-------

-------
            UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

                       WASHINGTON, D,C. £0450
April 26," 1988

The Honorable Lee M* Thomas                           „
Administrator
U.S. Environmental Protection Agency
401 M. Street, S.W.
Washington, D.C.  20460

Dear Mr. Thomass
                                                    t
     The Municipal  Waste  Combustion Subcommittee  of the Science
Advisory Board's Environmental Effects,  Transport and Fate
Committee has  completed  its report entitled "Evaluation of
Scientific Issues Related to Municipal Waste  Combustion".  The
evaluation was  initiated  at your requestt  along  with two  other
charges related to  municipal  waste  combustion,  all of which are
now complete.   The Subcommittee began  gathering  information in
April  of  1986  and has  achieved  consensus  on  a "number of
conclusions and  recommendations  in  the  intervening time.   These
findings are summarized below.'

     The  Subcommittee  recognizes  that  regardless  of  the
technologies a society employs to reduce or dispose of municipal
waste,-there will 'always be a degree of  residual risk  to  both the
public and the  environment.   Members of the Subcommittee do not
attempt to' evaluate all of  the  issues  that municipalities  must
weigh as they consider incineration'as a waste  management option,
but  instead  strive to  inform citizens  and decision  makers of
current  risks and  uncertainties . accompanied  by  recommendations
for increasing knowledge to reduce such  risks and  uncertainties.
The report  examines a series -of generic scientific issues  that
policy makers must address in an  order that  reflects the  movement
of  potential - pollutants  through  and  from a municipal  waste
combustion  facility,  in  particular, such  issues,  as combustor
feedstocks; the  design  and operation of municipal  incinerators;
the performance of incinerators with various degrees of pollution
control  equipment; stack emissions;   ash disposal;  operator
training and  certification;  environmental transport and  fate of
combustion residues and by-products; pathways  to and potential
for' exposures of  humans  and  ecosystems;  and  potential  public
health and environmental effects  are addressed.

     The Subcommittee concludes that, in general,  the performance
side of thek  technology,  including design and pollution  control,
has greatly' improved, and  is  likely to  continue to improve.  In
the Subcommittee's  judgment,  two .critical needs  at present are
expanded and  more  rigorous operator training  requirements, and

-------
data  collection  and analyses  aimed  at enabling  scientists  and
decision  makers to better estimate health  and environmental
exposures from this technology.

     Since  technological  improvements  have  created  highly
efficient stack emission control  systems, fly ash with relatively
smaller particle size and increased concentrations of pollutants
such  as  heavy  metals   and  trace  organics  has  resulted.    The
Subcommittee recommends that EPA develop  a  series of alternative
techniques  fors 1)  analyzing ash samples and  the compounds
present in ash extracts; 2)  assessing the toxic potential of ash;
and 3} managing ash disposal.

     The  subcommittee  also recommends  that  the potential  for
health and  environmental effects  be  addressed by  developing  a
more  comprehensive  data base through field  studies.    Little
information is presently available on the  fate of chemicals from
MWC facilities, and information is needed to estimate deposition
of  particulate  and  gaseous  emissions,  to  model transport  and
diffusions   operations,  and   to  understand  environmental
transformation and dispersal of  technology  by-products  that  may
pose risk.

     Finally,  the  Subcommittee recommends that  EPA  assist local
decision makers and the public by developing ways to collect and
analyze data that will allow more informed  choices regarding the
management  of municipal solid  waste.   Approaches should  be
developed for assessing exposure and risk and these tools should
be  transferred  to  the  parties  responsible  for  making  the
decisions.  Appropriate tools may include guidance for evaluating
waste management options,  and  means for comparing  exposure  and
risk between available  options.

     The Subcommittee appreciates the  opportunity to conduct this
scientific review.    We  request  that  the  Agency formally respond
to the scientific advice transmitted  in the  attached report.

                                       sincerely,
                                       Norton Nelson, Chairman
                                       Executive Committee
                                       Science Advisory Board
                                       Red i Hartung, Chmrman
                                       Municipal Waste
                                         Combustion Subcommittee
Enc, cc;  A. James Barnes
          J. Winston Porter
          Vaun Newill
          Donald £. Barnes

-------
                                         SAS-EBTFC-8 8-2 5
          EVALUATION OF SCIENTIFIC ISSUES
            RELATED TO MUNICIPAL WASTS
                    COMBUSTION
     REPORT OP THE MUNICIPAL WASTE COMBUSTION
                   SUBCOMMITTEE
ENVIRONMENTAL EFFECTS, TRANSPORT AND FATE COMMITTEE
              SCIENCE ADVISORY BOARD
       U.S. ENVIRONMENTAL PROTECTION AGENCY
                    APRIL 1988

-------

-------
                             NOTICE

     This report Has been written as a part of the activities  of
the  Science  Advisory Board,  a public , advisory  group ' providing
extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency.  The
Board is  structured to provide a  balanced expert assessment  of
scientific matters  related to  problems  facing the Agency. ' This
report has not been reviewed  for approval by the  Agency, and,
                                                     i
hence the  contents  of this  report  do not necessarily  represent
the views  and policies of the Environmental  Protection  Agency,*
nor  of  other  agencies  in the Executive Branch  of  the  Federal
government,   nor  does  mention  of  trade names  or commercial
products constitute endorsement- of recommendation  for use.

-------
              U.S.  ENVIRONMENTAL PROTECTION AGENCY
                     SCIENCE ADVISORY BOARD
       ENVIRONMENTAL EFFECTS, TRANSPORT AND FATE COMMITTEE
             MUNICIPAL WASTE COMBUSTION SUBCOMMITTEE

Chairman

Dr. Rolf Hartung,  Professor of Environmental Toxicology,  School
of  Public  Health,  University  of Michigan,  Ann Arbor,  Michigan
48109

Members

Dr. Martin Alexander, Professor, Department  of  Agronomy,  Cornell
University, Ithaca, New York  14853

Dr. Stanley Auerbach, Environmental  Sciences Division,  Oak Ridge
National Laboratory,  Oak Ridge, Tennessee  37831

Mr. Allen  Cywin, P.E.,  1126 Arcturus Lane,  Alexandria,  Virginia
22308

Dr. Walter Dabberdt, National  Center for Atmospheric  Research,
P.O. Box 3000, Boulder,  Colorado  80307-3000
                                                       n
Dr.  Robert  Huggett,  Professor of Marine Science,  Virginia
Institute  of  Marine  Science,  School of Marine  Sciences,  College
of William and Mary,  Gloucester Point, Virginia   230€2

Mr. Alfred Joensen, Associate Professor, Department of  Mechanical
Engineering, Iowa State University,  Ames,  Iowa   50011

*Dr. Renate  Kimbrough,  Centers for  Disease  Control,  Center  for
Environmental Health, 1600 Clifton Road, Atlanta,  Georgia  30333

Mr.  Raymond  Klicius,  Environment  Canada, 351 St.  Joseph's
Boulevard,  Hull Quebec,  Canada  K1AOE7

Dr. William  Lowrance,  Senior Fellow and Director,  Life  sciences
and Public Policy Program, The  Rockefeller University,  1230  York
Avenue, New York, New York  10021

Dr. John Neuhold, Professor of Fisheries and Wildlife,  College
of Natural Sciences,  Utah State University,  Logan, Utah  34322

Dr.  Adel   Sarofim,  Department   of   Chemical   Engineering,
Massachusetts  Institute  of Technology, Cambridge,  Massachusetts
02139

Mr.  Charles  0.  Velzy,   Charles R.   Velzy Associates,  355  Main
Street, Armonk, New York  10504


*DrTKimbroughr"served on the Subcommittee until  May 11,  1987.
                               ii

-------
Staff Director

*Dr. Donald Barnes, U.S. EPA, Science Advisory Board  (A101), 401  M
Street, S.w., 1145 WE, Washington, D.C.  20460

Executive Secretary

Ms. Janis  C.  Kurtz,  U.S.  EPA,  Science Advisory Board  (AIQ1-F),
401 M Street, S.W., Room 508, Washington, D.C.   20460

Staff Secretary

Mrs, Liitithia Barbee, U.S.  EPA,  Science Advisory Board  (A101-F) ,
401 M Street, S.w., Room 508, Washington, D.c.   20460
*Dr. ferry P* ¥osie served as Director until February 28, 1988,


                               ill

-------
                        Table of Contents
                                                    Page Number
 I.     Executive Summary                                 1

       A.   Municipal Waste Combustion:  Process and       3
               Technology
       B.   Operator Training and Certification           5
     *  c.   Air Pollution Control Technologies            6
       D.   Ash Characterization and Disposal             8
       £.   Environmental Transport and Fate              9
       F.   Assessment of Risk to Public Health and      11
               the Environment

II.     Introduction                                     12

       A.   Charge and Scope of the Review               12
       B.   Major Assumptions and Limitations of the     14
               Review

III.   The National Municipal Waste Management Problem  16

IV.     The Process and Technology of Incinerating       20
          Municipal Waste

       A.   Feedstock                                    20
       B.   The Incineration Process                     21
       C.   Description of Combustion Systems           *22
           1.  Mass Burning of Unprocessed Municipal    23
                 Waste
           2.  Modular/Starved Air Burning of           25
                 Unprocessed Municipal Waste
           3.  Dedicated Stoker Boilers Burning         25
                 Coarsely Processed Refuse
           4.  Cofiring of Coal and Municipal Solid  '   29
                 Waste Burning Processed Refuse
       D.   Combustion System Design and Operating       29
           1.  Stages of Combustion Operation, Old and  29
                 New Plant Designs
           2.  Emissions from the Combustion Chamber    31
               a.  Acid Gases                           31
               b.  Fly Ash and Residues                 32
               c-  Trace Metals                         33
               d.  Organic Compounds                    33
       E.   Operator Training and Certification          35
       F,   Conclusions and Recommendations              36
           1.  Conclusions                              36
           2.  Recommendations                          3?

V,     Performance of Air Pollution Control Technology  40
       A.   Potential Air Pollutants of Concern          40
       B.   Description of Air Pollution Control Systems 40
           I.  Electrostatic Precipitators              40
           2.  Fabric Filters                           44
           3.  Scrubbers                                44
                               iv

-------
V.    Performance of Air Pollution Control Technology
          (Continued)        .  ,   .,.,         -^i^*

      C.  Historical Perspective of Air Pollution       46
            Control for MSW Incinerators
      D»  Air Pollution Control Experience              50
          1.  Particulates                              50
          2.  Metals                                    51
          3,  Acid Gases                                S3
          4.  Trace Qrganies                            54
          5*  Conventional Combustion Gases           .  56
          6*  Ash Disposal                              57
          7.  Ongoing Research and Development          57
      E,  Conclusions and Recommendations               58
          1.  Conclusions                               S3
          2.  Recommendations                         ,  59
                                                      j
VI.   Environmental Transport and Fate                  60

      A.  Dispersal and Persistence in Environmental    60
            Media
          1.  The Atmosphere                            61
          2.  The Terrestrial Environment               67
          3.  The Aquatic Environment                   70
      B,  Conclusions and Recommendations              *73
          1,  Conclusions                               73
          2,  Recommendations                           73
      C.  Transport and Fate of MWC Ash                 74
          1.  Considerations                            74
          2.  Conclusions and Recommendations           76

•VII.  Potential Exposure and Effects                    78

      A.  Environmental Loadings                        79
      B.  Exposures                                     80
          1.  Human Exposures      ,                     80
          2.  Ecosystem Exposures                '       81
          3.  Approaches for Estimating Exposures       82
      C.  Effects        ,                               82
          1,  Human Health Effects                      32
          2.  Environmental Exposures                   83
      D.  Conclusions and Recommendations               84
          1.  Conclusions                               84
          2.  Recommendations                           35

VIII. Concluding Perspectives                           87

LITERATURE CITED                                        91

-------
                     APPENDICES

                                               P_as§ Number
As  Assessing EPA's Risk Assessment Methodology       A-l
      for Municipal Incinerator Emissions;
      Key Findings and Conclusions

B:  Dioxin Toxic Equivalency Methodology              B-l
      Subcommittee Report; Executive Summary

GJ  Review of the Municipal Waste Combustion          c-i
      Research Plan

D:  Description of Refuse Derived Fuel (RDF)          D-l
          Categories

EJ  Glossary of Terms and Units                       E-i
                         vi

-------
                         List of Tables
  Table
Table
Table
Table 3j
Table
Tab!*
Table 6;
                                      	 Number

Existing MWC Facilities by Design Type       19
Current and Predicted Composition of         20
     Discarded Residential and Commercial
     Solid Waste
Concentration of Stack Emission Components   52
     for MWC Equipped with Scrubber/
     Fabric Filters
Scrubber/Fabric Filter Performance         .  55
Distance in km where Dry Deposition          65
     Depletes the Mass of a Plume
     by 50 percent
Conparative Advantages of Selected Waste    •89
     Disposal Options                       -
Figure l:
Figure 2*.
Figure 3;
Figure 4;
Figure 5s
Figure 6j
Figure 7;
Figure 8!
Figure 9:
Figure 10s
               List of Figures

 Mass Burning Incinerator                    24
 MSW Grate                      •             26
 Starved Air Combustors                      27
 BDF-Fired Combustor                         28
 Electrostatic Precipitation Process         42
 Arrangement of Electrostatic Precipitators  43
 Fabric Filter                               45
 Dry Scrubber                                47
 Wet-Dry Scrubber                            48
 Transport of MWC Emissions from an          62
     Incinerator Facility Through the
     Ecosystem

-------

-------
                      I.   EXECUTIVE SUMM&RY

     The present problem  of  safely disposing of municipal solid
waste  (MSW) •• stems   from  practices of  modern industrial society
that  emphasize  storage of  wastes   over disposal  methods that
destroy wastes  or" minimize  them at  the source.  Regardless  of
which technologies a society uses to reduce or otherwise dispose
of  municipal solid wastes,  there will  always be • a  degree  of
residual risk to the public and the environment.    Incineration
offers particular environmental  advantages and disadvantages as a
waste disposal option, and  may  be applied as  a technology with
varying degrees of safety and effectiveness.   Safe and effective
application requires well  designed plants,  state-of-the-art
pollution control equipment  and appropriately skilled operating
personnel.

     Because different technological  options exist for^municipal
waste  disposal,  EPA  has  an  important  'task  —   prior  to
establishing a  comprehensive strategy for  regulating municipal
waste combustion — to generate and evaluate  data,  and develop
methodologies,  for assessing the relative risks  resulting from
incineration and other municipal  waste 'disposal  processes.  EPA
needs to perform, this task  to  enable citizens,  scientists  and
public officials to  compare risks across  various environmental
media  for  each  waste disposal and  management  option.   Such
comparative analysis can  provide  .the technical basis   for
choosing among technological options.

    Communities    must   evaluate    available   technological
alternatives for waste  disposal and  choose the technology (or
combination  of technologies) that presents  acceptable levels  of
public  health  and   environmental  risk.     Given  the  local
differences  in, waste  composition, available landfill capacity,
urban and  rural  'locations,  population density,  cost,  and other
factors, no -single disposal technology is likely to be uniformly
efficient  or safe in all regions of  the  country.    Thus,  the
overall goal,  for   individual  communities  and for  society  in

-------
general,  is to choose the particular technological option(a)  that
is both  cost-effective and  presents  the least risk  to the
population and the environment.

     The  Municipal  Waste  combustion  subcommittee  of  the
Environmental Effects,  Transport  and Fate  Committee  of the
Science Advisory Board has  evaluated a series of technical  issues
related to the performance  of waste  combustion  technologies.   it
has examined,  in particular, such issues as eombustor feedstocks?
the  design  and  operation  of  municipal  incinerators;  the
performance of  incinerators with various  degrees of  pollution
control  equipment; stack  emissions,*   ash disposal; operator
training and certification; environmental  transport and fate  of
combustion residues and  by-products;  pathways  to  and  potential
for  exposures  of humans and ecosystems;   and  potential  public
health and environmental  effects.
     Evaluating the  human  health  and  environmental impacts  of
municipal waste combustion  is a difficult task.  This- is true for
a number  of reasons  including!    1)  difficulty in  identifying
and/or  obtaining  a  representative or  "average" sample  of
municipal waste; 2)  variability  in the  conditions  of combustion?
3) limited information on the identity of emitted  compounds;  4)
lack of validation of transport  and fate models; 5}  the relative
lack  of data  on  the environmental  loadings contributed  by
incinerators compared to  other combustion sources {including coal
and oil  fired power  plants,  automobiles,  and  wood stoves  and
fireplaces)j and  6)  large uncertainties in  estimating  human
health and environmental effects  from  municipal incineration  in
comparison to other combustion sources.

     In evaluating the issues identifi*K  above,, the Subcommittee
has reached the following major conclusions and recommendations;

-------
A.   Municipal Waste Combustion!  greeess and Technology

     o  Municipal  solid   waste  (MSW)  is   heterogeneous   in
composition.   MSW composition is heavily dependent  on location,
time of  year  and patterns  of  consumption.    Even  simple
constituents  such as moisture content may  fluctuate widely,
Because of  the inherent  variability of the MSW feedstock,  it  is
difficult to predict the composition of  stack emissions that may
result from combustion.  Wide variations in feedstock composition
can  affect  combustion conditions in the incinerator furnaces,  and
can  cause  cycles of poor  combustion*  Poor  combustion conditions
have a direct impact  on  emissions.   It is  important  to  design
incinerators  with  state-of-the-art features  that will  provide
operators  with  the ability  to accommodate  wide variations  in
feedstock composition to reduce the potential for poor combustion
and  increased  emissions.
                                                       b
     o  Organic  materials  containing only carbon and hydrogen are
completely  combusted or burned in an oxygen-containing atmosphere
-theoretically  producing  water  vapor and carbon  dioxide as  the
products of combustion.   Municipal solid waste,  however,  is  not
composed  entirely of organic materials  or carbon and hydrogen,
and, therefore|.many other products of combustion are released to
the  environment.   In addition, combustion is not always complete,
resulting in release of products  of incomplete combustion (PIC).
Proper  or complete combustion  depends not only  on  sufficiently
elevated temperatures  but also on the  residence time needed for
the  materials  to burn   fully,   and  the  need  for  turbulent
conditions  in  the furnace in order to achieve proper  mixing  of
air  and  the gases evolved from the burning fuel, which all vary
with the composition of the waste.

    - o    Increased competition  for  the  growing   market  for
incinerators is  leading to improvements in engineering design and
especially  to an  increased understanding and  sophistication  of
the  technology  of  combustion.     Recognition of   environmental
problems has also  been a key  factor motivating the development of

-------
improved  combustors  and emission  control  equipment.  Earlier
designs of mass burners did  not  incorporate  the flexibility for
controlling the location and  amount  of introduced combustion air,
or the sophistication of instrumentation for control of feedback
combustion air that   newer  plant designs provide.   Thus,  older
plants generally  do  not  achieve the  efficiency  of combustion
attainable in modern plants.

     o  The  Subcommittee concludes  that EPA should investigate
the  hypothesis  that  polychlorinated  dibenzodioxins  (PCDD)  and
polychlorinated dibenzofurans (PCDF) can result  from  free radical
reactions that take  place  in fuel-rich zones of  incinerator
flames.    These   reactions  may   yield '•  polycyclic   aromatic
hydrocarbons (PAH) ,  oxygenated compounds  such  as phenol  and
perhaps,  in the presence of chlorine, some PCDD  and PCDF,

     These compounds may also be  present in MWC feed stock since
they are  by-product contaminants in a number of chemicals, most
notably chlorinated phenols and polychlotinated biphenyls  (PCS) .
These  compounds  may   persist  beyond   combustion   only  if
temperatures  are  sufficiently cooled by  excesses in  local air
flow.  Condensation reactions involving the  chlorinated phenols,
phenol ethers, and biphenyls  may  also produce PCDD and
     o  The  design  and operation  of an  incinerator combustion
chamber has major  impact  on  the concentration of the pollutants
entering the air pollution control devices.   In well-designed and
operated  incinerators,  the  emissions  of  organic  compounds
currently measured can be  reduced to levels close to the limits
of detection with existing analytical methods.

Rec.ommendat ions

     o  EPA  and private vendors  should fund  research  to gain a
better understanding of  1)" municipal solid  waste  composition? 2)
the  affects  of furnace design and operating conditions  on the
combustion process? 3)  the relation between  inorganic and organic

-------
emissions   and;   4)  PCDD and PCDF  formation..  To  obtain  this
information, research on full-,  pilot-,  and laboratory-scale units
is needed.

     o Well-planned  field  testing that  evaluates different
operating conditions  will generate  a  realistic correlation  of
emissions with operations,  and will provide data for establishing
emission indices.   Pilot-plant and  laboratory-scale  testing can
be used  to  critically investigate hypotheses  derived from field
studies.    Small-scale equipment  facilitates such testing because
of the greater ease of independently varying design parameters,
operating conditions,  and feed composition.           ,

     o   Research on the relationship .between  the composition of
                                             *
MSW  and  emissions  should be  carried out  over a wide  enough
temperature range to be useful in testing the various hypotheses
for formation  of PCDD and PCDF.   Specific research is required
to  understand  post-combustion  formation  of   PCDD/PCDF  by
condensation reactions that occur as the flue gas   cools  in the
heat recovery process.

     o  Continuous  monitors to detect  upsets  in operating
conditions should be developed.   Carbon monoxide and hydrocarbons
are currently being explored as potential indicators of emissions
of PCDD  and PCDF.   Alternatives "such as  polycyclic aromatics,
detectable by their fluorescence or ultraviolet irradiation, may
be more appropriate surrogates.

B.   operator Tr^ij^gg and Certification

     o The  combustion of municipal solid wastes at resource
recovery facilities is exempt from the Subtitle C requirements of
the- Resource  Conservation and Recovery  Act,  provided that the
owners or  operators  take  precautions  to ensure that hazardous
wastes are  not burned.  Because, of  this exemption,  no national
policy related to  operator  training and/or  certification is
required for  municipal solid  waste  combustion  facilities.

-------
However,  proper operation  of  municipal  incinerators requires a
thorough understanding  of the  complexities  of  the  combustion
process.  An understanding of the composition and variability of
the  feedstock,   fundamentals   of  the  combustion  process,
requirements  and consequences  of  adequate emission controls,
procedures  for handling upset conditions,  and  elements of  safe
operator practice  are required for efficient and ef-fective
municipal waste disposal*   At present, there are no recommended
criteria for selecting MWC staff nor is there an existing pool of
trained,  experienced personnel to  operate  municipal  waste
combustion  facilities.

     o  New facilities planned and/or under construction are  much
more  complex  than  existing  facilities.    The need for  proper
operation of new  plants is made even more critical by the rapidly
increasing  complexity of regulatory requirements and the need for
increases  in capacity and   efficiency  of  pollutant control
devices to  ensure environmentally safe plant operations.

Recommendations

     o  EPAf  the  states  and private vendors  should  support and
promote efforts  to  ensure that  adequate  training   programs are
developed  to  provide  a  reservoir of  technically competent
personnel   to staff  municipal waste combustors.   Training
programs should be readily available,  developed with appropriate
expertise,  and  tailored to  the  specific  technology  being
utilized, and the  programs  should lead  to  certification  when
sufficient expertise is demonstrated.
C.   Ajj: BallatioB Control  Technologies
     o  There  appear  to be trade-offs between .the  influence  of
combustor design  and  operation  and the technology  of  emissions
control.   For  example,  higher  incinerator  temperatures    more
thoroughly  destroy organic  compounds?  but  at those higher
temperatures certain metals volatilize more readily  creating the

-------
potential for  emissions  with greater metal  concentrations.
Increased nitrogen  oxide production  can  also r' result  at higher
temperatures.

     o    Until  about 1985, stack sampling of only very limited
scope   was   conducted  at   several   scrubber/fabric   filter
installations  in   Europe,   principally   for   emissions  of
particul'ates,  acid gases  (hydrochloric  acid  and  sulfur dioxide)
and certain metals.  These  studies  generated a narrow data base
of somewhat limited use,  since trace organic  compounds were often
not  studied  or, at  best,  only PCDD/FCDP  were.    Moreover,  the
operating conditions of  the incinerators and the  identities of
the pollution control devices  were often not  well documented, and
the  studies did  not  examine a range of different  operating
conditions.

     In 198S, Environment  Canada completed  an extensile testing
program providing the first thorough data base for  evaluating the
performance of these  control  systems  for  a  wide range  of
pollutants of concern.  Testing of  more limited scope conducted
in Denmark paralleled these efforts,  fhe results are  encouraging
and indicate that, at  appropriate temperatures,  scrubber/fabric
filter technology can significantly reduce not only particulates
and acid  gases, but  also a range  of trace organics (e.g., PCDD,
PCDF, chlorophenols, chlorobenzenes, PCB, and polycyclic aromatic
hydrocarbons),  and   a  host  of   metals   (including  cadmium,
chromium, lead  and  mercury).   Equipment  design  and operating
conditions necessary to achieve  high removal of these compounds
were identified in these studies on  a pilot scale.

     o    The  scrubber/fabric   filter  is currently  an effective
technology and  the data  base  is growing rapidly to substantiate
its- capability .to  reduce stack emission to  low  levels (in some
cases approaching the analytical detection limits for  compounds
such as  PCDD,  PCDF,  and certain metals).   This  conclusion does
not represent a Subcommittee endorsement that the scrubber/fabric
filter technology is the only one to use.  Other technologies may

-------
offer  equal  or even better performance  with less  associated
capital  cost.    Furthermore,  the  capability  for  continued
performance  at such  low emission levels under a  variety of
operating conditions remains  to  be demonstrated  for  full-scale
municipal solid waste incinerators*

Recommendations

     o IPA and  the  private  sector should  examine the long-term
performance  of air pollution control systems  under a variety of
operating conditions,

     o    EPA's  determination  of  "Best  Available  Control
Technology"  should be sufficiently  flexible to allow  adoption of
improvements  in control technologies,

D-   Ash Characterization and Disposal                 *

     o  The  concentration of various metals and organic compounds
in  ash is highly dependent on  whether it is bottom grate ash,
boiler hopper  ash,  or  ash  from  emission  control devices.  Most
compounds  of  concern  appear  to  become progressively  more
concentrated in the ash sampled  or removed  from the  flue gas
stream further downstream in the process.   Highly  efficient  stack
emissions control  systems  result  in  fly ash with  relatively
higher concentrations of  pollutants,  e.g. heavy metals and  trace
organics, since those substances  tend to concentrate  on the
smaller particles that  are more efficiently removed  by  these
systems.

     o   EPA  has  considered  requiring  compliance with  RCRA
Subtitle  C  if  ash  residues  from  municipal waste  combustion
contain wa«*e constituents  defined  as "hazardous11.  Alternatives
under consideration  include  regulating  municipal  incinerator ash
as non-haza-rdous waste.    Leachate tests on incinerator ash
conducted by EPA and other organizations have  identified lead and
cadmium  levels above the  Extraction  Procedure  (EP)  toxicity

-------
Units.    The  EP  test,  originally  developed  by  1PA  for
characterising the toxieity.of hazardous waste liquids, has not
been validated as  a  test  for municipal incinerator ash.

Recommendations

     o  State-of-the-art  analytical chemical techniques  should be
employed on  ash  samples, and as  many of  the  compounds  in the
extracts as feasible  should be identified  in order to  provide a
broad-scale data base,

     o EPA  should re-examine the appropriateness of using  the EP
test  or its successor,  the Toxicity characteristic  Leaching
Procedure (TCLP),  to assess the toxicity of municipal incinerator
ash*

     o EPA should  evaluate a number of alternative techniques for
managing ash disposal  from  municipal  incinerators.     These may
involve solidification or vitrification of the waste material, or
grouting  of disposal  trenches,  sometimes  in  combination with
liners*   The Subcommittee  recognizes  that these techniques may
need   to  be  modified   to  meet   the  particular  chemical
characteristics of  incinerator bottom ash and  fly  ash, although
the  experience  of  disposing of  fly  ash  from coal-fired  power
plants may have relevance.

£,   Environmental Transport and Fate

     o  The  atmospheric  transport and  fate of  emissions from
municipal solid waste  incinerators  involve  a broad spectrum of
physical and chemical processes.   The processes that need to be
addressed Include  stack emission phenomenon, including plume rise
and  downwash;  plume  chemistry,  involving  changes  of physical
state   and   chemical  reactions;  atmospheric  transport  and
diffusion?  gravitational  settling;  dry  deposition!  and wet
deposition  due  to  in-cloud and below-cloud  processes.   A
scientific  basis exists to support  model simulations  of the

-------
atmospheric: transport of  pollutants  provided  the emissions are
properly characterized along with the atmospheric and topographic
characteristics of each  site.   However,  considerable uncertainty
surrounds  the  ability to  properly  simulate  tooth  wet  and dry
deposition processes.

     o chemicals  which  are emitted  to  the atmosphere,  or are
deposited  on   soil  or   in   water,  undergo   a  variety  of
transformations.    Such  transformations can   result   in the
destruction of  the  parent  compounds  and  the  simultaneous
formation of one or more products.  Some of the  products may be
toxic.  The transformation may be photochemical, may .proceed in
the  dark,  or  may be mediated by  biological processes.   For
assessing potential effects,  the identity, quantity and  rate of
destruction of  the parent  chemical in various environmental  media
and the identity,  concentration, and  persistence  of  the  products
are  of great  importance.   Little  information   is presently
available on the fate of chemicals from MWC operations because of
the paucity of information on  the parent  compounds  released and
the  absence of  a research  program to  identify  and quantify
products  formed from  the  parent substances.   In  some  instances
where the parent compounds have been   identified,  scientists can
make reasonable'predictions of  fate based on published  studies.

RegpTOendations

     o  EPA  and  the private sector  should  develop  a  more
comprehensive   data base through   field studies   at   several
representative  MSW facilities.   The data base should  provide
information that can be used to estimate deposition (wet  and dry)
of  particulate and gaseous  emissions,  and  also  to evaluate
mathematical  and fluid  models  of  transport,   diffusion and
deposition in  urban  and suburbs  environments.    The data base
should include  measurements  of  MSW  emissions   (stack and
fugitive), plume rise, dispersion, wet and dry deposition.

-------
P*   As_s_ejs_sm.e.o.t. of Risk to Bateliq Health and .the Environment

     o The Subcommittee  concluded In a separate, previous report
that  the proposed EPA  methodology  for assessing risks  from
municipal  incinerators  through  multiple  environmental pathways
represents a considerable improvement over other multi-media risk
assessment methodologies previously developed  by EPA and reviewed
by  the  Science Advisory  Board  (See  Appendix A).    The current
methodology is  more  comprehensive  and,  in  general,  provides a
conceptual framework that should be expanded to other environmental
problems.  The Subcommittee identified areas  in this methodology
that need  further consideration  or  improvement,  including:  the
inappropriate  use of  the  Hampton  incinerator  facility  and
associated data to represent typical mass burn  technology;  the
failure  to  use  data  from  current best   available  control
technology facilities for model validation; separate treatment of
particulate and gaseous emissions and their fate, i.e. »downwash;
the  need to  use  best  available kinetics  in  predicting  soil
degradation? exposure resulting from the disposal  of  ash?  over
emphasizing the maximally  exposed individual  {Mil}  concept?  and"
the treatment of plant (and herbivore) exposure.
                           11

-------
                        It.  INTRODUCTION

A.   Charge and Scope of the Review

     At  the   request  of  the   Administrator   of  the  U.S.
Environmental  Protection Agency (EPA), the Science Advisory Board
(SAB) Executive Committee  agreed  on  April  23,  1986,  to   review a
number  o£ scientific issues  related to  the incineration  of
municipal  wastes.    The Executive  committee  assigned  the
responsibility for  conducting  the  review  to  its  Environmental
Effects, Transport and Fate Committee which, in turn, established
a Municipal Waste Combustion Subcommittee.

     The  subcommittee's  review   encompasses  current  municipal
waste  incineration  technologies/ the combustion process,  and
emissions to the  atmosphere, including associated  air  pollution
control equipment.   In  addition, it  covers  such issues  as  ash
disposal,  transport and  fate  of process residues,  and assessment
of  potential  effects on  human and  ecological  receptors.   The
Subcommittee   recommended   research  to   reduce  scientific
uncertainties  associated with incineration technologies.

     The Municipal Waste Combustion  Subcommittee reviewed several
separate documents  prepared by  EPA  on aspects of  the  municipal
waste  combustion  problem.    On November   10-11,  1986,  the
Subcommittee  reviewed  a  methodology jointly prepared  by  the
office  of  Air Quality  Planning  and  Standards  (OAQPS)  and  the
Environmental   Criteria  and  Assessment  Office (ECAQ)  entitled:
Methodology for  the Assessment  of Health Risks Associated with
Multiple Pathway Exposure  to Municipal Waste Combuator Emissions._
EPA  intends that  the  methodology  serve as  a  principal  technical
basis  for  its  decision on  whether  to  regulate  municipal
combustors. EPA was required  by a court settlement   *j  publish a
decision on this  issue  in the  Federal Register by July  2, 1987.
Because the Subcommittee desired  to  advise the Administrator in a
timely fashion, the  review of this methodology was  issued  as a
separate report on April i, 1987  (reprinted in Appendix A).

                          12

-------
     During  the  course  of  the  Subcommittee's  review,   the
Assistant Administrator for Air and Radiation, j,  Craig Potter,
requested that the Science Advisory Board   review  a  methodology
prepared  by EFA's Risk  Assessment  Forum entitled,   "Interim
Procedures  for Estimating Risk Associated with  Exposures  to
Mixtures  of  Chlorinated  Dibenzo-p-dioxins  and  Dibenzo-p-furans
(CDD and  CDF)".   The purpose of this methodology is to provide
EPA with a tool for risk assessment,  specifically to  address  the
toxicity  of  various congeners  of  CDD  and  CDF in relation  to
2,3,7,8-  TCDD.    The  SAB  Executive  Committee,  recognizing  the
relationship between  this procedure  and  the issues* undergoing
review  by   the  Municipal   Waste  Combustion   Subcommittee,
established a Dioxin  Toxic Equivalency  Methodology Subcommittee
to conduct a review of the former, and  provided for  overlapping
membership between the two Subcommittees.  This  enabled a  joint
consideration of  information pertinent to  assessing 
-------
included   the   evolution   of  municipal   waste   combustion
technologies, projected  scenarios  of  the  growth  in  demand for
municipal incinerators, problems  encountered  in  permitting
municipal waste combustors and the  potential of public health and
environmental risles resulting  from the use of  this  technology.
The Subcommittee made  site  visits  to operating incinerators; in
Hampton, Virginia on May 29-30, 1986, and in Baltimore, Maryland
on July 28-29,  1986.

1.   Major Assumptions and Limitations  of the Review

     Because of  the complexity  of the scientific issues under
review, the  data limitations  for many of these  issues,  and the
time constraints for  providing advice to EPA,  the Subcommittee
adopted  a   number  of  assumptions  and  recognized  several
limitations in defining its  charge.   They included  the following:
                                                       H
     Q  The   Subcommittee   considered  but  did  not  evaluate
information on alternatives to municipal  waste  combustion, such
as landfilling, recycling, and waste minimization.    Nor did it
assess  potential  risks  from  these  alternatives  in  a rigorous
manner.  In principle, the Subcommittee believes that MWC is one
of several acceptable  waste  management techniques.  However, it
recognizes that some degree  of risk or hazard is associated with
the application of any waste management technology.

     o The Subcommittee recommends that the potential effects of
municipal waste combustion be compared with those associated with
other  common combustion processes.  For example,  emissions from
coal-fired  or oil-fired  power  generators,  internal combustion
engines,  and  wood-burning stoves and  fireplaces should be
compared to  emissions  contributed  by waste combustors to  better
define no antecedents refers to what respective'contributions to
health and environmental risks. The Subcommittee did  not  compare
potential  emission  characteristics from the  various combustion
sources in common use.
                           14

-------
     o  The  Subcommittee  report    presents  an  evaluation of  a
series  of generic scientific issues that  policy, makers  at  the
national level  roust address.   This report  is 'not  designed to
evaluate all  the  issues,  such as  optimal  incinerator location,
that municipalities  typically  encounter as they  evaluate waste
combustion as a part of their local or regional  waste disposal
strategy.   The issues  that the  Subcommittee  reviewed-  for  the
purpose of advising  national policy makers  may not  be of equal
relevance- or importance  for making local decisions or site-*
specific assessments.

     This report  analyzes  issues  in an order  that  reflects  the
movement of  potential  pollutants   from  a specific  incinerator
source.  This analysis follows  the  process  from  combustion,
through emissions resulting from the  combustion process _ (either
directly through the stack, or  fugitive emissions),  through
environmental transport and fate  of emissions through  various
media (e.g., air, land  and  water)  and finally  through potential
human health and environmental effects.

     o  The  Subcommittee did not initiate or conduct any economic
analysis of  alternatives for  municipal waste  disposal  or
management.
                           15

-------
      III.  THE NATIONAL MUNICIPAL WASTE MANAGEMENT PROBLEM

     All societies produce municipal solid waste as  a  by-product
of  their industrial  activities  and  consumption  patterns.    In
general, the larger and more complex the society,  the greater and
more complex its municipal  wastes.   One of the key  technologies
available for managing the growing amounts of garbage or trash is
incineration.  This is  not  a new technology?  incinerator  plants
have  been in  use  in  both Europe  and  the  United  States  for
decades.

     EPA estimates that the United States generated approximately
126-159  million  tons of  municipal  solid waste  (MSW)  in  1980.
Only  about  6 million tons,  or approximately  4  percent of  such
wastes,  were incinerated in approximately  100  municipal  waste
combustors  (MWCs)  [1J»    In  comparison,   Sweden  currently
incinerates  approximately 50 percent of its municipal vaste and
Japan combusts approximately 70- percent.

     over 90 percent, or about 137 million tons of MSW  are buried
in  the United States each  year in  about 10,000  municipal,  and
privately operated sanitary landfills.   currently, SPA estimates
that  12.7 million tons/year of industrial solid waste is recycled
and recovered as  raw material for manufacturing.   There  is some
potential for waste reduction due to waste minimization efforts.

      In  the  past decade a number of intersecting events have com-
bined  to alter  the nation's  awareness,  and  the public  policy
framework, regarding municipal waste  management.   These includes
growing  amounts of municipal  waste to be  collected and disposed?
limitations  — such as the need for greater efforts by government
to  provide  technology  transfer  and  consumer  awareness and,  in
sou!'  areas,  economic disincentives  — in the current potential
for recycling waste  and reducing the  volume of waste  generated;
shrinking landfill  capacity  in many areas  of  the country;
escalating  costs  for transportation  and storage of  municipal
wastes?  and  stricter controls on landfills  increasing operating
                           16

-------
costs  and  owner's  legal  liability.     Public  health  and
environmental  concerns over  wasteVmanagement' alternatives,  such
as landfilling, incineration, and ocean dumping, have  also been
heightened,  as evidenced by  difficulties in  gaining public
acceptance to  new  landfill or incinerator sites and concerns over
the  potential  for  groundwater  contamination.    Through  the
Hazardous and Solid Waste  Act  Amendments of  1984,  congress
declared a national policy preference for more permanent methods
of disposal,  such as incineration,  over the  storage of wastes,
such as landfilling.  In  general,  these factors are stimulating a
wider reliance upon incineration technologies and are encouraging
expansion of this  industry.                          •

     The present municipal waste problem stems in large part from
the  fact  that, to date,  the nation  has  chosen  disposal methods
that favor   storage  of  wastes    over  methods  that  favor
destruction.   In the future,  waste minimiiation and recycling can
reduce the overall  volume of waste,  but ultimately there  is a
requirement for some form of  disposal. Municipal waste combustion
currently represents a technological alternative that  can reduce
the  volume  of waste by  over 90  percent.   In addition, it may
provide a source of energy recovery  under certa-in conditions.

     EPA  projects significant  growth  in the  use of  municipal
waste combustion  in  the United States between 1985  and the year
2000,   By that time the Agency  estimates  that  as many as 311
additional  MWCS  may be  in service with  a design  for  total
capacity  of about 252,000 tons of  SSW per  day.   This compares
with 1985  design  capacity  of  approximately  45,000   tons
incinerated per day in more  than 100  combustors  [1].   Table I
identifies currently operating  incinerators by design type.   Data
on facilities now  being planned or built suggest to EPA that MWCs
with a  design capacity of more  than 1,000  tons per day   will
constitute more than 50  percent  of the new  facilities built by
1990.
                           17

-------
     During the  course of  its review,  the  Subcommittee became
aware of the many changes under  way in combustion technology as
well as the   improvements in  emissions control technology.  The
Subcommittee is  also  aware  that  other related  waste disposal
technologies are under development or show  promise for  wider use,
given favorable  economics  and ease  of practice.    TWO of these
developing approaches—preprocessing and resource  recycling—have
been  considered  but  not  fully  evaluated  by  the  Subcommittee.
These approaches singly, or  in combination,  have the  capability
to  further minimize the generation  of potentially  hazardous
residues.  Minimizing solid wastes will reduce  the amount of land
needed for disposal  of waste,  and wi.ll increase the  potential for
returning materials to  the  economic cycle, potentially reducing
pressure on natural  resources.

     Ho matter which methods society uses to reduce and  dispose
of  municipal  wastes,  it will encounter some  degree  "of public
health or environmental risk.  ' In this respect,  waste disposal,
including   combustion,  is  no  different  than  most  other
technologies which serve our needs.  It is  important for  EPA  to
develop  the  means   for  and  to  undertake   comparative  risk
assessments 'across media for each waste management option.   Such
comparative analyses  would  provide a  basis  for  selecting among
the   different  options  and  would help to  identify  the option
presenting the least  adverse  risk.      State  and  local decision
makers  could also  utilize  this  technique  for  site-specific
assessments.  Furthermore,  comparative analysis would  facilitate
risk management,   taking economic,  societal,  and other factors
into account.

     It  is  necessary for individual  municipalities to evaluate
all available technological alternatives for waste  disposal, and
to choose the technology vies)  that presents acceptable levels of
risk  to  the  local  population and environment.    It  should  be
recognized  that  all  technological  alternatives  (including the
maintenance  of  the  status  quo)   impose  (voluntarily  or
involuntarily)  some form of risk.   The overall  societal objective

                           18

-------
is to select 'the most cost-effective technology  that imposes the

least adverse risk to the population and environment.

                             TABLE 1

             EXISTING MWC FACILITIES BY DESIGN TYPE*
DESIGN TYPE INSTALLED
DESIGN CAPACITY
fTONS/DAY)
MASS Burn13
With Heat Recovery
Without Heat Recovery
Total
MQPtflAR INCINERATOR0
With Heat Recovery
Without Heat Recovery
Total
EQE PROCESS*
With Heat Recovery
Without Heat Recovery
Total
GRAND TOTAL

20,900
9,800
30,700

3,300
500
3,800

10,700
0
10,700
45,200
NUMBER OF
FACILITIES

25
16
41
J.

33
16
49
M
"9
0
- 9
99
   aSource; Radian Corp*, [1]

   "Mass burn -  The burning  of unprocessed  MSW, typically  in
                      refractory or waterwall furnaces

   cModular incinerator -  Factory preassembled mass burn units
                    usually employing controlled air combustion
                    technology  to  incinerate  considerably lower
                    volumes of waste than  those  employed by mass
                    burn or IDF units

   %DF - Refuse derived fuel processes subject JfSW to varying
                    degrees  of processing to improve  fuel  quality
                    for better combustion efficiency and to achieve
                    some material recycling or recovery  (see Appendix D)
                           19

-------
   IV. THE PROCESS AND TECHNOLOGY OF INCINERATING MUNICIPAL WASTE
A.
Feedstock
     Municipal solid waste is extremely heterogeneous in nature,
and its composition is, in part, a  function  of consumption pat-
terns   that  differ   with  geographic  locations  and   vary
significantly with time of  the year.   There is a   substantial
data base  describing MSW  by  major constituents—paper, plastics,
glass,  wood,  cardboard and  ferrous  and nonferrous  metals (see
Table  2).  The  data contain not only proximate  and  ultimate
analysis, but also  chemical  analysis of ash.   This dnformation
can be useful  in  making  the  standard combustion  calculations,
including  combustion air  requirements,  inorganic stack  gas
emissions  (such  as  acid  gases and volatile  metals),  and bottom
ash characterization.
                             TABLE  2
         CURRENT AND PREDICTED COMPOSITION OP DISCARDED
    RESIDENTIAL AND COMMERCIAL SOLID WASTE  (WEIGHT PERCENT)a
Component
Paper and Paperboard
Yard Wastes
Food Wastes
Glass
Metals
Plastics
Wood
Textiles
Rubber and Leather
Miscellaneous

1980
33.6
18,2
9,2
11.3
10.3
6.0
3,9
2.3
3.3
1.9
Year
1990
38.3
17.0
7.7
8.8
9,4
8.3
3.7
2.2
2.5
'j: 1

2000
41.0
15,3
6.8
7.6
9.0
9.8
3.8
2.2
2.4
.2_iA
TOTAL
                                 100.0
                                       100.0
100.0
asources Radian Corp.,  [1]
                           20

-------
     Due to  the  inherent variability of MSW, and  shortcomings of
 current computer,  models  of  incinerator  combustion,   it  is
 difficult  to accurately  predict  the  composition  of  stack
 emissions,  -Accordingly,  it is important to consider the  impact
 of variation in MSW  composition  when  designing  furnace and
 emission/control  systems  in order to minimize solid and gaseous
 emissions.

     The  American  Society for  Testing Materials  (ASTM)  has
 classified municipal solid waste used as a fuel as Refuse Derived
 Fuel  (RDF-1,  RDF-2, R0F-3, EDF-4, or RDF-S)    based on the  degree
 of MSW processing  required.   Appendix  IV provides, a further
"description  of these categories.

 B»   The Incineration Process

     Organic materials that are completely combusted or*burned in
 an oxygen  atmosphere will theoretically produce water vapor and
 carbon  dioxide as gaseous, products  of combustion.  This assumes
 that the organic materials contain only carbon and hydrogen.

     Municipal solid waste, which is  usually  composed of 50-75
 percent organic materials,   is  a  fuel  that  contains many
 constituents other  than organic materials, such as  free moisture
 and inorganic materials  including  minerals  and  trace  metals.
 Thus, the products  of combustion, whether complete or incomplete,
 will leave the incinerator in various forms.   These forms include
 stack  emissions  as flue gas  and suspended particulates,  bottom
 ash falling  off  the grate  at  the  end of  the burning1 fuel bed, or
 fly ash removed by  pollution control devices.

     Complete' combustion  depends on temperature, turbulence and
 residence  time.   The temperature required for proper  combustion
 varies  with the "raw material.    The turbulence  required in the
 furnace to  ^achieve  the  proper  mixing  of  combustion  air and
 product gases evolved  from  burning  materials also  varies and
 influences  efficient -combustion.   Similarly,  the amount of time

                            21

-------
needed for Materials to  fully  combust depends on the  elemental
and  physical  characteristics  of  the  feedstock  and has  an
influence on combustion efficiency.

     Thermodynamic  properties  of  chemical  constituents  in MSW
indicate that, tinder excess air conditions  and the  temperatures
typical of incinerators,  emissions of organic compounds  should be
so low as  to be considered zero.   However, field sampling data
show significant emissions  of trace organic compounds.

     Organic compounds, which include hydrocarbons,  can  be  formed
during MSW combustion,   some  of these  hydrocarbons' may  raise
toxicological concerns  or may be  precursors to potentially toxic
compounds.   The heterogeneous  characteristics . of  the  fuel can
prevent complete and  uniform mixing of volatile gases and thereby
prevent complete combustion.   Fuel-rich  pockets  develop   in the
furnace  leading to  hydrocarbon  formation,  Chemical kinetic
considerations  indicate  that these hydrocarbons should  be
destroyed rapidly  in  the  presence of  oxygen  at  elevated
temperatures,

     fhe  objective  of  the combustion  control  process  is  to
provide  for  effective mixing of the fuel  with  oxygen at a
temperature sufficiently  high and for a time sufficiently long to
promote the  destruction  of all  organic  species.  Thus, organic
emissions can be eliminated or  reduced to minimal amounts  by the
proper  implementation of  combustion control, which  includes
efficient  furnace  design,  sufficient  instrumentation  for
combustion air control  and proper unit operation.

C.   Qescriptiona pj£  Combustion Systems

     Increased competition  in the growing incinerator market is a
prime motivation for continued improvements in the  engineering
design of  incinerators,  especially in the combustion process.
Soth  competitive  pressures and  concerns over environmental
                           22

-------
performance are forces leading to  improvements  in the design of
municipal waste  combustors  and emissions  control  equipment.
Major  municipal  waste combustion  systems can  toe  grouped  into
several categories:

1.   Mass Burning of, Unprocessed  Municipal Waste fRDF-l)

     Mass burning  usually  implies  an incinerator that employs a
waterwall furnace enclosure positioned over the combustor  grate.
The flue gases that are products of combustion leave the furnace
then flow through a convective (back-pass)  heat recovery  boiler
(see Figure  1}» _   Older  systems  aay also consist of' refractory
furnace  walls  combined  with  a  convective  (back-pass)  heat
recovery boiler*

     Early mass burning  units  introduced the waste  into the
furnace  and  onto  the grate  by  gravity  through  a  feed  chute.
Newer  units  utilize hydraulic 'rams  to  meter the  fuel onto the
grate.  Grate designs use some form of fuel bed  agitation through
reciprocating, oscillatory or rotary motion  or some  combination
of these movements.   This  bed agitation allows for more uniform
burning and 'maximum butmup.   Grate area is designed to maximize
the heat release rate.

     In  such units' combustion  air  is  introduced as  undergrate
(primary) air and as  ov'erfire (secondary) air.   Overfire  air is
introduced via'  nozzles  positioned in  the  front,  rear, and
sidewalls of the  furnace over sections of the grate.   Excess air
levels in such units  usually  range from approximately  80 percent
for waterwall plants  to  ISO percent or more for  refractory wall
units.   Flue gases exiting the furnace  usually  pass through a
convection heat transfer boiler.   the non-combustible matter in
the fuel,,  along  with unbumed carbon,  fall  off  the  end of the
grate as bottom (hopper)  ash, or will be carried up as fly  ash in
the flue gases passing through the burning fuel bed.   Bottom ash
is the residue remaining after nearly complete combustion  of the
organic matter achieved in  current design and operation.   Bottom

                           33

-------
                      Figure  1
        MASS BURNING INCINERATION
   WATER WALL-*-
   MEMBRANE
    FEED
   CHUTE
            COMBUSTION   FLUE
             CHAMBER  —» GAS
        PRIMARY
      AIR SYSTEM
 RAM
FEEDER
                    PRIMARY
                   AIR SYSTEM
                  24

-------
ash usually drops off  Into a water-filled hopper  (for quenching)
and is usually transported to a landfill for disposal (see Figure
2) ,

2.    Modular/Starved Air BiarjaJLtig of Unprocessed  Municipal Waste
fRDF-11

     In  small-scale facilities,  starved air  or  controlled  air
combustors use  two-stage or double  combustion chambers  (see
'Figure   3) .   Usually,  sub-stoichometrie air is  supplied  to  the
primary refractory  lined chamber  to control  exit  temperatures of
gases and to reduce particulate entrainment  by  the  flue  gas from
the burning  bed. A  variation  in  the system design of  the  two-
stage  combustor  is  known  as  a eontrolled-air  incinerator.   In
this  design,  excess combustion air is  supplied  to  both primary
and secondary  chambers*   To  minimize  fly  ash  carry-over,  the
excess air in the primary chamber is relatively- low.  la either a
starved  air  or  an  excess  air .unit,  a heat recovery boiler  is
located downstream,  followed  by appropriate  equipment  for
particulate removal.

3.    Dedicated Stofcer. Boilers Burning Coarsel' Processed Refuse
        -2  may  be combusted in  "conventional"  stoker  fired
boilers which  consist of  a waterwall  furnace  and a  convective
back-pass heat recovery boiler (See Figure 4) .   Fuel  is injected
into  the  furnace by  air  swept spouts  (or essentially pneumatic
injection) .  Traveling grates drop the  bed  ash into  hoppers  as
they  move towards the front wall of the boiler.   Optimum amounts
of  excess  air  range  from 70-§0  percent.    Several  levels  of
overfire air nozzles  are  normally positioned above the  grate  in
the front and back waterwalls.  These nozzles induce  turbulence,
providing the necessary mixing of partially combusted flue gas as
it exits the • grate bed*   New units are  also  being designed with
arches  located  in several  of  the waterwalls to  promote further

-------
                     Figure 2
Refuse
Drying
Grate
            Burning Grate
                   MSW Grate
                   26

-------
             Figure  3
CONTROLLED AIR
SECONDARY
SR>100%
                       WASTE HEAT
                       RECOVERY
    HYDRA
      CHARGING
                            CONTROLLED AIR TO
                            PRIMARY CHAMBER
      Starved Air Combustors
               27

-------
             Figure 4
              Refuse Bred Baiter
Trailer Storage Area
                             Electrostatic
                             PrecMlst
                         ss-      \
                             Ash Hopper
      RDF-Fired Combustors
                28

-------
turbulent mixing.   Some volatile  matter in the  RDF-2 undergoes
drying and ignition while in suspension.
4 .   Co-firing of SgaJL and Municipal Solid Sasjse Burning
     Processed Refuse (RDF-3 ,.. EDF-4 *. and RPF-51

     The practice of co-firing MSW with  coal  involves the use of
processed fuel such as EDF-3, RDF-4,  or  RDF-5.   Co-firing may be
accomplished either in  a spreader stoJcer  or  in a  utility steam
generator co-fired with pulverized coal.

     Fluff RDF (RDF-3)  is pneumatically injected into the furnace
of stoker units at firing rates  of up to 50 percent BTU of total
heat  input  from  fuel,  or  up  to 20  percent BTU  heat  input in
pulverized coal units.
D,   Cambustion System Design ajjd. Operation

1.   Stages of Combus.t_o_r; Operation, Old and Jfgw. giant Designs

     As previously stated,  completeness  of combustion depends on
oxygen  supply,  time,  temperature, and  turbulence.   sufficient
temperature  and  residence time  are  required for  the  fuel  to
undergo complete  oxidation.   Proper amounts of excess air aid- in
developing  necessary  furnace  flue  gas  temperatures  and
turbulence.    .          s

     Grate  combustion of  municipal solid  waste takes  place in
three often overlapping stages.   Multi-sectioned grates are used
to accomplish these- steps in both  American and European designs
for  mass  burn  systems.  These  three  stages are  illustrated in
Figure 1 and are  described below?

     o  Drying-Volatilization;  As  the waste is heated, moisture
and volatile matter  is released,  leaving a carbonaceous residue.
The combustible content of  the  volatiles burns partially  within

                           29

-------
the refuse bed and  partially  in suspension over the  grate (see
"refuse drying/  Figure 1),

     o  Fixed Carbon Combustion:   The carbon residue produced by
the devolatilization of the waste burns on the grate,  leaving an
inorganic residue (see "burning grate", Figure l).

     o  Final Ash Burnouts   Additional tine on the  burnout grate
is required to reduce the remnants of carbon embedded in the ash
to an acceptable  ( <5 percent  ) level  (see  "burnout grate, Figure
1).

     In older units,  undergrate  combustion air was  supplied  by
use of a single damper"Controlled compartment  (wind-box) for each
section  of the  grate system.   In  the  past,  the  incinerator
operator often depended on visual inspection to achieve a "good-
looking" fire, with  the  hope of maintaining  "good"  combustion.
Modern designs use sectionalized undergrate' air compartments  to
supply varying  amounts of  primary  combustion air to different
areas of the grate.  Automatic controls on the combustion system
integrate signals from CO,  Oj*  or COj/ waste  feed  and/or steam
production,' as  well  as  combustion  air  control  to  produce
optimized burnout of the  products of  combustion.

     Turbulent nixing of the flue gas  leaving  the  grate in mass
burners or dedicated boilers is essential  and can be obtained by
appropriate location of   the overfire  air  nozzles.   Proper
combustion   control   yields    the    correct    ratio    of
undergrate/overfire  air.   This  ratio provides good  flue gas
mixing at  a  constant  excess air  setting.   Too little combustion
air  will  result  in  the  generation  of  products  of  incomplete
combustion including soot, while  too much  overfire  air will
result in a quenching of  the combustion process causing  fon-^tion
of the products of incomplete combustion.   Sometimes the emission
of  a white  smoke will  result  from  the condensation  of the
          unburned hydrocarbons.
                           30

-------
     Earlier  designs  for  mass  burners  did  not  incorporate
flexibility in introducing combustion air, or the  sophistication
of combustion air control,  used  in newer plant designs.  Thus,
older  plants  generally did  not  achieve  the  efficiency of
combustion attainable in modern plants.

     Furnace designs feature arches over  the grate in the front
and rear walla to allow -for faster drying  and distillation of the
fuel volatile  matter,  in fuel and  also  improve  the  mixing of
stratified flue gases to permit more complete combustion and ash
buraup  (See Figure 1}.  Plugging or jamming  of  the drag conveyer
for bottom  ash has been  the  source of some  boiler  load upsets
resulting In the release'of products of incomplete combustion to
the environment.

2»   Emissions from the Combustion Chamber
                                                      i»
     Data  on  the  identity  and  concentration  of  different
pollutants  emitted  from  the  combustion  chamber  provide useful
information in  designing air pollution  control  devices, but  such
data are relatively  scarce,   fhe following information pertains
primarily to mass-burn incinerators and  is provided to illustrate
the relationship among emissions,  feed composition, and combustor
design and operation.

a.   Acid Gases

     HCI and S02 are produced as a result of chlorine and sulfur
containing materials in  the  feedstock.  Approximately 60 percent
of the chlorine in the waste, ends up as  HCI;  the remainder occurs
primarily in the solid residue  as inorganic  chloride and may be
combined with  trace amounts of gaseous organic compounds.   The
sulfur if* the  feedstock oxidiies to SO2.  Part of. the SQ2  will
react  further, ".such  'as with alkali  in the waste,  to   font
sulfates.   The  ash  residues  may retain  from 10 to 90 'percent cf
the sulfur  depending upon the alkali and sulfur  content of the
                           31

-------
waste.   The remainder of  the sulfur will  be emitted  with the
combustion products,  primarily as  SQ2.

     The oxidation of nitrogen, present either in  the air or in
organic compounds  in the feed, produces  nitrogen  oxides  (NOX).
The organically-bound nitrogen, found in  high concentrations in
proteins  and  some  plastics,  can  be  converted  to NOX with
efficiencies of  up to 50  percent.   The  NOX produced  from the
oxidation of  atmospheric nitrogen  in  combustion  air increases
markedly with  increases  in combustion temperature and  strongly
depends upon combustion conditions.
b«   Fly, ash. and Residues

     Unlike  organic  compounds,  elemental  or non-combustible
materials are not destroyed during the incineration process*  The
composition of feedstock or incoming wastes, therefore, provides
a measure of the total inorganic residue.  Most of the  inorganic
residue  and  the  products  of  incomplete combustion  of organic
compounds leave  the  furnace  as either  fly  ash or  bottom ash.
Bottom ash drops  off  the end  of   the grate and  is  conveyed to
hoppers,  while  fly ash  is  elutriated with  the  flue gas  to be
collected  by  air  pollution  control  devices  or  emitted  as
particulate  out  of the   stack.  ' The distribution  of elements
between bottom ash and fly ash carried over to the air  pollution
control  device(s)  depends  upon the design and  operation  of the
incinerator and the composition of the feedstock.   The  amount of
ash carried out with  the flue  gases leaving a burning refuse bed
increases with increasing  underfire  air  and with bed agitation.
For this reason, starved air incinerators with low underfire air
flow tend to have less  particulate  emissions  than  conventional
mass-burn  units. The  amount of  fly  ash  carried   from  the
combustion chamber will be influenced by  the  particle size  of the
inorganic content of the MSW,
                           32

-------
     The  distribution  of  elements   between  the  different
components  of  refuse  has  a  strong  influence  on  the  their
environmental fate.  For example,  TiO2 used as a pigment in paper
products has a particle size of about 0.2 urn and will be carried
off by  the  flue gases  passing through the  refuse  bed,  whereas
Ti02  in glass  will  accumulate  in  the  bottom ash.   Up  to  20
percent of the total_inorganic content of the waste will.usually
be entrained in the flue  gas causing the  burning refuse bed to
form  fly "ash  particles in  the  1  to  20  UM  size  range.    The
remainder will end up  in the bottom  ash,

c.   Trace Metals  ^                                  >

     Volatile  elements  and their compounds, usually  present  in
trace amounts  in the  feed, will vaporize from the  refuse and
condense in the cooler portions of a furnace.   They will  condense
either  as  ultra fine  aerosol  (less than  1  uk size) 
-------
the  grate  that may overwhelm  the  local  air supply.    The
capability of  modulating the  air  supply  and  changing  the
distribution of air can control the effects  of such sudden surges
in heat release.   Mixing  the  MSW in the receiving or collection
pit may also help control  surges of heat release  in the  furnace.

     As combustion takes place, polycyclic  aromatic hydrocarbons
(PAH) are formed during the fuel rich combustion  of gas, oil, and
coal, as a consequence of free radical chemical reactions in the
high temperature flame.  Quenching of partially combusted fuel due
to  interaction  with  cooled surfaces  is another  PAH  formation
mechanism  that  occurs with internal  combustion  engines,  diesel
engines   and   oil-fired   home  heating   furnaces.   In   such
circumstances a high fraction  of the  polycyclic  compounds are
oxygenated.  Upset conditions  leading to local air deficiency may
also result in the emission of organic compounds  such  as PAH.
                                                      i«
     One hypothesis  deserving further  analysis  is  that similar
free radical reactions take place in fuel rich zones of  incinera-
tor  flames yielding PAH,  oxygenated  compounds  such  as phenols,
dioxins and  furans and,  in the presence of chlorine, some PCD®
and  PCDF.   This hypothesis is supported fay  the observation of
PCDF in  the combustion products  of  pine wood only when  it had
absorbed HC1.  The argument for the high temperature synthesis of
PCDD and PCDF is also supported by the  demonstrated  increase in
the  concentration of the pollutants across  a heat  recovery
boiler.

     The   above free  radical  mechanism  should be  further
investigated to determine  if  it  is  the dominant source of PCDD
and PCDF in incinerators.   These compounds may also be present as
contaminants in a number  of   chemicals,  therefore  they may  be
present 'tu MWC•feedstock*   The presence of chlorinated phenols,
and  polychlorinated  biphenyls  (PCI)  may result  from  the  use of
these chemicals (uses that have been discontinued in  some cases)
as fungicides and bactericides (phenol derivatives),  and as heat
exchanger and capacitor fluids (PCS)  contaminated with low levels

                           34

-------
of PCDF.  These  compounds may persist beyond  combustion  only if
process gases are cooled to  temperatures below,;.those required, for
their decomposition and reaction by large excesses  in local air
flow.

     PCDD and PCDF may also  be produced by condensation reactions
involving the chlorinated phenols  and biphenyls.    The observed
formation of PCDD when fly ash from MSW incinerators is heated to
250°-300°"C  suggests  such catalyzed condensation reactions of
chlorinated phenols.  PCS can be a precursor  to PCDF? pyrolysis
tests with  PCB  in laboratory reactors  at elevated -temperatures
have yielded PCDF.                                   *
2.   Operator Training ^nfl Certification
     The  proceeding  sections  underline  the  importance  of
controlling combustion  air flow  rates,   air  distribution,  and
furnace  operating  temperatures for minimizing  emissions.     To
minimize the potential  for hazardous  emissions,  facilities must
be operated  properly.  The proper , operation  of MWCs  requires  a
thorough understanding of the  complexities  of incineration,
including  knowledge  of the  composition  and variability of  the
feedstock, the  fundamentals  of  the combustion  process,  and
requirements and consequences of adequate emission controls. .- In
addition,  operators  must  be trained  in  procedures  for managing
upset conditions in  order  to prevent  or  mitigate the release of
hazardous compounds,

     The  combustion of  municipal  solid  wastes  at  resource
recovery facilities is exempt from Subtitle C requirements of the
Resource Conservation and Recovery Act (RCRA) , providing that the
owners or  operators  assure the permitting authorities  that  the
burning  of hazardous  wastes  will be prevented.   Due to  this
exemption, -no  national  policy  on  operator  training and/or
certification for MWC facility  operators  has developed.
                           35

-------
     Since many  new resource  recovery  facilities are,  or will
soon be, under construction, there is an urgent need for approp-
riately trained,  and technically qualified operators.  Incinera-
tors  in  the planning  and/or construction stages  are generally
larger  and much  more  complex  than existing  facilities.    The
performance of these newer plants is becoming  more  critical in
view of   the increasing complexity  of  regulatory requirements,
the  requirement  for increased  efficiency of  pollutant  control
technologies in  newly  permitted  facilities  and  the heightened
public concern for environmentally safe  disposal of residue from
the combustion process.
                                                     i
       There  is  no existing pool  of  trained  plant  operating
personnel that private industry or  municipalities  can draw upon
to staff  MSW plants.   Some states  have promulgated regulations
requiring plants  to be  operated by certified personnel, but these
states  do  not  have  formal  training  programs  leading  to
certification.  At present, training courses for plant operating
personnel are available only to  a limited extent, and   moat of
these are one week general  training  programs.   Such programs do
not  provide the necessary  understanding of  the concepts and
details of combustion system design  and  operation,  and  emissions
control.   Vendors  must deal  with the normal problems  of plant
startup while  providing   extensive on-the-job  training  for
personnel who are basically unfamiliar with the  facilities.
F.   Conclusions and Recgmnendations

1.   Conclua iona

     o   The design and  operation  of combustion  chambers  has a
major influence on the type  an4 concentration of the pollutants
entering air pollution control devices.   In well-designed, well-
operated incinerators  with  state-of-the-art  systems  for  air
pollution control, the emissions of organic compounds of concern
                           36

-------
can  be  reduced  to  levels  close  to  the limits  detectable by
currently available  sampling and analytical methods.

     o   When  high  concentrations  of organic compounds are
emitted,  it  is  usually  a  consequence of  poor  mixing of
combustible volatiles generated in the burning refuse bed  with
air, or  from  quenching of  the  partially combusted products by
excess  quantities  of  air or  contact  with  cold  surfaces.
Inadequacies of design and/or operation of overfire air jets or
underfire  air  compartments   may  result  in  the  improper
distribution of  air  causing  inefficient mixing  and  quenching
before volatiles  are combusted.

     o  Extensive field testing has  been  conducted to  establish
general  emission  concentrations.    There are fewer  data on
systematic variation of operating and  design parameters to
provide insight into the mechanisms governing organic .emissions.
[2, 3, 4]                     ,

     o  The wide  variety of  polycyclic aromatic  compounds and the
large number  of  congeners  of  PCDO  and  PCDF   observed  in the
emissions  from incinerators appear  to be consistent  with the
pyrosynthesis  of these  compounds  in  the high  temperature  flame
zone.

     o   Feedstock composition has  an  important  impact  on the
emissions of inorganic'compounds.  Chlorine,  sulfur, and volatile
trace metals will be transferred with relatively high efficiency
to the gaseous and  fine particulate matter carried  out of the
combustion  chamber.   In  addition,  particulate matter  will be
carried from the combustor in amounts that will depend upon the
fineness  of  the  mineral  constituents in  the  refuse,  bed
agitation, and .the underfire air flow rate.'

     o   Municipal waste  combustion  is a  complex process  that
depends  on many factors . that begin  with  initial  feedstock
variability and end with emissions control.   Technologies  under

                          37

-------
development will  add  to this  complexity.  Operators are  seldom
trained to  operate  existing or new  incinerators,  nor are  they
required to be certified for incinerator operation.   The  lack of
trained operators may  slow  the application  of MWC technology and
may compromise efficient and safe plant operation.
2.   Recommendations

     There is a need for better understanding of the relationship
of  inorganic  and  organic  emissions,  and PCDD and  PCDF  in
particular,   to  MSW composition,   furnace design,  and' operating
conditions.    This requires research  on  full-*,  pilot-,  and
laboratory-scale  units.   Well-planned field  testing  under
different operating  conditions  will  generate  a  more realistic
correlation of emissions to operations,  in addition to providing
data  for  establishing emission  indices.   Pilot-plant  and
                                                      tt>
laboratory-scale  testing  can  be  used  to  critically  test
hypotheses on the  routes and mechanisms  of  pollutant formation,
because  of  the  ability   to independently  vary  operating  and
design parameters and feed composition in small-scale equipment.
The following specific tasks  need  to be undertaken:

     o The  relationship between undLerfire  and  overfire  air
distribution and  emissions  needs  to  be  understood in order  to
establish guidelines for adjustments in  air flow  rates that are
responsive to  changes  in  MSW composition  and  feed rate.    A
complementary study  is needed on  the  emission  of  combustible
volatiles  from  burning refuse   beds since  the  overfire  air
distribution should be  matched to the evolution  of  combustible
volatiles.    The  effect of  transient  operation is of particular
interest.

     o  The kinetics of pyrosynthesis and condensation reactions
as  they  relate  to the  formation of  PCDD  and  PCDF should  be
further investigated.   An understanding of the factors governing
the distribution of  congeners and isomers  of  these compounds

                           38

-------
would be  useful  for assessing health  effects  and as  an  aid in
diagnosing the  genesis of .emissions.    For  example,    -the
relationship of  the composition of MSW to emissions  should be
studied   over  a wide enough temperature  range to be  useful in
testing the various hypotheses  for  formation of PCOD and
     o    EPA and the private sector should worlc cooperatively to
develop  continuous monitors to  detect  upsets  in  operating
conditions.  Carbon monoxide and  total  hydrocarbons are currently
being explored as potential  indicators  of  emissions  of PCDD and
PCDP,   Alternatives,  such  as  polycyclic aromaties,   may be
appropriate surrogates.                              ,

     o  Private industry and Federal, State and local governments
should initiate efforts to plan and implement an operator train-
ing program leading  to certification.   This plan should  provide
the operator  with a basic  understanding of  the combustion  pro-
cess, management  of plant equipment,  and  impact  of operational
parameters on environmental emissions.   EPA and state authorities
should provide guidelines to facilitate operator training and to
maximize assurance that hazardous materials will not be burned in
MWC.  Certification should be valid nationally and transferable
from state to state.   Implementation of this recommendation  will
provide a  reservoir of appropriately trained personnel to staff
the increasing number of MWCs.

-------
      V.  PERFORMANCE OF AIR POLLUTION CONTROL TECHNOLOGIES

A.'   Pot.enfcial Air Pollutant a of Concern

     As outlined in the previous chapter, the  combustion  process
results in the generation of  flue gases and  particulates which
contain various pollutants.   These can  be  grouped into  several
categories;

               particulates
          -    Heavy Metals
          -    Acid gases
               Trace organics

     To prevent or reduce emission of  these compounds into  the
atmosphere,  various  air  pollution  control systems  can  be
installed between  the incinerator/boiler and  the  stack.   These
are discussed below.

     For some  pollutants, there appear to be  trade-offs  between
combustor design,  unit operation   and emission  controls.    For
example,  higher  incinerator temperatures  can  destroy trace
organic compounds, but also  cause an  increase  in NQ^  production.
In addition, metals like mercury volatilize more readily  and  are
carried  from the  combustor to  the  control  devices  in  greater
amounts at higher incinerator temperatures.

B.   Description of Air Pollution Control Systems

     The following  are the  main types  of air pollution  control
systems  or  devices  that can be  installed  on  municipal waste
combustor3:

1.   Eleetrp-itatig Preeipitators fESP)

     Electrostatic precipitators have demonstrated capability to
remove particulate matter  but do not remove  gaseous  pollutants.

                           40

-------
They can be  used alone,  linked in a series  of  2 - 5  fields or
linked with  other pollution control devices such  as  scrubbers.
The  precipitation process  follows  these basic steps:  (a)
development of  a current of  negative  ions from a  high voltage
corona discharge to charge dust particles in the gas stream; (b)
the presence  of an electric  field  in  the gas  space between the
high voltage discharge electrode wires  and the  collection plate
that propels the negatively charged  particulate matter toward the
positive  collection  plate? and  (c)  removal of the collected
particulate matter into  hoppers  by use  of  a rapping  mechanism.
Figure  5  illustrates the basic principles of electrostatic
precipitation.

     Electrostatic precipitation  occurs within  an enclosed cham-
ber.   A  high  voltage transformer  and  a rectifier modify  the
electrical power input.    Suspended within  the  chamber  are  the
grounded collection electrodes  (metal  plates)  connected  to  the
grounded steel framework of the supporting structure.   Suspended
between the  collection  plates  are the high voltage  discharge
(wire)  electrodes  (corona  electrodes)  insulated  from  ground and
negatively charged with voltages  ranging from 20 JcV to loo JCVDC,

     The last step of  this process involves dust removal from the
collection electrodes*   In  dry  ESFs,  this is  accomplished by
periodic  striking of  the  collection  plates and  discharge
electrode with a rapping device.  Hoppers collect the fly ash and
it is conveyed .to storage or disposal points.

     In North America, electrostatic precipitators  have  tradi-
tionally been  used alone  for  particulate control.    In Europe,
several installations use a  scrubber    in  combination  with an
electrostatic precipitator.   The physical arrangement  of  a
typical  electrostatic  precipitator   having  two  independent
electrical fields is illustrated  in  Figure 6.
                           41

-------
                      Figure  5
Earthed Collector
Eectrode at _
Posttive Polarity
Gas Flow
  Uncharged Particles
                                      Charged
                                      Partldt
                               Partfcte» Attracted
                               to Collector Sectrode,
                               Formir»fl a Dust Layer
                                                 Ctean Gas Exit
                                              Discharge Electrode at
                                               Negative Polarity
     Electrostatic Precipitation Process
                         42

-------
                       Figure 6
   Insditor Compartment
High Votttge System
Upper Support Rama
    Casing
                                           Trarafomer/Recfiftaf
                                               Reactor
Efectricai Equipnie
  Flatfonn
                                              Cotectlig Surfacos
                                             Cdecting Svface Rappe
                                          Hopper
            I
             F»fc3 ««*
         Arrangement of Electrostatic
                   Precipitators
                      43

-------
2.   Fabric. Filters

     Operation  of  this  technology  involves  impaction  of  dust
particles on  a fabric filter  to  form a  dust  cake on the  cloth
surface, with filtration of the gas as it passes through the cake
and  cloth.    The  fabric  typically  used is  a  woven  or  felted
material.   The  dust  cake formed  on the  filter  plays  a  key role
in   the  overall   efficiency   of   particulate   collection.
Periodically the dust cake is removed from the filter surface via
a cleaning cycle that may consist of shaking the.bag, reverse air
cleaning or blow-back by compressed air via pulse jets.  The cake
remaining after cleaning forms a base for collection of particles
as the bag is put back on line.

     The type of cleaning cycle used is,a factor in distinguish-
ing the different  designs of fabric  filter  type dust collectors.
Figure  7  illustrates  a small  pulse-jet cleaning  fabric filter.
Fabric filters have not been used  alone  on  MSW  incinerators,  but
are used in combination with lime  injection scrubbers, described
below.

3.   Scrubbers

     Three widely  used types of scrubbers exist.   These include
wet, dry, and wet-dry scrubbers.   A  wet  scrubber can be designed
with  several  different  configurations,  but they have in  common
an underlying principle of intimate  contact of  a gas stream with
a liquid that may  also contain some  absorbent and/or reagent  for
removal of  acid gases.   Although some  wet scrubbers have been
installed in the past, typically on older incinerators, these are
not likely to be used in the future due to several disadvantages,
including the  generation of a liguid waste  effluent  and  a  wet
plume from the stack.
                           44

-------
                       Figure  7
    Reduced Flow
                                         — Housing
                         To Clear Air Outlet
                          and Exhauster
Dirty Air Inlet and
    Diffuser
                                              Filter Tube
                                        Hopper
                                     Rotary Air Lock
                     Fabric Filter
                        45

-------
     Dry  scrubbers  are typically cylindrical  vessels  where
powdered  dry  sorbent  is  injected  into  the  gas  stream  by
compressed air.  Intimate mixing of the  sorbent  and gas occurs,
then  dry    gases   flow  into  a  highly  efficiency device  for
particulate removal,  such as a fabric filter or an electrostatic
precipitator.   For temperature  control,  a dry  scrubber is often
preceded toy a heat exchange  system which  may also  involve a water
spray system to cool the gases.  Figure 8 depicts a dry scrubber
system with water sprays and fabric filter.

     A wet-dry scrubber is also called a spray dryer or semi-dry
scrubber,  or  even a  dry scrubber.   In a  wet-dry scrubber   a
liquid sorbent stream is sprayed into  a gas stream and the amount
of  liquid  is carefully controlled  so that all  the  liquid
evaporates into the gas  stream, yielding a  dry fly ash product.
A high  efficiency  particulate removal device, such  as a fabric
filter or  an  electrostatic  precipitator, is  required -to remove'
the particulates from  the gas -stream prior to discharge up the
stack.  Figure § illustrates a wet-dry  scrubbing system with a
fabric filter.
C.   Historical Perspective of Ajyc pollution Control  for
      Incinerators

     The  air pollution control  systems used  to reduce stack
emissions from municipal solid waste incinerators are undergoing
continued design  improvement.   In post-19SO North America, two-
field electrostatic precipitators  were  succeeded  by three, four
and,  more  recently,   five  fields  for  enhanced  removal  of
particulate matter from flue gas.

     Beginning in the  late 1970s, several ai- pollution  control
systems, consisting of  a combination of a dry scrubber or a wet
dry  scrubber  followed  by  either  a  fabric  filter  or  an
electrostatic precipitator, were  installed  in Europe and Japan.
It  appears  that facilities adopting  this  technology  initially

                           46

-------
     Figure 8
            9
Wet-Dry Scrubber
      47

-------
 Figure 9
Dry Scrubber
   48

-------
sought improved acid gas control*   As concerns over trace organic
compounds and  toxic metals  emerged  during the  past few  years,
questions arose  regarding the capability  of  this equipment  for
removing these substances, as well as acid gases*

     Up  to   1985,  limited  sampling  occurred  at  several
scrubber/fabric  filter  installations 'in Europe, principally  for
emissions of particulates,  acid  gases  (hydrochloric  acid  and
sulfur dioxide}  and certain metals.   these studies generated  a
narrow data  base of somewhat, limited use, since the analysis of
trace organic compounds was  often omitted or, at  best,  confined
to  PCDD/FCDF or even  UCDD/TCDF.    In addition,  the  operating
conditions of the incinerators and pollution control devices were
often not  well  documented,  and the  studies did  not  examine  a
range of different operating conditions.

     In 1985, Environment Canada completed extensive testing  on
a pilot-scale unit with pollution control  equipment.   This
testing resulted in  the  first thorough data base  for  evaluating
the performance  of  these control systems  for a  wide range of
pollutants [4].  Testing  of  a more limited scope in Denmark  [5]
paralleled these efforts.  The results of these tests  indicate
that,  at appropriate  temperatures,-  the  scrubber/fabric  filter
technology can  significantly  reduce not   only particulates  and
acid  gases,  but also  a  range of  trace   organics  (FCDD,  PCDP,
chlorophenols,  chlorobenzenes,   £CB,   polycyclic   aromatic
hydrocarbons), and a host of metals (including mercury,  chromium,
cadmium, and lead)   in the stack emissions.  Equipment  design  and
operating conditions necessary to  achieve high removal  of these
compounds were identified on a pilot scale in these studies.

     The first, full  scale scrubber/fabric  filter installation on
a. waste-to-energy incinerator in North America was tested and  the
stack data  for SCDD/PCDF [6]  show comparable concentrations to
the emissions  found in the pilot-scale studies  discussed  above.
Several  municipal   solid  waste  incinerator facilities  are  now
                           49

-------
operating in North American with this  type  of pollution control
equipment*

     The scrubber/electrostatic  precipitator combination has been
installed at several  incinerators  in Europe,    However,  data to
evaluate this combination and compare its performance to that of
the  scrubber/ fabric  filter are  currently limited  to inorganic
compounds.    There is  a reluctance on  the part  of regulatory
agencies to  permit  such facilities because of  the limited test
data  on  scrubber/electrostatic  precipitator installations,
Equipment manufacturers and system  suppliers also have, some hesi-
tation  to guarantee  that  such  facilities  will  meet very  low
emission levels required for some new plants.  EPA may test a new
scrubber/electrostatic precipitator installation at an incinera-
tor in Massachusetts in 1988,

     The  scrubber/ fabric  filter  technology  redudes  stack
emissions to low levels  approaching  the detection  limits  for
certain compounds, such as FCDD, PCDF and some metals.  The data
base  to  substantiate  this  capability is  growing rapidly.
Nevertheless,   the   reader  should   not  construe  that  the
Subcommittee- endorses the  scrubber/ fabric  filter  as the  only
technology to use.   Other technologies may offer  equal  or even
better performance in the  future.   The potential development of
other improved systems should not be hindered by undue insistence
on the use of a scrubber/ fabric  filter.

D.   Air P-0-ljLu.t-loja Cgjiferol Experience
     Tests on  incinerators  equipped with  the  conventional two-
field  electrostatic  precipitator  have shown 'a wide  range of
particulate emissions, varying from  50  to  300 mg/Nm3.  The three-
and  four-field  electrostatic  precipitators achieve emissions of
20  to  75 mg/Nm3.   An  emission  level  below 20  mg/Nm3  is
technically possible.   However,  there is  a  high  capital cost

                           SO

-------
associated with  constructing a  precipitator  with  a  sufficient
number of  fields and  adequate ^ treatment  area  to  consistently
achieve  such performance.   Furthermore,  the  reliability  of
continued performance at such low emission  levels remains to be
demonstrated.

     Scrubber/fabric  filter control systems are  capable  of
operating  within a  range of  2  to  20  mg/Nm3.     The  material
selected for  the filter  bags  can  have  an important  effect  on
filtering  efficiency  and the  emission  levels achieved.   In
general,  test  results to  date  for the  scrubber/fabric  filter
indicate lower particulate emissions than those for el'ectrostatic
precipitators on municipal solid waste incinerators.   However,
there is considerable  controversy that electrostatic precipita-
tors can be as effective.  The  longer-term reliability and cost
effectiveness of the various control  processes also need to be
considered,
                                                      i»

2«   Metals

     Data exist on emission levels for approximately 30 different
elements.  Among those present  in stack emissions from municipal
waste incinerators are the followings   lead,  chromium, cadmium,
arsenic,  zinc, antimony, mercury, molybdenum,  calcium,  vanadium,
aluminum,  magnesium,  barium,  potassium,  strontium, sodium,
manganese, cobalt,  copper, silver, iron,   titanium,  boron,
phosphorus, tin,  and others.  (See fable 3}

     A number of sampling studies  for metal  emissions  were re-
viewed by M. Clarice [7],   Since  the  condensation point for metals
such as lead, cadmium, chromium,  and  zinc  is  above  300°C, ultra
fine aerosol  particles will form for which  removal  efficiency
depends  largely  on  the  efficiency of the particulate control
system used.  Efficient removal,  defined as  exceeding 99 percent,
has been observed for most metals with the scrubber/fabric filter
system.    Conversely,  relatively high  metal  emissions  are
associated with  lower  efficiency precipitators.   Many existing

                           51

-------
                             TABLE 3
         CONCENTRATION OF STACK EMISSION COMPONENTS FQH
           HWC EQUIPPED WITH SCRUBBER/FABRIC FILTERS3
         COMPOUND
CONCENTRATION IN  (UNITS}
0  TRACE ORGANIC CQMPQUMDS


PCDD                     '
PCDF
CB
CP
PAH
PCS  -
 ng/Nm3 I 12% C02b

 i	- 5
 1-5
 100 - 1000
 200 - 1000
 10 - 200
 1-10
o  METALS
Zinc
Cadmium
Lead
chromium
Niclcel
Arsenic
Antimony
Mercury
 ug/Nm3 € 12% C02
 5-10
 0.5
 1-6
 0.2 - 1
 1-2
 0.02 - 0.1
 0.2 - 0.6
 10 - 40
   PA1TICULATES
All partieulates
 aig/Nm3 § 12% C02
 2-10
O  ACID GASES
HC1
SO-»
   ppa
 10 - 30
 10 - 40
a Source;  Environment Canada  [3]

b To convert to mass  flow rates, use approximately 5000
      rar flue gas I  12% CO2 per ton of  refuse as  fired,
                            52

-------
facilities  have  Inefficient  partieulate  control  equipment,
resulting in higher levels of metal emissions.

     Since  mercury  is  a relatively volatile metal,  several
studies indicate  that  both  sufficient cooling of  the flue  gas
(below 14Q°C, based  on  tests  conducted to date)   and a  highly
efficient particulate removal system are required to achieve high
mercury removal.  The scrubber/fabric filter system can  achieve
efficient  mercury  removal,  provided  that  the  flue  gas  is
adequately cooled.

3.   ^sM Gases
                                                    >

     Municipal  solid waste incineration typically  generates
levels of 300-1000 ppra HC1,  50-200 ppm SO2, 1-10 ppm.HF,  and 75
to 320 ppm  NOX.  Lime  injection into a  scrubber/fabric  filter
system has resulted in removal  efficiencies of  90-99 percent for
                               *                       t%-
HC1 and 70-90 percent SQ2, provided that the flue gas temperature
and the stoichometric ratio are suitable.  This has reduced HC1
to levels below  20 ppm and SO2 to levels below 40 ppm.   This
technology has  also  been extensively  used  in other  applications
for acid gas removal [1,  2,  8 J * •

     The scrubber/electrostatic precipitator combination provides
about  90  percent HC1  removal,  but typically  less  SO2  removal
(about 50 percent).   Since precipitators and baghouses  alone have
no effect on HC1 and SO2 ^removal, lime injection into the furnace
has  been  tested with  some  success  (about  50-70  percent
efficiency). Some sampling to determine HF  removal has  been
reported.  In general, HF removal of approximately 50 percent has
been reported where HC1 removal exceeded 90 percent.

     The  Commerce  Waste-to-Energy  facility in  Los  Angeles
recently achieved significant HOX  reduction  through  the use  of
Selective Npn-Catalytic Removal technology  (SNCR).   Start-up
operation testing has shown  NOX reduction up to 50 percent.   Most
                          53

-------
MSW plants under permit review in California propose to use SNCR
for HOX control £9].

4.   Trace Organiga

     Organic  compounds  for  which  emission  data exist  include
PCDD, PCDF, chlorobenzenes (CB),  chlorophenola  (CP), PAH and PCB.
Available test data identify a number of other organic compounds,
including aldehydes,  chlorinated  alkanes, and phthalic  acid
esters.  Since public perception has  focused  on PCDD and  PCDF
emissions, there  are  more data  for these  compounds,, especially
for the tetra homologues,  and especially the 2,3,7,8 'substituted
isomers.   The other  compounds have  been analyzed at only  a few
facilities.

     Data clearly show that  chlorinated dioxins  and furans exit
the  boilers  and,  depending on the  emission  control devices
employed, some fraction enters the  atmosphere either as gases or
sorbed   onto particulates.   In addition,  the  solids  remaining
behind in   fly ash  or bottom ash contain most  of  the  same com-
pounds,  which become another potential source  of environmental
release of these substances.

     worldwide, there are  data  pertaining  to PCDD/PCDF in stack
emissions for about  35  incinerators.   It  is  important  to
recognize that  this data base was  developed  using somewhat
inconsistent  sampling and  analytical  techniques.   Reported
emission concentrations for PCDD fall  into three  ranges:

          -  low emissions, in the range of 20  to 130 ng/Nm3,

          -  typical emissions,  from 130 to 1000  ng/NM3, and

          -  high emissions,  over 1000 ng/Nm3.

     Average PCDD emissions from older plants may be expected to
range  from 500 to 1000  ng/Nm3.   Concentrations of  the  2,3,7,8

                           54

-------
isomers represent only small fractions  of  these  levels.  The "low
emission" levels tend to be achieved by newer, well operated mass
fired  facilities  such  as  waterwall plants  and  modular design
incinerators.  In moat testing programs, adequate operating data
were  not collected to  correlate  emissions  with  incinerator
operations.    Researchers in the  field theorize that combustion
conditions can  play a  role in minimizing PCDD emissions  [10].
Several studies are underway in  Canada and the united States to
define this role more exactly [11, 123.

     Recently, Environment Canada has evaluated  a  scrubber/fabric
filter  system control for PCDD emissions,  and   has reported PCDD
removal efficiencies exceeding 99 percent.  This has resulted in
PCDD  concentrations  at  the  stack that approach  the analytical
detection limit  of  the  sampling and  analytical equipment
currently available.  Emissions  of  S»CDF exhibit a similar range
of values, and  the  scrubber/high efficiency  particulate removal
combination can reduce PCDF to  very low or non-detectable levels.

     Some limited data on  emissions  of CB, cp,  PCS,  and PAH are
available.     Most  sampling  programs   for  PCDD/PCDF  have
unfortunately neglected to analyze for these compounds.   Maximum
levels from two Canadian studies follow in Table 4.

                             TABLE 4
               SCRUBBER/FABRIC  FILTER PERFORMANCE
COMPOUND
EMITTED
1HLBT
ng/m3
OUTLET
ng/m3
       CB                      17,000                 3,000
       CP                      30,000                 8,000
       PCB     .                   700            Non-detectable
       PAH                     30,000                   130
                           S5

-------
     The  scrubber/fabric  filter  technology  generally  achieved
removal  rates  of 80-99  percent  for these compounds in  the
Canadian studies.  Very few studies report  on other products in
the flue ^ gas.   Some data  from tests  on older  plants have been
reported  for  aldehydes and  certain volatile  hydrocarbons  [8].
Unfortunately,  no data  are available from newer  plants,

5.   Conventional Combustion Gases

     The  conventional  combustion  gas  measurements  include  CO,
total hydrocarbons  (THC) ,  CO2,  and H20.   Both CO  and  THC have
been considered as potential  surrogates or indices of combustion
efficiency for dioxin/furan production; however  studies published
before 1985  report  no  strong correlations.   Nonetheless,  a few
authors have  attempted to  correlate   CO  and  dioxin/furan data
obtained  from several different  facilities  [10J.  Prom such
comparisons,  low  CO  levels (below 100 ppm)  are associated with
low dioxin/furan emissions.  Higher CO levels,  (ranging from 100
ppm to more than 1000 ppm), indicate high dioxin/furan emissions,
but correlations are  not consistent.   During  poor or  upset
combustion conditions,  CO levels of 1000 ppm have been observed
and THC levels have risen from  a typical 1-5 ppm  to 100 ppm and
above.

     A few  studies  have  attempted to determine  CO  and  dioxin
emission  data  under varying  operating conditions  on the same
incinerator,  successfully  demonstrate  a direct correlation  [4r
11] .  Since one  of  the measures of optimized combustion that is
available to incinerator operators   is  minimal CO production, one
could  hypothesize   from  the   above   noted  correlations  that
dioxin/furan emissions  could also  be minimized.    However,
presence of high CO  has been used more  as an indicator of furnace
upset, alerting the  unit operator to take corrective action.
                           56

-------
6.   Ash Disposal

     MSW Incineration  facilities  generate several  residues  for
disposal.       These    include   incinerator   bottom    ash,
boiler/economiier ash, electrostatic  preeipitator  ash,  scrubber
ash, and fabric  filter ash.   The formation of  the latter three
types  of  ash  depends  upon  the type  of  air  pollution  control
equipment utilized.

     Environment Canada studies report that the concentration of
various organic  and metallic  compounds  in ash  greatly  depends
upon their source: bottom ash, boiler ash, or fabric filter ash.
                                                     *
In  general, most compounds present  in  ash  appear to  become
progressively more concentrated in ash found further downstream
in  the  combustion/pollution  control  process*   For example, PCDD
concentrations in fabric  filter ash  were  reported as 200 - 700
«g/g* whereas  30 - 150 ng/g were detected in boiler/economizer
                                                      **
ash, and PCDD were non-detectable in bottom ash*  PAH have shown
a  mixed trend,  with  highest  values in  the  bottom ash,  lower
values  in  the   scrubber  ash,  and  increased  values in  fabric
filter ash.    Host  metals  show a  progressive  increase  in
concentration  (i.e, more  in fabric  filter ash  than in scrubber
ash) ; however, some metals such as chromium  and nickel  show the
reverse trend.   Highly,  efficient air emission  control  systems
result  in  fly  ashes  with  relatively higher  concentrations  of
heavy metals and trace organlcs,  since air pollutant removal is
more efficient.

7*   ongoing Research  and(
     Recent research  results,  based upon the more  modern plant
design and  operation  in the United  States,  Canada,  Germany and
Japan have contributed  measurably to the existing knowledge base
relative to emission control capabilities.   Several  ongoing
studies in Canada, Germany, 'and the United states, will generate,
during  the next year,  data that will  provide  additional
Information on  the  role  of  incinerator design   in  limiting
                           57

-------
dioxin/furan emissions.  A  reduction  of these emissions  in the
furnace will  result  in  lower  concentrations to  be treated  in
subsequent air  pollution control devices.   This will  result  in
lower concentrations  of  pollutants  in  the  fly ash  removed and,
thus, can potentially reduce the concern with ash disposal.

B.   Conclusions and Recommendations

1.   Conclusions

     o  Recent studies of municipal waste incinerator particulate
emissions indicate that  state-of-the-art control devices reduce
these emissions to levels of 20 mg/Nm3 and less.  It  has yet to
be demonstrated whether  such  levels  of  control  can ba maintained
over long periods of time under all normal operating conditions.

     o  Acid gas  control for HC1, SQ2, and HF  can achieve 90-99
percent removal of HCL,  and  lower removal  of  SOj and  HF where
lime injection is used in conjunction with a  wet  scrubber, a dry
scrubber or a wet^dry scrubber.

     o   Removal  of  heavy  metals   (over 99  percent)  including
mercury (over 95 percent),  can be achieved provided that the flue
gas temperature is maintained below  140°C and a highly efficient
particulate control device  is used  (fabric filter or  a properly
designed electrostatic preeipitator).

     o  scrubber/high  efficiency  particulate  removal  technology
offers  the  possibility of  reducing  PCDD/PCDF emissions  to very
low  levels,  well  below  10  ng/Mm3*   This is 1  to  3  orders  of
magnitude below emissions data  reported for  incinerators lacking
this  type of  control technology.    In  addition, this  control
technology is  capable of removing  a significant  portion  of'.CB1,
CP, PCB, and PAH.

     Table  3 provides  a  summary of  emission  results  from  an
Environment  Canada study,  using  a pilot-scale  scrubber/fabric

                           58

-------
filter    control  system.    These  results  provide  further
significant documentation of the low  emission  levels  achievable
with state-of-the-art air pollution control equipment.

     o   Currently,  the  only New  Source Performance  Standards
[NSPS] for MSW incinerators  relate to particulate control.   There
are no  Federal  standards directly  controlling organic or  metal
emissions from MSW incinerators.  Several states have established
their own regulations and permit procedures as municipalities and
private industry have proposed facilities.  This inconsistency of
emission  requirements  among various  levels of  government  has
contributed to public  uncertainty ' regarding the  use of  this
technology and has caused further complexities  in the  permitting
process .

2.
     Adequate data may exist to begin to develop technology based
emission standards for municipal  Incinerators.   However,  EPA and
private Industry should continue  research  to better define trace
emissions,  and the  relationship  between  combustion,  control
technology   and  emission  of  these  hazardous  substances.
Conducting this research will provide an improved data base for
risk assessments  that  can lead to more  scientifically informed
decisions for  adequate protection  of public health and  the
environment.
                           59

-------
              VI.   ENVIRONMENTAL TRANSPORT AND FATE

A.   Dispersal and persistence jn Env ir onment a 1 Media
      Pollutants emitted  to the  atmosphere,  entering  soils  or
waters/ or  deposited  in landfills are  subject  to a  variety  of
dispersal    processes    and   fate    mechanisms,    including
transformation.   Transformations can result  in the destruction of
the parent  compounds  and the  simultaneous  formation of  one  or
more chemical products.   The parent compounds or the products may
have long or  short  periods  of  persistence.   The transformations
may be  photochemical  as can occur via  atmospheric processes  in
the airborne  plume, on  the  surfaces of  soil and vegetation, and
near the top  of water columns.  They may  be chemical  reactions
that can proceed in the dark.   They  may be mediated by physical
and biological processes at or  below soil surfaces and in surface
waters.  From the  viewpoint of assessing potential effects, the
degree  of  persistence  or  rate of  destruction  of the  parent
chemical in the various  environmental  media and  the  identity,
quantity,   and  persistence of  the    products  are  of  great
importance.

    Assessments of  fate — transformation being one of many fate
processes —  clearly  rely   on knowledge of the  identities and
quantities of the parent compounds,  tittle  information  is avail-
able on the fate of chemicals  from MWC  operations because of the
paucity of information on the parent compounds released  in either
stack  emissions or ash.   Furthermore, there  is  currently  no
substantive program designed to identify and  quantify products
formed from the parent  substances.  In  some instances,  in which
the parent  compounds  have been identified,  scientists  can make
reasonable predictions of fate  based on  published studies.

     Assessments of  the environmental transport  and fate  of
chemicals   also depend upon  the  availability  of  validated
mathematical models which can make efficient use of the  available
data.  The availability of such models is,  in turn,  dependent on

                           60

-------
an understanding of  the transport and fate processes occurring in
different media.    Unfortunately,  current levels  of  knowledge
regarding the transport and  fate  of chemicals vary greatly for
different   environmental  media.   In  general/   knowledge  of
atmospheric transport and fate processes  is  much more  advanced
than that for-other  media.  This factor is reflected in the pages
to follow where  discussions relating to atmospheric transport and
fate are more extensive than those for soil and water.

     Figure 10 schematically depicts the major transport pathways
that may disperse'MWC emissions through the ecosystem.   However,
current  understanding  of  complex  terrestrial  food  webs,
biotransformation  and  bioaccumulation processes,  and  the
influence of  these  environmental  processes  on the  quantitative
transport and fate of chemicals, is rudimentary at best.
1.  The Atmosphere

     The  atmospheric  transport  and  fate  of  emissions  from
municipal  solid'  waste incinerators are governed by  a broad
spectrum  of physical  and chemical  processes.   These  include
emission  dynamics,  such  as plume  rise and downwash;  plume
chemistry,   involving changes  of state  and  chemical  reactions;
atmospheric transport and diffusion;  gravitational  settlingi dry
deposition! and wet deposition, due  to  in-cloud and below-cloud
processes.    Model  simulations  are  scientifically  feasible
provided the emissions  are properly characterized along with the
atmospheric and topographic structure of each  site.  The use of
so-called generic  conditions can result in model  results  that are
not  representative  of  realistic potential impacts.    Model
simulations are impeded further by the scarcity of  information on
combustion products and the atmospheric transformations  of those
products from any  emission sources*

-------
 MUNICIPAL
  WASTE
INCINERATOR
 BOTTOM
  ASH
   DISPOSAL
    SITE
                Figure 10
  STACK
 EMISSIONS
FUGITIVE
EMISSIONS
    Transport of MWC Emissions
    From an Incinerator Facility
    Through the Ecosystem
                 62

-------
a.   Stack Smiaaions

     Understanding emission products'requires characterization of
the  plume constituents  and the  factors  pertaining  to their
release, including temperature, velocity,  and physical character-
istics of the materials exiting the stack.   Emissions need to be
characterized according to chemical state,  rate of  release,  and,
for  particulates,  size  distribution,  density,  and  mass.    The
description of the release must take into account factors such as
stack  height and  diameter,  efflux  velocity, and  sensible  and
latent  heat  content  in  order to  calculate  plume  rise and  the
effective release height. The  latter also requires  characteriza-
tion of the vertical  (and sometimes the horizontal)  structure of
the  ambient atmosphere—winds,   temperature,  and humidity.
Several  algorithms are  available  which  provide  representative
estimates of the effective plume rise [13].
                                                       (i
     Downwash can  occur  where  the  stack height is  low with  res-
pect to the  incinerator  or adjacent buildings.   This phenomenon
occurs due to aerodynamically generated, horizontal-axis vortices
•or eddies  in the lee  of  'the stack/structure that  transport the
stack plume to the ground, thereby effectively creating a ground-
level volumetric source.  Case-specific analyses  are required to
assure the absence or prevention of adverse  downwash effects in
the vicinity of  a  given  incinerator.   In  general, however, it is
possible to estimate  minimum stack-height requirements  using the
so-called  "2.5  times ' rule"  that  suggests that stacks  discharge
their emissions  at a  height  at least 2.5  times the  height of the
tallest  nearby structure,   "Nearby"  i»  interpreted to  include
structures whose horizontal separation  from  the  stack  is  less
than five  times  the height or  width of  these  structures
{whichever is greater).

    . EPA's Administrator has promulgated regulations  (40 CPS  Part
51)  that  define the  use of good  engineering practice  (GSP)  to
limit the stack  heights  that can be used  to avoid downwash.   EPA
has developed  guidelines for  determining GEP stack height [14],

                           €3

-------
and to provide guidance for the  use  of  fluid modeling  (i.e. wind
tunnel simulations)  to determine GEP  stack heights  [15].

b.   Atmospheric. Dispersion and Transformation

     The transport and  diffusion  of  gaseous and particulate MWC
emissions are  governed  by the magnitude and variability of the
wind, the thermal  stratification of the lower troposphere, and
the aerodynamic characteristics of the ground  surface  (including
manmade  structures).    These  factors  interact,  and  their
characterization  is  frequently difficult   (but  tractable),
especially in urban areas and complex terrain.

     Well established and  representative modeling techniques are
available from  the  Agency, especially  for determining  long-term
impacts rather  than  short-term  case  studies [16].   It  is  impor-
tant, however,  that  the atmospheric  measurements used "as  inputs
to  these models  be  representative  of  conditions  both  at the
source and downwind.  For  example, wind and temperature profiles
recorded from observing  instrumented weather balloons  located  at
airports are  frequently not  representative  of  the  urban
environment, similarly, stability  estimates based on airport
surface weather observations  may not be representative  in  urban
areas, or when extrapolated to areas that  are bounded by  large
water bodies.   A useful overview of dispersion  parameterization
methods is provided  by Hanna, Briggs and Hosker,  1982, [17]  as
well as many other sources. .

c.   Depos ition

     Removal of  particulate  emissions and  gaseous   constituents
by atmospheric deposition is also an  important  fate  process.  The
removal and deposition   at the earth-air interface c . jurs by dry
deposition and precipitation scavenging.  Although there has been
extensive theoretical  and observational research on these fate
mechanisms,  there  is still considerable uncertainty surrounding
                           64

-------
both physical processes and modeling. Hosker  and  Lindberg,  [IB]
have prepared a critical review and summary of these issues.

            Deposition

     Dry deposition  of  large  particles occurs, by gravitational
settling.  Several mechanisms  remove small particles and gases at
the  surface,  including impaction,  electrostatic attraction,
adsorption,  and  chemical  processes.   The  fall  velocity  of
particles  is determined by  the  balance between  gravitational
forces and aerodynamic drag forces, and depends on particle size,
density,  and shape.   Inert materials typically deposit  more
slowly  than reactive  materials or  charged particles,   and
vegetated surfaces effect  greater deposition than bare surfaces.
Dry  deposition  estimates for  MSW  emissions  are, therefore,
subject to considerable uncertainty.  Hanna,  Briggs and Hosker,
have estimated  the distance  from sources of  several  heights at
which 50 percent of the plume is depleted through dry deposition
for a wind speed of 1 m  s"1 and a  deposition velocity of i cm s""3"
[17 ].   These data are summarized in Table  5, and  they clearly
show that dry deposition can be  an effective removal process for
certain combinations of  source height and stability.
                            TABLE 5
          DISTANCE IN km WHERE DRY DEPOSITION DEPLETES
     THE MASS OF A PLUMS BY 50 PERCENT FOR A WIND OF 1 m s'1
              AND A DEPOSITION VELOCITY OF 1 cm s'1
Meteorological
Stability Class
A-B
C
D
E~
F
Source Height (m)
0 10 50..
>10 km
1.8 18 43
0.4 3.5 8.6
0*13 . 2.2 8.3
0.10 2.0 10.0
100

60
19
17
28
       Source;Hanna,  Briggs and Hosker, 1982

                           65

-------
     o Wet. Deposition

     Wet deposition of plume material  occurs both as the result
of in-cloud scavenging by  cloud droplets  (i.e. rainout)  and
below-cloud scavenging  by rain or  snow,   called  washout.
Scavenging  is  a function of  pollutant reactivity,   solubility,
size  distribution  (for particulates) ,  rainfall  intensity,  and
cloud  residence  time.    Therefore,  it  is  important  that  both
clouds and  pollutants  be carefully  characterized  in  order  to
provide accurate estimates of wet deposition.

     Washout and rainout are typically combined and expressed in
terms of either a scavenging coefficient or  washout  ratio.   The
scavenging  coefficient  is more  appropriate to  single episode
events and is used to  express the decrease of concentration with
time.   The washout ratio  averages  conditions  over  multiple
precipitation   events.    It   is defined  as  the * effluent
concentration   in  precipitation normalized  by  the  effluent
concentration   in  air, and tends  to  decrease with time  of
precipitation in a  given  storm and increases  with  the overall
precipitation   rate.    Worst-case  wet  deposition  rates  and
distributions could be obtained using data  from specific events
while  long-term wet deposition  amounts  may  be better  estimated
using the washout-ratio concept.
     o  Deposition on;

     The Agency has presented  a  number of  possible approaches to
evaluate exposures to  emissions  from  municipal  waste  combustion.
However, literature available  to the  subcommittee contained only
a   few  examples  that   sought  to   predict  environmental
concentrations of chlorinated  dioxins around municipal waste
incinerators.  These approaches,  although simplistic and limited,
may be helpful in indicating whether  more  sophisticated analyses
should be undertaken.   leychok made an effort to calculate
exposure  of  soil to PCPD based  on air  concentrations.   He
calculated a  value  of  7.5  x  !CT10g/g   of  soil  [20].   The

                           66

-------
Subcommittee also conducted its  own  evaluation  of possible soil
deposition which might  be associates with the  emissions  of the
Hampton,  Virginia  incinerator  facility.   The Subcommittee
concluded that  simplistic  environmental  transport models,  which
did  not include  any allowance  for the  possible breakdown  of
emitted compounds,  did indeed provide similar estimates to those
cited above.  Dispersion modeling of FCDD by Higgins,  predicted
maximum ground-level atmospheric concentrations on the order  of
10*12 to  10~13  g/m3 [21],   Average  ground-level concentrations
were predicted  to be approximately 5 orders  of magnitude lower
than  stack  concentrations.    However,  until  confirmed  by
measurements,  the  significance  of  these simplistic  modeling
exercises will  remain very controversial.

2.   The Terrestrial Environment

     The Subcommittee has  examined the terrestrial  fate of PCDD
                                                      K
and PCDF  as examples of  the  terrestrial fate  of MWC  emissions
because emissions of  PCDD/PCDF have  generally been studied more
intensively than the emissions of other compounds, such as metals
and  acid gases.    Unfortunately,  knowledge  of the  fate  and
transport of PCDF/PCDD  emissions  is largely based  upon  the
application of mathematical models.  In  this case,  knowledge of
the physical transport models is  superior to  our understanding of
the  fate of these  chemicals during transport and subsequent
deposition.
                        s
     Despite a  limited  understanding of  the  terrestrial fate of
dioxins and  furans  that have the potential  to be  deposited  on
ground  and  vegetation surfaces  as a result of municipal  waste
combustion,  there  is  some evidence that incinerators are  the
source of dioxins found in nearby surface soils [22].  Given this
evidence, major questions arise  concerning the fate and mobility
of dioxins  in  the  terrestrial  environment  due to the lack  of
definitive  information  concerning the  physical  nature of  the
stack  emissions,  atmospheric transformation,   photodegradation,
                           67

-------
volatilization, mobility in  soil,  and translocation  in and
retention by plants.

     Whereas particulate  dioxin emissions are  characterized by
condensation onto fly  ash,  virtually all  research  on the
environmental fate of dioxins has  focused on pure dioxin,  dioxins
in  solvents  and herbicides,  dioxins  in aqueous  solution, and
dioxins  from  the 1976  ICMESA  accident  in Seveso,  Italy.
Furthermore, most research on the  fate of  dioxins  has been
concerned with  2,3,7,8-tetrachlorodibenzo-p-dioxin  (TCDD)  which
is a solid at standard atmospheric temperature and pressure/ has
low solubility  in water but high solubility in organic solvents
and has  a  very low vapor  pressure.   Approximately half  of all
incinerator  emissions  are gaseous  and  dioxins  may  be  emitted
along with these gases, however, most relevant research has been
conducted with TCDD, a solid.   Therefore,  the fate of gas-phase
dioxin  emissions  is largely  unknown.   Except  as noted, the
following discussion refers to.the environmental fate of  2,3,7,8
- TCDD.

     In the atmosphere,  2,3,7,8-TCDD may  be subject to photolysis
and oxidation by the hydroxyl radical*  A photolysis half-life on
the order  of 5  to 24  days  is  estimated for  typical  sunlight
conditions.    No quantitative  estimates   of  oxidation  rate are
known to  be available, presumably  due in part  to the  lack of
information  on  atmospheric  abundance and  distribution  of the
hydroxyl radical,

     Photodegradation of dioxins appears  to be the principal loss
mechanism, although the findings are in some cases contradictory
and poorly understood.   Research indicates that TCDD  is unstable
when dissolved in  solvents and exposed to ultraviolet light  [23,
24]; while  thin  films  of pure TCDD applied to  glass  plates are
reported to  be  stable  in  sunlight [24].   Later studies indicate
that TCDD photodegrades when contained in a herbicide  solution;
the loss rate  is  greater when the  solution is  applied  to  plant
leaves and less when applied to soil surfaces (presumably due to

                           63

-------
shading)  [25],   subsequently, Plimmer confirmed  that pure  TCDD
can decompose by UV photolysis when applied to;glass in very thin
layers,  while  soil  appears to  protect  against  UV  degradation
[26].

     Studies  of  fate  in  soils  indicate that  TCDD   can be
persistent under certain conditions with  an apparent half-life qf
approximately one year [24, 27].  TCDD is  nearly  immobile in the
soils tested by Helling [28] and Matsumura and Benezet [29].  The
latter also found microbial degradation  of TCDD to be very slow
as did Isensee and Jonea  [30].   On the other hand.  Young et al.
report  biodegradation losses  with  apparent  half-lives  of  250
days.  They also report that plant uptake  and  transport  are'very
Slow [31].

     Perhaps highly relevant environmental studies regarding soil
transport and fate are those of DiDomenico et.  al*,  [32,,  33], who
made measurements  following the  ICMESA  chemical plant  accident
and  release of  2,3,7,8-TCDD.   Surface soil samples  (7  cm deep)
were analyzed  at  44 locations  1,  S  and 17  months after  the
accident.   Overall,  concentrations decreased  significantly
between  the  first and  second  surveys  with an equivalent  mean
half-life of 10 or 14 months.   Concentrations changed very little
between the second and third  surveys, and  the  apparent half-life
was  estimated  to be  greater  than- 10  years.    These changes, in
apparent  half-lives  with  time show  that  half-life,  or  first-
order kinetics are  inappropriate for describing  the persistence
of TCDD.   Concentration profiles were  also measured  at  about 32
sites to  depths  of 136 cm, up  to 17 months after  the accident*
significant   soil   penetration   was    observed,   although
concentrations decreased rapidly away from the soil  surface.  As
a generalization, the concentration  of TCDD more than 8  cm from
the- surface was. ten fold lower than that  in the .upper 3 cm*  Some
of the soil profiles showed maximum concentrations  in the 0.5 to
1,0  cm layer,  rather  than in the  uppermost  layer,  suggesting
degradation at the surface as  well as  migration away  from  the
        f
surface.  Loss mechanisms at the surface  were not identified, nor

-------
were the specific transport phenomena, although  it  was  suggested
that "migration  in  the soil  may have occurred  along with  soil
colloids and particles to which TCDD may have been bound."

3.   The Aquatic Environment

     Even less information is available on the transport and fate
of MWC emissions  and  residues in aquatic  systems.   Contaminants
associated  with  partlculates  emitted from  MWCs are subject  to
deposition on surfaces downwind  from  the  facility.   This fallout
is subsequently subject to  dissolution or suspension ,in rain  or
meltwater from  precipitation events.  Eroding  soil  and flowing
water enter nearby water bodies, and  suspended solids may settle
out  and become  incorporated  with  the  sediment  in such  water
bodies.   The  dissolved  portion may also  infiltrate   into  the
ground,  recharging groundwater and may be re-evaporated into the
atmosphere.                                            *

     Again, PCDD and PCDF emissions, specifically TCDD,  have been
studied  in  more detail than  other  identified contaminants  and,
therefore, they are used as  examples.  The transport of  dioxin -
contaminated soil into lakes  and streams  by  erosion is  evidenced
by the detection  of 2,3,7,8-TCDD in water samples  from a Florida
pond  adjacent  to  a  highly   contaminated  land  area  t34]-
Additionally, several laboratory studies have shown that lakes  or
rivers can  become contaminated  with  minute quantities  (ppt)  of
2,3,7,8-TCDD and  possibly  other dioxins  through  leaching  from
contaminated sediments.  In a study reported by Isensee and Jones
2,3,7,8-TCDD was  adsorbed to  soils,  which  were then placed  in
aquariums filled  with water and various  aquatic organisms  [30].
Concentrations of the dioxin  in the water  ranged from  0.05  to
1330 ppt.  These values resulted from initial soil concentrations
of  2,3,7,8-TCDD  rang^ig   from   0.001   to  7.45   ppm.    The
investigators concluded that dioxin adsorbed  to soil could lead
to  significant  concentrations  of  2,3,7,8-TCDD  in water if  the
dioxin-laden soil was washed into a pond  or other  small body  of
water.
                           70

-------
     Other  investigations  have  given  similar  results.    Using
radiolabeled 2,3,7,8-TCDD,  ward and Matsumura showed that, after
dissolution from  contaminated lake  bottom  sediment,  dioxin
concentrations in   water ranged  from  0.3 to  § percent  of  the
original  dioxin  concentration  found in  the  sediment  [35],
Results of  another  test  indicated that  a total  of  about  0.3
percent of  the applied  dioxin concentration  passed through sand
with  leaching water  [29].     In  some  cases,   the observed
concentration  of TCDD in the  water  was greater than  its water
solubility  (0.2 ppb).   The dioxin present in  the  aqueous phase
probably results from presence of 2,3,7,8-TCDO metabolites,  and
binding or  adsorption of fCDD onto  organic matter  or sediment
                                                     i
particles suspended in the water.  In another study, application
of 0,1 ppm TCDD to a silt loam soil led to TCDD  concentrations in
the water ranging  from  2.4  to 4*2 ppt over a  period of  32 days
(32].
                                                      t»
     The  findings  of such investigations  are  consistent with
recent  reports that  TCDD migrates to  nearby water  bodies from
industrial chlorophenol wastes buried  or stored in  various
landfills.  At Niagara Falls,  New York,  for example,  1.5 ppb TCDD
has been detected at an onsite lagoon at the Hyde Park  dump where
3300 tons of 2,4,5-TCP wastes  are buried*   Sediment from a creek
adjacent to the Hyde Park fill is contaminated with ppb levels of
the dioxin.   There  is  growing evidence  that  TCDD has migrated
from process waste containers  in the  landfill of a  former 2,4,5-T
production site in Jacksonville, Arkansas* The  dioxins have been
found in a  large pool of surface water on the site (at 500 ppb),
downstream  of  the  facility  in the local sewage treatment plant,
in bayou bottom sediments, and  in the flesh of mussels and fish.
TCDD  is  also apparently  being  leached  into   surface  and
groundwaters from  an 880-acre dump site  of  the Hooker Chemical
Company at Montague, Michigan*  Dioxins were found  on the site at
a level approaching 800 ppt [37]*
            t
     A  recent study,  [38],   considers  the  fate of 1,3,6,8-
tetrachlorodibenzo-p^dioxin in  aquatic  systems.   This congener,

                           71

-------
though  less  toxic  and  persistent  than  2/3,7,8-TCDD,  hag been
reported as a component of  fly ash from municipal incinerators.
A  major  portion  of  the  total  TCDD  isomers  entering  the
environment from  tooth  herbicide  and  combustion  sources  is
1,3,6,8-TCDD,    Movement and  accumulation  of  1,3,6,B-TCDD were
monitored in air, water, sediment  and vegetation over a 426-day
period,  sediments were shown  to  be the  major reservoir for the
compound, accounting for 5  to  14 percent  of the added compound
after 426 days.   Five to eight percent  of the applied compound
could be accounted for in the water. The experimenters concluded
that volatilization losses  nay represent a significant route of
release  following slow  release from sediment  and decayed plant
material.   TCDD isomers degrade  only under conditions  of high
microbial activity, such  as in  the presence  of  decaying plant
material, and in situations where the compound is bioavailable.
Direct and indirect photolysis are  major paths of chemical loss
in shallow water? however,  little of the compound is degraded in
waters shielded from sunlight.  •

     The above information on  the fate and dispersal  of TCDD in
waters is relevant to  assessments of MWC emissions and .residues,
but these compounds  are  not  the only ones likely to be present in
such emissions and residues, as  previously pointed  out.   Metals
and acidic gas components should  also  be assessed to adequately
characterize the transport and  fate  of MWC  by-products in aquatic
environments,   A considerable  data base gives insight  into the
participation-  of  metals  -in  environmental processes.  HWC
emissions,  however,  have not been well characterized, and  likely
constituents have not been  identified or  verified.   without an
understanding of the metal contaminants that  are likely to  result
from   MWC  emissions,   environmental   transport   and   fate
determinations cannot  be  scientifically supported*   similar
problems with  different complexity  surround acid gas emissions
and associated transport and fate determinations.
                           72

-------
B.   Conclusions and Recommendations

1.   Conclusions

     Atmospheric concentrations  and surface  depositions of  MSW
emissions result from  a complex relationship among  such factors
as  emission  rates,  plume rise,  downwash,  dispersion,  chemical
reactions, and wet and dry removal processes.   Casual attempts to
simulate  or  estimate  ambient  and  surface  pollutant levels  can
lead to  large  uncertainties  in the  results.   Notable gaps  in
current  knowledge  include identification  and quantification  of
the organic or  other chemicals contained in stack  emissions,  and
aguatic and terrestrial transport and fate processes.   There  are
other  cases  where current  information is  inadequate to permit
reliable predictions of concentrations and compounds  that result
from MSW emission*
                                                       K
2.   Recommendations

     o   A comprehensive  data  base should  be developed through
atmospheric  field  studies   at  several  representative  MSW
facilities.   The data base should be used to  estimate deposition
           /
(wet and dry)  of particulate and  gaseous  emissions,  and  the
organic compounds generated should be identified and quantified.
Such field studies and the  resultant data base  should also  be
used to  evaluate  mathematical and  fluid  models  of  transport/
diffusion and deposition in urban and suburban environments.   The
data base should include measurements of HSW emissions (stack  and
fugitive), plume rise,  dispersion, and wet  and dry  deposition,
and  should  incorporate  the use of  inert gaseous  tracers  and
soluble  particle  tracers; both  long-term  and case-study  (i.e.
intensive) measurement programs should be conducted.

     o  Fluid modeling studies  should be conducted for urban  MSW
incinerators, and those likely to be affected  by  complex terrain.
EPA should implement fluid modeling methods for GEP  stack height
determination  in  the design  and  siting of MSW  facilities.

                           73

-------
Fugitive  partieulate  emissions  also  need to  be considered.
Estimation of  their  impact  on areas downwind  is complicated by
th& need  to  model the processes  by which  the  particles  become
airborne.   Major  difficulties  include  description  of  the  source
term and specification  of the  accompanying local wind field.

     o  There  have been  significant advances in the  design and
performance  of municipal waste  incineration  facilities.   An
environmental monitoring program should be  instituted  that will
assess the occurrence of  combustion products from diverse sources
before  a  state-of-the-art  municipal waste incinera.tor begins
operation, and then assesses the  incremental   contribution  of
combustion products from  the municipal waste incinerator after it
begins to operate.

C.   Transport and Fate of MWC 4fJ3
                                                      t*
l.   Considerations

     The  solid  waste  or  ash generated  by  municipal   waste
Incinerators  potentially  contains any  and/or  all  of the same
chemical substances found in stack emissions,  The masses of most
Individual components  are likely to  be  greater In  the  solids
since  the chemicals  preferentially  partition  to  the  solid
particles.  Unfortunately, in  few data  sets  are  the solids well
characterized chemically.

     The ashes remaining after combustion, and those collected by
pollution control devices,  can pose  a threat to humans and the
environment if not properly handled and disposed.   For example,
ash from  particulate control  devices  at one incinerator visited
by the  Subcommittee was  poorly contained  and was observed  to be
partially dispersed  into the ambient air through  large, unsecured
openings  in  the  exterior walls  of the  incinerator  building,
Fugitive dust  (ash)  was  also  observed  to  be blowing off the top
of uncovered  dump trucks.
                           74

-------
     It Is  common practice to  dispose of  incinerator-generated
solids in municipal landfills.   Once landfilled,  the solids and
associated pollutants are mixed with other wastes, some of which
may be liquid.  These wastes vary considerably in their chemical
and solvent  characteristics.   They  are  composed  of mixtures of
organic solvents, decomposing organic materials,  and high or low
pH liquids.   Just how the pollutants will react and interact as
they  associate  with added MWC ash under such conditions is
unknown.  Some of the organic pollutants may leach from the ash
particles,  especially if  they  are  exposed to  organic  fluids.
Some of the trace metals may be dissolved under normal  conditions
in ground water.

     A Canadian  study has been recently completed which  focuses
on the  leachability of  metals  and  trace  organic compounds for
different ashes  from  a  dry scrubber/fabric filter system, and  a
wet/dry, scrubber/fabric filter system [3]*  Batch leaching tests
with  distilled water on a  laboratory scale indicate that there
was no organic contaminant mobility from any of the ashes, except
for  chlorophenols.    However,  long-term  leaching of  organic
compounds was not determined.

     An extensive investigation of the disposal of bottom  and fly
ash in  a separate ash  landfill subjected to acid  precipitation
showed no significant mobility of'metals  [39],  Although the ash
contained significant  quantities  of  metals,   most were  not
mobilized  at the  expected  pH  of  the ash/leachate  system.   In
contrast,  some  metals,  such  as copper,  lead, zinc,  and boron,
were  leached to varying  degrees  when  subjected to  waters  of
varying pH  in a Canadian Study  [3}*  Significant quantities of
cadmium, lead, zinc, and copper may be leached in  the short-term,
suggesting  that   further  investigation and  special  handling of
these ashes are needed for safe disposal.

     Landf i-lling  of  fly   and  bottom   ash,   without  some
stabilization, may or may  not pose  hazards  that surpass those
presented by  burying  wastes that have not  been incinerated.  The

                           75

-------
mass  of  municipal  waste  may  be  reduced  considerably  by
incineration,  but some  of  the constituents in the ash, especially
heavy metals, may become correspondingly more concentrated.
Also, the chemical form of many elements may change  during the
incineration process, affecting  the  subsequent transport and fate
of the resulting  compounds.   Such changes may result  in either
increases or decreases  in  the leachability of the substances.

     EPA  has  developed  two  leaching  tests  with  potential
applicability to assess transport and fate of  pollutants in ash.
These tests are  the   Extraction  Procedure  (EP)   test,   and  the
Toxicity Characteristic Leaching  Procedure  (TCLP).  * Such  tests
were designed  for purposes*  other  than assessing the pollution
potential of  MWC ash,  and,  therefore,  the salient  features  of
these tests should be evaluated to determine their applicability
for assessing MWC ash*
                                                      h
2.  Conclusions and Recommendations

a.   conclusion

     Insufficient data  exist on the identities and quantities of
chemicals  in ash residues, preventing a  rigorous scientific
evaluation of the transport  and fate of contaminants discharged
from municipal waste  incinerators,

b.   Recommendation

     State-of-the-art  analytical  chemical  techniques should  be
employed  to  characterize ash  samples,  and as  many of  the
extracted  compounds   from selected installations  should  be
identified as feasible, to provide a useful data base.  The goals
of this effort   should include determination of  the speciation
and mobility  of trace  metals  and  trace organics released  from
municipal solid waste combustion facilities.
                           76

-------
c.   Conclusion

The  practice  of disposing of  bottom  ash and  fly  ash from
municipal waste combustors by landfilling will be  of increasing
concern  as  more MWC  employ state-of-the-art pollution  control
technologies.   It is  time to consider the disposal of ash as  a
discrete problem,  independent of  the combustor itself.  A number
of approaches already being applied to the disposal of hazardous
residuals   could  be  utilised.    Generally   these   involve
solidification or vitrification of  the waste material.   Grouting
of disposal trenches,  sometimes  in combination  with  liners,  is
another technique that  may be  applicable.  The  Subcommittee
recognizes that these techniques  may need to be modified  to meet
the  particular  chemical  characteristics of  bottom ash  and  fly
ash, although much work has been  done on the utilization of coal-
fired power plant fly ash.
                                                      ft
d.   Recommendation

     The  present  handling and disposal  practices  of ash,
especially fly ash,  should be examined closely in  light  of  data
regarding the potential  for  movement  of heavy metals, contained
in  MWC ash',  into the environment.   This  examination  should
include  identification and quantification of the  inorganic  and
organic  chemicals that may leach from both fly and  bottom  ash.
Determination of the  transport and fate of identified chemicals
should  follow,   and   should   include  determinations   of
bioavailability.
                           7?

-------
                 VII.  POTENTIAL EXPOSURE AMD EFFECTS

     The ensuing discussion  of  potential exposures and effects
associated  with municipal  waste  combustion  should not  be
construed as a risk assessment of this technology.   Rather,  it is
a discussion of the data needed  to  further  improve  EPA's ability
to  conduct  a  risk assessment  of  municipal waste  combustion
emissions   and  discharges   and'  to   enhance  the  public's
understanding of this technology*  The Subcommittee's  critique of
EPA's risk assessment methodology (See Appendix I) indicates that
the  procedure   that  has  been developed  is  a   significant
preliminary  step that  aids  decision  makers  in  evaluating the
risks of municipal  waste combustion.

     It is  clear that  a certain fraction  of the components of
stack  gases and particulates,  fly and  bottom ash,  and   scrub
waters will  reach the ambient environment.  Their presence in the
ambient  environment  may  result  in  human  arid environmental
exposures.   The preceding chapters have identified research  needs
pertaining  to  the  quality and  quantity  of  emissions  and
discharges,  and their potential transformations and delivery to
sites  where ecosystems  and  humans could be  exposed.   In this
section, the  Subcommittee  examines the  need  for data that will
improve  the   characterization and  prediction of  potential
exposures and effects to both humans and ecosystems.

     On  the • basis of  current scientific  information,  the
Subcommittee cannot state  that  no  risk is posed from  municipal
waste  combustion.   From  both  a  scientific  and  a  policy
perspective,,  the two  most critical  unsolved questions  are as
follows:

     o  What  is .'..MS relative contribution  of pollutants emitted
by  municipal waste combustors relative  to other  combustion
sources?
                          78

-------
     o   what  is the  probability that  human  or  environmental
receptors  will be  exposed to .harmful amounts of incinerator
emissions or discharges?

     At best,  obtaining  answers to  these questions can  help
resolve the  issue  of the risk associated  with this technology.
At a minimum, such  answers  can reduce the current uncertainties
faced by scientists and decision makers as they seek to develop
environmentally  "safe"  alternatives  for  municipal  solid  waste
disposal.

A.   Environmental Loadings
                                                     *

     The by-products  of  municipal  waste incineration  (stack
emissions and ash discharges, for example), contain constituents
that are virtually all already present in the environment.   They
originate   from a  variety of  combustion   sources  including
vehicles,  smelters,  home wood  burners,   and  fossil  fuel  power
plants.  The emissions generated by municipal waste incineration
will  add to  emissions from  other  sources to  yield  the  total
environmental load.  The relative proportion that MWC discharges
will contribute to total ambient levels is open to question, and
will vary from site-to site.

     The Subcommittee  concludes.'that,  with ' state-of-the-art
combustor designs, controlled operating conditions, and effective
emission control  devices,/  the emissions  from MWC  alone  are not
likely to significantly increase total environmental loadings on
a national basis over the  next generation.  This conclusion rests
on  the observation  that  the reported environmental  levels  of
chemicals  known to be  emitted  from  MWCs do  not  appear to  be
significantly greater in Sweden or Japan,  countries that practice
a much greater degree of incineration  than that projected for the
United States.  -In  addition,  this conclusion  assumes that,  over
time, existing facilities, both controlled and uncontrolled, will
be  replaced  or  retrofitted with advanced  design  features,
controlled operating conditions,  and emission  control equipment.

                       .   ' 79

-------
     This conclusion can,  of course,  vary with the  total  loadings
generated from this and other sources in specific municipalities.
It should not  be expected that  loadings  from an  incinerator  in
Philadelphia or  Boston will  be  of the same proportion  (relative
to other combustion  sources)  as  loadings from an  incinerator  in
the rural southwest.  The policy issues faced by decision makers
in these two  regions of the  country  may  also be different.   in
the former case, a chief issue will be the incremental risk  that
is experienced fay urban populations relative to other  sources,-  in
the later instance,  a major issue will be the impacts  of a new
source, a risk in the absence of other major  combustion  sources.
Members  of this Subcommittee  cannot  answer- the question  of
whether such  risks  under either scenario are acceptable.    what
scientists and engineers can  do  is inform citizens and  decision
makers  of  current  risks   and  uncertainties,   and   develop
recommendations for further reducing them.

B.'   EXPOSURES

1.   Human Exposures

     The ambient environment at  any particular  MWC  site may
present a  hazardous  exposure to  either humans  or  ecosystems.
Assessing exposure  to MWC  emissions  is particularly difficult.
The compounds  present are generated  from a  variety of  sources,
and  isolating  the contribution  of  pollutants  from MWC  is a
complex undertaking.  The magnitude of exposure to MWC emissions
depends upon 1) the density of respective populations, and 2) the
extent to which  the environment  already receives discharges  from
other  sources.   Individual life styles  also  influence  the  body
burdens of these chemicals.   For instance, cigarette  smokers may
have higher levels  of cadmium in thel^- kidneys than  individuals
of  the same  age  who do  not smoke.    Cadmium may  also  be  a
component contributed by MWC ash residues.    The level  of human
exposure to municipal waste combustion discharges will be highly
                           80

-------
site-specific, .and  its  significance  will  be  dependent  on
contributions relative to  other combustion sources.

     Based on  present and  proposed  incinerator sitings,  it is
likely that municipal incinerator facilities will be sited in or
near areas with dense human  populations,  Siting  areas are likely
to be industrial and urban,  as opposed to agricultural and rural.

     A significant amount  of background   information on  the
toxicity to  and tolerance of  humans  to many  of the components
identified in  MWC  emissions has bean  collected.  Much  of this
information exists in the  general toxieologieal literature.  The
data also appear in secondary sources,  where the information has
been summarized and evaluated, such as in criteria documents and
health advisory documents  prepared by the U.S.  EPA,  in criteria
documents  prepared  by  the  National  Institute  of Occupational
Safety and Health,  and in various publications  of the National
Research  Council.     When   combined  with  existing" data  on
precombustion exposure levels,  adequate criteria  can be developed
to both  protect' human health and  facilitate  the permitting
process for plant construction and  operation,

2.   Ec_O-Sy_sjfcem Exposures

     When  compared to humans,  much less  is known about  the
exposures of plants and animals  to MWC  discharges and potential
toxicants.    Moreover,  even  less  is  known  about  how well  (or
poorly) ecosystems respond to, tolerate or recover from exposure
to these substances*

     EPA   supports  research   on  the   tolerance  of   aquatic
vertebrates  and invertebrates to  such  substances  dissolved in
water,   These  efforts,  along  with  work  in  structure/activity
relationships -among  chemicals,  toxic e
-------
necessary for  ecological  risk assessment in  general.    However/
little of this effort is directed specifically to municipal waste
combustion emissions.

     The ecosystem components most likely to  encounter risk  from
municipal waste incineration  emissions/  in  the  short-term,  will
be   aquatic and terrestrial  life  existing  near the facilities.
Ecosystems may experience  longer-term  or chronic risk depending
upon the persistence and accumulation of emitted substances,

3.   ,Approaches for Estimating Exposure

     EPA*s approach, to date, for estimating  exposures of humans
and  ecosystems has  been to develop models that make a number  of
assumptions, in the  absence  of  data, as to how  pollutants reach
human  and environmental receptors  through direct  and  indirect
pathways.   The Subcommittee believes that modeling represents  a
first but only a preliminary step towards answering  the two major
questions of this  chapter.    Modeling  in  the  absence  of  even
limited data  or validation is simply  too uncertain a tool  from
which  to present  statements  to the public  on  the presence  or
absence of  risk  from a technology.  What is  needed is  for  EPA,
the  private sector  and the public to take the next  step, that  is
to  develop  a strategy fors  1) measuring  the emissions  and
discharges  from major combustion sources and the proportion  of
such  emissions  and  discharges  attributable  to  municipal
incinerators, 2}  measuring  selected  human  and  environmental
receptors in  urban and rural areas to  discern impacts, and  3)
comparing the source-receptor relationships that emerge.

e.   Effects

1. •  Human Health Effects

     The  impact on health  from ingestion, inhalation and dermal
absorption  of  individual chemicals  or  chemical  mixtures emitted
by municipal waste  combustors depends on  the  dose humans receive

                           82

-------
and the  duration of  the dose*   The Agency,,has maintained  an
active program to develop health advisories, criteria documents,
and risk assessments  that  characterize  the  toxicology data base
and evaluate the dose-response  relationships in  humans for many
of the chemicals identified as  by-products of MWC,  Understanding
the Impacts  from HWC  was  not  the  primary motivation  for such
efforts.  Nonetheless,  pertinent  information  and analyses have
resulted.  Additional relevant information is available  in the
toxieologleal literature.  Very little  information  is available
on the effects  of specific  isomers  or on  effects  of mixtures.
Interactions between and/or  among  compounds  may enhance  or
eliminate their toxieity or bioavailability.

     Although a data base exists for many compounds, the effects
caused by  a significant number of substances  are  relatively
unknown.   There may also be chemical  constituents in  MWC
emissions or residues that  have not been identified.

2.   Environmental Effects

     The response of individual organisms to toxic substances is
a function of concentration, toxiclty and duration  of exposure.
Combinations of  these functions  may produce lethality or more
subtle responses,  such  as  behavioral  changes  or  reproductive
inhibition.  These sublethal effects can take  on a  multitude of
forms with varying effects.  They are often difficult to  detect
under field or laboratory conditions.

     Much of  the bioassay research  on  environmental pollutants
has been conducted  using fish  and aquatic  invertebrates.   Less
research has been  performed  with  plant,  mammalian  and  avian
species and with microbes.   To  a large extent,  this research has
yielded data on the levels of a toxicant  to which species produce
an acute response,  while research on  chronic and behavioral
responses is now under way.

                           83

-------
     Beyond studies  of individual  organisms,  the  responses  of
populations, communities and ecosystems are important to assess.
Because these  environmental units  are not easily  subjected  to
experimentation,  simulation models are often utilized.   However,
models  are  often  limited in  their capability  to predict
responses.

     Unlike the  human  health effects data base,  very  little
information is available on the environmental effects that result
from compounds contributed by MWC  technology.    While EPA  and
other research  organizations support  work on  the  toxicity  of
atmospheric toxicants to terrestrial plants  and animals,  little
of  this  effort  is  directed  to  specific  evaluation  of  MWC
emissions or residues,

D.   Conclusions  and  gecommendati ons
                                                      *
I-   Conclusions

     o The  Subcommittee concludes,  based  on "currently available
information,   that  emissions  from  state-of-the-art,   well-
controlled  and  operated municipal  waste  incinerators are  not
likely to significantly increase  total pollutant loadings to the
environment on a national basis.  However,  background levels will
vary with the sites  selected for  the incinerator plants,  and it
is important to distinguish background levels from new emissions
before  adequate  exposure. and  effects  assessments  can  be
developed,

     o  There are very limited data for evaluating both exposure
and  effects  of  MWC  emissions or  residues.   It is  clearly  not
feasible  to conduct toxicity  teats on  representatives  of  all
species, but effec*"  on animals,  terrestrial plants and microbes
have not been well  characterized  when  compared  to fish  and
aquatic invertebrates.
                           84

-------
     o  More  data are  available for  assessing environmental
effects on individuals  and populations  than.-...pn more  complex
communities and  ecosystems.    Very  little  is  known  about how
ecosystems  respond to  toxicants from any particular  source,
including  municipal waste incinerators.

     o  Exposure data  for some  selected  compounds that are
potentially emitted by  MWC are available  for  evaluating human
health effects;  however,  data on specific isomers  and other
compounds  identified  as  by-products  of  MWC have  not  been
collected.   In addition, mixtures of  compounds  and  associated
interactions  under  environmental  conditions have/ not  been
investigated or evaluated,

     o Throughout its report, the Subcommittee  has separated the
evaluation of  municipal  waste  combustion into various  components
including;   feedstock,  the incineration process,  combustion
                                                      BI
system design,  performance of  pollution  control technologies,
operator   capabilities,   environmental  transport  and  fate
processes,  and potential exposures and  effects.   Each of these
can also be thought  of  as a critical component  for assessing the
risk of this technology.

2.   Recommendations

     o  EPA, private industry  and  other interested organizations
should  initiate  efforts  to  characterize  emissions into the
ambient environment  or   conduct baseline  surveys  through  site~
specific sampling and field monitoring techniques.   BPA should
consider whether to  require such data collection  as part of the
permitting process.    By  accumulating  and  analyzing   such
background data,  the foundation for an accurate comparison of
post-combustion environmental effects can begin  to emerge.

     o  The, Subcommittee recommends  that a higher  priority be
placed on  evaluations   of  environmental  exposure  and  effects.
Individual or  species level toxicity testing should  be conducted

                         '  85

-------
with compounds potentially emitted by KWCs.   This  testing  should
occur in species that  are known  to  be  of primary  importance  in
community interactions,  including terrestrial plants,  animals  and
microbes for which  data are scarce.

     o  The exposure and effects  data  base  for  human risk
assessment should be expanded  to include more of  the  compounds
identified in MWC  emissions  and residues,  as well  as their
transformation products.  The  use of  toxic equivalency  and
structure/activity  relationships  should be expanded  and  refined
through collection of  such  data.    In  addition,  toxicity
evaluations of  mixtures  and products  that  are  predicted from
interactions should improve the risk assessment process.

     To improve the utility of such  investigations for decision
makers,  data should be  generated  and evaluated from  a range  of
incinerators (controlled and uncontrolled, as  well  as*both  new
and old designs) in a variety of locations.  From  such efforts,
scientists can obtain data that can be used to test  the  "ground
truth"  and the sensitivity of  previous  modeling  efforts.    In
combination,  modeling  and measurements will  provide decision
makers  with  more  powerful   tools to  assess  the  relative
contribution of pollutants from MWCs relative  to  other sources,
and the probability of  exposures reaching human and environmental
receptors *
                           86

-------
                VIII.  CONCLUDING PERSPECTIVES

     The Municipal  Waste Combustion Subcommittee  recognized  a
central fact throughout every phase of  its deliberations on the
incineration of municipal wastes  - the  existence of some degree
of hazard,  or  dsk,  associated  with  the • application  of any
technological   alternative  (landfilling,  preprocessing  and
recycling,  etc,)   for managing  the disposal  of  solid  waste.
Limitations  in  current scientific understanding make it difficult
to make precise statements on the  relative risk  from the various
alternatives on  a large scale  and  over time.  Since  every
recognized alternative practice has some associated risk,  it is
important to compare  the risks  posed  by  all  waste' management
alternatives.  Eisfc-based comparisons of the various options can
provide a valuable  perspective to  aid  local decision  makers in
choosing the most  appropriate option for their community.

     The  previous  chapters  have  pointed  out  som6  of  the
deficiencies in the data  base  for  conducting a  formal  MWC risk
assessment.    Nevertheless,  the  Subcommittee finds  that  a
significant amount of research  has been  carried out  on the
biological  and human health effects of  dioxins  and,  that
considerable data  are becoming available on levels of emissions
and on the impact  that control  technologies have on emissions at
incinerators currently in use*  In  addition,  government agencies,
industry  and other researchers are currently characterizing
potential human and environmental  exposures from  air  emissions
and from ash, and  have begun to establish research plans.

     Unfortunately,  the other waste management  options  are also
plagued  by  significant  data gaps,  preventing scientifically
rigorous  analyses  that could   lead  to  comparative  risk
assessments. On the basis of risk assessment  alone,  therefore,
the  Subcommittee  believes  that  no single  waste management
alternative is universally  applicable to  the range  of site-
specific solid  waste problems that  municipalities encounter.  At
the  same  time,  the Subcommittee  concludes that well-designed,
                          87

-------
well-operated and well-controlled incinerators  represent one
alternative that should be available for use by localities.

     As stated at the beginning  of  this  report, the  Subcommittee
did not  evaluate the  other waste  management technologies in
depth,  but it is possible to illustrate  certain commonalties and
differences  among   these  alternatives.    In  Table  6,  the
Subcommittee compares the  relative advantages of  landfilling
(without  prior  treatment),  municipal  waste  combustion and
recycling.   For the purposes  of this  illustration the reader
should  assume that  incinerator ash is  landfilled, that the
portion of waste that can not be recycled is landfille,df  and that
recycling procedures include processing  after  collection.   The
relative  importance of  each issue can differ significantly,
depending in  part upon  site-specific  conditions,   The  relative
advantage for  each  factor,  as  judged by  the  Subcommittee,  is
denoted as (a)  for most advantageous, (b) moderately  advantageous
and (c)  as least advantageous.

     While Table 6 considers only a  limited number of issues, it
illustrates that when different  management  options are compared,
they may  exhibit advantages  or disadvantages,  depending  upon
which issues  are  highlighted.   Comparisons among these options
for a single  issue are difficult.   For instance,  in the case of
groundwater contamination, incineration  may produce  a  major
impact as metals leach out  of ash.   The  same metals are present
in raw, landfilled waste,  but metals become more concentrated in
ash.  The form of metals  also affects their leachability.  In the
case of recycling, the potential problems are  intimately related
to  the  exact  recycling  process   used,  and  can  range  from
negligible to very significant.
                           88

-------
                             TABLE €
    COMPARATIVE ADVANTAGES OP SELECTED WAST1 DISPOSAL OPTIONS
ISSUES
INCINERATION   LANDFILL  RECYCLING
Air Emissions
Methane Generation
Land Volume Required
Transportation
Energy Recovery
Leachable ' Metals

Leachable Qrganics
c
a
a

a
b-c

a
b
c*
c
depends on
c
b-c

c
a
a
_j
f\
distance
b
a
i»
a
Reduction•of Infectious
Agents a
Reduction of Rodents a
Groundwater 'Contamination b
Surface Water Contamination a
Capital Cost c
a » most advantageous
b » moderately advantageous
c * • least advantageous
? * uncertain
+ « methane generation can be harnessed
c ?-c
c b
b a-»o
a a-c
a b-c
to advantage

-------
     o  Conclusion

     All waste management options entail some form of risk. Each
option lias environmental advantages and disadvantages.  Although
the assessment of municipal  waste incineration ggj; se generates
useful information,  the value of that  information  is very limited
when  one  seeks  to  develop  guidance  for  the  selection  of the
preferred waste  management options  in specific communities.  In
order  for decision makers  and  the public  to  make  informed
decisions on  the most appropriate waste  management  options for
their community,  it is important to have  comparative  assessments
of the options that are under consideration for specific sites,

     o   Recommendation

     EPA  can  assist  local  decision  makers and  the  public by
developing ways to collect and analyze data that will allow more
informed  choices regarding  the  management  of  munidipal   solid
waste.  Such support can be provided by  developing approaches for
assessing exposure and by generating  models for assessing  risk*
In addition, means should  be provided or improved to  facilitate
the  transfer  of such tools  and information  to the  parties
responsible for  making the decisions. The Agency should develop
guidance  for  evaluating  individual  waste management  options, as
well  as  comparative  exposure  and  risk  assessments*  between
available options.

-------
                        LITERATURE CITED

[1]  Radian Corporation.  Appendix.A:  Characterization of the
Municipal Waste' Combustion industry.  Draft of-'"October
31f 1986.

[2]  Environment Canada,  The National Incinerator Testing and
Evaluation Programs  Two-Stage Combustion.  Report EPS
3/UP/l.  September 1985.

[3]  Environment Canada.  The National Incinerator Testing and
Evaluation Programs  Air Pollution Control Technology,
Report EPS 3/UP/2.  September 1986.

[4]  Midwest Research Institute.  Results of the Combustion and
Emissions Research Project at the vlcon Incinerator Facility in
Pittsfield, Massachusetts.  Project No. 8649-L(12) for New York
State Energy and Development Authority.  May 1987.

[5]  Denmark National Environmental Protection Agency. Formation
and Dispersion of Dioxins, Particularly in Connection with
Combustion of Refuse.  December 1984*

[6]  Hahn, Jeffrey L.  Testimony before the California Air
Resources Board at a Public Hearing to consider the adoption
of a regulatory amendment identifying Chlorinated Dioxios
and Dibenzofurans as toxic air contaminants.  July 25, 1986.

[7]  Clarke, Marjorie J. "Emission Control.Technologies for
Resource RecoveryH, presented at the Symposium on Environmental
Pollution in the Urban Area, Brooklyn Polytechnic University.
March 15, 1986,

[8]  Howes, J.E.; Kohler, D.f.,* DeRoos, F.L*l Biggin, R.M.; and
Barbour, R.L.  Characterization of Stack Emissions from Municipal
Refuse-to-Energy Systems,  Prepared by Battelle Columbus Labora-
tories for U.S. EPA under Contract No. 68-02-34SS,  Draft of
May 16, 1985.

[9]  Hanson, James C.  U.S. EPA Region 9, 2IS Fremont St., San
Francisco, Ca  94105.  Personal communication to the Subcommittee,
April 30-Hay 1, 1987.

[10] Hasselriis, Floyd.  Technical Guidance Relative to Municipal
Waste Incineration.  Prepared for the Task Force on Municipal
Waste Incineration, New York State Department of Environmental
Protection. August 1985.

[11] Environment Canada.  The National Incinerator Testing and
Evaluation Program.  Mass Burning Technology Assessment*
September 1987, Statistics published separately IN:  NIT2P
Update, November/December/January, 1987.
                           il

-------
[12] U.S. EPA and Environment Canada.  Joint Project on Refuse-
Derived Fuel Combustion Technology.  Status Report INs  NITEP
Update i  November/December/January 1987.

[13] Briggs, G.A.  Plume Rise Predictions IN Lectures on Air
Pollution and Environment Impact Analysis,  Workshop Proceedings.
American Meteorological Society, Boston, Massachusetts.  1975,

[14] U.S. EPA. Guidelines for Determination of Good Engineering
Practice Stack Height (Technical Support Document for Stack
Height Regulations). Report Mo. EPA-45Q/4-8Q-Q23, Office of Air
Quality Planning and Standards, Research Triangle Park, N.C.
27711.  1981.

[15] U.S. EPA. Guideline for Use of Fluid Modeling to Determine
Good Engineering Practice Stack Height, IPA Report No.
EPA4SO/4-81-003, Office of Air Quality Planning and Standards,
Research Triangle Park,  N.C.  27711 (NTIS No. PB 82-145327).
1981.

[3,6] Turner, D.B.  Atmospheric dispersion modeling:  A critical
review, J. Air. Pollution Control Association 29:318-519, 1979.

[17] Hanna, S.R. ; G.A. iriggs, and R.P. Hosker, Jr.  Handbook on
Atmospheric Diffusion.  Pub. No. DOI/TIC-11223 (De 82002045) tJ.S.
Department of Energy, Oak Ridge, Tennessee.  1982.

[is] Hosker, R.P. and S*E. Lindberg.  Reviews  Atmospheric
Deposition and plant assimilation of gases and particles.
Atm Env. 16 f 5) t  899-910.  1982.

[19] McMahon, T,A. and P.J. Dennison.  Empirical Atmospheric
Deposition Parameters - A Survey   Ajfemoa^ Environ. 13 i
571-585.  1979.

[20] Beychok, Milton R.  A data base of dioxin and furan
emissions from municipal refuse incinerators.  Draft Manuscript
submitted to Atmospheric Environment. 1986,

[21] Higgins, Gregory M.  An Evaluation of Trace Organic Emissions
from Refuse Thermal Processing Facilities.  Systech Corp. Xenia,
Ohio.  1982,

[22] Amendola, Gary A.  Soil Screening Survey at Four Midwestern
Sites. U.S. EPA-905/4-8S-OQ5, 1194. June 1985.

[23] Isensee, A.R. and G.E. Jones.   Absorption and translocation
of root and foliage applied 2,4-dichlorophenol, 2,7-dichloro-
dibenzQ-p-di<-jdln, and 2,3,7,S-tetrachlorodibenzo-p-dioxins,
   Agr. Food Chen^ 19: 1210-1214.  1971.
[24] Crosby,, D.G.? A.S. Wong, J.R. Plimner and E.A. Woolson.
Photodecomposition of chlorinated dibenzo-p-dioxins. Science
173 i 748.  1971.
                           92

-------
[25] Crosby, D.G. and A.S. Wong. Environmental Degradation of
2,3,7,8»tetrachlorodibenzo-p-dioxin  (TCDD). Science
195! 1337. .1977,
[26] Plimmer, J.M.   Photolysis of TCDD and trifluralin on
silica and soil, Bull.... EjnvJ.ro.nm.j., eonfeam^ Toxicol..
20: 87-92. 1978.

[27] Kearney, P.C., E.A. Woolson, A.R. Isensee and C.S. Helling.
Tetrachlorodibenzodioxin in the environment;  Sources, Fate  and
Decontamination, Environ. Health Perspectives. 5; 273.  1973.

[28] Helling, C.S,  Pesticide mobility in soils It.  Applications
of soil thin-layer chromatography, Soil Science Soc. Aateg. Proc.
35: 737. 1970.

[29] Matsumura, F, and H» Beneiet.  Studies on the bioaccumila-
tion 'and microbial degradation of 2,3,7,8-tetrachlordifaenzo-p-
dioxin, Enjj.rojCbL Health Perspectives. 5: S23.  1973.  '
[30] Isensee, A.R. and 6.E. Jones.  Distribution of 2,3,7,3-
tetrmchlorodibenzo-p-dioxin (TCDD) in an aquatic ecosystem,
Environ. Sci. Technol.. 9(7) 667.  1975.
[31] Young, A*L*r et.al.  Fate of 2,3,7,8,-Tetraehlorodibenzo-p-
dioxin (TCDD) in the Environments Summary and Decontamination
Recommendations.  USAFA-TR-76-18.

[32] DiDomenico, A,, V. silano, G* Viviano and G. Zapponi,
Accidental release of 2,3,7,8,-tetrachloro-dibenzo-p-dioxin
(TCDD) at seveso, Italy, II,  TCDO Oistribution in tne soil
surface layer, leotoxic. Enviro. Safety.. 4:  298-320.
1930a.

[33] DiDomenico, A., V. Silano, G. Vivano, and G. Zapponi.
Accidental release of 2,3,7,3-tetrachlorodibenzo-p-dioxin
(TCDJ5) at Seveso, Italy, IVj  Verticle Distribution of TCOD
in soil, ficotoxic* Environ* Safetyt 4t  327-338.  I980b.

[34] Bartelson, F.D., D.D. Harrison, and J.B.  Morgan.  Field
Studies of Wildlife Exposed to TCDD Contaminated Soils. Air
Force Armament 'Lab.  Eglin Air Force Base, FL.  1975.

[35] Ward, C* and P* Matsumura* Fate of 2,4,5-T Contaminant
2,3,7,8-tetrachloro-p-dioxin  (TCDDJ in Aopiatic Environments.
NTIS Pl-264187. 1977,

[36] Yockim, R.S., A.R. Isensee, and G.T. Jones.  Distribution
and Toxicity of TCDD and 2,4,5-T in an Agtiatic Model Ecosystem.
Chenosphere. 7{3)s 215-220.  1978.

[37] U.S. -EPA.  Dioxins, Volume 1.  Sources, Exposure, Transport
and Control., .EPA-600/2-80-156. June, 1980,
                            93

-------
[38] Corbet, R.L., G.R. larrie Webster, and D.C.G. Mxiir.  Fate
of 1,3,6,8-TetrachlorQdibenzo-p-diQxin in an outdoor aquatic
system.  Environ. Tox*_ Chem. 7f 167-180. 1988.

[39] Seever, W.R., Lanier, W.S., and Heap, M.P,  Municipal Waste
combustion Studyi combustion Control of Organic Emissions.
Energy and Environmental Research Corporation, Irvine, CA.
Draft Report to EPA, January, 1987.

-------
                        Appendix A

ASSESSING  SPA'S RISK  ASSESSMENT METHODOLOGY FOR  MUNICIPAL
     INCINERATOR EMISSIONS? KEY FINDINGS AND CONCLUSIONS
                            A-l

-------
           ASSESSING EPA»S RISK ASSESSMENT METHODOLOGY
              FOR MUNICIPAL INCINERATOR EMISSIONS!
                 Kay Findings and Conclusions


Introduction

     On November  10,  1986,  the  Municipal  Waste Combustion
Subcommittee of the  Environmental Effects, Transport and Fate
Committee of EPA's  science  Advisory Board  reviewed a  draft
document entitled "Methodology for the Assessment of Health Risks
Associated  with Multiple Pathway Exposure to Municipal  Waste
Combustor Emissions"  prepared  by the  Office  of Air Quality
Planning and Standards  (OAQPS) and the Environmental Criteria and
Assessment  office'(ECAO). -This document will be referred to
hereafter as the "methodology".

    The purpose  of  the risk assessment and exposure methodology
developed  in the  document  under review is to  examine the
potential health and environmental effects exposed populations
are  liJcely  to  experience as  a  result of  municipal  waste
combustion (MWC) technologies. This asessment allows comparison
of variations in the efficiency of combustor design and operation,
and  is  also intended  to  predict  the effects resulting  from
multiple exposures to emissions  from more than one source,

     OAQPS and ECAO requested the Subcommittee to evaluate the
scientific validity of the methodology for assessing health
risks associated with multiple  pathway  exposures to  municipal
waste combustor emissions.  Specifically,  the Subcommittee was
asked to determine  whether the methodology provides a reasonable
scientific approach to evaluating effects on public health  given
the available data, the validity  of exposure assessments, and the
appropriateness  of  transport and dispersion  models.  The
Subcommittee's key  findings' are  reported  in the  following  pages;
detailed comments  and meeting transcripts have been provided to
appropriate Agency  authors.

General Comments and Methodology  Overview

     overall,  the Subcommittee considers the proposed methodology
to be conceptually thorough,  although it  identifies a number of
areas where specific technical  improvements  are needed.   Since
the methodology  will  be used  as  a technical support document for
regulatory  decision  making,  a thorough technical  effort is
necessary. The approach also makes reasonably effective  use of
existing scientific data  and exhibits the degree of accuracy and
understanding  needed for using models.  The Subcommittee
consensus is that the methodology is a credible effort towards
developing  a  tool for assessing multiple media exposures from
this source category.

     The Subcommittee commends the  authors  on both the tone and
the detail used in documenting  the assumptions  that support the
       kl nrrv.  f**»  tmrtArta i nt 1 £S and DOSSible  COtlSeCfUejlGeS Of

-------
using  the methodology are  clearly  presented in a number  of
instances,,  such as limitations created by  focusing  on stack
pollutants  rather than  total -pollutant  loadings  (e.g.,  ash
residues,  aqueous residues,  and stack emissions).  Another con-*
earn is tha uncertainty in identifying specific pollutants in
emissions from a municipal waste combustor, since characterizing
emissions improves the ability to predict the physical and
chemical  'properties  and effects of emitted substances.   the
authors are clearly aware that the methodology they have deve-
loped is'but a step in a development process to expand current
risk assessment methodologies  to include other pathways,  in
addition to  atmospheric, exposures beyond inhalation and non-human
effects.

     The Subcommittee has  several recommendations for placing Che
scientific issued raised  by the use  of this technology into
better perspective.  These  recommendations includes

         0 The methodology should attempt to predict tha risk
posed  from both combustion as  a  whole and  from  specific
activities, such as automobile use, industrial practices (e.g.,
coal combustion  for  energy production) /  and  both* hazardous
chemical and municipal incineration.

         0 While  individual scenarios are  modeled  in this
methodology,  calculating  dose  from  the source and dispersal
through various pathways does not lead tha reader to understand
the entire  risk perspective  that  incineration technologies
present.
          ^
          °  In applying the models,  the  methodology utilizes
two separata sites as examples: 1) Hampton,  Virginia, and 2) a
proposed,  or  hypothetical,  state-of-the-art facility to be
located  in  Florida.   Although  both  sites  are individually
discussed and  evaluated as  to tha,risks they presumably pose,
they are  not compared. Since risk assessment is a comparative
tool,  the. Subcommittee  recommends  that  the chosen  sites be
evaluated in comparison' to  ona another,  and for reasons to be
discussed later,  recommends that  facilities in addition to
Hampton bat .used for this comparison.

          °  Tha subcommittee believes  that the most appropriate
data for monitoring MWCs  may be derived from combining actual
field, measurements with predictions from mathematical models.  For
the field aaasuremants, .this presupposes that measurement^ nave
been made in appropriate  locations,  at appropriate times, and
with  appropriate methods. It  also presupposes,   for the
mathematical  models,  that they have been validated at least to
the extent'that their  limitations are understood and that the
range of divergence between modal  predications and reality can be
quantified. In this context it  is important to consider both
statistical  variability and its propagation through tha model, as
well as conceptual biases  which inherently result  from making the
                             A-3

-------
 simplifying assumptions required for the construction  of models.
 The Subcommittee recognizes that elements of this recommendation
 are best carried out through a longer-term research program.

     The document should definitely state that,  even when models
 are validated,  actual data are preferable to results predicted by
 models.  Also,  the methodology should caution that  the existence
 of a useful  model should not substitute for or discourage the
 collection of  site specific data. In addition, the methodology
 should encourage the use of field data and model application in
 concert.

     The  methodology  appropriately states  that  much o'f the
 information  needed to  further support the methodology does not
 exist,  and that some assumptions about non-existent 'data must be
 made to make initial predictions.  However,  the specific choices
 in such assumption raise several questions  for the Subcommittee,
 which are addressed in  the sections to follow.

     That Subcommittee recommends that uncertainties be identified
 as to  whether  they  are the  result  of limitations in the
 understanding  of the  MWC  process itself,  or a  result of the
 predictive capability of the modal.            .      *

 Technolocry and  Emissions

     The document cover attempts to represent a broad  perspective
 of exposure patterns.   However, the Subcommittee is concerned
 that the drawing depicts a worst-case exposure scenario without
 illustrating  the problem-solving aspects  of the technology.  This
 concern centers around the negative impression that  may result
 from the depiction of a particulate  emissions plume.  It was also
 noted that the illustration represents a rural setting,  and does
 not depict the  urban environment, where most incinerators may be
 built.

     The methodology reviews the state-of-the-art for existing
 and projected  municipal wasta combustors,  and provides useful
 background information* However,  various sections on existing
 and projected  facility sites are  inconsistent with  regard to
 future  locations.   In addition, projections for California may be
 misrepresented. The Subcommittee believes that  it is important to
 distinguish between the number of  facilities and  the number of
 incinerator  furnaces,  since most facilities consist of several
 incinerators  that can be operated independently.

     Using a  combination  of dry  scrucber "and  fabric filter
 technology for pollution control is reported to reduce mercury
 emissions  -by 50  percent. Data actually demonstrate  that at 140
degrees  Celsius  (C)  or below,  §5-97  percent  collection is
achieved,  while at 20i degrees C,  no collection is achieved. The
average may be 50 percent, but averaging this type of data does
 not accurately  represent  the performance of the control system.
                              A-4

-------
    The methodology discusses many factors that may influence
emissions*  The apparent!  and ultimata conclusion  is  that  the
efficiency  of the air pollution control system  determines the
emission level  of  particulate  natter  (PM) and associated
pollutants   from the -stack.  This conclusion should be clearly
stated«

     The-Subcommittee disagrees  with the use  of  the Hampton
facility to represent existing incinerators and their emissions.
Use of this inappropriate  example will yield a gross overestima-
tion of emissions from new incinerators. The Hampton data set may
be extensive,  but the technology used at the facility" is hardly
representative of typical mass burn technology. The design'and
operating practices used  at Hampton should be explained, -along
with  the fact that this design  is not  in  common  use.  This
facility provides a worst case scenario that  is  not representa-
tive of most  recant  installations.  The  results of modeling will
be very'  different when' best available control technology  (BACT)
is used*  The Subcommittee recommends that  EPA  develop more
scenarios,  including one for BACT,  that  can be used to evaluate a
more complete range of source and emission characteristics for
existing and proposed MWC  facilities.

     The methodology cites three reasons to explain the presence
of polychlorinated dibenzo-dioxins  and  furans  (PCDDs  and  PCQFs,
respectively)  in MWC flue  gases.  A fourth reason should bo  added,
since these organic compounds may be formed in the boiler  during
cooling,  in the presence  of  fly  ash (post-combustion formation).
It should also be stated that little is  known about reactions
that occur between gaseous species within emission plumes.

     The methodology recognizes that the available emissions, data
are limited in both quantity and quality.  Few specific chemicals
have been identified, although much of the total mass has been
characterized as silicates and forms of carbon.  There is  reason
to suspect that some of the chemical components of MWC emissions
that remain to  be identified may be toxic.   However,  these
chemical components, such as polyaromatic hydrocarbons (PAHs),
may be contributed by sources other than municipal  incinerators,
and background levels are not adequately established*  Major  data
gaps exist with regard to  chemical identity, toxic potential, and
total  environmental burden of MWC  emissions,  making the
assessment  of risk posed by  the  technology  itself,  and in
comparison to other alternatives, difficult to predict*

Exposure Models

          0  Industrial Source Complex (ISC)  Model

     The introduction to  the  ISC model would be improved  by  a
discussion  of the  likely uncertainties  of  the estimates for
models of gaseous dispersion, particle dispersion, and wet and
dry deposition of gases and particles.  This discussion  should


                            A-5

-------
address uncertainties  that arise both as  a  result of limitations
in the understanding  of the processes and those due  to random
variation in deposition  and dispersal processes.

     Although some of  the assumptions made  in parameterizing wet
deposition  may be rather crude (e.g., assumptions regarding the
spatial distribution  Qt precipitation), they are not likely to
present a  problem when annualized computations  are made.
However,  the parameterization of dry  deposition,  particularly for
emission of chemicals for which loss mechanisms are not under-
stood, is not  clear. The methodology seems  to  imply that gaseous
components are not considered*  This, point  needs to be  clarified.
The use of  data concerning  the size distribution of  particles
obtained from the Braintree  MWC  may not be representative, and
the data on emission rates seem to be conservative.

      The methodology  for atmospheric dispersion  and  deposition
of emissions  should separately consider  particulate and gaseous
emissions  and their fate.   The contribution  from chemicals in
different physical and chemical states should be  evaluated vita
respect to to direct and indirect routes  of exposure, variability
in the size and solubility of particles should be considered. The
biological  availability  of emitted materials is  also affected by
the degree of sorbtion to particles  that occurs.   The  discussion
should specify the assumptions made about emission characteristics.

    The effects of buildings on lateral  and vertical  dispersion
of emissions  has been considered in the methodology.  However,
careful consideration of  downwaah is  also  necessary. The
proximity of  other structures in urban  areas and the potential
for downvash are not treated in the methodology. Since one of the
strengths of  the ISC model  is the ability to consider multiple
sources,  the document  should  also address the  issue of the
proximity of other incinerator  facilities.

     The methodology does not consider the  exposure of  people who
do not reside at ground level.  This factor could be significant
for urban residents,  and id compounded by  the likely  concentra-
tion of incinerators in  urban settings.

          ° Human  Exposure Model (HEM)

     The HEM is used to  estimate the carcinogenic risk posed to
populations by inha."1' cion of predicted ambient air concentrations
of MWC emissions.  The  model assumes equivalency of  indoor and
outdoor concentrations,  an assumption that  the Subcommittee finds
suspect for  two  reasons:   1) the  finite length of typical
infiltration rates (>  1  hour, typically), and  2) the significance
of indoor sources  of certain chemicals,

     The HIM  estimates  do not consider the short or long-term
mobility of the population.   It also assumes  a  70-year lifetime
for MWCs.   in other parts of the methodology, a more realistic
30-year estimate  is  utilized.  The  assumption of  continuous
operation of MWC facility is also an unrealistic  assumption.

-------
     Specific aspects  of  the locality and siting  of the MWC
facility  n«ed to be considered because  of their significant
effect on concentration and dispersal of pollutants.

     The document should refer to the discussion of Quantitative
risk assessment  modeling  found  in EPA revised guidelines for
cancer in order to provide  the reader with a better understanding
of  the  range of  assumptions and models  used  in cancer  risk
assessment.

          0  Terrestrial Food Chain (TFC)  Model

     This model  is  used to predict the  deposition of MWC
emissions on soil  and vegetation*  Its pathways  assess the
exposure  to humans,  animals, soil  biota and vegetation, and
associated effects on the food chain.   Tha  TFC model has separate
components for examining the potential  for human  exposure from
ingesting contaminated soil and from consuming vegetation and
animal tissues containing the contaminants. The potential for
children  to be exposed as a result of ingesting soil ia also
estimated. However, pathways of human exposure via consumption  of
herbivorous animals  are not clearly explained. The assumption
that  herbivores are exposed  only by  ingesting  soil  or  by
consuming  plants that  have assimilated emitted materials
deposited on soil  neglects consideration of the component
presenting the highest exposure potential.   Herbivores are likely
to receive the highest  exposure  from  ingesting leaves of plants
upon which particulate emissions  have been deposited.

     The Subcommittee  questions the appropriateness  of  using
sludge  or pesticide amendment  practices as  surrogates for
predicting  fallout  from MWC emissions.   The burden  of  toxic
compounds and metals  that is created by applying sludges to soils
should be compared to that presented by the assumption that rates
of dioxin or furan emissions will equal or exceed  2.7 kg/ha over
50 km linear dimension as a result of MWC.

     This modal uses a hypothetical Florida MWC as  an example for
making predictions,  but  the input  factors,  such as  rates  of
emissions, soil characteristics,  and design and operation,  are
not documented.   It is  not clear whether tha Florida KWC
represent* a best or worst case illustration. More exposition is
neaded  with respect  to both input and output parameters*   These
improvements  would greatly enhance  the reader's understanding of
the methodology,
                              A-7

-------
          0 Exposure Pathways

     The assumptions  required  for determining the maximally ex-
posed  individual  (ME!) need to be considered' more carefully to
prevent th« overcanservatism which may result  from combining the
basic  MSI  concept with  those  resulting from  the multi-exposure
models. The  MEI concept estimates the effect on only one
hypothetical human subject,*  population effects and effects over
generations are not determined.  The ME1 concept also does not
consider  acute exposure  or exposures to other biota.   These
oversimplifications  result in conservative estimates of human
exposure.   A new  concept should also be developed • which includes
the cumulative probability of ME1 exposure.

    'Another flaw in the methodology is the assumption of flat
terrain.  Urban or hilly settings may, in actuality, result in
greater levels of  human exposure.

     The methodology  does  give  appropriate consideration to soil
type.  Soils differ greatly, making tha selection of a specific
standard soil density and penetration depth tenuous. Compounds
from MWC emissions will be deposited at different concentrations
and will be found at varying  depths in tha soil,  depending on
soil type.  Assumptions that-toxicants will ba concentrated in
tha upper centimeter  of  soil.may b*  incorrect for some locations
because of differences in soil  density, moisture and composition.
Some toxicants will be  concentrated  near tha soil surface, while
others may move down  from  tha surface and  be dispersed.

     Degradation of chemicals  in soil is oftan assumed to be a
first-order reaction,  even when data for specific chemicals
indicate that the degradation  rate  la not first order. The best
available kinetics should ba used, since first order kinetics may
often ba inappropriate*

     In the methodology, trace  metal contaminants are assumed to
persist  indefinitely unless loss constants ara available.  A
reasonable  loss  constant, which can ba derived  from  soil pH
values, should b* used  instead of making  a blanket  assumption
that contaminants will persist*

     Assuming that no dagradation and no retardation takes place
for chemicals in the plow depth layer is of concern when there is
a lack of data to support this  assumption.  Tha fata of chemicals
is known to b* altered  in  plow depth layers composed of organic
clays as a result  of  biologic activity.

          °. Surface/Ground Water Models

     Tiar one of the surface/ground water methodology assumes
that all material deposited  during a single year is incorporated
into the water in the same year.  This modal does not take into
account the potential for  build-up over periods of  more  than one
year,  or tha potential  for this large amount of material to be
released by a single storm event at some future time.  In drier

                               A-a

-------
climates  (i*e.^*-the Intermountaln-West and the<>>,Southwest deserts)
major storms or "gully washers" can occur as seldom as once in 10
years,  rendering doubtful  the  assumption that  all  toxicants
adhering  to  particulates are  flushed out in a ana year period.
Furthermore, in  wet climates  the opposite may be true, as some
toxicants may not build up appreciably.

          ° Other Exposures Not considered

     As the  authors point  out, no consideration  is given to
exposures from .land-filling ash* 'Similarly,  consideration  is not
given to  the potential for  change  in emission  characteristics
that  nay  result  from incinerator upsets.  These data gaps are
significant,  but consistent  with the inadequate  knowledge
regarding' HWCs. The  Subsommittee recommends that the methodology
address these issues.

Estimation of Risk to Humans .

     The equation used to calculate  the adjusted reference  intake
(BIA)  is logical  for application, since the use of the acceptable
daily intake (ADI) is veil established.   Also, the use of  excess
concentration over background  in the equation is  an established *
measure of the potential  for human  health effects.  However, the
definition of total  background intake (Til) of pollutants from "all
existing, sources  needs  some  clarification.

     Examples presented in the methodology use national averages
to  define  the  TBI, although  these  values may not be
representative of 'the particular  sites where  risk is  to be
evaluated.   The approach taken for risk  assessment is based on
the location with  the minimum RIA,  although  people at this
location  may not be those with the  maximum  exposure to the
pollutant. The  Subcommittee  believes that the values selected may
not be valid for the particular sites being evaluated.

     Defining the TBI aa the sum of contributions from individual
sources  assumes  that no.interactions,  such as synergism or
antagonism,  occur when sources are combined  and individuals are
exposed by multiple  routes.  There are  many instances  where  this
concept i* not supported  by  the available data.

     Thar* is inconsistency in the methodology's treatment of
exposure  to background  concentrations  of different chemical
substances.  For some chemicals, such as cadmium, contributions
from MWC emissions are added to contributions from all background
sources to give  total exposure.  For other substances, suen as
benzo(a)pyrene,   exposures to background  concentrations  are
ignored and  assessment is conducted in terms of additional  risk
posed by MWC contributions alone.   The methodology should  assess
exposure to chemical substances in a consistent manner*

     The prediction  of inhalation exposure, which assumes that
individuals  are  exposed to emissions  only in gaseous  form,
neglects  the potential for  particulate  absorption and particle

                              4-9

-------
deposition.  Pathways other  than inhalation, such  as dry
depositon of particulate emissions and related dermal absorption,
need to be considered.

     The methodology postulates that some nonearcinogenic  effects
that exhibit thresholds  occur only after nearly an entire  life-
time of exposure.  This assumption does not reflect the actual
situation.  For  example,  fibrotic lung diseases  occur after less
than a  full life  span  of exposure,  and  their onset is very
gradual.   For  many chemicals, the  reported latency periods tend
to be measured in terms of weeks or months,  rather than years,

     Relative effectiveness (RE)  is used in the methodology to
standardize effects of exposure by one route to the effects of
exposure by another.   There may not be  scientific justification
for this conversion factor. However,   the concept is  useful as
long as users realize that the effect  of  an exposure does not
relate solely to absorption efficiency, but is also related to
differences in the  sensitivity of absorption sites to damage, and
to differences in toxieokinetics between exposure routes. The
methodology should  acknowledge the assumptions required for using
this approach.

     Consumption  of fish  by the general population is discussed,
but the discussion  does not take into  account the fact that fish
may  come from a  variety of sources  with varying degrees of
contamination.  A  similar situation exists for drinking water,
Drinking water obtained from any  one tap  may  consist of  water
from a local source, may contain water that originates outside of
the  localized delivery  area,  or  may  be a  mixture of  both,
Alternatively, drinking water  may be obtained from individual
wells drawing  on  ground water from a large  source or deep aquifer.
Local contamination is not always represented in the  localized
supply of drinking  water.

       With regard to water consumption,  the  amount of  fluid
intake documented  ia  low.  It is  not clear whether this amount
represents total fluid intake or the intake of water alone. It is
usually assumed that  fluid intake for adults averages 2 liters
per day.  Ill is questionable, therefore,   that females between the
ages of 14  and 16  would only take in  586  ml water per day,  as
reported In the document.

Ecological Effects

     The treatment of  plant  uptake as a linear function  is
erroneous  unless  no  other information  ia  available.   Many
toxicants,  especially metal salts,  are  actively transported
across  membranes  or cell walls  and, therefore,  cannot  be described
by a linear function.

     The Subcommittee disagrees with  the assumption that  plants
are exposed to contaminants mainly through uptake from soil.
Greater  exposure  is  likely to occur  from  foliar deposition.
Estimates of  deposition  can be obtained  from acid deposition

-------
studies  and also from studies of the nuclear energy industry,
«.g.,  deposition  of radioiodine  (11J1).

     The Subcommittee also cpiestions  the  method used to  average
bioeoncentration data for aquatic species. Even when means are
calculated separately for bivalves  and fin fishes,  misleading
interpretations can result.  The bioconeentration data should be
correlated with human, dietary  factors.  For example, humans con-
sume  more  oysters than mussels, and oysters may  accumulate
significantly more contaminants  than mussels.  Averaging  bioeon-
centration factors together for oysters and mussels may create a
significant  source of  error  in  calculating  exposure  to
bioaccumulated chemicals.
                                                  ^
     The document summary  mentions measurement of adverse  effects
on  natural ecosystem vitality. The definition of  ecosystem
vitality is unclear,  as  are the endpolnts to be used  in measure-
ment.  Uptake from  water is modeled,  but few other environmental
endpoints are considered,  on*  important component not treated is
the highest  trophic  level,  predators.   Predators play an
important role in community regulation. There is also a  need to
consider  the  potential  for concentration of materials in
sediment,  since sediments  may  serve as a source of contamination
for overlying waters,,  and  materials concentrated  in sediment may
be biologically available to benthic organisms and organisms
dwelling in the water column.  Assessments of exposure cannot be
derived from water «|uality concentrations for benthic dwellers,
since they are exposed in  a totally different way.

     In closing,  the Subcommittee agrees that the methodology
represents an  appropriate step  towards  modeling and predicting
exposure from. MWC emissions.  Some conceptual assumptions can be
strengthened by closer examination of the complexities associated
with pollutant emission to and interaction with the environment,
while others must  await collection of actual field data  to fill
in knowledge voids and elucidate environmental interactions.
Finally,  the methodology, over time,  must be  validated with actual
data to evaluate and demonstrate its utility, and to guide its
further development and  refinement*'

-------

-------
                        Appendix B

DIQXIN  TOXIC EQUIVALENCY METHODOLOGY  SUBCOMMITTEE  REPORT'
                      EXECUTIVE SUMMARY
                             B-l

-------
A.  Major Subcommittee Conclusions

     EPA has proposed intarin procedures for estimatir*g health risks for

CDDs and CCFs based on the premises chats (a) toxicity equivalence factors

can be assigned'to untested (at incanpletely tested) compounds on the basis of

structure/activity relationships? and (b) the toxicity of mixtures of% these

compounds can be approximated foe policy purposes by the sm& of their TEF

times concentrations.  Empirically, the present proposal falls generally

between the positions adopted by certain European countries, which rank
                                                          *
2,3,7,8 TCDO far above any other congener in toxicity, and that initially

proposed by the state of California, which equates all the dioxin congeners.

All have,used similar scientific assumptions in developing policy.

     the Subcommittee agrees that the congeners of CDDs and CDFs constitute

a class o£ chsnical substances, that "share similar structural relationships

and qualitatively similar toxic effects and, therefore, can reasonably be

considered together.  Fran the limited toxicologic data available it seans

reasonable, too, to consider those tetra-to hexa-chlorinated compounds with

chlorine substitutions at the lateral 2,3,7,8 positions as a closely related

subclass in teens of biological activity and environmental fate.

     The Subcommittee also concurs that the problems in assessing the health

risks of difaenzo-g-dioxins and dibenzofurans are two-fold.  They include:

limited infonuatiofi from human or experimental studies about the hazards from

exposing to these cctipourtds (few of the 75 CDCB and 135 CDFs have been tested

at all) and even more limited  information about their possible  interactions

in nixtur^' *  Indications of interactions, mostly additive, are found in

certain experimental model systans  (e.g. binary combinations).  Not addressed

in the draft document, however,  is  the possibility of chemical and toxicologic

-------
 interactions with other types of  compounds  in complex envirensBp.cal mixtures,


 especially solvents that might affect uptake and retention by the body*


 EPA should address the latter subject in the TEF document, perhaps with


 warn specific reference to its recently published Risk Assessment Guidelines


 and to three National Acades^ of  Sciences'  reviews on toxieological interaction:


 the last of which is currently being prepared for SPA and the National


 Institute of Environmental Health • Sciences.  The Subeonraittee also questions


 the basis for including or excluding other  chemicals with effects siiailar


 to CDOs and CDFs, such as chlorinated biphenylenes.


      Based upon its review of the draft document, the Subeomittee concludes


 that the method proposed by SPA is a reasonable inte_riat approach to assessing


 the health risks associated with  exposure to mixtures of CDOs and CDFs for


• risk management purposes.  It Is  necessary, however, as lessons are- learned
                     *                 *                   S.               "

 fron toxicologic research and froa application, the -approach should be-


 re-evaluated systematically by EPA.  Moreover, attests should be made to


 validate the method by selected experimental testing of hypotheses.  For


 example,  more.data are needed on  in vivo potencies of additional PCDDs and


 PCDfs to caspare with in vitro test results,  the asstuption of additivity


 can be evaluated by conparing observed activities with predicted activities


 in selected tests.


      The Sufacamiitted reeomnends  that EPA place more erophasis on the interim


 nature of the method in the document.  The  Subcommittee anticipates that,


 eve? tl»,  the method will be modified and  eventually superseded as pore


 preeisa data becone available. Meanwhile,  the general method proposed
                        *
 appears to have utility for this  and for other classes of closely related


 compounds where toxicologic data  are  inccn^lete.  Application' of structure
        t
 activity relationships is an old  and established practice of demonstrated


 usefulness in pharmacology and toxicology.
                                B-3

-------
     However, EPA should  not abandon its exploration of other approaches to


            risks  foe substances  in mixtures.  For example, where variability


in the cctrposition of environmental sauries is not wide, a reference standard


approach might be used  (similar  to those used in toxicology for selecting a


reference cigarette or  a  representative blend of gasolines).  As another


example, the incorporation of a  staall anount of radiolabeled test compound
                                                          >

into a representative and defined mixture might be one useful way of determining


in vivo whether the uptake and metabolism of one congener is-greatly modified


by the presence of other  substances in a mixture.


     Sorts additional technical comments that the Subconaittef wishes to draw


to the Agency's attention include: 1)- perceptions b£ aany Subcommittee    •  -
                                               *                      ~"

members of an over-reliance upon the postulated mechanisms of the' Ah


receptor/AHH enzyme induction upon which to gauge relative and absolute


toxicity; 2) the  need to  discuss the work of Matsinura, Roaran, Greenlee,


Poellinger and others on  additional toxieologieally significant effects of


the dioxins other than  those associated with receptor binding or with


cytQchroti* p-450  induction? 3} observations of a disassociation between AHH


induction and cytotoxicity in studies on the gonacto toxicity of TCDDj and


4} examination of the extent to  which the longer biological half-life of


higher chlorinated dioxin iseraers, as compared to 2,3,7,a-TO»» counter-


balances their lesser in  vivo potency.


B.  Major Subcommittee  Recoimendatif j>


     The Subccranittee has several recenmendations for  improving the  report.


First, the draft  report narrative is relatively brief  and may not be


readily understood by those not  familiar with dioxins.  For example,  four

-------
possible approaches are  introduced and one (TEF) selected, but the document



dees not clarify what the other three ape-caches- are and the reasons for



their rejection".'  The first approach, long-term animal testing, might be



appropriate  for municipal incinerator fly ash, where analytic data suggest



there- is a characteristic pattern of composition.  The second approach



(short-tern ass^s) is not clearly described (no* even whether they are



in, vivo or in vitro).  The third approach, additivity of-the toxicity of



components, is at first  rejected in the narrative but then forms the



basis for handling the equivalents to 2,3,7,8-TCDD in mixtures.



     Because the draft document presents a procedure, it is essential



that the decision steps  be clearly articulated, the assmgtiens made*
               *               .     «    *  *               *
                       *                            *              *    - -i

explicit, and the mechanics of calculating be illustrated in a stepwise



fashion.  To approach the subject from Che viewpoint of studying the



whole class of pollutants and to avoid bias by selecting data, the Subcom-



mittee recommends that the tabular data be enlarged to include all compounds
             +


with zero to eight substituted chlorines.  Biological activity has been



reported for di- and tri-CDOs, and caccinogenicity studies exist for DD



and 2,7 DCDD, as exaiqples.  Moreover, the activity of brcminated and



other substituted cotpounds should also be indicated and a specific



effort encouraged to collect data on non-chlorine substituted coipQunds.



     la contrast witil the document's first priority on carcinogenic and



then on teratolcgic effects in animals, the Subcoiwaittee"recoroiinds that



the TEF methodology assign first priority to human data when it exists.  In



       ircj experimental  data,  EPA should continue to follow the current
                               B-5

-------
toxicologic practice of evaluating all endpoints, and selecting the ones



post reliable, "sensitive, and important for risk assessment.  Thus, columns



should be added to the tables in the document for other important toxic



endpoints including unmmotoxicity, thymic atrqphy, body weight, and enzyne



induction in vivo.  The limited data points frcm which TEFs are currently



derived (e.g. earcinogenieity of 2,3,7,8-TCDD, 2,3,7,8-Hx ,CDCs and repro-



ductive effects of those compounds plus 2,3,7»8-TCOF)> should be critically



re-examined and the range of experimental data and estimated potencies froo



all studies tabulated.  The Subccnanittee also reeotmends that EPA consider



assigning higher relative TEFs to CDFs in general, and 2,3,4/7,8-PeCD? in



particular*

                             *"                            «  *   * " *  "    *

     The Subcommittee' strongly believes that EPA should assign greater
                                              *


priority to obtaining and using data on toxicokinetics, including metabolism*



The rates of uptake and distribution of compounds alone and in mixtures



are important measures of bioavailability and dosimetry.  The kinetics of



metabolism and excretion, along with those of receptor kinetics anf



affinities, should be especially useful for interspecies ceraparisons and



for estimating risks for this particular class of cewpounds.



     The SubccRinittee wishes to enfjhasize that the method proposed may lack



scientific validity.  The associated errors have not been quantified.  It



is iu^ortant, therefore, that the Agency make every effort to validate



the method.  The Subcomittee recoranends periodic review and  .Dialysis as



better data are obtained, and that EPA make systematic efforts  to obtain



critically unportant data,  including that frcm in vivo tests on confounds
                                   1-6

-------
with representative positional substitutions.   Efforts should continue to
develop methods for assaying the biologic activity of important mixtures
(e.g. fly ash) in in vitro systems, using other cells in addition to
hepatocytes and" other endpoints in addition to MiH activity.  Until the
uncertainties ace reduced, the interim TEF method should be largely
reserved for specific situations where the components qf the mixture are
known, where the conposition of the mixture is not expected to vary such
with time, and where the extrapolations are consistent with existing
animal data.
                              1-7

-------

-------
               Appendix c
OP THE MUNICIPAL W&STS COMBUSTION RESEARCH KAN
                   C-l

-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
           WASHINGTON, D.C. 204SO
                                             OFFICE Of
                                          -rum
April 11, 1988


The Honorable Lee M.  Thomas
Administrator
U,S, Environmental Protection Agency
401 M. Street, S.W.
Washington, D.c.  20460
                                                    *
Dear Mr. ThomasI

     The Municipal Waste Combustion Subcommittee of the Science
Advisory Board's Environmental Effects,  Transport  and  Fate
Committee has completed its review of the Office of Research and
Development's (QRD's) "Municipal  Waste Combustion Research Plan".
Ihe review  was  initiated  at your request, along with  two other
charges  related  to municipal waste  combustion,  and wa"s reviewed
concurrently with other issues on March 10, 1987,

     The Subcommittee  concludes  that  the research plan  is  well
defined  and reflects-considerable thought, however, the proposed
level of funding for the  research appears grossly  inadequate in
view  of  the large number of  scientific uncertainties  associated
with  this  technology,  and  EPA's responsibility to  develop
scientifically  credible regulatory decisions.   Important areas,
such  as ecological effects,  are entirely  left out  or  are
addressed  in  a  cursory  fashion, which  is  understandable since
allocated funds  are  inadequate for the areas that are addressed.
Prioritisation  of research  emphasizes  avenues  with  short-term
goals which may  be  necessary to meet  the needs  for  technical
guidance in permitting the many MWCs that are  being planned or
are already in operation.

     The  Subcommittee  believes  that  emissions  should  be
characterized as a  first  priority  through  analytical  chemistry
projects,   methods  development,  and   field,   testing.    Risk
assessment, health effects prediction and emission control cannot
be  adequately  conducted without a thorough  knowledge  of  the
quality  and quantity of the emissions, both gaseous and residual.
                   C-2

-------
                                                       Page two
Following such  characterization,  environmental  transport,  fate
and bioavailability should be .determined,  since they are  key  to
assessing both risk and exposure to humans and the environment.

     Monitoring is also considered by the Subcommittee to be  an
important research priority,  and research directed towards
monitoring goals will insure the development of tools  to  ensure
compliance with  guidelines or standards  that  may be  set.    In
addition,  monitoring  is  important  for  the  validation  of
predictive models which have been developed  for  air transport  of
stack emissions.

     The  Subcommittee  agrees that  major  areas  of promising
research have  been  proposed  and  developed  to  investigate
important areas  of uncertainty  with,  respect to municipal waste
combustion technology.  However, budgetary constraints  shed
doubt, in the Subcommittee's opinion, on  EPA's  ability to reach
the  objectives  defined  in the  program.   Considerations  of
priority might be  revisited  to  allow identification  of research
areas with high priority and attainable objectives.

     The Subcommittee appreciates the opportunity to conduct  this
scientific review.  We  request that  the Agency  formally  respond
to the scientific advice transmitted  in the attached report.
                                                     t»
                                       Sincerely,
                                       Norton Nelson, Chairman
                                       Executive Committee
                                       Science Advisory Board
                                          .f Hartung,
                                        funicipal Waste
                                         Combustion Subcommittee
Enclosure

cc: A. James Barnes
    Vaun Hewill
    Alfred Lindsey
    Larry Fradkin
   • Terry Yosie
                               C-3

-------

-------
                    Appendix D



DESCRIPTION OF REFUSE DERIVED FUEL (RDF) CATEGORIES
                        D-l

-------
CLASS
FORM
       Appendix  D'

ASTM CLASSIFICATION OF RDFs

 DESCRIPTION
KDP-I     Raw       Municipal solid waste with minimal processing
                    to remove oversize bulky waste.
                                                   *
RDF-2     Coarse    MSW processed to coarse particle size with or
                    without ferrous metal separation such that
                    95% by weight passes through a i-inch-square
                    mesh screen,

RDF-3     Fluff     Shredded fuel derived from MSW processed for
                    the removal of metal, glass, and 
-------
        Appendix E
GXX3SSARY-QF TERMS AND UNITS
             S-l

-------
                   GLOSSARY Q£ TERMS AND UN^TS
Acid gases
As

Back-pass convective
 heat recovery boiler,
 eonveetive back-pass

BACT

Bed agitation
Bottom ash



BTU

Burning refuse bed

CB

Cd

CO

Congeners
CP
-  Compounds, such as hydrochloric acid
   (HC1),  sulfur dioxide (SO2),
   nitrogen oxides (HOX) and
   hydrofluoric acid (HF) that are in
   the gaseous state*

-  Arsenic

-  the heat recovery boiler at the
   furnace outlet generating steam by
   convective heat transfer.

-  Best Available control technology

-  Agitation of the burning fuel bed
   by mechanical movement of the
   furnace grate

-  Residual ash resulting from the
   burning of garbage, as discharged
   from the bottom of the incinerator

-  British Thermal Units

-  MSW bed burning on the furnace grate

*  Chlorobenzenes

-  Cadmium

-  Carbon Monoxide

-  A group of closely related chemical
   compounds such as the 75 chlorinated
   dibenzodioxins or chlorinated
   dibenzofurans

-  Chlor©phenols
DC

Oioxins

Downwash



Dry Deposition
-  Direct Current

-  See PCDD below

   Downward air movement in lee of
   buildings and structures due to
   aerodynamic forces

-  Turbulent exchange of gases and small
  • particles from  the  atmosphere to the
   ground surface
                               E-2

-------
EPA

ISP

Efflux velocity
Electrostatic
  precipitator (ESP)
Elutriate
Excursion
Fabric filter, or
 fabric , filter baghouse
Feed pit
Ply ash
Front, rear arch
 geometry

Gasification of MSW
Gravitational settling


HC1

HF

Hg
-  Environmental Protection Agency

-  Electrostatic precipitator

-  Plue gas velocity leaving the
 *  furnace grate passing up through
   the combustion chamber
-  An air pollution control device
   designed to remove particulate
   matter from a gas stream using
   electrostatic forces.

-  Particulate ash carried up from
   a furnace fuel bed by the gas
   velocity through the grate

-  Deliberation from normal
   operating conditions resulting
   in incinerator upset conditions

-  An air pollution control device
   used to remove particulate matter
   from a gas using filtration
   principles                 *

-  Receptor pit used for MSW
   storage from which the fuel is
   introduced into the MWC

-  General term for all ash carried
   up from the grate and out from
   the incinerator/boiler by the
   flue gas

-  furnace wall design configuration
   Heating of the HSW at the entry
   point of the furnace grate - which
   drives off MSW Moisture and Volatile
   hydrocarbon constituents of fuel

   Settling of particulate matter by
   force of gravity

   Hydrochloric acid

   Hydrofluoric acid

   Mercury
                               E-3

-------
Homologue groups
Hydrometeors


Isoraer


KV

mg/Nm3

MSW

MWC


Mass-bum units



Modular incinerator
Ha

ng/g

ng/Nm*
PAH

Pb

PCS

PCDD
-  Group of chemicals which vary in
   structure but have the same
   composition,  such as degree of
   chlorlnation

-  Solid and liquid water particles
   and droplets

-  2 particular members of a
   homologue group

-  Kilovolts

-  Milligrams per normal cubic meter

- .Municipal Solid Waste

-  Municipal Waste Combustion
   (or Combustors)

-  Incinerators that burn unprocessed
   MSW,  typically  in  refractory  or
   waterwall furnaces

-  Factory preassembled mass burn units
   usually  employing  controlled  air
   combustion technology to  incinerate
   considerably  lower  volumes of  waste
   than those employed by mass  bum or
   RDF Units

-  Sodium

-  Nanograma per gram

 -  Nanograms per  normal  cubic meter at
   normal   temperature  and  pressure
   conditions

-  Oxides of Nitrogen,  such as NO2,
   or nitrogen dioxide

-  Folycycllc aromatic hydrocarbons

-  Lead

-  Polychlorinated blphenyls

-  Total of all ^Corners and/or all
   homologue groups of polychlor-
   inated dibenzo dioxins
                               B-4

-------
PCDF
'plume rise
Pneumatic injection
ppra

Proximate Analysis



pyrolysis of MSW



RCRA


RDF


RDF processes
Rotary combustors
Scavenging coefficient
S crubber/baghouse
Se
-  Total of all isomers and/or all
   monologue groups of polychlor-
   inated dibenzo furans

-  A term used to describe the rise of
   a steady stream of flue gas due to
   buoyant effects after it leaves a
   stack

-  Air injection of MSW or processed
   refuse into the furnace

-  Parts per billion

-  Parts per million

-  The gravimetric composition of
   moisture, Ash, volatile matter and
   fixed carbon in a MSW fuel*

-  The heating of MSW in the absence of
   oxygen, which drives off moisture and
   volatile matter in municipal waste

-  Resource Conservation and Recovery
   Act                        *

-  Refuse-derived fuel (unprocessed
   or processed, municipal solid waste)

 -  Refuse  derived fuel processes that
   subject  MSW  to varying degrees  of
   processing to  improve fuel  quality
   for better combustion efficiency and
   to achieve some material recycling or
   recovery.

-  MSW combustion occurring in a
   rotating drum (or kiln)

N-  frhe coefficient describing the expo-
   nential decrease with time of
   atmospheric contaminants due to
   capture by rain and cloud droplets/
   usually applied to single precipita-
   tion events

-  An air pollution control system
   consisting of a scrubbing device
   (lime injector) followed by a
   fabric filter dust collection

-  Selenium
                               E-5

-------
so.
-  Sulfur dioxide
Spreader stoker
Stack
starved air incinerators -
Suto-stoichometric air



TCDD


TCDF


Tetrahomologues



THC

Tio

TFD

Ultimata analysis




Underfire air



Washout ratio (Wr)
Water-wall incinerator
Wat deposition
   Coal or RDF injection over a forward
   moving traveling grata

   Chimney through which gases and
   particulate residues are emitted

   MSW combustion occurring in primary
   combustion chambers supplied with
   aub-stoichometric air

   Combustion air supplied that is
   less than that theoretically
   required to burn the fuel completely

   Any tetra Isomers or the' tetra
   homologua.group of dioxins

   Any tetra isomers or the tetra
   homologue groups of furans

   Aa applied to PCDD and PCDF - those
   iaomars and homologuea which are
   chlorinated at 4 position^

   total hydrocarbons

   Titanium oxide
-  Tons per day

-  A gravimetric fuel analysis giving
   mass composition of fuel elements
   necessary to do combustion
   calculations

-  Combustion air Introduced under
  ,the grates of an incinerator or
   furnace

-  The effluent concentration in
   precipitation normalized by the
   effluent concentration in air
   used to describe average conditions
   over many precipitation events

-  Th« furnace of a MWC 'that is ^xned
   with tubes recovering heat for
   steam generation

*  Removal of atmospheric contaminants
   as a result of capture by cloud
   droplets as well as precipitation
      B-6

-------
Wlndboxes - multiple
  air compartments
Zn

2,3,7,8 TCDD
   The use of multiple compartments
   under a furnace grate to allow for
   better undergrate combustion air
   distribution
-  Zinc
   2,3,7,3 tetrachloroclibenzo-p-
   dioxin
                               1-7

-------

-------