addition to the employees involved in mining activities at Calaveras

Asbestos, there are 49 employees involved in milling activities at the same

site.  The breakdown of workers for milling is approximately as follows (Toney

1987):

        19 - Bagging or Packaging Operations Workers
         7 - Forklift Operators
         6 - Laboratory Technicians
         4 - Crusher Operators
         3 - Sizing and Washing Operations Workers
         3 - Janitors
         3 - Mill Operators
         2 - Dryer Operators
         1 - Gathering Arm Loading Operator
         1 - Front-End Loader Operator
        49   Production Workers and Supervisors

    Maintenance workers are intimately involved in the production of asbestos

fiber, but are not classified as either miners or millers.  Maintenance

workers could be included within the category where their job is performed

(e.g., screen repairman could be included with mill workers), but Calaveras

Asbestos, Ltd. claims that these workers are usually excluded from exposure

studies dealing with mining and/or milling due to the intermittent and varying

nature of the exposure that occurs in these occupations (Toney 1987).

    It is likely that maintenance employees will be engaged in activities that

will cause them to be exposed to high levels of asbestos.  For example, a


                                    -  39  -

-------
mechanic may be exposed to high asbestos fiber concentrations while repairing

drilling equipment or overhauling bagging or packaging equipment used in the

milling operations.  It is not likely, however, that maintenance employees

will be exposed to high levels of asbestos for extended periods of time.

Maintenance operations are of a changing nature, and there are two factors

that would affect the level and duration of actual maintenance exposure:

        (1)  The employee may be a skilled laborer (e.g., mechanic)
             and may be exposed to asbestos for only short periods
             (i.e., less than eight hours per shift); and

        (2)  The employee may have varying maintenance duties and may
             be exposed to very different levels of airborne asbestos
             fibers depending on activities at a given time during the
             shift.

It is, therefore, difficult to develop "average" eight-hour TWA exposure

levels for maintenance workers due to the probable fluctuation in exposure

levels and the transitory nature of the activities.

    The job categories identified as being part of the maintenance operations

include:

           clean-up crew workers;
           laborers/bullgang;
           screen repairmen;
           supply handlers;
           mechanics;
           engineer/technical services; and
           building repair crew.

It should be noted that maintenance workers are normally associated with the

mining and milling of asbestos, but their duties are considered by Calaveras

Asbestos to be independent of the actual mining and milling operations  (Toney

1987).  While it is true that the equipment used and the facilities employed

in the production of asbestos fiber from ore need to be maintained, these

activities are not specifically a part of the mining or milling operations.

Available HSHA and Calaveras exposure data for maintenance workers are,

therefore, not presented herein.


                                    -  40 -

-------
    The other asbestos mining sites and milling facilities (KCAC, Inc.  and

Vermont Asbestos Group) have specified that maintenance operations should not

be considered separately from mining and milling operations; and, therefore,

exposure data for maintenance workers have been included in the relevant

operation (mining or milling) for those sites (Hyers 1987, Phelps 1987).

    All mining and milling workers at Calaveras required by MSHA to wear

protective devices do so.  In addition, many workers that are not required to

wear air-purifying and air-filtering equipment wear them due to health

concerns. There are two types of protective devices worn by employees at

Calaveras Asbestos:

        •  negative-pressure, HEPA filter, half mask, self-contained
           breathing apparatus; and

        •  3M(R) model number W-344, battery powered helmets with a
           3M(R)-8710 disposable face plate.

The negative-pressure, half masks have a protective factor of 10X and are worn

by all employees exposed to asbestos, with the exception of those potentially

exposed to fiber concentrations greater that 2 f/cc (Toney 1987, Myers n.d.).

There is at least one job  (dryer operator) that would potentially expose

workers to asbestos levels greater than 2 f/cc.  All employees  in this

exposure category are required to wear the battery powered respirators  that

have been recommended for  a protection factor of 25X (Toney 1987, Myers n.d.).

Calaveras will be equipping workers  in higher exposure categories with  a more

efficient respirator in  1987; this model  is a full face, HEPA filtered,

battery powered respirator with a nose clip.  These respirators have been

approved for  a protective  factor of  SOX in the workplace environment and the

amount of protection provided is not affected by variation  in fit or movement

(Toney 1987,  Myers n.d.).

    A breakdown of the number and  type of employees that  actually wear  some

type of protective equipment  is not  available.  Therefore,  it will be assumed

                                     - 41 -

-------
that all employees except those identified as wearing more efficient




respirators (protection factor of 25X), are wearing half-face masks that have



a protective factor of 10X (Myers n.d.).




           c.  Vermont Asbestos Group




           The mining and milling operations at the Vermont Asbestos Group




(VAG) are of the conventional type and are very similar to the operations at




the Calaveras Asbestos facility.  The VAG facility (the Lowell mine and mill)




is located in Orleans County, Vermont and employs 75 people, 48 that were




involved in some aspect of the production of asbestos fiber from asbestos -




containing ore in 1985.  The number of production workers and supervisors was




also about 50 in 1986 and is expected to remain the same in 1987 (Phelps




1987).




    Previously, the Lowell mine and mill has employed more production workers




while operating at full capacity (35,000 tons of fiber per year).  Currently,




it is operating at about 1/4 capacity, producing 8,070 tons of asbestos fiber




in 1985 and roughly the same amount (less than 10,000 tons) in 1986 (Phelps




1987).




    There are two shifts per day at the VAG mining facility.  Most of the




operations at the mine are performed during the first shift.  Most operations




last eight hours/day and 186 days/year, although conveyor belt crew operators




work ten hours per shift.  The second shift at the mining site is a skeleton




crew of three (truck drivers and one shovel operator) that are responsible for




hauling ore to the milling facility (Phelps 1987).  Milling operations are




also performed for 186 days per year, but there is only one shift per day.




The shifts are ten hours each, except for four activities (lab technician,




supply handler, backhoe operator, and some supervisors) that are performed on




eight-hour shifts.
                                    - 42 -

-------
    There are approximately 16 mine workers involved in various activities  at

the Lowell mine that can be divided into various job categories.   An

approximate breakdown of workers for mining activities as provided by VAG

(Phelps 1987) is:

         6 - Conveyor Belt Crew Workers
         2 - Rotary Drill Operators
         2 - Quarry Truck Drivers
         2 - Bulldozer Operators
         2 - Drill Helpers
         1 - Front-End Loader Operator
         1 - Shovel Operator
        16   Production Workers and Supervisors


The number of workers involved in milling activities at VAG is 32 and can be

divided as follows (Phelps 1987):

         8 - Maintenance Workers
         6 - Bagging or Packaging Operations Workers
         5 - Clean-up Crew Workers
         3 - Laboratory Technicians
         3 - Administrative Supervisors
         2 - Dryer Operators
         2 - Crusher Operators
         1 - Dry Screening Plant Operator
         1 - Supply Handler
         1 - Backhoe Operator
        32   Production Workers and Supervisors

    MSHA personal exposure data are available for various activities

undertaken at the VAG facility (Table 7).  The breakdown of employees by job

category provided by VAG (Phelps 1987) is used to determine weighted exposure

values for mining and milling activities.

    No separation of maintenance workers has been performed because these

workers are considered part of the milling activities for VAG (Phelps 1987).

In addition, no breakdown of maintenance workers by specific activity (e.g.,

mechanic, welder) was available.  For some job categories provided by VAG,

there are no exposure levels available from MSHA; these job categories have,

therefore, not been included in the weighted-average exposure calculations.

                                    -  43  -

-------
                        Table 7.  Asbestos Exposure Profila for th* Mining and Milling Operations at the Vermont Asbestos Group
Number of
Job Category (MSBA Job Code) Workers"
Mining
Rotary Drill Operators (734 and 634)
Drill Helper (833)
Bulldozer Operator (368)
Conveyor Belt Crew (601)
Front-End Loader Operator (782)
Shovel Operator (367)
Truck Driver (376)
Total Number of Workers Involved in
Mining Operations
Weighted Average Exposure Level for Mining
Operations0
Hilling
Backhoe Operator (778)
Crusher Operator (079)
Dry Screening Plant Operator (488)
Dryer Operator (379)
Bagging or Packaging Operations Worker (879)
Clean-Up Man (613)
Administrative Supervisor (649)
Supply Handler (671)
Maintenance (513)
Laboratory Technician (514)
Total Number of Worker* Involved in
Milling Operations

2
2
2
6
1
1
_2
16




1
2
1
2
6
5
3
1
a
3
32

Current Exposure Level
Without Respiratory Protection
(f/cc)b
Geometric Mean

0.58
0.16
0.70
0.38
0.27
N/A
N/A


0.42


N/A
0.49
0.59
0.58
0.63
1.74
1.14
N/A
N/A
0,63



(3)
(1)
(2)
(3)
(2)








(6)
(13)
(5)
(10)
(13)
(1)


(1)


Arithmetic Mean

0.59
0.16
0.98
0.54
0.34
N/A
N/A


0.54


N/A
0.51
0.66
0.66
0.67
2.31
1.14
H/A
R/A
0.63


Current Exposure Level
With Respiratory Protection
(f/cc)d
Geometric Mean

0.058
0.016
0.070
0.038
0.027
N/A
N/A


0.042


N/A
0.049
0.059
0.058
0.063
0.174
0.114
N/A
N/A
0.063


Arithmetic Mean Frequency and Duration

0.059
0.016
0.098
0.054
0.034
N/A
H/A


0.054


H/A
0.051
0.066
0.066
0.067
0.231
0.114
H/A
R/A
0.063



. 8
8
8
10
8
a
8





8
10
10
10
10
10
10
8
10
10



hr/d,
hr'/d,
hr/d,
hr/d,
hr/d.
hr/d,
hr/d.





hr/d.
hr/d.
hr/d,
hr/d,
hr/d.
hr/d,
hr/d.
hr/d,
hr/d,
hr/d,



186
186
186
186
186
186
186





186
186
186
186
186
186
186
186
186
186



d/yr
d/yr
d/yr
d/yr
d/yr
d/yr
d/yr





d/yr
d/yr
d/yr
d/yr
d/yr
d/yr
d/yr
d/yr
d/yr
d/yr


  Weighted Average Exposure Level for Milling
  Operations0
                                                               0.93
                                                                                   1.09
                                                                                                    0.093
                                                                                                                      0.109
N/A - Not available.

•Phelpa 1987.

HcHA 1986.  Exposure levels are geometric and arithmetic mean,  8-hour THAI.   The number of data points  is  given in parentheses.

°The average la calculated by weighting exposure values by number of employees in each Job category.   Job categories for which no data on  number  at
employees or exposure levels are available are excluded from the exposure analysis.

dAssumes all employees use half-face masks with a protection factor of 10X.

-------
    All employees at VAG are required to have a respirator (Gerson disposable




Number 1710) with them at all times, and these respirators must be worn in




specified areas of the facilities as part of a mandatory respirator program




(ICF Exposure Survey 1986-1987).  The Gerson Number 1710 has a workplace




protection factor of 10X, but labelling for use with asbestos has been




voluntarily removed by the company at the request of NIOSH (Bellinger 1987).




While MSHA is not bound by law to follow NIOSH criteria and recommendations




for asbestos, VAG may wish to follow NIOSH recommendations and require some




other type of respirator.  For preliminary calculations, we have assumed that




the Gerson masks provide a protective factor of 10X, although it may be more




appropriate to use a lower protective factor for this type of respirator.




    The only type of exposure controls in place are negative draft from




baghouses located at various emission sources throughout the plant (ICF




Exposure Survey 1986-1987).




        4.  Summary




        Available data make it possible to break down the exposure for mining




and milling activities by job category.  The number of employees in each job




category for each facility and their respective 8-hour TWA exposure levels and




exposure duration/frequency are presented in Tables 5, 6, and 7.




    An overall profile of asbestos exposures in the mining and milling




industries for each facility, taking into account information on number of




workers by job category and protective equipment used, show that average




exposure levels are well within the MSHA permitted exposure, limit of 2 f/cc,




and with protective devices provided by the facilites (required in some




instances) could be well below the OSHA standard (0.2 f/cc).




    B.  Product Manufacture




    This section presents occupational exposure profiles for primary and




secondary product manufacture.  For product manufacture, a large quantity of




                                    -  45 -

-------
exposure monitoring data are available.  Both geometric  and arithmetic means

of the raw exposure data are presented  in this section.   The  raw exposure data

have been converted based on each worker's duration  of exposure such that the

effective duration of exposure  is 8 hours/day.

    Due to the limited  availability of  exposure  data on  any one product, we

have estimated exposure levels  for product categories.   However, these product

category exposures are  applied  to the worker populations for  eac.h  individual

product  (from the 1CF Market  Survey  1986-1987),  thus allowing for  some

distribution of  risk by product.  This  analysis, however, assumes  that job

category exposures  from all products  in a product category are identical.  The

product categories  are:

            Paper products;
            Asbestos cement products;
            Friction products;
            Textiles;
            Packings and gaskets;
            Coatings;
            Asbestos-reinforced plastics; and
            Miscellaneous uses.

 This analysis does not cover products no longer produced in  the U.S. or

 imported into the U.S. such as commercial paper, corrugated  paper,  rollboard,

 flooring felt,  roofing felt  (imported only), corrugated A/C  sheet  (imported

 only),  and vinyl asbestos  floor tile.  Occupational exposure levels and

 population factors  for products no longer produced or used in the  U.S. are
                                            i
 presented  in Appendix  A for use in sensitivity analysis.

      To distribute total populations  froa the IGF Market Study (1986-1987)  into

 job categories, 1981 TSCA  Section 8(a) data were used (Hendrickson and Doria

 1983, EPA n.d.).  The  percent of the total worker population in each job

 category for each product  in 1981 was assumed to be the same in 1985.  This

 percentage was applied to  the 1985 total population to estimate the number  of

workers in each job category.  Since job categories defined in this report  do


                                    - 46  -

-------
not always correspond to those presented in the TSCA Section 8(a) data,  it was




often necessary to attribute the percent population from the 8 (a) data evenly




among two or more of our job categories.




        1.  Paper Products




           a.  Product Descriptions




           Asbestos is used in papers primarily due to its chemical and heat




resistant properties.   This section provides descriptions of those asbestos




paper products currently manufactured in the U.S. and/or imported for




secondary manufacture or installation in the U.S.  These products include




millboard, roofing felt (imported only), pipeline wrap, beater-add gaskets,




high-grade electrical paper, and specialty papers.  Currently, rollboard,




corrugated paper, commercial paper, and flooring felt are no longer




manufactured in the U.S. nor are they imported (IGF Market Survey 1986-1987).




Sources indicate that no roofing felt is produced in the U.S.; however,




283,200 squares of roofing felt are imported annually  (ICF Market Survey




1986-1987).  Nevertheless, there are no production workers exposed to asbestos




roofing felt; thus, it is excluded from this section.




               (1)  Millboard




               Asbestos millboard is an asbestos paper product similar  in




appearance to heavy cardboard.  It is used as a  fire-resistant lining in




floors, walls, ceilings, and fire doors, as well as an insulating barrier in




commercial ovens and household appliances.  Different  grades  of  millboard are




available, each differing in their ability to withstand  elevated temperatures.




Standard  millboard may withstand temperatures up to 850*F high quality




millboard may withstand up  to 1,000'F.  Certain  "premium" grade  asbestos




millboards are manufactured to withstand temperatures well  above 2,000°F.




     The primary  constituent of asbestos millboard  is asbestos fiber, with the




balance consisting  of binders and  fillers.  The  asbestos content ranges from




                                    - 47 -

-------
60 to 95 percent by weight; 70 to 80 percent asbestos  is  considered  typical.




Frequently used binders  include starches,  elastomers,  silicates,  and cement;




mineral wool,  clay,  and  lime  are commonly  used fillers.




     There are many commercial, residential,  and industrial  uses  of asbestos




 millboard.   Specific industrial applications include thermal  protection in




 large circuit breakers in the electrical industry;  lining for covers and




 troughs in the aluminium, marine,  and aircraft industries;  and .insulation in




 glass tank crowns, melters. refiners, and sidewalls in the  glass industry.




 Very thin millboard is sometimes cut for use as gaskets.   Commercial




 applications include fireproof wallboard linings for safes, dry  cleaning




 machines, incinerators,  and spark and glare shields in welding shops.




 Residential applications include tent shields, stove mats and linings for




 stoves, heaters,  and electrical switchboxes.




                (2)  Pipeline Wrap




                Asbestos pipeline wrap is an asbestos paper product similar to




 asbestos roofing  felt.  It is made with approximately 85  percent asbestos and




 15  percent  cellulose fibers and starch binders.  Asbestos is  used in pipeline




 wrap because  it resists soil, chemicals, rotting, and decay,  while maintaining




 dimensional stability throughout its lifetime.




     The largest user of this product is the oil and gas industry with their




 extensive underground piping networks; there is some use  by the  chemical




 industry for underground hot water and steam piping.  The asbestos felt




 protects the pipe from moisture,  corrosion, rot, and abrasion.  Pipeline wrap




 is used minimally in above-ground applications, such as for special  piping in




 cooling towers.   Pipeline wrap is applied to the pipe by  high-speed  wrapping




 machines.   The wrap is usually attached or bonded to the  pipe surface by




 special adhesive  coatings or by hot enamels that are coated onto one side of
                                    - 48  -

-------
the wrap.  The pipeline wrap is designed to last the service life of the

pipeline to which it is applied.

               (3)  Beater-Add Gaskets

               Beater-add gaskets are installed to provide tight, non-leaking

connections in piping and other joints.  Asbestos is used in gaskets because

it is heat resistant, resilient, strong, and chemically inert.  Asbestos

beater-add gasket papers contain approximately 60 to 80 percent asbestos and
                                                                   (-
20 to 40 percent binders.  Binders that are used include latex, styrene-

butadiene, acrylic, acrylonitrile, neoprene, fluoroelastomeric polymers, and

silicone polymers.  The binder determines the material's suitability for use

in water, aqueous solutions, oil, fuel, or chemical environments.

    These gaskets are primarily used in the automotive industry as heat

gaskets, carburetor gaskets, and oil and transmission gaskets and in trains,

airplanes, and ships.  They are also used in a variety of industrial and

commercial equipment, such as boilers and furnaces.  Asbestos gaskets are also

used widely in the chemical industry because of their chemical inertness.

               (4)  High-Grade Electrical Paper

               Asbestos is non-flammable and has high thermal and electrical

resistance, properties which make it very useful as an electrical insulator.

Asbestos high-grade electrical paper is composed of 80 to 85 percent asbestos,

encapsulated in high-temperature organic binders.  Generally, the paper

contains asbestos fibers and cellulose bound with latex polymers.   Chemical

treatment is sometimes used to remove trace elements from asbestos  fibers.

    The major use of asbestos electrical paper is in insulation  for high

temperature, low voltage applications.  It may be used in motors, generators,

transformers, switch gears, and other heavy electrical apparatuses, usually at

operating temperatures of 266*F to 428°F.
                                    - 49 -

-------
               (5)  Specialty Pacers




               Specialty papers  are produced in small  volume for specific  end




uses.  The products  in this subcategory include transmission paper,  filter




paper, cooling tower fill,  metal lining paper,  electrical wire or cable




wrapping paper,  and industrial and decorative laminates (OSHA 1986b).




 Transmission paper,  because of its structural stability and oil resistance,




 has been used as a covering for metal transmission disks in automatic




 transmissions.




     Asbestos has been used in filters for the purification and clarification




 of liquids because  it offers an exceptionally large surface area per unit  of




 weight and has a natural positive electrical charge which is very useful for




 removing negatively  charged particles found in beverages (Krusell and  Cogley




 1982).



     Asbestos filters may contain, in addition to asbestos, cellulose fibers,




 various types of latex resins,  and occasionally, diatomaceous earth (Krusell




 and Cogley 1982).  The asbestos content of beverage filters ranges from a low




 of 5 percent, for rough filtering applications, to a high of 50 percent, for




 very  fine filtering.  In general, the higher the asbestos content, the better




  the  filtering qualities of the  filter  (Krusell and Cogley 1982).




      Applications of the asbestos  filter paper are found primarily in the beer,




  wine, and liquor distilling industries where they are used to remove yeast




  cells and microorganisms from liquids.  Asbestos filters are also used for




  filtration of some  fruit juices,  such as apple juice,  and for special




  applications in the cosmetics and Pharmaceuticals industries.




             b.   Process Descriptions




                 (1)   Primary Manufacture




                The main operations in all asbestos paper primary manufacturing




are receiving,  bag opening, mixing,  forming, and finishing.  In the fiber




                                    - 50 -

-------
introduction operation, raw asbestos is most often introduced in unopened




pulpable bags, although for certain types of paper the fiber is dumped from




the bags.  In cases where the fiber is dumped from the bags, asbestos is




obtained in non-compressed pulpable bags so that the bags may be slit and the




asbestos added directly to the mixer.   At the mixing stage the fiber is




immediately wetted.




    As in other manufacturing processes, the asbestos fiber is carried under




negative pressure by conveyor to the mixer.  There, the fiber is wet-mixed




with paper stock, binder, and other ingredients.  The stock slurry flows into




the papermaking machine and forms a sheet.  The solids content of this sheet




may be less than five percent; the moisture content of this sheet is reduced




greatly during transit through the paper machine.   The wet nature of the




material precludes the release of asbestos fiber.




    The forming of asbestos paper is completed during the drying, slitting,




and calendering stages.  The final operation involves rewinding in which the




paper products are bulk packaged on spools, reels, or beams from the larger




rolls.  Rewinding is a dry operation.




    There are certain primary manufacturing procedures that are specific to




the individual products within the paper products category, although most of




the operations are similar if not identical.




    Asbestos millboard j,s manufactured in essentially the same process that is




used in the general paper manufacturing industry (i.e., the operations




described above).  A cylinder is rotated in a vat of pulp, creating a thin




fiber coating.  The coating is removed from the cylinder and drawn through a




process for partial dewatering.  Sheets are wound continuously until the




desired board thickness is obtained.  The built-up layer of material is cut




lengthwise and removed for drying.  Herein lies the principle difference




between millboard and paper manufacture.  Paper is manufactured into a




                                    -  51 -

-------
continuous sheet while millboard is not.  Standard size millboards are 42 x 48




inches and 1/32- to 3/4-inch thick.  The most popular thicknesses are 1/4- and




1/2-inch millboards.




    Pipeline wrap is manufactured on conventional papermaking machines.  The




manufacture of pipeline wrap differs from the manufacture of other paper




products in that the felt is saturated with coal tar or asphalt before drying.




In addition, it is usually reinforced with parallel strands of fiberglass.




After saturation, the felt passes over a series of hot rollers which set the




coal tar or asphalt into the paper.  The felt then passes over a series of




cooling rollers that reduce the temperature and provide a smooth surface




finish.  The felt is then air-dried, rolled, and packaged.




    Beater-add gaskets are manufactured on conventional papermaking machines




by the same process as the other paper products and are, therefore, considered




paper products.  The binder is added during the beater process, which is how




the name "beater-add" gaskets originated.  The gasketing paper is usually




produced in a sheet or roll varying in thickness from approximately 1/64-inch




to 1/16-inch.  Most gasket paper is sold to fabricators who cut the beater-add




paper to customer-specified sizes and dimensions.  The gaskets may be further




processed by reinforcing the gasket with wire or by sheathing the paper with




various metals, foils, plastics, or cloth.




    Electrical paper is manufactured in rolls, sheets, and semi-rigid boards.




The rolls and sheets are manufactured on conventional papermaking machines.




Boards are formed on wet process board machines.




    Asbestos specialty papers (specifically filter paper) are made on a




conventional papermaking machine.  Due to the very low demand for certain




types of specialty paper, the machines are generally used to produce more




popular paper products, such as the non-asbestos filter substitutes (i.e.,




diatomaceous earth and cellulose fiber products  (Krusell and Cogley 1982).




                                    -  52 -

-------
                (2)  Secondary Manufacture




                For paper products, there is secondary processing during the




manufacture of  millboard, beater-add gaskets, electrical paper, and specialty




papers.  Final  fabrication of millboard usually involves cutting, trimming,




and shaping to  meet the requirements of the space into which the millboard is




to be installed.  Secondary manufacturers of gaskets cut the gaskets from




paper sheets using metal die stamping or processing machinery.




           c.   Production and Employment




           Table 8 presents the total production and asbestos consumption, and




estimates of total employees exposed to asbestos for each type of paper




product.  The data presented are based on 1985 figures, excluding companies




that no longer  produce asbestos products.  The estimates of total employment




are based on figures from the ICF Market Survey (1986-1987), supplemented by




figures from the ICF Exposure Survey (1986-1987).  For producers who did not




provide employment data, we estimated employment from the average number of




workers per ton of asbestos consumed for each product type,  based on available




data.  Employment figures presented in the table should be considered




estimates only.




    In addition, 2.7+ tons of asbestos millboard were imported into the U.S.




in 1985 (one company failed to provide data).  Secondary processors of




millboard employ a total of over 448 asbestos workers (ICF Market Survey




1986-1987).   Due to numerous refusals by secondary processors to supply either




asbestos mixture consumption or population data, it is not possible to




estimate populations for these facilities; therefore, the population estimate




for secondary manufacture of millboard is a lower bound estimate.  1981 TSCA




Section 8(a) data were not adequate to refine this estimate.




    In 1985, 2,898 tons of pipeline wrap were imported (ICF Market Survey




1986-1987).   There is, however, no secondary manufacturing.




                                    -  53  -

-------
      Table 8. . Production and Employment for Primary Manufacture
                           of Paper Products
                                 Total        Asbestos
                              Production3   Consumption"
                                (tons)         (tons)
    Total
 Population
  Exposed
to Asbestos0
Millboard
Pipeline Wrap
Beater-Add Gaskets
High-Grade Electrical Paper
Specialty Paper
581
276,949 sq.
16,505
698
434
435.8
1,333.0
12,436.4
744.0
300.3
12
27
227
27
6
a!985 production, excluding production by companies no longer producing
asbestos products.

"1985 consumption, excluding consumption by companies no longer
producing asbestos products.

Population based on IGF Market Survey (1986-1987) (1985 employment
excluding companies no longer producing asbestos products),  ICF
Exposure Survey (1986-1987),  and estimates described in the text for
companies with no population figures reported.

Sources:  ICF Market Survey 1986-1987, ICF Exposure Survey 1986-1987,
          and ICF estimates.
                                 -  54 -

-------
    Total 1985 imports of beater-add gaskets are estimated at over 5.6 tons.




Because few data were supplied by secondary manufacturers of beater-add




gaskets for the ICF Market Survey (1986-1987), total population for secondary




processing of beater-add gaskets was estimated by adjusting 1981 populations




from the TSCA Section 8(a) data (RTI 1985) to 1985 populations by multiplying




by the ratio of 1985 primary production (ICF Market Survey 1986-1987) to 1981




primary production (EPA 1986b, midpoint was used).  The exposed population was




estimated to be 1,264 for secondary manufacturing of beater-add gaskets.




    Secondary processing of high-grade electrical paper potentially exposes 20




workers to asbestos (ICF Market Survey 1986-1987).  There are no imports of




high-grade electrical paper.




    Secondary processing of specialty papers potentially exposes 145 workers.




An estimated 1 ton of specialty paper was imported in 1985 (ICF Market Survey




1986-1987).




           d.  Exposure Profile




           Airborne asbestos fibers are generated throughout the entire




asbestos paper manufacturing process.  This is the case for all five of the




products specified in this category.  Table 9 presents the exposure profile




for each paper product as determined from the raw monitoring data.  We




categorized the jobs performed by each worker monitored into one of several




job categories.  Geometric and arithmetic means were calculated for the data




by job category.  This summary table includes the results for primary and




secondary manufacturing of asbestos paper products.
                                    -  55  -

-------
                                                                 Table 9.  Exposure Profile for Paper Product*
ui
CT\
8-Hour TWA Exposure (f/cc)
Product
Job Category*
Population
Pre-p,^ f/cc PELC
Geometric Mean
Arithmetic Mean
Post-0.2
Geometric Mean
f/cc PEL"
Arithmetic Mean
Duration
(hr/day)'
Frequency
(days/year)1
Primary Manufacturing
Millboard



Pipeline Wrap



Beater-Add Gaskets



Hlgh-Grade Electrical Paper



Specialty Paper



Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
total
Fiber Introduction
Processing
Other
Total
3
3
6
12
8
9
	 i2
27
20
11*
93
227
*
13
	 18.
27
1
3
2
6
0.091 (6)
0.013 (23)
0,052 (5}
0.052
0.091 (6)
0.013 (23)
0.052 (5)
0.051
0.091 (6)
0.013 (23)
0.052 (S)
0.036
0.091 (6)
0.013 (23)
0.052 (5)
0.039
0.091 (6)
0.013 (23)
0.052 (51
0.039
0.13*
0.030
0.070
0.076
0.13*
0.030
0.070
0.076
0.134
0.030
0.070
O.OS6
0.13*
0.030
0,070
0.060
0.13*
0.030
0.070
0.061
0.079
0.013
0,05.2
0.0*9
0.079
0.013
0.052
0.0*7
0.079
0.013
0.052
0.03S
0.079
0.013
0.052
0.037
0.079
0.013
0.032
0.037
0.09*
0.030
0.070
0.066
0.09*
0.030
0.070
0.06*
0.09*
0.030
0.070
0.052
0.09*
0.030
0.070
0.05*
0.09*
0.030
0.070
0.05*
8
8
8
8
8
8
8
8
8
8
8
a
a
8

8
8
8
8
8
172
172
252
211
172
172
250
201
172
172
2JO
20*
172
172
8250
201
172
172
250
198
Secondary Manufacturing.
Millboard
Beater-Add Oeaketa
High-Grade Electrical Paper
Specialty Paper
H/A
H/A
H/A
H/A
**8+
1,26*8
20
1*5
0.016 (26)
0.016 (26)
0.016 (26)
0.016 (26)
0.022
0.022
0.022
0.022
0.016
0.016
0.016
0.016
0.022
0.022
0.022
0.022
8
8
8
8
250
250
250
250
         "job categorie* are baaed on a concise cetegorication of the job titles.

         bEstimatlon of total population and the distribution of the population into Job categorlea la described in the text.  Population by Job category la estimated
         from the 1981 TSCA Section 8(a) data (Hendrickson and Dorla 1983).

         cThes« values represent geometric and arithmetic means of the raw 8-hour THA exposure data.  The number of data points ia given in parentheses.   The values
         corresponding to the total populations are calculated as weighted average* baaed on the number of worker* exposed in each Job category.  Since the monitoring
         data la aggregated for all paper producta, the exposure values for each Job category are assumed to be equivalent for all products.

-------
                                                                              Table  9  (Continued)


          The** post-0.2 f/cc PEL exposure  value*  are calculated directly from the raw monitoring  data.   Each 8-hour  THA exposure value that 1* above 0.2 f/cc  1*
         reduced to exactly 0.2 f/cc.   Data that are already at or  below this  value  remain  unchanged.  The value  corresponding to the total population la determined in
         the game manner as the pre-0.2 f/cc PEL total exposure value.

         eTh* effective duration of exposure is  6  hours/day in  all  cases.   Where  no  duration  is  provided, 8  hours/day is assumed.   Exposure* for lea* than 8 hour*  are
         converted to 8-hour TWA* when not  already appearing aa such,  assuming zero  exposure  during perioda  Mien  the  worker la not handling asbeatoa.

          Frequency refera to the number of day* annually  that  the  workers are performing a task involving potential  exposure to asbestos.   The frequency la not
         assumed to be 250 days/year for all primary paper manufacturing since data  for  specific job type* indicates  otherwise.   Data from the ICF Exposure Survey
         (1986-1987) show* typical frequencies to  be 172 days/year  for stock preparation (categorized as  fiber introduction)  and paper machine operation (categorized
         as processing).

         8Thls total population was estimated by adjusting 1981 TSCA Section 8(a) population  (RTI  1985) to 198S by multiplying by the ratio of 1985'primary production
         (ICF Market Study 1986-1987)  to 1981 primary production (EPA  1986b).

         Sources:  ICF Market Survey 1986-1987,  ICF Exposure Survey 1986-1987,  OSBA  1987.
Ui

-------
    Since the monitoring data are aggregated for all paper products, the

durations, frequencies, and exposures  for each job category are assumed to be

equivalent for  all  paper products.  However, since the distribution of the

total populations into job categories developed from 1981 TSCA Section 8(a)

data (Hendrickson and Doria 1983) for primary manufacturing is different for

 each product, the weighted average  exposures and frequencies for each product

 vary among the various paper products.

     The total population values for each specific product are based on the  ICI

 Market Survey  (1986-1987).  The populations for manufacturing sectors for

 which population data were not available are estimated by using the ratio of

 population to amount of asbestos fiber  consumed  for all  other paper products.

 The distribution of the total number of workers  exposed  into specified job

 categories is based on 1981 TSCA Section 8(a)  data  (Hendrickson and Doria

 1983).

     For secondary manufacturing of paper products,  the total populations are

not disaggregated into job categories.   Population  data  were not available  foi

all secondary manufacturers of millboard; thus,  the population for secondary

processing of millboard is most likely larger than  the value indicated.  1981

TSCA  Section  8(a) data are not adequate to refine this estimate.   The

populations for secondary beater-add gasket companies for which population

data were not available are estimated from the ratio  of  population to asbestos

consumption for  the other  secondary beater-add gasket companies.   Due to gaps

in data necessary to estimate  total population for  secondary processing of

beater-add gaskets, this population was  estimated by  adjusting the 1981
      * In reality, however, sources indicate  that exposure levels  can vary
 widely depending on the asbestos content of the product (OSHA 1986b).   Levels
 of exposure at a plant producing beater-add gasketing containing 90  percent
 asbestos are expected to be higher than levels  at a plant producing  specialty
 paper with a 10 percent asbestos content.

                                    - 58 -

-------
population from the TSCA Section 8 (a) data (RTI 1985) to 1985 by multiplying




by the ratio of 1985 primary production (ICF Market Survey 1986-1987)  to 1981




primary production (EPA 1986b, midpoint was used).




    As indicated in Table 9, the primary manufacturing operation with the




greatest potential for causing asbestos exposure is fiber introduction.   The




fiber introduction procedure in paper manufacturing involves the dumping of




asbestos into a beater or hydropulper.  A local exhaust system with dust-




collection equipment is used to keep the processing area under negative




pressure.  A primary method used to reduce exposure during this step is the




acquisition of asbestos in noncompressed pulpable bags.  This enables the bags




to be slit open and the asbestos added directly to the mixer where it is




immediately wetted.  Emissions can vary greatly at this point, depending on




the physical process employed.  Some plants perform fiber introduction and




stock preparation in a single wet operation and others perform them as




separate operations.



    The wet-mixing of the fiber with paper stock, binder, and other




ingredients controls the release of airborne asbestos.  As in most other




industry sectors, the asbestos is transported to  the mixer under negative




pressure by conveyor.  Rigorous housekeeping and  clean-up measures are




critical during mixing to prevent spillage of material.  Central vacuum




cleaning systems and mechanical floor-sweeper-vacuum units are often used




during these operations.



    Canopy hoods and exhausts that are utilized to remove water vapor and heat




from steam-heated rolls in  the dry section also aid in asbestos dust control.




These devices augment the general ventilation in  the dry processing area.  At




the slitting and calendering stages,  local exhaust ventilation  (LEV), area




hoods, and central exhaust  collection systems are the  typical engineering




controls.  Housekeeping is  very important as well during these operations.




                                    - 59 -

-------
 LEV and hoods represent the normal dust  control measures at  the  rewinding




 step.  This dry operation involves the bulk packaging of paper products on




 spools, reels, or beams from larger rolls.  Control  of airborne  fibers in the




 workplace has been and should continue to be  achieved mostly by  LEV and strict




 housekeeping  and work practices.  Respirators are likely worn by personnel at




 bag  opening,  pulping, and scrap handling operations.




     All job categories comprising the manufacture of asbestos papers have been




 able to achieve mean exposure levels below 0.2 f/cc;  this  is indicated by the




 values exhibited in Table 9 for all job categories.   Out of  all  the monitoring




 data shown in Table B-l for primary manufacturing, only a  few  samples are




 above the 0.2 f/cc level; all of these data fall under the fiber introduction




 job category.   Assuming that in these few cases additional controls will be




 utilized to achieve 0.2 f/cc exposure levels, the projected  exposure under the




 new PEL will decrease.




    Table  9  indicates  that the average post-0.2 f/cc PEL exposure level for




 the total exposed  population of each paper product (calculated by the




proportion of  workers in  each job category) is only slightly lower than the




pre-0.2 f/cc PEL value.   The  decrease is solely due to changes in the fiber




introduction job category.




    No changes are projected  for secondary  manufacturing of  paper products.




As  indicated in Table 9,  the  pre-  and post-0.2 f/cc PELs  are identical; this




 is  due  to  the  absence of  any  monitoring  data  greater than 0.2  f/cc.  Job




 categories are not identified for secondary manufacturing because  all tasks




 fit under the general  category of fabrication.




         2.  Asbestos-Cement Pipe




            a.  Product Description




            Asbestos-cement pipe is used primarily for transporting drinking




 water (pressure pipes)  and for draining storm water,  sewage, and other liquid




                                     -  60  -

-------
waste  (non-pressure pipes).  Other applications are in industrial products,




air/gaseous products, and electrical conduit for heating, cooling, and




gas-venting.  The composition is generally 15 to 25 percent asbestos by




weight, 42 to 53 percent Portland cement, and 34 to 40 percent ground silica



sand.




    The use of A/C pipe appears to be regional, primarily occurring in the




Southwest.  The use of raw asbestos in the production of A/C pipe dropped




significantly between 1981 (129,800 metric tons) and 1982 (37,600 metric tons)




but has remained fairly constant since then (i.e., 26,100 tons in 1983, and




37,000 tons in 1984) (OSHA 1986b).   In 1985, approximately 32,691 tons of




asbestos fiber were consumed in this sector (ICF Market Survey 1986-1987).




Manufacturers produced 216,903 tons of A/C pipe in 1985, and employed 286




asbestos workers; there are no known imports (ICF Market Survey 1986-1987).



           b.  Process Description




           Asbestos-cement (A/C) pipe is composed of a mixture of Portland




cement, finely ground silica, and asbestos fibers (ICF 1984).  The process as




a whole consists of four distinct steps:  raw material blending, pipe forming,




curing/autoclaving, and finishing.   In the blending step, asbestos fiber is




fluffed and separated in a willow,  then silica and Portland cement are added




to form a dry mix.  Water is added to the dry mix, and the mixture is blended




to form a homogeneous slurry.  The slurry is fed onto a moving felt conveyor




and water is drawn through the felt by a vacuum to form a continuous




asbestos-cement sheet.  The sheet is wound onto a cylindrical mandrel until it




reaches the desired thickness, and thus, forms a pipe.  The pipe is loosened




electrolytically by producing gases between the mandrel and the formed pipe.




After a short time, the mandrel is removed and the pipe is cured.  Curing is




either by water immersion or pressurized steam called autoclaving.  This




autoclaving step enhances corrosion resistance to high sulfate soils and




                                    -  61 -

-------
 waters.   After the curing/autoclaving step,  the pipe is  passed to the




 finishing area where the pipe is  cut to size and the ends  are trimmed  and




 machined to  facilitate junctions.   A/C pipes are produced  in a variety of




 diameters, formulations,  and weights designed for different applications.




 Diameters may range from 4 to 42  inches,  and standard lengths are 10 to 13




 feet.




     A/C  pipe undergoes little or  no secondary processing (Anderson et.  al.




 1983).   The  user of this product  (i.e., the  construction industry) may perforn




 some cutting of the pipe at the sites of the installation.




           c.  Exposure Profile




           The total exposed population is based on the  ICF Market Survey




 (1986-1987).  The population for  companies for which no  data were available  is




 estimated by using the ratio of population to amount of  asbestos fiber




 consumed for the other companies.  The total population  is allocated into job




 categories using the 1981 TSCA Section 8(a)  data (Hendrickson and Doria 1983).




 The data presented in Table 10 are based on  1985 figures,  excluding companies




 that no longer produce asbestos products.




     Data indicate that exposure levels to asbestos fibers  in A/C pipe




 manufacturing may vary widely from process to process.  This is revealed in




 Table 10 which presents the exposure profile for A/C pipe  production as




 determined from the raw monitoring data.




     This summary table shows geometric and arithmetic mean exposure values




 ranging  from quite low for certain miscellaneous jobs (i.e., the "other" job




 category)  to close to 0.2 f/cc for finishing operations; this rather high




value for  finishing correlates to exposure during certain dry mechanical




operations.   Exposures at coupling cut-off operations in an A/C pipe plant




have been shown  to  average 0.369  f/cc,  higher than any other operations (Bragg
                                    - 62 -

-------
                                               Table 10.  Exposure Profile for A/C Pipe — Primary Manufacture
8-Hour TWA Exposure (f/cc)

Product Job Category*

Population
Pre-0.2 f/cc PEL0
Geometric Mean
Arithmetic Mean
Post-0.2
Geometric Mean
f/cc PEL"
Arithmetic Mean
Duration
(hr/day)*
Frequency
(days/year)
Primary Manufacturing
A/C Pip* Fiber Introduction
Pipe Forming
Finishing
Other
Total
14
49
109
114
286
O.OS3 (2)
0.077 (34)
0.171 (79)
0.047 (44)
0.100
0.103
0.129
0.311
0.080
0.178
O.OS3
0.071
0.117
0.046
0.078
0.103
0.096
0.141
0.072
0.104
8
8 '
6
e
8
250
2SO
250
250
250
"job categories are based on a concise categorization of the job titles.

bEstimation of total population and the distribution of the population into Job categories is described in the text.  Population by job category is estimated
from the 1981 TSCA Section 8(a) data (Hendrlckson and Oorla 1983).

cThese values present geometric and arithmetic means of the raw 8-hour TWA exposure data.   The number of data points is given in parentheses.   The value
corresponding to the total population Is calculated as a weighted average based on the number of workers exposed in each Job category.

 These post-0.2 f/cc PEL exposure values are calculated directly from the monitoring data.  Each 8-hour THA exposure value that is above 0.2 f/cc ia reduced
to exactly 0.2 f/cc.  Data that are already at or below this value remain unchanged.  The valua corresponding to the total population ia determined in the
same manner as the pre-0.2 f/cc PEL total exposure velue.

*The effective duration of exposure is 6 hours/day in all cases.  Where no duration is provided, 8 hours/day is assumed.   Exposures for less than 8 hours are
converted to 8-hour THAs, assuming cero exposure during periods when the worker is not handling asbestos.

 Frequency refers to the number of day* annually that the worker* are performing a task involving potential exposure to asbestos.   The frequency is assumed to
be 250 days/year unless data for specific job types indicates otherwise.

Sources:  ICF Exposure Survey 1986-1987, Clarke 1986, OSHA 1987, ICF Market Survey 1986-1987.

-------
 1986).   This  operation involves  the repetitious cutting of A/C pipe coupling




 into small sections for use  in pipe connections.




     As  indicated in Table 10,  the value for finishing operations can be




 expected to decrease significantly to  a post-0.2 f/cc PEL geometric mean




 exposure level of about 0.12 f/cc  (or  an  arithmetic mean exposure level of




 0.14 f/cc).  This can be accomplished  via engineering controls.  Engineering




 controls during the coupling cutoff operations and other A/C pipe dry




 finishing processes (e.g., drilling and lathing) typically include custom-




 engineered hoods, local exhaust  systems,  wet  sawing, and special single-point




 cutting tools (OSHA 1986b).   The control  most widely used for all phases of




 pipe production is local exhaust ventilation  (LEV) with hooding.  LEV  draws




 airborne asbestos fibers away from workers; the fibers are collected in




 baghouses.  These systems are used during bag opening, fiber introduction, dry




 mixing, willowing (fluffing), dry  finishing of both pipes and fittings, and




 scrap grinding for recycling.




     There is minimal generation of asbestos dust  for A/C pipe production




 processes other than dry finishing.   Fiber  emitted between  fiber introduction




 and wet mixing from material handling equipment  (e.g., screw conveyors and




 bucket  elevators) is controlled by using  continuous exhaust  and maintaining




 negative pressure within pneumatic conveying  systems.  The  fluffing, blending,




 and dry mixing of the asbestos fiber  (with  cement, silica,  and  scrap)  take




place automatically in closed blending tanks  maintained under negative




pressure by LEV.   These engineering controls  reduce the possibility of




exposure at these processing stages  (OSHA 1986b).  No  special control




equipment  is  typically used during pipe formation, air curing,  and steam




curing since  the  asbestos fibers in the pipe  become bound in a  cement  mixture,




thus restricting  fiber release.
                                    - 64 -

-------
    Good housekeeping practices are also important during pipe formation.




These practices include the use of wet vacuum machines and squeegees,  instead




of brooms, to clean up spills of slurry that could become a source of




emissions.  Central "vacuum systems with flexible-hose pickups are also used at




work stations vulnerable to asbestos spillage (e.g., bag opening).  Other




prevalent work practices include minimum handling of asbestos bags prior to




bag opening and fiber introduction, as well as not recirculating local exhaust




air.




    As expressed by the values exhibited in Table 10, no exposure level change




is expected during the fiber introduction activities.  A small decrease may




occur for pipe forming and "other" activities via the previously discussed




controls and practices.  The primary reduction, however, occurs due to the




expected decrease in exposure during finishing operations.




        3.  Asbestos-Cement Sheet




           a.  Product Descriptions




           Asbestos-cement (A/C) sheet includes flat sheet, corrugated sheet,




and roofing and siding shingles.  Corrugated A/C sheet is no longer produced




in the United States; however, it is imported (ICF Market Survey 1986-1987).




    Flat A/C sheet is used primarily in the construction industry as wall




lining in factories and agricultural buildings, fire-resistant walls,  curtain




walls, industrial partitions, soffit material (covering the underside of




structural components), and decorative paneling in both exterior and interior




applications.  In addition, it is used for special applications in cooling




towers, and as laboratory table tops and fume hoods, electrical equipment




mounting panels, and as a component of vaults, ovens, safes, heaters,  and




boilers.  Flat A/C sheet is found in schools as well as in residential




construction.
                                    - 65 -

-------
    A/C shingles are used as siding and roofing for both residential and




commercial buildings.  Roofing shingles account for approximately 70 percent




of the A/C shingle market, and siding shingles take the remainder.




    Production of A/C sheet products has declined significantly since 1981




(ICF Market Survey 1986-1987).  In 1985, 181,808 squares of A/C sheet materi




were manufactured in the U.S. (one square - 100 square feet),  consuming




4,481.8 tons of asbestos (ICF Market Survey 1986-1987).  Only ,three percent




this total production is flat A/C sheet, the remaining 97 percent is A/C




shingles.  An additional 3,395.7 squares of flat A/C sheet were imported in




1985.  The numbers of asbestos workers are 12 and 11 for flat sheet and




shingle primary processing, respectively (ICF Market Survey 1986-1987).  Ther




are currently no secondary manufacturers of A/C sheet or shingle in the U.S.




(ICF Market Survey 1986-1987).




           b.  Process Descriptions




           The production process for flat sheet and shingles is very similar




A/C sheet products are made from a mixture of Portland cement and asbestos




fiber.  Finely ground inert filler such as silica and pigments are sometimes




included.  In general, asbestos-cement sheet contains between 15 and 40




percent asbestos fiber.  However, for curing in short-time periods, a general




formulation of 12 to 25 percent asbestos, 45 to 54 percent cement,  and 30 to




40 percent silica is used (ICF 1984).  Similar to the A/C pipe process, the




raw materials are mixed with water to form a wet slurry of asbestos, cement,




and silica.  The slurry is then picked up by a screen-cylinder mold and




transferred to a felt conveyor.  The felt is then dewatered, passed to a




mandrel, and wound to the desired thickness.  After achieving the required




thickness, a cut is made across the width of the sheet.  The sheet is manually




peeled off the rotating mandrel onto a transfer roll conveyor.   The sheet is




then cured, and processed through embossing rollers and trimming/cutting




                                    -  66  -

-------
wheels.  Roofing and siding shingles are cut from A/C sheets; therefore,  the




composition and manufacturing process used are the same.




           c.  Exposure Profile




           The total population is based on the ICF Market Survey (1986-1987).




The distribution of the total number of workers exposed into specified job




categories is based on the 1981 TSCA Section 8(a) data (Hendrickson and Doria




1983).  The data presented in Table 11 are based on 1985 figures, excluding




companies that no longer produce asbestos products.




    Although the manufacturing processes of A/C sheet are very similar to




those of A/C pipe, asbestos dust is apparently less well-controlled in A/C




sheet plants, resulting in higher exposure readings.  This is exemplified by




the total pre-0.2 f/cc PEL level for primary manufacturing shown in Table 11.




The values for all primary job categories depicted in Table 11 are very high,




with two out of three of the job categories having exposure values above 1.0




f/cc.  Since the monitoring data are aggregated for A/C sheet and A/C




shingles, the exposure values and other information for each job category are




assumed to be equivalent for both products.  However, since the distribution




of the total populations into job categories developed from 1981 TSCA Section




8(a) data (Hendrickson and Doria 1983) for primary manufacturing is different




for each product, the weighted average exposures are different for the two A/C




sheet products.




    With the widespread employment of exposure controls, the post-0.2 f/cc PEL




exposures are assumed to fall at or below the 0.2 f/cc level.  The controls in




use for A/C sheet production, like the process itself, are similar to those in




A/C pipe manufacture.  LEV is the primary control technology.  It could be




used in such operations as bag opening, fiber introduction, punching,




trimming, and brushing, as well as along the forming line during rolling,




veneering, cutting, and embossing.  The willowing and mixing operations are




                                    -  67 -

-------
                                                       Table 11.  Exposure Profile for A/C Sheet -- Primary Manufacture
00
8-Hour TWA Exposure (f/cc)

Product Job Category8

Population
Pre-0.2 f/cc PEL0
Geometric Mean
Arithmetic Mean
Post-0 t2
Geometric Mean
f/cc PEL"
Arithmetic Mean
Duration
(hr/day)"
Frequency
< days/year )f
Primary Manufacturing.
A/C Flat Sheet Fiber Introduction
Proceeding
Other
Total
A/C Shingle Fiber Introduction
Processing
Other
Total
2
7
3
12
1
7
3
11
1.054 (7)
0.408 (6)
1.028 (6)
0.671
1.054 (7)
0.408 (6)
1.028 (6)
0.636
1.364
0.577
l.}43
0.850
1.364
0.577
},143
0.803
0.200
0.166
0.200
0.160
0.200
0.166
0.200
0.178
0.200
0.172
0.200
0.184
0.200
0.172
0.200
0.182
8
8
8.
8
8
8
8
8
250
250
250
250
250
250
250
250
"job categories are based on a concise categorization of the job tltlea.

Estimation of total population is described in the text.  Population by Job category is estimated froa the 1981 TSCA Section 8(a) data (Hendrlckson and Doria
1983).

°These values represent geometric and arithmetic means of the raw 8-hour THA exposure data.  The number of data points is given in parentheses.  The values
corresponding to the total population are calculated aa weighted averages baaed on the number of workers exposed in each job category.  Since the monitoring
data are aggregated for A/C sheet and A/C shingles, the Job category exposure values are assumed to be equivalent for both products.

 These post-0.2 f/cc PEL exposure values are calculated directly from the monitoring data.  Each 8-hour THA exposure value that la above 0.2 f/cc ia reduced
to exactly 0.2 f/cc.  Data that are already at or below this value remain unchanged.  The values corresponding to the total populations are determined in the
seme manner as the pre-0.2 f/cc PEL total exposure value.

*The effective duration of exposure ia 8 hours/day in all cases.  Where no duration ia provided, 8 hours/day is assumed.   Exposures for less than 8 hours are
converted to 8-hour TWAs, assuming sero exposure during periods when the worker is not handling asbestos.

 Frequency refers to the number of days annually that the workers are performing a teak involving potential exposure to eabestoa.   The frequency is assumed to
be 250 daya/year.

Sources:  OSHA 1987, ICF Market Survey 1986-1987.

-------
controlled with the use of negative pressure conveying systems, closed

vessels, and isolation in a restricted access area.   Host finishing operations

can be controlled by adopting tools equipped with exhaust systems or wet spray

devices, as well as LEV.  During sanding, however, supplemental respiratory

protection is needed to prevent excessive exposure to asbestos dust.

Housekeeping and work practices also greatly reduce the amount of airborne

dust and, as in all industries, provide very effective dust control.

    The fiber introduction, and dry and wet mixing stages of A/C sheet produc-

tion are virtually equivalent to the steps used in A/C pipe production; so

exposures in these processes can be kept low.  The advanced processing steps

are also similar to those of A/C pipe with relatively little fiber dust gener-

ation.  As with pipe manufacturing, the highest and most difficult exposures

to control occur after curing, during the mechanical finishing operations.

        4.  Friction Products

           a.  Product Descriptions

           The friction product category includes the following products:

           Drum brake linings;
           Disc brake pads (light and medium);
           Disc brake pads (heavy);
           Brake blocks;
           Clutch facings;
           Automatic transmission components; and
           Friction materials.

Each of these products is discussed briefly below.

               (1)  nfVTfl firake Linings

               Most light and medium vehicles (i.e., passenger cars  and light

trucks) are equipped with drum brakes on the  rear wheels.  Drum brake  linings

are pieces of molded friction material attached to curved pieces of metal

(brake  shoes), aligned to  a  cylinder or  drum  which rotates with  the wheel.

The vehicle is stopped by  friction between  the drum  and the  lined  brake shoes
                                    - 69 -

-------
when pressure is applied to the brake.   (The  term  "brake shoe" is sometimes




used to refer to the lined piece, not the metal piece alone.)




               (2)  Disc Brake Pads  (Light and Medium)




               Disc brakes are generally used on the front wheels of light an




medium vehicles, although some cars, primarily high-performance types, have




disc brakes on all four wheels.  In  disc brakes, two metal pieces, lined with




friction material called disc brake  pads, straddle the rotor,, or disc, in the




center of the vehicle's wheel.  Friction between the disc and the brake pad




slows or stops the vehicle when the  brakes are applied.




     In addition to disc brake pads containing asbestos fibers as a component




of the friction material, there are  some semi-metallic disc brake pads which




have asbestos-containing underlayers between  the plate and the pad (other




 semi-metallic pads have no underlayers or non-asbestos underlayers).  These




products are not  considered asbestos disc brake pads.




                (3)   Disc Brake Pads  (Heavy)




               Asbestos disc brake pads are rarely used for heavy vehicles.




Disc brakes for heavy  vehicles are similar to those described for light and




medium vehicles.  Asbestos disc brake pads for heavy vehicles differ from




those for  light and  medium vehicles  primarily in size; therefore, information




in the previous section on disc brake pads is applicable to disc brake pads



for heavy vehicles as  well.




                (4)   Brake Blocks




               Brake blocks are drum brake linings for heavy vehicles (trucks




and off-highway vehicles).  Drum brakes for heavy vehicles are similar to the




drum brakes described  for light and  medium vehicles.  Brake blocks are




generally considerably larger than automobile drum brake linings (0.75 inches




or greater in thickness, compared to about 0.3 inches for automobile drum




brake linings), but their operation  is similar.




                                    - 70 -

-------
               (5)  Clutch Facings

               Clutch facings are friction materials attached to the steel

disc in a manual transmission vehicle.  Pressure plates pressing against the

clutch facings keep the gears in position when the clutch is engaged.  Clutch

facings may be molded products, usually reinforced with yarn or wire, or woven

products (see Section B.5, Textiles, of this chapter).  Molded clutch facings

are more widely used than woven facings; woven clutch facings are more likely

to be used in high performance vehicles.

               (6)  Automatic Transmission Components

               Automatic transmission components are used in fluid-filled

automatic transmissions for automobiles.  The friction material is a type of

paper used to line a metal band around the gears and metal rings (friction

clutch plates) that fit over the gears.  The entire assembly is fluid-filled;

the fluid absorbs the heat of gear-changing and also any wear debris from the

friction material.

               (7)  Friction Materials

               This category covers a variety of non-automotive friction

materials, including the following  (Scott 1984):

        •  Woven band brakes (see Section B.5, Textiles) for
           heavy-duty use (e.g., in oil well drilling and
           construction);

        •  Molded brake and clutch  materials for industrial use;

        •  Railroad brakes;

        •  Paper-type friction materials for fluid-filled clutches for
           industrial use; and

        •  Other friction materials for industrial use and for
           equipment such as lawn mowers, washing machines, and
           machine tools.
                                    - 71 -

-------
           b.  Process Descriptions

               (1)  Primary Manufacture

                    Prvil Pra^e Linings.  Asbestos drum brake linings contain

resins, fillers, and other additives, as well as asbestos fibers.  Asbestos

fibers make up approximately 40 to 50 percent by weight of the linings (Cha

and Carter 1982).  Primary manufacture of drum brake linings usually is

carried out by a wet-mix process, consisting of the following steps, with

possible variations by manufacturers:

        •  Mixing of fibers, solid and liquid resins, property
           modifiers, and solvents;

        •  Extrusion or rolling;

        •  Molding and curing using heat and pressure;

        •  Finishing by grinding and drilling; and

        •  Packaging of finished product.

    The degree of automation in manufacturing may be highly variable (ICF

1986a).

                    Disc Brake Pads (Light. Medium and Heavy Vehicles).

Asbestos disc brake pads, like drum brake linings, are molded products

containing resins, fillers, and other additives, as well as asbestos fibers.

These products are approximately 40 to 50 percent asbestos by weight (Cha and

Carter 1982).  A dry-mix process is usually used in their manufacture;  the

basic steps in this process are as follows:

        •  Mixing of fibers, dry resins, and property modifiers;

        •  Molding and curing using heat and pressure;

        •  Finishing by grinding and drilling; and

        •  Packaging of finished product.

    The degree of automation of these steps may vary considerably from

manufacturer to manufacturer (ICF 1986a).


                                    -  72  -

-------
                    Brake Blocks.  Asbestos brake blocks are primarily molded




products usually produced by a dry-mix process, as described for disc brake




pads.  Some woven brake blocks are also produced (see Section B.5, Textiles).




                    Clutch Facings.  Molded clutch facings are usually made by




a dry mix process, as described for disc brake pads.  The mix may be molded




around strand or wire reinforcements.  Woven clutch facings are made by




running a strand through a wet mix to pick up the wet mixture and weaving




after drying.  The woven product is then hot-pressed, cured, and ground, as




other wet-mix products.




                    Automatic Transmission Components.  Paper for automatic




transmission components is manufactured by conventional paper-making




processes.  Raw materials are pulped and fed to a continuous papermaking




machine, and finished paper is cut from the machine.




                    Friction Materials.  Woven band brakes are produced from




asbestos cord, possibly reinforced with wire, which is passed through a




wet-mix to pick up resin and modifiers and then woven into tapes.  The tapes




are heated to partially cure the resin, and then may be further cured to form




flexible rolls or rigid segments (Jacko and Rhee 1978).




    Primary manufacture of molded brake and clutch materials is probably




similar to manufacture of molded automotive products.




    Railroad brakes are manufactured by methods similar to those described for




drum brake linings.




    Manufacture of paper-type industrial friction products is similar to the




manufacture of automatic transmission components for automobiles.




    The manufacturing methods for other friction materials, which may include




molded, woven, and paper-type products, vary depending on the type and




application of the material.
                                    -  73 -

-------
                (2)   ff^ondarv Manufacture




                     ppim Brake Linings.  Secondary processing of  drum brake




 linings  may  be  of several  types.  Some processors install new brake  linings




 into brake assemblies  for  new vehicles; others  repackage linings  for sale  as




 replacement  parts in the aftermarket.  Generally, neither of these secondary




 processes involves any grinding or drilling of  the brake linings; these




 operations are  usually performed by primary manufacturers.  Anothe-  distinct




 type of  secondary processing is automotive rebuilding.  Rebuilders receive




 used, worn brake linings,  attached to the shoes.  The old linings are removed




 from the shoes, the  shoes  are cleaned by abrasion, and new linings are




 attached.  The  rebuilt shoes with linings are then packaged and sold for the




 aftermarket.




                     Disc Brake Pads (Light. Medjyn) find Heavy Vehicles).




 Secondary processing of disc brake pads includes installation of  the pads  in*--




 new  brake assemblies and repackaging for sale to the aftermarket




 addition, rebuilders remove the worn pads from  the metal plates, clean tr._




 plates, and attach new pads for resale.




                    Brake Blocks.   Secondary processing for brake blocks




 includes installation in new brake assemblies and automotive rebuilding, as




 described for drum brake linings and disc brake pads.  There may       be some




 repackaging of brake blocks.




                    Clutch Facings.   Secondary processing of clutch facings is




 similar to secondary processing of the automotive friction products previously




 discussed.   Clutch facings may be  rebuilt,  as described for other automotive




 products.




                    Automatic Transmission Components.   Secondary




manufacturing for automatic transmission components  includes  installation c.
                                   - 74  -

-------
the componer.-._  .- new vehicle transmissions.  Rebuilding of transmissions is




another type  ; secondary processing.




                    Friction Materials.  Secondary processing of woven band



brakes may involve cutting the band material and installing it in new




equipment.  Rebuilding may include removing worn material and replacing it




with new material.  Secondary manufacturing processes for other types of




friction materials may be similar to those described for automotive -friction




products; little information is available on non-automotive products, however.




           c.  Production and Employment




           Table 12 presents total production and asbestos consumption and




estimates of total employees exposed to asbestos for each type of friction




product.  The data presented are based on 1985 figures, excluding companies




that no longer produce asbestos products.  The estimates of total employment




are based on figures from the ICF Market Survey (1986-1987), supplemented by




figures from the ICF Exposure Survey (1986-1987).  For producers who did not




provide employment data, we estimated employment from the average number of




workers per ton of asbestos consumed for each product type, based on available




data.  In the case of disc brake pads for heavy vehicles, the average for disc




brake pads for light and medium vehicles was used.  For some products, there




were wide variations between companies in number of employees per ton of




asbestos consumed.  This may indicate differences in the method of reporting




employees; for some companies, total employment may be reported, while for




others, only employees directly exposed to asbestos may be included.  In




addition, the level of automation may vary considerably from company to




company.  Employment figures presented in the table should be considered




estimates only.




    There are a number of secondary processors of friction products.  The ICF




Market Survey (1986-1987) includes automobile importers in the listing of




                                    - 75  -

-------
     Table 12.  Production and Employment for Primary Manufacture
                         of Friction Products



Drum Brake Linings
Disc Brake Pads (Light
and Medium Vehicles)
Disc Brake Pads (Heavy
Vehicles)
Brake Blocks
Clutch Facings
Automatic Transmission
Components
Friction Materials
Total
Production*
(thousand
pieces)
91,922.4
58,633.5

146.9

3,752.7
7,237.1
55.5

8,521.4

Asbestos
Consumption0
(tons)
12,645.5
6,323.3

110.1

2,337.0
993.5
0.2

1,523.0
Total
Population
'Exposed
to Asbestos0
1,115
815

14

232
239
1

-•
a!985 production,  excluding production by companies no longer producing
asbestos products.

°1985 consumption,  excluding consumption by companies no longer
producing asbestos products.

Population based on ICF Market Survey (1986-1987) (1985 employment,
excluding companies no longer producing asbestos products), ICF
Exposure Survey (1986-1987), and estimates described in the text for
companies with no  population figures reported.

Sources:  ICF Market Survey 1986-1987, ICF Exposure Survey 198*
          and ICF estimates.
                                 -  76  -

-------
secondary manuracturers; and, therefore, it is difficult to obtain from this




source an estimate of the number of employees that actually carry out




secondary processing.  In addition, the major emphasis of the Market Survey




was on primary manufacturing.  Because the survey did not focus on secondary




processing, the number of secondary processors may be underestimated.  Also,




brake rebuilders, who perform a type of secondary processing, were not




included in the survey.  Because of these uncertainties, the IGF Market Survey




data on worker populations were not used in this analysis.  To estimate the




potentially exposed worker population for secondary manufacture of each




friction product, populations from the 1981 TSCA Section 8(a) data (RTI 1985)




were adjusted to 1985 populations by multiplying by the ratio of 1985 primary




production (ICF Market Study 1986-1987) to 1981 primary production (EPA 1986b,




midpoints were used).  Based on this estimating technique, less than one




full-time employee would be involved in secondary manufacture of disc brakes




for heavy vehicles and automotive transmission components.  The total




population for secondary manufacture of all friction products was estimated to




be 2,295.




    Population estimates by OSHA (1986b) for automotive rebuilding were used




in this analysis; OSHA (1986b) identified 181 plants with 4,750 workers for




brake rebuilding.  The 1981 TSCA Section 8(a) data do not include data




specifically for automotive rebuilding.  It is not known, therefore, whether




or not automotive rebuilding is included with secondary friction product




manufacturing data.  If automotive rebuilding is included, there may be some




double counting using this method of estimating population.




           d.  Exposure Profile




           To determine exposure levels during primary and secondary




manufacture of friction products, data from the ICF Exposure Survey




(1986-1987), NIOSH reports, and OSHA surveys were obtained.  We categorized




                                    -  77 - •

-------
the jobs performed by each worker monitored into one of several job



categories.  Geometric and arithmetic means were calculated using the raw data




for each job category.



               (1)  Primary Manufacture



               Table 13 presents a summary of the data on primary manufacture



of friction products.  Processes used for the manufacture of the various types



of molded friction products are similar, and it was generally not possible to



categorize the available exposure data by product; therefore, we have assumed



that the geometric and arithmetic mean exposures for each job category apply



to each of the molded products.  For woven friction products, refer to Sectionr



B.5 for exposure levels.  For automatic transmission components, which are



paper products, exposures for specialty paper products apply (see Section



B.I); there were no exposure data specific to automatic transmission



components.  (For friction materials, both molded products and paper products



may be included.  No breakdown by type is available; we assumed for simplicity!




that the exposure values for molded products apply.)



    The total number of workers exposed to asbestos during manufacture



type of friction product was estimated as described earlier  (see Ta-



To estimate the number of workers by job category for each product, we used



the 1981 TSCA Section 8(a) data  (Hendrickson and Doria 1983).



    As Table 13 shows, the geometric mean, TWA exposures  for three job



categories (fiber introduction and mixing, forming, and scrap and waste



disposal)  exceed  the 0.2 f/cc PEL.  The arithmetic mean,  TWA exposures are at




or above the 0.2  f/cc PEL for all job categories.



    Based  on exposure surveys received, it is likely that worker



categories with exposures exceeding 0.2 f/cc are  required to wear  .       -^-



Some companies may  require respirators  for workers  in all job categories  VIC?



Exposure Survey 1986-1987).  Engineering controls currently  used in  the



                                    - 78 -

-------
                                                   Table 13.   Exposure  Profile  for  Friction  Products -- Primary Manufacture
Product Job Category" Population


Pre-0.2
Geometric
Mean

f/cc
8 -Hour
PELC
Arithmetic
TWA Exposure

Mean

(f/cc)
Post-0.2
Geometric Mean

f/cc PELd
Arithmetic Mean
Duration
(hr/day)'
Frequency
(days /year)
f
        Drum Brake Lining
        Disc Brake Pad*  (Light
           and Medium Vehicles)
        Disc Brake Pad*
           (Heavy Vehicles)
vO
         Brake Blocks
        Clutch Facings
Fiber Introduction/      212         0.251  (71)        0.487
  Mixing
Forming                  301         0.226 (301)        0.482
Finiehing                301         0.174 (336)        0.357
Inspection/Packing       101         0.104  (63)        0.404
Scrap/Haste Handling     100         0.240  (18)        0.515
Other                    100         0.060  (68)        0.197
     Total             1,115         0.192              0.420

Fiber Introduction/      180         0.251  (71)        0.487
  Mixing
Forming
Finishing
Inspection/Packing
Scrap/Wast* Handling
Other
     Total

Fiber Introduction/
  Mixing
Forming
Finishing
Inspection/Packing
Scrap/Haste Handling
Other
     Total

Fiber Introduction/       30         0.251  (71)        0.487
  Mixing
Forming                   60         0.226 (301)        0.482
Finishing                 60         0.174 (336)        0.357
Inspection/Pecking        27         0.104  (63)        0.404
Scrap/Waste Handling      27         0.240  (18)        0.515
Other   •               	28         0.060  (68)        9.197
     Totel               232         0.183              0.411

Fiber Introduction/       72         0.251  (71)        0.487
  Mixing
Forming                   82         0.226 (301)        0.482
Finishing                 82         0.174 (336)        0.357
Inspection/Packing         1         0.104  (63)        0.404
Scrap/Waste Handling       1         0.240  (18)        0.515
Other                  	I         0.060  (68)        0.197
     Total               239         0.215              0.439
236
236
55
54
_5i
815
2
4
3
2
2
_1
14
0.226 (301)
0.174 (336)
0.104 (63)
0.240 (18)
0.060 (68)
0.198
0.251 (71)
0.226 (301)
0.174 (336)
0.104 (63)
0.240 (18)
0.060 (68)
0.191
0.482
0.357
0.404
0.515
0.197
0.425
0.487
0.482
0.357
0.404
0.515
0.197
0.429
0.122

0.122
0.115
0.066
0.110
0.049
0.107

0.122

0.122
0.115
0.066
0.110
0.049
0.111

0.122

0.122
0.115
0.066
0.110
0.049
0.106

0.122

0.122
0.115
0.066
0.110
0.049
0.103

0.122

0.122
0.115
0.066
0.110
0.049
0.119
0.157

0.161
0.152
0.117
0.159
0.099
0.148

0.157

0.161
0.152
0.117
0.159
0.099
0.150

0.157

0.161
0.1S2
0.117
0.159
0.099
0.148

0.157

0.161
0.152
0.117
0.159
0.099
0.145

0.157

0.161
0.152
0.117
0.159
0.099
0.156
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8 >
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  8
                                                                                                                                                  &
                                                                                                                                                  8
8
8
8
8
8
8

8

250
8
8
8
8
8
250

250
250
Z5U
250
250
250

250

250
250
250
250
222
250

250

250
250
250
250
250
250

250

250
250
250
250
250
250

250
                                                                                                                                                              250
                                                                                                                                                              250
                                                                                                                                                              250
                                                                                                                                                              250
                                                                                                                                                              250

-------
                                                                              Table 13 (Continued)
00

O
8-Hour TWA Exposure ft/cal
Product
Automatic Transmission
Components

Friction Materials






Job Category*
(Saa Paper Products
Specialty Paper,
Section B.I)
Fiber Introduction/
Mixing
Forming
Finishing
Inspection/Packing
Scrap/Hast* Handling
Other
Total
Population
1

99

34
33
7
7
187
Pre-0.2 f/cc PELC
Geometric Mean
0.039

0.251 (71)

0.226 (301)
0.174 (336)
0.104 (63)
0.240 (18)
0.060 (68)
0.220
Arithmetic Mean
0.061

0.4S7

0.482
0.357
0.404
0.515
0.197
0.450
Post-0.2
Geometric Mean
0.037

0.122

0.122
0.115
0.066
0.110
0.049
0.115
f/cc PEL0
Arithmetic Mean
0.054

0.157

0.161
0.152
0.117
0.159
0.099
0.153
Duration
(hr/day)*
8 i

8

8
8
8
8
8
8
Frequency
(days/year)1
198

250

?sn
7«n
250
250
250
250
"Based on categorisation of job titles of workers monitored for exposure.



                                             and ICF estimates aa described in the text.   Population by Job category la estimated from the 1981 TSCA Section
          «r.i« F"«7t»T7ttl0|'?? >rittaitl? "•"• °f th6 rwf 8-hour ™ «*>•«• <»•«•••   »• nu«*>« «« ** point, i. given in parentheses.   A. value.
         corresponding to the total populations are calculated as weighted averages baaed on the number of workers exposed in each Job category.



         "Geometric and arithmetic .wans of raw data after reducing all value* exceeding the 0.2 f/cc PEL to 0.2 f/cc.   Data that are already at  or below this value

         remain unchanged.   The values corresponding to the total populations are determined in the same manner ea the pre-0.2 f/cc PEL total exposure  values.
         ™ff«      ™  °* Tr" is 8 h™"/0-* »» •" «=•«•»•  Mh«. no duration 1. provided.  8 hours/day Is  assumed.   Exposure, for 1... than 8  hours  are
         converted to 8-hour THAa,  assuming zero exposure during periods when the worker  is not handling asbestos.
         ».'««:,'  ™f*r'  to th* nuBlb»r  ot day*  annually that  the workers  are performing  a task  involving potential  exposure  to  asbestos.   The  frequency  is  assumed  to
         be 250 days/year unless  data for specific  Job  types  indicate  otherwise.                                                                 frequency  is  assumed  to



         Sources:  ICF Exposure Survey 1986-1987; ICF Market  Survey  1986-1987;  ICF  estimates; HIOSH  1982c, 1984a,  and 1985b; OSHA 1987.

-------
primary manufacture of friction products vary from company to company.   All




companies use exhaust ventilation.  Some companies use automatic bag-opening




machines; manual bag-opening is used by others.  The level of automation




varies (IGF 1986b) .  The use of automatic bag opening, emptying, and disposal




equipment can reduce asbestos exposure to well below the 0.2 f/cc PEL during




fiber introduction and mixing operations, according to a report by NIOSH




(1984a).                                                            '•




    To estimate the projected post-0.2 f/cc PEL exposures, all exposure levels




exceeding 0.2 f/cc were reduced to 0.2 f/cc.  As Table 13 shows, this reduces




the geometric and arithmetic mean TWA exposures to well below the 0.2 f/cc PEL




for all job categories.  Many companies report exposures below 0.2 f/cc in all




job categories; therefore, it is likely that exposures can be reduced to this




level by means of  improved engineering controls and housekeeping for most




workers.  Respiratory protection may be required during some operations.




               (2)  Secondary Manufacture




               Table 14 summarizes exposures for secondary processing of




friction products, including secondary manufacturing and brake rebuilding.




For secondary manufacturing, 1981 TSCA Section 8(a) data on population for




each friction product were adjusted to 1985 populations, as described above,




to obtain estimates of total population by product type.  To estimate the




population by product type for automotive rebuilding,  total employment from




OSHA (1986b) was used, and the ratio of population for rebuilding of each




friction product to total population for all friction product rebuilding was




assumed to be  the  same as determined for secondary manufacturing of friction




products from  the  8(a) data.  The  1981 TSCA Section  8(a) data (EPA  n.d.) were




used to distribute the total population by job category  for each friction




product  for both secondary manufacture and rebuilding.
                                    - 81 -

-------
                                            Table  1*.  Exposure  Profile  for  Friction  Product*  —  Secondary Manufacture and Rebuilding
oo
ho
6-Hour TWA , xuosure (f/cel
Product
Job Category*
Population
Pre-0.2
Geometric Mean
f/cc PELC
Arithmetic Mean

Post-0.2 f/cc PEL0
Geometric Mean
Arithmetic Mean
Duration
(hr/day)*
Frequency
(days/year)1
Secondary Manufacturing
Drum Brake Lining


Disc Brake Pads (Light
and Medium Vehicles)

Brake Blocks


Clutch Facings


Friction Materials


Processing
Other
Total
Processing
Other
Total
Processing
Other
Total
Processing
Other
total
Processing
Other
Total
949
988
1,937
166
101
267
a
8
16
36
	 12
48
25
2
27
0.016
0,011
0.014
0.018
0.011
0.015
0.018
0.011
0.015
0.018
0.011
0.016
0.016
0.011
0.017
(18)
(2)

(IB)
(2)

(16)
(2)

(18)
(2)

(18)
(?)

Automotive
Drum Brake Lining


Disc Brake Pads (Light
and Medium Vehicles)

Brake Blocks


Clutch Facings


Friction Mater


Processing
Other
Total
Processing
Other
Total
Processing
Other
Total
Processing
Other
Total
Processing
Other
Total
1,964
2.045
4.009
342
209
SSI
16
	 IZ
33
75
	 22
100
52
	 5
57
0.013
0.014
0.014
0.013
0.014
0.013
0.013
0.014
0.014
0.013
0.014
0.013
013
014
013
(44)
(?)

(44)
(?)

(44)
(8)

(44)
(8)

(44)
(8)

0.190
0.017
0.102
0.190
0.017
0.125
0.190
0.017
0.104
0.190
0.017
0.147
0.190
0.017
0.177
Rebuilding
0.033
0.063
0.058
0.033
0.063
0.037
0.053
0.063
0.038
0.053
0.063
0.056
f )


0.015
0.011
0.013
0.015
0.011
0.013
0.015
0.011
0.013
0.013
0.011
0.014
0.015
0.011
0.015

0.013
0.014
0.014
0.013
0.014
0.013
0.013
0.014
0.014
0.013
0.014
0.013
0.013
0.014
0.013
0.080
0.017
0.048
0.060
0.017
0.056
0.080
0.017
0.049
0.080
0.017
0.064
0.060
0.017
0.075

0.043
0.037
0.051
0.043
0.057
0.030
0.045
0.057
0.051
0.0.45.
0.057
0.048
0.045
0.057
0.046
8
8,
8
8
8
8
8

8
6
J
8
8
8.
8

6
8.
8
8
fi
8
8
fi
6
8
8
8
8
8
8
250
250
**^v
250
250
250
XIXpS,
250
250
8250
jcnm
250
250
250
SSJSiX.
250
250
250
xzs
250

250
230
sxs
250
250
250
XUUL
250
250
230
250
2SO
250
250
250
250
250

-------
                                                                          Table  1*  (Continued)


      "Based on categorization of job titles of workers monitored for exposure.

       Secondary manufacturing population totals estimated from 1981 TSCA Section 8(a)  populations  (RTI  1985)  adjusted to 1985 populations by multiplying by the
      ratio of 1985 primary production (ICF Market Study 1986-1987)  to 1981  primary  production (EPA 1986b,  midpoints used).   Automotive rebuilding population total
      from OSHA (1986b).   Population by job category estimated from 1981 TSCA Section 8(a) data (EPA n.d.).

      °These values present geometric and arithmetic means of the raw 8-hour TWA exposure data.   The number of data points is given in parentheses.   The values
      corresponding to the total populations are calculated as weighted averages based  on the number of  workers exposed in each Job category.

       Geometric and arithmetic means of raw data after reducing all values  exceeding the 0.2 f/cc  PEL to 0.2  f/cc.   Data that are already at or  below this  value
      remain unchanged.   The values corresponding to the total populations are determined in the same manner as the pre-0.2  f/cc PEL total exposure values.

      "The effective duration of exposure is 8 hours/day in all cases.   Where no duration is provided, 8 hours/day is assumed.   Exposures  for less  than 8 hours are
      converted to 8-hour THAs, assuming zero exposure during periods when the worker is not handling asbestos.

      frequency refers  to the number of days annually that the workers are  performing  a task involving  potential  exposure to asbestos.  The  frequency is assumed to
      be 250 days/year unless data for specific Job types indicate otherwise.  Some  companies may not do asbestos-related work  full-time,  but few data are available
      on frequency to make a better estimate.

      'included with automotive rebuilding because the rebuilding process would  be similar to rebuilding of  automotive products.

      Sources:  ICF Exposure Survey 1986-1987;  ICF Market Survey 1986-1987;  ICF  estimates;  HIOSH 1982d,  198
-------
    Disc brakes for heavy vehicles and automatic transmission components are




not included in Table 14 because production is very small compared to other




friction products; and, therefore, less than one full-time employee would be




exposed.




    Some secondary manufacturers of friction products may carry out




manufacturing processes on a part-time basis (e.g., see N10SH 1985c); others,




however, operate on a full-time basis (ICF Exposure Survey 1986-.1987).  We do




not have enough data to determine whether the average frequency of exposure is




less than 250 days per year, so we used the 250 days/year default value.




    As Table 14 shows, geometric mean, 8-hour TWA exposures are well under the




0.2 f/cc PEL for secondary manufacturing of friction products and brake




rebuilding, although in individual cases, exposures exceeded the PEL.  The




arithmetic mean exposure for "processing" is close to the PEL because several




raw data exposure points exceed 1 f/cc.




    Respirators or dust masks may be required for some job categories in some




facilities (ICF Exposure Survey 1986-1987).  Engineering controls such as dust




collection systems and local exhaust ventilation may also be used,  -"




companies may not use these controls  (ICF Exposure Survey 1986-i:




1982d).



    The post-0.2  f/cc PEL was estimated by reducing all exposure levels




exceeding 0.2 f/cc to 0.2 f/cc and determining new geometric and arithmetic




mean.   It is likely that most facilities will be able to achieve the 0.2 f/cc




level  through engineering controls and improved housekeeping, since in most




cases,  exposures  are already below this level.




         5.  Textiles




           a.  Product Descriptions




           Asbestos fiber is used to provide resistance to fire and heat in




 textile products.  Historically,  asbestos textiles have been used in a wide




                                    - 84 -

-------
range of produc-   jut many of the traditional products are no longer in

production.  Subs'itute fibers have taken up the bulk of the market for

electrical and thermal insulation, fire resistant materials, and protective

clothing.

    The products that continue to be made with asbestos textiles are:

        •  friction products;
        •  packings and gaskets; and
        •  specialty products.

Friction materials account for the majority of the asbestos textile products

made from asbestos yarn and include woven brake blocks and clutch facings.

Typically woven brake blocks are used in large industrial equipment such as

oil well drilling rigs and cranes.

    Packings and gaskets made from asbestos textiles include both yarn and

cloth products.  Asbestos yarn products, braid and rope, are used extensively

in pump and valve packings and as seals for oven doors, boilers, and furnaces.

Asbestos cloth is used to manufacture manhole and flange gaskets as well as

seals in incinerator (hot-air) piping, nuclear power plant cooling water

towers, and distillation columns.

    Specialty products are also made from asbestos cloth and asbestos yarn.

It is often difficult to find substitute materials for these specialized

applications, but products of this type are usually produced in relatively

small volume (less than 5,000 pounds).

    Some products made from asbestos textiles that can be classified as

specialty products are:

        •  mantles for gas lanterns (yarn);

        •  wicks for catalytic heaters  (yarn);

        •  rotor vanes for pumps and compressors used in air tools
           (cloth);
                                    - 85 -

-------
        •  ring type seals for valve' and compressor plat^     .m);  and

        •  bearings for high temperature applications requ..ing water
           lubrication (cloth).

    Primary manufacturing of asbestos textile materials produced 1,125 tons of

asbestos thread in 1985, consuming 558 tons of asbestos fiber and potentially

exposing 78 workers to asbestos (ICF Market Survey 1986-1987).  An additional

527.6 tons of asbestos textile mixture was imported in 1985.  Secondary

fabricators of asbestos thread, yarn, cloth, and other textiles' employ

approximately 208 asbestos workers (ICF Market Survey 1986-1987).

           b.  Process Descriptions

               (1)  Primary Manufacture

               Asbestos textile products can be classified into two main

product areas: asbestos cloth and asbestos yarn.  While there are a wide

variety of asbestos textile applications, each product will be assigned to one

of these two major categories for the purposes of simplicity.

    The manufacture of asbestos textiles shares characteristics common f~

textile production.  In particular, asbestos textiles are produc

asbestos fibers by standard textile production techniques.  Asbestos ri^-

can be blended with other types of fibers and often carrier yarns are added to

asbestos yarns to give the resulting textile products added tensile strength.

The manner in which asbestos fibers are processed into asbestos yarn and cloth

products includes the following steps:  milled fiber storage, pneumatic

grinding (addition of fibers), cording and combing, matting (mat and roving),

spinning (for yarn or cord), and braiding or weaving.

    There are two basic variations employed in asbestos textile manufacturing

the conventional and wet processes.  The methods employed in producing

asbestos textiles by each of these methods is described below, although most

textiles are manufactured by the conventional process.


                                    -  86  -

-------
    In the conventional process,  raw asbestos fibers of various grades are

blended and mixed, with the composition of the blends and fixing of the

formulation being governed by the fiber characteristics, manufacturing and

finished product requirements,  and intended use.   The different grades of

asbestos fiber received are moved to the rear of the blender where they are

mixed according to the requirements specified for the finished product.  The

selected fiber sizes then enter a hopper.  When filled, the hopper delivers

the blended material to the carding operation.

    The carding operation combs the fibers into a relatively parallel

arrangement called a fiber mat.  This mat is pressed and layered into a lap

consisting of alternating perpendicular arrangements of fiber mats.  The lap

is then separated into thin, continuous ribbons called roving. Cotton, rayon

or other material may be added at this stage to strengthen the roving.

    Roving, which has been mechanically twisted and spun to give it tensile

strength, forms a single yarn.   This yarn may be twisted with other single

yarns, wire or other material to produce plied yarn which can be coated to
                                         /•
produce thread or treated yarns.   Plied yarns may be woven to produce fabric,

sleeving, or tape.  Alternately,  plied yarns may be twisted to form wicking

and twisted rope, or braided to form braided rope or sleeving.

    The conventional process employs either the dry or damp method.  These two

methods are identical except that during the damp method, the yarn is

moistened either by contact with water on a roller or by a mist spray.  This

moistening of the yarns reduces the amount of fiber that becomes airborne and

also aids the processing of fibers into yarn.

    The wet process is based on forming single filament fibers by extrusion.

The process consists of making a gelatinous mixture of fine asbestos fiber in

water with a volatile dispersant.  The mass is then extruded through small

dies to form asbestos thread.  The extruded thread is spun to form yarn which

                                    -  87  -

-------
is fabricated into various plied yarn products as in the conventional process.



The textile products formed using this wet technique tend to hold asbestos



fibers better than those produced by the conventional processes,  thus reducing



workplace fiber levels, but the yarn formed has the disadvantage  of poor




adsorption and impregnation characteristics.



               (2)  Secondary Manufacture




               Asbestos cloth and yarns are used by secondary textile



manufacturers to produce fire and heat resistant materials and electrical




insulation.  Specific processes primarily involve cutting of asbestos cloths




and sewing with asbestos thread.




           c.  Exposure Profile



           The total population is based on the ICF Market Survey (1986-1987)




The distribution of the total number of workers exposed into specified job




categories is based on the 1981 TSCA Section 8(a) data (Hendrickson and Doria




1983).



    For secondary manufacturing of textiles, the total population -




disaggregated into job categories.  This total population value is based on




data from the ICF Market Survey (1986-1987).  Some of the companies surveyed



did not disclose the number of workers exposed.  The worker population for




these companies is estimated by using the ratio of population to amount of




asbestos mixture consumed for the other companies.



    Exposure levels for asbestos textile operations are exhibited in Table 15



These exposure levels are based on the raw monitoring data.  The .   "_  3



indicate that levels are highest during the carding and twisting operations




and lowest during preparation and winding.  Spinning and waste handling also




have high potential for producing airborne asbestos.



    Asbestos textiles are manufactured using either conventional dry or damp



processing, or a wet process in which the raw asbestos is dumped directly int<




                                    -  88 -

-------
                                                                    Table IS.  Exposure Profile for Textiles
00
vo
8-Hour TWA Exposure (f/cc)
Product Job Category*
Population
Pre-0.2 f/cc PELC
Geometric Mean
Arithmetic Mean
Post-0.2
Geometric Mean
f/cc PEL0
Arithmetic Mean
Duration
(hr/day)a
Frequency
(days/year)f
Primary Manufacturing
Textiles Preparation
Carding
Twisting
Winding
Total
16
30
30
	 2
78
0.168 (7)
0.402 (48)
0.613 (39)
0.217 (17)
0.430
0.241
0.565
0.730
0.260
0.554
0.126
0.184
0.197
0.161
0.177
0.149
0.187
0.198
0.169
0.183
8
8
8
a
8
240
240
240
240
240
Secondary Manufacturing
Textiles H/A
208
0.251 (12)
0.420
0.146
0.157
B
0250
"job categories are based on a concise categorization of the job titles.

 Estimation of total population is described in the text.   Population by Job category is estimated from the 1981 TSCA Section 8(a) data (Bendrickson and Dorla
1983).

°These values represent geometric and arithmetic means of the raw 8-hour TWA exposure data.   The number of data points is given in parentheses.   The value
corresponding to the total population is calculated as a weighted average based on the number of workers exposed in each Job category.

 These post-0.2 f/cc PEL exposure values are calculated directly from the raw monitoring data.  Each 8-hour THA exposure value that is  above 0.2 f/cc is
reduced to exactly 0.2 f/cc.  Oats that are already at or below this value remain unchanged.   The value corresponding to the total population is determined in
the same manner as the pre-0.2 f/cc PEL total exposure value.

*The effective duration of exposure is 8 hours/day in all cases.  Where no duration is provided, 8 hours/day is assumed.  Exposures for less than 8 hours are
converted to 8-hour TWAs, assuming tero exposure during periods when the worker is not handling asbestos.

'frequency refers to the number of days annually that the workers are performing a task Involving potential exposure to asbestos.   The  frequency is assumed to
be 240 days/year for primary manufacturing.  Secondary manufacturing is assumed to have a 250 day/year frequency.

Sources:  ICF Exposure Survey 1986-1987; OSHA 1987; HIOSH 1980a, 1980b; ICF Market Survey 1986-1987.

-------
a slurry tank with water and chemicals; the process utilized can greatly




influence the exposure levels.  Dust release is much greater during




conventional dry processing than damp processing, in which a roller or mist




spray is used to apply moisture to the yarn.




    Exposure is assumed to be reduced to post-0.2 f/cc PEL levels via




application of available engineering controls.  As in other industries, the




primary engineering control in textile plants using dry methods is LEV.  It is




typically used at such stages as bag opening, fiber introduction, willowing,




blending, carding, and winding.  The carding operation may be isolated in




addition to having exhaust ventilation.  Either LEV or general exhaust




ventilation with humidification may be employed during the spinning operation.




Nevertheless,  respirators, are usually required in the spinning and carding




areas.




    Wet processing, such as a damp loom for weaving, can great1%




exposure.  But this is not viable in many cases because wet processing  -




changes the nature of the textile.  Apparently, most dry asbestos textile




manufacturers using state-of-the art controls can achieve 0.2 f/cc exposure




levels.  Dry mechanical manufacturing, however, seems to present difficulties,




and certain operations (e.g., carding, spinning and twisting) require




supplemental respirator usage to reduce excessively high exposure levels.




    The pre-0.2 f/cc PEL exposure level determined for secondary manufacturing




of textiles is substantially lower than for primary manufacturing; however,




further reduction is also required for this sector to meet tht       '   level.




This reduction is assumed to occur by techniques similar to those       -.




primary manufacturing.  The dry mechanical operations involved in this sector




have high potential for exposures above 0.2 f/cc.  The specific processes of




concern are cutting of asbestos fibers and sewing with asbestos thread.
                                    -  90  -

-------
        6.  Sh?st: Gaskets and Packing




           a.  -roduct Descriptions




           Gaskets are devices which are used to seal the space between two




non-moving (usually metal) surfaces (OSHA 1986b).  They are used to prevent




leaks of liquids or gases in a system, often under conditions of high




temperature and pressure.  Because of these requirements, asbestos is a




desirable ingredient in gaskets due to its heat and chemical resistance as




well as its strength (Chemical Engineering 1986).  Asbestos gaskets are used




in large pipes in oil refineries and nuclear power plants,  and in automotive




engine blocks and oil pans.




    Packing materials are very similar to gaskets except that packings are




placed between a moving and a non-moving surface (Freimanis 1981).  Packings




are used in pumps, turbines, shafts, and other similar places.  The chemical




process and electric power industries are among the greatest users of packings




(Chemical Engineering 1986).




    According to the ICF Market Survey (1986-1987), primary producers of sheet




gaskets produced 3.5 million square yards of gasketing material in 1985,




consuming 5,301 tons of asbestos and exposing a total of 163 employees to




asbestos (ICF Market Survey 1986-1987).




    Packing manufacturers produced 4 tons of asbestos packing material in




1985, consuming 2.5 tons of asbestos fiber and exposing about 5 employees to




asbestos (ICF Market Survey 1986-1987).




    Secondary processors of sheet gaskets have an estimated total of more than




878 employees exposed to asbestos.  Total population exposed to asbestos was




estimated from the number of employees reported in the ICF Market Survey




(1986-1987) and an estimate based on the number of employees per square yard




of asbestos mixture consumed for those companies for which employment was not




reported.  Due to gaps in the ICF Market Study (1986-1987) data, the estimated




                                    -  91 -

-------
population TOT secondary manufacture of sheet Baskets        ^er limit .   1981


TSCA Section 8 (a) data are  not adequate to rerlne


    Secondary -processing of asbestos pac


(ICF Market Surwey 198fi-19£7>. -

                                  .*«
           "b.  Process Deseri-Dtlona


                         ary ManufacCugB
               The Tsixtare-nsefl

         ^       *         -* -af-^.j**        „ ,,- .
asbestos.  The asbestos Is -sized -wttS TSU
   metals^ anfl an organic  soTveot (e.^.J tetrsZluoroedkylme  (lFt.)l» __ aad
       together into a rather homogeneous jsheet (TrelmaniB 19B1).  'Vhen the
        .      7      -••*-

       i mLx and thickness axe .achieved,  the sliaet Is cut to « predetermined
length and vidrh  (ICF 15B7) .  flsjs/ »«nirf»rn«Sqg


Tollitig^iMd vejvijy «f £flsMX nifif gr fafTr  ^s^hcgro
                                   --   .••*•-    '."
hag-slitters in acate pracesses  (JUUftl 19B4c) .


    Manufacture of asbestos packings is siailar to that of gaskets.   «-'


is rolled into a honogenops mixture with plastic, rubber, neta1


other materials.  'Aiitaaateff^raidiss jfi^<»oeefte are used to strengthen the

                                 *              .               ~
entire paoV^Tip mat-»->H m\    rmnptHs-j f jam if nenaT^y 7Q £0 *99 SArcent or aoce


asbestos, depending xm the "heat resistance needed (Treljnanis T991).


    Prijiary manufacturers  of sheet gaskets and packing may pre-form the


Materials to fit a predetermined shape needed "by the user.." This; process1 Is
                               ~T - ;     - •                •"•••-.*••_•<..•

"frequently carried out l»y  seconAary nuniifai
                Secondary manufactnrrrs -of sheet gaskets and packing cut and


shape sheet gasketing and packing material to .meet user specific


uses such as automotive examine blocks, oil pans, and certain kinds of v<^


In some  rases ,  a gasket can be irot into Its -proper shape using a sharp knife


(OSHA 1966b) .


                                     -  92 -

-------
               "xposure Profile




               (1)  Primary Manufacture




               Table 16 presents geometric and arithmetic mean, 8-hour TWA




exposure levels and populations exposed by job category for primary




manufacture of sheet gaskets and packing.  Population by job category was




estimated from the 1981 TSCA Section 8(a) data (Hendrickson and Doria 1983).




    There are few exposure data available specific to packing; because




manufacturing processes are similar for sheet gaskets and packing, the mean




exposure levels for sheet gaskets are assumed to apply to packing manufacture




as well.




    From the limited data available, the geometric mean, 8-hour TWA exposure




level is well below the 0.2 f/cc PEL for all job categories, although




exposures may exceed the PEL in individual cases.  The arithmetic mean




exposure exceeds the PEL for the "other" job category.  Some companies may




require use of respirators for some operations, and there are  likely to be




engineering controls in place (ICF Exposure Survey 1986-1987,  NIOSH 1984c).




    To estimate exposure levels under the 0.2 f/cc PEL, all reported exposures




exceeding 0.2 f/cc were reduced to the 0.2 f/cc level.  New geometric and




arithmetic means are determined for each job category.  There  were few




reported exposures exceeding 0.2 f/cc for sheet gasket and packing




manufacture.  The highest exposures reported are for unspecified  "operators",




placed in the "other" job category; it is not known what type  of  operation was




involved at the time the monitoring was performed.




    Since in most cases current exposures are well below 0.2 f/cc, it is




likely that all facilities can achieve exposure levels below 0.2  f/cc by use




of engineering controls and good housekeeping practices.
                                    - 93 -

-------
                                                            Table 16.   Exposure Profile for Sheet Gaskets and Packings
vo
8-Hour TWA Exposure (f/cc)
Product
Job Category*
Population
Pre-0.2 f/cc PELC
Geometric Mean
Arithmetic Mean
Post-0.2
Geometric Mean
f/cc PEL"
Arithmetic Mean
Duration
(nr/day)*
Frequency
(days/year)1
Primary Manufacturing
Sheet Gaskets





Packing





Fiber Introduction
Mixing
Molding
Finishing
Other
Total
Fiber Introduction
Mixing
Molding
Finishing
Other
Total
62
11
10
38
42
163
1
1
1
1
I
5
0.044 (9)
0.062 (8)
0.022 (7)
0.015 (2)
0.035 (12)
0.035
0.044 (9)
0.062 (8)
0.022 (7)
0.015 (2)
0,035 (121
0.036
0.081
0.164
0.045
0.116
0.363
0.165
0.061
0.164
0.045
0.116
0.363
0.154
0.043
0.049
0.022
0.014
0.024
0.030
0.043
0.049
0.022
0.014
0.024
0.030
0.076
0.087
0.045
0.101
0.072
0.080
0.076
0.087
0.045
0.101
0.072
0.076
8'
8
8
8
8
8
8
8
8
8
8
8
250
250
250
250
250
250
250
250
250
250
250
250
Secondary Manufacturing
Sheet Gaskets


Packing


Processing
Other
Total
Processing
Other
Total
571+
307+
878+
16
9
25
0.061 (136)
0.127 (2)
0.084
0.061 (136)
0.127 (2)
0.085
0.128
0 . 128
0.128
0.128
0.128
0.128
0.055
0.127
0.080
0.055
0.127
0.081
0.094
0.128
0.106
0.094
0.128
0.106
8
fi
8
8
8
8
250
250
250
250
250
250
          'Based on categorization of job titles of workers monitored for exposure.

          bTotal population based on ICF Market Survey and ICF estimates as described in the text.  Population by Job category is estimated from the 1981 TSCA Section
          8(a) data (Bendrickson and Dorla 1983, EPA n.d.).

          cThese values present geometric and arithmetic means of the raw 8-hour TWA exposure data.  The number of date points is given in parentheses.  The values
          corresponding to the total populations are calculated as weighted averages based on the ntxofaer of workers exposed in each job category.

          dGeometric and arithmetic means of raw data after reducing all values exceeding the 0.2 f/cc PEL to 0.2 f/cc.  Data that are already at or below this valui.
          remain unchanged.  The values corresponding to the total population are determined in the same manner aa the pre-0.2 f/cc PEL total exposure values.

          eThe effective duration of exposure is 8 hours/day in all cases.  Where no duration <« provided, 8 hours/day is assumed.  Exposures for less than 8 hours are
          converted to 8-hour TWA*, assuming zero exposure during periods when the worker I      handling asbestos.
           Frequency refers to t h > .mitber of days annually that the workers are performing
          be 250 days/year tin let  •   « for specific Job types indicate otherwise.  Frequen<

          Sources:  ICF Exposuiu Sutvey 1986-1987, ICF Market Survey 1986-1987, ICF eatim*
volving potential exposure to asbestos.  The frequency la assumed to
  •omewhat from company to company.

    4c, OSHA 1987.

-------
                !2)  Secondary Manufacture

               Exposure data for secondary manufacturing of sheet gaskets and

packings are also presented in Table 16.  Total population was estimated from

the ICF Market Survey  (1986-1987) as discussed earlier; population by job

category was estimated from the 1981 TSCA Section 8(a) data (EPA n.d.).

    Geometric and arithmetic mean exposures are well below the 0.2 f/cc PEL

for secondary manufacturing.  The highest reported exposures are for

unspecified laborers (placed in the other category); it is not clear what

operations are involved in these cases.

    No mandatory use of respirators was reported by respondents to the ICF

Exposure Survey (1986-1987); however, engineering controls such as closed

process units in some  areas and exhaust ventilation were reported.

    To estimate exposures under the 0.2 f/cc PEL, all exposure levels

exceeding 0.2 f/cc were reduced to the 0.2 f/cc level and new geometric and

arithmetic means are determined.  The estimated, post-0.2 f/cc exposure level

is not significantly lower than the current level.  Many companies report

exposure levels well below 0.2 f/cc; it is likely that all companies can

achieve these levels through the use of engineering controls and improved

housekeeping practices.

        7.  Roof Coatings. Non-Roofing Coatings. Missile Liner and
            Sealant Tape

           a.  Product Descriptions

           Roof coatings, non-roofing coatings, missile liner, and sealant

tape are produced using similar methods and are, therefore, grouped together

in this section.

    Roof coatings containing asbestos are used to provide water,  weather,

heat, and corrosion resistance to roofs.  Roof coatings may also be applied to

sidewalks, concrete foundations, and brick.  These coatings are asphalt-based


                                    - 95  -

-------
and thinned with solvents;  asbestos fibers,  which matte up 7 to 10 percent of




the coating by weight (ICF 1987), provide body and reinforcing properties.




Roofing, tile, and flashing cements, also asphalt-based and solvent-thinned,




are used to repair roofs, seal around projections such as vent pipes and




chimneys, and bond horizontal and vertical surfaces.   These products contain




15 to 20 percent asbestos by weight (ICF 1987).




    Coatings for non-roofing applications include the following:




        •  Specialty coatings;




        •  Epoxy-based adhesives for automotive and construction use;




        •  Caulks and joint compounds;



        •  Sealants for equipment and building construction; and




        •  Vehicle undercoatings.




    Missile liner is a rubber compound used to coat Che interior of rocke-




motors.  Its main function is to insulate the outer casing of t'-




the intense heat generated while the rocket fuel is burned (ICF 1987).




    Sealant tape is made from a semi-liquid mixture of butyl rubber and




asbestos.  Asbestos usually constitutes 20 percent by weight of the mixture.




On exposure to air, the sealant solidifies forming a rubber tape.  The tape i




used for sealing building windows, automotive windshields, and mobile home




windows, and in the manufacture of parts for the aerospace industry and




insulated glass (ICF 1987).



    Table 17 shows production, asbestos consumed, and employees exposed to




asbestos for the coating-type products discussed in this sect!       ~ number




of employees exposed to asbestos for these products was estimated from the




total reported in the ICF Market Survey (1986-1987) and an estimate based on




the average number of employees per ton of asbestos consumed for those




companies for which employment was not reported.
                                    -  96 -

-------
      Table  17.   Production and  Employment for Coating Type  Products
  Product
 Total Production3
  Asbestos
Consumption^
  (tons)
   Total
 Population
  Exposed
to Asbestos0
Roof Coatings

Non-Roofing Coatings

Missile Liner

Sealant Tape
57.2 million gallons     22,215

8.1 million gallons       2,083

4.6 thousand tons           699

423 million feet          1,660
                   438

                   497

                   380

                   134
a!985 production, excluding production by companies no longer producing
asbestos products.

°1985 consumption, excluding consumption by companies no longer producing
asbestos products.

Population based on ICF Market Survey (1986-1987) (1985 employment,
excluding companies no longer producing asbestos products) and estimates
described in the text for companies with no population figures reported.

Sources:  ICF Market Survey 1986-1987, ICF estimates.
                                  -  97  -

-------
    In addition, secondary producers of non-roofing -eatings consumed a tota




of 25.4 gallons of asbestos mixture in 1985; 76 employees were exposed to




asbestos (ICF Market Survey 1986-1987).




           b.  Process Descriptions




           Asbestos roof coatings and cements are normally manufactured in




batches.  Asbestos fibers are dumped into a hopper or fluffing machine and tl




fibers are fluffed; the fibers are then transferred to a mixer and combined




with other dry ingredients.  Asphalt and solvents are added, and all the




ingredients are mixed.  The product is then packed in containers.  Bags of




fiber are generally opened and added manually (ICF 1986b), although some




companies may use bag-opening machines (NIOSH 1983c); the rest of the proces-




is automated (ICF 1986b).




    Manufacture of non-roofing coatings, missile liners, and sealant




carried out using processes similar to that described for roc




    Coatings and related products are usually shipped directly for use; u._




is little likelihood of secondary processing even though two secondary




processors were identified in the ICF Market Survey (1986-1987).




           c.  Exposure Profile



           Table 18 summarizes geometric and arithmetic mean exposure levels




for the primary manufacture of coatings and related products.  In most cases,




the exposure data could not be broken down by specific product, and the data




are limited for missile liner and sealant tape.  Since these products are




manufactured using similar processes, data related to all typ         -<-ngs




are combined.  The geometric and arithmetic mean exposure levels -        job




category are assumed to apply to all coating products.  There are no e.\.    :?




data for secondary processing of coatings; therefore, secondary processing  is




not included in the table.  In addition, it is not clear what type of




operations would be involved in secondary processing.




                                    -  98 -

-------
                                               Table 18.  Exposure Profile for Coatings -- Primary Manufacture
8 -Bout TWA Exposure (f/cc)
Product Job Category*
Roof Coatings Fiber, Preparation/
Fiber Introduction
Mixing
Other
Total
Non-Roof Coatings Fiber Preparation/
Fiber Introduction
Mixing
Other
Total
Missile Liner Fiber Preparation/
Fiber Introduction
Mixing
Other
Total
i
Sealant Tape Fiber Preparation/
vo Fiber Introduction
Mixing
Other
Total
Population
298

79
_il
438
338

20
139
497
2S8

15
12Z
380

91

5
_38
134
Pre-0.2
Geometric Mean
0.040 (12)

0.047 (50)
0.082 (34)
0.047
0.040 (12)

0.047 (50)
0.082 (34}
0.052
0.040 (12)

0.047 (50)
0.082 (34)
0.052

0.040 (12)

0.047 (50)
0.082 (34)
0.052 x
f/cc PELC
Arithmetic Mean
0.331

0.490
0.946
0.445
0.331

0.490
0.946
0.509
0.331

0.490
0.946
0.510

0.331

0.490
0.946
0.511
Post-0.2
Geometric Mean
0.024

0.029
0.04B
0.028
0.024

0.029
0.048
0.031
0.024

0.029
0.048
0.031

0.024

0.029
0.048
0.031
f/cc PEL"
Arithmetic Mean
0.103

0.104
0.114
0.105
0.103

0.104
0.114
0.106
0.103

0.104
0.114
0.106

0.103

0.104
0.114
0.106
Duration
(hr/day)e
8

8
a
8
8

8
8
8
8

8
fi
8

8

8
fi
8
Frequency
( days /y ear )f
250

250
250
250
200

200
200
200
200

200
200
200

200

200
200
200
"Based on categorization of job titles of workers monitored for exposure.

 Total population baaed on ICF Market Survey and ICF estimates as described in the text.  Population by job category la estimated from the 1981 TSCA Section
8(a) data (Hendrickson and Doria 1983).

°The values present geometric and arithmetic means of the raw 8-hour TWA exposure data.  The number of data points is given in parentheses.  The values
corresponding to the total populations are calculated as weighted averages based on the number of workers exposed in each Job category.

 Geometric and arithmetic means of raw data after reducing all values exceeding the 0.2 f/cc PEL to 0.2 f/cc.  Data that are already at or below this value
remain unchanged.  The values corresponding to the total populations are determined in the same manner as the pre-0.2 f/cc PEL total exposure values.

"The effective duration of exposure is 8 hours/day in all cases.  Mhere no duration is provided, 8 hours/day la assumed.  Exposures for less than 8 hours are
converted to 8-hour THAa, aasumlng zero exposure during periods when the worker is not handling asbestos.

fFrequency refers to the number, of days annually that the workers are performing a task involving potential exposure to asbestos.  The frequency is assumed to
be 250 days/year unless data for specific job types indicate otherwise.  It is likely that non-roofing coatings,  missile liner,  and sealant tape are produced
on a part-time basis, but insufficient data are available to determine average frequencies.

Sources:  ICF Exposure Survey 1986-1987; ICF Market Survey 1986-1987; ICF estimates; NIOSH 1981b, 1982e,  1983c, 1984d,  and 1984e; OSHA 1987.

-------
    The total number of workers exposed to asbestos during manufacture of eac




type of coating product was estimated from the ICF Market Survey (1986-1987)




and asbestos consumption,  as described earlier.   The number of workers per jo




category was estimated from the 1981 TSCA Section 8(a) data (Hendrickson and




Doria 1983).




    For non-roofing coatings and sealant tape, it is likely, based on the ICF




Exposure Survey (1986-1987) that workers are not exposed full-time.




Respondents reported production of these products 10 to 208 days per year.




There were insufficient data to determine an average frequency, but it seems




likely that it is less than 200 days per year.  It is likely that this is als




true for missile liner, based on the high ratio of employees to productio*-




asbestos consumption (see  Table 17).




    Few data are available on frequency for production of roof coaL*tlgs;




survey respondent reported production frequency of 100 days per year  (ICF




Exposure Survey 1986-1987).  Roof coatings are produced in larger quantities




than the other types of coatings (ICF Market Survey 1986-1987); therefore,




production is more likely to be full-time, although some companies are




probably part-time producers.



    Geometric mean, 8-hour TWA exposure levels are below the 0.2 f/cc PEL for




all job categories for coating products, although individual exposures may




exceed the PEL.  The arithmetic mean exposure levels for each job category




exceed the PEL due to high exposures reported.



    Mandatory use of respirators or masks during coatings manuf-	 —••.




reported by all respondents to the ICF Exposure Survey (1986-198/




processes were also reported.  OSHA (1986a) reports that fluffing and m~—.,




operations are kept under negative pressure, and housekeeping around  these




operations is continuous.
                                   - 100 -

-------
    Projectec., ccst-0.2 f/cc PEL exposures were estimated by reducing all




exposures thac exceeded 0.2 f/cc to the level of the PEL.




        8.  Asbestos-Reinforced Plastics




           a.  Product Description




           Asbestos fibers are added to a wide variety of plastics to improve




stiffness, strength, processability, and provide corrosion and heat




resistance.  The asbestos acts as both a mineral and fibrous binder carrier




with reinforcing action.  Asbestos-reinforced plastics are used in the




electronic, automotive, and printing industries with applications in




appliances, utensils, tools, automobiles (in the ignition, transmission, and




wiring systems), wiring devices, electrical switch gears, and communication




and electronics equipment.  Other uses include floor tiles, packing, and




gaskets, but these products are not discussed in this section.




    The amount of asbestos used in plastics varies widely from product to




product, but it is relatively small compared to other product categories.




Asbestos is now used only with phenolic resins, although traditionally




asbestos has been used with a variety of other plastic resins including




polyester, urea, diallyl phthalate, vinyl epoxy, polypropylene, and nylon.




    Primary manufacturers produced 4,250 tons of asbestos-reinforced plastics




in 1985.  These companies consumed 636.1 short tons of asbestos fiber in 1985




(ICF Market Survey 1986-1987).  In addition, 127.5 tons of reinforced plastic




was imported in 1985, and several companies perform secondary fabrication of




reinforced plastics (ICF Market Survey 1986-1987).  There are 138 and 456




asbestos workers involved in the primary and secondary manufacturing of




plastics, respectively.
                                   - 101 -

-------
           b.  Process Descriptions




               (1)  Primary Manufacture




               The processing of asbestos-reinforced plastics may be




implemented by either a primary or a secondary manufacturer.  Primary




processors introduce raw asbestos fibers and dry mix it with catalysts and




other additives.  The mixture is transferred via sealed containers or vacuum




conveyors to resin formation equipment.  Various types of equipment can be




used in this processing step, but in general the resin is formed by either




extrusion or internally heated Banbury mixing.  Both processes result in




pellets, powders, or some similar product which is known as "preform."  The




resin is drummed and employed in subsequent processes to form the end produr-




About 30 percent of primary processors fabricate the end products on-s-'




(Versar 1983).  The remaining 70 percent of the plastic preforrr




packaged and sold to secondary manufacturers (Versar 1983); these ta,.	




are usually remote from the primary processing facilities.




               (2)  Secondary Manufacturing




               At secondary manufacturing locations asbestos-reinforced




plastic, or preform, is received by railcar or truck and is transferred to




storage areas and forming process areas by forklifts.  The actual formation oi




an end-product is accomplished by rerneIting the preform and then submitting it




to rolling, molding, stamping, or pressing.  Dust control equipment used




during these steps includes exhaust hoods leading to the fabric filters and




partial enclosure of process vacuuming equipment.




    After forming the desired product, it is then cured.  This is normally




accomplished in an enclosed area furnished with a ventilating system.  When




air-curing processes are used, hoods and local enclosures are provided.  Final




product characteristics, such as strength and stiffness, are partially




controlled by the time and temperature conditions during curing.




                                   - 102 -

-------
    After curing, the product is finished by drilling, grinding, machining or




sawing, depending on the end use of the product.  Asbestos dust is released




when the plastic products are finished.  Hand tools employ local exhaust




systems vented through a central fabric filter.  Larger, stationary machines




also utilize local exhausts near the surface being finished, these are often




supplemented with hoods over the finishing machines themselves.  In summary,




the main steps in secondary manufacturing consist of (1) resin receiving and




storage, (2) resin introduction, (3) forming, (4) curing, and (5) finishing.




As a final step, the asbestos-reinforced plastic product is packaged and




shipped to the consumer.




           c.  Exposure Profile




           Due to the absence of monitoring data for primary manufacturing of




asbestos-reinforced plastics, the estimates for pre- and post-0.2 f/cc PEL




exposure levels are taken directly from OSHA's final regulatory impact




analysis (OSHA 1986b).   The total exposed population is obtained from the ICF




Market Study (1986-1987), and the population distribution among job categories




is based on 1981 TSCA Section 8(a) data (Hendrickson and Doria 1983).  Table




19 presents the exposure profile for this sector.  Worker exposures during the




manufacture of asbestos-reinforced plastics are below 0.2 f/cc for several




operation categories.  High exposure areas occur, however, in dry finishing




operations.




    Exposure data are available for secondary manufacturing from two NIOSH




studies.  Two personal samples collected for asbestos in the first NIOSH study




showed concentrations of 0.034 f/cc and 0.042 f/cc for operators working at




the standard injection molding line and at the HK standard compression molding




line, respectively (NIOSH 1983a).  While these samples are below the new OSHA




PEL (i.e., 0.2 f/cc), higher exposures to asbestos may occur during the
                                   - 103 -

-------
                                                 Table 19.   Exposure Profile for Asbestos-Reinforced  Plastics


Product


Job Category*


Population

Pre-0.2
Geometric Mean
8-Hour
f/cc PELC
Arithmetic
TWA Exposure (f/cc)
Post-0.2
Mean Geometric Mean

f/cc PELC
Arithmetic Mean

Duration
(hr/day)d

Frequency
(days/year)*
Primary Manufacturing
Asbestos-Reinforced
Plastics






Fiber Introduction
Wet-Mechanical
Operations
Dry-Mechanic al
Operations/
Finishing
Other
Total
35
4

12


_§Z
138
N/A
N/A

N/A


N/A
N/A
0.288
0.007

0.355


0.400
0.356
N/A
N/A

N/A


N/A
N/A
0.048
0.007

0.145


0.060
0.063
8
8

8


a
8
250
250

250


250
250
Asbsstos-Relnforced Plastics N/A
                                                    456
 Secondary Manufacturing

0.120 (3)          0.423
                                                                                                       0.066
0.092
                                                                                                                                                      250
N/A - Not Available.

*Job categories are based on the categorication presented by OSHA (1986b).

bThe total population is based on the ICF Market Survey (1986-1987).  The distribution of the total number of workers exposed into specified job categories is
based on the 1981 TSCA Section 8(a) data (Bendrlckson and Ooria 1983).

cThe values for primary manufacturing are taken directly from OSHA's final regulatory impact analysis (OSRA 1986b).   Ho other exposure information was
available for primary manufacturing of asbestos reinforced plastics.  OSHA'a data are presented as "means" which are assumed to be arithmetic means.   The
values corresponding to the total populations are calculated as weighted averages based on the number of workers exposed in each Job category.  The exposure
value for secondary manufacturing is determined from monitoring data.

 The duration of exposure is assumed to be 8 hours/day In all cases.

frequency refers to the number of days annually that the workers are performing a task involving potential exposure to asbestos.   The frequency Is assumed to
be 250 days/year for all Job categories.

Sources:  NIOSH 1983a, 1984f; OSHA 1986b; ICF Market Survey 1986-1987.

-------
filling of --  allon drums with asbestos-reinforced plastic pellets  (NIOSH




1983a).   This cask is performed only about once per week.




    In the second NIOSH study,  one personal and four area samples were




collected for airborne asbestos; this personal sample indicated an exposure of




1.2 f/cc for the grinder operator (NIOSH 1984f).




    Asbestos-reinforced plastics have a relatively small asbestos content




compared to the other ingredients involved in resin forming.  Manual bag




opening methods have normally been used to handle this quantity of dry




asbestos.  Central ventilating systems with exhaust hoods are the prevalent




controls used in the bag opening area.  Some large manufacturers use limited




enclosure of their areas for better control of exhaust air flow.




    Control equipment utilized at the dry blending step includes exhaust




hoods, local process exhaust equipment, and partial enclosures to control air




flow and minimize asbestos dust exposure in surrounding areas.  Housekeeping




and maintenance practices are identical to those used in fiber introduction




areas; they range from manual floor and equipment sweeping to central




vacuum-cleaning systems and mobile sweeper/vacuum machines.




    Engineering controls at the mixing and forming stages generally include




exhaust hoods and partial enclosure of process equipment.  Housekeeping and




maintenance practices are similar to those employed in the earlier processes.




Curing usually requires an enclosed area furnished with a ventilating system.




When air curing is involved in this process, hoods and local enclosures should




be provided.



    Following the curing process, the product is finished.  This could involve




sawing,  drilling, machining, and grinding, depending on the end use of the




product.  Asbestos dust is released when the plastic products undergo these




finishing processes.  Hand and portable tools are normally supplied with local




exhaust systems connected to the central ventilation/collection system.  LEV




                                   - 105 -

-------
is employed near the finishing surface of large, stationary machines; and, a




times, hoods are installed over the finishing machines as a supplementary




control.  Where large amounts of dust are released, area or machine partial




enclosures can be used.  Housekeeping practices are very important at these




final processing stages and are similar to those employed in previous dry




processing steps.




    Based on the exposure data presented in Table 19 and technologies




currently available in other primary sectors, projected asbestos exposure




levels are expected to be below the current PEL of 0.2 f/cc.  Wider use of




respirators is anticipated for dry-mechanical processes and the high




exposures, shown in Table 19, should be greatly reduced.  The 8-hour TWA




exposure exhibited for secondary fabricating in Table 19 further indicatp-




feasibility of achieving low levels in this sector.  The projected




value for secondary fabricating is substantially lower than the




PEL value due to the expected reduction via the technologies discussed abo,.




        9.  Miscellaneous Products




           a.  Production Data




           There is some information available regarding the status of




companies manufacturing miscellaneous asbestos products.  Asbestos production




data derived from the ICF Market Survey (1986-1987) indicate that three




asbestos products not previously categorized are manufactured in the U.S.




These products have been identified as acetylene cylinders,  battery




separators, and arc chutes.   Although production data does exist for these




products, no occupational exposure data has been located; nevertheless, it is




important to acknowledge the existence of these uncategorized,  y        ...~.iy




produced asbestos products which result in potential occupational exp.   •> to




asbestos.
                                   - 106 -

-------
    The primar  manufacture of these three products consumed 493.6 short tons




of asbestos in 1985 and potentially exposed 371 asbestos workers (ICF Market




Survey 1986-1987).




    Table 20 summarizes the production data for each one of these products.




There are no known secondary manufacturers or importers of these products.




           b.  Product Descriptions




               (1)  Filler for Acetylene Cylinders




               Asbestos is used to produce a sponge-like filler that is placed




in acetylene cylinders.  The filler holds the liquefied acetylene gas




(acetone) in suspension in the steel cylinder and pulls the acetone up through




the tank as the gas is released through the oxyacetylene torch.  The torch is




used to weld or cut metal and is sometimes used as an illuminant gas.  The




filler also acts as an insulator that offers fire protection in case the




oxidation of the acetylene becomes uncontrollable.  The desirable properties




of asbestos in this function include its porosity, heat resistance,




anti-corrosiveness and its strength as a binding agent (ICF Market Survey




1986-1987).




               (2)  Battery Separators




               In very specialized aerospace applications, asbestos functions




as an insulator and separator between the negative and positive terminals of a




fuel cell/battery.  The porous nature of the 100 percent woven-asbestos




material allows it to absorb the liquids used in fuel cells and batteries.




The liquids used in these fuel cells/batteries are highly corrosive and reach




high temperatures in this application.  The properties of asbestos that are




desirable in this function are its porosity, heat resistance, anti-




corrosiveness, strength and dielectric strength (ICF Market Survey 1986-1987).
                                   - 107 -

-------
        Table 20.  Production Data for Miscellaneous Products
                      Production Volume
  Asbestos
   Fiber
  Consumed
(short tons)
  Number
.of Workers
Potentially
  Exposed
to Asbestos
Acetylene Cylinders    308,121 pieces

Battery Separators       2,046 Ibs

Arc Chutesa                900 pieces
   479.1

     1.0

    13.5
   162

   207

     2
       processors of asbestos arc chutes manufacture plastic arc
chutes that have been classified in the asbestos-reinforced plastics
category.  Generally, the plastic arc chutes are smaller and are not
able to withstand as high a temperature (above 1500°F) as the
ceramic arc chutes.  The plastic arc chutes are used in smaller
electric motors, often in the automotive and appliance industries
(ICF Market Survey 1986-1987).

Source:  ICF Market Survey 1986-1987.
                               - 108 -

-------
               (3)  Arc Chutes




               Ceramic arc chutes containing asbestos are used to guide the




electric arc in motor starter units in electric generating plants.  The




asbestos is used in the arc chutes for its strength, heat resistance,  and




dielectric strength.




    C.  Chlorine Manufacture (Asbestos Diaphragyn Cells)




        1.  Process Description




        Asbestos diaphragm cells represent one type of electrolytic cell




employed in the chlor-alkali industry for the production of chlorine and other




primary products such as caustic soda.  There are presently three types of




electrolytic cells in commercial use:  asbestos diaphragm cells, mercury




cells, and membrane cells.  As of mid-1985, 77 percent of the U.S. industry's




total installed chlorine production capacity was in electrolytic cells




equipped with asbestos diaphragms (Chlorine Institute 1986a).  Mercury cell




technology accounted for about 17 percent of the capacity, while membrane cell




technology, recently developed, accounted for 2 to 3 percent (Chlorine




Institute 1986a).




    All electrolytic cells operate on the same principle in the production of




chlorine by brine electrolysis.  An electric current decomposes a solution of




brine into chlorine, caustic soda, and hydrogen; the former is liberated at




the anode and the latter two at the cathode.  The ratio of chlorine to caustic




soda produced during the process is 1 to 1.1 by weight (Chemical Week 1982).




During this process, it is necessary to keep the chlorine gas separated from




the alkali metal hydroxide co-product to prevent chemical reaction.




Separation is achieved by use of an asbestos diaphragm which basically acts as




a mechanical barrier between the two chambers.
                                   - 109 -

-------
    The proper operation of the diaphragm cell depends greatly on the

diaphragm material.  The diaphragm requires some of the following properties

in order to function properly (Chlorine Institute 1986a):

        •  Sufficient mechanical strength;

        •  High chemical resistance to acids and alkalies;

        •  Favorable electrical energy efficiency;

        •  A physical structure that permits percolation of depleted
           brine with minimum back-migration; and

        •  Feasible service life.

Asbestos exhibits an extremely favorable comb  ation of these properties

making it uniquely well suited as a diaphragm  acerial.

    In the application of asbestos in the diaphragm forming process, a layer

of asbestos slurry is drawn by vacuum techniques through a screen or

perforated plate.  Asbestos fibers are deposited on the screen, or plate,

forming a paper-like mat approximately an eighth of an inch thick (Coats

1983).  The resulting asbestos-coated screen is the diaphragm and is used as

the cathode in the electrolytic cell.  Asbestos diaphragms ar

at the chlorine plant site itself; they are not available as pi

products ready for use.

    Research has established that the deposited asbestos fibers do not only

function as filter-type mats.   Under proper cell diaphragm depositing and

operating conditions, a gel layer forms in the mat.  This gel layer greatly

aids in the diaphragm performance by optimizing power-efficiency.  Over the

past twenty years, many advances have been made in the design           3

diaphragms and in the design of the cell itself.   These have inciuued -r.s

introduction of novel types of anodes and the development of modified asbestos

diaphragms.  These resin bound diaphragms consist of chrysotile and polymeric

powders of fibers stabilized at high temperatures before use.  This


                                   - 110 -

-------
development increases the stability of asbestos diaphragms and extends their




service life.  These and other innovations have resulted in a significant




reduction in asbestos consumption per ton of chlorine over the years (Chlorine




Institute 1986a).   The majority of diaphragm cells currently in use in the




U.S. utilize these modified asbestos diaphragms; they consume 2,300 kwh of




power per ton of chlorine produced (Chlorine Institute 1986a, Chemical Week




1982).




    The surface of the diaphragm ranges from approximately 200 to 1000 square




feet for a cell with a volume of 64 to 275 cubic feet (Coats 1983).  Each




diaphragm may use up to 200 pounds of asbestos and have a service life of




three months to over one year, depending on the type of anode in use (Chlorine




Institute 1986b).   Using modified asbestos diaphragm technology, the




production of 1000 tons of chlorine as well as other primary products requires




about 250 pounds of asbestos (Chlorine Institute 1986b).




    Due to impurities in the brine solution and a variety of operating




conditions during cell operation, the asbestos diaphragm requires periodic




replacement.  When replacement is required, the cell is shut down, dismantled,




and the asbestos diaphragm is removed, thus maintaining the desired level of




chlorine processing efficiency.  The spent asbestos diaphragm is physically




separated from the cathode frame.  This procedure occurs in a designated plant




area while the diaphragm is still wet.  The liquid waste containing asbestos




is settled, decanted or filtered, and the separated solids are subsequently




disposed.




        2.  Manufacturers Using Asbestos Diaphragms




        Asbestos diaphragms are not marketed; the chlorine producers purchase




asbestos fiber and manufacture and install the diaphragm themselves.  Table 21




provides a list of chlorine manufacturers who use asbestos diaphragm cells.




In 1985, 19 manufacturers were operating 30 chlorine plants using asbestos




                                   - Ill -

-------
           Table 21.  Producers of Chlorine Using Asbestos Diaphragms*
                                               Plant
                       Annual Capacit
                        (Thousands off
                        Metric Tons)
Brunswick Pulp & Paper Company
  Brunswick Chemical Company, Division

Diamond Shamrock Corporation6
  Diamond Shamrock Chemicals Company
  Chlor-Alkali Division

Dow Chemical U.S.A.
E.I. duPont de Nemours & Co., Inc.
  Petrochemicals Department
  Freon Products Division

FMC Corp., Industrial Chemical Group

Formosa Plastics Corporation, U.S.A.

Fort Howard Paper Company

General Electric Company
  Plastics Business Operations

Georgia-Gulf Corporation

The B.F.  Goodrich Company6
  Convent Chemical Corporation,
  Subsidiary

Kaiser Aluminum and Chemical Corp.
  Kaiser Industrial Chemicals Division

LCP Chemicals and Plastics, Inc.
  LCP Chemicals Divisions

Occidental Petroleum Corporation
  Occidental Chemical Corporation,
    Subsidiary
  Hooker Industrial and Specialty
    Chemicals
(Location not known)

Deer Park, TX
La Porte, TX
Oyster Creek, TX
Pittsburg, CA
Plaquemine, LA
Freeport, TX

Corpus Christi, TX
Charlotte, NC

Baton Rouge, LA

Green Bay, VI

Mount Vernon, IN


Plaquemine, LA

Convent,  LA



Gramercy, LA


Solvay, NY
Montague, MI
Niagara Falls, NY
Taft, LA
Tacoma, VA
   27
  3601
  465
  320
  180
1,050
2,330

  297
  263

  180
  410

  261



  181


   831
   76
  279
  636
  181
                                   - 112 -

-------
                            Table 21  (Continued)
                                               Plant
                       Annual Capacity
                        (Thousands of
                        Metric Tons)
Olin Corporation
  Olins Chemicals Group

Pennwalt Corporation
Chemicals Crop
  Inorganic Chemicals Division

PPG Industries
Stauffer Chemical Company
  Chlor-Alkali Products Division

Vulcan Materials Company
  Vulcan Chemicals, Division

Weyerhaeuser
Mclntosh, AL
Federal Way, WA
Portland, OR
Wyandotte, MI

Lake Charles, LA
New Martinsville,  WV

Henderson, NV
Geismar, LA
Wichita, KS
  346
   82
  136
   91
l,041b
  255b
Longview, WA
                                                      TOTAL
  104
  220
  237C

  136
9,295d
aAs of January 1, 1985.

^Combined capacity for asbestos diaphragm cell and mercury cell.

cCombined capacity for asbestos diaphragm cell and membrane cell.

^Total assuming that half the capacity at facilities using asbestos diaphragm
cells and some other kind of cell is attributed to the asbestos cell.

Accidental Chemical bought all chlorine plants from Diamond Shamrock and The
B.F. Goodrich Company  (ICF Exposure Survey 1986-1987).

Sources:  SRI 1985, Chemical Business 1985, Chemical Engineering 1976,
          Vulcan Chemicals 1986, Chemical Week 1986, ICF Exposure Survey
          1986-1987.   (ICF's best estimates based on these sources).
                                   - 113 -

-------
diaphragm cells with an estimated total annual capacity of approximately 9.3




million metric tons (10.2 million short tons).  The largest of these chlorint




producers was Dow Chemical, with a combined annual capacity of 3.9 million




metric tons (4.3 million short tons), approximately 42 percent of the total




asbestos diaphragm cell chlor-alkali capacity, followed by Occidental




Chemical, accounting for about 13 percent of the asbestos diaphragm cell




chlorine production capacity.  Asbestos diaphragm cell chlorine production




accounts for about 77 percent of the total chlorine production (see Table 22.




    Chlorine production and asbestos fiber consumption information for the




period 1983-1985 is presented in Table 22.  Based on this infonaation, about




975 tons (i.e., short tons) of asbestos was consumed by the chlorine industrj




in the production of approximately 10 million tons of chlorine in 1985.



    Based on worker population data provided by the respondents of




Exposure Survey (1986-1987), approximately 0.07 workers are exp-




asbestos for every metric ton of asbestos diaphragm cell capacity.  Therefore




an estimated 650 workers are exposed to asbestos during chlorine manufacture.




(The Chlorine Institute (1986a) estimated that about 225 workers are involved




in asbestos handling operations.)




        3.   Exposure Profile




        The results for 8-hour TWA exposure to asbestos during the manufacture




of chlorine are presented in Table 23 for several job categories.  These




values are based on the raw monitoring data.  As revealed by the data and the




geometric and arithmetic mean values in Table 23, there is minimal




occupational exposure for chlor-alkali workers to asbestos fibr .




Apparently, exhaustive and continuous efforts are made by the industry to




minimize the number, frequency, and duration of worker exposure to asbestos.




    The Chlorine Institute, Inc. (1986a) states that workers in this industry




are assigned to periodic, short-term tasks involving the handling of asbestos



                                   - 114 -

-------
                                     Table 22.  Chlorine Production/Asbestos Fiber Consumption
1
rear
1983
1984
1985
2
Total
Chlorine Capacity
(millions of ton»)b
14.6
13.6
13.2
3
Capacity
Utilization
Rata
(on Average)
661
721
77X
I
Production of
Chlorine
(millions of tons)b
(2 x 3)
9.64
9.79
10.16
5
Percentage of
Production
Using Asbestos
Diaphragms
77. Oc
77. Oc
76.7'
6
Quantity
of Chlorine
Produced Using
Asbestos Diaphragms
(millions of tons)
(4 x 5)
7.42
7.54
7.80
7
Ratio of
Asbestos
Fiber
Consumption
to Chlorine
Production*
0.000125
0.000125
0.000125
8
Consunption of
Asbestos Fiber
(tons)
(6 x 7)
928
943
975
"Chlorine Institute 19B6b.




bChamlcal Week 1985.




cCoats 1963.

-------
                                                   Table 23.   Exposure Profile for Asbestos Diaphragm Cells
... ... 8-Hour TWA Exoosure (f/cc)

Product Job Category"
Asbestos Diaphragm Cells Brine/Sludge Worker
Cell Worker
Yard Worker
Other
Total

Population
165
410
38
_2Z
650
Pre-0.2
Geometric Mean
0.009 (8)
0.024 (252)
0.007 (6)
0.019 (18)
0.019
f/cc PELC
Arithmetic Mean
0.010
0.083
0.012
0.033
0.058
Post-0.2
Geometric Mean
0.009
0.022
0.007
0.019
0.018

f/cc PEL"
Arithmetic Mean
0.010
0.052
0.012
0.033
0.038

Duration
(hr/day)d
8

8
8
8

Frequency
(days/year)*
250
250
4
6
220
aJob categories are based on a concise categorization of the job titles.

bThe total population is estimated based on information provided'in the ICP Exposure Survey (1986-1987) and the asbestos diaphragm cell chlorine capacity.
The distribution of the total number of workers exposed into specified Job categories ia based on the raw data.  The total population is allocated into Job
categories by selecting one of each type of Job (i.e., Job titles, as shown in monitoring data) for each Job category,  and then totalling the individual Job
title populations which are either provided in the specific study or estimated.  The populations Include workers from all shifts.

cThese values represent geometric and arithmetic means of the raw 8-hour TWA exposure data.  The number of data points is given in parentheses.   The value
corresponding to the total population la calculated as a weighted average based on the number of workers exposed in each Job category.
d
 These post-0.2 f/cc PEL exposure value* are calculated directly from the monitoring data.  Each 8-hour TWA exposure value that is above 0.2 f/cc is reduced
to exactly 0.2 f/cc.  Data that art already at or below this value remain unchanged.  The valua corresponding to the total population la determined in the
same manner as the pre-0.2 f/cc PEL total exposure value.

"The duration of exposure is 8 hours/day in all cases.  Where no duration is provided, 8 hours/day is assumed.   Exposures for less than 8 hours  are converted
to 8-hour TWAs, assuming zero-exposure during periods when the worker is not handling asbestos.

 Frequency refers to the number of days annually that the workers are performing a task involving potential exposure to asbestos.   The frequency is not
assumed to be 250 days/year for all Job categories alnea data for specific Job types indicate otherwise.  For the "yard worker" and "other" categories,
information shows that the performance of these Jobs (yard work and maintenance pipe fitting) is very infrequent.

Sources:  ICF Exposure Survey 1986-1987, ICF Market Survey 1986-1987,  HIOSH 1983b, Chlorine Institute 1986a.

-------
fibers.  Thes-  workers are trained, equipped with respirators and/or work in a




non-exposure ervironment.  Asbestos diaphragms are wet for most of the time




they exist;' the cell in which they are contained is a fully enclosed system,




essentially precluding the release of airborne fibers (Chlorine Institute




1986a).  All "spent" diaphragms are subject to controlled disposal at the end




of their useful lives.




    The principal potential sources for exposure to asbestos among ohlor-




alkali workers are associated with receiving, storage, weighing, diaphragm




depositing, diaphragm rebuilding, and disposal (Chlorine Institute 1986a).




Thus, the cell worker receives the highest exposure, as shown in Table 23.




The weighing of asbestos in the dry state is the work activity with the




highest exposure potential (Chlorine Institute 1986a).  Dry handling




operations take place in dedicated plant areas with restricted access and




respiratory protection.  Weighing occurs in equipment engineered to control




asbestos emissions.




    Whenever feasible, dry asbestos is wetted at the earliest opportunity,  and




subsequently used as a liquid slurry.  The process of removing asbestos from a




diaphragm cell is conducted under wet conditions to avoid the possibility that




inner layers of dry asbestos exist.




    Strict housekeeping and monitoring practices are required by the industry.




Accidental spills of asbestos are cleaned up immediately, either by vacuum or




wet sweeping.  Wash down systems are employed; all equipment used in asbestos




handling is washed down in a trench system.  Complete individual exposure




evaluations are periodically conducted for all personnel potentially exposed




to asbestos dust.




    Hence, exposure levels are low for all .job categories.  Except for a few




cell workers, all monitoring data are below the 0.2 f/cc level.  Thus,
                                   - 117 -

-------
projected exposures are identical to pre-0.2 f/cc PEL   posures for all cases




except in the cell worker category where a slight decrease is expected.




        4.   Frequency and Duration of Exposure



        As indicated by the monitoring data, most operations require less th<




a full day of work and typically occur about 250 days per year.  The duratioi



presented in Table 23, however, are all 8 hours per day because the exposure




levels are converted to 8-hour TWAs.  The various sources provided the data



8-hour TWA levels.  Thus, the duration  is taken as 8 hours per day, not the



actual duration  of exposure, because  the short exposure duration  is already




accounted for  in the exposure level  (Donahue  1987).



     Nevertheless,  several jobs are carried  out quite  infrequently.   Data




 suggests that  the yard worker  is exposed only 4  days  per  year (ICF Ex-



 Survey 1986-1987).   Exposure durations  associated with the task cr



 asbestos in the  dry  state  range  from 10 to  120 minutes (averaging 30  to 60




 minutes) and occur from one to seven times  weekly.




     D.  Brake Repair Service



     Asbestos is currently used in drum brake linings for the  rear brakes of




 most automobiles.  Most new cars have drum brakes, and most new drum  brake




 linings contain asbestos while a  small percentage are non-asbestos; nearly al-




 older cars have asbestos drum brake  linings  in the rear brakes.  The  nv>   -;-




 of  disc brake pads, used in the front brakes of most automobiles an



 rear brakes of  a small number of  automobiles, are non-asbestos for newer  ca_



 however, a  sizeable percentage  of disc brake pads on older  cars  still contain




  asbestos.   A  majority of heavy  vehicles have asbestos brake linings.   Thus,



  most automobiles and  trucks have  asbestos  brake linings  that requ    periodic




  servicing (i.e., inspection,  adjustment, or  replacement).   In additi.     me



  automobiles have asbestos  clutch facings or  automatic transmission components




  which must be replaced periodically.  Industrial equipment may also  contain




                                      - 118 -

-------
asbestos fricL-on materials such as brake linings of various types,  clutch




facings, and transmission components that require periodic servicing.




    Exposure data on the repair of friction products are available only for




automotive brakes; no data have been found for other friction products such as




industrial brakes and clutches.  Therefore, this section focuses on exposures




during automotive brake servicing; automotive brakes are the largest segment




of the friction products category, consuming about 91 percent of the asbestos




used in this category (based on IGF Market Survey (1986-1987) estimates).




Repair of other types of brakes may lead to exposures similar to those found




for automotive brakes.  However, exposures from repair of paper-type friction




materials such as automatic transmission components and industrial wet




friction components, where wear debris from friction is trapped in a fluid,




are likely to be much lower than for automotive brakes.




        1.  Exposure Setting/Process Description




        Automobiles, trucks, and other vehicles with asbestos-containing brake




lining materials require periodic brake servicing and replacement of the brake




linings.  Brake servicing may be carried out in specialized brake service




facilities, full-service garages, service stations, dealer shops, or




self-serviced fleet shops (Hunter Publishing Co. 1985).  Brake servicing




includes removal of the wheels and drums (in the case of drum brakes),




cleaning of the brake assembly and drum, and replacement of the worn linings,




if necessary.  Worn linings are usually removed still attached to the shoe or




plate and new linings, already attached to shoes, are used for replacement;




mechanics rarely attach new linings to shoes during brake servicing.  There is




usually no grinding or drilling of brake linings for automobiles or trucks




during servicing (ATA 1985, Sears 1985).
                                   - 119 -

-------
    Rebuilding of brakes, by stripping off the old , :  ing material and

replacing it with a new lining, is considered a secondary manufacturing

procedure and is discussed in Section B of this chapter.

    Brake cleaning procedures vary.  In the past, it was common practice to

blow out dirt and debris from the brakes and drums using compressed air; th:

method of cleaning has been prohibited by OSHA in a rule which went into

effect on July 26, 1986 (OSHA 1986a).   Other common cleaning methods include

the use of liquid brake cleaners, wiping the parts and drums with a wet rag,

use of a wet or dry brush, and various combinations of methods.  Enclosed

vacuum systems are also available for brake cleaning.   Some facilities may

require the use of dust masks by brake repair workers (Sears 1985), but this

practice appears to be unusual.  Host brake repair workers do not use dust

masks or respirators (ATA 1985, Precision Import Service 1985).  B~

cleaning and asbestos control procedures and devices are desc

           a.  Compressed Air/Solvent Mist

           Brake cleaners, which contain solvents such as 1,1,1-trichloro-

ethane (PEI 1985), may be incorporated into the compressed air system and

sprayed on the brake assembly.  OSHA allows this method of cleaning,

presumably because the solvent mist captures the dust and asbestos fibers.*

PEI Associates (1985) points out, in a study of control methods, that it is

important that the solvent be collected for recycle.  If the solvent is

allowed to evaporate, the asbestos may be reentrained in the air.  It is not
                                                        »
known to what extent the solvent is recycled when this method is used.
       OSHA presents the compressed air/solvent system method as a preferred
method for brake repair even though use of compressed air is prohibited unles*
an enclosed ventilation system is used (OSHA 1986a, pp. 22753, 22758).

                                   - 120 -

-------
           b.  Hrush




           Dust and debris may be removed from brake assemblies by use of a




brush.  According to PEI Associates (1985), this method may cause dust to fall




on the worker's clothing and in the immediate area.  The amount of exposure




may be highly dependent on the worker's work practices.




    A variation of this method is use of a brush that is wet with water or




another liquid to keep the dust down.




           c.  Water Sprav/Rag




           Water may be applied to the brake assembly by squirt bottle or




hose, and the brake parts then cleaned with a rag.  This method, like brush




cleaning, may cause contamination of clothing and accumulation of dust on the




garage floor.  Exposure may be highly variable, depending on individual work




practices (PEI Associates 1985).




           d.  Brake Washer




           NIOSH (1987a) describes a brake washer assembly unit used to




control asbestos; other designs are commercially available.  The unit




described by NIOSH contains a water solution that is pumped through a flexible




tube with a brush attached to clean the brake parts.  A removable upper tray




holds small parts and catches the solution; the bottom tank is used for




cleaning brakes on larger vehicles.  The solution is filtered and




recirculated.




    Some brake washer units are connected to a compressed air gun.  Liquid is




siphoned into the air line and sprayed onto the brake parts for cleaning.  The




liquid runs off the parts into a perforated pan which catches debris; the




liquid that drains through is recycled (PEI Associates 1987).




    PEI Associates (1985) reports finding garages that do not use their brake




washers as intended, using them as parts washers instead.
                                   - 121 -

-------
           e.   Vacuum Unit Without Enclosure




           A vacuum unit with a High Efficiency Parti-ulate Air (HEPA) filte




for use in brake servicing is described by NIOSH (1987b).   The unit consists




of a dust removal hose,  used to suck loose dirt from the brake assembly




surface, connected to a three-stage vacuum dust filter assembly.  This unit




originally included an enclosure, which was removed by the user.  According




PEI Associates (1987), no HEPA vacuum units without enclosures are'currently




being marketed for collection of asbestos from brake maintenance; however,




units marketed for other purposes might be used for brake cleaning, and unit;




sold with enclosures might be-used without the enclosures, as described in tl




NIOSH study.




           f.  Vacuum Unit with Enclosure




           HEPA vacuum units with enclosures are commercially available.




These units typically have enclosures that surround the brake drum, and have



glove inserts for the worker's hands.  Compressed air is used inside  the




enclosure to clean the brake parts, and the dust is removed by vacuum.  A




number  of different models of this type of system are commerci




NIOSH  (1987c) studied a facility where mechanics performed brake t>c.




using a vacuum enclosure unit, consisting of a glove box enclosing the brake




assembly and a hose connection to a three-stage vacuum  dust filter assembly.




Using  the glove box, mechanics can perform brake cleaning and other operations




within  the  enclosure with a compressed air gun, a vacuum line with brush



attachment, a hammer or mallet,  and a separate brush.




    NIOSH  (1987b) reports that one company stopped using enclosures with  its




vacuum units because mechanics complained of dust escaping during  use.  PEI




Associates  (1985) mentions contamination due to buildup of asbestos inside  the




enclosure  and mechanics finding  the system cumbersome as additional problems



with vacuum enclosure systems.




                                    -  122  -

-------
        2.  Current Exposures

        Monitoring data for a number of brake servicing facilities, both for

the period of brake servicing and expressed as 8-hour TWAs for brake

mechanics,* are presented in Table 24.  The 8-hour TWA data are included for

comparison with the.OSHA PEL.  The data are grouped according to whether the

brake cleaning methods used are non-engineering or engineering controls, and

are presented by type of cleaning method or control used.  It should be noted

that the data by control type are limited.  For several of the controls, data

are available from only one facility; and only a few data points are available

for other controls.   Details of work practices followed when using the non-

engineering controls are not available,  and only a few models of the available

types of devices are represented by the engineering control monitoring data.

Because exposures may vary by type of vehicle (i.e., automobile versus truck),

type of brake (i.e., drum brake versus disc brake), type of facility,

individual work practices, and control device used, and because background

asbestos levels are not available, the data in Table 24 cannot be used to make

accurate comparisons of the effectiveness of the various control methods.

    The data presented in Table 24 are personal sampling results from NIOSH

studies.  Similar methods of sample collection and analysis were used in all

the studies.  The samples were analyzed by phase contrast microscopy, using a
       For short-term, non-manufacturing jobs such as brake repair, the
reported exposure levels are short-term TWAs (Reed 1987).  To calculate the
8-hour TWA exposure from the TWA exposure for the period of time sampled, the
following equation is used:

                    Sampling Time (minutes)
       8-Hour TWA -	x TWA for Period of Time Sampled
                         480 Minutes

One asbestos brake job per 8-hour period is assumed in this calculation,
although the number of brake jobs may vary widely by facility.  The sampling
time serves as a surrogate for the time of job performance.

                                   - 123 -

-------
Table 24.  Asbestos Exposure Daring Brake Servicing, By Control Method
Vehicle
Control and/or Brake Type
Non-EnftinaerinR Controls
Compressed Air/ MA
Solvent Mist
Compressed Air/ Front disc brakes
Solvent Mist 4 -Wheel disc brakes
Front disc brakes
Front disc brakes
and rear shoes
Front disc brakes
and rear shoes
Dry Brush NA






Brush Wet with NA
Gasoline



Hater Squirt NA
Bottle and Rag
TWA for
Time Sampled
(f/cc)

0.08
0.07
0.060
0.030
0.070
0.002

0.010

0.20
0.21
0.34
0.08
0.03
0.15
0.44
0.15
0.30
0.31
0.26
0.24
0.33
0.12
Sampling Time
(minutes)

343
283
77
239
54
162

210

129
68
124
177
47
162
184
ISO
196
175
194
135
136
190
8-Hour TWA*

-------
                                                                                 Table 24 (Continued)
NJ
Ln
Control
Vehicle
and/or Brake Type
TWA for
Time Sampled
(f/cc)
Sampling Time
(minutes)
8-Hour TWA8
(f/cc) Source Notes
Engineering ControU
Brake Hasher
Assembly Unit


















HEPA Vacuum
Unit Without
Enclosure










Jeeps



















Passenger Car
Passenger Car
Van (1/2 ton)
Van (1/2 ton)
Passenger Car
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
Van (1/2 ton)
0.0041
0.0042
0.0041
0.0042
0.0042
0.0041
0.0041
0.0041
0.0041
0.0042
0.0042
0.0041
0.0041
0.0042
0.0036
0.0043
0.0040
0.0040
0.0042
0.0041
0.0133
0.0078
0.0163
0.0033
0.0042
0.0040
0.0040
0.0034
0.0055
0.0070
0.0092
0.0040
0.0039
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
0.001° HIOSH 1987a Post Office motor vehicle maintenance
0.001° facility; source reports only one sample
0.001° a-bove the detection limit of 0.004 f/cc.
0.001C
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.004C HIOSH 1987b Fleet maintenance facility.
0.002°
0.00*c
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.002°
0.002°
0.001°
0.001°

-------
                                                                               Table 24  (Continued)
             Control
     Vehicle
and/or Brake Type
   TWA for
Time Sampled
   (f/cc)
Sampling Time
  (minutes)
6-Hour THA*
   (f/cc)
                                                                                                         Source
                                                                                                                                            Notes
             Engineering Control*  (Continued)
NJ
HEPA Vacuum
Unit with
Enclosure















BEPA Vacuum Unit
with Enclosure
Van
Van
Truck (1/2 ton)
Truck (1/2 ton)
Truck (1/2 ton)
Truck U/2 ton)
Automobile
Automobile
Truck (1/2 ton)
Truck (1/2 ton)
Salt Truck
Salt Truck
Automobile
Automobile
Truck (1/2 ton)
Truck (1/2 ton)
Truck (1/2 ton)
Truck (1/2 ton)
HA

0.0041
0.0041
0.0042
0.0042
0.0041
0.0041
0.0038
0.0038
0.0042
0.0042
0.0036
0.0036
0.0042
0.0042
0.0029
0.0029
0.0033
0.0033
0.01

b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
240

0.001° HIOSH 1987o Department of Transportation maintenance
0.001° facility; source reports no samples above
0.001° the detection limit of 0.004 f/cc.
0.001C
0.001C
0.001C
0.00lc
0.001C
0.001C
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.001°
0.005 NIOSH 1982b Private fleet service garage.

            Hotel:  Airborne fibers may be 30 percent (NIOSB 1982b) to 55 percent (NIOSH 1980c) chryaotile asbestos.  The remaining fibers may be up to 20 percent
                    foratarits, probably produced from chryaotile by the high temperatures of braking (HIOSH 1982b) and up to 50 percent unknown.  Some of the
                    unknown fibers are probably intermediate between chrysotile and forsterlte (NIOSH 19B2b).  Samples were analyzed by phase contrast microscopy.
                    Results presented are for single samples.

            * 8-hour THA was calculated from the THA for the time of the brake job aa follows:

                                8-hour THA - Semolina Time (minutes) x THA for Period of Time Sampled
                                                   480 Minutes

            One brake job per 8-hour period is assumed in this calculation.  The sampling time serves aa a surrogate for the time of job performance.

              Sampling time was 2 hours or duration of brake job, whichever was longer.  Exect time was not given.

            c Sampling time assumed to be 2 hours.

-------
procedure that requires counting of fibers greater than 5 urn in length and




with at least a 3 to 1 length to width aspect ratio.




    While exposures vary by facility and by cleaning method, the TWAs for the




time sampled and the 8-hour TWA exposures are generally less than 0.2 f/cc.




Exposure levels measured during use of the brake washer assembly unit and the




HEPA vacuum enclosure were nearly all at or below the detectable limit of




0.004 f/cc; the vacuum unit without enclosure also produced similar exposure




levels.  The brake washing device in use during NIOSH's study used water




pumped through a brush.  Exposure data for a brake washer that uses compressed




air are not available; it is possible that exposure levels might be quite




different with this type of device.  It is also possible that different




results might be obtained with other models of the vacuum enclosure system.




The data presented represent few specific sites.  Host of the exposure data




reported for the engineering controls are from large government fleet




maintenance facilities which may not be representative of the average




workplace because workers are paid by the hour and are, therefore, not under




time constraints to perform their work quickly.  In addition, potential




exposures during parts washer or vacuum unit servicing and clean-out have not




been addressed.




    Data from Table 24 are summarized, and the geometric and arithmetic means,




for both the time sampled and the 8-hour TWA, are presented by control type in




Table 25.  Because the data are very limited, both in the number of data




points available and the number and types of facilities monitored, it is not




possible to draw definite conclusions about the relative effectiveness of the




controls presented.  The range of asbestos concentrations reported for some of




the controls varies widely, and the ranges reported for various control types




overlap.
                                   - 127 -

-------
                                     Table 25.   Sunmary  of Asbestos Exposure by Control Method, With Calculated Means
Reported TWA Exposures
Number of
Cleaning Method Facilities

Compressed Air/Solvent Mist 2
Dry Brush 1
Wet Brush 1
Water Squirt Bottle/Rag 1
fo All Non-Engineering Controls 5
oo
i
Engineering Controls
Brake Washer Assembly Unit 1
HEPA Vacuum Unit Without 1
Range of
Nunber Sample Times
of Samples (minutes)
7 54-3*3
7 47-184
5 135-196
2 136-190
21 47-343

20 a
13 a
Range
of Values
(f/cc)
0.002-0.06
0.03-0.44
0.15-0.31
0.12-0.33
0.002-0.44

0.0040-0.0056
0.0040-0.0163
Geometric
Mean
(f/cc)
0.028
0.16
0.24
0.20
0.10

0.004
0.006
Arithmetic
Mean
(f/co)
0.046
0.21
0.25
0.22
0.17

0.004
0.007
Calculated
Range
of Values
(f/cc)
0.001-0.057
0.003-0.169
0.047-0.122
0.0*8-0.09*
0.001-0.169

0.001*
0.001-0.004*
8-Hour TWA
Geometric
Mean
' (f/cc)
0.010
0.038
0.086
0.067
0.031

0.001
0.002
Exposures
Arithmetic
Mean
(f/cc)
0.019
0.061
0.091
0.071
O.OSS

0.001
0.002
  Enclosure
HEPA Vacuum Unit With
  Enclosure
     All Engineering Controls

     All Cleaning Method* and
     Controls
                                                   19
                                                   52
                                                   73
120-240      0.0029-0.01       0.004        0.004       0.001-0.005       0.001       0.001
120-240      0.0029-0.0163     0.004b      0.00*b      0.001-0.005
                                                               *7-343      0.002-0.44
                               0.09
0.15       0.001-0.169
                            0.001       0.001
                                                                       0.028b      0.050b
 Sample time was 2 hours or period of brake Job,  Whichever was longer;  2 hours assumed for  8-hour TWA calculation.

 Weighted according to estimates of use by PEI Associates (1967)  (see text).

Source:  See Table 24.

-------
    No data are available on frequency of use of various non-engineering

cleaning methods, such as brush, rag, and solvent mist.  Combining the

exposure data for the various non-engineering cleaning methods,  giving equal

weight to each data point, the geometric mean TWA exposure for the time of

brake servicing with no use of engineering controls is 0.10 f/cc.   The

arithmetic mean TWA exposure for the time of brake servicing with no use of

engineering controls is 0.17 f/cc.

    PEI Associates (1987) has estimated use of engineering controls as

follows:

        •  Brake washer --6 percent of facilities (63 percent of
           facilities using engineering controls);

        •  HEPA vacuum unit without enclosure -- less than 1 percent
           of facilities (less than 10 percent of facilities using
           engineering controls);  and

        •  HEPA vacuum unit with enclosure -- 2.6 percent of
           facilities (27 percent of facilities using engineering
           controls).

If the geometric mean exposure level reported for each of the engineering

control methods is weighted by the percent of all facilities using engineering

controls that use each individual control, the geometric mean TWA exposure for

the time of brake servicing using engineering controls is 0.004 f/cc.  The

weighted arithmetic mean TWA exposure for the time of brake servicing using

engineering controls is also 0.004 f/cc.

    An overall geometric mean TWA exposure for the time of brake servicing was

estimated using the PEI (1987) estimates presented above for use of each type

of control.  It was assumed that the exposure level for all facilities that do

not use engineering controls (90.4 percent of the total) is represented by the

geometric mean calculated for non-engineering controls (0.10 f/cc); this

exposure level was weighted by 90.4 percent.  The geometric mean exposures for

each of the engineering controls were weighted by the PEI estimates of current


                                   - 129 -

-------
use.  The overall weighted geometric mean TWA exposure for the time of brake

servicing was calculated as 0.09 f/cc.  The overall weighted arithmetic mean

TWA exposure for the time of brake servicing, calculated by similar methods,

is 0.15 f/cc.

    We have not attempted to determine exposure for repair of each type of

automotive brake lining because of insufficient data.  It is likely that

airborne asbestos dust concentrations may be lower during disc brake pad

replacement than drum brake lining replacement because for disc brakes much o

the wear debris is released to the air during use; there is no drum to collec

brake dust and debris as there is for drum brakes.  However, there are no dat<

to verify differences in exposure levels; therefore, we have assumed that the

average exposure is applicable to all types of brakes.

        3.  Populations Exposed

        The Hunter Publishing Company (1985) lists four types of service

facilities that may perform brake work:

        •  Service stations;

        •  Independent repair shops;

        •  New car and truck dealer shops; and

        •  Self-serviced fleet shops.

Table 26 shows 1984 estimates provided by Hunter Publishing, based on survey

data, of the number of facilities of each type, the average number of

full-time employees per facility, the average number of full-time mechanics

per facility, and the total number of full-time mechanics.*

    The Hunter Publishing estimates of total facilities and employees do not

include facilities such as Sears and Midas Muffler and, therefore, may be low.
     Jp
       The Hunter Publishing data are based on a survey of a cross-section of
facilities.  The survey results are projected to give estimates for the whole
industry (Hunter Publishing Co. 1986).

                                   - 130 -

-------
      Table 26.  Facilities Where Brake Repair is Performed (1984 Data)

Type
Service

of Facility3
Stations
Independent Repair Shops
New Car
and Truck

Number of
Facilities
115,000
150,000
25,000
Full -Time
Employees
per Facility
(average)
4.3
4.3
19.0
Full-Time
Mechanics
per Facility
(average)
2.6
3.3
8.6

Total
Full -Time
Mechanics13
299,000
495,000
215,000
  Dealer Shops

Self-Serviced Fleet        39,000
  Shops (with 25 or
  more truck-type
  vehicles)               	

     Total                329,000
21.5
9.8
382,000
                          1,391,000
aDoes not include chains such as Sears and Midas Muffler.

"Calculated from average per facility.

Source:  Hunter Publishing Co. 1985 and 1986.
                                   - 131 -

-------
We do not use these total estimates of facilities or employees in our




population estimates, however.  We have instead chosen an approach which




estimates full-time equivalent (FTE) populations based on the total number c




expected brake replacements.  The only information we use from Hunter




Publishing is the distribution of brake jobs among the various types of




service facilities for each friction product and the average number of




full-time mechanics at each of those types of facilities, which allow us to




estimate the number of full-time mechanics indirectly exposed during asbesto




brake servicing (see below).  The total number of brake jobs which is used t




estimate FTE populations directly exposed during brake servicing is not




affected by the distribution of brake jobs among the various types of servic




facilities.  It is unclear what effect the omission of facilities such as




Sears and Midas Muffler has on the brake job distributions among facility




types.



    These facilities employ part-time mechanics and workers  in other




capacities not involving auto repair in addition to full-time mechanics.  No




information is available on the number of hours worked by part-time mechanic




or exposure to non-mechanics; therefore, we are not able to  estimate potenti.




exposure to part-time or non-mechanic employees.  Only full-time mechanics &'




used  in estimating the population exposed to asbestos during brake  repair.




While other workers might occasionally be exposed to  asbestos, full-time




mechanics are likely to be  exposed  on a regular basis.   Some mechanics  may dc



no brake work, some may do  very  little, and others may do brake work




full-time; however, monitoring data indicate that exposure  to asbestos  may




occur all over the workplace  and that workers  other  than those engaged  in




actual brake work may be exposed (NIOSH 1982b).  NIOSH (1982b) reports  that




according to  its  studies, all individuals in a workplace may be  exposed to  th




sane  fiber concentrations during a  work day.   Area monitoring data,  shown in




                                    -  132  -

-------
Table 27, indicate significant asbestos concentrations.  The geometric mean of

the area TWA concentrations for the period of time sampled (i.e.,  during a

brake job) is 0.04 f/cc (the corresponding arithmetic mean is 0.05 f/cc).

This is not significantly lower than the geometric mean for personal samples

for mechanics performing brake jobs; as discussed in Section D.2,  the

geometric mean TWA concentration of personal samples for the time of a brake

job was 0.09 f/cc (the corresponding arithmetic mean is 0.15).  [The geometric

mean of the area sampling results converted to 8-hour TWA concentrations (see

Table 27) is 0.03 f/cc; the corresponding arithmetic mean area concentration

is 0.04 f/cc.]

    From the Hunter Publishing Co. data shown in Table 26, the total

potentially exposed population is 1,391,000, the total number of full-time

mechanics; the actual full-time equivalent population  is smaller as discussed

below.

           a.  Duration of Exposure for One Brake Job

           The average length of time estimated for brake jobs (per axle) is

as follows:

        •  Automobile drum brakes -- 1.5 hours  (Chilton Book Co.
           1987);

        •  Automobile disc brakes -- 1.1 hours  (Chilton Book Co.
           1987); and

        •  Truck brakes  (disc and drum)  -- 2.5 hours  (see below).

The average  length of time given for automobile disc  and drum brake jobs are

average times used for estimating labor  charges for  servicing front pads and

rear  linings for rear wheel drive cars,  respectively;  the  time for  front wheel

drive cars may vary, but  the average is  approximately the  same (Chilton Book

Co. 1987).   For  a four-wheel brake  job,  the  time would be  2.5 hours,  somewhat

shorter  than for front  and rear jobs separately, but we have assumed  for

simplicity  that  each axle is serviced  separately.   The actual time  needed for

                                    - 133 -

-------
                                     Table 27.  Area Asbestos Concentration During Brake Servicing
Type of Facility
Auto Brake Service Shop
Sane Shop
Municipal Garage
Same Garage
Municipal Garage
Same Garage
Same Garage
Municipal Garage
Private Fleet Service Garage
Type of Service Performed
Brake
Brake
Brake
Brake
Brake
Brake
Brake
Brake
Brake
Service;
Service;
Service;
Service;
Service;
Service;
Service;
Service;
Service;
Compressed Air-Solvent Mist Cleaning
Compressed Air-Solvent Mist Cleaning
Dry Brush
Dry Brush
Wet Brush
Wet Brush
Wet Brush
Cleaning
Cleaning
Cleaning
Cleaning
Cleaning
Liquid Squirt Bottle Cleaning
Vacuum

TWA
Concentration
(f/cc)
0
0
0
0
0
0
0
0
0
.04
.03
.07
.03
.07
.07
.07
.06
.01
Sample
Time
(minutes )
231
222
414
382
360
360
360
395
378
8-Hour TWA
Concentration

-------
a brake job may vary; for example, NIOSH (1980c) gives 20 minutes per wheel or




90 minutes for a four-wheel job as the average at one facility, while another




NIOSH report (1982b) gives an average of 65 minutes per vehicle at an




automobile brake service facility (without specifying whether this time is for




two wheels or four wheels) and an average of five hours per vehicle at a




municipal garage.




    The time needed for truck brake jobs, in particular, may vary enormously.




For this reason, there was no average time available for estimating labor




charges for truck brake jobs, as there was for automobiles.  The time for a




brake job for trucks from pick-up truck size up to 1-1/2 tons may be in the




same range as for automobiles, while for 2- to 3-ton trucks the time varies




too much to make an estimate.  For a tractor-trailer, it may take an entire




day to do a tandem axle job (Chilton Book Co. 1987).  A representative of the




American Trucking Association suggested that a brake job might typically take




somewhere between one and four hours (ATA 1987); therefore, we decided to use




the average of 2.5 hours as a rough estimate.




    We assumed that workers would be exposed to asbestos only during the time




brake repair is taking place.  It is possible that asbestos could be airborne




for a longer period of time, or that asbestos that has settled could again




become airborne; however, there are no data available to allow for an estimate




of the extended periods for which asbestos levels would be elevated.




           b.  Full-Time Equivalent Populations




           The Hunter Publishing Company's 1985 Service Job Analysis gives an




estimate of the percent of total axle sets of drum brake shoes and disc brake




pads for automobiles and trucks installed in 1984 by the four types of service




facilities listed above.  Tzanetos et al. (1987) have estimated annual




replacement sales of asbestos drum brake linings and asbestos disc brake pads




for automobiles and light trucks for the years 1986 to 2000.  Based on an




                                   - 135 -

-------
average four-year life for brakes on automobiles and light trucks,  brakes




installed in 1985, the base year for this study, will be replaced in 1989.



have used the replacement sales estimates for 1989 to estimate the number of



brake jobs performed on brake linings and pads produced in 1985.   The




estimated, number of brake jobs and the Hunter Publishing data are used to




estimate the number of workers who might be exposed to asbestos during




installation and repair.




    As discussed above, all workers in a facility may be exposed to




approximately the same levels of asbestos when a brake job is performed.



Therefore, we believe it is reasonable to assume that when a brake job is



performed in a facility, all mechanics at the facility may be exposed during




the period of the brake job at approximately the same level as the mechanic




performing the brake work.  To estimate the number of full-time equivalent




workers exposed, we used the total number of brake jobs, the total number of



workers at facilities where brake jobs are performed, and the estimated



average time for a brake job for each type of brake lining.  We also assumed




that all full-time mechanics at the facility where the brake job was performe




would be exposed for the average time of the brake job.



               (1)  Drum Brake Linings for Automobiles




               Tzanetos et al. (1987) estimated replacement sales of asbestos




drum brake linings for automobiles and light trucks as 136,045,000 pieces in




1989, the replacement year for drum brake linings installed in the base year



of 1985.  This is equivalent to 34,011,250 axle sets (four lining pieces are




used per axle).




    Some brake jobs are performed by consumers who do their own brake repair.




Versar (1987) reports that 9,132,000 people purchased brake linings in 1982.




Assuming that approximately the same number of people buy brake linings




annually and that an axle set each of drum brake linings and disc brake pads




                                   - 136 -

-------
is included in the purchase, about 9,132,000 drum brake lining axle sets will




be bought and installed by consumers in 1989.  Ninety-eight percent of these,




or 8,949,360 axle sets, will be asbestos drum brake linings (Tzanetos et al.




1987).  Subtracting the number of asbestos drum brake lining sets installed by




consumers from total replacement sales, approximately 25,061,890 sets will be




installed in the workplace in 1989.  We applied the Hunter Publishing Co.




(1985) data on percent of total drum brake shoes replaced by each type of




facility to the total number of sets replaced in the workplace to estimate the




number of drum brake shoe sets replaced by facility type.




    Table 28 presents the breakdown by facility type of the sets of drum brake




shoes (axle sets) installed.  To estimate full-time equivalent workers, we




assumed that each brake job would take 1.5 hours (the average for rear brake




linings) and calculated the total time in hours spent on brake jobs at each




type of facility.  We calculated the number  of full-time equivalent workers




engaged in brake jobs by dividing the total  time in hours by 2,000 (one




working year).  As discussed earlier, all workers  in a workplace may be




exposed to asbestos; therefore, to estimate  the number of full-time equivalent




workers exposed, we multiplied the number of full-time equivalent workers




actually engaged in brake work by the number of full-time mechanics per




facility (see Table 26).  The results of our calculations are  shown  in Table




28.  A total of approximately 71,395 full-time equivalent workers  is  exposed




to asbestos during automobile drum brake shoe repair.




                (2)  Disc Brake Linings  for Automobiles




               Tzanetos et  al.  (1987) estimated replacement sales  of asbestos




disc brake pads for automobiles and light trucks as  96,273,000 pieces, or




24,068,250 axle sets  (four  pads per axle),  in 1989,  the year of  replacement




for disc brake pads  installed in  1985.
                                    -  137  -

-------
                                        Table 26.  Automobile Drum Brake Shoe Repair by Facility Type and
                                           Estimated Full-Time Equivalent Workers Exposed to Asbestos








1

H-1
oo
i



Type of Facility
Service Station*
Independent Repair
Shop*
New Cer and Truck
Dealer*
Self-Serviced Fleet
Shop* (with 25 or
more truck -type
vehicle*)

Total
Percent of Total
Number of Drum Brake Shoes
Facilities Installed
(1984) (1964)
115,000 39.0
150,000 46.8

25,000 12.6

39,000 1.6




329,000
Number of
Drum Brake Shoes
Time Spent on
(Axle Sets) Installed Drum Brake
as Replacements
(1989)
9,774,140
11,728,960

3,157,800

400,990




25,061,690
Jobs Per gear
(hours )
14,661,210
17,593,440

4,736,700

601,490




37,592,840
Full-Time ' Estimated Full-
Mechanics Time Equivalent
Estimated Full-Time
Equivalent Workers
Doing Brake Jobs
7,331
8,797

2,368

300




18,796
Per Workers Exposed
Facility to Asbestos from
(average) Drum Brake Jobs
2.6 19,060
3.3 29,030

8.6 20,365

9.8 2.940




71,395











*The average length of time to install a set of drum brake shoes 1* 1.3 hours.

 A full-time working year la 2,000 hours.

°Results rounded because of uncertainties in the data.

Sources:  Hunter Publishing Co. 1985 and 1966,  Tcanetos et al.  1987.

-------
    Versar (1987) reports 9,132,000 consumer purchases of brake linings in




1982.  Assuming that this number remains approximately the same in 1989,  and




that each consumer buys an axle set of both drum brake linings and disc brake




pads, 9,132,000 sets of disc brake pads will be replaced by consumers.   About




65 percent of these," or 5,935,800, will be asbestos disc brake pads (Tzanetos




et al. 1987).  Thus, approximately 18,132,450 disc brake pads (axle sets) will




be installed in the workplace in 1989.  Using this total for replacement brake




sets and the Hunter Publishing Co. (1985) estimates of percent of disc brake




axle sets installed by each type of facility, we estimated the number of disc




brake axle sets installed by facility type.  These estimates are shown in




Table 29.




    We used 1.1 hours, the average length of time for a front brake pad job




(ChiIton Book Co. 1987) to calculate the total time spent installing asbestos




disc brake pads annually.  We used the same methodology used for drum brakes




to estimate the number of full-time equivalent workers exposed to asbestos




during automobile disc brake repair; the results of our calculations are shown




in Table 29.  Ve estimate a total of approximately 38,890 workers exposed to




asbestos during automobile disc brake pad repair work.




               (3)  Drum Brake Linings for Trucks




               ICF Market Survey (1986-1987) estimated that 4,570,266 brake




blocks (drum brake linings for heavy vehicles) were produced in 1985.  Heavy




vehicles use four brake blocks per wheel (ICF 1987) or eight brake blocks per




axle; therefore, about 571,280 axle sets were produced in 1985.




    Truck brake blocks are assumed to need replacement every six months;




therefore, all brake blocks installed in 1985, the base year for this study,




will be replaced in 1985 (i.e., there is no lag time).  It is assumed that




there is no replacement of brake blocks by consumers and that all the brake




blocks produced are used for replacement.  Therefore, approximately 571,280




                                   - 139 -

-------
                                         Table 29.   Automobile Disc  Brake Fad Repair fay Facility Type and
                                            Estimated FuLl-Tima Equivalent Workers Exposed to Asbestos
Number of
Full-Time Estimated Full-
Percent of Total Disc Brake Pads Time Spent on Mechanics Time Equivalent
Disc Brake Sets (Axle Sets) Installed Disc Brake Estimated Full -Time Per Workers Exposed
Hunter of Installed as Replacements Jobs Per Year Equivalent Workers Facility to Asbestos from
Type of Facility Facilities (1984) (19S9) (hours) Doing Brake Jobs (average) Disc Brake Jobs0
Service Stations 115.000 37.8 6, 854,070 7,539,460 3,770
Independent Repair 150,000 46.2 8.377,190 9,214,910 4.607
Shops
Ne« Car and Truck 25,000 14.6 2,647,340 2,912,070 1,456
Dealers
Self-Serviced Fleet 39,000 1.4 253,850 279,240 140
Shops (with 25 or
more truck-type
vehicles)
Total 329,000 18,132,450 19,945,700 9,973
2.6 9.800
3.3 15,200
8.6 12,520
9.8 1,370

38,890
"The average length of time to install a set of disc brake pads la 1.1 hours per axle.

bA full-time working year la 2,000 hours.

°R«sults rounded because of uncertainties in the data.

Sources:  Hunter Publishing Co. 1985 and 1986,  Tzanetos et al.  1987.

-------
brake blocks (axle sets) were replaced in 1985.  We used this total and the




Hunter Publishing Co. (1985) estimates to estimate the number of truck drum




brake lining axle sets installed by facility type.




    Table 30 shows the total number of sets of truck drum brake linings and




the number of asbestos linings installed annually by facility type.  We




assumed that each brake job would take 2.5 hours  (note that there are large




uncertainties associated with this estimate) and used the same methodology




used for automobile drum brakes to estimate the number of full-time equivalent




workers exposed to asbestos during truck drum brake lining repair.  The




results of our analysis are shown in Table 30.  We estimated a total of




approximately 3,832 full-time equivalent workers  exposed to asbestos during




truck drum brake repair.



               (4)  Disc Brake Pads for Trucks




               ICF Market Survey (1986-1987) estimated production of asbestos




disc brake pads for heavy vehicles as 156,820 pieces or 19,600 axle sets




(eight pieces per axle) in 1985.  Disc brake pads  for heavy vehicles need




replacement about every six months; therefore, all disc brake pads for heavy




vehicles installed in 1985, the base year  for  this study, will be replaced in




1985 (i.e., there is no lag time).  It is  assumed that no replacement of brake




pads is done by consumers, and that all pads produced in 1985 were used for




replacement.  Thus,  19,600 disc brake pads (axle  sets) were  replaced in 1985.




We used this total and  the Hunter Publishing Co.  (1985) estimates  to estimate




the number of heavy vehicle disc brake pad axle sets  installed by  facility




type.



    Table 31 shows the  breakdown by facility type of  the sets of truck disc




brake pads  (axle  sets)  installed.  We assumed, as for brake  blocks, that each




job would take 2.5 hours.  We used the same methodology as for automobile drum




brakes to estimate the  number of  full-time equivalent workers exposed  to




                                    -  141  -

-------
                                           Table 30.   Truck Drum Brake Lining Repair by Facility Type and
                                             Estimated FuU-Time Equivalent Workers Exposed to Asbestos
Type of Facility
Service Stations
Auto Repair Shops
New Car and Truck
Dealer*
Salt-Serviced Fleet
Shops (with 25' or
more truck-type
vehicles)
Total
Number of
Facilities
115,000
ISO. 000
25,000
39,000
329,000
Percent of Total
Truck Drum Brake
Linings Installed
(1984)
21.7
41.7
13.0
23.6
Number of Asbestos
Drum Brake Linings
(Axle Sets) Installed
as Replacements
(1985)
123,970
238,230
74,260
134,820
571,280
Time Spent
on Truck
Drum Brake
Jobs Per Year
(hours)
309,920
595,580
185,650
337,050
1,428,200
Estimated Full-Time
Equivalent Worker*
Doing Brake Jobs
155
298
93
168
714
Full-Time
Mechanics
Per
Facility
(average)
2.6
3.3
8.6
9.8
Estimated Full-
time Equivalent
Workers Exposed
to Asbestos
from Truck
Drum Brake Jobs
403
983
800
1,646
3,832°
aThe average time per axle for a truck brake job is roughly 2.5 hours.

bA full-time working year is 2,000 hours.

°Results rounded because of uncertainties in the data.

Sources:  Hunter Publishing Co. 1985 and 1986, ICF Market Survey 1986-1987,  ICF 1987.

-------
                                                            Table 31.   Truck Disc Brake Pad Repair by Facility Type and
                                                            Estimated Full-Time Equivalent Workers Exposed to Asbestos
OJ




Type of Facility
Service Stations
Auto Repair Shops
New Car and Truck
Dealers
Self-Serviced Fleet
Shops (with 25 or
more truck -type
vehicles)
Total



Number of
Facilities
115,000
150,000
25,000

39,000



329,000

Percent of Total
Truck Disc Brake
Pads Installed
(1984)
27.0
45.6
17.0

10.4




Number of Asbestos
Disc Brake Pads
(Axle Sets) Installed
as Replacements
(19B5)
5.290
8,940
3,330

2,040



19,600
Time Spent
on Truck
Disc Brake
Jobs Per Year
(hours)
13,220
22,350
8.330

5,100



49,000


Estimated Full-Tlme
Equivalent Workers
Doing Brake Jobs
7
11
4

3



25
Full-Tine
Mechanics
Per
Facility
(average)
2.6
3.3
8.6

9.8




Estimated Full-
Time Equivalent
Workers Exposed
to Asbestos
from Truck
Disc Brake Jobs
18
36
34

29



117°
                *The average time per  axle for  a truck brake  job is  roughly  2.5 hours.

                 A full-time working year  is  2,000 hours.

                cResults rounded because of uncertainties  in  the data.

                Sources:  Hunter Publishing Co.  1985  and  1986,  ICF Market Survey  1986-1987,  ICF  1987.

-------
asbestos during repair of disc brake pads in heavy vehicles.  The results of

the calculation are shown in Table 31.  We estimated a total of approximately

117 full-time equivalent workers exposed to asbestos during truck disc brake

repair.

        4. " Freouencv and Duration of Exposure

        Since the analysis uses full-time equivalent populations, the

frequency and duration of exposure are 250 days/year and 8 hours/day (the

assumed full-time work year), respectively.

        5.  Summary

        The total exposed population for brake repair, in full-time equivalent

workers, is estimated to be about 114,234.  (This figure is based on

projections and estimates and should be considered approximate.)  The

breakdown by product type is as follows:

        •  Drum brake linings for automobiles -- 71,395 full-time
           equivalent workers;

        •  Disc brake linings for automobiles -- 38,890 full-time
           equivalent workers;

        •  Drum brake linings for trucks -- 3,832 full-time equivalent
           workers; and

        •  Disc brake pads for trucks -- 117 full-time equivalent
           workers.

This  total takes into account not only workers engaged in brake work, but all

full-time mechanics in the workplace at the time brake work is performed,

because there is evidence suggesting that such workers may also be exposed  to

asbestos at comparable levels.

    The projected exposure level during repair for all brake products (data

showing variations by product type are not available) is estimated to be

0.09  f/cc, the geometric mean TWA exposure level during brake servicing using

various brake cleaning methods weighted by the fraction of facilities

estimated  to use each cleaning method.  The weighted arithmetic mean TWA

                                   - 144 -

-------
exposure level during brake serviing  is 0.15  f/cc.  This exposure level may be




rather high for workers not actually  engaged  in brake work, but evidence




indicates that exposure levels for such workers may not be a great deal lower




than for those actually performing the brake  work.




    Implicit  in the  estimate of  full-time  equivalent workers is a frequency




and duration  of exposure equal to 250 days/year and 8 hours/day (the assumed




full-time work year), respectively.




    E.  Construction Industry Exposure




    Due to the many  favorable characteristics and uses of asbestos, the




construction  industry is the principal market for asbestos materials and




products in the United States.   This  industry accounted for more than




two-thirds of the  total asbestos demand in 1980,  and for 50 percent of the




demand in 1984  (OSHA 1986b).  Historically, construction materials and




products containing  asbestos fibers have  included asbestos/cement (A/C) sheets




and pipes, vinyl-asbestos  floor  tiles, papers,  insulation, coatings and




sealants.  Since the early 1970s, however, the overall demand  for these types




of products has declined due to  the availability  of adequate substitutes, and




the increased regulatory requirements and restrictions.  This  declining demand




has continued through the  present and, as a result, several asbestos products,




which have traditionally been used in the construction industry, are no longer




produced or sold in  the U.S.  Some of these construction products that have




'largely disappeared  from the U.S. marketplace are vinyl-asbestos floor tiles,




flooring felt,  insulation  materials,  electrical insulation, rollboard, and




other paper products.  (Available exposure data and population factors for




vinyl-asbestos  floor tile  and flooring felt construction activities are




presented in  Appendix A.)
                                    -  145 -

-------
    Currently, exposure to asbestos occurs in the construction industry due to




the handling of A/C pipe, A/C sheet (flat and corrugated), A/C shingle,




roofing felt, and roof coatings and cements.




    Exposure to asbestos in the construction industry occurs during several




activities.-  First, exposure can occur during new construction.  Although




concerns about the potential health hazards of asbestos have curtailed its use




substantially, new construction activities continue to account for the




majority of the consumption of asbestos materials (Anderson et al. 1982).




These activities are classified as "installation" operations.




    Second, many building owners or managers and industrial firms are




performing asbestos abatement projects.  This involves the removing and/or




encapsulating of asbestos materials in existing buildings.  Third, renovation




work in office buildings, schools, hospitals, residential and commercial




buildings, and industrial plants may release asbestos fibers.  This is due to




the widespread use of asbestos in construction prior to recent years resulting




in the existence of substantial amounts of asbestos materials in existing




building stock.  Fourth, routine maintenance and repair activities may also




involve disturbance of asbestos material.  Finally, demolition of all or part




of a building also disturbs the asbestos materials causing possible fiber




release.  These latter four activities (i.e., abatement, renovation,




maintenance, and demolition) are all classified as "removal" operations.




    Although there are a large number of uses and activities involved in the




application of asbestos products in construction, there is only a limited




amount of data concerning exposure to asbestos fibers during construction




projects.  Furthermore, the available exposure data is quite old, dating back




to the 1970's in most cases.




    One recent source that has been utilized and scrutinized in great detail




is a collection of health sampling results by inspection received from the




                                   - 146 -

-------
Occupational Safety and Health Administration (OSHA 1987).  This information,




provided in the form of a computer-file printout, consists of approximately




1,100 individual inspections conducted by OSHA compliance officers between




1979 and 1986.  Each inspection summary provides detailed information on the




location of the surveyed establishment, the date of the inspection, the job




titles of the potentially exposed workers, the number of similarly exposed




workers, as well as the exposure measurements (e.g., sample types, measurement




types, exposure values); in many cases there are up to ten distinct exposure




values given in a single inspection.  Nevertheless, these inspection reports




fail to furnish some vital information; the specific asbestos products used




are not ascertainable via the data supplied.  The only clues given to identify




the asbestos product used are the 4-digit SIC (standard industrial




classification) codes for which the work falls under and the workers' job




titles.  This limited descriptive information is too general to identify the




specific products to which workers were directly or indirectly exposed; only




four specific inspections could be identified as applicable to this study.




OSHA regional and area offices were contacted in an attempt to determine the




specific asbestos product under inspection in each of these four  inspections.




Because these OSHA offices keep this  detailed material on file for only three




to four years following the actual inspection (subsequently the reports are




archived), information was available  for only one of these inspections.




Hence, only one inspection report could be used, for A/C shingle  removal, for




this analysis (Durham 1987).




    Information is available on most  of the germane asbestos products used  in




the construction industry:  A/C pipe,  A/C sheet  (flat and corrugated), A/C




shingle, and built-up roofing.  Each  of these products have several end uses.




For the analysis of A/C sheet removal, the absence of data necessitated the




utilization of exposure information on drywall removal.  Drywall  is actually




                                    -  147  -

-------
one specific end use,  or application,  of A/C sheet;  hence, "drywall" may be




considered a subset of the A/C sheet category and serves as a surrogate for



exposure during all other applications.   A/C sheet is also used as roofing,




siding, curtain walls, insulating board, and trim (OSHA 1986b).



    A similar methodology is applied to the analysis of roofing felt and roof




coatings/cements.   Both of these items could be considered distinct product




categories.  However,  due to the lack of data (particularly regarding the



number of exposed workers) these two products are assessed together under the




categorization of "built-up roofing."




    The data for each product category are analyzed in terms of two activity




classifications, installation and removal.



        1.  Exposure Settings and Operations




        Asbestos-cement (A/C) pipe was the greatest single user of asbestos ir




1981 (CONSAD 1984).  This product category continued to consume the most



asbestos fiber in 1985 (ICF Market Survey 1986-87).   Most of its use is



distributed in the southwestern U.S. for use in sewer systems.  Its




installation involves laying pre-cut segments of pipe in a pre-dug trench and




connecting them with various joints.  Exposure to asbestos fibers occurs when



a pipe segment must be cut to fit into a specific space (hook-up for home or




business); hand or power tools are used for this purpose.  If repair is




needed, a certain segment (corresponding to a standard pipe length) would be




cut out, and a new segment would be installed.




    Sheet material is also made from asbestos-cement.  Manufacturers cut the




sheet into standard 4'x 8' sheets, or smaller if requested.  Flat A/C sheet is




used as drywall and sometimes roofing, siding, and insulation in various




industrial, agricultural, and commercial settings.  (Flat A/C sheet is also an




important component in laboratory fume hoods and surfaces because of its




resistance to corrosion.)  Corrugated A/C sheet is used mainly in industrial




                                   - 148 -

-------
and agricultural applications, serving as siding and roofing for factories,




warehouses, and agricultural buildings.  It is also used as a lining for




waterways and canal bulkheads, and for special applications in cooling towers.




Installers sometimes have to cut the sheet into smaller segments than those




sold by primary manufacturers.  Cutting segments and drilling holes can be a




source of fiber exposure during installation.  Removal of drywall is




accomplished by destruction of the wall; this would generally be done during a




building demolition or major renovation.




    Roofing material can also be made of A/C sheet.  A/C shingles, as they are




called, are used on sloped residential and commercial roofs (CONSAD 1984).




OSHA (1986b) claims that A/C shingle is no longer used in new construction,




but it is used for replacement in existing buildings.  Removal and repair of




A/C shingle is accomplished by shattering the old shingle and hammering in a




new one.  Asbestos dust release can be expected from this operation.




    Built-up roofing is used on commercial buildings with flat, horizontal




roofs.   Asbestos roofing felt is mixed with asphalt or tar and installed in




layers.  Cutting of the felts with knives or scissors is sometimes necessary




prior to installation; this operation can be a source of exposure for roofing




workers.  Built-up roofing is normally left in place for a period of decades.




Over this time,  some of the roofing material may become brittle, which in turn




may cause release of fibers upon removal.  For the most part, however, most of




the fibers are contained in asphalt or tar, leading to lower exposures during




removal than for installation.  Removal is accomplished by means of circular




saws or axes.




    Built-up roofing also involves use of another asbestos product, roof




coatings/cements.  Roof coatings are used in waterproofing (installation) and




covering cracks (repair).  These compounds are manufactured as sprays, pastes,




and other trowel-applied compounds (Anderson et al. 1982).  The greatest




                                   - 149 -

-------
chance for occupational exposure would exist during spraying applications.




Although up to 90 percent of roof coatings are applied using a trowel or




brush, some roofing contractors currently spray the material from either a




truck or a compressed-liquid container (Rose 1987).




    Plumbing and boiler maintenance (as well as heating, ventilation, and air




conditioning (HVAC) and lighting repair) can also be sources of asbestos




exposure if these systems are located in close proximity to old asbestos




ceiling tiles or insulation.  However, asbestos ceiling tiles and insulation




have not been used in new construction for many years now, so these exposure




classifications are not covered in this analysis.




        2.  Pre-0.2 f/cc PEL Exposures




        Various studies have been performed on construction-related (and




other) industries to determine exposure to asbestos products during




installation, repair, and removal.  Table 32 summarizes these current




exposures.  Samples were analyzed with phase-contrast microscopy.




    Both geometric and arithmetic means of the raw exposure data are presented




in Table 32.  For the purpose of these calculations, all short-term TWA data




points that have a value of zero (i.e., 0.0 f/cc) are assumed to be equivalent




to 0.05 f/cc.  A zero count merely means that the count was too low for any




fibers to be seen in the counting fields; these "non-detectable" levels are




consistent with very low, but rarely actually zero, concentrations of airborne




fibers.  Hence, the use of 0.05 f/cc values where 0.0 f/cc values are recorded




is chosen to enable geometric and arithmetic mean values to be calculated as




it appears to be the limit of detection for the affected construction studies.




    Asbestos-cement pipe unloading and laying causes little fiber exposure




(CONSAD 1984).  The greatest amount of exposure occurs during pipe-cutting




operations, especially when power tools are used.  No exposure data exist for
                                   - 150  -

-------
          Table 32.   Pre-0.2 f/cc PEL Exposures  to  Asbestos  Products
                         in the Construction Industry
Product
A/C Pipe Installation
A/C Sheet (Flat and Corrugated)
Installation
A/C Sheet (Flat and Corrugated)
Removal
A/C Shingle Installation
A/C Shingle Removal
Built-up Roofing Installation
Built-Up Roofing Removal
Short-Term
Level
Geometric
Mean
0.080b
0.215°
1.360d
0.0466
0.084f
0.128g
0.072h
TWA Exposure
Arithmetic
Mean
0.114b
0.405C
0.640d
0.0506
0.094f
0.1598
0.114h
8 -Hour TWA Exposure
Leve). (f/cc)a
Geometric
Mean
0.038b
0.148°
0.340d
0.0106
0.012f
0.032s
0.009h
Arithmetic
Mean
0.058b
0.195C
0.410d
0.0116
0.013f
0.0428
0.015h
aExposure estimates are calculated geometric and arithmetic means of all of
the available exposure data for each product and operation.

^Equitable Environmental Health 1977.

CNIOSH 1979a, NIOSH 1981a.
      1986b, CONSAD 1984.  Assumed a 2-hour job duration to calculate the
short-term TWA.

eNIOSH 1985a.

%IOSH 1985a, OSHA 1987.

SAnderson et al. 1982, based on data from Johns -Manville studies, reported in
Fenner 1980.

hAherne 1980, Anderson et al. 1982, Lebel 1985.
                                   - 151 -

-------
A/C pipe repair or removal; therefore, exposure for A/C pipe removal is not



analyzed.




    A/C sheet (flat and corrugated) material is installed as siding and



roofing.  Dust exposure is highly dependent on the use and control of power




tools in such operations.  Anderson et al. (1982) give a short-term exposure




figure of <0.2 f/cc for A/C sheet installation when proper dust collection



devices are attached to power tools, citing studies by Nilfisk (Argonne




National Laboratory 1981) and Johns-Hanville.  The lack of dust collection




systems has been shown to lead to short-term exposures of >2 f/cc, according




to studies by Cogley et al. (1981) and Rodelsperger (Anderson et al. 1932).



Our current exposure estimates are based on two NIOSH studies focusing on the



potential exposure to carpenters during the sawing and handling




of asbestos sheetboard used for enclosing air conditioning and physical plant:




(NIOSH 1979a) and for building patios and balconies (NIOSH 1981a).  It should




be noted that cutting and drilling operations constitute only a small fractioi



of an A/C sheet installer's time.




    No exposure data are available for the removal of general A/C sheet




material.  Information is available, however, for the removal of asbestos




drywall, one application of A/C sheet.  For demolition and repair (i.e.,



removal) of drywall, OSHA gives unsourced exposure numbers of 0.34 f/cc and




0.75 f/cc, respectively.  CONSAD gives figures of 0.41 f/cc and 0.13 f/cc for




demolition and cutting activities, respectively.   These values are used to




generate the geometric and arithmetic mean values presented in Table 32 for




A/C sheet removal.  Because no actual data are available for this sector, the



short-term TWA value is estimated assuming A/C sheet removal requires




approximately 2 hours of work per day.  Exposures during installation and




removal of flat and corrugated A/C sheet products are assumed to be identical.
                                   - 152 -

-------
    Breathing zone samples analyzed during the removal of an old asbestos




shingle roof from a residential building and the installation of a new




asbestos shingle roof are reported in a recent NIOSH report (NIOSH 1985a).




All members of the tear-off and clean-up crews wore half-face respirators with




high efficiency particulate filters (NIOSH 1985a).  Asbestos exposure data for




A/C shingle removal are also available for an inspection performed by the




Occupational Safety and Health Administration (OSHA 1987).  This planned




inspection, carried out from November 1983 to January 1984, involved roofers




tearing off old shingles from an apartment house in Waycross, Georgia (Durham




1987).  The sampling times are not provided for these OSHA values, thus, 2




hours per day is assumed to be the duration of worker exposure during A/C




shingle removal (NIOSH 1985a).




    Exposures to asbestos during construction with roofing felt products are




expected to be greatest during installation, as opposed to during repair or




removal.  Once built-up roofing has been installed, the fibers are bound with




the tar or asphalt in which they are saturated (Anderson et al. 1982, CONSAD




1984, OSHA 1986b);  this proves to be the case for current exposures.  As




discussed earlier,  the geometric and arithmetic mean calculations assume that




short-term TWA data points recorded as 0.0 f/cc are 0.05 f/cc (based on the




limit of detectability of the sampling methods utilized).  Additional data are




also available for the removal of asbestos built-up roofing from a high school




in Louisiana (Lebel 1985) and from a series of unidentified sites (Aherne




1980).




    Coatings and sealants are also applied to built-up roofing as




waterproofing material.  The occupational exposure for roof coatings during




construction is analyzed jointly with roofing felt under the categorization of




built-up roofing.
                                    - 153  -

-------
        3.   Projected Post-0.2 f/cc PEL Exposures




        Reduction of the PEL for asbestos from 2 f/cc to 0.2 f/cc means that




increased worker protection will be required in industries where asbestos is




still used.  In the construction industry,  many companies are small operators




who cannot afford extensive protective gear (such as respirators) or some of




the more sophisticated power tools with dust collection systems.  For all




product categories and activities, the projected exposures are based on the




assumption that all raw data points greater than 0.2 f/cc are adjusted to




exactly 0.2 f/cc.  This reduction may occur via the utilization of engineer in




controls (e.g., tool shrouding) or the use of respirators.  This approach




yields conservative estimates of projected post-0.2 f/cc PEL exposures; the




projections are presented in Table 33.




    Current exposure to asbestos from A/C pipe installation is below the 0.2




f/cc limit; only during pipe-cutting are large numbers of fibers released.




Therefore, the projected exposure equals the current exposure level.




    A/C sheet and shingle installation can lead to asbestos exposure during




cutting and drilling operations.  Proper shrouding of tools and dust




collection systems can reduce the estimated current exposure level for A/C




sheet installation.  The data suggest only a small reduction in overall




exposure levels for A/C sheet installation because most of the existing data




are below the 0.2 f/cc level.  No exposure reduction is projected for A/C




shingle installation or removal since all of the existing data are already




below 0.2 f/cc.




    Drywall demolition and repair involves cutting and/or shattering sections




of A/C sheet.  Repair and removal workers could easily be exposed to dust




levels for a longer period of time than installation workers.  OSHA (1986b)




recommends use of a half-mask negative-pressure respirator to reduce exposure




by a factor of 10.  CONSAD (1984) also recommends use of dust collection




                                   - 154 -

-------
             Table 33.  Projected Exposures to Asbestos Products
                         in the Construction Industry
                                   Post-0.2 f/cc PEL
                                Short-Term TWA Exposure
                                     Level ff/cc">a
 Post-0.2 f/cc PEL
8-Hour TWA Exposure
   Level ff/cc)a
Product
A/C Pipe Installation
A/C Sheet (Flat and Corrugated)
Installation
A/C Sheet (Flat and Corrugated)
Removal
A/C Shingle Installation
A/C Shingle Removal
Built -Up Roofing Installation
Built -Up Roofing Removal
Geometric
Mean
C.080b
0.173

0.800

0.046b
0.084b
0.128b
0.072b
Arithmetic
Mean
0.114b
0.278

0.800

0 . 050b
0.094b
0.169b
0 . 114b
Geometric
Mean
0.038b
0.119

0.200

0.010b
0.012b
0.032b
0.009b
Arithmetic
Mean
0.058b
0.140

0.200

0.011b
0.013b
0.042b
0.015b
Projections are calculated assuming  that  all  8-hour TWA raw data originally
greater than 0.2 f/cc would be reduced  to  this limit,  and new geometric and
arithmetic means are calculated.

bNo change from the pre-0.2 f/cc PEL  exposure.
                                    -  155  -

-------
systems and wetting agents to keep fiber exposures down.   Thus,  a substantia




reduction in exposure is expected for A/C sheet removal;  this is exhibited b




the lower projected geometric and arithmetic mean, 8-hour TWA values




(0.20 f/cc) compared to the pre-0.2 f/cc PEL geometric and arithmetic mean,




8-hour TWA exposures (0.34 f/cc and 0.41, respectively).




    Installation,  repair, and removal of built-up roofing material is a sour<




of asbestos fiber exposure (especially installation work).   OSHA and CONSAD




both recommend use of half-mask negative-pressure respirators to reduce




exposures during installation by a factor of 10.  This protection would also




automatically reduce exposures during application of asbestos roof coatings.




Additional protection is also recommended during removal and repair (tear-off




activities, especially where the roofing material has become brittle.




Recommendations for further exposure reduction included wetting the roof prio




to removal; evacuating the building occupants if feasible; sealing off all




doors, windows, and other openings; and proper handling of workers' clothing




(NIOSH 1985a).  No exposure reduction is projected for built-up roofing




installation or removal, however, since all of the existing data are already



below 0.2 f/cc.




        4.  Populations Exposed




        As the production of asbestos-containing products continues to




decline, so will the numbers of exposed workers.  Workers in the construction




industry often work with a variety of materials, depending on the needs of the




purchaser.  However, a determination can be made of the equivalent number of




workers who would work full-time (8 hours/day and 250 days/year) exclusively




with asbestos.  This number, the full-time equivalent (FTE) population, is




based on crew size, productivity, and total production plus imports of the



asbestos product.
                                   - 156 -

-------
    We assumed that workers would be exposed to asbestos only during the  time


construction activities are taking place.  It is possible that asbestos could


be airborne for a longer period of time or that asbestos that has settled


could again become airborne; however, there are no data available to allow for


an estimate of the extended periods for which asbestos levels would be


elevated.  It is also possible that other workers not handling asbestos


products would be in the vicinity during construction activities and would be


exposed to asbestos.  Data on the number of people likely to be in the


vicinity during asbestos construction activities and the levels of asbestos to


which they are exposed are not available.  It is likely that indirect


exposures to these people would be captured in the ambient exposure


assessment.


    In the construction industry, crew sizes can be estimated based on Means


Man-Hour Standards (Means 1983).  The construction industry is the biggest


user of asbestos products, so FTE populations are significant.


    A/C pipe accounts for the greatest single use of asbestos  (ICF Market


Survey 1986-1987).  A total of 216,903 tons of pipe, representing 15,062,709


linear feet, was produced or imported in 1985  (ICF Market Survey 1986-1987).


Given an average crew size of 3.5 workers  to install piping at a productivity


rate of 228.9 ft of pipe/crew/day  (Means 1983)  (working 250 days/year),  the


FTE population for A/C pipe installation is 921 person-years.


    Installation of A/C sheet  (flat  and  corrugated) is analyzed  separately


from A/C shingle, since separate production and import quantities have been

                                                                            n
obtained.  A typical crew of four persons  can  install an average of 877.5 ft'


of A/C sheet in one day  (Means  1983).  The total  1985 production plus imports


of A/C flat sheet was 856,070  ft2  (ICF Market  Survey 1986-1987).  Therefore,


the total FTE population  for A/C  flat sheet installation is 16 person-years


(assuming 250 days/year).  The  total 1985  imports of A/C corrugated sheet


                                    - 157 -

-------
(there is no domestic production)  was 385,900 ft2 (ICF Market Survey




1986-1987).  Therefore, the total  FTE population for A/C corrugated sheet




installation is 7 person-years.




    A/C sheet removal is based on the ability of a work crew of 3 persons tc




remove and replace 520 ft2 of A/C sheet per day (OSHA 1986b).   Assuming all




the 856,070 ft2 of A/C flat sheet produced and imported into the U.S. in 198




(ICF Market Survey 1986-1987) is repaired or replaced after its useful life,




the total FTE population for A/C flat sheet removal is 20 person-years.




Assuming all of the 385,900 ft2 of A/C corrugated sheet imported into the U.




in 1985 (ICF Market Survey 1986-1987) is repaired or replaced after its usef




life, the total FTE population for A/C corrugated sheet removal is 9




person-years.




    A total of 17,664,300 ft2 of A/C shingle was produced or imported in 198




(ICF  Market Survey 1986-1987).  Since this material is usually used in roofii




of residential housing, only one worker is assumed per crew (Means 1983).  A




worker can install 300 ft2 of shingle per day (Means 1983).  Therefore, the




total FTE population for A/C shingle installation is 236 person-years  (using




250 days/year).




    A/C shingle removal is assumed to require a 7-person crew, replacing 3,02




ft2 of shingle per day (as for built-up roofing).  If all of the A/C shingle




that  was produced or imported in 1985 is assumed to be repaired or replaced




after its useful life  (i.e., 17,664,300 ft2 of A/C shingle), then the total




FTE population for A/C shingle removal is 164 person-years.




    No companies manufacture primary roofing felt in the U.S.; however, one




company imported 1,625 tons of roofing felt in 1985 (ICF Market Survey




1986-1987).  Built-up roofing (i.e., roofing felt) can be assumed to weigh




11.5  lb/100 ft2; thus, 28,320,000 ft2 of roofing felt were imported into the




U.S.  in 1985 (ICF Market Survey 1986-1987).  Roofing felt is normally




                                   - 158 -

-------
installed by a 7-person crew at a productivity rate of 2,000 ft2/day (Means




1983).  Assuming a working year of 250 days, the total FTE population for




built-up roof installation is 396 person-years.




    Built-up roofing repair involves a 7-person crew that can replace




3,020 ft2 of roofing per day (OSHA 1986b).  Assuming all of the 28,320,000 ft2




of built-up roofing material imported in 1985 (IGF Market Survey 1986-1987)  is




repaired or replaced at the end of its useful life, a total of 263 FTE




person-years would be required.




    Use of adhesives and sealants can be assumed to be concurrent with use of




built-up roofing felt.  No specific productivity data are available, but time




spent applying sealant is probably insignificant compared to the time spent




laying the roofing felt.




    The above data are summarized in Table  34.




        5.  Frequency and Duration of Exposure




        In the construction industry, exposure duration and frequency are




effectively 8 hours/day and 250 days/year because  full-time equivalent




populations are being used in this analysis.  Conceptually, this is a




measurement of the total person-hours of exposure  involved  (much as




construction jobs require a certain number  of man-hours to do the work) and




not the actual number of workers who at some time  might install, repair, or




remove asbestos-containing construction materials.




    The actual duration of time required to perform the installation and




removal of these asbestos products ranges from approximately two to five hours




per day (Aherne 1980; Anderson et al. 1982; Equitable Environmental Health




1977; Lebel 1985; NIOSH 1979a, 1981a, 1985a; and OSHA 1986b).




        6.  Summary




        Table  35 summarizes FTE populations, projected exposure levels, and




duration and frequency of exposure.  The data presented are estimates of the




                                    - 159  -

-------
          Table 34.   FTE Populations in the Construction Industry
Workers Annual Consumption
Per Daily (Production +
Job Type Crew Productivity Imports )b
A/C Pipe . 3.5a 228.9 fta 15,062,709 ft
Installation
A/C Flat Sheet, 4a 877.5 ft2a 856,070 ft2
Installation
A/C Flat Sheet, 3C 520 ft2c 856,070 ft2
Removal
A/C Corrugated 4a 877.5 ft2a 385,900 ft2
Sheet,
Installation
A/C Corrugated 3C 520 ft2c 385,900 ft2
Sheet, Removal
A/C Shingle la 300 ft2a 17,664,300 ft2
Installation
A/C Shingle 7C 3,020 ft2c 17,664,300 ft2
Removal
Built-Up Roofing 7a 2,000ft2a 28,320,000ft2
Installation
Built-Up Roofing 7C 3,020 ft2c 28,320,000 ft2
Removal
FTE
Person-
Years
921
16
20
7
9
236
164
396
263
FTE - Full-time equivalent.  (Note:  All workers assumed to be working 8
      hours/day, 250 days/year.)

aMeans 1983.

bICF Market Survey 1986-1987.

COSHA 1986b.
                                  - 160 -

-------
         Table 35.   Summary of Occupational Exposure to Asbestos  in
                          the Construction  Industry
Projected
Post- 0.2 f/cc
PEL Short -Tern
TWA Asbestos



Job Category
A/C Pipe Installation
A/C Flat Sheet
Installation
A/C Flat Sheet Removal
A/C Corrugated Sheet
Installation
A/C Corrugated Sheet
Removal
A/C Shingle
Installation
A/C Shingle Removal
Built-Uo Roofine
FTE
Population
(person-
years)
921
16

20
7

9

236

164
396
Exposure
Levels
(f/cc)
Geometric
Mean
0.080
0.173

0.800
0.173

0.800

0.046

0.084
0.128
Arithmetic
Mean
0.114
0.278

0.800
0.278

0.800

0.050

0.094
0.169

Frequency and
Duration
of Exposure
8 hr/d, 250 d/yr
8 hr/d, 250 d/yr

8 hr/d, 250 d/yr
8 hr/d, 250 d/yr

8 hr/d, 250 d/yr

8 hr/d, 250 d/yr

8 hr/d, 250 d/yr
8 hr/d, 250 d/yr
  Installation

Built-Up Roofing
  Removal
263
0.072
0.114
8 hr/d,  250 d/yr
                                    -  161  -

-------
full-time equivalent exposures  of'workers in various construction jobs,  based,




on production and imports of the material and worker productivity.
                                   - 162 -

-------
III.
    Emissions measurement data from asbestos processing sources (i.e.,  mining,


milling, manufacturing, and trade uses) are limited; therefore, it is


necessary to use engineering techniques to estimate asbestos releases to the


ambient air from various processing sources.  The technical data base from


which emission estimates are derived contains significant data gaps as well as


other sources of uncertainty that require the use of simplifying assumptions.


Several judgments have been made in the absence of actual data and these are


stated throughout the discussion.  Therefore, the results of this analysis


represent estimates given the available data and should be used cautiously.


The emission estimates presented are by no means absolute values.


    The methodologies used to estimate air releases and the subsequent


emission estimates are presented for milling and primary manufacturing,


secondary manufacturing, mining and trade uses, and asbestos-containing waste


piles in the following sections.   (Descriptions of  products and processes are


presented in Chapter  II, Occupational  Exposure.)


    A.  Milling  and Primary Manufacturing Emissions



        1.  Methodology


        The basic approach for estimating emissions from milling  and primary


manufacturing  sources is adopted  from those  used by the EPA Office  of  Air


Quality Planning and  Standards  (OAQPS)  in establishing a National Emission


Standard for Hazardous Air Pollutants (NESHAP)  (OAQPS 1987) and  the EPA


Exposure Assessment Branch  (EPA  1986a) ,  with some  modifications.


    To  estimate  emission rates  from milling and primary manufacturing  sources,


the following  equation is used (adapted from EPA 1986a) :
         „  .   .      /n  wqw 2,000 Ib  W454 g   1 year
Asbestos Emission-  (l-e)O(_J _ )( _ _) (
                                                                    hr
                                  _
        Rate (g/sec)           e  1 short ton   Ib    8,760 hrs  3,600 sec
                                    - 163 -

-------
where, e - estimated collection efficiency of control device (fractional
           percent);

       q - total quantity of control device waste collected annually (short
           tons);  and

       a - average asbestos content of control device waste (fractional
           percent).

    The above equation shows that the asbestos emission rate is a function of

the collection or removal efficiency of the control device, the amount of

fiber collected by the control device, and the annual operating hours.  The

most comprehensive data base which provides these data elements is the data

submitted to EPA by asbestos manufacturers in response to the Toxic Substances

Control Act (TSCA) Section 8(a) asbestos reporting rule (EPA 1986b).   However,

the base year of the reported values is 1981 which is outdated for the present

estimates.

    Another data source available is the responses to the "Section 114

letters" provided by EPA's Office of Air Quality Planning and Standards

(Section 114 Letters 1985).  In 1985, Section 114 letters were sent to a

selected group of plants determined by OAQPS to be high maximum individual

risk plants.  The 114 data provides plant specific parameters on air pollution

control equipment, control device efficiencies, amount of waste collected and

asbestos content of the waste from control devices, operating hours,  and

stacks parameters.  These data are useful for estimating air releases but are

limited to 9 primary manufacturing plants that are still processing asbestos

(6 friction material plants, 2 reinforced-plastic plants, and 1 asbestos-

cement product plant) and one milling facility.

    A final data source available is the responses to the ICF Exposure Survey

(1986-1987).  The ICF Exposure Survey was sent to all the companies

manufacturing asbestos or asbestos products.  This survey contained a section

on stack data that was modeled after OAQPS' Section 114 Letter.  Seven


                                   - 164 -

-------
companies provided detailed stack data  for eight asbestos processing plants




(1 paper plant,  1  A/C  pipe  plant,  2  friction material plants, 1 textile plant,




1 adhesive and  sealant plant,  and 2  chlorine plants) and one milling facility.




In addition,  four  other plants (2  friction material plants and 2 adhesive and




sealant plants)  provided data  on the quantities of baghouse waste collected




but did not provide stack-specific information such as stack dimensions, and



exhaust gas flow rates and  temperatures.




    Due to limited data availability, many inputs to the emission rate




equation must be estimated;  the  derivation methods are discussed below.



           a.   Qperatine Schedule




           The  annual  operating  schedule at any plant is assumed to be 8,760




hours (i.e., 24  hours  per day  and 365 days per year).  Most manufacturing




plants operate  less than 8,760 hours per year (close to 6,000 hours); however,




air pollution control  systems  may operate continuously (PEI Associates 1986).




As a result, a  steady-state  release  rate is estimated over an entire year.




           b.  Quantity of Asbestos  Collected by the Control Device




           The  quantity of asbestos  fibers collected by the control device is




equal to the total quantity  of waste collected times the asbestos content in




the waste.  The  quantity of  waste  collected is derived by calculating the




ratio of waste collected by  the  control device in 1981 (EPA 1986b) to asbestos




fiber consumed  in  1981  (EPA  1986b) ,  and applying this ratio to the amount of




fiber consumed  in  1985  (ICF  Market Survey 1986-1987).  This proportioning




method, introduces  errors to  the  emission estimates because the Section 8(a)




data (non-CBI version  --  aggregated  data) gives quantity data per asbestos




product category instead of  per  plant.  Asbestos manufacturers employ slightly




different operating and housekeeping procedures from plant to plant thereby




generating different waste quantities.  However, this approach provides a




rough estimation of the waste  quantity generated.  In conjunction with Section



                                   - 165 -

-------
8(a) data, Section 114 data (1985) and ICF Exposure Survey (1986-1987) data or




waste are also used when available.




    The average asbestos content in the waste is available from Section 8(a),




Section 114, and ICF Exposure Survey data (EPA 1986b, Section 114 Letters




1985, ICF Exposure Survey 1986-1987).  Again, the Section 8(a) aggregated date-




gives asbestos content data per asbestos product category rather than per




plant.




           c.   Collection Efficiency of the Control Device




           The major types of collection devices available are cyclones, wet




scrubbers, electrostatic precipitators, and fabric filters (commonly referred




to as baghouses).  Each of these control devices has some applicability in the




asbestos industry; however, data collected under TSCA Section 8(a) shows that




baghouses are the predominant method (more than 90 percent) for controlling




asbestos releases to the ambient air (EPA 1986b, OAQPS 1987).  This analysis




assumes that baghouses are the control device used by all manufacturing




plants.




    As noted from the emission rate equation, collection efficiency is the




variable that is primarily responsible for determining the magnitude of the




asbestos emission.  Unless the collection efficiency estimates are accurate to




three or more significant figures, emission rates will vary by orders of




magnitude.  For example, an estimate of 99.90 percent collection efficiency




and one of 99.99 percent will alter emission rates by a factor of 10.  The




control device efficiencies reported by individual plants in 1981 and those




that responded to the Section 114 Letters or the ICF Exposure Survey were




based on vendor supplied estimates or design criteria and are often presented




as 99+ percent or 99.9+ percent (EPA 1986b, OAQPS 1987, Section 114 Letters




1985, ICF Exposure Survey 1986-1987).  Due to the sensitivity of air release




estimates with respect to the efficiency of the control device, it is




                                   - 166 -

-------
 necessary to estimate collection efficiency to four significant  figures.

 Since the Section 8(a) data on collection efficiencies is  insufficient and has

 a high degree of uncertainty,  it is necessary to develop more  reasonable

 efficiency estimates based on an engineering approach; such an approach was

 developed by OAQFS and is used in this analysis.

     OAQPS's methodology to estimate asbestos air releases  is presented  in the

 March 5,  1987 draft report entitled "National Emission Standards for  Asbestos

 -- Background Information for Proposed Standards" (OAQPS 1987).   OAQPS

 presents  three emission scenarios:  minimum, maximum,  and "best  estimate"

 emissions.  The normal mode efficiency for baghouses is 99.99  percent for the

 minimum and best estimate emission scenarios.  The maximum emission scenario

 gives two average baghouse efficiencies:   99.95 percent for asbestos  product

 categories with high inlet concentrations (greater than 0.1 grain/cu  ft) and

 99.67 percent for product categories with low inlet concentrations (less than

 0.1 grain/cu ft) (OAQPS 1987).  The asbestos product categories  with  high

 inlet concentrations are asbestos-cement sheet and pipe, friction materials,

 and reinforced plastics; those with low inlet concentrations are paper,

 coatings  and sealants, packings and gaskets, and textiles.*

     As a  comparision between the minimum and maximum emission scenarios, a

 collection efficiency of 99.95 percent compared to one of 99.99  percent

 changes the emission rates 5-fold, and a collection efficiency of 99.67

 percent compared to one of 99.99 percent affects the emission rates by

 approximately 33-fold.

     To estimate air releases from primary manufacturing sources, efficiency

 estimates under the maximum emission scenario are chosen for this analysis.

 It is believed that these efficiency estimates  (99.95 percent and 99.67
     * For additional information on the derivation of inlet/outlet concentra-
tions and collection efficiencies, see Appendix B of the OAQPS (1987) report.

                                    - 167 -

-------
percent) are reasonable estimates.  The selection of these efficiency




estimates is based on a conservative approach.  The collection efficiency of




99.99 percent under the minimum and best estimate emission scenarios is




achievable;  however, it may represent best demonstrated efficiency under




optimum operating conditions rather than conditions that are typically found




during normal operating mode at the asbestos plants.  A well designed air




pollution control system may have very high removal efficiency1, but gradual




deterioration of the equipment and improper operation and maintenance can lead




to a decrease in its removal efficiency (IIT Research Institute 1981).  No



attempt is made to re-estimate the collection efficiences developed by OAQPS




since it is very difficult to refine these values further without actual test




data.  In actuality, many factors may influence the collection efficiency of a




fabric filter system such as the dust properties (i.e., particle size



distribution and concentration), fabric properties, operating parameters




(i.e., pressure drop, gas volume, gas velocity, etc.), and filter cleaning




method (IIT Research Institute 1981).  These parameters are different for each




manufacturing plant; therefore, it is not possible to predict quantitatively



the performance of a filter system with much more confidence.  The generalized



approach above introduces some uncertainties into the estimation of air




releases, and this should be realized in estimating the downstream exposure.




    Emission estimates by OAQPS also take into account both the normal



operating mode efficiencies and the failure mode efficiencies of control




devices.  Since the efficiency values chosen are already estimated values with




uncertainties associated with the fourth significant digit, it is not




appropriate to take failure mode efficiency into account (failure mode




efficiency only affects the fourth significant digit).  The present estimates,




therefore, ignore this factor.
                                   - 168 -

-------
         2.  Emission Estimates

         Aggregate asbestos emission estimates from milling and primary

 manufacturing sources by product are presented in Table 36.*  Although not

 presented in Table 36, plant-specific emission estimates and supplementary

 data on plant locations and zip codes for each milling and manufacturing

 source (ICF Market Survey 1986-1987) were provided to Versar, Inc.  to model

 ambient exposures.  The 1985 asbestos fiber consumption (ICF Market Survey

 1986-1987) and the quantity of baghouse waste for each plant used to calculate

 emissions were also provided to Versar, Inc.  The quantity of baghouse waste

 represents the variable "q" in the emission rate equation discussed in Section

 A.I above.  A number of plants provided the quantity of asbestos fiber

 consumed in 1985 but noted that they phased out of asbestos product

 manufacture in 1986; asbestos emissions from these plants are not included in

 the aggregate emission estimates.  Therefore, only primary manufacturers that

 are currently (i.e., in 1987) producing asbestos-containing products are

 included in the aggregate emission estimates and were provided to Versar, Inc.

 for ambient exposure modeling.  In addition, there are plants identified as

 current producers of asbestos-containing products, but company officials have

 refused to provide data.  The 1985 asbestos fiber consumption for these

 refusal plants are based on estimated values calculated by ICF (ICF Market

 Survey 1986-1987).

     The quantity of waste collected from the air pollution control devices (q)

 is obtained by applying the waste/fiber consumption ratio to the 1985 fiber

 consumption.  This ratio is calculated from the TSCA Section 8(a) data (EPA

 1986b) as discussed earlier; the 1981 fiber consumption and waste information
     * Recall that the choice of baghouse collection efficiency greatly affects
the final emission estimates.  Emissions could be 5-fold lower for asbestos-
cement sheet and pipe, friction materials, and reinforced plastics and 33-fold
lower for paper, coatings and sealants, packings and gaskets, and textiles.

                                    - 169 -

-------
Table 36.  Asbestos Emissions from Milling and Primary Manufacturing Sources
TSCA 1985 A»be»to»
Identification Consumption
Asbestos Product/Mixture Number ( short tons)"
Raw Asbestos Fiber
Millboard
Pipeline Wrap
Beater-Add Gasketing
High Grade Electrical Paper
Asbestos Diaphragms
A/C Pipe
A/C Sheet. 'Flat
A/C Sheet, Shingles
Drum Brake Linings
Disc Brake Pads (UM)
Brake Blocks
Clutch Facings
Friction Materials
Yarn, Thread, Roving, and Rope
Sheet Gaaketing
Packing
Paints and Surface Coatings
Adheslves and Sealants*
Reinforced Plastics
~
03
04
05
06
13
14
15
17
IB
19
21
22
24
26
27
28
29
30
31
62,070b
435.8
1,333
11,840.4
744
985. 5C
32,690.8
588.8
3,893.0
19,869.0
4,130.7
2,137.1
1,663.0
1.521.8
558.0
5,301.1
1.1
22,215.5
2,082.9
636.1
Total
Asbestos
Content
of Waste1
(short tons)
9.890.98
21.3
42.7
173
81.6
0.1728
1,216.7s
129.5
43.25
6,648.3s
483. 8s
177.9s
269.0s
205.4s
50.5s
100.6
0.065
26.0
0.60s
36.8s
Baghouse Emission
Efficiency Rate
(e) (g/sec)
99. 9i
99.67
99.67
99.67
99.67
99.95
99.95
99.95
99.95
99.95
99.95
99.95
99.95
99.95
99.67
99.67
99.67
99.67
99.67
99.95
1.43E-1
2.03E-3
4.07E-3
1.6SE-2
7.78E-3
2.48E-6
1.75E-2
1.87E-3
6.23E-4
9.58E-2
6.97E-3
2.S6E-3
3.87E-3
2.96E-3
4.81E-3
9.59E-3
6.19E-6
2.48E-3
5.72E-5
5.30E-4
Hours Total
of Operation Emissions
(hr/yr) ,(kg/yr)
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
4,495
64
128
520
245
0.08
552
59
20
3,020
220
81
122
93
152
303
0.2
78
2
17

-------
                                                      TabU  36  (Continued)

*ICF Market Survey 1986-1987 and ICF estimates for non-respondents.

 This is the quantity of raw asbestos fiber production from the asbestos mill rather than asbestos consumption.

°The 1985 asbestos consumption for chlorine manufacturers using asbestos diaphragm cells is calculated using the capacity data from Table 21
(converted to "short" tons), the capacity utilization rate (77 percent) from Table 22,  and the ratio of asbestos consumption to chlorine
production (0.000125 tons asbestos/ton of chlorine) from Table 22.  (See Chapter II, Section C.)

dThls product category has been reclassified as roof coatings and cements (ICF Market Survey 1986-1987).

9Thls product category has been reclassified as non-roofing adhesives,  sealants, and coatings (ICF Market Survey 1986-1987).

 Estimated from 1985 asbestos consumption and 1981 Section B(a) data (EPA 1986b) on waste-to-consumption ratios  and asbestos content (see
Table 37) except as indicated.

'includes facility-specific data reported in Section 114 Letters (1985) or ICF Exposure Survey (1986-1987).

-------
used in the calculation of this ratio are summarized in Table 37.  For plants

that responded to Section 114 or the ICF Exposure Survey, the quantity of

asbestos waste reported per control device by each respondent was used to

calculate plant emissions and was provided to Versar,  Inc. for input into the

ambient exposure model (Section 114 Letters 1985, ICF Exposure Survey

1986-1987).  The base year for the Section 114 reported values is 1984.  Tabl

37 also lists the percent of asbestos in the waste for each product category;

this percentage represents the variable "a" in the emission rate equation.

    Since chlorine producers using asbestos diaphragms were not surveyed by

TSCA Section 8(a) or the ICF Market Survey (1986-1987), data such as asbestos

consumption and the ratio of baghouse waste to fiber consumption needed to

calculate emissions are not available.  However, chlorine plant capacities atu

general industry data on the capacity utilization rate and the amount of

asbestos consumed per ton of chlorine produced are available (see Chapter II,

Section C) and have been used to estimate the asbestos consumption for each

plant.   The factor used to estimate the total quantity of baghouse waste is

based on an average of the conversions calculated for the companies that

supplied waste data.

    Stack dimensions and the exhaust gas flow rate, velocity, and temperature

were provided to Versar, Inc. as part of the input for the ambient exposure

model.   For plants that responded to Section 114 or the ICF Exposure Survey,

all of these parameters except the exhaust velocity* are available.  For

plants that did not respond to Section 114 or the ICF survey, the stack

dimensions, gas velocity, and temperature are indicated as not available.  It

was possible to estimate the exhaust gas flow rate, however, based on OAQPS's

model plants which were developed for each product category (OAQPS 1987).
       The exhaust gas velocity is calculated based on the exhaust gas flow
rate and the cross-sectional area of the stack.

                                   - 172 -

-------
Table 37.  1981 Primary Manufacturer Asbestos
      Consumption and Waste Generation
Asbestos Mixture/
Product
Milling
Millboard
Pipeline Wrap
Beater-Add Gasketing
High Grade Electrical
Paper
A/C Pipe
A/C Sheet, Flat
A/C Sheet, Shingles
Drum Brake Lining
Disc Brake Pads (L&M)
Disc Brake Pads (H)
Brake Blocks
Clutch Facings
Automatic Transmission
Components
Friction Materials
Yarn
Sheet Gasketing
TSCA
ID#

03
04
05
06
14
15
17
18
19
20
21
22
23

24
26
27
1981 Total
Asbestos ,
Consumption
(short tons)
87,817e
898
2,347
26,073
826
59,985
11,062
4,315
20,296
9,021
80
14,924
2,612
143

3,105
4,741
9,544
Total
Quantity of
Baghouse Waste
(short tons)
33,220
53f
89s
492
113g
4,354f
4,400
436h
2,311
2,891
52
3,446
877h
N/A

l,788h
N/A
372
Percent of
Asbestos in
Waste (a)
(%)
5.0
83.0
85.0
76.8
80.0
13.9
55.3
11.0
48.8
47.9
10.9
38.4
32.5
40.0

38.3
N/A
48.8
                    - 173 -

-------
                            Table 37 (Continued)

Asbestos Mixture/
Product
Packing
Paints and. Surface
Coatings
Adhesives and Sealants^
Reinforced- Plastics

TSCA
ID*
28
29
30
31
1981 Total
Asbestos ,
Consumption
(short tons)
1,105
15,036
22,557
2,569
Total
Quantity of
Baghouse Waste
(short tons)
206h
96
16
837
Percent o:
Asbestos :
Waste (a)
(%)
31.7
18.3
21.2
31.3
N/A - Not Available.

aEPA 1986b.

bEPA 1986b (Table 3).   This is the total quantity of asbestos consumed per
product category in 1981.  The quantities were reported in ranges as "minimum1
and "maximum" quantities; the average value was calculated and reported in
this table.

CEPA 1986b (Table 15).  This is the total quantity of baghouse fines, dry
waste.  It does not contain 100 percent asbestos fiber.

dEPA 1986b (Table 14).  Assumed the percent of asbestos in baghouse waste is
equal to the percent of asbestos in the total waste.

eThis is the fiber production rather than fiber consumption.

^Baghouse waste was not reported.  Assumed baghouse waste is equal to the
"miscellaneous" waste reported in EPA 1986b (Table 15).

Sfiaghouse waste was not reported.  Assumed baghouse waste is half of the total
waste reported in EPA 1986b (Table 15).

"Baghouse waste was not reported.  Assumed baghouse waste is equal to baghouse
fines, wet reported in EPA 1986b (Table 15).

*-This product category has been reclassified as roof coatings and cements  (ICF
Market Survey 1986-1987).

JThis product category has been reclassified as non-roofing adhesives,
sealants, and coatings (ICF Market Survey 1986-1987).
                                    - 174  -

-------
For each model plant, the following plant parameters were estimated:  annual

production and capacity, asbestos consumed, solid waste generated, annual

operating hours, and number of baghouses and their flow rates.  The basis for

these model plants  is the Section 8(a) submittals, site visits, and general

knowledge of the industry (OAQPS 1987, EPA 1987).  For a few product

categories, more than one model plant is presented (e.g., small, medium, and

large plants), and  the selection of the appropriate model plant is based on

the comparison of the actual fiber consumption level for each plant with those

of the model plants.  Where the asbestos consumption rate is much lower than

that of the small model plant, it is assumed that the plant may have to shut

down a production line due to the decreased demand for the product.*  Thus,

the number of baghouses is decreased accordingly.  When the stack dimensions

and the exhaust gas temperature are not available, default values can be

applied.  A 2-foot  diameter stack is considered  to be typical,** and stack

height normally ranges from 4 to 10 feet above roof level (AIA 1986).  Ambient

temperature of 70°F is an appropriate default value for stack exhaust gas

temperature.

    Using the total quantity of waste collected  from the baghouses  (q) for

each plant, the assumed efficiency (e), and the  percent of asbestos in the

waste (a) from Table 37, emission rates for each plant are calculated using

the emission rate equation.  The result is an emission rate that represents

the total rate of release of asbestos fibers to  the ambient air from each

processing source.  To determine the asbestos release rate from each control
     * This plant may also  operate  intermittently; however, this assumption is
not used because continuous release rates are being estimated.

     ** Office of Research  and Development  (ORD) estimate based on experience
with emissions from  coal-fired boilers  (OAQPS 1987).   (Another source stated
that the stack diameter varies from 1.5 to  5 feet  (AIA 1986) but did not
indicate a typical value.)


                                    - 175 -

-------
device (baghouse), this total emission rate is divided by the number of




baghouses at each plant.  This is an arbitrary assumption since specific




information about each baghouse is not available.  For plants that responded




to the Section 114 letters or the 1CF Exposure Survey, the emission rate is




calculated separately for each baghouse since the inputs to the emission rate




equation were provided for each baghouse.  Although only aggregate emission




estimates are reported in Table 36, the plant-specific emission estimates and




supporting data were provided to Versar, Inc. for use in the ambient exposure




model.




    Due to lack of data, it is not possible to estimate asbestos emissions foi




a number of primary manufacturing sources.  These include primary




manufacturers of specialty paper, asbestos insulation materials (used as




missile liners -- not textile insulation materials), and miscellaneous




products such as acetylene cylinders, sealant tapes, battery separators, and




arc chutes.  Commercial paper, rollboard, roofing felt, flooring felt,




corrugated paper, V/A floor tile, and corrugated A/C sheet are no longer




produced in the U.S. (refer to Appendix A for a discussion of emission factors




for these products).




    B.  Secondary Manufacturing Emissions




        1.  Methodology




        Emission estimates for secondary manufacturing sources are also




derived based on the same emission rate equation used for primary




manufacturing (see Section A.I).  Section 114 data (Section 114 Letters 1985)




do not contain information on secondary manufacturers, and none of-the




secondary manufacturers who responded to the ICF Exposure Survey (1986-1987)




supplied stack data (i.e., stack dimensions, exhaust gas flow rate,




temperature).  The waste quantity collected from the control devices is




estimated based on the waste-to-mixture consumption ratio calculated from the




                                   - 176 -

-------
TSCA Section 8(a) aggregated data  (non-CBI version) for secondary




manufacturing (EPA 1986b).  The operating schedule is assumed to be 8,760




hours per year as in the case of primary manufacturing sources.   Control




device (i.e., baghouse) efficiencies are also based on OAQPS's estimates under




the maximum emission scenario which are 99.95 percent and 99.67 percent for




high and low inlet concentration loadings, respectively (OAQPS 1987).




        2.  Emission Estimates




        Aggregate asbestos emission estimates from secondary manufacturing




sources by asbestos mixture are presented in Table 38.  Although not presented




in Table 38, plant-specific emission estimates and supplementary data on plant




locations, zip codes, and the quantity of asbestos mixture consumed in 1985




(ICF Market Survey 1986-1987) were provided to Versar, Inc. for input into the




ambient exposure model.   In addition, the ratio of total baghouse waste to




asbestos mixture consumed (EPA 1986b) was also provided to Versar, Inc. for




each asbestos mixture.  Table 39 shows the 1981 8(a) data used to calculate




the waste-to-mixture consumption ratios and the percent of asbestos fibers (a)




in the waste (EPA 1986b).




    It should be noted that the Section 8(a) data, the only source of data




available for secondary manufacturers, may not be representative of the




secondary manufacturing plants in  the United States.  EPA was unable to




determine what percentage of the secondary processors actually submitted




reports in response to the Section 8(a) reporting rule.  Therefore, the values




used here may be inaccurate.  Furthermore, the reporting rule exempted certain




classes of potential respondents that process asbestos-containing products.




    A difficulty in analyzing the  values  reported in the 8(a) data is that all




secondary processors reported their  individual products according to the more




general categories of products listed on  the reporting form.  Therefore, a




variety of products are  reported within each product mixture category.  For




                                   - 177  -

-------
                                                        Table  38.  Asbestos Emissions  from Secondary Manufacturing Sources
00
Total Asbestos
TSCA Content
Identification 1985 Asbestos of Wasted
Asbestos Product/Mixture Number Mixture Consumption* (short tons)
Millboard
Beater-Add Gasketing
High Grade Electrical Paper
Drum Brake Linings (LSM)
Disc Brake Pads (LfiM)
Disc Brake Pads (H)
Brake Blocks
Clutch Facings
Friction Materials
Asbestos Cloth, Thread, Yarn, Roving,
Cord, Rope, Hick, etc.
Sheet Gasketing
Packing
Adhesive* and Sealants
Asbestos-Reinforced Plastics
03
05
06
ia
19
20
21
22
24
26
27
28
30
31
157
4,868
17
1,564,830
1,304,000
800
4,000
87,705
253,138
477
845.971
2,114
25
135
.3 tons
.9 tons
.6 tons
pieces
pieces
pieces
pieces
pieces
pieces
.1 tons
sq yd
Ibs.
.4 gallons
.1 tons
30
437
-
11
2
0
0
0
25
2
2
0
2
4
.0
.0

.2
.29
.0014°
.011
.23

.68
.81
.083
. 15E-4
.07
Baghouse
Efficiency
(e)
99
09
99
99
99
99
99
99
99
99
99
99
99
99
.67
.67
.67
.95
.95
.95
.95
.95
.95
.67
.67
.67
.67
.95
Emission Rate
(g/sec)
2.
4.

1.
3.
2.
1.
3.
3.
2.
2.
7.
2.
5.
66E-3
17E-2
-
6 IE- 4
30E-5
02E-8
57E-7
36E-6
60E-4
55E-4
68E-4
68E-6
05E-8
86E-S
Hours of
Operation
(hr/yr)
8,760'
8,760
-
8,760
8,760
6,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
8,760
Total
Emissions
(kg/yr)
90
1,315
-
5.1
1.0
6E-4
5E-3
0.1
11.4
8
B.4
0.2
6E-4
2
                'iCF Market Survey 1986-1987.

                 This product category has been raclasslfied as non-roofing adhesive*,  sealants,  and coatings (ICF Market Survey 1986-1987).

                °The data to calculate the waste-to-fiber consumption ratio for this product category was not available (missing front the 8(a) data --
                see Table 39).  The waste-to-fiber consumption ratio calculated for disc brake pads — light and medium vehicles (19) is assumed.

                Estimated from 1985 asbestos mixture consumption and 1981 Section 8(a) data (EPA 1986b) on waste-to-consumption ratios and asbestos content (see
                Table 39).

-------
                    Table 39.   1981 Secondary Manufacturer Asbestos Mixture Consumption  and Waste Generation"
Asbestos Mixture/
Product (ID*)
Millboard (03)
Beater-Add Gasketlns (OS)
Drum Brake Lining (18)

Disc Brake Pads (LSM) (19)

Disc Brake Fads (H) (20)
Brake Blocks (21)

Clutch Facings (22)
Friction Materials (24)

t
!"* Thread, Yarn, Roving, Cord, Rope,
vo Hick (26)
Sheet Gasketlns (27)

Packing (28)

Adheslves and Sealants* (30)
Asbestos-Reinforced Plenties (31)

End Product Description*5 (ID*)
Other Electrical Products (142)
Sheet Gasketlng (155)
Drum Brakes for Light-Medium
Vehicles (101)
Disc Brake Pads for Light-Medium
Vehicles (102)
N/A
Brake Blocks for Heavy
Equipment (104)
Clutch Facings (105)
Friction Materials (Coomercial
and Industrial) (107)

Brake Blocks for Heavy Equipment
(104)
Sheet Gasketing — Rubber
Encapsulate Compressed (1S6)
Automotive and Friction Components
(113)
Automotive (113)
Asbestos-Reinforced Plastics (203)

1981 Total Asbestos
Mixture Consumption
72 tons
3,941 tons
44,721,663 pieces

22,060,325 pieces

N/A
2,396,352 pieces

3,010,546 pieces
1,878,822 pieces


2,340,000 Ibs
(1,170 tons)
758,584 sq yd

295,320 U>s

1,341,099 gallons
359,156 U»
(179.58 tons)
Total
Quantity of
Baghouse Haste0
(short tons)
31.5
868.7
685.6

180.4

N/A
12.5

35.4
489.4


12.5

6.5

117.2

117.2
28.0

Percent of
Asbestos in
Haste (a)d
(Z)
43.5
40.8
46.7

21. ,5

N/A
52.5

22.5
38.0


52.5

38.8

9.7

9.7
19.3

N/A - Not Available.
'EPA 1986b.
DEPA 1986b (Table 8).
 EPA 1986b (Table 17 — Baghouse Fines Dry).



dEPA 1986b (Table 16).



*This product category has bean reclaisified  as non-roofing adhe.ives,  sealants,  and  coatings  (ICF Market Survey  1986-1987).

-------
example, product mixture number 155  represents "sheet gasketing" products.


All processors of sheet gasketing products provided their information under


this category including processors of beater-add gasketing paper which should

fall under paper products rather than gasketing products.  In this case, the


selection of the product mixture to  be used in calculating the waste-to-


mixture consumed ratio (Table 39) is based on the product mixture with the

highest consumption level in 1981 (Table 8 of EPA 1986b).


    In addition, respondents often reported production using a unit of measure

that differed from the standard unit of measurement for their product.  In


some cases, it is not possible to convert the reported value to a consistent

unit.


    OAQPS developed model plants for three types of secondary processors of


asbestos products:  friction products (i.e., brake rebuilders) , A/C building

products, and A/C or asbestos-silicate boards (OAQPS 1987).  No secondary


manufacturers of A/C building products were identified in the ICF Market

Survey  (1986-1987).  Only the number of baghouses and the exhaust gas flow


rates are available from the model plant.   Therefore, other plant parameters


such as stack dimensions, and exhaust gas velocity and temperature were not


provided to Versar, Inc.  for input into the ambient exposure model.*  There

are no plant parameters available for the other secondary processors which

OAQPS was unable to model.


    The approach used to calculate asbestos emissions from secondary sources

is the same as for primary sources.   The asbestos emissions are calculated for


each plant based on the emission rate equation presented in Section A.I.

Several secondary processors refused to provide the annual mixture consumption
     JU
       The default values discussed in Section A.2 can also be applied to
secondary manufacturing sources.


                                   -  180 -

-------
rate;  therefore, asbestos air releases could not be estimated for these

plants.  Since the number of control devices or baghouses at most of the

plants is not available, the emission rates provided to Versar, Inc. for

ambient exposure modeling represent the total emission of asbestos fibers per

plant rather than per control device.  Only aggregate emission estimates are

reported in Table 38.

    Secondary manufacturing sources for which we are unable to estimate air

releases are the specialty products such as specialty paper, insulation

materials (i.e., missile liner), and the miscellaneous products (e.g., cooling

tower fill, ceramic arc chutes, and fuel cells).  A/C pipe and asbestos

coatings do not require secondary processing.  Asbestos diaphragms are used

within the primary manufacturing plants and, therefore, do not undergo

secondary processing.  In addition, there are currently no secondary

processors of A/C sheet in the U.S.  V/A floor tile, commercial paper,

rollboard, flooring felt, and corrugated paper products are no longer produced

or sold in the U.S. (refer to Appendix A for a discussion of emission factors

for these products).

    C.  Mining and Trade Use Emissions

    This section discusses the methodology used to calculate asbestos air

emissions and the resulting emission estimates for three sources:  mining

activities, automotive brake servicing, and construction related operations.

A brief introduction to each asbestos-generating source is presented, followed

by the methodology and the emission estimates.
     * ICF  (IGF Market Survey 1986-1987) was  able  to estimate the annual
consumption rates  for the  primary processors  that  refused to respond to the
market survey; however,  ICF is unable  at this time to estimate the mixture
consumption rates  for those secondary  refusal plants due to limited data
availability  on the  secondary processors.

                                    - 181 -

-------
    Asbestos mines operate somewhere from 7 to 10 months per year, mainly




during the dry months.  The potential emission points during mining activities



include drilling, blasting, loading ore into trucks, and hauling and dumping




ore into stockpiles at the mills.  Blasting and drilling processes are the




main sources of emission; these processing steps are performed in open pit



mine areas.  There are three facilities currently mining asbestos; they are




the Vermont Asbestos Group, Calaveras Asbestos Ltd., and KCAC Incorporated.




    Automotive servicing includes repair of automotive brakes, clutches,



automatic transmission components, and other friction materials.   However,



exposure data are only available for automotive brakes which are the




automotive products of greatest concern with respect to asbestos exposure;




therefore, only emissions from automotive brake servicing are estimated.



Brake servicing may be performed at service stations, independent repair



shops, new car and truck dealer shops, and self-serviced fleet shops.  There




are approximately 329,000 facilities in the U.S. where brake repair is




performed (Hunter Publishing Co. 1985).  Location and facility-specific



information on automotive repair shops are not presented in this analysis due



to the large number of facilities involved.




    The asbestos-related activities performed in the construction trade




include installation and removal of asbestos material.  This analysis focuses



only on those construction materials that were still being produced or



imported in 1986 which are limited to roofing felt, roof coatings,




asbestos-cement pipe, and asbestos-cement shingles and sheet (flat and




corrugated).  These asbestos construction materials are used for exterior




applications such as roofing, sewer pipe, and exterior walls.




    The construction industry is different than general industry in that the




worksites are temporary in nature and seasonal.  The site conditions, size and




scope of tasks, methods of operation and environmental conditions are quite




                                   - -182 -

-------
varied (CONSAD 1984) .  Some general characteristics of the construction

industry are:

        •  Construction work is performed at temporary locations that
           vary in size, physical boundaries, and working surface;

        •  Construction work is usually performed in open air, subject
           to weather variability;

        •  Construction work varies as the project progresses from
           initiation to completion, demanding a variety of materials,
           equipment and skills; and

        •  Employment is transient in construction, permitting
           tradesmen and laborers to work for several different
           contractors at several different sites per year (CONSAD
           1984).

Due to highly variable nature of construction jobs, detailed information or

site-specific data on construction activities are not available.

    Unlike milling and manufacturing sources, engineering controls for

asbestos air releases are typically not feasible at mining, brake servicing,

or construction sites.  Pollution control equipment such as baghouses are not

applicable, thus there are no stacks.  The points of emissions cannot be well

defined due to the nature and large number of the work settings.  For indoor

jobs, for example, asbestos is emitted through a number of openings such as

windows, doors, and cracks.  For outdoor jobs, airborne asbestos can be

dispersed in any or all directions, depending on the wind vectors.  In other

words, the release configuration from these sources is unconfined with

multiple exits.  All of these factors contribute to the difficulty of making

an accurate assessment of asbestos air emissions from these sources.  As a

result, generalized assumptions must be made to estimate asbestos emissions.

    With respect to mining emissions, an attempt was made to estimate

emissions using the available data on particulate emissions from general

mining operations (non-asbestos) since asbestos emission menasurements are not
                                   - 183 -

-------
available.  A literature search* on the subject resulted in many mining

related documents.   The majority of these reports vere emission studies on

coal mines.  Other non-coal surface mining reports included open pit and strip

mining of clays, cement, sand and gravel, stone, iron, gypsum, and phosphate

rock.  One study which compiled particulate emissions from open sources

estimated a rough emission factor of 2 Ibs/ton of production for all surface

mining (Evans and Cooper 1980).   Another study (Axetell and Cowherd 1984)

developed fugitive dust emission factors for individual coal mining operations

in the form of equations with several correction factors to account for site-

specific conditions.  The resulting emission factors for significant sources

of particulate emissions at surface coal mines which were developed from

extensive sampling at three different Western mines are:  (1) drills --

overburden (1.3 Ib/hole), (2) blasting (35.4 Ib/blast), (3) loading -- coal

(0.037 Ib/ton), (4) dozers -- coal (46.0 Ib/hr), (5) dozers -- overburden (3.7

Ib/hr), (6) dragline (0.059 lb/yd3), (7) scrapers (13.2 lb/VHT**). (8) graders

(5.7 Ib/VMT), (9) light and medium duty vehicles (2.9  Ib/VMT), and (10) haul

trucks (17.4 Ib/VMT) (Axetell and Cowherd 1984).  There are diverse values for

emission factors for surface mining in the published  literature  (Axetell and

Cowherd 1984, EPA 1977) due to problems encountered in sampling mining

sources.  Each mine is unique in its emission characteristics, and the

selection of mines may have influence final emissions  factors.

    Use of emission factors from surface coal mining  sources  (or  non-coal

mining sources) to estimate emissions from asbestos mines would  result in

unreliable emission estimates with a high margin of error.  Asbestos dust is
     * A literature search was performed on-line using APTIC, Pollution
Abstracts, NTIS, and GPO.  Key words used vere mining emission
measurements/estimates, surface mining, and asbestos/non-asbestos.

     ** VMT - vehicle miles traveled.
                                    - 184  -

-------
not similar to coal dust in particle size, geometry, density, etc.  (Asbestos




is unique for its crystal structure/fibrous form.)  Coal dust particles are




quite different from asbestos "dust particles".  Generally, large dust




particles tend to settle out near the source whereas fine dust particles emit




and disperse over much greater distances.  Depending on the sample location,




this factor affects the sampling results and, thus, the emission factors.




Emission factors are extremely source and site-specific (Yocom 1976).  The




type of pollutant and its behavior in the environment are important factors.




There is no basis for assuming that the published emission factors for coal




mining would be appropriate for asbestos mining sources.




    Recognizing the difficulties and uncertainties associated with estimating




asbestos mining emissions, the "bubble" approach used to estimate air releases




in this study (described below) is highly generalized.  One weakness of this




approach is that it does not account for fugitive emissions which occur 24




hrs/day (this weakness applies to both mining and construction sources).  To




date, efforts have been directed at asbestos emissions from well defined




sources and little attention has been given to fugitive emissions from




asbestos mines (Roy 1987).  This is due to the fact that asbestos mines are




further removed from population centers and air emissions are suspected to be




rather low (Roy 1987).  Since there are only 3 asbestos mines in the U.S.,




actual air sampling at these mines should be considered if better estimates




are required.




        1.  Methodology




        The methodology to calculate the asbestos air release rate from




mining, brake servicing, and installation and removal work in the construction




sector is based on a bubble/air exchange rate approach.  The rationalization




for using this method assumes that a worker's activity at a worksite causes




the release of asbestos fibers to the air; these airborne asbestos fibers are




                                   - 185 -

-------
assumed to accumulate in a specified volume of air over a period of time.

This volume of asbestos-contaminated air is replaced periodically (as defined

by the air infiltration rate) by asbestos-free air, thereby transferring

airborne asbestos fibers downwind.  This simplified approach is used due to

the lack of actual emission measurement data.  The emission rates are

calculated based on the following equation:


                 Asbestos Emission - (c) (v) (ac)  (	)
                 Rate (g/sec)                      3600 sec

    where, c - typical asbestos airborne concentration (g/ft^);
                                          o
           v - volume of the work area (ft0); and

          ac - number of air changes per hour (air infiltration rate).

    As shown in the emission rate equation, there are three parameters that

can influence the downwind asbestos concentration levels.  The first

parameter, asbestos airborne concentration (c), is derived from the

occupational exposure analysis (refer to Chapter II).  The projected (i.e.,

based on the new 0.2 f/cc PEL) exposure level (geometric or arithmetic mean)

is assumed for c, which is defined as the typical work area concentration.

The exposure concentration is often expressed as fibers per cubic centimeter

(f/cc); therefore, a conversion factor of 30 fibers/ng  is applied to convert

fiber counts to fiber mass.

    The second parameter in the emission rate equation, the volume of the worl

area (v), is a function of the types of activities performed.  The types of

activities performed determine the work setting (i.e., indoors versus

outdoors) from which asbestos is released.  For indoor work such as brake

servicing, the assumed value for v is based on knowledge of the work setting

related to the activity under study.  For example, brake servicing is
       For additional infc "mation on the fiber-to-mass conversion factor, see
Section B of Chapter I.

                                   - 186 -

-------
typically performed inside a service garage; therefore, a "typical" garage




size is chosen as the work volume.  The work area volume for indoor activities




is assumed to be the volume of the structure in which the work is performed.




    For work performed outside, the selection of a work area or "bubble"




volume is less clearly defined.  Since there are really no physical boundaries




to the work area for outdoor activities (e.g., installation of roofing felt),




this is merely an arbitrary assumption thought to reasonably represent a work




space.  The "work area air volume" can be thought of as an imaginary room or a




cubic air bubble surrounding a worker where airborne asbestos fibers are being




accumulated over a period of time.  A 216 ft^ (or a 6-foot cubic) air volume




or "bubble" is assumed for all outdoor work to represent the work area.  The




selection of this 216 ft^ work area is based on the assumption that the area




should be relatively small and close to the worker's breathing zone.




    The third parameter in the emission rate equation is the air change rate




(ac).  The air change rate or air infiltration rate is dependent upon the




size, shape, and construction of the structure, and upon the pressure




difference between the inside and outside environments.  Infiltration is air




leakage through cracks and interstices, around windows and doors, and through




floors and walls into a building  (ASHRAE Handbook 1977).  One source estimates




the leakage rate for houses ranging from 0.5 to 1.5 air changes per hour in




the winter  (ASHRAE Handbook 1977).  Another source estimates 0.3 to 1.8 air




changes per hour as a typical leakage rate for buildings in the summer and 0.5




to 3.0 air changes per hour in the winter (Perry 1973).  The range of




estimates is due to the variation in the number of windows and doors in the




buildings studied.  For indoor activities, one air change per hour is assumed




to be typical  (ASHRAE Handbook 1977).




    For outdoor activities, the air change rate varies highly, and it is




virtually impossible to measure.  The only basis to estimate the outdoor air




                                   - 187 -

-------
change rate is the wind flow.  The wind flow is used as an indicator of the




dilution rate of asbestos contaminants at the source as they are emitted.  For




this analysis, 4,000 air changes per hour (approximately 66 air changes per




minute) is assumed.  This is a rough estimate based on a long-term average




horizontal wind flow of about 2 meters/sec and a 216 ft-* work area.  The




assumption for the air change rate is directly related to the assumption for




the work area air volume.  If a larger work air volume is selected, a lower




air change rate would be used.




    It is important to emphasize that there are many uncertainties in the




estimation of air releases from mining, brake servicing, and construction




activities.  Many factors can affect the release of asbestos.  Simple




assumptions are made when no data are available and there is a lot of room for




subjective judgment.  The way in which the assumptions are utilized to reach




quantitative estimates makes the numerical values appear more precise than




they actually are.  The asbestos emission rates may vary considerably between




worksites and processes (also during the course of a particular process); the




approach used in this analysis is a simplified one due to.the lack of actual




air emission data.




        2.  Emission Estimates




        Emissions are estimated below for mining, brake repair, and




construction activities using both the geometric and arithmetic mean workplace




asbestos concentrations.  As described in Chapter I (Section A.2), the




arithmetic mean of the raw exposure data is used to estimate health benefits




because it represents total exposure when multiplied by the exposed




population.  This is true for both occupational exposure estimates and ambient




exposure estimates, which are based on the emissions estimates presented in




this section.  Therefore, the emissions estimates for mining, brake repair,
                                   - 188 -

-------
and construction calculated using the arithmetic mean exposure are used to




estimate ambient exposures.




           a.   Mining




           Asbestos-air emissions from mining sources are estimated for the




three mining facilities presently in operation.   The mining methods employed




are the conventional open-pit stripping method (KCAC Incorporated) and the




open-pit bench method (Calaveras Asbestos, Ltd.  and Vermont Asbestos Group).




Open-pit stripping is a surface mining method where the asbestos-containing




ore lies near the surface; therefore, power shovels and bulldozers are used to




remove the ore such that blasting and drilling are not necessary.  In the




open-pit bench method, however, blasting and drilling are required because the




ore body goes deeper and is surrounded by hard rocks.  As mentioned earlier,




blasting and drilling processes are the main sources of dust releases;




however, these operations are performed intermittently as needed.




    The locations of the asbestos mines are shown in Table 40, along with the




emission estimates.  The emissions are calculated based on the emission rate




equation presented in Section C.I above.  The typical asbestos concentration




during mining operations (both geometric mean and arithmetic mean) is obtained




from the occupational exposure analysis (see Chapter II, Section A).  Because




mining operations are performed outdoors, an air bubble volume of 216 ft^ is




assumed with 4,000 air changes per hour.  The average work day for mine




workers is 8 hours; therefore, the 8-hour TWA exposure concentration is




applied.  The duration of asbestos releases is also 8 hours per day, 4 or 5




days per week, for approximately 7 to 10 months per year (refer to Table 40




for the appropriate duration for each mining facility).




    The asbestos emission rates are calculated per mine worker then multiplied




by the number of mine workers to obtained the total emissions per mining site.




As shown in Table 40, the asbestos emission rates calculated for mining




                                   - 189 -

-------
                                                      Table 40.  Asbestos Emissions from Mining Sources
     Mine and Location
                                      Asbestos
                                 Concentration (c)
                                     8-Hour TWA
                                      (f/cc)'
                         Work Area
Geometric   Arithmetic   Volume (v)   Per Hour
  Mean         Mean        (ft3)        (ac)
            Asbestos Emission Rate
Number of         per Worker
   Air      	(it/sec)	
 Changes
                                                                                           Total Asbestos
                                                                                       Bnlaslon Rate Per S
                                                                                       	(K/sec)
  Using       Using      Number of     Ueing       Using
Geometric   Arithmetic    Workers    Geometric   Arithmetic
  Mean         Mean      Per Site"     Mean         Mean
                                                                                                                                                Duration of
                                                                                                                                                 Emissions*
Calavaras Asbestos, Ltd.         0.06         0.06
Copperopolls Mine
Calavera's County, CA  95228

KCAC Inc.                        0.53         0.57
Joe 5 Pit
San Ban!to County, CA  93930

Vermont Asbestos Group, Inc.     0.42         0.54
Lowell Mine
Orleans County, VT  05661
                            216         4,000     1.36xlO~5   1.36xlO~5       25
                            216         4,000     1.20xlO~*   1.29xlO~*
                            216         4,000     9.51xlO~5   1.22xlO~*       16
                                                 3.40xlO~*   3.40xlO~*   8 hr/d, 5d/wk
                                                                         (10 mos/yr)
                                                 3.60x10**   3.87xlO~*   8 hr/d, 4d/wk
                                                                         (7 mos/yr)
                                                 1.52xlo"3   1.96xlo"3   8 hr/d, 186 d/yr
"Refer to Chapter II, Section A.  The weighted average exposure without usage of respirators is the appropriate exposure to use because we are Interested in
the concentration of asbestos fibers in the work area.

-------
                            -4                   -3
sources range from 3.40 x 10   g/sec to 1.52 x 10   g/sec.  It should be noted



that the rate of asbestos emission at any time may be higher than the



estimated value because the short-term peak exposure concentration is often



higher than the 8-hour TWA concentration.  Emissions from mining sources are



assumed to be area source emissions rather than point source emissions.  One



mine (Vermont Asbestos Group) estimated an area of 2300+ acres as the total



area of the facility including the area of asbestos deposits, with 10 acres as



the portion where asbestos operations are performed (ICF Exposure Survey



1986-1987).  Since there are no stacks, the release velocity depends on the



magnitude of the wind flow which is a function of the meteorological



conditions at a particular time.  Ambient air temperature is assumed at the



release point.  Fiber size distribution data are not available.



           b.  Brake Repair



           Asbestos air emissions are estimated for four types of brake



servicing jobs:  drum brake shoes for cars (and light trucks), drum brake



shoes/brake blocks for trucks, disc brake pads for cars (and light trucks),



and disc brake pads for trucks.  Based on a 1984 survey of a cross-section of



automobile and truck servicing facilities performed by the Hunter Publishing



Co. (1985), the majority of brake repair work is done at  independent repair



shops  (45.6 percent) and service stations (35 percent).  Table 41 shows the



distribution of brake repair work by types of facilities.  The installation of



disc brake pads and drum brake shoes on trucks represents a small fraction of



the total brake repair work.  Approximately 60 percent of the brake servicing



shops  are located in cities with population over 25,000 people and 40 percent



with population under 25,000  (Hunter Publishing Co. 1985).  Since this  is a



service industry, it is reasonable  to assume that brake servicing shops are



located mainly in populated areas.
                                    -  191  -

-------
                                                       Table 41.   Distribution of Brake Repair  Work  by  Types of Facilities
fO
Total Number
Type of Facility of Facilities
Service Stations 115,000
Independent Repair 150,000
Shops
New Car end Truck 25,000
Dealer Shops
Self -Serviced Fleet 39,000
Shop
Total 329.000
Number of
Drum Brake Shoes
(Axle Sets) Installed
As Replacements
on Cars (1989)"
9,774,140
11,726,960
3,157,800
400,990
25,061,890
Number of
Drum Brake Shoes
(Axle Sets) Installed
As Replacements
on Trucks (1985)'
123,970
238,230
74,260
134,820
571,280
Number of
Disc Brake Pads
(Axle Sets) Installed
As Replacements
on Cars (1989)'
6,854,070
8,377,190
2,647,340
253,850
18,132,450
Number of
Disc Brake Pads
(Axle Sets) Installed
As Replacements
on Trucks (1985)'
5,290
8,940
3,330
2,040
19,600
                          'These value* represent the number of asbestos lining* for drum brake shoe*  and disc brake  pad*.  The  total number  of  asbestos
                          linings for each category is based on Tzanetos et al.  1987 and ICF Market Survey (1986-1987)  (see Chapter  II, Section  D).  The
                          distribution of the total number of breke jobs by type of facility ie based  on the  results  of the 1964 survey performed by Hunter
                          Publishing (1985).

                          Source:  Hunter Publishing Co.  1985 and 1986,  Tzanetos et al.  1987,  ICF Market Survey  1986-1987, ICF 1987.

-------
    Using the emission rate equation presented in Section C.I above,  asbestos

air emissions from brake servicing shops are calculated.   The geometric mean,

8-hour TWA area concentration found at brake repair shops is 0.03 f/cc; the

arithmetic mean, 8-hour TWA area concentration is 0.04 f/cc (see Chapter II,

Section D).   The area concentration is used to calculate asbestos emissions,

rather than the personal breathing zone concentration (0.09 f/cc geometric

mean; 0.15 f/cc arithmetic mean).  Since airborne fibers are continuously

dispersed into a confined work area air volume, it is more appropriate to use

the area concentration.  Over a period of time, the area concentration can be

thought of as the equilibrium concentration of the personal concentrations.

The personal exposure concentration varies during brake servicing but the

8-hour TWA area concentration is assumed to be constant throughout the day.

    Whether it is a car servicing job or a truck servicing job, the typical

asbestos concentration level is assumed to be the same because of the

similarity of the brake servicing procedures (see Chapter II, Section D).  The

difference between car versus truck servicing estimates then is in the

assumptions of the shop areas.  Truck servicing facilities require more  shop

space than car servicing facilities.  A 3-bay garage is assumed to be  typical

for brake repair shops/service stations.*  The shop area is estimated  to be

about 900 square feet  (each bay is 15 ft x 20 ft) with a ceiling height  of 10

ft for car servicing shops.  Thus, the total work area air volume is 9,000

ft3.  Truck servicing  shops are also assumed to have 3 bays, each with a 25  ft

x 30 ft area and a 20  ft ceiling height, which yields a total work area  air

volume of 45,000 ft3.  Because of the variation in shop sizes, one air change
     *  Since over 80 percent of brake  repair work  is performed at  service
 stations and independent repair shops, the assumption  for  the work area
 is based on what is typical for these  facilities.

                                    - 193  -

-------
per hour is assumed for car servicing shops whereas 1.5 air changes per hour


is assumed for truck servicing shops.



    The average time for a brake job ranges from 1 to 3 hours (Chapter II,


Section D) -(the actual time that a mechanic would spend on brake jobs in a day



is not known).  It is assumed that mechanics (full-time equivalents) perform



multiple brake jobs a day for a full 8-hour day.  The duration of release is,



therefore, also 8 hours per day.



    Table 42 shows the emission estimates for brake servicing.  For both the



installation of drum brake shoes and disc brake pads on cars, the estimated


asbestos emission rates using geometric mean and arithmetic mean area


                            -8                    -8
concentrations are 7.08 x 10"  g/sec and 9.44 x 10" , respectively.  For both


drum brake shoe and disc brake pad installation on trucks, the estimated



emission rates using geometric mean and arithmetic mean area concentrations


are 5.31 x 10"  g/sec and 7.08 x 10" ,  respectively.



    Since the data from the Hunter Publishing Co. (1985) gives a breakdown of



the number of brake jobs performed in each regional area in 1984, it may be


more appropriate to model asbestos emissions for each regional area (since



facility-specific information is not available).  Tables 43, 44, 45, and 46



present the number of brake jobs performed in each region, and the amount of



asbestos released per region for drum brake shoes on cars, drum brake shoes on


trucks, disc brake pads on cars, and disc brake pads on trucks, respectively.



Based on the typical time required per brake job (discussed in Chapter II,



Section D),  the total time spent on brake repair work in each region is



calculated.   This number is then applied to the emission rate calculated


earlier (Table 42) to obtain grams of asbestos released for each region per


year.



    From Tables 43,  44, 45, and 46, it can be seen that the East North Central



region has the highest asbestos release rate for brake servicing work on cars,


                                   - 194 -

-------
                                                                 Table 42.   Asbestos Emissions  from Brake Repair
vo
Ul
                                                                    Asbestos Area
                                                                  Concentration (c)                                 Asbestos Emission Rate
                                                                      8-Hour TWA                      Number of    	U/sec)	
                                                                       (f/cc)a	   Work Area    Air Changes     Using      Using
Geometric
Job Description Mean
Drum Brake Shoe Installation on Cars 0.03
Drum Brake Shoe Installation on Trucks 0.03
Disc Brake Pad Installation on Cars 0.03
Disc Brake Pad Installation on Trucks 0.03
Arithmetic Volume (v) Per Hour Geometric Arithmetic Effective Duration
Mean (ft ) (ac) Mean Mean of ' Exposure
0.04 9,000 1 7.08xlO~8 9.44xlo"8 8 hr/d, 5 d/wk, 250 d/yr
0.04 45,000 1.5 5.31xlo"? 7.08xlo"7 8 hr/d, 5 d/wk, 250 d/yr
0.04 9.000 1 7.08xlo"8 9.44xlO~8 8 hr/d, 5 d/wk, 250 d/yr
0.04 45,000 1.5 5.31xlo"7 7.08xlo"7 8 hr/d, 5 d/wk, 250 d/yr
                       •Refer to Chapter II, Section D (Table 27).

-------
                                                       Table 43.  Asbestos Emission Estimates In g/yr for Each Region --
                                                                    Installation of Drum Brake  Shoes  on Cars
vO
Region
New England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Centra
Mountain
Pacific
Total
Number of Replacement
Asbestos Drum
Brake Shoes Installed
Per Year (1989)*
1.723,530
3,773,890
4,487,270
2,770,830
3,688,080
949,480
12,141,240
1,518,670
4, 004,, 900
25,061,890
Number of
Establishments
Per Region
22,650
49,570
58,910
36,370
48,420
12,460
28,110
19,940
52,570
329,000
Total Servicing
Hours Far Year
Per Region0
(hours)
2,588,300
5,663,840
6,730,900
4,156,250
5,532,120
1,424,220
3,211,860
2,278,000
«. 007. 350
37,592,840
Emission Estimate Per Region
(it/year)
Using
Geometric Mean
659.7
1,443.6
1,715.6
1,059.3
1,410.0
363.0
818.6
580.6
1.531.2
9,581.6
Using
Arithmetic Mean
879.6
1,924.8
2,287.4
1,412.5
1,880.0
484.0
1,091.5
774.2
2.041.5
12.775.5
                          *The total number of replacement drun brake shoes installed on cars in 1989 la based on Tzanetos et el. (1987) (see
                          Chapter II, Section 0).  The distribution of the total number of drum brake jobs on cars by region is based on the results
                          of  the 1984 survey performed by Hunter Publishing (1985).

                           This is an estimate.  The average number of brake jobs per establishment per year is calculated using the data presented
                          in  Table 41 (25,061,890 total Jobs/329,000 facilities equals approximately 76.176 brake jobs par establishment);  this
                          number is divided by the number of brakes installed per year par region to obtain the number of establishments for each
                          region.  Consequently,'all brake repair facilities (e.g., service station or self-serviced fleet shops) are assumed to
                          perform the same number of brake jobs a year.

                          °The total hours spent on brake repair work is calculated based on an estimated 1.5 hours per drum brake Job (see
                          Chapter II, Section D).

-------
                                                        Table 44.  Asbestos Emission Estimates In g/yr for Each Region —
                                                                    Installation of  Drum Brake Shoes on Trucks
vo
Region
New England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Central
Mountain
Pacific
Total
Number of
Asbestos Drum
Brake Shoes Installed
Per Year (1985)'
31 , B20
77,990
90,590
68.910
81,970
21,570
55,500
51,530
91.400
571,280
Number of
Establishments
Per Region
18,330
44 , 920
52,170
39,680
47,200
12,420
31,960
29,680
52.640
329,000
Total Servicing
Hours Per Year
Per Region0
(hours)
79,550
194,960
226,470
172,280
204,920
53,930
138,750
128,820
228.500
1,428,200
Bnisslon Estimate Per Region
(x/year)
Using
Geometric Mean
152.1
372.7
432.9
329.3
391.7
103.1
265.2
246.3
436,8
2,730.1
Using
Arithmetic Mean
202.8
497.0
577.2
439.1
522.3
137.5
353.6
328.3
582.4
3,640.2
                           "The total number of drum brake shoes installed on trucks in 1965 it based on the ICF Market Survey 1986-1987)  (see
                           Chapter II, Section D).  The distribution of the total number of drum brake Jobs on trucks by region is baaed on the
                           results of the 1984 survey performed by Hunter Publishing (1985).

                            This is an estimate.  The number of brake Jobs per establishment per year la calculated using the data presented In
                           Table 41 (571,280 total asbestos jobs/329,000 facilities equals approximately 1.736 brake Joba per eatabllahment);  this
                           number is divided by the number of brakes Installed per year per region to obtain the number of establishments  for each
                           region.  Consequently, ell brake repair facilitiea (e.g., service stations or self-serviced fleet  shops) are assumed to
                           perform the same number of brake Joba a year.

                           cThe total hours spent on brake repair work la calculated based on an estimated  2.5 hourr per drum brake Job (see
                           Chapter II, Section D).

-------
                                                        Table 45.  Asbestos Emission Estimates in s/yr for Each Region —
                                                                     Installation of Disc Brake Pads on Cars
vo
00
Region
Hew England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Central
Mountain
Pacific
Total
Number of Replacement
Asbestos Disc
Brake Shoes Installed
Per Year (1989)*
1,268,580
2,700,780
3,244,870
2,002,890
2,653,960
650,480
1,629,380
1,113,620
2. 867. 690
18,132,450
Nunber of
Establishments
Per Region
23,020
49,000
58,880
36,340
48,150
11,800
29,560
20,210
52.040
329,000
Total Servicing
Hours Per Year
Per Region0
(hours)
1,385,440
2,970,860
3,569,360
2,203,180
2,919,350
715,530
1,792,320
1,224,980
3.154,680
19,945,700
Emission Estimate Per Region
(«/vear)
Using
Geometric Mean
355.7
757.2
909.8
561.5
744.1
182.4
456.8
312.2
804,1
5,083.8
Using
Arithmetic Mean
474.2
1,009.6
1,213.0
7*8.7
992.1
243.2
609.1
416.3
1.072.1
6,778.3
                           *The total number of replacement disc brake pads installed on cars In 1989 ia based on Tzanetos  et al.  (1987)  (see
                           Chapter II, Section D).  The distribution of the total number of disc brake Jobs on cars by  region is based on the
                           results of the 1984 survey performed by Hunter Publishing (1985).

                            This is an estimate.  The number of brake jobs per establishment per year is calculated using the data presented  in
                           Table 41 (16,132,450 total asbestos Jobs/329,000 facilities equals approximately 55.114 brake Jobs per  establishment);
                           this number Is divided by the number of brakes installed per year per region to obtain the number  of establishments for
                           each region.  Consequently, all brake repair facilities (e.g., service stations or self-serviced fleet  shops)  are  assumed
                           to perform the same number of brake Jobs a year.

                           cThe total hours spent on brake repair work is calculated based on an estimated 1.1 hours per disc brake  Job  (see
                           Chapter II, Section D).

-------
                             Table 46.   Asbestos Emission Estimates in g/yr for Each Region --
                                         Installation of Disc Brake Pads on Trucks
Region
New England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Central
Mountain
Pacific
Total
Number of
Asbestos Disc
Brake Shoes Installed
Per Year (1985 )a
1,360
2,660
3,170
2,340
2,410
730
2,280
1,680
2.970
19,600
Number of
Establishments
Per Region*1
22 , 830
44,650
53.210
39,280
40,450
12,250
38,270
28,200
49.860
329,000
Total Servicing
Hours Per Year
Per Region0
(hours)
3,400
6,650
7,930
5.850
6,020
1,830
5,700
4,200
7.420
49,000
Emission Estimate Per Region
(K/year)
Using
Geometric Mean
6.5
12.7
15.2
11.2
11.5
3.5
10.9
8.0
1
-------
whereas the Pacific and East North Central regions have the highest annual




asbestos release rate for brake servicing work on trucks.




    Data on the fiber size distributions at brake repair facilities is




limited. -Only one industrial hygiene survey (NIOSH 1982b) reported fiber si;




distributions from vehicle brake servicing operations.  The fiber size




distribution performed for all airborne sample fibers, observed by




transmission electron microscopy (TEM),  showed a geometric mean length of




1.66 urn.  For airborne sample fibers identified as chrysotile, a geometric




mean length and diameter of 1.7 urn and 0.15 urn, respectively, were found.  TV




TEH analysis also showed that 80 percent of the total fiber population was




less than 5 urn in length and that 30 percent of the fibers were chrysotile,  i




percent forsterite, and 50 percent unknown.




    There are no stacks at brake repair shops.  Asbestos may be released




through a number of openings, depending on the worksites.  The multi-release




points and velocities are not known.  Exhaust temperature is assumed to be




ambient air temperature.




           c.  Construction




           Asbestos air emissions for each construction activity are




estimated.  Two types of activities are analyzed for each product category:




installation and removal.  Installation operations are analyzed for A/C pipe,




A/C flat sheet, A/C corrugated sheet, A/C shingle, and built-up roofing;




removal operations are analyzed for A/C flat sheet, A/C corrugated sheet, A/C




shingle, and built-up roofing (exposure estimates for A/C pipe repair or




removal are not available).  As discussed earlier, all of the construction




related activities are exterior applications (i.e., outdoor activities); thus




a 6-foot cubic air volume (v) (i.e., 216 ft^ air bubble) is assumed to




represent the work area, with an air change rate (ac) of 4,000 per hour.  The




geometric and arithmetic mean short-term TWA exposure concentrations (c) (as




                                   - 200 -

-------
derived in Chapter II, Section E) are also applied.  These three variables are

applied to the emission rate equation presented in Section C.I to obtain the

asbestos emissions for construction.  The emission rate calculated for each

activity is one that is generated as the result of a single worker's activity

(e.g., cutting or drilling) at a site.  To obtain the total annual emissions

for any construction activity, the emission rate is multiplied by the total

hours spent on construction per year.  Since the actual time required per job

and the number of workers per job are not known, the total hours spent on

construction activities are taken as the product of the number of full-time

equivalent workers and a full working year (8 hours a day for a total of 250

days per year).  Table 47 shows the estimated asbestos emission rates for

construction activities.

    No other information concerning the emission sources are available.  A/C

sheets and shingles, and built-up roofing materials are generally applied to

(or removed at) residential and commercial installations; therefore, the

locations where these activities take place are mostly in surburban areas.

A/C pipe installation and removal operations are likely to be in both

surburban and  rural areas.

    Based on location information provided in the  NIOSH,  industry, and

academic studies used to develop the geometric mean exposures,  several job

sites where asbestos construction work has been undertaken include:

        •  A/C Pipe -- Alameda, CA;

        •  A/C Sheet  -- Denver, CO  and Steamboat Springs, CO;

        •  A/C Shingle  -- Waycross, GA and Rockford, IL;

        •  Build-up Roofing -- New  Orleans, LA; Hobart, IN; Racine,
           WI; Indianapolis,  IN; Allentown, PA; Gary,  IN; Anderson,
           IN; and Chesterfield, IN.
                                    -  201  -

-------
                                                            Table 47.  Asbestos Emissions from Construction Activities
NJ
O
10
Short-Term TWA
Asbestos
Concentration (c)
(f/cc)a
Construction Activity
A/P Pipe Installation
A/C Flat Sheet Installation
A/C Flat Sheet Removal
A/C Corrugated Sheet
Installation
A/C Corrugated Sheet Removal
A/C Shingle Installation
A/C Shingle Removal
Built-Up Roofing Installation
Built-Up Roofing Removal
Geometric
Mean
0.080
0.173
0.800
0.173
0.800
0.046
0.084
0.128
0.072
Arithmetic
Mean
0.114
0.278
0.800
0.278
0.600
0.050
0.094
0.169
0.114
Work Area
Volume (v)
(ft3)
216
216
216
216
216
216
216
216
216
Number of
Air Changes
Per Hour
(ac)
4,000
4,000
4,000
4,000
4,000
4,000
4,000
4,000
4,000
Asbestos Emission Rate
(K/»»C)
Using
Geometric
Mean
i.aixio'5
3.92xlO~3
1.81x10'*
3.92xlO~5
l.Blxlo"*
1.04xlO~S
1.90xlO~5
2.90xlO~5
1.63xlO"S
Using
Arithmetic
Mean
2.58xlO~5
6.30xlo"3
1.81xlO~*
6.30xlo"5
1.81xlO~*
1.13xlo"5
2.13X10*5
3.83xlO*5
2.58xlO~5
Total Construction
Hours ,Pec gear
(hours )b
1,842,000
32,000
40,000
14,000
18,000
472,000
328,000
792,000
526,000
Total Asbestos Emissions
(k*/veer)c
Using >
Geometric
Mean
120.0
4.5
26.1
2.0
11.7
17.7
22.4
82.7
30.9
Using
Arithmetic
Mean
171.1
7.3
26.1
3.2
11.7
19.2
25.1
109.2
48.9
             'Projected under the new 0.2 f/ce PEL.   Short-term exposures are used to correspond with the full-time  equivalent worker population used to calculate
             total emissions.  (See Chapter II, Section E.)

             Tlefer to Chapter II, Section E.   The total construction hours per year is  the product of the number of full-time equivalent workers and the length of  a
             full-working year (2,000 hours/year).

             °The total emissions would be distributed throughout the U.S.

-------
Asbestos construction work is by no means limited to these locations,  but




these locations can be used to model likely ambient exposures and exposed




populations.




    D.  Emissions from Asbestos-Containing Waste Piles




    Asbestos-containing wastes are generated during a variety of processes,




including the mining and milling of asbestos ore, the manufacture and




fabrication of asbestos products, and the installation/demolition/renovation




of asbestos building materials.  The handling, transport, and disposal of




these wastes are regulated by the U.S. Environmental Protection Agency (EPA)




and the Occupational Safety and Health Administration (GSHA).  These




regulations are found in the Code of Federal Regulations (40 CFR Part 61,




Subpart M and 29 CFR Parts 1910 and 1926).  OSHA's requirements governing




asbestos-containing wastes are designed to protect employees from exposure to




asbestos fibers that are released at the worksite.  EPA also sets forth




regulations governing the handling of asbestos wastes onsite.  In general,




EPA's regulations are found to be very similar to OSHA's.  The disposal of




asbestos wastes off-site is regulated by EPA.  A discussion of the types of




asbestos-containing wastes that are generated and the regulations governing




their handling is presented below, followed by estimates of emissions from




mining/milling waste piles.




        1.  Regulations Affecting Management of Asbestos Wastes




           a.  Mining and Milling Wastes




           Asbestos is  'manufactured1 by mining the ore deposit and separating




the fibers from the non-asbestos rock.  The process of separating the fibers




from  the mined ore, and grading and packaging these fibers,  is called milling.




The following three types of wastes are generated from the mining and milling




process:
                                    -  203  -

-------
        •  Mine waste;
        •  Tailings; and
        •  Wastes from air cleaning devices.

    Mine wastes are defined as overburden, or waste rock, having insufficient

asbestos for additional processing (i.e., milling).  Approximately 4 million

metric tons of mine waste were generated in 1982 (EPA 1985b).  This waste is

typically piled in an area adjacent to the mine.  EPA does not regulate the

handling of these wastes since the asbestos content of mine waste is extremel;

low (approximately 0.1 percent).  Mills also generate waste rock, called

tailings, that contain residual amounts of asbestos.  The asbestos content of

tailings is higher than that of mine waste (approximately 1.4 percent).  It

has been estimated that between 1 and 2 million metric tons of tailings were

generated in 1982 (EPA 198Sb, EPA 1986d).  Wastes from air cleaning devices

generated approximately 30,000 tons of wastes in 1981; the average asbestos

content of these wastes was found to be 5 percent (EPA 1986b).  According to

one source, a ratio of 16.5 tons of asbestos waste is generated during millinj

operations for every one ton of asbestos produced (EPA 1986d).  Production

figures submitted to EPA in 1981 place this ratio at 15 tons of waste per ton

of asbestos produced (EPA 1986b).

    Tailings are disposed by loading on a conveyer belt and dumping on a wast*

pile,  usually located within 5 miles of the mill.  Under 40 CFR Part 61,

Subpart M, EPA requires that during the collection, processing, packaging,

transport or disposal of asbestos-containing waste, there either be no visibl'

emissions to the outsite air or a wetting agent or air cleaning device be use'

to control the dust.  Emission control during transport and dumping is usuall}

achieved by enclosed conveyors, negative air hoods, or the use of a wetting

agent (i.e., water) introduced via a mixing screw conveyor.  An alternative

disposal method nay be used if prior approval by the Administrator has been

received.

                                   - 204 -

-------
    Waste piles are regulated under 40 CFR §61.156.  A waste pile is an

acceptable site for the disposal of asbestos wastes if either there are no

visible emissions to the outside air or the waste is covered within 24 hours

of deposit.  The minimum required cover is 6 inches of non-asbestos material,

normally soil or a resinous or petroleum-based dust suppression agent.  An

alternative control method for emissions may be used with prior approval by

the Administrator.  In addition to these federal requirements, many state or

local agencies require more stringent waste handling procedures.

           b.  Manufacturing and Fabricating Wastes

           Asbestos products are manufactured by combining the milled asbestos

fibers with binders, fillers, and other materials.  The resultant mixture is

molded, formed, and then cured or dried.  Some products require further

machining or coating operations prior to their sale.  This process is called

fabrication or secondary manufacture.  Manufacturing and fabricating

operations generate the following four types of asbestos-containing wastes:

        •  Empty asbestos shipping containers;

        •  Process wastes such as cuttings, trimmings, and
           off-specification, reject material;

        •  Housekeeping waste from sweeping or vacuuming; and

        •  Pollution control device waste from dust capturing systems.

It has been estimated that approximately 88,500 tons of asbestos-containing

wastes were generated from manufacturing/fabricating operations  in 1981.

According  to one source, a ratio of 0.25 tons of asbestos waste  is generated

during manufacturing/fabrication operations for every one ton of asbestos

consumed  (EPA 1986d).

    The handling and disposal of manufacturing and fabricating wastes are

regulated  under 40  CFR  §61.152 and §61.156.  The same standards  apply during

the collection, processing, packaging,  transport,  and deposit of


                                   -  205  -

-------
asbestos-containing wastes generated by manufacturing and fabricating

operations as during mining and milling operations.   However, in addition,

both EPA and OSHA require that asbestos-containing wastes from

manufacturing/fabrication operations be containerized to avoid creating dust

during transport and disposal (EPA 1986e,  OSHA 1986a).   Both EPA and.OSHA

specify that the containers be labeled as  follows:

                                   CAUTION
                           Contains Asbestos Fibers
                     Avoid Opening or Breaking Container
                Breathing Asbestos is Hazardous to Your Health

                                     or

                                   CAUTION
                           Contains Asbestos Fibers
                             Avoid Creating Dust
                        May Cause Serious  Bodily Harm

    EPA recommends that process wastes and housekeeping wastes be wetted

before packaging.  Air pollution control device wastes are usually packaged

directly by connecting a container to the  waste hopper outlet.  Vacuum bags

and disposable paper filters should not be cleaned,  but rather wetted and

placed intact into a proper container.   EPA also recommends that empty

shipping drums not be cleaned; instead, the drums should be sealed and

properly disposed or used to contain other asbestos wastes for disposal.

Empty shipping bags can be flattened and packaged under hoods exhausting to a

pollution control device (EPA 1985a).

           c.  Installation Wastes

           OSHA regulates installation wastes under the general category of

'construction' wastes.  According to OSHA, construction operations include

installation, removal, and demolition activities.  These wastes are regulated

by OSHA under 29 CFR Parts 1910 and 1926.   The regulations state that "all

asbestos waste, scrap, debris, bags, containers,  equipment, and contaminated
                                   -  206 -

-------
clothing must be collected and disposed of in sealed impermeable bags or in

other closed impermeable containers"  (51 FR 22725) (OSHA 1986a).

    OSHA does not, however, address the ultimate disposal of these wastes

"since these wastes and their disposal are regulated by EPA" (51 FR 22726)

(OSHA 1986a).  EPA, on the other hand, does not regulate the disposal of these

wastes because it considers the quantities of asbestos-containing products

currently being installed in the U.S. to be insignificant (EPA 1986C).  EPA

does regulate, along with OSHA, wastes generated from demolition and

renovation operations; these regulations are discussed below.

           d.  Demolition and Renovation Wastes

           A large amount of asbestos-containing waste may be generated during

the removal of asbestos products from buildings.  As mentioned above, both EPA

and OSHA regulate demolition and renovation wastes.  EPA regulations

specifically address the removal of 'asbestos-containing waste'.  This term

must be understood properly in order  to understand exactly which types of

wastes are regulated by EPA.  According to EPA:

        "Asbestos-containing waste materials means any waste that
        contains commercial asbestos and is generated by a source
        subject to the provisions of  this subpart.  This term includes
        asbestos mill tailings, asbestos waste from control devices,
        friable asbestos waste material, and bags or containers that
        previously contained commercial asbestos.  However, as applied
        to demolition and renovation operations, this term includes
        only friable asbestos waste and asbestos waste from control
        devices" (40 CFR, §61.141, Subpart M) (EPA 1986e).

According to the above definition, asbestos waste generated during demolition

and renovation operations is considered 'asbestos-containing waste' (and is,

therefore, regulated) only if the waste is friable.*  For example, waste

generated during asbestos-cement pipe manufacture is regulated under 40 CFR,
     * Friable asbestos material refers to "any material containing more than
1 percent asbestos by weight that hand pressure can crumble, pulverize, or
reduce to powder when dry" (40 CFR, §61.141, Subpart M).

                                   - 207 -

-------
Subparc M; however, waste gent^acta during the removal of this same product is




not regulated.  It has been estimated that approximately 12,700,000 tons of




asbestos-containing wastes have been generated during demolition and




renovation operations between the years 1880 and 1985 (EPA 1986d).




    EPA and OSHA regulations require that when the asbestos materials are




removed, they be containerized to avoid creating dust during transport and




disposal.   The generally recommended containers are 6-mil thick plastic bags,




sealed in such a way to make them leak-eight.  More thorough containerization




may include double bagging, plastic lined cardboard containers, or plastic-




lined metal containers.   Asbestos waste slurries may be too heavy for plastic




bags, and can be packaged in leak-tight drums.  In situations where pipes or




other facility components are removed as sections without first removing the




asbestos,  6-mil plastic can be used to wrap the section sufficiently to create




a leak tight container (EPA 1985a).  The sane requirements for handling and




disposing of manufacturing wastes apply to demolition and renovation




'asbestos-containing wastes'.




    After the asbestos-containing materials have been removed, all plastic




barriers should be removed and the facility should be thoroughly washed.  The




plastic used to line the walls, floors, etc. should be treated as asbestos




waste and containerized appropriately.  Any asbestos-containing wastes




collected by cleaning devices must be appropriately bagged, labeled, and




disposed.




        2.  Emission Estimates from Mining/Milling Waate Piles




        Due to the regulated handling of wastes from manufacturing and




fabrication, installation, and demolition and renovation of asbestos products,




emission of asbestos from these wastes is not expected to be significant.




Wastes from these operations are containerized for disposal in landfills; and




should not, therefore, emit asbestos unless the containers are damaged.




                                   - 208 -

-------
    Asbestos emissions are most likely from mining and milling wastes which




are not containerized, but are instead disposed on-site in waste piles or




trenches.  Furthermore, mining wastes are not expected to emit significant




quantities of asbestos because of their low asbestos content (approximately




0.1 percent).  Therefore, only emissions of asbestos from milling wastes




(i.e., tailings and air cleaning device wastes) are estimated.  Note that the




quantity of air cleaning device waste is small compared to the quantity of




tailings; air cleaning device wastes only account for about 3 percent of the




total waste from the mills.




    Very little data on actual emissions from asbestos waste disposal




operations exist.  For this reason, the Emissions Standards and Engineering




Division of EPA has developed a methodology for estimating these emissions




(EPA  1986f).  Emissions from the disposal of tailings have been estimated for




the three active U.S. milling operations.  Following is a description of




potential emission points during asbestos waste disposal operations, the




methodology used to estimate emissions, and the resulting emission estimates.




           a.  Potential Emission  Points During Waste Handling Operations




           A summary  of current waste generation  and waste management




practices at the three milling facilities  is shown  on Table 48.  Waste




management practices  differ only slightly between the mills.




    Tailings generated by  the Vermont Asbestos Group are wetted (using  water




and a dust  suppression agent) and  then  dumped  onto  waste piles,  located




adjacent to  the mill.  The process is continuous, with tailings  being




transported to the waste pile via  conveyor (ICF Exposure  Survey 1986-1987,  PEI




Associates  1984).  The wetting agent prevents  blowing  and  solidifies the




wastes for  permanent storage.  The wastes  are, therefore,  not transported to a




landfill for ultimate disposal  (PEI Associates 1984, Vermont  Asbestos Group




1987).



                                    - 209 -

-------
   Table 48.  Summary of Current Waste Generation from Milling Operations
                        Vermont
                    Asbestos Group
                   (Morrisville,  VT
                        05661)
                                           Calaveras
                                        Asbestos,  Ltd
                                       (Copperopolis,  CA
                                            95228)
                        KCAC Inc.
                      (King City, CA
                           93930)
Annual Waste
Generation
                   219,366 TPY
Asbestos Content   <0.05%
Disposal Method
                   Waste pile --
                     tailings are
                     wetted
  Number of Files  3

  Size of Piles    100 Ac
  Location of
  Piles
                   Adjacent to the
                     mill
                                       645,000 TPY
Waste pile  --
  tailings  are
  wetted

1

110 Ac

Adjacent to the
  mill
                     13,000 TPY
                                                            Waste trench - -
                                                              covered with
                                                              earth seal

                                                            1

                                                            12 Ac

                                                            1/4 - 1/2 mile
                                                              from the mill
TPY - tons per year.
AC
     acre.
aNo value for the asbestos content was obtained; therefore, we assumed the
more conservative of the two values obtained for the other mills.

Sources:  ICF Exposure Survey 1986-1987,  PEI Associates 1984, Vermont Asbesto
          Group 1987, Calaveras Asbestos  1987, KCAC Inc. 1987, OAQPS 1987.
                                   - 210 -

-------
    Tailings from milling operations at Calaveras Asbestos are transported

from the mill via conveyor to a waste pile, located approximately 300 ft away.

Two wet screws are used to mix water and a binding agent into the tailings.

The last 150-foot section of the conveyor contains water sprays which are used

to saturate the tailings so that by the time the tailings reach the end of the

conveyor they are in the form of a slurry.  The tailings are then dumped onto

the waste pile for permanent storage (Calaveras Asbestos, Ltd. 1987).

    Tailings generated by KCAC Inc. are disposed in a trench, located between

1/4 and 1/2 of a mile from the mill.  The trench is essentially a hole used to

bury the wastes.  Due to its extensive use over the years, the stored waste

currently extends above ground.  The wastes are transported via conveyor from

the mill to a dump truck and then transported to the disposal site.  KCAC,

unlike the two other milling operations, uses a wet process for milling.   The

tailings leave the mill as a slurry, doing away with the need to add

additional wetting agents prior to disposal.  The wastes are continuously

delivered to the disposal site.  The tailings are covered with two  or  three

feet of topsoil approximately once a month to prevent erosion and exposure.

They are not covered more frequently since they  are wetted and solidify into a

crusty material when dry  (KCAC Inc. 1987).

    There are  three points along the waste disposal process  from which

potential emissions may occur; they are  as follows:

        1.  Transfer of tailings onto and between conveyor systems;

        2.  Dumping of tailings from  the  conveyor system/dump truck
            onto  the waste pile; and

         3.   Wind erosion  from  the  waste  pile.

 It is  assumed that as a result of  the  use of dust suppressants,  emissions

 during the  transferring and  dumping  of mill  waste would be nonexistent (OAQPS

 1987).   Emissions from waste piles would be  very low; however,  they are not


                                    -  211 -

-------
expected to be zero due to the effects of wind erosion.  The methodology used

to estimate these emissions is described below.

           b.  Methodology

           Emissions from tailings disposal are estimated using the

methodology presented by EPA in the publication 'National Emission Standards

for Asbestos -- Background Information for Proposed Standards' (OAQPS 1987).

The equations were derived empirically for operations other than asbestos

waste disposal and then modified based on comments submitted by industry,

experts on fugitive emissions, and environmental groups.  An equation used to

calculate emission factors for wind erosion from sand and gravel aggregate

storage piles was used to develop emission factors for wind erosion from

asbestos disposal sites; this equation is:


                    E - (1.9) (JL) (365'?) <1>
                               1.5    235    15

where;

        E - Emissions factor (kg/day/hectare (ha))

        S - Silt content (%)

        P - Number of days/year at site with more than 0.01 inches
            of precipitation

        F - Percentage of time wind speed exceeds 5.4 m/sec at the site.

The emission factor is then used to estimate emissions as follows:
                    Emissions (kg/yr) -   E (365 **?s^ G A
                                        (10,000 mVha) H D (100)

where;

        E - Emissions factor (kg/day/ha)

        G - Annual waste generation (Mg/yr)

        A - Asbestos content of waste (%)
                                   - 212 -

-------
        H - Height of waste pile (m)




        D - Bulk density of waste (Mg/m3).




The equation assumes no emission control.  It is likely that actual emission




values would be much less than the estimates calculated because of the use of




dust suppressants or earth seals, the growth of vegetation, and crusting




(OAQPS 1987).




           c.  Emission Estimates




           The parameters used to calculate emission factors for wind erosion




are shown in Table 49.  No measured data were available for the silt content




or the percentage of time wind speed exceeds 5.4 m/s; therefore, we used the




same estimates used by OAQPS (1987).  Data on the number of days with rainfall




exceeding 0.01 inches was found in the National Climatic Center's publication,




'Local Climatological Data' (NOAA 1979).  Climatological data summaries are




presented for various cities throughout the country.  Summary sheets were not




available for the towns in which the disposal sites were located; therefore,




data was taken for the closest city/town.  The cities chosen and their




distances from the disposal sites are also shown in Table  49.




    Resulting emission estimates are shown in Table 50.  As one can see from




the table, emissions are low.  As discussed earlier, the equations used to




calculate emissions do not factor in the  use of controls.  For this reason,




the values shown in Table 50 are actually higher than would be expected at the




disposal sites.



    The highest emission estimate from waste disposal operations was found to




be 7.99 kg/yr.  This is 35 times less than levels typically found during




milling operations  (approximately 280 kg/yr).  One of the  reasons for this is




the extensive use of dust suppressants and binding agents  during waste




disposal operations.  If the wastes are not already wetted as they leave the




mill, water  and other dust suppressants are added to create a slurry.  When




                                    - 213  -

-------
 Table 49.  Parameters for Estimating Emissions  Resulting  from Wind Erosion
 Emission Parameters
     Vermont
 Asbestos  Group
(Morrisville,  VT
     05661)
    Calaveras
  Asbestos, Ltd
(Copperopolis, CA
     95228)
  KCAC Inc.
(King City,  CA
    93930)
Percent silt content (S)

Number of days with
 rainfall exceeding
 0.01 in (P)a

Percent of time wind
  speed exceeds 5.4 m/s
  (F)

Asbestos emission
  factor (kg/day/ha)
      2

    153



     24
      3.657
      2

     52



     24



      5.399
     2

    44
     1.846
aClimatological data for Burlington, VT was used for the Vermont Asbestos
Group; Burlington is located approximately 30 miles from Morrisville.
Climatological data for Stockton, CA was used for the Copperopolis Mine;
Stockton is located approximately 35 miles from Copperopolis.  Climatological
data for Fresno, CA was used for KCAC Inc.; Fresno is located approximately 80
miles from King City.

-Sources:  NOAA 1979, OAQPS 1987.
                                   - 214 -

-------
            Table 50.  Emission Estimates from Milling Waste Piles
 Emission Parameters
     Vermont
 Asbestos Group
(Morrisville,  VT
     05661)
    Calaveras
  Asbestos,  Ltd
(Copperopolis,  CA
     95228)
  KCAC Inc.
(King City,  CA
    93930)
Emission factor
(kg/day/ha) (E)
Height of waste pile
On) (H)a
Bulk density of waste
(Mg/m3) (D)b
Annual waste generation
(Mg/yr) (G)
Asbestos content of
waste (%) (A')c
Emissions (kg/yr)
3.657

46

1.121
<
199,200

0.05

0.258
5.399

88

1.642

585,700

1.0

7.99
, 1.846

20

1.121

11,800

1.0

0.355
aThe heights of the waste piles at the Morrisville and Copperopolis milling
operations were obtained directly from the mills  (Vermont Asbestos Group 1987,
Calaveras Asbestos 1987); however, similar data was not available for KCAC
Inc.  Therefore, we assumed the value estimated by OAQPS (1987).

"The densities of the wastes generated at the two California mills were
obtained directly from the mills (Calaveras Asbestos 1987, KCAC Inc. 1987).
We used the most conservative number of the two for the Vermont mill, for
which we did not have data.

cThe asbestos content of the tailings was obtained directly from the Vermont
and Copperopolis mills (ICF Exposure Survey 1986-1987, Calaveras Asbestos
1987).  The most conservative of these two numbers was used for KCAC Inc., for
which data was not obtained.

Sources:  OAQPS 1987, Vermont Asbestos Group 1987, Calaveras Asbestos 1987,
          KCAC Inc. 1987, ICF Exposure Survey 1986-1987.
                                   - 215 -

-------
the slurry dries, it solidifies forming a crusty material which effectively




controls emissions.




    The low asbestos content of the wastes is also a contributor to the low




asbestos emission rates.  Fibers that become airborne during milling




operations are usually released from material being processed for use.  Fiber




released during waste disposal operations are typically released from reject




material (i.e., material whose asbestos content was too low for use).




    Only limited monitoring has been done by the milling companies.  KCAC Inc




has found that emissions downwind of the disposal site are less than




0.005 f/cc using phase contrast microscopy.   The sample was taken at the edge




of their property, approximately 1/4 of a mile from the site (KCAC Inc. 1987)




Area samples taken by Calaveras Asbestos yielded non-detectable levels, using




a method with a lowest detectable limit of 0.15 f/cc (Calaveras Asbestos, Ltd




1987).   We are not aware of any testing done by the Vermont Asbestos Group.
                                   - 216 -

-------
                               REFERENCES
Addingly CG.   1966.   Asbestos dust and its measurement.   Ann.  Occup.  Hyg.
9:73-82.

Aherne J.  1980.  Evaluation of asbestos exposure from asbestos-roofing
removal operations.   Submitted to Professor Lester Levin, Dept. of Industrial
Hygiene, Drexel University.  October 27, 1980.

AIA.  1986.  Asbestos Information Association response to EPA, Office of Air
Quality Planning and Standards.  Letter from Kenneth E.  Nyquist of AIA to Jack
Farmer of EPA.  March 21, 1986.

Air Resources Board (State of California).  1986.  Technical Support Document.
Public hearing to consider the adoption of a regulatory amendment identifying
asbestos as a toxic air contaminant.

Anderson PH,  Grant MA, Mclnnes RG, Farino VJ.  1982.  GCA Corporation.
Analysis of fiber release from certain asbestos products.  Draft final report.
Washington, D.C.:  Office of Pesticides and Toxic Substances,  U.S.
Environmental Protection Agency.  Contract 68-01-5960.

Anderson et al.   1983.  Asbestos industry profile, trends and outlook, and
substitute products analysis.  Draft Report.  Washington, D.C., U.S.
Environmental Protection Agency.  Office of Pesticides and Toxic Substances.

Argonne National Laboratory.  1981 (December 7).  Asbestos fiber measurements
during the Nilfisk Power and Vacuum Demonstration (August 30,  1979).  Memo
received by GCA/Technology Division, from Bruce Newman,  Nilfisk of America,
Inc., King of Prussia, PA.

Asbestos Information Association/North America.  1986.  Recommended work
practice procedures for asbestos-cement sheet.   Submittal to U.S.
Environmental Protection Agency, Office of Toxic Substances, in response to
Commercial and Industrial Use of Asbestos Fibers:  Advance Notice of Proposed
Rulemaking.  EPA Docket Number OTS 61005.

ASHRAE Handbook.  1977.  American Society of Heating, Refrigerating and Air
Conditioning Engineers.  New York, New York.  Chapter 21, Infiltration and
Ventilation.

ATA.  1985, 1987.  American Trucking Association.  Telephone conversations
with Mr. Larry Strawhorn, Engineering, Alexandria, VA, on December 13, 1985,
and February 4, 1987.

Augustyniak CM.  1987 (July).  Economics and Technology Division, Office of
Toxic Substances (OTS), U.S. EPA.  Choice of summary statistic for exposure
work relating to asbestos ban/phasedown.  Memorandum to Elizabeth Margosches
of the Exposure Evaluation Division, OTS, U.S.  EPA.

Autio JE.  1986.  Mine Safety and Health Specialist, Mine Safety and Health
Administration.  Telephone conversation on November 12,  1986.
                                   - 217 -

-------
Axecell K, Jr.,  Cowherd C.  1984.  Improved emission factors for fugitive dus"
from western surface coal mining sources.  Prepared by PEDCo Environmental,
Inc. and Midwest Research Institute for U.S. Environmental Protection Agency,
Office of Air Quality Planning Standards and Industrial Environmental Researcl
Laboratory in Cooperation with the Office of Surface Mining.  Publication No.
EPA-600/7-84-048.  March 1984.

Ayer HE, Lynch JF, Fanney JH.  1965.  A comparison of impinger and membrane
filter techniques for evaluating air samples in asbestos plants.  Ann. NY Aca<
Sci 132:274-287.

Beattie J, Knox JF.  1961.  Studies of mineral content and particle size
distributions in the lungs of asbestos textile workers.  In Inhaled Particles
and Vapours (CN Davies,  ed.), Pergamon Press, Oxford, England.

Bloomfield JJ, Dallavalle JM.  1935.  The detemination and control of
industrial dust.  Public Health Bull. 217.

Bellinger N.   1987.  Representative, KIOSH Safety Research.  Telephone
conversation on February 6, 1987.

Bragg G.  1986 (June 29).  Exposure to asbestos:  an analysis of the technica.
aspects of the Environmental Protection Agency proposal to ban and phase out
asbestos.

Calaveras Asbestos, Ltd.  1987.  Telephone conversations with Joey Toney,
Director - Environmental Affairs, Copperopolis, CA during March 1987.

Canadian Asbestos Information Center (A.I.C.).  1979.  Brochure explaining thf
use of asbestos flooring products.

Cha S, Carter P.  1982.   Estimation of urban automobile brake emissions
through simulation of wear dynamics.  Submitted to Mobile Source Emissions
Research Branch, Environmental Science Research Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, NC, under Contract 68-02-2566.
Research Triangle Park,  NC:  Northrop Services, Inc., Environmental Sciences,
P. 1.

Chatfield EJ.   1983.  Measurements of asbestos fiber concentrations in ambient
atmospheres.   Study No.  10.  Report of the Royal Commission on Matters of
Health and Safety Arising from the Use of Asbestos in Ontario.  Ontario
Ministry of Government Services, Publication Services Branch, Toronto,
Ontario, Canada.

Chemical Business.  1985 (September).  Chlorine/caustic:  anatomy of a
struggle.

Chemical Engineering.  1976.  New signs point to wider use of synthetic-
membrane chlor-alkali cells.  (March 29):61.

Chemical Engineering.  1986.  Makers of gaskets, packing try asbestos
replacement.   October 27:23-25.

Chemical Week.  1982.  A revolution in chlor-alkali membranes.  November 17,
1982:35-36.

                                   - 218 -

-------
Chemical Week.  1985.  Better times for chlor-alkali in 1986.   (February
1):28.

Chemical Week.  1986.  Dow shuts some chlorine capacity.  (July 23):24.

Chilton Book Co.  1987.  Telephone conversation with Mr. Robert McAnally,
Editor, Radnor, PA, on February 4, 1987.

Chlorine Institute.  1986a.  Comments of the Chlorine Institute, Inc.   The
Chlorine Institute, Inc.  70 West 40th Street, New York, NY.   June 30,  1986.

Chlorine Institute.  1986b.  Questionnaire to the Chlorine Institute".   The
Chlorine Institute, Inc., 70 West 40th Street, New York, NY.   Docket OPS
62036.

Clarke G.  1986.  Certainteed in profile:  Tackling the asbestos cement
dilemma.  Industrial Minerals.  April 1986.

Coats V.  1983.  Vulcan Materials.  Telephone conversation on July 1,  1983.

Cogley D et al.  1981 (May).  The experimental determination of asbestos fiber
size distribution during simulated product use.  Final Report prepared by
GCA/Technology Division for the U.S. Environmental Protection Agency.   Office
of Toxic Substances, Washington, D.C.

Commins BT.  1985.  The significance of asbestos and other mineral fibers in
environmental ambient air.  Scientific and technical report.   STR 2.  Commins
Associates, Berkshire, England.

CONSAD.  1984.  CONSAD Research Corporation and Clayton Environmental
Consultants, Inc.  Asbestos task order for construction alternatives.   Final
Report.  Washington, D.C.:  Occupational Safety and Health Administration,
U.S., Department of Labor.  Contract J-9-F-4-0024.

Cordova B.  1987.  Telephone conversation between Bob Cordova, Environmental
Division of J-M Sales Corp., Denver, CO, and ICF Incorporated.  April 6, 1987.

CPSC.  1983.  Chronic Hazard Advisory Panel on Asbestos.  U.S. Product Safety
Commission, Directorate for Health Sciences, Washington, D.C. (OSHA EX
84-256).

Diamond Shamrock.  1986 (June 26).  Letter to U.S. Environmental Protection
Agency from Diamond Shamrock Chemicals Co., Irving, TX.

Donahue R.  1987.  Telephone conversation between Robert Donahue, Safety
Superintendent, Occidental Chemical Corp., Tacoma, WA, and ICF Incorporated.
April 8, 1987.

Dunnigan J, LeBel J.  1985.  Asbestos exposure in the workplace:  recent data
for asbestros roofing felt handling and relationship to the existence of a
threshold.  New Orleans, LA.  February 10-13, 1985.  National Roofing
Contractors Association, 98th Annual Convention.

Durham J.  1987.  Telephone conversation between Joe Durham,  Compliance
Officer at OSHA Atlanta area office and ICF Incorporated, February 9,  1987.

                                   - 219 -

-------
Edminster D.  1987.  Telephone conversation between Dave Edminster,  Quality
Assurance Manager, Lydall Inc.. Covington, TN, and ICF Incorporated.   April 7
1987.

EPA.  1977.  Compilation of Air Pollutant Emission Factors.   Section 11.2.
Fugitive Dust Sources.  EPA Publication No.  AP-42, Part B.  Research Triangle
Park, North Carolina.  August 1977.

EPA.  1982.  Summary of information collected under the asbestos reporting
regulation (40 CFR Part 763, Subpart D).

EPA.  1985a (May).  Asbestos waste management guidance:  generation,
transport, disposal.  U.S. Environnental Protection Agency,  Office of Solid
Waste.
                                                            i

EPA.  1985b (December).  Report co Congress:  wastes from the extraction and
beneficiation of metallic ores, phosphate rock, asbestos, overburden from
uranium mining,  and oil shale.  U.S. Environnental Protection Agency, Office
of Solid Waste and Emergency Response.

EPA.  1986a (July 29).  EPA Memorandum entitled "Methodology for calculating
asbestos emission rates from Section 8(a) data."  Exposure Assessment Branch,
EPA.

EPA.  1986b (June 17).  EPA Memorandum entitled "8(a) Asbestos aggregate
statistics report."  Intra-agency memorandum from S. Shapiro to D. Devoe,
Confidential Data Branch. IMD, OTS.

EPA.  1986c.  Telephone conversations with John Copeland, Office of Air
Quality, Planning, and Standards, Research Triangle Park, NC during October
1986.

EPA.  1986d.  Asbestos waste estimates.  Office of Air Quality, Planning,  and
Standards, Research Triangle Park, NC.

EPA.  1986e (July 1).  Protection of the environment.  40 CFR Part 61,
Subpart M.

EPA. 1986f (July IS).  National emission standards for asbestos -- background
information for proposed standards.  Emission Standards and Engineering
Division, Research Triangle Park, NC.

EPA.  1986g.  U.S. Environmental Protection Agency.  Airborne asbestos health
assessment update.  Washington. D.C.:  USEPA.  EPA/600/8-84/003F.

EPA.  1987 (January 8).  Telephone conversation with John Copeland of EPA.
Office of Air Quality Planning and Standards, Research Triangle Park, North
Carolina.

EPA.  n.d.  Exposure to asbestos fron secondary processing.   Confidential
Business Information (Document Control No. 638600018).
                                   - 220 -

-------
Equitable Environmental Health, Inc.  1977 (December 15).   Dust exposures
during the cutting and machining of asbestos/cement pipe,  additional studies.
Prepared for the A/C Pipe Producers Association.   OSHA Docket No.  H-033C,
Exhibit No. 84-279.

Evans J, Cooper D.  1980.  An inventory of particulate emissions from open
sources.  Journal of the Air Pollution Control Association, Volume 30,  No. 12.
December 1980.

Fenner E.  1980.  Correspondence between Edmund Fenner, Vice President of
Health, Safety, and Environment Dept., Johns-Manville Sales Corp., and Lester
Levin, Drexel University, on January 14, 1980.

Freimanis I.  1981.  Packing materials.  In:  Kirk-Othmer Encyclopedia of
Chemical Technology, Third Edition.  Vol. 16.  New York:  John Wiley & Sons,
pp. 725-742.

Hammond YY, Diem J, Weill H.  1979.  Evaluation of dust exposure in asbestos
cement manufacturing operations.  Am. Ind. Hyg. Assoc. J. 40:490-495 (OSHA Ex.
84-248).

Hendrickson G, Doria J.  1983  (April 29).  Versar Inc.  Asbestos 8(a) data
evaluation.  Memorandum  to M. Callahan and L. Delpire, U.S. EPA.

Hunter Publishing  Co. 1985.  Service job analysis.  Des Plaines, IL:  Hunter
Publishing Co., pp. 24-25.

Hunter Publishing  Co.  1986.  Telephone conversation with Mr. Gene  Zack,
Promotion and Research Manager, Des Plaines,  IL, on November  26, 1986.

Hwang CY.  1981 (July ).  Size and  shape of airborne asbestos fibers in mining
and mineral processing environments.  Thesis  submitted to  the Faculty of
Graduate Studies  in partial fulfillment of the requirements of  Doctor of
Philosophy.  Department  of Mining and Metallurgical Engineering.  McGill
University.  Montreal, Quebec.

ICF.  1984.  Asbestos products and  their  substitutes.   Washington,  D.C.,  U.S.
Environmental Protection Agency, Office of Toxic Substances.  July  3, 1984.

ICF.  1986a.  Potential  occupational  exposure to durable  fibers due to
substitution  for  asbestos fibers.   Prepared  for Chemical  Engineering Branch,
Office  of  Pesticides and Toxic Substances, U.S. Environmental Protection
Agency, Washington, D.C.

ICF.  1986b.  Durable fiber exposure  assessment.   Prepared for  Monica Chatmon,
Chemical Engineering Branch, Office of Pesticides  and  Toxic Substances, U.S.
Environmental Protection Agency, Washington,  D.C.

ICF.   1987.   Use  and substitutes analysis  for asbestos.   Draft  report.

ICF  Exposure  Survey.  1986-1987.   ICF survey on asbestos  manufacturers for  EPA
to identify occupational exposure  and air release  information.

ICF  Market Survey.  1986-1987.   ICF survey on asbestos manufacturers for  EPA
to identify 1985  production of asbestos products.

                                    -  221  -

-------
IIT Research Institute.   1981 (July).  Characterizing baghouse performance t(
control asbestos manufacturing source emissions.   U.S. Environmental
Protection Agency, Office of Research and Development.  EPA-68-03-2558.

Jacko MG, Rhee SK.  1978.  Brake linings and clutch facings.   In:   Kirk-
Othmer Encyclopedia of Chemical Technology,  Third Edition.   Vol.  4.   NY:   Jol
Wiley and Sons, pp. 202-212.

Jenkins, J.  1985.  Asbestos fibers release from V/A tile.   From the NAC
Journal, summer issue 1985.

Johns-Manville Corporation.  1978 (November 8).  Testimony prepared for a
public hearing before the California Occupational Safety and Health Standard:
Board.  Source of testimony unknown.  Information supplied to GCA by
Johns-Manville Corporation, Denver, CO.

Johns-Manville Corporation.  1982.  Submission of test results in letter  of
May 11, 1982 from James Reis, Director,  Asbestos Policy, Johns-Manville
Corporation, Denver, CO to Marc Grant, GCA/T«chnology Division.

KCAC Inc.  1987.  Telephone conversations with John Myers,  Product and
Production Manager, King City, CA during March 1987.

Kemmer WF, Hall FD.  1986 (October 23).   Memorandum to Project File PN
3687-23:  Plant Trip to Calaveras Asbestos,  Ltd.

Kirk-Othmer.  1984.  Encyclopedia of chemical technology.  John Wiley and Sor
Publishing Co.  New York, N.Y.

Krusell N, Cogley D.  1982.  GCA Corp.  Asbestos substitute performance
analysis.  Revised Final Report.  Washington, DC:  Office of Pesticides and
Toxic Substances, U.S. Environmental Protection Agency.  Contract 68-02-3168.

Lebel J.  1985 (January 22).  Occupational asbestos exposure data when pullin
out asbestos roofing felt.  Jacques Lebel, Technical Director of the Research
Division, Asbestos Institute.

Lebel J.  1987.  Telephone conversation between Jacques Lebel, Technical
Director, Sherbrooke University, Quebec, Canada, and ICF Incorporated.  April
6, 1987.

Lynch JR, Ayer HE, Johnson DL.  1970.  The interrelationships of selected
asbestos exposure indices.  Am. Ind. Hyg. Assoc. J. 31:598-604.

Manosh H.  1986.  President, Vermont Asbestos Group.  Telephone conversation
on November 14, 1986.

Means.  1983.  Man-hour standards.  Kingston, MA:  Robert Snow Means Company,
Inc.

MSHA.  1986 (November 14).  Personal exposure samples.  Department of Labor.
Mine Safety and Health Administration.

Murphy RL, et. al.  1971.  Floor Tile installation as a source of asbestos
exposure.  From the American Review of Respiratory Disease, Volume 104.

                                    - 222 -

-------
MVMA.  1986.  Motor Vehicle Manufacturers Association.   MVMA motor vehicle
facts and figures '86.  Detroit:  MVMA.

Myers JL.  1985.  Product and Production Manager, KCAC Incorporated.   Letter
to Jack Fanner, U.S. Environmental Protection Agency, on June 6,  1985.

Myers JL.  1986a.  Product and Production Manager, KCAC Incorporated.   Letter
to Mark Wagner, ICF Incorporated, on October 21, 1986 with attachments.

Myers JL.  1986b.  Product and Production Manager, KCAC Incorporated.
Telephone conversation on November 17, 1986.

Myers JL.  1987.  Product and Production Manager, KCAC Incorporated.
Telephone conversation on February 17, 1987.

Myers WR.  n.d.  National Institute for Occupational Safety and Health.
Strategy for recommending respirators for control of exposures to substances
undergoing premanufacturing notice (PMN) review.  Washington, D.C.:  Office of
Toxic Substances, U.S. EPA.  Contract No. EPA DW 75932235.

National Roofing Contractors Association Newsletter.  1974.  Airborne
asbestos:  test results.

NIOSH.  1976.  National Institute for Occupational Safety and Health.   Revised
recommended asbestos  standard.  Cincinnati, OH:  NIOSH.  PB-273965.

NIOSH.  1977.  Manual of analytical methods.  2nd Edition, Vol. 1, P&CAM No.
239.

NIOSH.  1979a  (August).  National Institute for Occupational Safety and
Health.  Hazard Evaluation and Technical Assistance Report No. TA 79-31.
Poole Construction Company, Denver, CO.

NIOSH.  1979b.  National Institute for Occupational  Safety and Health, Report
No.  79-95-613, Federal-Mogul Corp., Gallipolis, OH.  September 1979.

NIOSH.  1979c.  National Institute for Occupational  Safety and Health.  Health
Hazard Evaluation Report HE 78-73-612, Kentile  Floors,  Inc., Chicago, IL.

NIOSH.  1979d.  National Institute for Occupational  Safety and Health.  Health
Hazard Evaluation Report HE 78-71-633, Kentile  Floors,  Inc., Brooklyn, NY.
NTIS.  PB80-195308.

NIOSH.   1979e.  National Institute for Occupational  Safety and Health.
Industrial hygiene  assessment of  seven brake  servicing  facilities.  By P
Johnson, RD Zumwalde, DR Roberts.  Cincinnati,  OH:   NIOSH.

NIOSH.   1980a.  National Institute for Occupational  Safety and Health.  Report
No.  79-151-657, Dearborn Rubber  Corporation,  Westmont,  IL.   January 1980.

NIOSH.   1980b.  National Institute for Occupational  Safety and Health.  Report
No.  78-79-736,  Cissell  Manufacturing  Co.,  Louisville, KY.  August  1980.
                                    - 223 -

-------
wioan.  J.yauc.  National Institute for Occupational Safety and Health.
Industrial hygiene report:  Asbestos at Allied Brake Shop, Cincinnati, OH.  B-
Roberts DR.  Cincinnati, OH:  NIOSH.  PB81-241879.

NIOSH.  1980d.  National Institute for Occupational Safety and Health.
Industrial hygiene survey report of the New York City sanitation, traffic, an
police brake servicing facilities, Queens, New York.  By DR Roberts, RD
Zumwalde.  Cincinnati, OH:  NIOSH.

NIOSH.  1981a (January).  National Institute for Occupational Safety and
Health.  HETA 81-038-801.  Hensel Phelps Construction Company, Greeley, CO.

NIOSH.  1981b.  National Institute for Occupational Safety and Health.  HETA
81-407-993, Grundy Industries. Denver. CO.  Springfield. VA:.  NTIS.
PBS-18701.

NIOSH.  1982a (October).  Walk-thru survey report of Calaveras Asbestos, Ltd.
at Copperopolis,  California.  Cincinnati, Ohio:  National Institute for
Occupational Safety and Health.

NIOSH.  1982b.  Industrial hygiene summary report of asbestos exposure
assessment for brake mechanics.  By Roberts DR, Zumwalde RD.  Cincinnati, OH:
National Institute for Occupational Safety and Health.

NIOSH.  1982c.  National Institute for Occupational Safety and Health.  Healtl
Hazard Evaluation Report No. HETA 81-100-1140. Raybestos Friction Materials
Company, Crawfordsville, IN.  Springfield, VA:  NTIS.  PB84-139880.

NIOSH.  1982d.  National Institute for Occupational Safety and Health.  Healtl
Hazard Evaluation HETA 82-270-1238, Drive Train Industries, Inc., Denver, CO.
Cincinnati, OH:   NIOSH.

NIOSH.  1982e.  National Institute of Occupational Safety and Health.  HETA
81-477-1192, Grundy Industries, Inc., Joliet, IL.  Springfield, VA:  NTIS.

NIOSH.  1983a.  National Institute for Occupational Safety and Health, Report
No. 83-275-1394.   Kautt & Bux Manufacturing, Mundelein, IL.  December 1983.

NIOSH.  1983b.  National Institute for Occupational Safety and Health, Report
No. 79-040-1381.   Pennwalt Corp., Portland, OR.  October 1983.

NIOSH.  1983c.  National Institute for Occupational Safety and Health.
Walk-Through Survey Report, Control Technology for Celotex Corporation, Perth
Amboy, NJ.  Springfield, VA:  NTIS.  PB84-184597.

NIOSH.  1984a.  National Institute for Occupational Safety and Health.
In-Depth Survey Report:  Control Technology for Solids Material Handling at
General Motors Corporation, Inland Division, Vandalia, OH.  Springfield, VA:
NTIS.  PB85-221265.

NIOSH.  1984b.  National Institute for Occupational Safety and Health.  Health
Hazard Evaluation Report HETA 84-028-1517, Drive Train Industries, Inc.,
Albuquerque, NM.   Springfield, VA:  NTIS.  PB85-208353.
                                   - 224 -

-------
NIOSH.  1984c.  National Institute of Occupational Safety and Health.
In-Depth Survey Report, Control Technology for Richard Klinger, Inc.
Cincinnati, OH:  NIOSH.  Report #144-15b.

NIOSH.  1984d.  National Institute of Occupational Safety and Health.   Health
Hazard Evaluation Report HETA 83-438-1479, Grundy Industries, Inc.   Joliet,
IL.  Springfield, VA:  NTIS.  PB85-220960.

NIOSH.  1984e.  Health Hazard Evaluation Report HETA 84-074-1476,  Grundy
Industries, Inc., Denver, CO.  Springfield, VA:  NTIS.  PB85-187698.

NIOSH.  1984f.  National Institute of Occupational Safety and Health,  Report
No. 82-280-1407.  Hoover Co., North Canton, OH.  January 1984.

NIOSH.  1985a  (May).  National Institute for Occupational Safety and Health.
HETA 84-321-1590, Asbestos Shingle Tear-Off.  Rockford, Illinois sampled on
May 12, 1984.

NIOSH.  1985b.  National Institute for Occupational Safety and Health.  Health
Hazard Evaluation Report HETA 84-151-1544, Nuturn Corporation, New Castle,  IN.
Springfield, VA:  NTIS.  PB86-138724.

NIOSH.  1985c.  National Institute for Occupational Safety and Health.  Health
Hazard Evaluation Report HETA 84-368-1624, FMC Corporation, San Jose,  CA.
Springfield, VA:  NTIS.  PB86-191269.

NIOSH.  1987a.  National Institute for Occupational Safety and Health.  In-
depth survey report:  evaluation of brake drum service at United States Postal
Service vehicle maintenance facility, Nashville, Tennessee.  By FW Godbey et
al.  Cincinnati, OH:  NIOSH.  Report No. ECTB152-20b.

NIOSH.  1987b.  National Institute for Occupational Safety and Health.  In-
depth survey report:  evaluation of brake drum service controls at Cincinnati
Bell maintenance facility, Fairfax, Ohio.  By JW Sheehy, WF Todd,  TC Cooper,
HD Van Wagenen.  Cincinnati, OH:  NIOSH.  Report No. ECTB152-21b.

NIOSH.  1987c.  National Institute for Occupational Safety and Health.
In-depth survey report:  control technology for brake drum service operations
at Ohio Department of Transportation maintenance facility, Lebanon, OH.
Cincinnati, OH:  NIOSH.  Report No. ECTB 152-18b.

NOAA. 1979.  Local climatological data, annual summaries for 1979.   National
Oceanic and Atmospheric Administration, Environmental Data and Information
Service,.National Climatic Center, Asheville, N.C.

NRC (National Research Council Committee on Nonoccupational Health Risks of
Asbestiform Fibers).  1984.  Asbestiform fibers.  Nonoccupational health
risks.  Washington, D.C.:  National Academy Press.

OAQPS.  1987 (March 5).  National emission standards for asbestos --
background information for proposed standards.  Draft Report.  U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards.
                                   - 225 -

-------
OSHA.  1986a (June 20).  Occupational Safety and Health Administration.
Occupational exposure to asbestos, tremolite, anthophyllite, and actinolite;
Final Rules. (51 ffi No. 119, p. 22612).

OSHA.  1986b (June 11).  Occupational Safety and Health Administration,  Off!
of Regulatory Analysis.  Final regulatory impact and regulatory flexibility
analysis of the revised asbestos standard.

OSHA.  1986c.  Telephone conversation with Ira Wainless, Office of Health
Standards, Washington, DC on October 23, 1986.

OSHA.  1987 (January 12).  Occupational Safety and Health Administration, U.
Department of Labor.  National inspection summary report by substance:  heal
sampling results by inspection.  Computer print-out for 1979-1986.

PEI Associates.  1984.  Site specific data summary forms.

PEI Associates, Inc.  1985.  Asbestos dust control in brake maintenance.
Draft.  Prepared for Office of Pesticides and Toxic Substances, U.S.
Environmental Protection Agency.

PEI Associates.  1986 (June).  Operation and maintenance manual for fabric
filters.  U.S.  Environmental Protection Agency, Office of Research and
Development.  EPA-68-02-3919.

PEI Associates.  1987.  Cost of engineering controls for brake
maintenance/repair.  Prepared for Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency.

Perry R.  1973.  Chemical Engineer's Handbook, 5th ed. New York:  HcGrav-Hil
p. 12-27.

Phelps LG.  1987.  Plant Engineer, Vermont Asbestos Group.  Telephone
conversation on January 30, 1987.

PPG.  1986 (June 24).  Letter to U.S. Environmental Protection Agency from P-
Industries, Pittsburgh, PA.

Precision Import Service.  1985.  Telephone conversation with Pete Vandervat*
Owner, on December 11, 1985.

Reed LV.  1937.  Telephone conversation between the office of L.V. Reed,
NIOSH, Cincinnati, OH, and ICF Incorporated.  April 7, 1987.

Research Triangle Institute (RTI).  1984.  Phase I Report.  Regulatory
analysis of the proposed OSHA standard on asbestos.  OSHA Docket No. H-033C.
Exhibit No. 84-473.  May 1984.

Rodelsperger K et al.  Estimation of exposure to asbestos-cement dust on
building sites.  Study supported by the Unwelfbundesant, Berlin, Project No.
10401023/11, by the Commission of the European Community, Project No. 298-781
ENVD, and by the Bau-Berufagenossenschaften, Frankfurt.

Rose T.  1987.   Telephone conversation between Tom Rose, Rose Roofing,
Arlington, Virginia, and ICF Incorporated, April 1, 1987.

                                   - 226 -

-------
Rosenthal FS, Paull JM.  1985.  The quality of respirator programs:   an
analysis from OSHA compliance data.  American Industrial Hygiene Association
Journal.  46(12):709-715.

Roy N et al.  1980 (February).  Asbestos product test results.   Draft Final
Report prepared by GCA/Technology Division for U.S. EPA, Office of Pesticides
and Toxic Substances, Washington, D.C.

Roy S.  1987.  Telephone conversation with Mr. Sims Roy, Office of Air Quality
Planning and Standards on July 6, 1987.

Royal Commission.  1984.  Report of the Royal Commission on matters of health
and safety arising from the use of asbestos in Ontario.  Ontario Ministry of
Government Services, Publication Services Branch, Toronto, Ontario,  Canada.

RTI.  1985 (August).  Research Triangle Institute.  Regulatory impact analysis
of controls on asbestos and asbestos products.  Vol: III.  Appendix J.
Prepared for U.S. EPA.

Scott SW.  1984.  Asbestos substitutes in friction applications.  Design News
40:6, 44-50.

Sears.  1985.  Sears Roebuck and Co., Automotive Service.  Telephone
conversation with Mr. Luther Budd, Brake Service, Washington, D.C.,  on
December 11, 1985.

Section 114 Letters.  1985.  Request for information by OAQPS on the subject
of air pollution control for asbestos in primary manufacturing facilities.

Shapiro S.  1986 (June 17).  Asbestos Action Program, Office of Pesticides and
Toxic Substances, U.S. Environmental Protection Agency, Washington, DC 20460.
8(a) Asbestos aggregate statistics report.  Intra-agency memorandum to D.
Devoe, Confidential Data Branch, IMD, OTS.

Spurny KR, Stober W.  1978.  Asbestos measurement in remote ambient air.  In
Clean Air, the Continuing Challenge; Proceedings of the International Clean
Air Conference, Brisbane, Australia, pp. 373-387.  Ann Arbor, Michigan:  Ann
Arbor Science Publishers, Inc.

SRI International.  1985.  Chlorine.  In:  1984 Directory of Chemical
Producers, United States:487-488.

Syracuse Research Corp. 1978 (August).  Chemical market input/output analysis
of selected chemical substances to assess sources of environmental
contamination:  Task III Asbestos.  Prepared for U.S. EPA." NTIS PB-285-531.

Toney J.  1986.  Director of Environmental Affairs, Calaveras Asbestos, Ltd.
Telephone conversation on November 13, 1986.

Toney J.  1987.  Director of Environmental Affairs, Calaveras Asbestos Ltd.
Telephone conversation on February 10, 1987.

Tzanetos et al.  1987  (July 21).  Forecasts of asbestos drum and disc brake
sales for light vehicles.  Memorandum from ICF Incorporated to Christine
Augustyniak of the Office of Toxic Substances, U.S. EPA.

                                   - 227 -

-------
Vermont Asbestos Group. 1987.  Telephone conversation with Lloyd Phelps,  Plan
Engineer, Morrisville, VT on March 17, 1987.

Versar.  1983.  Draft Final Report -- Exposure assessment for asbestos.
Prepared for the Environmental Protection Agency. Under EPA Contract No.
68-01-6271, Task No. 49.  Washington, D.C.. April 29. 1983.  OSHA Docket No.
H-033C.  Exhibit No. 333.

Versar.  1984.  Exposure assessment for asbestos.  Draft Final Report.
Washington, D.C. , U.S. Environmental Protection Agency, Office of Toxic
Substances.  EPA Contract No. 68-01-6271, Task No. 49.

Versar.  1987 (June 17).  Nonoccupational asbestos exposure.. Revised Report
Washington, D.C.:  Office of Toxic Substances, U.S. EPA.  Contract No.
68-02-4254.

Verta R.  1987.   Telephone conversation with Mr. Robert Verta, U.S. Bureau of
Mines. Washington. D.C. on July 6, 1987.

Vulcan Chemicals.  1986.  Comments of the Vulcan Materials Company.  Vulcan
Chemicals,  P.O.  Box 7609. Birmingham, AL  35253.

Wainless I.  1987.  Telephone conversation with Ira Wainless, Office of Healt
Standards,  Washington, D.C. in April of 1987.

Williams L.  1987.  Telephone conversation between Lewis Williams, Plant
Engineer, Raymark Corp.. Manville, NC, and ICF Incorporated.  April 7, 1987.

Wright MD et. al.  1984.  Phase I Report-Regulatory analysis of the proposed
OSHA standard on asbestos.  Prepared for U.S. Department of Labor, OSHA by
Research Triangle Institute, Research Triangle Park, N.C.  May 1984.

Yocom, JE.   1976.  Fugitive emissions problems in perspective.  Paper
presented at a Symposium on "Fugitive Emissions -- Measurement and Control."
Sponsored by the Environmental Protection Agency, Hartford, Conn., May 17-19,
1976.
                                   - 228 -

-------
APPENDIX A.  OCCUPATIONAL EXPOSURE PROFILES AND AIR PF.T.EASES FOR
             PRODUCTS NO LONGER PRODUCED OR USED IN THE U.S.

    Based on the ICF Market Survey (1986-1987), several asbestos products are

no longer made in or imported into the United States.  These products include

commercial paper, corrugated paper, rollboard, flooring felt, unsaturated

roofing felt (this product is imported), saturated roofing felt, vinyl

asbestos floor tile, and corrugated asbestos cement sheet (this product is

imported).  Commercial paper, corrugated paper, rollboard, flooring felt, and

unsaturated and saturated roofing felts, were they still manufactured in the

U.S.,  would be included in the paper products category.

    This appendix is divided into two sections:  1.  Occupational Exposure,

and 2.  Air Releases.  Section 1 presents product and process descriptions and

an exposure profile (using geometric means of the raw data) for each of the

products no longer manufactured or used in the U.S.  Relevant exposures during

primary manufacture, secondary manufacture, and construction are presented.

Since the products are not manufactured or used in the U.S. , populations would

not be exposed to the asbestos concentrations presented.  However, available

data on the number of workers expected per unit of product manufactured or

mixture consumed is estimated such that a sensitivity analysis may be

performed to determine potential risk associated with the reestablishment of

these product markets.

    Section 2 presents emission factors per unit of asbestos consumption for

primary manufacture, secondary manufacture, and construction.  Specifics such

as zip codes necessary for modeling purposes are not presented because no

facilities are currently manufacturing or handling these products.

    1.  Occupational Exposure

    This section presents our estimates of occupational exposure during

asbestos product manufacture and construction for the products no longer


                                   - 229 -

-------
manufactured or used in the U.S.  Each subsection presents a brief discussio




of each product, the processes used to manufacture the products or a



description of the operations, current and projected exposure levels,




populations exposed per unit of product manufactured (primary manufacturing)




or mixture consumed (secondary manufacturing),  and duration and frequency of




exposure.  All of the paper products are presented together because of the




similarity in the process by which they are all manufactured and, therefore,




the similarity in their associated exposure profiles.




    As was done in Chapter II, projected exposure levels under the new




asbestos standard are estimated assuming that for those operations where




8-hour TWA exposures are currently above 0.2 f/cc, work practices will be




changed either with the addition of engineering controls or respirators to




reduce the exposures to 0.2 f/cc.  Once the raw data have been manipulated b;




this methodology, new geometric and arithmetic mean exposures are calculated



this is the projected exposure.




        a.  Product Manufacture




        This section presents the exposure profiles for both primary and




secondary manufacture of paper products, corrugated asbestos cement sheet, af



vinyl-asbestos floor tile, which are no longer manufactured in the U.S.




           (1)  Paper Products




               Product Descriptions.  This section provides descriptions of




those asbestos paper products no longer manufactured in the U.S.  These




products include commercial paper, corrugated paper, rollboard, flooring fel<




and unsaturated (this product is imported) and saturated roofing felt.




    Commercial paper includes general insulation paper and muffler paper.




These papers differ in weight and thickness and usually range from 95 to 98




percent asbestos fiber by weight, and 2 to 5 percent starch binder.  Muffler








                                   - 230 -

-------
paper contains an even larger percentage of asbestos fiber and very little




starch binder (Krusell and Cogley 1982).




    Commercial paper was used to provide insulation against fire,  heat,  and




corrosion with minimum thickness.  Muffler paper was used by the automotive




industry for exhaust emission control systems.  The paper was applied between




the inner and outer skins of the muffler or converter to maintain the high




temperatures necessary for pollution control within the catalytic converter




reaction chamber and to protect the outer layer from the heat (Krusell and




Cogley 1982).




    General asbestos insulation papers were used in a variety of industries.




The steel and aluminum industries used it as insulation in furnaces, in trough




linings in the smelting process, and against hot metal and drippings of molten




metal.  Asbestos paper was also used in the glass and ceramic industry for




kiln insulation, in foundries as mold liners, and in the electrical parts and




appliance industry for electrical insulation.




    Corrugated paper is a type of commercial paper which is corrugated and




cemented to a flat paper backing, and is sometimes laminated with aluminum




foil.  It is manufactured with a high asbestos content  (95 to 98 percent by




weight) and a starch binder (2 to 5 percent)  (Krusell and Cogley 1982).




    Corrugated asbestos paper was used as a thermal insulator for pipe




coverings and as block insulation.  The paper could be used for appliance




insulation up to 270°F, hot-water and low-pressure steam pipe insulation, and




process line insulation.




    Rollboard is a thin and flexible material composed basically of two sheets




of paper laminated together with sodium silicate.  It can be cut, folded,




wrapped, and rolled, and can go around sharp  corners.  Asbestos rollboard was




used to protect against fire, heat, corrosion, and moisture.
                                    -  231  -

-------
    The primary constituent of asbestos rollboard is asbestos fibers, with the




balance consisting of binders and fillers.  The asbestos content ranges from




60 to 95 percent by weight; 70 to 80 percent asbestos is considered typical.




Frequently used binders are starches, elastomers, silicates, and cement;




mineral wool, clay, and lime act as fillers (Krusell and Cogley 1982).




    Flooring felt is a paper product which was used extensively as a backing




for vinyl sheet flooring products.  It was also used as an underlay for other




flooring surfaces.  The latter use accounted for a small fraction of the total




quantity of felt used.  Desirable features of this product are its dimensional




stability and high moisture, rot, and heat resistance.  Dimensional stability




refers to the ability of the flooring to stretch and contract with temperature




changes and "settling" of the floor deck.   The flooring should be able to




withstand these conditions without cracking, warping, or otherwise




deteriorating.  Asbestos backing is particularly useful in prolonging floor




life when moisture from below the surface is a problem (Krusell and Cogley




1982).




    Asbestos flooring felt is composed of approximately 85 percent asbestos




and 15 percent latex binder.  The latex binder is usually a styrene-butadiene




type, although acrylic latexes have been used.




    All unsaturated roofing felt is used in the production of saturated




roofing felt.  Unsaturated roofing felts used to be manufactured at centrally




located plants and then usually shipped to various geographical locations




nearer to demand.  The felts were then saturated with coal tar or asphalt at




these locations, rather than at the central location, because of the shipping




cost; it is much cheaper to ship unsaturated roofing felt.




    Unsaturated asbestos roofing felt is composed of 85 to 87 percent




asbestos, 8 to 12 percent cellulose fibers,  and 3 to 5 percent starch fibers
                                   - 232 -

-------
by weight.  Other materials such as wet and dry strength polymers,  Kraft




fibers, fibrous glass, and mineral wool may also be used as fillers.




    Saturated roofing felt is used for built-up roofing.  "Built-up"  refers to




the practice of layering paper lengths on top of each other while hot roofing




tar or asphalt is mopped between layers for adhesion and/or additional weather




protection (Krusell and Cogley 1982).  There are three basic types of built-up




roofing:  gravel surface, smooth surface, and mineral surface.  The felt is




either single- or multi-layer grade; fiberglass filaments or wire strands may




be embedded between felt layers for reinforcement.  The felt's thickness or




grade and the amount of asphalt coating required depend on the product's



intended use.




    Asbestos is used in roofing felts because of its dimensional stability and




resistance to rot, fire, and heat.  Rot resistance is particularly important




because of roofing felt's use on flat or nearly flat roofs with poor drainage.




Given the rapid heating and cooling of roof surfaces, some cracking may occur,




allowing water to penetrate, particularly in damper climates or in areas where




snow, subject to periodic melting, has accumulated on the rooftop.  Asbestos




felt resists cracking.




               Process Descriptions.  The main operations in all asbestos




paper manufacturing are receiving, bag opening, mixing, forming, and




finishing.  In the fiber introduction operation, raw asbestos is most often




introduced in unopened pulpable bags, although for certain types of paper the




fiber is dumped from the bags.  In cases where the fiber is dumped from the




bags, asbestos is obtained in non-compressed pulpable bags so that the bags




may be slit and the asbestos added directly to the mixer.  At the mixer stage,




the fiber is immediately wetted.




    As in other manufacturing processes, the asbestos fiber is carried under




negative pressure by conveyor to the mixer.  There, the fiber is wet-mixed




                                   - 233 -

-------
with paper stock, binder, and other ingredients.  The stock slurry flows into



the papennaking machine and forms a sheet.  The solids content of this sheet



may be less than five percent; the moisture content of this sheet is reduced



greatly during transit through the paper machine.  The wet nature of the



material precludes the release of asbestos fiber.



    The forming of asbestos paper is completed during the drying, slitting,



and calendering stages.  The final operation involves rewinding in which the




paper products are bulk packaged on spools, reels, or beams from the larger




rolls.  Rewinding is a dry operation.



    Final fabrication may involve cutting, trimming, and shaping to meet the



requirements of the space into which the paper product is to be installed.




Secondary manufacturing is performed on all of these paper products.



    There are certain manufacturing procedures that are specific to the



individual products within the paper products category, although most of the




operations are similar if not identical.




    Commercial and corrugated paper are both manufactured using conventional



papennaking machines.   Corrugated paper further passes through a corrugation




machine which produces the corrugated molding on the surface of the paper.



    Rollboard is manufactured in a process similar to that used for millboard,




but it is produced in a continuous sheet.  A conventional cylinder paper




machine, with minimal changes when necessary, can be adapted for rollboard




production.  The wet mixture of asbestos fiber and chemical additives is dried



before being cut and shaped to size.




    Asbestos felt-backed vinyl sheet flooring is composed of three products:




the asbestos flooring felt backing, the coating on the felt, and the wear




layer.  The major steps in the manufacture of asbestos felt-backed vinyl sheet



flooring are the production of flooring felt, coating, printing, fusion,




trimming, and packaging:  The flooring may be manufactured at the same plant




                                   - 234 -

-------
as the sheet backing or in a separate facility.  Asbestos flooring felt is




manufactured on conventional paperoaking machines.  During manufacture, the




asbestos fibers are coated with latex and are reported to be fully




encapsulated when the sheet backing is ready to be coated with vinyl.   The




asbestos felt roll is fed into the coating machinery, coated with vinyl,  and




possibly decorated by various printing techniques.  The vinyl plastisol can




then be colored by various additives or techniques.  The printed sheet then




goes to a fusion step where the sheet is coated with another layer of material




called the "wear layer."  The wear layer is a homogeneous polymer application




that provides an impervious surface for the finished product.  The coated and




printed sheet is dried by a fusion oven at temperatures of at least 250cF for




the copolymers, and 300"F for the homopolymers.  Chill rolls made of




chrome/plated steel remove most of the heat from the laminate.  After fusion,




these layers remain distinct but are no longer chemically or mechanically




separable.   The sheet flooring may be further decorated by various chemical or




printing methods before or after cooling (Krusell and Cogley 1982).




    Asbestos roofing felt is manufactured on conventional papermaking machines




and then saturated with asphalt or coal tar.  The felt is pulled through a




bath of hot asphalt or coal tar until it is thoroughly saturated.  After




saturation, the felt passes over a series of hot rollers to set the asphalt or




coal tar into the felt.  It may, on occasion, be coated with extra surface




layers of asphalt or coal tar.  After saturation and coating, the felt passes




over a series of cooling rollers that reduce the paper's temperature and




provide a smooth finish.  Paper given extra coats of asphalt or coal tar must




be treated to prevent adhesion between layers when the felt is rolled.  The




felt is then air-dried, rolled, and packaged for marketing (Krusell and Cogley




1982).
                                   - 235 -

-------
               Production and Employment.  Currently, no workers are exposed

to asbestos during primary and secondary manufacturing of commercial paper,

corrugated paper, rollboard, flooring felt, and roofing felt because these

products are no longer produced in the U.S. (IGF Market Survey 1986-1987).

However, based on historical data from the TSCA Section 8(a) submissions (EPl

19B6b),  the average number of workers who would be exposed per unit 'of

asbestos product manufactured or mixture consumed for each product is:
                                Primary Manufacture
                                (workers/1,000 tons
                                     produced)
               Secondary Manufacture
                (workers/1,000 tons
                     consumed)
     Commercial Paper
     Corrugated Paper
     Rollboard
     Flooring Felt
     Unsaturated Roofing Felt
     Saturated Roofing Felt
 78
N/A
N/A
  1.4
  2
 13
  955
  330
5,510
   20
   12
   22
Enough data were not available for primary manufacture of corrugated paper an

rollboard to estimate populations.

               Frequency and Duration of Exposure.  A full working year of 25

days/year and 8 hours/day can be assumed for the worker populations estimated

above.

               Exposure Profile.  Table A-l presents the exposure profile for

each paper product as determined from the raw monitoring data.  We categorized

the jobs performed by each worker monitored into one of several job

categories.  Geometric and arithmetic means were calculated for the data by

job category.  This summary table includes the results for primary and

secondary manufacturing of asbestos paper products.

    Since the monitoring data are aggregated for all paper products, the job

categories and exposures are assumed to be equivalent for all paper products.

                                   - 236 -

-------
Table A-l.  Exposure Profiles for Paper Products No Longer Manufactured in the U.S.*
8-Hour TWA (Exposure (f/cc)c
Product
Commercial Paper



Corrugated Paper



Rollboard



Flooring Felt



Unsaturated Roofing Felt



Saturated Roofing Felt



Job Category
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Fiber Introduction
Processing
Other
Total
Population
Distribution15
ax
581
34X
78/1,000 t
81
581
341
N/A
23Z
301
47X
N/A
21X
25X
541
1.4/1,000 t
461
191
351
2/1,000 t
OX
78X
22X
13/1,000 t
Pre-0.2
Geometric Mean
0.091 (6)
0.013 (23)
O.OS2 (5)
0.033
0.091 (6)
0.013 (23)
0.052 (5)
0.033
0.091 (6)
0.013 (23)
0.052 (5)
0.049
0.091 (6)
0.013 (23)
O.OS2 (5)
0.050
0.091 (6)
0.013 (23)
0.052 (5)
0.063
0.091 (6)
0.013 (23)
0.052 (5)
0.022
f/cc PEL
Arithmetic Mean
0.134
0.030
0.070
0.052
0.134
0.030
0.070
0.052
0.134
0.030
0.070
0.073
0.134
0.030
0.070
0.073
0.134
0.030
0.070
0.092
0.134
0.030
0.070
0.039
Post-0.2
Geometric Mean
0.079
0.013
0.052
0.032
0.079
0.013
0.052
0.032
0.079
0.013
0.052
0.047
0.079
0.013
0.052
0.04S
0.079
0.01
0.052
0.057
0.079
0.013
0.052
0.022
f/cc PEL"
Arithmetic Mean
0.094
0.030
0.070
0.049
0.094
0.030
0.070
0.049
0.094
0.030
0.070
0.064
0.094
0.030
0.070
0.065
0.094
0.030
0.070
0.073
0.094
0.030
0.070
0.039
Duration
(hr/day)*
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
6
8
§
8
Frequency
(days/yr)e
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
SECONDARY MANUFACTURING
Comnerclal Paper
Corrugated Paper
Rollboard
Flooring Felt
Unsaturated Roofing Felt
Saturated Roofing Felt
N/A
N/A
N/A
N/A
N/A
N/A
955/1,000 t
330/1,000 t
5,510/1,000 t
20/1,000 t
12/1,000 t
22/1,000 t
0.016 (26)
0.016 (26)
0.016 (26)
0.016 (26)
0.016 (26)
0.016 (26)
0.022
0.022
0.022
0.022
0.022
0.022
0.016
0.016
0.016
0.016
0.016
0.016
0.022
0.022
0.022
0.022
0.022
0.022
8
8
8
8
8
8
250
250
250
250
250
250

-------
                                                                                Table A-l (Continued)





              "This exposur. prof11. .ppli.s to comnerci.l paper,  corrugated paper,  rollboard.  flooring felt,  and roofing felt.



              manufaetnMlB>0hiC*h """'"'M  PoP"latl°n factor dependent upon the quantity of product manufactured  (primary manufacturing)  or mixture consumed (secondary

              nr rniihn.trf   TK  A,  » ?v J.     ,  5 """ ln 8ensitlvlty analysis.   Adequate data are not available to estimate population  factors for corrugated paper

              (Lndrlc^n^ndXrl. iS            tOt"1 ""**"  °f m*"' "^^ int° 8Peci£lc  -»ob  Categories is based on the 1981  TSCA Section 8(a)  data
                                             .c  and  arithmetic means  of the  raw 8-hour TWA  exposure data.   (Exposures  for  leas  than  8 hour,  are  converted  to  8-hour

                    ,        ..        -.      - *8 8ucn-  """"""ins  zero  exposure  during periods when the worker  is not handling asbestos.)  Th. numb.r of data points  is

                    in  parentneses.   Since  the monitoring  data is  aggregated for all paper  products, the exposure values for each Job category are assumed to be

                     in  each JobPcatego^y           "  C°rrespondlns  to tne  total Population, are calculated  as weighted averages based on the distribution of workers





                    'to^x.ctlv^ I^/^'n'r  V?iUr*  *" "lculated dl"«="y "o» the raw monitoring data.  Each 8-hour  TWA exposure value that i. above 0.2 f/cc is
                     to  exactly 0.2  f/cc.   Data that are  already  at  or below  this value remain unchanged.



             8A full working year (8 hr/day, 250 daya/yr) la assumed.



             Source.:  ICP Market Survey 1986-1987, 1CF Exposure  Survey 1986-1987. OSHA 1987. Bendrlck.on and Dorla 1983.
OJ
00

-------
In reality, however, sources indicated that exposure levels can vary widely




depending on the asbestos content of the product (OSHA 1986b).




    The distribution of workers exposed into specific job categories is based




on the 1981 TSCA Section 8(a) data  (Hendrickson and Doria 1983).  For




secondary manufacturing of paper products,  the total populations are not




disaggregated into job categories.




    The primary manufacturing operation with the greatest potential for




causing asbestos exposure is fiber  introduction.  The fiber introduction




procedure in paper manufacturing involves the dumping of asbestos into a




beater or hydropulper.  A local exhaust system with dust-collection equipment




is used to keep the processing area under negative pressure.  The wet-mixing




of the fiber with paper stock, binder, and other ingredients controls the




release of airborne asbestos.  Canopy hoods and exhausts that are utilized to




remove water vapor and heat from steam-heated rolls in the dry section also




aid in asbestos dust control.  At the slitting, calendering and rewinding




stages, local exhaust ventilation,  area hoods, and central exhaust collection




systems are the typical engineering controls.




    All job categories comprising the manufacture of asbestos papers have been




able to achieve mean exposure levels below 0.2 f/cc.  Out of all the




monitoring data, only a few samples are above the 0.2 f/cc level; all of these




data fall under the fiber introduction job category.  Assuming  that in these




few cases additional controls will  be used to achieve 0.2 f/cc  exposure




levels, the projected exposure under the new PEL will decrease  slightly, as




shown in Table A-l.



    No changes are projected for secondary manufacturing of paper products due




to the absence of any monitoring data greater than 0.2 f/cc.
                                    -  239  -

-------
           (2)  Corrugated Asbestos Cenent Sheets




               Product Description.  Asbestos is used as a reinforcing




material in cement sheet products because of its high tensile strength,




flexibility,  thermal resistance,  chemical inertness, and large aspect ratio




(ratio of length to diameter).   A cement sheet becomes strong, stiff, and




tough when asbestos fiber is added, resulting in a product that is -stable,




rigid, durable, noncombustible,  and resistant to heat, weather, and corrosive




chemicals.  Also, the asbestos  cement sheet has sufficient wet strength to




enable it to be molded into complex shapes at the end of the production




process (Krusell and Cogley 1982).




    Corrugated A/C sheet was used in the construction industry when the




additional strength afforded by corrugation was beneficial.  Flat and




corrugated A/C sheets are used in somewhat different applications.  Corrugate




A/C sheet is used mainly in industrial and agricultural applications, serving




as siding and roofing for factories, warehouses, and agricultural buildings.




It is also used as a lining for waterways and canal bulkheads, and for specia




applications in cooling towers  (Krusell and Cogley 1982).




               Process Description.  A/C sheet, both corrugated and flat, is




manufactured by using a dry, a wet, or a wet-mechanical process.  A/C sheet i




made from a mixture of Portland cement and asbestos fiber.  An additional




fraction of finely ground inert filler and pigments is sometimes included.  V'




general, sheets contain between 15 and 40 percent asbestos fiber.  However,




for curing in short time periods, a general formulation of 12 to 25 percent




asbestos, 45 to 54 percent cement, and 30 to 40 percent silica is used




(Krusell and Cogley 1982).




    Similar to the A/C pipe process, the raw materials are mixed with water t<




form a wet slurry of asbestos,  cement, and silica.  The slurry is then picked




up by a screen cylinder mold and transferred to a felt conveyor.  The felt is




                                   - 240 -

-------
then dewatered, passed to a mandrel, and wound to the desired thickness.




After achieving the desired thickness, a cut is made across the width of the




sheet.  The sheet is manually peeled off the rotating mandrel onto a transfer




roll conveyor.  The sheet is then cured, processed through embossing rollers




and trimming/cutting wheels, and finally corrugated.




    There is no secondary manufacturing for corrugated A/C sheet.




               Production and Employment.  Currently, no workers are exposed




to asbestos during the manufacture of corrugated A/C sheet because this




product is no  longer produced in the U.S. (ICF Market Survey 1986-1987).




However, based on historical data from the TSCA Section 8(a) submissions (EPA




1986b), the average number of workers who would be exposed per unit of




corrugated A/C sheet manufactured is one per 1000 squares (square - 100 ft^).




               Duration and Frequency of Exposure.  A full working year of 250




days/year and  8 hours/day can be assumed for the worker populations estimated




above.




               Exposure Profile.  Exposure levels for corrugated A/C sheet are




assumed to be  the same as those for flat A/C sheet and A/C shingles.  The




exposure levels for all job categories exhibited  in Table A-2 are high, with




two out of the three job categories having exposure levels above 1.0 f/cc.




With the widespread use of engineering controls,  the post-0.2 f/cc PEL




exposures are  assumed to fall at or below the 0.2 f/cc level.




    The distribution of workers exposed into specific job categories is based




on 1981 TSCA Section 8(a) data  (Hendrickson and Doria 1983).




           (3)  Vinvl-Asbestos Floor Tile




               Product Description.  Floor tile was one of the largest




industrial uses of asbestos fiber in the past.  Asbestos use in resilient




flooring includes viny1-asbestos "(V/A) floor tile and asbestos felt-backed




vinyl sheet flooring.  V/A floor tile provides a hard, durable surface and is




                                    - 241 -

-------
                                            Table A-2.  Exposure Profile for Corrugated Asbestos Cement Sheet — Primary Manufacture
NO
*>
fo
8-Hour TWA (Exnosura (f/cc)b
Job Category
Fiber Introduction
Processing
Other
Total
Population
Distribution*
13S
301
3ZI
1/1,000 sq
Pre-0.2
Geometric Mean
1.054 (7)
0.408 (6)
1.028 (6)
0.845
£/cc PEL
Arithmetic Mean
1.364
0.576
1.143
1.000
Post-0.2
Geometric Mean
0.200
0.166
0.200
0.190
f/cc PEtc
Arithmetic Mean
0.200
0.171
0.200
0.191
Duration
(hr/day)d
8
8
8
8
Frequency
(days/yr)d
250
250
250
250
'Based on historical data, a population factor dependant upon the quantity of product manufactured has been estimated for use
in sensitivity analysis.  The distribution of the total population exposed into specific Job categories is bated on the 1981
TSCA Section 8(a) data (Hendxlckson and Dorla 1983).

 These value* represent geometric and arithmetic means of the raw 8-hour IMA exposure data.   Exposure* for leas than 8 hours
are converted to 8-hour THAs, assuming tsro exposure during period* when the worker 1* net handling asbestos.  The number of
data point* is given in parentheses.  The values corresponding to the total population* are calculated as weighted averages
baaed on the distribution of workers exposed in each job category.

°These post-0.2 f/cc PEL exposure values are calculated directly from the raw monitorins data.   Each 8-hour THA exposure value
that is above 0.2 f/cc is reduced to exactly 0.2 f/cc.  Data that are already at or below this value remain unchanged.

dA full working year (8 hr/day, 250 days/yr) is asauned.

Sources:  OSHA 1987, ICF Market Survey 1986-1987, Hendrlckson and Doria 1983.

-------
suitable for most heavy traffic areas (e.g.,  supermarkets, department stores,

institutional/commercial settings) and radiant-heated floors (Wright 1984).

Asbestos felt-backed vinyl sheet flooring is produced as a floor cover for

general uses; the asbestos felt-backing that forms the underlaying of sheet

vinyl flooring is a paper product (refer to Section (1) above).

    Beginning in the early 1980s, most major manufacturers of V/A floor tile

switched to substitutes for asbestos (Wright 1984).  Since that time, asbestos

has been eliminated completely from V/A floor tile as there are currently no

domestic producers or importers of this product (ICF Market Survey 1986-1987).

    V/A floor tile are made from polyvinyl chloride polymers or copolymers and

are usually produced in squares 12 inches by 12 inches; it is commonly

manufactured in thicknesses of 1/16, 3/32, and 1/8 of an  inch (ICF Market

Survey 1986-1987).

    Although V/A tile composition varies by type and manufacturer, typical

compositions are as follows (on a percentage weight basis):

        •  asbestos: 5-25 percent,

        •  binder: 15-20 percent,

        •  limestone: 53-73 percent,

        •  plasticizer: 5 percent,

        •  stabilizer: 1-2 percent,  and

        •  pigment: 0.5-5 percent  (ICF Market Survey  1986-1987, Wright
           1984).

Asbestos fiber grades 5 through  7  are used to impact  wet-strength and

dimensional  stability (OSHA 1986b).

               Process Description.  The production of V/A floor  tile involves

fiber  introduction, mixing, melting, calendering,  embossing,  curing,  and

finishing.   Opened paper bags, or  unopened polyethylene bags, of  raw asbestos

are  dumped  into  Banbury mixers,  along with other  dry  ingredients  (e.g.,


                                    - 243  -

-------
binder, limestone, pigments).  The mixers combine the ingredients into an




agglomerated homogenized plastic mass at a temperature of about 300°F (Wright




1984).  This hot plastic mass encapsulates the asbestos fibers, thus reducing




the potential for exposure during later steps; the points with the most




potential for fiber release precede the mixing operation (Wright 1984).   The




hot mix is dumped into a conveyor and passes under negative pressure to a




two-roll mill.  The mill presses the plastic into a continuous slab, with an




initial thickness of 1 or 2 inches, which is fed through a series of




calendering rolls to achieve the desired final-product thickness.




    Next, the warm sheet passes through an embosser which adds design feature




and texture to the surface.  After partial cooling and waxing, the sheet is




cut into squares by a cutting press.  The tiles are separated from the excess




scrap, inspected, and packaged.  Scrap and rejected tiles are returned to the




mixer for recovery.  V/A floor tile undergoes no secondary processing.




    Specific job categories fit into the production process as follows.  The




weighing and mixing of chemicals are accomplished in designated process areas




(NIOSH 1979c).  A scale operator pulls a handle which gravity feeds a. tared




quantity of polyvinyl chloride resin to an empty hopper car.  A bag of




asbestos is then added to the hopper with water  (NIOSH 1979d).  The hopper ca




is sent to the next station  (scale operation) where it is positioned under a




hood.  Any recycled product returned from the end of the line due  to




unsatisfactory appearance is added to the hopper by a recycling process man.




At the next position, pigments from bulk samples are added to the hopper;




those pigments have been previously weighed by a pigment scale operator




(pigment operation).  The hopper then travels to the Banbury station where th<




bucket is emptied into the Banbury mixer along with other ingredients; this




job is categorized as Banbury mixing.  "Other" workers complete  the processing




of the tile.




                                   - 244 -

-------
               Production and Employment-.  Currently, no workers are exposed




to asbestos during the manufacture of V/A floor tile because this product is




no longer produced in the U.S. (ICF Market Survey 1986-1987).   However,  based




on historical data from the TSCA Section 8(a) submissions (EPA 1986b),  the




average number of workers who would be exposed per unit of V/A floor tile




manufactured is 8.3 per million yd2.




               Duration and Frequency of Exposure.  A full working year of 250




days/year and 8 hours/day can be assumed for the worker populations'estimated




above.




               Exposure Profile.   Table A-3 presents the exposure profile for




V/A floor tile production as determined from the raw monitoring data.  All job




categories achieve average fiber concentrations significantly less than 0.2




f/cc.  The encapsulation of asbestos fibers, due to the formation of an




agglomerated plastic mass, controls exposures.  The only jobs which have any




exposure values greater than the 0.2 f/cc level fall into the scale operation




and "other" categories; this is depicted by the average exposure reductions in




the post-0.2 f/cc PEL values from the pre-0.2 f/cc PEL values.  The projected




post-0.2 f/cc PEL exposures are based on the assumption that all raw data




points greater than 0.2 f/cc are adjusted to exactly 0.2 f/cc.  This reduction




may occur via the utilization of engineering controls or the use of




respirators.  Dust control is commonly achieved through enclosure of processes




and conveying equipment, exhaust ventilation, and good housekeeping and work




practice.  Local exhaust ventilation is provided at stations such as fiber




introduction and cutting (OSHA 1986a).  Mottling granulation and scrap




grinding may be isolated in enclosed rooms.  More specifically, slot exhaust




ventilation is used during Banbury mixing and pigment operations, open duct




exhaust ventilation is used during recycling processes, and canopy exhaust




ventilation is used during scale operations  (NIOSH 1979c, NIOSH 1979d).




                                    - 245 -

-------
                             Table A-3.   Exposure Profile for V/A Floor Tile  --  Primary Manufacture
   Job Category
 Population
Distribution"
                                                         8-Hour TWA (Exposure (f/cc)
                                             Pre-0.2 f/cc PEL
                                                                               Post-0.2 f/ce PELC
                                                                   Duration   Frequency
                Geometric Mean   Arithmetic Mean   Geometric Mean   Arithmetic Mean  (hr/day)   (daya/yr)
Scale Operation
Banbury Mixing
Pigment Operation
Recycling Process
Other

     Total
231
23X
 81
 7X
121

8.3/mll sq yd
0.029 (28)
0.028  (9)
0.005  (6)
0.048  (6)
0.047 (10)
0.095
0.039
0.020
0.055
0.094

0.073
0.200
0.028
0.005
0.048
0.046

0.034
0.068
0.039
0.020
0.055
0.086

0.064
8
8
8
8
I
250
250
250
250
250

250
*Ba«ed on historical data, a population factor dependent upon the quantity of product manufactured hai been estimated for use
In sensitivity analysis.  The distribution of the total population exposed Into specific job categories is based on the 1981
TSCA Section 8(a) data (Hendrlckson and Dor la 1983).

 These values represent geometric and arithmetic means of the raw 8-hour THA exposure data.  Exposures for less than 8 hours
are converted to 8-hour THAs, assuming zero exposure during periods when the worker is not handling asbestos.  The number of
data points is given in parentheses.  The values corresponding to the total populations are calculated as weighted averages
based on the distribution of workers exposed in each Job category.

°These post-0.2 f/cc PEL exposure values are calculated directly from the raw monitoring data.   Each 8-hour THA exposure
value that is above 0.2 f/cc is reduced to exactly 0.2 f/cc.  Data that are already at or below this value remain unchanged.

dA full working year (B hr/day. 250 days/yr) is assumed.
Sources:  OSHA 1987, ICF Market Survey 1986-1987, NIOSH 1979c, NIOSH 1979d,  Bendrickson and Dorla 1983.

-------
        b.   Construction Industry Exposure




        Historically, construction materials and products containing asbestos




fibers have included vinyl-asbestos floor tiles and asbestos felt-backed sheet




vinyl flooring.  Since the early 1970s, however, the overall demand for these




types of products has declined due to the availability of adequate




substitutes, and the increased regulatory requirements and restrictions.  This




declining demand has continued through the present and, as a result, these




products are no longer produced or sold in the U.S. (ICF Market Survey




1986-1987).  Exposure to asbestos in the construction industry would occur




during installation and removal of these products.  (Exposures during




construction of built-up roofing using roofing felts are covered in Chapter




II, Section E.)




            (1)  Exposure Settings and Operations




           Vinyl-asbestos floor tiles could be installed either by




professional floor installers or homeowners, while asbestos felt-backed sheet




vinyl flooring is usually laid down only by professionals due to the




difficulty  involved  in creating a perfect  fit with the one-piece sheet




(Anderson 1982).  The installation of  floor covering  is typically begun in the




center of a room and proceeds towards  the  walls.  For V/A floor tile, tile




adhesive is applied  over one section of the floor at  a time,  if the tiles are




not prebacked with adhesive, and  full  tiles are  pressed  in  place.   Once all




full  tiles  are applied, the perimeter  and partial pieces are  measured and cut




out.  Tile  pieces are hand  cut  either  by  scoring the  tile  (cutting  partially




through) and snapping  it, or by simply cutting entirely  through the tile in




one operation  (Anderson 1982).  All  cutting occurs  in the room being tiled.




Cutting  tools  for  installation include commercial tile cutters, utility




knives,  scissors,  and  razor knives  (Anderson 1982).
                                    - 247 -

-------
    Many installers consider subfloor preparation to be the most difficult and




time consuming part of flooring installation (Anderson 1982).  This highly




important activity commonly involves sanding and dry scraping, although




sanding in often explicitly warned against by the manufacturer.  Subfloor




preparation by sanding involves the greatest potential for asbestos exposure




if the existing subfloor is covered with an asbestos-containing floor product




    The majority of installers try to avoid sanding and either remove existir




flooring using flat-bladed putty knives or cover the floor with plywood,




fiberboard, or masonite (Anderson 1982).  Other tools utilized for removal of




old flooring (particularly vinyl*asbestos floor tile) include hammers,



chisels, scrapers, and stripping machines.  By far, the major release of




asbestos to the environment from flooring occurs at replacement time when old




flooring is removed (Syracuse Research Corp. 1978).  Service life for asbesto;




floorings depends upon the severity of use and may, thus, vary from 10 to 30




years (Syracuse Research Corp. 1978).




    The installation of asbestos felt-backed sheet vinyl flooring involves




adhesion techniques similar to those described for V/A tile.  Gluing




techniques for sheet flooring vary in three basic ways: fully pasted,




perimeter pasted, and "put down quick" (PDQ) technique (Anderson 1982).




    A fully pasted floor,  whereby the entire felt backing is glued, takes the




most time to install, but will last the longest.  When only the flooring




perimeter is glued, the installation is quicker and less expensive; this




technique is mostly used on concrete subfloors.  A PDQ installation, in which




the flooring is layed without any glue, is more common to flooring which is




solely vinyl than it is to the more expensive asbestos felt-backed product




(Anderson 1982).




    Sheet vinyl flooring must be cut to shape and size; a template is widely




used for this process.   The template, often composed of narrow strips of




                                   - 248 -

-------
asbestos-containing felt, is cut to match the exact perimeter details of the




room.  These details are then transferred onto the sheet vinyl product.




Utility and razor knives are used for cutting the template as well as for




cutting the actual flooring.  Preparation and installation of sheet vinyl




flooring involves minor mechanical disturbance to the product, minimizing the




potential for airborne asbestos fiber release.




    The standard procedure  in the removal of asbestos felt-backed vinyl




flooring is to  "strip" the  floor.  This operation involves removing the top




two layers of sheeting (wear and foam) and splitting the bottom asbestos




flooring felt layer  in half, along the horizontal plane  (Anderson 1982).  This




procedure is relatively  easy if the  subfloor covering is glued over its entire




area, since the felt will split when pulled.




    During flooring  felt removal, the covering is cut into strips




approximately 0.46 meters  (1.5 feet) wide and each  strip is pulled up and away




from  the subfloor.   This separation  process  leaves  half  of the felt in place




and provides a  uniform subfloor for  the new  flooring product; the cutting and




separation may  have  a high  fiber  release potential  (Anderson  1982).  One




manufacturer recommends  that  the  exposed felt be vacuumed  immediately after




each  strip is removed  to collect  loose  asbestos  fiber-containing  dirt




 (Anderson 1982).



     If  the asbestos  felt-backed sheet vinyl  flooring is  not  totally surface




glued,  then  stripping  will remove the entire covering,  leaving only split




asbestos  felt  around the perimeter  (assuming the flooring had only been pasted




along its perimeter).   The recommended practice for this process  is to  wet  the




 felt before  scraping it up; it can also be feathered to provide a smooth felt




 interface, however this involves the sanding of the felt,  with higher




 potential exposures.  Another scenario can arise if the existing  sheet




 flooring is  intact and relatively smooth.   In this case the  top layer can be



                                     - 249 -

-------
used as Che subfloor after being rough sanded Co provide for better adhesion;




Chis sanding will discurb only Che wear layer of Che sheec flooring and not



Che asbestos felt (Anderson 1982).




           (2)  Current Pre-0.2 f/cc PEL Exposures  .




           Various studies have been performed on construction-related




industries to determine exposure Co asbestos produces during installation and




removal.




    Alchough Chore are a large number of potential uses and activities




involved in the installation and renoval of asbestos flooring products, there




is only a limited amount of data concerning exposure to asbestos fibers during




construction projects.  ICF has utilized three sources of exposure data in




determining average values for the installation and removal of V/A floor tile




and asbestos flooring felt (Canadian A.I.C. 1979. Murphy 1971, Jenkins 1985).




The one study containing information on flooring felt (Canadian A.I.C. 1979)




refers to the product as sheet vinyl flooring.  As discussed earlier, the




asbestos flooring felt (or sheet backing) forms the underlayment of sheet




vinyl flooring.  Alchough the felt is manufactured separately from the sheec




vinyl flooring, it muse be considered together with sheec vinyl flooring for




purposes of exposure assessment, since the felt backing is typically not used




alone.   Table A-4 summarizes the pre-0.2 f/cc PEL asbestos exposure levels for




both V/A floor tile and sheet flooring.




    Except for those samples from the Jenkins (1985) study, all samples were




analyzed with phase-contrast microscopy.  The value reported in this study




(Jenkins 1985), comprised of condensed data from 40 asbestos fiber samples, is




based on transmission electron microscopy (TEM) testing.  Because the TEM




detects all fiber sizes, both greater than and less than 5 microns in length,




the value is adjusted so as to represent an equivalent value that only




consists of fibers greater than 5 microns in length.  The percentage of fibers




                                   - 250 -

-------
              Table A-4.  Exposure to Asbestos Floor Products in the
                          Construction Industry Under Pre-0.2 f/cc PEL
                                       8-Hour TWA          Short-Term TWA
                                    Exposure Level*        Exposure Level3
                                        (f/cc)	  	(f/cc)
Product
V/A Floor Tile
V/A Floor Tile
Asbestos Flooring
Felt
Asbestos Flooring
Felt
Activity
Installation
Removal
Installation
Removal
Geometric
Mean
0.017b
0.011C
0.016d
0.032d
Arithmetic Geometric
Mean Mean
0.026 0.049b
0.012 0.018C
0.036 0.016d
0.081 0.032d
Arithmetic
Mean
C.425
0.029
0.036
0.081
aExposure estimates are geometric and arithmetic means of all of the available
exposure data for each product and operation.

bCanadian A.I.C. 1979, Murphy 1971.

cCanadian A.I.C. 1979, Jenkins 1985.

dCanadian A.I.C. 1979.
                                    -  251  -

-------
greater than 5 microns in length is determined by Jenkins (1985) to be 27.5
percent of the total TEM count; thus, 27.5 percent of the value provided
(0.2382 f/cc) is utilized for the geometric and arithmetic mean values.
    Both geometric and arithmetic means of the raw exposured data are
presented in Table A-4.  For the purpose of these calculations, all 8-hour TWA
data points that have a value of zero (i.e., 0.0 f/cc) are assumed to be
equivalent to 0.001 f/cc.  A zero count implies that the count was too low for
any fibers to be seen in the counting fields; these "non-detectable" levels
are consistent with very low, but rarely zero, concentrations of airborne
fibers.  The low value (0.001 f/cc) is assumed to approximate the limit of
detection for the Canadian A.I.C. (1979) study, which reported very low values
(e.g., 0.007 f/cc) for other samples.
    Table A-4 presents two sets of data, 8-hour TWA and short-term TWA
exposure levels.  The effective duration of exposure is 8 hours/day in all
cases; and for two out of four product-activity average values, the 8-hour TW
equals the short-term TWA.  For these two cases, both sets of values were
provided in the study results (Canadian A.I.C. 1979).  But for the other two
cases, where data are supplied by two other sources (Jenkins 1985, Murphy
1971) , some exposure data (assumed to be sampled for less than 8 hours/day)
are converted to 8-hour TWAs.  The 8-hour TWA was calculated from the TWA for
the time sampled as follows:

    8-Hour TWA - Sampling Time (minutes) x ^ fn period Qf ^ ^^
                     480 Minutes
The worker is assumed to have no exposure for the remainder of the work day
(i.e., when not handling asbestos).
    Most V/A floor tiles are installed by simply applying adhesive, or
removing the protective cover layer of tiles prebacked with adhesive, and
pressing the tiles in place.  Only perimeter pieces are typically cut.  Both
                                   - 252 -

-------
tasks, applying adhesive/pressing tiles in place and cutting, result in




minimal asbestos fiber release as revealed by the low exposure level in




Table A-4.  The low energy input associated with the use of cutting tools




(mostly hand tools) and the binding properties of the tile matrix minimize




fiber release during installation (Anderson 1982).




    Although a significant amount of asbestos fiber can be released during




removal of V/A tile using conventional tile removal methods (Jenkins 1985),




Table A-4 reveals a low exposure level.  Apparently, good removal practices




were followed during the sampling.  Fiber control methods such as damp




removal, isolation of areas by plasticizing, and use of respirators should be




used for tile removal.  It is also important to avoid sanding, whenever it is




economically or practicably possible, as is the case with the removal




operations sampled (Canadian A.I.C.  1979, Jenkins 1985, Anderson 1982).




    Potential exposure levels during the installation of asbestos felt-backed




sheet vinyl flooring are also low.  There are only minor mechanical




disturbances to the product, minimizing the potential for airborne asbestos




fiber release.  The only action during installation that disturbs the asbestos




felt layer of the flooring is cutting, an activity that is short in duration




(5 to 20 minutes) and not energy intensive (Anderson 1982).




    As with removal of all asbestos floor products, the industry recommends




the use of wet scraping methods for the removal of sheet vinyl flooring.




Results of the study employed (Canadian A.I.C. 1979) in determining the values




in Table A-4, show that wet scraping methods result in very low concentrations




of airborne asbestos fiber.  All data points used to calculate the geometric




and arithmetic means were sampled during removal by the proper wet methods.




According to the Resilient Floor Covering Institute, dry scraping is not a




recommended work practice (Anderson 1982).
                                   - 253 -

-------
    Concentrations for the two sheet vinyl flooring activities are somewhat




higher than those for V/A floor tile.  The higher levels are expected because




the felt backing contains a much higher percentage of asbestos (85 percent)




than the tile product (8 to 30 percent) and is not as structurally cohesive as




the tile (Anderson 1982).




           (3)  Projected Post-0.2 f/cc PEL Exposures




           Reduction of the PEL for asbestos from 2 f/cc to 0.2 f/cc means




that increased worker protection is required in installation and especially




removal of asbestos flooring.  In the construction industry, many companies




are small operators who cannot afford extensive protective gear or certain




sophisticated power tools with dust collection systems.   However, for these




product sectors, extensive change in procedures or controls is barely needed.




As shown in Table A-5, almost all of the projected exposure levels are




equivalent to the pre-0.2 f/cc PEL values.  This is because, of all the raw




8-hour TWA data, only one data point is greater than 0.2 f/cc (Canadian A.I.C.




1979); thus, only projections for flooring felt removal differ from the




pre-0.2 f/cc PEL values (i.e., both 8-hour and short-term TWAs differ).




    The projected exposures presented in Table A-5 are based on the assumption




that all raw data points greater than 0.2 f/cc (in this case there is only one




data point) are adjusted to exactly 0.2 f/cc.  This reduction may occur via




the increased utilization of engineering controls (e.g., tool shrouding), the




total abandonment of sanding and dry scraping operations, or the use of




respirators.




           (4)  Population Exposed




           Workers in the construction industry often work with a variety of




materials,  depending on the needs of the purchaser.  However a determination




can be made of the equivalent number of workers who would work full-time (8




hours/day and 250 days/year) exclusively with asbestos products.  This number,




                                   - 254 -

-------
            Table  A-5.   Exposure to Asbestos Floor Products in the
                Construction Industry Under Post-0.2 f/cc  PEL
                                       8-Hour TWA          Short-Term TWA
                                    Exposure Level3        Exposure Levela
                                        (f/cc)	  	(f/cc)
Product Activity
V/A Floor Tile Installation
V/A Floor Tile Removal
Asbestos Flooring Installation
Felt
Asbestos Flooring Removal
Felt
Geometric
Mean
0.017b
0.011b
0.016b
0.029
Arithmetic Geometric
Mean Mean
0.026b 0.049b
0.012b 0.018b
0.036b 0.016b
0.051 0.029
Arithmetic
Mean
0.425b
0.029b
0.036b
0.051
Projections are calculated assuming that all 8-hour TWA raw data values
originally >0.2 f/cc would be reduced to this limit, and new geometric and
arithmetic means are calculated.

bNo change from the pre-0.2 f/cc PEL exposure.
                                   - 255 -

-------
the full-time equivalent (FTE)  population,  is based on crew size,




productivity, and total production plus imports of the asbestos product.




Since vinyl-asbestos floor tiles and flooring felt are no longer manufactured,




imported, or used in the U.S.,  the FTE population cannot be calculated.




However, the factors which allow calculation of FTEs are provided.   These




factors provide FTE population per unit of product installed or removed.   In




the construction industry, crew sizes can be estimated based on Means Man-Houi




Standards (Means 1983).




    A typical crew size for installation of V/A floor tile (i.e.,  categorized




as resilient flooring) is composed of a single tile layer.  Daily output (for




an 8 hour work day) is estimated at 520 square feet per day (Means 1983) .




Therefore, the population factor for installation of V/A floor tile is 0.008




person-years/I,000 ft2.   Anderson (1982) states that the installation of V/A




tile in a standard 9 by 12 foot (2.7 by 3.6 meter) room would take one profes




sional installer from 2 to 4 hours (i.e., 0.009-0.018 person-years/I,000\ft2)




The time spent in cutting the tile (usually only the perimeter pieces are cut




amounts to only about 10 minutes for a standard room (Anderson 1982); this




short duration is the period of time with the highest potential for asbestos




fiber release.




    Means (1983) estimates that a single installer (typical crew size) can la>




down between 325 and 650 square feet of vinyl sheet flooring (with backing)




per day (i.e., 0.006-0.012 person-years per 1,000 ft2).  This wide range is,




apparently, due to differences in the width of the vinyl roll  (1.8, 2.7, 3.6,




or 4.5 meters), the way in which the flooring is glued, and whether or not a




pattern is used (Anderson 1982).  The installation of asbestos felt-backed




sheet vinyl flooring in a standard 9 by 12 foot room is estimated to take




between 1.5 and 6 hours by Anderson (1982), depending on the roll width,




gluing technique, and pattern usage.  Based on this source, the population




                                   - 256 -

-------
factor for  installation of asbestos  felt-backed sheet vinyl flooring is




between 0.007 and 0.028 person-years/I,000 ft2.  The narrower the flooring




width, the  greater  the installation  time since the patterns of adjoining




sections must be matched  (Anderson 1982).  Installation time also depends on




the complexity of the perimeter details and the skill of the installer




(Anderson 1982).  As with tile installation, the cutting activity is short in




duration (5 to 20 minutes for the standard size room), taking only a fraction



of the total installation time.




    The time it takes to remove asbestos flooring products (i.e., either V/A




tile or vinyl sheet flooring) is approximately 4 to 8 hours for a standard 9




by 12 foot room (Anderson 1982); this figure varies considerably depending on




difficulties encountered in removing old tile or sheeting.  Subfloor




preparation (i.e., old flooring removal) is considered by many workers to be




the most time consuming part of a flooring construction activities.  The




productivity estimate (4 to 8 hours) is assumed to be based on a crew of only




one worker.  Therefore, the population factor for removal of V/A floor tile or




sheet vinyl flooring is between 0.018 and 0.037 person-years/I,000 ft2.




            (5)  Frequency and Duration of Exposure




           In the construction industry, exposure duration and frequency are




effectively 8 hours/day and 250 days/year because full-time equivalent




populations are being used in this analysis.  Conceptually, this is a




measurement of the total person-hours of exposure involved (much as




construction jobs require a certain  number of man-hours to. do the work) and




not the actual number of workers who at some time might install or remove




asbestos-containing materials.




    2.  Air Releases




    As stated in Chapter III, the technical data base from which emissions




estimates are derived contains significant data gaps as well as other sources



                                   - 257 -

-------
of uncertainty that require the use of simplifying assumptions.  Therefore,




the results of this analysis represent estimates given the available data and




should be used cautiously.  The emission estimates are by no means absolute



values.




    In this section, emission estimates are presented for:  (1)  primary and




secondary manufacture of commercial paper, corrugated paper, rollboard,




flooring felt, roofing felt (unsaturated and unsaturated), corrugated asbesto




cement sheet, and vinyl-asbestos floor tile which are all products no longer




manufactured in the U.S.; and (2) construction usage of vinyl-asbestos floor




tile and asbestos felt-backed sheet vinyl flooring which are no longer used in




the U.S.  Since these products are no longer manufactured in the U.S.,




information on the location of emission sources; stack dimensions; exhaust gas




flow rate,  temperature, and velocity are not relevant.




        a.   Primary and Secondary Kanufaeturing Sources




        The methodology used to calculate asbestos emissions from primary and




secondary manufacturing sources is discussed in detail in Chapter III.  A




steady state release rate is estimated over an entire year for primary and




secondary manufacturing;  a full year is 8,760 hours per year (365 days/year




and 24 hours/day).   The quantity of asbestos fiber or mixture consumed for




each product category is derived from the Section 8(a) data (EPA 1986b).  The




collection efficiency of the control device is estimated to be 99.67 percent




for paper products, and 99.95 percent for V/A floor tile and corrugated A/C




sheet.




    The quantity of asbestos fibers collected by the control device is equal




to the total quantity of waste collected times the asbestos content in the




waste.  The quantity of waste collected per unit of production is derived by




calculating the ratio of waste collected by the control device in 1981 (EPA




1986b) to asbestos fiber consumed in 1981 (EPA 1986b).  To determine the




                                   - 258 -

-------
quantity of waste collected, this ratio would be multiplied by the amount of




fiber consumed.  Since these products are no longer manufactured, only the




ratio of waste generation to fiber consumption is presented.  The average




asbestos content in the waste is also available from 8(a) data (EPA 1986b).




    For secondary manufacturing, the factor used to calculate the total waste




collected in the control device is presented as waste collected per unit of




asbestos mixture consumed; these ratios are also obtained from the TSCA 8(a)



data (EPA 1986b).




    Asbestos air emissions from primary and secondary manufacturing sources




are estimated for asbestos paper products (commercial paper, corrugated paper,




rollboard, flooring felt, and roofing felt), vinyl-asbestos floor tile, and




corrugated A/C sheet.  Table A-6 presents asbestos emissions from primary




manufacturing sources, and Table A-7 presents emissions from secondary




manufacturing sources.  There is no secondary manufacturing of V/A floor tile.




        b.  Construction Sources




        The asbestos-related activities performed in the construction trade




include installation and removal of asbestos material.  The construction




industry is different than general industry in that the worksites are




temporary in nature and seasonal.  The points of emissions cannot be well




defined due to the nature and large number of work settings.  For indoor jobs,




asbestos is emitted through a number of openings such as windows, doors, and




cracks.  For outdoor jobs, airborne asbestos can be dispersed in any or all




directions, depending on the wind vectors.  In other words, the release



configuration from these sources is unconfined with multiple exits.  All of




these factors contribute to the difficulty of making an accurate assessment of




asbestos air emissions from these sources.  The methodology used to estimate




emissions from indoor and outdoor construction activities is presented in




Chapter III, Section C.



                                   - 259 -

-------
                                                        Table A-6.  Asbestos Emission* from Primary Manufacturing Sources
                                                                 for  Products No Longer Manufactured in the U.S.
N)
ON
O
Asbestos Product/Mixture
Conner cial Paper
RoUboard
Unsaturated Roofing Felt
Saturated Roofing Felt
Flooring Felt
Corrugated Paper
Vinyl Asbestos Floor Tile
Asbestos Cement Sheet,
Corrugated
TSCA
Identification
Number
01
02
07
08
09
10
12
16
1981 Asbestos
Consumption*
(short tons)
1,038
213*
15.434
5,709
105, 7 821
54
29.6511
1,106
Total
Quantity
of Baghouae
Haste (198l)b
(short tons)
U2d
13*
44«
16h
8,838J
7d
3*2-*
11*
Percent
of Asbestos
in Waste0
(X)
80. Od
83.0*
62.5
82. 5h
57.3
80. Od
7.9
11. Oh
Emission Rate
(g/sec)
1.08E-2
1.03E-3
3.46E-3
1.26E-3
4.83E-1
5.34E-4
3.89E-4
1.77E-4
Emission Rate
Per Pound of
Fiber Consumed
(g/sec/lb)
5.22E-9
2.41E-9
1.12E-10
1.10E-10
2.28E-9
4.94E-9
6.56E-12
8.02E-11
*EPA 1986b (Table 3).  This is the total quantity of asbestos consumed per product category in 1981.   The quantities were
reported in ranges as "minimum" and "maximum" quantities; the average value was calculated and reported in this table.
                             1986b (Table 15).   This  is  the  total  quantity  of baghouse  fines, dry waste.  It does not contain  100 percent  asbestos  fiber.

                        CEPA 1986b (Table 14).   Assumed  the  percent  of  asbestos  in baghouae waste is equal to the percent of asbestos  in the total  waste.

                        'There is  no  data available for  this product category.   It is ana of the "missing categories" in the Section 8(a)  data, EPA 1986.
                        Assumed that  the  quantity  of  baghouae waste  end the percent of  asbestos in waste are relatively similar to electrical  paper
                        category,  and thus the quantity  of baghouse  waste was adjusted  accordingly based on the total quantity of asbestos consumed.

                        'This is the  quantity of asbestos consumed in 1979.  Asbestos rollboard was not produced in  1980 or 1981.

                         See footnote (d).  Calculations were made based on reported values for asbestos millboard.

                        'Baghouse  waste was not  reported.  Assumed baghouse waste is half of the total waste reported In EPA 1986b .(Table  15).

                        ^>ee footnote (d).  Calculations were made based on reported values for unsaturated roofing  felt.

                        1This is the  quantity of asbestos fiber consumed in 1981 obtained from EPA 1986b (Table 3).  A single  value rather than a range
                        is  reported.

                        ''Baghouse  waste (dry) was  not reported.  Assumed baghouse fines dry is equal to baghouse fines wet reported in EPA 1986b
                        (Table 15).

                         See footnote (d).  Calculations were made based on reported values for asbestos cement shingle.

-------
                                                       Table A-7.  Asbestos Emissions from Secondary Manufacturing Sourcaa
                                                                 for Products No Longer Manufactured In the U.S.
ro
cr»


Asbestos Mixture/
Product (ID 41)
Ccmnercial Paper (01)
Rollboard (02)
Unaaturated Roofing Felt (07)
Saturated Roofing Felt (08)
Flooring Felt (09)
Corrugated Paper (10)
A/C Sheet. Corrugated (16)



1981 Total Asbaatoa
Mixture Consumption*
508 tons
28.5 tons
27,714 tons
3,457.5 tons
98,420 tons
14. 5 tons
344 squares
(1 square ~ 100 sq ft)
Total
Quantity
of Baghouse
Waste (1981)b
(short tons)
17
1.6
291
150
3,413
0.3
3.4d


Percent
of Asbestos
in Waste0
(I)
46.6
55.3
54.7
54.7
39.0
£0.7
30.0



Emission Rate
(g/sec)
7.57E-3
8.45E-5
1.52E-2
7.B4E-3
1.27E-1
1.73E-3
1.47E-5


Emission Rate
Per Found of
Mixture Consumed
(g/aec/lb)
1.49E-5
2.96E-6
5.49E-7
2.27E-6
1.29E-6
1.19E-6
4.27E-B*

                             "EPA 1986b (Table 8).

                              EPA 1986b (Table 17).  Baghouse fines dry or one-half of the total waste adjusted by the percentage of the consumed
                             materials accounted for by the product under investigation.  Haste data are provided by mixture code* (101-220) rather
                             that product codes (1-34), and several products are identified as feed materials for the mixture classifications.

                             CEPA 1986b (Table 16).

                              Estimated based on values reported for A/C sheet, flat.

                             *Thls la the emission rate of asbestos per one square of corrugated A/C sheet rather than per Ib pof corrugated A/C
                             sheet.

-------
    Asbestos air emissions from construction sources are presented in




Table A-8, using both the geometric and arithmetic mean, short-term asbestos




concentrations.  The two types of activities analyzed for V/A floor tile and




asbestos flooring felt are installation and removal.  (Asbestos air releases




from roofing felt construction activities are estimated in Chapter III,




Section C.)  Since all of the construction activities for V/A floor tile and




flooring felt are performed indoors, a work volume of 9,000 cu ft is assumed




with one air change per hour.
                                    - 262  -

-------
                                                    Table A-B.  Asbestos Emissions from Construction Activities Using Products
                                                                            Ho Longer Used in the U.S.
NJ
Short Term TWA
Asbestos Concentration
(f/cc)a
Construction Activity
V/A Floor Tile Installation
V/A Floor Tile Removal
Asbestos Flooring Felt Installation
Asbestos Flooring Felt Removal
Geometric
Mean
0.049
0.01B
0.016
0.029
Arithmetic
Mean
0.425
0.029
0.036
O.OS1
Asbestos Emission Rate
((l/»*c>
Work Area
Volume
(ft3)
9,000
9,000
9,000
9,000
Mumber of
Air Changes
Per Hour
1
1
1
1
Using
Geometric
Mean
1.16E-7
4.25E-8
3.78E-6
6.8AE-8
Using
Arithmetic
Mean
l.OOE-6
6.84E-8
8.49E-8
1.20E-7
Person-Years
Per 1,000 sq ft
of Product
Consumed
0.011
0.028
0.013
0.028
Asbestos Emission Ratt
(*/so ft)
Using
Geometric
Mean
9.2E-6
8.6E-6
3.5E-6
1.4E-5
Using
Arithmetic
Mean
7.9E-5
1.4E-5
8.0E-6
2.4E-5
             *Refer to Section  l.b above.

             bAverage value  la  used.  Refer  to Section  l.b  above.

-------