EPA-560/3-75-007
      Materials Balance and Technology Assessment
            of Mercury and Its Compounds on
               National and Regional Bases
                          October 1975
                          Final Report
               U.S. ENVIRONMENTAL PROTECTION AGENCY
                      Office of Toxic Substances
                       Washington, D.C. 20460

-------
                                REVIEW NOTICE

        This report has been reviewed by the Office of Toxic Substances,
EPA, and approved for publication.  Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.

-------
EPA 560/3-75-007
                 MATERIALS BALANCE AND TECHNOLOGY ASSESSMENT

                        OF MERCURY AND ITS  COMPOUNDS

                                     ON

                         NATIONAL AND REGIONAL BASES
                                October 1975
                                Final Report
                             Contract 68-01-2930
                               Project Officer
                             David Garrett, P.E.
                                Prepared for
                       Environmental Protection Agency
                         Office of Toxic Substances
                           Washington, D.C. 20460

-------
                                   PREFACE


     This study was conducted under Contract  68-01-2930 for the Office of
Toxic Substances (OTS),  Environmental Protection Agency, Washington, D.C.
Mr. David Garrett was the EPA Project Officer.  Mr. Garrett and Dr. Herbert
Katz of OTS were most helpful in providing  guidance and assistance during
the course of the study.  Other branches of the Agency were also  cooperative
in providing advice and information on studies  and data sources used in  the
study.  Representatives of various industries were contacted  for  information,
and many were most generous in providing insights  and data regarding their
industries and technologies.  To all of these who  contributed to  this proj-
ect, we express our thanks.

     This work was conducted in the Hazards Assessment Directorate of URS
Research Company, under the administrative direction of William H. Van Horn,
who also served as project manager.  Other URS  personnel  who contributed
technical inputs to the report include Dr. Wesley Bradford,  Leon Crain,
Charles Hamman, Gary Kaufman, Lance Johnson,  Michael LaGraff, Eric Schrier,
Ruth Shnider, and William Penn.  Research, editorial, and typing support was
provided by Margaret Chatham, Mama Lieberman,  Frances Maurier, Lucy Vincent,
and Louise Zwick.

     The study covers a very broad spectrum of disciplines, and  information
was obtained from many published and unpublished  sources, as well as numerous
proprietary contacts.  While the authors have attempted to minimize errors,
the type and magnitude of the study mean that  some errors have undoubtedly
crept into the report.  We  assume  responsibility  for such inadvertent mis-
takes and will attempt to correct  important  deficiencies  upon receipt of
better data.
                                       iii

-------
                                   CONTENTS
Section
          EXECUTIVE SUMMARY  	      a

          Purpose of Study 	      a
          Method of Approach 	      a
          Uses of Mercury	      a
          Study Results	      b
          Technology Assessment	•	      e
          Regulatory Impacts 	      i
          Conclusions	      j
          Findings 	      m

     I    INTRODUCTION 	      1

          Scope of Work	      5
          Method of Approach	      6
          Report Organization  	      9

    II    HISTORICAL OVERVIEW AND OUTLOOK SUMMARY	     11

          Sources of Mercury	     11
          Present and Projected Mercury Consumption  	     11
          Major End Uses	     15
            Agriculture	     15
            Catalysts	     15
            Dental Preparations  	     15
            Electrical Apparatus 	    16
            Caustic-Chlorine (Chlor-Alkali)  	    16
            Paints	    17
            Pharmaceuticals	     17
            Other Uses (including Laboratories)   	     18
            Consumption by Federal Agencies  	     18
          Regional Distribution  	     18
          Final Disposal	     20
          Ultimate Fate of Mercury in the Environment	     25
          Legislation and Regulations Applicable to  Mercury  ...     29
            Mercury Crisis:  The Early  Response   	     29
            Regulatory Guidelines and Actions   	     31
              The Air Environment	     31
              The Land Environment	     31
              The Water Environment	     31
            Interstate Transportation   	 	    33
            Trends  in Legislation and Regulatory Action  	     33

-------
                                   CONTENTS
                                 (continued)
Section
Page
   III    INPUT/OUTPUT ANALYSIS 	     37
          Introduction  	     37
          Methodology	     38
            Development of an Input/Output Model  	     42
            Acquisition of the Data	     48
            Calculation of Contributions from Natural  Sources  .  .     54
               Degassing	     54
               Runoff	     58
            Contributions from Final Disposal 	     58
               Sewage	     58
               Groundwater	     60
          Results	     60
            Regional Analyses 	     62
               California Study Region  	     62
               Arizona Study Region 	     65
               Kentucky/Tennessee  Study Region  	     67
               Louisiana Study Region 	     73
               New York/New Jersey Study Region	     75
            National Inventory  	     79
            Risk  Analysis	     91
          Selection  of Technologies for Study   	     98

   IV     TECHNOLOGY ASSESSMENT	    105.
          Introduction  	    105
         Microtechnologies  	    107
            Sector I  - Mercury Input	    107
              Mercury Mining  and  Refining   	    107
            Sector II  - Mining	    116
              Copper Mining and Smelting	  .    116
            Sector III - Miscellaneous Unregulated Sources  . .  .    128
              Power Plants	    128
           Sector V - Manufacturing and Processing 	    138
              Chlor-Alkali Manufacturing 	    138
              Manufacture of Mercurials 	    153
              Battery Manufacturing  	    170
              Electric Lamp Manufacturing  	    178
              Industrial Instruments 	    185
              Paint  Manufacturing.	    195
           Sector VIII - Commercial and Final Consumption  .  . .    205
              Agricultural Pesticides  	    205
              Nonagricultural Pesticdes  	    212
              Laboratory Uses	    217


                                       vi

-------
                                   CONTENTS
                                 (continued)
Section
    IV    TECHNOLOGY ASSESSMENT (Cont.)
            Sector VIII - Personally Oriented Final Consumption     222
              Dental Applications  	    222
              Pharmaceuticals  	    227
            Final Disposal Sector	    230
              Municipal Waste Treatment  	    230
              Municipal Solid Waste Disposal 	    237
          Macrotechnology  	    246
          Implications 	    246

     V    REGULATORY IMPACT ANALYSIS 	    253
          Introduction	    253
          Methodology	    255
          Regulation in the Chlor-Alkali Industry  	    257
            Alternative I—Elimination of Substitution of
            Product	    257
            Alternative 2—Alternate Technologies  	    258
            Alternative 3—Increased Control 	    261
            Introduction of New Emission Control Technology  .  .    263
            Cost-Benefit Analysis  	    267
            Risk Analysis	    269
          Regulation in the Paint Industry  	    271
            Regulatory Alternatives and Effects  	    272
            Cost-Benefit Analysis  	    280
            Risk Assessment	    280
          Regulation in the Battery Industry 	    282
            Industry Regulation  	    284
            Cost-Benefit Analysis  . .  .	    289
            Risk Assessment	    292
          Overview	    292

          REFERENCES

          APPENDIXES

          A.   Mercury in  the  Environment
          B.   Selection of  Study  Regions
          C.   Computer Data Base  Format and Use
          D.   General Methodology
          E.   Point  Source  Emissions  of Mercury
                                        vii

-------
                                   FIGURES
Number                                                                Page

  A       Percent Distribution of Consumptive Uses of Mercury,
          1973	        e
  Bl      Percent Distribution of Estimated Emissions to Air of
          Mercury from Man-Related Sources, 1973  	

  B2      Percent Distribution of Estimated Emissions to Water
          of Mercury from Man-Related Emissions, 1973 	
          Annual Emissions of Mercury to the Environment from
          Man-Related and Natural Sources
  1       Study Regions   ....................        8
  2       Projected Mercury Consumption in the United States,
          by End Use  ......................       14
  3       Past and Probable Future Trends in Final Disposal
          of Mercurials .....................       22
  4       Mercury Cycling through Environmental "Reservoirs"   .  .       27
  5       Variations in Mercury Content of the Environment in
          Relation to Existing Standards  ............       28

  6       Sources of Mercury and Industrial and Commercial
          Usage in the United States, 1973  ...........       39
  7       Basic Mercury Inventory Flow Chart  ..........       43

  8       Materials Balance  of Mercury from Coal in the
          United States .....................       52
  9       Mineralized Areas  and Mercury Deposits in the
          United States .....................       59
 10       Mercury Emissions  to Land in Kentucky/Tennessee  Study
          Region,  by County,  1973 ................       70
 11        Mercury Emissions  to Air in Kentucky/Tennessee Study
          Region,  by County,  1973 ................       71

 12        Latitude/Longitude  Analysis of Mercury Emissions  along
          Tennessee  River, Kentucky/Tennessee  Study Region, 1973        72

13        Mercury  Emissions to Air in New York/New Jersey
          Study Region, by County,  1973  .............       78

14       Mercury  Emissions to Water  in  New York/New Jersey
         Study Region, by County,  1973  .............       80
                                     viii

-------
                                   FIGURES
                                  (Continued)
Number                                                               Page
  15      Mercury Emissions to Land in New York/New Jersey
          Study Region, by County,  1973	        81
  16.      Latitude/Longitude Analysis Areas in New York/
          New Jersey Study Region 	        82
  17      Annual Elevated Mercury Losses to Air	        92
  18      Annual Elevated Mercury Losses to Water 	        93
  19      Annual Elevated Mercury Losses to Land	        94
  20      Mercury Emissions as a Function of Population
          Density	        95
  21      Summary of Mercury Losses to the U.S. Environment
          by Sector and Class, 1973	        99
  22      Flow Diagram of Mercury Recovery Using 1973
          Technology	       Ill
  23      Mercury Recovery Using 1983 Technology	       113
  24      1973 Technology for Copper Mining and Smelting  . . .       119
  25      1983 Technology for Copper Mining and Smelting  . . .       122
  26      Maximum Mercury Vapor Concentrations for Cases 1-4  .       126
  27      Mercury Losses from Typical Power Plants, 1973  ...       133
  28      Mercury Losses from Typical Power Plants, 1983  . . .       135
  29      Mercury Losses from a Typical Chlor-Alkali Manufac-
          turing Plant, 1973	       143
  30      Mercury Losses from a Typical Chlor-Alkali Manufac-
          turing Plant, 1983	       148
  31      Mercury Balance for the Preparation of Mercuric
          Sulfide and  Mercuric Chloride 	       159
  32      Mercury Balance for Two Methods for Production of
          Red Mercuric Oxide	       160
  33      Mercury Balance for Synthesis of Phenylmercuric
          Acetate  	  .....       161
  34      Wastewater Flow in  a Plant Producing Mercurials and
          Nonmercurials  	      163
                                         ix

-------

Number
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
FIGURES
(Continued)

Manufacturing Process for Vinyl Chloride Monomer
Using a Mercury Catalyst 	
Mercury Loss in Manufacturing, by Type of
Battery, 1973 	
Plow Diagram of Mercury Losses during Manufacture
of Ruben Mercury Cell Batteries 	
Flow Diagram of Estimated Mercury Losses during
Manufacture of Fluorescent Lamps 	
Flow Diagram of Mercury Losses in Typical Control
Flow Diagram of Mercury Losses in Typical Switch
Flow Diagram of Estimated Mercury Losses in
Typical Paint Manufacturing Operation in 1973 . . .
Flow Diagram of Estimated Mercury Losses from
Typical Paint Manufacturing Operations in 1983 . .
Mercury Balance, Agricultural Pesticide Manufac-
Mercury Balance, Nonagricultural Pesticide Manu-
Mercury Losses from Laboratory Usage, 1973 ....
Mercury Losses from Laboratory Usage, 1983 ....
Mercury Balance, Dental Applications, 1973
Mercury Balance, Dental Applications, 1983
Mercury Balance, Pharmaceuticals Manufacture and
Use , 1973 Technology 	
Path of Mercury in a 100 MGD Sewage Treatment
Plant 	
Flow of Mercury in the Municipal Solid Waste Stream
of a Tvoical Community of 500,000 in 1973 	

Page
164
171
175
183
191
192
201
201
209
214
219
219
223
225
229
235
242

-------
                                   FIGURES
                                 (Continued)
Number                                                              Page
  52      Flow of Mercury in the Municipal Solid Waste
          Stream of a Typical Community of 500,000 in 1983 .  .      243
  53      Sequential Mercury Losses in Electrical Applica-
          tions, 1973	     248
  54      Sequential Mercury Losses in Paint Applications, 1973     249
  55      Rate of Introduction of Pollution Control Technology      264
  56      Time-Phased Reduction of Mercury Emissions from
          Mercury Cell Chlor-Alkali Plants, and Associated Costs    268
  57      Product and Income Flow for the Production and Con-
          sumption of Mercury-Containing Paints 	   273
  58      Price Elasticity of Demand in the Paint Industry  .  . .   276
  59      Consumer and Manufacturing Costs Associated with the
          Removal of Mercury from Paint.	   279
  60      Time-Phased Reduction in Mercury Losses from Paint
          Usage, and Associated Costs  	   281
  61      Product and Income Flow for  the Production and Con-
          sumption of Mercury-Containing Batteries	   283
  62      Time-Phased Reduction in Mercury Losses from Battery
          Recycling, and Associated Costs  	   290
                                        xi

-------
                                 TABLES
 Number                                                              Page

   A       Comparison of Regional  and  National Inventory Results .      d
   1       Mercury Consumption in  the  United States, 1965-
           1973,  by End Use	     13
   2       Sales  of Manufactured Products Containing Mercury ...     19
   3       EPA Mercury Regulations	     32
   4.       Input/Output Model  Sectors   	     44

   5       Inputs into the Input/Output Inventory  	     47
   6       Outputs from Various Programs in the Input/Output
           Model	     49

   7       Emissions and Emission  Factors for Mercury, by
           SIC and Sector	     55
   8       Natural Degassing Rates Calculated by Various
           Workers	     57

   9        Mercury Emissions to the Environment in the
           California  Study Region 	     63
 10       Mercury Emissions to  the Environment in the
          Arizona Study  Region  	     66
 11       Mercury Emissions to  the Environment in the
          Kentucky/Tennessee Study Region 	     68

 12       Mercury Emissions to the Environment in the
          Louisiana Study Region  	     74

 13       Mercury Emissions to the Environment in the
          New York/New Jersey Study Region  	     76

 14       Per Capita Distribution of Mercury Losses  to Air,
          Land,  and Water in Five Areas of New York/
          New Jersey Study Region 	     83

 15       Total Mercury Losses in 1973 for the Coterminous
          United  States, by  Sector and SIC Category  	      84

 16       Total Mercury Losses in  1973 for the Coterminous
          United  States, by  State  	      86

17       Comparison of Regional and National  Inventory
          Results	      90

18       Production of U.S. Mercury Mines  in  1973	     109
                                       Xii

-------
                                     TABLES
                                   (continued)
Number                                                              Page
  19      Estimated Total Production and Mercury Emissions
          from Mercury Mining and Refining:   1973  and  1983  .  .       115
  20      Estimated U.S. Smelter Production  of Copper, 1973  .       118
  21      Estimated Emissions from Copper Smelting,  1973  and
  •       1983	      124
  22      Power Plant Mercury Losses in the  United States,
          1972	      130
  23      United States Chlor-Alkali Manufacturers and Esti-
          mated Mercury Losses of Plants, by State,  1973   . .  .      141
  24      Materials Balance for Chlor-Alkali Plant,  1973   . .  .      145
  25      Materials Balance for Chlor-Alkali Plant,  1983   ...      149
  26      Estimated Mercury Losses from Typical Chlor-Alkali
          Plant Operations, 1973 and 1983	      152
  27      Estimated Mercury Consumption of Mercurials in  1973,
          by Plant and Product End Use Category	       155
  28      Estimated Mercury Losses from Production of
          Mercurials, 1973 and 1983	       169
  29      Estimated Mercury Losses from Use of Manufactured
          Catalysts in the Production of Vinyl Chloride
          Monomer and Vat Dyes	       169
  30      Major U.S. Battery Manufacturers' Mercury Losses by
          Plants and State, 1973	      173
  31      Estimated U.S. Battery Manufacturing  and Related
          Consumer Losses of Mercury, 1973  and  1983	      177
  32      Estimated Mercury Losses from Electric  Lamp Manufac-
          turing in the United States, by State,  1973	      180
  33      Estimated Mercury Losses from Lamp Manufacturing
          and Use, 1973 and 1983	       184
  34      Estimated Mercury Losses from Industrial  Instrument
          Manufacturing, by State, 1973	       187
  35      Estimated Mercury Losses from Industrial  Instrument
          Manufacture  and Use,  1973  and 1983	      194
                                      xiii

-------
                                   TABLES
                                 (Continued)
Number                                                               Page

  36      Mercury Losses from U.S. Paint Manufacturing
          Operations, by State, 1973	       197
  37      Estimated Mercury Losses from Paint Manufacture
          and Use, 1973 and 1983	       204

  38      Regional Distribution of Industrial Wastewater
          Discharges to Community Sewers in 1968	       232
  39      Wastewater Discharges to Community Sewers in 1968,
          by Industrial Category  	       233
  40      Estimated Mercury Emissions from U.S. Municipal
          Sewage Treatment Plants, 1973 and 1983	       238
  41      National Solid Waste Disposal Practices 	       239

  42      Estimated Mercury Losses to Environment from U.S.
          Municipal Solid Waste Disposal, 1973 and 1983 ....       245
  43      Mercury Losses for Various Applications, 1973 ....       247

  44      Emission Control Technology for Mercury Cell Chlor-
          Alkali Industry:  Estimated Annual Cost and Emissions,
          1973-1983	       265
  45      Estimated Product and Income Plow for Production and
          Consumption of Paints Containing Mercury,  1973  . . .       275

  46      Estimated Income and Employment Consequences of a
          Shift  to  Substitutes for Mercury in Paint  Manufacture
          and Use	       278
 47       Estimated Product and Income Flow for Production
          and Consumption  of Batteries Containing  Mercury,
          1973-1983	       286
                                    xiv

-------
                                                        EXECUTIVE SUMMARY
PURPOSE OP THE STUDY

     The toxic nature of mercury and its compounds and the relative ease
with which these materials may, when improperly discharged to the  environ-
ment, enter into man's food chain and affect him adversely have been rec-
ognized only in recent years.  Our understanding of mercury's transfer
mechanisms within nature and of its adverse effects on man is still lim-
ited.  Little reliable data are available on quantities of mercury in the
environment and it is difficult to assess whether mercury may actually be
building up in the biosphere.  The Office of Toxic Substances of the En-
vironmental Protection Agency has funded this study, recognizing that more
definitive data are needed concerning the amount of mercury entering the
environment each year and on the manner in which this mercury is redistrib-
uted  (particularly with regard to introduction into man's food chain).
The Agency also wishes to determine what requirements there may be for
regulatory actions to reduce unnecessary or unsafe discharges of mercury
to the environment.

METHOD OF APPROACH

     Quantitative data are generally not available to pinpoint the loca-
tions of mercury in the environment  (whether it be in air, water,  or land).
Therefore, URS Research Company used a study approach which traced all
sources of introduction of mercury and its compounds into the environment
and then determined where losses were incurred.  This input/output study
required that all mercury sources be identified and that the mercury be
traced through usage to final disposition for each source.  Five U.S.
regions were selected, each having some unique characteristics with re-
spect to usage of mercury, and an inventory was made of mercury usage in
these regions.  A similar inventory was made for the nation, and the
regional and national results were compared.  The data base was very  large,
so a computer-based inventory model was developed.  The regional and  na-
tional inventories formed the basis  for assessing which were the most im-
portant sources of discharges of mercury to the environment.   An addition-
al study was then made of the technology involved with respect to  these
identified emitters, and forecasts of possible  future trends were  developed.
Finally, three technologies were  selected  for evaluation of the economic
impact of regulatory actions with respect  to mercury.

USES  OF MERCURY

      Mercury and its compounds are  used in relatively small quantities  (par-
tially because of  their high cost),  but the uses are distinctive and varied.
Mercury metal is the only  common liquid metal;  because of its conductive

-------
 characteristics it has many specialized electrical applications.  Many mer-
 cury comppunds are chemically and biologically active and are uniquely use-
 ful as chemical reagents and biological pesticides and preservatives.   Fig-
 ure A shows major uses for mercury and its compounds in 1973.  The ways in
 which the slightly less than 1,900 tonnes* were used by industry and commerce,
 and the resultant emissions, are a major concern in this report.  The use
 of mercury has been declining in recent years both because of the concern
 over its possible hazards and because of its cost (currently about $4 per
 pound).  Production of mercury in this country has been particularly hard
 hit because of control technology requirements and depressed prices.  Some
 industries — notably paper and gold mining — formerly used mercury or mer-
 cury compounds but no longer do so, primarily because of the recognized haz-
 ards of the materials.

      Mercury also comes from some "unregulated sources" such as power plants
 and livestock manure.  Mercury occurs in many natural materials, and fossil
 fuels are among these sources.  Coal in particular is high in mercury (al-
 though we are talking about perhaps one part of mercury for every million
 parts of coal).  With the increasing use of coal as a power source, mercury
 emissions — primarily to air — are likely to increase.

      Mercury is present in trace amounts in virtually all soils, and these
 soils release mercury vapor to the air in trace amounts (the so called "de-
 gassing" phenomenon); over a large land mass this degassing effect accounts
 for large amounts of mercury.

      Mercury and its compounds are notorious for moving from one medium to
 another.   Thus mercury compounds discharged to a water stream may settle as
 insoluble compounds in the sediments, there to be acted upon by bacterial
 action ("methylation"),  forming soluble toxic mercury compounds which  have
 been demonstrated to  enter man's food chain via fish.  Although these  com-
 pounds generally enter the chain in small quantities, they are undeniably
 toxic; hence we must  always be concerned with the ultimate fate of mercury.

 STUDY RESULTS

     Most of the mercury  consumed each year eventually reaches the environ-
ment.  A summary of the findings of the regional and national inventories
is given in  Table A for major  man-related sources of emissions.   It can be
seen that the  study regions  deviate,  sometimes  markedly,  from national values*
these deviations are  related to  specific,  known sources of mercury emis-
sions.  For  example,  copper  mining  in Arizona causes the high emissions
value for "other mining."  Chlor-alkali manufacturing (for the production
of chlorine) accounts for the  high  "manufacturing" loss in the Louisiana
study region.  Certainly of  interest  are  the  high mercury  losses  accounted
* The (metric) tonne is about 10 percent heavier than the short ton of
  2,000 pounds.  One tonne contains 1,000 kilograms  (kg), or 2,205
  pounds.

-------
Lamps and
   Electrical ^
  3.2% (60.0)
                              Chi or-Alkali
                             24.1% (450.5)
              Instruments
             13.2% (246.6)
Pharmaceuticals
1.1% (20.9)
Miscellaneous
3.7% (69.2)
Laboratory/Use
1.2% (22.7)
Pesticides
3.4% (63.1)
Catalysts
1.3% (23.4)
      SOURCE: URS Research Company.
    Figure A  PERCENT DISTRIBUTION OF CONSUMPTIVE USES OF
              MERCURY, 1973 (Tonnes)

-------
                    TABLE A. - Comparison of Regional and National Inventory Results
                                                  (percent)
Sector
Name
Mercury Mining
Other Mining
Unregulated Sources
Manufacturing
Industrial Consumption
Personal Consumption
Total
National
0.5%
4.0
10.0
21.0
19.0
45.5
100.0%

California
11.5%
0.5
5.0
5.0
20.0
58.0
100.0%

Arizona
0 %
64.0
4.0
2.0
6.0
24.0
100.0%

Kentucky/
Tennessee
0 %
1.0
44.5
12.0
0.5
42.0
100.0%

Louisiana
0%
0
5
71
5
19
100 %

New York/
New Jersey
0%
0
21
7
6
66
100 %

Source:  URS Research Company

-------
for by the final consumption of mercury-containing products, both by  in-
dustry and individual consumers Cfor instance,  batteries/  dental fillings).
The high aggregate losses at the consumer level create  special  control prob-
lems, since they are made up of many small losses  by many  persons and users.

     We are concerned not only with size of the losses  but also with  points
of loss.  Mercury wastes directed to regulated  landfills,  which do  not leach
to groundwaters, are of less concern than are discharges to air or  to water,
which are more directly accessible to man and his  food  chain.   (It  should be
noted, however, that two-thirds of all solid wastes  are disposed of in un-
regulated sites, for which data are lacking.)  Figure B shows estimated  1973
losses of mercury and its compounds to these two reservoirs—air and  water—
from man-related applications or uses.

     The man-related emissions must be viewed in the perspective of all  mer-
cury emissions.  Figure C shows all mercury emissions to air, water,  and
land from both man-related and natural sources. Emissions from natural
sources, which are difficult to estimate, appear to  outweigh man-related
emissions.  However, these natural sources are  relatively  uniformly distrib-
uted and do not concentrate specifically in man's  habitat. Man-related  dis-
charges, on the other hand, concentrate in population centers,  and  a  direct
relationship between total mercury discharges and  total population  can be
demonstrated.  We found that, aside from industrial  and mining  applications,
as a rule of thumb the estimated emissions* of mercury  on  a per capita basis
are:  air, 1.80 g; water, 0.39 gj land, 4.25 g.  It is  noteworthy that the
emissions to water are by far the lowest on a per capita basis.  This indi-
cates that ongoing practices have been quite successful in reducing these
discharges.

TECHNOLOGY ASSESSMENT

     The report considers in some detail current  (1973) and forecast (1983)
technologies which may affect the rate of mercury discharge to the environ-
ment.  The technology assessment identifies  production  sites (where  appro-
priate) , the point of discharge, alternatives which might be employed to re-
duce reliance on mercury, and estimated total discharges  in 1973 and 1983.
Some highlights of the findings are  summarized below:

     o    Mercury mining may increase sharply but improved technology
          will prevent concurrent rises  in emissions.

     o    By adopting available technology,  copper  smelters can  re-
          claim appreciable quantities of mercury and  reduce the en-
          vironmental burden from  smelting.

     o    Manufacturers  of  chlorine who use the mercury cell process
          can, by applying  available technology,  reclaim much of the
          mercury that  is currently lost (primarily to land).  By
          1983 more  than 75 percent of the mercury used  in this process
          can be reclaimed.
* In grams  (g); 28 grams =  1  ounce.

-------
                                            Unregulated
                                           22.2% (104.5)
      Miscellaneous
        1.8% (8.76)
  Lamps, Tubes, Switches
             2.9% (13.6)
           Control Instruments
              3.5% (16.5)
              Chlor-Alkali
              3.1% (14.8)
Other Manufacturing
2.4% (1 1.2)
                                                                                             Mining
                                                                                             Unregulated
                                                                                             (Fossil fuels, etc.
                                                                                             Manufacturing
                                                                                             Final Consumption
                                        SOURCE:  URS Research Company.
             Figure B-1   PE RCENT DISTR I BUT ION OF ESTIMATED  EMISSIONS TO AIR OF
                          MERCURY FROM MAN-RELATED SOURCES, 1973 (Tonnes)
                                       Dental
                                     19.0% (16.6)
                  Pharmaceuticals
                   20.3% (17.8)
               Miscellaneous
               1 1.7% (10.3)
                                              Chlor-Alkal
                                              12.0% (10.5)
                              Pesticides
                             23.2% (20.4)
                                                               Laboratory/Use
                                                                 6.8% (5.9)
                              Mining
                              3.5% (3.1)
                              Unregulated
                              2.5% (2.2)
                                                          Mining
                                                                                            Unregulated
                                                                                            (Fossil fuels, etc.
                                                          Manufacturing
                                                         Final Consumption
                                                    Other Manufacturing
                                                    1.0", (0.9)
SOURCE:  URS Research Company.
          Figure  B-2  PERCENT DISTRIBUTION OF ESTIMATED EMISSIONS TO WATER OF
                      MERCURY FROM  MAN-RELATED SOURCES, 1973 (Tonnes)

-------
                             TOTAL: 2,732 Tonnes
                                   an-Related Emissions
                                  Natui al Emissions
     SOURCE.  URS Research Company.
Figure C  ANNUAL EMISSIONS OF MERCURY TO THE ENVIRONMENT FROM
         MAN-RELATED AND NATURAL SOURCES (Tonnes)

-------
 Manufacturers  of mercury compounds are currently minor
 sources of emissions and will become even less important
 in  future years.

 The battery  industry — one of the largest consumers of
 mercury — is  likely to grow in the future; unless the
 current rate of 5 percent mercury recycling is substan-
 tially increased, emissions from this source will con-
 tinue to rise  appreciably.

 Lamps are fairly substantial users of mercury, and there
 is  little possibility of decreasing this usage by re-
 cycling mercury from lamps.

 Dental usage of mercury will continue, but unnecessary
 losses will  be sharply curtailed.

 The use of mercury in Pharmaceuticals is declining, but
 there will continue to be some demand for mercury for
 unique applications.

 The use of mercury in paints is currently under attack
 by the Environmental Protection Agency; even if this
 does not result in total banning of mercury in paint,
 such usage is at least likely to decrease.

 The use of mercurials in agricultural applications has
 already been sharply curtailed,  and even lower usage is
 forecast for the future;  nonagricultural usage, primarily
 on golf courses, is also likely to decrease.

 Mercury emissions  from power plants will continue to
 grow,  and present  technology only partially removes
 mercury from stack gases.

 Sewage plants receive wastes containing mercury from
 both natural  and man-related sources,  concentrate a
 portion of this mercury in  the sludge,  and discharge the
 remainder to  the receiving  waters.   If the sludge is
 burned, mercury is released to the  air;  if the sludge
 is used as a  soil additive,  mercury (and other heavy
 toxic metals) may build up  to dangerous levels.

 Of the solid  wastes generated in this  nation,  an average
 of about 92 percent are disposed of in landfills and
 about 8 percent are incinerated.  The  trend is toward
greater material and energy recovery, with an  attendant
decrease in waste volume.  Greater energy recovery
could result  in more emissions to air, but increased

-------
          material recovery can be designed to decrease mercury
          emissions through the segregation and removal of
          mercury-containing wastes.

     The control of mercury emissions  is most feasible at the manufactur-
ing or processing level.   For this reason control at this point has progressed
ahead of controls for mercury losses from other sources of emissions.  Con-
tinuing attempts to reduce mercury emissions at all points of their entry
into the environment are  desirable.   (Of course, natural emissions, although
large, are not controllable; however,  because of their diffuse nature they
are of less concern than  the man-related emissions that collect where man
congregates.)

REGULATORY IMPACTS

     An analysis was made of several alternatives that might be imposed by
regulatory action, and of the economic impacts of such alternatives.  The
alternatives, designed to reduce  or eliminate mercury emissions, include
the use of a substitute product or  an  alternative technology that does not
require mercury or its compounds; improved emission controls  (which may
allow reclamation of mercury); and  recyling of mercury from mercury-containing
products.  Three case studies were  considered:  the chlor-alkali manufac-
turing industry; latex paint manufacture  and consumption; and battery manu-
facture and consumption.

     The portion of the chlor-alkali  industry that uses mercury cells, al-
though growing, is resisting attempts  to  convert  to other technologies that
do not use mercury — namely the  diaphragm cell.  Although the two tech-
nologies are competitive, the conversion  costs  are at the present time ex-
cessively high.  The cost of converting the 28  existing mercury cell plants
in this country to the diaphragm technology,  over the ten-year period  1973-1983,
would be roughly $600 million but would result in a  100-percent reduction of
mercury emissions  (up to 326 tonnes annually).   However,  at a cost of  a little
over $30 million, these same plants could be fully equipped with emission
control technology which would reduce emissions (and consumption)  of mercury
more than 75 percent  (still leaving 81 tonnes entering the environment each
year).  The best solution seems to lie in phasing out mercury cell technology
only as plant replacement becomes necessary.

     For the paint industry only one  alternative was considered — complete
removal of mercury from paint and the use of substitute materials.  (Mer-
curials are used primarily  in water-based paints for preservation purposes
while in the can, and for protection  against mildew formation in many ex-
terior paints.)  The most immediate impact of the alternative would be  an in-
crease in the cost of paint at the consumer level of at least 50 cents  per gal-
lon.  A likely  (but not totally  demonstrated) side effect would be lower paint
quality, which would necessitate more frequent painting.  The net effect of
substitution on  the paint industry and its ancillary industries is actually
a benefit,  since  additional capacity  would be  required because of the

-------
  demands  for more frequent painting tassuming the consumer tolerates the
  change).   However,  the cost to the consumer over the 10-year period would be
  an estimated $1.4 billion (accounted for by the increase of 50 cents per
  gallon).   The benefit derived would be  the elimination of more than
  180 tonnes of mercury from the environment each year, primarily in the form
  of reduced emissions to air.   (The loss of mercury from paint occurs pri-
  marily after the paint is applied, when the vapor transfers to the air.)

       The  use of mercury in high-energy-density batteries has become more
  prominent in recent years, and the forecast is for even greater usage
  in the future.   While some substitutes  are on the horizon, these are not
  seen as an important consideration for  the period under study.  Hence the
  only regulatory action considered  was the use of "forced" recycling,
  which would have the net effect of reducing mercury losses to the en-
  vironment—primarily to land.   To  "force" the necessary degree of partici-
  pation for our case study, a  regulatory screen was imposed upon the industry
  itself so that in one case 10 percent of all mercury purchases must come
  from recycled batteries and in the second case 25 percent of mercury pur-
  chases must come from recycling.   (The  details of the recycling were not
  explored,  but a variety of schemes, based upon price incentives for
  individual returns,  can be envisioned.)  For 10-percent recycling the esti-
 mated reduction in mercury losses  over  the 10-year period was about 5 per-
 cent,  while  for  the  25-percent recycling the reduction was about 20 per-
 cent  for  the  same period.  The associated costs were $3.8 million and
 $9.4 million  respectively.

      From  the results of these two cases it appears that, in the long term,
 regulatory policy decisions threaten not so much "costs" as new equilibrium
 within the industry.  The  net effect, then, is not loss or gain, but redis-
 tribution.  Thus regulations that force contraction of mercury-related inr
 dustry will inevitably result in the expansion of other industries (which
 may merely represent conversions of the defunct mercury-related industry).

 CONCLUSIONS

      The  results of this study lead to the following conclusions:

      1.   Mercury and its compounds are used in relatively low volume com-
 pared to  other industrial chemicals, but these uses impact on a variety of
 areas and  on all phases of the economy.   Consequently,  mercury discharges
 enter the  environment at a number of points.   Of particular concern is the
 characteristic of mercurials to move, after discharge,  from one environmental
 medium to  another and to enter man's food  chain through various mechanisms.

      2.    Present monitoring capabilities  for  determining the concentration
 of mercury in the environment are  inadequate,  in both scope and reliability.
 Therefore  the available data on ambient  mercury levels  do not provide a com-
plete picture of either natural or  elevated  levels.

-------
     3.   An input/output model of sources and uses of mercury and  its  com-
pounds in the economy has proved very useful in determining the location and
magnitude of discharge of material wastes to various environmental  receptors.

     4.   About 80 percent of the almost 1,900 tons of mercury used each
year in the United States is ultimately discharged to the environment;  an
additional 8 percent is recycled, and the remainder is permanently  in place.

     5.   Differences are evident in use and discharge patterns betvteen
the various regions studied and the nation as a whole; these differences
can generally be traced to industrial-type activities (which may include
mining and smelting, combustion of fossil fuels, and manufacturing). When
such perturbations are allowed for and discounted, mercury discharges can
be shown to be a function of population density; for the nation overall,
these discharges are estimated to be as follows:  to air, 1.80 g per capita;
to water, 0.39 g per capita; and to land; 4.25 g per capita.

     6.   For the United States as a whole, the percent distribution of mer-
cury discharges from man-related sources to the various environmental
receptors is:  31 percent to air; 6 percent to water; and 63 percent to land.

     o    Less than 5 percent of the man-related discharges to air
          are from regulated industries  (i.e., chlor-alkali and
          mercury mining).  Natural emissions  (from degassing of
          the earth's crust) are estimated to be three times greater
          than the man-related discharges.  The fate of mercury and
          its compounds in air is not clear  (although they do not
          appear to present a direct hazard to man because of the
          low ambient concentrations), but the mercury eventually
          returns to the  land and to the water.

     o    About 13 percent of the man-related  discharges to water
          are from regulated industries  (currently chlor-alkali
          and mercurials  manufacturing).  Natural  sources  (pri-
          marily runoff)  contribute about two  times as much mer-
          cury to water as do man-related discharges.   Some mer-
          cury is removed from water  in sewage treatment processes
          as sludge and is either returned  to land or incinerated
          and returned to air.   Most  mercury discharged to water
          appears to  be  incorporated  rapidly into sediments and may
          be released by  biological  and mechanical action over a
          period of time.

     o   Some 25 percent of the man-related discharges to land (as solid
          waste) are  from regulated industries.  Although mercury  and  its
          compounds appear  in trace amounts in virtually all soils, these
          cannot be compared with man-related discharges.  Solid wastes
          incorporated in regulated landfills  (which currently receive
          about one-third of all solid wastes) do  not discharge appreciable

-------
            amounts  of mercury to water; the discharge of mercury vapors
            to  air from  landfills is a possibility which needs further study.
            Also needed  is additional study of mercury losses from improper-
            ly  operated  or unregulated landfills.  Solid wastes which are
            incinerated  do release mercury to air and are of major concern
            with respect to pollution.

      7.    The often expressed "threat to man" associated with the current
 rate d$ man-related mercury discharges to the environment has not, in fact,
 been conclusively  demonstrated.  Similarly, there is no conclusive evidence
 to indicate that a long-term buildup of mercury in the biosphere is occur-
 ring.  Nonetheless, the potential consequences of a mercury buildup are of
 such magnitude that efforts should continue to decrease the quantities of
 mercury discharged each year.

      8.    Regulatory actions are most easily and best carried out at a "point
 source" such  as a  manufacturing or processing facility.  However, the final
 consumption sector, which represents many points of discharge, accounts for
 60 to 70 percent of all man-related discharges to air, water, and land.  Be-
 cause regulation of these "area sources" is extremely difficult, alternatives
 which impose  regulation at more accessible points — that is, the manufactur-
 ing or processing  level — will probably be more effective.  Future regula-
 tory actions  should address the following areas:

      o    Regulatory action should ensure that man-related
           discharges to air and water are minimized by re-
           quiring the adoption of the best available technology.

      o    Federal regulations should serve as the basis for
           related state regulations,  which can be adapted to fit
           with federal criteria.

      9.    Of the  present "unregulated"  industry-type discharges, copper
 smelters,  fossil-fuel-burning power plants,  and municipal incinerators are
 the most  likely candidates  for  regulation, particularly with respect to
 emissions  to air.

      10.   The  total elimination of  mercury as a way of reducing discharges
 is generally a most expensive choice; improved control over losses is much
 less expensive and  may  meet  the desired objectives.   Long-term regulatory
policy may cause redistribution of  industrial capacity but will have little
effect on  "costs" per se; however,  costs  are  passed on to the consumer, and
these are  inevitably higher.

     11.  Mercury emissions are almost invariably associated with other types
of toxic or noxious discharges.  Multiple discharges from a single source
should be treated as a  common problem and attacked as such both at the regu-
latory end and in the technological approach.  Mercury discharged to one

-------
environmental receptor may be transferred to another receptor.   Such inter-
media transfer is best handled by one cognizant agency to ensure that con-
trols on one medium do not adversely affect another.

     12.  Natural emissions, although appreciably greater than  man-related
emissions, are much more diffuse than are man-related discharges.  Thus the
fact that natural discharges are much greater in the aggregate  should not
be used as an argument that regulatory controls of man-related  discharges
are unnecessary.

     13.  This study indicates areas where regulations should be considered
and may prove of value.  However, such regulations should be developed only
after further refinement of the data, with in-depth analyses of the affected
industries and with the cooperation of these industries.

 TECHNICAL FINDINGS

      1.    At present,  there are  not  sufficient  limitations to ensure minimi-
 zation  of the quantities of mercurial wastes which  are discharged to the en-
 vironment.   The situation  is particularly  serious with respect to mercurials
 from consumer products (primarily batteries) discarded to land.  To affect
 significant  reductions it  may  be necessary for  the  manufacturing sector  (and
 specifically battery manufacturers)  to  assume  responsibility for recycling
 mercury-containing  products.  This would reduce manufacturing needs for
 virgin  metal and would also reduce the  amount of waste going to  landfills.
 If new  regulations  are required  to achieve such actions, they will need  to
 be formulated in the light of  several implications, including those listed
 below:

      o     Product substitution or elimination  of mercury usage
           is desirable only when a potential hazard can  be  cited
           and only  when it can be clearly demonstrated that emis-
           sion  control technology will  not produce a sufficient
           reduction to mitigate  the  hazard.

      o     Regulation of mercury discharges from copper smelting,
           fossil-fuel-burning power plants, and municipal incinera-
           tion would significantly reduce mercury discharges to air.
           (EPA has  the power,  under Section 112 of the Clean Air
           Act,  to develop mercury emission standards for these
           sources because mercury has already been designated as a
           hazardous air pollutant.)

      o    New regulations, if developed in cooperation with the
           affected industries, would ensure the lowest  redistribu-
           tion impact on the specific  industry, least cost to  the
           consumer, and maximum benefits  in terms  of reducing mer-
           cury losses to the environment.
                                         m

-------
     o    Mercury discharges  are  generally  associated with dis-
          charges of  other hazardous  substances;  furthermore,
          the movement of  mercury from  one  medium to another is
          a recognized phenomenon.  An  understanding of these
          interrelationships  is most  useful in  the  formulation
          of regulatory  actions and technological developments.

     2.   In-depth data  for all media (for  selected study areas) would
establish a baseline  for ascertaining whether mercury is building up in
the biosphere.   (Little  value is  seen in adding mercury monitoring to
routine analyses or on a nation-wide  basis,  although additional informa-
tion on the natural mercury content of  some widely  used raw materials,
such as natural gas,  would be useful.)

     3.   The input/output model  utilizing  a computerized data base manage-
ment system has been  demonstrated to  be a valuable  approach for assessing
losses of toxic materials  to  the  environment.   The  model appears to be
readily adaptable for more general use.
                                      n

-------
                                                                             I
                                                                  INTRODUCTION

BACKGROUND

     Mercury and its compounds have been used by society since antiquity but
did not come into widespread and varied usage until modern times.  Mercury,
which is unique because it is the only liquid metal in common use, finds numer-
ous applications which utilize its conductive properties.   Its compounds are
some of the most effective mildewcides,  pesticides, and preservatives  avail-
able; they also have many industrial and agricultural applications.  Mercury
compounds serve as unique catalysts for certain reactions and are  useful in
some pharmaceutical applications.  In addition,  mercury's ability  to amalga-
mate has led to its widespread use in dental applications.
     However, although the use of mercurials* has developed as a necessary
adjunct to society's growth, they have also earned a  notorious reputation  as
toxic substances, and it is widely recognized that they can cause  disabling
sickness and death.  While the mercurials have generally been treated  with
caution with respect to direct human exposure, it was not recognized until
recent years that when they are discharged to the environment they can enter
man's food chain and lead to illness and fatalities.   Hence the unfortunate
episodes in the late 1950s at Minamata and Niigata in Japan when industrial
discharges of mercury to water entered the human food chain via fish,  subse-
quently poisoning a significant portion of the population of these areas and
introducing "Minamata disease" into our lexicon.
     Alerted by the Minamata experience, governments and  industry began inten-
sive investigations of the pathway of mercurials into  the environment and of
the threat they might offer to man.  It was  quickly  established that  the
* For simplicity, the term "mercury"  or  "mercurials"  will be used henceforth
  to mean both mercury and its compounds,  unless otherwise specifically noted.

-------
  alkylmercury compounds were by far the most toxic to man;  they have a long
  residence time in the body (a biological half-life of 70 to 74 days) and cross
  both placental and blood-brain barriers.   In the  latter case, disabling damage
  can occur.   It was further found that the labile  substance mercury, through
  bacterial or biological action in the environment,  can undergo changes from
  relatively innocuous forms to the very toxic methylmercury, which can be con-
  centrated in the food chain.
       As a result of these findings,  industrial  discharges  and many pesticides
  and preservatives which heretofore had not been regulated  were identified as
  being harmful to the environment and to man,  and  immediate steps were taken both
  by the government and by industry to limit or eliminate unsafe and unnecessary
  applications.   Several problem areas were identified  (for  example discharges to
  water from  chlor-alkali plants* using mercury cell  technology) and corrective
  actions were  taken to limit such discharges.  Other, more  stringent actions
  were  taken  with  the suspension of all uses  (including manufacture) of any of
  the alkylmercury compounds.  In yet  other  instances the U.S.  Environmental
  Protection Agency (EPA), which had become very instrumental both in identifying
  the extent of the problem and  in ensuring that industrial discharges were
 brought into compliance with standards, had taken action to force the cancella-
 tion of the use of mercurials as biocides and preservatives.
      As a result of the recognition of the situation  and the actions taken to
 alter it, the total use of mercury in the United States has dropped significantly
 (from 73,560 flasks in 1965 to 54,289 flasks in 1973).**   Perhaps more signifi-
 cantly, a number of uses of mercury were completely eliminated,  in cases where
 substitute materials or alternate processes could readily be  adopted.   For
 example,  mercury was formerly  used (albeit in small quantities)  for the recovery
 of  gold by an  amalgam process;  mercury is  no longer used  in this application.
 * Throughout the report, when we refer to chlor-alkali plants,  unless
   otherwise noted we mean plants that use mercury cells.
** A flask of mercury contains 34.4 kilograms  (76 pounds);  there are
   approximately 29 flasks in a metric ton (2,205 pounds).

-------
Likewise, mercurials were long used as preservatives in the lumber,  pulp,  and
paper industries; today these uses have also been eliminated.
     Why, then, since the use of mercury is decreasing,  its discharge to the
environment is under better control,  and its properties of concentrating in the
food chain are now recognized, is an additional study of mercury necessary?
There are several reasons.  First, some researchers are very concerned about
the possibility that any mercury unnecessarily introduced into the environment
becomes available to the so-called "mercury cycle" (which is discussed in
detail later) and leads to increasing concentrations in the environment.  In
short, because of its unique properties mercury may continue to recycle indefi-
nitely within the biosphere.  According to adherents of this theory,  strict
limits should be placed on the introduction of mercury into the environment in
order to minimize the present and potential impacts created by the gradual and
continued increase of mercury in the cycle.  (For more detailed descriptions,
see Refs. 1-4.)*
     The concern regarding possible increases of mercury in the biosphere has
led to more monitoring for mercury and to its discovery, albeit in low levels,
almost universally.**  Appendix A gives more details concerning our present
understanding of mercury sources and cycling and of its effects on the air,
water, and land environments.
     The recognition that mercury is widespread, if not ubiquitous, has given
rise to a second school of thought which theorizes that mercury is naturally
present in the environment as a result of degassing from the earth's crust and
as runoff from natural erosion processes.  These natural processes add  to the
available mercury in the biosphere at a constant rate, but  as yet there is  no
 * Numbered references are  listed at the  end of  this report.
** Until recently, one of the difficulties  in evaluating the  presence of mer-
   cury has stemmed from the use  of analytical techniques which often were
   limited to 0.5 part per  million  (ppm)  or greater and whose reproducibility
   of results was poor.  More sophisticated analytical techniques—primarily
   thin layer chromatography  (TLC)—have  now been developed which can reliably
   detect mercury in  concentrations as low as one part per trillion  (ppt).

-------
  evidence ,-to support the belief that there has been a  sharp increase in global
  mercury concentrations in the last 50 to 100  years (Refs. 5-8).  Therefore, a
  second school of thought has arisen which holds  that  the processes that maintain
  an equilibrium with respect to natural emissions easily handle the small addi-
  tional discharges that are man-related.   Those who adhere to this theory feel
  that the natural processes tend to maintain some sort of equilibrium between
  mercury discharged into the environment  and that removed by natural processes,
  so that even rather large quantities of  man-related discharges do not adversely
  affect the  mercury balance worldwide or  over  a long period.   (However, these
  people are  not advocating any relaxation of controls on discharges which can
  directly or indirectly enter  man's  food  chain.)  On this premise, the need to
  reduce or eliminate mercurials  in certain industrial applications has been
  strongly questioned by some industry representatives.  In particular, the goal
  of eliminating mercurials  as biocides and preservatives is contested.   Industry
  representatives also point out  that  unregulated emissions from such sources as
 power plants burning fossil fuels are permitted—and the quantities involved
 are appreciable—whereas the known beneficial uses are to be prohibited.
      Man-made emissions, however, are usually localized and  thus may result in
 localized high concentrations.  In the past this was true of emissions to water
 from, say,  a chlor-alkali plant.  Emissions to water have recently come under
 strict regulation, but localized emissions to air (for instance from an incin-
 erator or power plant)  from unregulated sources may indeed pose hazards.
      Two questions seem to be still unresolved.  First, what  are the  sources
 that  contribute mercury to the biosphere, and  what are the quantities  involved?
 If  in answering this question it can be shown  that certain of these sources can
 be  limited or even eliminated,  then the second question must  be addressed.
 This  question is:   what will be the impacts of such restrictions?  (We will not
 deal  at this time  with  the suggestion that introduction of mercury into the
 environment  is beneficial rather than detrimental,  but will adhere  to  the
general  conclusion that unnecessary  introduction  of mercury into  the environment
is indeed undesirable.)  The EEA, seeking answers to these two basic questions,
lias funded the study whose  results are reported here.

-------
SCOPE OF WORK

     The EPA set forth the following scope of work  for the  study:   "The con-
tractor shall prepare a detailed materials balance  on mercury and  its com-
pounds.  He shall assess quantitatively the distribution on an input/output
basis nationally and for selected geographic regions.  He shall  include esti-
mates of the environmental fate of the quantities necessary to close, the
balance.  In addition, the contractor shall consider current and projected
process technologies for mercury products,  indicating how alternative technol-
ogies influence the materials balance.  He shall carry out  analyses of economic
and market trends that affect the materials balance, and assess  the feasibility
of substitute processes and materials that will significantly reduce the amounts
of mercury entering the environment."
     This scope of work is designed to answer the two questions  posed in the
previous section.  By identifying all sources that  discharge mercury into the
biosphere and the relative quantities discharged, and by deriving  a materials
balance indicating the final disposition of this mercury  (fixed  in place or
discharged to land, air, or water),  the relative importance of both sources
and discharges can be evaluated.  A part of this concern is the  ultimate fate
of mercury, recognizing that movement between the  earth, water,  and air is
characteristic of the cycling of mercury.
     The study considered both national and regional materials balances;
regional inventories were made to ascertain differences which may exist between
regions and to accent special problems which may be peculiar to certain regions.
The study also considered those points where mercury enters the environment for
which better controls (which may include substitution or elimination) may exist.
Since the introduction of controls is certain to have economic  impacts (with
possible social implications) and may also  entail  certain  risks to society,
both the costs and benefits of instituting  such controls were considered and
their impact on the marketplace was evaluated.  The basic  objective of the
study, then, was to determine the need  for additional controls  on mercury dis-
charges and the cost  to the individual  and to  the  overall  economy for imple-
menting such controls, as related to the benefits  accrued.

-------
  METHOD OF APPROACH

       A sequential study approach was used,  beginning with the inventory
  (including an input/output model with a materials balance),  which considers
  all sources in identifying major perturbations and differentiates major  from
  minor contributors.  Next a technology assessment was made which  considers,
  for major contributors only,  technological actions which might be taken  within
  the next 10 years to appreciably reduce the emissions of mercury  to the
  environment.   Finally,  the economic impact of regulatory controls on major
  users of mercury was evaluated.
       An input/output model for the mercury economy can be readily derived,
  since annual  production and consumption figures are available (these include
  data with some degree of reliability concerning recycling of mercury and
  releases  from the national stockpile).   At  the  national  level, a  number  of
  material  balances had been made  previously.  A  pioneer effort was completed
  by  Davis  in 1968  (Ref.  9),  and there have been  several updates (Refs. 10-13).
  Other  independent estimates have been made  by other investigators.  Thus the
  input/output model as a concept  is well  established and  the  study inventory
 was based upon these earlier results.  However, this study,  which is keyed to
 the year  1973 with estimates for 1983, also afforded an  opportunity to include
 later and better  estimates.
      The  introduction of a regional inventory is a new concept.   It permits
 making comparisons of various regions to determine if  significant variations
 occur, thus making it possible to identify  special problem areas.   At the same
 time,  making the regional inventory is very complicated,  because the values
 utilized for the national estimates must be adapted to the specific region
 under study, where national values no longer apply.   (The inventory and  its
 concomitant materials balance  utilize emission factors dealing with the propor-
 tions of mercury lost to air,  water,  land,  or recycling.   In some cases these
 factors must also be tailored  to the specific region studied, even though the
 variations may be slight.)
     Five  regions were selected for study on the bases of (1) known discharges
as derived from the EPA  permits program,  (2) known high discharges to air or
water, and (3)  to  a lesser  degree,  differing socioeconomic characteristics of

-------
the regions.  It was hoped that the discharges identified for a given region
might correlate with ambient mercury values in that region; however,  ambient
readings are generally inadequate to permit such correlation.  This problem
is discussed in greater detail subsequently.
     Study regions were selected on the basis of a  map prepared by the Office
of Toxic Substances of the EPA that shows elevated  mercury concentrations  in
the United States.  This extensively researched map showed:
     o  High and low discharges of mercury to water according to
        permit applications in the RAPP file
     o  High ambient air readings from unknown sources
     o  High ambient water readings from unknown sources
     o  Known mercury deposits
     o  Known gold and silver deposits
From this map it was possible to see that certain regions have a probability
of unusually high discharges of mercury to the environment from natural  and/
or man-related sources.  These "high discharge" regions, which are  shown on
Figure 1, were also selected with a view to geographical distribution, popu-
lation characteristics, and manufacturing activity.  A brief discussion  of
the regions is given in Appendix B.
     Technology assessment is concerned only with those sources  of  mercury
which can potentially be controlled  (and thus excludes natural contributions)
and which are important contributors.  This assessment, then, deals with loca-
tion of discharges and quantities involved and with control technology which
is or may become available to reduce discharges.  In addition,  alternative
materials or processes which would substantially reduce or entirely  eliminate
requirements for mercury are assessed.  The technology  assessment reported
here provides a state-of-the-art study for each source  of mercury considered
important and an  estimate of the degree to which further controls on that
particular  source might effectively  reduce the quantity of mercury  reaching
the environment.  This assessment indicates where  additional technological
improvement (including substitution of products or processes) might be of
considerable value.  The assessment also identifies  for more detailed study

-------
00
                                                 ^ Estimated 1973             .      .
                                                 • „                  O Area (sq. km)
                                                   Population
                                                                                                                           24,478,200
                                                                                                                        O  32,813
    • 206,343,000
~-  O 3,022,260
  |_
  f
                                                                                                       • 7,329,800
                                                                                                       6
                                        • 2,144,542  7
                                        O 299,272
                                                                           •"2,661,700
                                                                           O  59, 756    i
       SOURCE:  URS Research Company.
                                                    Figure 1   STUDY REGIONS

-------
the major man-related sources of mercury discharged into the  environment  that
are amenable to regulation.
     As an adjunct to the technology assessment,  an investigation was  made of
the ultimate fate of mercury in the environment—how it may move  endlessly
between land, air, and water.  Of particular  concern here  is the adequacy of
present waste disposal techniques (particularly with regard to landfills,
sewage treatment plants,  and incinerators)  in minimizing the  potential for
discharged mercury to recirculate in the environment.
     The study of the socioeconomic impacts of imposing regulatory  controls to
reduce mercury discharges considered both the assumed benefits and  the short-
comings of several use alternatives.  These alternatives include  using other
chemicals to replace mercury in existing processes and products,  alternative
processes to produce the desired products,  and alternative  final  use products
which might obviate the need for mercury precursors.  Any of  these  alternatives
may involve risks such as the introduction of other toxic or  undesirable  by-
products into the environment or a direct impact on the user  in terms  of
reduced yield, less satisfactory product performance,  etc.   Since benefits are
not derived directly but are based on the total reduction of  mercury discharged
to the environment, risks must be treated as part of total cost.  Total cost
thus includes, in addition to the cost represented by risk, a variety of
economic indicators and techniques which must apply to the specific industry
under investigation.  The cost versus benefit analysis, then, reflects the
cost of reducing the amount  of mercury entering the environment by a certain
fraction (or absolute number).  The willingness of the public to endure  such
costs to meet specific goals in reducing mercury emissions must then be
assessed.

REPORT ORGANIZATION

      Subsequent sections of the report provide the details  of the methodology
 and results from which the conclusions and recommendations  presented in the
 executive summary are derived.

      Section II is a historical overview  and outlook summary for  mercury
 usage.  This section provides insights  into current trends  in the consumption
                                        9

-------
  of mercury and lays the groundwork for  the remainder of the report by giving
  a first iteration of future trends.  The  "mercury economy," which ranges from
  mining to manufacturing to consumption  to final disposal and then to the ulti-
  mate fate of mercury in the environment,  is described, as are legislative im-
  plications.  The section provides  much  background information but is not essen-
  tial to the understanding  of the remainder of the report.

       Section III describes the  input/output model and the methodology employed
  in determining losses of mercury to the environment.  The results for the five
  study regions are presented and compared with those for the national inven-
  tory.   Implications  on  possible risk to the population and the importance of
  control over emissions  which may enter  man's food chain are discussed.  Fin-
  ally,  a description  is  given of the inventory results as the basis for selec-
  tion of technologies for more detailed  study.

      In Section  IV technologies which entail unusually heavy usage of mercury
 and/or  the possibility  of  substantial losses of mercury to the environment
 are considered.  Technologies from each sector of the mercury economy — that
 is, mining, manufacturing, final consumption, and final disposal — are con-
 sidered.  For each technology considered, current (1973)  and foreseeable
  (1983) technology is discussed in detail.  Where applicable, the sites of
 emissions are specified and alternatives which might be considered to reduce
 such emissions are indicated.  Finally,  estimates of total losses to the en-
 vironment for 1973 and 1983 are presented.

      Section V deals with the probable economic impact of regulatory actions
 designed to lower mercury emissions to the environment.   Chlor-alkali plants
 using mercury cell technology, which account  for large losses in the manufac-
 turing sector, are analyzed.  In the final consumption sector,  two high vol-
 ume users of mercury are considered:  water-based latex paints  and high energy
 density mercury-containing  dry cell batteries.

     Backup  materials and additional details of  results are  presented in
the appendixes.

                                      10

-------
                                                                            II
                                       HISTORICAL OVERVIEW AND OUTLOOK SUMMARY

     Efforts to reduce man's exposure to mercury require an  understanding of
ways in which it is lost to the environment and of its ultimate  fate.   This
section summarizes the recent history of,  and projected technical  and  commer-
cial outlook for, the principal sources,  consumption,  and final  disposal of
mercury.  The ultimate fate of mercury in the environment is also  summarized
here.  This information provides a basis for the analysis of the mercury
input/output balance and the investigation of means of reducing  losses of
mercury to the environment.  Finally, existing legislation on the  regulation
of mercury and its compounds and trends in regulation are discussed briefly
at the end of this section.

SOURCES OF MERCURY

     The principal sources of mercury entering U.S. commerce include:
     1.  Domestic primary production and stockpile
     2.  Net imports for consumption (imports less exports)
     3.  Recycled (secondary) mercury
     Between 1960 and 1973, domestic primary production dropped from 33,223
to 2,131 flasks; stockpiles ranged widely over the period, reaching a high of
31,764 flasks in 1965, but in 1972 fell to 512 flasks and in 1973 totaled
2,583 flasks.  Secondary sources reached highs of 10,000 or more  flasks in
some years during the 1960-1973 period, but overall averaged 5,000 to 8,000
flasks annually.  Net imports in 1960 were 18,814 flasks; by  1973, net imports
contributed 45,734 flasks.

PRESENT AND PROJECTED MERCURY CONSUMPTION

     Between 1962 and 1969, growth in mercury consumption averaged 2.1 percent
per year, but from 1969 through 1973 consumption actually declined by 30 percent

                                        11

-------
  overall, at an annual average rate of 8.5 percent.  This decline can be attrib-
  uted chiefly to  (1) the apprehension of mercury users concerning the probable
  tightening of legal restrictions, (2) the reduction of emissions by caustic
  chlorine (chlor-alkali) producers, and (3) the 1972-1973 business recession.
  The low point was reached in 1971, when consumption was 52,725 flasks.   In
  1973 consumption increased slightly, to 54,283 flasks,  and preliminary  data
  for 1974 indicate a 10.7-percent increase over the 1973 level.
       World market prices have been very volatile during the period 1960-1973,
  but price trends have not been the .controlling factor in the use of mercury.
  Instead, price has depended largely on the production costs of marginal pro-
  ducers who have had to demand higher prices for their output.   As the demand
  for mercury has declined,  the price  has fallen and mines have closed down.
  During the  1960s,  when prices topped $700 per flask,  149 mines  were in  produc-
  tion in the United States.   However,  the Bureau of Mines reports that only
  five mines  were producing  in the first quarter of 1974,  when the average price
  was about $286  per flask.   In part,  the decrease in producing mines was caused
  by  the  speculation which took place  in mercury (as well  as in  other metals),
  which led to excessive  inventories and ultimately drove prices  down.
      Forecasting  the demand for  mercury ten years in  the future is  a complex
  task.   Population growth is one  obvious factor that must be considered,  but
  the  development of  new manufacturing processes and substitutes  to avoid the
 adverse effects of mercury  in some applications has introduced  new  uncertain-
 ties in forecasting.  For one thing,  legal constraints may militate against
 the economics of using mercury.  In many instances, the  substitutes used in
 order to comply with governmental regulations are  less economic  (and  sometimes
 less effective)  than the mercury or mercury compound they replace.
      Recent consumption data for the principal end uses of mercury and mercury
 compounds in the United States are listed in Table 1,  according to use cate-
 gories established by the U.S.  Bureau of Mines.  Figure 2 gives a range of
 estimated values for 1985 made by the Bureau of Mines.  Forecasts derived by
 URS Research Company in the current study are also shown; for the aggregate
 demand,  they are probably accurate to between ±10 and ±15 percent of the
 figures  shown.  For  specific uses,  the accuracy is more likely to be in the
±10 percent area.

                                        12

-------
U)
                   TABLE 1. - Mercury Consumption in the United States,  1965-1973, by End Use
                                                (number of flasks)
End Use
Agriculture
Amalgams
Catalysts
Dental preparations
Electrical apparatus
Caustic chlorine
Laboratories
Industrial instruments
Paint (antifouling)
Paint (mildew proofing)
Pulp and paper
Pharmaceuticals
Other
Total
1965
3,116


3,
18,
8,
2,
10,

8,


16,
73,
••MM
268
924
196
887
753
332
330
255
211
619
418
251
560

1966
2,374
248
1,932
2,133
17,638
11,541
2,217
7,294
140
8,789
612
232
16,359
71,509

1967
3,732
219
2,489
2,386
16,223
14,306
1,940
7,459
152
7,026
446
283
12,856
69,517

1968
3,430
267
1,914
3,079
19,630
17,453
1,989
7,978
392
10,174
417
424
8,275
75,422

1969
2,689
195
2,958
2,880
18,490
20,720
1,936
6,655
244
9,486
558
712
9,134
77,372

1970
1,811
219
2,238
2,286
15,952
15,011
1,806
4,832
198
10,149
226
690
6,085
61,503

1971
1


1
16
12
1
3

8


,478
-
996
,871
,646
,252
,357
,906
414
,192
2
668
4,943
52,725

^^B^HM^HMHHH|
1972
1,836
-
800
2,983
15,553
11,519
594
6,541
32
8,190
-
578
4,280
52,907

1973
1,830
-
673
2,679
18,000
13,070
658
7,155
32
7,571
-
606
2,009
54,283

    Source:  Ref.  14.

-------
                 Low
             High         T Most Probable
Agriculture
Catalysts
for Plastics
Dental
Preparations
Electrical
Apparatus
Chlor-Alkali
Instruments
Paints
Pharmaceuticals
Other (including
laboratory use)

;.


T
W///////A 1

X


W/////////////////////////////////^^^^^^^




J.







T
i -i
w//////////////////////////////////////////^^^^^^ \
\

1
W/////////////////////////////M \

T




W////M \
i
J.
T-

m//////////////////////////m \


J.



r


           2.5
5.0
                                          7.5       10.0       12.5
                                             Thousands of Flasks
SOURCES: High and low forecasts: Ref. 14; "most probable" forecast: URS Research Company.
15.0
17.5
20.0
Figure 2  PROJECTED MERCURY CONSUMPTION IN THE UNITED STATES
         BY END USE (34.4-Kg Flasks)
                                    14

-------
MAJOR END USES

     The following summary discussion of the outlook for the principal use
categories for mercury elaborates on the forecasts presented in Figure 2.

Agriculture

     The principal agricultural use of mercury is for seed dressing;  lesser
amounts are used for disease control of certain crops.   No effective  and harm-
less substitutes for these uses have been found to date, but continuing
research should result in the development of equally effective but less  poten-
tially harmful substances over the next decade.  The EE& has already  banned
the use of alkylmercury compounds, and prohibitions  on the use of all other
mercury biocides are anticipated.  By 1985,  agricultural consumption  may have
declined to some 800 flasks a year, for some limited applications that will
undoubtedly be under strict governmental supervision.

Catalysts

     Catalysts containing mercury have been used in the production of vinyl
chloride monomers, urethane foams, and anthraguinone derivatives.  The produc-
tion of urethane foams is expected to continue to increase in the years ahead,
and no wholly satisfactory alternative catalyst has been found,  although for
some purposes the organotins have been substituted.  Vinyl chloride monomer
production has shifted to increasing  use of  ethylene rather than  acetylene.
It was the acetylene process which used mercury-based  catalysts.  The URS
forecast anticipates this decrease in use for vinyl chloride  monomers,  with
some offsetting increases for  the urethane  foams.

Dental Preparations

     Population growth and  rising affluence, coupled with the increasing number
of health care plans which  cover dental costs, account for much of the projected
increase in mercury use  for dental preparations.  Newer technologies in dental
                                       15

-------
  restorations could reverse this trend; at present,  however,  it does not seem
       t
  that replacements for mercury amalgams will be found during  the forecast
  period.  Furthermore, the use of mercury in dental preparations has been shown
  to be essentially harmless to humans,  and its effectiveness  is well known.
  Thus, the total probable demand forecast by URS for 1985  is  the same as the
  high figure shown in Figure 2.

  Electrical Apparatus

       The projected high demand for mercury in electrical  applications reflects
  population growth and continued demand for mercury  batteries and lamps.  Note,
  however,  that it is not significantly  greater than  the 1973  consumption figure
  shown in Table 1.   Declines in  military battery requirements have been more than
  offset by increased nonmilitary usage.   The low forecast  in  Figure 2 assumes
  significant replacement of  the  battery uses by other systems and further
  replacement of mercury rectifiers by solid-state devices.  The URS demand
  forecast  reflects  stabilization of  recent  demand trends and  the  probable lack
  of large-scale,  radical departures  from current manufacturing materials and
  processes over the next decade.

 Caustic-Chlorine (Chlor-Alkali)

      The Chlorine Institute has forecast a 6-percent annual increase in chlo-
 rine production through 1980.  In Figure 2, the high production of 19,000
 flasks of mercury consumption by 1985 is based on:   (1)  an extrapolation of
 projected chlorine production through 1985; (2) production of about 24 percent
 of total chlorine through mercury cells; and (3) a  continued makeup rate of
 0.37  pound (Ib) of mercury per ton of chlorine produced.   The low figure of
 9,000 flasks assumes that:   (1)  the 1985 output of  chlorine by the mercury
 process will remain at the current level of 2.4 million tons; and (2)  mercury
 consumption in this process  will decline to 0.10 Ib per ton of chlorine
produced.
      It  is quite likely that the makeup rate will drop to  around 0.30  Ib of
mercury per ton of  chlorine  produced by 1985,  due to implementation of EEA

                                      16

-------
standards.   In addition,  several mercury cell production units have  closed
in recent years and the mercury cell plant share of chlorine production has
declined 15 percent between 1968 and 1972.  On  the other hand, the 13-percent
jump from 1972 to 1973 suggests that the 9,000-flask  figure is probably too
low.  The 1985 projection of 14,000 flasks derived by URS  is believed to be
more realistic.  (Note, however, that  none of the new plants that will be
producing caustic-chlorine in the near future uses the mercury cell.)
Paints
     Increasing use of the less toxic organic fungicides and preservatives is
expected to restrain the use of mercury compounds in the manufacture of paints
over the next decade.  The high demand figure (13,000 flasks)  in Figure 2 is
based on increased, rather than decreased,  consumption of these compounds, in
line with the forecast 3.7-percent annual growth for paint.   The low figure
assumes a complete ban on mercury compounds in paints by Em,  while the URS
forecast assumes special uses under controlled conditions.
     Over the last decade, the development of emulsion paints has led to a
swift rise in the use of organomercurials for mildew prevention to extend the
shelf life and durability of the product.  This use will likely continue at
least for the near term—up to five years—since development of fully satis-
factory replacements will take nearly that long.  It should also be noted,
however, that this application generally has been declining in recent years.
In marine paints, the mercurials are antifoulants; this use of mercurials
declined, however, to a low of 32 flasks in  1972 and again in  1973,  in sharp
contrast to consumption during the preceding  five years.

Pharmaceuticals
     The high forecast of 700 flasks of mercury for pharmaceutical end uses
 in  1985  reflects the impact of expected population increases on present (1972-
 1973)  demand levels; the low figure assumes elimination of all mercurial
 preparations,  except for special uses where no substitutes are known.  It
 appears  to URS that a 1985 figure of 500 flasks for this end use is a reason-
 able estimate.
                                        17

-------
 Other Uses  (including Laboratories)

      The high and low demand levels shown in Figure 2 for 1985 were based on
 Gross National Product  (GNP) and population growth factors,  respectively.  The
 URS  forecast, which is  closer to the low figure, is based on the assumption
 that there will continue to be a steady demand for mercurials for general
 laboratory chemicals and pigments, and that new uses are also likely to arise.

 Consumption by Federal Agencies

      In 1971, in  a very comprehensive study (Ref. 82),  a survey was made of
 federal agency usage of mercury that indicated that the Department of Defense
 and  the Atomic Energy Commission were the largest users.  This study was
 sufficiently thorough that we were not able to improve upon it during the
 present work.

 REGIONAL DISTRIBUTION

      The two largest outlet categories for manufactured goods in the United
 States are merchant wholesalers and manufacturers'  sales branches and offices.
 The U.S.  Census of Business lists totals for product classes passing through
 these outlets,  and divides these totals into nine Census regions.   Table 2
 shows results of an attempt to determine regional sales of mercury-containing
 products  and to relate  these data to regional  population figures.   Some data
 are available from manufacturers'  associations in the East.   However,  it is
 not always possible to  obtain data that differentiate between mercury-containing
 and nonmercury-containing  products.   In such cases,  the "mercury portion"  of
 total production can be  estimated from the gross usage figures in Table 2 and
 from knowledge of the average mercury content  of one unit of product.
      In theory, the sum  of the  regional sales  should equal the U.S.  total for
 each  type of  outlet.  The  total of the  outlets plus  minor distribution systems
 should yield  an approximation of the  total domestic  market for any given
product class.  In fact, however,  relatively few sums "checked" closely or
exactly.  Fortunately, one  or the  other (merchant wholesalers or manufacturers'

                                       18

-------
TABLE 2. -
                                    Sales of Manufactured Products Containing Mercury
                                    (percentage sold, by Census region, compared to
                                           region's share of U.S. population)
U.S.
Census
Region
Hew England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Central
Mountain
Pacific
Totals3
Paints
3.7%
23.2
23.7
6.5
11.1
3.3
7.9
2.1
18.6
100.1%

Pharm.
4.4%
23.7
20.1
8.2
14.1
3.6
9.2
2.2
14.3
99.8%

Agric.
Chem.
0.3%
12.3
24.0
16.1
7.4
10.3
11.0
2.8
15.3
99.5%

Lamps
5.8%
17.8
19.4
8.0
15.5
6.3
9.7
4.4
13.1
100.0%

Switches
and
Rectifiers
6.2%
31.3
15.2
7.1
10.7
2.5
6.8
2.5
18.5
100.8%

Dental
Amalgam
6.4%
27.6
17.4
7.8
10.4
4.4
6.5
3.9
15.6
100.0%

Laboratory
and
Hospital
Supplies
5.7%
23.7
18.6
7.8
12.5
4.5
8.0
2.9
16.1
99.8%

U.S.
Population
(12/73)
5.8%
17.8
19.4
8.0
15.5
6.3
9.7
4.4
13.1
100.0%

a.  Totals may not add, due to rounding.
Source:  Derived from current Census data, USDA data, and information supplied by the following manu-
         facturers' associations:  National Paint and Coatings Association; American Hospital Associa-
         tion; American Dental Association; and Pharmaceutical Manufacturers Association.

-------
  sale? branches)  checked within reason for all product classes.  Only in pharma-
  ceuticals  did both column totals agree, yielding a consistent picture of the
  total market.
      Note  the significant differences between the consumption of agricultural
  chemicals  and the  proportion of national population in the Atlantic and Central
  regions, and  the disproportionately large consumption of mercury-containing
  products generally in the Middle Atlantic Region.  In spite of the statistical
  difficulties  in estimating the extent of mercury consumption in manufactured
  products,  the preliminary data in Table 2 are considered reasonable approxima-
  tions of the markets for such products.

 FINAL DISPOSAL

      Prior to the recognition in 1970 that mercury and its compounds had some
 unusual characteristics which  allowed it to enter man's food chain and
 adversely affect him,  mercurials  had received little  more attention than hal9
 the other heavy  metals.  Certain  industries had been  recognized as sources  of
 mercury in the environment,  and certain pollution controls were in effect.
 However,  because mercury was usually lost to the environment in smaller quan-
 tities  than other pollutants,  special  treatments aimed solely at mercury had
 not been developed. At  first,  the major  concern was  with mercury discharges
 to  water, because these  led most  directly into  the  food chain.   The develop-
 ment of  controls  on water emissions was also expedited because it was rela-
 tively easy to monitor for mercury in water  in  the parts per million (ppm)
 range.*
* Even at the parts per million level difficulties were experienced in obtain-
  ing consistent and reliable monitoring data (see Ref. 67 for the ROUND ROBIN
  evaluation made by the EPA).  However, the data obtained were considered
  sufficiently reliable to establish standards.   Further, water lends itself
  to continuous or intermittent sampling whose results, combined with known
  flow rates,  can be used to calculate total discharge.  Such a calculation
  cannot easily be made for either air or land emissions.
                                       20

-------
     Industrial emissions to air where exposure of workers was  a recognized
possibility had been under surveillance,  if not regulation,  for many years,
so that excessive discharges were at least discouraged (if not  prohibited).
However, with recognition of the threat and imposition of stricter control
on air emissions, the quantities emitted to air by regulated industries began
to decrease sharply in the early 1970s.  Even so,  a number of known mercury
emitters, such as coal-burning steam electric plants and copper smelters,  still
remain unregulated and continue to emit quantities of mercury to the air.
These industries are concerned with the control of other pollutants and have
as yet not been very concerned with concomitant emissions of mercury,  which on
a mass basis represents only a tiny fraction of the major pollutant,  sulfur
dioxide.  Thus, at the present time, the unregulated emitters probably repre-
sent the largest "industrial" source of mercury emissions to air,  and will
likely continue to do so in the future.  However,  the increasing concern  with
controlling the emission of major pollutants may well result in decreased
emissions of mercury.
     Industrial discharges of mercury-containing solid wastes were never  con-
sidered as a major problem, since the assumed insolubility and stability of
mercury compounds were well recognized.  However,  it is now recognized that
mercury can migrate through the soil to the groundwater or even be emitted
directly to the air.  This has led  to  increasing concern with isolating these
industrial landfills by more sophisticated methods,  such as the use of imper-
meable  liners  and other procedures  to  ensure  segregation from groundwaters.
It is anticipated that, as pressure continues to reduce the  loss  of mercury
to the environment in any form, industrial  discharges  to  land  will continue
to decrease—primarily through better  process control, which may include
internal recycling.
     Figure 3a is the URS estimate  of past and future trends with respect to
the discharges of mercurials from industrial and  unregulated sources to  air,
land, and water.  These  curves sketch out only trends and relative quantities
involved, but  they provide  some interesting insights.  For example, we have
reason  to believe that in 1963 most industrial discharges were directed  into
the watercourses.  This  trend was probably accelerated in the late 1960s by
                                        21

-------
 2
 CO
 to

 c •
.1
+j

.2
0)

DC
                                   Land
                                   Air
                                   Land
                                              (a) INDUSTRIAL AND

                                                 UNREGULATED SOURCES
                                              (b) MUNICIPAL SOURCES
                                   Water
  1963
1973
                                   Air
1983
                                              SOURCE: URS Research Company.
          Figure 3  PAST AND PROBABLE FUTURE TRENDS IN

                   FINAL DISPOSAL OF MERCURIALS
                                   22

-------
the sharp increase in the construction of new chlor-alkali plants using mer-
cury cells.  After the imposition of strict controls in early 1970,  discharges
to water dropped sharply, but at the same time the discharges to land (which then
became the recipient of much of the material previously discharged to water) rose
sharply.  Even today (in 1975) discharges to water are but a fraction of what
they were five years ago.  This trend will continue but will taper off by 1983,
because a near-irreducible minimum- will have been reached.  The forecast is that
industrial and unregulated discharges to land will level off and thereafter de-
crease sharply as better utilization of mercury in industrial processes is re-
quired.  Even so, we feel that land will continue to be the largest single
recipient of industrial discharges of mercury.
     Industrial discharges to air are believed to have increased somewhat in the
1963-1973 period, despite the imposition of regulatory standards on chlor-alkali
facilities and on mercury smelters.  This increase is primarily attributable to
the increased use of fossil fuels with higher mercury content  (for which no
regulatory action as yet has been proposed).  The increased use of fossil fuels
 (primarily coal) for electric power generation may increase emissions to air over
the next ten years, but regulatory standards will probably be imposed that will
lead to an ultimate decrease  in emissions.
     Municipal discharges of mercury to the environment had never been con-
sidered to be a significant problem.  After the 1970 alert, however, research
on mercury discharges  in municipal sewage was instigated and demonstrated that
appreciable quantities can be attributed to that  source.  It was recognized
that the incoming water  source often had a  detectable mercury  content, which
might  represent 10 to  20 percent  of the mercury content of  the untreated sewage.
     The emission of mercury in  sewage was  still  not considered a serious con-
cern.   In  general,  it  was thought that discharges through municipal sewage were
probably small  in comparison to  industrial  discharges to  water and that the dis-
charges from  the municipal  sources were in any event scattered uniformly through-
out populated areas (in contrast to industrial discharges,  which were not).
Further, the  removal of the low levels of mercury found in sewage presented a
formidable—and indeed impossible—task.
     Problems with mercury in sewage sludge, which accumulates about half  of
 the mercury in the influent (raw) sewage, have stirred the most recent  concern.
If this sludge  is used in landfill operations, no serious problems are  likely
 to be  created.   However, if it is incinerated (as is done  in  a number of
                                        23

-------
  municipalities)  or used as a soil  amendment, the mercury can then be trans-
  ferred to the air or to plants.  The  forecast is that the incineration
  of sludges will  decrease in the  future, because of the now known release of
  mercury as well  as other toxic materials  into the atmosphere, and that the
  total quantities released via this route  will be lowered.  The use of soil
  amendments is also likely to come  under stricter control.
       By far the  largest loss to  the environment from municipal sources is
  that to land, normally in landfill operations.  Most mercury used in consumer
  products such as batteries or lamps is finally relegated as solid waste to
  landfills.   The  quantities involved in this case probably exceed those from
  industrial sources (although,  of course,  the disposal sites for the two types
  of wastes are segregated).   There  has been little concern with the accumulation
  of mercury-containing  products in  landfills, since there is little or no
  evidence to indicate that mercury  escapes from properly designed landfills of
  types that are coming  into  common  use.  Hence the forecast is that municipal
  landfills will continue  to  receive and treat mercury-containing products in
  the  future  much  as they  have done  in the past.  However, the total quantity
  of mercury  disposal to landfills will decrease, for two reasons.  One reason
  is the probable  decrease in the quantity of mercury being incorporated in con-
  sumer products because of the  forced recognition by industry of mercury's
 potential for environmental harm.  Second, as recycling becomes more popular
 consumer  salvage of mercury-containing products, particularly batteries, is more
 likely.  Nevertheless, the total amount of mercury in landfills is now large
 and will continue to grow with the years.
      Mercury enters the air as a result of incineration of solid wastes or of
 sludges.  Incineration and other thermal conversion processes such as pyrolysis
 are not general practices today  (although incineration is common along the
 Eastern seaboard), but they are likely to become more popular as energy sources
 in the future. Even today, incinerator stack gases have been identified as con-
 taining excessive quantities of mercury.   Because they are currently unregulated
 sources of mercury, incinerators  have  no pollutant control devices specifically
 to  remove mercury,  although they  do have pollution control devices for such
primary air pollutants  as particulates,  and  these  do act to reduce mercury emis-
sions.  Hence  it  is forecast that in the near term discharges to air from
incineration of solid wastes and  sludges will continue to rise,  but that with
the recognition of  the  threat this  entails,  the mercury-containing products

                                      24

-------
will be segregated so that in the future such discharges from municipal sources
will decrease.
     Our evaluation of past and future trends with regard to municipal sources
is summarized in Figure 3b.  Again, the quantities indicated are purely for
illustrative purposes and, aside from those for land,  are considerably less than
the quantities related to industrial and unregulated sources.

ULTIMATE FATE OF MERCURY IN THE ENVIRONMENT

     The fate of mercury in the environment is determined by its origin,
the medium into which it is lost, the conditions imposed on that medium,
and transformations occurring within that medium (Ref. 15).  For instance,
airborne mercury lost from a chlor-alkali plant disperses rapidly in the
environment, due to diffusion and dispersive action of wind  (Ref. 16).
Some mercury   concentrates in nearby soil and water sediments, but because most
of it is in a volatile form, the largest percentage of the emitted mercury
becomes part of the regional or global atmospheric burden, with little effect
on ambient regional concentrations.  Emissions to the atmosphere from other
large sources  such as smelters, incinerators, and coal-fired power plants would
follow similar patterns.   In these cases, there would be relatively large
increases in atmospheric,  soil, and water concentrations in the local area
(Ref. 17).  The cumulative effects of many such sources measurably increase
regional concentrations and may indetectably increase global concentrations,
as shown by recent global modeling efforts  (Refs.  1,  15).  With regard to
natural atmospheric sources, high ambient air concentrations have been found
near mineralized and  geothermal areas (Ref.  18).   The fate of this  mercury is
similar to that from  point sources, but there is  less effect on ambient con-
centrations because of the generally  dispersed nature of natural sources.
Atmospheric mercury is returned to the terrestrial environment through pre-
cipitation and dry deposition;  the ocean and its  deep sediments constitute
the ultimate  sink  for some of this mercury (Ref.  15).
     Mercury  emitted directly into rivers is usually adsorbed onto suspended
particles in  the rivers  and consequently is found in the sediments  (Refs.  19,
20).  However,  such deposition is only a temporary fate, as  biological and

                                        25

-------
  chemical activity leads to volatilization that sometimes reduces the
  sediment concentrations to background (equilibrium)  levels (Ref. 21).
  Also, deposited sediment is continually resuspended  and is carried
  downriver by periodic floods, eventually reaching the ocean.  Mercury
  entering estuaries and bays generally remains in such waters  longer
  than in more rapidly moving streams.  Lakes experience even longer
  residence times for mercury and tend to accumulate higher concentrations
  in both the sediment and the water column.   The absence of turbulence
  in lakes also slows the exchange of mercury with the atmosphere  (Ref. 22).
       Man and nature contribute significant  quantities of mercury to
  the land environment.   Directly, mercury-containing  industrial and munic-
  ipal solid wastes  are  deposited in landfills.   Indirectly,  emissions
  from atmospheric sources add to the burden  of mercury in the  land; nat-
  ural degassing also contributes indirectly.
       Generally,  mercury losses from solid waste deposited in  landfills
  can be well controlled through proper disposal site  selection, design,
  construction,  operation,  and maintenance.   Improper  disposal practices
  can contribute mercury to water (primarily groundwater)  and air  environ-
  ments  (Refs. 17, 18).
      Figure 4  shows the various major environmental "reservoirs"  and the trans-
 fer paths of mercury among these reservoirs.  The critical role of biota in
 the mercury cycle is not indicated, but should be emphasized with respect to
 its effect on transfer rates and mechanisms and with respect to the potential
 health problems related to mercury contamination.  In terms of rates and
 mechanisms,  microbial communities catalyze the formation of the mobile  alkyl-
 mercurials and volatile methylmercury.   The production of methylmercury leads
 to  the public health concern, because members of the  food chain (especially
 fish)  have the ability  to concentrate methylmercury.
      Little  information has  been collected with regard to interreservoir
transfer  (except  on a global basis),  although much information has been col-
lected on  the concentrations of mercury  in the materials which make up  these
environments.   Figure 5 is a summary of measurements  for both  normal and

                                      26

-------
                             Man-Related
                               Sources
             Land
             Surface
          J2I  8
          LL i LU
           J	
i
                              Freshwater
            Freshwater
            Sediments
I
                  Continental
                    Rocks
                                                1
                   Air
                                                 J
                 Oceans
     Oceanic
    Sediments
      SOURCE: Ref. 19.
Figure 4  MERCURY CYCLING THROUGH ENVIRONMENTAL "RESERVOIRS'

-------
to
oo
          100,000,000
           10,000,000
            1,000,000
              100,000
        03

        Q.
        C
        o
C
CD
u
C
o
o
        u
        ^
        
-------
contaminated environments.   Most of these data  were  collected from special
studies.
     National monitoring data on a regular basis are available only for
mercury in water,  as the National Soils Monitoring Program does not monitor
mercury on a regular basis  and the SAROAD (Storage and Retrieval of Aerometric
Data) network contains only a few measurements.  The water quality monitoring
system, STORET  (Storage and Retrieval of Water  Quality Control Information),
contains a limited number of mercury measurements for streams and lakes,
with the best data available for the East Coast.

LEGISLATION AND REGULATIONS APPLICABLE TO MERCURY

     This analysis of legislation and regulations involving mercury and mercury
compounds deals primarily with federal agencies.  Prior to 1970 there were no
government regulations specifically applicable  to mercury or mercury compounds
as a health hazard to man.   Four federal agencies had the authority in existing
legislation for such regulation, but had not used it.  The U.S. Food and Drug
Administration  (FDA) had the authority, under the Federal Food, Drug, and
Cosmetics Act of 1938, to control mercury in food products.  The U.S. Public
Health  Service  (PHS) had authority to control mercury in public drinking water
supplies; however, current  standards (adopted in 1962) do not consider mercury.
The U.S. Department of Agriculture  (USDA) had authority to control mercury in
pesticides, under the Federal Insecticide, Fungicide, and Rodenticide Act of
1947.   The Federal Water Quality Administration had authority under  the Clean
Water Act of 1966 to establish  standards for interstate waters; however, only
the standards of Illinois have  a limit  on mercury—0.5 part  per billion (ppb).

Mercury Crisis;  The Early  Response

     In 1969, 1970, and 1971 a  series  of unrelated  incidents (see listing
below)  aroused  public  concern and prompted action by the U.S. Congress and
federal regulatory  authorities.  Congress responded to the "mercury crisis"
by incorporating more  specific  regulatory authority in a variety of new
                                       29

-------
  legislation.   A series of Congressional hearings in 1970 provided the first
  major public  forum for an airing of  the impacts of mercury on man.  Subsequent
  hearings on specific legislation also dealt in part with mercury pollution.
  The FDA's initial monitoring activities were directed toward mercury contamina-
  tion in fish,  and in 1970 the agency established a guideline of 0.5 ppm mercury
  in fish.   This guideline  was not a standard, and the only significant action by
  FDA was to recommend in 1971 that swordfish not be eaten.
           Year
           1969
           1970
                  Event
          1971
Alamagordo poisoning case
Tuna confiscated by FDA in fall
Mercury found in Lake St. Clair fish (note
sent to U.S. Secretary of State by Canada,
April 2)
FDA recommended that swordfish not be eaten
6,530 persons poisoned in Iraq by mercury-
coated seeds
First reported case of mercury poisoning
from fish consumption in United States
          Source:   Ref.  15

      The EPA was  formed in 1970,  and was given the responsibilities of the ,IHS,
 USDA,  and the Federal Water Quality Administration for control of mercury.  The
 EEA  responded to  the "mercury crisis" through several regulatory avenues (as
 outlined below).
      Existing legislation is directed toward controlling discharges of mercury
 to the environment as waste matter  or in pesticides.   Pending  toxic substances
 control  legislation  authorizes the  EPA to evaluate chemical  substances and to
regulate or prohibit the  use of those determined to be toxic.   Such substances
would include those  containing mercury.
                                       30

-------
Regulatory Guidelines and Actions

     Early EPA efforts to regulate the discharge of mercury are  outlined below,
followed by a. discussion of subsequent legislation that redefined or strengthened
EFA's regulatory authority in each of the environmental sectors  considered.

     The Air Environment

     The Clean Air Act of 1970 authorized EPA to establish national standards
for hazardous air pollutants.  The law did not specifically denote mercury as
such a pollutant.  However, on the basis of evidence that mercury is emitted
by coal-burning power plants, municipal incinerators,  and industrial plants,
in March 1971 the EPA designated mercury as a hazardous air pollutant (36  PR
5931).  Standards for mercury were proposed in December 1971 (36 PR 23239) and
were promulgated as regulations in April 1973 for mercury ore processing
facilities and chlor-alkali plants (see Table 3).

     The Land Environment

     Beginning in 1970,  EPA  took  a series of actions to cancel pesticides that
contained mercury, under authority obtained from the Department of Agriculture.
The Federal Environmental Pesticide  Control Act of 1972 strengthened EPA's
authority to regulate,  control,  and  cancel harmful pesticides,  including  those
 containing mercury.

     The Water Environment

     Under the Drinking Water Act of 1974,  EPA was given the authority to pro-
 pose and promulgate standards for mercury in drinking water.  The proposed
 standard is 0.002 ppm,  which is much more strict than the previous standard
 of 0.005 ppm.   The most widely known action taken by EPA in this early period
 involved application of an old law to a new situation.  In  1970, the agency
 conducted a nationwide survey to determine mercury levels in surface waters.
 Based on these results and on the Rivers and Harbors Act  of 1899,  EPA  proposed
                                        31

-------
                                           TABLE 3. - EPA Mercury Regulations
Federal
Register
38 PR 8820a
38 PR 35388
Date
4/06/73
12/27/73
Applicable to
Mercury ore processing facili-
ties and chlor-alkali plants
Paper and allied products; oil
Standard
2,300 gm/24-hour period
(1) Into streams, lakes,

or
  (proposed)

(39 PR 10603
 amendment)
38 PR 28610


39 PR 38064
10/15/73
10/25/74
              and gas extraction; industrial
              organic or inorganic chemical;
              alkalis and chlorine; ferrous
              metal production; nonferrous
              metal smelting and rerefining;
              lumber and wood products;
              bituminous coal and lignite
              mining; storage or primary
              battery manufacturing; or metal
              mining facility discharging in-
              to navigable water
Ocean dumping


Wastewater treatment plant
sludge incinerators
estuaries with flow less than 10
cfs or lakes less than 500 acres
—no discharges; (2) other streams
and lakes—20 yg/fc per discharge
or I/10th this concentration when
low flow is less than 10 times the
waste flow; (3) other estuaries
and all coastal waters— 100 yg/fc
per discharge or I/10th this
concentration where low flow is
less than 10 times the waste flow;
(4) stream—not to exceed 0.000162
times flow in cfs (or) 1.62 Ibs/day;
(5) lake—not to exceed 0.000135
times flow in cfs (or) 1.35 Ibs/day;
(6) estuary—not to exceed 0.00027
times flow in cfs (or) 2.70 Ibs/day;
(7) coastal water—not to exceed
0.000324 times  flow in cfs (or)
3.24 Ibs/day
No mercury except as trace con-
taminants

3,200 gm/24-hour period
a. 38 PR 8820 = Vol.  38, Federal Register, page 8820.
Note:  This  table does not include regulations dealing with pesticides.  There have been and continue to
       be many such regulations, all involving either cancellation or suspension of pesticide use; how-
       ever, limitations of time and space have precluded listing them here.

-------
standards for mercury discharge into surface waters, at  rates  dependent  upon
the type of receiving water body.   Since  then,  several states  have enacted
their own limitations on mercury discharges to  water from a given emitter  or
class of emitters.
     The Federal Water Pollution Control  Act amendments  of 1972 gave EPA the
basis for enforcing the regulation of mercury discharges into  the water  environ-
ment.  In September 1973, EPA designated  mercury as a toxic pollutant  (38  PR
24342), and in December 1973 the agency proposed toxic pollutant effluent  stan-
dards, which were amended in March 1974 (Table  3).  In October 1973, EPA
issued regulations prohibiting dumping any material containing mercury into
the ocean.  The agency has also designated mercury compounds  (-acetate,
-ammonium chloride, -bromide, -chloride,  -cyanide, -iodide, -nitrate,  -nitrate,
-oxide, -sulfate, and -thiocyanate) as hazardous substances, and any person  or
entity spilling such substances is liable for  cleanup and/or fine  (38  FR 30466,
August 1974).  In October 1974, standards for  mercury emissions from wastewater
treatment facilities were proposed.

Interstate Transportation

     One form of governmental regulation not mentioned previously concerns the
interstate transportation of hazardous substances.  Many mercury compounds are
designated as Group B poisons by the U.S. Department of Transportation,  and
their  shipment is  regulated in  accordance with applicable regulations.  Also,
because metallic mercury readily amalgamates with aluminum, the shipment  of
metallic mercury by  air is  restricted by the Civil Aeronautics Board.

Trends in Legislation and Regulatory Action

     Toxic  substances control legislation has  been introduced in Congress
 during the  last  three legislative sessions, but bills  introduced in 1971 and
 1973  failed to  survive  House and Senate  conferences.   In February 1975, a
revised bill (S.776)  was introduced in the  Senate.  The intent of this legis-
 lation is to provide more careful evaluation  of new chemicals prior to their
                                      33

-------
  commercial introduction and to provide adequate federal authority to deal with
  toxic substances now in the environment.   The EEA will have  the primary
  responsibility for implementing the legislation if it is passed.  The legisla-
  tion is being challenged,  primarily by the chemical manufacturing industry, as
  being too costly for the benefits to be gained,  and as restrictive of industry
  efforts to develop and market new products.   The EIA  has responded with much
  lower cost estimates and with arguments in support of the legislation as
  necessary for both human and environmental protection.
       The continuing efforts on the part of Congress and the EEA to enact
  legislation for the control of toxic substances  indicate that if the legisla-
  tion now before Congress is not enacted, other attempts will be made in the
  future.   In the long run the prognosis  for enactment  of such legislation is
  good;  however,  compromises  made to secure  its  passage  could reduce the effec-
  tiveness of the legislation from that now  being  contemplated.  If this did
  occur, the  legislation would doubtless be  tightened over the years,  as seems
  to be  the trend with most environmental legislation.
      The major  issue is  the  cost to be incurred by the chemical manufacturing
  industry to comply with  the  federal regulations.  The present economic environ-
 ment in  the United States, coupled with recent administrative positions taken
 on environmental legislation that cause industry to incur additional costs,
 indicate a very high probability that such legislation, if passed in the
 immediate future, would be vetoed.  If the economic situation improves,  the
 justification for vetoing such legislation would decline, of course.
      If no new legislation is enacted in the short term, the EEA can be expected
 to continue to use control mechanisms that now exist with respect to toxic
 substances.   The EE& has continued to regulate pesticides and has recently
 promulgated controls on other toxic substances through the use of provisions
 of the water and air acts.   While these are not source controls,  EEA actions
 taken after  introduction of substances into the environment (such as banning
 such  substances  in air and  water discharges)  will force consumers to consider
more  carefully the  use of substances with  toxics in them.
      In summary, the EEA  will continue to  use existing authority  to  control
toxic substances.  New legislation can be expected that will  more explicitly
                                       34

-------
define the EEA's authority in the control,  including source control,  of toxic
substances, but the probability that such legislation will be enacted in the
next two years is much lower than it is for the succeeding five years.
                                       35

-------
                                                                           Ill
                                                         INPUT/OUTPUT ANALYSIS

INTRODUCTION

     The usefulness of a national inventory of mercury sources and emissions
in deriving rough estimates of the quantity of mercury entering  the environment
from man-related sources has been apparent for some time.   Davis (Ref.  9),  in
a comprehensive effort for the year 1968,  used a simple model to estimate
mercury losses to air, land, and water from industrial,  consumer,  fuel-burning,
and sewage treatment activities; he introduced the concept of an emission factor
(the term "Davis emission factor" refers to values he established). Anderson
(Ref. 10) updated some of the industrial emission factors and provided consider-
able information on emission factors from combustion of fossil fuels.   Roy
(Ref. 11) provided additional inputs on the mercury discharges associated with
sewage and sludges.  An EPA internal document presented other inputs using the
same model, which were finalized in 1974 in a working paper, "Mercury Pollution
in the United States, by Source Category"  (Ref. 12), which served as the point
of departure for the  inventory reported here.
     These inventories have several points in common:
     o  Estimates are made  on a national basis  for  emissions  for a
        stated year.
     o  Bureau of Mines data on the number of  flasks  of mercury
        consumed in that year are  the basic  input data.
     o  Emission factors are applied to determine what fraction of
        the total available mercury is  lost.
In addition, some of  the inventories  include estimates for unregulated sources
 (primarily fossil  fuel burning,  sewage,  and sludges)  and natural sources.
     The URS inventory was expanded to incorporate an input/output (I/O)
analysis with  a materials balance to  ensure that all sources were included
and  yet that no double accounting occurred.   The basic goals in preparing
this inventory were:

                                        37

-------
      . o  To verify and update,  where required,  the latest available
          inventory (Ref. 12),  including all sources of  mercury to
          the environment on a national basis.
       o  To prepare a materials balance for selected regions which
          have been identified as having unusually  high  ambient
          mercury readings and/or an unusual number of known mercury
          emission sources.
       o  To determine from the  national and regional inventories
          where the major losses of mercury to the  environment have
          occurred and what the  nature  of these  losses was.  Under
          this item the ultimate fate of mercury in the  environment
          was to be considered to ascertain if final disposal
          (primarily to land) indeed represented the ultimate fate,
          or  if transfer to  other media  was a concern.

 METHODOLOGY

      The development of an input/output  (I/O)  model which would provide the
 information outlined above required the development of suitable methodology
 not only for the model but also for establishing a data base to support this
 model.  Because the two are so closely interrelated, the development of the
 methodology for one proceeded simultaneously with that for the other.
      First let us consider the state of the art of I/O models available for
 mercury at this time.  The several models described above were based almost
 entirely on data provided by the Bureau of Mines.   Our initial efforts also
 relied primarily upon the Bureau of Mines data. We first developed a flow
 chart (shown in Figure 6)  which shows  the sources  of mercury used  in the United
 States in 1973,  together with the industrial and commercial uses to which this
 mercury  was directed.   In 1973 the double impact  of lower  mercury prices on
 the world market and  restrictions on allowable  emissions from mercury mining
 and smelting operations in the  United  States had effectively reduced domestic
mercury production to very minor quantities (2,131 flasks,  or 3.9  percent of
total domestic usage).  The supply/demand gap created by this decreased
                                       38

-------
UI

U.S. 2.131
Production

'"JiP0"1 46,734
Mettl) in rhit
•————» "' ,, ,,„ 	 K
Secondary V4B 64J289
Metal

Gojjpnunent 2383
Releases

Inventory. 3,906
Exports

• Und primarily in thtlnonpnte form.








• Metal




42.482
Organic
10.430


7,397






•eaa
«e»


•mp
aiaMl
r


Industrial
Control*
SUM
7.155
Dental

2.679



668

Electrical


18.000
Chior-Alkali
13.07C
Other
900
' Paint
7.671
Agriculture
1,460
Pharma-
ceuticals
600



Catalyst!
500

aeiaM
Paint




.Pharma-
ceuticals


[

aHM
Switches,
; Relays


Control
Instruments

6,795
SM
.



uw L

Tubes

500
Batteries*


» —
r— Meters
• I Pumps
I— .Misc.
_ <
H

32

r1
L.
—

Turf
Treatment
Seed Treat-
ment, etc.
Urathane
Elastomers,
etc.
1.016
435
100



v-inylcnlorUa
Synthesis

—
Misc.

• Fluorescent

1,170
All Other
HgTypet

70
Ladanche
(Zinc)

4,550
Alkallne-Mn

8.780
Ruben and
Other
Hg Types

Vat Dyes
600 60
19




Agriculture

Other
409
380
r— Preservativet
1 • I 	 Adheilves
L— Misc.
106




Catalytts
179 700


Other
r— Pigments
-4— Stabilizers
L.MISC.
                    Figure 6  SOURCES OF MERCURY AND INDUSTRIAL AND COMMERCIAL USAGE
                            IN THE UNITED STATES, 1973 (All values in number of 76-pound flasks)

-------
 domestic production was primarily filled by imports from other countries.
 Secondary metal production  (from recycling within the nation) was little
 changed from previous years, accounting for 14.3 percent of consumption.
 Releases from government stockpiles accounted for another 4.8 percent of
 consumption.  (Some 3,905 flasks were diverted from supply into inventory
 buildup, export, or unaccounted losses.)
      Figure 6 also indicates the several categories of usage to which mercury
 is assigned and the relative quantities involved.   It does not indicate the
 losses to the environment from these categories (these losses will be discussed
 later in the report).   It should be recognized that this flow chart,  while
 useful,  has limitations and inconsistencies which  have been resolved to the
 degree possible.   For example,  the Bureau of Mines data reports only usage of
 the electrical applications. From a variety of references,  none of which agreed
 completely with  any other,  we were able to break this usage down further,  into
 lamps, power tubes,  and batteries.   Still further  division of lamp and battery
 usage was  possible.  However,  in many applications such detailed breakdowns are
 at best  tenuous.   In particular,  the "other"  category,  which represents a
 rather heavy usage  of mercury,  is so broad that  further tracing is impossible.
      Figure  6 indicates that less than 20 percent  of  the mercury used is in
 the organic  form.  The vast  majority is used  in  metallic form,  and the remainder
 is in the inorganic form.*   We needed  a comprehensive model that would allow
 us to track mercury from its introduction  (input)  into  the  economy, through
 its use at various stages within the economy  (including losses  during transfer
or by accident),  and thence  to final disposal of discarded  products.   In addi-
tion,  we had to recognize that a number of unregulated .sources  exist  that
  In the flow chart the battery industry is shown as using only virgin mercury
  metal,  which is not entirely correct.  Battery makers purchase only the
  virgin metal and convert it,  in their own facilities, to mercury oxide and
  other inorganic salts for incorporation into their products (with a minor
  quantity for sale to other users).   Thus when taking into account the
  inorganic mercury usage of the battery industry,  the form of mercury dis-
  charged to  the  environment is more  likely:  metallic 50 percent,  organic
  19 percent;  inorganic 31 percent.
                                       40

-------
contribute to mercury losses to  the  environment.  Finally, we knew that natural
sources also contribute mercury  to the  environment, although the extent of
their contribution was not clear.  And  in our  study, all of these losses to
the environment had to be traceable  not only on a national basis, but also on
a regional basis.
     The basic framework of the  methodology which we adopted to develop both
a suitable input/output (I/O) model  and the necessary  inputs to support it is
sketched below:
DATA
MANAGEMENT
SYSTEM

FORTRAN
PROGRAM

COM!
I/
<
TYPE OF I
BELIEVED
BE AVAIL;
"REHENSIVE .
'O MODEL
L
&TA
TO |
LBLE

k STMPT.TT?
I/O MOD
ti
1 1
14
DEVELOPMENT
„ OF DATA
COLLECTION
METHODOLOGY

IED
EL


^-



INVENTORY
MODEL
(COMPOTER)

COLLECTION AND
COLLATION OF
DATA BASE



Bureau of Mines
Open literature
Proprietary sources
Consultants, etc.

^ INVENTORY
OUTPUT

      As a first step,  a systems analyst,  working in conjunction with the data
 collection team, outlined a comprehensive I/O model,  which was "pretested" to
 determine its applicability.  It was discovered that the model was too cumber-
 some for adaptation to computer usage, so a simplified I/O model was derived.
 At the same time, the development of the simplified model solidified the  data
 requirements, which led to the development of the data collection methodology.
 Again interaction was maintained between the systems analyst  developing the
 simplified I/O model and the members of the data collection team (as shown by
                                        41

-------
  the dashed lines in the sketch),  to ensure compatibility of these two main-
  streams.   The simplified model was then adapted for use  on a PDP-10 computer.
  Meanwhile,  the data base for the  first study region (California) had been
  assembled.   These data were subsequently analyzed using  the inventory model
  and the desired output was obtained.   (As with any  such  model, debugging was
  necessary,  and as with any such large data base, newer information was fre-
  quently encountered and had to be incorporated.)  Of course, some modifica-
  tions were  made in both the data  collection methodology  and the inventory model
  during the  course of the study.   In the following sections, however, we report
  only the  final versions.

  Development of an Input/Output Model

      A  conceptual model  (see Figure 7)  was  designed to accommodate all possible
  inputs  and  outputs and to track man-related losses of mercury from both regu-
  lated and nonregulated sources in the environment.   The model is comprehensive
 and, with adequate data, it is possible to  trace all inputs,  regardless of
 source, through manufacturing, processing,  trade, final consumption, and thence
 to final disposition.  The model recognizes the possibility of mercury loss
 not only to air, water, and land,  but also to export (although this is .not a
 true loss,  of course).  In addition, the model considers the possibility of
 loss by accident  (the probability function) and in transit (the transfer
 function).
      This model, although useful for conceptual purposes, required a degree
 of knowledge concerning regional inputs and outputs  which proved unavailable
 (tracing the movement of a very small  volume product such as  the mercurials
 through interstate transfer was among  the most difficult problems encountered).
 The program  was therefore modified and simplified so that the materials  flow
 from  one sector to the other was eliminated but losses  from each sector  were
 maintained,  and these were then accumulated.  The final sectors in this  simpli-
 fied model are  shown  in Table 4.   (A complete list of all sources under  each
 section  is presented  in the  discussion of the data base.)  In addition to the
six sectors  shown  in Table 4 which are an integral part of the  I/O model,  two
                                      42

-------
                                    Input
   I  L
                                              II  L
Ill
4H
•*-
Losses
L,A,W,T
IV L
Losses
L,A,W,E
V L
Losses
L.A.W.E
VI L
Losses
E,P
VII L
Losses
E.P
VIII - L
Losses
L,A,W
*—


Basic Primary
Sources
1
IV


^








1
Secondary
Sources
|1

Manufacturing
Industries
V


Processing
Industries
/I


Wholesale
Stockpile
VII


Final Product
(Retail Stock)
VIII

1

Final
Consumption





«*-





lj



—
Losses Unregulated
L, A, W, T Sources

Recycle
1




t
L




Imports


















III . , L
Losses
* L,A,W
«-T


Losses (Legend)
L = Land
A = Air
SOURCE: URS Research Company.
                                                             W = Water
                                                             E = Export
                                                             P = Probability
                                                             T = Transfer
            Figure 7   BASIC MERCURY INVENTORY FLOW CHART
                                    43

-------
                    TABLE 4. - Input/Output Model Sectors
 Sector
             Name
           Example
 I

 II

 III
VIIIA


VIIIB
Mercury input

Mining

Misc. unregulated sources
Manufacturing and process-
ing

Commercially oriented final
consumption

Personally oriented final
consumption

Final disposal
           Natural sources
Mercury mining and refining

Copper mining and smelting

Livestock manure; fossil-
fuel-burning power plants

Battery manufacturing;
catalysts

Fluorescent tubes; control
instruments

Dental applications; batter-
ies

Sewage treatment; incinera-
tion

Degassing; runoff
Source:  URS Research Company
                                      44

-------
other sectors—final disposal and natural sources—are listed.   These sectors
are listed separately  because they represent neither a true input nor a true
output.  Final disposal,  which includes sewage treatment,  incineration,  and
landfill operations, encompasses the treatment and disposal of  those mercury
losses previously reported in the numbered sectors;  its inclusion in the
overall balance as a separate category could lead to double accounting.*
     The massive quantity of data expected to be derived for each study region
dictated that use of the I/O model, even in its simplified form,  would require
the speed and accuracy inherent in computer operations.  To this end URS
utilized a data management system leased from TYMSHAKE, INC. (an international
computer timesharing system) to store and analyze the mercury data collected
during the study.  This system, used on the POP-10 computer, is referred to as
System 1022; it is compatible with the conceptual model described above.  System
1022 is a sophisticated, general-purpose, data management, software system.
It allows the user to create, update, and maintain large data bases, features
a very fast retrieval capability, and allows interface with FORTRAN source
programs.  The 1022 system categorizes the input data into a form that can be
retrieved and updated quickly and efficiently.  Special command files can be
generated to make the system usable by individuals who have little or no com-
puter experience.
     Since a basic  requirement  in  the study was to determine the areal  distri-
bution discharges of mercury to the  environment, an  initial problem was to
determine the  smallest geographical  unit to be considered.  The best  definition
would, of course, result from  using  the  smallest unit,  say a Census tract;
however, the resulting number  of units for even a small region would be astro-
nomical.  Furthermore,  few  data are  available even  at the level of the
city, the Standard  Metropolitan Statistical  Area  (SMSA),  or the county (or
parish).  Some data, particularly  commercial and industrial data,  are available
  Both sewage and solid wastes do include some mercury in the form of a
  natural background; however,  these are small in comparison to man-related
  losses and do not influence the findings of the study.
                                      45

-------
  only at the state or regional level.   A compromise was reached in using the
  county as the base for which all data would be accumulated.   (As will be dis-
  cussed later,  for specific facilities location by latitude and longitude is
  possible for special analyses.)
       Just as the county served as the basic unit  of area, the Standard Indus-
  trial Classification (SIC)  served as  the basic unit of source.  Each known or
  suspected emitter was assigned to an  SIC which generally related to those found
  in the Standard Industrial Classification Manual  of 1972.  However, we modified
  these numbers  (which are generally four digits) to meet our own requirements
  where necessary.   For example,  the manual classes SIC  2851 as paints, varnishes,
  lacquers,  enamels,  and allied products and  includes establishments primarily
  engaged in their  manufacture.  However,  we  used 2851M  to indicate the manufac-
  ture  of the phenylmercuric  acetate used by  the paint industry, 285IF to indi-
  cate  the formulation of the paint itself, and  285IP to indicate final consump-
  tion  of the paint.
       Each  SIC was then  assigned to one  of the  sectors  shown in Table 4, in keep-
  ing with the I/O  model.  A  list of the  inputs  into  the  I/O data management
  system  is given in Table 5.  Those  concerned with location or classification
 have either been  discussed  or are  self-evident.   However,  those concerned with
 "hard" data on the quantities of mercury and emissions  to the environment
 require further clarification.  The "inventory" input accounts for mercury
 which is either stocked by  a user  or is part of a nonconsuming process (such
 as chlor-alkali plant).  (In practice, data for this listing proved to be
 difficult to derive and were generally not included.)  The total quantity of
 mercury which is "exposed" for loss to the environment, as estimated by the
 data collection team, was input as "PROD."  The inventory program then used a
 simple function to determine losses to the environment, namely:

                       PROD X emission  factor =  loss.*
* Initially, additional  entries  had been allowed for total loss for a given
  facility.  However, the  few cases where such usage was necessary did not
  warrant the retention  of this  option;  the  functions listed above were
  easily substituted in  its place.
                                        46

-------
                TABLE  5.  -  Inputs into the  Input/Output
                           Inventory Program, (for a given
                           program)
Input
          Description
Purpose
Sector


SIC


Name



County


State

Latitude/
Longitude


Inventory
Total
Quantities
Produced or
Processed
(PROD)

Form of
product or
process

Emission
Factors  (EF)
  Air
  Water
  Land
  Recycle
Generic description of sources
Standard Industrial Classifica-
tion

Used for manufacturer's facility
County (in some cases groups of
counties) or parish
In Mercator coordinates
Estimated quantity of mercury
kept but not consumed

An estimate of the total mer-
cury consumed, in product or
process, annually
 Designates major form of emis-
 sion — Hg,  inorganic or
 organic

 An EF represents the fraction
 lost to each output.  (The sum
 of all EF's  cannot exceed  1.0,
 but may be  less, due to in-
 creased inventory  or permanent
 retention of the mercury in
 PROD.)
To classify similar sources
together

To standardize, in common
usage, each individual source

To differentiate between sev-
eral SIC's at the same loca-
tion

To identify by geographical
location
To permit location, within
±4 km, where desired

To distinguish inventory
from losses

The  base value to which
emission factors  are
applied
 To allow a determination of
 the form in which mercurials
 are lost to the environment

 Used in program to determine
 losses to the environment.
 For examole:  PROD x EFA =
 loss to air; PROD x EFW =
 loss to water
 Source:  URS Research Company
                                        47

-------
       Emission factors are a widely used concept to represent  the  fraction of
  the total available mercury (that is,  EROD)  which is lost  to  a given environ-
  ment.  (Note that "recycle" has been treated as an emission since this simpli-
  fied the computer program.)  The sum of the  emission factors  is 1.0.  However,
  in those instances where mercury can be shown to be permanently retained (for
  example,  75 percent of the  mercury used in dental treatment is permanently
  retained in the teeth),  the true amount lost to the environment is accounted
  for by an "availability factor," in this case 25 percent.
       Six computer programs  were developed, in the FORTRAN  IV programming
  language,  which are used in conjunction with the 1022 system  to accept the
  indicated inputs and provide the desired outputs.   These programs are de-
  scribed briefly in Table 6.  The bulk  of the information which will be reported
  on here was derived from ALLSIC,  which lists,  for  a given  region, all losses
  (by SIC)  to the environment; CTYALL, which provides county-by-county listing
  of total mercury losses  to  the environment;  and URSLL, which provides,  within
  any given  coordinate  points, all mercury losses to the environment.  Changes
  in the  emission factors  can be made by simply  accessing System 1022 as out-
  lined in Appendix C.

 Acquisition of the Data

      It was recognized that it would be a monumental task to acquire data on
 the utilization of mercury and its compounds on any but a gross (that is,
 national) basis.  The first obstacle is the lack of substantive statistics
 on the use of mercury and its compounds; this lack is attributable to the very
 low volume of usage.  (The Bureau of Mines is undoubtedly the best source of
 information,  although the Tariff Commission provides some limited data on
 organic  mercurials.)  A second problem arises from the "scare" philosophy
 which has grown out of the highly publicized  incidents in the early 1970s,
 and which has resulted in the reluctance of many users to discuss their
present  or  projected utilization of mercurials.  Many manufacturers are
believed to underestimate their  usage of mercury.  In order to overcome these
two major obstacles  as far as possible, URS assembled a  data collection team
which included chemical engineers,  a chemical economist,  a  regional economist,
                                       48

-------
              TABLE 6.  -  Outputs  from Various  Programs  in the
                         Input/Output Model  (for a given region)
Program
Name
Output
Purpose
ALLS1C
CTYALL
URSLL
URSIC
URSCTY
MULFAC
Provides an enumeration of mer-
cury losses, by SIC, to air,
land, water, and recycling.
Provides an enumeration of mer-
cury losses, by county (or
state), to air, water, land,
and recycling.

Provides an enumeration by lat-
itude and longitude,"of all mer-
cury losses within an area desig-
nated by four coordinate points.
 Provides  a correlation, for a
 given SIC, between the number
 of emitters and their size.
 Five size ranges  are considered
 from 400 Ibs/yr.

 Provides  a correlation, for a
 given county, between the num-
 ber and size of emitters, by
 sector.  Five size ranges are
 considered from <5 Ib/yr  to
 >400 Ib/yr.

 Provides  the capability to
 change total mercury usage
 (PROD) for a given SIC.
Permits identification of
major sources of loss, by
SIC, to each environmental
recipient (air, water, or
land).

Permits identification of
major contributors, by county
(or state), to each environ-
mental recipient

Permits study of losses to
each environmental recipient
in geographical locations of
special interest (such as
a river basin).

Permits an  assessment of the
importance  of  size of emitter
as  compared to the total num-
ber of such emitters, by SIC.
 Permits assessment of rela-
 tive importance of contribu-
 tors to environmental losses
 from different types of
 counties
 Allows rapid revisions of
 input data
 Source:  URS Research Company
                                       49

-------
  an industrial engineer,  and an environmental engineer.  Members of this team
  had many years of marketing and forecasting experience.  The team utilized
  published data,  the open literature,  contacts with trade associations and
  individual industries,  and other proprietary sources in acquiring the necessary
  body of information.
       In many cases,  excellent sources of  data were found, but in others it was
  hard to locate and verify data.   Thus it  sometimes became necessary to make
  first estimates  for a given SIC  and later to revise these as better data
  became available.   Certainly the quality  of the data on the known or suspected
  uses of mercury  is variable,  but because  of the rigorous materials balance
  procedure  followed,  we estimate  that  most of  the values used are within plus
  or minus 30 percent  of actual, and  in many cases are much better.*
       Inputs on the quantity  of mercury entering the economy in 1973 were of
  course based on the Bureau of Mines data, which are the best available.**  .Tri
  determining the sources of mercury from Sectors I, II,  III,  and V for specific
  facilities such as mines, power plants, and manufacturing plants,  a number of
 industrial sources provided much of the necessary information on location and
 production.  These varied sources included the Bureau of Mines Mineral Year-
 book, SRI's Chemical Economics Handbook,  the Directory of Chemical Producers,
 and the magazines Chemical Week and Chemical Marketing.   Where conflicting
 information was encountered  (which happened frequently)  the  best judgment of
 the  team was exercised.   Where the number of facilities was  excessive (greater
 than 50), as for example paint manufacturers. Census of Manufacturers'  data
 * Some  sources  of  information initially considered to be most useful were
   soon  found to be of  little or no value.   For example,  permit application
   lists of potential mercury emitters  (based on RAPP data)  by EPA region
   were  found to contain spurious data,  or  data which could not be validated.
   We suspect that  much of  the data was  recorded in the wrong units.
** It is true the Bureau of Mines data have some shortcomings in that they
   are derived from proprietary reports  submitted by industry.   Thus  there
   may be small  errors  in reporting use  category.   However,  on the input
   end,  only secondary  mercury is subject to such error,  so overall,  the
   input data are considered  exceptionally  reliable.
                                       50

-------
were used on a state-by-state basis.   In all cases the production of a desig-
nated facility was used as a basis for estimating mercury uses and losses;  all
values were reported in terms of mercury metal.   As an example,  a materials
balance for coal usage in the United States is shown in Figure 8.  It indicates
total usage (input) of 73,923 kg of mercury, with detailed losses (output)  to
the environment thereafter.  This chart was derived from a large variety of
sources as well as from Bureau of Mines data (Ref. 23) and assumes an average
mercury content of 0.16 ppm.  Similarly, a materials balance for any fossil
fuel for any state or group of states can be made based upon these same data,
with adjustments for the 1973 base.
     Two categories have been identified which are amenable to either a case-
by-case study or to the use of regional statistics.  Discharges in the first
category are termed "point source emissions" and include mines,  smelters,
chlor-alkali plants, other chemical plants that manufacture mercurials, battery
manufacturers, etc.  For the second category, a number of "area source emis-
sions" were identified in the final consumption sector.  For example, dental
usage applies to the general population, industrial instruments can be applied
to a very wide variety of manufacturing industries, and so on.  Since it was
not possible to trace down these very widespread  uses, an equation which
derives from national consumption  figures was developed which allows us to
estimate regional  consumption, as  follows:

       National consumption X —*•:	  r°r   .— = regional consumption
                              national  population     *            *
     Population  serves  adequately as  a.  factor for general consumer products
such as dental preparations,  drugs, batteries, and heating fuel.  However,  it
does not adequately describe other types of consumption patterns which are
related to  commercial and industrial activity.   Therefore,  two other types of
factors—a  multifactor,  and a manufacturing employment ratio—were developed.
     The multifactor, which is  based on population and on commercial and
industrial activity,  is used to determine a region's usage of mercury in
commercial products (fluorescent lamps, nonfarm pesticides, laboratory items,
paints,  "other").   The manufacturing employment ratio is based  on manufacturing
activity (which can be either employment or manufacturing value added).   This
                                        51

-------
                             7,349
974

Export Losses
All U.S. Coal |
(1973)



Household and
Commercial Losses
45,249
66,574 46,226 | 45,249 ^ Electric Power Plant
i Losses
20,351
Industrial Losses
I
f 896


Food and
Products
*
Kindred
Losses

J 1,384
Stone, Clay,
and Concrete
Product Losses
1,587 \ 2,341
Paper and Allied
Product Losses
*

Chemical and Allied
Product Losses
10,766 | 3,378
Primary Metals
Losses

Other Mineral and
Manufacturing Losses

SOURCE: URS Research Company.
         Figure 8  MATERIALS BALANCE OF MERCURY FROM COAL
                 IN THE UNITED STATES (kilograms)
                                52

-------
ratio is used to determine a region's mercury usage for specific processes
only (paint formulation,  tubes/switches,  control instruments).
     Distribution of consumption within a study region can be further broken
down to the county level, based upon the same criteria and factors as for the
region.  Emitters, however, by their very definition are diffuse and must be
treated in the computer analysis as point sources.   Hence they are assigned
to a county or, in the latitude-longitude program (DRSLL),  they are assigned
to the population center of that county.
     The total amount of mercury available (PROD in the listing of inputs in
Table 5), whether from an area or point source, must be assigned to an environ^
mental recipient.  A second equation applies here:

          Total mercury X fraction available X emission factor
               =  loss to environment

     The fraction of mercury'available for use is normally 1.0, but it can be
used to account for the mercury which remains  in place  (for example, as noted
above, in dental  use 75 percent is assumed to  remain in the teeth indefinitely)
The emission  factor accounts for the fraction  of mercury lost to the environ-
mental recipient.  As an example, Figure 8 showed that  the total estimated
mercury in coal used by  electric utilities in  1973 was  45,249 kg, all of which
was assumed to be available.   The emission factors used to distribute this
mercury are:

          EF   =  0.9  (air)
            f\
          EF-. =  0.0  (water)
          EPT =  0.1  (land)
            jj
         EP    =  0.0  (recycle)

     Then the annual emissions to air nationwide are:

          45,249  kg X  1.0 X 0.9  =  40,724 kg

In a  similar  fashion,  the emissions to land can be found to be 4,525 kg  (with
nothing to water or recycling).
                                        53

-------
       Emission factors for air,  water,  land,  and recycle were  in most cases
  based upon prior studies,  especially Ref.  12.   Those which could not be
  validated satisfactorily were researched further,  and  in  some cases substan-
  tial changes were introduced.
       Table 7 summarizes,  for each sector and each  SIC  study:

            o  Total mercury available on a national basis, in  thousands
               of kilograms
            o  The fraction of this mercury available
            o  Emission factors for each  environmental recipient

  Further  information on the detailed  methodology followed  for  each particular
  SIC  is given in Appendix D.

 Calculation of  Contributions  from Natural  Sources

      Degassing

      The calculation of natural degassing rates was based on work performed
 by others and on observations and assumptions developed in this study.   Various
 workers have developed both global and regional degassing rates (see Table 8).
 These observations and calculations indicate that  (1)  the average land degas-
                                                                   2
 sing rate is probably 0.2 microgram per square  meter per day (|Jig/m /day);
 (2)  degassing rates higher than the average are associated with mineralized
 deposits; (3) because of biological and chemical activity, over igneous sedi-
 ments high in mercury the degassing rate is probably higher than over land;
 (4)  the ratio between global land and ocean degassing  is  unknown,  but Mackenzie
 and Wollast's assumption  of equality (Ref.  26)  is probably adequate.  Using
 these concepts of degassing,  URS  calculated a probable  national degassing  rate.
Assumptions  on which these  calculations  were based  are  summarized below:

     1.  The base national  degassing  rate was assumed to be
                  2
         0.2 ng/m /day.
     2.  For moderately mineralized areas a degassing rate of
                  2
         0.4 |jUj/m /day was assumed.

                                       54

-------
                        TABLE 7. - Emissions  and Emission Factors for Mercury,  by SIC and Sector
en
en

Sector





II








III





i .. **

V


SIC
1092M
1092P
1021
1031
3331
3332
3333
3241
3274
021
2911C
2911M
2951C
331 2M
4091C
49110

4924C

4911C
• •
2.5B
2819M
285 1M
2879M
2833M
2812
2261

Description
Mercury Mining
Mercury Processing
Copper Mining
Zinc and Lead Mining
Copper Smelting
Lead Smelting
Zinc Smelting
Cement Processing
Lime Processing
Livestock
Fuel Oil
Refineries
Tars and Asphalt
Coke Ovens
Coal
Utilities - Oil § Gas
Burning
Natural Gas - Res., Comm.,
Ind.
Utilities - Coal Burning
Caustic
Catalyst
Paint
Pesticides
Pharmaceutical Manuf .
Chlor-Alkali
Textiles
*
Kg(x 105) of Hg
Available
0.02
8.8
0.11
0.02
45.3
S.3
5.1
2.6
0.5
36.3
17.1
2.4
17.8
10.2
11.2

12.1

6.0
45.3
9.5d
348. 5e
20.9
452.4
7.8
Fraction of
Available
Hg Lost0
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.97
1.00
1.00
1.00
1.00
1.00

1.00

1.00
1.00
1.00
0.001
0.001
1.00
1.00
Emission Factors for
Air
0.20
0.95
0.20
0.20
0.90
0.90
0.90
0.20
0.20
-
0.999
0.50
0.06
0.70
0.90

0.999

0.999
0.90
-
0.05
0.05
0.035
—
Water
0.10
•"
0.10
0.10
0.05
0.05
0.05
0.10
QUO
-
-
-
0.09
0.05
-

-

-
-
.80
0.90
0.90
.004
0.02
Land
0.70
0.05
0.70
0.70
0.05
0.05
O.'OS
0.70
0.70
0.50
0.001
0.50
0.85
-
0.10

0.001

.001
0.10
.20
0.05
0.05
0.501
0.98
Inventory
and Recycle
—
~
_
-
-
-
-
-
-
0.50
-
-
-
.25
-

-

_
-

-
0.46
-

-------
                         TABLE 7.  - Emissions  and Emission Factors for  Mercury, by  SIC and Sector
                                                              (continued)
01

Sector



V






VIIIA



VIIIB



SIC
2851F
38291M
3079P
3.0
36292M
36410M
36420M
2.5A
2879C
2879F
38291C
36292C
36410C
7391
2834C
2851C
36420C
8021

Description
Paint Formulation
Control Instrument Manuf.
Vat Dyes
Other
Tubes/Switches Manuf.
Lamp Manuf.
Battery Manuf.
Urethane
Nonagric. Pesticides
Agric. Pesticides
Control Instrument Cons.
Tubes/Switches Cons.
Lamp Cons.
Laboratory
Pharmaceutical Cons.
Paint. Cons. -
Battery Cons.
Dental
3
Kg(x 10*) of Hg
Available
247.7
196.8
7.8
69.2
63.3
46.6
563.6
15.6
44.0
19.1
206.9
53.9
43.4
22.7
20.9
247.1
497.0
92.3
Fraction of
Available
Hg Lost
0.65
0.01
0.001
1.00
0.025
0.04
0.005
0.15
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.70
1.00
0.25g
Emission Factors for
Air
0.45
_
0.5
0.15
_
0.05
0.05
0.05
0.10
-
0.04
0.05
0.05
0.10
_
0.95
0.05
0.04
Water
0.55
_
0.90
0.15
_
-
0.02
«.
0.40
0.15
-
—
0.26
1.00
-
-
0.72
Land

1.00
0.05
0.40
1.00
0.95
0.93
0.95
0.50
0.85
0.56
0.95
0.95
0.07
_
0.05
0.90
"•
Inventory
and Recycle

-
_
0.30
-
-
-
_
-
-
0.40
_
0.57
-
-
0.05
0.24
      a.  Amount of mercury available on a national basis within the  given SIC.
      b.  Fraction of available mercury that is lost to air, land,  or water, or that is recycled or remains as part  of an in-plant in-
         ventory .
      c.  Includes secondary processing.
      d.  Based on 3.7 ppm mercury concentration in caustic sold in 1973.
      e.  This total represents all the mercury used in the manufacture of organic and inorganic compounds that will be incorporated
         in paint and pesticides and used as catalysts.                                                                       .
      f.  Emissions to air, land, and water are based on the proportion of solid waste incinerated.   On a national average, this is
         about 5 percent.
      g.  The remaining 75 percent stays in the teeth and is considered unavailable.

      Source:  URS Research Company

-------
               TABLE 8.  - Natural  Degassing Rates  Calculated
                                  by Various Workers
Source
Ref.
No.
     Degassing Rate
       (yg/m /day)
        Remarks
Kothny
Weiss
Erickson
Mackenzie and
  Wollast
Anderson et al,
McCarthy et al.
 24
 8
25
26
 18
 4.0  (Franciscan soils)
 0.8  (normal  soils)
 0.5
10.0
 0.5
       0.0014
 0.064 - 41.86

 0.2 (natural back-
     ground)
Calculated on the basis of
an apparent increase in reg-
ional ambient air concentra-
tion  as the air mass moves
across Western California.

Calculated from the total
global atmospheric burden
(based on global concentra-
tion) and the average global
turnover time (10 days).  Emis-
sions were assumed to occur
only over land.

Same calculation method as
Weiss, but assumed background
concentration of 20 yg/m
and about 150 days turnover
time.
Used Weiss's number but
assumed degassing rates over
air and land are equal.

Same calculation procedure
as Weiss but assumed back-
ground concentrations  of
0.5 yg/m3.

Measured  degassing rate from
both mineralized and unmin-
eralized  areas.
Source:  Compiled by URS Research Company
                                       57

-------
       3.  For highly mineralized areas and known mercury deposits
                             2
           a rate of 0.8 jJig/m /day was assumed.

       For the purposes of differentiating between areas,  Figure 9,  which  indi-
  cates general areas of known mercury and mineralized deposits,  was used.  Using
  the degassing rates shown above and the total surface area (both land and water
  for each state), a state-by-state inventory of total annual degassing was made.
  On a national basis,  the annual degassing rate was estimated,  on the basis of
           2
  0. 36 |Jig/m /year, to be about 1 million kg per year.

       Runoff
       The other natural source contribution,  that of  erosion, was determined
 by calculating the average annual runoff  from each state  (Ref.  27) and multi-
 plying this figure by the average statewide  sediment concentration (Ref. 28).
 The total annual  sediment load was then multiplied by the average sediment
 mercury concentration of  71 ppb  (Ref.  29).   The state totals were then summed
 for the national  inventory, for an overall total of  about 200,000 kg per year.
 An  attempt was also made  to determine  the contribution from urban runoff.
 According  to this  calculation, urban runoff  contributes about 10,000 kg per
 year to  the nation's  waters, or about  5 percent of the soil erosion total.

 Contributions from Final Disposal

      Sewage

      In determining mercury losses to air, land,  and water from sewage treat-
 ment plants,  the initial assumption was that the overall sewage generation
 rate for sewered communities was 100 gallons per capita per day (gpcd) and
 that the rate for  areas served by septic  tanks was 75 gpcd.   The population
percentages served by community sewer systems and septic tanks were found in
the  U.S.  Census  of Housing (Ref.  30).   The per capita generation rates were
then multiplied by the specific study area population in question to estimate
the  sewage  flow.
                                       58

-------
                 SOURCE: URS Research Company.
Mineralized Areas
(Copper, Lead, Zinc, Silver, Gold)
Figure 9  MINERALIZED AREAS AND MERCURY DEPOSITS IN THE UNITED STATES

-------
       Fpr each study area,  an influent concentration of  2.0 ppb  (Ref. 12) and
  a removal efficiency of 50 percent (Refs.  31,  32) was assumed.  By multiplying
  together the estimated total flow,  the assumed influent concentration, and a
  series of conversion factors,  the total annual amount of mercury flowing into
  the typical sewage treatment plant was determined.
       The environmental receiving medium (air,  land,  or  water) of this mercury
  was calculated by multiplying the total annual amount of mercury reaching the
  plant by the percentage of the effluent lost to land or water and sludge land-
  filled or incinerated.   These  percentages  were derived  from the latest national
  inventory of municipal treatment plants, which lists the effluent and sludge
  disposal practices  for each plant in the nation (Ref. 33).

      Groundwater

      Very little information is available on the mercury  concentration of
 groundwater;  the most widely quoted value is 0.05 ppb (see Refs. 2,  34,  35).
 This is a reasonable figure because the generally high adsorptive capacity of
 soils should reduce mercury concentrations as water containing mercury perco-
 lates to the water table.
      Quantities of groundwater withdrawn in a given region were taken from U.S.
 Geological Survey (USGS) publications for individual states;  USGS data were
 also used for national figures (Ref. 36).   The yearly quantity of mercury was
 determined by  applying a list of conversion factors to the average  mercury
 concentration and quantity of groundwater withdrawn.

RESULTS

     Tables 9 through 13 summarize the  mercury emission data  for each study
region.  These tables and the  discussion to follow point out  the significant
contributors  of mercury to  the  environment.
                                       60

-------
     These tables have been designed to present the major results of each
regional inventory in a concise form,  while still providing the reader with
an overview of important findings.   At the far left of each table are listed
the sectors,  which correspond to the sectors of the I/O model (Fig.  7)  and
which are described in Table 4.   In the next column are the sector names,  with
important contributors within each sector.   Since all results are reported on
a sector basis rather than as individual SIC contributors,  the fraction each
SIC contributes to the total emissions of a sector is shown in parentheses in
this column.   For example,  in Table 9 under Sector V,  paint manufacture
accounts for 4 percent of the total emissions (1,839 kg)  of that sector.   The
next three columns indicate emissions (in both kilograms and pounds)  to air,
water,  and land, from each sector,  with the fourth column showing the totals.
A final column indicates the amount of recycling associated with the sector.
Only recycling which extends beyond plant limits is considered.   Thus,  for
example, recycle of rejected batteries in a manufacturing facility is not
included.  Emissions to air, water, and land are summed under "Totals"  in the
row following Sector VIIIB, indicating, for the region, total losses to these
three environments from man-related sources.  (These totals, of course,
include mercury introduced into commerce as well as that from unregulated
sources.)  The totals have been used to calculate the fraction each  sector
contributes to the given environment (shown in each column).  (For example,
in Table 9 the total emissions to water are 2,666 kg; for Sector V the con-
tribution to water is 584 kg—or 22 percent of the total, as shown parentheti-
cally.)  A similar percentage has been derived for the total column also.
     Below the totals and under the heavy line are figures  for the two other
identifiable sources of mercury—final disposal and natural sources.  Final
disposal, which includes sewage treatment plants and  landfills,  is not
included as part of the man-related inputs,  in order  to  avoid double account-
ing.  In the case of landfills no  values have been listed,  but it is assumed
that virtually all emissions to land  from Sectors  V,  VIIIA,  and VIIIB will be
deposited in either industrial or  public landfills.   Similarly,  in  the case
of water, most of the emissions from  these  same sectors will find their way
into public sewage treatment plant influents.   We estimated the mercury con-
tent of sewage influents independently,  and these values have been  included,

                                       61

-------
  as  have estimates  on the  disposal of the resultant mercury-bearing sludge to
  either  landfills or  to  incineration  (see Section IV for additional detail).
  On  the  assumption  that  most water-borne mercury wastes are directed to sewage
  influents, we presumed  that the independent estimate of the mercury content of
  sewage  would agree,  within limits, with the emissions determined by our inven-
  tory results.  In  Table 9 this is the case; that is,  we estimate the emissions
  from Sectors V, VIIIA,  and VIIIB to be 2,605 kg and 1,679 kg was the indepen-
  dent estimate—in  reasonable agreement.  However,  this agreement was not found
  for the other study  regions.  We conclude that the error (if it is an error)
  is largely in the  estimate of the mercury content of sewage influents;  the
  values were obtained from many sources and the sampling and analytical tech-
  niques used were in many cases open to question.
       Each table includes estimates of the magnitude of three natural sources—
  degassing from the earth's crust,  surface runoff,  and release through under-
  ground waters.   The derivation of these estimates  was discussed previously;
  it was noted in that discussion that  these are "best  estimates," based upon
  rather sketchy data (and in the case  of degassing  on  tentative hypotheses).
  We have,  however, used conservative estimates  in all  cases.   The totals for
  natural  sources  do  not include  any man-related sources  of mercury and it can
  be assumed that  the natural sources are not controllable to  any degree  by man.

 Regional Analyses

      The following  paragraphs present a brief overview of the findings  for
 each study region and discuss the implications of these findings.

      California Study Region

      As indicated in Table 9,  three sectors (I, VIIIA, and VIIIB) account for
virtually all of the mercury losses in this region.  These findings are conso-
nant  with the  characteristics of the region, which has relatively little heavy
industry  (and contains no mercury or mercurial manufacturers or processors)
but which does contain a  fair amount of  agricultural activity and which is the
only active mercury  mining  region in the nation at  the present time.  Looking

                                       62

-------
          TABLE  9. - Mercury Emissions  to the Environment in  the California  Study Region
Sector
I
II
III
V
VIIIA
Mercury Mining & Smelting
Imports
Recycling
Air
(1.0)
(0.0)
(0.0)

Other Mining
Copper
Lime
Cement
(0.80)
(0.10)
(0.10)

Unregulated Sources
Livestock
Fossil Fuels,
Comm. 8. Ind.
Household
Electric Power
(0.17)
(0.46)
(0.35)
(0.08)

Manufacturing 8 Processing
Chlor-Alkali
Paint Mfr.
Foam-Caustic
Other
(0.00)
(0.04)
(0,23)
(0.77)

Final Consumption:
Comm. 8. Industrial
Lamps
Controls/ Instr.
(0.06)
(0.29)
3,795 kg
-~(8,360 lb)
(34.6%)*
•14 kg
	 -(30 lb)
(0.1%)
913 kg
—(2,010 lb)
(8.3%)
195 kg
— ~(430 lb)
(1.8%)
300 kg
	 -(660 lb)
(2.7%)
Water
5kg
(10 lb)
(0.2%)a
9 kg
(20 lb)
(0.3%)
45 kg
(100 lb)
(1.7%)
586 kg
(1,290 lb)
(22.0%)
468 kg
(1,030 lb)
(17.5%)
Land
232 kg
(510 lb)
(1.1%)
68 kg
(150 lb)
(0.3%)
676 kg
(1,490 lb)
(3.2%)
1,058 kg
(2,330 lb)
(5.0%)
6,229 kg
(13,720 lb)
(29.2%)
Total
4,032 kg
(8,880 HO
(11.5%)D
91 kg
(200 lb)
(0.3%)
1,634 kg
(3,600 lb)
(4.7%)
1,839 kg
(4,050 lb)
(5.3%)
6,997 kg
(15,410 lb)
(20.0%)
Recycle
409 kg
(900 lb)
3,682 kg
(8,110 lb)
/IIIB Final Consumption:
Consumer Goods
Pharmaceuticals (0.03)
Paint (0.31)
Batteries (0.63)
Dental (0.03)
Totals
FINAL DISPOSAL
Sewage
» *_ JJM l 1
li&naxiii
NATURAL SOURCES
• Degassing
• Runoff
• Groundwater
5,739 kg 1,553 kg
—(12,640 lb) (3,420 lb)
(52.4%) (58.3%)
10,956 kg 2,666 kg
(24,130 lb) (5,870 lb)

54 kg 876 kg
(120 lb) (1.930 lb)



13,066 kg
(28,780 lb)
(61.3%)
21,329 kg
(46,980 lb)

749 kg
(1,650 lb)



20,358 kg
(44,840 lb)
(58.2%)
34,951 kg
(76,980 lb)

1.679 kg
(3,700 lb)

26,900 kg
(59.300 lb)
5,200 kg
(12,700 lb)
270 kg
(600 lb)
922 kg
(2,030 lb)
5,013 kg
(11,040 lb)





a. Emission by sector as a percent of total emission to air (or water or land).
b. Percent of all emissions by sector to all emissions in the study region.
c. An estimated one-third (including mercury metal)  is in soluble form; remainder moves rapidly into sediments.

Source:  URS  Research Company
                                                           63

-------
  further at mercury mining and smelting (Sector I)f  it can be seen that these
  sources alone accounted for almost 35 percent of the total emissions to air;
  considering the depressed state of the mercury mining industry in the subject
  year (1973) it can be seen that if mining activity were greatly increased, the
  Sector I emissions would be even more important in the overall picture.  How-
  ever, it should be noted that the mercury mining activity in the study region
  at the present time is conducted in very remote areas and is likely to have
  little impact on populated centers.   (The one exception is the New Almaden
  Mine, which is adjacent to the San Jose metropolitan area but which is now
  being phased out of operation.)
       Contributions from miscellaneous mining and processing (Sector II) were
  trivial,  and even those from unregulated sources (Sector III)  were surprisingly
  low.   In this sector,  emissions  related to livestock manure were only marginal
  (despite a  rather extensive livestock industry in the Central  Valley) and
  emissions related to fossil fuel usage were  small.   These low  emissions from
  fossil  fuel usage are  to be expected,  however,  since the study region uses
  natural gas (with low  mercury content)  for most residential and commercial
  uses, and for  steam-electric  power generation.   Since the energy "crunch" has
  developed, boilers using only natural  gas  have  generally been  converted to
  dual usage—natural gas or  fuel oil—with  increasing reliance  on fuel oil.
 The use of coal, with  its relatively high mercury content,  is virtually unknown
 in the region.
      The contributions from the final consumption sectors  (VIIIA and VIIIB)
 accounted for most of the man-related losses in the region, and were quite
 typical for this sector.  These losses were primarily to  landfills (incinera-
 tion of solid wastes is not practiced in the region) which are well regulated,
 ensuring that movement of mercury from the final disposal sites is minimal.
      The region was found to contribute above-average quantities of mercury to
 the  air  from degassing of the earth's crust.   The region consists of mercuri-
 ferous soils in general, and is therefore estimated to degas at about two
 times the  national average,  resulting in emissions to air estimated at 26,900
kg annually—about two and  one-half times higher than the total man-made emis-
sions to air.   Runoff (primarily  from mercuriferous soils) was about two
times higher than  were  man-related water discharges.  Groundwater usage is

                                         64

-------
relatively high in some parts of the region,  but groundwater is very low in
mercury content and contributed little.
     Two potential sources of mercury were initially considered:   (1)  the
fumaroles and geysers in the region,  and (2)  the spent mercury used during the
Gold Rush days in the amalgamation of gold.   However,  neither of these sources
proved to be important.  Steam from the  geysers,  which is  currently being used
commercially by the local electric utility, was found to contain low concen-
trations of mercury, as was runoff associated with the geysers.   (The geysers
are surrounded by abandoned mercury mines,  but of course their source of steam
is far below ground.)*
     It has been estimated that millions of flasks of mercury were mined and
used in processing gold in the early days of  California.   Even today it is
reported that, in digging among the gold tailings,  metallic  mercury may be
uncovered.  However, we were unable to find any substantive  data which indi-
cated that this residual mercury impinges in  any way on man  or his food chain.
We hypothesize that most of this mercury has  either been leached away or has
vaporized over the intervening century.

     Arizona Study Region

     The State of Arizona was chosen for study primarily because of its signifi-
cant copper mining and processing industry.  Copper ores contain mercury in
trace quantities, and this mercury is released during the smelting process.
From Table 10, it is evident that copper smelting operations are the major
contributors to the air and water emissions in this region.   Nevertheless, the
copper industry poses no significant or even appreciable hazard in terms  of
adverse health effects.  Smelter emissions do  lead to ambient mercury levels
in air which are substantially above background but well below  levels con-
sidered deleterious to health.
* Some geyser activity and much volcanic  activity have high mercury content;
  however, the quantities emitted are  low,  except for extreme volcanic activity,
  and in general cannot be considered  to  be important mercury contributors to
  the natural environment.
                                        65

-------
                TABLE 10. -  Mercury  Emissions to the Environment in the Arizona Study Region
Sector

I

II

III

V

VIIIA

VIIIB





Mercury Mining ft Smelting

Copper Mining & Smelting
Copper Mining
Copper Smelting

Unregulated Sources
Livestock
Fossil Fuels; Res.,
Comm. 8 Ind.
Elec. Utilities

Manufacturing and
Processing
Foam Caustic
Other

Final Consumption:
Comm. 8 Industrial
Lamps
Controls/ Instr.
Tubes/Switches
Laboratory
Pesticides

Final Consumption:
Consumer Goods
Pharmaceuticals
Paint
Batteries
Dental

Totals
FINAL DISPOSAL
Sewage
NATURAL SOURCES
o Degassing
o Runoff
o Groundwater




(0.004)
(0.996)

(0.06)
(0.92)
(0.02)

(0.20)
(0.80)

(0.07)
(0.29)
(0.47)
(0.09)
(0.08)

(0.03)
(0.34)
(0.52)
(0.11)



Air

h -

19,622 kg
*.(43,168 Ibl
(84.3%)*

70S kg
-+ (1,552 Ib)
(3.0%)

114 kg
	 »• (251 Ib)
(0.5%)

20 kg
	 (44 Ib)
(0.0%)

2,825 kg
-»-(6,216 Ib)
(12.2%)

23,286 kg
(51,231 Ib)
23 kg
(51 Ib)

Water

--

1,088 kg
(2,393 Ibl
(53.8%)*

144 kg
(316 Ib)
(7.1%)

127 kg
(280 Ib)
(6.3%)

96 kg
(211 Ib)
(4.7%)

568 kg
(1,249 Ib)
(28.1%)

2,023 kg
(4,449 Ib)
227 kg
(499 Ib)

Land

--

1,119 kg
(2,461 Ib)
(12.8%)*

433 kg
(953 Ib)
(5.0%)

385 kg
(847 Ib)
(4.4%)

2,055 kg
(4,520 Ib)
(23.6%)

4,723 kg
(10,390 Ib)
(54.2%)

8.715 kg
(19,171 Ib)
278 kg
(594 Ib)

Total

—

21,829 kg
(48,022 Ib)
(64.2%)D

1.282 kg
(2,821 Ib)
(3.8%)

626 kg
(1,378 Ib)
(1.8%)

2,171 kg
(4,775 Ib)
(6.4%)

8,116 kg
(17,855 Ib)
(23.8%)

34,024 kg
(74,851 Ib)
528 kg
(1,144 Ib)
84,400 kg
(192,200 Ib)
5,600 kgC
(12,400 Ib)
296 kg
(652 Ib)
Recycle

--





144 kg
(317 Ib)
(7.6%)

1.552 kg
(3,414 Ib)
(82.3%)

190 kg
(417 Ib)
(10.1%)

1,886 kg
(4,148 Ib)

a. Emission by sector as a percent of total emission to air (or water or land).
b. Percent of all  emissions by sector to all emissions in study region.
c. An estimated one-third (including mercury metal) in soluble form;  remainder moves rapidly into sediments.

Source:   URS Research Company
                                                             66

-------
     Most of the major copper smelters are located some 40 to 100 miles east
of the two major metropolitan areas.  Phoenix and Tucson.   The prevailing winds
are from the east,  so the Phoenix and Tucson areas could be exposed to mercury
emissions.  However,  diffusion modeling results indicate that the amounts of
mercury reaching highly populated areas would be trivial.   (Further discussion
of this hazard is included in Section IV.)
     The region displayed no unusual  anomalies with respect to emissions
from other sectors.   However, the state has many ore deposits;  as discussed
elsewhere, these are a source of mercury via degassing and contribute  heavily
to the total quantity of 84,400 kg emitted in the region.   The total emis-
sions to air from natural sources are four times higher than those from
man-related sources.  The mercury content of runoff is unusually low,
chiefly because the region is so arid.  Because of the highly variable flow
rates of the few streams in the state (which are dry much of the year), the
probable fate of mercury in sediments may well be different from that  in
streams that maintain flow the year around.  This point was not investigated,
however.

     Kentucky/Tennessee Study Region

     This study region contains a number of large, coal-fired power plants of
the Tennessee Valley Authority  (TVA).  Table 11 shows that almost 45 percent of
the total mercury emissions to air are due to Sector III sources, and that the
bulk of the Sector III contribution comes from the utilities plants.  However,
despite the large volume of mercury from these sources, the resultant ground
level concentrations of mercury vapor are very small—in fact, the  levels are
not much greater than that contributed by natural background.
     This area also contains  two chlor-alkali plants  which account for more
than 90 percent of the Sector V emissions.  Manufacture  of various electrical
apparatus, and catalyst use,  which initially were thought to contribute sig-
nificantly to the mercury  loss  in this region,  proved to be only minor sources.
     By discounting the losses  assigned to the chlor-alkali plants in the
region  (whose major losses  are  to  industrial landfills),  a truer picture
of other  losses can be obtained.  Without chlor-alkali losses, man-related
                                       67

-------
             /TABLE  11. -  Mercury Emissions  to the  Environment  in the  Kentucky/Tennessee Study Region
Sector
I
II
III

Air

Mercury Mining ft Smelting
Other Mining 6 Smelting
Zinc Mining
Copper Smelting
(0.01)
(0.99)

Unregulated Sources
Livestock
Electric Utilities
Natural Gas
Tars and Asphalt
Refineries
(0.02)
(0.71)
(0.04)
(0.10)
(0.00)
— * -
257 kg
	 »• (566 Ib)
(1.0%) a
9,114 kg
».(20,050 Ib)
(44.5%)
Water
14 kg
(31 Ib)
(0.5%)a
120 kg
(264 Ib)
(4.0%)
Land
14 kg
(31 Ib)
(0.0%) a
2,112 kg
(4,646 Ib)
(4.0%)
Total Recycle
285 kg
(628 IbJ
(0.5%)°
11,346 kg
(24,960 Ib)
(16.0%)
Manufacturing 8
Processing
Foam-Caustic
Catalysts
Other
Chlor-Alkali
Paint


(0.01)
(0.01)
(0.07)
(0.91)
(0.00)
                                              2,428 kg
                                          —»-(5,341 Ib)
                                              (12.0%)
   857 kg
(1,885 Ib)
 (27.5%)
 20,620 kg
(45,364 Ib)
   (44.0%)
 23,905 kg
(52,590 Ib)
  (34.0%)
  VIIIA
Final Consumption:
Commercial and Industrial
Lamps
Controls/Instr.
Tubes/Switches
Laboratory
Pesticides
Foam/Plastics


(0.06)
(0.27)
(0.43)
(0.05)
(0.16)
(0.03)
                                                100 kg
                                              • (220 Ib)
                                               (0.5%)
   530  kg
(1,165  Ib)
   (17.0%)
  9,600 kg
(21,120 Ib)
  (21.0%)
 10,230 kg
(22,505 Ib)
  (14.5%)
 VIIIB
    458 kg
 (1,009 Ib)
   (6.0%)
  6,589 kg
(14,495 Ib)
  (81.0%)
Paint (0.34)
Batteries (0.59)
Dental (0.04)

Totals
FINAL DISPOSAL
Sewage
Landfill
NATURAL SOURCES
o Degassing
o Runoff
o Groundwater
8,497 kg 1,585 kg 14,191 kg
-K18.694 Ib) (3,487 Ib) (31,220 Ib)
(42.0%) (51.0%) (31.0%)

20,396 kg 3,106 kg 46,537 kg
(44,871 Ib) (6,832 Ib) (102,381 Ib)





24,273 kg
(53,401 Ib)
(35.0%)

70,039 kg
(154,084 Ib)

1,691 kg
p1720 Ib)

15,300 kg
(33,700 Ib)
7,400 kgc
(16,200 Ib)
32 kg
(70 Ib)
1,024 kg
(2,252 Ib)
(13.0%)

8,071 kg
(17,756 Ib)





a. Emission by sector as a percent of total emission to air (or water or land).
b. Percent of all emissions by sector to all emissions  in the  study region.
c. An estimated one-third (including mercury metal)  is  in soluble form; remainder moves rapidly into sediments.

Source:  URS Research Company
                                                              68

-------
emissions to air account for almost 38 percent of all emissions;  these air
emissions are almost entirely from the coal-burning power plants  (Sector III).
(However, the quantities involved are still far less than those emitted by
degassing, even though a degassing rate about one-fifth of that for California
and Arizona study regions was assumed.)  The large TVA coal-burning power plants
are in most cases located in or are contiguous to populated areas.   However,
our analyses indicate that ground level concentrations are well within safe
standards.  The added load of some 7,000 kg to the global burden  is,  of course,
another matter.
     The Kentucky/Tennessee region was also analyzed on a county-by-county
basis.  This analysis was carried out only for emissions to land  and air,  as
losses to water in the area are quite small,  because of the rigorous standards
in effect for mercury discharges into this medium.
     Figures 10 and 11 show where the greatest losses of mercury  are occurring
in the Kentucky/Tennessee region.  As would be expected,  losses to  land are most
substantial where population density is the highest,  particularly near Louis-
ville and Memphis.  The probable major source of these emissions  is disposal
to landfills.  The high values in southwestern Kentucky and southeastern
Tennessee (see Fig. 10) are due to chlor-alkali plant sludges that  are disposed
of in landfills.  Nashville, Chattanooga, Knoxville, and northern Kentucky are
high population centers where significant amounts of mercury are  lost to land,
primarily from battery disposal but also from disposal of mercury-containing
lamps, tubes/switches, and control instruments.
     Figure 11 is a graphic summary of air emissions for the region.  Most of
the emissions are from Kentucky, where the greatest number of large power plants
are located, as well as two chlor-alkali facilities.  Note that the highest
emissions are adjacent to areas  of greatest population  (e.g., Memphis,  Louis-
ville, Nashville, and Knoxville).
     In order to determine where mercury losses are  occurring  in areas that  do
not conform to political boundaries, a portion of the Tennessee  River between
Knoxville and Chattanooga was chosen for analysis.   The area was arbitrarily
broken up into five "boxes," each  covering approximately 250 square miles (see
Fig. 12).  Boxes 1 and 5 surround  the metropolitan areas of Knoxville and
                                         69

-------
<1000

1000-5000

5000-10,000

>10,000
                                                                  CINCINNATI
             KENTUCKY
                                                                                    TENNESSEE
SOURCE: URS Research Company.
                     Figure 10  MERCURY EMISSIONS TO LAND IN KENTUCKY/TENNESSEE
                              STUDY REGION, BY COUNTY, 1973 (pounds)   "

-------
                                                                  CINCINNATI
                   <500

                   500-1000

                   1000-2000
                                                                                TENNESSEE
SOURCE: URS Research Company.
                      Figure 11  MERCURY EMISSIONS TO AIR IN KENTUCKY/TENNESSEE
                              STUDY REGION, BY COUNTY, 1973 (pounds)

-------
                            1437 (A)
                            1078 (L)
                              10 (W)
  Losses
   to
  Air      1003
  Land     2116
/ Water    244
              SOURCE: URS Research Company.
                                                                         Scale:   I"  =  30 Miles
Figure 12  LATITUDE/LONGITUDE ANALYSIS OF MERCURY EMISSIONS
         ALONG TENNESSEE RIVER, KENTUCKY/TENNESSEE STUDY REGION, 1973
         (pounds)

-------
Chattanooga, respectively.  These two boxes account for most of the emissions
to air, primarily because of the presence of power plants that use mercury-
bearing coal for fuel.
     No emissions are shown for Box 4,  but this is not entirely accurate.
Emissions that are based on either population,  manufacturing employment,  or
a multifactor are represented by a point in the center of a county (or group
of counties) and may not always be picked up in a "box" representing two points
of longitude and two points of latitude.   In contrast,  specific emitters such
as power plants, chlor-alkali plants,  smelters,  and manufacturers of consumer
goods  (e.g., batteries) can be so located.

     Louisiana Study Region

     This region was selected because it contains one of the largest petro-
chemical and oil refinery complexes in the nation.  In support of this complex,
several chlor-alkali plants are also located in the region; among these are
three  large mercury cell plants.  As shown in Table 12, these three plants are
the chief contributors of mercury emissions in the study region,  particularly
of emissions to land, which represent 79 percent of the total.  However,  these
losses are to industrial landfills, which are normally lined with barrier
material to ensure that the high water table in the state does not cause leach-
ing problems.  Thus the emissions can be considered as nonavailable to man's
environment.
     Aside from the chlor-alkali plant emissions, the unregulated discharges,
which  include those from industrial uses of natural gas  (which are very high
because the state is a major producer) and from refineries are important con-
siderations, particularly with  respect to air.   However,  the total quantities
involved are relatively small,  and it  does not appear  that petrochemical an'd
refinery operations are of  special concern.
     Disregarding the  chlor-alkali contributions,  the final consumption sectors
(VTIIA and VIIIB) are  about on  a  par with those  in the rest of the nation.
Pesticides, primarily  for nonagricultural applications,  are important mercury
emitters.   Incineration is  fairly widely practiced,  so that mercury contribu-
tions  to air  (from Sector VIIIB in particular)  are a bit above the national
average.
                                        73

-------
              TABLE  12. - Mercury Emissions  to the  Environment in the Louisiana Study Region
Sector

I
II
III







V




VIIIA




Air Water

Mercury Mining $ Smelting


Other Mining 8 Smelting

Unregulated Sources
Livestock
Electric Utilities
Nat. Gas Industr,
Nat. Gas - Comm.
and Res. '
Tars and Asphalt
Refineries
Fuel Oil

(0.02)
(0.18)
(0.42)

(0.06)
(0.12)
(0.13)
(0.07)

Manufacturing and
Processing
Foam-Caustic
Other
Chlor-Alkali


(0.01)
(0.02)
(0.97)

Final Consumption:
Comm, 1} Industrial
Lamps
Controls/ Instr.
Tubes/Switches


(0.08)
(0.20)
(0.31)



1,519 kg 56 kg
-* 3,343 Ib) (124 Ib)
(21%) a (6%)







2,855 kg 310 kg
	 »>(6,282 Ib) (683 Ib)
(40%) (32%)


26 kg 163 kg
» (57 Ib) (379 Ib)

(0%) (17%)
Land


'.

333 kg
(733 Ib)
d%r







27,838 kg
(61,243 Ib)
(79%)


2,027 kg
(4,460 Ib)
(6%)
Total




1,908 kg
(4,200.1b)
(5%)b







31,003 kg
(68,208 Ib)
(71%)


2,216 kg
(4,876 Ib)
(5%)
Recycle


_


~







125 kg
(275 Ib)
(8%)


1,032 kg
(2,269 Ib)
(68%)
 VIIIB
Pharmaceuticals (0.02)
Paint (0.32)
Batteries (0.64)
Dental (0.04)
Totals
FINAL DISPOSAL
Sewage
Landfill
NATURAL SOURCES
o Degassing
o Runoff
o Groundwater
2.753 kg 426 kg
	 ».(6,056 Ib) (937 Ib)
(39%) (45%)
7,153 kg 955 kg
(15,738 Ib) (2,103 Ib)

43 kg 259 kg
(95 Ib) (570 Ib)


5,114 kg
(11,251 Ib)
(14%)
35,312 kg
(77,687 Ib)

273 kg
(601 Ib)


8,293 kg
(18,244 Ib)
(19%)
43,420 kg
(95,528 Ib)

573 kg
(1,266 Ib)

4,400 kg
(9,590 Ib}
550 kgc
(1,210 Ib)
264 kg
(580 Ib)
36,7 kg
(808 Ib)
(24%)
1,524 kg
(3,352 Ib)




a. Emission by sector as  a percent of total emissions  to air (of water or land).
b. Percent of all emissions by sector to all emissions in the study region.
c. An estimated one-third (including mercury metal) is in soluble form; remainder moves rapidly into sediments.

Source:  URS Research Company
                                                       74

-------
     Because the degassing rate assumed for the state was low,  degassing
from the earth's crust (which in this region includes a substantial
amount of water bodies)  contributed much less mercury to the environment
than did man-related sources.  Runoff contributions were also somewhat
less than man-related discharges.  (As discussed earlier, these runoff
values do not include any mercury content of water "imported" into the region.
Thus the Mississippi River,  which has a mercury burden of its own accumulated
through its passage through many states,  does not show up as a  major natural
source within the study region.  However,  it is estimated that  the total emis-
sions to water for the study region are actually less than 5 percent of those
carried through the study region by the Mississippi itself.)*
     Industry in Louisiana deserves credit for undertaking the  necessary control
measures to reduce mercury discharges to meet the standards established by the
EPA.  Nevertheless, it must be noted that the state itself seems to have a
laissez-faire attitude toward industrial waste dischargers.   Perhaps the reli-
ance of the state economy on the vast petrochemical and allied  industries is
responsible for this attitude.  In any case, there is a considerable contrast
in the matter of mercury emissions from industrial complexes in this region
compared with those in other states where regulatory actions are even more
strict than those mandated by the EPA.

     New York/New Jersey Study Region

     Table 13 gives the results  of the analysis of the New York/New Jersey
study region.  The study area  includes parts  of New  York, New  Jersey,  Pennsyl-
vania, Delaware, and Connecticut and contains a varied range of mercury
contributors.
  It has been argued by some dischargers of mercury to the Mississippi River
  that  for this reason alone it is foolish to be concerned about the discharge
  of low-level  mercurial wastes to the Mississippi.  We cannot condone this
  viewpoint,  since widespread adoption of this attitude could only act to
  increase rather than to stabilize or decrease the mercurial burden presently
  carried by receiving waters.
                                       75

-------
             TABLE  13.  - Mercury Emissions to  the Environment in  the New York/New Jersey Study Region
   Sector
                                                   Air
                                                       Water
  Land
                                                                                                    Total
                  Recycle
   I

   II

   III
  VIIIA
  VIIIB
I Mercury Mining and Smelting
 Other Mining and Smelting
Unregulated Sources
Livestock
Electric Utilities
Natural Gas
Tars and Asphalt
Refineries
Fuel Oil
Coal
Blast Furnaces

CO. 00)
(0.37)
(0.06)
(0.14)
(0.02)
(0.27)
(0.13)
(0.01)

Manufacturing and
Processing
Foam-Caustic
Catalysts
Other
Chlor-Alkali
Organic Chemicals
Inorganic Chemicals
Medicinal Chemicals
Tubes/Switches
Lamps
Batteries
Controls/Instruments


(0.03)
(0.01)
(0.28)
(0.58)
(0.01)
(0.00)
(0.00)
(0.05)
(0.01)
(0.01)
(0.03)

Final Consumption:
Comm. and Industrial
Lamps
Controls/Instruments
Tubes/Switches
Laboratory
Pesticides
Foam/Plastics


(0.17) •
(0.38)
(0.21)
(0.04)
(0.17)
(0.03)

Final Consumption:
Consumer Goods
Pharmaceuticals
Paint
Batteries
Dental


(0.03)
(0.27) -
(0.67)
(0.03)
                                                 10,473 kg
                                                 (23,040 Ib)
                                                        161 kg
                                                       (354 Ib)
                                                        (2%)a
  2,021 kg
 (4,447 Ib)
   (2%)a
 12,655  kg
(27,841  Ib)
  (8%)5
                                                  3,489 kg
                                                 (7,693 Ib)
                                                   (7%)
                                                      2,467 kg
                                                     (5,439 Ib)
                                                       (24%)
 16,608 kg
(36,621 Ib)
  (18%)
 22,564 kg
(49,753 Ib)
  (15%)
                                                  3,306 kg
                                                 (7,290 Ib)
                                                   (6%)
                                                       2,992 kg
                                                      (6,598 Ib)
                                                        (28%)
 26,394  kg
(58,198  Ib)
  (28%)
 32,692 kg
(72,086 Ib)
  (21%)
a. Emission by sector  as percent  of total  emissions to air  (or land or water).
b. Percent of all emissions by sector  to all  emissions in study region,
c. An estimated one-third  (including mercury  metal) in soluble form; remainder moves rapidly into sediments.
Source:  URS Research Company
  1,745 kg
 (3,840 Ib)
   (8%)
 16,339 kg
(36,028 Ib)
  (76%)
Pharmaceuticals (0.03)
Paint (0.27)
Batteries (0.67)
Dental (0.03)
Totals

FINAL DISPOSAL
Sewage
Landfill
NATURAL SOURCES
o Degassing
o Runoff
o Groundwater
32,901 kg 4,844 kg
— » (72,547 Ib) (10,680 Ib)
(66%) (46%)
50,169 kg 10,464 kg
(110,570 Ib) (23,071 Ib)

107 kg 5,063 kg
(235 Ib) (11,164 Ib)



47,985 kg
(105,806 Ib)
(52%)
93,008 kg
(205,072 Ib)

2,491 kg
(5,494 Ib)



85,730 kg
(189,033 Ib)
(56%)
153,641 kg
(338,713 Ib)

7,661 kg
(16,893 Ib)

4,800 kg
(10,550 Ib)
170 kgc
(375 Ib)
137 kg
(301 Ib)
3,360 kg
(7,391 Ib)
(16%)
19,875 kg
(47,259 Ib)





                                                             76

-------
     Chemical manufacturing facilities,  which are highly concentrated in north-
eastern New Jersey,  contribute relatively small amounts of mercury,  principally
because their water discharges are closely monitored by regulatory agencies.
Electrical equipment manufacturing,  although it too is concentrated here,  also
contributes relatively small amounts of mercury to the environment.
     Mercury emissions to air from other sources are more significant.   New
York City and communities in its vicinity incinerate up to 40 percent of their
solid wastes.  Via incineration, mercury-containing batteries,  lamps,  control
instruments, tubes,  and switches contribute significant amounts of mercury to
the atmosphere.  For example, in a test conducted of New York's 73rd Street
incinerator, mercury concentrations ranged from 4.8 to 9 ppm (Ref. 106).
With 635 metric tons of solid waste incinerated daily at the facility,  this
amounts to approximately 3.0 to 5.7 kg of mercury per day, or 1,100 to 2,100
kg of mercury per year released to the air from this incinerator alone.  (This
assumes that all the mercury becomes volatilized, which of course is not the
case, but the figures do provide a range that probably encompasses the true
value.)  For all seven New York City incinerators, the mercury emitted to the
air could range from 10,000 to  19,000 kg per year.  Our inventory of the city
and adjacent areas indicated that just over 17,000 kg of mercury are emitted to
the air from all sources.  Assuming our  inventory  is reasonably accurate,
incineration is obviously one of the prime contributors of emissions to air.
     In contrast to the other study regions, in  the New York/New Jersey region
natural sources contributed  substantially  less than did man-related sources.
This is attributable to the  very high population density of  this  relatively
small area.  The implications of population density, and of  population-at-risk
with respect to exposure to  mercury in one form or another,  will  be discussed
later.
     As for the Kentucky/Tennessee  study region, the New York/New Jersey region
was also analyzed on a county-by-county basis,  in order to pinpoint zones of
relatively  high or  low mercury  losses.   This analysis was completed for all
three environmental parameters, even though water losses, because of  rigorous
controls, are  quite low.
     Figure 13 shows that the greatest loss of mercury to the air occurs in the
Philadelphia and New York metropolitan areas.  One of the primary  sources is

                                         77

-------
                                                              CONNECTICUT
PENNSYLVANIA
          STUDY REGION
	•• STATE BORDERS
	 COUNTY BORDERS
   DELAWARE
 SOURCE: URS Research Company.
              Figure 13  MERCURY EMISSIONS TO AIR IN NEW YORK/NEW JERSEY
                       STUDY REGION, BY COUNTY, 1973
                       (thousands of pounds)
                                         78

-------
consumer paint usage.   Other major contributors are fossil fuels and incinera-
tion.  Figure 14 shows that emissions to water are  also  highest in the New York
and Philadelphia metropolitan areas and in northeastern  New Jersey.   Mercury
emissions to water are well controlled by law,  and  our analysis indicates that
dental and pharmaceutical uses,  "other," and vat dye manufacture contribute the
greatest quantities to this medium.
     Figure 15 shows that land emissions follow the same general pattern.   The
principal losses arise from the disposal of electrical apparatus and chlor-
alkali plant sludges.
     A latitude/longitude analysis was undertaken to compare various sections
within the area to identify "hot spots" of emissions. Figure 16 shows the
areas chosen for analysis and Table 14 presents the results,  showing mercury
losses per 1,000 population.  The most striking feature  of the table is the
large per capita loss to land in eastern New Jersey (Box C).   Primary contribu-
tors in the point source category are a chlor-alkali plant, sludges from chemi-
cal plants, and users of nonfarm pesticides.
     The upper and lower portions of the Long Island Sound estuary (Boxes D
and E) were compared.  The upper estuary contributes much more mercury, on a
per capita basis, then does the lower.  Since the upper estuary (primarily
Connecticut) is much more  industrialized than the lower  (Long Island), this
result is not surprising.

National Inventory

     Mercury emissions in the coterminous  United States were  determined in a
manner similar to that employed for each study region.   However,  these estimates
were reached independently and  are not a summation of results for the study
regions.  The results for the national inventory are found in Table 15, which
shows mercury losses  by  sector and SIC category, and in Table 16, which shows
losses by  state, with total losses to land, water,  air, and recycling  (with
contributions from sewage, urban runoff,  and natural sources).
     A study  of  Table 15 indicates that in 1973 the total man-related losses
in the United States  were an estimated 1,525,100 kg.  Of this total  31 percent
was  directed  to air,  6 percent to water, and 63 percent to land.  An additional

                                         79

-------
                                                                    CONNECTICUT

PENNSYLVANIA^.-}*'
           STUDY REGION
........... STATE BORDERS

	 COUNTY BORDERS
    DELAWARE
                                                                .'.•"•".•. •••"•'
                                                                '.•••.••• •(
                                                                • ••••••••'
                                                                .*.*».*.*«•«,
            1-1.5


            .5-1.0

            «=: .5
 SOURCE: URS Research Company.
             Figure 14  MERCURY EMISSIONS TO WATER IN NEW YORK/NEW JERSEY
                       STUDY REGION, BY COUNTY, 1973
                       (thousands of pounds)
                                             80

-------
                                                               CONNECTICUT
PENNSYLVANIA'
           STUDY  REGION
	••••••  STATE  BORDERS

	 COUNTY BORDERS
   DELAWARE
SOURCE: URS Research Company.
            Figure 15  MERCURY EMISSIONS TO LAND IN NEW YORK/NEW JERSEY
                     STUDY REGION, BY COUNTY, 1973
                     (thousands of pounds)
                                         81

-------
                                                              CONNECTICUT
PENNSYLVANIA'
          STUDY REGION
	 STATE BORDERS
	 COUNTY BORDERS
   DELAWARE
SOURCE: URS Research Company.
                  Figure 16   LATITUDE/LONGITUDE ANALYSIS AREAS IN
                            NEW YORK/NEW JERSEY STUDY REGION
                                        82

-------
oo
u»
                 TABLE 14. - Per Capita Distribution of Mercury losses to Air, Land, and Water

                              in Five Areas of New York/New Jersey Study Region*

                                       (kilograms per 1,000 population)


Losses from .
Point Sources to
Air
Land
Water
Totals
Losses from
All Sources to
Air
Land
Water
Totals
A
New York City
0.4
0.4
0.1
0.9
2.1
2.8
0.4
5.3

B
Philadelphia
1.0
0.4
0.1
1.5
2.0
3.9
0.4
6.3

C
Eastern
New ^Jersey
1.1
5.4
0.5
7.0
2.4
9.8
0.9
13.1

D
Upper
Estuary
0.6
0.9
0.1
1.6
2.1
3.8
0.5
6.4

E
Lower
Estuary
0.4
0.1
0.1
0.6
1.1
1.5
0.2
2.8

      a.  See Figure 17 for area delineation.


      b.  Includes losses from Sectors I, II, HI, and V.


      c.  Includes losses from all sectors.


      Source:  URS Research Company.

-------
TABLE 15. - Total Mercury Losses  in  1973  for  the  Coterminous  United States,
                             By Sector  and  SIC  Category
                               (thousands  of. kilograms)


Sector SIC No.
I
1092M
1092P



II
1021
1031
3331
3333
3241
3274
3332


III
021
29 11C

291 1M
2951C
3312M
4091C

49110
4924C

491 1C


r
2. SB
	 2819M
2851M
— 2879M
^ 2833M
2812
	 2261
28S1F
38291M
3079P
__ 3.0
36292M
36410M
^_ 36420M


[IIA

2.5A
	 	 2879N
— 2879A
38291C
36292C
26410C
7391




Description
Mercury Mining fi Smelting
Mercury Mining
Mercury Processing
(including secondary)
Subtotals

Other Mining
Copper Mining
Zinc and Lead Mining
Copper Smelting
Zinc Smelting
Cement Processing
Lime Processing
Lead Smelting
Subtotals

Unregulated Sources
Livestock
Fuel Oil Residential, Com-
mercial, Industrial
Refineries
Tars and Asphalt
Coke Ovens
Coal - Residential, Commercial,
Industrial
Utilities - Oil and Natural Gas
Natural Gas - Residential, Com-
mercial, Industrial
Utilities - Coal
Subtotals

Manufacturing 8 Processing
Caustic
Catalyst Manufacture
Paint Manufacture
Pesticide Manufacture
Pharmaceuticals Manufacture
Chlor-Alkali
Textiles
Paint Formulation
Control Instrument Manufacture
Catalyst Usage
Other
Tubes/Switches Manufacture
Lamp Manufacture
Battery Manufacture
Subtotals

Final Consumption:
Comn>. 6 Industrial
Urethane Pt Misc.
Nonagricultural Pesticide Use
Agricultural Pesticide Use
Control Instrument Consumption
Tubes/Switches Consumption
Lamp Consumption
Laboratory Usage
Subtotals



Total Losses
Air

0.01
7.84

7.85
(1.7%)

0.02
0.00
40.77
4.59
0.50
0.08
4.75
50.71
(10.8%)

0.00

16.94
1.15
1.10
7.16

9.97
11.99

15.46
40.71
104.48
(22.2%)

0.00
0.00
0.01
0.00
0.00
14.84
0.00
0.29
0.00
0.05
10.28
0.00
0.40
0.13
26.00
(5.5%)


0.12
4.39
0.00
16.54
7. S3
6.07
2.28
36.93
(7.8%)
Water

0.00
0.00

0.00
(0.0%)

0.01
0.00
2.26
0.25
0.25
0.04
0.26
3.07
(3.5%)

0.0

0.00
0.00
1.67
0.51

0.00
0.00

0.00
0.00
2.18
(2.5%)

7.61
0.02
0.20
0.06
0.02
2.93
0.15
0.35
0.00
0.10
10.23
0.00
0.00
0.05
21.72
(24.8%)


0.00
17.56
2.83
0.00
0.00
0.00
5.92
26.31
(30.0%)

to
Land

0.01
0.41

0.42
(0.0%)

0.08
0.01
2.26
0.25
1.76
0.29
0.26
4.91
(0.5%)

17.70

0.02
1.15
14.99
2.56

1.11
0.01

0.01
4.52
42.07
(4.4%)

1.90
0.00
0.05
0.00
0.00
226.83
7.63
0.00
1.97
18.85
8.70
1.57
1.34
2.49
271.33
(28.1%)


2.22
21.95
16.02
107.50
46.26
37.31
1.59
232.85
(24.1%)
Total
Mercury
Lost

0.02
8.25

8.27
(0.5%)

0.11
0.01
45.29
5.09
2.51
0.41
5.27
58.09
(3.8%)

17.70

16.96
2.30
17.76
10.23

11.08
12.00

15.47
45.23
148.73
(9.8%)

9.51
0.02
0.26
0.06
0.02
244.60
7.78
0.64
1.97
19.00
29.21
1.57
1.74
2.67
319.05
(21.0%)


2.34
43.90
18.85
124.04
53.79
43.38
9.79
296.09
(19.4%)

Total
Recycled

0.00
0.00

0.00
(0.0%)

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
(0.0%)

0.00

0.00
0.00
0.00
0.00

0.00
0.00

0.00
0.00
0.00
(0.00%)

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.97
0.00
20.66
0.00
0.00
0.00
20.66
(14.1%)


0.00
0.00
0.00
82.69
0.00
0.00
12.98
95.67
(65.2%)
                                           84

-------
         TABLE 15.  - Total Mercury Losses in 1973 for  the Coterminous United States,
                                          By Sector  and SIC category)
                                             (thousands of  kilograms)
                                                    (continued)
Sector
          SIC No.
Description
                                                                Total Losses to
Water
                                                                                  Land
                                                                       Total
                                                                      Mercury
                                                                        Lost
                                        Total
                                      Recycled
VIIIB
Final Consumption:
Consumer Goods:
	 2834C
2851C
36420C
8021
Total
Final Disposal
Sewage
Urban Runoff
Natural Sources
o Degassing
Pharmaceuticals Consumption
Paint Consumption
Battery Consumption
Dental Applications
Subtotals




o Runoff and Groundwater
1.04
173.61
69.71
0.93
245.29
(52.0%)
471.26

4.01
0.00

1,018.70
0.00
17.77
0.00
0.00
16.65
34.42
(39.2%)
87.70

19.92
11.70

0.00
188.30
2.09
9.14
403.30
0.00
414.53
(42.9%)
966.11

22.88
0.00

0.00
0.00
20.90
182.75
473.01
17.58
694.24
(45.5%)
1,525.07

46.80
11.70

1,018.70
188.30
0.00
0.00
24.89
5.55
30.44
(20.7%)
146.77

-


-
*•
 a. The values of losses  in some cases vary slightly from those given in Section  IV  (Technology Assessment) because of
   refinements in the data base that are reflected in the latter section.  However, such differences as do occur are
   small and do not affect the findings of the report; hence no attempt has been made to adjust the values in this
   table.
 Source:  URS Research Company.
                                                              85

-------
                     TABLE 16. - Total Mercury  Losses  in 1973  for the Coterminous  United  States,
                                                             by State
                                                     (thousands of kilograms)
00
Ol
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Losses
Land
41.02
6.88
7.09
65.22
7.76
11.72
10.58
20.59
28.17
2.63
47.60
21.12
10.87
11.02
29.45
46.15
7.36
13.25
20.93
33.03
from Man-Made Sources to
Water
1.56
1.76
0.72
7.57
0.94
1.29
0.35
2.41
1.92
0.40
4.56
2.27
1.04
1.41
1.35
1.53
0.45
1.57
2.32
3.72
Air
9.89
22.28
3.04
36.75
3.51
6.05
1.93
12.07
9.53
2.78
22.44
12.17
5.65
3.66
10.51
8.38
2.54
7.56
11.37
18.83
Recycling
2.46
0.96
1.32
13.10
1.17
2.85
0.44
3.35
3.46
0.48
9.21
4.79
1.77
1.25
2.12
1.76
0.82
2.64
4.71
7.77
Losses from
Natural Sources to
Degassing
9.59
85.77
9.82
118.25
39.24
1.84
0.37
10.25
21.96
31.27
10.54
6.82
10.58
15.46
7.50
8.49
5.85
3.74
2.96
10.74
Runoff
4.14
11.33
1.97
26.30
6.01
0.01
0.01
0.77
2.73
6.30
1.41
3.22
5.88
5.68
2.18
1.18
0.49
0.30
0.14
0.63
Losses
from Sewage
0.59
0.46
0.89
5.20
0.51
0.58
0.13
1.53
0.96
0.15
2.78
1.12
0.64
0.52
0.65
0.82
0.19
0.96
1.32
2.01
Losses from
Urban Runoff
0.39
0.19
0.15
1.40
0.20
0.15
0.05
0.71
0.35
0.08
0.64
0.36
0.28
0.15
0.11
0.27
Of A
.10
0.40
0.26
0.45

-------
                          TABLE 16.  - Mercury Losses for 1973 for the Coterminous United States,
                                                           by State
                                                    (thousand of kilograms)
                                                           (continued)
CD
>4
Losses from Man-Made Sources to
State
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
\f*l^*r »
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee

Texas
• •_•_ <_
Utah
Vermont
Virginia
Washington
Land
• 13.94
7.59
17.24
4.17
6.15
1.89
2.94
41.90
2.98
78.44
25.28
6.39
43.48
10.50
7.13
46.22
3.94
12.98
3.44
31.67
59.43
+}& • T^»*
3.62
IXC
• **o
15.33
22.43
Water
1.51
0.82
2.02
0.74
0.71
0.32
0.32
3.37
0.51
7.14
2.17
0.97
4.49
1.27
0.81
5.12
0.41
1.15
0.42
2.27
4.70
0.72
0.15
w • * **
1.76
1.50
Air
7.02
5.11
13.78
3.97
2.31
5.64
1.44
14.57
4.87
35.47
12.23
0.91
25.33
4.57
3.61
34.83
1.82
5.49
1.11
10.00
25.89
7.84
0.69
8.14
8.98
Recycling
2.51
1.41
3.29
0.28
0.78
0.20
0.67
6.18
0.37
13.03
4.65
0.19
9.51
1.38
1.33
10.27
0.90
2.27
0.27
3.18
6.68
0.57
0.29
3.05
2.13
Losses from
Natural Sources to
Degassing
14.94
8.94
13.04
55.01
14.46
83.10
3.42
1.42
45.91
18.10
18.46
13.10
7.75
13.00
72.74
17.00
0.20
5.71
14.36
7.81
49.56
31.03
.2.19
15.04
25.18
Runoff
0.81
5.97
4.28
4.56
1.87
4.18
0.10
0.17
20.66
0.61
1.25
0.91
1.80
25.47
2.60
0.83
0.02
1.14
1.27
5.24
18.53
2.98
0.15
1.07
0.87
Losses
from Sewage
0.89
0.42
0.95
0.15
0.36
0.13
0.15
1.80
0.24
4.38
0.94
0.14
2.49
0.61
0.45
2.72
0.21
0.49
0.15
1.05
2.78
0.28
0.09
0.99
0.74
Losses from
Urban Runoff
0.26
0.13
0.38
0.07
0.10
0.07
0.05
0.18
0.10
0.59
0.33
0.05
0.57
0.65
0.13
0.43
0.03
0.45
0.09
0.39
1.43
0.12
0.02
0.39
0.14

-------
                   TABLE  16.  -  Total Mercury Losses in 1973 for the Coterminous United States,
                                                           by State
                                                   (thousands of kilograms)
                                                         (continued)
State
West Virginia
Wisconsin
Wyoming
Totals
Losses
Land
31.22
20.71
1.18
966.08

from Man-Made Sources to
Water
0.80
2.31
0.11
87.73

Air
5.36
8.39
0.74
471.05

Recyling
1.01
3.68
0.13
146.60

Losses
from
Natural Sources to
Degassing
9.10
10.30
36.76
1,018.67

Runoff
0.58
0.92
2.90
192.42

Losses
from Sewage
0.34
1.02
0.08
46.55

Losses from
Urban Runoff
0.11
0.11
0.04
13.37

00
00
Source:  URS Research Company

-------
146,800 kg were recycled.   Sector VIIIB (personal consumption)  accounted for
45.5 percent of the total,  followed by 21 percent for manufacturing (Sector V)
and 19 percent for industrial and commercial consumption (Sector VIIIA).   Total
losses through sewage were estimated to be 46,800 kg,  as compared with a total
of 87,700 kg from all man-related emissions to water—a reasonable comparison.
The estimated emissions from natural sources were slightly more than 1 million
kg (as compared to 471,000 kg to air from man-related sources)  from surface
runoff and 188,000 kg from underground waters (as compared with 87,700 kg from
man-related losses to water).  Finally, on a national basis,  an estimated
12,000 kg of mercury are lost as a result of urban runoff.
     In Table 17 the results for the national inventory are compared with those
of the various regions.  In Part A of the table the comparison is by sector
and indicates the differences found in the regions studied.  As discussed
earlier, there were significant variations attributable to specific sources
in each region.  This suggests that the choice of regions for study was very
valuable in terms of highlighting different use and discharge patterns.
     In Part B of the table, total man-related losses (by environmental recep-
tor) are compared on a national and a regional basis.  Again significant
differences between regions can be noted.  However, if the major perturbations
in a given region are recognized and discounted  (for example copper smelters in
the Arizona region), the results for the regions approach the national norm.
     In Part C of the table, total man-related losses versus total mercury
losses from natural sources are  listed for both  the national and regional
results.  As anticipated,  the ratios between man-related and natural  (degassing)
emissions to air vary widely, as do those for man-related  emissions  to water
versus those for runoff plus groundwater.   On a  regional basis these  ratios are
strongly affected by the amounts of man-related  emissions  (amounts that are
primarily related to population density and to manufacturing activities that
utilize mercury) and by the relative  size of the region.   The New York/New
Jersey study region, with  the highest population density and the smallest total
land area of the selected study regions,  is greatly above the national average
with respect to both ratios.
                                        89

-------
           TABLE 17.  - Comparison  of Regional  and National  Inventory Results



                              A.  Total Man-Related Losses, by Sector
Percent of Total Losses
Sector
I
II
III
V
VIIIA
VIIIB
Name
Mercury Mining
Other Mining
Unregulated
Manufacturing
Consumption- Ind .
Consumption-Pers .
Total
B.
Losses to

Air
Water
Land
Total
National
0.5%
4.0
10.0
21.0
19.0
45.5
100.0%
California
11.5%
0.5
5.0
5.0
20.0
58.0
100.0%
Arizona
0%
64.0
4.0
2.0
6.0
24.0
100.0%
Kentucky/
Tennessee
0%
1.0
44.5
12.0
0.5
42.0
100.0%
Louisiana
0%
0
5.0
71.0
5.0
19.0
100.0%
New York/
.New Jersey
0%
0
21.0
7.0
6.0
66.0
100.0%
Total Man-Related Losses, by Environmental Receptor

National
31.0%
6.0
63.0
100.0%
C. Ratio of Total
Ratio
Man-Related (air)
Degassing
Man-Related (water)
Runoff

National
0.46%
0.47

California
31.0%
8.0
61.0
100.0%
Man-Related
California
0.41%
0.49
Percent of
Arizona
68.0%
6.0
26.0
100.0%
Losses to
Arizona
0.28%
0.34
Total Losses
Kentucky/
Tennessee
29.0%
4.5
66.5
100.0%
Losses from
Kentucky/
Tennessee
1.37%
0.42
Louisiana
16.5%
2.5
81.0
100.0%
New York/
New Jersey
33.0%
7.0
60.0
100.0%
Natural Resources
Louisiana
1.69%
1.17
New York/
New Jersey
10.47%
34.08
Source:  URS Research Company
                                                90

-------
     Because population density is an important consideration,  the emission
data in Table 16 were normalized on the basis of population.  The mean and
standard deviations,  in terms of kilograms of mercury emitted per 1,000 popu-
lation, were as follows:  air, 1.81 ± 0.53;  water,  0.41 ± .08;  land,  4.55 ±
1.20.  The results for each state were then plotted on maps that showed the
average case as ± 1 standard deviation,  variations  as greater than 15.0,  and
hot spots as greater than 25.0.  These results are  shown in Figures 17 through
19.  In all cases, it was possible to identify hot  spots as being related to
either chlor-alkali plants or copper smelting facilities.  However,  the
presence of such activities in a given state did not necessarily result in
overall elevated concentrations for the region.

Risk Analysis

     In general our findings  indicated that final consumption (Sectors VIIIA
and VIIIB) represented relatively constant emissions on a per capita basis both
nationally and for each study region.  Indeed, the more highly populated study
regions have proportionately  higher total mercury losses.  But at the same
time we recognized that total mercury emissions might also be related to land
area.  To test this possible  relationship we plotted, for various study
regions and for  selected  states, emissions to  land, air, and water (in terms
of kilograms of mercury per  square kilometer)  as a  function of population
density  (in terms of thousands of  persons per  square kilometer).  Initially
the  correlation was not obvious, but when we discounted the hot  spots identi-
fied in Figures  17 through 19,  we  obtained  the surprisingly good correlations
shown  in Figure  20.  Emissions per square kilometer increase proportionately
to the population, but the rate of increase is markedly different for land,
air, and  water emissions.
     The  relationships shown in Figure 20 can be reduced to a simple equation:
                               2                     2
          Emissions  (in kg)/km   =  K (population/km )
where  the value  of K is:
                                        91

-------
NO
                                                                    Kg per 1,000 population
                                                                    >4.0
                                                                    >2.34
                                                                    <1.28
                                                                    2.34-1.28
  Hot spots (due to copper smelting)
> Standard deviation
         I deviation
  Mean
        SOURCE: URS Research Company.
                                   Figure 17   ANNUAL ELEVATED MERCURY LOSSES TO AIR

-------
01
                                                                    Kg per 1 ,000 population
                                                                    >0.8
                                                                     <0.33
                                                                     0.49-0.33
       Hot spots (due to copper smelting)
?ig;gsj3 >Standard deviation
RTTT33 < Standard deviation
       Mean
        SOURCE: URS Research Company.
                                 Figure 18   ANNUAL ELEVATED MERCURY LOSSES TO WATER

-------
ID

                                                                   Kg per 1,000 population
                                                                   >9.0
                                                                   >5.7
                                                                   <3.4
                                                                   5.7-3.4
   Hot spots (due to chlor-alkali)
l > Standard deviation
]< Standard deviation
   Mean
       SOURCE: URS Research Company.
                                 Figure 19  ANNUAL ELEVATED MERCURY LOSSES TO LAND

-------
en
                                                         I     I    I    F~~lI
                          Note: O indicates selected study region

                               • indicates state data (from Table 16)
                                                                                             N.Y.-NJ.^    —I
                                                             I     I    I     I
I     I    I  II
7i  m _~  TJ
                   0   40   80  120  160  200  240  280  320  360 400  440  480  520  560 600  640  680  720 ~760

                                               Number of Persons/Square Kilometer


             SOURCE: URS Research Company.
                       Figure 20   MERCURY EMISSIONS AS A FUNCTION OF POPULATION DENSITY

-------
            Air    =  0.00180

            Water  =  0.000385

            Land   =  0.00425

  This equation can be further  simplified by eliminating areal considerations,

  so that we obtain the following per capita emission factors:

                                           Per Capita
                        Receiving        Emission Factor
                         Medium          (grams/per son)

                          Air                  1.80

                          Water                0.385

                          Land                 4.25

 These factors make it possible to estimate quickly the expected emissions  from

 any population base.*

      From this simple model it is possible to show the impact on a community

 of new activities that emit mercury and to assess the importance of emissions

 from natural sources on a given population—notably the population at risk.

 To illustrate, consider two cities with the following characteristics:

                                             City A       City B

              Area (km2)                         100          200

              Population                      50,000      500,000
                              2
              Density (pop./km )                  500        2,500

              Per capita air
                emissions  (kg/person)        0.00180      0.00180

              Total annual  mercury
                contribution to air (kg)           90          900
* The usefulness of the regional and national inventories is again demonstrated
  by these per capita emission factors.  Based on national values only, the
  per capita emission factors were found to be 1.40g/person for air, 0.304 for
  water, and 3.24 for land.  The values derived from Figure 20 then more nearly
  represent the average case (which includes the effects of metropolitan areas)
  and can safely be used to estimate per capita emission factors for the higher
  population densities found in metropolitan areas.
                                      96

-------
     Now consider the addition of a new industrial activity designed to release
to the air no more than 0.1 kg of mercury per day (36.5 kg per year).   For City
A this added burden represents an increase of over 40 percent,  but for City B
the added burden is only 4 percent.  In other words,  the added burden has a
much more noticeable effect on the quality of the air in the smaller community.
But contributions from natural sources must also be considered. Using the
national average degassing rate of  0.337 kg/km2 per year,  City A  will add
another 33.7 kg from this source to its man-related base contribution of 90 kg
per year—or an additional 37 percent.  For City B the additional  burden from
natural degassing within its own boundaries will equal 67.4 kg per year,  an
increase of only 7.5 percent over man-related contributions.*
     In terms of a population exposed to an additional mercury burden,  it would
appear that City B  (high population density) would be least affected.   However,
if mercury vapor concentrations are already high, a small incremental increase
would merely perpetuate an already bad situation, at the same time exposing a
large group of people.  On the other hand, if a mercury-emit ting activity is
established in an area of lower population density, this new source adds sig-
nificantly to the assumed low concentrations which are already present.  There-
fore the impact is  small in terms  of people exposed and resultant concentra-
tions, although the relative increase  in mercury burden is unquestionably high.
The problem then, in evaluating  new or existing  emitters, is to limit emissions
to the environment  to the greatest extent possible,  so that  the resultant
impacts are as negligible as possible.
     This case study illustrates the  fact that  the population itself  could be
the major source  of the mercury  pollution it experiences (obviously point
source and  industrial discharges,  if not well controlled,  could be much
greater—at least locally—than  population-derived emissions). The case study
  We recognize that metropolitan areas will almost certainly degas at different
  rates from those for the general countryside,  and that "incoming" air carries
  a global mercury burden.  However, we use degassing within the corporate
  boundaries for illustrative purposes and to indicate the importance of various
  sources of mercury to the inhabitants.
                                        97

-------
  also suggests that natural sources and background  can be important factors in
  the total picture.  On a regional basis natural  sources also have some interest-
  ing implications.   Figure 20 shows that, the average annual degassing rate for
  the nation is 0.34 kg/km ,  which can be equated  with the contribution from a
  population density of 190 persons per  square kilometer—that of the State of
  Ohio,  say.   The average annual  contribution from runoff from surface waters
                                                                   2
  plus groundwaters  accounts,  on  the national basis, for 0.062 kg/km ,  equivalent
  to the contribution from a population  density  of 150 persons per square kilo-
  meters.*  Thus,  even in smaller but heavily populated states, natural sources
  are important.

  SELECTION OF TECHNOLOGIES FOR STUDY
      One of the reasons for developing regional and national inventories is to
 identify emitters that, on either a local or a national basis, generate suffi-
 cient waste products that are emitted to the environment to be of special con-
 cern.  For mercury,  these emissions are summarized, for the national basis, in
 Figure 21.  From the figure certain obvious emitters can be identified; these
 require further consideration as far as technologic and economic evaluation is
 concerned (for example, chlor-alkali in the manufacturing sector and battery
 consumption in the final consumption sector).   A brief discussion of the infor-
 mation shown in Figure 21 and detailed in Table 15 follows.
      In Sector I,  mercury mining and processing contribute only 0.5 percent of
 the total mercury lost in the nation annually,  although for the State of Califor-
 nia this  value rises to 11.5  percent.   Because  mercury production is an essential
 link in this  study and because its production in the United States may well
 increase  significantly in future years as market demand grows,  it was selected
* We hesitate to place much importance on applying a  contribution factor from
  natural waters to even regional comparisons because of  the  variations in
  stream flow, geology, etc. of different regions.  However,  the value for the
  national average, encompassing all types of situations,  is  believed valid and
  can be used cautiously for purposes of making comparisons.
                                      98

-------
to
to
 SECTOR I
 Mercury Production
 SECTOR II
 Smelting and Processing
 SECTOR III
 Electric Utilities
 Refineries and Coke Ovens
 All Other Fossil Fuels
 Livestock
 SECTOR V
 Chlor-Alkali
 Chemical Manufacturing
   and Processing
 Electrical Manufacturing
 Miscellaneous ("Other")
 SECTOR VIII
Control Instrument Use
Tubes, Switches, and Lamps
Nonagricultural Pesticides
           Laboratory
          Dental Uses
            lint Consumption
                                                                                                        -**-
                                           2
                                    :•*••
                                   ;j_
                                   .•A'.-'-'--'/-.''Vr:?:.-''.-.--'.;M '    h££gSESftS&:Ctlvv T-'^-y^J
        Water
I      |  Air
ES£oEl  Land
                                                                                                                      254
                                                                                            3C

                                                                                                      3€3*EF2ZSS3 473

                                       5   10   15    20  25    30   35   40   45   50   55   60   65
                                                     Thousands of Kilograms of Mercury
        SOURCE: URS Research Company.
                              Figure 21   SUMMARY OF MERCURY LOSSES TO THE U.S. ENVIRONMENT
                                         BY SECTOR AND CLASS, 1973

-------
  few: study.   However,  the secondary mercury processing industry—that is, the
  recovery of mercury from its  final use—is not considered.  This industry was
  well covered in a previous study (Ref.  38).  Also, contacts with representatives
  of the  industry indicated that  no  substantial changes have occurred in recent
  years.   Certainly the industry  is  well  aware of the hazards of mercury, and
  total (estimated)  emissions are very  small.
       In Sector  II,  copper smelting accounts for more than 75 percent of all
  emissions.   For this  reason,  copper mining and smelting were selected for
  further study.   However,  it was not considered necessary to study zinc and
  lead mining and smelting,  since the mercury content of these ores is similar
  to that for copper and the technology is not very different with respect to
  the ways in which  mercury is  released to the environment.  (It might be noted
  that some recovery of mercury from zinc smelters is currently under way in the
  country.)   Cement  and lime processing were considered initially as a possible
  source  of mercury  release  to  the atmosphere, but it was soon found that the
  total quantities involved  are low  and that the impact, even on a regional basis,
  is  insignificant.   Phosphate  rock, which may be heated in the kiln or applied
  directly  to  land as a  fertilizer,  was also considered as a possible source of
  serious losses to the  environment.  While it is true that phosphate rock does
 have a background mercury  content, even with the high volume of usage the total
 quantities of mercury involved do  not present any problem with respect to
 processing or concentration in soils.   A more serious problem,  discussed under
 Final Disposal Technology fin the next section,  is the possible accumulation of
 heavy metals (one of which;is mercury)  in soils in which sewage sludges are
 used as  an additive.
      Sector III, which includes miscellaneous unregulated mercury sources,
 considers emissions from the numerous applications of fossil fuels in the
 economy; it also includes the item of livestock manure.   This manure,  which is
 estimated to account for  losses of almost 18,000 kg of mercury per year, pri-
 marily to land,  represents about 1 percent of the total mercury losses in the
 nation.   However,  this loss is in a dilute form and is generally restricted to
 areas  of low population.   Also,  the livestock industry is actively pursuing
means  of preventing animal discharges  from reaching public watercourses where
                                        100

-------
they might have serious consequences  (the  emphasis  in this  search, of course,
is on disease and BOD).  For these reasons it  did not seem  profitable to
investigate this particular "technology."
     Of the emissions associated with fossil fuel consumption, electric utili-
ties account for the largest portion—over 55,000 kg per year.  Most of this is
emitted to air.  This usage includes  natural gas  (which is  on the decline)  and
fuel oil and coal (which are on the upswing).  Electric power utilities are
likely to become even more important  as emission  sources in the future as coal,
which typically has the most mercury  of any of the  fossil fuels on a weight
basis, comes into more prominent use.  Hence utilities were selected for tech-
nological evaluation.
     Fossil fuels are also used for residential and commercial heating and in
miscellaneous industrial applications.  While  these uses represent a significant
source of mercury emissions to the environment (again primarily to air), these
emissions in general are so diffuse and individually small  that the possibility
of controlling them at the source of  release is poor.  Hence  it was decided
that further study of these sources was not warranted.
     Refineries, which process crude  oil containing traces  of mercury, and coke
ovens, which destructively distill coal containing mercury, represent a  rela-
tively small source of mercury emissions.   Moreover,  these  emissions are some-
what controlled by the emission control devices  or vapor recovery systems that
accompany each process.  Therefore,  emphasis upon mercury release and control
did not seem warranted, and these sources were dropped from further consideration.
     It appears that a fair quantity of the mercury in crude oils remains in
the residuals  following distillation; these residuals directly or indirectly
supply tars and asphalts for numerous applications, such as highway and building
construction.  We hypothesize that in such  usage a fraction of the mercurial
content is released each year by continuing erosion; most  of this eroded material
would end up on a land surface,  but  some  of the  mercury would reach watercourses.
(Certainly a portion of the mercury  in  the  11,700  kg estimated to be present in
urban runoff comes from this source.)   Although  the slow loss of mercury from
tar and asphalt  surfaces might  be of concern  in  some special cases, we cannot
see that there would be  serious consequences  in  general, because of the very
diffuse nature of this loss.  Perhaps of  most concern would be direct losses to

                                        101

-------
 watersheds,  but we estimate that  such  loss  is relatively small; in addition,
 new highway  construction techniques tend to channelize runoff to minimize
 possible  pollution of  receiving waters.
       In Sector V,  manufacturing and processing, the chlor-alkali industry
 clearly predominates as  the major contributor of mercury emissions.  Further
 consideration of this  technology  and of possible regulatory controls was
 indicated.   Losses from  other manufacturing activities involving mercury (these
 include catalyst,  paint,  pesticide, and Pharmaceuticals manufacture) were
 found to  be  surprisingly low.  The apparent bases for this finding were that
 the size  of  these  manufacturing facilities  is small (allowing better manage-
 ment of emissions), the  quantities of mercury are low, and the high cost of
 the mercury  itself leads to close attention to control of losses.  Nonetheless,
 because the  manufacture  of  mercurials is an essential intermediate step in
 control of mercury throughout the economy,  these technologies were selected
 for study.
      Processing, which included paint formulation, catalyst usage (such as in
 vat dye manufacture and  vinyl chloride production), was dominated by the large
 "other" category.  This  "other,"  or miscellaneous, use was found to consist of
 many small,  diverse uses, often by small manufacturers or processors; conse-
 quently losses are very  difficult to trace.  However,  it can be assumed that
 since such usages are widespread and diverse,  there will be no appreciable
 local concentrations.   Hence no attempt was made to trace final usages or to
 discuss the  technologies involved with such usage.  It may well be that, since
 the "other"  category represents a considerable mercury loss to the environment,
 a  toxic  hazard label might be required for sale of mercury and its compounds to
 such consumers.  Such a label would state the hazardous nature of the material
 and emphasize the importance of proper disposal of wastes.
      Electrical manufacturing,  which includes tubes,  switches,  lamps, and
 batteries, releases relatively small quantities of mercury to the environment
 in the manufacturing process.  (A  great deal of internal recycling is charac-
 teristic of these industries.)   Nevertheless,  because  these industries repre-
 sent a major  source of  final disposal  losses,  it was decided to look further
at their technologies and perhaps  to determine if regulatory actions might be
                                       102

-------
taken at the manufacturing level to reduce the quantities of mercury which
ultimately enter the environment from these products.
     In Sectors VIIIA and VIIIB, the final consumption of various electrical
products accounts for very large losses in final dispoal—primarily to land.
Hence these uses were lumped together under final disposal,  which is a primary
concern.
     Pesticide usage accounts for rather substantial losses, particularly in
nonagricultural applications, and considerable quantities of mercurial pesti-
cides may move to receiving waters.  Both agricultural and nonagricultural
applications were considered in the technology assessments.
     Laboratory usage, while relatively small, was believed to be another
point where changes in technology might well reduce the quantities of mercury
emitted to the environment.  Therefore, laboratories were also considered in
the technology assessments.
     Pharmaceutical usage represents fairly substantial losses of mercury to
water.  This usage is already slight, and demand for mercurial drugs seems to
be dropping as new and equally  effective substitutes are discovered.  Even so,
it seemed advisable to consider trends  in pharmaceutical usage in the technology
assessment.
     The use of mercurials  in paint is  decreasing,  and an EPA action is  now
under way to curtail  this usage even more.  However,  this action has drawn
strong criticism  from the paint industry and  others.   Hence the  technology  of
paint and the  implications  of regulatory  actions were considered.
                                        103

-------
                                                                           IV
                                                        TECHNOLOGY ASSESSMENT

INTRODUCTION

     The DRS inventory identified a number  of man-related  sources of mercury
which can and do enter the environment.  All of  these  sources can be identified
in terms of an SIC category.   Each of these industries (including those  in the
trades and services) can be identified with respect  to a technology or process
involved.  In this section these technologies are  described, with emphasis on
the mercury emissions that accompany the technology  and on changes foreseen in
the technology by the year 1983.
     Technologies are discussed by sector,  as described in Section III of the
report.  A detailed technology assessment is reported only for  major contribu-
tors for which regulatory controls might prove advantageous.  Because the con-
cern here is focused on loss of mercury to the  environment,  the technologies
of the individual industries are described only  in general terms and for a
typical facility or case.  (Minor technologies  are generally not discussed.)
     Composite technologies are described by use category.  This presentation
considers those industries and users of mercury that are  found  in  several  sec-
tors and possibly in disparate geographical locations, but that are  related to
a single use category.  Paint is an example of such an industry:  phenylmercuric
acetate  (PMA) as a preservative is discussed under mercurial compound manufac-
turing, and the losses in a typical facility are described.  But PMA is used by
many paint formulators across the nation, and additional losses occur in the
formulation process.  Finally, paint is sold to and used by a large segment
of the population, and the emission characteristics for such usage are, again,
different.  Hence, in this presentation the overall losses as of 1973 at each    ,
of these various steps, for the entire paint industry, are aggregated to clearly
identify the point of major losses, and to  indicate where control measures might
be of value.
     Finally, 1973 and 1983 technologies and emissions are  compared,  to see what
possible requirements for regulatory  actions may  arise.   Since the  concern in
                                        105

-------
  this entire report section is to provide an overview of many industries and
  many emitters,  a broadbrush approach must be taken.
       In the discussion of technology by facility, the background of the indus-
  try of interest is given,  including  its relative size and importance in the
  economy,  its general location,  and regulatory status with respect to mercury
  (if any).   Pertinent references are  cited to give additional information to
  the interested  reader.
       Facility sites are also discussed.  For mining and manufacturing,  where
  the number  of sites is  relatively small and where information is available,
  tabular data are given, including name  of company, location (county and state),
  estimated production or output,  and  type of process used.  While these data
  are generally reliable, many have been  derived from a variety of sources whose
  agreement often seems to be  merely fortuitous.  In many cases,  particularly
  with respect to production and  processes, the figures are best estimates.
       Next, the  1973  (current) technology is shown in a simplified flow sheet
 which indicates  only the basic  process  or technology most commonly found in
  that  industry—that  is, a  typical facility.  Superimposed on this technology
 is a best estimate of mercury input and losses throughout the process.   For
 simplification,   and  to allow the  reader to  readily compute fractional losses
 throughout the process, we have used a  nominal input such as 100 or 1,000.
 These inputs are tied in each case either to a specific plant size or to a
 specific usage value.  Outputs always equal inputs;  no unaccounted losses have
 been recognized, even though such losses are frequently encountered in the real
 world.
      The 1983 (projected)  technology discussion presents a similar flow sheet
 but includes our best estimate of changes that may occur during the 1973-1983
 period and of the possible effects of emissions of mercury to the environment.
 Often the changes are less in the technology than in the pollution control
 which is exercised.   At best it is difficult to forecast technological changes.
 Therefore we  have generally adopted the  procedure of considering that the
 latest and best  technology available  (that  is,  in 1974)  in terms of minimizing
mercury losses or emissions is that which can and will be generally utilized
in 1983.  This may be a  bit optimistic,  but it is equally possible that
                                       106

-------
regulatory requirements and technological advances may result  in even greater
reductions than those indicated.
     Alternative processes or products which could be  substituted to  either
reduce or eliminate the use of mercury are discussed briefly.   In most cases,
unless the transition is already under way (as  in pesticides),  alternative
technologies which are only now appearing would be unlikely to have much impact
by 1983.  However, where a novel or interesting technology has been encountered,
it is mentioned.
     In a final subsection the estimated emissions for 1973 and 1983  by all
users of a particular technology are presented.   In  some  instances, high and
low forecasts are included to account for variations which are foreseeable.
Usually the 1983 estimates do not take into account  market increases,  since
these would be small in comparison with errors  which otherwise accompany the
technological forecast.  In seme cases, the values for the 1973 emissions may
differ slightly from those derived by the URS inventory (see Section  III).
These differences usually reflect modifications based  upon better information
obtained after the completion of the inventory* but  in no way affect  the over-
all findings of the report.

MICROTECHNOLOGIES

Sector I - Mercury Input

     Mercury Mining and Refining  (SIC  1092)

     Cinnabar ores  (which have an average mercuric  sulfide content of less  than
5 pounds per ton  of ore) are refined by  heating  the ore  in a  furnace and condens-
ing the mercury vapors that are released.  Mining and refining operations are
nearly always adjacent to each other.  Major mercury-containing ore  bodies  are
found primarily in the Western states, although  some  deposits are found else-
where.  At present, mining is largely concentrated  in California and Nevada.
     Domestic mercury production  has  plummeted from a high of 29,640 flasks in
1969 to 2,131 flasks in 1973  (Ref.  37).   This  decrease is primarily attributable
                                       107

-------
 to the imposition of regulations on the quantity of emissions allowed and to
 the general decrease in the price of mercury, from a high of about $600 per
 flask to around $250 per flask.  As will be discussed later,  the introduction
 of new technology and increased economic incentives could rapidly reverse this
 picture.  Industrial consumption of mercury decreased somewhat over the 1969-
 1973 period, but there is still a gap between domestic production and demand
 which has been filled by imports from other producing nations.  Secondary
 mercury (from recycling)* and release of mercury from stockpiles are also
 important supply sources.
      The U.S.  Geological Survey estimates that mercury is one of the scarcest
 domestic natural mineral resources (Ref. 39).  It is estimated that domestic
 resources can supply approximately 10 percent to 35 percent of the minimum
 anticipated cumulative domestic demand through the year 2000.   However,  this
 estimate is price-sensitive,  and the same source indicates that if the value
 of mercury were to rise sharply,  the minable reserves would almost triple.
      In April  1973,  EPA promulgated regulations permitting the release of no
 more  than 2.3  kg (5.1 Ibs)  of mercury per 24 hours for any size of operation.
 For the mining and refining industry as a. whole,  it has been estimated that
 2  percent to 3 percent of the mercury is lost as vapor from stack emissions
 (Ref.  13).
      In addition to the publications noted above,  additional in-depth informa-
 tion  on mining history and  development,  particularly for the State of California,
 can be  found in Refs.  40  and  41.
      Production Sites.  Table 18  lists the U.S.  mercury mines which were active
 in 1973 and  their  estimated production rates.   Since production rates were at
 an all-time  low in that year,  it  is  obvious that a number of mines were shut
* The mercury reprocessing industry has not been identified as  an important
  contributor of mercury to the environment and  therefore has not been con-
  sidered in detail in this study.  Clark and Fulkerson (Ref. 38)  conducted
  a survey in 1968-1970 of this industry and their  report provides some
  interesting and definitive information on the  location of reprocessors  and
  the source of scrap.
                                      108

-------
             TABLE 18. - Production of U.S. Mercury Mines in  1973
Mine
County
State
Estimated Annual
   Production
     (flasks)
New Almaden
Guadalupe
Chilino Valley
Corona
Oat Hill
Manhattan-One-Shot
Mt. Jackson
Culver- Baer
Abbott
Cardero
Ruja
Study Butte
Alice and Bessie
Total
Santa Clara
Santa Clara
Marin
Napa
Napa
Napa
Sonoma
Sonoma
Lake
Humboldt
Humboldt
Brewster
Kuskokwim

California |
California 1
California
California \
California I
California )
California \
California )
California
Nevada \
Nevada *
Texas
Alaska

720

80

110

250

59

678
180
54
2,131
 Source:   Ref.  42.
                                       109

-------
  down;  if the economic climate were  suitable  these, as well as the presently
  operating mines,  could undoubtedly  become more active.  To assess the proper-
  ties which might become active under such a  situation, we compared the largest
  mines  reported in the Bureau of Mines Mineral Yearbook in 1971 with those
  reported in 1966 (Ref.  43).   The  turnover in mines is highly volatile.  Of the
  24  mines producing 100 or  more flasks each in 1966 (five of these produced
  1,000  flasks or more),  only  five  are retained on the list in the 1971 census.
  Of  the 13 producing mines  in 1973,  all were  on the 1971 list, although their
  production in all cases was  greatly reduced.*
      Future production of mercury in the United States is projected by some
  sources  (including the  Bureau of  Mines) to  zemain near its 1973 low point.
 However,  a  very large mine has recently opened in Humboldt County,  Nevada.
 The McDermitt Mine, which will be described  in greater detail below, is expected
 to produce  at least 20, 000 flasks per year of virgin mercury, with a maximum
 capacity of  up to  60,000 flasks per  year.  This large mine will obviously have
 a significant  effect  on both  domestic and world markets.
      1973 Technology.  Figure  22  is  a flow diagram of a typical mercury recovery
 unit in operation  in  1973.**   The ore may be obtained by either surface or
 underground mining and in recent years has contained about 2.3 kg (5 Ibs) of
 mercury per ton of ore.  Dust  from the mining operation is generally a very
 minor problem and since the tonnages are low, emissions from this source are
 trivial (estimated in Ref.  12 to be  less than 0.02 ton annually).   Generally,
 the  ore is crushed and fed from a storage bin into a rotary kiln (in a few
 locations, screen or table concentration is employed).   Hot gases are fed into
 the  lower end of the kiln;  the ore is thus heated to well over 1, 200°F and the
 mercuric  sulfide is decomposed with an efficiency of about 96 percent.  Mercury
 * A small amount  of mercury is recovered as a by-product from a gold mine in
   Eureka County,  Nevada and from a zinc smelter in Monoca,  Pennsylvania.
** Although the size of the  units involved for any given operation may vary
   considerably, the same basic configuration is used in all recovery,
   facilities.
                                       110

-------
 Ore
 from
 Mine
                                 Mercury Vapor, Gases
                                                          Iron or Steel
                                                          Condensers
   Combustion
        Gases
Redwood    ,
Expansion
Chamber
                                                              Liquid Mercury
                                                               Collected in
                                                              Rubber Buckets
                                                              under Water Seal
  Crushing
   House
       Prime Virgin
      Mercury Flask
      34.6 Kg (76 Ibs)
      Net, 99.9% Pure
                                                                            Bottled
                                                           Calcines Dump
              1000 Units Kg
J
-»»
Crush
1000


(5 Ib Hg/ton of ore)
i


K
•In

0.5 |10
Dust
I


Land
989.5

Tailings





Condenser
1
Stack
1
Air
986.5 virgin
Mercury

SOURCE: Ref. 40; URS Research Company.
       Figure 22  FLOW DIAGRAM OF MERCURY RECOVERY USING 1973 TECHNOLOGY
                                                    111

-------
  vapor,  sulfur dioxide,  combustion products, and dust are then passed off.  The
  dust is separated in a  cyclone  unit and the vapors are passed through a simple
  water-cooled iron or steel condenser.   Liquid mercury is collected as shown on
  Figure  22 and is bottled on-site  in 34.4-kg (76-Ib) flasks.  Exhaust gases
  contain sulfur dioxide  and traces of mercury vapor (which are not allowed to
  exceed  2.3 kg per day total discharge,  by EEA standards) and are then discharged
  through a low stack.  The calcined tailings are collected and dumped in a nearby
  waste pile;  they typically contain about 5 ppm of residual mercury (Ref. 40).
  The  tailings do not  appear to present any leaching problems.  The water used
  in the  condensers does  not come in contact with the mercury and is discharged
  or reused.   The mercury exhausted from  the stack can be assumed to remain pri-
  marily  in the air, although there is doubtless some local deposition on the
  land  surface; however,  no  harmful effects from this have been reported.
       In the past,  some  very small operations have relied on the retort,  or
 batch, processing  of the ore.  These operations are less efficient and probably
 are not in use at all today.
      The simplicity of  recovering mercury using the modern technology,  coupled
 with low operating costs when controls were less rigid,  accounts for the many
 small operations that were active in the recent past.   Many marginal operations
 were activated briefly in response to higher prices and were just as quickly
 abandoned when prices declined.
      1983 Technology. Figure 23 is  a flow diagram of a mercury recovery unit
 which can be considered  the prototype for 1983  technology.   This flow chart is
 for the  new McDermitt Mine, which has several unique  features.   The ore,  which
 has an unusually high mercury content (about 4.6 kg per ton),  is obtained from
 an open  pit.*  After  being fed through crushers and a  mill,  it is concentrated
* Added emissions resulting from open pit mining techniques  (as  contrasted
  with the more conventional underground techniques)  are  unlikely.   Emissions
  resulting from dust are trivial and can be easily controlled.  The vaporiza-
  tion of mercury or its compounds from surface  works is  predicted by McCarthy
  (Ref. 18) to be of the same order as vaporization from  the underground
  deposits.
                                       112

-------
         Front End Loader
                          Rock
                         Media
                         Feeder
     Clean Gas
        to
    Atmosphere
        A
          50-ton
           Clay
          Hopper
  Gas
Cooler
        Venturi
  gQ    Scrubber
Scrubber       Mercury
     Impinger Condenser
             Mercury
             Cleaning
                                                                        No. 1 Cleaner Flotation
                                                                               t2£j|B77733ZR
                                                                                 Ulo.

                                                                         -   +  T
                                                                     I  1 a     •-'	
                    2 Cleaner Flotation

                          O'Flo
  Screw    Filter
Conveyors   m
                                6 Hearth
                                 Furnace
                                  Sealed
                                  Calcine
                                                            Stock
                                                            Tank
                                 \ O'Flo   1 Discharge _             ]
   }   U' Flo
   T         Mercury Con
              Thickener
                                   Waste
                            Mercury
                              Bulk
                             Storage
                                                 Solutions
                                        Mercury
                                        Bottling
                                                   Reclaimed Water
                                                       to Mill
                  Tailing to Storage
                      Pond for
                     Decanting &
                     Evaporation
Tailing Dam

1000

Water





Concentrate
i
100
f
Tailing
Pond
i
I
Land
(fixed)
900

Furnace
I
f
Stack

«
r
Air
899



• Maximum discharge of 2.3 kg (5.1 lbs)/day. For a production rate of 2,000 kg/day
  of mercury (equivalent to 20,000 flasks per year), the discharge to air represents a
  loss of  0.1%.

SOURCE: Placer-Amex Corporation; URS Research Company.
                    Figure 23   MERCURY RECOVERY USING 1983 TECHNOLOGY
                                                    113

-------
  by flotation processes to about 60  percent mercury content.  It is then fed at
  a rate of some 2,200 kg (1,000 Ibs) per hour through a six-hearth furnace and
  the calcined tailings are discharged,  through a  seal, to a waste solution system
  and carried to a tailing pond for permanent storage.  The reclaimed water is
  returned to the mill for further use.   (All makeup water comes from underground
  wells.)   The vapor from the furnace goes  through tube condensers that continu-
  ously collect most of the mercury,  which  is then fed through a cleaning process
  to bulk storage (a unique feature of this mine is that it will use 1,000-kg
  shipping containers).
       The vapor is further cleansed  of  mercury and the sulfur dioxide is removed;
  final traces of mercury are removed in a  refrigerated gas cooler (utilizing the
  principle that mercury is very insoluble  in air  at temperatures of less than
  100°F) and are discharged to the atmosphere.  The discharge is designed to
  meet  the EPA standard of 2.3 kg per day maximum.  Although the technology
  used  in  this new  mine  is not yet proven,  it appears to meet high standards in
 minimizing discharges  to the environment.  The cost of the mine and facilities
  construction was  approximately $9 million, according to the developers (Placer-
 Amex Corporation).
      The technology outlined above for 1983 obviously applies only to new
 installations.  Capital  costs  to  control  emissions for installations using
 1973 technology are estimated  to  range from $228,000 to $260,000,  with annual
 operating costs of about one-third of capital costs (Ref. 15).   The authors
 of Ref. 15 conclude that  "... considering the 'fragile1 condition of this
 industry, it is doubtful  if they can sustain even minimal control costs."  They
 also state that break-even production costs average between $300 and $400 per
 flask, which is well above the current average price per flask of about $250.
 Therefore it can be expected that small mercury producers will not be an impor-
 tant factor in the 1983 marketplace.
     Estimated total emissions with high and low values from mercury mining and
 refining  for 1973 and 1983 are shown in Table 19.  Assuming that larger opera-
 tions  are the trend of the future and that stringent regulations will minimize
 the  number of active operations, it is likely that total emissions from mercury
mining and refining will not increase  appreciably by 1983 despite the possibil-
ity of a  10-  to 50-fold increase in production.   However,  if for some reason

                                      114

-------
         TABLE 19.  - Estimated  Total  Production and Mercury Emissions
                     from Mercury Mining and Refining:   1973 and 1983
                                 (kilograms)
Value Range
High
Low

1973
Production
73
73
,600
,600


Emissions
11
8
,000a
,290C

1983
Production
2,000
700
,000
,000


Emissions
21,
12,
150b
700d
Source:  URS Research Company
a. Assuming 2.3 kg/day discharge from each of 13 sites.
b. Assuming 2.3 kg/day discharge from each of 15 sites.
c. Assuming a lower than allowable discharge, and averaging 3 per-
   cent of throughput.
d. Assuming 2.3 kg/day discharge from each of 25 sites.
                                      115

-------
 small operators were to increase appreciably in number, they could contribute
 unduly to the total discharge of the industry (due to the clause in present
 regulations which regulates discharge by a given facility on total quantity,
 not as a fraction of the throughput of the facility).  A simple change in the
 regulations, however, would suffice to correct such a situation were it to
 arise.
      Alternative Technologies.  Some study has been made of the possibility of
 obtaining mercury either by leaching the ore or by electrolytic oxidation of
 sulfide ores (the latter is a project of the U.S. Bureau of Mines).   However,
 these processes have not advanced beyond the pilot stage and do not appear to
 be serious contenders for the forecast period.

 Sector II - Mining

      Copper Mining and Smelting (SIC 1021 and 3331)

      The United States accounts for about one-quarter of world copper produc-
 tion.   Copper is mined in eight states,  primarily in the West,  with Arizona
 alone accounting for almost half of the total U.S.  production.*  Copper is
 produced primarily from low-grade sulfide ores  from open pit mines.   The ores
 are concentrated mechanically,  smelted,  converted to blister copper,  and finally
 electrolytically refined.   Copper is also produced from low-grade copper oxide
 ores  using hydrometallurgical methods which employ leaching,  usually followed
 by electrolytic  refining.   Of major concern are the pyrometallurgical (smelting)
 methods,  since they  represent the most commonly used technology,  utilize ores
 with higher average mercury content,  and in the course of heating and smelting
 emit mercury as  a vapor to 'air.   Therefore the  concern in this section is with
 smelter operation.
* The URS regional inventory indicated that  the copper  industry (mining  and
  smelting) in Arizona accounted  for almost  all of the  mercury emissions to
  air in the state and contributed appreciably to those to  land and water.
                                       116

-------
     Little concrete information was found on the manner  in which  the  mercury
of copper ores is ultimately released to the  environment.   Attempts  to discuss
this problem area with representatives of the industry proved fruitless;  in
general the response was that mercury was not a problem in copper  smelting.
However, a rough estimate of the total quantities of  mercury  likely  to be
released by pyrometallurgical operations is 100 metric tons per  year of mercury
(as Hg).  This value is based on 203,411 metric tons  of ore processed  in  1973,
with an average mercury content of 0.5 ppm (Ref.  87 indicates a  range  for copper
sulfide ores of 0.1 to 40 ppm).
     In recent years the copper industry has invested heavily in pollution
control technology to reduce sulfure dioxide (SO  )  emissions  to  the  environ-
ment.  EEA regulations requiring the industry to  lower SO  discharges  to  air
have also aided in somewhat reducing the mercury  emissions, because  the SO
recovery system removes a small portion of the mercury in the gas  stream.
Since data are not available on this removal rate,  we have made  estimates which
are presented in the technology sections below.
     Production Sites.  Table 20 lists the producer,  location, and estimated
production for copper smelter operations in 1973.  The production values  for
this year are fairly typical, although the industry1s output fluctuates con-
siderably from year to year depending upon demand,  world supply, and labor
relations.
     1973 Technology.  A simplified flow diagram showing estimated losses of
mercury in the various major processes  in the mining and production of copper
is presented in Figure 24.  The input value of 100 could represent the 100
metric tons of mercury estimated to be  present in the copper ores mined  in the
United States in 1973, or it could represent  the flow in a typical  facility.
     Typically, copper ore is  obtained from  large, open-pit  mines and trucked
to the mill.  A small amount of mercury is no doubt  lost in  the dust  generated
from this operation.   (An additional small amount of mercury might  be released
directly to the air as a result  of the increased degassing due  to the exposure
of the mineral deposit; however, as discussed elsewhere,  this increase over
background is small.)  At the  mill the ore is crushed,  ground,  and mixed with
water and selective reagents;  selective flotation concentrates the copper
                                        117

-------
          TABLE 20. - Estimated U.S. Smelter Production of Copper,  1973
 State and Producer
 Location
Thousands of
  Tonnes
 Arizona

   Phelps Dodge
     it      ti
   Magma
   Kennecott
   ASSRC
   Inspiration

 Montana

   Anaconda

 Michigan

   White Pine

 Nevada

   Kennecott

 New Mexico

   Phelps Dodge
   Kennecott

 Utah

   Kennecott

 Texas

   ASSRC

 Washington

   ASSRC

Total
                                791
Morenci
Ajo
Douglas
San Manuel
Hayden
Hayden
Miami
Great Falls
Ontonagon Co,
McGill
Hidalgo
Hurley
Bingham
El Paso
Tacoma
                                118
                                 63
                                 96
                                131
                                244
                                100
                                108
                                                                1,651
Source:  URS and U.S. Bureau of Mines.
                                       118

-------
                                                     Air
                                                                  Air
  (Mercury)

  100
i
                              Land
                              .1".
                              Fly Ash
   Dry Rock
    Mining
                Grinding
                                                     1
                                                  50
                                            Electrostatic
                                            Precipitator
                                    1
3    (Sales)
                                  Reactor
                                                    Cyclone
                                                 Flux
                               Concentration
              Dust
                I
                                             95
                                                  1
           Sulfuric Acid
             (<20ppm
             Mercury)
                                   Stock
                                 (4%S02)
                                                              Flux
                                            Reverberatory
                                              Furnace
                                           (Smelter) 1480° C
                             Tailings
              <1
T
               Air.
SOURCE: URS Research Company.
                              Land
                                                               I
                                 Converter
                                                             10 (Matte)
                                             <1
            Electrolytic
              Refining
                                                                       (Blister Copper)
                                    Slag
               Metallic
               Copper
                                                      Slag
                                                               Land
                   Figure 24   1973 TECHNOLOGY FOR COPPER MINING
                               AND SMELTING (showing mercury input and output)
                                                119

-------
 content in the  froth—the  concentrate—which is then recovered.  The worthless
 residue (gangue), which contains  some mercury, is discharged to a tailing pile.
 The concentrate, which has about  a 20-percent copper content, can then be
 economically moved to the  smelter even if it is not located nearby.
      The copper concentrate is fed into a reverberatory furnace (smelter) which
 is heated  with  hot combustion gases and which is also charged with slag from
 the copper converter and with limestone and siliceous fluxes.  (The use of
 electrical furnaces, another form of smelting, to replace the reverberatory
 furnace  is a rather  recent innovation.)  The operating temperature of about
 1480° Celsius drives off some of  the sulfur, which is oxidized to sulfur
 dioxide  and escapes  with the flue gases.  Excess iron is eliminated in the
 slag.  The  copper matte,  which is periodically withdrawn from the furnace,
 consists of a mixture of copper sulfide and iron sulfide.  In Figure 24 we
 indicate that most of the mercury goes up the stack along with the sulfur
 dioxide, but that a  small amount remains in the slag and some is carried
 forward in the copper matte.   Considering the operating temperature of the
 reverberatory furnace, it is more likely that almost all of the mercury is
 emitted from the stack at this point,  but for want of definitive information,
 the division shown in the figure was used.*
      The stack gases from the reverberatory furnace are approximately 1 percent
 sulfur  dioxide  (SOo)>  this is too dilute for economical production of sulfuric
 acid.   On the  other  hand,  the stack gases from electric furnaces usually con-
 tain  3  percent to  4  percent S02 ,  from which sulfuric acid can be economically
 produced.   The  gases  from the furnace  are passed through cyclonic dust collec-
 tors and electrostatic precipitators.   Thus,  an estimated two-thirds of the
 mercury content  is released to the air (through the typically high stacks used
by copper smelters) and the remaining  one-third is assumed to be caught in the
 fly ash and deposited on  land.
* Older technologies often had a  roasting  step prior to the smelter to reduce
  both the sulfur and iron contents to  desired levels.   Roasting is rarely
  used in newer installations, although it does have the advantage of producing
  a more concentrated sulfur dioxide  stack gas which is more readily converted
  into sulfuric acid.
                                       120

-------
     Prom the smelter the molten matte is transported to  the converter, where
it is treated with air or oxygen which reacts with the copper sulfide to  dis-
charge S02 (in concentrations of about 4 percent out  the  stack).  As part of
the conversion process, iron sulfide is oxidized,  creating additional S02 ,  and
the iron combines with the siliceous flux to form a slag  which contains copper
and other precious metals and is returned to the reverberatory furnace.   The
resultant "blister" copper is 99.5 percent pure and is normally further puri-
fied, usually by electrolytic refining.  It can be assumed that at this point
the mercury content of the copper is virtually  nil, since most of the mercury
was exhausted to the stack during the smelting  stage  in the reverberatory or
electric furnace.  Any remaining mercury in the gases would find its way  into
the sulfuric acid, probably in concentrations of less than 20 ppm.  A minute
quantity of mercury would go out the stack.
     It should be noted that approximately 80 to 90 percent of the mercury
exhausted remains airborne for at least 10 km  downwind and may become part  of
the global burden (Ref. 16).  However, the remaining  10 to  20 percent is  assumed
to be deposited evenly on either land or water surfaces.  At  least half of  this
mercury eventually reaches the watercourse, either directly or by runoff  from
land surfaces  (Ref. 44).
     1983 Technology.  Figure 25 is a flow diagram of the technology which will
probably be in use in  1983 for the pyrometallurgical processing of  copper.
Again, an input of 100 units of mercury  is assumed, since better forecasts are
not available.  The major changes for  1983 will occur in the pollution control
technology associated with the treatment of the stack gases and in the use of
electric furnaces which produce a higher SO. gas  concentration.  In this flow
diagram it is assumed  that the SO2~laden gases from the  smelter and converter
are combined and that  a considerably  larger fraction  of  these gases is then
reacted to form sulfuric acid.  The remainder  of  the  stack gas stream undergoes
further treatment for  recovery of mercury, and discharges to air and land  are
greatly reduced.  Mercury is also recovered from the sulfuric acid  produced,
thus adding to the revenues  of the  facility.
     While it  is impossible  to  describe in detail the technology likely to be
used in 1983 for control of  mercury emissions  and for recovery of  mercury itself.
                                      121

-------
    (Mercury)

       I 100
    Dry Rock
     Mining
                                                           Air
                  Land
                                 Air
                                 1
            Mercury
            Recovery
              (32)
                                          40
    Grinding
                                 Electrostatic
                                  Preciptator
                                                      Cyclone
                                 Concentration
               Dust
                T
                                               95
                                                          84
                                   Electric
                                   Furnace
                Tailings
<1
T
               Land
SOURCE: URS Research Company.
                 Land
                                              10
                  Slag
                                                     Mercury
                                                     Recovery
                                                       (49)
                                    Reactor
                                   Converter
                                     Slag
                                                   Land
                                                                                           H2S04
                                                                                            3  I (Sales)
                                                                                    50
                                                      Purifier
                                                                                    53
                                                                                               53
 Sutfuric
   Acid
Electrolytic
 Refining
 Metallic
 Copper
                    Figure 25   1983 TECHNOLOGY FOR COPPER MINING
                               AND SMELTING (showing mercury input and output)
                                                 122

-------
current literature suggests various possibilities.   For example,  the Gortdrum
mines in County Tipperary,  Ireland, now recover mercury from copper concen-
trates.  However, more interest has been displayed in a technology involving
recovery of mercury from flue gases.  For example,  Ref.  45  suggests the  use
of a countercurrent sulfuric acid wash to recover up to 90  percent of  the
mercury content of the stack gases.  The Outokumpu Oy zinc  plant  at Kokkolla,
Finland, is reportedly recovering 600 flasks per year of mercury  as a  by-
product of the roaster gases in a process which should be adaptable for  copper
smelting (Ref. 46).
     As indicated above, the mercury content of the stack gases may be removed
prior to the sulfuric acid unit.  However, if such technology is  not employed,
mercury will appear in the concentrated sulfuric acid.  This problem has
apparently been  successfully attacked by the Bunker Hill Company,  which  reports
that it has reduced the mercury content of by-product sulfuric acid to less
than 1 ppm at its Coueur d'Alene  (Idaho) smelter (Ref. 47).  Other processes
are reported to be under development for removal (and recovery) of mercury
from by-product  sulfuric acid.
     Alternative Technologies.  Smelting appears to be the preferred route for
extraction of copper sulfide ores.  However, a hydrometallurgical approach has
been proposed by the Power Plate  Corporation of Salt Lake City (Ref.  48).  In
this patented  (but not yet proven)  process, the sulfide ore is mined,  crushed,
ground, concentrated, and then processed  in the Power Plate cell.  "Here, copper
dissolution and  electrodeposition occur simultaneously at high current densi-
ties, and in the presence of ferric sulfate, sulfuric acid, and oxygen."  The
author claims that this process results in a cost  of  $376 per  ton of  installed
capacity per year, which is only  34 percent of the smelter  cost  it replaces.
However, the  likelihood is remote that this process,  even  if proved commercially
feasible, will become an important part of the production  pattern by  1983.
     Implications.  Table  21  shows the estimated total emissions from all U.S.
copper  smelters  to various  receiving media for the year 1973 to 1983.  Drastic
changes are forecast which  are primarily related to the continuing concern of
the copper industry with cleaning up major pollutants,  especially sulfur dioxide.
It is presumed that this concern will coincide with the realization that recovery
                                       123

-------
TABLE 21. - Estimated Emissions from Copper Smelting, 1973 and 1983
                              (tonnes)
    Receiving Medium            1973              1983


    Air                        53a              3a

    Land                       40 (2.6)a        8 (0.15)a

    Water                         (2.7)a          (0.15)a

    Product                     7               4

    Reclaim                     -              85
    Total                     100             100
    a.  Ten percent of emissions to air are expected to
       fall out locally and to be equally distributed
       between land and water.
    Source:  URS Research Company
                               124

-------
of the mercury associated with SO  is economically feasible and will serve to
improve the environment.  If this forecast is correct,  an estimated 85 percent
of the mercury currently lost (primarily to air and land) may be reclaimed for
resale.
     However, it must be noted that a case cannot be made for regulation of
these mercury emitters on the basis of safety or hazard alone.  To assess the
potential airborne hazard which might be represented by each smelter,  isopleths
of mercury vapor concentration were constructed from the results of a simple
Gaussian diffusion model.   (Complete results are shown in Appendix E.)  Three
runs were made; two were of maximum short-term concentrations under worst-case
and average meteorological conditions, and the third was for an 8-hour averag-
ing time under worst-case meteorological conditions.  Pertinent stack informa-
tion was provided by the EPA National Emission Data System (NEDS) Program.  The
assumed mercury content in the stack gas was approximately 167 micrograms per
cubic meter  (jig/m ), corresponding to an average concentration in the ore of
0.5 ppm, all of which was assumed to reach the atmosphere  (i.e., the cyclone
and electrostatic precipitators  did not remove any of the mercury and all of
it was volatilized during the smelting process in the reverberatory or electric
furnace).
     As  shown  in Figure 26, all  three cases  studied  exceed the average back-
ground level for nonmineralized  areas.  However, only the short-term  worst case
exceeded the background level expected  for mineralized  areas such as  would be
found near copper smelters  (~30  ng/m near A jo,  California;  see  Ref.  18).  And
most importantly, even the  maximum recorded level,  at  approximately 2.5 km
downwind, was  approximately l/25th of the EPA designated exposure level for 30
days of  continuous  exposure.  Thus,  in  light of the fact that copper smelters
are  located  in the  very lightly populated areas, neither the quantity of
material emitted nor the population exposed to these somewhat elevated  (above
background)  levels  is actually of great concern.
                                       125

-------

                                                                                  ACGIH* Occupational
                                                                                  Exposure Standard
                                                                                  (8 hours)

                                                                                  EPA Designated
                                                                                  Exposure Level
                                                                                  for 30 days
        75
  c
  o
  c
  8
  o
 o
50
                              Average
                           Meteorological
                                Case
        25
                     Case 1          Case 2
*  American Conference of Government Industrial Hygienists.
SOURCE: URS Research Company.
                                          Case 3
Case 4
                                                                         Average over Copper
                                                                         Mine at
                                                                         Ajo, Arizona

                                                                         Average Background
                                                                         Levels over
                                                                         Nonmineralized Areas
             Figure 26  MAXIMUM MERCURY VAPOR CONCENTRATIONS FOR CASES 1-4
                                                   126

-------
     However,  since it is suspected that continued deposition of  toxic materials
from airborne plumes can lead to an accumulation of such materials  in the soil,*
an estimate of such a cumulative effect was made.   Certainly,  elevated values
for land have been observed around copper smelters (greater than  200  ppb versus
71 ppb for background on similar terrain; see Ref.  29),  which suggests that
buildup may be occurring.  In our estimate (see Appendix E)  the principal
assumptions were that 10 percent of the mercury in the plume reaches  the ground
within a radius of 10 km from the smelter,  remains there,  and accumulates for
a period of 20 years.
     Using these relatively gross assumptions and the average case  shown in
Figure 26, we calculate that the smelter would contribute an additional  111  ppb
to the existing background of 71 ppb.  By changing the assumptions  somewhat  to
assume that 10 percent to 100 percent of the mercury released from  the  smelter
reaches the ground within 10 km, and 20 percent to 100 percent of this  remains
in the soil, the range of mercury concentrations above background could be
from 70 to 890 ppb.  However, even the highest of these values has  not  been
demonstrated to present a health hazard  to people or animals.  (Uptake  by
plants and presumably from there to animals is certainly likely,-  again,  however,
the impact cannot be demonstrated to be  of great significance.)
     Aside from the mercury that settles locally,  the balance of the emissions
from copper smelters is assumed to become part of  the global mercury burden.
Even so,  the quantities  involved for the area bounded by the coterminous United
States are less than 10 percent of all emissions  to  air from man-related
sources,  and are insignificant  in  comparison with degassing emissions.
* Lawsuits have resulted from the identified buildup of lead in vegetation and
  people located downwind of lead smelters.  However, it is most difficult to
  establish the smelter as the cause of these observed effects; hence the use
  of  the word "suspected."
                                       127

-------
 Sector III - Miscellaneous Unregulated Sources

      Power Plants  (SIC 49110 and 4911C)

      The electricity generated by power plants for transmission and distribu-
 tion to consumers is obtained from several energy sources.   Fossil fuels account
 for 81 percent, hydropower for 15 percent, and nuclear power for 4 percent of
 this energy (Ref. 49).  Of these, the only significant unregulated source of
 environmental mercury pollution is fossil fuels.
      Electric power is usually generated by superheating water in a boiler and
 passing the resultant steam through turbine generators.  The three types of
 fossil fuels used to heat the water are coal,  oil (distillates or residuals),
 and natural gas; these fuels account for 44 percent, 16 .percent,  and 21 percent
 respectively of total U.S. electrical power generation in 1973 (Refs.  49, 50).
 Local utilities' choice of fuel usually depends on what is most available and
 most economical in the immediate area.  For example, the Appalachian and Mid-
 west regions depend on coal as the major, fuel, but the Gulf states and the Far
 West depend on petroleum or natural gas.
      The rate at which mercury is emitted to the environment by a power plant
 depends on the size of the plant, the type of fuel,  and the mercury content of
 the fuel.   Mercury content varies according to geographical region and even
 according to layer within the fuel bed.  Appalachian coal,  for example, has an
 average mercury content of 0.2 ppm (range of 1.46 ppm to 0.07 ppm), but coal
 from the mountains of the West has a mercury content of 0.06 ppm (Ref. 51).
 In general,  the national average content of mercury in fossil fuels is as
 follows:-  coal,  0.2 ppm;  distillate oil, 0.066 ppm (Ref.  12); residual oils,
 0.13 ppm (Ref.  10);  natural gas,  0.04 ppm (Ref.  12).
     The cost  of petroleum and natural gas is  escalating because of the deple-
 tion of  domestic reserves and increased dependence on foreign supplies.  There-
 fore, other  fuel sources  are being increasingly utilized for electric power
 generation.  Coal is unquestionably the leading contender for the short term,
because there are abundant reserves in the United States—-estimated at 421
billion tons in 1973 (Ref.  49).   The other possible energy sources are breeder
                                       128

-------
and thermal reactors and oil shale deposits (in Colorado,  Utah,  and Wyoming).
It is difficult at this time to determine what fuel source will  dominate in the
long term.  Environmental constraints,  the long lead times required for research
and development, uncertainties in predicting capital investment  by utilities
or government, and mining constraints all combine to make forecasting in this
area extremely hazardous and unreliable.
     Production Sites.  Table 22 lists power plant emissions by  type of fuel
and by state.  Most large power plants are located near heavily  populated and
industrialized centers.  Plants that use coal and natural gas tend to be very
large facilities (in the megawatt generation range), but petroleum-using plants
vary from small municipal to large facilities.
     Most power plants in the Atlantic region use coal-fired furnaces because
of the rich Appalachian coal deposits.  States that rely primarily on this
energy source are Florida, Georgia, Maryland, New York, North Carolina, Pennsyl-
vania, South Carolina, Virginia, and West Virginia.  The other Atlantic states
balance coal use with petroleum and natural gas.
     In the North Central region, power generation  relies chiefly on coal, with
lesser amounts  of petroleum and natural gas.  The leading coal-using states are
Illinois, Iowa, Kentucky, Michigan, Minnesota, Missouri, Ohio, Tennessee, and
Wisconsin.  In  contrast to the Atlantic region,  the North Central region has
many small municipal utility companies  that rely on diesel power.
     The predominant energy source in the South is  natural gas,  because of the
rich deposits found in the Gulf  states.   Petroleum also supplies  some Southern
power generation needs.  The Mountain  region has a relatively good balance  of
energy sources.  The West derives most of its energy from petroleum and natural
gas, since  large reserves of these resources are available there.
     1973 Technology.  Two  flow  diagrams for typical power plants are shown in
Figure 27.  The only difference  between the two diagrams is in the mercury
emission factors of the  fuel  consumed for electricity generation.  The emission
factors  for coal are 90  percent to air and 10 percent to land (Refs.  51, 52,  53)
Natural  gas and petroleum,  because of their cleaner burning characteristics,
emit 99  percent to air and 0.1 percent to land  (Ref.  54).  The principal reason
for the  high air emissions is the low boiling temperature of mercury—685°F
                                        129

-------
                        TABLE 22. - Power Plant Mercury  Losses  in  the  United States,  1972

                                                   (kilograms)
U)
o
Fossil Fuel Source


Region and State
Atlantic
Connecticut
Delaware
Florida
Georgia
Maine
Maryland
Massachusetts
New Hampshire
New Jersey
New York
North Carolina
Pennsylvania
Rhode Island
South Carolina
Vermont
Virginia
West Virginia
Subtotal
North Central
Illinois
Indiana
Iowa
Kansas
Kentucky
Total Energy
Output
(millions of kwh)

15,619
5,400
67,081
30,204
2,985
27,802
29,327
4,121
30,327
68,133
51,162
96,034
1,284
17,262
163
28,441
48,466
523,811

70,214
53,843
14,652
16,953
49,891

No. of
Plants

9
5
4
6
2
12
0
2
4
22
18
33
2
9
1
15
12
156

28
24
23
8
16
Coal
Mercury
Loss

15
99
1,598
1,029
59
324
0
45
221
1,209
3,063
6,994
56
931
27
323
1,258
17,251

2,165
1,760
1,159
45
4,452

No. of
Plants

17
4
51
10
18
14
33
8
29
44
6
64
7
21
12
11
2
351

48
21
138
111
5
Oil
Mercury
Loss

537
80
1,248
79
107
525
888
44
852
1,632
87
447
52
35
2
470
11
7,096

197
14
10
10
6
Natural Gas
No. of
Plants

3
4
33
6
5
5
18
0
8
18
7
4
1
6
2
3
1
124

18
7
11
9
4
Mercury
Loss

<1
2
138
31
<1
6
6
0
20
61
14
5
<1
20
<1
4
<1
308

59
* f*
15
49
146
8

-------
TABLE 22. - Power Plant Mercury Losses in the United States,
                           (kilograms)
                           (continued)
1972
Fossil Fuel Source


Region and State
North Central
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Ohio
Oklahoma
South Dakota
Tennessee
Wisconsin
Subtotal
South
Alabama
Arkansas
Louisiana
Mississippi
New Mexico
Texas
Subtotal
Mountain
Colorado
Idaho
Montana
Total Energy
Output
(millions of kwh)

57,339
15,752
32,404
7,066
5,495
92,553
25,571
705
40,690
25,096
508,224

39,550
9,539
39,348
12,141
18,192
129,912
248,682

12,204
I
1,245
Coal
No. of
Plants

26
19
16
2
3
42
2
7
9
22
247

8
2
0
8
2
3
23

12
1
2
Mercury
Loss

1,482
854
2,861
30
181
4,738
41
195
3,424
759
24,146

1,742
69
0
1,115
124
267
3,317

154
29
27
No. of
Plants

62
114
61
71
28
47
32
39 •
1
42
820

3
6
31
8
13
57
118

15
3
5
Oil
Mercury
Loss

147
29
•9
6
<1
40
1
5
3
10
487

7
55
21
42
9
21
155

11
<1
<1
Natural Gas
No. of
Plants

11
6
7
5
4
5
25
4
5
14
135

5
14
27
10
10
89
155

14
0
2
Mercury
Loss

48
42
48
40
<1
14
211
3
13
23
719

2
59
310
82
48
1,043
1,544

55
0
1

-------
                   TABLE 22. - Power  Plant Mercury  Losses  in the United States, 1972
                                              (kilograms)
                                              (continued)
Fossil Fuel Source


Region and State
Mountain
Utah
Wyoming
Subtotal
West
Alaska
w Arizona
California
Hawaii
Nevada
Oregon
Washington
Subtotal
Totals
Total Energy
Output
(millions of kwh)

2,145
7,065
22,659

1,271
9,279
90,357
4,756
11,441
55
23
117,182
1,420,558


No. of
Plants

4
2
21

1
5
0
0
2
0
0
8
455

Coal
Mercury
Loss

22
206
438

12
58
0
0
•27
0
0
97
45,249


No. of
Plants

13
19
55

28
9
32
15
10
7
12
113
1,457

Oil
Mercury
Loss

24
2
37

5
28
896
155
2
<1
2
1,088
8,863

Natural Gas
No. of
Plants

4
1
21

11
10
30
0
7
2
0
60
495

Mercury
Loss

3
2
61

10
63
492
0
33
<1
0
598
3,230

Source:  URS Research Company, based on Ref. 49 and "Electrical World - Dictionary of Electric
         Utilities," 1972-1973.

-------
1.4x 1013 BTU
                    Air
     I
   900 Kg
Mercury
in Coal
1,000 Kg

Boiler
                     \
   100 Kg
Landfill
    (a) COAL-BURNING PLANT
3.8 x 1014BTU
     jo.
 Air
  1
                                       999 Kg
Mercury
in Fuel
1000 Kg

Boiler or
Diesel
Engine
     t
  I
                       Natural Gas
                1.3x1015 BTU        Landfill

         (b) PETROLEUM- OR NATURAL GAS-BURNING PLANT

          SOURCE: URS Research Company.


Figure 27   MERCURY LOSSES FROM TYPICAL POWER PLANTS, 1973
             133

-------
  (Ref.' 55).  The air emissions are primarily in the metallic form, with minor
  losses in the oxide form.   Losses to land come from bottom ash, which ranges
  between 4 percent and 20 percent by weight of  coal  (Refs. 54, 56) , and fly ash
  (stack emissions).   To date very little effort has been expended by the govern-
  ment or the utilities to reduce mercury losses from fossil fuel power plants.
       Technology in 1983.  Figure 28 is  a flow  diagram for the partial recovery
  of mercury from power plant operations.  These diagrams, which represent proto-
  types for 1983, are patterned after a sulfur dioxide  (SO2) scrubber pilot plant
  operated by the Tennessee  Valley Authority (TVA).  The Colbert plant, which
  burns high sulfur coal,  was equipped with a S02 scrubber to remove the sulfur
  gases from stack discharges, but preliminary tests conducted by the EPA found
  that it also removed about 33 percent of the stack mercury discharges (Ref. 57).
  The  reasons for the high mercury recovery are  that the scrubber removes the
  fly  ash (total  emission  0.7 kg mercury  per 10   kg ash) and that heavier mer-
  cury vapor particles are found in the flue gas  (total emission 0.08 kg mer-
  cury per 10  kg gas)  (Ref.  10).
                                                   2
       Because  of increased  constraints placed on SO  emissions to meet clean air
  standards,  scrubbers probably will be installed on most fossil fuel power plants
 by 1983.   This  should result in about a one-third decrease in uncontrolled
 mercury emissions to the atmosphere.  This  quantity of mercury will then accrete
 in controlled landfills.
      Fossil fuel consumption is  expected to  increase about 50 percent between
 1973 and 1983 (at a  3.4-percent  annual  rate  compounded), if current demands
 and usage continue, with coal  increasing 55 percent and natural gas and oil de-
 creasing 5 percent  (Ref. 144).  Mercury  emissions, however, should increase
 only 15 percent for the low prototype technology and should decrease by 14 per-
 cent for highly developed technology.  Mercury emissions are not expected to
 increase at the same rate as fuel consumption because mandated control of S02
 emissions by 1983 will coincidentally lead to removal of part of the mercury.
      Alternative Technologies.  Due to the cost of developing an effective re-
 covery system for mercury alone, it appears that complete mercury recovery from
 stack discharges will not occur in the foreseeable future.  However, fuel sources
which do not release appreciable amounts of mercury to the environment may be
developed before the year 2000.  Nuclear power  (especially breeder reactors) is
                                      134

-------
1.4xltf3 BTU

   J_
                     Air

                      t
                     SO,
                                     603 Kg
                   Scrubber
   Mercury
   in Coal
                        1000 Kg
                            297 Kg
                                 Landfill
                                  Boiler
                            lOOKg
Landfill
        (a) COAL-BURNING PLANT
                     Air
                        669 Kg


3.8 x 1014 BTU
S02
Scrubber

1 Oil

Mercury
in Fuel
1000 Kg

I
Boiler
                            330 Kg
                                 Landfill
                            1Kg
                                 Landfill
     T
                    Natural Gas

             1.3x1015  BTU


          (b) PETROLEUM- OR NATURAL GAS-BURNING PLANT

         SOURCE: URS Research Company, based on TV A Colbert plant process.



Figure 28  MERCURY LOSSES FROM TYPICAL POWER PLANTS, 1983
                  135

-------
 an  excellent  long-term possibility for supplementing fossil fuels.
 A minor development is that mercury is used in sterilizing the
 control rods  in  the reactors  (Ref. 58).  However, the chief constraints
 to  full development of nuclear power plants are other environmental
 considerations*—radioactive waste disposal, mine and smelter tailing
 wastes, plant safety, thermal heat dissipation, and so on.
      Another  alternative to fossil fuel usage is geothermal power.
 Geothermal steam has proved to be an available energy source (albeit
 small to date) in the United States.  Mercury is released along
 with the spent steam, but the amounts are insignificant.  Like
 most of the alternatives, constraints other than mercury content
 may limit the  full development of this energy source.  Also,
 thermal energy is not evenly distributed across the United States.
 Solar energy also offers an alternative to fossil fuels.  The
 full development of this energy source, however, will require
 large expenditures for research and development.  Other energy
 sources for electric power plants are possible, but are not expected
 to materialize in the foreseeable future.
      The energy sources most likely to be used to alleviate the
 present shortage appear to be shale oil and coal.  These materials,
 which are  found abundantly in the United States, can be treated
 by a variety of processes to yield synthetic crude oil or synthetic
 natural gas.   Although both materials contain mercury, no informa-
 tion is available on the  fate of mercury in shale oil or coal
 which is converted to a synthetic product.   However, based upon
 our  knowledge  of the  candidate processes, we can make some very
 preliminary estimates  about the final disposition of such mercury
 to the  environment.
     The processes  for recovering product from shale oil or coal
 involve heating to relatively high temperatures.   Since mercury
boils at about 360°C  (and its  compounds  either boil or decompose
at or below this  temperature) mercurials are likely to be found
                                      136

-------
increasingly in the gaseous phase as the process  temperature
increases.  For the recovery of oil from oil shale,  some  of the
older processes (such as the U.S. Bureau of Mines gas combustion
retort) which operate at over 900°C should certainly drive most
of the mercury off with the combustion gases.  However, some
more recent processes, such as TOSCO II, operate  at much  lower
temperatures (480°C), so that considerably more mercury is likely
to be found in the residual wastes.  Nevertheless, the presence
of mercurials in waste from shale oil processing  is of minor
concern compared with the major problem of disposing of the vast
quantities of wastes themselves.  If oil is a major processing
product, the likelihood that mercury will concentrate therein
is fairly small, but if a synthetic natural gas  (SNG) is  a major
product, the probability of mercury occurrence therein is consider-
ably higher.
     Some consideration has also been given to in_ situ retorting
of deep-lying beds of shale.   It appears unlikely that the mercury
content would concentrate in the resulting oil product,  and the
mercury that did appear there  would probably be vaporized with
stack gases in subsequent refining processes.
     Three general classes of  coal  liquefaction processes typify
the manner in which mercury might be distributed to the environment.
In pyrolysis processes, pulverized  coal is  heated at atmospheric
or low pressures to release  its oil content.   In several  such
processes  (for instance that developed by FMC Corporation) tempera-
tures above 800°C  are reached; this suggests that the mercury
content is almost  totally released, probably ending up in the
gaseous fraction  (which is a low heating  value SNG). The TOSCOAL
retorting process  operates at around 500  C, so considerably more
mercury may remain in the oil product or  in the  residual matter.
                                       137

-------
       The direct hydrogenation of coal occurs  at rather low tempera-
  tures (450 C)  and at very high pressures  (greater than 2,000
  psi).  Mercury would probably concentrate in  the oil product,
  and to a lesser degree in residual matter, with little appearing
  in lighter fractions.  In the solvent extraction of coal the
  temperatures are low (somewhere above 400°C)  and pressures are
  moderate to high.   Thus little mercury is likely to be found
  in off-gases.   Mercury and its compounds  are  unlikely to be vapor-
  ized  and as a  result, most of the mercury content could remain
  in the waste material.
       The removal of sulfur from solid fossil  fuel materials is a part
  of the developing  technology described above.  Mercury can be expected
  to "accompany" sulfur and may be extracted with it.  While the recovery
  of mercury from such processing does  not  appear feasible at this time,
  it should be noted that mercury is now being  recovered from the sulfur
  residues from  at least  one copper mining  operation (see discussion
  under  Copper Smelting Technology).  It is certainly not inconceivable
  that,  if extensive  development of shale and coal resources occurs,
  the recovery of  the mercury content may become economically justifiable.
  It is most unlikely, however,  that  this would happen in the period
 prior to  1983  (see  Ref. 145) .

 Sector V - Manufacturing and Processing

      Chlor-Alkali Manufacturing  (SIC 2812)

      In 1973 the chlor«-alkali manufacturing industry consumed 13,070
 flasks (450,967 kg), or 24 percent, of the mercury used in the United
 States (Ref. 59).  According to a report by Versar, Inc.  (Ref. 60),
 the loss of mercury to land was approximately 7,834 flasks, or 60
percent of the  mercury used by this industry.   The largest loss,
6,590  flasks, occurred when the chlor-alkali sludges  were deposited
in  industrial landfills.  The remaining environmental mercury losses
were to air  and water,  representing 672 flasks and 125 flasks
                                      138

-------
respectively.  The remaining 5,683 flasks of mercury consumed by the industry
were used in plant expansions,  system buildups,  inventories,  etc.
     The production of chlorine gas (C12) and caustic alkali  (called caustic
soda or sodium hydroxide) is accomplished using the mercury cell,  which was
discovered in 1892.  However, the diaphragm cell plant (the diaphragm cell
process filters a salt brine and then electrolyzes the brine  without mercury
to form chlorine and caustic alkali) dominated the world's chlor-alkali indus-
try until the late 1930s.  In that decade rapid progress in mercury cell tech-
nology was made by I. G. Parbenindustrie of Germany because of the growing
demands for rayon-grade caustic.  It was not until the end of World War II,
however, that other nations realized the technical superiority of the mercury
cell over the diaphragm cell.  In the United States, the diaphragm cell process
still dominates, but since 1945 mercury  cell plants have increased.  In 1945,
mercury cells accounted for only 4 percent of the chlorine production in the
United States; by 1959, chlorine production using the mercury cell had increased
to 19 percent of the total, and by 1973 this figure had increased to 24 percent
(Ref. 61).  Because of recent problems with mercury discharges to the environ-
ment, however, it is not certain how much the mercury cell process will grow.
The key to such growth will be the ability of the industry to control mercury
emissions.
     Mercury cell production of chlorine and caustic alkali is accomplished
using an electrolyzer and a decomposer.  In the electrolyzer a sodium chloride
solution  (brine) is subjected to an electric current, making use of an  anode
(positive pole) and a flowing mercury cathode  (negative pole).  Chlorine  gas is
evolved and collected at the anode, and a liquid alkali metal amalgam (NaHg)
is formed with the flowing mercury cathode.  The alkali metal is transported
to a decomposer where the amalgam reacts with  water to form  caustic (NaOH)  and
hydrogen gas  (from which excess mercury is  removed);  the  purified gas is
normally flared.
     Production Sites.   Fifteen companies in the United States used the mercury
cell to manufacture chlorine in 1973.  Of these.  Allied Chemical Corporation,
                                        139

-------
 Olin Corporation,  Stauffer Chemical Company, Diamond Shamrock Corporation,  and
 PPG  Industries,  Inc. accounted  for approximately 62 percent of the chlor-alkali
 production using the mercury cell process  (Ref. 62).  Table 23 lists the loca-
 tion of mercury  cell plants by  state and gives their approximate mercury losses
 to the environment in  1973.*  Thirty-eight percent of the mercury loss in
 chlorine manufacturing occurred in the Atlantic region, 20 percent in the North
 Central region,  and 38 percent  in the South.  Five states account for 54 per-
 cent of the mercury loss:  New  York (8 percent), Kentucky (8 percent),  Alabama
 (13  percent), Louisiana  (16 percent), and Texas (9 percent).
       Most of the chlorine produced in Alabama,  Georgia, Maine, North Carolina,
 Washington, West Virginia, and Wisconsin is used in paper production.  The
 chlorine produced  in the other  states is used by the industrial chemical indus-
 try,   except that in Louisiana chlorine sales are balanced between paper produc-
 tion and industrial chemicals.
      Chlor-alkali plants are generally found on navigable rivers or near ports
 because chlorine is most safely and economically transported by barge.   However,
 some  users  (for example paper and pulp producers)  are located in rather remote
 areas, and since transporting chlorine for great distances is not practice1,
 small chlor-alkali  plants,  generally with mercury cell technology,  are found
 in such scattered locations.**   Larger plants (with more than 500 tons/day
 capacity)  are more  efficient  and are built wherever demand and transportation
 factors permit.
      As shown in  Table  23, there are 28 plants  in the United States that manu-
 facture chlorine  using  mercury cells.   The types of mercury cell used by these
 * The land losses shown  in  this  table  are  due to sludge only.   Sludge losses
   are singled out because the remaining  losses,  representing approximately
   2,035 flasks, cannot be treated and  remain fixed through 1983.
** The direct production  of  50 percent  caustic which is  also used by the paper
   industry from mercury  cells probably accounts  for the popularity of the
   mercury cell process in these  remote locations.   The  absence  of  salt in  the
   caustic, which was at  one time the major advantage of mercury cell technology,
   does not apply in this case nor for  that matter,  in any  but a very few other
   chemical processes.  (The production of  rayon,  once the  major user,  is no
   longer an important factor.)

                                        140

-------
   TABLE 23.  - United States  Chlor-Alkali Manufacturers  and Estimated
                     Mercury Losses of Plants, by  State, 1973
                                            Annual
                                           Estimated
                                        Chlorine Prod.
Estimated Mercury Losses to the
       Environment (kg)
Region and State
Atlantic
Delaware
Georgia

Maine
New Jersey
New York


North Carolina
West Virginia

Subtotals
North Central
Illinois
Kentucky

Ohio
Tennessee
Wisconsin
Subtotals
South
Alabama



Louisiana


Texas

Subtotals
West
Washington

Subtotals
U.S. Total
Company

Diamond
Olin
Allied
Sobin
Linden
Hooker
Olin
Allied
Allied
Allied
PPG


Monsanto
Goodrich
Pennwalt
Sobin
Olin
BASF


Stauffer
Olin
Diamond
Diamond
PPG
Stauffer
BASF
Diamond
Al. Co. of Am.


Georgia- Pac.
Weyerhaeuser

28 Plants

(thousands of tons)

133
93
63
71
154
56
56
56
56
176
56
970

84
100
104
33
124
57
502

116
70
70
71
218
167
48
56
157
973

45
88
133
2,578

Land9

8,385
5,776
5,611
3,778
15,059
5,171
5,171
6,895
9 »— •*
3,867
20,441
5,110
85,264

5,990
8,674
9,021
2,120
16,004
2,966
44,775

7,897
8,017
4,523
8,137
17,838
12,538
5,240
7,621
13,910
85,721

3,624
7,671
11,295
227,055

Air

827
827
827
827
827
827
827
827
827
827
827
9,097

827
827
827
827
827
827
4,962

827
827
827
827
827
827
827
827
827
7,443

827
827
1,654
23,156

Water

235
150
90
115
280
80
80
80
80
325
80
1,595

130
170
175
50
220
80
825

200
110
110
115
415
310
55
80
285
1,680

55
140
195
4,295

Total

9,447
6,753
6,528
4,720
16,166
6,078
6,078
7,802
4,774
21,593
6,017
95,956

6,947
9,671
10,023
2,997
17,051
3,873
50,562

8,924
8,954
5,460
9,079
19,080
13,675
6,122
8,528
15,022
94,844

4,506
8,638
13,144
254,506

a. Land losses included in this table are sludge only and do not include other mercury waste mate-
   rials such as filter cake,  anode graphite, etc.

Source:  URS Research Company; Ref. 60, 62, 63.
                                              141

-------
  manufacturers are the DeNora (13 plants),  Olin (6 plants), Solvay  (4 plants),
  and Uhde (5 plants)  (see Ref.  63).
       1973 Technology.   Figure  29 is a flow diagram of a "typical" chlorine and
  caustic soda plant which manufactures 500  tons of chlorine gas and 570 tons
  of caustic per day.   The basic unit in the plant is the mercury cell, which
  theoretically recirculates the same mercury continuously.  However, mercury
  losses do occur from this cell,  and makeup mercury must be added occasionally.
  The Uhde cell,  which may contain 2,651 kg  of mercury, is typically  2.0 meters
  by 17.6 meters  and is constructed of heavy gage steel to which hard rubber-
  covered channels are bolted to form the sides.  The bottom has a slight slope
  (4 to  5 millimeters/meter)  so  that  mercury, which is introduced at the top of
  the cell through an  end box seal, flows uniformly down the flat bottom of the
  cell to a depth of about 0.003 meter and is removed (with an amalgam content
  of less than  0.5 percent sodium)  through an outlet end box.  Together with the
  mercury,  purified brine is  introduced at the top of the cell, covering the
  flowing mercury to a depth  ranging  from less than 0.03 meter at the top of the
 cell to almost  0.12  meter at the bottom of  the cell.  Multiple anodes extend
 down from the cell cover  (which may be  attached to the body of the cell with a
 flexible gasket  or with a firm connection).  A high-density,  low-voltage current
                      2
  (less than 0.5 amp/cm ), which is carried by massive bus bars from cell to cell
 in series connection, activates these anodes.   The anodes then electrolyze the
 brine to form chlorine, which  is collected at the top of the cell or at slots
 in the anode.  The sodium amalgam is deposited in the moving mercury cathode.
 After leaving the end box, this amalgam enters a decomposer (also called a
 denuder) which is typically affixed to the end of the cell.
      The decomposer is a small reactor that is normally packed with pieces of
 graphite which,  in the presence of trickling water,  cause the sodium amalgam
 to  decompose,  forming 40- to 60- percent caustic and releasing the mercury for
 return  to the inlet of the cell.   Hydrogen is  also formed (99.9 percent pure).
 Its temperature  of over 80°C allows  mercury vapor  to be incorporated in the
gas.  To minimize losses,  the hydrogen is cooled at  the decomposer to about
55°C; this removes over 98  percent of the mercury, which is then returned to
the cell.  The remainder of the hydrogen stream is further  purified so that
                                       142

-------
Mercury
Input
10,000
Air Lai
< • 0)'
0) ^
^ in
If) CO
t"~ »••
id Wai
1 >
CO
^ Electrolvzer
Kg «*»>
T
Return System
Air
'? Kg


276

t.
Returr

Brine Saturator
and Purifier

Return
ter
System
	 	 1


i System

79 Kg
Hydrogen » Air
i
231 Kg Land
1 *" La0d |2709Kc
Decompo!
and
Cooler


r 2627 Kg
System Buildups,
Theft, etc.





er _
Caustic
* Purifier
I 200 Kg
Product
8 Kg 1 36 Kg
 Land      Water

SOURCE: URS Research Company.
   Figure 29  MERCURY LOSSES FROM A TYPICAL CHLOR-ALKALI
            MANUFACTURING PLANT, 1973
                              143

-------
  99 percent or more of the mercury is ultimately removed.  The gas is then
  normally flared.
       The brine which leaves the cell is depleted of its sodium chloride con-
  tent and it carries some mercury with it.  This depleted brine is returned to
  the brine saturator and is treated to remove  undesirable chemicals (primarily
  calcium and magnesium salts) which are introduced in that operation.  This
  purification step results in the formation of sludges which incorporate signifi-
  cant quantities of mercury.   These sludges must be segregated for further treat-
  ment and disposal.   The purified and saturated brine is recycled to the cell.
       A materials balance for mercury in a  "typical" chlor-alkali plant is shown
  in Table 24.   With the 1973  technology,  mercury consumption is assumed to be
  0.4 pound of mercury per ton of chlorine produced, or 31,780 kg per year for
  the 500 tons/day plant.   Table  24 shows the environmental mercury losses of
  the "typical" plant by source and form of  emission.
       In this  materials balance  certain assumptions have been made to account
  for all mercury losses.   For example,  a buildup of the mercury "inventory"
  within  each cell occurs over time as  a result of changing operating conditions;
  these may include the  formation of a  mercury  "butter" which effectively reduces
  the mercury available  for reaction temporarily,  or the need to operate at higher
  current  densities to increase production, which  requires a higher mercury level,
 etc.  Mercury also accumulates  in piping and  other equipment and is an apparent
 loss.  Other poorly defined  losses, for example  theft,  can occur.  We have made
 estimates to cover all known or  suspected losses so that the amount of mercury
 introduced (i.e., 0.4 Ib per ton of chlorine produced)  is accounted for.
 Actually, such accounting is not possible in a typical mercury cell chlor-
 alkali plant.   Instead, for the industry as a whole,  only about 50 percent of
 its annual mercury consumption can be accounted for.   This does not imply that
 mercury is indiscriminately lost to the environment;  rather,  it is most diffi-
 cult to estimate where mercury may accumulate in the system and to what extent.
 For example,  we estimate that each year, in this typical plant,  523 kg of
 mercury  enter structural members and can only be recovered when the plant is
disassembled after its active lifetime.  The  industry  is concerned with this
lack  of  accountability and is continuing to seek better results.   Many plants
now run  radioactive  inventories  of the mercury content of each cell at least
                                        144

-------
TABLE 24. - Materials Balance for Chlor-Alkali Plant (500  tons of chlorine per  day)
                with a Mercury Efficiency of 0.40  Pound per Ton Chlorine, 1973
                                            (kilograms)



Mercury Remaining in
Process Loop
Cell

Stored Inventory
Uncontrolled Traps

Equipment Holdups
. Gases

. Liquids

Within the Brine

Mercury Consumed dur-
ing Manufacture
Product
. Caustic Soda &
Sodium
Hypochlorate
. Chlorine
Air
. End Boxes \
. Hydrogen ft |
Waste Air )
. Cell Room
Water
Solids
. Sludges
. Filter Cake
. Anode 6 De-
composer
Graphite
. Misc. Solids
Other
. Recycling
System
. Mercury Flask
. Theft
Long-Term Structural

Subtotal (kg)
(percent)
total (31,780 kg)
(percent)
Plant Losses
Air
xjsses























350
455










22




827
(2.6)
Water
Losses

























276














276
(0.9)
Land
Losses



























14,695
61


895
3,541







19,192
(60.3)
Other Losses
Product



















636
1



















637
(2.0)
Theft





































184


184
(0.6)
HLwa










































21,116
(66.4)
Plant Inventory
System
Buildups


(265,227)
8,859
(9,171)
(22V)
65

(2,724)
783
(1,362)
392'
(136)
0


























' 10, 099

Escapage




































42

(5,312)
523
565
(31.8) | (1.8)
10,664
(33.6)















































Source:  URS Research Company

a.  High-Level Waste

Note:  Numbers in parentheses indicate total mercury in system in previous years.  The number below
      is the amount of mercury added during the study year.  For example, the cell contained
      265,227 kg in 1972 and 271,586 kg in 1973.
                                             145

-------
  once a year,  but this expensive technique is  only accurate to about ±1 per-
  cent.
       Losses to air from a typical chlor-alkali plant occur chiefly at the end
  box,  the electrolyzer cell,  from hydrogen and waste air flaring, and from the
  brine recycling system.   Except for the brine recycling system, air losses are
  assumed to be within EE& requirements of  2,300 grams per day (Ref. 64).  The
  small recycling system losses,  however, occur at  a variety of uncontrolled
  locations within the plant (brine purifier and saturator, leaks at fittings
  and seals,  and settling ponds).
      As a part of good housekeeping practice, the inevitable mercury leaks
  which  occur under the cell house are regularly collected, by hosing with water,
  into  sumps.   The collected mercury is returned to the system and the mercury-
  containing water is  sent to  the  water purification plant where it is joined by
  the supernatant liquid from  the  sludge settling pond.  The collected waters
  are usually treated  with sodium  sulfide,  forming  the very insoluble mercuric
  sulfide which is easily  removed  by filtration prior to discharge to a local
  receiving water body.
      The amount of mercury that  can be discharged to the aquatic environment
 depends on the  type and  size of  the water body.   Proposed EPA regulations
 limit the maximum mercury discharge to 0.735 kg/day for streams, 0.612 kg/day
 for lakes, 12.2 kg/day for estuaries,  and 14.7 kg/day for coastal waters
  (Ref.  65).  The mercury discharge for  the typical plant was conservatively
 assumed to be  0.79 kg/day  (276 kg per year).
      Mercury losses to land are the most significant waste problem for chlor-
 alkali plants.  Most of this waste is  discharged  in sludge (approximately 200
 ppm of mercury) generated by the brine and caustic purifiers.   The sludge is
 concentrated in settling ponds and deposited in local industrial landfills.
 Other  solids with higher mercury levels include spent graphite (anodes and
 decomposer), filter cakes (sludge and water treatment residues),  and miscella-
 neous solids (discarded wood  and plastic cell room flooring,  machinery,  pipes,
 fittings,  etc.).  These solids may be incorporated in the sludge or shipped to
high level  industrial disposal sites.
     Chlorine  gas contains, at most,  trace amounts of mercury (assumed to be
0.005 ppm).  However,  caustic,  even after  treatment,  contains approximately
                                        146

-------
3.7 ppm of mercury.   Most of this  caustic is  shipped out of  the plant to other
manufacturers and is consumed in their products.
     Theft of mercury from traps,  flasks, sumps, and other sources is a problem
at most chlor-alkali plants.   It is believed  that most of this mercury is  sold
to local salvage firms for resale.  Complete  elimination of  this problem is
impossible because of the quantity used at  the plants and the accessibility of
mercury.
     The system buildups shown in  Table 24  indicate the amount of mercury  known
to be in the system the previous year (number in parenthesis) and the unknown
amount of mercury trapped within or added to  that system.  For example, uncol-
lected traps contain 227 kg of mercury on inventory day, but an additional 65
kg are in "miscellaneous" and "other" traps or adhere to equipment.  The system
buildup is not an environmental mercury loss.
     Small amounts of mercury will remain in  empty mercury flasks, because they
are discarded rather than recycled.  We estimated this loss  at approximately
0.0454 kg per flask (3 to 4 drops).  The other loss in this  category, long-term
structural, occurs as mercury vapor or metal  is absorbed into a building's
superstructure.  This loss remains within the plant until the structure is
discarded.
     1983 Technology.  Figure 30 is a flow  diagram  for the technology which
will probably be used in 1983 by the  "typical" chlor-alkali  plant  (500  tons
of chlorine per day).  Table 25 is a  materials balance  for this  facility.
Mercury consumption in chlor-alkali plants  by 1983  is estimated to be reduced
to 0.10 Ib of mercury per ton of chlorine produced,  for a total of 7,945 kg
per year.  This reduction is expected to be brought about by the industry's
efforts to control mercury emissions to conform with federal,  state, and  local
regulations.  Individual plants are expected to have almost complete account-
ability of their mercury consumption, which will enable them to control internal
plant discharges.  Also, technological changes by 1983 are  expected to help
chlor-alkali plants in the recovery and final disposition of mercury waste.
     Mercury losses to air may be  reduced  40 percent, or to 492 kg, for the
typical plant by 1983.  Most  of this  reduction will occur at the  end boxes and
hydrogen streams when the waste gases are  chilled, demisted, and  put through
                                       147

-------
Mercury
Input
L_
2,500
Return
Air
1 Kg



456 H
\ •
Land
Air Land W
ii ii i
O) O)
^ Electrolyzer
Kg (cell)
t t
System Retu
Brine Saturator
and Purifier
Cg |4Kg
Water
Return System
ater
i
«•- 	 1

141 Kg
Hydrogen __». Air
i
78 Kg U"d
1 *" La"d |27Kg
Decompo
and
Cooler

rn System J g27Kg
System Buildups,
Theft, etc.





ser _ 
-------
  TABLE 25.  - Materials Balance for Chlor-Alkali  Plant  (500 tons of chlorine per day)
                  with a Mercury Efficiency of 0.1 Pound Per Ton of  Chlorine,  1983
                                                  (kilograms)



Mercury Remaining in
Process Loop
Cell

Stored Inventory

Uncontrolled Traps

Equipment Holdups
. Gas

. Liquids

Within the Brine

Mercury Consumed dur-
ing Manufacture
Product
. Caustic Soda ft
Sodium
Hypochlorate
. Chlorine
Air
. End Box |
. Hydrogen ft I
Waste Air J
. Cell Room
Water
Solids
. Sludges
. Filter Cake
. Anode ft De-
composer
Graphite
. Misc. Solids
Other
. Recycling
System
. Mercury Flask
. Theft
Long- Term Structural

Subtotal
(Percent)
Total (7,945 kg)


Air
Losses
























35

455










2




492
(6.2)
Plant Losses flee)
Water
Losses



























16














16
(0.2)
Land
Losses






























403 ••-
61 —


249 "

3,541 '







4,254
(53.6)
Other Losses
Product




















75
1






















76
(1.0)
Theft









































19


19
(0.2)
HLW3





























-to a
^to> a


-to a

-to a









Plant Inventory (kg)
System
Buildups


(265,227)
2,703
(4,585)
0
(272)
13

(2,724)
156
(1.362)
79
(136)
0



























2,951
C37.1)

Escapage






































10

(1,326)
127
137
(1.7]
4,857 3,088
(61.2) (38.8)

















































Source:  URS Research Company

a. jpf (High-Level Waste) - Some or all of the final wastes which are greater than 400 ppm and cannot be
   further treated are sent to controlled disposal sites or scavengers.

N6te:  Numbers in parentheses indicate total mercury in system in previous years.  The number below is the
      amount of mercury added during the study year.  For example, the cell contained 265,277 kg in 1982
      but 267,930 kg in 1983.
                                                 149

-------
  microsieve filters.   Also,  with tighter  internal plant controls on mercury
  consumption,  losses  in the  recycling system will be reduced.
       Losses to water will be  reduced substantially because of waste water puri-
  fication technology  advances.   This  advancement will be brought about by govern-
  ment regulations  requiring  industries to meet best practical waste treatment
  standards by  1977 and best  available waste treatment standards by 1983.  There-
  fore,  the mercury discharge for the  typical plant should be approximately
  0.05 kg/day,  or 16 kg per year.
       The major changes in the  1983 technology occur in pollution control tech-
  nology associated with the  treatment of  sludges.  Two attractive processes are
  now  under development for recovery of mercury from sludges.  One, developed
  and  patented by Georgia-Pacific, involves retorting the sludge with condensa-
  tion of  the released  mercury.  The other, which is reported in the literature,
  involves  the dissolution  of the mercury in the sludge in a solution of sodium
 hypochlorite,  with possible further purification steps and subsequent return
 to the system.  It is  forecast that  some high level wastes will continue to be
 generated; these will continue to be  segregated in special industrial reposi-
 tories.  It is expected that sludge treatment will reduce the mercury entering
 the environment to 403 kg in 1983 for the typical plant.
      Other significant land loss reductions (Table 25)  will occur in decomposer
 graphite emissions.   (Anode losses are expected to be eliminated by 1983 as the
 industry completes the switch from the graphite anode to the dimensionally
 stable anode.)   Also,  reductions in decomposer graphite discharges to the
 environment are likely to result from adoption of good housekeeping practices
 such  as treating the  waste graphite with hypochlorite solution.
      Mercury losses to product (almost entirely caustic soda) will decrease
 significantly.   This  reduction will take place at the caustic purifier where
 the mercury enters the sludge  system for treatment.   Technological changes are
 expected  to reduce the residual mercury content of caustic alkali to 0.4 ppm
by 1983.
                    »
      Losses  due to theft,  systems buildups, and miscellaneous escapage will
decrease by  1983 because more  attention will be concentrated on locating these
sources and eliminating them.  Also,  leaks will be fixed  promptly,  traps will
                                       150

-------
be collected regularly,  and plant structures will be sealed properly.   Generally,
the attitude will change from moderate concern to major concern with controlling
internal plant mercury consumption.
     Estijnated mercury losses from a typical chlor-alkali plant's operation in
1973 and 1983 are compared in Table  26.   The 1983 technology values are based
on a mercury consumption rate of 0.1 Ib per ton of chlorine produced and  no
expansion or decrease in mercury cell chlorine production.   Mercury loss  to the
environment is expected to decrease  by 78 percent during the decade.
     Alternative Technologies.  Chlorine and caustic alkali cannot be  economi-
cally or practically eliminated from the U.S. inorganic chemical industry.
However, two alternatives could be implemented in the future to eliminate or
reduce mercury pollution.  (The economic impact of these two alternatives will
be dealt with at some length in the next section.)
     The first alternative is to switch from the mercury cell process  to  another
process.  The diaphragm cell process is the most attractive, because it is the
major chlorine process used in the United States  (accounting for 69 percent of
chlorine production in 1973).  Other processes generate chlorine as a  by-
product (7 percent of 1973 chlorine production).*  Like the mercury process,
the diaphragm process generates hazardous wastes—most significantly asbestos,
lead, and chlorinated hydrocarbons.   Asbestos waste, however, is expected to
be largely eliminated by 1983 as plastic membranes are developed to replace
asbestos.  Continuing development of improved diaphragm cells, accelerated by
the current disfavor of mercury cells, is likely to result  in even higher
efficiencies and lower discharges of noxious or  hazardous  substances,  further
enhancing the competitive position of diaphragm  cells.  However,  complete
replacement of existing mercury cell  facilities  within a 10-year period  is
most unlikely.
     The second alternative—and the more practical—is to regulate the  dis-
charges to air, water, land,  and product.   By 1983 mercury pollution  control
 * The most significant of these processes is the Downs' cell (electrolysis
  of molten sodium chloride to obtain sodium as the principal product and
  chlorine as the minor product).
                                        151

-------
     TABLE 26. - Estimated Mercury Losses from Typical Chlor-Alkali Plant
                           Operations, 1973 and 1983
                                  (kilograms)
Mercury Losses to
Air
Water
Land
Product
Totals
1973
15,869
1,951
244,875
8,661
271,356

1983
12,527
113
48,310
1,035
61,983

Source:  URS Research Company
                                      152

-------
technology is likely to be well developed, and, as indicated by Table  26,
environmental mercury losses should be  reduced to acceptable limits.   A signi-
ficant reduction in mercury consumption by 1983 is likely; several plants
already have consumption rates  approaching 0.1 Ib of mercury per ton of chlorine.

     Manufacture of Mercurials

     The manufacture of mercury compounds  (mercurials) accounted for almost 20
percent of total U.S. consumption of  mercury  in 1973.  Some 10,718 flasks were
consumed in making organic and  inorganic mercurials in the following Bureau of
Mines usage categories:  agricultural,  17.1 percent; catalyst, 6.3 percent;
paint, 70.9 percent; and Pharmaceuticals,  5.7 percent.  Mercurials for any or
all of these categories are often manufactured in a single facility.   Hence, in
this section we consider the manufacture of inorganic and organic mercurials as
a single technology, but one with broad applications.
     The manufacture of mercury compounds  can be assigned to several SIC groups
(for example, 2818 organics; 2819 inorganics; 2833 medicinals).  The process
initially involves conversion of the  metal into one of several inorganic
mercurials which in turn is used to produce other inorganic, or a variety of
organic, mercurials.  There are only  about 16 manufacturers of mercury compounds
in the country, and often the same facility manufactures  both  inorganic and
organic compounds.
     Mercury compounds, although essential in a  variety of uses,  are a very low-
volume chemical for which statistical data are virtually unobtainable.  However,
we have made an estimate, based on available information, that the total value
of shipments for all mercury compounds (including both intermediates  and final
product) was about $30 million in 1973 and required about 500 production workers.
In contrast, the relatively small inorganic pigments industry (SIC 2816) employed
8,900 production workers -in 1967, with a total value of  shipments of  over  $500
million.*  Despite their  low volume,  some 45 mercury compounds are regularly
manufactured in the United  States, according to the Directory of Chemical
  The production of mercury compounds  is not an industry per se.   In most cases
  the production of mercurials is almost a  sideline—to a much larger chemical
  facility.  Even in facilities that specialize in mercurials', the production
  of nonmercurials is usually an important  consideration.

                                        153

-------
  Producers (Ref. 88).  Of these, however, only four (mercuric  chloride,  red and
  yellow mercuric oxide, mercurous chloride,  and ammoniated mercury)  are  of
  sufficient commercial interest to merit a listing in the Chemical Marketing
  Reporter.
       The price of mercury compounds is quite dependent upon the  selling price
  of mercury metal (although a time lag between changes in the  price  of the
  metal and of the compound can be observed);  as a rule of thumb,  inorganic
  mercurials sell for one and a half to two times the billing price for mercury
  metal, whereas organic compounds sell for three and a half to four  times the
  cost of the metal.   The higher price for the organics is primarily  attributa-
  ble to their usage of intermediate inorganics and of more complicated syntheses.
  For example,  the quoted price of mercuric chloride (with a 73.9  percent mercury
  content) varied between $7.17 and $8.66 per  pound in July 1975,  whereas the
  price for phenylmercuric acetate (with a 59.6 percent mercury content)  is
  estimated to be $14 per pound.
       Because  the mercurials are mature chemicals with a  long  history  of manu-
  facture and because,  in most cases,  their production is  a simple process,
  manufacturing facilities are generally simple and use standard technology.
  These facilities tend to be small—for example,  a 1,000-gallon reactor  is con-
  sidered average for the industry;  the  adoption of small batch processes was
  dictated by the high cost of the materials (primarily of  mercury) and the
  instability of the  market.   For reasons which are not completely clear,  the
  major manufacturing center  for  mercury compounds  lies within  a rather small
  region in New  Jersey which  accounts  for almost 90  percent of  all mercurials
  made  in the nation.
      Catalysts  represent a  small, but  important,  secondary manufacturing usage
  of mercury compounds.  Mercury  catalysts  are  still used to a  small extent in the
  synthesis of vinyl chloride  (from acetylene and hydrogen chloride) and  in the
 synthesis of vat dyes.  These secondary manufacturing usages are  described
 briefly in the following subsections.
      Production Sites.  The 16 U.S. mercurials manufacturing plants and their
 estimated production in terms of use category are listed in Table 27.   As
mentioned above,  these production sites are highly concentrated in the New
Jersey industrial region,  which, except for pharmaceutical end uses, accounts

                                      154

-------
          TABLE  27. - Estimated Mercury Consumption of Mercurials in 1973,  by Plant and Product End Use Category'
                                                          (kilograms of Hg consumed)
en
Location
Plant
Tenneco
Eli Lilly
Becton, Dickinson
Strem
Mallinckrodt
City Chemical
W. A. Cleary
Cosan
Mallinckrodt
Merck
Richardson-Merril 1
Tenneco
Troy
Ventron
RSA-Ardsley
Pennwalt

Totals
State
California
Indiana
Maryland
Massachusetts
Missouri
New Jersey
New Jersey
New Jersey
New Jersey
New Jersey
New Jersey
New Jersey
New Jersey
New Jersey
New York
Oklahoma


City
Long Beach
Indianapolis
Baltimore
Danvers
St. Louis
Jersey City
Somerset
Clifton
Jersey City
Hawthorne
Phillipsburg
Garfield
Newark
Woodridge
Ardsley
Tulsa


Catalysts
1,378
-
-
-
-
3,447
-
-
1,723
3,447
1,723
3,447
3,447
3,447
345
999
23,403


Agriculture
3,447
1,723
-
1,723
5,170
3,447
10,340
-
4,137
8,961
-
8,617
10,340
5,170
-
—
63,075


Paint
17,234
-
-
-
-
-
20,680
79,274
-
6,893
-
68,934
41,464
27,574
-
"•
262,053


Pharmaceuticals
-
2,964
4,136
-
3,447
-
-
-
5,170
5,170
-
-
-
-
-
~
20,887


Total
22,059
4,687
4,136
1,723
8,617
6,894
31,020
79,274
11,030
24,471
1,723
80,998
55,251
36,191
345
999
369,418


          a. Values are based on estimated usage, in flasks, and were rounded to the nearest kg.

          Source:  URS  Research Company

-------
for about 88 percent of total mercurials production.   (The known concentration
of these producers in New Jersey was one reason for selecting that state as
part of a study region; however, as discussed earlier in connection with the
inventory results, the combined impact of these numerous producers was found
not to be significant for that region.)  More quantitative data on the tech-
nology involved, by-products produced, etc., are not  available and would in
any case have little meaning, since these statistics  can readily change as
markets change for specific compounds.
     Mercury compounds are used as catalysts in two major types of chemical
syntheses—the manufacture of vinylchloride monomer (VCM)  and that of anthro-
quinone "vat" dyes.  (Mercury catalysts are also used in foaming polyurethane
plastics in place and in other minor applications related to urethane.  How-
ever, the use of mercurials in these operations is small when compared to the
whole industry,  and the manufacturing plants are widely dispersed around the
nation; therefore, a use-site tabulation has not been attempted for them.)  The
few production sites for the two principal uses of mercury catalysts are shown
in the tabulations below.   These rough estimates indicate again that New Jersey
is an important manufacturing center (for vat dyes),  but that the production of
VCM is limited to the Southeast petrochemical complex.   In both cases production
involving mercury as a catalyst is small in comparison with the manufacture of
other mercurials.

                     Use Sites for Production of Vat  Dyes
                                                      Estimated Use of
                                                      Mercury in 1973
            State        	Company	   	(kg)
       New  Jersey      American Cyanamid                   270
                        G.A.F.  Corporation                  270
                        Otto B.  May Company                 270
                        Mobay Chemical                      270
                        Toms River Chemical                 270
                           (Ciba-Geigy)
       Pennsylvania     American Color &  Chemical           250
       Massachusetts    Nyanza,  Inc.                         240
       North Carolina   Soll-Tex Chemical                   230
         Total                                             2,070
                                      156

-------
       Use Sites for Production of Catalysts for Vinyl Chloride Monomer
                                         1973 Production     Estimated Use of
                                              of VCM         Mercury in 1973
         State (City)        Company     (millions of kg)     	(kg)
      Louisiana (Geismar)    Monochem           136                9,800
      Texas (Houston)        Tenneco            102                7,400
        Total                                   238               17,200
     1973 Technology.  Although mercury compounds have been synthesized since
time immemorial, the quantities involved are small and the general interest
level in commercial processes is likewise small.   Therefore the literature
essentially ignores commercial processes for the  production of mercurials.
But at the same time, manufacturers are reluctant to release details concern-
ing their processes, possibly because of this very dearth of information.
However, based upon an analysis of similar inorganic reactions and on the
information that could be extracted from the available literature,* we have
concluded that the inorganic and organic syntheses involved are straightforward
and require only simple equipment such as corrosion-resistant atmospheric
reactors, column stills, and so on.  We therefore felt that consideration of a
few commercial processes for which details are known should suffice to typify
the industry as a whole.
     The main concern in this portion of the study was to determine the emissions
to air, water, and land from the various processes used in making mercurials.
Accordingly, we selected four inorganic processes and one organic process for
consideration.  The inorganics selected are among those most commonly  used,  and
they provide intermediates for production of organic mercurials.
* Ref. 68, which is one of our basic sources of information, provides additional
  flow charts for alternate ways of producing mercuric  sulfide,  mercuric
  chloride, and red mercuric oxide.  It also includes flow charts for the
  manufacture of yellow mercuric oxide and mercurous chloride.
                                       157

-------
       Figure 31a is a flow diagram of a common production method for black mer-
 curic sulfide  which utilizes  an aqueous phase reaction.  In this and subsequent
 diagrams  we have assumed yields of between 95 percent and 98 percent,  although
 some  sources claim even higher yields.  We then assigned mercury losses to air,
 water,  or land,  depending on  the  type  of reaction.  As can be seen in Figure
 31a  (a typical case)  most of  the  loss  is recaptured in the treatment process
 where the mercury wastes are  precipitated by sodium sulfide treatment, removed
 by filtration,  and then recycled  (after further purification) back into the
 process.  Losses to water from the treated aqueous wastes are low,  and losses
 to air are  minimal.
       Figure 31b shows a common production method for mercuric chloride (known
 as corrosive sublimate)  in which mercury metal is reacted in a closed, glass-
 lined or  stainless steel vessel with an excess of chlorine.   Heat is applied
 to promote  a gas  phase  reaction which results in high yields of the mercuric
 chloride.   Unreacted mercury and chlorine are trapped in a caustic  scrubber
 and this  waste  effluent  is treated to recover small amounts  of mercury (as
 mercuric  sulfide).
      Figure 32  shows two alternate, and common,  methods for  producing  red
 mercuric  oxide  (widely used in battery manufacture and as an intermediate in
 the synthesis of other mercurials).  Both of these aqueous phase reactions
 produce high yields, and waste effluents are treated to recover most of the
 mercury losses.
      Figure 33 is a flow diagram of the manufacture of the most widely used
 organic mercurial—phenylmercuric  acetate.   Here again yields are high and
 mercury losses are small, and most of the mercury that is  lost is recycled
 from  the waste effluent.  (It should be noted that the very  high value of the
 waste  mercury is in itself an incentive to recover it;  in this case the
 recovered mercury represents  a savings of over 2 percent on  raw material costs
 alone.   In contrast,  in  the chlor-alkali industry the  amounts of mercury losses
 are much larger,  but the additional cost per  kilogram  of product produced is
 insignificant.)  The finished  phenylmercuric  acetate is  sold either as a dry
powder  (100  percent)  or  as a dilute suspension.
                                       158

-------
    1000(asHg)<
         H2S Gas-
SOURCE: Ref.60.
                        Reactor
                                     1000
              Filter
                           45



                         HgS





                          46
            950   Black Mercuric

                  Sulfide  •
                 HCI

                 50
Na2S-
Treatment
          1        I'
        Solid Waste  Water
                  Effluent
            1
                                             Land
                            (a) MERCURIC SULFIDE
Vent to Air

M.,r\l I _i



mm /**l-ll *



1 ,



1



<


.
bber




CI2

Reactor

1



Land
i .
«1
980



I .
Heat
19


20

Waste Effluent 4
NaCI + NaOCI UU
-------
C02 Air
    «1
                                     Water
                                                       Land
                                      I

luuu (as rig;
HgCI2 *
Water »









i
Reactor

i




t

Na2S

35




1000




Filtration
and
Washing
i




960 960


40
t
Waste
Treatment
36
Solid Waste
HgS
1-
Land


\- .
NaCI 4 » Water
Effluent


(a) FROM MERCURIC CHLORIDE

1000
Hg >
CI2 *
NaOH— »•

2



Air
t«,
IHCI
Glass-Lined
Reactor

i


i
35



Air Air
]
1000


Wash

«1 |<1
Tank

1000 Filter 960
Purification

40


Solid
Wastes
J
_ 36 Wastewater 4
* Treatment * Wdlt)l
1
Land
SOURCE: Ref. 60; URS Research Company.
                             (b) FROM MERCURY METAL
      Figure 32  MERCURY BALANCE FOR TWO METHODS FOR PRODUCTION
               OF RED MERCURIC OXIDE
                                   160

-------
                      «1
                                          n •     485 ^ Phenylmercuric
          Mercury 1QOO
            Oxide
          Benzene
        Acetic Acid
L—^|  Mixing  |  470   Phenylmercuric
                   Acetate (10%-30%)
                   Acetate (100%)
                                                                  955
                                                       Water
                                                       Land
       SOURCE: URS Research Company.
Figure 33  MERCURY BALANCE FOR SYNTHESIS OF PHENYLMERCURIC ACETATE
          (using 1973 technology)
                                      161

-------
       In general,  mercurials are manufactured in small reaction vessels, pro-
  ducing a slurry in an inert liquid which  can conveniently be filtered to
  recover the desired product (which may be further purified); after treatment,
  the filtrate is discharged.   In the 1973  technology, the filtrate, which may
  contain unreacted mercury compounds,  is treated with sodium sulfide to pre-
  cipitate mercuric sulfide and,  after a second filtration, is discharged to the
  sewer.  Because these effluent  flow rates are so small and the removal rate
  for residual mercury is so high,  the total quantities of mercury discharged
  are low.  Figure 34,  which was  supplied by the Ventron Corporation, shows the
  control capability currently in use by this major manufacturer of mercury com-
  pounds which results in a final effluent  containing less than 0.23 kg of
  mercury per day.   We estimate that,  on the average, about 0.4 percent of the
  total mercury used in a given plant will be lost to the environment, primarily
  to water.
       As mentioned earlier, vinyl  chloride monomer (VCM)  was still being produced
  in 1973 from acetylene  and hydrogen chloride,  using a mercuric chloride catalyst.
  Figure 35  is the  flow diagram for this reaction,  and indicates the resultant mer-
  cury losses to the environment.   It is believed that most of the mercury losses
  occur when the carbon pellets impregnated with mercuric  chloride are discarded
  but,  as indicated  on the figure, other small losses to product and to air can
  occur.   However, the overall loss is of the same  order as that for other chemi-
  cal  reactions  involving mercury—less than one-half of 1 percent of throughput.
  It seems likely that the losses  for synthesis  of  the anthroquinone dyes would be
  of the  same order, although this belief cannot be verified because of lack of
 knowledge concerning the processes actually employed.
      Both of these uses of mercury as a catalyst  are on  the decline.   According
 to available information, the Tenneco plant in Texas has changed its production
 route for vinyl chloride monomer and stopped using mercurial catalyst in 1974.
 Usage of anthroquinone,  which is sulfonated in the presence of mercuric sulfate
 to  yield two important dyestuff  intermediates—1,5-  and  1,8-disulfonic acid—is
 also  declining.  (The anthroquinone vat dyes are of a  class that can be easily
 reduced to  a soluble,  colorless  form in which the  fibers are readily impregnated;
after this  the  dyes are  oxidized to produce the insoluble color  in the fibers.)
                                       162

-------
Utility
Water
Utility
Water


Organic
Mercurials
Production

Inorganic
Mercurials
Production
Process Water
1000 to 2000 GPD
20 to 80 ppm Hg
Process Wa
5,000 to 12,00
50 to 150 ppi
\
Men
Rem
Pro
ter '
OGPD
TiHg

Continuous
Sample and
Flow Measurement
i .
i
cury
cess Total Plant 1
' * c nnn tn At
' Less than 0.5 L
Continuous
Sample and
Flow Measurement
< ,
Effluent -
),OOOGPD;
b/Day Mercury
Process and
 Utility
 Water
Nonmercurials
 Production
                         SOURCE: Ventron Corporation.
               Figure 34  WASTEWATER FLOW IN A PLANT PRODUCING
                        MERCURIALS AND NONMERCURIALS
                                      163

-------
                         Acetylene and HCI
Catalyst (as HgCI2)

             1000
"1
Re
       Acetylene -»•
Hydrogen Chloride—
                      1000
efrigerated
Column
                                              c
                                              3
                                     *• Land
                                                     t
                                                    Phenol
                                                            Vinyl chloride (stabilized)
                                                             Bottoms to Pot Still
Note: Use of 1,000 kg of mercury as catalyst would account for vinyl production of about 26 million kg —
     about 10 percent of the production of the industry by this route in 1973.

SOURCE: URS Research Company.
   Figure 35   MANUFACTURING PROCESS FOR VINYL CHLORIDE MONOMER
              USING A MERCURY CATALYST
                                      164

-------
The emergence of new synthetic fibers and new dyeing technologies is expected
to reduce demand for anthroquinone dyes,  with a corresponding decline in
mercury usage.
     1983 Technology.  No important changes or improvements in the technology
for making mercurials are forecast.  Individual facilities may upgrade equip-
ment in an effort to reduce possible sources of mercury loss to the environment
and to improve process efficiency.  However,  the production of mercurials is
small and is not in a growth phase, so that there are no pressures to improve
the technology beyond its present limits.  Also, the costs involved in the
small batch processes now generally in use are dominated by the raw materials
cost (primarily for mercury) and small batch sizes allow a degree of flexibil-
ity which is quite important to this small industry.
     Improvements in the pollution control technology associated with the indus-
try are likely.  Although the mercury losses are presently small, continued
emphasis upon a clean environment will undoubtedly lead to even more reduction
of these losses.  For example, rather than directly discharging liquid wastes
to sewers after treatment, some plants are now using holding tanks which effec-
tively isolate any accidental spills and prevent their inadvertent discharge
to sewers.  For 1983, then, we predict that total quantities will be lower and
that discharges from production will be about half as great as those for 1973—
that is, 0.2 percent of throughput.  Examples of the technology which we see as
helping to effect this change are  discussed briefly below.
     Improved methods for removing traces of mercury from water  effluents are
now in use (see chlor-alkali technology  discussion).  Basically,  these  systems
depend on the precipitation of mercuric  siilfide (by the  injection of sodium
sulfide) from the waste stream, with collection of  the precipitate on a suitable
filter.  The mercury content  of the precipitate is  usually recovered either
by a retorting process  (with  control of  air  emissions)  which returns mercury
metal to the system, or by a  dissolution process which returns mercuric chloride
to the system.  With either of these technologies,  discharges of mercury to the
receiving water are reduced to the parts per billion range, and total quantities
discharged are well below E£A standards.
                                        165

-------
       Removal of mercury from gas  streams  (which is not a particular problem in
  the manufacture of most mercurials) has been of special interest to the chlor-
  alkali industry.   Final traces of mercury in the hydrogen stream from these
  operations are frequently removed by a final pass through a molecular sieve.
  Another possible  technology involves the use of sulfur-impregnated carbon
  filter bed.   The  manufacturer (Pittsburgh Activated Carbon Division of Calgon
  Corporation)  claims its Type HGR  activated carbon will reduce mercury concen-
  trations in hydrogen streams from 3 ppm to 1 ppb or less at all gas velocities
  up to 59 feet per minute (fpm).  A mercury recovery of over 95 percent is
  obtained by retorting.
       Development  of pollution control devices for a variety of heavy metals
  is continuing and is likely to accelerate as regulations and standards are
  developed and applied.   Mercury manufacturing and using industries will bene-
  fit from these developments, which will likely be aimed at heavy metal emitters
  with far larger total losses  than those involving mercury.   As processes are
  developed which can remove  a variety of pollutants in a single pass,  the con-
  cept of centralized industrial waste treatment facilities may perhaps become
  a  reality; this would increase removal efficiency and decrease overall costs.
      Other Considerations.  The manufacture of mercurials will continue and
 will of necessity involve the use of mercury,  even though alternative
 technologies and products are now reducing the total demand for mercurials.
  (For example, the production of vinyl chloride monomer using other tech-
 nologies will eliminate the need for mercuric  chloride catalysts.)   Mer-
 curials are presently a nongrowth industry, with  further recession likely
 in the future.
      Mercury losses through theft, a particular problem when a high-value
 product is involved, will undoubtedly continue  to plague the industry.
 Iricreased security is one possible answer.  Another, perhaps more practical
 and considerably cheaper answer, is  for manufacturers  to place increased empha-
 sis upon  yields and material balances.  Since the processes  involved  are well
known and well  documented, expected yields are also well known.  Furthermore,
analytical techniques are now well developed for tracing mercury losses  in
                                        166

-------
waste matter (mostly as an inorganic  or  organic material  in water).  Thus the
bases for tracking all mercury into and  out  of a process  are now available.
If such simple balances were maintained  at all times within the plant, abrupt
or even subtle changes in balance would  immediately alert management to  the
existence of trouble—probably theft.   (While theft may not necessarily  indi-
cate a loss of mercury to the environment, the consumer of stolen mercury or
mercury compounds will probably have  little  regard for regulations concerning
the control of mercury losses to the  environment.)
     The storage, handling,  and shipping of  mercury and its compounds involves
a very low volume flow, with only a limited  possibility of accidental spills
or losses.  Mercury metal is almost universally  shipped in 76-pound lots, in
iron flasks equipped with screw plugs.   Flask breakage is virtually impossible,
although reports have been received of plugs being loosened with subsequent
loss of mercury in transit.  However, because of the value of mercury, every
effort is made to minimize such losses.   According to Ref. 89,  the regulations
are minimal for surface transport; however air transport  is prohibited because
of the vulnerability of the aircraft's aluminum structure  to amalgamation.
     Mercury compounds are generally-shipped as solids, typically  in  50-pound
fiberboard drums.  Because of the density of the mercurials  and the consequent
ease of handling, there seem to be very few accidents.  The  Interstate Commerce
Commission  (ICC) requires a poison B label on all solid mercury compounds, with
a maximum container size of 200 pounds.   The International Air Transport Asso-
ciation  (IATA) also requires the poison B label for a maximum shipment of 25 kg
in the passenger section and 95 kg in the cargo section of a plane.  Liquid mer-
curials  (normally a solution or suspension of solid material) are usually ship-
ped in small containers such as 55-gallon drums).  According to the ICC they
must carry the poison  B label, but there  is no restriction on container size.
The IATA also requires the poison B  label, with the same weight limitations  as
for solid mercury compounds.
     Mercury and most  of  its compounds  do not present fire or  explosion hazards.
 (The most obvious exception to this  is  mercury  fulminate, which is a prohibited
explosive and is no longer used or available in this country.)   However, fires
or similar catastrophes involving mercury or its  compounds  could generate
mercury vapor clouds,  which could be very dangerous  to firefighters or other
                                       167

-------
  persons in the immediate area.  The use of chemical gas masks, while  unlikely
  to be fully effective against mercury vapor,  would certainly be  indicated
  for people downwind of a fire.  (In other types of fires  involving mercury-
  containing products, for example a paint warehouse fire,  mercury vapors might
  be released, but the concentrations would probably be  so  dilute  as not to be
  of major concern.)
       Implications.  Total estimated mercury losses to  the environment from the
  production of mercurials and from the use of mercurial catalysts in the manu-
  facture of vinyl chloride monomer and vat dyes  are shown  in Tables 28 and 29,
  respectively.  The estimate for the manufacture of mercurials assumes good
  control in the present (1973)  time frame,  with  an  average loss for each step
  of production of 0.4 percent of the mercury input.   Virtually all of  this loss
  is assigned to water;  even for the largest plant identified the  estimated dis-
  charge to water would not exceed 1 kg per day on this basis (and this does not
  imply that this manufacturer's losses are necessarily that high).  Losses to
  air are small and,  because of  industrial  safety requirements,  these discharges
  are vented to the atmosphere.   Very little solid waste accumulates (as in
  discarded filter materials), and the solids that do accumulate would be directed
  to industrial disposal sites.
       By 1983 the production of mercurials  will have fallen sharply as demand
  declines, particularly for paint and agricultural usage.   Also by 1983, control
  technology will  be  adopted that should halve total mercury losses,  particularly
  with  respect to  water.  Thus by 1983  losses to water will be well regulated and
  controlled.
      The losses  from the use of mercurial catalysts are only estimates and are
 based on the assumption that no catalyst recovery is undertaken but that  the
 spent catalyst is discarded to industrial landfills.  Hence the total loss for
 1973 from catalyst usage is seen as higher than that from all mercurial manu-
 facture (Tables 28 and 29).  Of course the final disposal site for catalysts—
 industrial landfills—is less likely to impinge on man's  habitat than is  the
 case for the other mercurials.   By 1983 the use of mercurial catalysts for
 producing VCM and vat dyes will certainly be greatly curtailed,  but in our
 estimates we  again assume total loss of the mercury content of the spent
catalyst.  If recovery processes are instituted,  these  losses  would be greatly
decreased.

                                       168

-------
            TABLE 28.  -  Estimated Mercury  Losses  from Production
                        of Mercurials,  1973  and  1983
                                 (kilograms)
Losses to                            1973                 1983


Air                                    80                   10

Water                               2,800                  450

Land                                   80                   20
Totals                              2,960                 480
a. Estimated losses are based on 0.4 percent and 0.2 percent of through-
   put for 1973 and 1983 respectively.  Throughput is calculated as the
   total usage of mercury in the Bureau of Mines' use categories of agri-
   culture, catalysts, paint, and Pharmaceuticals, which for 1973 totaled
   369,419 kg and for 1983 are estimated at 120,000 kg.  However, these
   values have been doubled to account for the numerous intermediate
   steps in the manufacture of mercurials, which result in double han-
   dling (by our estimate) of mercury consumed.
Source:  URS Research Company.
       TABLE 29. - Estimated Mercury Losses from Use of Manufactured
                    Catalysts in the Production of Vinyl Chloride
                             Monomer and Vat Dyes,
                              1973 and 1983
                                (kilograms)
Losses to                     .        1973                 1983


Air                                     50                   15

Water                                  100                   30

Land                                18,850                6,955
Totals                              19,000               7,000
Source:   URS Research Company


                                      169

-------
       Battery Manufacturing (SIC 3692)

       The battery manufacturing industry consumed approximately 16,260 flasks,
  or 30 percent,  of the mercury used in the United States in 1973.*  According
  to a recent report on the industry by Versar, Inc.  (Ref. 69), in the battery
  manufacturing process the loss of  mercury to the environment was only about 82
  flasks,  or 0.5  percent of the mercury used by the industry.  Most of the
  mercury lost in this manner was deposited in landfill sites near the manufac-
  turing plants.   Some 95 percent of the mercury lost from the battery manufac-
  turing and use  cycle occurs when the batteries are thrown away by the consumer,
  with 92 percent of this going to landfill and 8 percent to public incineration.
  The remaining wastes are  recycled  by the government, industry, or manufacturers.
       Mercury is used in the manufacture of zinc-carbon dry cells, alkaline-
  manganese dioxide dry cells,  mercury cells (Reuben,  Weston, and mercury-cadmium
  cell),  zinc-silver oxide  cells,  and carbon-zinc air cells.  All of these
  batteries are manufactured within  SIC group 3692 (primary batteries), except
  for the  zinc-silver oxide battery,  which is in SIC group 3691 (storage batter-
  ies).  In the following technological analysis,  however,  only the zinc-carbon,
  alkaline-manganese  dioxide, and Ruben mercury batteries are considered,  because
  they account  for 99.5 percent of the industry's  mercury losses (Ref. 69).
 These mercury losses, by battery type,  are shown in Figure 36.
      The average annual growth rate of the battery industry from 1963 to 1973
 was 4.3 percent.  The industry is expected to grow at about 4.5 percent annually
 from 1973 to 1985 (Ref. 71).   Developments in battery technology and new appli-
 cations for batteries will be  the prime motivators for this continued growth;
 for example,  battery uses  will expand with the development of electric vehicles
 as a means of avoiding the increasing costs of fossil-fuel burning vehicles.
* Best estimates based on Ref. 14.  This value is approximately one-third
  higher than that reported in Ref. 69.
                                      170

-------
757 Kg
(28.3%)
Mercury Purchased
564,140 Kg
I
Battery
Industry 2,674 Kg
1 Loss
561 ,466 Kg
Mercury in
Product
1,446 Kg
(54.1%)
457 Kg
(17.1%)
469 Kg 10 Kg
(17.54%) (0.39%)
2 Kg
(0.05%)
1Kg
(0.01%)
2 Kg -
(0.05%)







SOURCE: Ref.70.
                                           Zinc-Carbon Dry Cell





                                           Alkaline-Manganese Dioxide Dry Cell
                                           Ruben Cell
                                           Weston Cell
-•»• Mercury Cell
                                           Mercury-Cadmium —
                                           Zinc-Silver Oxide Cell
                                            Carbon-Zinc Air Cell
Figure 36  MERCURY LOSS IN MANUFACTURING, BY TYPE OF BATTERY, 1973
                                     171

-------
       The three major consumer categories for the battery industry are the
  public,  industry, and government.   In 1973,  70.5 percent of  U.S.  battery
  shipments went to the public,  19.4 percent to industry,  and  10.1  percent to
  the government.  By 1983,  shipments to these sectors are expected to  be  72.7
  percent to the consumer,  19.7 percent to industry,  and 7.6 percent to the
  government (Ref. 71).
       For over a century,  until the early 1960s,  the primary  battery market
  was dominated by the zinc-carbon battery.   Recently, however superior elec-
  trode materials have been developed that pose a  competitive  threat to the
  zinc-carbon cell.  This type of battery represented 54 percent of the primary
  battery market in 1973, but it is  expected that  by  1985  it may account for
  only 36  percent of the market  (Ref.  71).
       In  contrast, the alkaline-manganese dioxide battery's popularity increased
  very rapidly during the last decade.   In 1973  this battery accounted  for 22
  percent  of the mercury battery market,  and it  is expected to account  for 34
  percent  of the market by  1985  (Ref.  71).   The  chief causes of this growth are
  the superior energy characteristics  and less expensive costs.
       Mercury cells are expected to account for some 7 percent per year of the
  primary  battery market between 1973  and 1985  (Ref.  71).  These cells  will have
  a steady share  of the  market because of their  specialized uses as small,
  reliable,  and  long-life power  sources.  The remaining primary batteries  should
  continue  to account for about  23 percent of the market over the next  decade
  (Ref. 71).  The batteries included in this group are silver-zinc,  ammonia-
 activated, cadmium-mercury, metal-air, solid-state,  organic electrolyte,
 lithium,  and sodium-bromine.
      Production Sites.  In the United States there  are  21 plants that manufac-
 ture zinc-carbon batteries, 7 plants that manufacture alkaline-manganese  dioxide
 batteries, and 7 plants that manufacture Ruben mercury batteries (Ref. 69).
 Of these,  Union Carbide, ESB Inc. (Ray-0-Vac),  P. R. Mallory, and Gould Corpora-
 tion (Burgess)  accounted for 85 percent of the battery  sales  in 1973  (Ref.  71).
Table  30  shows plant locations  of the principal manufacturers,  by state,  and
indicates  the approximate mercury losses to the environment from these plants
                                       172

-------
           TABLE  30.  - Major U.S. Battery Manufacturers'  Mercury
                         Losses, by Plants  and State, 1973
                                     (kilograms)
Region and State
Atlantic
Connecticut
Vermont

New Jersey


New York
North Carolina


Pennsylvania
Rhode Island
South Carolina
Subtotal5
North Central
Illinois
Iowa
Missouri
Ohio

Tennessee

Wisconsin


Subtotal5
Total
Manufacturer and
Type of Battery
Manufactured

C-Z
C-Z, A-M, Hg
C-Z, Other
C-Z
C-Z, Other
C-Z
Hg
C-Z, A-M, Other
C-Z, A-M, Hg

C-Z, A-M
Other
C-Z, Hg
•

C-Z, Hg
C-Z
Other
C-Z, Other
C-Z
A-M
A-M
C-Z, A-M, Hg
C-Z
C-Z, Other


Losses
Land

<1
29
3
9
<1
8
127
129
439
36
43
3
6
833

101
27
3
224
93
135
34
998
14
14
1,643
2,476

to the
Environment
Water Air

_
<1
-
-
12
_
3
3
9
1
1
-
-
29

2
<1
-
5
2
3
1
20
-
—
33
62


_
2
<1
<1
<1
<1
7
7
19
2
2
-
<1
42

5
2
-
12
5
7
2
54
1
1
89
131

Total
Losses

<1
32
4
10
13
8
136
138
467
39
46
3
7
902

109
29
3
241
101
145
36
1,073
15
15
1,767
2,670

a.. In some instances, a manufacturer may have more than one plant in a state,
   Battery types are designated as follows: C-Z = carbon-zinc dry cell; A-M =
   alkaline-manganese dioxide dry cell; Hg = Ruben mercury cell.
b. Totals may not add, due to rounding.
Source:  Derived by URS from Ref. 70.

                                      173

-------
  in 1973.  (As previously mentioned,  only plants manufacturing  zinc-carbon,
  alkaline-manganese dioxide, and Ruben mercury cells  are  considered in this
  analysis.)  As the table shows,  all of the significant mercury-consuming plants
  are located in the Atlantic and North Central regions of the country.  About  34
  percent of the mercury losses in battery manufacturing occurred in the Atlantic
  region and 66 percent in the North Central region.   (Note the  predominance of
  manufacturing and losses in Wisconsin and North Carolina.)
       1973 Technology.  The present battery manufacturing technology is most
  readily discussed in terms of types of batteries.
            Zinc-Carbon Dry Cells.   Although mercury is frequently used to amal-
  gamate zinc,  its primary use in  this process  is as a paste  (mercury chloride)
  applied to paperboard separators.  Mercury losses during these operations are
  comparatively small.
            The major loss of mercury  during the  manufacture of  zinc-carbon
  batteries occurs when batteries  are  rejected  during inspection, testing, and
  packaging.  Normally,  these rejects  are  sent  to local landfill sites  for dis-
  posal.   The amount of mercury wastes from rejected batteries represents
  7.3 x  10~  kg of mercury per kilogram of batteries produced.  The total annual
  mercury loss  from all plants that manufacture zinc-carbon is 756.4 kg  (1,668
  Ibs)  (Ref. 70).   Presently  there is  no technology to allow economical  recycling
  of  these rejected batteries.  However, the  zinc-carbon batteries that  are
  thrown  away by the ultimate  consumer represent  the most  significant loss of
  mercury to the environment  (mostly to landfills or public incineration).
           Alkaline-Manganese Dioxide Dry Cells.  Mercury is amalgamated with
 powdered zinc to form the anode.   This is normally done in a closed system,  so
 that there is very little mercury loss.  Again,  the primary loss of mercury
 occurs when rejected batteries are disposed of in landfill  sites.   This esti-
 mated  loss is  7.6 x 10*  kg of mercury per kilogram of battery (Ref. 70).
 Annual  mercury losses of all the  U.S. alkaline-manganese  dioxide battery plants
 are  estimated  to  total 1,445 kg (3,187 Ibs).  Rejected batteries are not
 recycled by the manufacturers, because such recycling is  not economical.
          Ruben Mercury Batteries.  A typical  flow  diagram  for a Ruben mercury
cell battery plant is  shown  in Figure 37.  The anode  of the mercury cell is
                                       174

-------
  Mercury
    Input
  100,000
           Land
           HgO
                16    4

SOURCE: URS Research Company.
                          Water
 Figure 37  FLOW DIAGRAM OF MERCURY LOSSES DURING
          MANUFACTURE OF RUBEN MERCURY CELL BATTERIES
                            175

-------
  produced by amalgamating zinc with metallic mercury.  The cathode is fabricated
  by pressing red mercury oxide with graphite.  As  shown in the diagram, during
  these operations very little mercury is  lost to the environment.  In contrast
  to the manufacture of the other two types of batteries, however, 95 percent of
  the rejected Ruben cells are processed to reclaim mercury.  The other 5 percent
  are sent to landfill sites.   Mercury is  recovered from the rejected cells by
  incineration in a reclaim furnace;  the recovery of mercury here is approximately
  1.6 x 10~3 kg mercury oxide  and 0.39 X 10"3 kg mercury per kilogram of product
  (Ref. 69).   Any unreacted residue  containing mercury is disposed of in a land-
                                                                           • 3
  fill.  In this process,  the  total  loss of mercury to landfill is 1.88 x 10   kg
  of mercury per kilogram of product (1.49 X 10"  kg from mercury oxide and
  0.39 x 10~   kg from metallic mercury).  Estimated total annual mercury losses
  from Ruben mercury cell plants  are  457.4 kg (1,009 pounds).   Again,  however,
  the most significant losses  occur when consumers  discard the spent batteries
  (in public  landfill sites or incinerators).
       Technology in 1983.   Because relatively small amounts of mercury are lost
  during battery manufacture,  few changes in manufacturing processes are expected
  by 1983.  However,  there  will probably be more recycling of rejected batteries
  during the  next five to ten  years.  The major reason for this is that the cost
  of zinc  is  rising  so rapidly that recycling is becoming economically justifiable.
  In addition, if industry and consumers follow the lead of the many federal
  agencies that now  sell spent mercury batteries for scrap value,  the total
 mercury loss to landfill and incineration should decrease markedly in the next
 ten years.
      Estimates of mercury losses from primary battery production and consumption
 for 1973 and 1983 are given in Table 31.   Primary battery production is expected
 to increase 45 percent during the decade,  but tight  mercury  controls should keep
 annual manufacturing losses down to about 0.5 percent of the mercury used (for
 an estimated manufacturing loss of 3,876  kg  in 1983).  Mercury losses to land-
 fill that result from consumers disposing of  old batteries will  grow by 31 per-
cent  if recycling approaches  10 percent  (currently it is  about 5 percent)  and
decrease by  27  percent  if extensive recycling  (50  percent) is implemented.
                                       176

-------
                        TABLE  31.  -  Estimated U.S.  Battery Manufacturing and Related  Consumer
                                              Losses of Mercury,  1973 and 1983
                                                        (kilograms)
-a
-4
Type of Primary 1973 Losses
Battery Manufacturer Consumer
Zinc-Carbon 756 15 7 , 84 7

Alkaline-Manganese 1,445 301,752

Mercury 469 98,165

Total 2,671 557,764

1983 Losses
Consumer
Technology
Manufacturer Level
926 High
Low
2,270 High
Low
680 High
Low
3,876 High
Low


Losses
97,190
174,940
240,170
445,200
71,640
116,060
409,000
736,200

      Source:  URS Research Company

-------
       Alternative Technologies.  Battery manufacturers have done an excellent
  job of controlling mercury losses during manufacture.  However,  substantial
  losses to the environment occur when the consumer disposes of old batteries.
  If these losses cannot be controlled by recycling,  and if such losses cannot
  be tolerated, substitutes may have to be found for most mercury cells.
       Batteries which do not require mercury are expected to account for 23
  percent of the primary battery market by 1983,  even in the absence of controls
  on mercury use (Ref. 71).  If environmental controls are enacted to eliminate
  mercury from batteries,  of course,  the manufacturing picture will be drasti-
  cally altered.  Primary batteries which might be substituted for most of the
  mercury cells include silver-zinc,  ammonia-activated,  metal-air,  solid-state,
  organic electrolyte, lithium,  and sodium-bromine cells.   However,  many of the
  materials (cadmium,  chromium,  copper,  lead,  and zinc,  among others)  used in
  these cells also have adverse environmental properties,  and no other material
  has been developed that  has the electrolytic properties  of mercury.   Extensive
  research and development will be required before alternate types of cells will
  be able to compete seriously with mercury.

       Electric Lamp Manufacturing (SIC  3641)

       The  electric  lamp manufacturing industry consumed approximately 1,258
  flasks  (43,360 kg),  or 2.3  percent, of the mercury used  in the United States
  in 1973  (Ref.  14).   During  electric lamp  manufacture,  however, only  50  flasks
  (1,723 kg) of the mercury used by the  industry were  lost to the  environment
  (Refs. 10, 12).  Most  (95 percent) of  the mercury lost in manufacturing went
 to. nearby industrial landfill sites; the remaining losses were to  air.   The
 greatest losses occurred when lamps were discarded by  consumers, with 95 per-
 cent going to landfill and 5 percent to incineration,  according  to the  URS
 inventory.
      Mercury has been used in lamp manufacture since the late 19th Century, but
 its use  in this application was not significant until 1938  (Ref.  72).  Prom
 then  until 1963 total lamp shipments increased at an annual rate of 6.5 percent.
From  1963  to  1973,  however,  lamp shipments increased only 2.9 percent annually,
                                       178

-------
and by 1983 they are expected to increase only 2 percent annually (Ref.  73).
The chief reasons for the slowdown in lamp shipments are the increased popular-
ity of long-life lamps and the use of larger lamps with more lumens per watt
(more light output at lower cost).  It is important to note,  however,  that these
larger lamps (most of which are fluorescent) have had an annual rate of increase
in shipments of 5.5 percent, and this figure is expected to increase to 5.7
percent (Ref. 73).
     Discharge lamps which use mercury are of the fluorescent,  mercury vapor,
metal halide, and high-pressure sodium types.  These lamps,  which are manufac-
tured within SIC group 3641, are used primarily for lighting streets and high-
ways, high-ceiling rooms, motion picture projection, photography,  dental exami-
nations, photochemistry, heat lamps, and water purification.
     Typically, discharge lamps require the following amounts of metallic
mercury:  a 40-watt, 4-foot fluorescent lamp requires 40 mg of mercury and a
75-watt, 8-foot lamp uses 50 mg; a 400-watt mercury vapor lamp contains 60 to
76 mg of mercury; a 400-watt metal halide lamp requires between 50 and 60 mg
of mercury; and a 400-watt high-pressure sodium lamp needs 20 to 30 mg of
mercury (Ref. 74).
     In the following technological analysis, only fluorescent lamps are con-
sidered, because they utilize 95 percent of the mercury used in lamp manufactur-
ing.  Also, fluorescent  lamps comprise 30 percent of total lamp shipments, com-
pared to about 2 percent for any of the other mercury lamps; all mercury lamps
contain about the same amount of metallic mercury; and manufacturing procedures
for all the mercury lamps are fairly similar.
     Production Sites.   Approximately 69 U.S. plants manufactured  electric lamps
in 1973 (Ref. 75).  Of these, 47 manufactured mercury lamps.  GTE-Sylvania,
Westinghouse Electric, General Electric, and North American  Phillips  accounted
for about 90 percent of  the total SIC 3641  sales  (Ref.  73).  Table 32 lists the
number of mercury lamp manufacturers, by state, and shows approximate mercury
losses to the environment.  As the  table indicates,  most of the mercury lamp
manufacturing plants are located  in the Atlantic  and North Central U.S. regions
of the United States.  The  five states that together account for 73 percent of
the mercury losses are Ohio (23 percent), Massachusetts (15 percent), Kentucky
(13 percent). North Carolina  (11  percent),  and New Jersey  (11 percent).
                                        179

-------
                      TABLE 32. - Estimated Mercury Losses from Electric Lamp Manufacturing
                                       in the United States, by State, 1973
                                                    (kilograms)


                         No. of Major
                         Mercury Lamp     Estimated Losses to the Environment    Total Environmental
     Region and State    Manufacturers          Air            Land                      Loss

     Atlantic

       Massachusetts
       New Hampshire
       New Jersey
       New York
       North Carolina
       Pennsylvania
       West Virginia

^      Subtotal               20                   41              776                       817
oo
     North Central

       Illinois
       Kansas
       Kentucky
       Missouri
       Ohio
       Oklahoma
       Tennessee

       Subtotal               22                  41             767                       808
3
1
6
1
1
7
1
13
3
9
2
1
10
3
244
57
180
38
14
185
58
257
60
189
40
15
195
61
3
1
6
1
8
1
2
4
1
12
1
20
_
3
76
14
219
20
386
5
47
80
15
231
21
406
5
50

-------
                         TABLE 32.  - Estimated Mercury Losses from Electric Lamp Manufacturing
                                          in the United States, by State,  1973
                                                       (kilograms)
                                                       (continued)
00


Region and State
South
Alabama
Arkansas
Mississippi
Texas
Subtotal
West
California
Subtotal
United States Total
No. of Major
Mercury Lamp
Manufacturers

1
1
1
1
4

1
1
47


Estimated Losses
Air

2
2
1
1
6

-
0
88
-

to the Environment
Land

31
33
20
14
98

4
4
1,645


Total Environmental
Loss

33
35
21
15
104

4
4
1,733

      Source:  URS Research Company

-------
       1973 Technology.   A flow diagram for a typical fluorescent lamp plant  is
  shown in Figure 38.  The metallic mercury, starting gas  (normally argon), and
  other materials are  injected into a quartz tube which is then sealed.  During
  this process,  very little mercury is lost to the environment.  Small losses to
  air occur when the mercury is spilled (and later vaporized) and when tubes  are
  broken during manufacture and testing, which represent perhaps 5 percent  of
  total manufacturing  losses.   The greatest manufacturing losses occur when
  rejected tubes are sent to industrial landfills for final disposal.  Such esti-
  mated annual losses  total about 1,733 kg, or 95 percent of the manufacturing
  losses.   The fluorescent tubes that are thrown away by the ultimate consumer,
  however,  represent the most  significant loss of mercury—some 41,660 kg
  annually—to the environment (mostly to landfills or public incineration).
       1983 Technology.   Because mercury losses during manufacturing are rela-
  tively small,  few changes in the use  of mercury during lamp manufacture are
  expected by 1983.  Of  course,  recycling or careful disposal of spent fluores-
  cent tubes could occur by that date, but it seems likely that the economics of
  recovering the  mercury from  spent tubes would not warrant recycling.  The only
  possible  reduction of  mercury usage by this industry would come from substitu-
  ting some other, as yet unknown,  material for mercury.
      A comparison  of estimated mercury losses from lamp manufacturing and use
  in  1973 and 1983 is given  in Table 33.  The proportion of electric lamps using
 mercury is expected to  increase about 50 percent by 1983,  but due to tight
 controls, mercury  loss  during manufacture is expected to be only 4 percent  of
 the total input, or perhaps 2,700 kg.  If recycling approaches 5 percent of
 all lamps sold to consumers,  mercury losses to landfill by consumers will
 increase by 43 percent.
      Alternative Technologies. Manufacturers of mercury lamps have done an
 excellent job of controlling internal plant losses  of mercury.   Substantial
 losses to the environment occur, however, when consumers discard spent tubes.
 If such losses  are not  tolerable,  substitutes for mercury or mercury lamps
will have  to be  developed,  but most  mercury  lamp manufacturers indicate that
no other satisfactory substitute for mercury  in discharge lamps  has been found
(Refs. 74, 76).  With extensive research and  development and increased
                                       182

-------
                          Air
                          20kg
   Metallic
   Mercury
10,000 Kg
Fluorescent
  Tube
Manufacture
9,600 Kg
                                 Consumer
                         Landfill
                         380 Kg
SOURCE: URS Research Company.
Figure 38   FLOW DIAGRAM OF ESTIMATED MERCURY LOSSES
          DURING MANUFACTURE OF FLUORESCENT LAMPS
                           183

-------
00
                             TABLE 33.  -  Estimated Mercury  Losses from Lamp
                                 Manufacturing  and Use,  1973  and 1983
                                               (kilograms)

1973 1983
iyptt ut Lamp Manufacturing Losses Consumer Losses Manufacturing Losses Consumer Losses
Fluorescent
Mercury Vapor
Total
1,671 40,227 2,623 57,324
62 1,437 74 2,048
1,733 41,664 2,697 59,372

H*     a.  Assuming a 5-percent recycling rate.
      Source:   URS Research Company

-------
expenditures, a substitute may be developed in the future, but the efficiency
of such a substitute is only conjectural at this time.
     Other lamps which do not require mercury are available,  and in fact com-
prise most of the current market.  Most of these are incandescent metal fila-
ment lamps and cannot match mercury lamps in electrical efficiency or for many
specialized purposes.  Although fluorescent lamps operate  under a higher voltage
than other nonmercury lamps, they give off more lumens  per watt of energy,  thus
requiring less overall energy consumption.  Fluorescent lamps also portray
colors more accurately, have a longer life than most other tubes,  and have a
higher efficiency than conventional lamps (Refs. 72, 74, 77,  78).   If other
lamp types without improved efficiency were substituted for  fluorescent lamps,
the United States would require more energy output from its  electric power
plants in order to maintain current lighting levels.
     Another advantage which conventional lamps have not duplicated is that
mercury vapor lamps can give off high levels of ultraviolet  radiation.  Ultra-
violet ray lamps are used extensively for therapeutic purposes, improved photo-
copying, photochemistry, and as germicides (most notably for water purification)
(Ref. 77).  If these special-purpose lamps were removed from the market, sub-
stitutes would be needed to replace them.

     Industrial Instruments (SIC 3613 and 3821)

     The industrial instrument industry in the United States comprises many  SIC
groups, including electric measuring instruments  (3611); switchgear and  switch-
board apparatus (3613); industrial controls  (3622); engineering and scientific
instruments  (3811); mechanical measuring and control instruments  (3821);
environmental controls  (3822); process control  instruments  (3823);  fluid meters
and counting devices (3824); and photographic equipment and supplies (3861).
This report, however, deals directly with only  two of  these groups,  SIC 3613
and SIC 3821.
     In 1973 the entire industrial instrumentation industry consumed approxi-
mately 7,155 flasks, or 13 percent, of the mercury used in  the United States
(Ref. 14).  For several reasons, this mercury consumption was limited almost
exclusively to metallic mercury.  Metallic mercury is  liquid at ordinary
                                       185

-------
  temperatures (melting point -38.87°C),  with high electrical  conductivity;  it has
  excellent high thermal conductivity and regular  thermal expansion.  Because  of
  these properties,  mercury metal is readily adaptable  for use in many instru-
  ments.  Some of the major types of instruments that use mercury are thermometers,
  thermostats, and thermoregulators;  flowmeters  (manometers);  pressure-sensing
  devices and barometers (now largely the nonmercury aneroid types); gages;  valves;
  pump seals; switches and relays;  navigational devices (gyroscopes, artificial
  horizons,  etc.); and medical devices (red  blood  cell  counters, blood CO2 analy-
  zers,  etc.).
       URS estimates that 81 percent,  or  199,932 kg, of the 246,612 kg of mercury
  consumed in instrument manufacturing in 1973 was used by the control instrument
  industry (SIC 3821)  and 19 percent (46,680 kg) was consumed  by the switch  and
  relay industry (SIC  3613).   About 55 percent (133,908 kg) of the mercury con-
  sumed by  the instrument industry was recycled or reclaimed  (secondary) mercury
  (Ref.  10).   The control instrument  industry losses during manufacture are
  estimated  to be 1  percent (2,000  kg), and  the switch  and relay instrument  losses
  are estimated to be  2.5 percent (1,169  kg).
      Overall,  the  mercury-using industrial instrument manufacturing industry is
  expected to grow 14  percent between 1973 and 1983.  With the introduction  of
  solid-state devices  and new designs,  however, the consumption of mercury for
  manufacturing switches  is expected  to decline 20 percent between 1973 and  1983.
  The relay and other  instrument  industries,   on the other hand, are expected to
  increase their consumption  of mercury by 14 to 20 percent.
      Production  Sites.  Table 34  lists  the  industrial instrument manufacturers
 for SIC 3613 and 3821 by state and  shows estimates of their mercury losses to
 the environment in 1973.  It is difficult to determine specifically what instru-
 ment manufacturers are of concern, because of the number of SIC groups involved
 and the varied types of instruments marketed within each SIC group.   The basis
 for this table,  however, was the value of shipments added by manufacture for
 the switch  and relay  industry (SIC 3613) and for  industrial control instruments
 (SIC 3821).   The results indicate that although the switch  and relay manufac-
turers used  only 19 percent of the two groups'  total mercury consumption in
1973, their  losses  (1,169 kg)  represented 37 percent of the total mercury losses
                                       186

-------
                             TABLE 34. - Estimated Mercury Losses from  Industrial Instrument
                                              Manufacturing, by State,  1973
                                                        (kilograms)
00


Region and State
Atlantic
Connecticut
Florida
Georgia
Maine
Maryland
Massachusetts
New Hampshire
New Jersey
New York
North Carolina
Pennsylvania
Rhode Island
South Carolina
Vermont
Virginia
Subtotal
North Central
Illinois
Indiana
TfMiffl
JtUM 0>
Kentucky
Michigan
Losses

Land

79
5
38
12
12
28
5
71
28
28
271
-
4
-
12
593

111
28
28
52
13
from Switch and
Manufacturing
Air

4
1
1
1
1
1
1
4
1
1
15
-
-
-
1
32

6
1
1
3
1
Relay

Total

83
6
39
13
13
29
6
75
29
29
286
-
4
-
13
625

117
29
29
55
14
Losses

Land

161
8
19
-
-
310
-
57
298
19
298
19
-
19

1,208

218
19
21
6
46
from Control Instrument
Manufacturing
Air

9
-
1
-
-
16
-
3
16
1
16
1
-
1

64

12
1
1
2

Total

170
8
20
-
-
326
-
60
314
20
314
20
-
20

1,272

230
20
22
6
48
Total
Losses

253
14
59
13
13
355
6
135
343
49
600
20
4
20
13
1,897

347
49
51
61
62

-------
00
00
                             TABLE 34. - Estimated Mercury Losses from  Industrial  Instrument
                                              Manufacturing, by State,  1973
                                                        (kilograms)
                                                        (continued)
Region and State
North Central
Minnesota
Missouri
Nebraska
Ohio
Oklahoma
Tennessee
Wisconsin
Subtotal
South
Arkansas
Mississippi
New Mexico
Texas
Subtotal
Mountain
Colorado
Subtotal
Losses from Switch and Relay
Manufacturing
Land Air Total

12
61
12
39
-
5
28
389

5
12
-
26
43

-


1
3
1
2
-
1
1
21

1
1
-
1
3

-


13
64
13
41
-
6
29
410

6
13
-
27
46

-
.
Losses from Control Instrument
Manufacturing
Land Air Total

*•
6
6
133
4
4
34
497

_
-
6
40
46

40
40

_ ..
6
6
7 140
4
4
2 36
25 522

-
^-
6
2 42
2 48

2 42
2 42
Total
Losses

13
70
19
181
4
10
65
932

6
13
6
69
94

42
42

-------
                             TABLE 34.   Estimated Mercury Losses from Industrial Instrument
                                              Manufacturing,  by State, 1973
                                                       (kilograms)
                                                       (continued)
00
Losses from Switch and Relay Losses from Control Instrument
Manufacturing Manufacturing Total
Region and State Land Air Total Land Air Total Losses
West
Arizona
California
Oregon
Washington
Subtotal
Total

6
68
4
5
83
1,108


1
3
-
1
5
61


7
71 104
4 6
6
88 110
1,169 1,901


7
6 110 181
6 10
6
6 116 204
99 2,000 3,169

      Source:  Derived by URS from 1967 Census of Manufacturers.

-------
  to the environment.  The control instrument manufacturers,  on the other hand,
  used 81 percent of the mercury but their in-plant losses were only 63 percent
  (2,000 kg) of the total.
       1973 Technology.  Figure 39 is a flow diagram of mercury losses during a
  typical control instrument manufacturing process,  and Figure 40 is a flow dia-
  gram of relay and switch manufacturing losses.   During control instrument manu-
  facture, mercury is lost primarily in the filling process,  when the tubes or
  receiver bodies are loaded with accurately measured amounts of mercury.   Losses
  occur through spillage (to land),  volatilization (to air),  and in cleaning old
  instruments that require servicing (to land).   Minor mercury losses,  mostly
  air, also occur when leaks develop during testing and storage of the instru-
  ments.  In total,  however,  we estimate that only  about 1 percent of the mercury
  consumed by the control industry is lost in the manufacturing process.   About
  95 percent of this loss is to land,  and the remaining 5 percent is to air.
  (Mercury losses to water are insignificant,  because water is not required during
  the filling process and the number of instruments  that must be water tested is
  very small.)
       In the switch and relay manufacturing process,  the principal mercury losses
  occur during  filling and testing (see Fig.  40).  Because switches and relays are
  smaller and use less mercury than  do most  of the control instruments,  losses to
  land during filling  total only about 50 percent  of the  manufacturing  loss,  and
  this loss  is  only  1.2 percent of the total mercury used in  the  manufacturing
  process.  About  the  same  percentage  of the total mercury used  in manufacturing
  is lost to land  when switches  and  relays are tested.  The remaining very slight
  manufacturing losses  (0.1 percent) are to  air and  are evenly divided between
  the filling and testing processes.
      The overwhelming majority of mercury  losses to the  environment that are
 associated with industrial instruments occur during consumer use  and disposal
 of  the finished products.  Industrial consumers recycle substantial amounts  of
 mercury back to the manufacturers  (approximately 55 percent), usually when
 instruments are traded in for new models or are serviced.  The most significant
 consumer losses occur when individual small consumers discard instruments—such
 as thermometers,  small manometers,  or switches—after they have served their
purpose.
                                       190

-------

Mercury
Input

Land Air Air
"t t4 'I
10.000 Filling 9,901 Testing

i " Process " Storage
t
\ '


9,900
5,445


Industrial 4,455 ^ Environmental
Consumer Loss



SOURCE: URS Research Company.
        Figure 39 FLOW DIAGRAM OF MERCURY LOSSES IN TYPICAL CONTROL INSTRUMENT
                MANUFACTURING PROCESS

-------
Is)

Mercury
Input


Land Air
'-t t-
10.000 _ FHIina £
i
Process
1 1



Land Air
119} fe
'•875 Testing and
" Storage
1 1
1 1


9,750
5,362


Industrial 4,388 Environmental
Consumer Loss



      SOURCE: URS Research Company.
                   Figure 40  FLOW DIAGRAM OF MERCURY LOSSES IN TYPICAL SWITCH AND
                            RELAY MANUFACTURING PROCESS

-------
     1983 Technology.  Little or no change is  expected  in the mercury-related
technology of instrument manufacturing by 1983.   Except for  increases  that
accompany the normal and expected growth of the  industry, mercury consumption
and losses to the environment should remain stable through 1983.   A 14-percent
growth rate is expected for the industry (and  hence for its  consumption of
mercury) between 1973 and 1973.  Therefore, Table 35 is based on  the assumption
that mercury losses to the environment will also increase by about 14  percent.
     Alternative Technologies.  Very little mercury is  discharged to the environ-
ment during industrial instrument manufacturing  (1.0 percent of the mercury con-
sumed by the control instrument industry and 2.5 percent of  that  consumed by the
switch and relay industry).  The significant mercury losses  to  the environment
occur when consumers dispose of instruments to local landfills.  Either mercury
would have to be replaced by other materials,  or mercury-containing instruments
would have to be completely recycled, in order to reduce present  loss  rates
significantly.
     Many substitute devices for certain mercury instruments are  presently
available.  Temperature-sensing devices which can be substituted  for mercury-
in-glass thermometers  (probably the most significant mercury-consuming instru-
ments) include alcohol-in-glass thermometers;  thermistors (devices in which
current changes are directly proportional to temperature);  thermocouples  (most
popular are copper-constantan, iron-constantan, chromalumel, and platinum/
platinum-rhodium); bimetallic  strips; and pyrometers (optical and radiation).
Replacements for mercury manometers  (a device for measuring fluid pressures
under steady-state condition)  and mercury barometers (a type of manometer used
to measure atmospheric pressures) include Bourdon tubes, diaphragms and bellows,
Bridgman and Pirani gages  (using electrical resistance), ionization gages (in-
cluding a vacuum tube), and alphatrons  (radioactive ionization tubes).  Many
solid-state devices are also  in development or  production which  may replace
other mercury instruments.  The problem in the  use of  these devices is to get
consumers to accept them  as desirable alternatives. Because mercury  instru-
ments are generally cheaper and  are highly reliable, instrument  users are
reluctant to change quickly to newer, more expensive,  and less proven instru-
ments .
                                     193

-------
            TABLE 35.  -  Estimated Mercury Losses  from Industrial
                     Instrument Manufacture  and Use,  1973 and 1983
                                  (kilograms)
Losses
to
Land
Air
Total
1973 Losses
from
1983 Losses from
Manufacturing Consumer Use
3,009
160
3,169


100
8
109


,780
,763
,543


Manufacturing
3,430
182
3,612


Consumer Use
114,895
9,991
124,886


Source:  URS Research Company
                                     194

-------
     The other alternative for controlling mercury losses  to  the  environment—
complete recycling of used mercury instruments—would be difficult to accom-
plish.  As indicated earlier,  most industries already recycle many of their
instruments.  If more emphasis is placed on recycling,  industries may increase
their recycling rate from the 1973 figure of 55 percent to as much as 75 per-
cent by 1983.  However, many private and small consumers  (who break  instruments
and spill mercury on the ground or down the sink)  cannot recycle  instruments,
and such losses will doubtless continue to represent a  major  discharge to the
environment.

     Paint Manufacturing (SIC 2851)

     The paint manufacturing industry consumed 7,603 flasks,  or 14 percent, of
the mercury used in the United States in 1973 (Ref.  14).   A small amount  (32
flasks) of this was used for mildewproofing substances, and the remainder
(7,571 flasks) went into paint additives.  According to the URS inventory
results, the total loss of mercury to the environment  from paint manufacturing
was only about 95 flasks (3,272 kg), or 1.2 percent of  the mercury used by the
industry.  The largest environmental loss occurs when  the  paint is used by the
ultimate consumer; overall losses for all paint usage  are  discussed  under
"Macrotechnology."

     The use of phenylmercuric compounds as preservatives and mildewcides dates
back to the late 1930s, but their use in latex paint did not become significant
until after World War II.  In the early 1950s, when paint film defects were
diagnosed as mildew growth, the paint industry started using phenylmercuric
compounds extensively.  However, latex paint did not become popular as a house-
hold paint until the early 1960s.  Various ingredients based on  other nonmercury
compounds were used but were abandoned because of adverse physical  characteris-
tics and inadequate performance  (most noticeably fluorescence, yellowing, and
viscosity changes within the paint).  Phenylmercuric compounds were found to be
more reliable and less expensive, and it was possible  to  use them in  low con-
centrations.  Today, phenylmercuric compounds  are  used chiefly as a preserva-
tive in water-thinned paint  (latex) to prevent bacterial  growth, and as a

                                       195

-------
  mildewcide in both latex and solvent paints for exterior surface protection to
  prevent fungus growth on the applied paint film.
       Phenylmercuric compound additives are used in small amounts (less  than
  0.02 percent by weight in interior paint and less  than 0.2 percent  in exterior
  paint) with supposedly low orders of toxicity.   The most common  mercuric addi-
  tives in paint are phenylmercurie acetate and phenylmercuric oleate (Ref. 79).
  Phenylmercuric acetate contains approximately 18 percent mercury and is
  slightly soluble in water (about 600 ppm),  while the phenylmercuric oleate
  contains approximately 11 percent mercury and is insoluble in water.
       The Federal Insecticide, Fungicide,  and Rodenticide Act requires that  all
  pesticide products be registered with the EPA before they can be marketed in
  interstate commerce.   Phenylmercuric compounds  used in paints are included  in
  this category.   Because of the  high mercury content of many U.S.  harbors (Ref.
  13),  which was suspected to be  connected  to losses  of  mercury from  antifouling
  paints,  most  mercury  compounds  used in marine antifouling paints  had been
  removed prior to March 22,  1972,  when the EPA published  an order  in the Federal
  Register (PR  Notice 72-5)  to suspend or cancel all  alkylmercury pesticides
  because  of their highly toxic and cumulative effects on  living organisms.
  Recently the  EPA has  reviewed this  use and has taken action to ban  the  mercu-
  rials completely.  The paint industry and other groups, however,  are appealing
  this ruling, and hearings are under way to determine if  the use of mercury
  should be  curtailed.
      Nonmercury paint additives have been recently marketed;  most of these
 products contain tin,  copper, or boron.  There are presently several disadvan-
 tages to using these substitutes in paint:  the cost of the substitutes is
 higher (estimated to be as much as 25 percent higher);  they must be used in
 high concentrations;  their effectiveness in paint,  both in the can and on applied
 surfaces,  has  not matched that of mercury; and their long-term environmental
 safety has  not been established.
     Production Sites.  Table 36 lists approximate  mercury losses to the environ-
ment from U.S. paint manufacturing operations in 1973.   As the table shows,
there are a great many manufacturing plants,  relatively evenly distributed in
                                      196

-------
TABLE 36 . - Mercury Losses from U.S. Paint
   Manufacturing Operations, by State, 1973
                 (kilograms)
Number of Establishments with

Region and State
Atlantic
Connecticut
Delaware
Florida
Georgia
Maine
Maryland
Massachusetts
New Hampshire
New Jersey
New York
North Carolina
Pennsylvania
Rhode Island
South Carolina
Vermont
Virginia
West Virginia
Subtotal
North Central
Illinois
Indiana
Iowa
Kansas
Kentucky
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Less than
20 Employees

15
4
71
37
3
28
52
3
154
151
18
68
10
4
2
18
3
641

150
30
10
6
23
67
24
62
2
1
More than
20 Employees

2
2
14
17
1
18
21
1
69
47
11
36
3
2
1
12
1
258

73
11
6
3
18
31
9
28
0
0
Mercury Losses to
Land

52
7
40
55
15
40
80
11
106
201
84
177
15
40
7
45
15
990

158
84
26
15
33
136
40
52
7
2
Air

6
1
5
7
2
5
10
1
13
25
10
22
2
5
1
5
2
122

19
10
3
2
4
16
5
6
1
<1
Water

14
2
11
15
4
11
23
3
30
56
24
52
4
11
2
12
4
278

44
24
7
4
9
33
11
14
2
I
Total

72
10
56
77
21
56
113
15
149
282
118
251
21
56
10
62
21
1,390

221
118
36
21
46
190
56
72
10
3
                       197

-------
TABLE 36. - Mercury Losses from U.S. Paint
   Manufacturing Operations, by State, 1973
                  (kilograms)
                  (continued)
Number of Establishments with

Region and State
North Central
Ohio
Oklahoma
South Dakota
Tennessee
Wisconsin
Subtotal
South
Alabama
Arkansas
Louisiana
Mississippi
New Mexico
Texas
Subtotal
Mountain
Colorado
Idaho
Montana
Utah
Wyoming
Subtotal
West
Alaska
Arizona
California
Hawaii
Nevada
Less than
20 Employees

101
15
1
19
31
542

14
8
9
6
2
89
128

12
5
2
5
1
25

0
2
217
1
2
More than
20 Employees

52
5
0
7
9
252

6
2
3
2
0
42
55

5
2
0
3
1
11

0
1
90
0
0
Mercury
Land

172
15
2
55
62
859

40
22
22
22
2
91
199

15
7
2
7
2
33

0
9
191
0
2
Air

21
2
<1
7
8
104

5
3
3
3
<1
11
25

2
1
<1
1
<1
4

0
1
23
0
<1
Losses
Water

48
4
1
15
17
239

11
6
6
6
1
26
56

4
2
1
2
1
10

0
3
53
0
1
to
Total
•
241
21
3
77
87
1,202

56
31
31
31
3
128
280

21
10
3
10
3
47

0
13
267
0
3
                    198

-------
                TABLE 36.  - Mercury Losses  from U.S.  Paint
                   Manufacturing Operations., by State,  1973
                                 (kilograms)
                                 (continued)
Number of Establishments with
Region and State
West
Oregon
Washington
Subtotal
Total
Less than
20 Employees
16
25
263
1,599

More than
20 Employees
9
11
111
687

Mercury Losses
Land
17
33
252
2,333

Air
2
* 4
30
285

Water
5
9
71
654

Total
25
46
353
3,272

Source:  Derived by URS from 1973 Census of Manufacturers.
                                      199

-------
  proportion to population across the United States.   Analysis  of the table
  indicates that 42 percent of the mercury losses in  paint manufacture occurred
  in the Atlantic region,  37 percent in the North Central region,  9  percent in
  the South, 1 percent in the Mountain region,  and 11 percent in the West.
       The U.S. paint industry has a large number of  small "mom and  pop"  manu-
  facturers (less than 7,000 gallons per day).  However,  the larger  manufacturers
  account for the majority of production.   Establishments with  an average of 19
  or less employees account for only about 8 percent  of the value of shipments
  added by manufacture,  even though such plants comprise  70 percent  of all manu-
  facturing establishments.
       In 1973 more than 180 million gallons of exterior  paint  (98 million gal- .
  Ions of latex and 82 million gallons of  oil/alkyd)  and  160 million gallons of
  interior latex paint were  manufactured by the paint industry.*  The industry is
  expected to grow at a  5-percent annual rate through 1983.  On this  basis,  178
  million gallons of exterior paint and 211 million gallons of  interior paint will
  be formulated and sold in  1983.
       1973  Technology.  Figure 41 is  a flow diagram  showing mercury  losses during
  typical paint manufacturing operations.   The  principal  loss occurs  during formu-
  lation when  the ingredients are  mixed in large metal vessels.    (The  size  of
  these  vessels varies from  a few  hundred  to several  thousand gallons.)  Two
  types  of losses occur  during the process—bad batch losses and losses due to
  cleaning of the vessels.   Industry representatives  indicate that upwards  of  3
 percent of product is  lost during formulation.  However, we estimate  (conserva-
 tively)  that only 1.2 percent of the product  contains mercury.   (This lower
 level of loss was selected because phenylmercuric compounds, which are very
 expensive compared to the other paint ingredients,  are added in the later steps
 of paint formulation; therefore most of the bad batches will have been detected
 before the phenylmercuric compounds are added.)
      The largest share  of the manufacturing loss of mercury (some 71 percent  of
 total manufacturing losses) is to land.  Most of this type  of  loss is from the


* Interior oil-based  paints are not included because they contain relatively
  little mercury.
                                       200

-------


Phenylmercuric 1000 _
Compound


Air Water
't t«
. - Q8

•I
Landfill


9 Inventory 9E
Spoilage
. 1
Landfill


8



SOURCE: URS Research Company.
     Figure 41  FLOW DIAGRAM OF ESTIMATED MERCURY LOSSES
              IN TYPICAL PAINT MANUFACTURING OPERATIONS IN 1973
              (for manufacture of 250,000 gallons of latex paint)
                    Air   Water

                   J
fill   vvaiczf
iL_t
Phenylmercuric 1000
Compound
Formulation
886

Inventory
Spoilage
880

Consumer
10o| J6
Landfill Landfill
SOURCE: URS Research Company.
      Figure 42  FLOW DIAGRAM OF ESTIMATED MERCURY LOSSES FROM
               TYPICAL PAINT MANUFACTURING OPERATIONS IN 1983
               (for manufacture of 250,000 gallons of latex paint)
                                  201

-------
  settled solids collected from bad batches and vessel washwater.   Other  losses
  to land are from inventory spoilage, spillage, and discarded washwater  filters.
  These solids are generally disposed of in nearby landfill  sites.
       Approximately 20 percent of paint manufacturing mercury losses  are to
  water.  Most washwater and bad batch water is allowed  to settle  (to  remove
  solids) and is then filtered (for more solids removal)  and discharged into the
  sewer system.  Phenylmercuric acetate, the mercury additive used  most in paint,
  is slightly soluble.  Phenylmercuric oleate,  however,  is nonsoluble  in  water.
  Cumulatively, the most significant offenders  in water  discharges  are the small
  "mom and pop" manufacturers, who tend to flush waste down  sewers  rather than
  collecting the solids for proper disposal.  Mercury dissolved in  solvents  is
  generally not a major problem,  because most manufacturers  store these waste
  liquids in 55-gallon drums for  recycling back to solvent distributors.   Except
  in very large plants (80,000 gallons of paint per day)  it  is not  economically
  feasible to recycle washwater for use in subsequent batches.
       The remaining 9 percent of the manufacturing loss  is  to air.  These losses
  are brought about by volatilization of mercury from water, paint  formulation,
  spillages, etc.
       Final consumption accounts for the major loss  of the mercury used  in
  paint.   Of the 7,571 flasks  used as water-based paint additives in 1973, an
  estimated 65 percent of the  mercury was emitted,  probably as Phenylmercuric
  acetate  or similar compound,  to air.   Minor quantitities were discharged to
  water (via sewers)  and to landfills;  the balance  remained  in the  applied
  paint.   Estimated  losses  to  air after paint application are 3.8 to 4.0  yg/m
  immediately after painting,  1.3 to  1.7 yg/m  up to  200 hours after painting,
  and 0.07 yg/m  after  three years  (Ref.  80).  Mercury emissions from  latex
 paint have been observed  to occur seven  years after application;  the rate  of
 emission is proportional  to temperature  increases.  Losses to land occur when
 paint is spilled in the soil, or when paper, rags,  cans, and brushes are dis-
 carded,  or when old, partially used paint is thrown away.  Losses to water
 occur when paint brushes and cans are cleaned; these residues are flushed
 directly into sewer systems.
      1983 Technology.  Figure 42 is a flow diagram of estimated mercury losses
during paint manufacture in 1983.  It is assumed that the 1983 technology  is
essentially the same as the 1973 technology, but that paint manufacturers  will

                                      202

-------
add flocculents to washwater to settle out dissolved mercury.   This  residue,
along with other solids, will be disposed of in nearby sanitary landfills.   It
is estimated that in total, the manufacturing mercury losses will be about  89
percent to landfill, 8 percent to air, and 3 percent to water.
     It must be noted here that the EEA is considering the complete  curtailment
of all mercury pesticide use in the United States.   If this occurs,  mercury
will no longer be used by the paint industry in 1983, and Figure 42,  of course,
would not apply.
     Table 37 compares estimated losses of mercury from paint  manufacturing and
consumer use for 1973 and 1983.  The "low technology" level shown in the table
would occur if the paint industry grows at an annual rate of 5 percent and
mercury usage decreases by 50 percent.  If this situation did indeed material-
ize, the use of mercury would remain stable at approximately 7,600  flasks
annually.
     Alternative Technologies.  Very little mercury is discharged during paint
manufacture (1.2 percent of the mercury consumed),  but substantial  discharges
to the environment occur when consumers use the paint.  If these losses cannot
be tolerated in the future, less toxic or nontoxic additives will have to be
substituted for mercury.
     Many nonmercury substitutes are currently being used by manufacturers;
the more promising  substitutes contain tin, copper, or boron.   Some of the
additives that are considered less toxic than mercury are tributyl tin oxide
(or esters), .triaryl bismuth, triaryl boron, cuprous oxide, copper 8-hydroxy-
quinalinalate.  zinc chromate,  and metal  chelates of acetoacetic ester (Ref. 77)
However, many paint representatives are  not completely  satisfied with mercury
substitutes.  The disadvantages cited include:  substantially  higher  cost;
higher concentrations required  (1 to  2 percent by paint weight); substitutes
are not as effective in preserving paint in the can;  paint properties are
sacrificed  (pigment discoloration, viscosity  changes, bad odors, fluorescence,
etc.); substitutes  are  not as  effective  in preventing mildew  growth on painted
surfaces; and sufficient field testing has not  been done.  Another  problem in
finding a quick pesticide  substitute  for mercury  is that the  product should
have no adverse long-term  environmental  properties.  However,  it should be
                                       203

-------
                TABLE  37.  -  Estimated Mercury Losses
                   from Paint  Manufacture and Use
                            1973 and 1983
                              (kilograms)
Losses
to
Land
Air
Water
Total
1973 Losses
from
Manufacturing Consumer Use
2,333
285
654
3,272
8,140
173,610
1,000
182,750
1983 Losses
from
Manufacturing Consumer Use
2,890
285
97
3,272
8,140
173,610
1,000
182,750
Source:  URS Research Company
                                    204

-------
noted that most paint manufacturers recognize that effective substitutes
will be developed and are actively—even though reluctantly—looking for
them.
     Most states have ignored the use of mercury in paint,  although these
same states generally have stringent limits on the use of mercurial pesti-
cides.  According to Ref. 146, only Connecticut has an outright ban on the
use of mercurials in paint (however, this ban is not being  rigorously en-
forced) .  In New York all mercurial pesticides are banned,  except those in
paint.  In Illinois all manufacturers who use 15 or more pounds of mercury
product per year must register with the Air Pollution Board; also, discharges
to water must not exceed 1 ppb.  However, neither of these  constraints is
an outright ban and mercurial-containing paints continue to be manufactured
(by obtaining exceptions to discharge criteria) and used in Illinois.  In
general it appears that the use of mercurials in paint is either  ignored or
discouraged.  But, coupled with the action of the federal government, the
manufacturers themselves have moved to switch to nonmercurials in expectation
of increasingly stringent regulations.  But as discussed elsewhere, manu-
facturers-have had problems in using the substitute additives, and as a result
the state has already allowed a number of exceptions.  Texas also has dis-
charge requirements on the release of mercury to water which are not as strin-
gent as those in Illinois.  No adverse impact on paint manufacturers in Texas
has been reported, possibly because of nonenforcement of the law.

Sector VIII - Commercial Final Consumption

     Agricultural Pesticides

     Agricultural use of pesticides for seed dressing  is based on the
necessary control of about 1,100 disease-producing (pathogenic)  and 95 dis-
coloring organisms found in or on seed  (Ref.  2).   Of all chemicals developed
in the last 50 years, none possess as broad a spectrum of  fungicidal activity
as do the mercurials.  Organomercurial dressing of wheat,  barley, oat, and
rye seed is generally recognized as a factor  contributing  to increased cereal
grain production throughout the world.   Though effective substitutes
exist for all antipathogenic  functions of the organomercurials,  the
                                      205

-------
  mercurials are effective  at considerably lower concentrations  (Ref.
  81)  and therefore cost less than most  (if not all) alternatives.
       After the 1970 USDA  ban on all alkylmercury pesticides, phenylmer-
  curic acetate (PMA) emerged as the dominant organomercurial used
  in agriculture.  Though PMA does not have the antifungicidal range
  of the alkylmercurials, it is demonstrably less persistent and
  less toxic to terrestrial and aquatic  life, and it has efficiency
  advantages over substitute compounds in many limited seed treatment
  applications.   Nonetheless, awareness  of the environmental factors
  involved,  together with the development of viable alternatives, has
  led to a significant decrease in use of mercurial chemicals in agri-
  culture.   Even before  the 1970 ban the agricultural use of mercury was
  lessening—it declined 10 percent in 1968, 20 percent in 1969, and
  33 percent in 1970 (Ref.  35).  Mercury use in 1971 was only one-third
  as much as in 1966 (Ref.  83).  The 1970 suspension of alkylmercurials
  and  the probability of additional action against the remaining organo-
  mercurials have  accelerated this trend.  In 1973 usage was only about
  21 tons* and an  even lower tonnage is expected in 1975.
       PMA is manufactured  almost exclusively in the New Jersey region,
  and most of the  PMA for seed treatment is ultimately dispersed through-
  out the major wheat and barley producing states in the Midwest.  Accord-
  ing to the URS national inventory, the largest mercurial pesticide
 users in 1973 were North Dakota, Kansas, Montana, and Oklahoma.
      At present, PMA and all other mercurial pesticides are under
 cancellation by EPA.  These compounds can be distributed, however, un-
 til the EPA administrator officially suspends them, providing that
 evidence presented at the pesticide hearings in Washington does not
 warrant reversal of the cancellation.   Whatever the result, the can-
 cellation has  had a marked effect upon the mercurials.  Manufacturers
 have  dropped most of their mercury registrations and dealers have
 curtailed their supplies to avoid holding quantities of "inevitably
restricted" material.
* ORS reapportionment of Bureau of Mines"  statistics.
                                     206

-------
     As more efficient and less expensive compounds  are developed,
and as the mercurial threat to the environment is more clearly
demonstrated, usage will decline with or without the ban.   By 1983,
this usage could be less than 9 tons annually, even  without further
EPA restrictions.
     The Bureau of Mines predicts a 75.8-ton annual  "agricultural"
use by 1985 without any further EPA restrictions (Ref. 14).  When
the URS 70/30 nonagricultural/agricultural ratio is  applied to the
75.8—ton estimate  (which includes both agricultural  and  nonagri-
cultural mercury biocides), a 1985 forecast of about 23  tons for
seed treatment is established.  The URS 10-ton forecast  for 1983
is based primarily on the  current downward trend in agricultural seed
treatment use of the mercurials.  Because of the efficiency and eco-
nomics of mercuric pesticides use, that trend would gradually be re-
versed, unless there is an EPA suspension.  Thus, consumption can be
expected to exceed 9 tons  and to approach 23 tons after 1983.
     Assuming a further EPA action against the mercurials, the
Bureau of Mines estimates  a probable usage of 17 tons and a low of
10.3 tons for all mercuric pesticides.  These figures are based on
continued utilization in  some applications under strict governmental
control, almost exclusively for industrial purposes where  foodstuffs
are not involved.  Thus,  there will probably be no  consumption  for
seed treatment by  1983, if the 1975-1976 EPA  suspension materializes.
     Because there is a large  supply of organomercurials,  elimination
of these pesticides from  agriculture will not be abrupt.   The least
damaging means of  disposing of them would be  through the  standard
treatment of wheat and barley, seeds.   Draw-down of  the  organomer-
curial  stores could take  several years, but by 1983 mercurial pesti-
cide usage  in agriculture would  be effectively zero, if there is an
all-inclusive EPA  suspension*
                                       207

-------
      The beneficial and detrimental factors associated with the seed
 treatment are more thoroughly discussed in the following references:

      Ref. No.
         2      Overview of mercuric pesticide usage—especially
                translocation in plants and bioaccumulation
        84      Technical summary of chemical form, method of
                application and fate of the organomercurials in
                agriculture
        85      Translocation and bioaccumulation
        86      Efficacy, application, bioaccumulation, and re-
                sults of the methylmercury ban in Sweden
        81      Relative effectiveness of PMA

      1973 Technology.   The path and fate of mercury in agricultural pesticide
 usage are diagrammed in Figure 43.   The manufacture and formulation of mercuric
 pesticide compounds (principally PMA)  result in a limited loss to water.   In
 the figure,  the 98-percent efficiency value assigned to formulation represents
 a worst-case situation.   Development of new processes in the last 20 years
 which are responsive to the properties of the chemicals used has made the seed
 treatment process extremely efficient.   The 2-percent loss to soil is attribut-
 able to cleansing of the seed-treater (a "mistomatic" when PMA is used),  dis-
 posal of PMA-30 containers,  and washing of trucks used to transport treated seed.
      Most  of the  mercury lost to the environment is lost after the seed is
 planted.   Depending  on the  chemical composition of the pesticide used,  and thus
 on differing affinities  for individual components of the soil,  a portion  of
 the organomercurial compound on the seed will be leached out of the soil  and
 lost to water.  The phenylmercurials are not  immobilized in the upper few inches
of soil as are the alkylmercurials.  Consequently,  the phenylmercurials penetrate
                                     208

-------
                                                                              t
t
1
28
Wildlife
1 '
1000 999 . 979 959

1 1 20 20 144 1 J57
Water Water Soil Water Soil
!O • 	 _fT... _..__ 	 , „_-„, -I, - 	 _ , M
O
VD
1
1
2
Man
•L
211

5 19
1
Water !
•





190
i !
Soil

      Losses to
      Water   185
      Soil    785
      Wildlife
      and man  30
            1000
—— Secondary losses to air =145
SOURCE: URS Research Company.

                                                                                 *
                    Figure 43  MERCURY BALANCE, AGRICULTURAL PESTICIDE MANUFACTURE
                             AND USE, 1973 TECHNOLOGY

-------
  deeper into the surface and can be leached more readily.  Approximately 22
  percent of the mercury on the seed is incorporated in the plant tissue,  though
  only 1 percent of that quantity is ultimately translocated  into the  edible
  grain portion (Ref. 85).  More significant translocation occurs after foliar
  application of mercuric pesticides to control a wide  range  of  fruit  and vege-
  table pests (residues in crops four to six times higher than unsprayed controls;
  Ref. 2).  Recognition of the environmental factors involved led to an EPA  sus-
  pension of PMA as a foliar crop spray.  Thus,  the use of mercury compounds for
  this purpose is assumed to have disappeared by 1973.
       Despite leaching,  translocation,  and  loss to seed-eat ing birds,  more  than
  half of the mercury on the treated seed remains in the soil.  In addition,  most
  of the mercury transferred to the  plant is plowed under after harvest and
  returns to the soil via decomposition.  Overall,  more than  80 percent of all
  mercurials used for agriculture in 1973 were  incorporated in the soil.
       Not all the mercury lost to the soil  or  to water remains there  indefinitely.
  Approximately 15 percent of the mercury in the soil is lost as  volatile  mercury
  vapor to air each year (Ref.  90).  A comparable loss  figure is  postulated  for
  mercury in water.   Similarly,  the  quantities of mercury assigned to  wildlife
  and man in Figure 43 do not connote  the ultimate  fate  of the mercury.  Most of
  the mercury entering the biological  food chain is  lost as waste, with the per-
  sistence  of the  phenylmercurials substantially below  that of the alkylmercuric
  compounds commonly regarded as hazardous to aquatic and terrestrial  life.
       1983  Technology.  Without a mercuric pesticide ban,  a  flow  sheet for  1983
 would reflect  1973  types and percentage losses.  Because agricultural pesticides
 are produced expressly to be applied to the environment,  little can be done to
 control mercury  dispersion  through that environment short of restricting the
 initial application.  Control of mercury loss during formulation and  seed treat-
 ment would be counterproductive, in view of the negligible  loss  and  the  high
 payoff in increased yields.  The only major difference between  the 1973  and
 1983 technologies would be in the initial quantities of mercuric pesticides
 entering the system.  As mentioned previously, it is likely that the  1983
 figure will be  less than half of the 1973  consumption.
     Alternative  Technologies.   The EPA contends that there are viable alterna-
tives to organomercurial seed treatment for disease control.  In most seed
                                       210

-------
treatment applications, the American farmer has switched to substitute com-
pounds.  These substitutes and effective combinations thereof are registered
with the EEA (FIFRA Docket No. 246 et al.f  USEEA,  before the Administrator:
Notice of Substitutes for'Mercurial Seed Treatment,  1975).   Unfortunately,  the
safety of many of the alternative compounds presently used in agriculture
remains in question (Ref. 15).  A growing number of farmers have also opted to
plant untreated seeds in recent years and,  as a result,  reports of unharvested
acres due to a variety of diseases (e.g., smut) that are normally controlled
by seed treatment have increased.  As more effective alternatives are developed
and produced and as these alternatives gain acceptance,  the loss in cereal
grain yield will be diminished and/or eliminated.   Until the efficacy and quan-
tities of substitutes required match the efficiency of the mercurials, though,
there will be increased costs to the commercial seed treater (e.g.,  machine
replacement), the farmer  (e.g., mercuric pesticide/substitute price difference),
and the consumer.  The cost of mercury substitutes for seed treatment can
range from a savings of $1 million to a $15 million increase (for 108 million
acres of grain, assuming  50 percent treated seed).  Costs to consumers are
estimated at 10 cents per capita per year.*
     Most, if not all, of the problems with mercuric pesticides in agriculture
have been a result of improper use of, or accidental exposure to, alkylmercury-
treated grain.  Undeniably, methylmercury in seed dressings did represent a
direct hazard to wildlife and an indirect hazard to man.  Comparable  problems
with the other organomercurials  (FMA) are feared.  The EEA contends that the
risk associated with the  use of mercurial pesticides outweighs any benefit  of
such use and that the existence of preferable  alternatives mandates mercuric
pesticide removal.  For most uses of the mercurials, such  a  contention is
realistic, but so far not all available  alternatives are "preferred"  by the
farmer.  Farmers sold  on  the  efficiency  of  mercury treatment contend that,
until the economic and efficacy gap between the mercurials and best substitutes
is closed for all seed treatment applications, a  governmentally supervised
* Hertzmark  (1974),  as quoted in Ref.  15.
                                       211

-------
  "limited use" approach might be more productive,  in view  of  the  world food
  shortage.  At the International Congress on Mercury (1974),  L. J. Goldwater
  summarized the basis for this contention:

            "More than twenty years devoted to studying  the effects
            of phenylmercurials on humans had convinced  me  that these
            compounds,  when properly used,  present  no hazard to the
            user nor to the environment.   In agriculture the phenyls
            can replace the alkyls in protecting  seed before germina-
            tion and in controlling a number of plant diseases caused
            by fungi.  I am not an expert in plant  pathology,  but  I
            am told by competent authorities that the phenylmercurials
            cannot be matched by other fungicides and that  their use
            adds substantially to yields  of  food  and  fibre.  Senseless
            and unnecessary curtailment of the use  of these valuable
            chemicals resulting in decreased production  and consequent
            increase in prices would be one  more  victory for the anti-
            mercurialists."

       Nonagricultural  Pesticides

       In 1973,  larger  quantities of mercury-bound  fungicides were used for non-
  agricultural turf treatment than for  agricultural seed treatment.  The nonagri-
  cultural/agricultural (70/30 percent) pesticide breakdown in the UBS national
  inventory reflects the ubiquity and frequency of nonagricultural use.  Mercurial
  fungicide treatment of turf is  practiced throughout the United States and
  requires  a continued  series of  applications to be effective.   The cost, range,
  and effectiveness  of  the mercurials have led to their widespread use as turf
  fungicides.  In addition mercury-containing pesticides have a limited applica-
 tion to control Dutch elm disease.  Except in paint, other nonagricultural uses
 of mercuric pesticides (e.g., as wood preservatives) are assumed to have declined
 to insignificant proportions by 1973.
      Parasitic fungi are the major disease-causing agents that are effectively
 controlled by mercuric compounds.  The mercurials  act as protective-contact
 fungicides.  Application to seed, foliage,  or soil can provide a surface shield
against  fungus spores. Foliar spraying, which treats the turf soil indirectly,
is the principal means of application; treatment of  grass seed is extremely
rare.  Commercial  turf operations depend heavily upon the organomercurials.
                                       212

-------
Golf courses, parks, cemeteries,  and lawns historically have received most  of
the nonagricultural inorganic and/or organic mercury fungicide treatment.   By
1973, only golf course greens, tees,  and fairways were still receiving  signifi-
cant amounts of mercury.
     The 1970 alkylinercury ban, the current mercuric pesticide cancellation,
and the rapid emergence of substitute compounds have reduced the nonagricul-
tural use of mercury-based fungicides.  Recognition of mercury accumulation in
the fish in golf course lakes due to runoff and percolation, coupled with a
general concern about the hazards of mercury's entrance  into the biological
food web, has accelerated the abandonment of  these  chemicals.
     The 1973 figure of 44.0 tons of mercurial fungicides assigned to nonagri-
cultural use is considerably below the consumption  averages of the 1960s.   User
dependence upon these fungicides will decrease still further during the next
few years as major producers continue their cutbacks but,  without  an EPA ban,
it is likely that this use will stabilize at a level only slightly below the
1973 total.  Nonagricultural turf treatment with the mercurials  probably will
not decrease as much as agricultural seed treatment, principally because (1)
the nonagricultural threat to man and the environment is not  as  direct, (2)
considerably more money is invested in the turf industry, and (3)  the activity
range of the alternatives has not yet been as thoroughly proven as have the
seed-treatment alternatives.  The Bureau of Mines  (Ref.  14) forecasts a high
of 53.1 tons of mercurial fungicides for 1985.
     If a suspension is imposed, usage will definitely diminish, but not all
organomercurials may disappear immediately.  Use of  stored mercurials  to combat
severe disease problems will  continue at  least through the 1970s.  Though
alkylmercurials were banned in 1970, these stored  mercury compounds are still
used occasionally in 1975.  As mentioned previously, gradual use  of the organo-
mercurial stores in this manner  is the  least  damaging method of removal.   The
Bureau of Mines does not expect  a complete elimination of all nonagricultural
uses of mercuric pesticides and  consequently  predicts a  probable  consumption
of 14 tons per year for limited  nonfoodstuff  applications.
     1973 Technology.   Figure 44 shows  the movement of  mercury  into the environ-
ment as a result of nonagricultural pesticide uses. Because other pesticide
                                      213

-------
10
»••
*.
                   Losses to
                   Air    98
                   Soil   519
                   Water  383
                       1000

            "•—•"•  Secondary losses to air« 135

            SOURCE: URS Research Company.
                                               *
                   Figure 44  MERCURY BALANCE, NONAGRICULTURAL PESTICIDE MANUFACTURE
                            AND USE, 1973 TECHNOLOGY

-------
sources contribute a relatively insignificant amount of mercury  to  the  environ-
ment, the figure shows only the mercury lost in turf treatment.  Manufacture
and formulation of the inorganic and organic (FMA)  chemicals  required for  turf
application parallel the agricultural loss figures.   After formulation,  though,
the similarities between agricultural and nonagricultural usage  end.
     As the inorganic and organic fungicides are applied,  a portion of  the
mercury is lost in aerosol form directly to air during the spraying process.
Similarly, mercury is lost to air as volatile mercury vapor during  the  weeks
following application.  The total amount lost to air depends  on  climatic condi-
tions during application and on the chemical form of the fungicide.   The URS
10-percent estimate used in Figure 44 represents a maximum average  loss to air.
     Uptake of mercury by plants treated with protective-contact materials is
much less than plant absorption of the systemics.  Though as  much as  50-percent
absorption of PMA has been reported following foliar application to rice plants
(Ref. 2), a 20-percent figure is more realistic for grasses,  considering their
morphology and the limited residence time of the pesticide on their surface due
to runoff loss.  Most of the 20 percent may actually remain at the  base of the
grass blade or in the upper portion of the turf mat.  Nonetheless,  even signifi-
cantly reduced levels of the pesticide spray on the plant's surface can afford
effective protection.  Repeated applications ensure the continued presence of
adequate pesticide residues.
     The key aspect of Figure 44 is the amount of mercury lost to water (approxi-
mately 40 percent).  The initial presence of the pesticide on exposed surfaces,
the high frequency of watering, and the unusual configuration of golf course
terrain combine to make mercury-loaded surface runoff a  significant  factor.
Results from a University of Missouri study on golf course lake contamination
indicate .significant pollution of the lakes due to  greens treatment  (Ref.  91).
A portion of the mercury not immediately  carried to the water environment moves
down through the soil.  A fraction of this  mercury will,  in  turn,  be leached
from the soil and enter the groundwater  system.  Leaching from  the soil is sub-
stantially increased when organically bound mercury is used  in the treatment.
Inorganic mercury tends to be  more  efficiently bound by the  soil  and to remain
in the top few inches of earth.  Because of these factors that encourage mercury
                                        215

-------
 movement into the water, equal and significant loss to land and water during
 and immediately following application is postulated.
      The overall loss to soil (> 50 percent) exceeds the mercury loss to water
  (< 40 percent) because most of the mercury tied to turfgrass is mowed and
 removed every few weeks.  The grass cuttings are either returned to the soil
 as mulch or dumped.  Approximately 10 percent of the mercury associated with
 the cuttings is leached from the soil during and after decomposition.
      Though there is no direct consumption of mercury by man,  there is evidence
 that a portion of the mercury which enters the water environment is eventually
 accumulated in the aquatic food web of golf course streams and lakes.  The type
 and extent of such bioaccumulation are not significant enough to threaten man
 but, if preventive measures are needed,  they are readily available (e.g.,
 prohibition of fishing in such lakes and streams).
      1983 Technology.  No changes in the ultimate fate pattern of nonagricul-
 tural pesticides are expected by 1983.   As previously stated,  initial amounts
 of mercuric pesticides used for turfgrass treatment may be slightly lower by
 1983,  but control of the mercury as it is dispersed through the environment is
 not possible,  due to the general nature  of pesticide use.
      Alternative Technologies.   Successful growth and upkeep of disease-free
 turfs depend more upon organomercurial availability than do growth and upkeep
 of cereal  grains.   Nonetheless,  alternatives for turfgrass fungicides do exist
 and are being used throughout  the  country (FIFRA Docket No.  246,  et al.,  USEFA,
 before the Administrator:  Notice  of  Substitutes for Mercurial Turfgrass
 Fungicides).  Unfortunately, the economic burden and doubts concerning efficacy
 of these alternatives hinder their complete acceptance.  Though no health
 problems have been attributable  to turfgrass use of mercuric pesticides,  the
 EEA position on mercury use for  nonagricultural  purposes is consistent with
 its position regarding agricultural seed  treatment:   the risks associated with
 the use of mercurial pesticides  outweigh  any benefit of  such use,  and the
existence of preferable alternatives mandates mercuric pesticide  removal.
                                      216

-------
     Laboratory Uses

     Metallic mercury and mercurial compounds are used for many  laboratory pur-
poses.  They are standard laboratory items as reagents and indicators,  for
calibration anc3 sealing, and in vacuum pumps (Ref.  68).  Mercury is  used  pri-
marily for radioactive diagnosis and as a fixative for tissues in hospitals.
According to Ref. 68, radioactive mercury is available for brain scanning for
tumors and for renal scanning.  Mercuric chloride has been used  in the  preven-
tion of suture line recurrences in anterior resections for colonic cancer
(Ref. 68).
     University and high school chemistry laboratories use mercury in class
experiments and independent research projects.  Due to a growing awareness of
the potential adverse effects of mercury flow into the natural environment,
research into the chemical properties and toxicity of mercury apparently  has
increased during the last decade.  Use of mercury compounds  in university and
high school chemistry programs, however, is generally acknowledged to be
decreasing.
     According to Ref.  12, 22,680 kg of mercury were consumed in U.S. labora-
tories in 1973.  This tonnage was considerably below the 1971 use (62,325 kg)
and about one-half the  1965-1968 average annual consumption of 43,546 kg
(Ref. 43).  The marked  consumption decrease between 1971 (62,325 kg) and 1972
(20,503 kg) is not explained by any governmental restrictions or variations in
method of accounting.   Significant fluctuations, though not of this magnitude,
were evident in other years.  The general  decline  in  consumption  in the  last
ten years probably reflects a more conservative approach to mercury use  in the
laboratory and more  efficient recycling of the mercury that is  used.
     Current laboratory uses  for mercury will not  diminish markedly by 1983.
Consumption will decline gradually as  efforts to recover and recycle mercury
efficiently are accelerated.  This decline may be countered, however,  by
increased demands from  hospital and  research laboratories caused by population
growth and expanded  research  efforts on mercury.   Considering all these factors,
URS forecasts a  1983 laboratory consumption of  approximately 18,000 kg.
                                        217

-------
       1973 Technology.   The  flow of mercury into and through the laboratory is
  diagrammed in Figure 45.  The  diagram represents a cumulative accounting of
  the different types of  laboratory usage:  university and high school, indepen-
  dent research,  and hospitals.  Thus, the percent loss to a particular medium
  is indicative of laboratory use in general and not of specific laboratory
  applications.
       Mercury emissions  to air  resulting from hospital uses probably exceed 10
  percent of the mercury  consumed in these applications because of the hospital
  practice of incinerating tissues, bandages, and other used supplies, some of
  which may contain traces of metallic mercury or a mercuric compound.  Univer-
  sity and high school laboratories and independent research laboratories prob-
  ably lose less  than 10 percent of their mercury to air.  Overall,  therefore,
                                                                  •
  the 10-percent  loss to the  atmosphere assumed in the URS inventory is probably
  realistic.
       Mercury losses to land could occur through the disposal of mercury con-
  tainers,  toweling used to absorb spilled mercury,  and defunct vacuum pumps,
  among other  substances.  Such  losses would tend to be greater for university
  and high  school laboratories, where incineration is not practiced and inexper-
  ience with the handling of chemicals is greatest.   URS estimates that as much
  as  7  percent of the mercury used in laboratories may ultimately be lost to land.
      Most laboratory mercury that is not recovered and recycled is lost to
 water.  Fixative solutions  (mercuric chloride)  are commonly flushed down sinks.
 Antiseptic applications of mercuric compounds in the hospital operating room
 are also eventually washed away.  Accidental spills and the cleansing of labora-
 tory containers and instruments used in mercury experiments also contribute to
 mercury discharges to water.  Extensive precautions against dumping mercury
 solutions down sinks can prevent the largest potential source of mercury contami-
 nation from reaching the water environment.   University and high school labora-
 tories,  cognizant of the environmental  factors  involved,  are extremely efficient
 at collecting and recycling  most of  the mercury they use.   However,  occasional
 dumping of residual  mercury-containing  solutions will continue to occur as long
as mercury and mercury derivatives are  used  in  the  laboratory.
                                       218

-------
    Losses to
    Air    100
    Land    70
    Water  260
    Recycle 570
1000
        1,000
        SOURCE: URS Research Company.
570
                                    260
Figure 45   MERCURY LOSSES FROM LABORATORY USAGE, 1973
    Losses to
           30
   Water   120
   Recycle 800
         1000

                                                 800
                                     120
        SOURCE: URS Research Company,
Figure 46  MERCURY LOSSES FROM LABORATORY USAGE, 1983
                             219

-------
       Laboratory recycling prevents more than half of the incoming mercury
  stock from being discharged to air,  land,  or water.   University laboratories
  have achieved an efficient level of mercury recovery through the use  of stock-
  room regulations requiring that fresh mercury be  issued only in amounts equal
  to the mercury being returned.   Though initial requests for mercury are filled,
  control of any additional requests is facilitated by this type  of exchange
  approach.  In many instances,  individual laboratories in chemistry and biology
  building are monitored in a similar fashion,  though  the use of  mercury in small
  research projects is more easily controlled than  are miscellaneous and class-
  room uses.
       With the emphasis on recycling during the last  decade,  the university
  stockrooms can survive on a "fresh"  mercury supply purchased every five to  ten
  years.   Thus,  for a typical usage of 200 pounds of mercury  and  its compounds,
  the individual university laboratory loses 20  to  40  pounds  per  year (10 to  20
  percent).   These yearly losses  appear in the national inventory only  in the
  years when restocking occurs.   The 200 pounds  which  would appear in those
  years,  however,  would not represent  the  real mercury use, just  as zero consump-
  tion in the years between restocking does  not  represent actual  mercury use.
  Therefore,  the yearly national  consumption total  is  influenced  by the  number of
  laboratories that must replace  their mercury stock in that  year,  and by the
  amounts  needed.   The  fluctuations  in annual laboratory mercury  consumption
  mentioned  previously might be partially  explained by  an uneven  distribution of
  "fresh"  mercury requests.
      Hospital  recycling efficiency is considered  to be below that for  non-
 hospital laboratory use.  In general, hospital requirements for mercury are
 more constant and replacement of lost mercury is more  frequent  than for uni-
 versity and contract research laboratory uses.
      After mercury and mercury-containing wastes are collected,   they are  sent
 to a chemical recovery center for triple distillation.  This process means
 that ultimate losses are trivial.
      1983 Technology.   The conclusions reached in Ref. 68 regarding the need
and ability to control mercury use in the laboratory are in accordance with
the URS  findings and fully express the basis upon which Figure 46 was
established:
                                       220

-------
          "Because of the nature of laboratory and hospital operations,
          there is almost no danger of continuous large effluent flow
          containing unacceptable concentrations of mercury.  The dan-
          ger is in careless or accidental  disposition of relatively
          small amounts of expended metallic mercury.  In a laboratory
          and hospital context,  the best available level of treatment
          and control consists of conscientious practice of routine
          methods such as:
          (1)  Precipitation of dissolved mercury compounds or
               safe disposal as solid wastes.
          (2)  Separation of suspended mercury compounds for
               recovery or safe disposal of solid waste.
          (3)  Chemical treatment or fixation of spilled metallic
               mercury before vacuuming or  washing, with subse-
               quent precipitation or separation of mercury for
               recovery or safe disposal as solid waste."

     As these practices are more completely incorporated into labora-
tory operations, the recycle factor will increase and mercury losses
to air, land, and water will decrease.  Any restrictions placed upon
hospital incineration would further increase the 1983 recycle esti-
mate of 80 percent.  To achieve 100 percent recycling is not possible
because of uncontrollable emissions to air, accidental spills, and user
carelessness, which will always characterize the laboratory use of
mercury.  Nonetheless, mercury losses to air, land and water will
probably drop 50 percent or more by 1983 as control measures are in-
creased during the late 1970s and  early 1980s.
     Alternative Technologies.  Alternatives to general laboratory use of
mercury either do not exist or are not readily available.  Mercury is used
sparingly and only in the absence  of  suitable substitutes.  Ref. 68 mentions
that nonmercurials might be able to replace particular mercury compounds used
as chemical  reagents, but no specific  examples are cited.  Regarding hospital
use, Ref. 68 reports that techneitium (  ""TC), radioactive potassium carbonate
(42K), and diiodof luorescein  (   I) may be used in place of  radioactive
mercury for  brain scans.  Centrimide  (1-percent solution), which has been shown
to be effective in preventing the  growth of tumor cells  in experimental wounds,
could serve  as an alternative to mercuric  chloride.
                                        221

-------
       An alternative for a specific mercury application in the hospital may be
  feasible in certain cases and not in others.   Thus,  any restrictions based on
  alternative availability and directed toward a particular mercury use would
  possibly interfere with best medical practice.

  Sector VIII - Personally Oriented Final Consumption

       Dental Applications

       Mercury has been used as an amalgam in the dental profession since the
  early 1800s.   Historically,  this use has been the  subject of  controversy (Ref.
  70), particularly with regard to the potential hazard  it poses to the health
  of the dentist and the patient.   Dentists are especially subject  to  chronic
  mercury poisoning,  because they  usually prepare the  fillings  and  thus have the
  most direct contact with the mercury.   Patient contact is limited to a few
  minutes during amalgam preparation,  and only  minute  quantities are ingested
  once the restoration is in the teeth.   Estimates place the average daily inges-
  tion for people in  the United States at 0.2 to 1.5 micrograms ([Jig) per day per
  person  from dental  restorations, based  on an  average of 0.5 gram per year per
  person  of mercury placed in  the  teeth and on  normal erosion of the restoration
  (Ref. 92).
      The U.S.  dental profession's demand  for mercury has  fluctuated  quite
  erratically over the last 15 years (Refs. 41,   93).   Current demand is  estimated
  at roughly  4 percent of  total U.S. mercury consumption.  Demand for  the  years
  to 1983 will probably be tied to population growth; for  1983 a high  of 3,600
  flasks and a low of 2,100 flasks for dental uses are projected.  The most likely
 value for 1983 is on the high side—approximately 3,500 flasks—because  the
 dental profession is expected to continue using mercury in tooth restorations.
 However,  by 1983,  mercury recycling should somewhat slow the net growth  rate.
      1973 Technology.   Figure 47  illustrates the procedure in manufacturing a
mercury  amalgam for  tooth restoration, and indicates the associated losses of
mercury at various stages of the  process.  Restorations are prepared by combin-
ing a silver-tin alloy  in powdered form with metallic mercury.  These ingre-
dients, and  a mortar, are placed  in a tightly  closed capsule mounted on a
                                       222

-------
               Silver-Tin
                Alloy
               (powder)
Grinding
Old
Fillings
necycie
Air
Water

 Combine
Ingredients
                                 Capsule
                                   with
                                  Mortar
                                 Shaking
to
to
OJ
Mercury
 Metal
 1000
                                  Open
                                Containers
                                 Air 10
Silver-Mercury Cpd.
 Tin-Mercury Cpd.
    Silver-Tin
  Alloy (residual)
                                                                         Metallic Putty
Permanent Tooth
  Restoration
     710
                                                                                                             Squeezing
                                                                                                             Excess Hg
                                                                  Condensing
                                                                  and Polishing
                                                                        » Water 210
                                             Total Use (Loss)
                                             Teeth    710
                                             Recycle   60
                                             Air      20
                                             Water    210
                                                   1000
                                                                                           Excess
                                                                                          Amalgam
                                                                                          Recycled
                                                                                             10
                                                                                           Air 10
                                                              Excess
                                                             Amalgam
                                                             Recycled
                                                               50
                                               SOURCE: URS Research Company.
                                 Figure 47  MERCURY BALANCE, DENTAL APPLICATIONS, 1973 TECHNOLOGY

-------
  machine which vigorously shakes the ingredients.  A metallic putty  is  formed,
  composed of silver-mercury and tin-mercury compounds and  some residual silver-
  tin alloy.   This putty is then placed in the  drilled tooth cavity.  The semi-
  hardened putty is later condensed and polished, and the excess amalgam is
  removed during the condensing process.
       Typically,  the mercury is stored in and  dispensed from a one-  to  five-
  pound container.   If the container is improperly sealed or is uncovered, there
  will be some vaporization.   The powder and liquid mercury are mixed in a closed
  capsule,  so no losses occur at this stage.  However, after the shaking is com-
  plete,  the  dentist may knead or squeeze out any surplus mercury in  the amalgam.
  If these droplets fall to the floor or onto the work bench, the mercury can
  become volatilized.   Also,  when the patient expectorates during shaping of the
  filling,  unless a trap is installed in the plumbing, the excess amalgam and
  mercury  are released into the  wastewater system.                        4
       1983 Technology.   The  postulated 1983 technology (Pig. 48) assumes that
  all dental  practitioners  are  well aware of the potential mercury hazard and
  have taken  all possible steps to reduce mercury loss.  "Premixes" are  already
  available in which the  metallic mercury and silver-tin alloy can be purchased
  in  measured quantities, so  that no  open containers of mercury are needed.  The
  mercury is  contained  in a capsule that, upon shaking,  ruptures and permits
  amalgamation with the alloy.  Thus, no mercury need be lost while making the
 metallic putty.
      During installation of the filling,  excess amalgam will be trimmed off
 and must be spit out by the patient.  To minimize mercury loss,  water basins
 should be equipped with traps to collect much of the excess amalgam; the
 mercury so recovered could be recycled.
      Mercury can be lost to air and/or water  when a filling is removed.  Removal
 is accomplished by drilling through the old filling;  small particles of the
 amalgam are  thus  emitted to the air and may be inhaled by  the patient and
 dentist.   Other bits of the restoration are spit out into the basin.  Drilling
 old  fillings is not frequently done, however,  as mercury amalgam restorations
are  intended to stay in place for up  to 25 years.
                                       224

-------
          Silver-Tin Alloy
          (Powder Form)
                 Combine
                Ingredients
ro
to
en
                Capsule with
                   Mortar
Shaking
 Silver-Mercury
  Compound
 Tin-Mercury
Silver-Tin Alloy
                                        Metallic Putty
Mercury
 Metal
 1000
Permanent Tooth
  Restoration
     750
                                                                                                                  Condensing
                                                                                                                  and
                                                                                                                  Polishing
                                                                                                                           I
                                                                                   Water
                                                                                   100
                                                                                    Excess Amalgam
                                                                                       Recycled
                                                                                          10
                                                                                              Excess Amalgam
                                                                                                 Recycled
                                                                                                   140
                                                         Total Use (Loss)
                                                         Teeth     750
                                                         Recycle   150
                                                         Water     100
                                                                1000
Grinding
Old
Fillings
neuycie
Air
Water

       SOURCE: URS Research Company.
                               Figure 48   MERCURY BALANCE, DENTAL APPLICATIONS, 1983 TECHNOLOGY

-------
      Alternative Technologies.  In terms of actual quantities of mercury lost
  to the environment,  dentistry-associated losses are very small.  However,
  since  these losses  can  occur  in confined work and office space, harmful con-
  centrations could build and pose hazards to dentists, technicians, and (less
  likely)  patients.   Proper handling techniques and an acute awareness of the
  potential danger associated with mercury use should mitigate this hazard.
      There are  a number of ways in which dentists can prevent unusually high
  mercury concentrations  in offices or clinics.  The Council on Dental Research
  sponsored a review  of mercury in dental practice which indicated that there
  will be no hazard to dental personnel or their patients if proper office pro-
  cedures are followed and scrap amalgam is collected and salvaged.  The proce-
  dures  suggested include storing mercury in unbreakable, sealed containers,
  confining any inadvertent spills to an easily cleaned tray or similar work
  area,  and working in an uncarpeted area.  A drain trap to collect solids should
  be installed and all amalgam  scrap should be collected and stored in a tightly
  sealed container.  Other suggestions include being sure that the area is well
  ventilated;  avoiding heating of mercury or amalgams;  using water spray and
  suction when grinding dental amalgams;  coating the restoration surface with
 a varnish  or base;  and eliminating the indiscriminate use of mercury-containing
 solutions.
      In sum, careful handling and use of mercury in dental offices are neces-
 sary to prevent accidental mercury poisoning,  but given present technology,
 the usefulness of mercury as an amalgam,  and the relative ease with which
 mercury losses can be controlled,  it should not be necessary to curtail mercury
 use in  dental restorations.   There are,  of course,  a  number of substitutes
 available for mercury amalgams,  although none is as permanent or as economical
 as  mercury.  Possible substitutes  include  silicate/zinc phosphate cements or
 aerylate and epoxy resins  (Ref.  41),  as  well as gold.   None of these is expected
 to  supplant mercury  amalgams  unless the  price of mercury becomes prohibitively
high or its use  is banned.
                                       226

-------
     Pharmaceuticals

     The use of mercury in medicine is a part of  the history of man  (Ref. 70).
Many of the historical applications of mercury have retained a surprising
level of popularity and use throughout the 1960s  and early  1970s.  Organic
mercury compounds are used in diuretics and antiseptics;  inorganic mercury
salts are used in solutions for sterilizing instruments;  ammoniated  mercury,
oxides of mercury, and metallic mercury are used  in skin  preparations; and
phenylmercury compounds are used as preservatives in cosmetics and soaps  (Ref.
68).  The antimicrobial properties of mercurials  make  these compounds  desirable
components of drug, cosmetic, and soap base emulsions,  while the  fast  action
and effectiveness of the mercurial diuretics have contributed to  their survival.
Phenylmercuric acetate is used as a contraceptive jelly.
     The pharmaceutical industry used some 20,900 kg of mercury in 1973  (Ref.
37).  This level of usage reflects a return to a declining  trend  in  mercury
consumption that characterized this industry between  1950 and  1973  (from  207,000
kg in 1950 to 20,900 kg in 1973).  The decline is primarily due to the discovery
and effectiveness of nonmercurial substitutes, but the toxicity  of mercury
compounds has also contributed to the marked decrease in their use.   In  general,
mercury compounds have been replaced by drugs which are safer  (fewer side
effects) and/or as effective as the mercurials.  Drugs more specific in their
activity are available in many cases.
     Mercury-containing Pharmaceuticals are produced chiefly in New Jersey.
The U.S. market for mercurials as antiseptics has generally followed population
growth.  As long as mercurial antiseptics are available to the public, there
will probably be a constant demand for them, at  least through 1983.  Mercurial
diuretics probably will be maintained as a  last-resort alternative  to non-
mercurial diuretics; thus their use will not decline to  zero.  However, mercury
use in skin preparations and preservatives  will  almost vanish by 1985 (Ref.  14).
     Cammarota predicts a consumption high of about 27,000 kg of mercury for
Pharmaceuticals in 1985 and  a  low  of  about 14,000 kg  (Ref.  14).  The high fore-
cast is based on  expected population  growth,  and the  low forecast  is based on
elimination of mercurials in the pharmaceutical  industry except for special
uses for which there is no substitute.   A probable consumption level would be
                                        227

-------
  about 17,000 kg;  this is  contingent upon trends in pharmaceutical applications
  and the possible development of  new uses for mercury.  This consumption level
  would be slightly below the  present pharmaceutical mercury use of 20,900 kg.
       1973 Technology.   Figure 49 shows estimated  (relative) mercury usage for
  pharmaceutical applications  and  corresponding losses to air, land, and water.
  The loss attributable to  production of mercury Pharmaceuticals* is limited to
  residue amounts not exceeding 0.1 percent of the quantity of mercury used in
  the end product (Ref.  68).   These residues are disposed of using a state-
  approved system of dry-fill  burial.  Groundwater contamination from the buried
  mercury residues is assumed  to be negligible.
       The use figures in Figure 49 approximate the 1968 marketing research find-
  ings of the  Panel on Mercury of  the National Materials Advisory Board of the
  National Research Council for 1968 (Ref s. 94, 95).  The losses assigned to each
  category represent URS  estimates  of mercury emissions to air, indiscriminate
  disposal of  mercury-containing drugs to land, and waste treatment plant disposal
  to  water.  The  loss to  air from mercury antiseptics and skin preparations
  probably does not  exceed  10 percent of the total mercury content.  Most of the
  mercury used in topical applications is ultimately lost to the water environ-
  ment through cleansing  of the mercury compounds attached to the body's surface.
  URS estimates that  95 percent of the mercury in mercurial diuretics is carried
  to water via human wastes.  In all applications,  a 5-percent loss to land
 reflects indiscriminate disposal of mercury-based drugs and containers by the
 American consumer.  Overall,  almost 90 percent of the mercury used by the
 pharmaceutical industry is lost to the water environment.
      1983 Technology.  No  changes in the  ultimate fate of mercury pharmaceuti-
 cals are expected by 1983. The use figures  for the different applications
 will doubtless shift as the availability  of  alternatives,  demands,  and govern-
 mental restrictions change.   The  relative losses to air,  land,  and water,  how-
 ever, will probably be much the same as the  1973  figures.
* The manufacture of Pharmaceuticals is  discussed under organic and inorganic
  mercurial manufacturing.
                                       228

-------
Mercury 1000
  Input

Manufacture
1
Land
Losses to
Air 80
Land 52
Water 868



399 "
150
350
100

Antiseptics

Diuretics

Skin
Preparations

Miscellaneous
(Preservatives,
contraceptives, etc.)
tu

__, -._ ^ MHtrr
339 W3tCr
8

142 W"
35
"• *•» -^ i ^.^^i

297
5_K. 1 Tnrl

on
                   1,000

SOURCE: URS Research Company.
    Figure 49  MERCURY BALANCE, PHARMACEUTICALS MANUFACTURE
             AND USE, 1973 TECHNOLOGY
                               229

-------
       Alternative Technologies.  According to Ref. 68, one factor which hinders
  development of nonmercurial substitute Pharmaceuticals is the need for certi-
  fication of products by the Food and Drug Administration (FDA).  Aside from the
  costs of development and testing, there is a significant time lag in processing
  new drug applications.   Nonetheless, effective nonmercurial substitutes are
  available (Ref.  68).  Considering that only limited amounts of mercury are
  used and eventually lost to the environment, though, it appears that the need
  for immediate controls  is not critical.
       The FDA has already ordered the removal of mercury from skin bleaching
  products and has limited the mercury content in cosmetics to less than 1 ppm
  in most cases.   Because of  the FDA fear that mercury compounds used for cos-
  metic preservation  may  be harmful to man as a result of permeation through the
  skin,  controls on cosmetics would depend on the efficacy of weak static action
  substitutes.  Similarly,  the availability of antiseptic mercurials should be
  examined in terms of  the  antimicrobial alternatives that already exist.  In
  cases  where a particular  mercurial pharmaceutical offers the physician a viable
  and unique  alternative  to a  nonmercurial solution, restrictions might be
  counterproductive.  As Goldwater (Ref.  70)  explains,  however,  promotional
  advertising and the "mystery, magic,  and mysticism of quicksilver" have con-
  tributed more to the perpetuation of the mercurials than have considerations
  of the irreplaceability and effectiveness of mercury medicine.   Further
  research would probably reveal a number of areas where the use of mercury
 Pharmaceuticals could be reduced or eliminated with no significant loss to
 "best practical" medical treatment.

 Final Disposal Sector

      Municipal Waste Treatment

      The development of  municipal waste  treatment plant technology grew out of
the  need to  protect  community health.  As citizen and governmental concerns
broadened, these  plants  were improved to protect  the  environment and particu-
larly the water that receives sewage discharges.   This environmental awareness
                                       230

-------
led to studies to determine the fate of heavy metals  in such plants.  Most  of
this Work has been directed at lead, copper,  cadmium,  chromium, and nickel;
very little work has been done on. mercury.
     In 1968, there were 12,565 municipal (community)  treatment plants  in the
United States, treating the residential and industrial wastes  of  some 65  per-
cent of the U.S. population (Ref. 96).   Of these plants,  2,384 provided only
primary treatment (removal of particulate matter only), 9,951  provided  secon-
dary treatment (primary treatment plus biological oxidation of most of  the
remaining organic matter, and 75 provided intermediate (between primary and
secondary) treatment.  These plants accounted for more than 99 percent  of the
total number of U.S. treatment plants,  and served about 64.5 percent of the
total population.
     According to Ref. 97, in 1968  slightly more than 7 percent  of industrial
wastewaters were discharged into municipal treatment systems.  Table 38 shows
that community sewers in the Eastern Great Lakes region received the  largest
amounts of industrial wastes, and sewers in the Great Basin region (Nevada,
Utah) received the least.  The actual amounts may be a bit higher than the
figures in Table 38  indicate, because those figure reflect only industries
whose water  intake was greater than 20 million gallons a year.  As shown in
Table 39, food and kindred products plants discharge the greatest amounts of
wastewater to sewers, while lumber  manufacturers discharge the least.
     Although industry is generally responsible for large concentrations of
heavy metals in sewage influents, the URS inventory results indicate that in
the United States about  50 percent  of the mercury in  influents probably  derives
from the disposal practices of hospitals, dentists, and laboratories.  A lesser
proportion (possibly 25 percent) comes  from  the  discharge  of  mercury-using
manufacturing industries.  Even  less  (15 to  20 percent)  is attributable  to
human additions and  other industrial contributions,  and approximately  5  per-
cent to 10 percent of the mercury  in  sewage  influent is a  result of natural
background concentrations.  Recent measurements (Ref. 98)  have  shown that  the
concentration of mercury averages  about 1.5 ppb in sewers that  receive resi-
dential discharges exclusively;  in contrast, the national average mercury
concentration in sewage  influent as used in this report is 2.0 ppb.
                                      231

-------
to
u>
10
                     TABLE  38.  -  Regional  Distribution of Industrial  Wastewater  Discharges  to
                                              Community Sewers  in 1968
                                                (billions of gallons)

Industrial Water Use Region
New England
Delaware and Hudson
Chesapeake Bay
Ohio
Eastern Great Lakes
Tennessee-Cumberland
Southeast
Western Great Lakes
Upper Mississippi
Lower Mississippi
Missouri
Arkansas -White-Red
Western Gulf
Colorado Basin
Great Basin
California
Pacific Northwest
National*
Total Industrial
Wastewater
Discharges
558.4
1,191.9
754.7
2,295.4
1,459.7
535.9
1,099.6
1,811.3
581.6
744.6
141.9
184.6
1,899.1
18.3
26.8
314.1
532.5
14,150.4

Percent Dis-
to Community
Sewers
8.4%
7.3
4.3
7.5
13.9
2.6
5.2
7.4
18.5
3.1
17.8
7.9
0.8
20.2
6.3
16.8
5.7


Total Industrial
Discharge to
Community Sewers
46.9
87.0
32.5
172.2
202.9
13.9
57.2
134.0
107.6
23.1
25.2
145.8
15.2
3.7
1.7
52.8
30.4
1,152.1

       a.  Excluding Hawaii and Alaska.

       Source:   Ref. 97

-------
                           TABLE  39.  -  Wastewater Discharges  to Community Sewers  in 1968,  by
                                                  Industrial Category
                                                 (billions  of  gallons)
OJ
U>
       Industrial Category
Total Industrial
   Discharges
Percent Discharged
to Community Sewers
Total Industrial
 Discharges to
Community Sewers
Food and kindred products
Textile mill products
Lumber
Paper
Chemicals
Petroleum and coal
Rubber
Leather
Stone, clay, and glass
Primary metals
Fabricated metals
Machinery
Electrical equipment
Transportation equipment
752.8
136.0
92.7
2,077.6
4,175.1
1,217.0
128.4
14.9
218.4
4,695.5
65.0
180.8
118.4
293.1
31.6%
37.2
2.7
3.5
4.3
0.6
17.4
68.0
9.4
3.1
59.4
24.6
62.8
26.3
237.9
50.6
2.5
72.7
179.5
73.0
22.3
10.1
20.5
145.6
38.6
44.5
74.4
77.1
       Source:  Ref. 97

-------
       1973 Technology.   There are various  ways of  treating municipal  sewage, but
  as of 1973 most plants used secondary treatment processes—either activated
  sludge or trickling filters.  As indicated above, about 65 percent of the
  community sewer systems surveyed by the Federal Water Quality Control Adminis-
  tration (FWQCA) in 1968 provided secondary treatment  (Ref. 96).  To  comply with
  Public Law 92-500 (the Federal Water Pollution Control Act Amendments of 1972),
  all municipal waste treatment plants will have to provide secondary  treatment
  by 1977.
       Figure 50, a schematic of a typical  activated sludge plant, shows the
  ultimate fate of influent mercury.   Because mercury tends to adsorb  on particu-
  late matter,  about 45  percent of the influent mercury is removed from the waste
  stream by the primary  clarifier.  The remaining mercury enters the aeration
  basin where aerobic bacteria metabolize the organic components of sewage.  It
  has long been known that slugs of heavy metals can degrade or halt the proper
  function of these bacteria.   However,  experiments with mercury indicate that
  only abnormally high concentrations  of influent mercury (probably over 10 ppm)
  would reduce  bacterial efficiency (Refs.  99,  100).  Similarly,  high  concentra-
  tions of mercury (between 1,000 and  2,000 ppm) in the biological sludge inhibit
  the sludge.
       Because  most of the biological  sludge produced in the aeration basin is
  recaptured  in the secondary clarifier and recycled,  very little mercury (about
  5 percent as  a  result of activated sludge wasting) is actually removed from
  the waste stream.  The remaining mercury in the waste stream (about 50 percent)
  is then  lost  to the receiving waters  (Refs. 101,  102).   This mercury is both
 in solution and adsorbed onto the particulates that are not captured in the
 secondary clarifier  (Ref. 103).
      The digester, then, receives about 50 percent of the incoming mercury.
 Recent experiments have shown that almost all of the mercury remaining in the
 digester is in or on the solid phase (Ref. 104);  consequently Figure 50 shows
 only a 1-percent recycle of mercury in digester supernatant (liquid formed on
 top  of the  digested solids as a result of settling and biological degradation)
to the primary clarifier.   Work by Bisogni and Lawrence (Ref.  100)  indicates
that at growth rates common to most  digesters  (residence  times between 10 and
                                       234

-------
Influent (2 ppb)^ Prjmary 154 Kg _ Aeration _ Secondary 140 Kg _ f ™U6nt .
280 Kg /yr '
1 Kg
N)
U>
in
Clarifier 55% ' Basin ' Clarifier 50% ' w.iuimiuun
&!>% 50% and Disposal
127 Kg
Pri
Slu

i
45% Waste Activated
Sludge
14 Kg
141 Kg
mary '
dge 91 Kg


Return ,
Sludge
-^^- Incineration (100% to Air)
— ^- Land Disposal (Most Remains in Land)
                              49 Kg
                                  *
                           To Air as Off-gas Flare
SOURCE: URS Research Company.
                    Figure 50  PATH OF MERCURY IN A 100 MGD SEWAGE TREATMENT PLANT

-------
  30 days)  about 35 percent of the mercury coming into the digester on solids
  would be  lost through volatilization; the remaining 65 percent would remain
  with the  digested solids.
       The  remaining mercury in digested  sludge may be disposed of in one of
  three ways—in landfill,  as a soil amendment, or by incineration.  The ultimate
  fate of this mercury depends on the disposal practice chosen.  In brief, the
  mercury in digested sludge that is disposed of in sanitary landfills has reached
  its ultimate fate.   Mercury in digested sludge applied as a plant fertilizer or
  soil amendment is bound  somewhat less tightly as some may be lost through
  volatilization or erosion.   Published studies indicate that all of the mercury
  in sludge that is incinerated reaches the atmosphere,  and that assumption is
  used here.
       1983 Technology.  The  municipal waste treatment technology that will be
  in effect by 1983 is also mandated by PL 92-500.  According to this law,
  municipal plants  must provide  treatment that will protect the fish,  shellfish,
  and wildlife of receiving waters and that will allow recreation in and on such
  waters.   If  present  funding and construction trends continue, in 1983 the bulk
  of these  plants will still be  secondary treatment plants but a few plants will
  provide tertiary  treatment  (mainly phosphorus removal and multimedia filtration)
  in order  to  achieve  higher quality of certain receiving waters (Ref. 105).
  Therefore, there will probably be no major changes in 1973 process technologies
 by 1983.  However, the trend toward regionalization means that plants will tend
  to be larger, but there will be fewer of them than in 1973.
      The major process difference will be in the final disposition of digested
 sludge. Greater use of sludge for fertilizer and soil amendment is expected,
 depending  on the results  of research on the fate of sludge-applied heavy metals
 and on the amount of vegetative uptake.   Initial research indicates that the
 buildup and subsequent uptake of cadmium,  copper,  zinc  and boron may limit the
 land application of sewage sludge, but that mercury causes no problem (Ref.
 106).  Greater control of industrial wastewaters influent to sewage treatment
plants and recycling of mercury-containing compounds could reduce the concen-
trations of such heavy metals in sewage  sludge.  As  industrial emissions account
for about  35  percent  of the mercury input, industrial reductions could have a
                                       236

-------
significant effect.  An even greater reduction would occur  if institutional
mercury users recycled more mercury.  Although it  is difficult to  determine
the extent of such reductions, URS assumes a 25-percent reduction  in municipal
sewer mercury disposal by 1983 as a result of the  above reductions.  URS  also
assumes that sludge incineration will decrease by  50 percent because of air
pollution regulations (mercury and others) and energy  conservation problems.
Using these assumptions and the 1973 national inventory results, we have  pre-
pared Table 40, which lists estimated mercury losses to air, land, and water
from sewage treatment facilities for 1973 and 1983.

     Municipal Solid Waste Disposal

     Municipal solid waste disposal systems are responsible for  the collection,
processing, and ultimate disposal  (either in landfills or by incineration) of
solid waste from residential, commercial, and light industrial areas.  As with
sewage treatment plants, these systems were developed to protect public health.
Recently the design and operation  of these facilities has also been directed
at minimizing environmental harm,  especially in the control of leachates  from
landfills and particulate emissions from municipal solid waste (Ref.  24).
Because mercury has been found in  landfill leachates  (Ref. 107)  and incinerator
emissions  (Ref. 108), these disposal practices must be examined in terms of
their ability to introduce mercury in the environment.  Collection and process-
ing, on the other  hand, contribute little (if  any) mercury to the environment
and therefore will not be discussed further.
     Table 41 summarizes a recent  national  survey of  solid waste management
practices, listing the number and  types  of municipal  landfill disposal sites
by states and the  number of municipal incinerators  over 100 tons  capacity (Ref.
109).  These figures  indicate that over  half of the landfill  sites do not meet
regulatory requirements and might  be  apt to lose  mercury to groundwater  at  a
greater rate than  do  sanitary (regulated) landfills.   The table also shows that
incineration, which emits mercury  directly to the atmosphere, is a common
disposal practice  on  the East Coast.
     The components of municipal solid wastes and their percent contribution by
weight are paper products  (60 percent),  food wastes  (8.5 percent), glass and

                                       23*7

-------
         TABLE 40. - Estimated Mercury Emissions from U.S. Municipal
                     Sewage Treatment Plants, 1973 and 1983
                                  (kilograms)
Emissions to
Air
Land
Water
Total
1973a
9,563
17,327
19,923
46,813

1983b
7,255C
16,346
17,482
41,083

Percent
Change
-24%
- 6
-12
-12%

Source:  URS Research Company.
a. Derived from URS inventory.
b. Based on expansion of total U.S.  population by 17 percent during
   1973-1983 period and 25 percent reduction in mercury losses to muni-
   cipal sewers by 1983.
c. Assumed 50 percent reduction in sewage sludge incineration.
                                     238

-------
   TABLE 41. - National Solid Waste Disposal Practices



States
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Total

Total
Landfill
Sites
143
160
450±
441
255+
144
150
500
625
190
404
151
500
•500
147
387
367
91
324
888
600
274
390
514
400
120
180
307
1,000
800+
162
412
290
507
241
379
38
276
369
250
1,525
272
95
188
397
300+
1,314
80 ±
18,539+

Landfill
Sites
Meeting
Regulations
134
70
47
302
380
209
39
204
275
50
225
220
36
160*
147
61
0
106
171
379
131
54
136
70
123
120
180
276
450
800+
156

270
354
313
129
7
276
32+
112
346
7
75
346
73
66
1,200
~
9,212



Municipal
Incinerators
0
0
6
1
0
23
0
10
0
0
4
1
0
0
2
4
0
4
18
10

0
3
0
0
-
4
3
0
39
0
0
13
2
0
20
2
0
0
1
4
1
1
4
0
0
6
0
189

Source:  Ref. 109
                                239

-------
  ceramics (8.0 percent), metals (8.0 percent),  plants and grass (6.5 percent),
  plastics (3.5 percent), and others (5.5 percent).   Although most of these
  materials may contain traces of mercury,  the major contributors—all found in
  the "other" category—are batteries,  control instruments,  fluorescent and
  other mercury-containing lamps, and tubes and switches.   They account for only
  a small percentage of the "other" solid waste category yet are believed to
  account for virtually all of the estimated 2 to 6  ppm mercury content of solid
  wastes.  (Paper products,  before mercurials were banned  in this usage,  also
  represented an appreciable source of mercury in solid wastes,  but by the year
  1973 the amount of such treated paper products in  solid  wastes was  assumed to
  be low.)  Furthermore,  these "concentrated" solid  wastes represent,  on the
  average (based on all study regions inventoried by DRS),  96 percent of all
  mercury wastes which are rejected to land.
       1973 Technology.   The 1973 technology for solid waste disposal is based
  upon the usage of well accepted sanitary  landfill  and municipal incinerator
  technology in simple combinations.   Important  aspects of sanitary landfill
  practice with regard to potential mercury contamination  include site selection,
  design,  and operation.   Landfill  sites  should  be selected to ensure minimum
  contact with groundwater.   Designs should provide  for adequate drainage and, if
  groundwater is a  problem,  should  provide  for the placement of  impermeable clay
  barriers and gravel  drains  to prevent groundwater  intrusion.   Impermeable
  linings  are also  being  used more  extensively to  seal  landfills, but as of 1974
  only 21  (out of more than 18,000) were  in place.
      Operations and maintenance activities  should be  such  as to guarantee the
  placement of adequate daily and final cover which ensures  sanitary  operations.
 While there is no evidence  that ground cover will stop the  leakage  of mercury
 to air, the topsoil layer has been shown to be effective in controlling the
 release of decomposition gases  (primarily methane and carbon dioxide) which
 undoubtedly contain trace amounts of mercury.
      Recent work done on sanitary landfills indicates that the mercury  concen-
 tration of leachate within the landfill varies between 0.05 and 16.3 ppb, with
 a considerable number of data points above the proposed EEA Drinking Water
 Standard  of  2.0 ppb (Ref. 107).   However,  other work indicates that the attenu-
ating action of soil  effectively lowers the mercury content of leachate as  it
                                        240

-------
passes through the lower soil boundary.   These results  lend credence  to the
acceptability of the modern sanitary design,  operation,  and maintenance princi-
ples now being followed by the individual states and to the ability of sanitary
landfills to provide safe, long-term storage sites for  mercury contained in
municipal solid waste.
     The municipal incinerator, on the other hand, is a direct emitter of
mercury to the atmospheric environment.   Recent work indicates that about 90
percent of the incoming mercury is volatilized and lost to the atmosphere (Ref.
109).  The remaining 10 percent remains in the ash for  ultimate disposal in a
sanitary landfill.  The flow of mercury in municipal solid waste from a typical
city of 500,000, using typical 1973 technology, is shown in Figure 51,  which
indicates that most of the mercury remains bound within the landfill.
     1983 Technology.  As shown in Figure 52, solid waste disposal technology
and practice by 1983 will probably be considerably different from what was
commonly being practiced in 1973.*  The greatest difference will be in the
fields of material and energy recovery, hazardous material disposal and fate,
and the regulations under which solid waste disposal now operates.  Increased
resource recovery—particularly materials recovery—would have the greatest
effect on mercury emissions by reducing the amount of concentrated mercury-
containing wastes going to incinerators and possibly to landfills.  As was
mentioned previously, the mercury content of solid waste appears to be concen-
trated in discarded batteries, fluorescent lamps, tubes and switches, and  con-
trol instruments.  Therefore separation of the light and heavy  solid waste
fractions influent to a municipal incinerator  would result in a significant
reduction in atmospheric mercury emissions.  At the present time there  are
about 38 major resource recovery facilities, not all of which are connected
to incinerators  (Ref. 109).  The main process  for such separation would be the
air classification method.
* The  same  input of mercury into solid wastes is assumed for 1983 as for 1973
  because,  although total quantities of solid waste are projected to grow, the
  amount of mercury-containing wastes discarded is likely to level off or
  decline.
                                        241

-------
                                                100*
     Community
  500,000 Population
                   3.32x10s Kg/yr
                    of Municipal
                    Solid Waste
1330


Processing
• Shredding
• Air Classification
• Metal Recovery
• Glass


10

1220

Incinerator

Material
Recovery
10
i
S
:

1 I
anitary 1
Land
Msposal
                                                                                Air
                                                       10
                                                                           10
                                                                          1220
                                                                               • Air
                                                                               •Water
                                                                               "Remains in Land
* Assumes 8 percent of all solid waste is incinerated.

SOURCE: URS Research Company.
          Figure 51   FLOW OF MERCURY IN THE MUNICIPAL SOLID WASTE STREAM
                     OF A TYPICAL COMMUNITY OF 500,000 IN 1973
                     (kilograms per year)
                                            242

-------

Community
500,000 Population

3.32x10° Kg/yr
of Municipal
Solid Waste
1330
100 Municip


Processing
• Shredding
• Air Classification
• Metal Recovery
• Glass


Conve
230

1000

al Incinerator —
nal or Other
Tsion Process

Material
Recovery
100
(

*
Land
Disposal
50
MMMMM
Ash

10
—10
1131
•*• Air
50
-».Air
•*• Water
r*" Remains in Land
SOURCE: URS Research Company.
         Figure 52  FLOW OF MERCURY IN THE MUNICIPAL SOLID WASTE STREAM
                  OF A TYPICAL COMMUNITY OF 500,000 IN 1983
                  (kilograms per year)
                                       243

-------
       Whether increased material recovery would result in significantly  less
  mercury loading to landfills would depend upon the possibilities  for  extracting
  mercury from the nonferrous metal waste stream.  As mercury is highly volatile,
  it is possible to extract mercury metal from that  fraction  by roasting  and sub-
  sequent condensation.  However,  the economics of this process versus  virgin
  sources and other recycled material are not  promising, and  its adoption appears
  unlikely.
       Much research is now being funded to determine the  fate of hazardous
  materials in landfills and to develop procedures for  reducing their adverse
  impacts,  if any.  Measurements of mercury from landfill  leachates outside  of
  the landfill have not shown anomalously high mercury  concentrations,  which
  indicates that sanitary landfills,  when properly designed and operated, effec-
  tively  remove the mercury contained in municipal  solid  waste from the  environ-
  mental cycle.
       Changes in government regulations could affect the  kinds of  material  in
  solid waste and the disposal practices.   The success  of  sanitary  landfills in
  retaining mercury from municipal  solid waste probably indicates that  regulations
  removing  mercury from solid waste destined for such landfills are not needed;
  however,  upgrading unregulated landfills  should be speeded  up.  On the  other
  hand,  forces fostering the construction and  operation of resource recovery and
  thermal conversion facilities would  cause more mercury to reach the atmosphere.
  If regulations  were promulgated which mandate  the removal of most of  the mercury
  from the process  stream, mercury  emissions to  air could  be  reduced significantly.
      Because sanitary  landfills are more economical to run  than incinerators,
 and because strong  demand  for recycled waste streams is  still unproven,  sanitary
 landfills will probably remain the dominate final disposal method.  In  1973
 land disposal accounted for about 93 percent of all solid wastes,  while in 1983
 land disposal will probably receive only about 75 percent of total municipal
 solid wastes.  This reduction in the percent of disposal to land  is a result
 of increases in material and energy recovery and reclamation.  URS combined
 these  trends in solid waste disposal with improvements in pollution control
 (better capture of mercury emissions from thermal conversion units,  landfill
upgrading)  to derive Table 42, which shows estimated environmental losses  of
mercury from  solid waste disposal activities  for 1973  and 1983.
                                       244

-------
      TABLE 42. - Estimated Mercury Losses to  Environment from U.S.
                  1  Municipal Solid Waste Disposal,  1973 and  1983
                               (kilograms)


Losses to                          1973                       1983a


Air                               40,000                     24,000b

Water                             18,000C                     6,000

Land                             466,000                     448,000

Reclamation                         -                         52,000
Totals                           530,000                     530,000
Source:  URS Research Company
a. Assuming total amount of mercury in solid waste stream will remain
   constant, even though the total quantity of solid waste will in-
   crease.
b. Assuming a  100 percent  increase in thermal conversion processes  (in-
   cineration, pyrolysis,  etc.), all operating with front-end material
   recovery units and air  pollution control units capable of 50 percent
   mercury recovery.
c. Assuming improperly  operated  landfills  and dumps lose mercury  to
   water at twice the rate of  regulated  fills.
                                    245

-------
  MACROTECHNOLOGY (1973)

       For industries that manufacture primarily consumer goods, the losses of
  mercury to the environment may extend  from the manufacturing process itself,
  to other (secondary)  manufacturing and processing, to final consumption and
  disposal.   To obtain the necessary perspective on these losses from the various
  sectors involved,  composite flow charts showing overall losses have been pre-
  pared for the applications of mercury  that involve sequential losses.  Table
  43 summarizes total mercury losses for five such applications.  Flow charts
  indicating the details of these losses to the environment are given in Figures
  53 and 54 for the  two largest consumers of mercury—electrical applications
  and paint.   (Details  on  the source of  discharges for other applications can be
  obtained by referring to the appropriate subsections in the preceding discus-
  sion of microtechnologies.)
       With the exception  of catalyst usage, most losses occur in the final con-
  sumption sector.  Recycling is an important consideration only for industrial
  instruments;  here recycling accounts for almost half of the mercury used by
  the  industry.   In general  it can be concluded that while manufacturing and
  processing represent  the most logical and the easiest points at which to apply
  controls over mercury losses, they also are the least rewarding in terms of
  significant reductions in mercury emissions.

 IMPLICATIONS

      With respect to the various technologies and composite technologies des-
 cribed in this section,  it is generally forecast that there will be appreciable
 reduction in mercury emissions between 1973 and 1983.   Most of these reductions
 will come about from improved utilization of  mercury (through recycling either
 within the plant or of products  containing mercury)  or by the imposition of
 effective pollution controls.  In  some  of  the  lesser technologies,  for example,
 livestock manure, specific reductions were not discussed,  the implication
being  that the loss  to the environment, while  large  in aggregate,  is diffused
geographically and hence  pose no threat to human safety.
                                       246

-------
TABLE 43. - Mercury Losses for Various Applications, 1973
Application
Electrical
Paint
Industrial Instruments
Pharmaceuticals
Catalysts
M
Total Usage
(kilograms)
620,408
260,330
246,612
20,887
23,403
Total Losses in Kilograms to
Environment during
Manufacture
4,932
1,312
3,169
126
17,473
Processing Final Consumption
587,433
3,272 182,750
124,040
126 20,000
1,766 1,945
Ratio of
Total Loss
to Total Use
0.955
0.720
0.516
0.970
0.905


-------
itk
00
 Mined
Mercury
                                                                       232,505
                                     Land
                                          2476
                                  563,537
620,408
                                     Water



                                      63
                                    Air
                               131
                                     Battery
                                  Manufacturing
                                     560,867
                                             Land     Air
                                          1637
                                   43,360
                                            86
                                    Electric
                                     Lamp
                                  Manufacturing
                                   13,511
                                     Power
                                     Tube
                                  Manufacturing
       SOURCE: URS Research Company.
                                513

                                  Land
                                              r
                               27

                             Air
                                                                                  Government
                                                                       131,345
                                          41,637
                                                                  12,971
                                                                                                     210,005
                                                                                   Recycled
                                                                                   (28,043)
                                                                                   Industrial
                                                                          Other
   All
Consumers
                                                              All
                                                           Consumers
                                                                                                  22,500
                                                                                                   5543
                                                                                                     125,802
                                                                                                  197,017
                           Disposal
                            Sites
                           (587,432)
41,687
                   12,971
                                                                                                                            Air
                                                                                                   Land
                            46,995

                            540,437
                          Figure 53  SEQUENTIAL MERCURY LOSSES IN ELECTRICAL APPLICATIONS, 1973 (kilograms)

-------
                                     Air





to
•Ck
VO







Mined
Mercury








262,054





UdllU "
1103






260,951

^"•»™
1








261
I
1
I
Inorganic
Mercury




Organic
Mercury
I
2
I
•
4








1044
>
^ water
1098






259,644
>• Watpr


Air
Air «« 1 1 » Water A
285 654 173,610
260,742 Paint 257,470 A)| 1000
Manufacturing Consumers
2333 18140
| 1
Land Land


                                    Air
SOURCE: URS Resaarch Company.
                 Figure 54 SEQUENTIAL MERCURY LOSSES IN PAINT APPLICATIONS, 1973 (kilograms)

-------
       In general, it appears that users and emitters of mercury have been
  thoroughly alerted to the inherent dangers of such emissions and have taken
  the necessary actions to limit or eliminate them entirely.   While this trend
  is indeed praiseworthy, it must be recognized that it may not be sustained
  without effective enforcement.  Hence,  even in industries which can demonstrate
  or forecast adequate self-policing,  regulations may be necessary to ensure
  continued compliance.
       Even with improved technology,  increased recycling,  and concern on the
  part of the industries involved,  high losses to the environment will still
  persist in some categories:
       o  The chlor-alkali industry,  assuming the continued operation of
          existing plants with vastly improved pollution control technology
          and utilization of mercury,  will in 1983 continue to emit appre-
          ciable quantities of mercury to the environment in localized areas.
       o  The paint industry,  assuming continued use of mercurials and even
          with the best of controls, will continue to be a  major contributor
          of man-related emissions  to  air in 1983.
       o  Electrical usage (including  batteries,  lamps,  and tubes and
          switches),  even with vastly  improved recycling, will continue to
          contribute vast quantities of mercury to landfills and,  unless
          stringent regulations are imposed,  to incinerators and thence to
          the  air.  To  a  lesser extent, the  same comments apply to control
          instruments.
      Since these  three categories have been identified as important and continu-
 ing emitters of mercury  to the environment, albeit to safe disposal sites,  these
 industries are considered in  Section V in a detailed analysis with respect to
 the impact of regulatory actions which would serve to further reduce  their
 emissions to the environment.
      It is difficult to assess the hazard posed to man and his environment by
 man-related emission sources as contrasted to the hazard posed by natural
 sources.  As  has been shown,  continuing action has reduced man-related emissions
 significantly and will continue to do so for some years in the future.  However,
natural emissions  of mercury,  from degassing and runoff, will remain unchanged
                                       250

-------
(or could even increase appreciably iff  for example,  unusual volcanic activity
occurred).  It is legitimate to ask:  Why spend millions of dollars to reduce
mercury emissions from, say, a power plant when such emissions to air are very
small compared to those from degassing from the earth's crust?  First,  we must
admit that man can control only his own release of mercury to the environment,
and that if he believes that mercury releases should be minimized, he must
concentrate on that which is controllable.  Second, and perhaps more important,
natural sources of mercury into the environment are diffuse and relatively
uniformly distributed  (across the nation and the world), and have not been
demonstrated  (or even  inferred) to concentrate to the extent of interfering
with man's food chain.  On the other hand, man's wastes do follow his habitat.
They do concentrate where he lives, and have on certain unusual occasions
concentrated  sufficiently to enter the food chain and threaten man's well-being.
Hence, man-related discharges of mercury, even though small, interact with man
and his environment more directly than do natural  sources.
                                        251

-------
                                                                            V
                                                   REGULATORY IMPACT ANALYSIS

INTRODUCTION

     In this section we discuss the ways in which regulatory actions might be
used to reduce the quantities of mercury discharged to the  environment by man-
related activities.  Three major emissions contributors—chlor-alkali plants,
final consumption of paint, and final consumption of  batteries—are examined.
These three examples also provide insight as to how controls might be applied
in related industries or usages.
     Chlor-alkali plants represent a manufacturing industry that uses large
quantities of -mercury each year and loses appreciable amounts of this mercury
to the environment (although in a form that has not been shown to be unsafe).
In this small but important industry mercury losses are limited  to the manufac-
turing process only; thus this example can be used to illustrate how controls
might be applied to clearly defined point source emitters.
     Final consumption  (and disposal) of batteries creates mercury  losses
primarily to landfills and incineration.  This example typifies  the  solid
waste losses of several other electrical applications, including tubes/switches,
lamps, and industrial control instruments.  In the case of batteries,  manufac-
turing losses are small but overall losses, including those created by ultimate
disposal of the used product, are large.  These  latter types of discharges are
considered to be area sources  (as contrasted with point sources), and regula-
tion at their points of release  (or disposal) may be  impractical.  Therefore
we explore the usefulness of regulation at  the "point" source—that is, at the
manufacturing level—as a way of reducing losses at  the consumer's end.  This
type of regulation relies heavily on recycling.  However,  the impact of such
regulation of point sources affects the entire economic chain from mercury
production through final sales,  and is therefore considered as  a part  of  the
derived costs in our cost-benefit analysis.
                                        253

-------
      Relatively little mercury is  lost in the manufacturing and processing
 of paint.  Most of the mercury emissions from paint occur after its final
 application,  and most of these losses are to air.  No control is possible at
 this point (and again there  is little or no evidence to indicate a hazard to
 health).   Therefore only one regulatory action—the elimination of mercurials
 from paint—can be envisioned.  The impacts of such a regulation would be felt
 by  suppliers  of mercury, manufacturers of mercurials, paint formulators, whole-
 sale and  retail distributors, and  ultimately the consumer.
      Three cost-benefit  analysis  (CBA) alternatives have been proposed  (although
 not all are applicable to any given case):
      o Substitution of  a nonmercury-containing material or end product
         for one that contains mercury or uses mercury in its production.
      o Use of alternative processes that can produce the desired mate-
         rial  or end product  without the use of mercury.
      o Imposition of mercury emission controls that will substantially
         decrease the losses  of mercury to the environment, either in
         production processes or in end uses.
      Because  there is virtually no knowledge concerning the health hazards
 which  may  be  associated with exposure to low levels of mercury (that is, levels
 far  below  the published  standards shown in Figure 5), it is not possible to
 demonstrate a benefit for reduction of mercury emissions when they are initially
 below  established  "safe" limits.   In lieu of a demonstrated benefit we have
 elected to use a stated benefit—the estimated reduction in the quantity of
 mercury emitted that  is attributable to the implementation of a given alterna-
 tive.  This accrued benefit may be expressed as a fraction, thus:
           Total mercury release with the alternative
           Total mercury release without the alternative
      Each alternative has costs attached to it.   These costs, both direct and
 indirect, are determined as the second step of the CBA.
      Finally, each alternative is considered in the light of the possible risks
it may entail.   One type of risk is that the alternative may in itself lead
to the introduction of other toxic or hazardous  materials into the environment;
the second type  of risk is that disbenefits, in  terms of acceptability or
effectiveness of the  substitute,  may result.  Since benefits are not treated

                                      254

-------
METHODOLOGY

     Benefits are extremely difficult to identify, much less to quantify.  In
the case of mercury this is doubly true, since its toxicity can only be demon-
strated at relatively high emission levels—far beyond those considered safe
for continued exposure—yet the quantifiable benefit of reducing intake below
"unsafe" levels cannot be demonstrated.  Also, mercury enters the environment
from natural sources that contribute to the total mercury burden.  Although
definitive data on benefits are lacking, it can be presumed that unnecessary
introduction of mercury into the environment, and thence into the food chain,
is undesirable.  We have therefore chosen as our criterion for benefit the
reduction in the amount of man-related mercury which enters the environment
that can be effected by employing a given alternative.  This reduction is
estimated in terms of projected mercury consumption by the contributor under
study over the  10-year period  1973 to  1983.   (Since mercury consumption  in
general is going down for certain consumers,  it is necessary to include  this
declining curve.)  The reduction is then compared with the estimated consump-
tion for each alternative.  An example of how such a curve might  look  is
sketched below.
                         1973 TECHNOLOGY
               cs
               oc
                I
                o
                cc
                UJ
                
-------
      For a given year, the reduction in the amount of mercury used under the
 control situation versus that for the alternative is plotted.  A second curve
 can then be derived  (see sketch below) which indicates the percent reduction
 in mercury use versus time.  For each curve a cost can be attached for the
 total period  (in this case, 10 years).  The CBA results then become a problem
 of choice for the planner in terms of the willingness and ability to allocate
 sufficient money to reach the desired objectives.
         100X  -
         rs «/»
         O *H
         ui a
         as

         UJ O
         o oc
               0
                                                  ALTERNATIVE 2
                      ALTERNATIVE 1
                                                BASE CASE
i      i     i
468
   YEAR
10
      Costs are not as difficult to  estimate  as  are benefits in this case,  but
 cost estimating is still difficult.  Costs may  include  costs to the manufacturer
 who  may have to change his  technology  or his product, costs to society in terms
 of displaced workers  and equipment, cost to  the consumer  in terms of a less
 effective  product,  etc.   In short,  a ripple  effect can  be traced.   However,
other alternatives  (which are  less  complicated  but which  will suffice for the
present analysis) are used  as  appropriate for each of the three major industries
studied.
                                      256

-------
     Cost estimates for the chlor-alkali industry were limited to those that
could-be identified as manufacturers'  capital investments for adoption of new
technology which could be mandated specifically  for the industry.  Costs for
the battery and paint industries were extended to include all associated costs,
since regulations, even though imposed at the manufacturer's level, have impli-
cations at all levels from mercury mining through final consumption of the end
products.
     The risks associated with each alternative  could reduce the overall bene-
fit associated with that alternative.  But we are not directly identifying
benefit, so risks identified for an alternative  are treated as part of that
alternative's costs.  For example, the removal of  lead discharges from diaphragm
cells that replace mercury cells for the production of chlorine would be treated
as a  cost — primarily an operating cost.

REGULATION IN THE CHLOR-ALKALI INDUSTRY

     Chlor-alkali producers who use mercury cell technology are already  subject
to regulation with respect to allowable emissions  of  mercury to the environment
and,  although some producers are still not in compliance,  the  necessary  actions
to comply are under way.  But the industry still emits large quantities  of
mercury, primarily to land, and in this section we consider regulatory actions
that might be taken to further decrease such emissions.   The alternatives that
might involve regulatory actions include:
      1.  Eliminating the demand for chlorine and caustic soda by
         substituting other chemicals and/or other end-use technologies.
      2.  Altering the technology of producing chlorine and caustic soda
         to end the use of mercury altogether.
      3.  Tightening enforceable controls  or  regulations to minimize  the
         emissions of mercury to  the biosphere.

Alternative 1—Elimination or Substitution of Product

     Chlorine is  one of the basic "building  blocks"  of  chemistry and is one of
the largest volume industrial chemicals.   More  than 10  million tons of chlorine

                                       257

-------
  were used in the United States in 1973; about one-fourth of this was produced
  using mercury cell technology.  Almost 50 percent of all chlorine produced is
  used in the synthesis of miscellaneous organic chemicals, and an additional
  19 percent in the synthesis of vinyl chloride—another organic.  Elimination of
  chlorine or substitution of another material for these uses is not feasible.
  Substitutes for chlorine have been developed for some smaller volume uses.
  Sanitation and water treatment use an  estimated 5 percent of all chlorine pro-
  duction.  The use of ozone  and iodine  (or compounds) has been adopted by some
  users in place of chlorine  for water treatment, but to date the acceptance level
  is low,  and there has been  only a minor impact on the demand for chlorine.
       Caustic soda and caustic potash,  which are produced simultaneously with
  chlorine in approximately equivalent quantities,  are also basic chemicals that
  are used in a variety of industrial applications.  Some substitution is possible
  here;  for example,  the choice can often be made between caustic soda (NaOH) and
  soda ash (Na2CO-).   However,  the  choice of caustic soda is often determined by
  economics:   its supply is dictated by  chlorine production patterns and the
  resultant caustic  soda must be  sold,  often at prices which beat out other
  competitive chemicals.
      It  is  our conclusion that  elimination of these products—chlorine and
  caustic—or substitution  of other materials is not feasible and can be discarded
  as a possible  alternative.

 Alternative  2—Alternate Technologies

      The use of alternate technologies  is  already feasible,  because more than
 three-quarters of the chlorine and caustic soda produced comes from diaphragm
 electrolytic cells which use no mercury.   Mercury cells and diaphragm cells
 have often been compared, and the  mercury  cell is the most favored choice.
 However,  the advantage of the mercury cell (which is basically the production
 of  a  concentrated,  salt-free caustic) has  been largely  negated by the major
 disadvantage of mercury emission problems.  Mercury  cell technology does not
have a  clear economic advantage over  diaphragm cell  technology and does have a
disadvantage with  respect to requirements  for  control of mercury emissions—
some unregulated discharges prior  to an understanding of the problems entailed
                                       258

-------
resulted in some bad publicity for the industry.  Therefore the  industry  has,
on its own, shifted to the diaphragm cell for  new plant technology.  In 1968
about 30 percent of the chlorine produced came from mercury cells.   In 1970
this had dropped to around 28 percent; in 1972,  to  24 percent; and  in  1973, to
less than 24 percent.  It is noteworthy that all new caustic  chlorine  units are
diaphragm cells (Ref. 140).  And even with the technology of  the diaphragm cell,
there are substitute materials that improve power utilization while eliminating
the asbestos (itself a pollutant source) in the diaphragm.  DuPont's Nafion,
a nonasbestos material, is now in commercial use, and in Japan both Asahi Glass
Company and Tokuyama Soda Company have their own nonasbestos  synthetic membranes
under development.
     In this case straight substitution of the diaphragm technology for the
mercury cell process is deemed technically feasible.  Since no unreasonable
differences in the final products result, there are no impassable barriers to
the changeover other than the costs of making the change.  The major disadvan-
tage of the conventional asbestos diaphragm cell is that it produces a more
dilute caustic which contains salt.  To produce a generally useful product,
concentration of the caustic from 11 percent to 50 percent and lowering the
salt content from 15 percent to  2 percent must be achieved.  This process
requires large amounts of thermal energy, offsetting much of the cost advan-
tage of the higher electrical efficiency of the diaphragm cell.
     To change the mercury cell  chlor-alkali plants to diaphragm or synthetic
membrane plants would involve significant costs.  These  costs would include
new capital investment for land  and facilities, the assumption being  that
existing mercury cell plants would  be kept operating  to  meet market demands
for product until the replacement units were  on-stream.
     Data  for 1971 from Ref. 63  give the capital investment  required  for a
100-ton/day plant as $8 million.   (This includes a $540,000  credit for mercury
recovered.)  Also for  1971, the  annual operating costs  for such a plant  are
given as being in the  range of  $2.4 million,  equivalent to about $73 .per daily
ton of chlorine.  Later  data  (1974)  from unpublished sources give an operating
cost  (raw  materials, energy,  labor and maintenance) for such a diaphragm cell
                                       259

-------
 plant of $84 per daily ton of chlorine.*  These sources  also provided cost data
 on mercury cell plants which indicated that operating costs per ton of product
 are virtually identical for either route when the  costs  include those for the
 purification and concentration of caustic from diaphragm cell plants.
      Using these 1974 data, it was assumed that the only major cost for convert-
 ing from mercury cells to diaphragm cells would be the capital investment.  At
 $8 million per 100-ton/day capacity,  it would require new investment on the
 order of $610 million to produce the 2.5 million tons now produced by mercury
 cell technology; assuming a 15-year depreciation period, the depreciation cost
 would average $16 per ton of chlorine** (compared  to the current estimated pro-
 duction cost of $84 per ton).   The changeover to diaphragm technology would
 represent a hardship on mercury cell producers only if imposed over a brief
 and arbitrary period, forcing simultaneous monetary requirements for old plant
 operation and new technology investment.
      The benefits of the conversion would be  the removal of as much as 326,000
 kg of mercury discharges to the environment each year.   Since the total man-
 originated contribution for all sources studied in this  analysis amounts to
 1,525,000 kg annually,  this would represent a decrease of about 21 percent.
 At the same time, the dismantling of  all  mercury-using chlor-alkali plants
 would release an estimated 3.6  million  kg of  the metal to the market.  (As
 previously noted, the value of  this mercury has been credited in accounting for
 the capital investment  required to convert to diaphragm  cells.)
  * Confidential information obtained during field contacts by URS Research
    Company.
** Depreciation costs are, of course, included in total cost.  In the case of
   chlorine from mercury cell plants it must be recognized that most of these
   plants are at least six years old and on the average are probably at least
   half depreciated.  Also, a portion of the $16 per ton might be charged to
   the co-product, caustic soda, but since chlorine demand is normally the
   determining factor, such an allocation would be difficult and ultimately
   arbitrary.
                                     260

-------
Alternative 3—Increased Control

     The present technology of control of mercury emissions is  recognized as
being capable of reducing mercury emissions to the biosphere from the 1973
average 0.37 Ib/ton  (0.185 kg/metric ton) of chlorine to about  0.25  Ib (0.125
kg) and, with more stringent controls, to a level of 0.04 Ib/ton (0.020 kg/
metric ton) (Ref. 141).  However, a more realistic goal for 1983 is  0.10 Ib/ton
of chlorine (0.05 kg/metric ton).  To relate such a reduction to the individual
losses to air, water, and land, we have examined certain control technologies
which will reduce emissions:
     Air.  Molecular sieves for use on hydrogen and end box ventilation streams
have the capability of reducing the mercury content of these streams by 95
percent over current EEA standards (from 1.0 kg/day to 0.045 kg/day).  Cell
room ventilation losses, which currently are set at a maximum of 1.3 kg/day,
are unlikely to be affected by new technology (although the amount of daily
emissions may be lowered through better housekeeping).*  Hence daily overall
losses to air, assuming an operating removal efficiency for the molecular sieve
unit of 90 percent, cannot be reduced by more than 40 percent.
     Water.  Sulfide precipitation treatment for final polishing of water
effluent from the plant has the capability of reducing discharges to water by
more than 94 percent.
     Land.  High temperature roasting of sludges in multiple hearth furances
can recover more than 99 percent of the mercury content.  Other solid waste
losses, such as graphite from decomposers and miscellaneous contaminated wastes,
are not considered amenable to such recovery practices, however;  since  these
represent about 23 percent of solid waste losses, the maximum  reduction possible
is 78 percent, assuming a conservative recovery of mercury from sludge  of 97
percent.
* Concentrations of mercury vapor in cell rooms might be appreciably reduced if
  the  industry as a whole were to adopt the "open air" cell room concept now in
  use  at PMC's Squamish,  British Columbia,  plant (Ref. 142).  However, this
  concept would not reduce total emissions to air.
                                       261

-------
       The cost of pollution control equipment was  derived  from  several  sources
  and the total cost (in constant dollars)  for the  industry was  estimated using
  the familiar "0.6" rule (Ref.  63).   Operating costs were  similarly estimated
  where possible.   The basis for these cost estimates is discussed briefly below.
       Reference 63, which provides  a comprehensive analysis of  the cost of con-
  trol and conversion for the chlor-alkali  industry, indicates a capital cost in
  1973 of $349,000 for a molecular sieve unit  for a 100 ton/day  facility, with
  annual operating costs of $105,000.   The  total cost of equipping all 28 mercury
  cell chlor-alkali plants with  such devices is calculated  to be $8.6 million,
  with total annual operating costs  of $2.6 million.  A review of the trade
  journals and contacts with representatives of the industry indicates that these
  are reasonable estimates.   The use  of other  control technologies, such as
  treated activated carbon,  would have similar values attached for cost  and
  efficiency.
       The use of  sodium sulfide to precipitate the very insoluble mercuric
  sulfide seems to be generally  accepted as the least expensive, most trouble-
  free,  and most effective  method for  removing traces of mercury from effluent
  waters prior to  discharge.  References 68, 141, and 143 describe several varia-
  tions  of this treatment,  including results for both pilot and  operating plants.
  Cost data were conflicting—Ref. 141 indicates a  capital investment of $500,000
  for a  sulfide treatment unit for a 175 ton/day (chlorine)  capacity plant,
  whereas Ref.  143  indicates a capital cost of only $143,900 for what appears to
 be a plant of the  same size.  Hence in our estimate we assume  an average capital
 cost of $300,000, with an annual operating cost of about $100,000.  Thus the
 estimated  total capital cost for equipping all 28 plants in the nation with
 this type  of technology would be $10.1 million, with total annual operating
 costs of $1.5 million.
      The use of sodium borohydride  treatment is considered an acceptable alter-
 nate to the use of sodium sulfide (Ref.  143), but the costs are somewhat higher
 and the efficiency is slightly less.  Activated carbon has also been tested
but appears to be less suitable than sodium sulfide.
     Two competitive technologies have been identified for the recovery of
mercury from  sludges:   high-temperature roasting of  sludges (the so-called
Georgia-Pacific process,  described  in detail  in Ref.  143)  and solubilization
                                       262

-------
of mercury and its compounds by a hypochlorite solution and subsequent reduc-
tion of the dissolved mercury to elemental mercury in the chlorine  cell (as
described "in U.S. Patent 3,691,037,  12 September 1972,  assigned to  FVC. Corpora-
tion, Squamish, B.C.).  Capital costs for the roasting technology (from Ref.
143) are estimated at $364,500 for a 200-ton/day (chlorine)  facility with  annual
operating costs of about $100,000.  Costs for the alternative technology,
hypochlorite dissolution, were not considered, but since some plants are even
now adopting this or similar technologies, it can be assumed to be  cost-
competitive.
     The total investment over the 10-year period 1973-1983 for control of
mercury emissions to land is an estimated $12 million, with annual opera-
ting costs of about $3.3 million.

Introduction of New Emission Control Technology

     Reduction of mercury emissions by using the newer recovery techniques is
not likely to occur suddenly.  Some time will be required for the engineering
design, construction, and bringing on-stream of the added facilities.  It is
estimated that, in general, three years would be a maximum time for these new
recovery and control facilities to be put into operation.  It is also probable
that not all of the mercury-cell plants would undertake or complete the instal-
lation of such new facilities simultaneously.*
     Figure 55 indicates the assumed rate of  adoption of pollution control
technology, and Table 44 summarizes the total cost of such technology and the
total emissions, prorated in the  manner shown in Figure  55.  As can be seen
from Table 44, over the 10-year period the total cost  for control  of  emissions
to each receiving environment is  on the same order  ($8  million  to  $12 million).
* Plants suffering the greatest losses would probably be the first to make the
  improvements.   However,  this is an assumption and cannot be proven before the
  fact.   Many plants will not have to make additional investments in order to
  meet  the projected average 1983 emission levels.
                                        263

-------
  100
   90
   80
   70
c
O
•o

£  50
   20
   10
                                                                76%
                                                                                                         81%
                                                   67%
                                         - 56%
                                32%
                      21%
                                                                                    1         1
1974       1975       1976      1977      1978      1979

                                            Year
                                                                         1980
1981
1982
1983
SOURCE: URS Research Company. '
                     Figure 55   RATE OF INTRODUCTION OF POLLUTION CONTROL TECHNOLOGY

-------
      TABLE 44. -  Emission Control Technology for Mercury Cell Chlor-Alkali Industry:   Estimated Annual Cost and Emissions, 1973-1983
Year
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
Water Land
Percent Emissions Capital Cost
Reduction (kg) ($ x 10 )
0% 4,295 $ 0
20 3,436 2.26
38 2,663 4.47
54 1,976 6.34
68 1,374 8.07
80 859 9.51
90 430 9.85
94 250 10.10
94 250 10.10
94 250 10.10
94 250 10.10












Percent Emissions Capital Cost
Reduction (kg) ($ x 10 )
0% 298,310 • $ 0
22 232,682 3.47
37 187,935 5.72
54 137,223 8.32
64 107,392 9.81
73 80,544 11.29
78 66,362 11.98
78 66,362 11,98
78 66,362 11.98
78 66,362 11.98
78 66,362 11.98
Air Total
Percent Emissions Capital Cost
Reduction (kg) ($ x 10 )
0% 23,156 $ 0
11 20,610 2.42
18 18,988 4.54
27 16,904 6.24
32 15,746 7.50
36 14,820 7.94
38 14,357 8.30
40 13,894 8.60
40 13,894 8.60
40 13,894 8.60
40 13,894 8.60
Percent Emissions Capital Cost
Year Reduction (kg) ($ x 10 )
1973 0% 325,761 $ 0
1974 21 256,728 8.15
1975 36 209,586 14.73
1976 52 156,103 20.90
1977 62 124,512 25.38
1978 70 96,223 28.74
1979 75 81,149 30.13
1980 75 80,506 30.68
1981 75 80,506 30.68
1982 75 80,506 30.68
1983 75 80,506 30.68
to
01
in
     Annual

     Operating

     Cost
$1,500,000
$3,300,000
                                                                $2,600,000
$7,400,000
     Source:  URS Research Company

-------
 but greatest overall  reductions are achieved for water  (94 percent) and the
 least  for  air  (40 percent).  However, the extent of reduction is of less impor-
 tance,  overall,  than  are the absolute quantities involved.  Thus the total
 quantity of mercury discharged to air drops from some 23,156 kg in 1973 to
 13,894 kg  in 1983;  this represents a decrease in the air burden of almost
 10,000 kg.  On the  other hand, the absolute decrease in emissions between 1973
 and 1983 with  respect to water is less—about 4,000 kg—but of course their
 removal from receiving waters is an important contribution (note that further
 decrease of emissions to water is most unlikely, and probably most unnecessary).
      Emissions to land, while not of serious concern as far as exposure to man
 and his environment are concerned (at least in the short term),  are reduced
 significantly  (78 percent) with an absolute value of more than 230,000 kg
 "saved."*  In  terms of cost-effectiveness the industry gains most,  dollar-wise,
 by investing in  control technology for its mercury-laden sludges.  And because
 these  sludges  are removed from a repository which might eventually release
 their  contents to the environment,  the general public also benefits.
      In Table  44 capital costs,  although calculated on a year-to-year basis,
 are shown as cumulative,  whereas operating costs are shown only on an annual
 basis.   The omission of operating costs from the year-to-year values is accept-
 able because it is the capital equipment that is the more significant cost and
 the one which weighs heavily in management decisions.**  The availability of
 internal capital resources and the  cost of capital in external money markets
 also enter the  picture.  A calculation was made to assess the inflationary
 effects of two  interest rates—6 percent and 10 percent (these rates consider
 the  time value  of money in constant  dollars).   At the expenditure rate indicated
 * A recent report  (Ref.  60) projects  that mercury cell solids will decrease
   from 270,000 kg  in  1973  to  120,000  kg  in 1978  and to 4,000 kg in 1983.   This
   latter value, which is far  below  our estimate,  appears to include effective
   treatment of high-level  mercury solid  wastes  (the URS estimate considers such
   wastes as untreatable).
** Also,  much of the new  equipment represents an  upgrading of present practices
   and this new equipment is very likely  to be operated by little or no addi-
   tional labor (a potential disbenefit).
                                      266

-------
in Figure 55 and for the total cost of -new emission control technology in
current dollars ($30.7 million), the effective investment for the 6 percent
rate at the end of the 10-year period is $48.7 million, and for the 10 percent
rate it is $65.7 million.
     The imposition of additional emission controls is  seen by the industry as
representing an investment of funds which does not contribute to productivity—
or to profits.  However, the savings of mercury recovered through control
measures can offset some of the operating costs.   Over  the ten-year period
considered, more than 2 million kg of mercury would be  recovered; at a cost of
$8.80/kg this amounts to almost $18 million dollars.

Cost-Benefit Analysis

     Two practicable alternatives have been identified—conversion of mercury
cell plants to diaphragm cell technology, and the reduction of emissions from
present mercury cell plants by the imposition of stiffer emission controls.
The cost-benefit results for these two alternatives are presented in Figure
56.  The figure also includes a base case which assumes no growth within the
industry and emissions continuing to meet present EIA standards.  The benefit
accrued is shown as the percent reduction in emissions of  mercury to environ-
ment and the costs as total capital costs.*
     Alternative 1, conversion to diaphragm technology over a 10-year period,
ultimately produces the maximum benefit—100 percent reduction in mercury
emissions.  However, the associated cost of more than $600 million is  clearly
of a magnitude that would have major impact on money markets,  to say nothing of
the effects on corporate growth.
* It must be recognized that capital costs do not necessarily equate with
  social costs.  For example, if the capital costs can be absorbed through
  higher prices  (a common occurrence), the "cost" will only appear at the
  consumer's end, probably magnified—appearing as a reduction of prior
  consumer surpluses  (also known as "foregone benefits").  If the consumer  is
  unwilling to pay for these nonproductive investments,  several other cost
  effects might occur, including plant closures and resultant unemployment;
  lower net income; or no plant expansion and a resultant constriction  of
  supply.
                                       267

-------
 100 r
                                           Alternative 1
                                      (Convert to Diaphragm Cell)
                                         Cost: $610 million
                 Alternative 2
        (Reduce Emissions to 0.1 Ib/ton CL2
              Cost: $31 million
-10
  1973
1974
1975
1976
1977
1978
Year
1979
1980
1981
1982
1983
   SOURCE: URS Research Company.
       Figure 56  TIME-PHASED REDUCTION OF MERCURY EMISSIONS FROM
                 MERCURY CELL CHLOR-ALKALI PLANTS, AND
                 ASSOCIATED COSTS
                                    268

-------
     The second alternative can be effected more rapidly, since it represents
only a minor modification of an existing structure, at a cost which can probably
be assimilated easily (if not willingly).   However, even with this vastly
improved control over emissions of mercury, the residual losses (primarily to
land and air), would still be significant,  and pressure might continue for the
removal of even these amounts.
     A third alternative (not shown) is likely if  pressures  for control continue.
This alternative involves the gradual phasing out  of  mercury cell plants, start-
ing with marginal plants which cannot economically justify the cost of higher
level emission control technology and/or which cannot compete with chlor-alkali
produced in diaphragm cells.*  To some extent this third alternative  is in
effect even now, since all new expansions in chlor-alkali are in diaphragm cell
technology.  However, any effort to force too short a time  frame on this  third
alternative could have serious consequences for the industry and its  customers.
But if reasonable latitude is allowed in recovering investment  on  existing
plants, the impact on the industry should be minimal.

Risk Assessment

     Alternative 2,  increased use of emission control technology,  has no fore-
seeable risks associated with its adoption.  But Alternative 1, conversion to
diaphragm  cell technology,  does  carry  risks, .since diaphragm cells are in them-
selves a source of  toxic pollutants.   Reference 60 estimates that chlor-
alkali plants using diaphragm cell technology will produce  solid wastes
as follows:
 *  The competitive advantage of one type of cell over the other is not clear and
   appears to depend heavily upon utility costs and the type of product required.
   Also,  advances in the efficiency of the diaphragm (both in terms of electrical
   efficiency and the concentration of the caustic produced) may give it an even
   better competitive advantage.  Even so, mercury cell technology may still be
   the obvious choice for plants where steam costs are prohibitive, or where
   customers require salt-free caustic, and so on.
                                       269

-------
                                    Thousands of Kilograms
                      Year          Asbestos          Lead
                      1973            2,300             980
                      1978            3,800             900
                      1983            2,000             490

      If mercury cell production of  chlorine  accounted  for  20 percent of total
 production by 1983,  complete conversion to diaphragm cells would apparently add
 400,000 kg of asbestos  and 98,000 kg  of lead to these  solid waste totals.
 However,  new technology in diaphragm  cells is eliminating  the use of asbestos
 diaphragms and reducing the reliance  on lead.   Hence the additional waste load
 could be rather small.   (As an aside, it is  interesting that discharges from
 the diaphragm cell technology are decreasing less rapidly  than those from
 mercury cell technology,  even accounting for the growth of the diaphragm cell
 technology.   The difference,  we suspect, has to do with the value of the
 recovered waste; neither asbestos or  lead has sufficient value to spur its
 recovery,  but mercury does.)
     Emissions  from mercury cell chlor-alkali plants to water and air are of
most concern, since these can  directly affect the surrounding environment.
Therefore, we made an estimate of the population which might be affected by
such discharges.  We determined the resident population (using 1970 Census data)
within a 5-mile radius of each of the 28 mercury cell chlor-alkali plants (a
total of 8,029,000 persons) and then estimated per capita emissions as follows:

                                               Milligrams
                      Year      Kilograms      per Capita
                                                  1.86
                                                  1.50
                                                  1.17
                                                  0.89
                                                  0.62
                                                  0.44
                                                  0.39
                                                  0.37
                                                  0.37
                                                  0.37
                                                  0.37
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
14, 910
12, 007
9,361
7,195
4,948
3,567
3,131
2,973
2,973
2,973
2,973
                                      270

-------
     The per capita emission factor (primarily from final consumption of
mercury-containing products) amounts to 2.185  mg/person for  air  and water
combined.  Thus the 1973 emissions of 1.86 mg/person from the chlor-alkali
plants do unpose an appreciable added burden on the affected 8 million popula-
tion.  However, with reduction of chlor-alkali emissions to  a projected value
of 0.37 mg/person by 1983, the added burden would be only 17 percent,  and is
of less concern.

REGULATION IN THE PAINT INDUSTRY

     The use of mercurials in the paint industry is largely  confined  to water-
based product manufacture.  The scope of this  analysis, accordingly,  is  limited
to two product groups:  exterior water-type paint products and tinting bases
(SIC 28512) and interior water-type paint products and tinting bases  (SIC 28514),
     Figure 57 relates mercury to its host economic process, exterior and
interior latex paint manufacture.  In terms of product flow, the integration
of this element into the paint manufacturing sequence may be viewed in five
distinct stages.
     1.  Production.  Approximately 7,600 flasks of mercury were produced
during 1973 for use in water-based paint manufacture.  This represents about
14 percent of the total 1973 mercury supply.  Since only a small portion of
the U.S. supply derives from domestic mining operations, it can be safely
assumed that a similarly  small U.S. mining revenue  (perhaps $100,000 in 1973)
generates from paint-related requirements/ and that very little mining employ-
ment  (perhaps the equivalent of  10 to 15 workers)  is paint-dependent.
     2.  Process.  Before it is  incorporated  in water-based paint, mercury is
subjected to an intermediate process.   It  is  converted into phenylmercuric
acetate  (PMA)* for use as a preservative and  mildewcide.  Approximately  one
million pounds of PMA were processed from  mercury during 1973 for use in paint
manufacturing.  The related value realized by the PMA industry is estimated as
$13 million to  $14 million.  Related employment may range  from 200  to 250
workers.
* Other types of phenylmercurics are less commonly used,  but the differences
  are  of  no  importance in this analysis.
                                        271

-------
       3.   Paint Manufacture.   Mercury,  reconstituted as PM&, contributes to the
  manufacture of water-based interior  and  exterior paints.  Approximately 260
  million gallons of these paint products  were produced in 1973.  Mercury (as
  PMA)  therefore contributed to the  generation of nearly $1 billion in revenues
  and to employment support for approximately 15,000 workers.
       4.   Paint Distribution and Marketing.  The distribution and marketing of
  mercury-bearing paints  occupies another  economic sector.  Wholesaling and
  retailing functions are intermediate between the manufacturing process and the
  consumer.   Following the same 260  million gallons of water-based paint through
  this activity reveals total  revenues of  $1.2 billion in 1973, and an implied
  associated employment of perhaps 25,000.
       5.   Consumption.   U.S.  consumers, through their demand for latex paints
  or for products or services  which  require such paints, stimulated the produc-
  tion and provision of the 260 million gallons of water-based paints in 1973.
  Consequently,  consumers ultimately had to pay the bill.  This bill may be
  disguised by  the  indirect channels through which it reaches the individual
  consumer (for instance  as a  component of furniture cost,  or housing cost,  or
  of his government tax bill), but eventually the entire revenue flow—from
  $1.2 billion  in final product cost to $100,000 in mercury mining revenues—
  will generate  from the  individual consumers.

 Regulatory Alternatives and Effects

      For purposes of this analysis, only one regulatory option is examined:
 the staged (phased) elimination of all mercury use in water-based paint manu-
 facture.   The analysis assumes the accomplishment of the transition to substi-
 tute technology within five years.  This regulating policy is examined relative
 to a theoretical base condition defined as "no regulation."
      Mercury regulation  would have  the  effect  of  cutting off the flow of a
 significant ingredient,  PMA,  in the water-based paint manufacturing process.
 In Figure 57,  a barrier  may be visualized midstream in the product flow diagram.
Two consequences are immediately apparent.  First,  all product flow,  revenue,
and employment that previously were realized behind that  regulatory screen will
                                       272

-------
to
-J
CO
         PRODUCT
         FLOW
         INCOME
         FLOW
         DIRECT
         EMPLOY-
         MENT
                     MERCURY
                    PRODUCTION
                      MERCURY
                     PROCESSING
        INTEGRATION TO
       MANUFACTURING
                                                 MERCHANDISING
                                                  OF PRODUCT
                       Convert
                         to
                        PMA
          Water-Based
437,800 Kg}   Paint
          Manufacture
7571 Flasks
                    (U.S. Mines)
                                                              Water-Based
U.S. Mines  (  $90 000
                                                              Manufacture
                             CONSUMPTION
          Distribution- j        \    Water-
258,000,000\ Wholesale- 1258,000,000\   Based
 Gallons /   Retail   |  Gallons /    Paint
       /
                                                     Consumer
                                                      Cost
         SOURCE: URS Research Company.
                                                   Regulation
                             Figure 57  PRODUCT AND INCOME FLOW FOR THE PRODUCTION AND
                                       CONSUMPTION OF MERCURY-CONTAINING PAINTS

-------
  be jeopardized.   Second,  if product  supply is to be maintained, substitutes
  for the regulated ingredient must be incorporated into the process.
       As stated earlier,  the mercury  production and processing aspects of water-
  based paint manufacture  generated total revenues in excess of $13 million and
  total paint-related employment of perhaps 265.  Table 45 indicates the scale
  of product,  income,  and  employment that would attend a base-case condition
  expanded at a 5 percent  rate of growth between 1973 and 1983.
       The elimination of  mercury and  mercury-dependent processes from the paint
  industry might eventually threaten the jobs of 430 workers and a business
  income of approximately  $22 million  per year.  Under worst-case circumstances
  and within a very narrow frame of reference, this may in fact reflect some
  small part of the industry's future.
       In a broader sense,  though, other compensatory changes will be created by
  these disruptions.   Some PMA plants, together with their revenue and employment
  bases,  will  be preserved by accomplishing the transition to substitute processes.
  In addition,  a major new  domestic industry—that of substitute process—will be
  catalyzed around  them.
       Few  secure predictions  can be made regarding the development of a substi-
  tute  process  industry and its incorporation into the water-based paint industry.
 Recent  substitute technology experience,  however,  indicates several areas of
 potential impact:
      o  Increase in Manufacturing Cost/Product Price.   Paint industry repre-
 sentatives estimate that the utilization of current mercurial substitutes would
 result in increases ranging from $0.15  to $0.25 per gallon in manufacturers'
 product prices.  Retail prices  may increase $0.60  to $1.20 per gallon.
      o  Degradation in Product  Quality.   Tests of  the  service life of water-
 soluble paints containing substitutes for PMA. indicate potentially significant
 reductions in paint permanence.
      o  Market Ramifications. The  demand characteristics of the American paint
 consumer have been relatively stable  over the last 10  years.   Figure 58 portrays
 the historic  pattern of consumer demand behavior in the face of price levels.
For purposes  of comparison,  approximately 10  years of  monthly price/supply
relationships  were calculated and normalized  to  1973 market proportions and
                                       274

-------
         TABLE 45.  -  Estimated Product and Income Flow for
                      Production and Consumption of Paints
                      Containing Mercury, 1973-1983:  Base
                             (No Regulation) Case
A.  Product Flow
Year
1973
1978
1983
B . Income
Year
1973
1978
1983
Total
Supply
(flasks)
7,571b
9,663b
12,333b
Flow (thousands
U.S. Mine
Production
$ 90.0
114.9
146.6
Process
(000 Ibs)
965. 4
1,232.1
1,572.5
Mfg.
(000 gal)
258,000.0
329,285.4
420,267.0
Retail/Wholesale
(000 gal)
258,000.0
329,285.4
420,267.0
of 1973 dollars)
Related
Process
(PMA)
$13,500.0
17,230.1
21,990.7
Related
Mfg.
$ 912,000.0
1,163,985.6
1,485,594.8
Related
Retail/Wholesale
$1,200,000.0
1,531,560.0
1,954,730.0
C. Related Employment (number of workers)
Year
1973
1978
1983
U.S. Mine
Production
15
19
24
Related
Process
(PMA)
250
319
407
Related
Mfg.
15,000
19,145
24,434
Related
Retail/Wholesale
25,000
31,908
40,724
 a. Assumes a growth rate of 5 percent per year.
 b. This total includes foreign and domestic sources,
   uted only 4 percent.
 Source:  URS Research Company
U.S.  mines contrib-
                                      275

-------
a\
          5.10
           4.85
           4.60
        5
        r-
        o>
        J
        ro
        01
           4.35
           4.10
           3.85
           3.60
                         *  Adjusted using historic consumer price indexes for all items from "Survey of Current
                            Business," U.S. Department of Commerce.
                         **  Adjusted to neutralize historic differences in the scale of the marketplace.
                                                                                               September-December
                                                         Units
                                                       30
60
78
90
99
                                                                   Relative Volume of Demand**
        SOURCE: Derived by URS Research Company from data in "Currant Industrial Reports;
                 Volume and Sales of Paint and Varnish/' 1964-1974.
                                         Figure 58  PRICE ELASTICITY OF DEMAND IN THE PAINT INDUSTRY

-------
1973 dollar values.   As shown,  the  consumer's price thresholds vary with the
seasons:  during the summer he  will tolerate price changes of up to 25 percent
before his demands will shift;  during the  fall/winter period his demands are
naturally lower but his willingness to accommodate price changes is also
reduced—perhaps to 15 percent;  during winter/spring his demands are typically
higher, but so is his willingness to tolerate price changes—down to 10 to 12
percent.
     From these data it would appear that  demand  for paint products may be
adversely affected by imposed substitutes, particularly in off-peak seasons.
     It is ironic that another  adverse quality  of mercurial substitution—
shortened product service life—may (at  least as  far as the industry is con-
cerned) serve to offset the price-stimulated contraction of demand.  It remains
to be seen, though,  whether the consumer's desire for equal product service
can be reconciled with his limited  disposable income and his reluctance to
surrender other valued goods and services. As  of this moment  there are not
enough behavioral precedents to enable quantification of this  aspect of con-
sumer behavior.
     Table 46 estimates the income  and employment consequences of a  transition
from current mercurial processes to substitute  technology.  The higher cost
of mercurial substitutes results in significantly greater  income to  the process
industry in general, and through the effects  of marginal pricing also results
in greater income to the activities that follow manufacturing.  Employment is
expected to be tripled by 1983 from the  base  case within the process industry
(from about 400 for the PMA process to more  than 1,200 for substitute processes),
while remaining essentially unchanged in related manufacturing and merchandising.
These gains are both small (relative to  the industry that surrounds them) and
highly tentative until the price-pull, quality-push forces on consumer demand
are resolved.  The estimated 1983 price  per gallon penetrates the $5.10 per
gallon (1973 dollars) threshold beyond which demand diminution has historically
been experienced.  To place this threat in perspective,  a 10-percent  contraction
in demand in 1983 would jeopardize the jobs of more than 6,000 industry workers.
     By way of summary, Figure 59 graphically displays these  various  effects in
net terms.  Relative to the baseline  condition detailed in Table  45,  mercury
                                       277

-------
   TABLE  46.  -  Estimated  Income and Employment Consequences of a Shift
                to  Substitutes for Mercury in Paint Manufacture and Use

 A.   Income Flow  (thousands of 1973 dollars)
U.S. Mine
Year Production
1973 $90.0
1978
1983
Related Processes
PMA Substitute
$13,500.0 $ —
83,090.0
106,050.0
Related
Mfg.
$ 912,000.0
1,253, 100. Ob
1,539. 135. Ob
Related
Retail/Wholesale
$1,200,000.0
1,691,980.0°
2,159,063.0°
B. Employment (number of workers)
U.S. Mine
Year Production
1973 15
1978
1983
Related Process
PMA Substitute
250
978d
l,248d
Related
Mgf.
15,000
19,145
24,434
Related
Retail/Wholesale
25,000
31,908
40,724
a. Assuming substitute price 20 cents/gallon above PMA.
b. Assuming maintenance at an operating margin of 35 percent.
c. Assuming that the collective margin requirement with the merchandising
   chain will approximate 80 percent.
d. Assumes an average value per worker of $85,000 per year.

Source: URS Research Company
                                   278

-------
  1973 0
to
-J
ID
                                             Consumer Costs — dollars
                                                                                                          -431
                                                                                                          204.3M
                SOURCE: URS Research Company
                             Fiaure 59  CONSUMER AND MANUFACTURING COSTS ASSOCIATED WITH

                                      THE REMOVAL OF MERCURY FROM PAINT

-------
  regulation can be seen to impose negative consequences upon mercury miners and
  PMA plant workers but at the same time to make a positive contribution in terms
  of substitute technology employment.  Ultimately, though,  the consumer will
  incur the largest unrelieved disbenefit.   By 1978,  if mercury substitutes are
  adopted, consumers will have spent $160 million more,  for products that may
  yet be inferior in service value,  than would be true for mercury-containing
  paints.  By 1983, this annual net cost (loss of prior consumer surplus or
  "foregone benefit") would amount to more  than $200  million.

  Cost-Benefit Analysis

       One alternative—the elimination  of  mercury from  paint—and two base cases
  have been considered.   Base  case 1 assumes that mercury usage  in water-based
  paints will remain constant  despite a  forecast  growth  in paint sales;  that is,
  all new market growth will be  for  substitutes.   Base case  2 assumes that mercury
  use will grow  at the same rate as  total paint sales.  These three curves are
  shown in Figure  60, which indicates  the percent  reductions (or  increases)  in
  mercury emissions  over  time.  No costs are assigned to the base  cases, and only
  costs to the consumer are assigned to the alternate.   (As noted  in the previous
  discussion the elimination of mercury from paint will cause some  redistribution
  of revenues within the mining, manufacturing, and final consumption sectors,
  but the net change will be an increase in the total number of workers in the
  associated industries.)   These consumer costs are estimated to amount to more
  than $1.4 billion over the 1973-1983 period.   (This is reflected as an additional
  constant-dollar cost of 50 cents per gallon to the consumer by 1983.)  The
 benefit,  of course, is  the elimination of  paint as a source of mercury emissions
 to the environment—a source  that discharged some 183,000 kg of mercury  in 1973
 alone.

 Risk Assessment

     No risks associated with the elimination of mercury from paint have been
quantified, but some possible risks which merit  further consideration have been
noted.   The materials used in place of the mercurials in paint  are, of themselves,

                                      .280

-------
                                              Alternative 1
                                              (Complete Elimination of Mercury)
                                              10-Year Cost to Consumer: $1.4 billion
                                                       Base Case 1
                                                 (Stability of Mercury Use)
                                            Base Case 2
                                    (Continued Increase of Mercury Use)
                              1977     1978    1979     1980    1981    1982    1983
SOURCE: URS Research Company.
           Figure 60  TIME-PHASED REDUCTION IN MERCURY LOSSES
                     FROM PAINT USAGE, AND ASSOCIATED COSTS
                                      281

-------
  toxic to a degree, but unlike mercurials they do not appear to migrate from
  the painted surface to air.  The introduction of these mercurial substitutes
  into receiving waters is undesirable, but as shown by our inventory results,
  the amount of paint that enters sewers is small.  Dermatitis resulting from
  the use of latex paints containing mercurial substitutes has been reported by
  some persons, but the extent of the problem is rather limited.
       Some representatives of the paint industry report the growth of
  Pseudomonas species on walls painted with nonmercurial,  water-based paints.
  They suggest that such growth could have serious consequences in some settings,
  such as hospitals.  However,  confirmed instances have not been  recorded.
       Virtually all loss of mercury associated with paint occurs soon after the
  paint is applied,  and these losses are almost entirely to air.   Concentrations
  can be high during and immediately following  painting, but the  consequences,  if
  any,  do not appear to concern the  general population,  whose frequency of
  exposure to newly  painted  rooms  is  undoubtedly low.   However, we can compare
  the per capita  emission to  air with the per capita values  for paint  alone.
  The first value is 1.80 mg/person/year; the emissions  to air from paint usage
  account for 0.87 mg/person/year—roughly half of the first  value.

  REGULATION  IN THE BATTERY INDUSTRY

      The use of mercury in the battery industry centers in three product areas:
  zinc-carbon, alkaline-manganese dioxide, and Ruben mercury batteries.  Collec-
  tively, these manufacturing activities required 30 percent of the mercury used
  in the United States in 1973.   The larger industry group from which these
 products derive  is  SIC 3692,  Primary Batteries,  Dry and Wet.
      Figure 61 relates mercury to these three  principal product  manufacturing
 processes.   Mercury can be  traced from its  source of  supply or production to
 its  terminus (consumption)  in  five  developmental stages.
      1.  Mercury Production.   Approximately 16,260  flasks of mercury were
produced by  domestic and foreign  suppliers for use  in battery manufacture in
1973.  Only a small fraction of that supply was  developed from then-current
domestic mining operations.  Indeed, U.S. producers contributed  only 4 percent
                                       282

-------
00
LJ
              MERCURY
             PRODUCTION
 BATTERY MANUFACTURING
          SIC3692






16,260







455


,
8 Workers
	
878(
y
U.S. Mines ^S
f"
16 Workers


2927
3 Flasks
1 *
$60,000
J~

mm* j
) Flasks
1i»l
mo,oo<

ip

™» f»
Flasks
                                         Process
         160,000
                             MERCHANDISING
              Mfg.
           Carbon-Zinc
                                                ^ $4.3 M

                                               -Y
           78 M Kg \
                                       CONSUMPTION
                                    >
                                                                           78 M Kg \
                       *«or,    i  Wholesale/  _
                       $187M  '   Retail   <  $243M
                                                              Consumer
                                                                Cost
Process
300,000 Kg^
  Mfg.
 Alkali-
Manganese
                                                      6.7 M Kg
                                               ^ $8.0 M
                                               -Y—
                               <$77M
                             ,  	
           6.7 M Kg ^
                                Wholesale/
                                  Retail
                                        Process
                      Mfg.
        100,000 Kg >  Ruben
                     Mercury
                        esa'e/ /$^7T
                        tail   X 	
                        —V
                         2100
                                                                                     Consumer
                                                                                      Cost
                      1.9 M Kg \
                               1.9 M Kg \
U.S. Mines ^  $36,000
                                                  $2.7 M
                               $29 M
Wholesale/
  Retail
                                           $38 M
                                         Consumer
                                           Cost
                            Regulation
     SOURCE:  URS Research Company.
                          Figure 61   PRODUCT AND INCOME FLOW FOR THE PRODUCTION AND
                                     CONSUMPTION OF MERCURY-CONTAINING BATTERIES

-------
  to the total U.S.  supply  in  1973.  Therefore, although approximately $5 million
  was  spent  for mercury by  the battery industry in that year, only slightly more
  than $200,000 accrued to  U.S. entrepreneurs and workers.  Battery-associated
  mining employment  was probably in the range of 30 persons.
        2.  Processing.  Most preliminary processing of mercury is accomplished
  within the vertically integrated structure of the battery industry itself.
  Processing, though legally invisible,  may be statistically approximated.  The
  scale of this activity is of significance to this analysis.
           o Carbon-zinc battery manufacture requires the processing of approxi-
              mately 160,000 kg of mercury.   Employment for this purpose may be
              70 persons.
           o Alkaline-manganese dioxide  battery manufacture requires approxi-
              mately 300,000 kg of mercury and provides support for perhaps
              130 workers.
           o  Ruben mercury cell manufacture  requires  approximately 100,000 kg
              of mercury and occupies approximately 40 workers.
       3.   Manufacture.  Mercury contributes  to the production  of almost 85
  million kg  of  carbon-zinc, alkaline-manganese, and Ruben mercury  cells.
  Collectively these  batteries  represent an economic value of nearly  $300 million
  at producers' prices.  Almost 7,000 workers derive their support  from  the manu-
  facture of  these types of  cells.
       4.  Merchandising.  Distribution,  wholesaling,  and retailing activities
  related to  these products  generate almost $400 million in revenues and  support
  for approximately 8,000 workers.
       5.  Consumption.   Once again, it is the consumer who must generate  the
 $400 million product need  that will initiate the  entire transaction sequence.

 Industry Regulation

     Research and  development  activity has yet to produce a satisfactory replace-
ment for mercury in  electrical applications.  As of now,  therefore,  only recycling
is envisioned as a viable regulatory goal.   This analysis,  accordingly, assumes
that regulations will have  the effect of  constricting the  flow of  virgin mercury
                                      284

-------
into battery manufacturing processes and of replacing that flow with recycled
material previously lost to  landfill and incineration.  A possible alternative,
the imposition of regulations  on disposal of mercury-containing solid wastes at
the final disposal site, was rejected as too cumbersome and not amenable to
analysis.
     Relative to a base case of "no regulation," a recycle regulation would
serve to diminish the support  capacity of mining operations allied to the
battery industry and, second,  would encourage the development of new support
capacity in the battery recycling  industry.  In order to judge the potential
socioeconomic effects of this  economic redistribution, the product, income,
and employment accounts of Figure  61 were extrapolated using differential
growth rates.  Carbon-zinc battery manufacture was assumed to increase at a
rate of 5.3 percent per year.   Alkaline-manganese cell manufacture was assumed
to grow at a rate of 12.8 percent  per year.  Ruben mercury battery production
was assumed to increase at  a rate  of 7 percent per year.  Table  47 shows the
product, income, and employment projections  that result.
     As indicated, recycling regulations would threaten  the  jobs of perhaps
70 U.S. mercury mine workers in 1985.  Beyond that,  recycling will  serve to  re-
duce our dependence on mercury imports  and thereby improve  our  balance of
trade.  Such regulations will also foster  the growth of improved scrap recovery,
reprocessing, and reuse technology.   A result of this process will undoubtedly
be higher mercury materials cost to the battery industry, at least in the near
term.  The ultimate implication of such higher costs, naturally, is higher
consumer prices.  Although no cost or price impact can be estimated at this
stage, some interval approximations may be advanced.*
     A.  10 Percent Recovery;  10 Percent Higher Input Cost
         o  Carbon-zinc.  Recovery of 16,000 kg in 1973 to 27,000 kg
            in 1983  (at $9.70/kg).  The total consumer price impact
            might approach  $60,000 in 1973 and $100,000 in 1983.
         o  Alkaline-manganese.  Recovery of 30,000 kg in 1973 with
            a price impact  of  $100,000 and 101,000 kg in 1983 with a
            price impact of $333,500.
* For these estimates the values given for  1973  represent a potential for re-
  covery.  Similarly, this analysis assumes that the  technology for recovery
  of the mercury content of carbon-zinc  and alkaline-manganese batteries is
  now available, but such is not the  case.   Modification of current retorting
  practices would probably suffice for the  recovery of mercury from alkaline-
  manganese batteries, but suitable technology would  have to be developed for
  carbon-zinc batteries.
                                      285

-------
     TABLE 47. - Estimated Product and Income Flow for Production and
                     Consumption of Batteries Containing Mercury
                                  1973-1983
A. Product Flow

                                 Carbon-Zinc

Year
1973
1978
1983
Mercury
Supply
(no. of flasks)
4,606
5,927
7,673
Thousands of Kilograms
Process
160.1
207.1
268.2
Alkaline-Manganese
1973
1978
1983
8,805
16,124
29,445
300.0
548.0
1000.5
Mfg.*
78,000.0
100,980.0
130,727.2
Dioxide
6,700.0
12,236.0
22,344.5
Merchandising a
78,000.0
100,980.0
130,727.2
6,700.0
12,236.0
22,344.5
Ruben Mercury
1973
1978
1983
2,349
4,128
5,790
100.0
140.3
196.7
1,900.0
2,665.0
3,737.8
1,900.0
2,665.0
3,737.8
                                  286

-------
     TABLE 47. - Estimated Product and Income Flow for Production and
                     Consumption of Batteries Containing Mercury
                                  1973-1983
                                 (continued)

B. Income Flow (thousands of 1973 dollars)

                                 Carbon-Zinc

Year
1973
1978
1983
U.S. Mine
Production
$ 60.0
77.7
100.6
Process
$ 4,300.0
5,566.8
7,206.8
Mfg.
$187,000.0
242,000.0
313,409.9
Merchandising
$243,000.0
314,600.0
407,300.0
Alkaline-Manganese Dioxide
1973
1978
1983
1973
1978
1983
110.0
200.0
367.0
36.0
50.0
70.8
8,000.0
14,600.0
26,700.0
Ruben
2,700.0
3,787.0
5,311.6
77,000.0
140,600.0
256,800.0
Mercury
29,000.0
40,700.0
57,050.0
100,000.0
182,600.0
333,500.0
38,000.0
53,300.0
74,800.0
                                      287

-------
        TABLE 47. - Estimated Product and Income Flow for Production and
                        Consumption of Batteries Containing Mercury
                                     1973-1983
                                    (continued)

  C. Employment (number of workers)

                                   Carbon-Zinc

Year
1973
1978
1983
1973
1978
1983
U.S. Mine
Production Process
6 70
8 91
10 117
Alkaline-Manganese
16 130
29 237
53 434
Mfg.
4,500
5,800
7,500
Dioxide
1,900
3,470
6,337
Merchandising
5,000
6,500
8,400
2,100
3,935
7,004
Ruben Mercury
1973
1978
1983
5 40
7 56
10 79
700
982
1,377
800
1,122
1,574
a. Total weight of batteries produced

Source:  URS Research Company
                                    288

-------
         o  Ruben mercury.   Recovery of  10,100 kg in 1973 with a net price
            effect of $35,000 and 20,000 kg  in 1983 with a price impact of
            $70,000.
     B.  10 Percent Recovery;  25 Percent  Higher Costs
         o  Carbon-zinc.   The price impact in 1973 would be $150,000 and in
            1983 it would be $250,000.
         o  Alkaline-manganese.   The price impact would be $250,000 in 1973
            and $833,000 in 1983.
         o  Ruben mercury.   The price impact in 1973 would be $87,500; the
            impact in 1983 would be $175,000.
     C.  25 Percent Recovery;  10 Percent  Higher Costs
         o  The net price impacts would be identical to those quantified
            under B.

Cost-Benefit Analysis

     Because of the increased projected usage of mercury-containing batteries
in the future, the base case projected for such batteries was found to result
in an increase of 260 percent in the emissions of  mercury to the environment
over the next ten years.  (Since the introduction of substitute batteries which
could reduce reliance on mercury-containing batteries is unlikely, a base case
which considered less spectacular growth could not be considered.)  Because
this increase masked any real benefits  (that is,  reductions in relative mercury
usage), the base case was normalized, as shown in Figure 62, to a "no growth"
base.  Two alternatives were then plotted:  10 percent recycling with an atten-
dant 10 percent cost increment based on the amount of recycling, and 25 percent
recycling, with an attendant 10 percent cost increment based on the amount of
recycling.  Alternative 1, 10 percent recycling,  reduces the discharge of
mercury to the environment 5 percent over the base case by 1983.  The total
reduction for Alternative 1  for the 10-year period is about 4 percent over the
base case and represents an absolute reduction in the amount of mercury  lost
of about 400,000 kg, with an attached cost to the consumer of  $3.8 million.
                                       289

-------
   32
   28
I  24

°  20
 Q>
 N
   16
o

la
OC
   -4
                                                        Alternative 2
                                                        (25% Recycle)
                                              10-Year Cost to Consumer: $9.4 million
                                                             Alternative 1
                                                             (10% Recycle)
                                                   10-Year Cost to Consumer: $3.8 million
                                                          Normalized Base Case
     1973    1974    1975    1976     1977    1978    1979     1980    1981    1982    1983
                                          Year
      SOURCE: URS Research Company.
            Figure 62   TIME-PHASED REDUCTION IN MERCURY LOSSES
                      FROM BATTERY RECYCLING, AND ASSOCIATED COSTS
                                      290

-------
     Alternative 2,  25 percent recycling, reduces the discharge of mercury to
the environment over the base case  22 percent by 1983.  The total reduction
for the 10-year period for this alternative  is  15 percent over the base case,
representing an absolute reduction  in the amount of mercury lost of more than
1.4 million kg, with an attached cost to the consumer of $9.4 million.
     The cost-effectiveness of these two alternatives was determined in terms
of dollars (consumer costs) per kilogram of  mercury removed (that is, reduced
below that for the base case).  For Alternative 1, the  cost per kilogram is
$9.47 for the 10-year period, and it is $6.55 for Alternative 2.  However, by
1983 the initial recycling costs will have been largely absorbed, so that  the
cost, for 1983 only, is $7.21/kg for Alternative 1 and  $4.06/kg for Alterna-
tive 2.

Risk Assessment

     Recycling, which has been delineated as the only viable  alternative,  does
not appear to have risks associated with its implementation.  As  has  been dis-
cussed, recycling of mercury-containing products either by the battery manu-
facturer or by an independent agent will not involve any appreciable losses to
the environment.
     Of particular concern (although not specifically addressed in the previous
                                                                                i
discussion) is the diversion of mercury-containing solid wastes such as batter- I
ies from final disposal involving incineration.  Some municipal incinerators
have been reported as the source of unusually high discharges of mercury to
air, and since these are usually located in urban areas,  the stack discharge
impinges on a large population.  Control of these emissions from the stack is
a possibility, but if the regulatory actions outlined above were properly
designed, special attention would be paid to the recovery of mercury-containing
solid wastes at recycling centers  (a growing phenomenon), thereby reducing  the
quantities of such wastes that would be incinerated.  Recycling might also
involve a "grass roots" collection  system,  such as now operates  for aluminum
cans and newspapers, with a monetary incentive to enlist a large segment of
the population to participate, possibly through recycling centers  established
                                       291

-------
and maintained by the battery industry  (and other producers of mercury-
containing solid wastes).   No risks are involved with such schemes.
        i
OVERVIEW

      Viewed in a truly macro  perspective and in the long term, regulatory
policy decisions threaten "costs" not so much as they do new equilibrium.
Unless regulatory policies  alter the magnitude of total personal income in the
United States  or have  some  differential effect on foreign competition, the
net effect of  such policies will not be loss or gain but redistribution.
      Thus,  losses to the mercury mining industry will be offset by gains in
other mineral  development areas; losses in mercury processing will accrue
gains to substitute processing activities;  costs to the consumer in terms of
higher prices  (foregone benefits)  will appear  as revenue to producing indus-
tries  and as salaries (realized benefits)  to their  labor forces.
     At such a macro level,  social judgments regarding such policy decisions
may be distilled to a few philosophical  considerations:
     1.  Regulations per  se  distort the  "normal" market  process;  the
         natural allocation  of goods and services is altered;  consumer
         discretion is reduced;  consumer values are imposed rather
         than elected.
    2.  Regulations  "take"  from one sector and "give" to another.
         Often  the total of  revenues or workers affected remain
         unchanged; however,  the fact remains that otherwise employed
        workers  are made unemployed and that previously income-rich
        regions may be made income-short in the process.
    3.  Regulations impose their effects on other activities than
        the targeted activities.  The forced contraction of mercury-
        related industry will inevitably affect support industries
        which have previously provided materials or services and have
        indirectly realized  mercury-related revenue and employment
        support.   Thus, the  magnitude  of economic redistribution which
                                    292

-------
may be authored by regulatory policy is substantially under-
estimated by quantification of direct effects alone.   Indirect
effects may equal or exceed those anticipated within the
regulated economic process.
                              293

-------
REFERENCES

-------
                                  REFERENCES
1.   Anderson, A. et al.,  "System Simulation to Identify Environmental
          Research Needs:  Mercury Contamination," OIKOS 24:231-238, 1973.

2.   D'ltri, F. M., The Environmental Mercury Problem, The Chemical Rub-
          ber Company,  Cleveland, Ohio,  1972.

3.   JernBlov, A. and T. FagerstrBm, "Some Aspects of the Quantitative
          Ecology of Mercury,"  Water Resources, Vol. 6, p. 1193-1202.

4.   Hartung, R. and D. Dinman, eds., Environmental Mercury Contamination,
          Ann Arbor Publishers, Inc., Ann Arbor, Michigan, 1972.

5.   Siegel, S. M. et al., "Icelandic Geothermal Activity and the Mercury
          of the Greenland Icecap," Nature, Vol. 241, p. 526, February 23,
          1973.

6.   Kevorkian, J. et al., "Mercury Content of Human Tissues during the
          Twentieth Century," A.J.P.H.,  April 1972.

7.   Kevorkian, J. et al., "Methylmercury Content of Selected Human Tissues
          over the Past 60 Years," A.J.P.H., Vol. 63, No. 11, 1973.

8.   Weiss, H. V. et al.,  "Evidence of Recent Input by Man in a Greenland
          Ice Sheet," Science.   174: 4010, pgs.  692-694.

9.   Davis, W. E. and Associates,  "National  Inventory of Sources and Emis-
          sions:  Mercury  — 1968," U.S. Environmental Protection Agency.
          APTD-1510.

10.  Anderson, D., "Emission Factors for Trace  Substances," U.S. Environ-
          mental Protection  Agency, EPA-450/2-73-OO/, December  1973.

11.  Wallen, I. E., "Mercury Pat a from Simms  L.  Roy," U.S. Environmental
          Protection Agency, March 22,  1974.

12.  U.S. Environmental Protection Agency,  "Mercury Pollution in the
          United States, by  Source Category," Internal  Document, 1971.

13.  Lambou, V., "Problems of Mercury Emissions into the Environment of
          the United States," EPA Position Paper,  1972.

14.  Cammarota, V. A.,  U.S.  Department of the Interior, Bureau of Mines,
          "The Outlook  for Mercury in the United States," Proceedings of
          the First International Mercury Congress, May 6-10, 1974.


                                      R-l

-------
  15.  Holt, B. R., P. G. Kruzier, S.  L.  Brown,  and L. D.  Spraggs,  "Inter-
            media Transfers of Pollutants,  Illustrated with a Case  Study of
            Mercury," Stanford Research Institute,  prepared for the U.S. En-
            vironmental Protection Agency,  Washington, D.C., 1975.

  16.  JerneltSv, A. and H. Lann,  "Studies in Sweden on Feasibility  of Some
            Methods for Restoration of Mercury-Contaminated Bodies  of Water,"
            Environmental Science and  Technology, 7:8, pp. 712-718.


  17.  Environmental Measurements, Inc.,  "Monitoring Mercury Vapor  near Pol-
            lution Sites," Prepared for the EPA  Water Pollution Control Re-
            search Series, 16080  HTD,  May 1971,  p.  135.

  18.  McCarthy, J. H., Jr.,  W. W. Vaughn,  R. E. Learned, and J. L. Mueschke,
            "Mercury in Soil, Gas, and Air  - A Potential Tool in Mineral Ex-
            ploration," U.S.  Geological Survey Circular 609, Washington, D.C.,
            1969.

  19.  Peyton,  T. 0.,  B.  R. Holt,  and  D.  L. Haynes,  "Transport, Transporta-
            tion, and  Fate of Mercury  in  Freshwaters," Southeast Environmental
            Research Laboratory, prepared for U.S. Environmental Protection
            Agency, Washington, D.C.,  1975.

  20.  Fulkerson, W.,  W.  R. Shults, and R. F. Van Hook,  "Ecology and Analysis
            of  Trace Contaminants," Oak Ridge National Laboratory, Oak Ridge,
            Tennessee,  Report No. ORNL-NSF-EATC-11,  1974.

  21.   Holm, H. W.,  "Mercury in Aquatic Systems:   Methyl at ion,  Oxidation Re-
            duction, and Bioaccumulation," Environmental  Protection Agency,
            August  1974, pg. 35.

  22.   Baughman,  G. L., J. A.  Gordon, N. L.  Wolfe,  and R.  G.  Zepp, "Chemistry
           of Organomercurials in Aquatic Systems," prepared for the U.S.
           Environmental Protection Agency, EPA-660/3-73-012,  September 1973.

 23.  Fitzhugh, 0. et al., "Mercurial  Pesticides, Man and the  Environment,"
           Environmental Protection Agency, p. 118, 1971.

 24.  Kothny,  E., "The Three-Phase Equilibrium of Mercury in Nature," Trace
           Elements in the Environment, p.  17.                              '

 25.  Erickson, "Rock  Sludges May Purify  Mercury-Polluted Lakes," The
      Engineer, 1968.

 26.   Mackenzie,  F.  T.  and R. Wollast,  "Sedimentary Cycling Models  of Global
          Processes," in The Sea, Vol. 6,  E. D. Goldberg, ed.,
          John Wiley  § Sons, New  York  (to be published).

27.  Geraghty, J. J.,  D.  W. Mitler, F. Van der  Leeden, and F- L. Troise,
          Water Atlas of  the United States, a Water Information Center Publi-
          cation, New York, 1973.


                                       R-2

-------
28.  U.S.  Department of the Interior, Geological Survey, The National Atlas
          of the U.S.A., Washington, D.C., 1970.

29.  Shacklette, H. T., "Mercury in the Environment - Surficial Materials
          of the Coterminous United States," USGS Circular 644, January
          1974.

30.  U.S.  Department of Commerce, 1970 Census of Housing, Detailed Housing
          Characteristics, 1972.

31.  Mytelka, A. L., J. s. Czachor, W. B. Guggino, and H. Goludo, "Heavy
          Metals in Wastewater and Treatment Plant Effluents," Water Pol-
          lution Control Federation Journal, Vol. 45, No. 9, September
          1973, pp. 1859-1864.

32.  Oliver, Barry G.  and E. G. Cosgrove, "The Efficiency of Heavy Metal
          Removal  by a Conventional Activated Sludge Treatment Plant,"
          Water Research, Vol. 8, 1974, pp. 869-874.

33.  U.S.  Environmental Protection Agency,  "1968 Inventory of Municipal
          Waste Facilities," A Cooperative  Report, 1971.

34.  Jonasson, I.  R. and R. W. Boyle,  "Geochemistry of  Mercury and Origins
          of Natural Contamination of  the Environment," Canadian Institute
          of Mining and Metallurgy,  1972, pp. 32-39.

35.  Priberg, L. and J. Vostal,  eds.,  Mercury in the Environment, Chemical
          Rubber Company, Cleveland, Ohio,  1972.

36.  Murray, C. R., and E.  B. Reeves,  "Estimated Use of Water in the United
          States," U.S. Geological Survey,  U.S. Department of the Interior,
          1972.

37.  Cammarota, V. A., ed.  "Mercury,"  Mineral Yearbook 1970, Vols.  1-3,
          Metals and Fuels,  U.S.  Dept. of the  Interior, Bureau of Mines,
          U.S. Congress,  House  of Representatives,  Government Operations
          Committee,  Stream Channelization, Part 1 and 2, Y4-7:St8, p. 1-2,
          92nd Congress.

38.  Clark, W. E.  and W.  Fulkerson,  "Mercury Reprocessing Industry," Oak
          Ridge National Laboratory, ORNLNSF-EP.22, October 1972.

39.  Brobst, D. A. and W.  P. Pratt,  "United States Mineral Resources," U.S.
          Geological  Survey Professional Paper 820, United States Govern-
          ment Printing Office, Washington, D.C., December 1973, p.  722.

40.  State Department of Public Health, Bureau of Sanitary Engineering,
          Mercury in California, June 1973, p. 127.
                                    R-3

-------
  41.   U.S.  National Materials Advisory Board,  "Trends in Usage of Mercury,"
            National Academy of Science, NMAB-158, September 1969.

  42.   Cammarato,  V. A.,  "Mercury,"  in  Mineral  Yearbook for 1973, (preprint),
            U.S.  Department of Interior, Bureau of Mines.

  43.   Cammarota,  V. A.,  "Mercury,"  in  Mineral  Yearbook for 1971  (preprint),
            U.S.  Department of Interior, Bureau of Mines.

  44.   National Air Pollution Administration, U.S. Department of Health, Edu-
            cation and Welfare,  "Control Techniques for Sulfur Oxide Air Pol-
            lutants," Washington, D.C., January 1969.

  45.   Singh,  T.  et al.,  "Study  on Recovery of Mercury from Flue Gases of
            Sulfide Ore Roasting and Smelting," Vol. 1, Proc. First Interna-
            tional Congress on Mercury, Barcelona, Spain, 1974.

  46.   Metal Bulletin,  "Quicksilver  from Autokumpo," 1971.

  47.   Engineering/Mining Journal, "Bunker Hill Develops Process to Remove
            Mercury from  Acid,"  1972.

  48.   Pearson, E.  W.,  "A New, Pollution-Free Copper Process via Simultaneous
            Leaching and  Electrowiring," Mining Engineering  ,  December 1974.

  49.   Crump,  L. H.  and C.  L. Readling,  "Fuel and Energy Data:   United States
           by States and Regions,  1972," U.S. Bureau of Mines, Washington,
           D.C.

  50.   Diehl, J. M., "Sales of Fuel Oil  and Kerosene in 1973,"  Mineral In-
           dustry Survey, U.S. Bureau of Mines, September  5, 1974.

 51.  Gorman,  P.  G. et al., "Determination of Concentration of Toxic and
           Hazardous Vapors and Particulates  in Fossil-Fuel Waste Gases,"
           Interim Report 2, Midwest Research Institute, August 1974.

 52.  Billings, C. E. et  al., "Mercury  Balance  on a  Large,  Pulverized Coal-
           Fired  Furnace,"  Journal of Air  Pollution  Control Association.
           23:9,  773-777, 1973^"~

 53.  U.S.  Environmental  Protection  Agency, "Background  Information  on De-
          velopment of National Emission  Standards for Hazardous Air Pol-
           lutants:  Asbestos, Beryllium,  and Mercury," 1973.

54.  Magee, E. M.  et al.,  "Potential Pollutants  in Fossil Fuels,
          EPA-R2-73-249, June 1973.

55.  Handbook of Chemistry and Physics. 41st ed., Chemical Rubber Publish-
          ing Company, Cleveland,  Ohio, 1960.
                                       R-4

-------
56.  Diehl,  R.  C., E. A. Hattman and H. Schultz, "Fate of Trace Mercury
          in the Combustion of Coal," Pittsburgh Energy Research Center,
          for U.S. Bureau of Mines, Technical Progress Report 54, May 1972.

57.  Kelly,  W.  P., "SO  Scrubber Pilot Plant:  Tennessee Valley Authority,
          Colbert Steam Plant," prepared for the Environmental Protection
          Agency, p.  76.

58.  Oak Ridge  National Laboratories, Environmental Science Division,
          "Annual Progress Report - September 1972," February 1973.

59.  U.S. Department  of the Interior, Bureau of Mines, Division of Non-
          ferrous Metals, Washington, D.C., 1975.

60.  Versar, Inc., "Assessment of Industrial Hazardous Waste Practice of
          the Inorganic Chemical Industry," prepared for the Environmental
          Protection  Agency, March 1975.

61.  MacMullin.  Chlorine;  Its Manufacture, Properties, and Uses, Reinhold
          Publishing  Company, New York, 1962.

62.  Stanford Research  Institute, 1975 Directory of Chemical Producers,
          United States of America, Menlo  Park, California, 1975.

63.  U.S. Environmental Protection Agency, Office  of Air Quality Planning
          and Standards,  "Control Techniques for Mercury Emission  from
          Extractions and Chlor-Alkali Plants," No. AP-118,  1973.

64.  Environmental Protection Agency,  "Asbestos and Mercury:   Proposed
          Amendment  to  National Emission  Standards," Federal Register,
          October 25, 1974.

65.  Environmental Protection Agency,  "Water Programs:   Proposed Toxic Pol-
          lutant Effluent Standards,"  Federal Register, Part II, December 27,
          1973.

66.  Faith, W.  L.,  Industrial  Chemicals,  John Wiley & Sons, New York, 1966.

67.  Personal communication,  J. Southerland, Environmental Protection Agency,
          Research Triangle Park,  September 1974.

68.  Litton Systems, Inc., "Industrial Waste Studies:  Mercury-Using  In-
          dustries," Project #805-19000 HIP, prepared for U.S. Environmental
          Protection Agency,  1971.

69.  Versar, Inc., "Assessment of Industrial Hazardous Waste Practices:
          Storage and Primary Batteries Industries," prepared  for Environ-
          mental Protection Agency Office of Solid Waste Management Programs
          under Contract No. 69-01-2276,  January  1975.
                                     R-5

-------
  70.  Goldwater, L. J., Mercury;  A History of Quicksilver, New York Press,
       Baltimore, Maryland, 1972.

  71.  Predicasts, Inc., "Electronic Trends, Batteries and Electric Vehicles,"
            December 1974.

  72.  Fluorescent Lamps Subcouncil Report, The Environmental Compability of
            Fluorescent and Other Mercury-Containing Lamps, National Industrial
            Pollution Control Council, February 1972.

  73.  Predicasts, Inc., "Lighting Devices," Cleveland,  Ohio, February 1974.

  74.  Electric World, "Mercury-Lamp Pollution:   How Much of a Problem?"
            February 15, 1971.

  75.  U.S. Department of Commerce,  "1972 Census of  Manufacturers  -  Electric
            Lamps, SIC 3641,"  January 1974.

  76.  Jenne, E. A.  and W.  Sanders,  "Literature  on Mercury,"  Water Pollution
            Control  Federation,  September 1973.                             ~

  77.  Lutz, G.  A. et  al.,  "Design of  an  Overview System  for  Evaluating the
            Public Health Hazards  of Chemicals in the Environment.   Test Case
            Study, Vol.  1:  The  Overview  System, Vol. II," U.S. Department
            of Commerce,  July  1968.

  78.  Larson, D.  A. and M. C. Unglert, "A New Era for Mercury Lamps," Westing-
            house  Engineer, Vol. 25, No.  4, pp. 116-120.

  79.   U.S.  Congress, Senate Committee on Commerce, "Effects of Mercury on
            Man and the Environment," Y4, C73:91-73,  Part 1, 91st Congress,
            Senate Public Works Committee, Public Works Water Pollution,
            Y4, C96-10:W29-2-970, Parts 1 through 5.

 80.  Foote, R.  S.,  "Mercury Vapor Concentrations inside Buildings," Science,
           Vol.  177, pp. 513-514.                                     	

 81.  Fitzhugh,  0. et al.,  "Mercurial Pesticides, Man and the Environment,"
           Environmental Protecton Agency,  p.  118, 1971.

 82.  Clark, W.  E.,  "Survey of Mercury Usage by  Agencies  of the United States
           Government  during 1971," Oakridge National Laboratory,  January
           1974.

83.  U.S. Department of Agriculture,  "Fanners' Use  of Pesticides  in 1971 "
          1974.

84.   International Atomic Energy Agency, "Mercury Contamination in  Man  and
          His Environment," Vienna, Technical Report 137, p. 181, 1972.
                                    R-6

-------
85.  Selikoff,  I.  J.,  ed., An International Journal of Environmental Medi-
          cine  and the Environmental Sciences - Hazards of Mercury, Vol. 4,
          No. 1, Academic Press, New York, 1971.

86.  U.S.  Department of Public Health, Education and Welfare, "Hazards of
          Mercury," Special Report to Secretary SIR Advisory Committee,
          Chairman Norton Nelson, November 1970.

87.  Boyle, R.  W.  and  I. R. Jonasson, "Geochemistry of Mercury and Origin
          of Natural Contamination of the Environment," Canadian Institute
          of Mining and Metallurgy.  65:717, 1972, pp. 32-39.

88.  Stanford Research Institute, Chemical Information Service, Directory
          of Chemical  Producers, United States of America, 1974.

89.  Sax,  N. I., ed.,  Dangerous Properties of Industrial Materials, 3rd Edi-
          tion, Van Nostrand Reinholdt, 1968.

90.  Bache, A.  et  al., "Mercury and Methylmercury Content of Agricultural
          Crops Grown  on Soil Treated with Various Mercury Compounds,"
          Journal  of Agricultural Food Chemistry, 21:4, pp.  607-613.

91.  Koirtyohann,  S. R. et al., "Mercury Pollution of Golf Course  Lakes,"
          Missouri Water Resources Research  Center, NTIS:PB213  508,
          August  1972.

92.  Peyton, T. (Southwest Research  Institute), Personal  communication,
          June  1975.

93.  Hammerstrom,  R. J. et al., "Mercury in  Drinking  Water Supplies,"
          American Water Works Association Journal, 64:1, p. 60.

94.  Jones, H.  R., Mercury Pollution Control, Heartridge: NOYES,  Data
          Corporation, 1971, p. 245.

95.  Stanford Research Institute,  "Mercury*   Salient  Statistics,  Nonferrous
          Metals," Chemical  Economics Handbook, December 1971.

96.  Bond, R. C.  and C. P. Straub, Handbook of Environmental Control.
          Vol.  IV: Wastewater Treatment  Disposal,  CRC Press. Inc.. 1974.

97.  U.S.  Environmental Protection Agency,  "The Economics of Clean Water,"
          Washington,  D.C.,  1972.

98.  Evans, R.  L.  et  al.,  "Mercury in Public Sewer Systems," Waste and
          Sewage Works,  120:2,  February 1973.

99.  Ghosh, M.  M.  and P.  D.  Zugger,  "Toxic Effects of Mercury  on  the Acti-
          vated Sludge Process," Jour.  Water Pollution Control Federation.
          Vol.  45, No. 3, March 1974.                        """""


                                       R-7

-------
  100.   Bisogni, J.  J.  and A.  Lawrence,  "Kinetics of Mercury Methylation in
            Aerobic and Anaerobic Environments," Jour. Water Pollution Cdntrol
            Federation, Vol.  47,  No.  1,  January 1975.

  101.   Oliver, B.  G.  and E.  G. Cosgrove, "The Efficiency of Heavy Metal Re-
            moval by a Conventional Activated Sludge Treatment Plant," Water
            Research,  Vol. 8,  1974.

  102.   Mytelka, A.  I., T. S.  Czachor, W. B. Guggino, and H. Golub, "Heavy
            Metals  in Wastewater  and  Treatment Plant Effluents," Jour. Water
            Pollution Control  Federation, Vol. 45, No. 9, September 1973.

  103.   Neufeld, R.  N.  and E.  R.  Hermann, "Heavy Metal Removal by Acclimated
            Activated Sludge," Journ. Water Pollution Control Federation
            Vol. 47, No.  2, 1975.

  104.   Lingle, J. W. and E. R. Hermann, "Mercury in Anaerobic Sludge Disposi-
            tion," Jour.  Water Pollution Control Federation, Vol. 47; No. 3,
            1975.

  105.   Environmental Reporter, "Current Developments," Vol. 6, No. 12,
            July 18, 1975.

  106.   U.S. Environmental Protection Agency, "Fate and Effects of Trace Ele-
            ments in Sewage Sludge when Applied to Agricultural Lands,"
            EPA-670/2-74-005,  January 1974.

  107.   Emcon Associates, "Sonoma County Solid Waste Stablization Study,"
           prepared for the Environmental Protection Agency,  1974.

  108.  Environmental Engineering, "Source Test Report for  the  73rd Street
           Municipal Incinerator," New York,  New York,  EPA Test  No.  71-C1-14,
           1971.

 109.  Eldridge, R., "An Overview of  the Land Disposal  Problem"  Waste Age.
           Vol. 6,  No.  1, 1975.                                  	*"

 110.  Minerals Yearbook, "Mercury:   Domestic Production,"  U.S.  Bureau of
           Mines, 1972.

 111.  Weston,  R. P.,  Inc., "Source Testing Report:  Sonoma Mines,  Inc.,"
           prepared  for EPA, July 1972.

 112.   Weston,  R. F., Inc., "Emissions from Wet Process Cement Kiln and
           Finish Mill  Systems at  Ideal Cement Company," prepared for EPA
          March 31, 1974.

113.  Minerals Yearbook, "The Mineral Industry of Arizona," U.S. Bureau of
          Mines, 1971.
                                       R-8

-------
114.  Minerals Yearbook,  "Copper:  Domestic Production," U.S. Bureau of
          Mines,  1972.

115.  Lesemann, R. H.,  "World Copper Use up 7 Percent in 1972, May Be
          Matched in  1973," Engineering and Mining Journal, March 1973.

116.  Minerals Yearbook,  Vol. II,  "Area Reports:  Domestic, U.S.," U.S.
          Bureau  of Mines, 1971.

117.  U.S. Department of  Agriculture,  "Farmers Use of Pesticides in 1971,"
          Report  AER-252.

118.  Friedman, I. and  N. Peterson,  "Fossil Fuels as a Source of Mercury
          Pollution," Science.   172:3986,  1971.

119.  Rook, H. L. et  al., "Mercury in  Coal:  A New Standard  Reference  Mate-
          rial,"  Environmental  Letters, 2(4), 195-204, 1972.

120.  U.S. Environmental  Protection Agency,  "Background  Information on De-
          velopment of  National Emission  Standards for Hazardous Air Pol-
          lutants: Asbestos, Beryllium and  Mercury," NTIS,  PB222-802,
          March 1973.

121.  U.S. Department of  Interior, Geological Survey,  "Mercury in  the  En-
          vironment," Geological Survey Professional Paper 713, 1970.

122.  Musche, R., "Unweltschutz-McBrahmen gaged Quecksilberund Cadmium-
          dontaminanten in der Bundesreouklik Deutschland,"
          Bundesgesundhedsblatt, September 7,  1973.

123.  Environmental Protection Agency, "Emission Factors for Trace Sub-
          stances," EPA,  1973.

124.  Guthrie, V. B., ed., Petroleum Production Handbook, McGraw-Hill Co.,
          New York,  1960.

125.  Oil and Gas Journal, April 1, 1974.

126.  Diehl, "Sales of Asphalt in 1973," U.S. Bureau of Mines, 1974.

127.  "Fluorescent Lamps:  The Environmental Compatibility  of  Fluorescent
          and Other Mercury-Containing Lamps," U.S. Department of Commerce,
          February 1972.

128.  Penn, William, "Preliminary Economic Analysis - Mercury  Inventory,"
          prepared for URS Research Company, August 1974.

129.  Chemical Marketing Newspaper, "Mercury Concentration Linked to  Coal
          Burning," Vol. 199, No.  24, p.  7, June 14,  1971.
                                     R-9

-------
  130.   Kangas,  J.  et al.,  "Smelter Gases Yield Mercury," Chemical Engineering.
            pp.  55-57, 1971.

  131.   Jepsen,  A.  P., "Measurements of Mercury Vapor in the Atmosphere," in
            Trace Elements  in  the Environment, 1973, pp. 81-95.

  132.   Jenne, E. A., "Mercury in the Environment," U.S. Geological Survey
            Professional  Paper 713, 1970.

  133.   Williston,  S. H., "Mercury in the Atmosphere," Journal of Geophysical
            Research, Vol.  73,  No. 22, November 15, 1968.

  134.   Holn, "Studies Concerning the Mercury Concentration of Natural Gases,"
            (text in  German),  Gaswasserfach, Gas/Erdgas.

  135.   Krenkel,  P. A.  et al., Mechanisms of Mercury Transformation in Bottom
            Sediments,  Part I  and II, Environmental Water Resources Engineer-
            ing,  Vanderbilt University, October 1973.

  136.   Johnson,  D. L.  and  R.  S. Braman, "Distribution of Atmospheric Mercury
            Species near Ground," Environmental Science and Technology, p. 7,
            November  1974.

  137.   Schlesinger,  W. H.,  W. A. Reiners,  and D.  S. Knopman, "Heavy Metal Con-
            centrations and Deposition in Bulk Precipitation in Montane Eco-
            systems of New Hampshire," Environmental Pollution (6), pp. 39-47,
            1974.

  138.  Wood, J.  M.,  "Metabolic Cycles for  Toxic Elements in the Environment:
           A Study of Kinetics and Mechanism." Presented at Symposium of
           Heavy Metals in  the Aquatic Environment, Vanderbilt University,
           December 4-7,  1973.

 139.  JerneldV, A. and H.  Lann, "Studies  in Sweden on Feasibility of Some
           Methods  for Restoration of Mercury-Contaminated Bodies of Water,"
           Environmental  Science and  Technology,  7:8,  pp.  712-718.

 140.  Kline, Guide to the  Chemical Industry,  2d  Edition,  1974,  p.  73.

 141.  U.S.  Environmental Protection Agency, "Development Document for
          Effluent Limitation Guidelines and New Source Performance  Stand-
          ards  for the Major  Inorganic Products  Segment of the  Inorganic
          Chemicals  Manufacturing Point Source Category,"  EPA-440/l-74-007a,
          March  1974.

142.  Ross, R.  E.,  "A Case Study of Mercury Losses in a Mercury Cell Plant,"
          presented at the  1973 Canadian Chemical Engineering Conference,
          Vancourver, B.C., September 1973.
                                    R-10

-------
143.  Perry, R., "Mercury Recovery from Contaminated Wastewater and Sludges,"
          Environmental Protection Agency, National Environmental Research
          Center Office of Research and Development, Corvallis, Oregon,
          December 1974.

144.  Energy Policy Project of the Ford Foundation, "A Time to Choose:
          America's Energy Future," Ballinger Publishing Co., Cambridge,
          Massachusetts, 1974.

145.  Class, D. L., "Synthetic Crude Oil from Shale and Coal," Chemtech,
          August 1975, pp. 449-510.

146.  U.S. Environmental Protection Agency,  'Transcript of Proceedings in
          the Matter of Public Hearings to Determine Whether the Registra-
          tion of Pesticide Products Containing Mercury Should be Can-
          celled or Amended," FIFRA Document No. 246, et al., October 17,
          1974, pp.  714-813.
                                     R-ll

-------
APPENDIXES

-------
                                                              APPENDIX A
                                              MERCURY IN THE ENVIRONMENT
     Improper usage of organomercurials or their careless disposal has
led to human suffering and death  in Iraq, the United States, and Japan.
These tragedies have forced the scientific and economic communities to
examine the sources of mercury in the environment; the nature of the physi-
cal, chemical, and biological interactions involving mercury; mercury's
environmental fate; and the effects on the biological community as mer-
cury and its compounds cycle through the environment.  This appendix re-
views recent information in terms of the air, land, and water environ-
ments.

MERCURY AND THE AIR ENVIRONMENT

Sources

     The air environment receives both natural  and man-made inputs of mer-
cury  (see Fig. A-l).  These inputs are shown in Table A-l, with typical
emission and ambient air concentrations.   The production of mercury vapor
during the combustion of fossil fuels (especially coals)  represents a
potentially important contribution because of the anticipated increase in
coal-powered generating facilities.  Several studies have indicated that,
with the present technology for particulate capture, approximately 90 per-
cent of the influent mercury from such facilities is lost to the atmo-
sphere as mercury vapor.  A wide variety of mercury concentrations is
found in coal, depending on the geographical area and position within the
deposit.  The mercury concentration in coal varies from 0.012 to  450 ppm
(Refs. 10, 51, 52, 54, 118, 119, 120, 129).  Eastern coals are relatively
higher than western coals  in mercury content.  The average concentration
in the United States seems to be 0.2 ppm.  In a study completed by the
                                     A-l

-------
      Inputs  (Sources')
                                                  Transformations
 Removal Mechanisms
JO
Industrial Mercury Users	

Volatilization from Zinc,
Lead, Copper and Gold Processing

Fossil-Fuel-Powered Generating
Facilities and Oil Refining

Vaporization from Mining and
Mercuriferous Areas	

Vaporization from Contaminated
Waters	

Geothermal Emissions	

Solid Waste and  Sewage
Sludge  Incineration	

Landfill Emissions	

Volatilization of Agricultural
Insecticides
                                                          Atmosphere
                                                    1.   Diphenylmercury
                                                        breakdown into mercury
                                                        vapor
                                                    2.   Particulate adsorption
                                                    3.   Possible oxidations?
       Precipitation
     Soil Adsorption^
    Plant Adsorption
Water Solubilization
                              FIGURE A-l.  -  Sources, Transformations, and Mechanisms
                                            for Removal  of Mercury  in the Atmosphere
      Source:  URS Research Company

-------
                              TABLE A-l.  - Concentrations of Atmospheric Mercury Emissions
        Source
                                         Concentration
                                                References
trt
Coal-powered electrical
generation

Nonferrous ore  (zinc,
copper, lead) processing

Geothermal emissions

Natural degassing


Garbage incineration

Contaminated water bodies


Background
  Urban

  Rural
15-31 yg/m  in flue gas
0.012-150 ppm

       3        3
10 rag/m -80 ng/m, before control
0.106-0.212  g/m  after

10-28,000 ng/m3

10-1,600 ng/m3   .
2,000-20,000 ng/m

Few to several thousand ng/m

400-4,000 ng/m3
(downwind of chlor-alkali pond)
                                        1-25 ng/m3 winter,
                                        2-50 ng/m  summer
                                        1-10 ng/m3
                                                                              10,  51,  52,  54,  118, 119, 120,
                                                                              129

                                                                                           130
                                                                                            31

                                                                                            34
                                                                                           132

                                                                                           131

                                                                                           131
                                                  133
                                                  133

                                                34, 132
        Source:  Compiled by URS Research Company

-------
  TVA,  a scrubber system designed  for sulfur dioxide removal was found to
  remove approximately 30 percent  of the mercury from the collected fly ash
  and gas stream (Ref. 57).

       The average concentration of mercury in U.S. crude oil is 0.05 ppm.
  Thus  the combustion and refining of crude oil for electrical energy gen-
  eration and gasoline production  may also lead to significant mercury emis-
  sions.  Mercury emission factors for the combustion of imported fuel vary
  from  0.000008 lb/10  gal to 0.001 lb/10  gal, and for U.S. crude oil the
  factors range from 0.00001 to 0.001 lb/10  gal.  Most of the influent mer-
  cury  probably reaches the atmosphere as mercury vapor (Ref. 2).  One study
  that  considered mercury in natural gas (Ref. 134) reported that the forma-
  tion  of insoluble compounds (chiefly sulfides) on pipeline walls reduced
  delivered gas mercury concentrations down to 1 to 2 ppb.  Because of
  increased energy consumption, crude oil contributions will probably in-
  crease .

       Considerable amounts of atmospheric mercury vapor result from the pro-
  duction  of xzinc, lead, copper, and gold (Ref. 34).  Where efforts have been
  made  to  recover ore-concentrated mercury, the outgoing processed gas still
  contained  high levels — 106 to 212 yg  of mercury (Ref. 130).  Thus, the
  applicability of this process to waste gas streams containing 15 to
  31 pg/m  , as found in coal-powered generating plants, is open to question.

     Mercury vapor also enters the atmosphere from mining and land develop-
 ment activities through natural degassing.   Because metallic mercury has
 a very low vapor pressure,  some mercury vapor is  always  being driven off.
 Decreases in.barometric pressure  and increases in temperature increase the
 rate of emission.  Weiss (Ref.  8) has estimated that  mercury released
 through natural degassing is 2-1/2 to 15  times that lost from industrial
activities.  Kothny  (Ref.  24) has estimated natural degassing rates of
        2                                       2
4.0 iig/m /day for Franciscan soils and 0.8 yg/m /day  for normal soils.
Using entirely different assumptions, Weiss developed an average world
                                      A-4

-------
                          M
degassing rate of 1.5 yg/m /day.  McCarthy  (Ref. 18) showed that the de-
gassing rate correlated better with mercury content of underlying mineral
deposits than with the mercury concentration of the soil.  He measured
                                  2
emission rates as low as  0.064 yg/m /day over unmineralized areas to as
                 2
high as 41.9 yg/m /day over  cinnabar veins.

     Although insoluble mercuric sulfide is found  in most natural de-
posits, chemical and biological  processes produce  mercury vapor.  An ex-
ample of a chemical process  is the oxidation of sulfides to sulfates by
ferric ions to free divalent mercuric  ions, followed by the reduction of
the mercuric ion to volatile elemental mercury and mercurous  ion  (Ref. 135).
Bacteria are also able to oxidize mercuric  sulfides, although the methyla-
tion rate of mercuric sulf ide is 1,000 times  slower than for  mercuric
chloride.  Similar emissions could be  expected  from contaminated water
bodies; measurements taken to date verify this  supposition (Ref.  131).

     Anticipated development of geothermal areas for power generation will
increase local atmospheric levels of mercury,  since likely areas for geothermal
energy — that is, zones of deep faulting and shearing — are also zones of
high mercury concentrations  (Refs.  34, 132).   Incinceration of sewage
sludge and municipal refuse will contribute gaseous mercury vapor, as
sewage sludge contains 4 to 8 ppm mercury and flue gas concentrations of
several parts per million have been measured in a study of a New York City
incinerator.  As it appears that all of the mercury in sewage sludge and
most of that in solid waste is volatilized, control devices able to capture
more than particulate matter will be needed.

     Other atmospheric emissions include those from chlor-alkali plants,
agricultural fields  sprayed with organomercurial  insecticides, landfills
containing mercury batteries and lamps, and walls covered with paints  con-
taining organomercurials.
                                      A-5

-------
 Transformations and Fate

      Although ambient concentrations are usually in the range of 1 to
 10 ng/m3, they vary greatly with the degree of mineralization and alti-
 tude, since the highest concentrations are found near the ground surface.
 Possible transformations involving airborne mercury are particulate ad-
 sorption and photodecomposition.

      Means of atmospheric mercury removal include precipitation and plant,
 soil, and water adsorption and solubilization.  Airborne concentrations
 of 20 vg/m  have been reduced to indetectable levels after precipitation
 (Ref. 132).  Particulate adsorption of mercury compounds is frequently
 cited as the reason for the complete mercury removal achieved by rainfall.
 Other workers, however, have shown that considerable amounts of airborne
 mercury remain after precipitation; this finding results from their obser-
 vation that most of the airborne mercury (more than 90 percent)  is volatile
 and not particulate (Ref.  136).

      Because of its volatile nature in air,  once mercury is introduced into
 the atmosphere from a point source it tends  to have a regional or global
 fate rather than a local one.  That is, although the ambient air concentra-
 tions will be highest locally, local deposition is relatively minimal.
 Therefore point sources  generally contribute more mercury to the total
 regional or  global budget  than to the local  mercury budget.   Most of this
 ends up  in the  oceans  or the polar ice cap.

 Environmental Effects

     The effects of airborne toxicants depend upon (1)  the amount and rate
of adsorption and desorption, (2) the  physiochemical and biological prop-
erties of the toxicant, and (3)  individual susceptibility.   No definitive
                                      A-6

-------
human dose-response reactions have been observed, however.  Therefore allow-
able environmental concentrations have been difficult to establish.  The
present EPA standard of 1  yg/m  of mercury vapor for 24-hour exposure is
not based on inhalation studies, but was determined from knowledge of the
total allowable human daily mercury intake and the average mercury consump-
tion from food sources. The concentration of many emission sources clearly
is greater than the EPA allowable standard.  However, because of the dis-
tinct localization of the  sources, most of the population inhales only
background levels, which at an  average of 10 ng/m  would only represent one-
sixtieth of the allowable  daily mercury consumption.

     In summary, the air environment provides a convenient path for mercury
and its compounds to cycle between  land and water, and  also acts as a dis-
persing and diluting medium for anomalous gaseous industrial and natural
sources.  The predominant mercury form may affect localized populations,
but does not appear to affect populations exposed to background levels.

LAND ENVIRONMENT

Sources

     Mercury enters the land environment in a variety of forms and from a
variety of sources.  Mercury sources, transformations, and outputs are
shown in Figure A-2.  Precipitation, deposition from industrial and natural
degassing plumes, spraying and spreading of sewage effluents and sludges,
filling of sanitary landfills, dredging of bottom sediments, and applica-
tion of insecticides all add mercury to the environment.

     Average reported rainfall concentrations are 0.2 ppb, although higher
concentrations may occur in areas of abnormally high ambient mercury con-
centrations.  In New Hampshire, an average value of 0.06 ppb was  reported
(Ref. 137); this is equivalent to a bulk precipitation rate of
         2
0.23 yg/m /day.  Mercury concentrations in landfills have ranged  from
0.05 to 16.3 ppb.  Average mercury concentrations of bottom sediments
                                     A-7

-------
      Input Sources
                                           Trans formations
                                           Removal Mechanisms
it*
00
      Precipitation

      Industrial and Natural
      Anomalous Plumes	

      Sewage Effluent and
      Sludge Disposal
Sanitary Landfill
Practices	

Dredging	

Application of Agri-
cultural Pesticides
                                                Soil
1.   Oxidation-Reduction
2.   Methylation
3.   Demethylation
4.   Adsorption-Desorption
5.   Volatilization
                                                                                          Evaporation
                                                                                         Plant Uptake
                                                                                                   Leaching
                                                                                       Hydraulic Removal of
                                                                                         Suspended Sediment
                             FIGURE A-2,
                                   -  Sources,  Transformations  and Mechanisms  for
                                     Removal of Mercury in the Land Environment
       Source:  URS Research Company

-------
range from 0.073 ppm to 0.3 ppm, depending on the nature of the body of
water and proximity of industrial  activity.  Because organomercurial in-
secticide usage is generally restricted  to croplands and golf courses,
highest concentrations can be expected in such areas.

Transformations and Fate

     Krenkel (Ref. 135) has stated that  the  average ambient soil mercury
concentration is around 60 ppb,  a  value  that depends on the degree of min-
eralization and the location of  pollution sources.  In some areas, soil
concentrations can reach 150 to  200 ppb  (with an extreme in Swedish  soils
of 922 ppb).  The ultimate fate  of this  mercury  will depend on  the physio-
chemical, biological, and hydrological processes acting on the  soil.  These
processes are not well understood, however,  because soil is a complex and
varying aggregation of inorganic gravel, silts,  and clays  and of  organic
humic and plant substances, which  exhibit varying  attractions for mercury
compounds under different conditions. More  understanding  is  being  gained
through modeling and through actual environmental  measurements.

     The physiochemical processes  are adsorption,  desorption, oxidation,
reduction, and disproportionation.  Because  soil does contain complex
constituents, mercury adsorption onto clays, sulfides, and humic water is
promoted under mildly reducing conditions and lessened under others.  Bio-
logical methylation may occur in heavily organic soils, leading to sub-
sequent removal through volatilization under alkaline conditions.  Plant
uptake of mercury does not appear to be significant in terms of concen-
tration or toxicity effects.

     The average mercury concentrations in U.S. foods are quite low, rang-
ing from 6 to 40 ppb, although they could be higher in crops grown  in
highly mineralized, polluted, or insect-infested areas  (Ref. 2).

     Mercury may also be removed from the soil  environment through leach-
ing and erosion of mercury-containing particulates.   Leaching  might be

                                      A-9

-------
 particularly important with regard to landfill operations and would de-
 pend upon the soil composition,  chloride concentration, landfill design
 and operation, and oxidation-reduction potential.  Measurements of leachate
 from landfills indicate  that soil passage effectively attenuates mercury
 through adsorption, precipitations and exchange reactions.  Because soil
 mercury is usually concentrated  in the upper few centimeters of the soil
 column, erosion can transport a  significant amount of mercury contamina-
 tion to receiving waters.

 Environmental Effects

       Although few qualitative or quantitative data have been reported, it
 appears that the land  environment offers a relatively permanent depot for
 mercury;  the processes of metallic mercury volatization, plant uptake,
 leaching,  and removal  from  surface layers by erosion serve to keep only
 a  small portion of the land-applied mercury in the environmental cycle.
 Because mercury is not absorbed  significantly by plants, its effects on
 the terrestial biological community are generally minimal.

 WATER ENVIRONMENT

      Mercury in the hydrosphere  (and particularly in the freshwater envi-
 ronment) seems to offer the greatest threat to the human environment be-
 cause of its continuous cycling within the water environment and the abil-
 ity of aquatic organisms to concentrate mercury in their tissues.
 Figure A-3 shows the various sources, transformations,  and removal mech-
 anisms of mercury in the water environment.

 Sources

      Although chlor-alkali plants have contributed  to significant  localized
mercury  problems,  improved housecleaning practices  and  efficient waste
treatment processes will  continue to  decrease these plant  emissions.   The
contribution of mercury-containing leachate is hard to  quantify, but it has
                                     A-10

-------
 Inputs (Sources)
  Trans format ions
Removal Mechanisms
 Industrial Effluents	
 Leaching from Mining of
 Mercuriferous Belts	
 Precipitation
 Solubilization of Anomalous
 Mercury Vapor Plumes	
 Solubilization of Volatized
 Agricultural Insecticides
 Urban Runoff	
 Rural Runoff	
 Sewage Effluents
 and  Sludges      	
    Hydrosphere
1.  Oxidation-Reduction
2.  Adsorption-Desorption
3.  Methylation-Demethylation
4.  Photodecomposition
5.  Disproportionation
 Biological Uptake
   Sediment Burial
    — Evaporation
         Suspended
     Sediment Loss
                       FIGURE A-3.  -  Sources,  Transformations  and Mechanisms for
                                     Removal of Mercury in the Water Environment
Source:  URS Research Company

-------
  been estimated that such processes amount to  less  than one-tenth the emis-
  sions from coal combustion.   Precipitation may  contain an average mercury
  concentration of 0.2 ppb. No information regarding the solubilization of
  inorganic and organic mercury-containing  plumes from unregulated sources
  and foliar insecticide application was  found, although one worker  (Ref. 135)
  mentioned that high values found in Clear Lake, California, may be due to
  such a process.  The contribution from  urban runoff may be a significant
  localized source, and rural  runoff contains mercury adsorbed onto sus-
  pended sediment.   Sewage effluent and sludge may also be a significant lo-
  calized source, in that higher ocean sediment mercury concentrations occur
  near ocean outfalls that discharge both sewage  effluent and sludge.  Sludge
  disposal is a particular problem,  because  of the ability of organic sludges
  to concentrate mercury and other heavy metals.

  Transformations and Fate

       As  shown in  Table A-2, typical ambient water column concentrations of
  mercury  are  low,  and are well within the 2.0 ppb maximum allowable value
  set by the new EPA Drinking Water Standards.   Although the transformation
  reactions occurring  in water and sediment are similar to those occurring
  in soil, more  information is available on water-sediment reactions, espe-
  cially with regard to me thy liner cury and its compounds.   The production of
 methylmercury  in aquatic systems has been proven,  and the factors deter-
 mining its importance—that is, pH, oxidation-reduction (redox)  potential,
 temperature, total mercury content, and microbiological community—have
 been elucidated (Ref. 138).  Both mercuric ion and methylmercury are toxic
 to some bacterial species, which reduce  both  compounds  to metallic vapor.
 Metallic vapor is  less toxic  than either compound  because of its high vola-
 tility.   This demethylation reaction acts  to keep  the concentration of sedi-
 ment methylmercury relatively constant,  which  shows that  the total mer-
 cury present  and the methylmercury uptake  by fish  are better indicators
of a contaminated  sediment than is  methylmercury concentration,  which is
about 1 percent of the  total mercury value.
                                     A-12

-------
     TABLE A-2. - Typical Ambient Concentrations of Mercury in the
                                 Hydrosphere
(Concentration (ppb)
Character! zat ion
Streams, rivers, lakes

Oceans and seas
Groundwaters
Hotsprings and mineral
waters
Hot springs and mineral
waters at Sonoma, Cali-
fornia
Range
0.01

0.03
0.01
0.01
0.7
- 0.10a

- 5.0
- 0.10
- 2.5
- 0.13
Average
0.03

0.2
0.05
0.10
0.10
Reference
6
7
6
6
6
19
a. Of 273 samples, 261 <1 ppb,  11 = 1 to 5 ppb,  and 1 >5 ppb.
Source:  Compiled by URS Research Company
                                      A-13

-------
       The natural mechanisms for mercury removal are all well documented.
  Although dlmethylmercury has a vapor pressure only a third to a half as
  great as that of metallic mercury, water bodies subjected to turbulent
  water and overlying air could transfer  significant amounts of dimethyl-
  mercury and mercury vapor to the atmosphere.  However, recent airborne
  mercury measurements over water bodies  did not detect dimethylmercury
  (Ref. 136).  Because sediment stabilization leads to decreased aquatic
  organism mercury uptake, mercury contained in buried, undisturbed sedi-
  ments may have reached its ultimate fate.  It is generally agreed that
  the half-life of mercury in such sediments varies from one to three years.
  In shallow, turbulent lakes, the removal of suspended sediment would tend
  to reduce mercury concentrations.  The  fourth removal mechanism, by aquatic
  life forms, has been clearly demonstrated.  In one study, the ambient
  aqueous mercury concentration ranged from 0.05 to 0.13 ppm, while the algal
  mercury concentration ranged from 3,000 to 180,000 ppm.

       The biological  accumulation of methylmercury by fish is the most im-
  portant aspect of the biological uptake  system as far as man is concerned.
  Possible sources  of  the methylmercury may be the lower links of the food
  chain,  methylmercury in the water column, or the innate ability of the
  organism to methylate inorganic  and metallic mercury.  The relative im-
  portance of the first two sources is still unquantified; the third does not
  appear  to be important.  Once in the fish, methylmercury is tightly re-
  tained with a half-life of from one to three years.   No significant effects
 on fish containing normal amounts of methylmercury have been noticed, al-
 though extreme concentrations such as found in Minamata Bay fish led to
 anomalous behavior and metabolism.

      D'ltri (Ref. 2)  lists nine possible decontamination procedures for the
 removal or inactivation of mercury-containing  sediments in aquatic eco-
 systems:  (1)  dredging or pumping, (2) conversion  of  all available mercury
 to mercuric sulfide,  which is biologically unavailable  for methylation;
 (3) evaporation of dimethylmercury into  the atmosphere;  (4) biological re-
moval by bacteria;  (5) burial under sand or other  inorganic material;

                                     A-14

-------
(6) adsorption and burial with  silicates or inert clays; (7) plastic coat-
ing applied to the bottom of mercury-contaminated lakes; (8) amalgamation
with aluminum or other active metals; and  (9) "biological mining" by clams.
All methods are costly and  only one  — dredging — has been evaluated full-
scale (Ref. 139).  The results  of that investigation indicate that high re-
moval efficiencies are possible, although  complete removal of all sedi-
ments must be accomplished  for  an appreciable effect on aqueous concentra-
tions to occur.

Environmental Effects

     Biological methylation and concentration are the  important processes
in the effects of mercury on humans. Different mercury compounds have
different physiological effects, body retention times, and organ concentra-
tions.  Methylmercury and dimethylmercury are  strongly held in brain tissue,
whereas ingested mercuric ion tends to be concentrated in  renal tissue  and
ingested metallic mercury has little effect at all.   Because methylmercury
is 5 to 20 times more toxic than other mercury forms,  its  concentration
in food products, especially fish, is critically important.

     The effect of methylmercury on aquatic organisms and predatory birds
has been well documented (Ref. 135).  The apparent difference between chemi-
cally and biologically synthesized mercury, however, needs further clarifi-
cation.  First, fish given chemically synthesized methylmercury are more
sensitive to stress than those given biologically produced methylmercury;
and second, fish accumulate 40 to 50 percent of chemically synthesized
methylmercury versus 10 to  15 percent of biologically synthesized methyl-
mercury.

     In summary, the hydrosphere may be the ultimate  fate for some  mer-
cury, a way station in the  continuing atmospheric recycle for some, and
the arena for biological methylation and  concentration.   However,  the
importance of methylmercury in sediments  may not be as great as was once
believed.
                                       A-15

-------
                                                              APPENDIX B
                                              SELECTION OF STUDY REGIONS


     Figure 1,  presented in the Introduction to this report, is a nap
showing the selected study regions.  This appendix describes the relevant
characteristics of each region.

CALIFORNIA STUDY REGION

     This study region extends from the San Francisco-Oakland metropoli-
tan area through the Sierras  and  encompasses Reno  (Nevada) and surrounding
counties.  It includes several metropolitan areas, relatively little
heavy industry, considerable  agricultural activity,  and — uniquely —  some  of
the major mercury-producing areas (and the only ones currently active  in
the nation).  Considerable gold and silver are also  found in this  region,
although mining activity has  slowed in recent  years.

ARIZONA STUDY REGION

     All of Arizona, two desert  counties in California, and the Las
Vegas (Nevada)  SMSA were selected for study,  primarily because of the high
ambient mercury air readings found throughout the region.  These high
readings were assumed to be associated with the several large copper smelt-
ers in Arizona.  The region is relatively sparsely settled and has only
two major metropolitan areas; heavy manufacturing is not a serious consid-
eration.

LOUISIANA STUDY REGION

     The southern half of Louisiana was included in this study  region be-
cause it incorporates virtually  all industrial activity in  the  state.   This
                                     B-l

-------
 includes one of the most massive petrochemical complexes in the world as
 well as three large chlor-alkali plants.*  The study region also included
 the Mississippi River,  so that its role as a recipient of industrial and
 municipal wastes could  be studied.   (The Houston area, which is contiguous
 with the Gulf Coast of  Louisiana, was initially considered for inclusion
 in the study region.  However, its industrial composition was similar to
 that in Louisiana and its inclusion would have complicated the data anal-
 ysis, so it was not included.)

 KENTUCKY/TENNESSEE STUDY REGION

      This region has a number of identified emitters, including two chlor-
 alkali plants, but of primary interest is the large number of unregulated
 emitters in the TVA power system.  More than 20 large, coal-fired power
 plants were included, as were four metropolitan areas.

 NEW YORK/NEW JERSEY STUDY REGION

      The industrial complex of the State of New Jersey contains six or
 eight small manufacturers who account for almost 90 percent of the nation's
 mercury compound manufacturing.   In addition,  there are other chemical
 manufacturers  and a number of refineries.   The New York megalopolis, with
 its very high  population density,  has a number of coal-burning power plants
 and incinerates  a large  fraction  of its solid  wastes,  including some sew-
                                                   •
 age sludges.   The study  region also includes some electrical and control
 instrument manufacturers.

     The five study regions  not only have unique  characteristics,  but in
 toto they account for virtually every type  of  mercury  emission of concern.
 In addition, they represent  different socioeconomic  concerns,  climatic
differences, and so on.  Thus  they provide  a very diverse data base, which
* When we speak of chlor-alkali plants, unless otherwise noted, we mean
  plants that use mercury cells.

                                      B-2

-------
is at the sane time all-inclusive  for  comparisons of regional variations
and of regions with the national average.

UTILIZATION OF EXISTING MONITORING EMISSION DATA

     In addition to the EPA map showing elevated mercury concentrations
in the United States, the EPA also furnished  an area ranking report  for
each EPA region which indicates average mercury emissions,  in pounds per
day, for each applicant under the  RAPP permit program.  These computer
printout data were further "massaged"  and, in most  cases, obviously  erro-
neous emissions were deleted; in some  cases,  locations  which were not  on
the printouts were ascertained. However, attempts  on the part  of URS  to
verify reported emissions proved almost totally fruitless.  In  the Cali-
fornia study region, for example,  we went through each  of the original
applications to determine the applicant's estimate  of likely mercury emis-
sions and generally found that these were stated  as "background" or
slightly greater.  In other regions obvious  emitters, such  as  chlor-alkali
plants, did not appear.  (It has since been reported that  the  EPA con-
siders that the files on these early permits are  no longer useful.)

     The Hazardous Air Polluting Emissions  Monitoring System (HAPEMS), an-
other computer printout that was obtained,  reported air emissions for mer-
cury mines and chlor-alkali plants.  In general the data on these forms,
which included location and a space for listing quantity of emission, could
be verified.  However, quantitative data usually were not  given, and  thus
this program was of little use to us.   (This program too has now been dis-
continued .)

     Storage and Retrieval of Air Quality Data (SAROAD) were obtained to  de-
termine mercury concentrations  in air.  However, these data comprise  only
two points  (both in Arizona) and were of little value.  STORET printouts
were obtained for each  study region to aid in determining  the  average mer-
cury content of surface waters  and groundwaters.   These data were generally
                                     B-3

-------
useful, although the mercury contents displayed may be deceptively low
because of the long standing time between sample collection and analysis
and the infrequency of sample collection.
                                     B-4

-------
                                                              APPENDIX C

                                       COMPUTER DATA BASE FORMAT AND USE
     During this study a vast amount of information has been collected con-
cerning mercury uses and losses.  Because of the massive volume of this
data, a data base management system was employed.  The system used (called
System 1022) is basically  a device to turn raw data into useful informa-
tion.  The system permits  making changes, updating, reordering, and analyz-
ing the raw data.  The software is owned by Tymshare, Inc., a world-wide
computer time-sharing system, and is therefore proprietary.

FORMAT OF DATA BASE INPUT

     The initial task in constructing any data base is to  establish param-
eters or attributes that will be used in the  final analysis.   In  the  case
of the mercury data base,  12 parameters have  been chosen.   A partial  data
coding form is shown in Table C-l.  The titles above  the actual data  are
the attribute names for each piece of data  and are used  when accessing that
particular piece of information in the  data base. Each  line is a new set
of information and corresponds  to a single  computer  card or record.

     •    Columns 1-6 contain information pertaining to the SIC code.
SIC stands for Standard Industrial Classification and is simply a means of
categorizing the mercury emitters considered (see Section II).  In this
case, SIC 8021 stands for dental uses and SIC 49110 is power plants.

     •    Column 7 contains the sector within which the particular SIC
falls.  In the case of dental applications this  is Sector VIII,  and  for
power plants it is Sector III.
          This allows extraction of mercury  losses by economic sector,
which one of the programs (to be described later) accomplishes.
                                     C-l

-------
 ttJ
 a
 o
                                             TABLE C-l. -  Data  Coding  Form
-p
!
.s
^








1
i










SIC, £
02
1
02 1

91
I

10

i ;
i •




; |



1 i ;
: ' i
i !
>
1 i
• i
;
: 1 •
I I '
i
I
i i
! i
























« NAME
£>£NTA
\D£N
1
\PE\N
1 1

• .
t ,

. !
• .
' !
; i




! i
i i
] t
> i
i
i •
i .
. i


L
LCD LGM
i -
TAL

N$,G RO
—t — : —
i
• ' I
, ! '
|
•

,



i •




' i ' i
, I
4
'


II




:

i i
\ \ i
1} 14 15


v\e




i
i
t


•




i
i
i
!
j
i
i
16 1
T3
73

75
1 !


i ,
l
l i
i

1
1
1


; 1




)
I


f

11

19

To
49
5:6


28


LD LM STATE CONTY EINV
q
0
to


,3
i ' ;
' i • .
: ' ;
i : ;
i
i
;
1




! : !
i
!
t





: . \

! !
i

j


i :
, i .
1
; : i

71
l
i

!
; 1
i !
72 ?3 24 23

9



















-
f

38

*3
i
i

,

i


!
,







'

•
1

77


N
N

N



















; x
y
y
•
y
'
•


:

|

i
l
1
i
I
I
i
1
J

1

1
31 :
Q\iTE
KIN
\
s;
AL
l '
1 !
i r
i
i

\


<

• i
i i


i
1
l
J !
! i
1 !


1
XI
1 i
14 .15
M 17 M ;9
ENS
GS
\
EM
i ' i





i
• '
' ' '

1 i
' ;
i !- i
! ! !
! i
: 1


•i ! '
06 37 M ;























+
PROD FPROD EFA EFW EFL ERE CODE
'

i

! i
i
— , 	 , —




i ' !





; i



• i
i i
1 i
i !
41 Aj *3


31 . 0
08 .0

88 .0
\ '

* '



i t -
1 I , !
l ! i .
i '
*
i
!
,


1
;
i i ,
'. : i
«5 44 47 *8 *» '.
HG

HG

HG



' t i


|

, ;
1 1
' ) 1
i '
t ' .
.
1
i
1
> 1
I ,
i ! 1
t ; ,
I '
i * t

' i
! ! 1
51 5? 53 3
04
\04


90
i j




i
i
I
i

'•
<


1
1
,
' i

j
|
15 M

7
7


<
i





!



1
l
1

;
•

1
i
*9 ac
2
2
'


(
-

1

i
1
i
1

!
1
i
1
1
1

i
j
i
»l 6

,
It
1*0
(

— 1 	
1
1

— f 	
1 •
i
1
> !


1 1
' i
i i
i




1 :
i
63 *4 6!























6
24
24
— •——"'»• —

! : 1
j '
I
1

.

) 1
1
i i
1 '- — '— T"
' \
1 I
i . .
! , ;
! !
!


j





' 1











1



'







i i
1





i
'. i


i
i ;
•
;



i

KO1 A
/C0'2/\
*
F01A



•
•
,
'' t
I
' : '

•

i
• ; '




l

t. lIM «l«
-------
     •    Column 8-17  contain the name or location of the particular
emitter or group of emitters.  In this case, "dental" is simply the name
of SIC category 8021,  and Pennsgrove is the name of the power plant.  When
evaluating a copper smelter, for instance, these columns would contain the
name of the particular smelter.

     •    Columns 18-20 and 21-22 are degrees and minutes, respectively,
for the longitudinal coordinates of the mercury source.  Columns 24-28
are the same for the latitudinal coordinates.  These columns are used
when specific emitters such as power plants, chlor-alkali plants, or mer-
curial chemical manufacturers can be rather precisely located.  In a reg-
ional analysis, many SIC groups fall within an area source category, in
which case the coordinates  are the central point of a county.  This is the
situation in the case  of dental losses, which cannot be defined by  indi-
vidual emitter.  Columns 23 and 29 are not used.

     •    Columns 30-32 are used  for a state abbreviation.  Any abbrevia-
tion up to three characters can be used,  as  long as  that use  is consistent.

     •    Columns 33-39 give the  county in which the mercury emissions occur.
In the case of the national inventory the state names  may go in  these
columns, which allows one of the  FORTRAN programs  to delineate mercury
losses on a state-by-state basis.

     •    Columns 40-44 contain,  if applicable, the amount of mercury a
user may have on inventory.

     •    Columns 45-50 contain the total amount of mercury lost or re-
cycled within a given county, state, or facility.  For example, Queens
County in the State of New York loses 631 pounds of mercury per year
through dental usage, and  the Pennsgrove power plant in Salem County, New
York, emits 188 pounds per year.  The totals in these columns could also repre-
sent the mercury lost during chlorine manufacture by a specific plant,  or
mercury lost in a given county from the disposal of batteries.  In any

                                     C-3

-------
  case,  the amount of mercury that is placed in these columns is that which
  is lost to air,  land,  or water or is  recycled.  The value represents the
  amount of raw mercury  that is  lost, although it may not be emitted in that
  form.

      •    Columns 51-54 are allocated for the form of mercury product
  emitted.  Basically, this can  be divided into three categories:  organic
  (ORG), inorganic (INO), and metallic  (H6).

      •    Columns 55-58,  59-62,  63-66, and 67-70 are the emission factors
  for mercury losses to  air,  water,  land, and recycling, respectively.  Add-
  ing across these four  columns,  the total should equal 1.0; i.e., 100 per-
  cent of the mercury available  for loss is accounted for.  These factors,
  therefore,  can be multiplied by  the value in columns 45-50 to obtain the
  amount of mercury lost  to each parameter for the given SIC and geographical
  area (or mercury-using  facility) .  In the case of the example for dental
  consumption, 4 percent of the mercury lost is vaporized to air, 72 percent
  is  washed down the drain  and therefore goes into water, 0 percent is lost
  to  land,  and 24 percent is recycled.   In the burning of coal,  90 percent
  of  the mercury is lost to air and 10  percent to land.

      •    Columns 71-75 are blank and can be used for adding other param-
  eters,  if necessary.

      •     Columns 76-80 are a numbered and  lettered coding sequence to
 facilitate keeping the  cards in order  when  reviewing the information con-
 tained  on them.

 LOADING THE DATA  BASE FOR USE BY SYSTEM 1022

     The  keypunched data is  loaded on  a disk file and given  the name
URSHG.DMI.  This  data file is now ready to be loaded into System 1022;  this
                                      C-4

-------
is accomplished using the  code  shown below  (assuming the user has logged
into the Tymshare network).  Note:  The user must type the underlined
characters:

     -LOAD URSHG.DMI

     Several lines of information will be printed out, including how much
storage is allocated for each parameter described above.  This loading
sequence is done in conjunction with a file called URSHG.DMD, which is  a
description of how the data appear  on the cards  and under what name each
series of columns (discussed above) is referred  to.  For example,
Columns 45-50, which contain the total amount  of mercury lost to the en-
vironment (plus recycling) have a length of 6  columns  and are referred  to  as
PROD.  A listing of this file  is shown in Table  C-2.

PURPOSE OF DATA BASE MANAGEMENT SYSTEM AND  FEATURES  OF THE  OUTPUT

     Once the data are successfully loaded  into  System 1022, the data base
management system can be utilized to update or change the  data with very
simple commands.  Running the  FORTRAN source programs on the data base is
a simple task.  Assuming the programs are compiled and saved, only one
command is necessary.  The example below will be to run ALLS1C.

     -RUN ALLSIC

     The program will then summarize mercury  losses by SIC code to air,
land, water, and recycling.  The sample output  for this and the remainder
of the FORTRAN source programs  follows this discussion.

     To access System 1022, the user types the  commands shown below:

          RUN  (UPL)  1022
          OPEN. URSHG
                                     C-5

-------
          TABLE C-2. - Listing of File URSHG.DMD
LOADING SECTION
INPUT URSHG.DMI
STRUCTURE SECTION
     ATTR SIC TEXT COL 1 6 KEYED
     ATTR ICODE TEXT LENGTH 1 KEYED
     ATTR NAME TEXT LENGTH 10 KEYED
     ATTR LGD INTEGER LENGTH 3 KEYED
     ATTR LGM INTEGER LENGTH 2 KEYED
     ATTR LGDR TEXT LENGTH I KEYED
     ATTR LD INTEGER LENGTH 3 KEYED
     ATTR LM INTEGER LENGTH 2 KEYED
     ATTR LDR TEXT LENGTH 1 KEYED
     ATTR STATE TEXT LENGTH 3 KEYED
     ATTR CONTY TEXT LENGTH 7 KEYED
     ATTR EFNV REAL LENGTH 5
     ATTR PROD REAL LENGTH 6
     ATTR PPROD TEXT LENGTH 4
     ATTR EFA  REAL LENGTH 4
     ATTR EFW  REAL LENGTH 4
     ATTR EFL  REAL LENGTH 4
     ATTR ERE  REAL LENGTH 4
     FILLER 12
                          C-6

-------
     The system responds with an asterisk, indicating that it is ready to
accept another command.  The  following command must be to open the data
base.  The system responds with another asterisk and now it is ready to
make any changes, additions,  or deletions to the data base.  It should be
noted that these changes will not be made to the original data that were
entered into the computer, but  rather to the loaded data base, which is in
a binary form.

     To make a change in the  binary data base, the user must first lo-
cate the particular record in question.  This is done by a FIND command.
For example, to change the emission factor to air for dental to 6 percent,
the following sequence of  commands  would be used:

     *    FIND SIC 8021
     *    21 records
     *    CHANGE EFA .06
     *
     All records for SIC 8021 would have  the attribute EFA changed to .06.

     Adding data is relatively simple,  as the series of commands below will
show.  To add data, the system will interactively request the values for
each attribute after the user has typed ADD.

     *    ADD
          Supply other attributes
          SIC 8021
          ICODE £
          NAME  DENTAL
          etc.

     To delete, a FIND command  is  again necessary to locate the particular
record  (or records) to be  deleted.  Below, only one record, that  for dental
losses in Queens County, New York,  is deleted.
                                     C-7

-------
     *    FIND SIC 8021 CONTY QUEENS
          POUND
     *    DELETE

     For any other specialized applications, the user should see the Sys-
tem 1022 Reference Manual.  Leaving System 1022, the user simply types
QUIT, and control is returned to the executive.
     *    QUIT
     The programs URSCTY, URSLL, MULFAC, and URSIC request specific in-
puts by the user.  URSCTY requests a county and a sector;  URSLL requests
two points of longitude and two points of latitude; MULFAC requests an
SIC code and a multiplying factor; and URSIC simply asks for an SIC.   The
output presented in the balance of this appendix illustrates the results
of these programs.
                                    C-8

-------
UR8
                         SICALL -  California  Region
SIC
Code
021
1021
1092M
1092P
2,5
2834C
2851 C
2851M
2879
2911 C
2951 C
3.0
3241
3274
36292
36410
36420
38291
4091 C
49110
4924C
4952
7391
8021
Total
Total
Losses to
Land
0.62
0.00
0.07
0.44
0.69
0.00
1.87
0.00
1.35
0.00
0.86
1.95
0.14
0.01
5.32
0.80
27. 02
4.17
0.01
0.00
0.00
0.00
0.80
0.00
46.11
Total
Losses to
Water
0.00
0.00
0.01
0.00
0.14
1.18
0.93
0.02
0.36
0.00
0.10
0.73
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.53
1.36
5.38
Total
Losses to
A1r
0.00
0.01
0.02
8.34
0.05
0.00
12.14
0.02
.0.09
0.32
0.07
0.73
0.02
0.00
0.00
0.00
1.15
0.00
0.08
0.29
1.25
0.00
0.53
0.07
25.19
Total
Losses to
Exports
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Total
Losses to
Recycling
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.98
0.00
0.00
0.00
0.00
1.60
6.84
0.00
0.00
0.00
.0.00
0.00
0.43
9.84
                                        C-9

-------
URS
                          SICALL  -  Arizona  Region
SIC
CODE
021
1021
2.5
2834C
2851 C
2879
29C
3.0
3331
36292
36410
36420
38291
49110
7391
8021
TOTAL
LOSSES TO
LAND
0.15
0.07
0.21
0.00
0.24
0.89
0.80
0.63
2.39
2.29
0.38
8.20
1.54
0.00
0.01
1.05
18.56


TOTAL
LOSSES TO
WATER
0.02
0.00
0.04
0.40
0.00
0.11
0.30
0.24
2.39
0.00
0.00
0.00
0.00
0.00
0.20
0.84
4.53


TOTAL
LOSSES TO
AIR
0.00
0.10
0.01
0.09
4.63
0.00
1.50
0.24
43.07
0.00
0.00
2.11
0.02
0.05
0.04
0.02
51.88


TOTAL
LOSSES TO
EXPORTS
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00


TOTAL
LOSSES TO
RECYCLING
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.32
0.00
0.00
0.00
0.50
3.40
0.00
0.18
0.12
4.51


                                       C-10

-------
UR8
                       SICALL - Louisiana Region
SIC
Code
021
2.5
2821
2834C
2851 C
2879
2911 C
2911M
2951 C
3.0
36292
36410
36420
38291
49110
4924C
49241
7391
8021
Total
Total
Losses to
Land
0.00
0.65
59.98
0.00
0.29
1.34
0.00
0.28
0.44
0.61
1.52
0.38
10.96
0.96
0.01
0.00
0.00
0.26
0.00
77.69
Total
Losses to
Water
0.07
0.12
0.33
0.28
0.00
0.17
0.00
0.00
0.05
0.23
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.19
0.66
2.10
Total
Losses to
A1r
0.01
0.04
6.01
0.07
5.49
0.00
0.29
0.28
0.02
0.23
0.00
0.00
0.47
0.01
0.74
0.25
1.75
0.05
0.03
15.74
Total
Losses to
Exports
0.00
0.00
0.00
0.00
0.00
0.00
0.00
9.76
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
9.76
Total
Losses to
Recycling
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.27
0.00
0.00
0.65
2.27
0.00
0.00
0.00
0.00
0.16
3.35
                                        C-ll

-------
URS
                          SICALL  - Tennessee Region
SIC
Code
021
1099 v
2.5A
2.5B
2812
2834C
2851 C
2851M
2879
291 1C
2911M
2951C
3.0
3331
36292
36410
36420
38291
4091 C
49110
4924C
7391
8021
9199
Total
Total
Losses to
Land
0.39
0.00
0.71
0.10
43.08
0.00
0.91
0.00
3.14
0.00
0.03
2.13
2.02
0.03
9.65
1.26
30.31
6.00
0.28
1.75
0.00
0.37
0.00
0.06
102.21


Total
Losses to
Water
0.00
0.00
0.00
0.38
0.14
1.65
0.00
0.00
0.53
.0.00
0.00
0.24
0.76
0.03
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.63
1.84
0.02
6.23


Total
Losses to
Air
0.00
0.01
0.05
0.00
4.54
0.00
17.32
0.04
0.00
0.57
0.03
0.16
0.76
0.56
0.00
,0.00
1.30
0.00
2.54
15.81
0.93
0.17
0.07
0.00
44.87


Total
Losses to
Exports
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00 ..
0.00
0.00
0.00


Total
Losses to
Recycling
0.00
0.00
0.00
0.00
0.00
0.00'
0.00
0..00
0.00
0.00
0.00
0.00 .
1.01
0.00
0.00
0.00
1.80
14.00
0.00
0.00
0.00
0.50
0.45
0.00
17.76


                                         C-12

-------
   SICALL - New York/New Jersey
SUMMARY  OF  TOTAL LOSSES  BY SIC CODE
      (IN THOUSANDS OF  LBS.)
SIC
CODE
021
2.5A
2.5R
2812
2818M
2819M
283-^M
2834C
2851
2879
291 1C
291 1M
2951C
3.0
3079P
3 12M
36292C
36292M
3641 OC
TOTAL
LOSSES TO
LAND
0.10
1.66
0.27
23.92
0.04
0 . 0 0
0.00
0.55
2.54
6.20
0.01
0.22
3.18
7.68
O.U2
0.06
12.69
2.56
10.04
TOTAL
LOSSES T.O
WATER
0.00
0.00
1.09
0.30
0.64
0.09
0.03
4.68
0.00
4.96
0.00
0 , 0 f )
0.35
2.89
0.40
0.00
0 . 0 (j
0.00
' 0.0*)
TOTAL
LOSSES TO
AIR
0.00
0.39
0.0"=
4.49
0.04
0.00
0 . 0 0
0.28
48.31
1.24
7.36
0.2?
0.23
2.89
0.02
0.31
2.78
0.13
2.24
TOTAL
LOSSES TO
EXPORTS
0.00
0.00
0 . 00
0.00
o . o n
0.00
0.00
0.00
0.00
0.00
0.94
0 . 0 n
I
\
0.00
0.00
o.oc
o . o n
0.0 • '
0.00
0.00
TOTAL
LOSSF < TO
RECYCLING
O.Oii
0.0!
O.f'T
25.02
0.0
0 . 0 i
0 . 0 0
0 . 0 i
O.i'' •
O.Or
0 . n i '
0 . f ' s •
O.Of
3.84
0.0-
0 . f < ' •
0.0
0.0'-
0 . 0 n
                 C-13

-------
URS
                     SICALL - New York/New Jersey  (cont.)
36410M
36420C
36420M
382Q1C
38291N
UOQ1C
'491 ' 0
4024C
7391
HO.'l

0.60
102.71
0.31
27.16
1.2
0.37
0.51
O.f i
0.4'.
O.t'l'
205.07
0.00
0.0('
0.01
0 . 0 r
0 . 0 0
0.00
0.0'
O.OC
1.64
6.0ii
?3.07
0.03
23.72
0.02
o.on
O.Oo
3.^,
9.80
1.76
0.63
0.24
1 0.57
0.00
0.00
0.00
o.on
0.0 (~
o.oo
O.Oc
O.Oi1
o.u^
O.Oi
0.94
o.on
5.94
0.0<
32.4-,
0 . 0 :
0.0!
0.0"
0 . f • •'<
3.59
1.45
72.28
                                      C-14

-------
URS
                          SICALL - National
                SUMMARY OF TOTAL LOSSES BY  SIC  CODE
                      (IN THOUSANDS OF LBS.)
          TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
       7<-
sic
CODE
021
1021
1031
1092M
1092P
2.5A
2.5P
2261
2812
2819M
283-' M
2834C
2851C
2851F
2851M
2879C
2879F
2879M
LOSSES TO
LAND
39.02
0.17
0.03
0.03
0.91
4.89
4.20
16.83
500.17
0.00
0.00
4.61
20.15
0.00
0.10
48.39
35. 3^
0.01
LOSSES TO
WATER
0.00
0.02
0.00
0.00
O.Of
0.00
16.79
0.34
6.46
0.i'5
0.04
39.18
0.00
0.7/
0.45
38.72
6.23
0.13
LOSSES TO
AIR
0.00
0.05
0.01
0.01
17.29
0.26 -
0.00
0.00
32.73
0. n
0.00
2.30
382.80
0.63
0.03
9.68
O.Of
0.01
LOSSES TO
EXPORTS
0.00
0.00
0.00
O.on
O.Of;
0.00
0.00
O.Of
o.oo
O.Of
O.On
0.00
0.00
O.Ot
O.On
O.Ot
0.00
O.Of:
RECYCL]
0 . 0 P
O.OC
O.C ;
0.0 (l
0.0(*
0.0'
O.fM
O.Oi
O.fti
o.o.
0.0'-
O.Of
o.o
o . n (
0.0'
0.0-
0.0"
O.Oi
                                 C-15

-------
SICALL -National  (cent.)
2911C
2911M
2951C
3.0
3079P
3241
3274
3312M
3331
3332
333-
36292C
36292M
36410C
36410M
36420C
3642.0M
38291C
38291M
4091C
491 JO
4911C
4924C
7391
8021

0.04
2.54
33.06
60.74
0.00
\ 3.88
0.64
5.64
4.99
0.58
0.56
102.00
3.47
82.26
2.98
889.27
5.48
237.03
4.34
2.44
0.03
9.97
0.03
3.51
0 . (.1 n
2130.32
0.00
0.00
3.68
22.78
0.01
0.55
0.09
1.13
4.9rj
0.58
0.56
0.00
O.OU
O.OH
0 . 0 0
0.0"
0.12
O.Ofi
0 . 0 U
0 . 0 U
0 . 0 0
o.ou
0.00
13.05
36.72
193.47
37.35
2.54
2.43
22.78
0.00
1.11
0.18
15.79
89.89
10.48
10.11
16.60
0.00
13.39
0.84
153.70
0.29
36.47
o.on
21.98
26.43
89.77
34.09
5.02
2.04
1039.08
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0 . 0'O
o.on
0.00
0 . 0 t.i
0.0 I!
0.00
0.00
0 . 0 f :
O.OU
0.00
0.00
0 . 0 (;
0 . on
o.on
0.00
0 . 0 0
0 . o n
O.on
0.00
0.00
0.00
45.55
0.00
0.00
0.00
0 . 00
o.on
o.on
0.00
o.on
o.on
o.on
0 . 0 0
54.89
/ 0.00
182.33
0.00
O.on
0 . 0.0
o.on
0 . 0.0
28.62.
12.24
323.64
           016

-------
CTYALL - California Region
County
Alameda
Alpine
Amador
Calaveras
Church
Contra Costa
Douglas
Eldorado
Lyon
Marin
Merced
Napa
Placer
Sacramento
Santa Clara
Santa Cruz
San Francisco
San Joaquin
San Mateo
Solano
Sonoma
Stanislaus
Storey
Tuolumne
Washoe
Yolo
Total
Total
Losses to
Land
7.96
0.00
0.08
0.11
0.08
3.66
0.05
0.25
0.06
1.22
0.71
0.60
0.46
3.69
9.99
0.92
5.08
1.97
3.88
0.98
1.42
1.47
0.00
0.15
0.77
0.55
46.11


Total
Losses to
Water
0.96
0.00
0.01
0.01
0.01
0.45
0.00
0.03
0.01
0.15
0.07
0.06
0.05
0.44
1.19
0.11
0.55
0.21
0.48
0.12
0.16
0.15
0.00
0.02
0.10
0.06
5.38


Total
Losses to
Air
2.78
0.00
0.03
0.04
0.03
1.64
0.02
0.11
0.03
1.64
0.25
1.79
0.20
1.56
6.51
0.46
1.77
0.71
1.41
0.45
2.69
0.49
0.00
0.05
0.33
0.20
25.19


Total
Losses to
Exports
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00 -
0.00
0.00
0.00
0.00
0.00
0.00
0.00


Total
Losses to
Recycling
1.90
0.00
0.00
0.00
0.00
0.72
0.00
0.01
0.00
0.06
0.03
0.03
0.03
0.60
2.61
0.20
1.46
0.40
0.95
0.21
0.07
0.30
0.00
0.01
0.20
0.03
9.84


              C-17

-------
URS
                        CTYALL - Arizona Region
                    bUMMAKY OF TOTAL LOSbtb BT COUNTY
                          (IN THOUSANDS OF
COUNTY

ccrn

CLAKK

CO-MIbE



Gil.A

GHE NL
TOTAL         TOTAL
     :"> TO     LOS'.t> TO
              WAI EH
LAND



 2.Ub

 0.32

 0.41
                            0
                            U.cb
TOIAL
LOS ES  10
AIH

 u. ^ i

 1.34

 5»7U "

 U.26

IB. 1
TOIAL
LOS'.-tb TO
EXPOKIS

 U.OJ

 O.Q'J
                                                       0..3 ••
                                                                   TOIAL
                                                                          Tt
                                                        U."

                                                        U. lb

                                                        0.1 (
JMI'Er< fl.M
MAhlCOP H.^.O
'-10'IAVr O.J°
PI MA 3.01
FINAL O.t;6
c./\f IHt •' li.' 
-------
UR8
                          CTYALL  - Tennessee Region
COUNTY
KEN01
KEN02
KEN03
KEN04
KEN05
KEN06
KEN07
KEN08
KEN09
KEN 10
KEN11
KEN12
KEN13
KEN14
KEN15
KEN16
KEN! 7
KEN18
KENT 9
KEN20
KEN21
KEN22
KEN23
KEN24
KEN25
KEN26
TEN01
TEN02
TEN03
TEN04
TEN05
TEN06
TEN07
TEN08
TEN09
TEN10
TOTAL
LOSSES TO
LAND
0.78
27.27
0.61
0.71
0.78
0.96
0.84
0.76
0.87
0.57
0.54
0.62
0.63
0.80
0.79
0.74
6.09
0.89
0.75
0.73
1.27
0.76
0.78
1.50
0.68
0.96
0.89
0.61
0.73
0.75
0.92
0.89
0.82
0.63
3.38
0.94
TOTAL
LOSSES TO
WATER
0.07
0.17
0.07
0.07
0.08
0.08
0.08
0.08
0.08
0.07
0.07
0.07
0.07
0.07
0.10
0.07
0.61
0.07
0.08
0.08
0.15
0.08
0.08
0.10
0.07
0.10
0.09
0.06
0.07
0.08
0.09
0.09
0.08
0.06
0.38
0.09
TOTAL
LOSSES TO
AIR
1.55
3.15
0.43
0.75
0.36
2.02
0.74
0.31
1.05
0.30
0.22
0.92
0.25
2.31
0.42
0.28
4.57
2.04
0.34
0.72
0.70
0.33
0.33
0.34
0.30
0.44
0.36
0.24
0.28
0.28
0.35
1.11
0.30
0.24
1.56
1.05
TOTAL
LOSSES TO
EXPORTS
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
. 0.00
0.00
0.00
0.00
0.00
TOTAL
LOSSES TO
RECYCLING
0.17
0.25
0.14
0.16
0.25
0.23
0.25
0.22
0.22
0.11
0.07
0.07
0.13
0.09
0.18
0.23
2.00
0.21
0.21
0.19
0.32
0.20
0.23
0.78
0.21
0.29
0.31
0.18
0.24
0.24
0.36
0.31
0.31
0.19
0.97
0.29
                                         C-19

-------
CTYALL - Tennessee Region (cont.)


COUNTY
TEN11
TEN! 2
TENT 3
TEN14
TEN15
TEN! 6
TEN! 7
TEN18
TEN! 9
TEN20
TEN21
TEN22
TEN23
TEN24
TEN25
TEN26
TEN27
TEN28

TOTAL
LOSSES TO
LAND
0.74
0.90
0.81
0.80
0.84
0.56
0.93
2.17
0.84
1.02
17.69
0.55
1.23
1.17
0.76
0.54
5.27
2.18
102.88
TOTAL
LOSSES TO
WATER
0.09
0.07
0.09
0.08
0.08
0.06
0.09
0.27
0.08
0.09
0.18
0.06
0.12
0.11
0.08
0.05
0.61
0.23
6.26
TOTAL
LOSSES TO
AIR
0.22
0.38
0.32
0.34
0.30
0.23
1.02
1.00
0.31
1.44
2.67
0.22
0.91
0.47
0.30
0.18
2.71
0.94
45.43
TOTAL
LOSSES TO
EXPORTS
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
TOTAL
LOSSES TO
RECYCLING
0.26
0.31
0.27
0.26
0.27
0.19
0.27
0.59
0.30
0.34
0.39
0.19
0.41
0.45
0.45
0.24
1.39
0.74
17.76
              C-20

-------
UR8
CTYALL - Mew York/New Jersey
                  SUMMARY OF TOTAL LOSSES BY COUNTY
                       (IN THOUSANDS OF LBS.)
           TOTAL
           LOSSES  TO
 TOTAL
 LOSFES  TO
TOTAL
LOSSES TO
TOTAL
LOSSES TO
TOTAL
LOSSES TO
COUNTY
BERGEN
BRONX
BUCKS
BURLING
CAMDEN
DELAWAR
ESSEX
FAIRFIE
Gl 01 iCES
HAPTFOP
HUDSON
HI INTER
KINGS
LITCHFI
MERCEP
MIDDLES
MOMMOUT
MONTGOM
MORF'IS
NASSAU
LAND
8.03
11.83
3.02
2.72
4.05
4.04
8.36
5.08
1.5'.-
5.36
5.67
0.79
14.76
0.98
2.80
5.60
3.72
4.50
3.49
9.74
WATER
1.11
1.18
0.40
0.28
0.41
0.53
0.95
0.75
0.3o
0.76
0.73
0.20
2.11
0.13
0.29
O.bb
0.39
0.58
0.40
1.2
AIR
2.8H
3.59
2.01
0.92
1.38
3.79
2.94
«U3*4
G.b7
3.98
1.95
0.95
11.9'-
O.fc«
1.02
2.53
1.24
3.37
1.25
7.12
EXPORTS
0.07
0.05
O.I;
0.02
0.03
o.oo
0.07
0.01
0.01
0.01
0.04
0.01
0.09
O.n<
0.02
0.04
0.03
O.OG
0.03
0.05
PFCYCLII
1.61
1.81
1.46
0.48
0.76
1.76
1.62
2.78
0.30
2.80
1.17
0.12
3.46
0.5h
0.51
1.21
0.60
2.11
0.70
1.89
                                  C-21

-------
URS





                      CTYALL  -  New York/New Jersey  (cont.)
NEWCAST
NEWHAVE
MEWYORK
OCEAN
PAS^AIC
PHILADE
QUELNS
PICHMON
ROCKLAN
SALEM
SOMERSE
SUFf- OLK
TOLI AND
UNION
WAPT-EN
WESTCHE

3.50
5.31
8.81
O.OP
4.47
16.87
12. 00
0.02
1.74
0.69
1.82
7.7b
0.69
29.06
0.00
6. 23
205.07
0.37
0.70
1.33
0.04
0.76
1.71
1.76
0.00
0.20
0.07
0.23
0.95
0.10
0.91
0.01
0.78
23.07
1.17
3.43
8.94
o.on
1.56
8.89
11.11
0.21
6.91
0.40
0.6i
4.21
0.48
6.37
O.or
3.72
110.57
0.03
0.01
0.05
0.00
0.03
0.00
0.07
0.00
0.01
0.00
0.01
0.04
0.00
0.04
0.00
0.03
0.94
0.64
2.60
2.09
0.00
0.99
5.38
2.91
0.00
0.31
0.14
0.39
1.43
0.33
26.13
0.00
1.24
72.28
                                       C-22

-------
UR8
                      CTYALL - National

                 SUMMARY  OF TOTAL LOSSES BY COUNTY
                      (IN THOUSANDS OF LBS.)
           TOTAL
           LOSSES TO
TOTAL
LOSSES TO
TOTAL
LOSSES TO
TOTAL
LOSSES TO
TOTAL
LOSSES TO
COUNTY
ALABAMA
ARIZONA
ARKAN5A
CALIFOR
COLORAD
COM' ECT
DCOLUMB
DEL.AWAR
FLORIDA
GEORGIA
IDAHO
ILL inoi
INDIAN/*
IOWA
KAMSAS
KtNTUCK
LOUISIA
MAINE
MARYLAN
MASSACH
MICHIGA
Mir ii it so
LAND
90.44
15.17
15.64
143.80
17.10
25.85
0.06
23.32
4b.41
62.11
5.80
104.96
46. 5t>
23.97
24.31
64.94
101. If
16.23
29.2-'
46.15
72.84
30.74
WATER
3.44
3.89
1.58
16.69
2.07
2.84
O.U1
0.77
5.31'
4.24
0.8H
10.06
5.01
2.30
3.12
2.97
3.38
0.9"
3.46
5.11
8.21
3.34
AIR
21.80
49.12
6.71
81.03
7.73
13.33
0.52
4.25
26. bl
21.02
6.13
49.48
26.83
12.46
8.07
23.17
18.47
5.61
16.67
25.06
41.51
15.47
EXPORTS
0.00
0.00
o.on
o.on
o.oc
0.00
0.00
0.00
o.oo
o . o r>
O.On
O.Od
O.Od
O.PO
o.on
0.00
0.00
O.Ol
O.On
O.On
0.00
0.00
RECYCLII
5.43
2.11
2.91
28.8
2.57
6.28
0.00
0.97
7.68
7.63
1.06
20.31
10.57
3.91
2.76
4.67
3.89
1.80
5.82
10.38
17.13
5.53
                                 C-23

-------
UR8
MISSISS
MISSOUR
MONTANA
MEBRASK
NEVADA
NEVYHAMP
MFWJLRS
NEWMEXI
NEWYORK
HOPTHCA
MOPTHDA
OHIO
OKLAHOM
ORE GOM
PENt :SYL
PHODEIS
SOUTHCA
SOUTHDA
TEMI iESS
TEXAS
UTAH
VERMONT
V1RGINI
WASHING
WISCONS
WVIRGIN
WYOMING
•

16.74
38.02
9.19
13.55
4.17
6.48
92.40
6.56
172.97
55.74
14.09
95.87
23.16
15.73
101.92
8.69
28.62
7.58
69.84
131.05
7.99
3.19
35.81
49.45
45.66
68.85
2.60
— • • »»fc». — •«•
1.80
4.45
1.63
1.56
0.71
0.70
7.42
1.12
15.74
4.79
2.14
9.89
2.79
1.79
11.29
0.90
2.53
0.93
5.0'
10.36
1.59
0.34
3.8K
3.31
5.10
1.7-
0.25
••••••i %%wnv« /
11.26
30.39
8.75
5.10
12.44
3.17
32.13
10.73
78.2*
26.97
2. On
55.86
10.07
7.97
76.80
4.02
12.13
2.4*1
22.04
57.08
17.28
1.53
17.94
19.80
18.49
11.81
1.64

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
o.on
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
o.on
0.00
0.00
0.00
0.00

3.12
7.26
0.61
1.72
0.44
1.48
13.63
0.81
28.75
10.25
0.42
20.97
3.04
2.94
22.65
1.98
5.01
0.59
7.02
14.73
1.25
0.65
6.72
4.70
8.11
2.22
0.28
2130.32
193.47
1039.08
0.00
323.64
                         C-24

-------
URS
         URSIC  New York/New Jersey - SIC 36420C:   Battery Consumption
SUMI'AP.Y OF LOSSES

LOSl.Eb
TO AIR
LOS-: EL
To 1 AMD

MO.
LBS.
NO.
LRS.
1 OS' Ki. MO.
TO WATER LHS.
LOS'.FS
TO EXfT
LOS bV
10 RECY
MO.
LHS.
NO.
LBS.
TOTAL
31
23720.71
102712.?.'
0
O.l'ii
0
O.Oi-
30
b937.40
<5
LBS
0
0.00
0
O.Ci,
0
o.ou
0
O.OC'
0
0.00
URSIC New York/New Jersey -
SUMMARY

LOSSES
TO AIR
LOSSES
TO LAND
LOSSES
TO WATER
LOSSES
TO EXPT
LOSSES
TO RECY
5-2 b
LPS
2
30.87
0
l> . 0 . .
0
O.OU
0
O.I-'
3
59.37
SIC 3079
26-100
LBS
101-400
LBS
LhS
6 10 13
396.22 15bO. 8721742. 7b
0
c.ur
0
0 . i. :;
0
O.Oi
8
P: Vat Dyes
3
0
O.Of
0
O.nr
19
37V0.28 It.

30
Li.l
0
fi . ,-. i
•»-'

OF LOSSES

NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
MO.
LBS.
TOTAL
10
22.16
10
22.16
10
396.93
0
0.00
0
0.00
<5
LBS
10
22.16
10
22.16
0
0.00
0
0.00
0
0.00
5-25
LBS
0
0.00
0
0.00
2
39.87
0
0.00
0
0.00
26-100
• LRS
0
0.00
0
0.00
8-
359.06
0
0.00
0
0.00
101-400
LBS
0
0.00
0
0.00
0
o.oo
0
0.00
0
0.00
>400
LBS
0
0.00
0
0.00
0
0.0(
0
0.00
0
0.00
                                       C-25

-------
UR8
     URSIC  New York/New Jersey - SIC 2833 M:  Pharmaceutical Manufacture
 SUMMARY OF  LOSSES


LOSSES NO.
TO AIR LBS.
LOSSES NO.
TO LAND LBS.
LOSSES NO.
TO WATER LBS.
LOSSES NO.
TO EXPT LBS.
LOSSES NO.
TO PECY LBS.
EXIT
URSIC New
SUMMARY OF LOSSES


LOSSES MO.
TO AIR LBS.
LOSSES NO.
TO LAND LBS.
LOSSES NO.
TO WATER LBS.
LOSSES NO.
TO EXPT LBS.
LOSSES MO.
TO RECY LBS.
TOTAL

2
1.58
2
1.58
2
28.51
0
O.Ou
0
O.Oc

York/New Jersey

TOTAL

7
4.74
7
'4.74
7
85.33
0
o.on
0
o.no
<5
LBS
2
1.58
2
1.58
0
0.00
0
0.0 1
0
O.Ou

- SIC 2819

<5
LBS
7
4.74
7
4.74
2
9.48
0
0.00
0
0.0 )
5-25 26-100 101-400
LBS
0
0.00
0
O.OP
2
28.51
0
0. c
0
o.oc.

M: Catalyst

LBS
0
0.00 0
0
0.00 0
0
0.00 0
0
o.on o
0
o.on o

Manufacture

LBS
0
.00
0
.00
0
.00
0
.00
0
.00



5-25 26-100 101-400
LBS
0
o.oo
0
0.00
4
48.37
0
o.on
0
0.00
LBS
0
0.00 0
0
0.00 0
1
27.49 0
0
o.on o
0
0.00 0
LBS
6
.00
0
.00
0
.00
0
.00
0
.00
>400
LBS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00



>400
LBS
0
p. uo
0
0.00
0
0.00
0
0.00
0
0.00
                                    C-26

-------
UR8
           URSIC New York/New Jtrsey - SIC 2818 M:
SUMMARY OF  LOSSES

LOSSES
TO AIR
LOSSES
TO LAND
LOSSES
TO WAThP
LOSSES
TO EXPT
LOSSES
TO RFCY


NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LRS.
MO.
LBS.
URSIC
TOTAL
7
35.63
7
35.63
7
641.26
0
0.00
0
0.0
<5
LBS
4
9.67
4
9.67
0
0.00
0
0.00
0
0.00
New York/New Jersey -
5-2b
LBS
3
25.96
3
25.96
1
14.15
0
0.00
0
o.on
SIC 7391 :
26-100
LBS
0
0.00
0
0.00
3
159.88
0
o.on
0
0.00
Laboratory
101-400
LBS
0
0.00
0
0.00
3
467.24
0
0.00
0
0.00

>400
LBS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00

SUMMARY OF LOSSES

LOSSES
TO AIR
LOSSES
TO LAND
LOS1.ES
TO WATER
LOSbtS
TO EXPT
LOSSES
TO PECY

NO.
LBS.
NO. .
LBS.
MO.
LBS.
NO.
LBS.
NO.
IRS.
TOTAL
3o
629.00
3;^
440l'
LBS
0
0.00
0
0.00
0
o.on
0
0.00
c
0.00
                                        C-27

-------
URS
        URSIC  New York/New Jersey - SIC 364COM:  Battery Manufacture
 SUMMARY OF LOSSES

LOSSES NO.
TO AIR LBS
LOSSES NO.
TO LAND LBS
LOSSES NO.
TO WATER LBS
LOSSES NO.
TO EXPT LBS
LOSSES NO.
TO RECY LBS
URSIC New
TOTAL
1
. 16.64
1
. 309.54
1
. 6 .6* >
0
0.00
0
York/New Jersey -
<5
LBS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
SIC 2834
5-25
LBS
1
16.64
0
0.00
1
6.66
0
o.oo
0
0.00
26-100
LBS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
101-400
LBS
0
0.00
1
309.54
0
0.00
0
0.00
0
0.00
C: Pharmaceutical Consumption
SUMMARY OF LOSSES

TOTAL
LOSSES NO. 3o
TO AIR LBS. 275.25
LOSSES NO.
TO LAND LBS.
LOSFES NO.
TO WATER LBS.
LOSSES NO.
TO EXPT LBS.
LOSSES NO.
TO PECY LBS.
33
5b0.50
3'
4679.25
0
0.00
0
0.00
<5
LBS
12
32.55
6
17.20
0
0.00
0
0.00
0
0.00
5-25
LBS
20
213.80
20
262.60
3
47.60
0
0.00
0
0.00
26-100
LBS
1
28.90
7
270.70
12
782.00
0
0.00
0
0.00
101-400
LBS
0
0.00
0
0.00
17
3358.35
0
V0.00
0
0.00
                                                                          LBS

                                                                            0
                                                                        0.00

                                                                            0
                                                                        O.Ofi

                                                                            0
                                                                        0.00

                                                                            0
                                                                        0.00

                                                                            0
                                                                        0.00
                                                                         >400
                                                                          LBS
                                                                        0.00
                                                                        0.00
                                                                        tr.oo
                                                                        0.00
                                   C-28

-------
URS
             URSIC  New York/New Jersey - SIC 2879   :  Pesticides
 SUMMARY OF  LOSSES
LOSSES
TO AIR
LOSSES
TO LAND
LOSSES
TO WATER
LOSSES
TO EXPT
LOSSES
TO RECY

TOTAL
NO. 3/
LBS. 1240.70
NO. 33
LBS. 6203.50
NO. 33
LRS. 4962.80
NO.
LBS.
MO.
LBS.
URSIC
0
0.00
0
o.r >
<5
LBS
1
4.20
0
0.00
0
0.00
0
0.00
0
0.00
New York/New Jersey
5-25
LBS
12
181.00
1
21. Of
1
16.80
0
0.00
0
O.On
- SIC 021 :
26-100 101-400 >400
LBS LBS LBS
18 2 0
840.70 214.80 0.00
9 20 3
582.50 4040.00 1560.00
12 18 2
724.00 3362.80 859.20
0
0.00
0
0.00
Livestock
0
0.00
0
0.0'"

0
0.00
0
0.00

SUMHARY OF LOSSES

TO AIR
LOSSES
TO LANt)
LOSSES
TO WATER
LOSSES
TO EXPT
LOSSES
TO PECY

NO.
LBS.
NO.
LRS.
NO.
LBS.
NO.
LBS.
MO.
LRS.
TOTAL
0
0.00
25
96.40
0
0.0
0
0.00
0
O.Of
<5
LBS
0
0.00
16
27.20
0
0.00
0
0.00
0
0.00
5-25
LBS
0
0.00
9
69.20
0
0.00
0
0.00
0
0.00
26-100
LBS
0
o.oo
0
0.00
0
0.00
0
0.00
0
0.00
101-400
LRS
0
0.00 •
0
o.on
0
0.00
0
0.00
0
0.00
>400
LBS
0
0.00
0
0.00
0
• o.on
0
0.00
0
0.00
                                        C-29

-------
URS
      URSIC  New York/New Jersey - SIC 49110 :  Utilities - 011  and Gas
SUMMARY OF  LOSSES

LOSSES NO.
TO AIR LBS.
LOSSES NO.
TO LAND LBS.
LOSSES NO.
TO WATER LBS.
LOSsf NO.
TO EXPT 1 BS.
LOS ES MO.
TO PECY LBS.
TOTAL
52
9795.56
52
512. 4U
0
0.00
0
0.00
0
0.00
URSIC New York/New Jersey -
<5
LBS
17
36.57
44
16.64
0
0.00
0
0.00
0
0.00
SIC 4924 C:
5-25
LBS
9
100.04
4
83.10
0
o.on
0
0.01'
0
o.on
Natural Gas
26-100
LBS
9
406.56
2
125.80
0
0.00
0
0.00
0
O.On
101-40C
LBS
10
1898.98 7353
2
286.90 0
0
0.00 0
0
0.00 . 0
0
0.00 0
>400
LBS
7
.42
0
.on
0
.00
0
.on
0
.00
• Residential, Commercial
SUM! APY OF LOSSES

LOS ES NO.
TO AIR LBS.
LOSC.ES NO.
TO LAND LBS.
LOS1 Ec- NO.
TO WATER LBS.
LOS'tS NO.
FO EXPT LBS.
.OS.SES NO.
ro RECY LBS.
TOTAL
33
1758.34
3
1.76
0
0.00
0
0. n
0
0.00
<5
LBS
2
8.89
33
1.76
0
O.on
0
0.00
0
o.on
5-25
LBS
6
77.82
0
O.Of
0
o.or
0
0.00
0
0.00
26-100
LBS
22
1152.75
0
0.0
0
o.cn
0
O.OC;
0
o.on
101-400 >400
LBS LBS
3
518.88 o
0
0.00 0
0
0.00 0
0
0.00 0
0
0.00 0
0
.00
0
.00
0
.00
0
.00
0
.00
                                     C-30

-------
URS
          URSIC  New York/New Jersey - SIC 2951 C:   Tars  and Asphalt
SUMMARY OF LOSSES

LOSSES
TO AIP
LOS :EC
TO LAND

NO.
LBS.
NO.
LBS.
LOSc-t^ NO.
TO WATbP LBS.
LOSSES
TO EXPT
LPSr Ef.
TO PtfY

SI WARY

LOS'.FC
TO AIP
LOSSES;
TO LAUD
LOS' ES
TO WATER
LOS' ES
TO EXPT
LOS' ES
TO PECY
NO.
LBS.
NO.
LBS.
URSIC
OF LOS\

NO.
LBS.
K>.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
TOTAL
3o
23?.. 56
3.^
3179.43
3
35*4.11
0
0.00
0
0.0 '
<5
LBS
9
21.04
0
O.OU
7
17.20
0
o.on
0
O.Ur
New York/New Jersey -
ES
TOTAL
7
216. bO
7
21b.bf>
0
0.0«»
0
0.0 r
0
O.ni-

<5
LBS
0
O.'i'
0
O.Oi
0
O.Of
0
0.00
0
o.oc
b-2b
LBS
24
21?. 52
b
77.06
25
303.84
0
O.Oi.
0
0.0'.
SIC 2911 M:

b-2b
LBS
4
60. bO
4
60. hO
0
0.0'
0
0 . 0 r.
0
O.on
26-100
LBS
0
o.on
lb
1188.61
1
3^.07
.0
0.00
0
0.00
Refineries

26-100
LHS
3
Ib6.00
3
156. Oo
0
0.00
0
0.00
0
0.00
101-400
LBS
0
O.Oi'
13
1913. It
0
0.00
0
0.00
0
O.Oi:


101-400
LBS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
>400
LBS
0
O.Of
0
0 . 0 f •
0
0.0(
0
0.00
0
O.Of


>uor:
LBS
0
O.Of
0
0.00
0
0.00
0
0.00
0
0.00
                                       C-31

-------
URS
              URSIC  New York/New Jersey - SIC 2911 C:  Fuel  011$
 SUMMARY  OF LOSSES

LOSSES
TO AIR
LOSSES
TO L AND

NO.
LBS.
NO.
LBS.
LOSSES NO.
TO WATER LBS.
LOSC.FS
TO FXPT
LOS^-tS
TO RECY

SI WARY

LOSSES
TO AIR
LOSSES
TO LAND
LOSSE^
TO WATER
LOSSES
TO EXPT
LOS' Ff,
TO PECY
NO.
LBS.
NO.
I BS.

TOTAL
33
7360.03
33
7.37
0
0.00
29
936.60
0
O.Or
URSIC New
<5
LBS
0
0.00
33
7.37
0
0.00
3
6.70
0
0.00
York/New Jersey
5-25
LBS
2
46.25
0
0.00
0
0.00
9
108.90
0
0.00
- SIC 4091
26-100
LBS
5
279.52
0
0.0 ;
0
0.00
17
821. 00
0
0.00
C: Coal
101-400
LBS
>400
LBS
21 5
4436.86 2597.40
0
0.00 0
0
0.00 0
0
0.00 0
0
0.00 0

0
.00
0
.00
0
.00
0
.00

OF LOSSES

NO.
LBS.
no.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LRS.
TOTAL
32
3->64.47
32
371;. 83
0
0.00
0
O.Oi
0
O.Oi
<5
LBS
10
25.38
21
15.25
0
0.00
0
0.00
0
O.on
5-25
LBS
11
111.87
6
89.75
0
0.00
0
o.or
0
0.00
26-100
LBS
1
86.13
4
139.94
0
0.00
0
0.00
0
0.00
101-400
LBS
9
1981.08 1160
1
128.89 0
0
0.00 0
0
0.00 0
0
0.00 0
>400
LBS
1
.01
0
.00
0
.00
0
.00
0
.00
                                     C-32

-------
URB
              URSIC  New York/New Jersey - SIC 3312 M:  Coke Overt
SUMMARY OF LOSSES


LOSSES
TO AIP
I OSSEC
TO LAND
LOS' ES
TO WATER
LOSSES
TO EXPT
L(.S EV
TO PECY



NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
URSIC
TOTAL

2
311.64
2
59.36
0
0.00
0
0.«M'
0
0.0.
<5
LBS
0
o.on
0
O.Od
0
0.00
0
0.00
0
0.0;
New York/New Jersey -
5-25
LBS
0
0.00
1
21.28
0
O.OG
0
o.on
0
0.0 '. i
SIC 2.5 B:
26-100
LBS
0
0.00
1
38.08
0
0.00
0
0.00
0
0 . d 0
Caustic
101-400
LBS
2
311.64
0
0.00
0
O.OC
0
0.00
o
0.00

>400
LBS
0
0.00
0
0.00
0
0.00
0
0.00
n
\J
0.00

SUMMARY OF LOSSES


LOSSES
TO AIP
tOSl>Fc
TO LAND
LOS' L
TO WATER
1. OSf E S
TO E.XPT
LOS F1-
TO RLCY


NO.
LRS.
MO.
LBS.
NO.
LBS.
MO.
LHS.
NO.
LhS.
TOTAL

0
o.oo
3...
271.42
3:-
1085.68
0
0.0"
0
O.OC
<5
LBS
0
O.Oi'
1 i
28.0.'
2
7.60
0
O.Od
0
0.00
5-2S
LBS
0
O.I 0
2?
243.4^
13
146.80
0
O.Od
0
0 •'••<'
26-10,
LBS
0
o.ou
0
O.Oi
20
931.28
0
0.00
0
0.00
101-40P
LBS
0
0.0(
0
0.00
0
0.00
0
0.00
0
0.00
>4on
LhS
0
o . : j
0
O.Of
0
o.on
0
0.00
0
o.or
                                     033

-------
URS
                URSIC  New York/New Jersey - SIC 3.0   :  Other
SUMMARY OF  LOSSES

LOSSES
TO AIP
LOS ES
TO LAND

NO.
LBS.
NO.
LBS.
LOSSES NO.
10 WATER LBS.
LOSSES
TO EXPT
I OS^LS
TO PF.CY
FXI1
NO.
LBS.
NO.
LBS.
TOTAL
33
280 ,b3
33
7679. 6<-»
3 •
28M"..b3
0
O.Oi
33
3839.85
<5
LBS
0
0.00
0
0.00
0
0.00
0
O.Oi'
0
0.00
URSIC New York/New Jersey -
SUMi'APY

LOSSES
TO AIR
LOSf £•
TO LAND
LOSSES
TO WATER
LOSSES
TO EXPT
LOSSES
TO RECY
OF LOS

NO.
LBS.
MO.
LBS.
NO.
LBS.
MO.
LBS.
NO.
LBS.
<;ES
TOTAL
1
4^93.28
1
23923. 6b
1
303. bO
0
o.uo
1
25016.64
400
LBS
0
.00
5
.93
0
.00
0
.00
0
.00
: Chlor-alkall
26-1 00
LOS
0
0.00
0
0.00
0
0.00
0
0.00
0
0.00
101-400 >400
LBS LBS
0
0.00 4493.
0
0.0023923.
1
303.60 0.
0
0.00 0.
0
0.0025016.
1
28
. 1
68
0
00
. 0
00
1
64
                                    C-34

-------
 UR8
              URSIC  New York/New Jersey - SIC 2.5  A:  Urethtnt
SUMMARY OF  LOSSES

LOS' ES
TO AIR
LOSSES
TO LAND
LOSSES
TO WATER

NO.
LBS.
NO.
LBS.
NO.
LBS..
TOTAL
32
390.81
33
1657.09
0
o.on
<5
LBS
10
29.29
0
0.0
0
o.nc
5-25
LBS
19
253.00
9
114.84
0
0.00
26-100
LBS
3
108.52
21
1108.18
0
0.0ft
101-400
LBS
0
0.00
3
434.08
0 -
0.00
>4on
LBS
0
0.00
0
o.or.
0
0.00
LOS ES    NO,
                               0
                                                                0
TO EXPT   LBS.
LOS^F'.
TO RECY
          NO.
          LBS.
O.OU

0
O.on
0.0-


0
0.0?
0.00

0
o.oo
o.on

o
0.00
0.00


  0
0.00
                                                                       0.00
                                                                       o.o.n
                URSIC  New York/New Jersey - SIC 8021  :  Dental
 SUMMARY OF  LOSSES  -

LOSSES
TO AIR
LOS( ES
TO LAND
LOSSES
TO WATER
LOSSES
TO EXPT
LOSSES
TO PECY

NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
TOTAL
33
238.49
0
O.OL
3?.
6000.70
0
0.00
33
1454.01
<5
LBS
15
42. ?r
0
o.uu
0
0.00
0
0.00
2
8.73
5-25
LBS
17
171.23
0
O.Oi
3
60.06
0
0.00
9
136.68
2b-100
LBS
1
25.04
0
0.00
8
540.07
0
0.00
19
923.28
101-400 >400
LBS LRS
0
0.00
0
0.00
19
3810.38
0
0.00
3
385.31
0
• 0.00
0
0.00
3
159-0.19
0
0.00
0
0.00
                                     C-35

-------
         URSIC  New York/New Jersey -   SIC 36292M:  Tubes/Switches Manufacture
SUMMARY OF  LOSSES

LOSSES
TO AIR
LOSst S
TO LAND
i. over
TO KvATER
LOS9ES
TO EXPT
LOSSES
TO PECY
URSIC
SUMMARY

LOSr,Fc:
TO AIP
1 OSi.ES
TO LAMD
LOS1 Ff
TO WATER
LOSSES
TO EXPT
1 OSF>Ei
TO PECY

NO.
IBS.
NO.
LPS.
no.
LPS.
NO.
LBS.
NO.
LRS.
'**—
TOTAL
19
134.51
19
251J1-.79
0
o.on
0
o.r'1
0
o . r i;
New York/New Jersey -
OF LOS

NO.
LBS.
no.
LBS.
no.
1 HS.
MO.
LBS.
MO.
LBS.
5ES
TOTAL
21
31. 5b
21
59.^.45
0
0.00
0
o.on
0
0.0!'
<5
LBS
16
33 . 76
0
O.OP
0
0,00
0
0.00
0
O.Oi!
Sie 36410M:
"
101-400
LBS
0
o.on
2
212. 1*
0
0.00
0
O.Ofi
0
O.on
>400
LBS
0
0.00
1
1702.12
0
0.00
0
O.OC;
0
O.on
Lamp Manufacture
5-25
2
11.47
5
b8.97
0
O.Oi
0
0.0(-
0
O.Oc
26~10u
LHS
0
O.Oi
7
2K7.28
0
0.00
0
O.OC
0
0.00
101-^Oh
LBS
0
0.00
2
217.93
0
0.00
0
0.00
0
o.on
>400
LBS
0
0.00
0
O.OU
0
0.00
0
O.OP
0
o.m.
                                         C-36

-------
 UR8
      URSIC  New York/New Jersey -  SIC 38291M:  Control Instrument Manufacture
 SUMMARY  OF LOSSES
 LOSSES    NO.
 TO AIR    LBS.

 LOS^EL    NO.
 TO LAND   LBS.

 LOS^EC>    MO.
 TO WATER  LBS.

 LOSsEf.    NO.
 TO EXPT   LBS.

 LOS-tL    NO.
 TC. RFCY   LBS.
TOTAL
16
64.07
16
1217.36
0
O.ni,
0
O.fi.
0
O.I
<5
LBS
1J
13.50
2
8.41
0
O.CO
0
O.Oi
0
o.o.
5-25
LBS
5
50.57
2
25.24
0
O.Oi'
0
O.l-r
0
O.Ui
26^100
LBS
0
0.0"
7
2r;'.87
0
o . n r
•0
o.on
0
I1-. I'.:
101-400
LBS
0
0.00
5
960.84
0
0.0'
0
0.00
0
C'.0l:
LBS
0
0.00
0
0.0'
0
0.<<;
0
0 . n i
0
O.c;
            URSIC  New York/New Jersey -  SIC 36410C:  Lamp Consumption
SLJMf ARY OF LOS1 E';

LOS'.L1
TO AIR
LOS' ES
TO 1 AMD
LOS- F'
10 WATER
L OSSEC.
TO EXPT
LOS' F:
TO HtCY

MO.
LBS.
NO.
LHS.
i;o.
LBS.
NO.
LBS.
MO.
LBS.
Tt-lAL
32
2.'4-,.73
3o
10( 38.9M
0
0 . 0 ,.
0
O.Or
0
0.0 1<
<5
LHS
0
O.f'f
0
0.0.'
(•
O.f,'-
0
O.I n
0
O.Oi
LBS
8
12L-.H1
I
O.H;
0
0.( .-
0
O.Oi.
u
0 . 0 ;
LBS
15
7H3.26
7
43^ .60
0
O.fM.
G
0.00
0
O.Oi!
I01-40f. >40.
LliS LBb
9
1378.67
17
4179.02
0
0.0;.
0
o.ou
0
0.00
0
O.I.M.
a
5425.33
0
O.r::;
• 0
.0.00
0
O.O.-
                                    C-37

-------
UR8
        URSIC  New York/New Jersey - SIC 36292C:  Tubes/Switches Consumption
   SUMMARY  OF LOSSES

LOSSES
in AIR
LOSSES
70 LAND
1 OS" ET.
70 WA7ER
LOSSES
70 hXP7
LOS' ES
70 PECY

NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
T01AL
31
2784.63
12693.27
0
c .1'.,
0
o . c r
0
0.!",.
<5
LBS
0
0.00
0
0.00
0
0.00
0
o.oc
0
0.00
5-25
LBS
5
75.99
0
O.Oi,
0
o.o.
0
O.Oi
0
26-100
LBS
16
949.16
4
237.23
0
0.00
0
0.00
0
C.r .
101-400 >400
LBS LBS
9
1346.96
14
3406.57
0
O.Oii
0
0.00
0
1
15
9049.47
0
0.00
0
0.00
0!
O.f-i
    URSIC  New York/New Jersey -  SIC 38291C:  Control Instrument Consumption
 SUK'APY OF LOSSES
 LOS'.ES    NO.
 7" AIP    LHS.
     es    MO.
 70 LAND   LBS.

 LOS U,    FlO.
 f" WA7ER  LBS.

 LOS'.EL    MO.
 70 EXPT   LBS.

LOS
27164.67
0
0.0(-
0
O.Oi;
3';-
324" 2. 4 1
40U
LBS
0
0.00
20
2841.3024213.84
0
0.00
0
- o.oo
7
0
0.00
0
. 0.00
24
1812.3930462.80
                                  C-38

-------
   URS
            URSIC  Nnr York/New Jersey -  SIC 2851  C:   Paint consumption
SUMMARY OF LOSSES


LOS'.ES
TO AIR
Lf>^ f- -
TO LAI ID
TOTAI

NO. 3:->
LBS. 48312.25
NO. 3'
LBS. 2b42.7h
<5
LBS
0
0 . f 1 0
0
0.0.':
5-25 :-t>-!Or 101-40T;
LBS LhS LhS
005
O.OC O.Or 1 163. 75471'. J
7186
106. 7h 1 kv.**2 1313.59 1
>MO.
i.hs
2H
i.t'O
C
D.».
I.PS'-Ls    HO.         0
0
                                   0
TO  WATFP LBS.
     F-'    f-0.
Tn FiXf'T  LBS.

LCb'  F '•,    NO.
lc FtC V  LBS.
o.o.

0
0.01

p
0.!
o.o.

0
O.Oi-
p.p..
O.f-

0
0.0'

0
O.fj.
0
o.t.r

(I
0 . U i
0 . (. (i

  0
                                   0
                                o.o.
                                                                           ll.l'
                                              0
                                              II
                                         C-39

-------
 URS
URSCTY  New York/New Jersey, Queens County:  Sector 3 (Unregulated) Sector 5
       (Manufacturing and Processing) and Sector 8 (Final Cons unction)


             FOR COUNTY  =QUEENS  AND BOX  NOS.= 358
                 TOTAL LOSSES TO LAND=        11995.
                 TOTAL LOSSES TO WATER=        1761. 4U
                 TOTAL LOSSES TO AIR=         11107.71
                 TOTAL LOSSES TO EXPORTS=        72.70
                 TOTAL LOSSES TO RECYCLING=    2910.08


   URSCTY  New York/New Jersey, Kings County:  Sector 3 (Unregulated) Sector 5
          (Manufacturing and Processing), and Sector 8 (Final Consumption)
             FOR COUNTY  =KINGS   AND BOX NOS.= 3 b 8

                 TOTAL LOSSES  TO LAND=        1^757.13
                 TOTAL LOSSES  TO WATER=        2110.00
                 TOTAL LOSSES  TO AIR=         11986.77
                 TOTAL LOSSES  TO EXPOPTS=        93.50
                 TOTAL LOSSES  TO RECYCLING=    3456.63
URSCTY  New York/New Jersey, Philadelphia County:  Sector 3 (Unregulated)
       Sector 5 (Manufacturing and Processing) and Sector 8 (Final Consumption)



             -OR COUNTY  =PHILADE  AND BOX NOS.=  358

                TOTAL LOSSES TO LAND=        16872.63
                TOTAL LOSSES TO WATER=       1709.56
                TOTAL LOSSES TO AIR=         8-90.95
                TOTAL LOSSES TO EXPORTS=         0.00
                TOTAL LOSSES TO PECYCLING=   5382.63
                                    C-40

-------
   UR8

   URSCTY  New York/New Jersey, Bergen County:  Sector 3 (Unregulated) Sector 5
          (Manufacturing and Processing) and Sector 8 (Final Consumption)

              FOR COUNTY  =BEPGEN  AND BOX NOS.= 358

                  TOTAL LOSSES TO LAND=        8029.90
                  TOTAL LOSSES TO WATER=        1110.91
                  TOTAL LOSSES TO AIR=         2878.94
                  TOTAL LOSSES TO EXPORTS=       65.10
                  TOTAL LOSSES TO RECYCLINGS   1614.20
       URSCTY New York/New Jersey, Queens County:  Sector 3 (Unregulated)
                FOR COUNTY  =QUEENS   AND BOX NOS.= 3

                    TOTAL LOSSES TO  LAND=         274.72
                    TOTAL LOSSES TO  WATER=         17.03
                    TOTAL LOSSES TO  AIR=         1783.75
                    TOTAL LOSSES TO  EXPORTS=       72.70
                    TOTAL LOSSEf TO  RECYCLING=      0.00
URSCTY  New York/New Jersey, Queens County:  Sector 5  (Manufacturing and Processing)


               FOR COUNTY   =QUEbNS  AND BOX  NOS.= b

                   TOTAL LOSSES TO LAND=          893.23
                   TOTAL LOSr.FS TO WATER=         321.93
                   TOTAL LOSSES TO AIR=           283.90
                   TOTAL LOSSES TO EXPORTS=         O.Ou


                   TOTAL LOSCF_F TO PtCYCLING=     326.72


     URSCTY  New York/New Jersey,  Queens County:   Sector 8 (Final Consumption)
               FOR  COUNTY  =QUEI NS   AND BOX NOS.=  8

                    TOTAL LOSSES TO  LAND=        10827.47
                    TOTAL LOSSES TO  WATER=        1422.47
                    TOTAL LOSSES TO  AIR=          9040.06
                    TOTAL LOSSES TO  EXPORTS::         0.00
                    TOTAL I.OSSEH TO  RECYCLING=    2583.37
                                     C-41

-------
  URS


        URSCTY  Net* York/New Jersey, Kings County:  Sector 3 (Unregulated)



               rOR COUNTY  =KINGS   AND  BOX NOS.= 3

                   TOTAL LOSSES TO LAND=          224.92
                   TOTAL LOS .ES TO WATER=          21.90
                   TOTAL LOSSES TO AIR=          1138.08
                   TOTAL LOSSES TO EXPORTS=        93.50
                   TOTAL LCSCES TO RECYCLlNGr       O.OC


URSCTY  New York/New Jersey, Kings County:  Sector 5 (Manufacturing and Processing)


               FOR COUNTY  =KINGS   AND  BOX NOS.= 5

                  TOTAL LOSSES TO LAND=         987.49
                  TOTAL LOSSES TO WATER=         362.11
                  TOTAL LOS-ES TO AIR=           318.49
                  TOTAL LOS-E^ TO EXPORTS=        0.00
                  TOTAL OS'F'- TO RECYCLING=    361.31

    URSCTY  New York/New Jersey. Kings County:  Sector 8 (Final Consumption)


               FOP COUNTY  =KINGS   AND  BOX NOS.= 8

                  TOTAL LOSSES TO LAND=       13544.72
                  TOTAL LOSSES TO WATEK=        1725.98
                  TOTAL I OS ES TO AIR=         10530.20
                  TOTAL LOSSES TO EXPORTS=        0.0"
                  TOTAL LOS' Ec. TO RECYCLING=   3095.32


       URSCTY  New York/New Jersey, Philadelphia:  Sector 3 (Unregulated)


              FOP COUNTY   =PHILADE AND  BOX NOS.= 3

                  TOTAL ' OSSEE  TO LAND=          685.82
                  TOTAL I OSC.E^ TO WATERr          3^.07
                  TOTAL LOSrEr- TO AIR=          3770.41
                  TOTAL LOS'ES TO EXPORTS=        O.OC
                  TOTAL LOS'F.S TO RECYCLING=      O.OiT
                                      C-42

-------
   UR8
 URSCTY  New York/New Jersey, Philadelphia:  Sector 5 (Manufacturing and Processing)


              FOR  COUNTY  =PHILADE AND BOX  NOS.= 5

                  TOTAL LOSSES TO LAND=         1130.55
                  TOTAL LOSSES TO WATER=         322.36
                  TOTAL LOSSES TO AIR=           284.99
                  TOTAL LOSSES TO EXPORTS^         0.00
                  TOTAL LOSSES TO RECYCLINGS     311.33



      URSCTY  New York/New Jersey, Philadelphia:  Sector 8 (Final Consumption)


               FOR COUNTY   =PHILADE AND BOX NGS.= 8

                   TOTAL LOSSES TO LAND=        15056.26
                   TOTAL LOSSES TO WATER=        1354.13
                   TOTAL ; OSSES TO A1R=          4835.bb
                   TOTAL LOSSES TO EXPORTS=        0.00
                   TOTAL LOSSES TO RECYCLING=   5071.30
          URSCTY  New York/New Jersey, Bergen County:  Sector 3 (Unregulated)

              FOR COUNTY  =BERGEN   AND BOX NOS.= 3

                  TOTAL LOSSES  TO  LAND=         184.19
                  TOTAL LOSSES  TO  WATER=         20.26
                  TOTAL I OSSES  TO  AIR=          397.85
                  TOTAL LOSSES  TO  FXPORTS=       65.10
                  TOTAL LOSSES  TO  RECYCLING=      0.00


URSCTY  New York/New Jersey, Bergen  County:   Sector 5 (Manufacturing and Processing)

              FOR COUNTY   =BEPGEr»  AMD BOX NOS.= 5

                  TOTAL LOSSES TO LAND=         424.62
                  TOTAL LOSSES TO WATER=         44^.37
                  TOTAL LOSSES TO AIR=          141.61
                  TOTAL LOSSES TO EXPORTS^        0.00
                  TOTAL LOSSES TO RECYCHNGr    149.89
                                      C-43

-------
URSCTY  New York/New Jersey, Bergen County:  Sector 8  (Final Consumption)
         FOR COUNTY  =BERGEN  AND BOX NOS.= 8

             TOTAL LOSSES  TO LAND=        7421.09
             TOTAL LOSSES  TO WATER=        646.29
             TOTAL LOSSES  TO AIR=         23^9.47
             TOTAL LOSSES  TO EXPORTS=        0.00
             TOTAL LOSSES  TO RECYCLING=   1464.31
                               C-44

-------
 UR8
URSLL  for New York/New Jersey Region:  Point Sources for Sectors I, II, III, and f
    THt FOLLOWING  LOCATION POINTS:
          41 DEG
          41 DEG
          72 DEG
          73 DEG
 0 MIN
27 MIN
12 MIN
42 MIN
LATITUDE
LATITUDE
LONGITUDE
LONGITUDE
INCLUDE THESE COUNTIES!

 FAIRFIF MEWHAVE
SUMMARY  OF  LOSSES
LOSSES    NO...
TO  AIR    LBS.

LOSSES    NO.
TO  LAND   LBS.

LOSSES    NO.
TO  WATER LBS.

IOSCE!,    NO.
TO  EXP7   LBS.

LOSSES    NO.
TO  RECY   LBS.
TOTAL
41 •
7790.73
47
12396.98
16
1493.03
2
12.60
10
5267.76
<5
LBS
7
16.21
18
10.21
0
0.00
0
0.00
0
0.00
5-25
LBS
16
183.95
4
47.29
2
21.34
2
12.60
0
o.rn
                             26-100    101-400    >400
                                 LBS        LBS     l.HS

                                 675
                               357.06 1159.69 6073.83

                                 8        12         5
                               51.2.91 3231.45 8351.IV

                                 861)
                               503.78  967.91     O.Oi

                                 OOD
                                 0.00    O.Oi;     O.Oii

                                 442
                               2P0.61  635.44 4M11.71
                                   045

-------
URSLL for New York/New Jersey Region:  Point Sources for Sectors I, II, III, and ft
    THF FOLI OWING LOCATION  POINTS!
          40 DEG
          41 DEG
          72 DEG
          73 DEG
42 Mir1
 0 MIN
12 Mirt
42 Min
LATITUDE
LATITUDE
I Ot'GlTUDE.
I OhGITUME
INCLUDE THESh  COUNTIES!

 SUH'OLK NASSAU
 SUMMARY OF LOSSES

LOSSES
TO AIR
LOSSES
TO LAND
LOSSET,
TO WATER
LOSSES
TO EXPT
LOSSES
TO PECY
FXIT'

MO.
LE1S.
NO.
LBS.
MO.
LPS.
NO.
LHS.
MO.
LBS.

TOTAL
16
5l'07.G5
19
7941.86
7
RK7.50
1
43. Oi1
b
1354.91

<5
LBS
3
11.18
4
1.7b
0
O.I i
0
n . n 1 1
(i
O.Oi'

5-2b
LBS
6
37.75
3
34.28
1
10.07
0
0.0''
0
O.On

26-100
LBS
3
196
5
34J
3
206
1
43
2
147


.89

.75

.83

.Of!

.80

101-400
LBS
4
667.80
3
816.19
3
670.61
0
0.00
2
4'l2.60


4593

6745

0

0

764

>4on
LBS
3
.36
4
• 8h
0
.00
0
,0(
1
.51

                                     C-46

-------
URS
   URSLLfor New York/New Jersey Region:  Point Sources for Sectors  I, II,  III, and f


           THE FOLLOWING  LOCATION POINTS:

                 40 DEG   31 MIN  LATITUDE
                 40 DEG   55 MIN  LATITUDE
                 73 DEG   42 MIN  LONGITUDE
                 74 DEG    4 MIN  LONGITUDE


           INCLUDE THESE  COUNTIES:

            HAS:.AU  QUEI MS KINGS   NEWYORK  PlCHMOf! WESTCHl  HUlJbOl i
f-UM-ARY OF  LOSSES


LOS' E'-
TO A IP
LOSLES
TO LAND
IOS5-ES
TO WATER
LOSSES
TO EXPT
LOS^ EG
10 RECY


MO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LBS.
NO.
LRS.
TOTAL

t>7
37463.67
76
501J 0.6<">
31
6245.78
4
272.20
20
9827.2:
<5
LHS
li
13.97
16
19.04
0
O.Od.
0
o.ou
0
O.Oi
5-25
LPS
13
168.16
10
172.32
4
63.76
0
0.00
0
O.f c
26-10r.
LHS
le
bb7.2U
1]
698.32
6
387.52
• 4
272.20
2
173.03
101-40T >4Q
i
LBS LhS
19 1
378M.3/ '29P6.03
13 ^
±1 ^7.91469^3.07
17
3645.39 2151.1''
0
0.00 O.Oi
12
2714.58 -693''.hl
4

6

<+

0

6

                                    C-47

-------
  URS
URSLL  for New York/New Jersey Region:  Point Sources for Sectors I, II, III, and V,
    THE FOLLOWING LOCATION POINTS:
          39 DEG
          40 DEG
          75 DEG
          7ft DEG
55 MIN
10 MIN
 0 MIf!
18 Mil!
LATITUDE
LATITUDE
LONGITUDE
LONGITUDE
INCLUDE THESE COUNTIES:

 PHILADE
 SUMMARY  OF  LOSSES
 LOS'.F'-,    NO.
 10 Mf    LBS.

 L OF' K:,    NO.
 TO LAND   LBS.

 LOSst'.-,    NO.
 TO WATER  LBS.

 LOSSES    NO.
 TO tXPT   LBS.

 LOS'fcf,    MO.
 To P.FTY   I MS.
TOTAL
19
8772.30
21
17251.29
7
1587.25
0
0.00
b
524' .93
<5
LBS
0
0.00
3
0.83
0
0.0 u
0
0.00
0
0 . r i .
5-25
LBS
4
59.26
2
43.32
0
0.00
0
0.00
0
o.ot.
26-10
LB
6
348.
1
62.
2
121.
0
0.
0
o.
                                            LBS      LBS

                                            4          5
                                348.87  698.77 76t5.40

                                            8          7
                                 62.0(. 1406.5415738.60
                                            3          2
                                121.23  568.94  897.08

                                            0          0
                                  0.00    0.00     0.00

                                            3          2
                                        577.75 4667.18
                                     C-48

-------
URSLL  for New York/New Jersey Region:  Point Sources for Sectors I, II,  III, and V.
 THE  FOLLOWING LOCATION POINTS!

        40 DEG  27 MIN   LATITUDE
        41 DEG   0 MIN   LATITUDE
        74 DEG   4 MIN   LONGITUDE
        74 DEG  48 MIN   LONGITUDE
  JI'CLJ'DF. THESr COUf IT Its:

  HS' bX    HUDSON   MORN.IS   PASi.AIC  UNI OH
SOMERSK  HLRGEN  MIDDLES ULLAWAR
       APY  OF LOS! t'.

I OS'.-ti.
TO AIR
LOS F'
TO LAND
TO WATER
LOSSES
TO FiXPT
1 05-: FS
TO RtCY

no.
LBS.
MO.
LBS.
NO.
LBS.
NO.
LBS.
ho.
LBS.
T(. lAL
125
18984.42
143
76362.84
70
70('2.36
7
294.40
36
32308. M8
LBS
23
54.45
31
20.59
1
4.b6
0
O.D-'1
0
O.r-i
5-25
LBS
38
429.46
16
20d.78
13
206.84
1
14.70
3
51.39
26-101'
LBS
3.'.
173".^
32
1814.49
32
17/1 .87
6
279.70
18
1039.18
101-400
LBS
23
3732. 731 ".OP
29
6313. e. 26801
24
C
0.00
8
>40f
LHS
8
7 .M11
3.^7
0 . •'• i
0
o.ou
7
h"..t.5
                                        C-49

-------
  uns
               URSLL for New York/New Jersey Region:  All  Sources
                 for Sectors I, II,.III, V, VIIIA, and VIIIB.
          THF FOLLOWING LOCATION  POINTS!
                41 DEG
                41 DEG
                72 DEG
                73 DEG
 P M1N
27 MIN
12 MlfI
42 MIN
LATITUDE
LATITUDE
LONGITUDE
LONGITUDE
INCLUDE THESE COUNTIES:

 FAIRF1E NEWHAVE
SUMMARY OF LOSSES

LOS1 I L-
TO All.
LOS' F-
TO LAUD
LOSr.t<
TO WATtP
LOSi-.ES,
TO EXPT
Lose. EL
TO RECY

HO.
LHS.
no.
LBS.
no.
L ns .
1 10 .
LPS.
no.
LI'S.
TOTAL
27
182^.12
31
2989 . 0 1
8
M86.37
2
12.60
2
261.38
<5
LBS
6
lb.37
18
10.21
0
0 . 0 i
0
O.Ol!
0
0 . ( M
b-2b
LBS
11
120.6^
3
31.M2
2
21.34
2
12.60
0
O.nr.
26-100
LBS
6
357. P6
4
2Hb.7fj
4
2P1.19
0
O.Pd
0
O.Of
101-400
LBS
3
613.79
5
H 13.91
2
243. fib
0
0 . C 0
2
261.38
>400
LBS
1
717.28
1
Ib47.71
0
0.00
0
0.00
0
O.OP
                                   C-50

-------
              URSLL for New York/New Jersey Region:  All Sources for
                   Sectors I, II, III, V, VIIIA, and VIIIB.
        THE FOLLOWING LOCATION POINTS!
              40 DEG
              41 DEG
              72 DEG
              73 DEG
42 MIN
 0 MIN
12 MIN
42 MIN
LATITUDE
LATITUDE
LONGITUDE
LONGITUDE
INCLUDE THESE COUNTIPS!

 SUFI-OLK NASSAU
SUMMARY OF  LOSSES

LOS'.ES
TO AIR
LOS E'
TO LAND
I OS' EE
TO 1/vATFP
LOfv Ef
TO fcXPT
LOS'.FC
If; RfcCY

NO.
LBS.
NO.
LBS.
MO.
LHS.
NO.
LRS.
MO.
1. BS .
TOTAL
10
2024. 3U
12
637.78
3
163.^2
1
43.00
1
1.V7.60
<5
LBS
3
11.18
4
1.76
0
0.00
0
o.on
0
O.on
b-2b
LBS
2
26.30
3
34.28
1
lO.i 7
t
0
0.0.
0
0.( i'
26-100
LHS
1
71.63
4
286. bl
1
34.80
1
43.0'.-
0
0.0'
101-400
LHS
3
b63.b4 U
1
31b,20
1
llR.bJ.
0
o.uu
1
Ib7.b0
>4on
Lhi.
1
>t;l.t,'..
0
O.Oi
f!
lj..'k
0
o.u
(;
(.'.!:.
                                  C-51

-------
             URSLL for New York/New Jersey Region:  All Sources for
                   Sectors I, II,  III, V,  VIIIA, and VIIIB.
           THE FOLI OWING LOCATION POINTS:

                 40 DEG   31 MIN  LATITUDE
                 40 DEG   55 MIN  LATITUDE
                 73 DEG   42 MIN  LONGITUDE
                 74 DEG    4 MIN  LONGITUDE
           IMCLUUL THEsi  COUNTIES:

            MAS' AU  QUFF'MS   KINGS   NEWYOKK  RICHMOM WERTCHE HUDSOf
   " ARY  Of-  LOSSES
I OS: E'i.    NO.
tO  EXPT   LBS.

I OS',El:    NO.
if  pprv   l hs.
                   TOTAL
l.f'S1 F«    MO.       43
To AIR    LBS.    68J-.P.12

l.cS* F'    110.       48
TO I AMD   LBS.    732:'.65
                   Ib
 K,  WATER LBS.   1474.46
  <5       5-25   .26-101'    101-400    >400
  LBS       LHS       LHS        LBS     LBS

11          8          7         12          b
13.97    120.23    351.66 2477.49 3916.78

16         10          9          7          6
19.04    172.32    526.72 1210.31 5394.27
  O.or    ' 63.76    387.52 1023.18     0.0'
4
272.20
4
j 64. <>3
0
0.00
0
O.O.:
0
O.or "
0
0.0!.
4
272.20
0
O.On
0
O.Ou
4
1164.63
                                            0
                                        0.00
                                            0
                                        O.Oi
                                 C-52'

-------
             URSLL for New York/New Jersey Region:  All  Sources for
                     Sectors I,  II, III, V, VIIIA, and VIIIB.
        THE FOLLOWING LOCATION POINTS:
              39  DEC
              40  DEG
              75  DEG
              7b  DEG
h!  Mil.
10 MIN
 (i Mil;
LATITUDE
LATITUDE
LONGIIUDh
LOIIGITUDL
INCLUDE  THESL COUNT If-

 PH1LADF
SUM  APY OF LOS'-Lb
I OS1 F-1-    NO.
TO AIR    I US.

I OS' **"•    NO.
TO LAND   I. HS.
    E'    NO.
TO WATER LBS.

L f c . L '    NO .
TO FXI'T   LBS.

I OS' f-'    NO.
TO PFCY   LHS.
TOT Al

12
405S.40
13
183^.38
3
35'... 43
0
C.l".
1
311. 3:.

-------
UR8
             UtSLL  for New York/New Jersey Region:  All Sources for
                    Sectors I. II, HI. V. VIIIA, and VIIIB.
  THE FOLLOWING LOCATION  POINTS:

         40  DEG  27 MIl-l  LATITUDE
         41  DEG   0 Mlh  LATITUDE
         74  DEG   4 Mill  LONGITUDE
         74  DEG  48 MIN  LONGITUDE
  INCLUDE  THESIS coum its:

   FS'f-X   HUDSON   MOP' IS   PASr.AK UMION   SOMfcRSE RLRGEN  MIDDLES DELAWAP
 SUM AKY OF  LOS
                    TC! AL
 l.('c.' ES   NO.
 TO  AIR   IJ'-S.    8621.78
                    V.4
LOS'fcL.    1.0.
TO LAND   LHS.

L( S F.S    NO.
Id WATFh  LHS,.

L Pc/ t '-    |,o.
TO FXf'T   LHS.
lOE'.El    NO.       8
TO PECY   U S.  ii
  9  29('.792b016.64
                                  C-54

-------
                                                              APPENDIX D

                                                     GENERAL METHODOLOGY
     The development of a general methodology for the project involved
two tasks.   The first was to determine the quantity of mercury used and
subsequently lost to the environment by a particular user.  The second
task was to determine the distribution of the expended mercury, in what-
ever form,  to any or all of the following categories:  air, land, water,
recycling.

     Data on general use of mercury by some consumers was rather easy to
obtain.  For instance, the Bureau of Mines has reliable figures on mer-
cury used by the dental profession, electrical apparatus industry, and
paint industry.  However, emissions of unregulated sources such as fossil
fuel combustion and incineration are difficult to quantify.

     Another problem arises  in determining how much  of the mercury used
actually enters the environment and at what step in  the process.  For ex-
ample, in paint manufacture  the preservative  used in the paint (phenyl-
mercuric acetate, or PMA)  is manufactured at  one site and the paint is
formulated at another and is applied  at  yet other locations.   At each of
these steps mercury is lost  in one form or another.   In the first step
(PMA manufacture), close scrutiny and regulation of the industry which
manufacturees this chemical  and other mercurials have reduced emissions
to insignificant quantities.  The same is true of the paint formulation
stage.  Once the paint is applied, some of the PMA vaporizes from the
coated surface and enters the air, and this amount must be estimated.
Therefore, Table  7 of the main body of this report has a column which
represents the total mercury available to that process or use which could
potentially be lost to the environment.  The next column, however,  illus-
trates what proportion of this available mercury actually does enter the
environment  (or is recycled).
                                     D-l

-------
      The figures in Table 7 represent national totals; when looking at a
 particular section of the country, a proportion of some sort must be
 applied to estimate the usage within the region of interest.  For this pur-
 pose population proportions, a manufacturing employment ratio, or a multi-
 factor ratio were used.  A manufacturing employment ratio is the total num-
 ber of people employed in manufacturing in the region divided by the na-
 tional total.  The multifactor ratio is similar to the manufacturing em-
 ployment ratio except that it also includes wholesale and retail trade,
 services  (including education), and government employment.  These ratios
 are applied to the national usage to obtain a total for the study region.
 This total is then distributed within the region by the ratio of the coun-
 ty  (or appropriate area) total to this regional total.  Such a methodology
 was used, for example, for many of the consumption categories in
 Sectors VIIIA and VIIIB.  In other cases, specific plants or emitters
 could be located and mercury emission data could be obtained.  Examples
 are chlor-alkali plants and the manufacture of products containing or need-
 ing mercury.

      Once total mercury usage for a particular use within a particular
 geographical area was determined, the second task was to distribute this
 mercury to air, land, water,  or recyling.  This task required the estab-
 lishment of emission factors.   Such factors were determined in a number of
 ways.   The primary source was  published data or information.  In some
 cases  these data proved incorrect and appropriate changes were made where
 necessary.   In other cases, plant visits and oral and written correspon-
 dence  were used.   This method  was used successfully in evaluating the
 chlor-alkali  industry.   Also,  materials balance flow diagrams were con-
 structed  to assist in verifying  the derived emission factors.

     Other  sources of information initially believed to be valuable were
found to be completely useless.   For  example,  a permit application list of po-
tential mercury emitters, by EPA  region,  was found to contain spurious
data, as the list was hurriedly compiled  and consequently the  methodology
                                     D-2

-------
and analysis were very poor.   In  fact, much of the data was recorded in
the wrong units.

     Both man-made and natural losses are covered in the methodology used
for this study, as each is important in determining where, how much, and
in what form mercury is released  to the environment.  Most emitters were
classified by an SIC (Standard Industrial Classification) code as  an aid
in identifying sources.

     Following are descriptions of how each  source of mercury emissions
was considered.  For more detailed explanations  of some  source categories,
the reader is referred to the section on Technology  Assessment.

MERCURY MINING  (SIC 1092M) AND PROCESSING  (SIC 1092P)

     Most active mercury mines are located in California or Nevada; to-
gether, these states account for 88 percent of the present national pro-
duction.  Mines in each region were identified and the total mercury mined
was determined  (Ref. 110).  Losses of mercury during the mining process are
very trivial  (Ref. 12) and result primarily from the dust produced in ex-
tracting the ore.

     Most mercury emissions occur during processing of ore to recover ele-
mental mercury.  The ore must go through a smelting operation in  which mer-
cury becomes volatilized  and is therefore lost  to the air.   The overall
loss of mercury is approximately 9 percent  (Ref. Ill) , but  it may be as
much as 20 percent or  as  little as  5 percent, depending upon recovery meth-
ods employed at the facility.  Emission standards are not presently tied
to the volume of ore processed, but to daily maximum release rates.
                                         D-3

-------
 LIME PRODUCTION  (SIC 3274)

      The mining  and processing of limestone ore cause very little mercury
 to be released to air, water, or land, as the concentrations of mercury
 in the ore is small.

 CEMENT PRODUCTION  (SIC 3241)

      Cement manufacture is another relatively small source of mercury emis-
 sions  (Ref. 112) that derives from the mining and processing of clay lime-
 stone to produce cement.  Emissions accrue primarily to land, as collected
 dust is disposed of in landfills.  Mercury concentrations at the kiln stack
 range from 0.0047 to 0.0082 ppm by volume.

 COPPER SMELTING  (SIC 3331) AND MINING (SIC 1021)

      Copper smelters are a major source of mercury emissions to the en-
 vironment.  Copper mining is an additional, but smaller, source.  In de-
 termining mercury emissions from these sources, a number of paths were
 pursued.   General disagreement among literature sources (Refs. 113-115)
 prompted us to telephone selected smelter operations in order to obtain
 more accurate information on the ores smelted and copper recovered.  Once
 figures were established on the amount of copper produced, an emission fac-
 tor of 0.011 pound of mercury released per ton of concentrate processed was
 applied (Ref.  12).   This increases to 0.055 pound of mercury per ton of
 copper produced,  as there is a concentration factor of 5 in going from the
 concentrated ore  to the  metal (Ref.  113).   The emission factor is then
 multiplied by the amount of copper produced by a given smelting operation
 (Refs.  113,  114)  and the  mercury released is apportioned to air, water,
 and land;  the principal emissions, of course, are to air.

     Very small — almost negligible  — amounts of mercury are lost during
the mining of copper ore.
                                     D-4

-------
ZINC AND LEAD MINING (SIC  1031)

     Because zinc and lead are normally found in the same deposit, the two
minerals are mined together.  Some mercury is lost in minute quantities as
a result of the mining process.  However, this loss is trivial and of little
concern.

LEAD AND ZINC SMELTING (SIC 3322 AND 3333)

     Mercury contained in lead  and  zinc ores is volatilized during the
smelting process.  Although not as  important as copper smelting  as a  source
of mercury, these processes nevertheless  contribute a. significant quantity.
The various plant locations and production rates were found from the
Minerals Yearbook  (Ref. 116).
AGRICULTURAL USES (SIC 2879A) , LIVESTOCK (SIC 021) ,  AND NONFARM
PESTICIDES  (SIC 2879N)
     The agricultural uses of mercury are primarily as seed dressings.  The
nonagricultural sources of mercury emissions include such things as golf
course and commercial turf fungicide applications.  Livestock losses are
from animal manures.

     Mercury lost from animal manures was determined from Census of Agri-
culture data by county, and emissions from agricultural seed dressings were
estimated using data on purchased seed, by type, for farms with cash  sales
of  $2,500 per year and over  (Ref. 117).  These large farms account for a
significant proportion of the total.  A ratio of 0.142 pound of mercury
per thousand tons of feed  (based on 50-percent recycle of manures) was
applied to these annual tonnage figures in order to calculate mercury loss.

     Data on agricultural pesticide use come from the U.S.  Department of
Agriculture and the U.S. Bureau of Mines.  Personal contacts with various
state, EPA, and Department of Agriculture personnel,  as well as with plant
                                      D-5

-------
 pathologists and pesticide distributors, gave us the information needed
 to estimate the distribution of mercurials used for seed treatment.  The
 estimates of distribution of mercurials to water and land were also based
 on information secured from these contracts.

      The remainder of the pesticides are assumed to be used for nonagricul-
 tural purposes.  The multifactor ratio was used to allocate usage within
 a defined region.  As these chemicals are topically applied, emissions are
 to air as well as to land and water.  All mercury lost from livestock
 manures is assumed to go to land.

 COAL-BURNING POWER PLANTS (SIC 49110) AND OTHER (SIC 4091C)

      Electric utilities consume 67 percent of the coal used in the United
 States.  All coal consumed within a study region was assumed to be mined
 in or near that area (Ref. 49).  (For example, coal consumed in California
 was assumed to come from Montana, Colorado, Arizona, or Utah.)  The mer-
 cury concentration assumed for coal is an average of the concentrations
 found in coals adjacent to the area in question (Refs. 51, 54, 118, 119).

      Coal consumed by households and by the commercial, industrial, and
 transportation sectors was apportioned on a population basis for geographi-
 cal regions within states.

      Power plants  that use  coal were located geographically and the coal
 consumed by each was  determined (Ref. 49).   The average concentration of the
 mercury in the  coal was then used to obtain figures on total mercury loss
 from  this  sector of the economy.  The mercury lost during combustion is
 primarily  to air.   However,  some remains in the fly ash, bottom ash, and
 water settling ponds; disposal  of this portion is  to land.  As temperatures
within a power plant boiler  (800 P)  are  higher than the boiling point of
mercury  (685°), the majority of the  mercury lost is in the form of the
                                      D-6

-------
vaporized metal or mercuric  oxide.  With this in mind, proportional mer-
cury losses to air and land  were determined  (Refs. 12, 52, 120, 121).

NATURAL GAS COMBUSTION — RESIDENTIAL, COMMERCIAL, INDUSTRIAL  (SIC 4924C),
AND UTILITIES (SIC 49110)

     The same method used in determining coal usage within a particular
region was also used for natural gas.  Natural  gas consumed for electric
power generation was assigned to the SIC 49110  group  and the remainder to
SIC 4924C.  Adequate backup  data were  lacking,  but the mercury concentra-
tions was assumed to be 0.04 ppm  (Ref.  12).

     Because natural gas burns very cleanly and with  a very high (theore-
tical) flame temperature  (3,700°F), the bulk of the mercury is lost to
air, primarily as the metal but with slight emissions as the metallic ox-
ide.  However, evidence indicates  that some of the metallic mercury in nat-
ural gas will combine with sour gas (H S)  and precipitate onto pipeline
                                      **
walls  (Refs. 121, 122).  Mercury may be lost to land as pipe or fittings
are replaced, but in only minor amounts.   Therefore,  total losses have
been assigned to air, with a minute quantity to land.

PETROLEUM  (SIC  2911C)

     There  are varying grades of fuel oils, each with correspondingly vary-
ing amounts of mercury.   Light distillates, because  of  their  low flame
temperatures, are assumed to release insignificant amounts of mercury.
Distillate-type oils  are  found to  contain 0.066 ppm  (Ref.  12)  and residual
oils 0.13 ppm (Ref. 123).   A mass  balance was  performed to determine the
remaining mercury in  tars and asphalt  (Ref.  124), and the  mercury concen-
tration  of  these  by-products was estimated  to  be  0.7 ppm.  Total petroleum
usage  within  a given  region is determined employing  the same methodology
as that  used  for  coal and natural  gas.
                                      D-7

-------
      Mercury losses from the combustion of petroleum fuels are primarily
  i
 to air.  Because the flame temperature of most fuels is above the boiling
 point of mercury, the mercury is lost chiefly as the metal or metallic
 oxide.

 REFINERIES (SIC 2911M)

      Mercury losses from refineries can be expected to occur at  oil  sludge
 ponds, from spillage, and from burning of waste materials.   This loss is
 assumed to be 5 percent (URS estimate)  of all the mercury contained  in the
 crude oil.  The amount of crude oil processed at a particular plant  is
 found (Ref. 125) and the total amount of mercury lost is determined.  Little
 is known about these internal plant losses, so factors were arbitrarily
 applied and emissions were assigned to air and land.

 TARS AND ASPHALT (SIC 2951C)

      Tar and asphalt mercury loss is a complex matter, because an initial
 loss during application (normally to air)  occurs and then there  is a gradual
 loss due to erosion.   The  5-percent figure for initial loss to air is based
 on  current consumption data (Ref.  126),  whereas the gradual loss is  based
 on  a life span of 30  years for tar- and  asphalt-paved surfaces.  Erosion of
 the surface due to use and rainfall will cause some of the  mercury to go to
 water, with the remainder  finding its way to  land from further erosion,
 street cleaning,  etc.   The majority of the mercury loss to  air is in an or-
 ganic form or  as  mercury oxide, because  of the low melting  point of  tar and
 asphalt.   Losses  to both water and land  are in organic forms only.   In our
 study the  mercury loss  was distributed on  a population basis, the assumption
being that usage of tars and asphalts correlates  satisfactorily  with this
factor.
                                    D-8

-------
DENTAL APPLICATIONS (SIC 8021)

     Mercury is used as an amalgam in dental filling preparations.  Total
national use was apportioned to a given region on the basis of population.
It has been determined that approximately 75 percent of the mercury used
in dental preparations remains in the teeth (Ref. 12) and will not reach
the environment.  Primary losses are to water, with the remainder to  air
(volatilization) and recycling (Ref. 12).

CONTROL INSTRUMENTS (SIC 329LM AND 38291C)

     Mercury is used in flowmeters, thermometers, barometers, and similar
devices.  Total national use was determined by comparing various sources
(Refs. 121, 125).  Approximately 1 percent of this  mercury is lost during
manufacture of the instruments.  Some mercury-containing instruments are
brought into the country, while others  are exported,  and this net trans-
port was also accounted for.  The resulting total was multiplied by the
manufacturing employment ratio for a region and this figure was apportioned
to land and recycle for final consumption.  (All losses during manufacture
are assumed to go  to an industrial  landfill.)

     Plants that manufacture control instruments in a given  study region
were located by field  contacts and  through a survey of the literature.
Employment at each plant was divided by  national employment  for the
particular SIC category.  This ratio was multiplied by total manufacturing
losses to arrive at the plant's  emissions to air,  land and water.

BATTERIES  (SIC  2642M AND  3642C)

     Mercury is used  in a number of types of batteries.   National usage
was determined from various sources and was adjusted for imports, exports,
and losses.
                                      D-9

-------
      During manufacture of batteries approximately 0.5 percent of the
 mercury used is lost.  Most of this loss is to land, with smaller amounts
 to air and water  (Ref. 69) .  We assumed that 5 percent of the batteries
 disposed of in this country are incinerated; an adjustment to this figure
 was made wherever a significant difference in solid waste disposal methods
 was found.  For example. New York City incinerates approximately 40 per-
 cent of its solid waste, thereby putting more mercury into the air than
 would otherwise be lost to land.

 TUBES AND SWITCHES (SIC 3629C AND 3629M)

      Power tubes, rectifiers, relays, and switches are mercury-consuming
 devices that contribute to environmental losses of the metal.  The method
 used to determine these losses was similar to that for batteries and con-
 trols, in that national totals were derived and appropriate adjustments
 were made for imports, exports, and manufacturing losses (Refs. 24, 25).
 Consumptive losses in a given region of the United States were arrived
 at by using the manufacturing employment ratio.  All tubes and switches
 were assumed to end up eventually in landfills (with some undoubtedly be-
 ing incinerated, in which case mercury would be emitted to the air).

      Manufacturing losses are quite trivial? the assumption is that approxi-
 mately 2.5 percent of the devices rejected are not recycled and disposed
 of in an industrial landfill.

 CATALYSTS  AND MISCELLANEOUS  COMPOUNDS (SIC 2819M, 2851M, 2879M)

     The three SIC codes  shown above refer to a group of chemical manu-
 facturers who produce compounds which are precursors for use as catalysts
 and in paint and pesticides.   The  location of the manufacturers and types
 of mercurial chemicals produced are  rather easily obtained (Ref. 62).  How-
 ever, it is difficult to allocate  actual  mercury usage and concomitant
losses.  From a knowledge of the relative size of the plant and maximum
allowable discharge rates, approximations can be  made with regard to
                                     D-10

-------
mercury loss.  Prom plant visits and correspondence, it was found that
actual losses are less  than  0.1 percent of the mercury used.

     Bureau of Mines data were used to sum total mercury usage by the
paint and pesticide industries as well as for catalysts.  By applying the
0.1 percent loss, total mercury losses for this group of manufacturers can
be estimated.

CHLOR-ALKALI (SIC 2812)

     Sources of data regarding mercury consumed and lost within the chlor-
alkali industry included various Bureau of Mines publications, as well as
actual plant visits.  In the case  of  the Bureau of Mines data, only a national
total for the industry's mercury use  is given.  Therefore  the national total
was multiplied by the ratio of  SIC 2812 employment in mercury cell plants
in each study area to the national employment in SIC 2812.

     The Ref. 14 data on emissions were used, as  they pertain  to mercury
consumption by specific plants.   To validate the procedures used and to
obtain some specific figures, visits were made to certain plants (PPG Plant
in Lake Charles, Louisiana, and the BASF Wyandotte Plant,  Port Edwards,
Wisconsin).  These visits revealed that as much as 68 percent of the mer-
cury used may be unaccounted for.  It is of major concern, we feel, that
so much mercury is apparently vanishing.   The chlorine manufacturers are
aware of this problem  and are making strides toward greater accountability,
but so far have not achieved total material balances.

     The mercury that  is lost escapes to air  (hydrogen stream and end box
ventilation), water  (brine  water), land (disposal of sludges), and a small
portion  (0.42 - 0.50 percent) into product  (caustic).  The values shown in
text Table 7 represent an average loss from all U.S. plants.  Where  speci-
fic information regarding losses could be obtained, this  was used in lieu
of the averaged data.   Other adjustments were made for age of plant,  total
chlorine produced,  and history of operation.
                                      D-ll

-------
 LABORATORY USES  (SIC 7391)

      Mercury use within a given region is, of course, a proportion of
 national use  (Ref. 12).  This proportion was calculated using an American
 Chemical Society list of laboratories in a given region for which the in-
 formation was considered adequate.  However, for most regions, the figures
 obtained were judged to be out of balance with the likely levels of use.
 For these regions the multifactor ratio was used, because it corresponds
 more logically with the probable distribution of both school and indus-
 trial laboratories where mercury would be used and lost.

      Mercury losses to the environment from laboratory usage have been
 greatly reduced in the last few years.  Therefore, in contrast to the
 assignment given in Ref. 12, we have assigned the greatest proportion of
 the mercury balance to recycling.
                                          •
 PHARMACEUTICALS  (SIC 2933M AND 2834C)

      Mercury losses during Pharmaceuticals manufacture are confined almost
 exclusively to Baltimore, Maryland; Chicago, Illinois; Indianapolis,
 Indiana; and the State of New Jersey.  New Jersey alone contains approxi-
 mately 80 percent of the U.S.  plants that produce mercurial Pharmaceuticals.
 The total consumption of these mercurials in a given study region was
 prorated from national figures and distributed on a population basis.  Some
 small error was  introduced in  this procedure in that veterinary uses are
 not specifically known and could not be assigned.

      During the  manufacture of mercury-containing drugs, only about 0.1 per-
 cent  of  the total mercury is lost.   Therefore,  the bulk of the mercury used
 in Pharmaceuticals is  lost  during bathing (in the case of an externally
 applied cosmetic  or medicine),  and  for an internally applied drug mercury
 is excreted in the urine.  Essentially all  the  mercury is therefore lost  to
water.
                                     D-12

-------
PAINT (SIC 2851F AND 2851C)

     Mercury, usually in the form of phenylmercuric acetate  (PMA),  is used
in paints as a preservative and mildewcide.  Mercury  losses  during  manu-
facture are very trivial (0.25 percent  of the  PMA used).  However,  during
and after application, approximately 65 percent  of the mercury becomes
vaporized—most of it within the first  24 hours  after application.   An-
other 5 percent is assumed to remain in the  can  and to end up ultimately
in a landfill.

     Total national mercury use in paints was  determined from Bureau of
Mines figures; for regional distribution, the  multifactor ratio  was used.
The emission factors for paint formulation were  assumed  to be almost
equally divided between air and water,  and application losses were assumed
to be primarily to air.

OTHER (SIC 3.0)

     This category includes mercury used in miscellaneous products such as
pump seals, together with chemicals not specifically addressed elsewhere.
Total national usage of mercury in this category was estimated using
Ref. 12, and URS applied the multif actor to determine regional totals  and
to establish apportionments with regions.
                <
URETHANE PRODUCTION  (SIC 2.5A) AND FOAM CAUSTICS (2.5B)

     The use of mercury as a  catalyst  in urethane production was determined
from national totals.  Of the total reported  by the  Bureau  of Mines,  approxi-
mately two-thirds goes into urethane foam manufacture and the other one-
third into vat dyes.  A regional proportion was assumed simply  in terms of
the population percentage of  the  area, because  the urethanes are used pri-
marily as a  soundproofing agent in walls.   Distribution of  mercury to the
                                       D-13

-------
  appropriate environmental parameters was based on the means of solid waste
     »
  disposal of these foam products.

       Caustic (sodium hydroxide),  recovered as a by-product of the chlorine
  manufacturing process, is used in the manufacture of a number of products,
  particularly foams.   Mercury is contained as a trace contaminant (4 ppm)
  and therefore finds  its way into  the foam, and is lost primarily to water.
  The mercury lost was distributed  within regions on a manufacturing basis.

  LAMPS (SIC  36410M AND SIC 36410C)

       Each fluorescent lamp contains approximately 50 mg of mercury (Ref. 127).
  About 4.6 percent of the  mercury  consumed nationally is used in the electri-
  cal and  control  instrument industries.  Proper adjustments were made for im-
  ports and exports to arrive at consumptive losses, most of which are to land,
  with  a slight amount to air if spent bulbs reach the municipal waste.  The
  manufacturing employment  ratio was used to obtain regional totals and ac-
  complish the  apportionment.  Approximately 4 percent of the lamps are assumed
  to be  rejected during manufacture, with approximately 5 percent of the mer-
  cury lost to  air  due  to vaporization during testing, and the remainder going
  to industrial  landfills.

 GOVERNMENT USAGE  (SIC 9199)

      Two of the largest government consumers of mercury are the Atomic Energy
 Commission (AEC)  and the Department of Defense (Ref. 82).  Much of this use
 is for reprocessing spent nuclear fuel.   Because it is very difficult to ob-
 tain information  regarding mercury use by these and similar agencies, approxi-
 mations were made. For example, in the Kentucky-Tennessee study region, a
 fractional proportion of the total U.S.  mercury use was assigned to the mili-
 tary installations there,  based on the military population of each state.  In
 the  case  of Oak Ridge National Laboratory (ORNL),  assumptions were made con-
cerning its potential mercury losses.   It was  assumed that ORNL uses 5 per-
cent of the U.S.  total consumed by the AEC.  It was also assumed that
                                       D-14

-------
4 percent of this is lost to the environment, of which 30 percent goes to
water and 70 percent to land. This was purely a conjectural attempt to
account for mercury used by this sector of our economy, and our figures are,
of course, subject to revision.

DEGASSING

     The calcultion of natural degassing  rates was based on work performed
by others and on observations and  assumptions developed in this study.
Various workers have developed both global and regional degassing rates.
These data are summarized in text  Table 8.

     These observations and calculations  indicate  that:   (1)  the average
                                       2
land degassing rate is probably  0.2 yg/m  /day;  (2) degassing rates  higher
than the average are associated  with  mineralized deposits;  (3)  over aqueous
sediments high in mercury, the degassing  rate is probably higher than over
land;  (4) the ratio between global land and ocean degassing rates is un-
known, but MacKenzie's assumption of equality is probably adequate.  Using
these concepts, URS calculated a probable national degassing rate.   The
assumptions for these calculations are summarized below:

     1.   The base-national degassing rate was assumed to be
          0.2 yg/m /day.
     2.   For moderately mineralized areas, a degassing rate of
          0.4 yg/m /day.
     3.   For highly .mineralized areas and known mercury deposits,
          a 0.8 yg/m /day was assumed.

     For the purposes of differentiating between areas, text Figure 9,  which
indicates general areas of  known mercury and mineralized deposits,  was  used.
Using  the degassing rates and the total  surface area (both  land and water)
for each state, a state-by-state inventory  of total annual  degassing was
made.  On a national basis, the annual degassing  rate was estimated to be
about  1 million kg per  year, equivalent  to  a national unit rate of
          2
0.36 yg/m /year.

                                      D-15

-------
      The other contribution from natural sources, that of erosion, was de-
      t
 termined by calculating the average annual runoff from each state and
 multiplying this  figure by the average statewide sediment concentration
  (Ref.  128).  The  total annual sediment load was multiplied by the average
 sediment mercury  concentration of 71 ppb (Ref. 29).  These state-by-state
 totals were then  summed for the national inventory.  This amounted to about
 200,000 kg per year.  An attempt was also made to determine the contribu-
 tion from  urban runoff.  According to this calculation, urban runoff con-
 tributes about 10,000 kg per year to the nation's waters, or about 5 per-
 cent of that from soil erosion.  Therefore, on an areal basis, the contri-
 bution of  urban sources is about four times higher than that of natural
 sources  (but urban areas represent only about 1.2 percent of the total land
 area).
 Runoff
      Runoff contributions were calculated by performing a surface water
 balance on major streams within each region.  The resulting net flow rate
 was then multiplied by 1 ppb (an average mercury content of most major
 streams) to determine the average mercury input rate.

 Solid Waste

      Information on the disposal of solid waste was obtained from personal
 communications with state and federal solid waste management personnel.
 This information,  coupled with average values for mercury in solid wastes,
 was utilized in distributing mercury losses due to incineration in various
 regions.

 Groundwater

     Very little information is available on mercury concentrations in
groundwater, but the most widely quoted value is 0.05 ppb.   This is a
reasonable figure, because the  general  high adsorptive capacity of soils
                                    D-16

-------
doubtless reduces mercury concentrations as water percolates to the water
table.  Data on quantities of groundwater withdrawn in a given region were
taken from USGS publications on water  supply for individual states.  The
yearly quantity of mercury was obtained by applying a list of conversion
factors to these values.

Septic Tanks

     The amount of mercury lost to land from septic tanks was calculated
by first assuming an average mercury content of 2 ppb in the sewage
(Ref. 12).  The total water volume was determined by subtracting from the
total population of a region the  number of people served by municipal treat-
ment systems and multiplying the  result by  75  gallons per  capita per day
(average loading from single family residences).  There is also some loss
of mercury to land due to irrigation,  leaks, and  so on.  The amount depends
on the region, but these increments are trivial.
Sewage
     Data were obtained on mercury lost in disposal of sewage and sludge by
municipal water users.  The total net output of mercury was calculated by
subtracting the mercury input from water supply and the mercury output ex-
pected from sewage treatment plants.  Mercury quantities in various flows
were calculated by multiplying the mercury concentration and water quantity
data using appropriate conversion factors.
                                      D-17

-------
                                                              APPENDIX E
                                       POINT SOURCE EMISSIONS OF MERCURY


     Two primary sources of mercury vapor that enters the atmosphere are
coal-fired power plants and copper smelters.  Such sources may subsequent-
ly also contribute to relatively high ground level concentrations.  With
this in mind, dispersion analyses were performed for a large coal-burning
power plant for which mercury emissions data were available, and for a
copper smelter.

     Figures E-l through E-3  are isopleths  of mercury vapor concentrations
which result from a copper smelting operation in Arizona.  Three cases
were analyzed for mercury vapor concentration:   (a) adverse meteorologi-
cal conditions (very unstable and  1 meter per second wind  speed) and an
8-hour averaging time; (b) average meteorological  conditions  (neutral sta-
bility and 3 meters per second wind  speed)  and  a 3-minute  averaging time;
(c) adverse meteorological conditions  (very unstable  and 1 meter per sec-
ond wind speed) and 3-minute  averaging time.  The  three cases were run to
compare with the occupational eight-hour average,  and to determine what
short-term concentration might exist under average and adverse conditions.
The highest concentration  (MO yg/m )  occurs  in Case 3, where the averag-
ing time is shortened and the meteorological  conditions are most conducive
to producing high ambient levels.

     Case 4  (Fig. E-4) pertains to a large coal-burning power plant
(M.2,000) tons of coal consumed per day).  Stack gas concentrations of
                                                   3
mercury were determined to be approximately 18 yg/m .   (This compares  to a
concentration of approximately 167 yg/m  in the stack gases of the copper
smelters.)  The maximum concentration generated from two stacks (300 and
                  3
500 feet is 7 yg/m  — well below any dangerous levels.
                                     E-l

-------
    45
    60
    75
  o
 Q
  12
 £
90
  105
  120
  135
                                       Wind Speed » 1 meter/second
                                       Stability = Very unstable (Class A)
                                       2ng/rrr
     0          15

SOURCE: URS Research Company.
                         30          45
                            Kilometers
60
75
Figure E-1  CASE 1: COPPER SMELTER MERCURY VAPOR CONCENTRATION
          UNDER ADVERSE METEOROLOGICAL CONDITIONS
          (8-hour average)
                                E-2

-------
                                       Wind Speed - 3 meters/second
                                       Stability = Neutral (Class O)
SOURCE: URS Research Company.
Figure E-2 CASE 2: COPPER SMELTER MERCURY VAPOR CONCENTRATION
         UNDER AVERAGE METEOROLOGICAL CONDITIONS (3-m1n avg)

-------
    1.5
    2.0
     2C
     .*
  I 3.0
  c
  o
  O
    3.5
   4.0
   4.5
   5.0
                 0.5
SOURCE: URS Research Company.
                                         Wind Speed = 2 meters/second
                                         Stability - Very unstable (Class A)
                20 ng/m2
1.0          1.5
   Kilometers
2.0
2.5
Figure E-3  CASES: COPPER SMELTER MERCURY VAPOR CONCENTRATION
          UNDER ADVERSE METEOROLOGICAL CONDITIONS
          (3-minute average)
                                E-4

-------
  1.5
  2.0 —
  2.5
I 3.0
Q
e
I
  3.5
  4.0
  4.5
  5.0
                0.5
SOURCE: URS Research Company.
                                         Wind Speed = 1 meter/second
                                             4ng/m2
1.0          1.5
    Kilometers
2.0
2.5
  Figure E-4  CASE 4: POWER PLANT MERCURY VAPOR CONCENTRATION
            UNDER ADVERSE METEOROLOGICAL CONDITIONS
            (3-minute average)
                               E-5

-------
      The mercury vapor concentrations generated from these two point
  sources are therefore going to be relatively small, even under the most
  adverse meteorological conditions.  Although appreciable quantities of
  mercury are released from these two sources each year (1,372 kg from the
  smelter and 589 kg from the power plant in 1973), the emissions occur over
  a widespread area and are well dispersed before they could pose a hazard
  to  the general population at ground level.

      Another concern is the potential buildup of mercury in the soil ad-
  jacent to such sources as power plants and smelters.  In order to arrive
  at  an estimate of mercury in the soil, various assumptions had to be
  made:

      o    The mercury will be found to a soil depth of 10 cm.
      o    The average density of the soil is 2.5 gm/cm .
      o    Approximately 10 percent of the mercury emitted from
           an elevated point source is locally deposited within
           a radius of 10 km.
      o    Some 80 percent of the deposited mercury remains in
           the soil.
      o    The average background soil concentration of mercury
           is  71 ppb.
      o    The  average life of mercury in soil is 20 years.

      Using these relatively gross  assumptions,  a mercury concentration in
 soil adjacent to the  copper smelter  discussed previously was calculated.
 It was found that the smelter  could  contribute  an additional 111 ppb to
 existing background.  However, by  changing the  assumptions above to not
 unreasonable figures, the range of mercury concentrations in soil could
be 70 to 890 ppb.  Unfortunately,  no specific data are available to corre-
late with these findings, although data points  in the  general area of the
smelter do show elevated levels  (>200 ppb).
                                     E-6

-------
     Based on the information presented above, it appears that copper
smelting operations do not pose a health hazard in terms of airborne emis-
sions of mercury.  It is very difficult to predict what happens to the mer-
cury once it is released from the stack, as it may adhere to particles and
settle out close to the plant, or it may be transported long distances, in
which case it becomes part of a regional transport phenomenon.  In any
case, apparently no hazard to health will be produced as a consequence of
air?emissions from the smelter studied.  Based on these findings and on
mercury emissions data from other smelting operations, it seems that no
further curtailment of operations,"•and no new controls, are necessary to
reduce mercury emissions.
                                      E-7

-------
BIBLIOGRAPHIC DATA
SHEET
                    1. Report No.
                           EPA 560/3-75-007
           3. Recipient's Accession No.
4. Title and Subtitle
 Materials Balance  and Technology Assessment of Mercury and Its
 Compounds on National and Regional  Bases
                                                                     5. Report Date
                                                                          October, 1975
                                                                     6.
7. Author(s)
  William Van Horn
                                                                     8. Performing Organization Rept.
                                                                       No.
9. Performing Organization Name and Address
                                                                     10. Project/Task/Work Unit No.
      URS Research  Company
      155 Bovet Road,
      San Mateo, California
                               94402
           11. Contract/Grant No.

                68-01-2931
12. Sponsoring Organization Name and Address

     U.S.  Environmental  Protection Agency
     Office of Toxic Substances
     Washington, D.C.  20460
                                                                     13. Type of Report & Period
                                                                        Covered
                                                                           Final Report
                                                                     14.
 15. Supplementary Notes
16. Abstracts
      The role of mercury and its compounds in the environment and  the economy  of the
United States was studied.   A detailed  material balance for mercury and its compounds wa
developed on a national  basis and for selected geographical regions, including estimates
of the environmental  fate of all emissions.

      Current and projected process technologies for mercury products were examined, and
estimates of environmental  losses for 1973 and 1983 were presented.  A set of  regulatory
alternatives was developed for each of  the major technologies involving substantial
losses of mercury to  the environment, and the economic impact of these alternatives
was examined.
17. Key Words and Document Analysis.  17o. Descriptors
           Mercury
           Mercury Compounds
           Mercury Production
           Mercury Compounds Production
           Mercury Use
           Mercury Compounds Use
           Mercury Control  Technology
           Substitutes  for  Mercury
17b. Identifiers/Open-Ended Terms
                                          Mercury in Air
                                          Mercury in Water
                                          Mercury in Land-Destined Wastes
                                          Mercury in Batteries
                                          Mercury in Electrical Equipment
                                          Mercury in Chior-Alkali  Industry
                                          Mercury in Dentistry
                                          Mercury in Fossil Fuels
                                          Mercury in Paints
                                          Mercury in Pesticides
                                          Mercury in Pharmaceuticals
I7e. COSATI Field/Group
18. Availability Statement

     Release Unlimited
                                                          19.- Security Class (This
                                                             Report)
                                                                        fb
20. Security Class (This
   Page
     UNCLASSIFIED
                      21. No. of Pages
                                                                                 22. Price
FORM NTIS.SB (REV. 10-73)  ENDORSED BY ANSI AND UNESCO.
                                                    THIS FORM MAY BE REPRODUCED
                                                                                 USCOMM'DC B26S-P74

-------