United States     Office of Air Quality
          Environmental Protection  Planning and Standards      EPA - 450/3-89-019
          Agency       Research Triangle Park NC 27711    May 1989

          "Air
?/EPA     Hazardous Waste
          TSDF - Fugitive
          Particulate Matter
          Air Emissions
          Guidance Document

-------
       HAZARDOUS WASTE TSDF
    FUGITIVE PARTICULATE MATTER
AIR EMISSIONS GUIDANCE DOCUMENT
                     by

C. Cowherd, P. Englehart, G. E. Muleski, J. S. Kinsey
           Midwest Research Institute
             425 Volker Boulevard
          Kansas City, Missouri  64110
          EPA Contract No. 68-02-4395
             Work Assignment 21
              MRI Project 8986-21
        William L. Elmore, Project Officer
          Emission Standards Division

    Office of Air Quality Planning and Standards
      U.S. Environmental Protection Agency
   Research Triangle Park, North Carolina  27711
                  May 1989
                           U.S. Environmental Pro
                           Eegion 5, Library • y- !.
                           230 S. Dearborn SV.^ot.
                           Chicago,. IL  6060-:

-------
This report has been reviewed by the Emission Standards Division of the Office of Air Quality Planning and
Standards,  EPA, and  approved for publication. Mention of trade names or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this report are available through
the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park NC
27711, or from National Technical Information Services, 5285 Port Royal, Springfield VA 22161.

-------
                             ACKNOWLEDGMENTS


     We wish to acknowledge the significant contributions of a number of
Individuals to the success of this project.  These individuals and their
organizational affiliations are listed below.


     Technical Contributors                  Word Processing Staff
     Chat Cowherd, MRI                       Janice Evans, MRI
     Dennis Doll, SRAB                       Kristy Henry, MRI
     Ken Durkee, ESD                         Cindy Melenson, MRI
     Michael Dusetzina, PAB                  Susie Powell, MRI
     Larry Elmore, ESD
     Phil Englehart, MRI
     John Kinsey, MRI
     Greg Muleski, MRI

-------
                                 CONTENTS

Acknowledgments	     Ill
Figures	      ix
Tables	      xi

     1.0  Introduction	     1-1
          1.1  Regulatory requirements for fugitive dust
               control under RCRA	     1-1
          1.2  Particulate emission factor models	     1-4
          1.3  Preventive and mitigative control options	     1-7
          1.4  On-site versus off-site facilities	     1-9
          1.5  Organization of this document	    1-10
          1.6  References for Section 1	    1-10

     2.0  Paved and Unpaved Roads	     2-1
          2.1  Paved roads	     2-2
               2.1.1  Source description	     2-2
               2.1.2  Estimation of uncontrolled emissions	     2-4
               2.1.3  Demonstrated control techniques	     2-7
               2.1.4  Control performance estimation/
                      specification	    2-11
               2.1.5  Procedures for compliance determination....    2-20
          2.2  Unpaved roads	    2-24
               2.2.1  Sourc  description	    2-24
               2.2.2  Estimation of uncontrolled emissions	    2-26
               2.2.3  Demonstrated control techniques	    2-29
               2.2.4  Control performance estimation	    2-34
               2.2.5  Procedures for compliance determination....    2-47
          2.3  Example calculation	    2-51
               2.3.1  Paved road estimates	    2-53
               2.3.2  Unpaved road estimates	    2-55
          2.4  References for Section 2	    2-57

     3.0  Open Waste Piles and Staging Areas	     3-1
          3.1  Source description	     3-1
          3.2  Estimation of uncontrolled emissions	     3-2
               3.2.1  Materials handling	     3-5
               3.2.2  Wind erosion	     3-6
               3.2.3  Wind erosion of continuously active
                      piles	    3-19
               3.2.4  Level of contamination	    3-20
          3.3  Demonstrated control techniques	    3-21
          3.4  Control performance estimation	    3-22
               3.4.1  Chemical stabilization	    3-23
               3.4.2  Enclosures	    3-24
               3.4.3  Wet suppression systems	    3-29

-------
                     CONTENTS  (continued)
     3.5  Procedures for compliance determination	    3-39
     3.6  Example calculation	    3-39
     3.7  References for Section 3	    3-47

4.0  Dry Surface Impoundments	     4-1
     4.1  Source description	     4-1
     4.2  Estimation of uncontrolled emissions	     4-3
          4.2.1  Wind erosion	     4-3
          4.2.2  Removal of surface material	    4-16
          4.2.3  Level of contamination	    4-18
     4.3  Demonstrated control techniques	    4-18
     4.4  Control performance estimation	    4-20
          4.4.1  Physical stabilization	    4-21
          4.4.2  Chemical stabilization	    4-21
          4.4.3  Wind fences/barriers	    4-22
          4.4.4  Watering of unpaved surfaces	    4-28
          4.4.5  Wet suppression for materials handling	    4-33
     4.5  Procedures for compliance determination	    4-38
     4.6  Example calculation	    4-39
     4.7  References for Section 4	    4-40

5.0  Landfills	     5-1
     5.1  Source description	     5-1
     5.2  Estimation of uncontrolled emissions	     5-6
          5.2.1  Particulate emission rates	     5-6
          5.2.2  Level of contamination	    5-12
     5.3  Control techniques	    5-12
          5.3.1  Preventive controls	    5-14
          5.3.2  Mitigative controls	    5-15
     5.4  Control performance estimation	    5-17
     5.5  Procedures for compliance determination	    5-18
          5.5.1  Preventive controls	    5-18
          5.5.2  Mitigative controls	    5-21
     5.6  Example calculation	    5-22
     5.7  References for Section 5	    5-27

6.0  Land Treatment	     6-1
     6.1  Source description	     6-1
     6.2  Estimation of uncontrolled emissions	     6-4
          6.2.1  Particulate emission rates	     6-4
          6.2.2  Level of contamination (a)	     6-7
     6.3  Control techniques	     6-7
          6.3.1  Preventive controls	     6-9
          6.3.2  Mitigative controls	    6-10
     6.4  Control performance estimation	    6-10
     6.5  Procedures for compliance determination	    6-11
          6.5.1  Preventive controls	    6-14
          6.5.2  Mitigative controls	    6-16
     6.6  Example calculation	    6-17
     6.7  References for Section 6	    6-18

                           vi

-------
                          CONTENTS  (continued)
     7.0  Waste Stabilization	     7-1
          7.1  Source description	     7-1
               7.1.1  Site characteristics	     7-1
               7.1.2  Emission sources	     7-5
          7.2  Estimation of uncontrolled emissions	     7-7
               7.2.1  Particulate emission rates	     7-7
               7.2.2  Level of contamination (a)	    7-11
          7.3  Control techniques	    7-12
               7.3.1  Preventive controls	    7-13
               7.3.2  Mitigative controls	    7-15
          7.4  Control performance estimation	    7-17
               7.4.1  Wind fences or barriers	    7-17
               7.4.2  Capture/collection systems	    7-19
               7.4.3  Wet suppression	    7-23
               7.4.4  Watering of unpaved surfaces	    7-24
               7.4.5  Paved surface cleaning	    7-28
          7.5  Procedures for compliance determination	    7-30
               7.5.1  Permit systems	    7-30
               7.5.2  Indirect measures of control performance...    7-37
          7.6  Example calculation	    7-38
               7.6.1  Process description	    7-38
               7.6.2  Raw material handling	    7-39
               7.6.3  Vehicle traffic on unpaved surfaces	    7-40
          7.7  References for Section 7	    7-42
Appendices
     A.  Open dust source emission factor rating and control
         efficiency terminology	     A-l
     B.  Estimation of control costs and cost effectiveness	     B-l
     C.  Screening techniques, modeling information, and
         health risks information	     C-l
     D.  Sampling and analysis procedures	     0-1
                                vn

-------
                                 FIGURES

Number                                                               Page

 2-1  Traffic count log	    2-22
 2-2  Sampling data form for paved roads	    2-23
 2-3  Mean annual number of days with at least 0.01 in of
      precipitation	    2-28
 2-4  Annual evaporation data for the contiguous United States
      (as diagrammed in the "Climatic Atlas of the United
      States," June 1968)	    2-38
 2-5  Watering control effectiveness for unpaved travel
      surfaces	    2-39
 2-6  Average PM10 control efficiency for chemical suppressants..    2-42
 2-7  Example chemical suppressant application log	    2-49
 2-8  Example completed log	    2-50
 2-9  Roads at a hypothetical TSDF	    2-52
 3-1  Open waste piles and staging areas	     3-4
 3-2  Illustration of logarithmic velocity profile	     3-8
 3-3  Relationship of threshold friction velocity to size
      distribution mode	    3-13
 3-4  Scale of threshold friction velocities	    3-14
 3-5  Contours of normalized surface wind speeds, us/ur	    3-16
 3-6  Decay in control efficiency of latex binder applied to
      coal storage piles	    3-23
 3-7  Example.  Pile surface areas within each wind speed
      regime	    3-41
 3-8  Daily fastest miles of wind for periods of interest	    3-42
 3-9  Field procedure for determination of threshold friction
      velocity	    3-43
 4-1  Surface impoundment activities	     4-2
 4-2  Illustration of logarithmic velocity profile	     4-6
 4-3  Roughness heights for various surfaces	     4-7
 4-4  Relationship of threshold friction velocity to size
      distribution mode	     4-9
 4-5  Field procedure for determination of threshold friction
      velocity	    4-10
 4-6  Increase in threshold friction velocity with I	    4-12
 4-7  Scale of threshold friction velocities	    4-14
 4-8  Decay in control efficiency of latex binder applied to
      coal storage piles	    4-22
 4-9  Annual evaporation data for the contiguous United States
      (as diagrammed in the "Climatic Atlas of the United
      States," June 1968)	    4-30
 4-10 Watering control efficiency effectiveness for scraper
      travel on unpaved surfaces	    4-31

-------
                           FIGURES (continued)

Number                                                               Page

 4-11 Typical form for recording watering program control
      parameters	    4-34
 5-1  Line drawings of common equipment types at HW landfills....     5-3
 5-2  (a) Hypothetical landfill unit plan view; (b) profile
      view; (c) hypothetical landfill with master cell/subcell
      configuration	     5-4
 6-1  Land treatment units—physical configurations	     6-3
 6-2  Flowchart for evaluating control associated with waste
      appl icati on	    6-12
 6-3  Procedure for estimating control efficiency with O&G
      content and moisture content measurements	    6-13
 7-1  General pozzolanic process flow diagram	     7-6
 7-2  Mean annual number of days with at least 0.01 in of
      precipitation	    7-10
 7-3  Performance evaluation of capture/collection systems	    7-22
 7-4  Annual evaporation data for the contiguous United States
      (as diagrammed in the "Climatic Atlas of the United
      States," June 1968)	    7-26
 7-5  PM10 control efficiency for watering unpaved surfaces	    7-27
 7-6  Typical form for recording watering program control
      parameters	    7-35

-------
                                  TABLES

Number                                                               Page

 2-1  Possible default values for paved road emission estimates
      at TSDFs	     2-5
 2-2  Selection of paved road emission factor	     2-6
 2-3  Measured efficiency values for paved road controls	     2-8
 2-4  Outline of preventive paved road control measures
      applicable to TSDFs	     2-9
 2-5  Miscellaneous operation/design and cost data for vacuum
      sweeping paved roads	    2-15
 2-6  Miscellaneous operation/design and cost data for
      flushing/broom sweeping paved roads	    2-19
 2-7  Possible default values for unpaved road estimates at
      TSDFs	    2-28
 2-8  Control techniques for unpaved travel surfaces	    2-31
 2-9  Chemical stabilizers	    2-32
 2-10 Summary of unpaved road chemical dust suppressant costs....    2-46
 2-11 Recordkeeping requirements for unpaved road controls	    2-48
 2-12 Summary information for roads A, B, and C	    2-53
 3-1  Typical silt and moisture content values of materials at
      various industries	     3-3
 3-2  Erosion potential function	    3-11
 3-3  Threshold friction velocities—industrial aggregates	    3-12
 3-4  Threshold friction velocities—Arizona sites	    3-12
 3-5  Subarea distribution for regimes of us/ur	    3-17
 3-6  Control techniques for storage piles	    3-22
 3-7  Capital and O&M items for chemical stabilization of
      waste pi les	    3-25
 3-8  Typical form for recording chemical dust suppressant
      control parameters	    3-26
 3-9  Typical form for recording delivery of chemical dust
      suppressants	    3-27
 3-10 Summary of available control efficiency data for water
      sprays	    3-32
 3-11 Summary of available control efficiency data for foam
      suppression systems	    3-33
 3-12 Wet suppression system capital and O&M cost elements	    3-36
 3-13 Typical costs for wet suppression of material  transfer
      poi nts	    3-37
 3-14 Compliance determination for waste piles	    3-40
 3-15 Example calculation of uncontrolled PM10 emissions	    3-45
 3-16 Example calculation of controlled PM10 emissions	    3-47

-------
                           TABLES  (continued)

Number                                                               Page

 4-1  Threshold friction velocities	    4-13
 4-2  Threshold friction velocities—Arizona sites	    4-13
 4-3  Control techniques for dry surface impoundments	    4-20
 4-4  Capital and O&M items for chemical stabilization of
      open area sources	    4-23
 4-5  Typical form for recording chemical dust suppressant
      control parameters	    4-24
 4-6  Typical form for recording delivery of chemical dust
      suppressants	    4-25
 4-7  Wet suppression system capital and O&M cost elements	    4-37
 4-8  Compliance determination for dry surface impoundments	    4-39
 5-1  Emission factors for hazardous waste landfill unit
      operations	     5-8
 5-2  Emission factor correction terms and source extents for
      sampled landfill units	     5-9
 5-3  Mean wind speed [U] for selected United States stations	    5-10
 5-4  Summary of RCRA metals concentration (vg/g)	    5-13
 5-5  Example observation/audit checklist for landfill
      preventive control practices	    5-19
 6-1  Model plant size distributions for land treatment units....     6-2
 6-2  Summary of RCRA metals concentrations (ug/g) for land
      treatment units handling refinery wastes	     6-8
 6-3  Example observation/audit checklist for land treatment
      preventive control practices	    6-15
 7-1  PM-10 emission factors and correction parameters for
      waste stabi 1 ization	     7-9
 7-2  Summary of RCRA metals concentration for waste
      stabilization samples	    7-13
 7-3  Measured efficiency values for paved surface controls	    7-29
                                XI

-------
                            1.0  INTRODUCTION

     Earlier EPA studies that evaluated fugitive partlculate matter (PM)
emissions from hazardous waste Treatment, Storage, and Disposal Facili-
ties (TSOFs) indicated that risks from contaminated PM emissions from
these facilities are potentially significant.  Sources of particular con-
cern are landfills, land treatment, waste stabilization, dry surface
impoundments, and roads.  These sources of emissions are regulated by a
site-specific permitting system for TSDFs that has been established under
the authority of the Resource Conservation and Recovery Act (RCRA).
     The purpose of this document is to provide regulatory and industrial
personnel with sufficient information to identify sources of contaminated
fugitive PM emissions, estimate the magnitude of emissions, select viable
control measures, and estimate the effectiveness of those measures in
order to ensure that high risks from these facilities do not occur.  The
PM particle size fraction of interest, designated as PM10, consists of
particles with an aerodynamic diameter equal to or less than 10 micro-
meters (urn).
     The remainder of this introduction addresses some of the general
issues associated with TSDF PM emissions and their control.  Separate
subsections discuss the regulatory basis for fugitive PM control at
TSDFs, the procedures used to estimate fugitive PM emissions, fugitive PM
control strategies, and potential differences in on-site and commercial
TSDFs.  The final subsection outlines the organization of the remaining
sections of the document.

1.1  REGULATORY REQUIREMENTS FOR FUGITIVE DUST CONTROL UNDER RCRA
     Regulations developed under the authority of the RCRA provide for
control of hazardous waste (TSDF) releases that are potentially harmful
to human health and the environment.  Contaminated fugitive (PM) emis-
sions are by definition included in those releases controlled under
                                   1-1

-------
RCRA.  Generally, this regulatory control is accomplished via a two-step
process.  The interim and final status regulations promulgated in 40 CFR
Parts 264 and 265 provide general performance requirements for TSDFs.
However, the primary focus under RCRA is the facility-specific permit
requirements issued as a part of the Part B (or final status) permit that
is issued under 40 CFR Part 270.  Permit writers generally have wide
latitude in defining the permit conditions that are needed to meet the
Part 264 performance requirements, and effective control of fugitive PM
emissions depends upon these facility-specific requirements.
     Contaminated fugitive PM emissions from land-based TSDFs are
addressed in three places—the general facilities standards, the
facility-specific regulations, and the permit provisions.  The omnibus
provisions in the general facilities standards of Part 264 generally pro-
hibit any releases to the environment that are harmful to human health or
the environment.  Specific provisions that apply to fugitive PM emissions
include the following:
     1.   40 CFR 264.15 requires that owners/operators inspect facilities
          for malfunctions, deterioration, operating errors, and dis-
          charges which cause or lead to discharge of hazardous constitu-
          ents to the environment.
     2.   40 CFR 264.31 requires that facilities be designed, con-
          structed, maintained, and operated to minimize unplanned sudden
          or nonsudden release of hazardous constituents to the environ-
          ment (including atmospheric releases).
     3.   40 CFR 264.11 requires that owners/operators close facilities
          in a manner that minimizes or eliminates, to the extent neces-
          sary to protect the human health and environment, the release
          of hazardous constituents.
     The facility-specific regulations in Part 264 address wind dispersal
of dust.  The regulations for waste piles (40 CFR Parts 264.251[f] and
264.254[bl), land treatment  (40 CFR Parts 264.273[fl and [g]), and land-
fills  (40 CFR Parts 264.301[f] and 264.303[b]) specifically address wind-
blown  dust.  They require that the owner/operator cover or otherwise man-
age  the facility to control windblown dust.  In addition, the owner/
operator must inspect the wind dispersal control system weekly.  Relative

                                   1-2

-------
to closure of surface impoundments, the process-specific regulations (40
CFR 264.228) require that the owner/operator either remove contaminated
liners and subsoils from the facility or cover them with a final cover
that minimizes erosion and abrasion.
     The permit requirements of Part 270 also address fugitive dust.  The
regulations require that for waste piles (40 CFR Part 270.18[c][5j),
landfills (40 CFR Part 270.21[bJ[5J), and land treatment facilities (40
CFR Part 270.20[c][6]), owners/operators provide detailed plans or engi-
neering descriptions of methods to control windblown particulate matter
in their Part B application.
     The regulations described above require that dust emissions be
controlled and that plans be provided for control of those emissions.
They do not establish specific emissions limits or control technology
requirements.  These specific requirements are established in the RCRA
Part B permitting process.  General EPA policy relative to specific
requirements often is presented in permit guidance materials.
     According to the "Permit Applicants' Guidance Manual"1 prepared by
the Office of Solid Waste (OSW), Part B permit applications for land-
based TSDFs should include, as an attachment, a "Wind Dispersal Flow
Control Plan."  The plan should contain (a) a description of methods for
controlling wind dispersal and (b) data to estimate the effectiveness of
control techniques.  The discussion of the types of techniques that could
be used is brief and general in nature (e.g., suggestion of watering or
windbreaks).  No specific design or operating requirements are provided.
     Thus the overall objective of this document is to provide detailed
information on applicable control techniques, methods of determining
control technique effectiveness, possible methods of ensuring compliance,
and general cost information to regulatory agency personnel plus owners
and operators of TSDFs.  This information should assist the TSDF permit
writer in determining whether the Wind Dispersal Flow Control Plan sub-
mitted by a TSDF owner/operator is adequate, given site-specific fac-
tors.  Also, this information should assist TSDF owners and operators in
developing a Wind Dispersal Flow Control Plan.
                                   1-3

-------
1.2  PARTICULATE EMISSION FACTOR MODELS
     In developing particulate control strategies for "traditional"
pollutant concerns (i.e., to meet National Ambient Air Quality Standards
[NAAQS]), gross particulate emissions from open sources are estimated
using the predictive emission factors presented in Section 11.2 of EPA's
"Compilation of Air Pollutant Emission Factors" (AP-42).2  These factors
cover the generic source categories:
     •    Unpaved travel surfaces
     •    Paved travel surfaces
          Exposed areas (wind erosion)
          Materials handling
Note that these factors are updated periodically and that users of this
document should use updated factors as they become available.  All fac-
tors cited in the document represent the most current versions as of
September 1988 (Reference 2).
     These emission factors share many common features.  For example, the
models are formulated as empirical expressions that relate variations in
emission factor (e) to differences in the physical properties  (p) of the
material being disturbed and the mechanical energy (m) responsible for
the generation of particulate according to the general form:
                                                                     (1-1)
As empirical models, open dust source factors have adjustable coeffi-
cients  (K,a,b) that reflect relationships determined from actual open
dust  source testing.
      As  indicated  in Section  1.1, much of this guidance document centers
on the  application of  the open dust factors to estimate particulate
emissions  from permitted TSDF units.
                                    1-4

-------
Estimation is accomplished according to the general model form:
                              n
                          R = z     a. •  e. •  A.                    (1-2)
where:     R = emission rate of contaminated airborne particulate (kg/yr)
               for a given TSDF, consisting of n identifiable unit opera-
               tions
          a..- - fraction of contaminant in particulate emissions for the
           J
               jth operation (wg/g); for uncontaminated particulate emis-
               sions, Oj = 0
          e,- = emission factor(s) (mass/source extent)
          A.- = source extent(s) (source dependent units)
           J

     This approach is analogous to that used by MRI in recent work on
estimation of emissions from surface contamination sites.3  The approach
is also consistent with techniques used in air pollution assessments.
     In the TSDF context it is important to recognize the following:
     1.   The o term refers to the "level of contamination" and is
          usually represented as the concentration (e.g., yg/g) of the
          metal(s) or organic compound(s) of concern in the disturbed
          material.
     2.   Permitted TSDF units may consist of multiple, identifiable unit
          operations; in the case of landfills, for example, unit
          operations include loadout of bulk hazardous waste, loadout of
          temporary cover, lift construction, and general vehicle traffic
          proximate to the landfill face.
     3.   The major assumption implicit in Eq. 1-2 is that TSDF processes
          that generate particulate emissions can be adequately
          represented by existing open dust source emission factor
          models.
Item 2 above is relatively straightforward and is demonstrated in the
individual chapters.  Items 1 and 3 require further clarification as
given below.
                                   1-5

-------
     Specification of the level of contamination (a) probably is the
single most difficult aspect of applying the emission factors.  For a
given TSOF source, "representative" value(s) of a may depend upon a
number of factors including principally:
     1.   The nature of waste streams handled and whether the wastes are
          metal-containing or organic waste streams.
     2.   Operational practices that influence the availability of waste
          material for entrainment into the atmosphere.
     3.   The age of the facility to the extent that it reflects "resid-
          ual" contamination associated with long-term disposal and typi-
          cal waste stream volumes.
Note that the term representative  is used here to connote a long-term
value, such as an annual mean, that adequately defines a over the entire
TSDF source.
     There are essentially two methods for specification of a.  The pre-
ferred method is through source-specific sampling and analysis (S&A).
Appendix D outlines generic elements of a source-specific program for
characterization of both physical and chemical  (i.e., a) properties of
surface materials at TSDFs.  In some cases it may be feasible to develop
estimates of a for a specific source based on existing information other
than actual S&A data.  In general, these estimates should be developed
from the perspective of a "worst-case scenario," because of the greater
uncertainty involved.  In other words, they should be conservatively high
and thus tend to protect against any potential underestimation of risk
associated with particulate emissions.
     The principal sources of information for worst-case estimates are
expected to be:
     1.   Waste manifests and any  "tracking" information that a facility
          routinely generates to characterize its receipts.
     2.   Conversations with facility personnel.
     Because the emission factor models are empirical expressions, in a
strict sense their degree of applicability relative to TSDF sources is
related to  the degree of compliance with two conditions:
     1.   How closely a given TSDF unit operation resembles the source
          conditions underlying test data used  in development of the
          emission factor.
                                    1-6

-------
     2.   How closely the physical characteristics of the disturbed
          material resemble those tested in development of the emission
          factor.
     Based on surveys described in reference 4, the general conclusion is
that TSDF unit operations are reasonably comparable to those tested in
development of the emission factor relationships.  Of course, this agree-
ment varies from source to source.  For example, the emission factor for
vehicle traffic on unpaved roads encompasses most common traffic condi-
tions—vehicle types, operating speeds, and weights—found at TSDFs.  On
the other hand, the emission factor applied to one of the important unit
operations at landfills—lift construction—is based on source testing
that does not bear strong resemblance to TSDF operations.
     Results of the surveys described 1n reference 3 also clearly point
out that the physical characteristics of waste and disturbed surface
materials at TSDFs do not always conform closely to those materials
tested in development of the emission factors.  More specifically, the
moisture and/or "oily nature" of waste streams may be markedly different
from the dry, finely divided materials upon which the field tests were
performed.  The current emission factor models do not fully account for
the inherent mitigation provided by the physical binding effect of oily
substances, which may add conservatism to the predictive emission
rates.  However, at any given facility, the potential for increased
spreading of material may offset this conservatism.

1.3  PREVENTIVE AND MITIGATIVE CONTROL OPTIONS
     Typically, there are several options for control of fugitive
particulate emissions from any given source.  This is clear from the
mathematical equation used to calculate the emissions rate:

                             R = A e (1 - c)                        (1-3)

where:     R = estimated mass emission rate
           A = source extent (i.e., surface area for most open dust
               sources)
                                   1-7

-------
           e = uncontrolled emission factor, i.e., mass of uncontrolled
               emissions per unit of source extent
           c = fractional efficiency of control

To begin with, because the uncontrolled emission rate is the product of
the source extent and uncontrolled emission factor, a reduction in either
of these two variables produces a proportional reduction in the uncon-
trolled emission rate.
     Although the reduction of source extent results in a highly predict-
able reduction in the uncontrolled emission rate, such an approach in
effect usually requires a change in the process operation.  Frequently,
reduction in the extent of one source may necessitate the increase in the
extent of another, as in the shifting of vehicle traffic from an unpaved
road to a paved road.
     The reduction in the uncontrolled emission factor may be achieved by
process modifications (in the case of process sources) or by adjusted
work practices (in the case of open sources).  The degree of the possible
reduction of the uncontrolled emission factor can be estimated from the
known dependence of the factor on source conditions that are subject to
alteration.  For open dust sources, this information is embodied in the
predictive emission factor equations for fugitive dust sources as pre-
sented in Section 11.2 of EPA's "Compilation of Air Pollutant Emission
Factors" (AP-42).
     Control techniques can be divided into two broad categories—preven-
tive and mitigative.  Although differences between the two are not always
clear, in general, preventive measures involve techniques that reduce
source extent or improve mechanical source operations relative to the
generation of particulate emissions.  By contrast, mitigative techniques
typically focus on altering the surface/material conditions that
constitute the source of particulate emissions.
      In the  TSDF context an example of a preventive control technique
involves traffic routing of haul truck vehicles in the staging areas
proximate to an active  landfill face.  The objective of traffic routing
is  to minimize the contact between vehicle wheels and waste material and
thus  limit the potential spreading of contamination material to the

                                    1-8

-------
facility roadways.  In this fashion the source extent of contaminated
material is effectively reduced.
     An example of a mitigative measure involves the application of water
to unpaved travel surfaces in order to suppress the entrainment of gross
or contaminated particulate by vehicle traffic.  The important point here
is that relative to contaminated particulate, it is assumed that the
roadways are characterized by some (albeit unknown) level of contamina-
tion (a) associated with long-term waste disposal operations, and that
the objective is to minimize the entrainment of this material into the
atmosphere.

1.4  ON-SITE VERSUS OFF-SITE FACILITIES
     Land-based hazardous waste TSDFs generally are classified as on-site
facilities, i.e., those facilities that generate and treat or dispose of
waste streams on-site, versus off-site facilities—those that accept
wastes from a variety of generators for treatment, storage, and
disposal.  According to recently released information,s 165 on-site
facilities have received Part B permits; there are approximately
30 facilities with Part B permits still pending.  According to the avail-
able information 35 off-site facilities have applied for Part B status,
22 facilities have received a permit; and 6 facilities have been ordered
to close.
     Relative to this document, the distinction between on-site and off-
site facilities would appear to have several important implications.  For
example, as a general rule, one might expect on-site facilities to handle
a more  limited range of waste streams than off-site facilities.
Similarly, it is reasonable to expect that on-site facilities will have
more complete knowledge of waste stream composition.  Thus, in the
absence of detailed sampling and analysis (S&A) programs, specification
of a for the calculation scheme (e.g., Eq. 1-2) may prove easier for on-
site facilities than for off-site ones.
     In evaluating particulate emissions from on-site facilities, one of
the most difficult problems may be in separating that portion of the
total facility associated with RCRA-related activities from the remainder
                                   1-9

-------
of the facility which presumably is not subject to regulation under
RCRA.  Undoubtedly, this makes the evaluation very site-specific.

1.5  ORGANIZATION OF THIS DOCUMENT
     Each section of this guidance document focuses on a different source
category:
     Section 2.0~Paved and Unpaved Roads
     Section 3.0—Open Waste Piles and Staging Areas
     Section 4.0—Dry Surface Impoundments
     Section 5.0—Landfills
     Section 6.0—Land Treatment
     Section 7.0—Waste Stabilization
     Each section begins with an overview of the source category,
describing emission characteristics and mechanisms.  Following this,
available emission factors are presented to provide a basis for analyzing
the operative nature of control measures.  Next, demonstrated control
techniques are discussed in terms of estimating efficiency and determin-
ing costs of implementation.  Suggested regulatory formats explain the
"philosophy" used in implementing the preceding technical discussions in
viable regulations and compliance actions.
     In addition, a series of appendices are included.  These provide
information on (a) terminology used in this manual, (b) a general costing
procedure used for open dust source controls, (c) screening atmospheric
modeling techniques and health risk information, and (d) generic methods
for  sampling and analysis of material samples from TSDFs.

1.6  REFERENCES FOR SECTION 1
  1.  U.S. Environmental Protection Agency.  Permit Applicants' Guidance
     Manual for Hazardous Waste Land Treatment, Storage, and Disposal
     Facilities.  Final Draft, EPA 530 SW-84-004, Washington, D.C.  May
     1984.
  2.  U.S. Environmental Protection Agency.  Compilation of Air Pollution
     Emission Factors  (AP-42), Fourth Edition, September 1985; Supple-
     ment A, October 1986; Supplement B, September 1988.  Research
     Triangle Park, North Carolina.  September 1988.
                                   1-10

-------
3.  Cowherd, C., G. E.  Muleski, P.  J.  Englehart, and D.  A.  Gillette.
    Rapid Assessment of Exposure to Participate Emissions from Surface
    Contamination Sites.  EPA/600-8-85/002,  prepared for U.S.  Environ-
    mental Protection Agency, Office of Health and Environmental  Assess-
    ment, Washington, D.C.  February 1985.

4.  Englehart, P., and D.  Wallace.   Assessment of Hazardous Waste TSDF
    Participate Emissions.  Final Report,  EPA Contract No.  68-02-3891,
    Work Assignment Nos. 5 and 13.   October  1986.

5.  Business Publishers, Inc.  "Hazardous  Waste News," Vol. 10, No.  49.
    p. 42.  December 12, 1988.
                                  1-11

-------
                       2.0   PAVED AND  UNPAVED ROADS

     Unlike the sources described in other sections of this manual, the
emissions from paved and unpaved roads at hazardous waste (HW) treatment,
storage, and disposal facilities (TSDFs) are not due to an identifiable
unit operation at each facility.  Rather, the roads serve as "linkages"
between the various unit operations (e.g., access from the gate to the
active disposal area, transfer of solidified wastes from the solidifica-
tion unit to a disposal unit, etc.).  Consequently, the approach taken in
this section will be somewhat unique in that the emission sources will be
discussed in very general terms and not in the context of a specific TSDF
unit operation.
     Particulate emissions occur whenever a vehicle travels over a paved
or unpaved surface, such as a road or an area proximate to a TSDF unit
operation.  Note that travel areas adjacent to TSDF operations may have a
high potential for contaminated particulate emissions, with the potential
inversely related to the level of "good operating practices" at the unit
operation.  In the case of paved roads, emissions originate with
(a) resuspension from vehicle tires and undercarriages or (b) material
deposited onto the surface.  Material  tracked onto the road from poten-
tially contaminated travel areas surrounding TSDF operations will be
spread along the road's length and eventually some of that material will
become airborne.  While track-on of material is usually more noticeable
on paved roads, the spreading of potentially contaminated material also
occurs on unpaved surfaces.
     Past surveys of TSDFs have in fact indicated that unpaved road
surfaces at these types of facilities may have RCRA metal and semi vola-
tile organic contamination levels essentially comparable to levels in
samples taken from process-related surfaces (e.g., landfill areas, areas
surrounding a stabilization/solidification unit, etc.).1  Because of the
large volume of traffic associated with many facilities, those surveys

                                   2-1

-------
suggest that general vehicular traffic on plant roads may be the princi-
pal source of contaminated emissions associated with TSDFs.  While the
source of road surface contamination cannot be definitively identified,
the two most likely causes are (a) spillage from vehicles hauling the
waste or (b) "track-out" of material from surfaces surrounding process
operations at the TSDF.
     In many respects, control measures commonly applied to alleviate
"gross" (i.e., uncontaminated) particulate emissions may actually com-
pound contaminated emission^.  For example, travel areas surrounding an
active TSDF unit operation may be watered to control emissions.  However,
watering tends to increase the amount of track-out onto an adjacent paved
road, and the deposited material soon dries out and becomes airborne
under normal plant traffic conditions.

2.1  PAVED ROADS

2.1.1  Source Description
     Particulate emissions occur whenever a vehicle travels over a paved
surface such as public and industrial roads and parking lots.  These
emissions may originate from materials previously deposited on the travel
surface or from the resuspension of material from tires and under-
carriages.  In general, emissions are due to the surface material loading
(measured as mass per unit area of the travel lanes) and that loading  is
continuously replenished by other sources (e.g., spillage, pavement wear,
carryout from adjacent areas, etc.).  Because of the importance of sur-
face loading, available control techniques either attempt to prevent
material from being deposited on the surface or to remove any material
that has been deposited onto the travel  lanes.2

     2.1.1.1  Public vs.  Industrial Roads.  Although publicly owned roads
are not common in HW TSDFs, transport to commercial facilities will
almost always entail some travel on public roads.  Thus, public roads  may
also be subject  (although to a  lesser degree than in-plant roads) to
carry-out of  possibly  contaminated material from the TSDF.  Should con-
taminated material  reach  the publicly owned system of roads, substantial
                                    2-2

-------
increases in metals or organics emissions might be expected because all
traffic (and not just that related to the TSDF) will cause resuspension.

     2.1.1.2  Site Characteristics.  While the mechanisms of material
deposition and particulate resuspension are largely the same for public
and industrial paved roads, there are often major differences in surface
loading characteristics, emission levels, traffic characteristics, and
viable control options.3  For the purposes of estimating both gross and
contaminated particulate emissions and developing control programs, the
distinction between public and industrial roads is not so much a question
of ownership/maintenance as one of surface loading and traffic character-
istics.  Decisions of this sort are discussed in Section 2.1.2 of this
report.

     2.1.1.3  Dust Emitting Activities.  Although loose material on
roadways and parking lots may become entrained from activities other than
vehicular traffic (erosion during high wind events, for example), the
principal long-term source of emissions for active roads and parking  lots
is undoubtedly vehicle travel.  Vehicle-related activities are the only
dust emitting activities considered in the rest of this chapter.

     2.1.1.4  Potential Contamination Problems.  As mentioned in the
introduction to this chapter, significant contaminated air emissions from
paved travel surfaces may occur if the surface loading contains measur-
able fractions of RCRA metals or organic compounds.  Contaminants in the
loose surface material on the roadway may be due to spillage during
transport of HW or track-on of material which has become contaminated
because of the various unit operations at the facility.  Both sources may
be at least partially controlled by instituting better operating prac-
tices at the facility.  Again (as discussed in this section's introduc-
tion), it is important to note that control measures designed to reduce
emissions at a unit operation may actually compound contaminated emis-
sions on adjacent roadways.
                                   2-3

-------
2.1.2  Estimation of Uncontrolled Emissions

     2.1.2.1  Particulate Emissions.  Estimation of the total uncon-
trolled participate emission rate R from paved TSDF roadways follows the
general emission model  presented earlier as Eq. 1-2.

                          R *  z  R, =  z  e
-------
     The product sL represents the mass of silt-size dust particles per
unit area of the road surface and is usually termed the "silt loading."
As is the case for all predictive models, the use of site-specific values of sL
is strongly recommended.   Recognizing that this  is not always feasible, a
summary of measured values is given as Table 2-1.

       TABLE 2-1.  POSSIBLE DEFAULT VALUES FOR PAVED ROAD EMISSION
                            ESTIMATES AT TSDFs

   Parameter (units)              Probable range     Possible default
   Silt loading (g/m2)                 0.3-30a               5a'b
   Vehicle weight (Mg)
     —General plant vehicles           2-9                  3^
     —Commercial haulers               9-45                20^
     —Plant haul trucks and           20-50                30°
         equipment
   aBased on measurements taken from paved plant and adjacent public
    roads for 5 HW and municipal solid waste (MSW) landfills.
    As noted in text, use of site-specific data is strongly
    recommended.
     The appropriate emission factor model (i.e., the constants a, b, and
c) for a road segment depends upon:
   • The mass of loose aggregate material less than 200 mesh per unit
     area of a paved road surface.
   • The average weight of vehicles traveling on the road.
Selection of the appropriate emission factor is summarized in Table 2-2.
An example problem illustrating the use of emission factor models is
given in Section 2.3.
                                   2-5

-------
                  TABLE 2-2.  SELECTION OF PAVED ROAD EMISSION FACTOR

Silt loading^ (sL)
g/m2
sL < 2
sL < 2
sL > 2d
2 < sL < 15
sL > 15d
oz/yd2
< 0.06
< 0.06
> 0.06
0.06 < sL < 0.44
> 0.44
Range weight
(W)

W
W
W
W
W
Mg
> 4
< 4
> 6
< 6
< 6
Ton
> 4.4
< 4.4
> 6.6
< 6.6
< 6.6
Applicable PM]Q emission
f actor
g/VKTa
. ,b
220 (sL/12)0'3
c
2.28 (sL/0.5)0'8
t>
220 (sL/12)0'3
220 (sL/12)0'3
93
lb/VMTa
b
0.78 (sL/0.35)0"5
c
0.0081 (sL/0.015)0'8
_b
0.78 (sL/0.35)0"5
_b
0.78 (sL/0.35)0-5
0.33

aVKT = Vehicle kilometers traveled, VMT = vehicle miles traveled.
^Commonly referred to as the "industrial" paved  road model.
cCommonly referred to as the "urban" paved road  model.
 For heavily  loaded surfaces (i.e., sL > ~ 300 to 400 g/m  (9 to 12 oz/yd ), it  is
 recommended  that the resulting estimate be compared to that from the unpaved road models
 (Section 2.2 of this manual), and the smaller of the two values used.


      2.1.2.2  Contaminated Emissions.  The emission rate  Rj of particu-
late  contaminated with a compound  j can be represented  as
      n
R. =  z   R.
 J    1.1   1
n
z
                                              lij
                                                   e.A.
(2-3)
where terms are  as  defined earlier and:
      R^J = emission rate of  contaminant  j  from road  segment i
      a.jj = the fraction of contaminant j in the participate emissions
            from  road segment i

      Because road surface contamination  may result from spillage and/or
track-on of material from a  wide variety of wastes and  areas around  unit
operations, no general rule  of  thumb for an appropriate value of a^j
(such as conservatively assuming 50% of  the highest  waste stream
transported over the road) should be applied. Consequently, it is
                                      2-6

-------
strongly recommended that material  sampling be undertaken at operating
facilities, and that those samples be analyzed for compounds of concern.
     Information on o^,- can be determined either by direct or surrogate
measurement.  Direct measurement of o^,- requires the determination of
specific contaminants in roadway emission samples collected at the
TSDF.  A surrogate measure of a^- is the fraction of contaminant j  in the
silt loading for road i.  An alternate surrogate measure would treat o^,- as
a constant, set equal to the average fraction of contaminant j in all
available silt loading samples.  In that event,.Eq. 2-3  above simplifies
considerably to the form

                                 R  . =  ^TR                             (2-4)
                                  J    J

where terms are as defined in Eqs. 2-1 and 2-3 and:

                       o7 = mean value of  the a^-'s
                        J                       J

     Determining a^j by the alternate surrogate measurement technique
will provide sufficient  information to conduct an  initial risk assessment
for a TSDF.  If the initial risk assessment  indicates that fugitive  PM
emissions result in unacceptable risks, the  TSDF owner/operator should
conduct an approved fugitive PM emission monitoring program in order to
directly measure a.,-,-.

2.1.3  Demonstrated Control Techniques
     As mentioned earlier, because emissions  are directly related to the
amount of material present on the  road surface, control  techniques for
paved roads either attempt to prevent material from being deposited  onto
the surface (preventive controls)  or attempt  to remove any deposited
material (mitigative controls).  Preventive  and mitigative control mea-
sures are discussed separately below.  Measurement-based control effi-
ciency values for control methods  are presented in Table 2-3.  Note  that
all values presented refer to mitigative controls.
                                   2-7

-------
       TABLE 2-3.  ESTIMATED METHODS FOR AVERAGE EFFICIENCY VALUES
                         FOR  PAVED  ROAD  CONTROLS3
   Control
Average efficiency, C
        (*)b
      Comments
Vacuum sweeping


Water flushing


Water flushing
followed by
broom sweeping
   168/D, D > 2
 _J 69-0. 116 V, V < 299
   I 10,300/V,   V > 299
   l 96-0.132 V, V < 365
   I 17,500/V,   V > 365
340 m3/m1n (12,000 cfm)
blower tested

Water applied at
2.2 L/mz (0.48 gal/yd*)


Water applied at
2,.2 L/m2 (0.48 gal/yd*)
 Based on PM-15 field emission measurements as given in Reference 2.
 PM-10 efficiency may be assumed equal to that for PM-15.

•'in the expressions, D and V represent the number of days or vehicle
 passes, respectively, since application.
     2.1.3.1  Preventive Control Techniques.  Preventive paved road

controls play a very important role in reducing contaminated particulate

emissions at TSDFs.  The two most likely contributors to contaminated

surface loadings—spillage and carry-out from nearby operational areas--

are undoubtedly better controlled at their source rather than by attempt-

ing to clean the surface of the deposits.  Furthermore, in the case of

gross particulate emissions, preventive measures may also provide a more

cost-effective control program than mitigative measures.  An outline of

preventive measures applicable for TSDFs is provided in Table 2-4.  Each

alternative is discussed in Section 2.1.4.
                                    2-8

-------
          TABLE 2-4.  OUTLINE OF PREVENTIVE PAVED ROAD CONTROL
                      MEASURES APPLICABLE TO TSDFs
      • Spills from haul  truck

        —require loads  to be covered
        —wet loads prior to transport
        —require a minimum amount  of freeboard  to be maintained
        Carry-out from other operational  areas

        —semicontinuous  cleanup of  access  point
        —vehicle wash or grizzly at exit
        —pave access point
        —redistribution  of traffic  patterns  (e.g.,  strict
          enforcement of  one-way routes)  in operational  areas  to
          reduce trackout
        Wind/water erosion from adjacent areas

        —vegetative stabilization of open areas
        —physical controls (e.g., wind breaks  or storm water
          ditches, etc.)
     The major drawback in designing and assessing  the  performance  of

preventive control  programs is due to the fact  that there  are  few,  if

any, efficiency values associated with the measures.  Because  these types

of controls attempt to prevent deposition of additional  material  onto

paved surfaces, quantitative measurements before  and  after control  imple-

mentation are generally not feasible.  Furthermore, interpretation  of

what data are available is often complicated.

     Instead of requiring control effectiveness estimates  for  preventive

measures at TSDFs,  regulators may prefer to require responsible  parties

to list in their Part B "Wind Dispersal  Flow Control  Plan" all paved road

preventive measures contemplated for use at their facility.  While  data

are generally not available to document  the effectiveness  of these  mea-

sures, regulators may in that case choose to require  a  systematic field

sampling program (of the type discussed  in Section  2.1.5)  to demonstrate
                                   2-9

-------
that the mean silt loading does not exceed a permit-specified value (for
example, 5 g/m*).  (Note that the regulatory agency should also reserve
the right for their own spot inspections of silt loading.)  Should the
specified value be exceeded, the permit would contain provisions calling
for more strenuous implementation of preventive (and possibly mitigative)
controls as well as a more intensive field sampling program to demon-
strate compliance with the permitted limit.

     2.1.3.2  Mitigative Control Techniques.  While preventive controls
are to be preferred for contaminated particulate emission control, pre-
ventive measures may not be able to provide the gross particulate control
effectiveness desired at a TSDF.  Furthermore, in the context of the
mitigative vs. preventive distinctions made in Section 1 of this docu-
ment, mitigative measures are required to reduce contaminated emissions
due to the background metals and organics levels inherent with the long-
term waste disposal operations at the TSDF.
     Generally recognized mitigative controls for paved roads were
presented in Table 2-4, with the controls falling in the categories of:
     •  Broom sweeping;  This control employs mechanical cleaning of the
travel surface.  Much of the effect may be viewed as "cosmetic," in the
sense that, while the roadway appears to be much cleaner, a substantial
fraction of the original loading may be emitted to the atmosphere during
the cleaning process.1*  Because there is credence to the claim that broom
sweeping is as much a source as it is a control of particulate emissions,
the possible increase in contaminated particulate emissions without provision
for capturing that mass may preclude further consideration of this con-
trol.  Regulators should be advised, however, that broom sweeping may
provide the only mitigative paved road control measure that is feasible
throughout the year in all portions of this country.
      •  Vacuum  sweeping:  Material may also be removed from the travel
surface by entraining particles in a moving air stream.  Hoppers contain
the material collected.   In an open system, air exhausts through a
filter;  in closed  systems, the air is continuously recycled."•
      •  Water flushing:  Street flashers remove surface loadings under
the  action of high-pressure water sprays.  Some systems supplement the

                                   2-10

-------
cleaning process with broom sweeping after flushing (sweeping emissions
are thus wet-suppressed).*»

2.1.4  Control Performance Estimation/Specification

     2.1.4.1  Preventive Measures.
     Spillage from haul trucks and waste carriers—Because spillage
during the transport of wastes and other materials causes some of the
road loadings present on paved roads at TSOFs, it is clear that better
control of spillage will reduce particulate emissions from those roads.
In terms of contaminated particulate, it is equally clear that control of
waste haulers (including plant traffic hauling stabilized and other
treated waste) is of primary importance.  What is not clear in these
situations, however, is how much of the surface silt loading on a paved
road can be attributed to spillage.  Without that final piece of informa-
tion, reliable estimation of the control efficiency afforded by spillage
controls is unlikely.
     Because of the difficulty in assessing the effectiveness of this
type of control and because reasonable requirements in permits to prevent
spillage should not be considered overly restrictive, it is recommended
that all wind dispersal flow control plans contain provisions to effec-
tively eliminate spillage of wastes onto paved roads.  This could be
accomplished by (1) specifying covered or enclosed trucks or minimal
freeboard requirements, (2) accepting no loads from haulers exhibiting
visible spillage, (3) requiring all accidental spills to be reported
immediately to the operations manager, and (4) rerouting traffic around
any accidental spills which will be cleaned as soon as possible (and in
any case less than, say 8 hours).
     Carryout from adjacent unpaved areas—Mud and dirt carryout from
operational areas, as well as the spreading of contaminated material from
those areas, can account for a large fraction of the total surface
loading on paved roads at TSDFs.  Preventive control measures applied to
this source of road loadings can result in substantially lower gross and
contaminated particulate emission levels.
                                   2-11

-------
     As was the case for spillage controls, reliable estimation of the
effectiveness of carryout control programs is not generally possible.
Quantification of control efficiencies for preventive techniques is
essentially Impossible using the standard "before and after" measurement
approach.  Furthermore, tracking of material onto a paved roadway results
in a great deal of spatial variation in loading about the access point.
This variation in loading complicates the modeling of emission reductions
as well as the simple estimation of the efficiency.  The methodology
described below will result in upper bounds of emission reductions.  That  is,
the control afforded by carryout prevention cannot be easily described in
terms of percent reduction but rather is discussed in terms of mass
emissions prevented.
     For an individual access point from an unpaved operational area to
an adjacent paved road, let N represent the daily number of vehicles
leaving the area.  Let E be given by*

                            E+ .   36 (N < 25)                        (2.5)
                            1  ~   91  (N > 25)                        V    '

Here, E* is the unit  PM10  emission increase in  g/vehicle.2   Finally,  if M
represents the daily number of total vehicle passes on the adjacent paved
road, then E+ x M is the total increase in PMi0 emissions due to carry-
out.**  Assuming complete prevention of carryout, then E"1" x M also
represents the effective emission reduction.
     The emission reduction calculated above assumes that essentially  all
carryout from the operational area  is controlled and, as such, should  be
viewed as an upper limit.  In use,  regulators may choose to assign an
effective  level of carryout control by assuming that only a fraction of
the total  is controlled.
  *The  carryout  emission  factors  presented  in  Reference  2 were based on
   the  urban  paved  road model  (presented  in Table  2-2).  The  values given
   in Equation 2-5  are based on a revision  of  the  underlying  data using
   the  industrial paved road model  (as  given in  Table  2-2).
 **The  emission  increase  in contaminated  particulate matter can be esti-
   mated  as  a,- x E"1" x M,  where o^ is  the  contamination level  for the com-
   pound  j  in the operational  area's  surface material.
                                   2-12

-------
     Methods used to control carryout consist of mltigative measures on
the paved surface or preventive measures applied at the operational
areas.  Discussion of mitigative approaches are discussed in Sec-
tion 2.1.4.2 of this chapter; preventive measures applied at operational
areas are discussed in the remaining sections of this report.
     As noted earlier, instead of requiring control effectiveness
estimates for preventive measures as TSDFs, regulators may prefer to
require responsible parties to list in their Part B "Wind Dispersal Flow
Control Plan" all paved road preventive measures contemplated for use at
their facility.  Regulators may choose to require in the permit that the
responsible parties conduct a systematic field sampling program (of the
type discussed in Section 2.1.5) to demonstrate that the mean silt load-
ing does not exceed a permit-specified value (for example, 5 g/m2).
Should this value be exceeded, the permit would contain provisions
calling for more strenuous implementation of preventive (and possibly
mitigative) controls as well as a more intensive field sampling program
to demonstrate compliance with the permitted limit.
     Other preventive measures—Numerous other preventive controls could
be applied to paved roads at TSDFs, ranging from wind fences to prevent
material from operational areas being blown onto the roadways to the
installation of stormwater controls.  No data are known to exist that
quantify the PM10 emissions reductions attributable to these measures.
It is recommended that regulators require responsible parties to list in
their Part B "Wind Dispersal Flow Control Plan" all paved road preventive
measures contemplated for use at their facility.  Because data are gen-
erally not available to document the effectiveness of these measures,
regulators may again require in the permit that the responsible parties
conduct a systematic field sampling program of the type discussed
above.  The permit should contain provisions for more strenuous preven-
tive  (and possibly mitigative) controls, should the field sampling pro-
gram demonstrate noncompliance with the permitted limit.
                                   2-13

-------
     2.1.4.2  Mitiqative Control Techniques.  Although the preventive
measures described earlier can be viewed as very cost-effective, these
measures alone may not be able to provide the total paved road emissions
reduction sought (see the discussion in Section 2.1.5).  Furthermore, in
the context of controlling contaminated particulate emissions, mitigative
controls can be considered as a "second line of defense" in safeguarding
public health and the environment.  This section discusses the effective-
ness of mitigative paved road control techniques.
     Broom sweeping of roads—Mechanical street cleaners employ rotary
brooms to remove surface materials from roads and parking lots.  Much of
their effect is cosmetic in that, while the roadway appears much cleaner,
a substantial fraction of the original loading is emitted during clean-
ing.  Thus, "dry" broom sweeping is not recommended as a means of
removing surface material from roads.
     Vacuum sweeping of roads—Vacuum sweepers remove material from paved
surfaces by entraining particles in a moving air stream.  A hopper is
used to contain collected material and air exhausts through a filter sys-
tem in a open loop.  A regenerative sweeper functions in much the same
way, although the air is continuously recycled.  In addition to the
vacuum pickup heads, a sweeper may also be equipped with gutter and other
brooms to enhance collection.
     Control efficiency values were given earlier in Table 2-3.  Avail-
able data show considerable scatter, ranging from a field measurement
showing no effectiveness (over baseline uncontrolled emissions) to
another field measurement of  58%.  An average of the field measurements
would  indicate a efficiency of 34%.  In addition, the estimated upper
limits for PM10 control of urban roads compare fairly well with that
average.  Because adequate controlled emission estimates can be obtained
using  the models given in Section 2.1.2, it is recommended that material
loading samples be employed,  if possible, in conjunction with the model
to  obtain better estimates of control effectiveness.
     Cost elements involved with vacuum sweeping include the following
capital and  operating/maintenance  (O&M) expenses:
     Capital:  Purchase of truck or other device
     O&M:   Fuel, replacement  parts, truck maintenance, operator labor
cost
                                   2-14

-------
Data presented in Reference 5 provides the following estimates for a
vacuum sweeping program:
     Initial capital expense:  36,800 $/truck
     Annual O&M expense:  34,200 $/truck
     All costs are based on April 1985 dollars.  Determination of the
number of trucks necessary can be made by assuming that 4 mi can be swept
per unit per 8-h shift.6  Additional cost data for a vacuum sweeping
program is provided in Table 2-5.  Note that, unlike the public and
industrial paved road estimates given above, the cost data in Table 2-5
are based on actual values reported by an industrial plant during the
first few years of an intensive road-cleaning program.

       TABLE 2-5.   MISCELLANEOUS OPERATION/DESIGN AND COST DATA FOR
                       VACUUM SWEEPING PAVED ROADSd

Purchase price:                                       $72,000 (1980)
Estimated life expectancy:                            5 yr
Approximate annual operating cost during 1981:        $214,000
Fuel consumption:                                     4 mi/gal
Hopper capacity:                                      10 yd3
Vacuum blower capacity:                               12,000 ftVmin
Vehicle weight:                                       32,000 Ib
Width of area cleaned per pass:b                      5 ft
Normal sweeping speed:                                5 mi/h
Velocity at suction head:                             NA
Type of dust control system  (i.e., wet or dry):       Wet
Reference 6.  Purchase cost is actual cost in year purchased; other
 costs in 1981 dollars.
^Multiple passes required.
                                   2-15

-------
     Enforcement of a vacuum sweeping dust control program would ideally
consist of two complementary approaches, record keeping and field inspec-
tion.  The first requires the owner to maintain adequate records that
would document to agency personnel's satisfaction that a regular cleaning
program is in place.  The second approach involves agency spot checks of
controlled roads by taking a material sample frcm the road.  As before,
the second approach is discussed in greater detail in Section 2.1.5.
Note that some sample collection may be necessary to estimate control
performance.
     Records must be kept that document the frequency of vacuum sweeping
paved surfaces.  Pertinent parameters to be specified in a control plan
and to be regularly recorded include:
     General Information to be Specified in the Plan
     1.  All road segments and parking locations referenced on a map
available to both the responsible party and the regulatory agency.
     2.  Length of each road and area of each parking lot.
     3.  Type of control applied to each road/area and planned frequency
of application.
     4.  Any provisions for weather (e.g., 1/4-in of rainfall win be
substituted for one treatment, etc.).
     Specific Records for Each Road Segment/Parking Area Treatment
     1.  Date of treatment.
     2.  Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
     3.  Start and stop times on a particular segment/parking lot, aver-
age  speed, number of passes.
     4.  Qualitative description of loading before and after treatment.
     5.  Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
     General Records to be Kept
     1.  Equipment maintenance records.
     2.  Meteorological log  (to the extent that weather  influences the
control program—see above).
     3.  Any equipment malfunctions or  downtime.
                                   2-16

-------
     In addition to those items related to control applications, some of
the regulatory formats suggested in Section 2.1.5 require that additional
records be kept.  These records may include surface material samples or
traffic counts.  Traffic counts may be recorded either manually or using
automatic devices.
     Water flushing of roads—Street flushers remove surface materials
from roads and parking lots using high pressure water sprays.  Some sys-
tems supplement the cleaning with broom sweeping after flushing.  Note
that the purpose of the program is to remove material from the road sur-
face; in some industries, water is regularly applied to roads to directly
control emissions (i.e., as in unpaved roads).  Unlike the two sweeping
methods, flushing faces some obvious drawbacks in terms of water usage,
potential water pollution, and the frequent need to return to the water
source.  However, flushing generally tends to be more effective in con-
trolling particulate emissions.  Caution must be exercised in flushing
near access points to operational areas.  As noted above, watering may
increase the spreading of contamination from operating areas onto paved
roads.  Furthermore, if the paved roads are already contaminated, flush-
ing may pose additional problems in terms of surface and ground-water
contamination.
     Equations to estimate instantaneous control efficiency values were
given in Table 2-3.  Note that water flushing and flushing followed by
broom sweeping represent the two most effective control methods (on the
basis of field emission measurements) given in that table.
     Cost elements involved with broom sweeping include the following
capital and operating/maintenance (O&M) expenses:
     Capital:  Purchase of truck or other device
     O&M:  Fuel, replacement parts (possible including brushes), truck
maintenance, operator labor, water
     Cost data presented in Reference 5 provides the following estimates
for a flushing program:
     Initial capital expense:  18,400 $/truck
     Annual O&M expense:  27,600 $/truck
All costs are based on April 1985 dollars.  Determination of the number
of trucks required can be based on the assumption that 3 to 5 mi can be
                                   2-17

-------
flushed or flushed and broom swept per unit per 8-h shift, respectively.6
Additional cost/design data are provided as Table 2-6.  As before, the
data in the table represent actual reported costs at an industrial plant.
     Enforcement of a road flushing (possibly supplemented by broom
sweeping) program could consist of two approaches, as before.  The first
requires the owner to maintain adequate records that document that a
regular cleaning program is in place.  The second involves agency spot
checks of controlled roads by taking a material sample from the road.
While resulting estimates (using the models in Table 2-2 in
Section 2.1.2) of controlled emissions should be adequate for a flushing
program, the estimates are probably substantially overestimated in a
flushing/broom sweeping program.2
     Records must be kept that document the frequency of water flushing
of paved surfaces.  Pertinent parameters to be specified in a control
plan and to be regularly recorded include:
     General Information to be Specified in the Plan
     1.  All road segments and parking locations referenced on a map
available to both the responsible party and the regulatory agency.
     2.  Length of each road and area of each parking lot.
     3.  Type of control applied to each road/area and planned frequency
of application.
     4.  Provisions for weather (e.g., program suspended for periods of
freezi ng temperatures).
     Specific Records for Each Road Segment/Parking Area Treatment
     1.  Date of treatment.
     2.  Operator's initials (note that the operator may keep a separate
log whose  information is transferred to the environmental staff's data
sheets).
     3.  Start and stop times on a particular segment/parking lot, aver-
age speed,  number of passes.
     4.  Start and stop times for refilling tanks.
     5.  Qualitative description of  loading before and after treatment.
     6.  Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
                                   2-18

-------
       TABLE 2-6.  MISCELLANEOUS OPERATION/DESIGN  AND  COST  DATA  FOR
                  FLUSHING/BROOM  SWEEPING  PAVED ROADS*
Purchase price:
Estimated life expectancy:
Approximate annual operating cost during 1981:
Vehicle weight (dry):
Water tank capacity:
Normal vehicle speed:
Water pressure at nozzles:
Vehicle weight (wet):
Fuel consumption:
Water flow at nozzles:
Hopper capacity:
Daily water consumption:
Degree of water treatment:
$68,000 (1976)
10 yr
$57,000
NA Ib
8,000 gal
4 mi/h
50 psig
NA Ib
7 mi/gal
188 gal/min
40 yd 3
30,000 gal
1,800 gal/mil
Reference 6.  Purchase cost is actual cost in year purchased; other
 costs in 1981 dollars.
                                   2-19

-------
     General  Records to be Kept
     1.  Equipment maintenance records.
     2.  Meteorological log (to the extent that weather influences the
control program—see above).
     3.  Any equipment malfunctions or downtime.
     In addition to those items related to control applications, some of
the regulatory formats suggested in Section 2.1.5 require that additional
records be kept.  These records may include surface material samples or
traffic counts.

2.1.5  Procedures for Compliance Determination
     Record review of control programs (e.g., vacuum sweeping, etc.) and
field checks (i.e., road silt loading sampling) will provide the likely
means of compliance determination for these sources.  Because paved road
emissions are directly related to the surface silt  loading, the most
reliable regulatory formats are based on loading.  Formats viable for
other open dust sources—including opacity measurements, visible
emissions at the property line—are generally not applicable for paved
roads because of the lower unit emission levels involved (e.g., there are
usually no visible plumes from a vehicle pass).
     An alternative format is presented below to suggest how a quantita-
tive method could be incorporated in a regulation.  If the silt loading
on a road ever exceeds the "action level" silt  loading specified in a
permit, the regulatory agency may require the responsible party to reduce
the silt loading to a  level less than the action level.  The action level
may be an agency-supplied multiple of either baseline measurements or
values provided  in Table 2-1 and should correspond to maximum allowable
emission factors.  The means of reduction will  be left to the discretion
of the responsible party and could consist of either preventive or miti-
gative controls.  The maximum allowed silt loading requirement could be
made part of an  operating  permit.  Note that these  action levels could be
specific to  individual roads, apply to all roads  in a plant, or be based
on plant traffic  levels.   Because some TSOFs may contain many paved
roads, the regulatory  agency may choose to set  plant-wide goals (such as
vacuum sweeping  each  road  twice per week) rather  than source-specific
programs.
                                   2-20

-------
     The field measurement of silt loading could either be made a
requirement of the responsible party or be assigned to agency inspection
personnel, or a combination of the two could be used.  In either event,
certain features of the measurement technique must be specified:
     1.  The sampling method used to determine silt loading for compli-
ance inspection should conform to the technique used to develop the paved
road equations.  That technique is specified 1n Appendix D and should be
made part of a standard operating procedure (SOP) for regulatory person-
nel or part of the operating permit.
     2.  Arrangements must be made to account for spatial variation of
surface silt loading.  Possible suggestions include:  (a) visually deter-
mining the heaviest loading on the road and selecting that spot for sam-
pling, (b) sampling the midpoint of the road length segment of interest,
and (c) sampling preselected (possibly on the basis of safety consid-
erations) strips on the road surface (note that the samples may be
aggregated).
     3.  Provision should be made to grant a "grace period" following a
spill or other accidental increase in loading and to reroute traffic
around spills  (if possible).  An 8-h period is suggested to allow time
for the responsible party to clean the affected area so that measurements
of silt loading will not be biased because of this occurrence.  This time
allowance snould be made part of a permit.
     The control efficiency equations presented in Table 2-3 provide
another potential regulatory format for TSDF paved road sources.  This
approach involves inspection of both plant road cleaning records and
traffic counts.  By combining the two sets of information, regulatory
personnel would be able to determine average efficiency values for the
plant's controlled paved roads.  Provision must be made to collect traf-
fic information.  The traffic data may require more frequent inspection
visits than surface loading samples; however, analysis is more easily
accomplished.  Surface loading sampling provides an additional means for
checking the success of achieving the estimated control efficiency.  Sam-
ple data forms for traffic counts and silt loading measurements given as
Figures 2-1 and 2-2, respectively.
                                   2-21

-------
                       TRAFFIC COUNT LOG
Date	        By.
Road Location:	
Road Type:	
Sampling Start Time:	   Stop Time:.
Vehicle Tvoa   Axles/Wheels   1	2_-2-_£_-5	6  7   8   9  10    Total
                  Figure  2-1.  Traffic count log,.
                                 2-22

-------
PAVED ROAD SURFACE SAMPLING
Data Bv
Site of Sampli
No. of Traffic La
Surface Condit
Sample No.

























na:
tnes:
on:
Type of Pavement: Asphalf Concrete



Vac. Bag
No.

























Time

























Location*

























Sample Area

























Broom
Swept?
(y/n)

























Use code given on plant map for segment identification and indicate sample location on map.
          Figure  2-2.   Sampling data form  for paved roads.
                                   2-23

-------
2.2  UNPAVED ROADS

2.2.1  Source Description
     Participate emissions occur whenever a vehicle travels over an
unpaved surface such as roads and parking lots.  Unlike a paved road,
however, the unpaved road itself is the source of emissions rather than
any "surface loading."  Of the various categories of open dust sources,
unpaved travel surfaces have historically accounted for the greatest
share of particulate emissions in industrial settings.  In a 1987 inven-
tory of four municipal solid waste (MSW) landfills in the Chicago area,
for example, vehicle travel on roads and access areas (i.e., exclusive of
compaction, lift construction, and covering) accounted for about 70% to
80% of the overall PM10 emissions from each facility.7  Furthermore, as
noted in the introduction to this chapter, a recent "scoping" study of
particulate emissions from HW TSOFs suggested that unpaved roads rather than
process operations may be the major source of contaminated emissions  from these
types of facilities.1
     Recognition of the importance of unpaved roads led naturally to an
interest in their control by the iron and steel, mining, and other  indus-
tries.  As a result, the portion of total open dust emissions attribut-
able to unpaved surfaces in those industries has decreased dramatically
over the past 10 years.8  However, it is important to note that MSW and
HW disposal sites have only recently attracted the general interest of
air pollution regulatory personnel.  Consequently,  one might expect that
unpaved road control programs are not as widely  implemented at TSDFs as
in many other industries and that the portion of total open dust emis-
sions at TSOFs attributable to this source category is high.

     2.2.1.1  Site Characteristics.  Travel surfaces  at TSDFs may be
unpaved for many reasons.  For example,  roads used by very heavy vehicles
(such as haul trucks) or subject to considerable amounts of spillage of
cover and  other materials during transport operations are not suitable
candidates for paving.  Some roads are used for  only  short periods  of
time to access certain disposal  areas; other more permanent roads may
have poorly constructed bases.   In either case,  paving is impractical.

                                   2-24

-------
Temporary roads can often be controlled more cost-effectively by regular
watering, while permanent roads could be controlled using regular water-
ing or chemical dust suppressants.
     In addition to roadways, TSOFs generally contain important unpaved
travel areas.   Examples  include access  areas proximate to treatment and
disposal areas and haul  truck traffic patterns related to the stockpiling
of cover and other materials.  These areas may account for a substantial
fraction of the total traffic-generated emissions at a facility.  Fur-
thermore, travel areas tend to be more difficult to control than roadways
(e.g., changing traffic patterns make semipermanent controls impractical,
increased shear forces from cornering vehicles rapidly deteriorate chem-
ically controlled surfaces, etc.).

     2.2.1.2  Dust Emitting Activities.  Loose material on roadways and
travel areas may become entrained from activities other than vehicular
traffic (erosion during high wind events, for example).  The principal
long-term source of emissions for roads is,  of course, vehicle travel.
Only travel-related emissions are considered in the remainder of this
chapter.  Other travel areas can be considered open areas, with wind ero-
sion emissions estimated following the discussion in Section 3.0 of this
manual.

     2.2.1.3  Potential  Contamination Problems.  As mentioned above, an
earlier study of HW TSDFs suggested that unpaved roads may be the prin-
cipal source of metals and semivolatile organic emissions at these types
of facilities.  Contaminants in the loose surface aggregate of the road-
way may be due to spillage during transport of HW or track-on of material
which has become contaminated because of the various unit operations at
the facility.  Both of those sources may be at least partially controlled
by instituting better operating practices at the facility.  Again (as
discussed in this section's introduction),  it is important to note that
control measures designed to reduce emissions at a unit operation may
actually compound contaminated emissions on adjacent roadways.
                                   2-25

-------
2.2.2  Estimation of Uncontrolled Emissions

     2.2.2.1  Estimation of Uncontrolled Paniculate Emissions.
Estimation of the uncontrolled parti cu late emission rate R from  unpaved
TSDF roadways follows the general emission model  presented as Eq.  1-2:

                          R =  z  R.  =  i  e,A.                      (2-6)
                              i=l  n    1=1  1 ^
where:    R^ = emission rate associated with the  i-th road segment
          e.j = emission factor applicable for the i-th road segment
          A.J = source extent, In terms of vehicle distance traveled on
               road segment 1 over the averaging  time of interest
           n = number of road segments

Unlike paved roads, emission estimates for unpaved surfaces do not
require a "decision" process involving surface/vehicle parameters  because
the AP-42 emission factor equation takes source characteristics  into con-
sideration.^
          «   n ci  / s\  / S\ / W \°-7 /w\°-s  (365-p)
          e = °*61  \12/  (W fcr)    U)       365
                                                                    (2-7)
          Q   91  / s\  /s\ /W\°-7 /w\o.s  (365-p)
          6 = 2A  (W  (SB) (3)    V?)       365
where:    e = PM10 emission factor in units stated
          s = silt content of road surface material, percent
          S = mean vehicle speed, km/h (mi/h)
          W = mean vehicle weight, Mg (ton)
          w = mean number of wheels (dimensionless)
          p = number of days with > 0.254 mm (0.01 in) of precipitation

Site-specific input parameters are strongly recommended; if this is not
feasible, a summary of measured values is presented as Table 2-7.
                                   2-26

-------
           TABLE  2-7.   POSSIBLE  DEFAULT  VALUES  FOR  UNPAVED  ROAD
                           ESTIMATES AT TSDFsa

Parameter (units)
S1H content (%)
Average vehicle
—Speed (kph)
—Weight (Mg)
—No. of wheels
Probable range
2-20

8-45
8-40
4-18
Possible default
8b
u
20°
25b
10b

        aBased on observations during inventories of HW and MSW
         landfills.
         As noted in text, use of site-specific data is strongly
         recommended.
     The number of wet days per year, p, for the geographical area of
interest should be determined from local climatic data.  Figure 2-3 gives
the geographical distribution of the mean annual number of wet days per
year in the United States.  Maps giving similar data on a monthly basis
are available from the U.S. Department of Commerce.9
     It is important to note that for the purpose of estimating annual or
seasonal controlled emissions from unpaved roads, average control effi-
ciency values based on worst case (i.e., dry, p = 0 in Eq. 2-7) uncon-
trolled emission levels are required.  This is true simply because the
predictive emission factor equation for unpaved roads, which is routinely
used for inventorying purposes, is based on source tests conducted under
dry conditions.3  Extrapolation to annual average uncontrolled (including
natural mitigation) emissions estimates is accomplished by assuming that
emissions are occurring at the estimated rate on days without measurable
precipitation, and conversely are absent on days with measurable precipi-
tation.  This assumption has never been verified in a rigorous manner;
however, MRI's experience with hundreds of field tests indicate that it
is a reasonable assumption if the source operates on a fairly "continu-
ous" basis.
                                   2-27

-------
                                   •o
                                   o> «
                                      v 2
                                    • S. o;
                                    c   >
                                    O  •f~

                                   -i*

                                   Sf £
                                   T- = CO
                                    O. .- Ol
                                    0,=

                                    C^fe
                                   t- i Q.

                                   —4 ,  00


                                   °.°^
                                   o ^~ -o


                                   •>-> ZZ o

                                    SS0^
                                    §=
                                      ou
                                    = V)

                                    ^E
                                        O

                                        C
2-28

-------
     The uncontrolled emission factor for a specific unpaved road will
increase substantially after a precipitation event as the surface
dries.  However, in the absence of data sufficient to describe this
growth as a function of traffic parameters, amount of precipitation, time
of day, season, cloud cover, and other variables, uncontrolled emissions
are estimated using the simple assumption given above.  Prior MRI testing
has suggested that for unpaved travel areas, surface moisture levels
approximately twice that for dry conditions afford control of roughly 75%
to 9Q%.2  Between the dry, uncontrolled moisture level (typically < 2%)
and approximately 3% to 4%, a small increase in moisture content may
result in a large increase in control efficiency.  Beyond this point,
control efficiency grows slowly with increased moisture content.  These
relationships are discussed in greater detail in Section 2.2.3.

     2.2.2.2  Estimation of Contaminated Particulate Emissions.  The
emission rate FL- of particulate contaminated with compound j can be
represented as

                       R. =  r  R.. =  "  a.. e.A.                  (2-8)
                        J   i=l  1J   i=l  1J  1 1
where the terms are as defined above and:
     R.JJ = emission rate of contaminant j from road segment i
     a.,-,- = fraction of contaminant j in the particulate emissions from
           road i

The guidelines for estimating a^ • presented earlier in Section 2.1.2.2
are applicable here.  It is strongly recommended that  material  samples be
taken from the road surfaces in the TSDF, and that the silt (i.e.,
< 200 mesh) fraction be analyzed for the compounds of interest.

2.2.3  Demonstrated Control Techniques
     While there are numerous control options for unpaved roads, the
differences between preventive and mitigative measures are not as easily
defined for this source category as compared to paved roads.  Paved road
emissions are directly related to the amount of  loose material present
                                   2-29

-------
per unit area of the road surface, and preventive measures reduce that
amount.  Emissions from uncontrolled unpaved roads, on the other hand,
depend on the texture of  the road  surface rather  than  the  amount  of
material present.  Thus, the term "preventive," as applied to unpaved
roads, does not connote the same idea of necessarily reducing a variable
which directly affects emissions.
     Furthermore, in the context of contaminated particulate emissions
from unpaved surfaces, "preventive" is possibly best considered as repre-
senting those techniques which reduce carryout and spillage of contami-
nated materials onto the unpaved road surface, and any other measure that
reduces both gross and particulate emissions from the roadway could be
considered "mitigative."

     2.2.3.1  Preventive Measures.  While the effects of carryout and
spillage are much more pronounced on paved rather than unpaved surfaces,
preventing contamination of the travel surface is equally important in
either case.  As such, the discussion of preventive measures in Sec-
tion 2.1.3.1 is equally  applicable here, with sole exception that con-
tamination of the silt fraction of the road surface rather than  the  "gross"
surface loading should be the variable of interest in field deter-
minations.  This is discussed in greater detail  in Section  2.2.4.

     2.2.3.2  Mitigative Measures.  Three broad  categories  of unpaved
road control options are presented in Table 2-8; each is discussed in
greater detail  in  the following  sections.
     Source extent  reductions—These controls either  limit  the amount of
traffic on a road  to reduce the  PM10 emission rate or lower speeds to
reduce  the emission factor value  given by Eq. 2-7.  Examples might
include restriction of  some roads to only certain vehicle types, or
strict  enforcement  of speed limits.   In any instance, the control
afforded by  these  measures  is readily obtained by the application of
Eq.  2-7.
                                   2-30

-------
               TABLE 2-8.  CONTROL TECHNIQUES FOR UNPAVED
                             TRAVEL SURFACES
          Source extent reduction:      Speed reduction
                                       Traffic reduction
          Source improvement:           Paving
                                       Gravel surface
          Surface treatment:            Watering
                                       Chemical stabilization3
                                         —Asphalt emissions
                                         —Petroleum resins
                                         —Acrylic cements
                                         —Other
          aSee Table 2-9.

     Surface improvements—These controls alter the road surface.   Unlike
surface treatments (discussed below),  these improvements are largely
"permanent"; that is, frequent periodic retreatments are not normally
requ i red.
     The most obvious surface improvement is,  of course, paving an
unpaved road.  This option is expensive and is probably most applicable
to high volume (more than  a few hundred passes per day) roads at TSDFs
which are used every day (i.e., not those providing temporary access to
areas).  Clearly, control  efficiency estimates can be obtained by  apply-
ing the information of Section 2.1 of  this manual.
     Other improvement methods cover the road  surface material with
another material of lower  silt content (e.g.,  covering a dirt road with
gravel or slag, or using a "road carpet" under ballast).  Because  Eq. 2-7
shows a linear relationship between the emission factor and the silt con-
tent of the road surface,  any reduction in the silt value is accompanied
by an equivalent reduction in emissions.  This type of improvement is
initially much less expensive than paving; however, periodic maintenance
(such as grading and spot  reapplication of the cover material) may be
required.1*
                                   2-31

-------
                   TABLE 2-9.  CHEMICAL STABILIZERSa
A.  Type:  Bitumens

    Product

    AMS 2200, 2300®
    Coherex®
    Docal 1002®
    Peneprime®
    Petro Tac P®
    Resinex®
    Retain®
Manufacturer

Arco Mine Sciences
Witco Chemical
Douglas Oil Company
Utah Emulsions
Syntech Products Corporation
Neyra Industries, Inc.
Dubois Chemical Company
B.  Type:  Salts

    Product

    Calcium chloride
    Dowflake, Liquid Dow®
    DP-10®
    Oust Ban 8806®
    Dustgard®
    Sodium silicate
Manufacturer

Allied Chemical Corporation
Dow Chemical
Wen-Don Corporation
Nalco Chemical Company
G.S.L. Minerals and Chemicals Corporation
The PQ Corporation
C.  Type:  Adhesives

    Product

    Acrylic DLR-MS®
    Bio Cat 300-1®
    CPB-12®
    Curasol AK®
    DCL-40A, 1801, 1803®
    DC-859, 875®
    Dust Ban®
    Flambinder®
    Lignosite®
    Norlig A, 12®
    Orzan Series®
    Soil Gard®
Manufacturer

Rohm and Haas Company
Applied Natural Systems, Inc.
Wen-Don Corporation
American Hoechst Corporation
Calgon Corporation
Betz Laboratories, Inc.
Nalco Chemical Company
Flambeau Paper Company
Georgia Pacific Corporation
Reed Lignin,  Inc.
Crown Zellerbach Corporation
Walsh Chemical
 aSource:   Reference  10,  as  cited  by  Reference  11.
                                  2-32

-------
     Surface treatments—Surface treatment refers to those control
techniques which require periodic reapplications.  Treatments fall into
the two main categories of (1) wet suppression (i.e., watering, possibly
with surfactants or other additives), which keeps the surface wet to
control emissions, and (2) chemical stabilization, which attempts to
change the physical (and, hence, the emissions) characteristics of the
roadway.  Necessary reapplication frequencies may range from several min-
utes for plain water under hot, summertime conditions to several weeks
(or months) for chemicals.
     As is the case for water flushing paved roads, care must be exer-
cised in order to prevent contamination of the road surface by track-out
from operational areas.  Furthermore, watering an already contaminated
road and adding chemicals to the road surface may pose serious ground and
surface water problems.  Consequently, the use of surface treatments must
be assessed on a site-specific basis.
     Water is usually applied to unpaved travel areas using a truck with
a gravity or pressure feed.  This is only a temporary measure, and peri-
odic reapplications are necessary to achieve any substantial level of
control efficiency.  Some increase in overall control efficiency is
afforded by wetting agents which reduce surface tension.
     Chemical dust suppressants (Table 2-9), on the other hand, have much
less frequent reapplication requirements.  These suppressants are
designed to alter the roadway, such as cementing loose material into a
fairly impervious surface (thus simulating a paved surface) or forming a
surface which attracts and retains moisture (thus simulating wet suppres-
sion).
     Chemical dust suppressants are generally applied to the road surface
as a water solution of the agent.  The degree of control achieved is a
function of the application intensity (volume of solution per area), dilution
ratio,  and frequency (number of applications per unit time) of the
chemical applied to the surface and also depends on the type and number
of vehicles using the road.
                                   2-33

-------
2.2.4  Control Performance Estimation
     The discussion of paved road preventive measures in Section 2.1.3.1
applies equally to control techniques to prevent spreading of contamina-
tion onto unpaved surfaces.  The primary exception in the case of unpaved
roads deals with the use of field samples to assess the effectiveness.
While silt loading samples afford a relatively inexpensive way to gauge
the efficiency of preventive measures applied to paved roads, contami-
nant-specific analyses of road silt samples would be required to directly
assess the effectiveness of the same techniques applied to unpaved
roads.  The expense of this type of analysis may preclude implementation
of a routine sampling program.  One suggested alternative to estimate the
effectiveness of preventive techniques applied to unpaved roads would be
(i) require a paved road sampling program in the permit and (ii) assume
that the same level of efficiency applies to the potential contamination
of unpaved roads.  Note that this suggested approach assumes that (a) the
facility has both paved and unpaved roads subject to spillage and/or
carryout from operational areas, (b) analogous control programs are
implemented to prevent spillage/carryout onto both the paved and unpaved
roads, and (c) the degree of potential (i.e., uncontrolled) spillage and
carryout onto paved and unpaved surfaces is comparable.
     As with paved roads, it is recommended that all wind dispersal  flow
control plans contain provisions to effectively eliminate spillage of
waste onto unpaved roads.  This could be accomplished by (1) specifying
covered or enclosed trucks or minimal freeboard requirements, (2) accept-
ing no loads from haulers exhibiting visible spillage, (3) requiring all
accidental spills to  be reported immediately to the operations manager,
and  (4) rerouting traffic around any accidental spills which will be
cleaned as soon  as possible  (and in any case less than, say, 8 h).
      Instead of  requiring control effectiveness estimates for measures
preventing carryout onto unpaved roads at TSDFs, regulators may prefer to
require responsible parties  to  list in their Part B "Wind Dispersal Flow
Control Plan" all measures contemplated for use at their facility.  Regu-
lators may choose to  require  in the permit that the responsible parties
conduct systematic field  sampling programs for either contaminant-
specific  determinations of unpaved road silt samples  (or possibly of

                                   2-34

-------
paved road silt loadings as a surrogate, as described above).  Should the
sampling program ever show that agency-supplied values are exceeded, the
permit would contain provisions calling for more strenuous implementation
of preventive (and possibly mitigative) controls as well as a more inten-
sive field sampling program to demonstrate compliance with the goals.

     2.2.4.1  Source Extent Reductions.  These control methods act to
reduce the emission rate due to traffic on a road.   Control efficiency
values are easily obtained by use of Eqs. 2-6 and 2-7.
     The reduction may be obtained by banning certain vehicles or
strictly enforcing speed limits.  Some of these methods may require capi-
tal plus operating and maintenance (O&M) expenditures, while others (e.g,
speed reductions) may only require indirect costs associated with
increased travel times.  Consequently, identification of cost elements
and estimation of costs are highly dependent upon the option(s) selected
to reduce source extent, and no attempt is made here to generalize costs.

     2.2.4.2  Surface Improvements.
     Paying--Contro1 efficiency estimates for paving previously unpaved
roads may be based on the material presented as Section 2.1 of this
manual.  Inherent in this process is estimating the silt loading on the
paved surface; It is recommended that the reader use either values mea-
sured from presently paved roads in the facility or the values given in
Table 2-1.
     Cost elements identified for paving are as follows:
     Capital:  Operating equipment (graders, paving equipment), paving
material (asphalt, concrete), base material
     O&M:  Patching materials, labor for patching,  equipment maintenance
     Reference 5 provides the following cost estimates (April 1985
dollars) for asphaltic paving:
     Initial capital expense:  $44,700-$80,200/mi
     Annual O&M costs:  $6,600-$ll,900/mi
These estimates are based on resurfacing every 5 years and "15 percent
opportunity costs."  Reference 6 estimates a cost of $140,000/mi (1983
dollars) to pave unpaved roads in the iron and steel industry.  Because
                                   2-35

-------
of the variety of cost estimates, it is strongly recommended that the
reader obtain quotes from local paving contractors.
     Gravel/slag improvements:  As noted earlier, these types of improve-
ments replace the present road surface material with a lower silt content
material.  Note that this method may increase road maintenance costs as
the new aggregate fractures.  This cost may be avoided by installing a
"road carpet."  Because Eq. 2-7 indicates a linear relationship between
silt content and emission levels, control efficiency can be estimated by
determining the reduction in silt content.  For example, if a road with a
1251$ silt content is recovered with a gravel (with an equilibrium silt
content of 5%), then a 58% control efficiency would be expected.
     Identified cost elements for these improvements follow:
     Capital:  Material (including "road carpet," if applicable), appli-
cation equipment, labor
     O&M:  Periodic grading including equipment and labor
No cost estimates were found in the reference documents used as the basis
for this document.  Because of the differences in local availability of
cover materials (and civil engineering fabrics) and the amount of surface
preparation, compaction, and maintenance required for various road types,
it is recommended that the reader obtain quotes from local contractors.

     2.2.4.3  Surface Treatments.
     Watering—The control efficiency of unpaved road watering depends
upon (a) the amount of water applied per unit area of road surface,
(b) the time between reapplications, (c) traffic volume during that
period, and  (d) prevailing meteorological conditions during the period.
While several investigations have estimated or studied watering effi-
ciencies,  few have specified all the factors listed above.
     An empirical model for the performance of watering as a control
technique  has been developed.1*  The supporting data base consists of
14 tests performed  in  four  states during five different summer and fall
months.
                                   2-36

-------
The model Is:

                           C . 100 - °'8 P d t                      (2-9)

where:    C = average control efficiency, percent
          p = potential average hourly daytime evaporation rate, mm/h
          d = average hourly daytime traffic rate, (h~»)
          i = application intensity, L/m2
          t = time between applications, h

     Estimates of the potential average hourly daytime evaporation rate
may be obtained from:
!              0.0049  x  (value in  Figure  2-4)  for  annual  conditions
                                                                   (2-10)
              0.0065  x  (value in  Figure  2-4)  for  summer  conditions

Note that no data are available for Alaska and Hawaii in Figure 2-4.
Readers responsible for those portions of the country should consult
local meteorological  resources (e.g., state universities, local weather
stations, etc.)
     An alternative approach (which is potentially suitable for a regula-
tory format) is shown as Figure 2-5 which show that between the average
uncontrolled moisture content and a value of twice that, a small increase
in moisture content results in a large increase in control efficiency.
Beyond this point, control efficiency grows slowly with increased mois-
ture content.  The relatively simple bilinear relationship shown in the
figure is applicable to all particle size ranges considered:2
                             (75  (M-l)    1  <  H <  2
                        c  = 1                                      (2-11)
                             (62+6.7M    2  <  M <  £
where:    c = instantaneous control efficiency, percent
          M = ratio of controlled to uncontrolled surface moisture
              contents
                                   2-37

-------
                                       O)
                                         in


                                       O
                                       «O
                                       11
                                       1-
                                       3
                                       01
2-38

-------
       100%
   o
   c
   o

   I     75%

   uj

   o

   c
   o
   o
   °    50% -
   Q.

   (0
   3
   O
   O
   c

   2
   c
   ra

   CO
   c
25% -
         0%J
                                                             95%
                   Ratio of Controlled to  Uncontrolled
                       Surface  Moisture Contents
Figure 2-5.  Watering control  effectiveness for unpaved travel surfaces,
                               2-39

-------
Costs for watering programs include the following elements:
     Capital:  Purchase of truck or other device
     O&M:  Fuel, water, truck maintenance, operator labor

Reference 5 estimates the following costs (1985 dollars):
     Capital:  $17,100/truck
     O&M:  $32,900/truck
     The number of trucks required may be estimated by assuming that a
single truck, applying water at 1 L/mz, can treat roughly one mile of
road every hour.
     Enforcement of a watering program would ideally consist of two
complementary approaches.  The first requires the owner to maintain ade-
quate records that would document to agency personnel's satisfaction that
a regular program is in place.  The second approach would involve agency
spot checks of controlled roads by taking either traffic counts (for use
in conjunction with adequate records in Eq. 2-9) or material grab samples
for moisture analysis from the road.  For example, the moisture content
of the traveled portion of the roadway would be measured and compared
against a minimum acceptable value.  Estimates of the PMi0 control effi-
ciency could then be obtained from Eqs. 2-9 through 2-11, respectively.
     Records must be kept that document the frequency of water applied to
unpaved surfaces.  Pertinent parameters to be specified in a control plan
and to be regularly recorded include:
     General Information to be Specified in the Plan
     1.  All road segments and parking locations referenced on a map
available to both the responsible party and the regulatory agency.
     2.  Length of road and area of each parking lot.
     3.  Amount of water applied to each road/area and planned frequency
of application  (alternatively, a minimum moisture level could be speci-
fied).
     4.  Any provisions for weather (e.g., 1/4-in of rainfall will be
substituted for one treatment; program suspended during freezing periods;
watering frequency as a function of temperature, cloud cover, etc.).
     5.  Source of water and tank capacity.
                                   2-40

-------
     Specific Records for Each Road Segment/Parking Area Treatment
     1.   Date of treatment.
     2.   Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
     3.   Start and stop times on a particular segment/parking lot, aver-
age speed, number of passes.
     4.   Start and stop times for tank filling.
     General Records to be Kept
     1.   Equipment maintenance records.
     2.   Meteorological log (to the extent that weather influences the
control  program, see above).
     3.   Any equipment malfunctions or downtime.
     In  addition to those items related to control applications, some of
the regulatory formats suggested require that additional records be
kept.  These records may include surface material samples or traffic
counts.   Traffic counts may be recorded either manually or using auto-
matic devices.
     Chemical treatments—As noted earlier, some chemicals (most nota-bly
salts) simulate wet suppression by attracting and retaining moisture on
the road surface.  These methods are often supplemented by some water-
ing.  It is recommended that control efficiency estimates be obtained
using Figure 2-5 and enforcement be based on grab sample moisture
contents.
     The more common chemical dust suppressants form a hard cemented sur-
face.  It is this type of suppressant that is considered below.  Average
performance curves have been generated for four common chemical dust sup-
pressants:  (a) a commercially available petroleum resin, (b) a generic
petroleum resin, (c) an acrylic cement, and (d) an asphalt emulsion.  The
results  of this program were combined with other test results to develop
a model  to estimate time-averaged  PM10  control  performance.   This model
is illustrated as Figure 2-6.  Several items are to be noted:
                                   2-41

-------
      0        0.05      Oil       0.15       0.2       0.25       0.3
                    Ground Inventory  (gal/sq  yd)


Figure 2-6.   Average PM10 control efficiency for chemical  suppressants.
                                2-42

-------
     •     The term "ground  inventory"  is  a measure of residual  effects
          from previous  applications.   Ground inventory is found by add-
          ing together the  total  volume (per unit area) of concentrate
          (not solution)  since the start  of the dust control  season.   An
          example is provided below.
     •     Note that no credit for control is assigned until the ground
          inventory exceeds 0.2 L/m2  (0.05 gal/yd2).
     •     Because suppressants must be periodically reapplied to unpaved
          roads,  use of  the time-averaged values given in the figure are
          appropriate.  Recommended minimum reapplication frequencies (as
          well as alternatives) are discussed later in this section.
     •     Figure  2-6 represents an average of the four suppressants given
          above.   The basis of the methodology lies in a similar model
          for petroleum  resins applied to unpaved roads in the iron and
          steel industry.  However, agreement between the control effi-
          ciency  estimates  given by Figure 2-6 and available  field mea-
          surements is reasonably good.
     As an example of the use of the  figure, suppose that Eq. 2-7 has
been used to estimate a  PM10 emission factor of 2.0 kg/VKT.  Further,
suppose that starting on May 1, the road  is treated with 0.25 gal/yd2 of
a (1 part chemical to 5  parts water)  solution on the first of each month
until October.  In this  instance, the following average controlled emis-
sion factors are  found:



Period
May
June
July
August
September
Ground
inventory,
gal /yd 2
0.042
0.083
0.12
0.17
0.21
Average control
efficiency,
percent3
0
68
75
82
88
Average controlled
emission factor,
kg/VKT
2.0
0.64
0.50
0.36
0.24

    aFrom Figure 2-6; zero efficiency assigned if ground inventory
     is less than 0.05 gal/yd2.
                                   2-43

-------
A form which could be used as part of a record keeping format is pre-
sented in Section 2.2.5.
     In formulating dust  control  plans for chemical dust suppressants,
additional topics must be considered.  These are briefly discussed below.
     Use of paved road controls on chemically treated unpaved roads—
Repeated use of chemical  dust suppressants tend, over time, to form
fairly impervious surfaces on unpaved roads.  The resulting surface may
admit the use of paved road cleaning techniques (such as flushing, sweep-
ing, etc.) to reduce aggregate loading due to spillage and track-on.  A
field program showed that treated surfaces could be successfully flushed
and vacuumed.13  PM10 control efficiency values of 90% or more (based on
the uncontrolled emission factor of the unpaved roads) were found on sur-
faces that had been last  chemically treated 70 days earlier.
     The use of paved road techniques for "housekeeping" purposes would
appear to have the benefits of both high control (referenced to an uncon-
trolled unpaved road) and potentially relatively low cost (compared to
follow-up chemical applications).  Generally, it is recommended that
these methods not be employed until the ground inventory exceeds approxi-
mately 0.2 gal/yd2 (0.9 L/m2).  The use of paved road techniques on chem-
ically treated surfaces should, of course, be examined in limited areas
prior to implementing a full-scale program.
     Minimum reapplication frequency—Because unpaved roads at TSDFs are
often used for the movement of materials and are often surrounded by
additional unpaved travel areas, spillage and carryout onto the chem-
ically treated road require periodic "housekeeping" activities.  In addi-
tion, gradual abrasion of the treated surface by traffic will result in
loose material on the surface which should be controlled.
      It  is recommended that at least dilute reapplications be employed
every month to control loose surface material unless paved road control
techniques are used  (as described above).  More frequent reapplications
would be required if  spillage and track-on pose particular problems for a
road.
      Weather considerations—Roads generally have  higher moisture
contents  during  cooler periods due to decreased evaporation.  Small
 increases  in surface  moisture may result  in large  increases in control

                                   2-44

-------
efficiency (as referenced to the dry summertime conditions inherent in
the AP-42 unpaved road predictive equation).3  In addition, application
of chemical dust suppressants during the winter may be inadvisable for
traffic safety reasons.
     Weather-related application schedules should be considered prior to
implementing any control program.  Responsible parties and regulatory
agency personnel should work closely in making this joint determination.
     Compared to the other open dust sources discussed in this manual,
there is a wealth of cost information available for chemical dust sup-
pressants on unpaved roads.  Note that many salt products are delivered
and applied by the same truck.  For those products, costs are easily
obtained by contacting a local distributor.
     For other chemicals, identified cost elements include:
     Capital:  Distributor truck, tanks, pumps, piping
     O&M:  Chemical suppressants, water, fuel, replacement parts, labor
Many plants contract out application and thus have minimal capital
expenditures.
     Because each plant faces a unique set of needs, no attempt has been
made here to include all possible costs involved in a dust control pro-
gram.  For example, some facilities may be forced to install new storage
tanks while others may only need to refurbish unused tanks in the
plant.  Still others may find it more efficient to retain an outside
contractor to store and apply the suppressants.
     In order to provide preliminary estimates of costs associated with
chemical dust suppressants, the reader may employ the average costs shown
in Table 2-10.  Delivery and contracted application costs may be esti-
mated by increasing bulk costs by 10% and 15%, respectively.
     For treatments at the intensities generally recommended by suppres-
sant vendors, the corresponding unit cost is approximately $5,000 per
treatment per mile.12  At application intensities and dilution ratios
common in the iron and steel industry, applied unit costs for chemical
suppressants may be estimated as $3,000 per treatment per mile of unpaved
road.13  Note that in the iron and steel industry, lighter application
intensities have been found to be more cost-effective over typical time
intervals between treatments.
                                   2-45

-------
                  TABLE 2-10.  SUMMARY OF UNPAVED ROAD
                     CHEMICAL DUST SUPPRESSANT COSTS
                               Chemical  suppressant cost,
                                       1985 $/gal

Salts
Other
Small lot
0.70a
2.60b
Bulk
0.46a
1.48C
                aCost includes delivery and application.
                bFOB costs for 55-gal drums.
                CFOB; note that at the time this manual
                 was prepared, bulk costs of suppressants
                 are slightly lower than that stated.

     Enforcement of a chemical dust control program would ideally consist
of two approaches.  The first requires the owner to maintain adequate
records that would document that a regular program is in place.  The sec-
ond approach involves agency spot checks of controlled roads by taking a
material sample from the road.  That approach is discussed in detail in
Section 2.2.5.
     Records must be kept that document the frequency of chemicals
applied to unpaved surfaces.  Pertinent parameters to be specified in a
control plan and to be regularly recorded include the following.
     General Information to be Specified in the Plan
     1.  All road segments and parking locations referenced on a map
available to both the responsible party and the regulatory agency.
     2.  Length of each road and area of each parking lot.
     3.  Type of chemical applied to each road/area, dilution ratio,
application intensity, and planned frequency of application.
     4.  Provisions for weather.
     Specific Records for Each Road Segment/Parking Area Treatment
     1.  Date of treatment.
                                   2-46

-------
     2.  Operator's initials (note that the operator may keep a separate
log of whose information is transferred to the environmental staff's data
sheets).
     3.  Start and stop times on a particular segment/parking lot, aver-
age speed, number of passes, amount of solution applied.
     4.  Qualitative description of road surface condition.
     General Records to be Kept
     1.  Equipment maintenance records.
     2.  Meteorological log (to the extent that weather influences the
control program--see above).
     3.  Any equipment malfunctions or downtime.
     In addition to those items related to control  application, some of
the regulatory formats suggested in Section 2.2.5 require that additional
records be kept.  These records may include surface material samples or
traffic counts, and are discussed below.

2.2.5  Procedures for Compliance Determination
     There are numerous regulatory formats possible for unpaved roads.
For example, some states have developed opacity measurement techniques by
which compliance may be determined.  It is important to note that opacity
has yet to be related to emission levels from roads.  One often-raised
question deals with prevailing wind speeds during opacity readings; ambi-
ent air concentrations (and hence, opacity levels)  tend to be greater
under low wind speeds.  Consequently, for a road with even a constant
emission rate, opacity readings would vary indirectly with wind speed.
     Recordkeeping offers another compliance tool for unpaved road dust
controls.  The level of detail needed varies with the control option
employed. Table 2-11 summarizes the level of detail required for the
various controls discussed in Section 2.2.4.  Eqs.  2-9 and 2-10, together
with traffic, application, and meteorological records, would allow one to
estimate average control efficiency.  Moreover, use of Figure 2-6,
together with the form shown as Figure 2-7 allows estimation of chemical
suppressant efficiency between applications.  Figure 2-8 shows a com-
pleted form corresponding to the example in Section 2.2.4.
                                  2-47

-------












_J
o
ce.
y
O
tJ
o
*L
Jj^
Uw
a
*™r
^T,
a.
^"*
^—
r^
0
u_
oo
I—
z
LbJ
z:
~T^
Q^
_
o-
LkJ
Q»
CD
Z

^•»
LkJ
LkJ
~~
•3
S
LkJ
O£


*
»-H
*.
1
O4

LkJ
03
H-































t/1
^>
c
QJ
C
0
o


















id
4_>
OJ
TJ

^_
0

^~
OJ

(U
~_i




^""
o
s-
4"i
c
o

TJ
id
O
TJ
QJ
(0
a.
<*-
•^

2!
«f»
3
cr
QJ
TJ
V) QJ
E&
O f—
U Q.
OJ E
U QJ

i— in
c "o
O S-
*^" 4-*
4-> C
f- o
TJ 0
TJ





<4_
o

QJ
id
TJ
a>
c
•F*
•M •
I- 0
id -i->
+j a;
en
•» QJ
i— CL
QJ ^>
> ^j
QJ

en
r— C
id t-
E >
•i- id
c a.
f






en
c
*f—
>
fQ
a.



^J
•^
1/1
o

t/1
g^
g
QJ
t»
3
a; TJ
E C

O B
4J O
M- U
10 QJ
I-
"^
* "c
QJ
QJ 4->
S- C
o o
<4- 0
QJ
CO


en
C
o
*4— "f~
O 4^
10 CO
QJ O QJ
10 *| tO
TJ f- TJ
0
C 0. 0
•^ trt *^
-4-> 4-*
S. f— (T3
*d a) u
•4—* > **"•
W  en u
QJ C "s.
i— -r- en
•— c

id > TJ
E id id
••- i- i.
c en en
•f-



CT)
C
•^
r~*
O)
>
to
v_
CD


QJ
3
(/)
i
^
o
l*-
0)

a.
1
03
id
i-
o>
««- TJ
O OJ
TJ
C C
0 0)
•^ B
0 0
(U O
r— QJ
•— i-
O




„,
1 M- TJ
O 0 S-
^ S- 0
>> O *-> 0
Id QJ C OJ
TJ -M 3 i-
QJ O
J= E E 0
O id i-
10 • VI-
Qj c »U-
O 1/1 id
cn-f- c s-
C -4-> O 4-J
•t- id i-
S- U *J •
> r- TJ OJ
O Q. C ••-
O Q. O •—
10 u a.
• a.
QJ U_ r— (0
> O 10
i- U S-
1/1 QJ -r- O>
c E cn-u
QJ •!— o id
4-» -M •— S
X
LU



cn
C
•^
^_
01

fo
3


01
2
CO
i
t.
o
M-
co
O)

"a.
§
en
•^
id
o>
«*- TJ
o at

c c
o QJ
-M 5
u o
QJ (J
i— QJ

o



id
(J
•r- cn
r^ cn
Q. C

id s-
01
Lj ^ l
O id
3
QJ ^J
-M C
id QJ
TJ -3
cr
. a)
QJ I/I
> -O
•»- 3
eo (/)
g~
O) TJ
4-> C
X 10
QJ

>i C.
<— o
J- 1-

10
u_
(d
i/i
cn
r^
id
o

E
01
.C
(_)





1
1


o
TJ
0)
TJ
QJ
0 C
U QJ
$•» O
I/) H-
O) 14-
F- QJ

id o
en U

TJ C
<— o
QJ U

U.



c
o
•f—
( *
cn <0
c u
•(-> r—
id a.
s- a.
QJ id
Q.
o ^
o
TJ Id
C 01
10
^
en o
QJ D-
id uo
TJ i-
QJ

QJ QJ
•U E
id id
i- S-
QJ id
TJ Q.
O



cn
r**
id
u

E
QJ
.C
CJ
2-48

-------
 «
o»o ^
   0
 SI
 w
c o -o
Iff
0
    211
    £ = 3
5

Q.

O

0)
Q

®

a
   9
   75
   O
  £ 6
                                                        c
                                                        O
                                                        to
                                                        O
                                                            Q.
                                                            CL
                                                            re
                                                        C
                                                        to
                                                        in
                                                        in
                                                        
-------
UJ
HI
co *i

II
8  «
   N,
a.
<
CO
UJ
cc
Q.
Q.
<
o
 O
   \
: c
 8
•1.2-1
3 flj vJ
f= 01 ®
O   03
         =
         £=§>
           a
                      1   1    1
                      til
                      1   1
          1    I   I   i    I

          I    I   I   I    I
                    I    I
                 NJ
I   I    I
                    I    I
                                                                            en
                                                                            O
                                                   -a
                                                    
-------
     While recordkeeping affords a convenient method of assessing long-
term control performance, it is important that regulatory personnel have
"spot-check" compliance tools at their disposal.  One such tool was men-
tioned earlier in connection with Figure 2-5.  Rules could be written
specifying a minimum surface moisture content (thus, corresponding to a
minimum control efficiency) to be maintained on an unpaved surface which
is watered or treated with salts.  Inspection personnel would then col-
lect grab samples for moisture analysis following the procedures in
Appendix D to determine compliance.
     For chemically (other than salts) controlled surfaces, it has been
found that industrial paved road model (see Table 2-2) tends to over-
estimate the controlled  emission factor (and thus,  underestimate instan-
taneous control efficiency).13  In this way, an inspector could collect
an unpaved sample with a whisk broom and dust pan, and after laboratory
analysis for silt content, have a conservatively low estimate of control
efficiency due to the chemical treatment.   If a rule is written to
maintain a certain level of efficiency, the inspector could then instruct
the responsible party to reapply the chemical or use paved road controls
(if feasible).

2.3  EXAMPLE CALCULATION
     An example application of the emission estimation methods is
presented in this section.  The example makes use of the hypothetical
TSDF shown in Figure 2-9.  Summary information on paved roads A and 8,
and the unpaved road C shown in the figure  is given in Table 2-12.
     Also, 20 vehicles exit the stabilization unit each day onto paved
road B; 15 vehicles per day exit the unit onto the unpaved road C.  Road
C is 0.3 km long, unpaved access route to the landfill unit with the
following characteristics:
          Silt content of 10%
     •    Average vehicle speed of 30 kph
     •    Average vehicle wheels is 12
     The facility operates 5 d/wk (8 h/d) and 261 d/yr in a location with
100 d/yr of > 0.01 in (0.254 mm) of precipitation (on average) and 60 in
(1,500 mm) of annual Class A pan evaporation.

                                   2-51

-------
                  LANDFILL UNIT
                        I
                        Ic
                        I
                            STABILIZATION
                                UNIT
   PAVED
PARKING LOT
	X—
A    IB
 -X-
-X
                               FENCE LINE
           Figure 2-9.  Roads at a hypothetical TSDF.
                           2-52

-------
         TABLE  2-12.  SUMMARY  INFORMATION FOR ROADS A, B, AND C


Road
A
B
C

Length
(km)
0.2
0.3
0.3

Vehicle passes
per work daya
80
100
90
Average
vehicle weight
(Mg)
2
20
30
Silt
loading
(9/m2)
1.0
5.0
N/Ab

    facility operates 261 days per year.
     Road is unpaved; see additional information in Section 2.3.2.

2.3.1  Paved Road Estimates
     The "gross" PM10 emission factors are estimated using Table 2-2
based on the mean vehicle weights and silt loading given above.   Because
road A's mean vehicle weight is less than  4 Mg with a silt loading  less
than 2 g/m2, a PMi0 emission factor of

                              2.28(1/0.5)0-8

or, 4.0 g/VKT is found.  On the other hand, because road B has a mean
vehicle weight greater than 6 Mg and a silt loading greater than 2  g/m2,
a PM10 emission factor of

                              220(5/12.0)0-3

or, 170 g/VKT is found.  Corresponding emission rates are:
     Road A:  4.0 g/VKT x 80 veh/d x 0.2 km x 261 d/yr,  or
              17 kg/yr
     Road B:  170 g/VKT x 100 veh/d x 0.3  km x 261 d/yr, or
              1,300 kg/yr
                                  2-53

-------
     In order to estimate contaminated emissions from paved road travel
using the suggestions 1n Section 2.1.2.2, 1t 1s necessary to first esti-
mate contamination levels in the roadway surface loadings on the roads.
In this example, 1t is assumed that no site-specific sampling and analy-
sis (S&A) program has been undertaken.  In addition, 1t is assumed that
because road A is not directly used to access operational areas, that
contamination levels for that road's surface loading are negligible.
Finally, because road B is used to access both the landfill and
stabilization areas, it 1s conservatively assumed that the long-term
loading on the road 1s composed of material carried out from those units
and, as such, has contamination levels identical to travel areas in those
units.  Here, it 1s assumed that the Pb concentration of 114 ppm
(cf. Table 7-2) is applicable.
     With the above assumptions, the estimated Pb emission rate is
estimated as
                                 x 1,300 kg/yr

or, 0.15 kg/yr.  Note that because it is assumed that the long-term
loading is due to carryout, additional calculations using Eq. 2-5 are not
considered necessary.

Specification of Dust Control Program
     Suppose that it is determined that 20% average control of paved road
Pb emissions from road 8 is necessary to achieve a goal (perhaps based on
the risk assessment methodology presented in Appendix C).  Based on the
equation

                                      SL °-3
                             e =  220  ^

presented  in Table 2-2, 20% control implies that

                        (l-0.2)(sL0)0'3 =
                                   2-54

-------
where:    sL0 » uncontrolled silt loading
          sLi * controlled silt loading

Solving the above equation yields

                           sLi
                                  °'48

Thus, a PM10 control efficiency of 20% Implies that silt loadings must be
reduced 52%.  Because it 1s not believed that preventive measures can
achieve that silt loading reduction, a m1tigat1ve program is
envisioned.
     The water flushing expressions presented in Table 2-3 indicate that
20% efficiency occurs long after 299 or 365 vehicle passes for water
flushing or flushing and broom sweeping, respectively.  The models thus
suggest that 20% average efficiency is achieved when
                              20 > 10, 300 /V
or, V < 515 vehicle passes for water flushing and
                              20 > 17,500/V
                       •         ~~
or, V < 875 vehicle passes for flushing followed by broom sweeping.
Because road B experiences 100 passes per day, the controls would require
application roughly every 5 or 9 days, respectively.

2.3.2  Unpaved Road Estimates
     Using Eq. 2-7 and the values given above, a PM10 emission factor of
or 2.2 kg/VKT is found.  At 90 VP per day and 261 working days per year,
an annual emission rate of

                 2.2 kg/VKT (90 veh/d)(261 d/yr)(0.3 km)

or 15,000 kg/yr is obtained.
                                   2-55

-------
     Because no on-site sampling and analysis (S&A) program is conducted,
assumptions must be made on contamination levels on the road.  Because
the road is used by landfill-related traffic it is reasonable to expect
that a conservatively high estimate may be obtained by setting the
a-levels in Eq. 2-8 equal to the mean value measured for landfill sur-
faces (cf. Table 5-3).  Thus, an estimate for the annual Pb emissions is

                              I*1|P- x 15,000 kg/yr

or 22 kg/yr.

Additional Emissions Due to Carryout From Operational Areas
     Because material carryout from the stabilization and landfill areas
is onto an unpaved road, it is reasonable to expect that "gross"
emissions directly related to the carryout is relatively unimportant.
The more important concern, of course, is the increase in contaminated
emissions.  Assuming that the carryout emissions from the unpaved road is
analogous to that on paved roads, Eq. 2-5 suggested that

                            E+ = 36 g/vehicle

for the stabilization unit (N=15 veh/day) and

                            E+ = 91 g/vehicle

for the landfill area (N=45 veh/day).  Thus, for M=90 veh/day on road C,
an annual  increase of

                   (36+91) g/veh x 90 veh/day x 261 d/yr

or 3,000  kg/yr  in  "gross" emissions is estimated.  The increase in
contaminated emissions may be estimated using the  information presented
as footnote **  in  Section 2.1.4.1.  If it is assumed that the Pb
contamination  levels  in  the  landfill (Table 5-3) and stabilization units
                                   2-56

-------
(Table 7-2) are 1,490 and 114 ppm, respectively, then Pb emissions due to
carryout may be estimated as
                36 -i- ^p^ 91) g/veh x 90 veh/day x 261 day/yr

or 3.3 kg/yr.

     In summary, a total of 25 kg/yr of Pb is estimated to be emitted
from traffic on road C.  Note that in light of assumptions made, this
estimate is probably conservatively high.

Specification of Dust Control Program
     Assume that 80% average control of emissions is required for road C,
and that watering will be used to achieve this control.  Based on
Eq. 2-9, 80% control requires that

                        80 > 100 -  0-8(p).(90/8)t

and, using Eq. 2-10, it is seen that
                     1     9       0.13 L/m2/h (annual)
                     ¥ ~ JQ p =
                     1  ^       0.18 L/mVh (summer)

Thus, any watering program which applies at least 0.13 (or 0.18) L/m2 of
water for every hour between application will achieve the desired goal.
Suitable plans include:
          0.13(0.18) L/m2 every hour
          0.25(0.35) L/m2 every 2 h
          0.066(0.88) L/m2 every 1/2 h

2.4  REFERENCES FOR SECTION 2
 1.  Englehart, P., and D. Wallace.  Assessment of Hazardous Waste TSDF
     Particulate Emissions.  Final Report, EPA Contract No. 68-02-3891,
     Work Assignment Nos.  5 and 13.  October 1986.
 2.  Cowherd, C., G. E. Muleski, and J. S. Kinsey.  Control of Open
     Fugitive Dust Sources.  EPA-450/3-88-008, U.S. Environmental
                                   2-57

-------
     Protection  Agency,  Research  Triangle  Park,  North  Carolina.
     September 1988.

 3.   Environmental  Protection  Agency.   Compilation  of  Air  Pollution
     Emission Factors  (AP-42).  Research Triangle Park,  North  Carolina.
     September 1988.

 4.   Cowherd, C.,  Jr.,  and  J.  S.  Kinsey.   Identification,  Assessment  and
     Control of  Fugitive Particulate Emissions.  EPA-600/8-86-023, U.S.
     Environmental  Protection  Agency,  Research Triangle  Park,  North
     Carolina.   1986.

 5.   Kinsey, J.  S., et  al.   Control  Technology for  Sources of  PM10.
     Draft Final Report, EPA Contract  No.  68-02-3891,  Work Assignment
     No.  4.  September  1985.

 6.   Cuscino, T.,  Jr.,  et al.   Iron  and Steel Plan  Open  Source Fugitive
     Emission Control Evaluation, EPA-600/2-83-110, U.S. Environmental
     Protection  Agency,  Research  Triangle  Park,  North  Carolina.
     October 1983.

 7.   Muleski, G.,  and 0. Hecht.   PM10  Emission  Inventory of Landfills in
     the  Lake Calumet Area. Final Report, EPA Contract  No. 68-02-3891,
     Work Assignment No. 30.  September 1987.

 8.   Muleski, G. E. Update of Fugitive Dust Emission  Factors  in AP-42
     Section 11.2.   Final Report, EPA  Contract No.  68-02-3891, Work
     Assignment  No. 19.   July  1987.

 9.   Climatic Atlas of  the  United States.   U.S.  Department of  Commerce,
     Washington, D.C.   June 1968.

10.   Rosbury, K. D. Fugitive  Dust Control Techniques  at Hazardous Waste
     Sites.  Interim Technical Report  No.  I—Proposed  Field Sampling
     Plan, Contract No.  68-02-3512,  Work Assignment No.  61, U.S.
     Environmental  Protection  Agency,  Municipal  Environmental  Research
     Laboratory, Cincinnati, Ohio.  March  1984.

11.   Turner, J.  H., et  al.   Fugitive Particulate Emissions From Hazardous
     Waste Sites.   Prepared for the U.S.  Environmental Protection Agency,
     Cincinnati, Ohio.   September 1984.

12.   Muleski, G. E., T. Cuscino,  Jr.,  and  C. Cowherd,  Jr.   Extended
     Evaluation  of Unpaved  Road Dust Suppressants  in the Iron  and Steel
     Industry.   EPA-600/2-84-027, U.S. Environmental Protection Agency,
     Research  Triangle Park, North Carolina.  February 1984.

13.   Muleski,  G. E., and C. Cowherd, Jr.   Evaluation of  the Effectiveness
     of Chemical Dust  Suppressants on Unpaved Roads.  EPA-600/2-87-102,
     U.S. Environmental Protection Agency, Research Triangle Park, North
     Carolina.   November 1987.
                                   2-58

-------
                 3.0  OPEN WASTE PILES AND STAGING AREAS

     Open waste piles constitute a direct source of atmospheric particu-
late emissions resulting from the action of mechanical disturbance or the
forces of wind on the exposed surface.  In addition, contamination from
the waste pile may spread to surrounding surface materials through
successive deposition/ entrainment processes, mechanical track-out by
vehicle movement, spillage during pile formation/removal, or migration of
leachate from the pile.

3.1  SOURCE DESCRIPTION
     Short-term waste piles may be found in several types of TSOF
operations.  Such piles may be disturbed frequently as part of a par-
ticular disposal cycle.  At the other extreme, some waste piles may
remain substantially intact over periods of months to years.  Not
unexpectedly, pile size (i.e., volume, height/depth, and exposed surface
area) tends to increase with the average waste retention time.1  For
example, based on model waste pile characteristics, a pile with a
retention time of several months may have a height up to 4 m and a
surface area exceeding 10,000 mz.  On the other hand, piles that are
disturbed daily have heights of less than 1 m and surface areas of less
than 100 m2.
     Dust emissions occur at several points in the waste retention cycle:
during loading of material onto the pile, during disturbances by strong
wind currents, and during removal of material from the pile.  The move-
ment of trucks and loading equipment in the waste pile area is also a
substantial source of dust.  The potential for direct emissions from the
pile surface increases substantially whenever portions of the pile are
disturbed.  This tends to occur when material is either added to or
removed from the pile or when the pile is otherwise reshaped.
                                3-1

-------
     The quantity of dust emissions from waste pile maintenance opera-
tions varies with the volume of aggregate passing through the waste
retention cycle. Also, emissions depend on three correction parameters
that characterize the condition of a particular waste pile:  age of the
pile, moisture content of the waste, and proportion of aggregate fines.
It should be noted that organic chemical constituents of the waste may
provide an inherent mitigative effect by binding otherwise suspendable
particles to the bulk material.
     When finely divided waste (without constituents that provide
inherent mitigation) is loaded onto a storage pile, its potential for
dust emissions is at a maximum.  Fines are easily disaggregated and
released to the atmosphere upon exposure to air currents from transfer
operations or high winds.  As the waste material weathers, however, its
potential for dust emissions may be greatly reduced.  Moisture from
precipitation and condensation causes aggregation and cementation of
fines to the surfaces of larger particles.
     Field investigations have shown that emissions from certain
aggregate storage operations vary in direct proportion to the percentage
of silt (particles < 75 ym in diameter) in the aggregate material.2"9
The silt content is determined by measuring the proportion of dry aggre-
gate material that passes through a 200-mesh screen, using the ASTM-C-136
method.  Table 3-1 summarizes measured silt and moisture values for
industrial aggregate materials.

3.2  ESTIMATION OF UNCONTROLLED EMISSIONS
     Total dust emissions from waste piles are the sum of several dis-
tinct source activities within the storage cycle (see Figure 3-1):
     1.   Loading of waste onto the pile  (batch or continuous drop
          operations).
     2.   Equipment traffic  in waste pile area.
     3.   Wind  erosion of pile surfaces and ground areas around the
          pile.
     4.   Loadout of waste for final disposal or for return to the
          process stream for reclamation  (batch or continuous drop
          operations).

                                3-2

-------












(/)
UJ
^_
(/)

!
i

<
en
<
e
LU
^
^C
^
*
LU
3
_J
<
*
t-
LU
h-
!
IV
3
V)
"•
i

i

h™
^j
—

_i
o

D.
*~
•
|
fO

LU
^

<
H-






















•*-
C
0
u
L.
£
«
«
4-
i








c
41
U

4)
Q.

+-
V>


















































C
ID







O
o>
c
ID
a


4. a
0+--
1/1 a.
O 4- O
Z fc
U
4-
3
•a
c
-—
r** ^ v CO CM ^ v *o ^ < r*»
* • • • 0\ z * • Z* Z *
^— CM m ^f * 10 ^o ^3
— o


in *• ^ CM ^ ^ ^ ^
• • ^^ • z 2 Z *

1 1 1 00 1 1
i v to *in ft
vO • CM CM •
.— o
0 0



^r ^0 *O ^O f^ ^3 ^* ^* ^d ^3 ^^


C^ O^ U^ lA fO ^3 ^f C5 f^ *T sO
O T o> inflommoo —


ro CT> r- to K\ 01
— — • • CM •
1 V CO 1 1 «T T
• • CM fl •— (O
—  C
4- t_ 3 U T3 O 'a
O— 4-O T3 r> 0) L. 4- 4)
§-^ 4) T3 4) I/? /—
O ~~&-~O)4)4)C4~4) 
~—
01
jU (^
C O
O O
•- 0
4- L.
U CL

0 c
L, <0
a.
ID CT
4- — C
CO) —
4) 4) >-
E 4- L.
4- in L.
10 ID
4) "D 3
C. C 0-
4- nj
41
JO O
l_ 4-
— (/>
to in
C^ f^
• •
OO


to
•

1
s

o*



f** f~r


•»0
Kl —


T <
«'Z
1
CM
*
CM




O> CM









in
in oi
4- C

_ ^

4) ID
0.1-




•o

.^
I/I
VI
4>

Q
L.
a.

•o
c
ID

Ol
C

C
"e

4)
4-
••.
C
U
10
h-
Ot ^ T
* Z *
to '^



O ^ T
^^ z *

* 00
o




r>. O K>


CM ino
10 r~ in


10 in —
T T i
«T OO —

10 ro in





in in 10






•o
c
o
c u
4) O)
TJ

3 4)
.0 in
— L. O
ID 41 a.
O O LU






4)
0)
C
.w
c

*i

_
s
u

41
u
ID
L.

in

c
i_
41
m
4)
2











































,
41

JD
ID
U
Q.
r*
UJ
i

"
^
z



p*.

1—
D)
O
L.

4-

• ^ • • •
CM fO CO Ot
in
4> 41 0 41 4)
O U O 0 0
c c c c c
4) 4) 4) 4) flj
L. L. l_ L. l_
4) 4) 4) 4) <])
41 4) 4) 41 41
Of OC Q? Q£ Q^
ID JD O T3 H)
3-3

-------
CO
LU
DC

CO

Q
z
<

CO
LU
-J

Q.

LU

CO


I

Z
LU
a.
O
                                               (O
                                               a;
                                               s_
                                               



                                               O)
                                               I/)
                                               OJ
                                               a;
                                               +j
                                               CO
                                               (O
                                               CD
                                               a.
                                               o
                                                i
                                               ro
                                               OJ
                                               01
                  3-4

-------
Items 1 and 4 (pile formation and removal) fall under the general cate-
gory of materials handling as discussed below.  Equipment traffic 1n the
waste pile area occurs primarily in association with pile formation and
removal.

3.2.1  Materials Handling
     Adding waste material to a storage pile or removing it usually
involves dropping the material onto a receiving surface.  Truck dumping
on the pile or loading out from the pile to a truck with a front-end
loader are examples of batch drop operations.  Adding material to the
pile by a conveyor stacker is an example of a continuous drop operation.
     The following equation is recommended for estimating emissions from
transfer operations (batch or continuous drop):

                                  / u \1'3
                    e = k(0.0016)^/L4  (kg/Mg)

                                                                    (3-1)
                                     1.3
                    e = k(0.0016) w/ 1>4  (Ib/ton)
where:  e = emission factor
        k = particle size multiplier (dimensionless)
        U = mean wind speed, m/s (mph)
        M = material moisture content, %

The particle size multiplier k varies with aerodynamic particle diameter
as shown below:

                 Aerodynamic Particle Size Multiplier, k
     <30 urn        <15 um        <10 urn        <5 urn          <2.5 urn
       0.74         0.48          0.35          0.20            0.11
                                3-5

-------
     For emissions from equipment traffic (trucks, front-end loaders,
dozers, etc.) traveling between or on piles, it is recommended that the
equations for vehicle traffic on unpaved surfaces be used (see Sec-
tion 2.0).  For vehicle travel between storage piles, the silt value(s)
for the areas among the piles (which may differ from the silt values for
the stored materials) should be used.

3.2.2  Wind Erosion
     Dust emissions may be generated by wind erosion of open waste piles
and exposed areas within a disposal facility.  These sources typically
are characterized by nonhomogeneous surfaces impregnated with nonerodible
elements (particles larger than approximately 1 cm in diameter).  Field
testing of coal piles and other exposed materials using a portable wind
tunnel has shown that (a) threshold wind speeds exceed 5 m/s (11 mph) at
15 cm above the surface or 10 m/s (22 mph) at 7 m above the surface, and
(b) particulate emission rates tend to decay rapidly (half life of a few
minutes) during an erosion event.  In other words, these aggregate
material surfaces are characterized by finite availability of erodible
material (mass/area) referred to as the erosion potential.  Any natural
crusting of the surface binds the erodible material,, thereby reducing the
erosion potential.

     3.2.2.1  Emissions and Correction Parameters.  If typical values for
threshold wind speed at 15 cm are corrected to typical wind sensor height
(7-10 m), the resulting values exceed the upper extremes of hourly mean
wind speeds observed in most areas of the country.  In other words, mean
atmospheric wind speeds are not sufficient to sustain wind erosion from
flat surfaces of the type tested.  However, wind gusts may quickly
deplete a substantial portion of the erosion potential.  Because erosion
potential has been found to increase rapidly with increasing wind speed,
estimated emissions should be related to the gusts of highest magnitude.
     The routinely measured meteorological variable which best reflects
the magnitude of wind gusts is the fastest mile.  This quantity repre-
sents  the wind speed corresponding to the whole mile of wind movement
which  has passed by the 1-mi contact anemometer in the least amount of
                                3-6

-------
time.  Daily measurements of the fastest mile are presented in the
monthly Local Climatological Data (LCD) summaries.10  The duration of the
fastest mile, typically about 2 min (for a fastest mile of 30 mph),
matches well with the half life of the erosion process, which ranges
between 1 and 4 min.  It should be noted, however, that peak winds can
significantly exceed the daily fastest mile.
     The wind speed profile in the surface boundary layer is found to
follow a logarithmic distribution:

                        uW-folnf-  (z>z0)                  (3-2)
                                       0
where:    u = wind speed, cm/s
         u* = friction velocity, cm/s
          z - height above test surface, cm
         ZQ = roughness height, cm
        0.4 = von (Carman's constant, dimensionless

     The friction velocity (u*) is a measure of wind shear stress on the
erodible surface, as determined from the slope of the logarithmic
velocity profile.  The roughness height (z0) is a measure of the rough-
ness of the exposed surface as determined from the y-intercept of the
velocity profile, i.e., the height at which the wind speed is zero.
These parameters are illustrated in Figure 3-2 for a roughness height of
0.1 cm.  The roughness height (ZQ) is needed to convert the friction
velocity to the equivalent wind speed at the typical weather station
sensor height of 7 to 10 m above the surface.
     Emissions generated by wind erosion are also dependent on the fre-
quency of disturbance of the erodible surface because each time that a
surface is disturbed, its erosion potential is restored.  A disturbance
is defined as an action which results in the exposure of fresh surface
material.  On a storage pile, this would occur whenever aggregate mate-
rial is either added to or removed from the old surface.  A disturbance
of an exposed area may also result from the turning of surface material
to a depth exceeding the size of the largest pieces of material present.
                                3-7

-------
 °
i

I
5
                                                                                          O
                                                                                          u
                                                                                          o

                                                                                          

                                                                                          u
                                                                                         1-
                                                                                         
-------
     3.2.2.2  Predictive Emission Factor Equation*!.  The emission factor
for wind-generated participate emissions from mixtures of erodible and
nonerodible surface material subject to disturbance may be expressed in
units of g/m2-yr as follows:
                                             N
                       Emission factor = k   ^   P.                  (3~3)
                                           i  =  1   n
where:     k » particle size multiplier
           N * number of disturbances per year
          P^ = erosion potential corresponding to the observed (or
               probable) fastest mile of wind for the i-th period between
               disturbances, g/m2

     The particle size multiplier (k) for Equation 3-3 varies with
aerodynamic particle size, as follows:

          Aerodynamic Particle Size  Multipliers  for  Equation  3-3
                   <30 ym    <15 ym    <10 ym    <2.5 ym
                     1.0       0.6       0.5       0.2

     This distribution of particle size within the < 30 ym fraction is
comparable to the distributions reported for other fugitive dust sources
where wind speed is a factor.  This is illustrated, for example, in the
distributions for batch and continuous drop operations encompassing a
number of test aggregate materials (see AP-42 Section 11.2.3).
     In calculating emission factors, each area of an erodible surface
that is subject to a different frequency of disturbance should be treated
separately. For a surface disturbed daily, N = 365/yr, and for a surface
disturbance once every 6 mo, N = 2/yr.
                                3-9

-------
     The erosion potential function for a dry, exposed surface has the
following form:

                      P = 58 (u* - u*)2 + 25 (u* - u*)
                      P = 0 for u* < u*                             (3-4)

where:  u* = friction velocity (m/s)
        u? = threshold friction velocity (m/s)
     Table 3-2 presents the erosion potential function in matrix form.
Because of the nonlinear form of the erosion potential function, each
erosion event must be treated separately.
     Equations 3-3 and 3-4 apply only to dry, exposed materials with
limited erosion potential.  The resulting calculation is valid only for a
time period as long or longer than the period between disturbances.
     For uncrusted surfaces, the threshold friction velocity is best
estimated from the dry aggregate structure of the soil.  A simple hand
sieving test of surface soil (adapted from a laboratory procedure pub-
lished by W. S. Chepil12) can be used to determine the mode of the
surface aggregate size distribution by inspection of relative sieve catch
amounts.  This procedure  is specified in Section 4.2.1.  The threshold
friction velocity for erosion can be determined from the mode of the
aggregate size distribution, following a relationship derived by
Gillette,13 as shown in Figure 3-3.
     Threshold friction velocities for several surface types have been
determined by field measurements with a portable wind tunnel.13-16  These
values are presented in Tables 3-3 and 3-4 for industrial aggregates and
Arizona sites.  Figure 3-4 depicts these data graphically.
     The fastest mile of  wind for the periods between disturbances may be
obtained from the monthly LCD summaries for the nearest reporting weather
station that  is representative of the site in question.1"  These sum-
maries report actual fastest mile values for each day of a given month.
                                3-10

-------






FUNCTION
»oivor^
                                   •-t CO UD
OOOOOOOOOOr^OtvOPxCO
                                i~i OO If) 00
                             -H CO tfl 00 «-•
                           »-l CO If) 00 ^^
                        ^H CO IT) 00 «-i «• 00
                     t-HCOlfJCO'—ITOOCO
                                f—I t-H •—I CM
                     CO UO CO •••* ^ CO OO CO
                             T-I i—I i—I CM CM
                  COlf)OO
                           «—I «—t «—I CM CM CO
                        i—I r-l r-( CM CM CO CO
                     »—1«—ti—ICMCMCOm^"
           itOOO«—l^OOCOCOCOCTvtOCM
                  «-Ht—li-HCMCMCOCO^-lO
OOOO>-H<— (»— !•—(•— ICMCMCMCMCMCO
                          3-11

-------
    TABLE 3-3.  THRESHOLD FRICTION VELOCITIES—INDUSTRIAL AGGREGATES
Material
Overburden*
Scoria (roadbed
Threshold
friction
velocity,
m/s
1.02
1.33
Roughness
height,
cm
0.3
0.3
Threshold wind
velocity at
10 m (m/s)
actual
21
27
zo *
0.5 cm
19
25
Ref.
9
9
  material)3
Ground coal
  (surrounding coal
  pile)
Uncrusted coal pile
Scraper tracks on
  coal pilea'D
Fine coal dust on
  concrete padc
                          0.55
                          1.12
                          0.62

                          0.54
0.01
0.3
0.06

0.2
16
23
15

11
10
21
12

10
 9
 9

15
dWestern surface coal mine.
bLightly crusted.
cEastern power plant.
        TABLE  3-4.   THRESHOLD  FRICTION  VELOCITIES—ARIZONA  SITES



Location
Mesa - Agricultural site
Glendale - Construction site
Maricopa - Agricultural site
Yuma - Disturbed desert
Yuma - Agricultural site
Algodones - Dune flats
Yuma - Scrub desert
Santa Cruz River, Tucson
Tucson - Construction site
A jo - Mine tailings
Hayden - Mine tailings
Salt River, Mesa
Casa Grande - Abandoned
agricultural land
Threshold
friction
velocity
(m/s)
0.57
0.53
0.58
0.32
0.58
0.62
0.39
0.18
0.25
0.23
0.17
0.22
0.25


Roughness
height
(cm)
0.0331
0.0301
0.1255
0.0731
0.0224
0.0166
0.0163
0.0204
0.0181
0.0176
0.0141
0.0100
0.0067

Threshold
wind velocity
at 10 m
(m/s)
16
15
14
8
17
18
11
5
7
7
5
7
8

                               3-12

-------
                                                                  tu
                                                                  •o
                                                           E,

                                                           0)
                                                          •o
                                                      o   o
to

•o

0)
M

(/I

o
+j
                                                                  at
                                                          m

                                                          N
                                                          (Q


                                                          0>



                                                          O)
                                                                  o
                                                                  
-------
          •o
          4>
                                                          i
3] = o o
1 " m o o _
*•* T- T- m o


>,
"c
o
v>
.2
'c
o
OB
£
•a
V
•a
i 1 1 i
E"
0>
0) 'D
_N O
'«> E
a> c
S =
a> ^
Cl'C
< .2
•o

E,
I 1 I i 1 1 1 1 1 1 1 1 1
CO P1** CO irt *sf CO ^^ ^" U") v~ C^
o 6 S
"c"
^~"
II II 1
co CM T— m T-
d o c> °. °.
0 0


O
c
'fi

•d

N
(0
C
                                         2
                                        O
                                                                               CD
                                                                               52 -o
                                                                               CO C
                                                                               O CO
                                                                              o w
                                                                                                                             o

                                                                                                                             
                                                                                                                             o

                                                                                                                             OJ

                                                                                                                             
-------
Because the erosion potential is a highly nonlinear function of the
fastest mile, mean values of the fastest mile are inappropriate.  The
anemometer heights of reporting weather stations are found in Refer-
ence 17, and should be corrected to a 10 m reference height using
Equation 3-2.
     To convert the fastest mile of wind (u+) from a reference anemometer
height of 10 m to the equivalent friction velocity (u*), the logarithmic
wind speed profile may be used to yield the following equation:

                              u* = 0.053 uto                         (3~5)

where:      u* = friction velocity (m/s)
          uto = fastest mile of reference anemometer for period between
                 disturbances (m/s)

     This assumes a typical roughness height of 0.5 cm for open
terrain.  Equation 3-5 is restricted to large relatively flat piles or
exposed areas with little penetration into the surface wind layer.
     If the pile significantly penetrates the surface wind layer (i.e.,
with a height-to-base ratio exceeding 0.2), it is necessary to divide the
pile area into subareas representing different degrees of exposure to
wind.  The results of physical modeling show that the frontal face of an
elevated pile is exposed to wind speeds of the same order as the approach
wind speed at the top of the pile.
     For two representative pile shapes (conical and oval with flat-top,
37 degree side slope), the ratios of surface wind speed (us) to approach
wind speed (ur) have been derived from wind tunnel studies.11*  The
results are shown in Figure 3-5 corresponding to an actual pile height of
11 m, a reference (upwind) anemometer height of 10 m, and a pile surface
roughness height (z0) of 0.5 cm.  The measured surface winds correspond
to a height of 25 cm above the surface.  The area fraction within each
contour pair is specified in Table 3-5.
                               3-15

-------
  Flow
Direction
                    Pile A
Pile B1
                     Pile B2
                                                              Pile B3
          Figure 3-5.  Contours of normalized  surface wind  speeds,  us/ur<
                                    3-16

-------
          TABLE 3-5.  SUBAREA DISTRIBUTION FOR REGIMES OF us/ur
Pile subarea
0.2a
0.2b
0.2c
0.6a
0.6b
0.9
1.1
Percent
Pile A
5
35
_
48
_
12
-
of pile surface
Pile Bl
5
2
29
26
24
14
-
area (Fiqure
Pile B2
3
28
_
29
22
15
3
3-3)
Pile 83
3
25
_
28
26
14
4
     The profiles of us/ur in Figure 3-5 can be used to estimate the
surface friction velocity distribution around similarly shaped piles,
using the following procedure:
     1.   Correct the fastest mile value (u"1") for the period of interest
          from the anemometer height (z) to a reference height of 10 m
          (uto) using a variation of Equation 3-2, as follows:

                          u+   -  u+ In  (10/0.005)                     (3.6)
                          ui° "  u    In  (z/0.005)                     v   '

          where a typical roughness height of 0.5 cm (0.005 m) has been
          assumed.  If a site specific roughness height is available, it
          should be used.
     2.   Use the appropriate part of Figure 3-5 based on the pile shape
          and orientation to the fastest mile of wind, to obtain the
          corresponding surface wind speed distribution (u^), i.e.,
                                          o
     3.   For any subarea of the pile surface having a narrow range of
          surface wind speed, use a variation of Equation 4-2 to calcu-
          late the equivalent friction velocity (u*),  as follows:
u* =
0.4 IL
- 2T
1nOT5
                               3-17
                                                                    (3"8)

-------
From this point on, the procedure is identical to that used for a flat
pile, as described above.
     Implementation of the above procedure is carried out in the
following steps:
     1.   Determine threshold friction velocity for erodible material of
          interest (see Tables 3-3 and 3-4 or use Figure 3-3 to determine
          the mode of the aggregate size distribution).
     2.   Divide the exposed surface area into subareas of constant fre-
          quency of disturbance (N).
     3.   Tabulate fastest mile values (u"1") for each frequency of dis-
          turbance and correct them to 10 m (uto) using Equation 3-6.
     4.   Convert fastest mile values (uto) to equivalent friction
          velocities (u*), taking into account (a) the uniform wind expo-
          sure of nonelevated surfaces, using Equation 3-5, or (b) the
          nonuniform wind exposure of elevated surfaces (piles), using
          Equations 3-7 and 3-8.
     5.   For elevated surfaces (piles), subdivide are~s of constant N
          into subareas of constant u* (i.e., within the isopleth values
          of us/ur in Figure 3-5 and Table 3-5) and determine the size of
          each subarea.
     6.   Treating each subarea (of constant N and u*) as a separate
          source, calculate the erosion potential (P^) for each period
          between disturbances using Equation 3-4 and the emission factor
          using Equation  3-3.
     7.   Multiply the resulting emission factor for each subarea by the
          size of the subarea, and add the emission contributions of all
          subareas. Note  that the highest 24-h emissions would be
          expected to occur on the windiest day of the year.  Maximum
          emissions are calculated assuming a single event with the
          highest  fastest mile value for the annual period.
     The recommended emission factor equation presented above assumes
 that all of  the erosion potential corresponding to the fastest mile of
 wind is  lost during the period between disturbances.,  Because the fastest
 mile event typically  lasts only about 2 min, which corresponds roughly to
 the half-life  for  the  decay of actual erosion potential, it could be

                               3-18

-------
argued that the emission factor overestimates participate emissions.
However, there are other aspects of the wind erosion process which offset
this apparent conservatism:
     1.   The fastest mile event contains peak winds which substantially
          exceed the mean value for the event.
     2.   Whenever the fastest mile event occurs, there are usually a
          number of periods of slightly lower mean wind speed which
          contain peak gusts of the same order as the fastest mile wind
          speed.
     Of greater concern is the likelihood of overprediction of wind
erosion emissions in the case of surfaces disturbed infrequently in
comparison to the rate of crust formation.

3.2.3  Wind Erosion of Continuously Active Piles
     For emissions from wind erosion of active storage piles, the
following total suspended particulate (TSP) emission factor equation is
recommended:
                                                                    (3-9)
where:  E = total suspended particulate emission factor
        s = silt content of aggregate, %
        p = number of days with >0.25 mm (0.01 in.) of precipitation per
            year
        f = percentage of time that the unobstructed wind speed exceeds
            5.4 m/s (12 mph) at the mean pile height

The PM10 fraction of TSP is estimated to be 0.5, consistent with the
PMio/TSP ratios for materials handling (Section 3.2.1) and wind erosion
(Section 3.2.2).
     The coefficient in Equation (3-9) is taken from Reference 3, based
on sampling of emissions from a sand and gravel storage pile area during
periods when transfer and maintenance equipment was not operating.  The
                               3-19
                       1.9 (~) ()  () (kg/d/hectare)
                   E = 1.7 (~) (-)  () (Ib/d/acre)

-------
factor from Reference 3, expressed in mass per unit area per day, is more
reliable than the factor expressed in mass per unit mass of material
placed in storage, for reasons stated in that report.  Note that the
coefficient has been halved to adjust for the estimate that the wind
speed through the emission layer at the test site was one half of the
value measured above the top of the piles.  The other terms in this
equation were added to correct for silt, precipitation, and frequency of
high winds, as discussed in Reference 2.
     Worst case emissions from storage pile areas occur under dry windy
conditions.  Worst case emissions from materials handling (batch and
continuous drop) operations may be calculated by substituting into
Equation (3-1) appropriate values for aggregate material moisture content
and for anticipated wind speeds during the worst case averaging period,
usually 24 h.  The treatment of dry conditions for vehicle traffic
(Section 2.0) and for wind erosion (Equation 3-9), centering around
parameter p, follows the methodology described in Section 2.0.  Also, a
separate set of nonclimatic correction parameters and source extent
values corresponding to higher than normal storage pile activity may be
justified for the worst case averaging period.

3.2.4  Level of Contamination
     Although published data on waste pile contamination levels are
sparse, the data presented in Reference 2 suggest several points that may
be useful  in specifying the level of contamination (a) for permitted
waste piles.  These include:
          Source-specific S&A probably will be required to reliably
          characterize a levels for organic compounds.  If a source-
           specific S&A plan is being developed (see Appendix D for
          generic procedures), the data in Reference 2 indicate that
          organic compounds with  lower vapor pressures tend to be found
          more often than compounds with  high vapor pressures (i.e.,
           volatile compounds).  As a result then the S&A plan should be
           directed toward organic compounds with relatively low vapor
           pressures  (<  10 mm Hg).
                                3-20

-------
     •    As an alternative to source-specific S&A, a "worst-case"
          scenario could be developed based on the ranges of available
          data.  For example, this could be accomplished by taking the
          maximum values (upper end of the range) and applying a "safety
          factor" of 5 or 10.  In other words, multiply the upper end
          values by 5 or 10 as a conservatively high estimate of o.

3.3  DEMONSTRATED CONTROL TECHNIQUES
     The control techniques applicable to storage piles fall into dis-
tinct categories as related to materials handling operations (including
traffic around piles) and wind erosion.  In both cases, the control can
be achieved by (a) source extent reduction, (b) source improvement
related to work practices and transfer equipment (load-in and load-out
operations), and (c) surface treatment.  These control options are
summarized in Table 3-6.  The efficiency of these controls ties back to
the emission factor relationships presented earlier in this section.
     In most cases, good work practices which confine freshly exposed
material provide substantial opportunities for emission reduction without
the need for investment in a control application program.  This statement
applies to areas around the pile as well as the pile itself.  In
particular, spillage of material caused by pile load-out and maintenance
equipment can add a large source component associated with traffic-
entrained dust.  Emission inventory calculations show, in fact, that the
traffic dust component may easily dominate emissions from transfer of
material and wind erosion.  The prevention of spillage and subsequent
spreading of material by vehicle tracking is essential to cost-effective
emission control.  If spillage cannot be prevented because of the need
for intense use of mobile equipment in the waste pile area, then regular
cleanup should be employed as a necessary mitigative measure.
     The evaluation of preventative methods which change the properties
or exposure of transfer streams or surface material are discussed in the
following section.
                               3-21

-------
            TABLE 3-6.   CONTROL TECHNIQUES FOR STORAGE PILES
    Material  handling
      Source  extent reduction
      Source  improvement


      Surface treatment
    Wind erosion
      Source  extent reduction


      Source  improvement

      Surface treatment
Mass transfer reduction
Drop height reduction
Wind sheltering
Moisture retention
Wet suppression
Disturbed area reduction
Disturbance frequency reduction
Spillage cleanup
Spillage reduction
Wind exposure reduction
Wet suppression
Chemical stabilization
3.4  CONTROL PERFORMANCE ESTIMATION
     Preventive methods for control of emissions from waste piles include
chemical stabilization, enclosures, and wet suppression.  Physical
stabilization by covering the exposed surface with less erodible aggre-
gate material and/or vegetative stabilization are seldom practical con-
trol methods for waste piles, unless piles are to remain in place for
long periods of time.
     To test the effectiveness of chemical stabilization controls for
wind erosion of aggregate storage piles and tailings piles, wind tunnel
measurements have been performed.  Although most of this work has been
carried out in laboratory wind tunnels, portable wind tunnels have been
used in the field on storage piles and tailings piles. iB»i9  Laboratory
wind tunnels have also been used with physical models to measure the
effectiveness of wind screens in reducing surface wind velocity.11*
                               3-22

-------
3.4.1  Chemical Stabilization
     A portable wind tunnel has been used to measure the control  of coal
pile wind erosion emissions by a 17% solution of Coherex® in water
applied at an intensity of 3.4 L/m2 (0.74 gal/yd*), and a 2.8% solution
of Dow Chemical M-167 Latex Binder in water applied at an average
intensity of 6.8 L/mz (1.5 gal/ydz).i«  The control efficiency of
Coherex® applied at the above intensity to an undisturbed steam coal
surface approximately 60 days before the test, under a wind of 15.0 mi/s
(33.8 mph) at 15.2 cm (6 in.) above the ground, was 89.6% for TP and
approximately 62% for IP and FP.  The control efficiency of the latex
binder on a low volatility coking coal is shown in Figure 3-6.
      100
       80
    o
    03
    O
       60
    uu
    "o
o  40
    o
    O
       20
                                        6.8   /m2 (1.5 gal/yd2) of
                                        2.8% Solution in Water
Tunnel Wind
Speed * 17 m/s (38 mph)
at 15 cm (6.0 in)
Above the Test Surface

Key:
                     1
                     1           2            3
                         Time After Application (Days)
   Figure 3-6.
            Decay  in  control efficiency of  latex binder applied to
                     coal  storage piles.18
                               3-23

-------
     Cost elements for chemical stabilization are presented in Table 3-7.
The cost of a system for application of surface crusting chemicals to
storage piles is $18,400 for the initial capital cost and $0.006 to
$0.Oil/ft2 for annual operating expenses based on April 1985 dollars.20
Tables 3-8 and 3-9 provide recordkeeping forms for application of chem-
ical dust suppressants.

3.4.2  Enclosures
     Enclosures are an effective means by which to control fugitive
particulate emissions from open dust sources.  Enclosures can either
fully or partially enclose the source.  Included in the category of
partial enclosures are porous wind screens or barriers.  Practically any
means that reduces wind entrainment of particles from a dust-producing
surface or dispersion of a dust plume generated by a source (e.g., front-
end loader in a three-sided enclosure) is generally effective in
controlling fugitive particulate emissions.
     Partial enclosures used for reducing windblown dust from large
exposed areas and storage piles include porous wind fences and similar
types of physical barriers (e.g., trees).  The principle of the wind
fence/barrier is to provide an area of reduced wind velocity which allows
settling of the large particles (which cause saltation) and reduces the
particle flux from the exposed surface on the leeward side of the fence/
barrier.  Wind fences/barriers can either be man-made structures or
vegetative in nature.
     With the exception of wind fences/barriers, a review of available
literature reveals no quantitative information on the effectiveness of
enclosures to control fugitive dust emissions from open sources.  Types
of passive enclosures traditionally used for open dust control include
three-sided bunkers for the storage of bulk materials, storage silos for
various  types of aggregate material (in lieu of open piles), open-ended
buildings, and similar structures.
     A  number of studies have  attempted to determine the effectiveness of
wind fences/barriers for the control of windblown dust under field
conditions.   Several of these  studies have shown both a significant
decrease in wind velocity as well as an increase in sand dune growth on
the lee  side  of the  fence.21"21*
                               3-24

-------
 TABLE  3-7.   CAPITAL  AND  O&M ITEMS  FOR  CHEMICAL  STABILIZATION
                        OF WASTE PILES
Capital equipment

  •  Storage equipment
       Tanks
       Railcars
       Pumps
       Piping

  •  Application equipment
       Trucks
       Spray system
       Piping (including winterizing)

O&M expenditures

  •  Utility or fuel costs
       Water
       Electricity
       Gasoline or diesel fuel

  •  Supplies
       Chemicals
       Repair parts

  •  Labor
       Application time
       Road conditioning
       System maintenance
                           3-25

-------
5

i



£

S
<:
o
0.
CO
 I
to
               ID


              +-

            O —
            o
            E •O
            a. «
              •a
              
              i.
              ID
              a
        o >-
        — -KM
        •t- — -O
        10 in x
        — -t- ID
        a. c 01
           — o


           3 -t-
           — IO
           •*• ID
           O JJ

           O E
           a a)
           >-j=
           i— o
              10
             a
                                                             3-26

-------
a.
T

UJ
          in

          •g
          u
          (O
        It
       — 4)
       — CT
          •o
        >- a
       4- L.
        10 —
        3 a
       O"0
        10 t_
       jj 
-------
     A 1988 laboratory wind tunnel study11* of windbreak effectiveness for
coal storage piles showed area-averaged wind speed reductions of - 50% to
70% for a 50% porosity windbreak with height equal to the pile height and
length equal to the pile base.  The windbreak was located three pile
heights upwind from the base of the pile.  This study also suggested
"that fugitive dust emissions on the top of the pile may be controlled
locally through the use of a windbreak at the top of the pile."
     Based on the 1.3 power given in Equation (3-1), mean wind speed
reductions of - 50% to 70% would correspond to - 60% to 80% control of
material handling PM10 emissions.  Estimation of the wind erosion control
efficiency requires source-specific evaluation because of the dependence
of erosion potential on ut and u* (for both controlled and uncontrolled
conditions) as shown in Equation (3-4).
     This same laboratory study^ showed that a storage pile may itself
serve as a windbreak by reducing wind speed on the leeward face (Fig-
ure 3-5).  The degree of wind sheltering and associated wind erosion
emission reduction is dependent on the shape of the pile and on the
approach angle of the wind to an elongated pile.
     One of the real advantages of wind fences for the control of PM10
involves the low capital and operating costs.21'23  These involve the
following basic elements:
     •  Capital equipment
        —  Fence material and supports
        --  Mounting hardware
     •  Operating and maintenance expenditures
        —  Replacement fence material and hardware
        —  Maintenance labor
     The following cost estimates (in 1980 dollars) were developed for
wind screens applied to aggregate storage piles:26
      •  Artificial wind guards
        —   Initial capital cost = $12,000 to $61,000
     Due to the  lack of quantitative data on costs associated with wind
 screens, it is recommended that  local vendors be  contacted to obtain more
 detailed data for capital and operating  expenses.  Also, since wind
 fences  and  screens  are relatively "low tech" controls,  it may be possible
                                3-28

-------
for the site operator to construct the necessary equipment using site
personnel with less expense.
     As with other options mentioned above, the main regulatory approach
for wind fences and screens would involve recordkeeping by the site
operator.  Parameters to be specified in the dust control plan and
routinely recorded are:
     General Information
     1.   Locations of all materials storage and handling operations to
          be controlled with wind fences referenced on a plot plan
          available to the site operator and regulatory personnel
     2.   Physical dimensions of each source to be controlled and
          configuration of each fence or screen to be installed
     3.   Physical characteristics of material to be handled or stored
          for each operation to be controlled by fence(s) or screen(s)
     4.   Applicable prevailing meteorological data (e.g., wind speed and
          direction) for site on an annual basis
     Specific Operational Records
     1.   Date of installation of wind fence or screen and initials of
          installer
     2.   Location of installation relative to source and prevailing
          winds
     3.   Type of material being handled and stored and physical
          dimensions of source controlled
     4.   Date of removal of wind fence or screen and initials of
          personnel involved
     General Records to be Kept
     1.   Fence or screen maintenance record
     2.   Log of meteorological conditions for each day of site operation

3.4.3  Wet Suppression Systems
     Fugitive emissions from aggregate materials handling systems are
frequently controlled by wet suppression systems.  However, use of water-
based dust suppression systems may result in unacceptable increases in
contaminated leachate production.  Thus, before instituting a watering
program, the facility must demonstrate that addition of water to the

                               3-29

-------
waste pile area does not produce undesirable impacts on ground and
surface water quality.
     Wet suppression systems use liquid sprays or foam to suppress the
formation of airborne dust.  The primary control mechanisms are those
that prevent emissions through agglomerate formation by combining small
dust particles with larger aggregate or with liquid droplets.  The key
factors that affect the degree of agglomeration and, hence, the perfor-
mance of the system are the coverage of the material by the liquid and
the ability of the liquid to "wet" small particles.  This section
addresses two types of wet suppression systems—liquid sprays which use
water or water/surfactant mixtures as the wetting agent and systems which
supply foams as the wetting agent.
     Liquid spray wet suppression systems can be used to control dust
emissions from materials handling at conveyor transfer points.  The
wetting agent can be water or a combination of water and a chemical
surfactant.  This surfactant, or surface active agent, reduces the
surface tension of the water.  As a result, the quantity of liquid needed
to achieve good control is reduced. For systems using water only, addi-
tion of surfactant can reduce the quantity of water necessary to achieve
a good control by a ratio of 4:1 or more.27'28
     The design specifications for wet suppression systems are generally
based on the experience of the design engineer rather than on established
design equations or handbook calculations.  Some general design guide-
lines that have been reported in the literature as successful are listed
below:
     1.  A variety of nozzle types have been used on wet suppression
systems, but recent data suggest that hollow cone nozzles produce the
greatest control while minimizing clogging.29
     2.  Optimal droplet size for surface impaction and fine particle
agglomeration  is about 500 ym; finer droplets are affected by drift and
surface tension and appear to be less effective.30
     3.  Application of water sprays to the underside of a conveyor belt
 improves the performance of wet suppression systems at belt-to-belt
transfer points.31
                                3-30

-------
     Micron-sized foam application is an alternative to water spray
systems.  The primary advantage of foam systems 1s that they provide
equivalent control at lower moisture addition rates than spray sys-
tems.31  However, the foam system is more costly and requires the use of
extra materials and equipment.  The foam system also achieves control
primarily through the wetting and agglomeration of fine particles.  The
following guidelines to achieve good particle agglomeration have been
suggested:32
     1.  The foam can be made to contact the particulate material by any
means.  High velocity impact or other brute force means are not required.
     2.  The foam should be distributed throughout the product
material.  Inject the foam into free-falling material rather than cover
the product with foam.
     3.  The amount applied should allow all of the foam to dissipate.
The presence of foam with the product indicates that either too much foam
has been used or it has not been adequately dispersed within the
material.
     Available data for both water spray and foam wet suppression systems
are presented in Tables 3-10 and 3-11, respectively.  The data primarily
include estimates of control efficiency based on concentrations of total
particulate or respirable dust in the workplace atmosphere.  Some data on
mass emissions reduction are also presented.  The data should be viewed
with caution in that test data ratings are generally low and only minimal
data on process or control system parameters are presented.
     The data in Tables 3-10 and 3-11 do indicate that a wide range of
efficiencies can be obtained from wet suppression systems.  For conveyor
transfer stations, liquid spray systems had efficiencies ranging from 42%
to 75%, while foam systems had efficiencies ranging from 0% to 92%.  The
data are not sufficient to develop relationships between control or
process parameters and control efficiencies.  However, the following
observations relative to the data in Tables 3-10 and 3-11 are noteworthy:
     1.  The quantity of foam applied to a system does have an impact on
system performance.  On grizzly transfer points, foam rates of 7.5 ft3 to
10.5 ft3 of foam per ton of sand produced increasing control efficiencies
ranging from 68% to 98%.33  Foam rates below 5 ft3 per ton produced no
measurable control.
                               3-31

-------









I
oc
is
§


s
s
c


u.
S:
g
E
§
M

i
|
<
i
O
A
hU
9
H-

















e* E
JJi S
t *



® **


1
1
i
1




%
9
^
*
vi
VI
I
e
<3



„
^ 1
•21 «
S a
' ?
v» —
%n **
S 2
II


"a
*
3
O
1


jj
S
1
•5
a
*~
i*

-


o v


slsl
!s!s
U!;
i*i*

* ^ 0

*° S -S
fiT c "v»
• « fc

as,3
«! -! §
»< x r* x
.7.2
s ° s, *
2 £ 2 * —
a. « a e 41
«» a vt o -9
00 03

5
i
s
u
(30
,,
&

^
n






1


O k
k "S
II
**•
|5
u
s

a *
s s


e v


tl.J
« u • w
1- t "
s c 5 e
jijl

« s *.
S -o °

C c '•i
^ 5 fc

a i.3
(A W> §
tNi >« r«I x
» e * u
2£ 2 Hi
SK * g J!
00 03





•8
|^
"w
2L
VI
1






3



^
s t
s s

0


* a 5 s
oe S S or


S S S S


tflrleicl
• Si g£ Si ii
Is Is Ic I c
HHHli
f " e? " £ " £ "



V *

^ V « V
<\J 






J


Vt

i-tl S
xf S
§: g 2
s »-
u
0,

I
$
u
1
i
£
•5
1
o
M
8
3
i

^
u
5
,„
ai
«
ft
H-

|
|
|

g1
fc

2
VI
£
.2"
~.
§
S
|
*o
$
1
=
15
tSf


«*.
I

s
1
tft
1


























*
•S
^
S
t
o
* •M
£ .' €
1 ! 2
5 fi t
fe 3 «
a C o
w u e
is ^
•t 2 H-
2 i . 1
^ r i s r
* S. tS « 8
i; a .z £ „
*^ k u u
« « a 3 5
"3 *< " "" "5
t 1 § § t!
« f k k 4f
i ^ o> «i —
- S "• tt -
« a in u> J
° 1 1 't f
S*J *J T3
" 5 —
^" U ^1 <» "^
3-32

-------



1
£
5
trt
*
g
i
I

i

w
it
g
K
S
S
5
5
fc
fc
S
i

5
^
i

















ill
ji
9
1
1
2


i
«
*

8

|






J5

"e S
" 1
1 2
% "Z
3 t
II




•j
5

^
o
N?

VI
i
a
e
V
ft
>•_ •
8
S

1 8
11
£ "





£

1
*4






U_
S

ft
Of
1

^


„,
jg
•5,
f
1
s


T u
^O Of
1 **•
2 g
i 2
i"
a
a
s

ll
5 C
Personnel
Type 1 i





|

I
~






S-
§

ft
V
•o

1


„
s

J.
•
s


S u
I **.
O V>
** e
i n
*rf U
(D

U
a

samplers.
test scheae
Personnel
Type 1 1





|
<*-
I







J^
S

ft

i

?
•0
VI

1

JC
s
8


I

^o
Jtf
i

s
-

li
!:s


i
&
u>
CM
e

^
VI
i
X

u.
1
'""
V
^

VI
VI
s
p**



1
•s
e v
s s
•0 "
* J
S "•
u


I u
0 ijl
r s
2 2
b-

»
m
*
II
i>


1
&
0
o

|
VI
i
X

u.
s
"
I
•o

"
^"
s
5




•s

i 1

i •
u

^
1 Sf1
*0f «

0) IA
I5

i
-
c
fj
s s
•— *i*
f *
p


g
1
«0
•«*
O*
H
t
VI


t*.
S
1
i
•w
s
*
^"
§
2




•s
§^
S
VI
1 2
S "m
0


*« u
•0 £
2 S
i 2
S"

•**
•
c
I!
2 S
P


4^
&

2
•
9
2"


U.
S
•""
!
^
s
M
<-X.
VI
5
S




•s
e v
s s
-0 "
if 2
S 1
u


k 8.
2 te

*1
h«V

8i 5 8
» - -

el .J el
• y • u v u
iiiii?
!- «- §-
ciiii!
£ £ £
•a
5 c S B • e
llllll
£ «!: s"c a
"! e 
-------

















_
^8
i
|


•tut
<*»
yj
S



























i-h
Ir g
3 Z a.



o -
fi'S


*»
e
:
!
I!


£
I
*
Ck
1
Wl
J?

e
e
o
O


|
41
1.5

s a
'1

S 2
II

—
•••
01
a

e
i


i
2
a
Of
ft



2s

e e S C S
3 5 & at 5



- S 5 5 5



elrlef g|e|
i r § r i - § r § r
ScScScJcSs
S&S&S|S|S|
£"£1~£>"£'~£'~

1 cl e
jyi; |
"c si: 2 1 § »
00 O ^ O ~+ ~* *
» • ~ • 1 1 Jj
• « « * 3 9 9
«4rf ^ ^ ••
« * e o e
*£ «^ » » »

flit! ! !






2

u

*rt
|
*
£
•
O
e
2
ll
5"




—
N

(5


rsi

«*
S
u
1
"5
l
s
•5
5
M
I
e
s
i
«
8
|
£
M
8
—
41
ft

VI
1
|
>i
£
b

S
wt
2

^
S
s
X
"o
8

1 a
*n £
£ 3
• I
»te a
g «
S i
*a y*
1 £
VI
m
1 J


































•
!
=
. c
i £
5 |

I -
5 S
2 C
€ -
S •£
s a.
s I

1 -
>, I
u V
•£ *o
^ £

1 J
3-34

-------
     2.  Material temperature has an impact on foam performance.  At one
plant where sand was being transferred, control efficiencies ranged from
20% to 65% when 120°F sand was handled.  When sand temperature was
increased to 190°F, all control efficiencies were below 10%."
     3.  Data at one plant suggest that underside belt sprays increase
control efficiencies for respirable dust (56% to 81%).3l
     4.  When spray systems and foam systems are used to apply equivalent
moisture concentrations, foam systems appear to provide greater con-
trol.33  On a grizzly feed to a crusher, equivalent foam and spray
applications provided 68% and 46% control efficiency, respectively.  Cap-
ital and O&M cost elements for wet suppression are shown in Table 3-12.
     In estimating the wind erosion control effectiveness of wet
suppression, it can be assumed that emissions are inversely proportional
to the square of the surface moisture content.  The emission/moisture
dependence is embedded in the agricultural wind erosion equation.  It
also appears in the observed relationship between the rate of emissions
from an unpaved road and the surface moisture content.
     In addition, a relationship between surface moisture content and
daily moisture addition has been developed from field studies of storage
piles exposed to natural precipitation.  The results of that research are
illustrated in the example problem to be presented at the end of this
section.
     Costs associated with wet suppression systems include the following
basic elements:
     •  Capital equipment
        —  Spray nozzles or other distribution equipment
        —  Supply pumps and plumbing (plus weatherization)
        —  Water filters and flow control equipment
        ~  Tanker truck (if used)
     •  Operating and maintenance expenditures
        ~  Water and chemicals
        —  Replacement parts for nozzles, truck, etc.
        —  Operating labor
        —  Maintenance labor
                               3-35

-------
      TABLE  3-12.   WET SUPPRESSION SYSTEM CAPITAL AND O&M
                         COST ELEMENTS
Capital equipment

  •  Water spray system
       Supply pumps
       Nozzles
       Piping (including w1nter1zat1on)
       Control system
       Filtering units

  •  Water/surfactant and foam systems only
       A1r compressor
       Mixing tank
       Metering or proportioning unit
       Surfactant storage area

O&M expenditures

  •  Utility costs
       Water
       Electricity

  •  Supplies
       Surfactant
       Screens

  •  Labor
       Maintenance
       Operation
                            3-36

-------
     Reference 6 estimates the following costs (in 1985 dollars):
     •  Regular watering of storage piles
        —  Initial capital cost = $18,400 per system
     *  Watering of exposed areas
        —  Initial capital cost « $1,053 per acre
        —  Annual operating cost » $25 to $67 per acre
     The costs associated with a stationary wet suppression system using
chemical surfactants for the unloading of limestone from trucks at
aggregate processing plants (in 1980 dollars) have been estimated at:


                  capital  = $72,000;  annual  = $26,000.


Typical costs for wet suppression of materials transfer operations are
listed in Table 3-13.
        TABLE  3-13.   TYPICAL COSTS  FOR WET SUPPRESSION OF MATERIAL
                             TRANSFER POINTS
Source method
Initial  cost,
 April  1985
  dollars*
Unit operating cost,
April 1985 dollars*
Railcar unloading station       48,700
  (foam spray)
Railcar unloading station
  (charged fog)
Conveyor transfer point         23,700
  (foam spray)

Conveyor transfer point         19,800C
  (charged fog)
                NR
  168,000b      NR
                0.02 to 0.05/ton material
                  treated

                NR
Source:  Reference 20.  NR = not reported.

aJanuary 1980 costs updated to April 1985 cost by Chemical Engineering
 Index.  Factor = 1.315.

bBased on use of 16 large devices at $10,500 each.

cBased on use of three small devices at $6,600 each.

                               3-37

-------
     As with watering of unpaved surfaces, enforcement of a wet suppres-
sion control program would consist of two complementary approaches.  The
first would be record keeping to document that the program is being
implemented and the other would be spot-checks and grab sampling.  Both
were discussed previously above.
     Records must be kept that document the control plan and its
implementation.  Pertinent parameters to be specified in a plan and to be
regularly recorded include:
     General Information
     1.   Locations of all materials storage and handling operations
          referenced on plot plan of the site available to the site
          operator and regulatory personnel
     2.   Materials delivery or transport flow sheet which indicates the
          type of material, its handling and storage, size and composi-
          tion of storage piles, etc.
     3.   The method and application intensity of water, etc., to be
          applied to the various materials and frequency of application,
          if not continuous
     4.   Dilution ratio for chemicals added to water supply, if any
     5.   Complete specifications of equipment used to handle the various
          materials and for wet suppression
     6.   Source of water and chemical(s), if used
     Specific Operational Records
     1.   Date of operation and operator's initials
     2.   Start and stop time of wet suppression equipment
     3.   Location of wet suppression equipment
     4.   Type of material being handled and number of loads (or other
          measure of throughput) loaded/unloaded between start and stop
          time (if material is being pushed, estimate the volume or
          weight)
     5.   Start and stop times for tank filling
     General Records
     1.   Equipment maintenance records
     2.   Meteorological  log of general conditions
     3.   Records of equipment malfunctions and downtime
                                3-38

-------
3.5  PROCEDURES FOR COMPLIANCE DETERMINATION
     There are several possible regulatory formats for control of dust
emissions from waste piles.  Commonly used opacity standards at the
property line are applicable as are opacity standards at the point of
emissions for continuous drop of material on to a pile by a stacker.
     For wet suppression and chemical stabilization, suitable record-
keeping forms, such as those presented in Tables 3-8 and 3-9, would
provide evidence of control plan implementation.  In addition, simple
measurements of moisture level in transferred material or of the crust
strength of the chemically treated surface could be used to verify
compliance.  If the crust is more than 0.6 cm thick and not easily
crumbled between the fingers, then the surface may be considered
nonerodible.  In addition, the loading as well as the texture of material
deposited around the pile could be used to check whether good work
practices are being employed relative to pile reclamation and maintenance
operations.  The suitability of these measurements of surrogate
parameters for source emissions stems from the emission factor models
which relate the parameters directly to emission rate.  Potential
regulatory formats for control of waste pile emissions are listed in
Table 3-14.

3.6  EXAMPLE CALCULATION

     Source Description
     •  Conically shaped pile (dry, finely divided material)
     •  Pile height of 11 m; 29.2 m base diameter; 838 m2 surface area
     •  Pile formation followed by 30-d exposure period before removal
        (Figure 3-7)
     •  LCD as shown in Figure 3-8 for a typical month
     •  Contamination level, a = 12,000 ppm Pb
     Calculation of Uncontrolled Emissions
     Step 1;  A composite sample of approximately 0.5 L of waste material
was obtained by collecting nearly equal amounts from six areas of the
pile surface.  This sample was hand-sieved according to the procedures
shown in Figure 3-9.
                               3-39

-------























UJ
H-x«
d.
UJ
H"


^^
o
U—
_
0
,_
v"^
^^
^^
^T
™-p
~~
J—
UJ
UJ

•o

"3

T3

O
u_


01

3
(/I
o>
E

'o
i.
+J
c
o








a)
i *
*o
en
0
b.
b.
3
to
c
o

CO
42
E
UJ












•o c
(U O
OJ U
a.
U) 0)
i.
•0 3
C +J
*3 •^*
O
(U E
0
a cu
M- -4-1
1- V)
3 «





O) C
c o
i. 
•^ QJ
33


+-> 1
1.
b. •»->
o c
01
OJ O 5) ^
S- 1— S-
c; •*••. a>
CL OQ.
^ ?T i»
Q.Q- a.
*J •••M
(O QJ t3
U
UJ J- C
> 3 O
O *^
•** 
en
^
^
•o
c
ra
JZ

(D
i >
(/)
(T3
3
a;
b. 4->
3 C
4-> 
0) C
4-> O
u
^5
C flU
<0 t-
3

 V
+•} ft^ ^)

JZ n- 0>
023

• «
>,-M 1
if * 2
(J S- J->
o o c
aj a> o a;
> c c c
•f» Q *^-
C r— U i—
O
•«-> *J OO 4J
U J- 1— b.
•»—  3 O
JZ 0 f-
t-yi. 
-------
       Prevailing
       Wind
       Direction
Area
ID
A
B
Ci
C2

us
0.9
0.6
0.2
0.2

Pile
12
48
5
35

Surface
Area (m2)
101
402
42
293
838
Figure 3-7.  Example.  Pile  surface areas within each wind speed  regime.
                                   3-41

-------
             Local  Climatologlcal Data
                    Monthly Summary
         For the Month:
30
	
3.3
	
11.1
	
31
29
Date: 1 1
Wind

5
1
CO

c
13
30
01
10
13
12
20
29
29
22
14
29
17
21
10
10
01
33
27
32
24
22
32
29
07
34
31
30
30
33
34
29

.c
d.
|E
^ W
CD 8
occo
14
5.3
10.5
2.4
11.0
11.3
11.1
19.6
10.9
3.0
14.6
22.3
7.9
7.7
4.5
6.7
13.7
112
4.3
9.3
7.5
10.3
17.1
2.4
5.9
11.3
12.1
8.3
8.2
5.0
3.1
4.9
-o
CO
o
2 si
9 6.

-------
     FIELD PROCEDURE FOR DETERMINATION OF THRESHOLD FRICTION VELOCITY*

1.  PREPARE A NEST OF SIEVES WITH THE FOLLOWING OPENINGS:  4 mm, 2 mm,
    1 mm, 0.5 mm, 0.25 mm.  PLACE A COLLECTOR PAN BELOW THE BOTTOM SIEVE
    (0.25-mm OPENING).                                                	

2.  COLLECT A SAMPLE REPRESENTING THE SURFACE LAYER OF LOOSE PARTICLES
    (APPROXIMATELY 1 cm IN DEPTH FOR AN UNCRUSTED SURFACE), REMOVING ANY
    ROCKS LARGER THAN ABOUT 1 cm IN AVERAGE PHYSICAL DIAMETER.  THE AREA
    TO BE SAMPLED SHOULD NOT BE LESS THAN 30 on x 30 cm.

3.  POUR THE SAMPLE INTO THE TOP SIEVE (4-ran OPENING), AND PLACE A LID ON
    THE TOP.

4.  ROTATE THE COVERED SIEVE/PAN UNIT BY HAND USING BROAD SWEEPING ARM MO-
    TIONS IN THE HORIZONAL PLANE.  COMPLETE 20 ROTATIONS AT A SPEED JUST
    NECESSARY TO ACHIEVE SOME RELATIVE HORIZONTAL MOTION BETWEEN THE SIEVE
    AND THE PARTICLES.

5.  INSPECT THE RELATIVE QUANTITIES OF CATCH WITHIN EACH SIEVE AND
    DETERMINE WHERE THE MODE IN THE AGGREGATE SIZE DISTRIBUTION LIES,
    I.E., BETWEEN THE OPENING SIZE OF THE SIEVE WITH THE LARGEST CATCH AND
    THE OPENING SIZE OF THE NEXT LARGEST SIEVE.


*AOAPTED FROM A LABORATORY PROCEDURE PUBLISHED BY W.  S.  CHEPIL (1952).12
                                  Figure 3-9
                                  3-43

-------
Visual inspection of the quantities retained on each sieve showed that
the largest amounts were contained on the 1 mm and 2 mm sieves.
Therefore, a particle size mode of 1.5 mm was assumed.  This size mode
corresponds to a threshold friction velocity of 75 cm/sec as obtained
from Figure 3-3.
     Step 2;  The entire pile surface is disturbed only once (i.e., at
the beginning of the 1-mo exposure period).  This corresponds to a value
of N = 1/yr.
     Step 3:  The calculation procedure involves determination of the
fastest mile for each period of disturbance.  Figure 3-8 shows a repre-
sentative set of values (for a 1-mo period) that are assumed to be
applicable to the geographic area of the pile location.  The highest
value for the 1-mo period is 31 mph (13.9 m/s).  In this example, the
anemometer height is 7m, so that a height correction to 10 m is needed
for the fastest mile values.
     From Equation (3-6)

                            +  .  + In  10/0.005)
                           u
                             i°
                            +
                                       (10/0.
                                    In (7/0.005)
                                 1  nr-
                           u10 = I. 05 u7
                               = 1.05(13.9) = 14.6 m/s

     Step 4;  The next step is to convert the fastest mile value for the
1-mo period into the equivalent friction velocity for each surface wind
regime (i.e., us/ur ratio) of the pile, using Equations 3-7 and 3-8.
Figure 3-7 shows the surface wind speed pattern (exjpressed as a fraction
of the approach wind speed at a height of 10 m).  The surface areas lying
within each wind speed regime are tabulated below the figure.
     The calculated friction velocities are presented in Table 3-15.  As
indicated, the friction velocity exceeds the threshold value of 0.75 m/s
within the us/ur = 0.9 regime and the 0.6 regime of the pile surface.
     Step 5;  This step is not necessary because there is only one
frequency of disturbance used in the calculations.
                               3-44

-------
     Steps 6 and 7:  The final set of calculations (shown 1n Table 3-15)
Involves the tabulation and summation of emissions for each disturbance
period and for the affected subarea.  The erosion potential (P) is
calculated from Equation (3-4).
     For example, the calculation for the 0.9 regime is:

          P2 = 58(1.31 - 0.75)2 + 25(1.31 - 0.75)
             = 32.2 g/m2

     The PM10 emissions generated by the fastest mile event are found as
the product of the PM10 multiplier (k = 0.5), the erosion potential (P),
and the affected area of the pile (A).

     TABLE 3-15.   EXAMPLE  CALCULATION OF UNCONTROLLED PMlo EMISSIONS3



A
B
Ci
C2

us/ur
0.9
0.6
0.2
0.2
+
(m/s)
1.31
0.88
0.29
0.29
ir ie
u -ut
(n/sl
0.56
0.13
-
-
Pile surface
P
(g/m2)
32.2
4.2
0
0
A
(m2)
101
402
42
293
kPA
(kg)
1.626
0.844
0
0

  aWhere ut = 0.75 m/s and k = 0.5 for PM10.
     As shown in Table 3-15, the results of these calculations indicate
an uncontrolled PM10 emission total of 2.47 kg.  Multiplying by the
contaminant fraction gives 2.47 kg x 12,000/1,000,000 = 29.6 g of Pb
emissions for the 1-mo exposure period.
     Target Control Efficiency;  60%
     Method of Control;  Application of a latex binder to the pile
surface immediately after pile formation.
                               3-45

-------
     Demonstration of Control Program Adequacy:  Uncontrolled PM10 wind
erosion emissions, Eu, from the storage pile were shown to be 2.47 kg for
the month.  To achieve a control efficiency of 60%, calculate the con-
trolled emissions, Ec, using the following relationship.

          Ec - Eu (1 - 0.60)
             = 2.47(1 - 0.6)
             = 1.00 kg
     The desired control efficiency of 60% may be achieved, in effect, by
increasing the threshold friction velocity through the application of a
latex binder.  A field study of the effectiveness of various chemical
dust suppressants on finely divided tailings material indicates that a
threshold velocity of approximately 1.06 m/s over a period of 1 month can
be achieved by a single application of latex binder.  In this study,
Nalco 655 or 656 (full strength) was applied at an application intensity
of 0.25 gal/yd2.i9
     As indicated in Table 3-16, the effect of the latex binder applica-
tion in reducing the threshold friction velocity to 1.06 m/s is to reduce
the PM10 emission rate to a value of 1.00 kg resulting from wind erosion
over the 1-mo exposure period.  This yields the desired control effi-
ciency of 60%, i.e., the PM10 emissions are reduced to 40% of the uncon-
trolled value of 2.47 kg (Table 3-15).  Contaminated PM10 would be
reduced correspondingly to 1.00 kg x 12,000/1,000,000 = 12 g Pb
emissions.
                                3-46

-------
      TABLE 3-16.  EXAMPLE CALCULATION OF CONTROLLED PM10 EMISSIONS4



us/ur
0.2
0.6
0.9

*
u
(m/s)
0.29
0.88
1.31

* *
u -ut
(•/$)

_
0.25

Pile surface
P
(g/m2)
0
0
9.88
Total
A
(m2)
335
402
101
PMio
kPA
(kg)
0
0
0.997
1.00 kg

      aWhere u£ * 1.06 m/s and k = 0.5 for PMJ0.
3.7  REFERENCES FOR SECTION 3

 1.  U.S. Environmental Protection Agency.  Hazardous Waste TSDF-Back-
     ground Information for Proposed RCRA Air Emission Standards, Draft
     EIS.  Office of Air Quality Planning and Standards.  Research
     Triangle Park, North Carolina.  March 1988.

 2.  Englehart, P., and D. Wallace.  Assessment of Hazardous Waste TSDF
     Particulate Emissions.  EPA Contract No. 68-02-3891, Midwest
     Research Institute.  October 1986.

 3.  Cowherd, C., Jr., et al.  Development of Emission Factors for Fugi-
     tive Dust Sources.  EPA-450/3-74-037.  U.S. Environmental Protection
     Agency, Research Triangle Park, North Carolina.  June 1974.

 4.  Bohn, R., et al.  Fugitive Emissions from Integrated Iron and Steel
     Plants.  EPA-600/2-78-050.  U.S. Environmental Protection Agency,
     Research Triangle Park, North Carolina.  March 1978.

 5.  Cowherd, C., Jr., et al.  Iron and and Steel Plant Open Dust Source
     Fugitive Emission Evaluation.  EPA-600/2-79-103.  U.S. Environmental
     Protection Agency, Research Triangle Park, North Carolina.  May
     1979.

 6.  Bohn, R.  Evaluation of Open Dust Sources in the Vicinity of
     Buffalo, New York.  U.S. Environmental Protection Agency, New York,
     New York. March 1979.

 7.  Cowherd, C., Jr., and T. Cuscino, Jr.  Fugitive Emissions Evalua-
     tion.  Equitable Environmental Health, Inc., Elmhurst, Illinois.
     February 1977.
                               3-47

-------
 8.   Cuscino, T., et al.  Taconite Mining Fugitive Emissions Study.
     Minnesota Pollution Control Agency, Roseville, Minnesota.  June
     1979.

 9.   Axetell, K., and C. Cowherd, Jr.  Improved Emission Factors for
     Fugitive Dust from Western Surface Coal Mining Sources.  2
     Volumes.  EPA Contract No. 68-03-2924, PEDCo Environmental, Inc.,
     Kansas City, Missouri.  July 1981.

10.   Local Climatological Data.  Monthly Summary Available for each U.S.
     Weather Station from the National Climatic Center.  Asheville, North
     Carolina 28801.

11.   Cowherd, C., Jr.  "Background Document for AP-42 Section 11.2.7 on
     Industrial Wind Erosion."  EPA Contract No. 68-02-4395, Midwest
     Research Institute.  July 1988.

12.   Chepil, W. S.  "Improved Rotary Sieve for Measuring State and
     Stability of Dry Soil Structure."  Soil Science Society of America
     Proceedings.,  16:113-117.   1952.

13.   Gillette, D. A., et al.  "Threshold Velocities for Input of Soil
     Particles into the Air by Desert Soils."  Journal of Geophysical
     Research, 54(C10):5621-5630.

14.   Studer, B. J. B., and S. P. S. Arya.  "Windbreak Effectiveness for
     Storage Pile Fugitive Dust Control:  A Wind Tunnel Study."  Journal
     of the Air Pollution Control Association.  38:135-143.   1988.

15.   Muleski, G. E.  "Coal Yard Wind Erosion Measurements.   Final Report
     prepared for Industrial Client of Midwest Research Institute, Kansas
     City, Missouri.  March 1985.

16.   Nickling, W. G., and J. A. Gillies.  "Evaluation of Aerosol Produc-
     tion Potential of Type Surfaces in Arizona."  Submitted to
     Engineering-Science.  Arcadia, California,  for EPA  Contract No.
     68-02-388.  1986.

17.  Changery, M. J.  National Wind Data Index Final Report.  National
     Climatic Center, Asheville,  North Carolina, HCO/T1041-01 UC-60.
     December 1978.

18.  Cuscino, T., Jr., G. E. Muleski, and C. Cowherd, Jr.   Iron and Steel
     Plant Open Source  Fugitive Emission Control Evaluation.  EPA-600/2-
     83-110, NTIS No. PB84-1110568.  U.S. Environmental Protection
     Agency, Research Triangle Park, North Carolina.  October 1983.

19.  Bohn, R. R.,  and J.  D. Johnson.  Dust Control on Active Tailings
     Ponds.  Contract No. J0218024.  U.S. Bureau of Mines, Minneapolis,
     Minnesota.  February 1983.
                                3-48

-------
20.  U.S. Environmental Protection Agency.  Control Techniques for
     Participate Emissions From Stationary Sources—Volume 1.
     EPA-450/3-81-005a.  Emission Standards and Engineering Division,
     Research Triangle Park, N.C.  September 1982.

21.  Chepil, N. S., and N. P. Woodruff.  "The Physics of Wind Erosion and
     Its Control."  In Advances in Agronomy, Vol. 15, Academic Press, New
     York.  1963.

22.  Carnes, D., and D. C. Drehmel.  "The Control of Fugitive Emissions
     Using Windscreens."  In Third Symposium on the Transfer and Utilization
     of Particulate Control Technology (March 1981)t  Volume IV, EPA-600/9-82-
     005d, NTIS No. PB83-149617.  April 1982.

23.  Larson, A. G.  Evaluation of Field Test Results on Wind Screen Effi-
     ciency.  Fifth EPA Symposium on Fugitive Emissions:  Measurement and
     Control, Charleston, South Carolina.  May 3-5, 1982.

24.  Westec Services, Inc.  Results of Test Plot Studies at Owens Dry
     Lake, Inyo County, California.  San Diego, California.  March 1984.

25.  Radkey, R. L., and P. B. MacCready.  A Study of the Use of Porous
     Wind Fences to Reduce Particulate Emissions at the Mohave Generating
     Station.  AV-R-9563, AeroVironment, Inc., Pasadena, California.
     1980.

26.  Ohio Environmental Protection Agency.  Reasonably Available Control
     Measures for Fugitive Dust Sources.  Columbus, Ohio.  September
     1980.

27.  U.S. Environmental Protection Agency.  Non-Metallic Processing
     Plants, Background Information for Proposed Standards.  EPA-450/3-
     83-OOla, NTIS No. PB83-258103.  Research Triangle Park, North
     Carolina.  March 1983.

28.  JACA Corporation.  Control of Air Emissions from Process Operations
     in the Rock Crushing Industry.  EPA-340/1-79-002.  U.S.
     Environmental Protection Agency, Washington, D.C., p. 15.  January
     1979.

29.  U.S. Bureau of Mines.  Dust Knockdown Performance of Water Spray
     Nozzles.  Technology News, No. 150.  July 1982.

30.  Courtney, W., and L. Cheng.  Control of Respirable Dust by Improved
     Water Sprays.  Published in Respirable Dust Control Proceedings,
     Bureau of Mines Technology Transfer Seminars, Bureau of Mines
     Information Circular 8753, p.  96.  1978.

31.  Seibel, R.  Dust Control at a Conveyor Transfer Point Using Foam and
     Water Sprays.  Bureau of Mines,  Technical Progress Report 97.
     May 1976.
                               3-49

-------
32.  Cole, H.  Microfoam for the Control of Source and Fugitive Dust
     Emissions.  Paper 81-55.2.  Presented at the 74th Annual  Meeting of
     the Air Pollution Control Association, Philadelphia, Pennsylvania.
     June 1981.

33.  Volkwein, J. C., A. B. Cecala, and E. D. Thimons.  Use of Foam for
     Dust Control in Minerals Processing.  Bureau of Mines RI  8808.
     1983.
                                3-50

-------
                      4.0  DRY SURFACE IMPOUNDMENTS

     Dry surface impoundments are sources of PM10 emissions resulting
from wind erosion of exposed surface material or excavation of waste
residue from the surface.  A hazardous waste lagoon may approach dry
surface conditions if the impounded liquid is pumped out or evaporated.
Unless the lagoon has reached the end of its usefulness, 1t will be
refilled with additional liquid waste material.  Otherwise, the removal
and disposal of dry contaminated bottom material must be dealt with as a
source of PM10 emissions associated with redesign or closure of the
facility.
     At some waste disposal facilities an array of open surface impound-
ments may be used to store and/or treat oily waste for an extended
period.  In these cases, wind erosion tends to be less of a problem
because of the inherent binding effect of organic constituents which
remains until the residue is disturbed.  On the other hand, the removal
of waste residue is an integral part of the waste treatment cycle so that
it occurs with regularity.  Here again, however, some inherent mitigation
of the residue removal process may be expected.

4.1  SOURCE DESCRIPTION
     The removal of residual material from a dry surface impoundment
typically involves a scraping/dozing operation.  The residual material
may be (a) added directly to existing impoundment berms or (b) if a
freeboard problem in the impoundment is anticipated, the material may be
loaded into dump trucks and transferred to an on-site landfill.  The
subject of dust control in landfill operations is addressed in Sec-
tion 5.0.  Dust generating activities associated with the removal of
waste residue from dry surface impoundments are depicted in Figure 4-1.
     The uppermost layer of the low permeability soils (e.g., compacted
clay), which typically are used to line a dry surface impoundment,
                                4-1

-------
            T3 *.
            0> C
            •C 
-------
contains the highest contaminant concentration.  Therefore, particulate
emissions from this layer contribute the highest percentage of
contaminant emissions.

4.2  ESTIMATION OF UNCONTROLLED EMISSIONS
     The two major sources of emissions from a dry surface impoundment
are (a) wind erosion and (b) mechanical disturbance associated with the
removal of waste residue.

4.2.1  Wind Erosion
     With regard to estimating particulate emissions from wind erosion of
exposed surface material, site inspection can be used to determine the
potential for continuous wind erosion.  The two basic requirements for
wind erosion are that the surface be dry and exposed to the wind.
     Nonhomogeneous surfaces impregnated with nonerodible elements
(stones, clumps of vegetation, etc.) are characterized by the finite
availability ("limited reservoir") of erodible material.  Such surfaces
have high threshold wind speeds for wind erosion, and particulate emis-
sion rates tend to decay rapidly during an erosion event.  On the other
hand, bare surfaces of finely divided material such as sandy soil
(uncharacteristic of dry surface impoundments) are represented by an
"unlimited reservoir" of erodible particles.  Such surfaces have low
threshold wind speeds for wind erosion, and particulate emission rates
are relatively time independent at a given wind speed.
     The clay soils that usually line surface impoundments have a
pronounced tendency for crust formation as the surface dries.  Crusted
surfaces are regarded as having a "limited reservoir" of erodible
particles.  Crust thickness and strength should be examined during the
site inspection, by testing with a pocket knife.  If the crust is more
than 0.6 cm thick and not easily crumbled between the fingers (modulus of
rupture > 1 bar), then the soil may be considered nonerodible.  If the
crust thickness is less than 0.6 cm or is easily crumbled, then the
surface should be treated as having a limited reservoir of erodible
particles.  If a crust is found beneath a loose deposit, the amount of
                                4-3

-------
this loose deposit, which constitutes the limited erosion reservoir,
should be carefully estimated.
     In the case of surfaces characterized by a "limited reservoir" of
erodible particles, even the highest mean atmospheric wind speeds are
usually not sufficient to sustain wind erosion.  However, wind gusts may
quickly deplete a substantial portion of the erosion potential.  Because
erosion potential has been found to increase rapidly with increasing wind
speed, estimated emissions should be related to the gusts of highest
magnitude.
     The routinely measured meteorological variable which best reflects
the magnitude of wind gusts is the fastest mile.  This quantity repre-
sents the wind speed corresponding to the whole mile of wind movement
which has passed by the 1-mi contact anemometer in the least amount of
time.  Daily measurements of the fastest mile are presented in the
monthly Local Climatological Data (LCD) summaries.1  The duration of the
fastest mile, typically about 2 min (for a fastest mile of 30 mph),
matches well with the half life of the erosion process, which ranges
between 1 and 4 min.  It should be noted, however, that peak winds can
significantly exceed the daily fastest mile.
     The wind speed profile in the surface boundary layer is found to
follow a logarithmic distribution:

                        u(z). £,!„!-  (*>!„)                  <«-!>

where:    u = wind speed, cm/s
         u* = friction velocity, cm/s
          z = height above test surface, cm
         z0 = roughness height, cm
        0.4 = von (Carman's constant, dimensionless

     The friction velocity (u*) is a measure of wind shear stress on the
erodible surface, as determined from the slope of the  logarithmic
velocity profile.  The roughness height  (ZQ) is a measure of the
roughness of the exposed surface as determined from the y-intercept of
the velocity profile, i.e., the height at which the wind speed is zero.
                                4-4

-------
These parameters are illustrated in Figure 4-2 for a roughness height of
0.1 cm.
     The roughness height, ZQ, which is related to the size and spacing
of surface roughness elements, is needed to convert the friction velocity
to the equivalent wind speed at the typical weather station sensor height
of 7 to 10 m above the surface.  Figure 4-3 depicts the roughness height
scale for various conditions of ground cover.2  The conversion to the 7-m
value is discussed below.
     Emissions generated by wind erosion are also dependent on the
frequency of disturbance of the credible surface because each time that a
surface is disturbed, its erosion potential is restored.  A disturbance
is defined as an action which results in the exposure of fresh surface
material.  On a storage pile, this would occur whenever aggregate mate-
rial is either added to or removed from the old surface.  A disturbance
of an exposed area may also result from the turning of surface material
to a depth exceeding the size of the largest pieces of material present.
     The emission factor for wind-generated particulate emissions from
mixtures of credible and nonerodible surface material subject to
disturbance may be expressed in units of g/m2-yr as follows:
                                             N
                       Emission factor = k  2.   P.                  (4~2)
                                           1  = 1   1
where:     k = particle size multiplier
           N = number of disturbances per year
          P^ = erosion potential corresponding to the observed (or
                  probable) fastest mile of wind for the i-th period
                  between disturbances, g/m*

     The particle size multiplier (k) for Equation 4-2 varies with
aerodynamic particle size, as follows:

          Aerodynamic Particle Size  Multipliers  for  Equation 4-2
                   <30 ym    <15 ym    <10 ym    <2.5 ym
                     1.0       0.6       0.5       0.2
                                4-5

-------
                                                       o

                                                       a.
                                                       u
                                                      •^
                                                       E
                                                       en
                                                       O
                                                       o

                                                       c
                                                       o

                                                      4->
                                                       (0
                                                       s_
                                                      •*->
                                                       I/)
                                                      (\J
                                                       I
                                                      0)
                                                      i-
4-6

-------
        High Rise Buildings.
        (30+Floors)   '
          Suburban
          Medium Buildings-
          (Institutional)
 E
 u
 o
M
I
O
uu
I
UJ
z
X
o
o-
    Suburban
    Residential Dwellings
            Wheat Field
            Plowed Field
                                 Zo (cm)
                                  1000
—800-
—600-
—400-
                                —200—
                                   100
           Natural  Snow
—80.0-
—60.0.
—40.0—
 -20.0-

  10.0
  .8.0-
  -6.0-
  -4.0—
                                  • 2.0-

                                   1.0
                                 -0.8-
                                 -0.6-
                               I—0.4.
                                  -0.2-
                                  0.1
                                          » Urban Area
                                           Woodland Forest
                                           Grassland
  Figure 4-3.  Roughness heights for various surfaces.
                        4-7

-------
     This distribution of particle size within the < 30 ym fraction is
comparable to the distributions reported for other fugitive dust sources
where wind speed is a factor.  This is illustrated,, for example, in the
distributions for batch and continuous drop operations encompassing a
number of test aggregate materials (see AP-42 Section 11.2.3).
     In calculating emission factors, each area of an erodible surface
that is subject to a different frequency of disturbance should be treated
separately. For a surface disturbed daily, N = 365/yr, and for a surface
disturbance once every 6 mo, N = 2/yr.
     The erosion potential function for a dry, exposed surface has the
following form:

            P = 58 (u* - u*)2 + 25 (u* - u*)                        (4-3)
            P = 0 for u* < u*
where:  u* = friction velocity (m/s)
        u* = threshold friction velocity (m/s)
Because of the nonlinear form of the erosion potential function, each
erosion event must be treated separately.
     Equations 4-2 and 4-3 apply only to dry, exposed materials with
limited erosion potential.  The resulting calculation is valid only for a
time period as long or longer than the period between disturbances.
     For uncrusted surfaces, the threshold friction velocity  is best
estimated from the dry aggregate structure of the soil.  The  threshold
friction velocity for erosion can be determined from the mode of the
aggregate size distribution, following a relationship derived by Gillette
(1980) as shown in Figure 4-4.3  A simple hand-sieving test of surface
soil is highly desirable to determine the mode of the surface aggregate
size distribution by inspection of relative sieve catch amounts, follow-
ing the procedure specified in Figure 4-5.
     A more approximate basis for determining threshold friction velocity
would be based on hand sieving with  just one sieve, but otherwise follows
the procedure  specified in Figure 4-5.  Based on the relationship
                                4-8

-------
                                                         o>
                                                         co
                                                         r-
                                                         CO

                                                         in
                                                         CO
                                                         CM
                                                         O>
                                                         00
                                                         to
                                                         in
                                                         m
                                                         CM
                                                         O)
                                                         eo
                                                         co
                                                         n
                                                         CM
                                                            O
                                                            O
                                                                 CD
                                                                •o
                                                                 O
                                      c
                                      O
                                      (ft
                                     5

                                      Q>
                                      N
                                     CO
                                      (0
                                      O)
                                      CD

                                      O)
                                      O)
  cneot«»co in
              n
                             o>oor>.cD in
                                          co
                                              CVJ
O
O
O
O
O
                                              O)
                                              •o
                                              c
                                              o
                                                                         1-
                                                                         •M
                                              •o

                                              QJ

                                              •i—
                                              

c
o
•r~
4->
O
•r—
S-


T3

'o
                                                                         
                                                                         re
                                                                         O)
                                                                         s_
(oas/uio)  )  -n  'Aj!00|9A uoi;ouj
                                 4-9

-------
     FIELD  PROCEDURE FOR  DETERMINATION OF THRESHOLD  FRICTION  VELOCITY*

1.  PREPARE A NEST OF SIEVES WITH THE FOLLOWING OPENINGS:  4 mm, 2 mm,
    1mm, 0.5mm, 0.25 mm.  PLACE A COLLECTOR PAN BELOW THE BOTTOM SIEVE
    (0.25-mm OPENING).

2.  COLLECT A SAMPLE REPRESENTING THE SURFACE LAYER OF LOOSE PARTICLES
    (APPROXIMATELY 1 cm IN DEPTH FOR AN UNCRUSTED SURFACE), REMOVING ANY
    ROCKS LARGER THAN ABOUT 1 cm IN AVERAGE PHYSICAL DIAMETER.  THE AREA
    TO BE SAMPLED SHOULD  NOT BE LESS THAN 30 cm x 30 cm.

3.  POUR THE SAMPLE INTO  THE TOP SIEVE (4-rom OPENING), AND PLACE A LID ON
    THE TOP.

4.  ROTATE THE COVERED SIEVE/PAN UNIT BY HAND USING BROAD SWEEPING ARM MO-
    TIONS IN THE HORIZONAL PLANE.  COMPLETE 20 ROTATIONS AT A SPEED JUST
    NECESSARY TO ACHIEVE  SOME RELATIVE HORIZONTAL MOTION BETWEEN THE SIEVE
    AND THE PARTICLES.

5.  INSPECT THE RELATIVE  QUANTITIES OF CATCH WITHIN EACH SIEVE AND
    DETERMINE WHERE THE MODE IN THE AGGREGATE SIZE DISTRIBUTION LIES,
    I.E., BETWEEN THE OPENING SIZE OF THE SIEVE WITH THE LARGEST CATCH AND
    THE OPENING SIZE OF THE NEXT LARGEST SIEVE.
*AOAPTED FROM A LABORATORY PROCEDURE PUBLISHED BY W.  S.  CHEPIL (1952).
                                Figure 4-5
                                    4-10

-------
developed by Bisal and Ferguson (1970), 1f more than 60 percent of the
soil passes a 1-mm sieve, the "unlimited reservoir" model will apply; if
not, the "limited reservoir" model will apply.s  This relationship has
been verified by Gillette (1980) on desert soils.3
     If the soil contains nonerodible elements which are too large to
include in the sieving (I.e., greater than about 1 cm in diameter), the
effect of these elements must be taken into account by increasing the
threshold friction velocity.  Marshall (1971) has employed wind tunnel
studies to quantify the increase in the threshold velocity for differing
kinds of nonerodible elements.6  His results are depicted in terms of a
graph of the rate of corrected to uncorrected friction velocity versus LC
(Figure 4-6), where Lc is the ratio of the silhouette area of the rough-
ness elements to the total area of the bare loose soil.  The silhouette
area of a nonerodible element is the projected frontal area normal to the
wind direction.
     A value for l_c is obtained by marking off a 1-m x 1-m surface area
and determining the fraction of area, as viewed from directly overhead,
that is occupied by nonerodible elements.  Then the overhead area should
be corrected to the equivalent frontal area; for example, if a spherical
nonerodible element is half embedded in the surface, the frontal area is
one-half of the overhead area.  Although it is difficult to estimate LC
for values below 0.05, the correction-to-friction velocity becomes less
sensitive to the estimated value of l_c.
     The difficulty in estimating l_c also increases for small nonerodible
elements.  However, because small nonerodible elements are more likely to
be evenly distributed over the surface, it is usually acceptable to
examine a smaller surface area, e.g., 30 cm x 30 cm.
     Threshold friction velocities for several surface types have been
determined by field measurements with a portable wind tunnel.  These
values are presented in Tables 4-1 and 4-2 and Figure 4-7.
     The fastest mile of wind for the periods between disturbances may be
obtained from the monthly LCD summaries for the nearest reporting weather
station that is representative of the site in question, i  These summaries
report actual fastest mile values for each day of a given month.  Because
the erosion potential is a highly nonlinear function of the fastest mile,
mean values of the fastest mile are inappropriate.
                               4-11

-------
o>

CO

N.

CO


U)
n
CM
CD
oo
CD

in
                                                                            2
                                                                             OJ
                                                                             O
 U
•f—
 J-
CO
CM
                                                                             (U
O)

00
 o>
 eor«-«DiO'j-
                               4-12

-------
                  TABLE 4-1.  THRESHOLD  FRICTION  VELOCITIES



Material
Overburden*
Scoria (roadbed
material)*
Ground coal
(surrounding coal
pile)
Uncrusted coal pile
Scraper tracks on
coal pilea*°
Fine coal dust on
concrete pad
Threshold
friction
velocity
(m/s)
1.02
1.33

0.55


1.12
0.62

0.54


Roughness
height
(cm)
0.3
0.3

0.01


0.3
0.06

0.2




Threshold wind
velocity at 10
20 » Actual z0
21
27

16


23
15

11

m (m/s)
* 0.5 cm
19
25

10


21
12

10


Ref.
8
8

8


8
8

9

aWestern surface coal mine (Reference 7).
bLightly crusted.

cEastern power plant (Reference 8).


           TABLE 4-2.  THRESHOLD FRICTION VELOCITIES—ARIZONA SITES
Location
Threshold
 friction
 velocity
  (m/s)
             Roughness
              height
               (cm)
                Threshold
              wind  velocity
                 at 10 m
                  (m/s)
Mesa - Agricultural site
Glendale - Construction site
Maricopa - Agricultural site
Yuma - Disturbed desert
Yuma - Agricultural site
Algodones - Dune flats
Yuma - Scrub desert
Santa Cruz River, Tucson
Tucson - Construction site
Ajo - Mine tailings
Hayden - Mine tailings
Salt River, Mesa
Casa Grande - Abandoned
  agricultural land
  0.57
  0.53
   .58
   ,32
   .58
   .62
   .39
   ,18
  0.25
  0.23
  0.17
  0.22
  0.25
0.
0.
0.
0.
0.
0,
0.0331
0.0301
0.1255
0.0731
0.0224
0.0166
0.0163
0.0204
0.0181
0.0176
0.0141
0.0100
0.0067
16
15
14
 8
17
18
11
 5
 7
 7
 5
 7
 8
Source:  Reference 9.
                                     4-13

-------
i i
1 1
                             4
                             1
                             1
     •4!
                                                              vt
                                                              0)
                                                              >
                                                              o

                                                              *J
                                                              u
           §
ed materials only
0
a T3
N O
« E
0 C
a °
rat:
S£
< «
<5
i 1 1 i
"i"
E,
1 1 1 1 1 1 1 1 1 1 1 I t
oo i^ 
                                                               u
                                                              C/)
                                                               (U


                                                               01
                                4-14

-------
The anemometer heights of reporting weather stations are found in
Reference 10, and wind speeds should be corrected to a 10 m reference
height using Eq. 4-1.
     To convert the fastest mile of wind (u+) from a reference anemometer
height of 10 m to the equivalent friction velocity (u*), the logarithmic
wind speed profile may be used to yield the following equation:

                              u* = 0.053 uto                         <4-4)

where:      u* = friction velocity (m/s)
          uto = fastest mile of wind from reference anemometer (m/s)

This assumes a typical roughness height of 0.5 cm for open terrain.
Equation 4-4 is restricted to large relatively flat piles or exposed
areas with little penetration into the surface wind layer.
     Implementation of the above procedure is carried out in the
following steps:
     1.   Determine threshold friction velocity for erodible material of
          interest (see Tables 4-1 and 4-2 or use Figure 4-4 to determine
          the mode of the aggregate size distribution).
     2.   Divide the exposed surface area into subareas of constant
          frequency of disturbance (N).
     3.   Tabulate fastest mile values (u+) for each frequency of
          disturbance and correct them from the anemometer height  (z) to
          10 m (u"J"0) using a variation of Equation 4-1, as follows:

                          . +     + ln(10/0.005)                      (4-5)
                          ui° " u  ln(z/0.005)

          where a typical roughness height of 0.5 cm (0.005 m) has been
          used.
     4.   Convert fastest mile values (uto) to equivalent friction
          velocities (u*), taking into account uniform wind exposure of
          nonelevated surfaces, using Equation 4-4.
     5.   Treating each subarea (of constant N and u*) as a separate
          source, calculate the erosion potential (P^) for each period

                               4-15

-------
          between disturbances using Equation 4-3 and the emission factor
          using Equation 4-2.
     6.   Multiply the resulting emission factor for each subarea by the
          size of the subarea, and add the emission contributions of all
          subareas. Note that the highest 24-h emissions would be
          expected to occur on the windiest day of the year.  Maximum
          emissions are calculated assuming a single event with the
          highest fastest mile value for the annual period.
     The recommended emission factor equation presented above assumes
that all of the erosion potential corresponding to the fastest mile of
wind is lost during the period between disturbances.  Because the fastest
mile event typically lasts only about 2 min, which corresponds roughly to
the half-life for the decay of actual erosion potential, it could be
argued that the emission factor overestimates partial!ate emissions.
However, there are other aspects of the wind erosion process which offset
this apparent conservatism:
     1.   The fastest mile event contains peak winds which substantially
          exceed the mean value for the event.
     2.   Whenever the fastest mile event occurs, there are usually a
          number of periods of slightly lower mean wind speed which
          contain peak gusts of the same order as the fastest mile wind
          speed.
     Of greater concern is the likelihood of overprediction of wind
erosion emissions  in the case of surfaces disturbed infrequently  in
comparison to the  rate of crust formation.

4.2.2  Removal of  Surface Material
     At present, the only emission factor available in AP-42 for
construction-related activities  is 1.2 tons/acre/month (related to
particles < 30 \an  Stokes1 diameter).  However, PM10 emission factors have
been developed for construction  site  preparation using test data  from a
study  conducted  in Minnesota  for topsoil removal, earthmoving  (cut-and-
fill),  and truck haulage operations.  For these operations, the PM10
emission  factors based on the  level of vehicle activity  (i.e., vehicle
kilometers traveled  or VKT)  occurring on-site are:11
                                4-16

-------
          Topsoil removal:    5.7 kg/VKT for pan scrapers
     •    Earthmoving:        1.2 kg/VKT for pan scrapers
          Truck haulage:      2.8 kg/VKT for haul trucks
            (uncompacted
            surfaces)
     The following equation is recommended for estimating emissions from
transfer operations (batch or continuous drop) involving aggregate
materials:
                                 /U \
                    e = k(0.0016)^2i1_ (kg/Mg)
                                  /M\ K4
                                  \2/
                                                                    (4-6)

                                 /U\ 1'3
                    e = k(0.0032)^ i A  (Ib/ton)
where:  e = emission factor
        k = particle size multiplier (dimensionless)
        U = mean wind speed, m/s (mph)
        M = material moisture content, %

The particle size multiplier k varies with aerodynamic particle diameter
as shown below:

                 Aerodynamic Particle Size Multiplier, k
     <30 ym        <15 ym        <10 urn        <5 ym          <2.5 ym
       0.74         0.48          0.35          0.20            0.11

Truck dumping or loading a truck with a front-end loader are examples of
batch drop operations.  Adding material to a pile by a conveyor stacker
is an example of a continuous drop operation.
     Procedures for estimating emissions from vehicle traffic on paved
roads and compacted unpaved roads are presented separately in Section 2
of this document.

                               4-17

-------
4.2.3  Level of Contamination
     The data presented 1n Reference 12 suggest several points that may
be useful in specifying the level of contamination (a) for dry surface
impoundments.  These include:
     •    The question of whether or not a given organic compound is
          present and if so at what levels is site-specific.  This sug-
          gests that source-specific S&A would, in fact, be needed to
          characterize a levels for organic compounds.
     •    If a source-specific S&A plan is being developed (see Appen-
          dix 0 for generic procedures), the data in Reference 12 indi-
          cate that organic compounds with lower vapor pressures tend to
          be found more often than compounds with high vapor pressures
          (i.e., volatile compounds). As a result then, the S&A plan
          should be directed toward organic compounds with relatively low
          vapor pressures (< 10 mm Hg).
     •    As an alternative to source-specific S&A, a "worst-case"
          scenario could be developed based on knowledge of the waste
          streams historically placed in the surface.  Applying a "safety
          factor" of 5 or 10 may be warranted.  In other words, multiply
          the upper end values by 5 or 10 as a conservatively high esti-
          mate of a.

4.3  DEMONSTRATED CONTROL TECHNIQUES
     The control techniques applicable to dry surface impoundments fall
into distinct categories as related to materials handling operations
(including  removal of waste residue) and wind erosion.  In both cases,
the control  can be achieved by (a) source extent reduction, (b) source
improvement related to work practices and transfer equipment, and
(c) surface treatment.  These control options are summarized in
Table  4-3.   The efficiency of these controls ties back to the emission
factor relationships presented earlier in this section.
      In most cases, good work practices which confine freshly exposed
material provide substantial opportunities for emission reduction without
the need for investment  in a control application program.  This statement
applies to  areas around  the  impoundment as well as the  impoundment

                               4-18

-------
Itself.  In particular, spillage of material during removal of waste
residue can add a large source component associated with traffic-
entrained dust.  Emission inventory calculations show, in fact, that the
traffic dust component may easily dominate emissions from transfer of
material and wind erosion.  The prevention of spillage and subsequent
spreading of material by vehicle tracking is essential to cost-effective
emission control.  If spillage cannot be prevented because of the need
for intense use of mobile equipment in the surface impoundment area, then
regular cleanup should be employed as a necessary mitigative measure.
     Wind erosion control of soil surfaces 1s accomplished by stabilizing
credible particles.  The stabilization process is accomplished 1n three
major successive stages:  (a) trapping of moving soil particles,
(b) consolidation and aggregation of trapped soil particles, and
(c) stabilization of the surface.13
     The trapping of eroding soil is termed "stilling" of erosion.  This
may be effected by roughening the surface, by placing barriers in the
path of the wind, or by burying the erodible particles during tillage.
Trapping is accomplished naturally by soil crusting resulting from
rain.  It should be stressed that the stilling of erosion is only
temporary; to effect a permanent control, surface stabilization must be
established and maintained.
     In bare soils containing a mixture of erodible and nonerodible
fractions, the quantity of soil eroded by the wind is limited by the
height and number of nonerodible particles that become exposed on the
surface.  The removal of erodible particles continues until the height of
the nonerodible particles that serve as barriers to the wind is increased
to a degree that affords complete shelter to the erodible fractions.  If
the nonerodible barriers are small, such as fine gravel, a relatively
large number of pieces are needed for protection of soil from wind
erosion.  The gravel in such a case would protect the erodible portion
more by covering than by sheltering from the wind.  Thus all nonerodible
materials on the ground that control erosion have an element of cover in
addition to the barrier principle which protects the soil.
                               4-19

-------
       TABLE 4-3.   CONTROL TECHNIQUES FOR DRY SURFACE IMPOUNDMENTS

    Removal  of waste residue
      Source extent reduction        Mass transfer reduction
      Source improvement             Drop height reduction
                                     Wind sheltering
                                     Moisture retention
      Surface treatment              Wet suppression
    Wind erosion
      Source extent reduction        Disturbed area reduction
                                     Disturbance frequency reduction
                                     Spillage cleanup
      Source improvement             Spillage reduction
                                     Wind exposure reduction
      Surface treatment              Wet suppression
                                     Chemical stabilization

     The above principles extend to almost all elements used in wind
erosion control.  All of these control methods are designed to (a) take
up some or all of the wind force so that only the residual force, if any,
is taken up by the credible soil fractions; and (b) trap the eroded soil,
if any, on the lee side or among surface roughness elements or barriers,
thereby reducing soil avalanching and intensity of erosion.
     Preventative methods which change the properties or exposure of
transfer streams or surface materials are discussed in the following
section.

4.4  CONTROL PERFORMANCE ESTIMATION
     Preventive methods for control of erissions from dry surface im-
poundments  typically consist of physical stabilization, chemical stabil-
ization, and wind  fences/barriers.
      In controlling emissions from the removal of dry waste residue, two
operations  must be  considered:  the movement of vehicles over the
surface; and  the excavation of  the residue itself.  Watering and wet
                               4-20

-------
suppression, respectively, are possible control measures for these two
activities.
     However, use of water-based dust suppression systems may result in
unacceptable increases in contaminated leachate production.  Thus, before
instituting a watering program, the facility must demonstrate that
addition of water to the waste residue does not produce undesirable
impacts on ground and surface water quality.
     In the sections that follow, these methods are discussed with
respect to their characteristics and effectiveness in controlling fugi-
tive particulate emissions from dry surface impoundments.

4.4.1  Physical Stabilization
     Physical stabilization by covering the exposed surface with less
credible aggregate material and/or vegetative stabilization are also
practical control methods with respect to surface impoundment closure
operations.
     The effectiveness of control is evaluated in terms of the increase
in the threshold friction velocity provided by the armoring aggregate or
vegetative cover.  Data on threshold friction velocities for various
aggregate materials are presented in Section 4.2.

4.4.2  Chemical Stabilization
     A portable wind tunnel has been used to measure the control of coal
surface wind erosion emissions by a 17% solution of Coherex® in water
applied at an intensity of 3.4 L/m2 (0.74 gal/yd2), and a 2.8% solution
of Dow Chemical M-167 Latex Binder in water applied at an average
intensity of 6.8 L/m2 (1.5 gal/ydz).^  The control efficiency of
Coherex® applied at the above intensity to an undisturbed steam coal
surface approximately 60 d before the test, under a wind of 15.0 m/s
(33.8 mph) at 15.2 cm (6 in) above the ground, was 89.6% for TP and
approximately 62% for IP and FP.  The control efficiency of the latex
binder on a low volatility coking coal is shown in Figure 4-8.
                               4-21

-------
       100
       80
       60
     o
    LU
    o 40
     o
    o
       20
                                        6.8 ^/m2(i.5gal/yd2)of
                                        2.8% Solution in Water
Tunnel Wind
Speed » 17 mis (38 mph)
at 15 cm (6.0 in)
Above the Test Surface

Key:
                                  1
    Figure 4-8.
       1234
          Time After Application (Days)
   Decay in control efficiency of latex binder applied
          to coal storage piles.11*
     Cost elements for chemical stabilization are presented in Table 4-4.
The cost of a system for application of surface crusting chemicals to
storage piles is $18,400 for the initial capital cost and $0.006 to
$0.Oil/ft2 for annual operating expenses based on April 1985 dollars.^
Tables 4-5 and 4-6 provide recordkeeping forms for application of chemi-
cal dust suppressants.

4.4.3  Wind Fences/Barriers
     Wind fences/barriers are an effective means by which to control
fugitive particulate emissions from open dust sources.  The principle of
the wind fence/barrier is to provide an area of reduced wind velocity
which allows settling of the large particles (which cause saltation) and
reduces the particle flux from the exposed surface on the leeward side of
the fence/barrier.  They can be used to provide a sheltered area for
materials handling operations to reduce entrainment during load-in/load-
out, etc.
                               4-22

-------
       TABLE 4-4.  CAPITAL AND O&M ITEMS FOR CHEMICAL STABILIZATION
                           OF OPEN AREA SOURCES
Capital equipment

  •  Storage equipment
       Tanks
       Rail cars
       Pumps
       Piping

  •  Application equipment
       Trucks
       Spray system
       Piping (including winterizing)

O&M expenditures

  •  Utility or fuel costs
       Water
       Electricity
       Gasoline or diesel fuel

  •  Supplies
       Chemicals
       Repair parts

  •  Labor
       Application time
       Road conditioning
       System maintenance
                                 4-23

-------
I
o
a.
ca
           •t-

           t.
             t-o
             o
          — in
           3 3
           er
             10
             4)
             10
        o  >.
       — -KN
       •t- — -o
        ID  m >.
        OCX
       —  4) _
       — +- 10
        a.  c oi
        a.—
           •3 •>-

          — (0
          •*• ID
          O  J
          0 E
          a u
             a
            a
                                                        4-24

-------

  C
         ._  Q
         —  cn
         4)  IO
         +-  >
          c —
          (0 —
         —  «)
         10  L.
         ^O  4)


         W —
             54)
            •O
             10
            c
O)
ia
                                                                                                                                                    O
                                                                                                                                                    -*-
                                                                                                                                                    in
                                                                                                                                                    •o

                                                                                                                                                    8
                                                                                                                                                    IO
                                                                                                                                                    b
                                                                                                                                                     )
                                                                                                                                                    U
                                                                                                                                                    4)
                                                                                                                                                    o
                                                                                                                                                    Q.
 X.


 41
4-
 K3
                                                                                                                                                    •o
                                                                                                                                                    4)
                                                                                                                                                    4!
                                                                                                                                                    J3
Q.
a
                                                                                                                                                    4)
                                                                                                                                                    £
                                                                                                                                                    Z

                                                                                                                                                    
-------
     Wind fences/barriers can either be man-made structures or vegetative
in nature.  Wind screens or fences reduce the mechanical turbulence
generated by ambient winds in an area the length of which is many times
the physical height of the fence.  Fences and screens can be portable and
thus capable of being moved around the site, as needed.
     A number of studies have attempted to determine the effectiveness of
wind fences/barriers for the control of windblown dust under field
conditions.  Several of these studies have shown both a significant
decrease in wind velocity as well as an increase in sand dune growth on
the lee side of the fence.16~ia  The degree of emissions reduction varied
substantially from study to study depending on test conditions.17*19
     Control Efficiency.  The control efficiency of wind fences is
dependent on the physical dimensions of the fence relative to the source
being controlled.  In general, a porosity (i.e., percent open area) of
50 percent seems to be optimum for most applications.  According to a
recent field study of soil storage piles, a screen length of five times
the pile diameter, a screen-to-pile distance of twice the pile height,
and a screen height equal to the pile height was found best.20  Also, a
1988 wind tunnel study of windbreak effectiveness for coal storage piles
showed 49% to 97% control efficiencies for a 50% porosity windbreak with
height equal to the pile height and length equal to the pile base.21  The
windbreak was located three pile heights upwind from the base of the
pile.
     Control Costs.  As  stated above, one of the real advantages of wind
fences for  the control of PM10 involves the  low capital and operating
costs.   These involve the following basic elements:
           Capital equipment:
               Fence material and  supports
               Mounting  hardware
      •     Operating  and  maintenance expenditures:
               Replacement fence material and  hardware
               Maintenance labor
      The following  cost  estimates  (in  1980 dollars) were developed for
wind  screens  applied  to  aggregate  storage piles:22
                                4-26

-------
          Artificial wind guards:
          —   Initial capital cost * $12,000 to $61,000
     •    Vegetative wind breaks:
               Initial capital cost * $45 to $425 per tree
     Due to the lack of quantitative data on costs associated with wind
screens, it is recommended that local vendors be contacted to obtain more
detailed data for capital and operating expenses.  Also, since wind
fences and screens are relatively "low tech" controls, it may be possible
for the site operator to construct the necessary equipment using site
personnel with less expense.
     Compliance Issues.  As with other options mentioned above, the main
regulatory approach involved with wind fences and screens would involve
recordkeeplng by the site operator.  Parameters to be specified in the
dust control plan and routinely recorded are:
     General Information to be Specified in Plan
     1.   Locations of all materials handling operations to be controlled
          with wind fences referenced on a plot plan available to the
          site operator and regulatory personnel
     2.   Physical dimensions of each source to be controlled and con-
          figuration of each fence or screen to be installed
     3.   Physical characteristics of material to be handled or stored
          for each operation to be controlled by fence(s) or screen(s)
     4.   Applicable prevailing meteorological data (e.g., wind speed and
          direction) for site on an annual basis
     Specific Operational Records
     1.   Date of installation of wind fence or screen and initials of
          installer
     2.   Location of installation relative to source and prevailing
          winds
     3.   Type of material being handled and stored and physical
          dimensions of source controlled
     4.   Date of removal of wind fence or screen and initials of
          personnel involved
                               4-27

-------
     General Records to be Kept
     1.   Fence or screen maintenance record
     2.   Log of meteorological conditions for each day of site operation

4.4.4  Watering of Unpaved Surfaces
     Watering of unpaved surfaces is one form of wet dust suppression.
This technique prevents (or suppresses) the fine particulate from leaving
the surface and becoming airborne through the action of mechanical
disturbance or wind.  The water acts to bind the small particles to the
larger material thus reducing emissions potential.
     Control Efficiency:  The control efficiency of watering of unpaved
surfaces is a direct function of the amount of water applied per unit
surface area (liters per square meter), the frequency of application
(time between reapplication), the volume of traffic traveling over the
surface between applications, and prevailing meteorological conditions
(e.g., wind speed, temperature, etc.).  As stated previously, a number of
studies have been conducted with regard to the efficiency of watering to
control dust, but few have quantified all parameters listed above.
     The only specific control efficiency data which are available for
vehicle travel on uncompacted surfaces involve the use of watering to
control truck haulage emissions for a road construction project in
Minnesota.23  Using the geometric means of the important source charac-
teristics  (i.e., silt content, traffic volume, and surface moisture) and
the regression equation developed from the downwind concentration data, a
PM10 control efficiency of approximately 50 percent was obtained for a
water  application intensity of approximately 0.2 gal/yd2/h.
     For more compacted unpaved surfaces, an empirical model for the
performance of watering as a control technique has been developed.  The
supporting data base consists of 14 tests performed in four states during
five different summer and fall months.  The model  is:15

                           c .  100  - °-8  P d t                      (4"7)

where      C =  average control  efficiency, in percent
           p =  potential  average  hourly daytime evaporation rate in mm/h

                                4-28

-------
          d = average hourly daytime traffic rate in vehicles per hour
          1 = application intensity in L/m2
          t = time between applications in h

The term p in the above equation is determined using Figure 4-9 and the
relationship:

         <• 0.0049 e (annual average)                               (4-8a)
           0.0065 e (worst case)                                   (4-8b)

where     p = potential average hourly daytime evaporation rate (mrn/h)
          e = mean annual pan evaporation (inches) from Figure 4-9

Note that no data are available for Alaska and Hawaii in Figure 4-9.
Readers responsible for those portions of the country should consult
local meteorological resources (e.g., state universities, local weather
stations, etc.).
     An alternative approach (which is potentially suitable for a
regulatory format) is shown as Figure 4-10.  Figure 4-10 shows that,
between the average uncontrolled moisture content and a value of twice
that, a small increase in moisture content results in a large increase in
control efficiency.  Beyond this point, control efficiency grows slowly
with increased moisture content.  Furthermore, this relationship is
applicable to all size ranges considered:

           75 (M-l)     1 < M < 2
     c = {                                                          (4-9)
           62 + 6.7 M   2 < M < 5

where     c = instantaneous control efficiency in percent
          M = ratio of controlled to uncontrolled surface moisture
              contents
                               4-29

-------
                                                                18

                                                                «O

                                                               •5

                                                                  c

                                                               VI •"3

                                                               O s
                                                               3   »
                                                               a> i«
                                                               •r"  <1J

                                                               ^K  (0
                                                               O -M
                                                               0 01

                                                               CD "O
                                                               f  O)

                                                                  f—
                                                               J-  C
                                                               O =3

                                                                   (U
                                                               
                                                               *
                                                               O
                                                               o
                                                               a. u
                                                               «o •»-
                                                               > •(->
                                                               ai re
                                                                  E
                                                               r— f-
                                                               «o •—
4-30

-------
    100%
o
c
.2
'o
m   75%	
LLJ

"2
++

O
O
     50% -
Q.
(0

O

C

c
(Q
"S5
     25% -
      0%
                                                        95%
              Ratio of Controlled to Uncontrolled
                  Surface Moisture Contents
      Figure 4-10.  Watering control efficiency effectiveness
            for scraper travel on unpaved surf aces. 21»
                         4-31

-------
     Control Costs.  Costs for watering programs include the following
elements:
     •  Capital:  Purchase of truck or other device
     •  O&M:  Fuel, water, truck maintenance, operator labor
Reference 15 estimates the following costs (1985 dollars):
     Capital:  $17,100/truck
     O&M:  $32,900/truck
The number of trucks required may be estimated by assuming that a single
truck, applying water at 1 L/m2, can treat roughly 4 acres of unpaved
surface every hour.
     Compliance Issues.  Verification of the effectiveness of a watering
program would ideally consist of two complementary approaches.  The first
facet would require the owner to maintain adequate records that would
document to agency personnel's satisfaction that a regular program is in
place.  The second approach would involve agency spot checks of con-
trolled surfaces by taking material grab samples for moisture analysis.
     Records must be kept that document the frequency of water applica-
tion to unpaved surfaces.  Pertinent parameters to be specified in a
control plan and rigorously recorded include:
     General Information to be Specified.
     1.   All areas to be treated referenced on a plot plan available to
          both the site operator and regulatory personnel
     2.   Application  intensity (gal/sq yd) and frequency (a minimum
          moisture content may be specified as an alternative)
     3.   Type of application vehicle, capacity of tank, and source of
          water
     Specific Records  to be Kept by Truck Operator
      1.   Date and time of treatment
     2.   Equipment used  (this should be referred back to dust control
          plan specifications)
     3.   Operator's  initials  (a separate operators  log may be kept and
          transferred  later to permanent records by  site operator)
     4.   Start and stop time, average speed,  and number passes
      5.   Start and stop time  for filling of water tank
                                4-32

-------
     Specific Records to be Kept by Site Operator
     1.   Equipment maintenance logs
     2.   Meteorological log of general conditions (e.g., sunny and warm
          vs. cloudy and cold)
     3.   Records of equipment breakdowns and downtime
An example permanent record form which may be used to record the above
information is shown in Figure 4-11.  In addition to the above, some
regulatory formats require that records of surface sample moisture
content also be kept.

4.4.5  Wet Suppression for Materials Handling
     Control Efficiency.  Wet suppression of materials handling opera-
tions is similar to that used for unpaved surfaces.  However, in addition
to plain water this technique can also use water plus a chemical surfac-
tant or micronized foam to control fugitive PM10.
     Surfactants added to the water supply allow particles to more easily
penetrate the water droplet and increase the total number of droplets,
thus increasing total surface area and contact potential.  Foam is gener-
ated by adding a chemical (i.e., detergent-like substance) to a rela-
tively small quantity of water which is then vigorously mixed to produce
small bubble, high energy foam in the 100 to 200-ym size range.  The foam
uses very little liquid volume, and when applied to the surface of the
bulk material, wets the fines more effectively than untreated water.
     As with watering of unpaved surfaces, the control efficiency of wet
suppression for materials storage and handling is dependent on the same
basic application parameters.  These include:  the amount of water, water
plus surfactant, or foam applied per unit mass or surface area of mate-
rial handled (i.e., liters per metric ton or square meter); if not
continuous, the time between reapplications; the amount of surfactant
added to the water (i.e., dilution ratio), if any; the method of appli-
cation including the number and types of spray nozzles used; and
applicable meteorological conditions occurring onsite.
     Wet suppression can be applied to material  handling operations by a
variety of methods depending on the material and how it is being
handled.  If aggregates are batch transferred using loaders or by truck
dumping, water (with or without chemicals) could be applied with a water
                               4-33

-------

















§

2.
Q
_i
Q.
Q.
<
OC
UJ
1



















m
I









fm
. a
8.1
<§•-

I'
Is
'5 D
cr
LU


"o "5
8 ji
«1 =
C 2 UJ
I*J
O.
£ £
to E
E « -^
•j= 1- LL
"I1-
^V

"O

w
o •
U 
£ (0
O 0.
<4- C
33
*io
O ••
•^ (/>
O. (LI
>> 0
t— i.
3
0
• 4— '
y^^
tj"
o;
3
O)
•r-
u_




4-34

-------
cannon or spray bar to the material prior to or during load-in or load-
out.  Foam may be a good alternative in such instances when the material
is handled repeatably over the period of a day.  Foam can be applied once
in the handling process (e.g., as it 1s Initially loaded into trucks) and
the binding action of the bubbles will carry through subsequent handling
operations.
     Control of transfer operations can also be augmented using portable
wind fences to provide a wind break to reduce dust generation and improve
application of water to the load during transfer to haul trucks.  Wind
fences are discussed later in this discussion.
     Available control efficiency data for wet dust suppression for
materials handling and storage are limited.1*  For suppression using
plain water, the most applicable efficiency information available is for
feeder to belt transfer of coal in mining operations.  Control effi-
ciencies of 56 to 81 percent are reported for respirable particulate
(particles < - 3.5 ymA) at application intensities of 6.7 to 7.1 L/Mg
(1.6 to 1.7 gal/ton), respectively.  Assuming that respirable particulate
is essentially equivalent to PM10, the above control efficiencies may be
representative of similar controls for waste material handling.  (The
above application intensities were estimated assuming 5 min to discharge
7 Mg of coal and 1.4 L/min/spray nozzle.)
     In the case of foam suppression, the most appropriate data available
are for the transfer of sand from a grizzly.  Using the respirable
particulate control efficiencies at various foam application intensities
(and assuming respirable particulate is equivalent to PM10), the
following equation was developed by simple linear regression of the data
compiled by Cowherd and Kinsey:15

                         C = 8.51 + 7.96 (A)                       (4-10)

where:    C = PMlo control efficiency in percent
          A = application intensity in ft3 foam/ton of material

A coefficient of determination (r*) of 99.97 percent was obtained for the
above equation based on the three data sets used in its derivation.
                               4-35

-------
     An alternate approach (which is potentially suitable for regulatory
formats) involves the use of Eq. 4-6.  By determining the moisture
content of the material before and after wet suppression, the control
efficiency can be determined by:

                         CE = 100(eu-ec)/eu                         (4-11)

where     CE = PM10 control efficiency in percent
          eu * "uncontrolled" PM10 emission factor from Eq. 4-6
          ec = "controlled" PM10 emission factor from Eq. 4-6

The above calculations would necessitate the determination of the
moisture content of the material.  This could be accomplished by taking
grab samples of the material before and after application of the wet
suppression technique being employed and analyzing for moisture content.
     Control Costs.  Costs associated with wet suppression systems  are
presented in Table 4-7.  Reference 15 estimates the following costs (in
1985 dollars):
     •    Regular watering of storage piles:
               Initial capital cost = $18,400 per system
     •    Watering of exposed areas:
               Initial capital cost = $1,053 per acre
               Annual operating cost = $25 to $67 per acre
     The costs associated with a wet suppression system using chemical
surfactants for the unloading of limestone from trucks at aggregate
processing plants  (in 1980 dollars) have been estimated at:  capital =
$72,000; annual = $26,000.  These costs are based on a stationary system
and may not be indicative of those used at construction and demolition
sites.22
     Compliance  Issues.  As with watering of unpavecl surfaces, verifica-
tion of the effectiveness of a wet suppression control program would
consist of two complementary approaches.  The first would be record
keeping to document that the program is being implemented and the other
would  be  spot-checks  and grab sampling.  Both were discussed previously
above.
                                4-36

-------
      TABLE 4-7.  WET SUPPRESSION SYSTEM CAPITAL AND O&M
                         COST  ELEMENTS
Capital equipment

  •  Water spray system
       Supply pumps
       Nozzles
       Piping (Including wlnterlzation)
       Control system
       Filtering units

  •  Water/surfactant and foam systems only
       Air compressor
       Mixing tank
       Metering or proportioning unit
       Surfactant storage area

O&M expenditures

  •  Utility costs
       Water
       Electricity

  •  Supplies
       Surfactant
       Screens

  •  Labor
       Maintenance
       Operation
                            4-37

-------
     Records must be kept that document the control plan and its imple-
mentation.  Pertinent parameters to be specified in a plan and to be
regularly recorded include:
     General Information to be Specified in Plan
     1.   Locations of all materials handling operations referenced on
          plot plan of the site available to the site operator and
          regulatory personnel
     2.   The method and application intensity of water, etc, to be
          applied to the various materials and frequency of application,
          if not continuous
     3.   Dilution ratio for chemicals added to water supply, if any
     4.   Complete specifications of equipment used to handle the various
          materials and for wet suppression
     5.   Source of water and chemical(s), if used
     Specific Operational Records
     1.   Date of operation and operator's initials
     2.   Start and stop time of wet suppression equipment
     3.   Location of wet suppression equipment
     4.   Type of material being handled and number of loads (or other
          measure of throughput) loaded/unloaded between start and stop
          time (if material is being pushed, estimate the volume or
          weight)
     5.   Start and stop times for tank filling
     General Records to be Kept
     1.   Equipment maintenance records
     2.   Meteorological log of general conditions
     3.   Records of equipment malfunctions and downtime
     In  addition to the above, some of the regulatory formats suggested
 in Section  5.4 below require that records of material samples be kept.

 4.5  PROCEDURES FOR COMPLIANCE DETERMINATION
     Potential regulatory  formats for control of emissions from dry
 surface  impoundments are  listed in Table 4-8.  These focus on appropriate
 measures for compliance determination.
                                4-38

-------
      TABLE 4-8.   COMPLIANCE  DETERMINATION  FOR  DRY  SURFACE  IMPOUNDMENTS
Source
Emission surrogate    Control measure    Field audit parameter
Wind erosion
Removal of
waste residue
Threshold friction
velocity; % VE at
property line or at
source; PM10/TSP
concentration at
property line
% VE at property
line or at source;
PM10/TSP concentra-
tion at property
Chemical
stabilization

Wind sheltering

Wet suppression


Wind sheltering

Wet suppression
Surface crust and
texture

Surface wind speed

Surface moisture
content

Surface wind speed

Waste moisture
content
4.6  EXAMPLE CALCULATION

     Source Description

     •    Dry surface impoundment of dimensions 100 m x 100 m

     •    Waste residue density of 1.3 g/cc with a 2-cm dry surface layer

          having a moisture content of 0.5% and with subsurface material

          at a higher moisture level (exceeding 1.0%)

     •    Use of pan scraper to remove waste residue for transfer to

          landfill

     •    Scraper operation on the surface impoundment for 4-h at an

          average speed of 3 km/h yielding a source extent of 12 VKT.
          (This entails an average of three passes over each subarea of

          the impoundment surface)

     •    Site specific S&A (see Appendix D) indicates that phenanthrene

          is the single organic compound found in the highest
          concentration—60 pg/g

     Calculation of Uncontrolled Emissions.  The uncontrolled PM10

emissions from the waste residue removal operation may be estimated using

the emission factor for top soil removal by pan scrapers, as presented in

Section 4.2.2.  The total uncontrolled PM10 emissions are given by
                               4-39

-------
                    eu = 5.7 kg/VKT (12 VKT)                       (4,12)

                       » 68.4 kg

The estimated uncontrolled emissions of a particular contaminant would be
obtained by multiplication of the above figure by the contaminant
concentration (expressed as a mass fraction) in the waste residue:

                    68 4 ka x 60 yg x  if000 g x
                    oe.* Kg x   g   x   [ |
-------
 2.   Cowherd,  C.,  and  C.  Guenther.   Development of a Methodology and
     Emission  Inventory  for Fugitive Dust for the Regional Air Pollution
     Study.  EPA-450/3-76-003.   Prepared for U.S. EPA, Office of Air and
     Waste Management, Office of Air Quality Planning and Standards,
     Research  Triangle Park, North  Carolina.  1976.

 3.   Gillette,  D.  A.,  et al. Threshold Velocities for Input of Soil
     Particles Into the  Air by  Desert Soils.  Journal of Geophysical
     Research,  85(C10),  5621-5630.   1980.

 4.   Chepil, W. S.  Improved Rotary Sieve for Measuring State and Sta-
     bility of Dry Soil  Structure.   Soil Science Society of America Proceed-
     ings, 16,   113-117.  1952.

 5.   Bisal, F., and W. Ferguson.  Effect of Nonerodible Aggregates and
     Wheat Stubble on  Initiation of Soil Drifting.  Canadian Journal of Soil
     Science, 50,  31-34.  1970.

 6.   Marshall,  J.   Drag  Measurements in Roughness Arrays of Varying
     Density and Distribution.   Agricultural Meteorology, 8, 269-292.
     1971.

 7.   Axetell,  K.,  and  C.  Cowherd, Jr.  Improved Emission Factors for
     Fugitive  Dust from  Western Surface Coal Mining Sources.  2 Volumes.
     EPA Contract No.  68-03-2924, PEDCo Environmental, Inc., Kansas City,
     Missouri.   July 1981.

 8.   Muleski,  G. E.  "Coal  Yard Wind Erosion Measurements.  Final  Report
     prepared  for Industrial Client of Midwest Research Institute, Kansas
     City, Missouri.  March 1985.

 9.   Nick!ing,  W.  G.,  and J. A. Gillies.  "Evaluation of Aerosol Produc-
     tion Potential of Type Surfaces in Arizona," Submitted to
     Engineering-Science.  Arcadia,  California,  for EPA  Contract
     No. 68-02-388.  1986.

10.   Changery, M.  J.  National  Wind Data Index Final Report.  National
     Climatic  Center,  Asheville, North Carolina, HCO/T1041-01 UC-60.
     December  1978.

11.   Grelinger, M. A.  Gap  Filling  PM10 Emission Factors for Selected
     Open Area Dust Sources.  EPA-450/4-88-003, U.S. Environmental
     Protection Agency,  Research Triangle Park, North Carolina.  1988.

12.   Englehart, P., and  D.  Wallace.  Assessment of Hazardous Waste TSDF
     Particulate Emissions.  Final  Report, EPA Contract No. 68-02-3891,
     Assignments 5 and 13,  U.S. Environmental Protection Agency, Research
     Triangle  Park, North Carolina. October 1986.

13.   Chepil,  N. S., and  N.  P. Woodruff.  The Physics of Wind Erosion and
     Its Control.   In  Advances in Agronomy,  Vol. 15, A. G. Norman, Ed.,
     Academic  Press, New York,  New York.  1963.
                               4-41

-------
14.  Cuscino, T.f Jr., et al.  Iron and Steel Plant Open Source Fugitive
     Emission Control Evaluation.  EPA-600/2-83-110.  U.S. Environmental
     Protection Agency, Research Triangle Park, North Carolina.
     October 1983.

15.  Cowherd, C., Jr., and J. S. Kinsey.  Identification, Assessment and
     Control of Fugitive Particulate Emissions.  EPA-600/8-86-023, U.S.
     Environmental Protection Agency, Research Triangle Park, North
     Carolina.  1986.

16.  Carnes, D., and D. C. Drehmel.  The Control of Fugitive Emissions
     Using Windscreens.  In Third Symposium on the Transfer and Utilization of
     Paniculate Control Technology (March 1981), Vol.  IV, EPA-600/9-82-005d,
     NTIS No. PB83-149617.  April 1982.

17.  Larson, A. 6.  Evaluation of Field Test Results on Wind Screen
     Efficiency.  In Fifth EPA Symposium on Fugitive Emissions: Measurement
     and Control, Charleston,  South Carolina.  May 3-5, 1982.

18.  Westec Services, Inc.  Results of Test Plot Studies at Owens Dry
     Lake, Inyo County, California.  San Diego, California.  March 1984.

19.  Radkey, R. L., and P. B. MacCready.  A Study of the Use of Porous
     Wind Fences to Reduce Particulate Emissions at the Mohave Generating
     Station.  AV-R-9563, AeroVironment, Inc., Pasadena, California.
     1980.

20.  Zimmer, R. A., et al.  Field Evaluation of Wind Screens as a Fugi-
     tive Dust Control Measure for Material Storage Piles.  EPA-600/7-86-
     027, U.S. Environmental  Protection Agency, Research Triangle Park,
     North Carolina.  July 1986.

21.  Studer, B. J. B., and S.  P. S. Arya.   Windbreak Effectiveness for
     Storage Pile Fugitive Dust  Control:  A Wind Tunnel Study.  Journal of
     the Air Pollution Control Association,  38:135-143.   1988.

22.  Ohio Environmental Protection Agency.  Reasonably Available Control
     Measures  for Fugitive Dust  Sources.  Columbus, Ohio.
     September  1980.

23.  Kinsey, J. S.,  et al.   Study of Construction Related  Dust Control.
     Contract  No. 32200-07976-01, Minnesota Pollution Control Agency,
     Roseville, Minnesota.   April  19, 1983.

24.  Cowherd,  C., G.  E. Muleski, and J. S.  Kinsey.  Control of Open
     Fugitive  Dust Sources.   EPA-450/3-88-008, U.S. Environmental
     Protection Agency, Research Triangle  Park,  North Carolina.
     September 1988.
                                4-42

-------
                              5.0   LANDFILLS

5.1  SOURCE DESCRIPTION
     The placement of hazardous bulk and/or containerized solid wastes
Into permitted landfill units may Involve a number of activities with the
potential to generate PM10 emissions that may be contaminated by toxic
constituents.  The "mix" of potential dust emitting activities and, con-
sequently, the level of particulate emissions can be expected to vary
significantly from facility to facility.  This variation likely depends
upon a number of site attributes including:
          Whether the site is an off-site facility handling a variety of
          waste streams or an on-site facility confined to relatively few
          specific waste streams.
          The physical and chemical characteristics of the waste streams
          accepted at the facility, as well as the volumes of waste
          handled.
          The physical size of the site, its geographic location, and
          associated climate.
          The available equipment, operating practices, and perhaps the
          "age" of the disposal area (both in terms of years and relative
          to its expected permitted capacity).
     In general terms, previous surveys 1 of hazardous waste (HW) treat-
ment, storage, and disposal facilities (TSOFs) suggest that there are
four unit operations common to HW landfills.  These operations, as
described below, are:
     •    Loadout of solid hazardous waste.
     •    Loadout of temporary, clean (i.e., presumably uncontaminated)
          cover material.
          Lift construction involving the placement and possibly
          compaction of hazardous waste, followed by spreading and
          compaction of clean cover.
                                5-1

-------
     •    General vehicle traffic proximate to the active face of the
          landfill.
     Wind erosion from open areas represents a fifth potential dust emit-
ting mechanism associated with landfill units.  However, as a general
rule it is expected that the contribution of wind erosion to total land-
fill emissions will be negligible 1n comparison to the emissions produced
by the above unit operations.  Thus, it 1s recommended that landfill
emissions calculations initially neglect wind erosion; if refinements are
desired, wind erosion estimates can be obtained using the procedures
described 1n Section 3.0.
     The unit operations may involve a variety of vehicles/equipment.
For convenience, Figure 5-1 provides line drawings of some of the common
equipment types.  These are referenced as A, B, C, and D in the discus-
sion below.
     Figure 5-2a and 5-2b show generalized versions of a landfill in plan
and profile.  As indicated, at a given point in time, a "typical" unit
would consist of (1) a completed lift area with a "typical" height of 2
to 4 m; (2) the active face and associated loadout zone where waste  is
being placed; and  (3) an area proximate to the face where the majority of
vehicle movement occurs.
     The first operation—loadout of solid hazardous waste—involves the
placement of waste into the loadout zone (see Figures 5-2a and 5-2b)
using vehicles with varying dumping capacities.  Figure 5-1A is intended
to represent a commercial hauler used to transport solid hazardous
waste.  These vehicles typically have capacities on the order of  15 m*
(20 yd3), and usually are equipped to hydraulically remove waste  material
from the cargo "box."  Note also that smaller dump trucks may also be
used to transport  solid waste.  A reasonable range of average dumping
capacity at the  surveyed TSDFs is on the order of - 1.5 to 15 m*  (2  to
20 yd3); a reasonable midpoint (i.e., mean or median) value for the
distribution is  probably 11.5 m3 (15 yds).  Note that it is quite
possible that "typical" dumping capacity at a facility varies depending
upon whether that  facility  is primarily an on-site or off-site TSDF.
Because of the generally smaller volumes handled at on-site facilities,
these  may  tend to  have  smaller capacity waste haulers than off-site
facilities.
                                5-2

-------
                                                       r\
         B
                                              i
                                           "•^•^ i
         D
Figure 5-1.   Line drawings of common equipment types at HW landfills,
                                 5-3

-------
                                  Landfill Unit

                                    PLAN VIEW
Gate/Scale
                    Facility Haul Road
                               (b)
                PROFILE
                                                                \
                                                                Load-out "Zone"
                                                        \
                                                       Disposal Area
                                                                    Lift Height
                                                 . Ground
                                                 Level
              (c)
Landfill Unit-Master Cell with
     Designated Subcells
                                     PLAN VIEW
1
Y////////////,
A
'//////
\
B
/BERMs
\
'//////.
C
'///////
I
            Figure 5-2.   (a) Hypothetical  landfill unit plan view; (b) profile view;
                         (c) hypothetical  landfill with master cell/subcell
                         configuration.
                                          5-4

-------
     The second operation—loadout of temporary cover—is essentially
identical to loadout of hazardous waste.   The typical  dumping capacity of
the vehicles used for this operation probably is greater than for haz-
ardous waste loadout with a reasonable midpoint value  on the order of 15
to 19 n»3 (20 to 25 yd3).  If the landfill  is operating properly,  the com-
pleted lift area should have clean temporary cover—i.e., earth material
with negligible contaminant levels.  This  material  often is loaded out
using vehicles such as Figure 5-1D—an off-road haul truck.
     Methods for the third operation—lift construction—tend to be site-
specific and in most cases have evolved to meet unique site parameters
like waste stream volume, physical/chemical characteristics of the
wastes, and site climate.  The objective  of lift construction is simply
to move and compact the waste material in  a manner  that creates a struc-
turally sound area.  This lift area then  forms the  basis for successive
lifts as the landfill capacity is gradually used.  Based on observation
of activities at several operating HW landfills, it appears that actual
lift construction usually is accomplished  by a tracked vehicle fitted
with either a blade or a bucket.  Other heavy equipment, such as a pan
scraper or motor grader, may be used to spread temporary cover and/or
waste material.  In some cases a sheepsfoot compactor  is used to attain
the desired compaction of the material, while other facilities depend
upon the routine heavy equipment movement  for compaction.  A typical
vehicle used for lift construction is depicted in Figure 5-1C.
     The fourth unit operation—general vehicle traffic--encompasses a
broad spectrum of activities including hauler traffic  on the lift or
proximate to the landfill face, placement  of containerized waste by
operating landfill equipment, and light-duty vehicle traffic (i.e.,
pickup trucks). Containerized wastes typically are  handled by a forklift
or comparable equipment depicted as Figure 5-1B.
     Figures 5-2a and 5-2b depict a situation in which all wastes are
placed in the same area, regardless of their physical/chemical character-
istics.  While this may be the case at some facilities, field observa-
tions suggest that some landfill units are configured  as in Figure 5-2c—
a master cell with subcells.  One reasonably common practice apparently
is to designate placement of wastes in subcells according to the RCRA

                                5-5

-------
characteristic criteria for ignitability (40 CFR Part 261.21), cor-
rosivity (40 CFR 261.22), and reactivity (40 CFR 261.23).

5.2  ESTIMATION OF UNCONTROLLED EMISSIONS
     The following section presents the emission factor models used to
estimate participate emissions from landfills,  It includes a summary of
available data on both the physical and chemical characteristics of
disturbed surfaces as well as source activity levels found at operating
HW landfill units.

5.2.1  Particulate Emission Rates
     Estimation of uncontrolled particulate emissions from landfills
follows the general model (Eq. 1-2) presented in Section 1.2.  For
convenience, it is reiterated below with terms potentially applicable to
landfill units:
                                n
                           R =  I  a.
where:     R = emission rate of contaminated airborne particulate (kg/yr)
               for a given landfill, consisting of n identifiable unit
               operations

          a,- = fraction of contaminant in particulate emissions for the
           J
               jth operation (wg/g); if only PM10 emissions are of
               concern, a..- = 1
                         J

          6j = emission factor(s)  (mass/source extent)

          A-- = source extent(s)  (source dependent units)
           J

 For landfills, n=4 with the unit operations corresponding to those out-
 lined  in  Section  5.1—loadout of HW; loadout of temporary cover; lift
 construction;  and general vehicle  traffic.

                                5-6

-------
     Table 5-1 provides an emission factor (e,-) for each of the unit
operations that may occur at a landfill.  These emission factors are
included in EPA's "Compilation of Air Pollutant Emission Factors"
(AP-42).2  Excepting the a,- term, there are features of each emission
factor that should be clearly recognized before any estimation is per-
formed for a given landfill unit.  These features are discussed below.
     For the first two unit operations (Table 5-1)—loadout of bulk waste
and cover, respectively—the emission factor is based on testing of com-
mon industrial aggregates (coal, crushed limestone, sand, etc.).
     For the third unit operation—lift construction and maintenance—the
unit operation is assumed to generate particulate emissions at rates com-
parable to those determined in field source testing of a dozer operation
involving overburden removal at western surface coal mines.  Note further
that the emission factor model developed for dozer operations actually
relates to emission measurements of particulate matter less than 15 ym
(PM1S).  However, Supplement B of AP-42 recommends a PM15 to PM10 conver-
sion factor which is reflected in Table 5-1.
     For the fourth unit operation—general vehicle traffic—the activ-
ities proximate to the landfill face are assumed to be analogous to those
tested in development of the unpaved road emission factor relationship
(presented in Section 2.2.2.1).  Although this emission factor was devel-
oped for "wheeled" vehicles on we11-compacted, unpaved surfaces, it is
assumed to apply to tracked vehicles as well.
     Application of the emission factor expressions in Table 5-1 for a
given site requires appropriate data on physical properties (e.g.,
vehicle speed, weight, etc.) as well as mechanical activities.  Table 5-2
provides summary statistics for the various emission factor correction
terms; these data were developed as part of an initial survey of land-
based TSDFs.  Because most of the terms exhibit significant variability
from site to site, use of site-specific data to represent these param-
eters is highly recommended.
     Table 5-3 provides mean wind speed (U) data (m/s) for selected
National Western Service (NWS) stations.  In general, the value of U for
the reporting station located closest to the facilty in question, may be
taken as data input for Eq. 5-1 to estimate emissions associated with
loadout of bulk hazardous material or temporary cover.
                                5-7

-------








0
4-
X
4)

U
l_
1





I 0
^™ ^E
^
cr o-
LU LU
&
t—




_J
J
^

LU
^v
<
X
VI

1 L.
< O
(SI 4-
< U
Z ID

§ §

§m
in
f- E
U LU
u.

^^
_
(/)
v*
z
LU

•
w—
1
tn
LU
3
*~







8

4-
IO
i_
4)
a
O
4-
.«
C
3






T3 TJ
4) O

•0 -0
c c
ID ID
* *
«J ID
«• •«
L. L.
. «
g i

O L.
4- «
in > 43
ID 0 E
* 0 —
"£ t- V.
O O in
in
i1 f f
T ^
I i
tn in

_
ID

L,
4)
^B
•0 ID
4) E

ID «
4- 4-
m in
ID
m x
— *•* 4)
C £ £
3 a 4-
E
£—5

f § fc-xS
X 4- 4- E —
Ol X O »*
.X .O ID 4- — .
— • — •«- •a c i_
— 4) 41 £
C 41 4- X
0 a c o>
— mo JL
mo w
^O (fl T3
V — C 4)
— E — L.
^— ^ -- 	 IO 4) * 3

3] • 2I 4)


^
4)
N
L.
4)
C L.
_ «
ID >
+- 0
§
U >>
C L.
3 ID
~- L. C
O O
je 4) a —
— 4- 64-
3 tn 4) u
^5 ID 4-3
X L.
^ «• ^
O *n O in
3 — C
4- O 4- ID 0
3 T3 3 — O
O L. O L.
•a 

4>

ID
+-
in

0
i
5

I
41
U ••>

.C ^

o to
— 1
L. in
4-
IO
•4- P E
— 41 >
> X V
c o O) -~ in
•^ ^^ o ^ ^^ (*^
^— 1H' 3E 0 *
•O ID •« > O O
T3 4) — C X -O 1C
4> tn i_  •*- AI
+-34) — 4- t.
O 4-4) ^— • ^ -* 10 3 C
V— ID4-I3J 4-lfl O
in 10 E tn i lin m —
— ID ml >o — +-
o in 0) ID
+- 4) 4) tol ^— ~^ 4- > ^ •»-

caO£ (in cuxo) a.
3 6 u 4- u-\ m\o 34-EZ —
• \O tO it *-" O
C 4) T3 -^ O to C 4) — 4)
— £ C O x-1 " » •««-»• — £ 4- in L.
4- ID »|in 4- •o £ — a.
u <-. ^.^^ in L. 4) O) 4)
O — 4) M. • O •»• 0 — 4) £
4-Q4-'w p. O 4-OQ.4)£4-
o in • -" ' «^ o m f i —
^ a <- a 4- o »l» ID -~ x
L. -4. •* X C ^— — ^ *** i •" •- *« 4) 4) 4)

x c 4-ae|'f^ c o u u >.
A O4-CC {\4 • O *- — — —  > > 3 O
^- LU m U E "*i ^" ^•••'•^ Q^ in ID ID ID C •-*
-«. «r CM
in • u u M in — u H u u u u
^mr ^ ^K ^I.B^M^
— 4) in I — 4>in
-------








t/)
H"
Z
2
«J
i
«|J
>—«
U*
0
3

C3
UJ
«J
o.
5
fl.
^0
f^
Q£
0
U»

v)
K*
Z
i_ij
h^j
H—
X
UJ
UJ
y
Ur
—^
O

C c
O> (0
4-> ee.

c
o
4-»
U
QJ
1_
1_
O
°T<





vl
^ C
U 0
<*- VI i-
O QJ 4->
r— (0
• Q. >
O £ i-
Z <0 QJ
«/l (/I
ja
o


vi
4->
•f—
CO
<4- 3 T3
O OJ

O ^ §"
Z <4- (0
T3 ol
C
03
^~




-*J
•^-
C
13






i- 4J
0 C
-*-> OJ
O 4->
<0 X
It- Q)
C 0)
o o
VI 3
bl O
i- ol
E^-
OJg
C Q)
0 4->
•T—
4-> C
to o
I. f-
QJ -4->
Q. (J
O 0)
s_
4-> i_
•r- O
C U
13
in
  «•  •> O"i
5 s: 2 s:
VP >>P
o*t o^




4-* *J
C C
 -M
X X
OJ OJ QJ
• • ••
4-» QJ 4-» OJ
E if 8 if
^3-03
(O o 10 O
O co O 
4-> f— 4-> i—
O) C <0 I- C <0
•M O) 3 QJ OJ 3
VI 4-> C > 4-> C
jg c c o c c
X O <0 U O (0
U (J
> T3
SO) OJ U OJ OJ
i_ 4-> (0 S- 4->
•J3 3 <0 fc. 3 *
fc- 4-> E O 4-> E
10 CO -I— Q. (/I -i—
N •<— 4-> E •<- -t->
10 O co  ^O
ff< o1^












t •
C
0 4->
f- C.
4-> QJ
U 4->
240 C
C O
4-1 QJ 0

§C QJ
O i.
U 0 3
4->
4J 4-J tO
1*- ^- -I-
•r« f~ O
— 100 S


•
HM
H^

§
CM •
• r^ co co
CM CM v
•o
O CM 00 O CM
0 ^ CM^
t>
«— I









O>
CO C^ C7^ Ot CT>













LO VO VO CO VO









2
f "^"rt O» i
-f^ .» ^^ '
^£
>p
w^




4^
C
QJ
4->
X
O)
*«
OJ U
U f- 4->
s- *^« -o o= 01
BC|_ a) C3»i—
10 Q) -^ QJ
VI J- O. O) 0)
4-> CO 2 J=
10 4) QJ O)
c 1j "o 'u o
C •«" 4-* f- •*-
(0 J= C JZ JZ •
O) OJ QJ QJ O
T3 > 4-> > > Z
01 C
4-i r- O QJ O) 0)
io ie o cr» o» o>
E k. (0 (0 ro
•<- 0) 4-> i. i- i-
4-> C r- QJ QJ  > >
UJ CO OO 
HH










































•
«— 1

O)
u
c
OJ
s-
QJ
Ci_
QJ
0£

E
0
s-
<4_
c
O)
j<:
10
4->

(0
4->
(0
•o
•1
T3

0
01
01
OJ
f™"
c
(0


CO
10
2

r—
^^
QJ
U
,_
^1

J^
•^
(J
10
<4_

E
O
01
4->
10

• *>
•o
QJ
>>
QJ

I.
3
Ol

01
QJ
•T™
-^
•f—
r—
•^~
u
(0
Ct-
^~
(0
3
T3
•r-
>
T3
-I~
<4-
O
s_
QJ
J3
E
C
O
4-> •
T3
01 QJ
1- r—
QJ Q.
'-*- E
QJ <0
at. co
n
QJ
3
4V
01
•^
O


t-
0
•4-
•o
OJ
N
>>

10
(0
VI
<0
2

Q)
pM
Q.
E
<0

QJ
O


•
01
^_
QJ
U

QJ


01

i.
O
C(_
• •!
T3
QJ

O.
E
(0
Ol
CO
r—
Q)
U

^~>
•^
tt-
T3
C
10
•~
Cl_
O
S-
QJ
J3
E
C
0
4-> «
4->
01 C
S- OJ
QJ -M
ci- C
QJ O
u020
































































•
C
(0

(J
4->
QJ
JZ
4->
s_
<0
0
4->
Ol
S-
OJ
14_
QJ
a:
T3











§
t-
«»-

c
Q)
.*
rO
4-»

10
4->
10
•o

• 0
VI
r~
•f—
<4_
•a
c
(0


3*
00
z^

QJ
4-»
CO
(0
3
T3
•r-
^—
O
01

f— "
at
a.
u
*^»
c

p

4->
r0
^— «
^
QJ
to
3
1_
QJ

O
O
>>
«0
O
IV
QJ
4-> ^O
C
4-> (0
C
Q; oo
to
QJ CO
S- QJ
a. u
QJ C
t- QJ
S_
10 QJ
4-> U-
(0 QJ
a OL.
QJ





































*
VI
OJ
4-1

^_
*^-
U
<0
14-

3
oo
2:
E
0
S-
n-

to
QJ
3
f*™
(0
>
C
O

-o
QJ
to
(0
.a
(0
4->
>0
-a
01
<0

QJ
^
<0
(J
Q.
a.
(0

4_)
O
z
C|_












•
4->
C
o
E

QJ

u
»^-
.c
QJ
>
3
Q
^_
VI

>»

10
W
S
QJ
a>
4->
U
QJ
<*-
QJ


0
4->

•o
QJ
T3
Ql
4->
C
•f—

• «
c.
o
•^*
4->
10
>
i_
QJ
to
JO
o
c
o
•o
QJ
to
(0
O2
to
QJ
^
iQ
>
•D
QJ
4->
(0
E

4->
CO
UJ
a>
5-9

-------
TABLE 5-3.  MEAN WIND SPEEDa [U] FOR SELECTED UNITED STATES STATIONS
Station
Birmingham
Montgomery
Tucson
Yuma
Fort Smith
Little Rock
Fresno
Red Bluff
Sacramento
San Diego
Denver
Grand Junction
Pueblo
Hartford
Washington
Jacksonville
Tampa
Atlanta
Macon
Savannah
Boise
Pocatello
Chicago
Mo line
Peoria
Springfield
Evansville
Fort Wayne
Indianapolis
Burlington
Des Moines
Sioux City
Concord i a
Dodge City
Topeka
Wichita
Louisville
Shreveport
Portland
Baltimore
Boston
Toledo
State
AL
AL
AZ
AZ
AR
AR
CA
CA
CA
CA
CO
CO
CO
CT
DC
FL
FL
GA
GA
GA
ID
ID
IL
IL
IL
IL
IN
IN
IN
IA
IA
IA
KS
KS
KS
KS
KY
LA
ME
MD
MA
OH
[U]
(«/$)
3.3
3.0
3.7
3.5
3.4
3.6
2.8
3.9
3.7
3.0
4.1
3.6
3.9
4.0
3.4
3.8
3.9
4.1
3.5
3.6
4.0
4.6
4.6
4.4
4.6
5.1
3.7
4.6
4.3
4.6
5.0
4.9
5.4
6.3
4.6
5.6
3.8
3.9
3.9
4.2
5.6
4.2
Station
Detroit
Grand Rapids
Lansing
Sault St. Marie
Duluth
Minneapolis
Jackson
Columbia
Kansas City
St. Louis
Springfield
Billings
Great Falls
Havre
Helena
Missoula
North Platte
Omaha
Valentine
Ely
Las Vegas
Reno
Winnemucca
Concord
Albuquerque
Ro swell
Albany
Binghamptom
Buffalo
New York
Rochester
Syracuse
Cape Hatteras
Charlotte
Greensboro
Wilmington
Bismarck
Fargo
Cleveland
Columbus
Dayton
Dallas
State
MI
MI
MI
MI
MN
MN
MS
MO
MO
MO
MO
MT
MT
MT
MT
MT
NE
NE
NE
NV
NV
NV
NV
NH
NM
NM
NY
NY
NY
Ny
NY
NY
NC
NC
NC
NC
ND
ND
OH
OH
OH
TX
[U]
(m/s)
4.6
4.5
4.6
4.3
5.1
4.7
3.4
4.4
4.6
4.2
5.0
5.1
5.9
4.5
3.5
2.7
4.6
4.8
4.8
4.7
4.0
2.9
3.5
3.0
4.0
4.1
4.0
4.6
5.5
5.5
4.3
4.4
5.1
3.4
3.4
4.0
4.7
5.7
4.8
3.9
4.6
4.9
                             (continued)
                                  5-10

-------
                             TABLE  5-3 (continued)
Station
Oklahoma City
Tulsa
Portland
HarHsburg
Philadelphia
Pittsburgh
Scranton
Huron
Rapid City
Chattanooga
Knoxville
Memphis
Nashville
Abilene
Amarillo
Austin
Brownsville
Corpus Christi
State
OK
OK
OR
PA
PA
PA
PA
SD
SD
TN
TN
TN
TN
TX
TX
TX
TX
TX
[U]
(m/s)
5.7
4.7
3.5
3.4
4.3
4.2
3.8
5.3
5.0
2.8
3.3
4.1
3.6
5.4
6.1
4.2
5.3
5.4
Station
El Paso
Port Arthur
San Antonio
Salt Lake City
Burlington
Lynchburg
Norfolk
Richmond
Quillayute
Seattle
Spokane
Green Bay
Madison
Milwaukee
Cheyenne
Lander
Sheridan
Elkins
State
TX
TX
TX
UT
VT
VA
VA
VA
WA
WA
WA
WI
WI
WI
WY
WY
WY
WV
[U]
(m/s)
4.2
4.5
4.2
3.9
3.9
3.5
4.7
3.4
3.0
4.1
3.9
4.6
4.4
5.3
5.9
3.1
3.6
2.8
Source:  Adapted from Rapid Assessment of Exposure to Particulate Emissions
         from Surf ace Contamination Sites,  EPA/600/8-85/002, U.S. EPA, Office
         of Health and Environmental Assessment, Washington,  DC.

aData taken from Local Climatologicai Data - Annual Summaries for 1977.
 U.S. Department of Commerce, National Oceanic and Atmospheric Administration/
 Environmental Data Service/National Climatic Data Center.
                                     5-11

-------
5.2.2  Level of Contamination
     The data presented 1n Reference 1 suggest several points that may be
useful 1n specifying the level of contamination (a) for landfill units.
These include:
     •    The question of whether or not a given organic compound is
          present and 1f so at what levels 1s far more site-specific for
          organics than for RCRA metals.  This suggests that source-
          specific S&A would, 1n fact, be needed to characterize a levels
          for organic compounds.
     •    If a source-specific S&A plan is being developed (see Appen-
          dix D for generic procedures), the data in Reference 1 indicate
          that organic compounds with lower vapor pressures tend to be
          found more often than compounds with high vapor pressures
          (i.e., volatile compounds). As a result then, the S&A plan
          should be directed toward organic compounds with relatively low
          vapor pressures (< 10 mm Hg).
          For off-site facilities, the available data indicate that all
          the RCRA metals (with the possible exceptions of silver [Ag]
          and selenium [Se]) typically are present at levels in excess of
          background values.  Thus, source-specific S&A generally should
          not be required to establish which metals are present, but only
          to determine what are representative a levels.
     •    As an alternative to source-specific S&A for metals, a "worst-
          case" scenario could be developed based on the data shown in
          Table 5-4.  For example, this could be accomplished by taking
          the maximum values  (upper end of the range) and applying a
          "safety factor" of 5 or 10.  In other words, multiply the upper
          end values by 5 or  10 as a conservatively high estimate of a.

5.3  CONTROL TECHNIQUES
     While  there has been no actual field testing of particulate control
techniques  at HW TSDFs, for  "first-cut" analyses, it is reasonable to
assume  that much of the experience and information developed in more
traditional industries (e.g., iron and steel) can be applied to HW TSDFs.
                               5-12

-------
          TABLE 5-4.   SUMMARY OF RCRA METALS

               CONCENTRATION  (yg/g)a'b'C
 RCRA metal              Xd             Range
Arsenic (As)
Barium (Ba)
Cadmium (Cd)
Chromium (Cr)
Lead (Pb)
Mercury (Hg)
Selenium (Se)
Silver (Hg)
12.7
771
35.8
1,200
1,490
1.7
1.5
13.2
4.7-21.3
86.9-3,340
NDe-150
91.2-1,500
54.4-6,870
NO-9.75
NO-4.80
ND-46.4
aData taken from Reference 1.

 At some facility units,  more  than one sample was
 analyzed; in this case the mean value for the unit is
 used to compute the overall mean shown.

Statistics based on samples collected from 8
 different landfill cells.

dRefers to arithmetic mean; nondetected values assumed
 to be equal to 0.0.

eND refers to not detected.
                     5-13

-------
This section discusses the use of both mitigative controls typical of
traditional heavy industry, as well as preventive controls potentially
applicable to HW landfill units.

5.3.1  Preventive Controls
     As applied to HW landfills, preventive controls might be more appro-
priately termed "gopd operational practices."  For HW loadout (unit
operation I), the principal preventive control would involve use of
containers, including double bags, shrink wrapping, drums, etc., to
reduce the potential for finely divided materials to become entrained
into the atmosphere during the loadout operation.
     The emission factor relationship to be used for waste loadout indi-
cates that emissions are expected to vary inversely with moisture content
of the material.  This suggests that addition of moisture to the waste
stream prior to loadout could be a feasible control,  However, this con-
trol cannot be generally recommended to cover all waste streams and sites
because it may contradict guidance designed to minimize leachate produc-
tion from the unit.  In other words, the decision as to whether or not
moisture is added must be made on both site- and waste stream-specific
bases.
     For the unit operation involving loadout of temporary cover, it is
assumed that the cover material is uncontaminated, and thus only gross
particulate emissions would be of concern.  Reductions in the volume of
material and/or addition of moisture would be expected to reduce gross
emissions associated with this operation.  In practice, however, reducing
cover material may  lead to situations where contaminated material is
inadequately covered, thereby increasing the potential for contaminated
particulate emissions.  For this reason, the feasibility of reducing
cover material must be evaluated on a unit-specific basis.
     For the unit operation of lift construction and maintenance, prac-
tices vary widely from facility to facility.  Despite this variation, the
use of dedicated landfill  equipment is  a preventive  control  practice  that
should be  applicable at almost all facilities.  The term "dedicated"
means simply that a given piece of landfill equipment (e.g., bulldozer)
operates only within the  landfill unit  and is not routinely removed from

                               5-14

-------
the landfill to perform other facility activities.  The rationale for
this practice is simply that restricting the movement of the landfill
equipment, in effect, decreases the likelihood that contaminated material
will be spread over other facility travel surfaces.  To control material
spreading when dedicated equipment must be removed from the landfill
unit—periodic maintenance, for example—a wash station should be
established.
     Another potential mitigative practice for contaminated partlculate
involves "housekeeping" on travel surfaces proximate to the landfill
face.  Housekeeping refers to the use of a scraper or motor grader to
remove contaminated material that may be deposited in travel areas during
normal lift construction.  Again, this practice is intended to reduce the
spreading of contaminated material.
     For the unit operation encompassing general vehicle traffic, preven-
tive control may be initiated by construction of temporary roadways with
gravel or other aggregate material.  The replacement of a travel surface
with relatively coarse aggregate reduces silt content and thus particu-
late emissions.  In control of contaminated particulate, this practice is
intended mainly to aid traffic routing.   The objective  is  to  restrict
general traffic to areas that are not used actively by operating landfill
equipment.  Separating the activity patterns of operating equipment from
other vehicles may potentially reduce the spread and subsequent
reentrainment of contaminated material.

5.3.2  Mitiqative Controls
     As discussed in Section 2.2.3.2, there are various mitigative mea-
sures that potentially could be used to control particulate emissions
from operating landfill units.   These measures fall under the category of
surface treatments and include watering as well as chemical stabiliza-
tion.  Chemical suppressants should be evaluated on a unit-specific
basis, with particular attention paid to compaction of the surface,
drainage, suppressant composition, etc.
     Watering as a control measure for landfill operations does not
appear to suffer from the same limitations as chemical suppressants.
That is, water alone presumably would not markedly increase the levels of
                               5-15

-------
undesirable organic compounds added to the landfill surface.  (Note that
leachate production due to frequent watering may be a concern; this is
discussed below.)  Similarly, curing time and its impact on operations is
not a relevant concern as it may be with chemical suppressants.  The
central issues in determining whether or not water represents a feasible
control at landfills involve:
          Application intensity (see Section 2.2.4.3).
          Frequency of application (see Section 2.2.4.3).
     •    Indirectly the potential spreading of contaminated material due
          to trackout.
     •    Climatic conditions.
     As discussed in Section 2.2.4.3, available field data indicate that
while application intensity (i.e., amount of water applied per unit area)
directly influences particulate control efficiency, the relationship
probably is not linear (cf. Eq. 2-11).  The long-term level of control
for a given average intensity can depend, for example, on activity
levels, climate, antecedent surface conditions, etc.  The relationship
between frequency of application and control may be assumed to depend on
essentially the same factors.
     In a qualitative sense, low application intensities and infrequent
applications generally may provide little effective control.  On the
other hand, high intensities/frequency may create undesirable results in
terms of both contaminated particulate control as well as ground- and
surface-water quality.  More specifically, addition of excess water
either directly to the bulk waste material or to the landfill surface
could result in:
     •    Reduction in the structural integrity of the lift material.
          Increases in leachate volume.
     •    Increase in the amount of contaminated material spread to
          surrounding areas.
     Each of the above represents an undesirable condition that has long-
term implications for the proper management of the landfill unit.  For
this reason, like chemical dust suppressants, use of water should be
recommended on a unit-specific basis.  If it can be demonstrated that a
watering program does not produce undesirable "side effects" (i.e., the

                               5-16

-------
points noted above), estimated control efficiencies can be developed by
extrapolating the bilinear relationship cited in Section 2.2.4.3 to cover
the variety of activities common to landfills.   For convenience, this
bilinear relationship is reiterated below.
                     c =
                          75 (M - 1)    1 < M < 2
                          62 + 6.7 N    2 < M < 5
(5-2)
where:    c = instantaneous control efficiency, %
          M = ratio of controlled to uncontrolled surface moisture
              contents

Note that this approach requires data on controlled and uncontrolled
surface moisture contents.  A potential format for producing this
information is outlined in Section 5.5.2.

5.4  CONTROL PERFORMANCE ESTIMATION
     Most of the control measures identified above as appropriate for
landfill operations involve the reduction of source extent.  The control
efficiencies to be achieved are predictable in that emissions are
directly proportional to source extent.  Therefore, the control effi-
ciency value, as a percentage of uncontrolled emissions, is numerically
equal to the percentage of source extent reduction.  The control effi-
ciency value would apply both to gross and contaminated emissions alike.
     In the case of containerizing the hazardous waste to reduce dust
emissions during loadout, a high efficiency of control (i.e., exceeding
90%) would be expected.  This assumes that container integrity is main-
tained during transport of waste to the site and during the waste
dumping.
     In the case of general vehicle traffic, the control efficiency
obtained through the use of coarse aggregate for temporary roadways is
predictable based on the decrease in road surface silt content achieved.
In the case of contaminated particulate emissions, adjustment of the
control efficiency for gross particulate emissions would have to be made
                               5-17

-------
to reflect any difference between the surface silt contamination level
with and without the surface improvement in place.

5.5  PROCEDURES FOR COMPLIANCE DETERMINATION
     The regulatory approach to control particulate emissions from land-
fills is by definition predicated on the RCRA site-specific permit
system.  Control agencies would issue construction, operation, and use
permits to facility operators.  In this section it is assumed that
permits specify a set of conditions or activities that must be provided
or undertaken by the source to ensure that PM10 emissions are kept to
acceptable levels.  Compliance determinations for landfill units will
then involve field observations/audits, as well as inspection of control
implementation records.  Additionally, compliance monitoring may require
collection and analysis of surface samples of dust-emitting material
using the procedures outlined in Appendix D.

5.5.1  Preventive Controls
     Field observations are the principal basis for determining whether
or not the preventive controls cited as permit conditions are in fact
being followed.  Table 5-5 provides an example checklist form for audits
of landfill preventive control practices.  There are several points
concerning the checklist that should be noted including:
     1.   The length of time for field observation/audit as well as audit
          frequency should be sufficient to view "typical" operations.
          While this certainly will vary from unit to unit, 1- to 2-day
          unannounced audits conducted twice annually can be recommended
          as a minimum program.
     2.   To the extent possible, the audits should not be scheduled in
          advance, so that a "clear picture" of typical practices is
          obtained rather than one potentially biased by advance
          notice.
     3.   In checking temporary cover, the objective is to look at that
          position of the landfill unit that was used on the day
          preceding the audit since most permits require "daily cover to
          a specified depth."

                               5-18

-------

UJ
fr#4
H-
ee.
o.
_j
o
Of
h—
y
O
0

LU
>
N««
h-
Z
LbJ
>
U4
Q-
•M!
«J
H«4
Lu
o
^
<£
_J
a:
O
Uu

h—
V/l
>«•*
«J
^^
t-P
LU

O
»—
o
2
^c
^*
o
^*
JI




O"
4->
•r-
g;
S

^
•^
it
*^
TO
IO



Ol
C
•r-
ID
D
I/I
•^

4->
C
O)
E
Q.

3


T5

4->
• • .C
^-v 4->
00
^_^ QJ
fl> O
E C
i- a;
4-> T3
C >
0 0)
•^
4_) (U
> a>
• •I- -C
ai a> 4->

1C J2 
IO
ai
o.
o
c
o

4->
03
4->
1/1
a;
i/1
c
.^
S-
4->
C
OJ
E
a.
3
cr
ID

00
»—H
«

C
aj

00
^

c "c
o -3

*-
IO TJP
3 *
<«- r—
O
OJ

C
a) o
T- C
(D
+J
(U C
O) E
JC O)
4-> (J
IO













C"
•o
^
IO
E
>>
T
IO
V

u
^^
V)
s-

4->
ia
0.
01
-f—
4->
o
i-

u
M-
«4-
lO
L.
4->

O)
i-
(/>
IV

(U
JZ
4->

oo
t— (
*










4)
>
o
c*
^
r-^
ai
>
IO
4->

a*
s-


J-
(U
10
J=

s-
d)

o
o

^
^
10
0
a.
HI
4->

o
"-s.
•o
c
10
ft)
4->
I/I
IO <»••
5 •—
IO
10 °£
^^ ftj
4-> 4->
IO
0) E
u
C d)
•0 to
t- (O
> 2
ai
•a
4) aj
S- 4->
Q^ »^
-C 00
4-^ O
a.
^




































_
"?
•f—
a.
(U

Vf
Si
3
O
.C
^
0
>4-
0}
1/1
3
C
**"
s-
O)
•a

en
$-
o
4->
O
E

IO

oo
""^
•
oo




c
>> a>
Trt Jo
3
O oo
"^ W
a>
t. 4->
0.10
O 4->
IO
0,0,
3 10
a. «
X
0) S-
o
O) <4-
N
*E ~V
i- C
C -r-
^ ^

+•> OJ
^3
^j
C 
(U jQ
•»~

> f^ .^
O t— T3
U 3
ia
J-  •—
WOO)
o z •*••
Q. «4-
4-4 ^* 4->
10
<*- *r- i.
O i- O
 O)
00 4->
TO iO IO
C 3 -i-
IO TO
X) a>
J=  E
^L f*" »^
a* oo
•O O T3
Ot OJ
vi a> oo
•— "O 3
•
5-19

-------
     Inspection of records is a necessary complement to the audit/
observation checklist.  For example, as indicated in Section 5.3, site
operators must remove dedicated equipment from the landfill unit for
periodic maintenance.  Maintenance records are routinely kept for heavy
equipment.  Site operators should be required to provide corresponding
documentation to indicate that the equipment has been thoroughly washed
prior to its removal from the landfill area.
     As indicated in Sections 5.3 and 5.4, construction of temporary
roadways using coarse aggregate is a recommended technique for reducing
particulate emissions.  At the same time, it is useful in clearly defin-
ing the traffic routes that are desired to minimize the spreading of
waste material.  If gravel is used as a preventive control, the operator
should be required to document the following items:
     1.   Type of gravel purchased, delivery dates, and quantities (tons)
     2.   Quantity stockpiled on-site at time of delivery (excluding
          current purchase)
     3.   Estimate of the material bulk density (Ib/yd3)
     4.   Application dates
     5.   Area treated (length and width)
     6.   Nominal depth of application (in)
     7.   Site map showing road orientation after each application
     8.   Any marking precautions (such as barrels, rope, pennants, etc.)
     The purpose of this information is to allow the field inspector to
perform a "rough" mass ba'.ance.  That is, the following expression should
be verifiable:

     Total quantity applied (tons) = total material purchased (tons) -
                                     total material stockpiled (tons)

Two points should be noted here:
     1.   The quantity stockpiled refers to the amount of material stock-
piled  on-site at the time of each delivery, exclusive of the delivered
material  itself.  Thus, each delivery should have a corresponding
stockpile value.
                               5-20

-------
     2.   Based on the estimated bulk density, area treated, and depth,
the inspector can approximate the quantity applied (tons) in each appli-
cation.  The sum of these individual values then becomes the total
quantity applied.

5.5.2  Mitigative Controls
     Watering is the principal mitigative control anticipated for use in
landfill units.  As indicated in Section 5.5.1, determining the control
efficiency associated with watering requires uncontrolled and controlled
values of surface moisture content.  If watering is being used as a con-
trol, the required data can be obtained during the field audits (speci-
fied under Section 5.5.1).
     Generic procedures for collection of surface samples and for mois-
ture content analyses are given in Appendix D.  The uncontrolled sam-
ple(s) should be taken early in the morning before the first water appli-
cation.  The controlled sample(s) should be taken at the approximate
midpoint in time between successive applications.  For example, if the
facility routinely applies water at 2-h intervals, the control moisture
sample should be taken after 1 h.  For purposes of this watering eval-
uation, it is critical that the application intensity (i.e., amount/unit
area) and frequency be the same as that used under "typical" operating
conditions.
     In order to verify that the watering done during the evaluation
constitutes typical practice, the facility operator must specify the
parameters of the control plan and make provisions to document the fact
that the plan is being followed.  The parameters to b€ specified in the
plan and to be regularly recorded are comparable to those for unpaved
roads.  They include:
     General Information
     1.   Area to be treated
     2.   Amount of water applied to each road/area and planned frequency
          of application (alternatively, a minimum moisture  level  could
          be specified)
                               5-21

-------
     3.   Any provisions for weather  (e.g.,  1/4-in of  rainfall with  be
         substituted for one treatment;  program  suspended  during  freez-
         ing periods; watering frequency as  a  function  of  temperature,
         cloud cover, etc.)
     4,   Source of water and tank capacity
     Specific Records for Each Road Segment/Parking  Area Treatment
     1.   Date of  treatment
     2,   Operator's initials (note that  the  operator  may keep a separate
         log whose information  is transferred  to the  environmental
         staff's  data sheets)
     3.   Start and stop times in a particular  area, average speed,
         number of passes
     4.   Start and stop times for tank filling
     General Records
     1.   Equipment maintenance  records
     2.   Meteorological  log  (to the  extent  that  weather influences  the
         control  program,  see above)
     3.   Any  equipment malfunctions  or downtime

5.6  EXAMPLE CALCULATION
     An  example  application of the emission  estimation procedures  for
landfill units is  presented below.   This  example  focuses on contaminant-
specific (i.e.,  lead)  emissions, but the  procedures  would be essentially
the same for gross particulate emissions.
     The example landfill  unit  is  taken  as an off-site facility  located
in a semiarid  climate with a mean  annual  wind speed  of 3 m/s and
323 dry d/yr.   The facility operates 5 d/wk  (261  d/yr) and accepts the
following listed waste streams  likely to  contain  appreciable concentra-
tions of lead (Pb).

                   Waste Stream      Annual  Receipts  (Mq)
                       K061                 5,000
                       K052                    100
                       K002                    15
                               5-22

-------
     Based on waste manifests and information supplied by the facility
operators, pertinent characteristics of waste streams are specified as:

                     Pb Concentration   Silt Content   Moisture Content
      Waste Stream        (ppm)              (%)              (%}
K061
K052
K002
3,000
1,500
300
30
8
8
1
20
20
Because the specified Pb concentrations are based on waste manifests and
information supplied by the facility, the values given above incorporate
a "safety factor1  of 5, in accordance with the approach indicated in
Section 5.2.2.  These Pb concentrations are applicable to loadout of the
raw wastestreams (unit operation I; see Eq. 5-1 of Table 5-1).  Because
no site-specific S&A is undertaken, the conservative assumption is made
that the Pb concentrations for surfaces subject to dozer activity and to
vehicle traffic are equal to the highest volume (K061) "raw" waste stream
levels (incorporating the safety factor of 5).  Default values for silt
and moisture content (averages presented in Table 5-2) are assumed
applicable to lift construction (unit operation III) and to general
vehicle traffic (unit operation IV).
     Based on site observations and conversations with facility person-
nel, annual source extent parameters are estimated as follows:
     Unit operation I~(annual receipts above);
     Unit operation II—(clean cover containing negligible concentrations
       of Pb);
     Unit operation 111 — 1,566 h (one dozer operating - 6 h/d); and
     Unit operation IV—260 VKT (50 vehicles/d traveling - 20 m round
       trip in the area proximate to the landfill face).
     For unit operation I—receipt/loadout of wastes—the uncontrolled
emission rate is calculated from Eq. 5-1 (see Table 5-1) which can be
expressed as:

 epb = a • (0.00056) • (^1<3 / (I)1'4 - (source extent) .
                               5-23

-------
It is simplest to perform the  calculations  for  individual waste streams
so that for K061:

 epb * itif • (°-00056)  • (M)1*3  /  (i)1'4  •  5>000 = °-03318 k9  ;
for K052:

             • (°-00056>  •      1'3  '     1'4 •  10° - 5-OE-6 k* 1
Uncontrolled Emission Rates
for K002:
        300    ,n nnncA\    /3.0U.3  /  /20U.4    lc    , cc -, ,,„
 ePb   OE6 '  (°-00056)  '  (o)    '  \~2l     '  15 =:  1'5E'7 kg  '
The total estimated Pb emissions  from unit  operation  I are then given as:

 el = eK061 + e052 + eK002 = °'0332  k9 •

     For unit operation II,  Pb  emissions  are  assumed  negligible because
the temporary cover material is specified as  "clean"  or uncontaminated.
     For unit operation III--dozer activity—the emission rate is
calculated from Eq. 5-2 (see Table 5-1) which can be  expressed as:

 epb  =  a - 0.34 - (s)1'5 / (M)1*4  • (source extent)  .

Given the specifications for source  extent  and a levels, the emission
rate is given by:
 epb «       -  0.34 -  (12.7)1-5 / (19.6)1-4  -  1,566  ; or

 em - 1.12 kg
     For unit operation IV--vehicle traffic— the  emission rate  is calcu
lated from Eq.  5-3 which can be expressed  as:
  "Pb
= a (0.61)  h4j  U|)  (J-T)  *   m *  (—3^5") ' (source extent) .
                               5-24

-------
Given the specifications for source extent and a levels, the emission
rate is then:

       3.000 ,n fin / 8\ / 8\ / 20\0.7 /12\0.5 /320\   9fin . n
 epb = TtoE6 (0-61) (izJ l48/ (TJ)    N    1365/ ' 26° ' or

 eiv = 0.326 kg .
     Total Pb emissions from the landfill unit, then, are estimated as
the sum of values for the individual unit operations.  That is:
 el "*" ell + eIII * eIV =  -     "*" "   "*"  *   +  -; or
                       = 1.48 kg
Thus it is estimated that 1.48 kg of lead is emitted annually due to
landfill operations.

Specification of Dust Control Program
     Suppose that it is determined that the Pb emission rate for the
landfill unit must be reduced to an annual value of less than 1 Kg.  This
result could be indicated for example by the application and results of
the risk assessment methodology presented in Appendix C.  Given the
relative contribution of each of the unit operations to the total
uncontrolled Pb emission rate (i.e., the uncontrolled calculations cited
above) it is apparent that emissions associated with dozer activity (unit
operation III) contribute the largest fraction to the total rate.  In
this sense, it is logical to first examine controls for this activity.
     As indicated in Section 5.3.2, water may be considered as a control
measure for landfill units on a site-specific basis subject to
demonstration that it does not produce undesirable "side effects" (e.g.,
reduction in structural integrity of the lift material).  Both the raw
wastestream moisture content values and the default moisture content for
                               5-25

-------
surface material disturbed by dozer activity (19.6% from Table 5-2) are
relatively high.  This suggests that application of additional water
would not be practical in this example.  The most easily Implemented
control for dozer activity is simple source extent reduction; in effect,
this control represents a change in operating practice.  In the equation
for estimating emissions associated with dozer activity, the emission
rate is directly proportional to the number of hours per day (source
extent) that the vehicle operates on the lift, I.e.,
 e * °-|ffii;-8 x source extent;
where s refers to silt content and M to the moisture content of the
disturbed waste and cover material.  Assuming no change in physical
characteristics, reduction of source extent from 6 h (in the uncontrolled
case) to 3 h (as a control) represents a control efficiency (CE) of 50%
(i.e., a reduction by one-half).  Application of this CE in the emission
factor model produces a corresponding reduction in emission rate as shown
below.

 ePb  (controlled) " « ' °'34 ' <»>1-8 '
                           * 0.34 •  (12.7)1-5 / (19.6)i.* .  (0.50
                  - 0.56 Kg

      Combining the above result  (for unit operation  III — dozer activity)
 with  uncontrolled values for unit operations I (loadout of bulk hazardous
 materials),  II (loadout of cover), and  IV (general vehicle traffic)
 produces  a total annual Pb emission  rate as follows:
  el  + ell  + eIII +  eIV = °-0332 +
                       = 0.859 Kg
                                5-26

-------
Thus, the total Pb emission rate with source extent reduction for dozer

activity produces the required decrease in estimated annual emission

rate.


5.7  REFERENCES FOR SECTION 5

1.   Englehart, P., and D. Wallace.  Assessment of Hazardous Waste TSOF
     Particulate Emissions.  Final Report, EPA Contract No. 68-02-3891,
     Assignments 5 and 13, U.S. Environmental Protection Agency, Research
     Triangle Park, NC.  October 1986.

2.   Environmental Protection Agency.  Compilation of Air Pollution
     Emission Factors (AP-42).  Research Triangle Park, NC.  September
     1988.

3.   Muleski, G., and D. Hecht.  PM10 Emission Inventory of Landfills in
     the Lake Calumet Area.  Final Report, EPA Contract No. 68-02-3891,
     Work Assignment 30.  September 1987.

4.   Muleski, G., F. Pendleton, and D. Hecht.  Chicago Area Particulate
     Matter Emission Inventory—Sampling and Analysis.  Final Report, EPA
     Contract No. 68-02-4395, Work Assignment 1.  May 1988.
                               5-27

-------
                           6.0  LAND TREATMENT

6.1  SOURCE DESCRIPTION
     Unlike disposal practices at hazardous waste landfills which have
not been studied extensively, a large body of literature is available
concerning various facets of land treatment of industrial wastes.1"2  The
American Petroleum Institute (API) has done extensive work on performance
of land treatment systems for refinery wastes.3"1*  API describes land
treatment as a managed technology that involves the controlled applica-
tion of a waste on the soil surface, followed by incorporation of the
waste into the upper soil zone.  The objectives of the land treatment
process are to degrade, immobilize, or transform the constituents of the
applied waste into environmentally acceptable components.
     The size (treatment area) and physical configuration of these units
will, of course, vary from facility to facility and may depend upon fac-
tors such as:
     •    Land availability.
     •    Waste stream physical and chemical characteristics.
     •    Disposal schedule and waste volume.
     •    Soil characteristics.
          Climate.
Table 6-1 presents "model plant" size distributions developed by the
U.S. EPA.s
     Physical configurations for land treatment units may be divided into
three broad categories—single plot, progressive plot, and rotating
plot.  These categories are depicted in Figure 6-1.  In design terms, a
single plot configuration is one in which the treatment area is not sub-
divided, so that the waste is spread "uniformly" over the available
area.  This approach is considered feasible when applications are made
only during one season of the year, or on only a few specific occa-
sions.  A progressive plot configuration is one in which the treatment

                                6-1

-------
area is subdivided into smaller cells, with one or more of the cells
designated as active and the remaining cells as inactive or "fallow."
The latter may be vegetated.  The rotating plot configuration involves
subdivision of the treatment area into plots which are then loaded
sequentially; that is, all plots may be considered active and application
of waste is made at intervals judged sufficient to achieve the desired
treatment effect(s).

    TABLE 6-1.  MODEL PLANT SIZE DISTRIBUTIONS FOR LAND TREATMENT UNITS

            	Land treatment size distributions	
                        Surface area (m2)
ID
ot
                        20th        50th        80th                No.  in
            Minimum  Percentile  Percent!le  Percentile   Maximum   sample
On-site
Off -site
Overall
180
20,000
180
14,000
40,000
16,000
54,000
81,000
57,000
1.5 x IDs
1.2 x 105
1.4 x 105
7.6 x 105
4.0 x 105
7.6 x IQs
34
8
42
     All land treatment operations can be viewed as involving two princi-
pal activities:  (1) waste application followed by (2) incorporation and
subsequent cultivation.  Additionally, some facilities practice soil
amendment (adding agricultural lime or commercial fertilizers to aid the
treatment processes).
     The waste application practices, including frequency of application,
employed at a specific facility have likely evolved to account for the
factors cited above.  Field observations suggest that choice of applica-
tion method probably depends  largely on waste  stream physical character-
istics  (i.e., land  availability, waste stream  characteristics, etc.).   In
cases where the waste streams are  "fluid" or pumpable, a vacuum truck
equipped with a delivery system (hose, spray bar, etc.) may be the pre-
ferred  application  system.  A variation on this approach involves an
intermediate step in which the waste is first  transferred from the vacuum
truck to a conventional  liquid manure spreader.  The spreader is attached
to a tractor and then pulled  across the treatment plot.
                                6-2

-------
Land  Treatment Units-Physical Configurations
                                    A. Single Plot -
                                       Uniform infrequent
                                       application
                                     B.  Progressive Plot --
                                        H Active subplots

                                        D Inactive (fallow)
                                          non-vegetated

                                        BB Fallow vegetated
      Loading Sequence
C. Rotating Plot -
   M Most recent
      loaded subplot
      (t0)

   D Subplot  loaded
      at time t-i

   E3 Subplot  loaded
      at time t-2
  Figure  6-1.  Land treatment units—physical  configurations.
                         6-3

-------
     For wastes with high solids and/or oil and grease (O&G) content,
application necessarily will Involve an Intermediate transfer and spread-
Ing operation.  In a generic sense, this can be envisioned as waste load-
out from a dump or haul truck onto a small portion of the treatment
plot.  The waste would then be spread evenly across a designated area
using a tractor/dozer fitted with a blade.
     Incorporation and cultivation typically are accomplished using
standard farm implements—harrows, plows, rototillers.  Cultivation of
the treatment surface is Intended to promote soil aeration.
     In terms of the potential for particulate emissions from land
treatment units, the most important operating parameters are probably
frequency of waste application, frequency of cultivation, and waste
application rates.  Application frequency will, of course, be site-
specific; frequency values ranging from 2 times per year up to 52 times
per year have been reported for petroleum refining wastes.3  Frequency of
cultivation, or more generally frequency of surface disturbance after
initial waste incorporation, may be expected to vary from one time per
waste application up to as many as four or five, depending upon site
conditions.
     Many terms have been used to identify application rate, that is, the
quantity of oily waste applied to a land treatment site.  Common defini-
tions include:  (a) percent oil in the soil/unit time, and (b) barrels or
tons of oil/acre/unit time.3

6.2  ESTIMATION OF UNCONTROLLED EMISSIONS
     This section outlines the emission factors to be used in estimating
particulate emissions from land treatment units.  The section also pro-
vides available data on RCRA metals concentrations found in the surface
material at active land treatment units.

6.2.1  Particulate Emission Rates
     Aside from vehicle traffic on roads servicing the treatment unit
(see Section 2.2), the principal dust producing operations would
involve:   (1) waste incorporation and cultivation; and (2) waste applica-
tion in cases where the delivery vehicle actually travels on the treat-
ment surface.  Under certain conditions, wind erosion from the treatment
                                6-4

-------
surface may constitute a third potential  dust-emitting mechanism.   How-
ever, this contribution can be expected to be negligible in comparison
with emissions generated by mechanical  activities.
     The emission factors for incorporation and cultivation is as
follows:
                      e = k(5.38)(s)°*6     kg/ha

                      e = k(4.80)(s)°'6     Ib/acre
where:     e » PM10 emission factor in stated units of treated surface
               area
           k = particle size multiplier * 0.21 for PM10
           s = silt content of the disturbed surface material  (%)

For waste application the emission factor would correspond to  travel  on
unpaved surfaces (see Section 2.2) and/or the dozer relationship (see
Section 5.2.1) depending upon the specific nature of equipment used  in
the operation.  These emission factor relationships are reiterated
below.
     For the case where the delivery vehicle travels on the treatment
unit surface, the emission factors are:
             e =
(it)
                                                                    (6-2)
                    1 / s\ / S^ /W\o-7/Wxo.s /365-p\  1KA/MT
                   A te) liff) is/   (4)    her)  lb/VMT
where:    e = PMlo emission factor in units stated
          s = silt content of road surface material,  %
          S = mean vehicle speed, km/h (mi/h)
                               6-5

-------
          W = mean vehicle weight, Mg (ton)
          w = mean number of wheels (dlmensionless)
          p » number of days with £ 0.254 mm (0.01 in) of precipitation

     For the case where the higher solids content wastes are spread over
the treatment area using a tractor/dozer fitted with a blade, the emis-
sion factors are:
                          e -     R       CWO

                                                                    (6-3)

                              0.74
                                 (M)
where:     e = PM10 emission factor in units stated
           s = silt content (%} of the surface material
           M = moisture content (%} of the surface material

     In application of the incorporation and cultivation emission factor
(Eq. 6-1) it is important to recognize the following features:
     1.   The emission factor is based on field source testing of a
          variety of agricultural operations including disking, harrow-
          ing, and land planing.7  It is assumed that waste incorporation
          and cultivation generate particulate emissions at rates compa-
          rable to those found in the source testing.
     2.   In effect, the initial waste application probably should be
          thought of as a particulate control measure, the control effec-
          tiveness of which would decrease through time in a manner
          linked to the rate of decomposition of the waste material in
          the soil.  At present, however, there is no information
                                6-6

-------
          available to specify the relationship between waste decomposi-
          tion and participate control efficiency.  For this reason, it
          is recommended that any initial emissions calculations
          attribute zero control (i.e., uncontrolled) to the waste-
          stream.  If under this conservative assumption, a subsequent
          risk assessment indicates the potential for unacceptable risk
          levels, further in-depth analysis probably will be required.
          Generic requirements for this analysis are provided in
          Section 6.5.2.
     Equation 6-1 requires only a single correction parameter—silt
content of the disturbed material.  Silt content values (determined
according to the procedures given in Appendix 0) for samples collected at
three facilities treating refinery wastes, ranged from 2.0 to 11.3%.
Implicit in the argument above concerning control efficiency versus waste
decomposition, is the idea that silt content of the surface material also
will vary as a function of time after waste application.

6.2.2  Level of Contamination (a)
     Unlike many TSOFs, there are considerable data available to charac-
terize a levels particularly in the case of RCRA metals for treatment
units handling refinery wastes.  Table 6-2 presents data compiled by the
API for several active treatment units.  For comparison, values based on
the initial TSOF survey6 also are included.  Note that the latter units
also were disposing of refinery wastes.
     Specifications of a for organic compounds probably will require
site-specific S&A (see Appendix D for generic procedures).  Alterna-
tively, information on the waste stream characteristics as supplied by
facility operators may form the basis for worst-case estimates of a.  In
the latter case, it may be desirable to multiply waste stream concentra-
tions by a safety factor of 5 to 10 in order to conservatively (i.e.,
overestimate) assign values of a for the treatment surface.

6.3  CONTROL TECHNIQUES
     Although potential control measures for particulate emissions have
not been evaluated under field conditions at specific land treatment
                                6-7

-------








g
LL.

$
§>
3 £
- ^
<£. £
E<
UJ
er
« i-
7 ^
>o
UJ
_I
3
t—





















2
«




S




m
i


£








u



•o
u





in
<:







U)
u
41
4-
4)
ID
(.
(O
^
0
o
u
8
tfl
ID
4-
IO
0
n»
•V ^
1 'i ". CM in \o
' S
CM
i ii °. T m 10
CM tn
O
o
•n 2
*. ' ' 8 3 * *
CM •
<0
o
CM
*™ t^
~" V *o ^
^* S oj c»i ""I i u*» *o
^ °. CN £
o .
0
0 °.
= ? s ? if--

pg ^
K>


CO °.
r, S OS
I^TS* CM *° in «
^ ^ R| ^
S

O
""• "~ 01 01 QQlfllO
in ^ 22
o


"1 «
* £ K, «N*
ro^j.2^ «• « « «
CBO  . .
•n" 0 «o
O 4)
0 C >J
— 3 «
•f: 7 >
U 1 I.
ID 1 in 3 „,
t- >> 4- en A.
Q. u 0 •" ^
+- — in u. — «
*: 2 a. ID Q o. S
S 5 - to _
41 ^ — Q. t— — Q.
EC ID E E — min4) _
i-= >— 10 oo E >-
1-4) —10 XC* _ra
_T •O4- — O T54-
^O c o-o in— COTJ
C^ID - +.(U 3^4- _ ^5
•O^^O "^ -^ O 0) •• « ». o -Sg ^-0^-u
0. ID U O— O4) (-•£ 4) O— Offl
'(.en a. — ID «• +- o> ex —
— 4) — c 'E • —
^fe^<2iSi3 5l€iSiSi8








O)
01
.
o
o
II

1
4-
ID
(.
4-
C
4)
U
O
y

ID
£
_
X

£. 3
a. "a
41 >
•a
t |
1 T3
O U
4-
4- 0
IO 0)
4-
c o
4) -0
j£
ID 4-
4- O
C
in
v ai
a —
E E
ID 3
in m
in
U • ID
O • -0
•^ ^ 4) ^
4) 4- 4)
C 4- O 4-
ID U 4) ID
4) O 4- —
E O. 41 3
41 'O O
U U. —
— 4- ID
4- 4- O O
4) O C
EC C
.C O ID
4- O 4- 4)
- 4- E
L. V)
10 in L. u
U 41 —
T3 0) *- +-
4) -4- 0) 
-------
sites, observations of several active units suggest a number of alterna-
tives for control.  These options are discussed below.

6.3.1  Preventive Controls
     As indicated in Section 6.1, the typical scope of operations at land
treatment units is likely to be more limited than at landfill units.  In
this sense it probably is easier to envision how the implementation of
preventive controls can produce emissions reductions, especially for con-
taminated material.  Like the landfill case, the principal emphasis
should be on source extent reduction—minimizing the spread of contami-
nated waste material from the treatment surface to unit service roads and
adjacent areas.
     A very basic preventive control involves design and implementation
of features to control surface water runoff.  The rationale here is that
under inappropriate design, particulate removed from the treatment sur-
face by high intensity rainfall events may tend to collect in areas sub-
ject to mechanical disturbance.  In turn, this eroded material may con-
stitute a significant source of contaminated particulate emissions.
     A related simple control measure involves revegetation of fallow
areas.  This measure would provide effective wind and waste erosion con-
trol and could support at least intermittent vehicle traffic.  The
applicability of this control may be limited to progressive plot
configurations.
     Most preventive control measures for land treatment units are
similar to those outlined for landfill units.  For example, dedicated
cultivation equipment should be required for land treatment units.  A
rinse station should be operated to clean equipment prior to removal for
any periodic maintenance.
     Other preventive controls include efforts to minimize travel of
cultivation equipment on unit service roads.  This measure is also
applicable to any equipment used to spread higher solids content waste
(e.g., dozer).  Perhaps the most critical preventive control involves
minimizing contact between the treatment surface and waste delivery
vehicles.  The rationale here is simply that a delivery vehicle generally
cannot be restricted to the treatment unit area, and thus efforts to
                                6-9

-------
reduce contact with the treatment surface are required so that material
will not be tracked out and deposited over all facility roadways.
     A final preventive measure involves reducing the frequency of
cultivation.  This measure would in effect reduce source extent (i.e.,
area disturbed in Eq. 6-1), and thus linearly reduce calculated emis-
sions.  This approach might adversely affect waste treatment effective-
ness, and thus the practice must be evaluated on a site-specific basis.

6.3.2  Mitigative Controls
     Application of water probably 1s the only viable mitigative control
for particulate emissions from land treatment surfaces.  As indicated in
Section 6.2.1, the initial waste application may be viewed as a mitiga-
tive control measure, the effectiveness of which would decrease through
time in a manner linked to the rate of decomposition of the waste mate-
rial.  Following this argument, it is apparent that at some time after
waste application, the treatment surface will in effect return to an
essentially uncontrolled state.
     If this time-dependent behavior can be specified (for a given
facility), a watering program could then be instituted to control
particulate for the time interval preceding another waste application.
Alternatively, it might be more practical to simply reapply waste at the
point in time where particulate control is considered to be effectively
zero.
     Note that use of a watering program for a land treatment unit should
require a demonstration much like that cited in connection with landfills
(see Section 5.3.2).  That is, before instituting a watering program, the
facility must demonstrate that addition of water to the treatment unit
does not produce undesirable impacts on ground and surface water quality.

6.4 CONTROL PERFORMANCE ESTIMATION
     A potentially applicable approach to specifying the time-dependent
behavior of the waste material, and  indirectly, its (waste) control
effectiveness is outlined below.  This approach involves the collection
and analysis of surface samples using the S&A procedures outlined in
Appendix D.

                               6-10

-------
     Figure 6-2 provides a schematic of the approach.  In essence, it
requires the collection of representative surface samples from:
     1.   The treatment surface.
     2.   A "background" area outside the treatment surface.
Note that the latter area should be chosen such that its physical and
chemical properties are comparable to those of the treatment surface
prior to any waste application.
     Treatment surface and background samples should be collected at a
minimum of two intervals (i.e., two points in time) between successive
waste applications.  The principal decision point in the procedure
involves whether or not meaningful silt and moisture analyses can be
performed (see Figure 6-2).  Experience indicates that if O&G content in
the treatment surface is relatively high, the material cannot be dry
sieved using the procedures given in Appendix D.  Thus silt content
values needed for Eq. 6-1 would not be available.  In this case, it is
necessary to do O&G content analyses.  Note that there are several pro-
cedures available to characterize O&G; because in the present context O&G
data are used only as a relative measure, the important point is that the
facility use a consistent and fully documented procedure to obtain this
quantity.
     In cases where the silt content can be obtained (a Yes decision in
Figure 6-2), use of Eq. 6-1 provides estimates of the controlled emission
rate.  In cases where O&G content are performed (a No decision in Fig-
ure 6-2), estimates of control efficiency can be obtained using Fig-
ure 6-3.  The latter estimates represent an extrapolation of the data
cited in Section 2.4.3, to cover the cultivation operation.

6.5  PROCEDURES FOR COMPLIANCE DETERMINATION
     This section assumes that RCRA permits specify a set of conditions
or activities that must be provided or undertaken by the facility to
ensure that PM10 emissions are kept to acceptable levels.  Compliance
determinations for land treatment units will then involve field observa-
tions/audits as well as inspection of control implementation records.
Additionally, compliance monitoring may require collection and analysis
                               6-11

-------

4J
•^
V)
V)
8,2
25
l§
< u

s




*4
£
•
§
LU
O>
we
2"-
Ol 4->
> e
10 01
+J
01 C
«/l O
a u
                                                                c
                                                                o
                                                                1C
                                                                o
                                                                o.
                                                                O.
                                                                id



O)
•^
0)
** 4«>
o> e
•o 01
*J
o e
•M O
U
I/I
0)
•5.S
1°
i/i
e
O) IQ
(/I O>
= I








•w
S




PO
ft 1
• I/I VO
^ *^
§S£
24^ 3
e ei
4J Ol O f
e j^ u LU
01 u
•M (O 4) ^
cZ C c
O » 3 «
U 4J
£ I/I »—•
Cfl 4J -^ i— (
•B 'p* O 1
o i ivo
e i. -o""*
(O 0) C •
£ .e re c
E -M o-
4) •*-> LU
0) O)P-
Irt O -P- C
3 +J (/)•!-
                                                                O)
                                                               •«->
                                                                id

                                                               "o
                                                                o
                                                                (/i
                                                                i/i
                                                               o

                                                               *J
                                                               c
                                                               o
                                                               u

                                                               en
                                                               id
                                                               >
                                                               a;
                                                               id
                                                              C-sJ


                                                              SO


                                                               O)
6-12

-------
      100^.
      75	
u.
LU
o
u
ID
       50-
      25
                                                                             95%
                    2            345



                      RATIO OF 0&6 CONTENT TO BACKGROUND


                          SOIL MOISTURE  CONTENT
        Figure 6-3.   Procedure for estimating control  efficiency with

               O&G content and moisture content measurements.
                                    6-13

-------
of surface samples of dust-emitting materials using the procedures out-
lined 1n Appendix D.

6.5.1  Preventive Controls
     Field observations are the principal basis for determining whether
or not the preventive controls cited as permit conditions are in fact
followed.  Table 6-3 provides an example checklist form for audits of
land treatment preventive control practices.  There are several points
concerning the checklist that should be noted including:
     1.   The length of time for field observation/audit as well as audit
          frequency should be sufficient to view "typical" operations.
          This certainly will vary according to the factors cited in
          Section 6.1.  In general, a set of field observations should be
          scheduled to view two points (in time) of the application
          treatment cycle.  The first point should correspond to an
          actual application.  The principal concern here is to document
          that waste spillage and trackout problems associated with
          application are minimal.  The second point (in time) should
          correspond to the last scheduled cultivation of the treatment
          surface prior to waste application.   The principal  concern here
          is to assess whether or not the waste applied at the beginning
          of the cycle is still serving as a control measure.  This can
          be done in a qualitative way simply by determining whether or
          not visible emissions are being released from the surface by
          the cultivation activities.
     2.   As a minimum program, at least two sets  of field observations
          should be conducted annually.  One set should correspond to
          hot/dry conditions  (summer through early fall in most parts of
          the country), while the other to a cooler period in which sur-
          face moisture and the O&G content of the surface may be
          expected to remain  high for longer time  intervals.
     3.   To the extent possible, the field audits should not be sched-
          uled  in advance so  that a "clear picture" of typical practices
          is obtained rather  than one potentially  biased by advance
          notice.
                                6-14

-------

o
I
o.
1
\~~
o
o
UJ
2*
t—
^
UJ
a?
O.
K-
i
^,
^^
•^j
~~
*™
a
i
^-
o
U»
h-
(/>

~~
^^
UJ
O
.^
;"^
a
™3
^^

o
t _
«?
^^
*s
^1
~^
§

"•
a.
X
UJ

CO
1
I
VJJV
I , j
U»l
J
J
CO
t—

£1

toi
ail





!o
£_
c
3
4->
C
4J
id
QJ
^J

QJ
.c
,— . ^j
01
¥ f
-(-> id
QJ
O *~
t- to
id **"
> ^J)
I- C
QJ QJ
oi E
JD a.
O n-
cr
a>
CM
•a
QJ

K3
^f
•D
HI
13

+J
id
j=
4-)
QJ

QJ
T3
C >
•^ ^
O
Q. QJ
C (U
O «• .C
*•" O> 4->
4H> 4_>
id id co
U O "-H
*Q.
a «-t ro






























C"
id
o

-*->
id
s-
Ol
Q.
0
C
0




(V.
•o
id
2
0)
u
*J
t-
O)

o
o
O)
«o
id
2

<4_ x"'».
O 1/1
en 01
_C ^E
•C 4->
id
Ol C
J- O
a. T-
03
^ >
O J-
•^ 01

id o
Ol
cr>
"O ^V
^»
•^"
a.








1%
^•*
C

4_>
id
0)
i *
^»
O)
£
4->

^
01
>
o
I/I
4->
O)

^
a.
^E

C
o

4->
Id
c •>-
O 4->

4-* C 3
id O U
U i-
••- *J JC
r— Id 4-»
in •• CL o ••—
•0 Q. i- 2
Ol O)
4^ 4^
id i—
o. «/*















C"
^
4->

>
•r—
4->
0
id
C
o
4->

>
4->

3
U
O)
J=
4->

>k
-°
^^
01

id
Ol
c
t- IO id O> Q. Ol O)
4^)
id
4H>
o>

id o) id
t - * ** DL
O id 01
O> r- (d "O
Ol i-
u id
c
O) <*-
•a o
> c c
O) i- O
o •<-
O) CU 4^
i- a
0) C t-
^ o •• v-
4-> i- Ol L>
4-* 4^ (/
1/1 id  o a
^f . .
LO 3 ^O OC
« 2 01
i- >
4-> 4-> S-
i- X Ol
C Ol CO
•^ C -Q
O

(J -1- «4-
C 4-> O
T- C O)
oi id
Ol Ol *""> «|
E E E S-
•t- ^- 33
t— H- Z CO
• • •
O^ O *"H
t—* »-H
in
c
o
(O
t/1
*E
Ol
O)

JD
42
^

Ol

•*
•
CM
«— 1
6-15

-------
6.5.2  Mitlqatlve Controls
     If under the conservative assumption of no control associated with
waste treatment (i.e., as in Section 6.2.1), calculated risks due to
particulate emissions are unacceptably high, development of site-specific
data may be warranted.  The objective here is to indirectly characterize
the potential mitigative effects of waste stream addition throughout the
course of the typical waste treatment cycle.  Generic requirements for this
approach are described below.
     The unit operators must fully document the following items for a
time period sufficiently long (- 1 yr) to adequately characterize the
expected range in unit operations:
     •    The date of each waste application.
     •    The type and amount of each waste stream applied to given
          treatment area.
Note that because a variety of terms are used to characterize application
rate (see Section 6.1), the facility operators are required to provide
conversion data so that data can be expressed as O&G (%) in the soil/unit
time.
     The above data defines length of the treatment cycle (t).  At  this
point, the facility must demonstrate that for the maximum period (tm)
given in the documentation, cultivation produces no visible emissions.
This will require field observation similar to that recommended under
Section 6.5.1.  Note that the field observations should not be made
immediately after significant rainfall.  If visible emissions are
observed at tm, a similar demonstration should be required for the
interval tm/2.  The interval should be "halved" until the criterion of no
visible emissions is met.  In effect, based on the conservative criterion
of no visible emissions, the above procedure provides a simple, qualita-
tive way to establish the length of time over which waste application
provides effective mitigative control.
     Following this argument, the established cycle length represents the
point at which the facility needs to  institute a mitigative control mea-
sure.  This mitigative  control may simply be waste reapplication, or
alternatively, a watering program could be  instituted to cover the  time
period until waste reapplication.

                               6-16

-------
     It 1s Important to recognize  that  the  above approach addresses only
risks associated with air emissions.   It  1s possible that the cycle
length (t) indicated by this approach may not  be compatible with other
requirements designed to safeguard ground and  surface water quality.

6.6  EXAMPLE CALCULATION
     An example application of the emissions estimation procedures for
land treatment units is presented  below.  Note that this calculation uses
the no control assumption discussed in Section  6.2.1.
     The hypothetical unit 1s taken as  a  generator facility treating
refinery wastes (K048,K051) and located 1n  a continental climate.  The
facility operates a 6 hectare (- 15 acres)  treatment unit in a rotating
plot configuration (six plots). Wastes are applied to each subplot at
approximately 2 month intervals.  A vacuum  truck with spray bar is used
for application; the subplots typically are cultivated four times after
initial waste incorporation at approximately 2 week intervals.
     Based on information supplied by  the facility, it is determined that
chromium (Cr) is the principal constituent  of  concern, and that Cr levels
in the treatment surface can be conservatively (high) estimated at
4,000 ppm (ug/g).  Note that this  includes  a safety factor of 5 as
suggested for worst case estimates (see Section 6.2.2).
     The annual source extent for  the  waste application operation is
estimated as 120 vehicle kilometers traveled (VKT).  This figure assumes
a uniform 3-m spray pattern from the vacuum truck, and that the wastes
are applied to square treatment plots  at  a  frequency of once every
2 months.  The annual source extent for cultivation is estimated at
144 hectare (ha) given four cultivation steps  after initial waste incor-
poration.  Note that it is assumed that each of the cultivation steps
consists of a single pass over the plot(s).
     The Cr emissions associated with  waste application are given by the
expression:
                       (it) (w) (o)°-7 (?rs  <•««• •**•*>
                               6-17

-------
Inserting the appropriate values yields the expression:

           p   = 4,000 ,n „. /10\ / 8\ / 8 \o.7 /I0\o.s
           RCr " ltoT6 (0*61) II2J \48) (zTr)    (~4J    (120)

Thus, Cr emissions for the application operation are estimated as
0.1375 kg.
     The Cr emissions associated with cultivation are given by the
expression:

               RCr = o(0.21) (5.38) (S)°-« (source extent)

Inserting the appropriate values yields the expression:

                 RCr

Thus, Cr emissions for the cultivation operation are estimated as
2.59 kg.
     Total Cr emissions from the treatment unit (exclusive of service
roads) is then estimated as the sum of the two unit operation values:

                     RCr = 0.1375 + 2.59 = 2.7275 kg

In other words, it is estimated that the unit emits 2.7276 kg of Cr
annually.

6.7  REFERENCES FOR SECTION 6
1.   Overcash, M. J.t and D. Pal.  Design of Land Treatment Systems for
     Industrial Wastes—Theory and Practice.  Ann Arbor Science
     Publishers,  Inc., Ann Arbor, Michigan.  1979.  684 pp.
2.   Brown,  K. W., 6. B. Evans, and B. D. Frentrap.  Hazardous Waste Land
     Treatment.  Ann Arbor Press, Ann Arbor Michigan.  1983.  692 pp.
3.   American Petroleum Institute.  Land Treatment Practices in the
     Petroleum Industry.  Washington, DC.  June 1983.
                               6-18

-------
4.   American Petroleum Institute.  The Land TreatablHty of
     Appendix VIII Constituents Present in Petroleum Industry Wastes.
     Document B-974-220, Washington, DC.  May 1984.

5.   GCA Corporation.  Preliminary Model Plants for Treatment, Storage,
     and Disposal Facilities.  Draft Technical Note, EPA Contract
     No. 68-01-6871, Assignment No. 13, prepared for U.S. Environmental
     Protection Agency, Research Triangle Park, North Carolina.

6.   Englehart, P., and D. Wallace.  Assessment of Hazardous Waste TSDF
     Particulate Emissions.  Final Report.  EPA Contract No. 68-02-3891,
     Assignments 5 and 13.  U.S. Environmental Protection Agency,
     Research Triangle Park, North Carolina.  October 1986.

7.   Environmental Protection Agency.  Compilation of Air Pollution
     Emission Factors (AP-42).  Research Triangle Park, North Carolina.
     September 1988.
                               6-19

-------
                         7.0  WASTE STABILIZATION

7.1  SOURCE DESCRIPTION
     In this section the basic types  of waste stabilization processes are
discussed along with resulting PM10 emissions.   Pozzolanic processes are
emphasized based on a 1986 survey of  TSDF facilities.

7.1.1  Site Characteristics
     "Solidification" and "stabilization" both  refer to treatment systems
designed to accomplish one or more of the following:  (a)  improve han-
dling and physical characteristics of the waste; (b) decrease the surface
area across which transfer or loss of contained pollutants can occur;
and/or (c) limit the solubility of, or to detoxify, any hazardous con-
stituents contained in the wastes.  Solidification implies that these
results are obtained primarily (but not exclusively) via the production
of a monolithic block of treated waste with high structural integrity.
Stabilization techniques are beneficial because they limit the solubility
or detoxify the waste contaminants, even though the physical characteris-
tics of the waste may or may not be changed or  improved.  Stabilization
usually involves addition of materials that ensure the hazardous con-
stituents are maintained in their least soluble and/or toxic form.
     The term "fixation," has fallen  in and out of favor but is widely
used in the waste treatment field as  generally  meaning any treatment sys-
tem which solidifies and/or stabilizes the waste as described above.
This is the restricted use of the term as contained in this document.
     "Surface encapsulation" is a technique of  waste treatment involving
isolation of the waste material by placing a jacket or membrane of
impermeable, chemically inert material between  the waste and the environ-
ment.  Ideally the jacket is bonded to the external surface of a solidi-
fied waste.  Encapsulation of small particles is sometimes called "micro-
encapsulation," but this term is used by processors to describe a wide
array of different techniques and therefore has no specific meaning.
                                7-1

-------
     There are seven major categories of industrial waste stabilization
systems:l
     1.   Cement-based processes.
     2.   Pozzolanic processes (not including cement).
     3.   Thermoplastic techniques (including bitumen, paraffin, and
          polyethylene incorporation).
     4.   Organic polymer techniques (including urea-formaldehyde,
          unsaturated polyester).
     5.   Surface encapsulation techniques (jacketing).
     6.   Self-cementing techniques (for high calcium sulfate sludges).
     7.   Classification and production of synthetic: minerals or
          ceramics.
Each will be described briefly below.

     7.1.1.1  Cement-Based Processes.  Five types of Portland cement are
generally recognized, based on variations in their chemical composition
and physical properties:1
     1.   Type I is the typical cement used in the building trade, and
          constitutes over 90% of the cement manufactured in the United
          States.
     2.   Type II is designed to be used in the presence of moderate
          sulfate concentrations (150 to 1,500 mg/kg) or where moderate
          heat of hydration is required.
     3.   Type III has a high early strength and is used where a rapid
          set is required.
     4.   Type IV develops a  low heat of hydration and is usually
          prescribed for large-mass concrete work but has a long set
          time.
     5.   Type V is a special low-alumina, sulfate-resistant cement used
          with high sulfate concentrations (> 1,500 mg/kg).
The types that have been used for waste solidification are Type I and, to
a  smaller extent, Types  II and V.
     Most hazardous waste slurried in water can be mixed directly with
cement,  and the suspended solids will be incorporated into the rigid
matrices  of the hardened concrete.  This process is especially effective
                                7-2

-------
for waste with high levels of toxic metals, since at the pH of the cement
•ixture, most multivalent cations are converted into insoluble hydroxides
or carbonates.  Metal ions may also be incorporated into the crystal
structure of the cement minerals that form.  Materials in the waste such
as sulfides, asbestos, latex, and solid plastic wastes may actually
increase the strength and stability of the waste concrete.

     7.1.1.2  Pozzolanic Processes (not containing cement).  Waste fixa-
tion techniques based on lime products usually depend on the reaction of
lime with a fine-grained siliceous (pozzolanic) material and water to
produce a concrete-like solid (sometimes referred to as a pozzolanic con-
crete).  The most common pozzolanic materials used in waste treatment are
fly ash, ground blast-furnace slag, and cement-kiln dust.  All of these
•aterials are themselves waste products with little or no commercial
value.  The use of these waste products to consolidate another waste is
often advantageous to the processor, who can treat two waste products at
the same time.  For example, the production of a pozzolanic reaction with
power plant fly ash permits the flue gas cleaning sludge to be combined
with the normal fly ash output and lime (along with other additives) to
produce an easily-handled solid.

     7.1.1.3  Thermoplastic Techniques (including bitumen, paraffin, and
polyethylene).  The use of thermoplastic solidification systems in radio-
active waste disposal has led to the development of waste containment
systems that can be adapted to industrial waste.  In processing radio-
active waste with bitumen or other thermoplastic material, the waste is
dried, heated, and dispersed through a heated plastic matrix.  The mix-
ture is then cooled to solidify the mass, and it is usually buried in a
secondary containment system such as a steel drum.  Variations of this
treatment system can use thermoplastic organic materials such as paraffin
or polyethylene.
     The above process requires some specialized equipment to heat and
mix the waste and plastic matrices, but equipment for mixing and extrud-
ing waste plastic is available.  The ratio of matrix to waste is gen-
erally quite high--a 1:1 to 1:2 ratio of incorporation material to waste
                                7-3

-------
(on a dry-weight basis).* The plastic in the dry waste must be mixed at
temperatures ranging from 130°C to 230°C, depending on the melting char-
acteristics of the material and type of equipment used.1

     7.1.1.4  Organic Polymer Processes.  Organic polymer techniques were
developed as a response to the requirement for solidification of waste
for transportation. The most thoroughly tested organic polymer solidifi-
cation technique is the urea-formaldehyde (UF) system.  The polymer is
generally formed in a batch process where the wet or dry wastes are
blended with a prepolymer in a waste receptacle (steel drum) or in a
specially-designed mixer.  When these two components are thoroughly
mixed, a catalyst is added, and mixing is continued until the catalyst is
thoroughly dispersed.  Mixing is terminated before the polymer has formed
and the resin-waste mixture is transferred to a waste container, if
necessary.  The polymerized material does not chemically combine with the
waste—it forms a spongy mass that traps the solid particles.  Any liquid
associated with the waste will remain after polymerization.  The polymer
mass must often be dried before disposal.

     7.1.1.5  Surface Encapsulation Techniques (jacketing).  Many waste
treatment systems depend on binding particles of waste material
together.  To the extent to which the binder coats the waste particles,
the wastes are encapsulated.  However, the systems addressed under sur-
face encapsulation are those in which a waste that has been pressed or
bonded together is enclosed in a coating or jacket of inert material.  A
number of systems for coating solidified industrial wastes have been
examined.

     7.1.1.6  Self-Cementing Processes.  Some industrial wastes such as
flue-gas cleaning or desulfurization sludges contain large amounts of
calcium sulfate and calcium sulfite.  At least one technology has been
developed to treat these types of wastes so that they become self-
cementing.  Usually a small portion (8% to 10% by weight) of the
dewatered waste sulfate/sulfite sludge is calcined under carefully con-
trolled conditions to produce a partially dehydrated cementitious calcium
                                7-4

-------
sulfate or sulfite.  This calcined waste is then reintroduced into the
bulk of the waste sludge along with other proprietary additives.  Fly ash
is often added to adjust the moisture content.  The finished product is a
hard, plaster-like material with good handling characteristics and low
permeability.

     7.1.1.7  Classification and Production of Synthetic Minerals or
Ceramics.  Where material is extremely dangerous or radioactive, it is
possible to combine the waste with silica and either fuse the mixture in
glass or to form a synthetic silicate mineral.  Glasses or crystalline
silicates are only very slowly leached by naturally occurring waters, so
these waste products are generally considered to be safe materials for
disposal without secondary containment.  There are no known applications
of glassification processes currently in operation.i
     Of those described above, pozzolanic processes will be emphasized in
the following sections.  Other stabilization processes should be evalu-
ated on a site-specific basis using the basic information contained in
Cowherd et al.2

7.1.2  Emission Sources
     The exact sources and character of the particulate emissions
generated by waste stabilization operations depend on the type of process
and disposal method used.  However, in general, pozzolam'c waste stabili-
zation processes involve the following basic sequence of operations:
(a) initial load-out of the liquid material into a mix bin; (b) addition
of solid material to the bin; (c) mixing of liquid and solid material;
and (d) removal of the mixture (i.e., stabilized waste) from the bin.3
At some sites, after mixing, the material is transferred from the bin to
off-highway trucks by an excavator.  The off-highway trucks then trans-
port the material to the landfill.  At other sites, material is trans-
ferred directly from the bin to temporary storage piles adjacent to an
active landfill surface.  Landfill equipment then spreads the material
over the landfill surface.  A general flow diagram of pozzolanic stabili-
zation processes is shown in Figure 7-1.
                               7-5

-------
I
•§
 m
53
                                                55
                        £
5T
I
 a
 r
                I
                         CO
                        I
                        .ff
                                                                        (0
                                                                       GC
                                                       o
                                                                               to
                                                                               to
                                                                               a>
                                                                               u
                                                                               o

                                                                               o.
                                                                                M
                                                                                N
                                                                                O
                                                                                ex
                                                                                10
                                                                                s-
                                                                                O)

                                                                                (U
                                                                               o
                                                        cu

                                                        en
               7-6

-------
     In the above process flow, fugitive PM10 can be generated during the
transfer and handling of solid materials prior to being mixed with the
liquid waste, during transfer and handling of the stabilized waste, and
from truck and loader traffic operating near the process equipment.  The
surface over which the vehicles operate may either be paved or unpaved as
the case may be.
     The extent of the fugitive emissions generated from stabilization
processes depend to a major extent on whether the actual mixing of liquid
and solid materials occur within a building or enclosure.  If the process
Is completely enclosed, occupational exposure would be of greater concern
than that of emissions to the atmosphere.  However, spillage of stabi-
lized waste from vehicles onto travel surfaces can be of concern regard-
less of whether the process is enclosed or completely or partially
exposed to the atmosphere.

7.2  ESTIMATION OF UNCONTROLLED EMISSIONS
     The estimation of uncontrolled emissions from waste stabilization
processes involves both the determination of PM10 emission rates from the
dust-generating source(s) and their associated level(s) of contamination.
Both topics are discussed below.

7.2.1  Participate Emission Rates
     Estimation of uncontrolled particulate emissions from waste stabili-
zation follows the general model (Eq. 1-2) presented in Section 1.  For
convenience, it is reiterated below with terms potentially applicable to
stabilization units:
where  Rn-  = emission rate of contaminated airborne particulate (kg/yr)
             for a given stabilization process,  consisting of n
             identifiable unit operations
       o-jj = fraction of contamination in particulate emissions for the
             jth stabilization operation (yg/g)
                               7-7

-------
       e^j * emission factor(s) (mass/source extent)
       A.JJ = source extent(s) (source dependent units)

For waste stabilization, the n unit operations correspond to those out-
lined in Section 7.1—transfer and handling of solid raw materials;
transfer and handling of stabilized waste; and vehicle traffic.  Wind
erosion emissions are considered to be negligible in the vicinity of
waste stabilization processes.
     Subject to the limitations/uncertainties discussed in Section 1,
Table 7-1 provides an emission factor (E^j) for each of the unit opera-
tions that may occur at or near a waste stabilization process.  Except
for the a^j term, there are certain aspects of each emission factor that
should be noted prior to estimation of PM10 emissions for a given
source.  These aspects are discussed below.
     For the transfer and handling of solid raw materials, the materials
used to mix with the waste (e.g., fly ash, cement kiln dust, etc.), gen-
erally consist of fine particles which fall outside of the range of silt
values for the load-in/load-out emission factor equation published in
Section 11.2.3 of AP-42.*  Therefore, it is recommended that the PMi0
emission factor developed by Muleski et al. for the loading of fly ash
into open trucks be used instead.5  Although this emission factor is not
currently published in AP-42, it would be expected that the nature of the
material and dust-generating process more closely represent those used in
waste stabilization.
     For the transfer of stabilized waste either into haul trucks or into
temporary storage, the load-in/load-out emission factor equation pub-
lished in Section 11.2.3 of AP-42 can be used.1*  Since this factor was
originally developed for dry aggregate materials, it should only be used
with caution.  An "A" quality rating is applicable to materials with silt
contents between 0.44 and 19% and moisture contents of 0.25 to 4.8%.  If
the characteristics of the stabilized waste do not conform to these
ranges, the quality rating is reduced by one level.  It is recommended,
therefore, that waste-specific data be used in lieu of the "default"
parameters  listed in Table 7-1.
                                7-8

-------









1
•w
_*
i
s
i
§

a
ON PMUHETE
*—
u
u
o
t/>
s
u.
o
 K 8
S £ o
Z K
^
0 g
Of —
U M



u
o
•M
g
VI
VI
5
o
S













a
i


88
77 < < 7S
M X ' Z M X




***• M
S*fa ?HS Sii'
«r csj PW ^  oj

il tl ii ll
in I *o — * •*•»• — •>
SI f^ (^ ^ w
11' I
in * ~ ~ ~
o*
s
o
1 Ca^ > J<
• 5" ^ ^
v»ico **> ""T""
O 9 *[ t^
• *^ x ^ ***
"•12 S. " 2J ^

• "A in a
s ^ a I g
o -? o o e




1 £
4J g
5 S « 2
fe » g
«* a 2 2
1 - - i T
v v n e
Z S 2 ? .2 i
> ^ S ? '- J
5 <2 5 2 y< %


• v '
. Is fe
X ~-J! S
M Ot C
s . S & !
1? ** g
li 8! s
S M <• O
i] ]i |
— X « < *^
u a. S
\l {j I
31 U. O •
S ^^ • * ^
«» +•» » O *
8 I c
• «•» 0 *
ii I! J
£2 * "e> 7
« r S g i
* fc 1 f •
U 4^ «f
!c 5 o £ «S
S " S • 1'
g * 2 g 1 '
i"? S - | '

5. 1 ^ § z ilf.
•o "" *o Z > *•- :
$ 4-> 0 « O
4b ^— ^~
MM ftl O k C
41 N M £ * • *•
— VI « *
W -1 S _ ** g
"£ .. J? "c "° S
IN* • *** K
J^ «— k "y •
« • ^ £ • 4
k > 5. •-
« a *• c M *
O» * IU I <.
VI -^ 41 *»
.. t^. r vo* £ * .
«•— • tU • M *
Me 5 • "
5 • o w
• "^ & • C
4^ *^ <^ • •
*^ 4J 4^ f"
« • '
If 1*1 J°
00. 2 V4 > 3
w u *rf x
c c • ••
s - .. 1 2 .. "
?s sis s.
•» is« ii
MA (/i 5 • 1/1 :
o .a o •







|
*
£
c
1
1
M
W
4^
M
1 character
tv
^
• Si
I 1
i i
•» >*
-• i
o -a
•• o
• ft
J £

i s
i 1
>
i 2
•» w
§ -

S f
' S

.
?t trained by di
3 ,fi
7-9

-------
                                                            go"
                                                            c«»-

                                                           Ss~
                                                                  o»
                                                              - o ia

                                                                  >-
                                                               ••- o
                                                                  -0
                                                            ^ « C
                                                            *
                                                               C  •
                                                            ±1 X
                                                             di
   CD
g^
                                                             c  c -a
                                                            _  1-  C
                                                                
-------
     Finally, for vehicle traffic operating near the stabilization
process, the activities are considered to be similar to those tested
during the development of the paved and unpaved road equations published
in Sections 11.2.1 and 11.2.6 of AP-42.*  It should be noted, however,
that unlike generic roads, the emissions from vehicle traffic near stabi-
lization processes can be contaminated with organic compounds and/or RCRA
metals.
     Also included in Table 7-1 are material "correction parameters" for
each of the above emission factors.  One column in the table provides
those material characteristics used in the derivation of the PM10 emis-
sion factor, and the second (and final) column presents available data
from a recent survey of TSDFs.3  These final values can be used as
default parameters in lieu of source-specific data, if necessary.  Please
note, however, that the default values provided in Table 7-1 are based on
very limited measurements and thus should not be considered indicative of
the industry as a whole.

7.2.2  Level of Contamination (a)
     For waste stabilization processes, a depends to a major extent on
the types of liquid waste being treated and the type of process being
used (Section 7.1.1).  Because of the great variability from site to
site, it is recommended that site-specific sampling and analysis be
performed to characterize contamination levels of the source material.
The following discusses potential contamination as related to the major
dust-emitting sources found in stabilization processes.  Results of a
recent survey are also provided to supply some indication as to the
contamination found in the two stabilization processes surveyed.
     With regard to potential contaminated emissions created during the
transfer and handling of solid raw materials (prior to being added to the
waste), it would be expected that the PMi0 emissions would contain only
those compounds inherent in the solid bulk material.  As these materials
are generally not considered to be hazardous, it would be expected that
the resulting fugitive emissions should not be contaminated to any
significant extent.  This is not the case, however, during handling and
                               7-11

-------
transfer of the stabilized waste prior to disposal and for vehicular
traffic operating in the vicinity of the process.
     Fugitive PM10 generated during the handling of stabilized waste can
be contaminated with various organic compounds and/or RCRA metals.  For
example, during a limited survey of two TSDFs, an average total concen-
tration of RCRA metals and semivolatile organic compounds in the samples
collected was found to be 919 and 103 yg/g, respectively.3  These concen-
trations represent average values obtained from one sample taken in the
vicinity of the process equipment at one facility and one sample of
stabilized waste collected at a second site.
     Contaminated emissions can also be generated by truck and loader
traffic operating in the vicinity of the stabilization process.  Con-
tamination of paved and unpaved surfaces can result from spillage during
handling and transport of the stabilized waste which is subsequently
reentrained as fugitive dust.  Contamination can also result from migra-
tion of contaminated material from other on-site treatment areas which is
transferred to and from plant roads, etc.  This has been discussed previ-
ously in Section 2.2.3 of this manual.
     Finally, Table 7-2 provides data for RCRA metals determined in a
survey conducted in 1986.3  Again, it should be cautioned that the values
presented in Table 7-2 are based on limited sampling and thus should not
be considered indicative of the industry as a whole.  Also note that the
data are for silt fractions only since this parameter is used in most
AP-42 emission factor equations for fugitive sources.1*

7.3  CONTROL TECHNIQUES
     While there has been no actual field testing of particulate control
techniques at HW TSDFs, it is reasonable to assume that much of the
experience and information developed in more traditional industries
 (e.g.,  iron and steel) can be applied.  In practice, however, complica-
tions may arise because of competing (and potentially contradictory)
objectives in controlling overall particulate emissions vs. contaminated
fractions thereof.
                                7-12

-------
          TABLE  7-2.   SUMMARY OF  RCRA METALS  CONCENTRATION  FOR
                       WASTE  STABILIZATION  SAMPLES3
                                 Concentration 1n sample (ug/q)b
RCRA Metal
Arsenic (As)
Barium (Ba)
Cadmium (Cd)
Chromium (Cr)
Lead (Pb)
Mercury (Hg)
Selenium (Se)
Silver (Hg)
Unpaved surface
11.3
191
2.1
67.4
114
0.8
< 1
< 10
Stabilized waste
14.8
202
18.8
337
866
0.11
1.7
11.5
        aFrom Reference 3.
        Concentration of specific RCRA metal in silt (< 74 ym
         physical diameter) fraction.  < indicates below detection
         limit.

7.3.1  Preventive Controls
     Preventive controls, as applied to waste stabilization, consists of
two basic approaches:  work practice controls, and reduction in source
extent.  Work practice controls include those measures which reduce emis-
sions potential either through a modification of the process or its
operation.  Reduction in source extent implies some decrease in the level
of process operation which lowers either gross or contaminated PM10 emis-
sions to the atmosphere.
     The type of preventive controls used at a specific facility depend
to a large extent on the individual process and its operation with
respect to other activities occurring on-site.  The following are pos-
sible preventive measures to reduce fugitive dust emissions from waste
stabilization processes:

     7.3.1.1  Transfer and Handling of Solid Raw Materials.
     •    Work practice controls;  The principal method would be to use
          elevated hoppers or silos to store and dispense solid stabi-
          lization materials to reduce emissions potential  during
                               7-13

-------
     load-in/load-out and to eliminate loader travel  to and from the
     process area.   Finely divided materials should be pneumatically
     or mechanically conveyed to fill  hoppers or silos instead of
     using some type of wheeled or tracked loader.
•    Source extent  reduction:  Restrict vehicle traffic in process
     area using work practice controls indicated above.
•    Comments:   Although the emission  factor for raw materials
     handling and transfer (Table 7-1) is not directly related to
     the particle size of the material, finely divided stabilizers
     would tend to  have a higher emissions potential  than coarser
     products.   It  is recommended that coarse raw materials be used
     whenever possible.

7.3.1.2  Transfer and Handling of Stabilized Waste.
     Work practice  controls:  The use  of containers (e.g., roll-
     off s) for the  transfer of stabilized waste to haul trucks or
     for temporary  storage near the process area is recommended.
     Enclosure of mix trough or bins is also suggested to reduce
     entrainment of finely divided materials during mixing of solids
     and liquid waste.
     Source extent  reduction:  Reduce  number of transfer operations
     which stabilized waste must undergo prior to disposal.
     Comments:   Elimination of temporary open storage of stabilized
     waste would be expected to reduce handling emissions as well as
     control the transfer of contamination to travel  surfaces and
     migration to other parts of the facility.  Containerization
     would also eliminate the use of wheeled or tracked loaders with
     their associated travel across potentially contaminated paved
     and/or unpaved surfaces (see below).

 7.3.1.3  Vehicle Travel in Process Area.
     Work practice controls:  The main preventive technique is to
     pave unpaved surfaces in the general vicinity of the process
     for delivery of raw materials and removal of stabilized
     waste.  For paved  surfaces, any method to reduce spillage and
                          7-14

-------
          improve general housekeeping to minimize silt loadings is
          recommended.  Speed limitations for all vehicles can also
          reduce emissions.
     •    Source extent reduction;  Eliminate the use of loaders by
          automation of the process operation.  Reduction or elimination
          of unnecessary traffic in process area is also suggested.
     •    Comments:  If loaders are used, this equipment should be
          "dedicated" to the stabilization process and not used for other
          operations within the facility.  By restricting the use of
          equipment to one operation, the spread of contamination is
          limited.
     It should be noted that the measures listed above are by no means
all-inclusive but are intended to provide guidance as to the general
types of practices which could be used to reduce emissions from stabi-
lization processes.  Also, the techniques listed are based on a limited
survey of stabilization processes performed prior to the issuance of RCRA
Part B permits for TSDFs.3  It would be expected, therefore, that a num-
ber of the preventive techniques described above may have already been
implemented in currently operating systems.

7.3.2  Mitiqative Controls
     In addition to preventive measures, certain mitigative controls can
also be applied to reduce emissions from waste stabilization processes.
As was stated previously, these measures are essentially identical  to
those applied at other industrial facilities to control  gross PM10.
Caution should be exercised, however, to apply the control  in an appro-
priate manner such as not to exacerbate any potential for further con-
tamination of source material (and the resulting PM10 emissions).  The
following describes mitigative control techniques applicable to waste
stabilization.
     For transfer and handling of solid raw materials, a number of
control options are available depending on how the waste is processed.
If open storage is used, the only viable alternative is  the use of  wind
screens or barriers to provide a sheltered area for transfer and handling
of the solid materials.6  Either portable screens,  permanent barriers, or

                               7-15

-------
some type of partial enclosure (e.g., three-sided bunker) could be used
for this purpose depending on site conditions.  It should also be noted
that wet suppression is also available to control fugitive PM10 from raw
materials handling but this technique may not be compatible with either
the liquid waste or the stabilization process being used and thus is not
generally recommended.
     If a silo or hopper is used to store and dispense solid raw
materials, a traditional capture/collection system could be used to con-
trol the PM10 generated.*  As the name implies, capture/collection sys-
tems consist of some form of hood or other ventilated device to capture
the dust emissions from the source with the suspended particles being
removed from the air stream using some type of dust collector (e.g.,
fabric filter).  These systems are reasonably expensive, but, if properly
designed and maintained, are also very effective in reducing dust emis-
sions.  Additional information on the design and operation of capture/
collection systems can be found in Cowherd and Kinsey.6
     With regard to the transfer and handling of the stabilized waste,
the number of control alternatives is limited.  In general, the waste is
relatively moist with little dust emission potential as it is initially
generated by the process.  The greatest dusting potential is present when
the waste begins to dry prior to disposal.  If temporary storage piles
are used, wet suppression using water or water plus a surfactant could be
employed to control dust emissions.  A similar system could also be
utilized during the loading of waste into haul trucks, if required.  The
major emissions potential during transfer and handling of stabilized
waste really comes from vehicle traffic in the process area as discussed
below.
      In most cases, vehicle travel on paved and unpaved travel surfaces
is  the major source of fugitive PM10 at TSDFs.  Spillage of waste
material onto these surfaces contaminates the surface with RCRA metals or
organic compounds which results in the generation of contaminated PM10.
      Available controls for unpaved travel surfaces include speed and
traffic reduction, paving, graveling, watering, and chemical stabiliza-
tion.  However,  in many cases, the only practical technique to control
                                7-16

-------
dust generation from unpaved travel surfaces is the use of watering.  The
use of chemical dust suppressants at hazardous waste TSOFs is also possi-
ble but, in general, not practical in most applications because of water
quality considerations.  Care must be exercised in the use of water for
dust control in that it too has the potential to spread contamination and
cause potential problems with regard to surface and groundwater.
     With regard to vehicle travel over paved surfaces, some type of
surface cleaning is required to reduce silt loadings and thus dust emis-
sions.  Either vacuum sweeping, flushing, or a combination of broom
sweeping and flushing can be used to clean paved surfaces. MRI has found
that the most effective technique seems to be water flushing followed by
broom sweeping.2*6  Again, care must be exercised when flushing paved
surfaces so as not to spread contamination from the process area to other
parts of the facility.  Therefore, a site-specific evaluation should be
performed to determine the most applicable method for reducing silt
loadings on paved surfaces.

7.4  CONTROL PERFORMANCE ESTIMATION
     The following sections provide available techniques to estimate the
performance of the dust control methods discussed above and their asso-
ciated cost elements.  Additional data on these methods can be found in
Cowherd and Kinsey and Cowherd et al.2'*

7.4.1  Wind Fences or Barriers
     Wind fences and screens are applicable to a wide variety of fugitive
dust sources.  They can be used to control wind erosion emissions from
raw material storage piles as well as to provide a sheltered area for
materials handling operations to reduce entrainment during load-in/load-
out, etc.  Fences and screens can be portable and thus capable of being
moved around the site, as needed.
     The control efficiency of wind fences is dependent on the physical
dimensions of the fence relative to the source being controlled.  In
general, a porosity (i.e., percent open area) of 50% seems to be optimum
for most applications.  (Note that no data directly applicable to waste
stabilization activities were found.)  According to a recent field study

                               7-17

-------
of small soil  storage piles, a screen length of five times the pile diam-
eter, a screen-to-pile distance of twice the pile height, and a screen
height equal to the pile height was found best.2
     A 1988 laboratory wind tunnel study of windbreak effectiveness for
coal storage piles showed area-averaged wind speed reductions of - 50% to
70X for a 50% porosity windbreak with height equal to the pile height and
length equal to the pile base.7  The windbreak was located three pile
heights upwind from the base of the pile.  Based on the 1.3 power given
in Table 7-1,  speed reductions of - 50% to 70% would correspond to - 60%
to 80% control of material handling PM10 emissions.
     One of the real advantages of wind fences for the control of fugi-
tive PM10 involves the low capital and operating costs.  These involve
the following basic elements:
     •    Capital equipment
               Fence material and supports
               Mounting hardware
          Operating and maintenance expenditures
               Replacement fence material and hardware
               Maintenance labor
     The following cost estimates (in 1980 dollars) were developed for
wind screens applied to aggregate storage piles:2'7
     •    Artificial wind guards
               Initial capital cost = $12,000 to $61,000
     •    Vegetative wind breaks
               Initial capital cost = $45 to $425/tree
Because of  the lack of quantitative data on costs associated with wind
screens, it is recommended that local vendors be contacted to obtain more
detailed data for capital and operating expenses.  Also, since wind
fences  and  screens are relatively "low tech" controls, it may be possible
for the site personnel to construct the necessary equipment with less
expense.
                                7-18

-------
7.4.2  Capture/Collection Systems
     Many industrial fugitive emissions have traditionally been
controlled by capture/collection, or industrial ventilation systems.
These systems have three primary components:  (a) a hood or enclosure to
capture emissions that escape from the process; (b) a dust collector that
separates entrained particulate from the captured gas stream; and (c) a
ducting or ventilation system to transport the gas stream from the hood
or enclosure to the air pollution control device.  Capture devices (or
hoods) generally can be classified as one of three types:  enclosure;
capture hood; or receiving hood.  These are described elsewhere.6
     The selection of a suitable capture device is source-specific and
depends on both operating and emissions characteristics.  Factors
influencing selection include location of the source with respect to
other operations, degree of process movement (if any), space needed for
worker or equipment access to the process, physical size of the operation
or process, and momentum of the particulate plume due to buoyancy or
inertia applied by the process.
     The two primary parameters involved in the design of effective local
exhaust or hooding systems are:  (a) locating the hood (defined in the
broad sense described above) to contain emitted particulate as much as
possible, and (b) providing adequate flow to capture any particulate not
contained by the hood and prevent the escape of all particulate from the
hood.  The goal of hood design is to install a hood or enclosure that
provides effective particulate control at the minimum exhaust volume.
     In the case of raw materials handling in waste stabilization
processes, some form of local capture hood or enclosure would generally
be used to control fugitive PM10 emissions where hoppers or silos are
employed.  Airflow exhausted from a local capture hood installed on an
operation involving material movement serves two purposes:  the exhaust
must overcome induced airflow created by material motion; and the exhaust
must provide sufficient velocity to capture particulate which escapes the
confines of the hood.  The predominant function is dependent on hood
type.  If an enclosure is used, control of induced airstreams is the pri-
mary objective.  If the operation requires an exterior hood, particulate
capture is the primary airflow function.
                               7-19

-------
     For those systems which can be controlled by complete or partial
enclosure, the airflow at the hood should be sufficient to overcome
induced air currents inherent to the process and to provide an inward air
velocity through all openings of about 50 to 200 ft/min.«  The volumes
needed to overcome induced air currents associated with specific pro-
cesses are discussed below.  The flow needed to provide adequate veloci-
ties at openings can be calculated by the formula:*

                                 Q = A V                            (7~5)

where  Q = required airflow (ftVmin)
       A = cross-sectional area of openings (ft2)
       V = required velocity at openings (ft/min)

     Material transport creates an induced airflow which must be overcome
to effectively control fugitive emissions.  The following equation is
used for calculating induced airflow at transfer points:*

                          Q = 10.0 Au

where   Q  = induced airflow (ftVmin)
        Au = feed opening  (ft2)
        R  = rate of material flow (tons/h)
        S  = height of fall  (ft)
        D  = average particle diameter (ft)

     The objective of a capture hood is to provide a capture velocity of
50 to  75 ft/min at  the farthest capture point from the  hood.  The total
flow required to achieve this velocity is:6

                             Q = V  (10 X2 + A)                        (7'7

where   Q = required airflow  (ftVmin)
        V  = required capture  velocity (ft/min)
                               7-20

-------
       X * distance from hood to farthest null point (ft)
       A = cross-sectional area of hood (ft2)

     For a capture/collection system, the controlled emissions are
comprised of:  (a) that portion of the uncontrolled emissions which are
not captured, plus (b) that portion of the uncontrolled emissions which
are captured but not collected.  To determine the overall control effi-
ciency of the system, both the capture efficiency and collection effi-
ciency must be known as illustrated schematically in Figure 7-3.
     Based on limited information, a capture efficiency of 70% to 98% was
measured for a close capture hood used on a metallurgical process.6  This
range of values may or may not be applicable to raw materials handling in
waste stabilization processes.
     Costs associated with capture/collection systems include the follow-
ing basic elements:
     •    Capital equipment
               Dust collector (baghouse or scrubber; concrete work; dust
               removal system; control instrumentation; monitoring
               instrumentation)
               Hood(s)
               Ventilation system (fan; electrical wiring; ductwork; con-
               crete support work; damper system; expansion joints)
               Dust storage system
     •    Operating and maintenance expenditures
               Utilities (electricity; water)
               Supplies (replacement bags; fan motors; chemical additives
               for scrubber)
               Labor (system operation; control device maintenance and
               cleaning; ductwork maintenance)
               Disposal of collected particulate
     Reference 8 estimates the following costs (1980 dollars) for a
pneumatic conveying system vented to a fabric filter (this also includes
a ventilated enclosure for unloading):  capital = $130,000 and operat-
ing = $33,000/yr.
                               7-21

-------
             Not
          Captured
                           Captured
                                   Collection
                                     Device
                    Capture
                     Device
                                                     Captured
                                                      but not
                                                     Collected
  Uncontrolled

Control Efficiency (%)

where rh-j  = rh2 •+• r?u
                                     m-,
                                               x 100
Figure 7-3.   Performance  evaluation of capture/collection systems.
                              7-22

-------
7.4.3  Wet Suppression
     Wet suppression of stabilized waste handling operations is similar
to that used for storage piles.  However, in addition to plain water this
technique can also use water plus a chemical surfactant or micronized
foam to control gross PM10.
     Surfactants added to the water supply allow particles to more easily
penetrate the water droplet and increase the total number of droplets,
thus increasing total surface area and contact potential.  Foam is gen-
erated by adding a chemical (i.e., detergent-like substance) to a rela-
tively small quantity of water which is then vigorously mixed to produce
small bubble, high energy foam in the 100 to 200-wm size range.  The foam
uses very little liquid volume, and when applied to the surface of the
bulk material, wets the fines more effectively than untreated water.
     The control efficiency of wet suppression for materials storage and
handling is dependent on the basic application parameters which
include:  the amount of water, water plus surfactant, or foam applied per
unit mass or surface area of material handled (i.e., liters per metric
ton or square meter); if not continuous, the time between reapplications;
the amount of surfactant added to the water (i.e., dilution ratio), if
any; the method of application including the number and types of spray
nozzles used; and applicable meteorological conditions occurring on-site.
     The only viable approach to estimating the control efficiency of wet
suppression in stabilization processes involves the use of the recently
developed materials handling equation published in AP-42.1*  This equation
was presented previously in Table 7-1.  By determining the "uncontrolled"
moisture content of the material and again after wet suppression, the
control efficiency can be determined by:

                           CE  = 100(eu  -  ec)/eu                      (7-8)

where  CE = PM10 control efficiency in percent
       eu = "uncontrolled" PM10 emission factor
       er = "controlled" PM10 emission factor
                               7-23

-------
The above calculations would necessitate the determination of the amount
of water added to the material by laboratory analysis.  This could be
accomplished by taking grab samples of the material before and after
application of the wet suppression technique being employed.  Sampling
and analysis techniques are described in Appendix D.
     Costs associated with wet suppression systems include the following
basic elements:
     •    Capital equipment
               Spray nozzles or other distribution equipment
               Supply pumps and plumbing (plus weatherization)
               Water filters and flow control equipment
               Tanker truck (if used)
     •    Operating and maintenance expenditures
               Water and chemicals
               Replacement parts for nozzles, truck, etc.
               Operating labor
               Maintenance labor
     Reference 2 estimates an initial capital cost of a stationary wet
suppression system using plain water to be $18,400 per system (1985 dol-
lars). The costs associated with a wet suppression system using chemical
surfactants for the unloading of limestone from trucks at aggregate pro-
cessing plants (in 1980 dollars) have been estimated at:  capital =
$72,000; annual = $26,000.  These costs are based on a stationary system
and may not be indicative of  those used at waste stabilization sites.

7.4.4  Watering of Unpaved Surfaces
     Watering of unpaved surfaces prevents (or suppresses) the release of
fine particulate from the surface through the action of mechanical dis-
turbance or wind.  The water  acts to bind the smaller particles to the
larger material thus reducing emissions potential.
     The control efficiency of watering of unpaved surfaces  is a direct
function of the amount of water applied per unit surface area (liters per
square meter), the frequency  of application (time between reapplication),
the volume of  traffic traveling over the surface between applications,
and prevailing meteorological conditions  (e.g., wind  speed,  temperature,

                               7-24

-------
etc.).  A number of studies have been conducted with regard to the effi
ciency of watering to control dust, but few have quantified all param-
eters listed above.  Additional details can also be found in Sections
2.2.3.2 and 2.2.4.3 of this document.
     For compacted unpaved surfaces found near waste stabilization
processes, an empirical model for the performance of watering as a con-
trol technique has been developed.  The model is:
                               100 -

where  C = average control efficiency (%)
       p = potential average hourly daytime evaporation rate (mm/h)
       d = average hourly daytime traffic rate (vehicles/h)
       i = application intensity (L/m2)
       t = time between applications (h)

The term p in the above equation is determined using Figure 7-4 and the
relationship:
                           0.0049 E  (annual  average)               (7-10a)
                      P =
                           0.0065 E  (worst case)                   (7-10b)

where  p = potential average hourly daytime evaporation rate (mm/h)
       E = mean annual pan evaporation (inches) from Figure 7-4

Note that Figure 7-4 does not present data for Alaska and Hawaii.
Readers responsible for those portions of the country should consult
local meteorological data from local weather stations, state universi-
ties, etc., for input to Eqs. 7-10a and 7-10b.
     An alternative approach (which is potentially suitable for a regula-
tory format) is shown as Figure 7-5.  Figure 7-5 shows that, between the
average uncontrolled moisture content and a value of twice that, a small
increase in moisture content results in a large increase in control effi-
ciency.  Beyond this point, control  efficiency grows slowly with
increased moisture content.
                               7-25

-------
•r/Vi'-K,Xr1
                                          (0
                                          ^•^ •


                                          sis


                                          4-» *~*
                                          00 a;



                                          I _> J


                                          c%


                                          .- 
-------
o
c
.2
'o
            Watering Control
         Efficiency Estimates
   100%
                                           95%
    75% I	
LLJ

"5
i.
*-
C
O
O
(0
13
O
O
c

2
c
    50% -
    25% -
           Ratio of Controlled to Uncontrolled
              Surface Moisture Contents
      Figure 7-5. PM10 control efficiency for watering unpaved surfaces.
                        7-27

-------
Furthermore, this relationship is applicable to all size ranges con-
sidered:
                          [75 (M-l)        1 < M  < 2                (7_n)
                          (62 + 6.7  M     2 < M  * 5
where  c = instantaneous control efficiency (%)
       M = ratio of controlled to uncontrolled surface moisture contents
           (dimensionless)

     Costs for watering programs include the following elements:
     Capital:  Purchase of truck or other device
     O&M:  Fuel, water, truck maintenance, operator labor
     Reference 2 estimates the following costs (1985 dollars):
     Capital:  $17,100/truck
     O&M:  $32,900/truck
The number of trucks required may be estimated by assuming that a single
truck, applying water at 1 L/m2, can treat roughly 4 acres of unpaved
surface every hour.

7.4.5  Paved Surface Cleaning
     Other than housekeeping, the only method available to reduce the
surface loading of fine particles on paved surfaces is through some form
of street cleaning practice.  The three major methods of street cleaning
are:  mechanical (broom) cleaning; vacuum cleaning; and flushing.   Broom
sweeping itself does remove  some debris from the pavement thus preventing
it from becoming airborne by the action of passing vehicles; but it can
also generate significant amounts of finer particulate by the mechanical
action used  to collect the material.  Thus, broom sweeping without  prior
water flushing is not normally recommended for removal of fine particu-
late from paved surfaces.
     Measurement-based efficiency values for paved  surface control
methods are  presented in Table 7-3.2  Note that all values in this  table
are for mitigative measures  applied  to industrial paved roads and reflect
reductions  in PM-15 not PM10.  It can be assumed, however, that similar
reductions  in gross PM10 may be expected for the  surface cleaning methods
listed  in Table 7-3.
                               7-28

-------
      TABLE 7-3.  MEASURED EFFICIENCY VALUES FOR PAVED SURFACE CONTROLS3
    Method
   Cited
 efficiency
                 Comments
Vacuum sweeping    Q%-58%
Water flushing

Water flushing
followed by
broom sweeping
46%


69-0.231 Vc'd

96-0.263 Vc'd
Field emission measurement (PM-15)
12,000-cfm blower0

Based on field measurement of
30 urn particulate emissions

Field measurement of PM-15 emissions0

Field measurement of PM-15 emissions0
aFrom Reference 2.  All results based on measurements of air emissions from
 industrial paved roads.

 PM-10 control efficiency can be assumed to be the same as that tested.

cWater applied at 0.48 gal/yd2.

 Equation yields efficiency in percent, V = number of vehicle passes since
 application.
                                     7-29

-------
     Cost elements Involved with vacuum sweeping include the following
capital and O&M expenses:
     Capital:  Purchase of truck or other device
     O&M:  Fuel, replacement parts, truck maintenance, and operator labor
     co%ts
     Data presented in Reference 2 provides the following estimates for a
vacuum sweeping program in April 1985 dollars:
     Initial capital expense:  $36,800/truck
     Annual O&M expense:  $34,200/truck
Determination of the number of trucks necessary can be made by assuming
that one unit can sweep 6 mi/12 h.2
     Finally, cost elements involved with flushing in combination with
broom sweeping include the following capital and O&M expenses:
     Capital:  Purchase of truck or other device
     O&M:  Fuel, replacement parts (possibly including brushes), truck
     maintenance, operator labor, water
     Cost data presented in Reference 2 provides the following estimates
for a flushing program:
     Initial capital expense:  $18,400/truck
     Annual O&M expense:  $27,600/truck
All costs are based on April 1985 dollars.  Determination of the number
of trucks required can be based on the assumption that 3 to 5 mi can be
flushed or broom swept per unit per 8-h shift, respectively.2

7.5  PROCEDURES FOR COMPLIANCE DETERMINATION
     In this section, two basic procedures will be discussed for the
determination of dust control compliance for waste stabilization pro-
cesses.  These procedures are:  permit systems; and  indirect measures of
control performance.  Each is addressed below.

7.5.1  Permit Systems
     The first approach  to compliance determination  involves the
implementation and  enforcement of  a permit program for waste stabiliza-
tion processes.  This would generally be accomplished as part of the RCRA
Part B permit for the site issued  by the appropriate regulatory agency.
                                7-30

-------
A permit system would require the site owner or operator to file a spe-
cific dust control plan with the appropriate regulatory agency having
jurisdiction.
     As part of the permit application, record keeping should be one of
the main conditions for approval.  Records of site activity and control
should be submitted to the regulatory agency on a monthly basis.  These
records must be certified by a responsible party as to their completeness
and accuracy.  All site records should be permanently maintained by the
agency.
     To enforce the dust control plan submitted as part of the permit
application, field audits of key control parameters should be made by
regulatory personnel.  The results of these audits would then be compared
to site records for that period to determine compliance with permit con-
ditions.  If differences are found between application of the control(s)
observed on-site and those recorded by site operating personnel, this
would constitute a violation and would be grounds for further enforcement
action.  This approach is, however, predicated on the fact that strict
implementation of the dust control plan will achieve certain reductions
in PM,0 emissions associated with site operation.  General information to
be included in a dust control plan, specific operational records, and
general records to be kept by the source for each of the control measures
discussed in Section 7.4 are provided below.2

     7.5.1.1  Wind Fences/Barriers.
     General Information
     1.  Locations of all raw materials storage and handling operations
to be controlled with wind fences referenced on a plot plan available to
the operator and regulatory personnel.
     2.  Physical dimensions of each source to be controlled and
configuration of each fence or screen to be installed.
     3.  Physical characteristics of material to be handled or stored for
each operation to be controlled by fence(s) or screen(s).
     4.  Applicable prevailing meteorological data (e.g., wind speed and
direction) for site on an annual basis.
                               7-31

-------
     Specific Operational Records
     1.  Date of installation of wind fence or screen and initials of
installer.
     2.  Location of installation relative to source and prevailing
winds.
     3.  Type of material being handled and stored and physical
dimensions of source controlled.
     4.  Date of removal of wind fence or screen and initials of
personnel involved.
     General Records
     1.  Fence or screen maintenance record.
     2.  Log of meteorological conditions for each day of site operation.

     7.5.1.2  Capture/Collection Systems.
     General Information
     1.  Locations of hood/enclosure, ductwork, and dust collector
referenced on a plot plan available to the responsible party and regu-
latory personnel.
     2.  Plans, drawings, and specifications for storage silo(s),
capture/collection system (including all available test data), and
ancillary equipment.
     3.  Physical characteristics of all materials to be transferred to
silos and stabilization process.
     4.  If a scrubber  is used, source of water and chemical additives to
be used, if any.
     5.  Location and type of permanently mounted instrumentation for
monitoring operation of capture/collection system.
     Specific Operational Records
     1.  Date of operation and operator's initials.
     2.  Start and stop time of dust control equipment.
     3.  Notations of malfunctions and corrective actions taken including
date and time of each.
     4.  Type of material and number of loads or weight transferred to
silo(s) and/or process  between start and stop time.
      5.  Notations of any visible emissions from dust collector stack or
vent  including date and time of each event.
                               7-32

-------
     General Records
     1.  Equipment maintenance records.
     2.  Records of equipment malfunctions and downtime.
     3.  Purchase orders, etc., of maintenance supplies and replacement
parts (e.g., bags, fan motors, etc.).

     7.5.1.3  Met Suppression.
     General Information
     1.  Locations of all stabilized waste storage and handling opera-
tions referenced on plot plan of the site available to the operator and
regulatory personnel.
     2.  Materials transport flow sheet which indicates the type of
material (if variable), its handling and storage, size and composition of
storage piles, etc.
     3.  The method and application intensity of water, etc., to be
applied to the various materials and frequency of application, if not
continuous.
     4.  Dilution ratio for chemicals added to water supply, if any.
     5.  Complete specifications of equipment used to handle the various
materials and for wet suppression.
     6.  Source of water and chemical(s), if used.
     Specific Operational Records
     1.  Date of operation and operator's initials.
     2.  Start and stop time of wet suppression equipment.
     3.  Location of wet suppression equipment.
     4.  Number of loads (or other measure of throughput)  loaded between
start and stop time.
     5.  Start and stop times for tank filling.
     General Records
     1.  Equipment maintenance records.
     2.  Meteorological log of general conditions.
     3.  Records of equipment malfunctions and downtime.
                               7-33

-------
     7.5.1.4  Watering of Unpaved Surfaces.
     General Information
     1.  All travel routes to be treated references on a plot plan
available to both the site operator and regulatory personnel.
     2.  Length and area of surfaces to be watered.
     3.  Application Intensity (L/m*) and frequency (a minimum moisture
content may be specified as an alternative).
     4.  Type of application vehicle, capacity of tank, and source of
water.
     Specific Records to Be Kept by Site Operator
     1.  Equipment maintenance log.
     2.  Meteorological log of general conditions (e.g., sunny and warm
vs. cloudy and cold).
     3.  Records of equipment breakdowns and downtime.
     An example permanent record form which may be used to record the
above information is shown in Figure 7-6.
     Specific Records to Be Kept by Truck Operator
     1.  Date and time of treatment.
     2.  Equipment used (this should be referred back to dust control
plan specifications).
     3.  Operator's initials (a separate operator's 'log may be kept and
transferred later to permanent records by site operator).
     4.  Start and stop time, average speed, and number passes.
     5.  Start and stop time for filling of water tank.

     7.5.1.5  Vacuum Surface Cleaning.
     General Information
     1.  All road segments (and parking areas, if applicable) referenced
on a map available to both the responsible party and the regulatory
agency.
     2.  Length of each road (and area of each parking area).
     3.  Type of control applied to each road/area and planned frequency
of application.
     4.  Any provisions for weather (e.g., % in of rainfall will be
substituted for one treatment; etc.).
                               7-34

-------
S3

                               7-35

-------
     Specific Records for Each Road Segment/Parking Area Treatment
     1.  Date of treatment.
     2.  Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
     3.  Start and stop times on a particular segment (or parking area),
average speed, number of passed.
     4.  Qualitative description of surface loading before and after
treatment.
     5.  Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
     General Records
     1.  Equipment maintenance records.
     2.  Meteorological log  (to the extent that weather influences the
control program—see above).
     3.  Any equipment malfunctions or downtime.

     7.5.1.6  Flushing/Broom Surface Cleaning.
     General Information
     1.  All road segments (and parking areas, if applicable) referenced
on a map available to both the responsible party and the regulatory
agency.
     2.  Length of each road (and area of each parking area).
     3.  Type of control applied to each road/area and planned frequency
of application.
     4.  Provisions for weather (e.g., program suspended for periods of
freezing temperatures).
     Specific Records for Each Road Segment/Parking Area Treatment
     1.  Date of treatment.
     2.  Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
     3.  Start and stop times on a particular segment (or parking area),
average  speed, number of passes.
     4.  Start and stop times for refilling tanks.
                               7-36

-------
     5.  Qualitative description of surface loading before and after
treatment.
     6.  Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
     General Records
     1.  Equipment maintenance records.
     2.  Meteorological log (to the extent that weather influences the
control program—see above).
     3.  Any equipment malfunctions or downtime.

7.5.2  Indirect Measures of Control Performance
     The final compliance format to be presented relates to various
indirect measures of control performance.  These could be used in con-
junction with or in lieu of permit enforcement discussed above.  They
will, however, require more effort and expense to implement but should be
at least somewhat defensible as measures of control efficiency.
     The most obvious approach to indirectly measuring control
performance involves the collection and analysis of material samples from
the various dust-emitting sources.  For example, if some form of paved
surface cleaning is performed, collection of surface samples for silt
content would indicate the efficacy of control for this particular
source.  The silt loadings obtained could be compared with "typical" sur-
face loading values for similar uncontrolled surfaces to determine the
degree of loading (and thus emissions) reductions achieved.  This would,
of course, necessitate the availability of a data base of "uncontrolled"
silt loadings for comparison with site-specific data.  The silt samples
could also be chemically analyzed to determine the degree of contamina-
tion control being achieved.
     Another indirect measure of control compliance is the collection and
analysis of material samples from unpaved surfaces and stabilized waste
storage and handling operations.  In this case, analysis of the moisture
content of these samples would indicate the amount of water applied and
thus the degree of control achieved by wet suppression.  Appropriate
equations (presented above) would be used to determine control efficiency
                               7-37

-------
based on the sample data.  Appropriate sampling and analysis methods to
implement the above approach are included in Appendix D.

7.6  EXAMPLE CALCULATION
     An example application of the emission estimation and control
performance procedures for waste stabilization units is presented
below.  Although this example focuses on contaminant-specific (i.e., Cr)
emissions, procedures are also provided for gross PMlo emissions as well.

7.6.1  Process Description
     A commercial facility is located in a semiarid climate with 320 dry
days per year.  The facility operates 5 days per week (261 d/yr), 8 h/d,
and processes the following listed waste streams likely to contain appre-
ciable concentrations of chromium:

                   Waste Stream      Annual Receipts (Mg)
                       K001                 1,000
                       F006                 5,000
The characteristics of the waste are specified as:
                                         Cr Concentration
                    Waste Stream           (ppm by wt)	
                        K001                   200
                        F006                   500

In the stabilization process, power plant fly ash is added to the liquid
waste in sufficient quantities to obtain a 50% (weight) solids content  in
the stabilized waste.  Fly ash is added to the liquid waste from elevated
silos with the process equipment exposed to ambient winds.  The stabi-
lized waste  is discharged directly into roll-offs which are carried by
trucks across a 0.25 mile (0.40 km) of unpaved surface to the main plant
road.  Liquid waste is also delivered by trucks traveling across this
same  unpaved  surface.  A capture/collection system is used for control  of
raw materials transfer and handling and watering is used for control of
vehicle-generated dust.
                                7-38

-------
Since no site-specific S&A has been conducted, contamination of unpaved
travel surfaces is assumed to be similar to that shown  in Table 7-2.
Silt content of the unpaved surface material is assumed to  be  identical
to that shown in Table 7-1.  All calculations are for annual average
emissions rounded to two significant figures.


Based on facility records the following extent values are calculated:
   Throughput of raw material:

   (1,000 + 5,000) M

   Delivery of raw material:  Assuming 41 Mg capacity tractor trailers
   6-000
   Liquid waste delivery:  Assuming 10,000 gal (37,850 L) capacity tank
   trucks and 8 Ib/gal (0.96 kg/L) specific weight
   37,850 ±-3 x 0.96 £3 x . * "9.   = 36
          load        L    1,000 kg      load
   6,000 M(? waste x yg^ x 0.8 J9D- = 130 km/yr
            yr      3o Mg        load

   Stabilized waste haulage:  Assuming 5.5 Mg load capacity

   (6,000 fly ash + 6,000 liquid waste) M9 waste x I ]oa.d x 0.8 J2L. = 1700 km/y
                                           yr      3.0 Mg        load

   Total truck traffic:

   (120 + 130 + 1700) km/yr = 1950 km/yr

7.6.2  Raw Material Handling

Assuming:  Fly ash is pneumatically conveyed from trucks to silo(s).
           Silo and discharge are ventilated to a baghouse dust collector
           (99.9% collection efficiency = 0.999)
           The capture efficiency of the ventilated discharge is 95%
           (0.95).
           The emissions are uncontaminated (no RCRA metals).
                               7-39

-------
UE10 = 0.00085  k9 PM'10  x 6,000 M9 f1v ash x 2 transfers
                  *                     "
The uncontrolled emissions are calculated:

                      PM
                     * 'y

      = 10 kg/yr uncontrolled PM-10 (or 5.1 kg/yr for each operation)

The controlled emissions are calculated:
Load- in— 0.00085 £ x 6,000 ^ x (1-0.999) * 0.0051
                                 v       '
                                                          PM-10
                          ,             .        .
                    Mg         yr   v       '             yr
                          (assumes  100* capture)

   Load-out— [5.1 x (1-0.95)] ^3 + [5.1 x 0.95 x (1-0.999)] ^3 = 0.26 ^
                              yr                 v       " yr        yr
               (not captured)           (not collected)

   Total — (0.26 + 0.0051) k3 PM"10 = 0.26 k3 PM~1Q * CE10
                        '    yr              yr        10

7.6.3  Vehicle Traffic on Unpaved Surfaces

Assuming:  The gross weight of the delivery trucks is 59 Mg (65 tons).
           The gross weight of the disposal trucks is 12 Mg (13 tons).
           The vehicle speed is 24 km/h (15 mph).
           Watering of unpaved surfaces is used at an application
           intensity of 2 L/m2.
           The annual vehicle traffic is 3000 vehicles/yr.

The uncontrolled emissions are calculated:

   Waste and raw material delivery = 1,950 km/yr (total) - 1,700 km/yr
     (stabilized waste)
                                   = 250 km/yr delivery traffic

Using the fractional proportion of waste and raw material traffic to
total traffic (Section 7.6.1):
   Average vehicle weight   =  ^^ x 59 Mg  +  f4^ x 12 Mg
   (weighted based on vol.)    iybu             iyb0

                            = 7.6 + 10 = 18 Mg
   Average number of wheels =  T||§ x 18  +  i4^ x 10
   (weighted based on vol.)    iysu          iytju

                            = 2.3 + 8.7 = 11 wheels
                               7-40

-------
Therefore:

                         s   /Uv0.7      0.5
UECr * o x 0.612 x -r| x ^! x (=£=)    x (J)    x ^§|iE x source extent
     = .67   x 0<612 x 1| x 24 x / 18\  *  x  /11\    x |f° x  1,950 veh-km/yr
       1(11);*           if.   «+o   \f.tll       \ 4/      jOO


     = 0.33 *a_£r. (or 4,900 kg total PM-10)
             yr   v               yr       '



   The controlled emissions are calculated assuming watering  is conducted

   every 4 h (t = 4):

   For e = mean annual pan evaporation (from  Figure 7-4) = 90 in

       p = potential average hourly daytime evaporation rate  (from Eq.
           7-7a)
         = 0.0049 e = 0.0049 (90) = 0.44 mrn/h

   and d = average hourly traffic rate

         = 3000 Vehic1es x .LSI- x 14 =  1.4 veh,!c1es
                   yr      261 d    8 h          h


   Thus C = average control efficiency = 100  - °*8 P d t (from Eq. 7-6)


          = 100 . |r0.8(0.44)g.4)(4)1 = gg% = Oi9g


   Therefore:

   CECr = 0.33 ^J£ x (1 - 0.99) = 0.0033 ^LJ>  or
   CElo = 4,900 J«L^ilO x (i _ Q.99) = 49 k^ ^10 total emissions
                               7-41

-------
7.7  REFERENCES FOR SECTION 7

 1.  Malone, P. G., L. W. Jones, and R. J. Larson.  Guide to the Disposal
     of Chemically Stabilized and Solidified Waste.  SW-872, U.S.
     Environmental Protection Agency.  Cincinnati, OH.  September  1982.

 2.  Cowherd, C., G. E. Muleski, and J. S. Klnsey.  Control of Open
     Fugitive Dust Sources.  EPA-450/3-88-008.  U.S. Environmental Pro-
     tection Agency, Research Triangle Park, NC.  September 1988.

 3.  Englehart, P., and D. Wallace.  Assessment of Hazardous Waste TSDF
     Particulate Emissions.  Final Report.  EPA Contract No. 68-02-3891,
     Assignments 5 and 13.  U.S. Environmental Protection Agency,
     Research Triangle Park, NC.  October  1986.

 4.  U.S. Environmental Protection Agency.  Compilation of Air Pollution
     Emission Factors, AP-42.  U.S. Environmental Protection Agency,
     Research Triangle Park, NC.  1988.

 5.  Muleski, G. E., F. J. Pendleton, and  W. A. Rugenstein.  Measurement
     of Fugitive Emissions in a Coal-Fired Power  Plant.  Proceedings:
     Sixth Symposium on the Transfer and Utilization of Particulate Control
     Technology, EPRI CS-4918, Electric Power Research Institute.  Palo
     Alto, CA.  November  1986.

 6.  Cowherd, C., Jr., and J. S. Kinsey.   Identification, Assessment,  and
     Control of Fugitive  Particulate Emissions.   EPA-600/8-86-023.  U.S.
     Environmental Protection Agency, Research Triangle Park, NC.  August
     1986.

 7.  Studer, B. J. B., and S. P. S. Arya.  Windbreak: Effectiveness for
     Storage Pile Fugitive Dust Control:   A Wind  Tunnel Study.  Journal of
     the Air Pollution Control Association.   38:135-143.   1988.

 8.  Ohio Environmental  Protection Agency. Reasonably Available Control
     Measures for Fugitive Dust Sources.   Columbus, OH.  September 1980.
                                7-42

-------
                    APPENDIX A.

OPEN DUST SOURCE EMISSION FACTOR RATING AND CONTROL
              EFFICIENCY TERMINOLOGY

-------
      APPENDIX A.   OPEN  DUST SOURCE EMISSION FACTOR RATING AND CONTROL
                           EFFICIENCY TERMINOLOGY

A.I   EMISSION FACTOR RATING  TERMINOLOGY
      In AP-42,  the reliability of  emission  factors  is  indicated by an
overall Emission  Factor Rating ranging from A  (excellent)  to E (poor).
These ratings take into account  the  type  and amount of data from which  the
factors were  calculated.  Note that  measurements  underlying each emission
factor are rated  on a similar scale  of A  to D.
      The use  of a statistical confidence  interval may  seem desirable  as a
more  quantitative measure of the reliability of an  emission factor.
Because of the  way an emission factor data  base is  generated,  however,
prudent application of statistical procedures precludes the use  of
confidence intervals unless the following conditions are met:
      •  The sample of sources from which the emission  factor was
        determined is representative of the  total population of  such
        sources.
      •  The data  collected at an individual  source  are representative of
        that  source (i.e., no temporal variability  resulting from  source
        operating conditions could have biased the data).
      •  The method of measurement was properly applied at each source
        tested.
Because of the almost impossible task of assigning a meaningful  confidence
limit to the  above variables and to other industry-specific variables, the
use of a statistical  confidence interval  for an emission factor  is not
practical.
     The following emission factor ratings are  applied  to  the emission
factors:
     A - Excellent.   Developed  only from  A-rated  test data taken from many
     randomly chosen  facilities  in the industry population.  The source
     category is specific  enough  to minimize variability within the source
     category population.
     B -  Above average.  Developed  only from A-rated test  data  from a
     reasonable number of  facilities.  Although no specific bias  is
     evident,  it is not  clear if  the  facilities tested  represent  a random
                                 A-l

-------
     sample of the industry.  As in the A-rating, the  source  category  is
     specific enough to minimize variability within  the  source  category
     population.
     C - Average.  Developed only from A- and B-rated  data  from a
     reasonable number of facilities.  Although no specific bias is
     evident, it is not clear if the facilities tested represent a random
     sample of the industry.  As in the A rating, the  source  category  is
     specific enough to minimize variability within  the  source  category
     population.
     D - Below average.  The emission factor was developed  only from A-
     and B-rated test data from a small number of facilities, and there
     may be reason to suspect that these facilities  do not  represent a
     random sample of the industry.  There also may  be evidence  of
     variability within the source category population.  Limitations on
     the use of the emission factor are footnoted in the emission factor
     table.
     E - Poor.  The emission factor was developed from C- and D-rated test
     data, and there may be reason to suspect that the facilities tested
     do not represent a random sample of the industry.   There may be
     evidence of variability within the source category population.
     Limitations on the use of these factors are always footnoted.
Because the application of these factors is somwhat subjective,  the
reasons for each rating are documented in the background files maintained
by the Office of Air Quality Procedures and Standards (OAQPS).
A.2  CONTROL EFFICIENCY TERMINOLOGY
     Some control  techniques often used for open dust sources  begin to
decay in efficiency almost immediately after implementation.  The most
extreme example of this is the watering of unpaved  roads where the
efficiency decays  from nearly 100 percent to 0 in  a matter of  hours (or
minutes). The control  efficiency for broom sweeping and flushing applied
in combination on  a paved road may decay to  zero in  1 or 2 days.  Chemical
dust suppressants  applied to unpaved roads can yield  control efficiencies
that will decay to zero in several  months.   Consequently, a  single-valued
control efficiency is  usually not adequate to  describe  the performance  of
most intermittent  control  techniques for open  dust  sources.  The control
                                 A-2

-------
 efficiency must  be  reported  along with a time period over which the value
 applies.   For  continuous  control  systems (e.g.,  wet suppression for
 materials  transfer),  a single control  efficiency is usually appropriate.
      Certain terminology  has been developed to aid in describing the time
 dependence of  open  dust control efficiency.  These terms are:
      1.  Control  lifetime is the  time  period (or amount of source
 activity)  required  for the efficiency  of an open dust control  measure to
 decay to zero.
      2.  Instantaneous control efficiency is the efficiency of an open
 dust  control at  a specific point  in  time.
      3.  Average  control  efficiency  is  the  efficiency of an open dust
 source control averaged over a given period of time (or number of vehicle
 passes).
      From  the  above definitions,  it  is  clear that  average control
 efficiency is  related  to  instantaneous  control efficiency by the following
 general equation:

                        C(X)  = x  0.  In that case, average control
efficiency  is given  by

                             c(X)  « a  x
                                  A-3

-------
                    APPENDIX B





ESTIMATION OF CONTROL COSTS AND COST EFFECTIVENESS
                    B-l

-------
     Relative costs of alternative control measures are used in the
development and evaluation of particulate fugitive emissions control
strategies.  Cost analyses are used by control  agency personnel to
develop overall strategies or to evaluate plant specific control strate-
gies.  Industry personnel perform cost analyses to evaluate control
alternatives for a specific source or to develop a iplant-wide emissions
control strategy.  Although the specifics of these analyses may vary
depending upon the objective of the analysis and the availability of cost
data, the general format is similar.
     The primary goal of any cost analysis is to provide a consistent
comparison of the real costs of alternative control measures.  The objec-
tive of this section is to provide the reader with a methodology that
will allow such a comparison.  It will describe the overall structure of
a cost analysis and provide the resources for conducting the analyses.
Because cost data are continuously changing, specific cost data are not
provided.  However, sources of cost information and mechanisms for cost
updating are provided.
     The approach outlined in this section will focus on cost effective-
ness as the primary comparison tool.  Cost effectiveness is simply the
ratio of the annualized cost of the emissions control to the amount of
emissions reduction achieved.  Mathematically, cost effectiveness  is
defined by:
                                       .                             (B-l)
where:    C* = cost effectiveness, $/mass of emissions reduction
          Ca = annualized cost of the control measure, $/yr
          AR = reduction  (mass/yr) in annual emissions
 This  general methodology was  chosen because  it  is equally applicable to
 different  controls  that achieve equivalent emissions reduction on a
 single  source  and to measures  that achieve varied reductions over
 multiple  sources.
                                B-2

-------
     The discussion is divided into three sections.  The first section
describes the general cost analysis methodology, including the various
types of costs that should be considered and presents methods for calcu-
lating those costs.  The second identifies the primary cost elements
associated with each of the fugitive emissions control systems.  The
final section identifies sources of cost data and discusses methods for
updating cost data to constant dollars, and includes example calculation
cases for estimating costs and cost effectiveness.

B.I  GENERAL COST METHODOLOGY
     Calculation of cost effectiveness for comparison of control measures
or control strategies can be accomplished in four steps.  First, the
alternative control/cost scenarios are selected.  Second, the capital
costs of each scenario are calculated.  Third, the annualized costs for
each of the alternatives are developed.  Finally, the cost effectiveness
is calculated, taking into consideration the level of emissions
reduction.
     The general approach for performing each of the above steps is
described below.  This approach is intended to provide general guidance
for cost comparison.  It should not be viewed as a rigid procedure that
must be followed in detail for all analyses.  The reader may choose or
may be forced through resource or informational constraints to omit some
elements of the analysis.  However, for comparisons to be valid, cautions
that should be observed are:  (1) all control scenarios should be treated
in the same manner, and (2) cost elements that vary radically between
cost scenarios should not be omitted.

B.I.I  Select Control/Cost Scenarios
     Prior to the cost analysis, general control measures or strategies
will have been identified.  These measures or strategies will fall into
one of the major classes of fugitive emission control techniques that
were identified.  The first step in the cost analysis is to select a set
of specific control/cost scenarios from the general techniques.  The
specific scenarios will include definition of the major cost elements and
identification of specific implementation alternatives for each of the
cost elements.
                                B-3

-------
     Each of the general control techniques Identified in this document
has several major cost elements.  The first step in any cost analysis is
definition of these major cost elements.  These elements include capital
equipment elements and operation/maintenance elements.  Information is
provided in Section B.2 on the major cost elements associated with each
of the general control techniques.  For example, the major cost elements
for chemical stabilization of an unpaved road include:  (a) chemical
acquisition; (b) chemical storage; (c) road preparation; (d) mixing the
chemical with water; and (e) application of the chemical solution.
     For each major cost element, several implementation alternatives can
be chosen.  Options within each cost element include such choices as
buying or renting equipment; shipping chemicals by railcar, truck tanker,
or in drums via truck; alternative sources of power or other utilities;
and use of plant personnel or contractors for construction and mainte-
nance.  The major cost elements and the implementation alternatives for
each of these elements for the chemical stabilization example described
above are outlined in Table B-l.
     Note that the following discussion is applicable only to "in-house"
open fugitive dust control programs.  Many industrial facilities prefer
to retain outside contractors to implement paved and unpaved road dust
control programs.  Some contractors provide complete services, from
ordering and storing raw materials through control application; others
may only supply the vehicles and drivers required for application.  As a
result, the reader may need to consider several implementation alterna-
tives (such as those given in Table B-l) prior to selection of a cost-
efficient dust control program.

B.I.2   Develop Capital Costs
     The capital costs of an open dust emissions control program are
those direct and indirect expenses incurred up to the date when the
program is  implemented.  These capital costs include actual purchase
expenses for capital equipment,  labor and utility costs associated with
installation of the control system, and system startup and shakedown
costs.   In  general, direct capital costs are the costs of control
                                B-4

-------
      TABLE  B-l.   IMPLEMENTATION  ALTERNATIVES FOR STABILIZATION OF AN
                               UhPAVED  ROAD

Cost elements/implementation alternatives

  I.  Purchase and Ship Chemical

      A.  Ship in railcar tanker (11,000 to 22,000 gal/tanker)
      B.  Ship in truck tanker (4,000 to 6,000 gal/tanker)
      C.  Ship in drums via truck (55 gal/drum)

 II.  Store Chemical

      A.  Store on plant property
          1.  In new storage tank
          2.  In existing storage tank
              a.  Needs refurbishing
              b.  Needs no refurbishing
          3.  In railcar tanker
              a.  Own railcar
              b.  Pay demurrage
          4.  In truck tanker
              a.  Own truck
              b.  Pay demurrage
          5.  In drums
      B.  Store in contractor tanks

III.  Prepare Road

      A.  Use plant-owned grader to minimize ruts and low spots
      B.  Rent contractor grader
      C.  Perform no road preparation

 IV.  Mix Chemical and Water in Application Truck

      A.  Put chemical in spray truck
          1.  Pump chemical from storage tank or drums into application
              truck
          1.  Pour chemical from drums into application truck, generally
              using forklift
      B.  Put water in application truck
          1.  Pump from river or lake
          2.  Take from city water line

  V.  Apply  Chemical Solution via Surface Spraying

      A.  Use plant owned application truck
      B.  Rent contractor application truck
                                 B-5

-------
materials and equipment as well as the labor and utilities needed to
install the equipment.  Indirect costs are overall costs to the facility
incurred by the system but not directly attributable to specific
equipment items.
     Direct costs cover the purchase of equipment and auxiliaries and the
costs of Installation.  Capital costs also include any cost of site
development necessitated by the control system.  For example, if the
storage tanks for chemical unpaved road dust suppressants require that an
access road be constructed, this access road 1s included as a capital
expense.  The types of direct costs for typical fugitive emissions con-
trol programs at TSDFs are associated with items such as vehicles (e.g.,
spray or flusher trucks) and storage tanks, pumps, and piping.
     Indirect costs cover the expenses not attributable to specific
equipment items.  Items in this category include:
     1.   Engineering costs—including administrative, project, and
general; labor and other costs associated with specification of the
control program parameters; cost analysis; purchasing and accounting;
consultant services.
     2.   Field and construction expenses—includes equipment rental,
repair, fuel, or lubricants; costs associated with grading or other road
preparation; field supervision; storage areas.
     3.   Shakedown/startup—includes costs associated with control
startup and shakedown.
     4.   Contingency costs—the excess account set up to deal with
uncertainties in the cost estimate, including unforeseen escalation in
prices, malfunctions, equipment design alterations, and similar sources.
     The values for these items will vary depending on the specific
operations to be controlled and the types of control systems used.  Typi-
cal  ranges for  indirect costs1 based on the total installed cost of the
capital equipment are shown in Table B-2.

B.I.3   Determine Annualized Costs
     The most common basis for comparison of an alternative control
system is that  of annualized cost.  The annualized cost of a fugitive
emission control system includes operating costs  such as labor,
                                B-6

-------
Cost item
          TABLE B-2.  TYPICAL VALUES FOR INDIRECT CAPITAL COSTS
Ranges of values
Engineering



Construction and field
  expenses

Contractor's fee

Shakedown/startup

Contingency
8 to 20 percent of installed cost.  High
  value for small projects; low value for
  large projects

7 to 70 percent of installed cost
10 to 15 percent of installed cost

1 to 6 percent of installed cost

10 to 30 percent of total direct and indirect
  costs dependent upon accuracy of estimate.
  Generally, 20 percent is used in a study
  estimate
 Source:   Reference 1.
                                 B-7

-------
materials, utilities, and maintenance items as well as the annualized
cost of the capital equipment.  The annualization of capital costs is a
classical engineering economics problem, the solution of which takes into
account the fact that money has time value.  These annualized costs are
dependent on the interest rate paid on borrowed money or collectable by
the plant as interest (if available capital is used), the useful life of
the equipment and depreciation rates of the equipment.
     The components of the annualized cost of implementing a particular
control technique are depicted graphically in Figure B-l.  Purchase and
installation costs include freight, sales tax, and interest on borrowed
money.  The operation and maintenance costs reflect increasing frequency
of repair as the equipment ages along with increased costs due to infla-
tion for parts, energy, and labor.  On the other hand, costs recovered by
claiming tax credits or deductions are considered as income.  Mathe-
matically the annualized costs of control equipment can be calculated
from:
                       ca ' CRF  (CP} + Co + °'5 Co
where:     Ca = annualized costs of control equipment, $/yr
          CRF = Capital Recovery Factor, 1/yr
           Cp - installed capital costs, $
           CQ = direct operating costs, $/yr
          0.5 = plant overhead factor

The  various components of this equation are briefly described below.
      The  annualized cost of capital equipment is calculated by using a
capital recovery  factor (CRF).  The CRF combines interest on borrowed
funds and depreciation into a single factor.  It is a function of the
interest  rate and the overall life of the capital equipment and can be
estimated by the  following equation:
                                B-8

-------
en
O
O
LU
2
O
O
z
    o
             Equipment. Installation, Freight, Tax, and Interest
Depreciation Tax Deduction
                      LIFE OF EQUIPMENT
                        Scrap
                        Value
     Figure B-l.   Graphical  presentation of fugitive emission control costs,
                                   B-9

-------
                              CRF  .  1(1+1)"                          (B-3)
                                    (l+i)n-l
where:    1 = annual interest rate expressed as a fraction
          n = economic life of the control system (yr)

The other major components of the annualized cost are operation and
maintenance costs (direct operating costs) and associated plant overhead
costs.  Operation and maintenance costs generally include labor, raw
materials, utilities, and by-product costs or credits associated with
day-to-day operation of the control  system.   Elements typically included
in this category are:
     1.   Utilities—includes water, electricity.
     2.   Raw materials—includes chemical unpaved dust suppressants,
surfactants, etc.
     3.   Operating labor—includes supervision, skilled and unskilled
labor required by the control program.
     4.   Fuel costs—includes fuel  required by vehicles and other
equipment used in the control program.
     5.   Maintenance and repairs—includes the manpower and materials to
keep equipment operating efficiently.
     Another component of the operating cost is overhead, which is a
business expense not charged directly to a particular part of the process
but allocated to it.  Overhead costs include administrative, safety,
engineering, legal, and medical services; payroll, employee benefits;
recreation; and public relations.  As suggested by Eq. B-2, these charges
are estimated to be approximately 50% of direct operating costs.

B.I.4  Calculate Cost Effectiveness
     As discussed in the introduction to this section the most informa-
tive method for comparing control measures or control strategies for
particulate fugitive emissions sources  is on a cost-effectiveness
basis.  Mathematically, cost effectiveness is defined as:
                               B-10

-------
                                                                    (B-D
where:    C* » cost effectiveness, $/mass of emissions reduction
          Ca » annual1zed cost of the control measure, $/yr
          AR » reduction (mass/yr) 1n annual emissions

     The annual1zed cost of control equipment can be calculated using
Eq. B-2.  The annual reduction 1n part1culate emissions can be calculated
from the following equation:
                                                                     (B-4)
                                AR = M e c                               '
where:    M * annual source extent
          e = uncontrolled emission factor  (i.e., mass of uncontrolled
              emissions per unit of source  extent)
          c = average control efficiency expressed as a fraction

B.2  COST ELEMENTS OF FUGITIVE EMISSIONS CONTROL SYSTEMS
     The cost methodology outlined in Section B.I requires that the
analyst define and select alternative control/cost scenarios and develop
costs for the major cost elements within these scenarios.  The objective
of this subsection is to assist the reader  in identifying the implementa-
tion alternatives and major cost elements associated with the emission
reduction techniques.  For open dust sources, the control techniques
addressed are:  wet dust suppression, surface cleaning, and paving.
                               B-ll

-------
     Implementation alternatives for open dust source emission control
measures are presented in Tables B-3 through B-5.  Table B-3 presents
Implementation alternatives for water and chemical dust suppressant
systems.  Table B-4 presents alternatives for three types of street
cleaning systems—vacuuming, flushing, and a combination of flushing and
broom sweeping.  Table B-5 presents alternatives for streets or parking
lot paving.
     After the control scenarios are selected, the analyst must estimate
the capital cost of the Installed system and the operating and mainte-
nance costs.  The indirect capital costs elements are common to all
systems and were identified 1n Table B-2.  The direct capital cost ele-
ments and direct operation and maintenance cost elements which are unique
to each type of fugitive emission control system are identified in
Tables B-6 through B-9.  These costs are provided for dust suppressant
programs for open dust sources in Table B-6, street cleaning programs in
Table B-7, paving in Table B-8, and wet suppression systems for process
sources in Table B-9.

B.3  SOURCES OF COST DATA
     Collection of the data to conduct a cost analysis can sometimes be
difficult.  If a well defined system is being costed, the best sources of
accurate capital costs are vendor estimates.  However, if the system is
not sufficiently defined to develop vendor estimates, published cost data
can be used.  Cost data are available for both paved and unpaved roads in
References 2 through 8.
     Often published cost estimates are based on different time-valued
dollars.  These estimates must be adjusted for inflation so that they
reflect the most probable capital investments for a current time and can
be consistently compared.  Capital cost indices are the techniques used
for updating costs.  These indices provide a general method for updating
overall costs without having to complete in-depth studies of individual
cost elements.   Indices that typically are used for updating control
system  costs are the Chemical Engineering Plant Cost; Index (available in
any issue of Chemical Engineering Magazine), the Bureau of Labor
Statistics Metal Fabrication Index, and the Commerce Department Monthly
Labor  Review.
                               B-12

-------
 rABLE 8-3.  IMPLEMENTATION ALTERNATIVES FOR OUST SUPPRESSANTS APPLIED TO
                             AN UNPAVED ROAD
Program implementation alternative
      Dust
suppressant type
ChemicalsWater
  I.   Purchase and  Ship Dust Suppressant

      A.   Ship in railcar tanker (11,000 to                  X
            22,000  gal/tanker)
      B.   Ship in truck tanker  (4,000 to 6,000 gal/          X
            tanker)
      C.   Ship in drums via truck (55 gal/drum)

 II.   Store dust  suppressant

      A.   Store on  plant property
      1.   In new  storage tank                                X
      2.   In existing  storage tank                           X
          a.  Needs  refurbishing                             X
          b.  Needs  no refurbishing                          X
      3.   In railcar tanker
          a.  Own railcar                                    X
          b.  Pay demurrage                                  X

III.   Prepare Road

      A.   Use plant-owned grader to minimize ruts  and low    X
            spots
      B.   Rent contractor grader                             X
      C.   Perform no road preparation                        X

 IV.   Mix  Dust Suppressant/Water in Application  Truck

      A.   Put suppressant in spray truck
          1.  Pump  suppressant  from storage tank or drums    X
               into application truck
          2.  Pour  suppressant  from drums into application   X
               truck,  generally using forklift
      B.   Put water  in application truck
          1.  Pump  from river or lake                        X
          2.  Take  from city water line                      X

  V.   Apply suppressant solution via surface spraying

      A.   Use plant  owned application truck                  X
      B.   Rent contractor application truck                  X
              X
              X
              X
              X
              X
              X
                                 B-13

-------
       TABLE B-4.  IMPLEMENTATION ALTERNATIVES FOR PAVED ROAD CLEANING



Program implementation alternative
I. Acquire Fl usher and Driver
A. Purchase fl usher/sweeper and
use plant driver
B. Rent f lusher/sweeper and
driver
C. Use existing unpaved road
watering truck


Vacuum
sweeping Flushing

X

X

X

Flushing
and
broom
sweeping

X

X



 II.   Acquire Vacuum Sweeper and Driver

      A.   Purchase sweeper and use
          plant driver
      B.   Rent sweeper and driver
X

X
III.   Fill  Flusher Tank With Water

      A.   Pump water from river or lake
      B.   Take water from city line
             X
             X
X
X
 IV.  Maintain Purchased Flusher
  V.  Maintain Purchased Vacuum Sweeper
                           X


                           X
                                     B-14

-------
      TABLE B-5.  IMPLEMENTATION ALTERNATIVES FOR PAVING

Program implementation alternative


  I.   Excavate Existing Surface to Make Way for Base and
       Surface Courses

       A.  2-in. depth
       B.  4-in. depth
       C.  6-in depth

 II.   Fine Grade and Compact Subgrade

III.   Lay and Compact Crushed Stone Base Course

       A.  2-in. depth
       B.  4-in. depth
       C.  6-in depth

 IV.   Lay and Compact Hot Mix Asphalt (Probably AC120-150)
       Surface Course

       A.  2-in. depth
       B.  4-in. depth
       C.  6-in depth
                           B-15

-------
  TABLE  B-6.   CAPITAL EQUIPMENT AND  O&M EXPENDITURE ITEMS FOR
                   OUST SUPPRESSANT  SYSTEMSd
                        (Open Sources)

Capital  equipment

  •  Storage equipment

       Tanks
       Railcar
       Pumps
       Piping

  •  Application equipment

       Trucks
       Spray system
       Piping (including winterizing)

Q&M expenditures

  •  Utility or fuel costs

       Water
       Electricity
       Gasoline or diesel  fuel

  •  Supplies

       Chemicals
       Repair parts

  •  Labor

       Application time
       Road conditioning
       System maintenance

aNot all items are necessary for all systems.  Specific  items
 are dependent on the control scenario selected.
                           B-16

-------
  TABLE  B-7.   CAPITAL  EQUIPMENT  AND  O&M  EXPENDITURE  ITEMS  FOR
                       STREET CLEANING

Capital  equipment

  •  Sweeping
       Broom
       Vacuum system
  •  Flushing
       Piping.
       Flushing truck
       Water pumps

O&M expenditures

  •  Utility and fuel  costs
       Water
       Gasoline or diesel  fuel
  •  Supplies
       Replacement brushes
  •  Labor
       Sweeping or flushing operation
       Truck maintenance
  •  Waste disposal
                           B-17

-------
 TABLE B-8.  CAPITAL EQUIPMENT AND O&M EXPENDITURES ITEMS FOR
                            PAVING

Capital  equipment

  •  Operating equipment
       Graders
       Paving application equipment
       Materials
       Paving material  (asphalt or concrete)
       Base material

O&M expenditures

  •  Supplies
       Patching material
  •  Labor
       Surface preparation
       Paving
       Road maintenance
       Equipment maintenance
                          B-18

-------
  TABLE B-9.   CAPITAL EQUIPMENT AND O&M EXPENDITURE ITEMS FOR
          WET SUPPRESSION SYSTEMS  (PROCESS SOURCES)

Capital equipment

  •  Water spray systems
       Supply pumps
       Nozzles
       Piping (including winterization)
       Control system
       Filtering units

  •  Water/surfactant and foam systems only
       Air compressor
       Mixing tank
       Metering or proportioning unit
       Surfactant storage area

O&M expenditures

  •  Utility costs
       Water
       Electricity

  •  Supplies
       Surfactant
       Screens

  •  Labor
       Maintenance
       Operation
                           B-19

-------
     Operation and maintenance cost estimates typically are based on

vendor or industry experience with similar systems.  In the absence of

such data, rough estimates can be developed from References 4 and 7.


B.4  EXAMPLE CALCULATIONS

     Table B-10 lists the steps necessary to calculate the cost effec-

tiveness for two control alternatives for stabilizing unpaved travel

surfaces.  An example problem illustrating the calculations is presented

in Table B-ll.  Tables B-12 through B-15 are referenced in the

calculations of Tables B-10 and B-ll.


REFERENCES FOR APPENDIX B

1.   PEDCo Environmental, Inc.  Cost Analysis Manual for Standards
     Support Document.  U.S. Environmental Protection Agency.  November
     1978.

2.   Cuscino, T., Jr., G. E. Muleski, and C. Cowherd, Jr.  Iron and Steel
     Plant Open Source Fugitive Emission Control Evaluation.  EPA-600/2-
     83-110, NTIS No. PB84-110568, U.S. Environmental Protection Agency,
     Research Triangle Park, NC.  October 1983.

3.   Muleski, 6. E., T. Cuscino, Jr., and C. Cowherd, Jr.  Extended
     Evaluation of Unpaved Road Dust Suppressants in the Iron and Steel
     Industry.  EPA-600/2-84-027, NTIS No. PB84-154350, U.S. Environ-
     mental Protection Agency, Research Triangle Park, NC.  February
     1984.

4.   Cuscino, T., Jr.  Cost Estimates for Selected Dust Controls Applied
     to Unpaved and Paved Roads in Iron and Steel Plants.  EPA Contract
     No. 68-01-6314, Task 17, U.S. Environmental Protection Agency,
     Region V, Chicago, IL.  April 1984.

5.   Richardson Engineering Services, Inc.  The Richardson Rapid Con-
     struction Cost Estimating System:  Volume I—Process Plant Construc-
     tion Estimating Standards.  1983-84 Edition.

6.   Robert Snow Means Company, Inc.  Building Construction Cost Data.
     1979.

7.   Neveril, R. V.  Capital and Operating Costs of Selected Air Pollu-
     tion Control Systems.  EPA-450/5-80-002.  GARD, Inc.  December 1978.

8.   Turner,  J. H., et al.  Fugitive Particulate Emissions from Hazardous
     Waste Sites.  Prepared for the U.S. Environmental Protection Agency
     under Contract No. 68-03-3149, Cincinnati, Ohio.  September 1984.
                               B-20

-------
          TABLE B-10.   COST AND COST-EFFECTIVENESS ESTIMATION FOR CHEMICAL  STABILIZATION
                                        OF AN UNPAVED ROAD
     This table lists the steps necessary to calculate the cost-effectiveness for two control
alternatives for stabilizing unpaved travel surfaces.

     Step I—Specify Desired Average Control Efficiency (e.g., 50. 75, or 90 percent)

     Step 2—Specify Basic Vehicle, Road and Climatological Parameters for the Particular Road of
Concern

     Required vehicle characteristics include:

     1.  Average Daily Traffic (ADT)—this  is the number of vehicles using the road regardless of
direction of travel (e.g., on a two-lane road in an iron and steel plant, 100 vehicles in one
direction, and 100 in the other direction during a single day yields 200 ADT);
     2.  Average vehicle weight in short tons;
     3.  Average number of vehicle wheels; and
     4.  Average vehicle speed in mph.

     Required road characteristics include:

     1.  Actual length of roadway to be controlled in miles;
     2.  Width of road to be controlled;
     3.  Silt content (in percent)—for an existing road, these values should be measured;
however, for a proposed plant, average values shown in AP-42 can be used;
     4.  Surface loading (for paved roads)  in Ib/mile—this is the total loading on all traveled
lanes rather than the average lane loading; and
     5.  Bearing strength of the road—At this time, just a visual estimate of low, moderate, or
high is required.
     Required climatological characteristics (applicable only to watering of unpaved roads):
potential evaporation in mm/h—the value depends on both the location and the month of concern.
Control efficiency data in this report for watering unpaved roads assume a location in Detroit,
Michigan, in the summer.

     Step 3—Calculate the Uncontrolled Annual Emission Rate as the Product of the Emission
Factor and the Source Extent


     The emission factor (E) should be calculated using the equations from AP-42.

     The annual source extent (SE) is calculated as 365 x ADT x average one way trip distance.

     Step 4—Consult the Appropriate Control Program Design Table to Determine the Time Between
Applications and the Application Intensity

     Select the appropriate table

                                                                    Table
                                                                 containing
               Control technique                                 information

               Coherex* applied to unpaved roads                 Table B-12
               Petro Tac applied to unpaved roads                Table B-13


                                                                                      (continued)
                                               B-21

-------
                                     TABLE B-10.   (continued)
     Verify that the vehicle and road characteristics listed in Step 2 are similar to those
listed in the footnotes of the selected table.  If they are significantly different, the table
cannot be used.

     Step 5—Calculate the Number of Annual Applications Necessary by Dividing 365 by the Days
Between Application (from Step 4)

     Step 6—Calculate the Number of Treated Miles Per Year by Multiplying the Actual Miles of
Road to be Controlled (from Step 2) by the Number of Annual Applications (from Step 5)

     Step 7—Consult the Appropriate Program  Implementation Alternatives Table and Select the
Desired Program  Implementation Plan

                                                                    Table
                                                                 containing
               Control technique                                 information

               Coherex* applied to unpaved roads                 Table B--14
               Petro Tac applied to unpaved roads                Table B-15

     Step 8—Calculate Total Annual Cost by Annualizing Capital Costs and Adding to Annual
Operation and Maintenance Costs

     To annualize capital investment, the capital cost  is multiplied by a capital recovery  factor
which  is calculated as follows:

                                CRF = [i(l  +  i)n] / [(!  + i)n - 1]

where     CRF =  capital recovery factor
             i *  annual interest rate  fraction
            n =  number of payment years

     Scale total annual cost  by ratio of actual road width  in  feet divided by 40 ft.

     Step 9—Calculate Cost-Effectiveness by  Dividing Total Annual Costs (from Step 8) by the
Annual Uncontrolled Emission  Rate  (from Step  3) and by  Desired Control Efficiency Fraction  (from
Step  1)
                                               B-22

-------
        TABLE  B-H.   EXAMPLE  CALCULATION CASE:  APPLICATION OF COHEREX* TO AN UNPAVED ROAD
     This table is an example cost-effectiveness  calculation for controlling PM10  using Coherex*
on an unpaved road in a Detroit, Michigan,  plant.

     Step 1—Specify Desired Average Control  Efficiency

     Desired average control efficiency * 90 percent

     Step 2—Specify Basic Vehicle.  Road, and Cl iinatological Parameters for the Particular Road
of Concern

     Required vehicle characteristics:

     1.  Average daily traffic = 100 vehicles per day;
     2.  Average vehicle weight = 40 ST;
     3.  Average number of vehicle wheels = 6;  and
     4.  Average vehicle speed * 20  mph

     Required road characteristics:

     1.  Actual length of roadway to be controlled * 6.3 miles;
     2.  Width of roadway = 30 ft;
     3.  Silt content = 9.) percent;
     4.  Bearing strength of road =  moderate

     Step 3—Calculate Uncontrolled  Annual  Emission Rate as the Product of the Emission Factor
and the Source Extent


                                                   0.7/M\0.5
          / s\/ S\/W\°-'/w\0'5 /565-o\
E •k 5-9 (liXslXs)    6)   (lr)
where     E = emission factor
          k = 0.36 for PMjg  (from Section 2.0 of this manual)
          s = 9.1 percent (given in  Step 2)
          S = 20 mph (given in Step  2)
          W = 40 ST (given in Step 2)
          w - 6 (given in Step 2)
          p = 140 (as per Figure 2-4 for Detroit, Michigan)

     E = 4.98 Ib/VMT

     SE * 365 x ADT x average one-way trip distance

     SE = 365 42*1 x ,00 vehicles  x  6'3 miles
              year         day      2 vehicle

     SE = 115,000 VMT/year

Emission rate = E x SE


     Emission rate = 4.98 Ib/VMT x 115,000 VMT/year  x   1 short ton


     Emission rate = 286 tons of PM10  per year

                                                                                     (conti nued)
                                              B-23

-------
                                     TABLE 8-11.   (continued)
     Step 4—Consult the Appropriate Control Program Design Table to Determine the Times Between
Applications and the Application Intensity

     Use Table B-12.

     The vehicle and road characteristics listed in Step 2 are similar to those In the footnotes
of Table 2-1.

     From Table B-12:

     Application intensity * 0.83 gal. of 20 percent solution/yd2
                             (initial application)

                           » 1.0 gal. of 12 percent solution/yd2
                             (reapplicat ion)

     Application frequency = once every 47 days

     Step 5—Calculate the Number of Annual Applications Necessary by Dividing 365 by the Days
Between Applications (from Step 4)


     No. of annual applications - ^ = 77.7 applications
                                   47            year

     Step 6—Calculate the Number of Treated Miles Per Year by Multiplying the Actual Miles of
Road to be Controlled (from Step 2)  by the Number of Annual Applications (from Step 5)

     No. of treated miles per year - 6.3 miles x 7.77 applications
                                                          year

     * 49 treated miles/year

     Step 7—Consult the Appropriate Program  Implementation Alternatives Tables and Select the
Desired Program  Implementation Plan

     From Table B-13, the following  implementation plan and associated costs are anticipated:



1.
2.
3.
4.
5.





Selected alternative
Purchase Coherex* and ship in truck tanker
Store in newly purchased storage tank
Prepare road with plant owned grader
Pump water from river or lake
Apply chemical with plant owned application
truck (includes labor to pump water and
Coherex® and apply solution)


Capital
investment, $

30,000

5,000
70,000


105,000
Unit
$/Treated
mi le
4,650


135



17785
cost
$/Actual
mile


630




630
                                                                                       (continued)
                                               B-24

-------
                                     TABLE B-ll.   (continued)
     Step 8 — Calculate Total Annual Cost by Annual! zing Capital Costs and Adding to Annual
Operation and Maintenance Costs

     Calculate annual capital investment (PI) » capital investment x CRF

     CRF » [i(1 «• !)"]/[(! * i)n - 1]
     CRF = capital recovery factor
       i « 0.15
       n « 10 years
     CRF « 0.199252
      PI * 105,000 x 0.199252 * S20,900/year

     Calculate annual operation and maintenance cost (MO)

     MO * $4, 785 /treated mile x 49 treated miles/year + $630/actual mile x 6.3 actual. mi les
                                                                                  year
        = 5238,000/year

     Calculate total cost (D) * PI + MO

     D = $20,900/year + $238,000/year
       = $258,900/year

     Scale total cost by actual  road width

     Actual total cost for a 30- ft wide road = $258,900/yr x 2P_±L
                                                             40 ft

                                             = J)94,200/yr

     Step 9 — Calculate Cost-Effectiveness Dy Dividing Total Annual Costs (from Step 8) by the
Annual Uncontrolled Emission Rate  (from Step 3) and by the Desired Control  Efficiency Fraction
(from Step 1)

     Cost effectiveness
                          ,Q              ft
                          286 ST/year x 0.9

                        = $754/short ton of PM10  reduced
                                               B-25

-------
 TABLE B-12.  ALTERNATIVE CONTROL PROGRAM DESIGN FOR COHEREX*
                APPLIED TO TRAVEL SURFACESa B
Average
percent
control
desired
50
75
90
Vehicle
passes between
applications
23,300
11,600
4,650
Days
as
100
233
116
47
between applications
a function of ACT
300
78
39
16
500
47
23
9
Calculated time and vehicle passes between application are
 based on the following conditions:

   Suppressant application:
   •  3.7 L of 20 percent solution/m2  (0.83 gallon of
      20 percent solution/yd2)  initial  application
   •  4.5 L of 12 percent solution/m2  (1.0 gal.  of 12 percent
      solution/yd2);  reapplications
   Vehicular traffic:
   •  Average weight—Mg (43 tons)
   •  Average wheels—6
   •  Average speed—29 km/h (20 mph)
   Road structure:  bearing  strength—low to moderate
DPM-10 = Particles <10 umA.
 For reapplications that span time  periods greater than
 365 days, the effects of the freeze-thaw cycle  are not
 incorporated in the reported values.
                          B-26

-------
 TABLE B-13.   ALTERNATIVE CONTROL  PROGRAM  DESIGN  FOR  PETRO TAG
                APPLIED TO TRAVEL SURFACES4 D
Average
percent
control
desired
50
75
90
Vehicle
passes between
applications
92,000
47,800
21,200
Days between applications
as a function of ACT
100
920
478
212
300
307
159
71
500
184
96
42
Calculated time and vehicle passes between application are
 based on   the following conditions:
   Suppressant application:   3.2 L of 20 percent solution/in^
   (0.7 gal of 20 percent solution/yd2)? each .application
   Vehicular traffic:
   •  Average weight—Mg (30 tons)
   •  Average wheels—9.2
   •  Average speed—22 km/h (15 mph)
   Road structure:  bearing  strength—low to moderate
°PM-10 = particles <10 umA.
cFor reapplications that span time periods greater than
 365 days, the effects of the freeze-thaw cycle are not
 incorporated in the reported values.
                          B-27

-------
        TABLE B-14.   IDENTIFICATION AND COST  ESTIMATION  OF  COHEREX*
                          CONTROL ALTERNATIVES
Program implementation alternatives
Cost
  I.   Purchase and ship Coherex®

      A.  Ship in railcar tanker (11,000-22,000
            gal/tanker)
      B.  Ship in truck tanker (4,000-6,000
            gal/tanker)
      C.  Ship in drums via truck (55 gal/drum)

 II.   Store Coherex®

      A.  Store on plant property
          1.  In new storage tank
          2.  In existing storage tank
              a.  Needs refurbishing
              b.  Needs no refurbishing
          3.  In railcar tanker
              a.  Own railcar
              b.  Pay demurrage

          4.  In truck tanker
              a.  Own truck
              b.  Pay demurrage
          5.  In drums
      B.  Store in contractor tanks

III.   Prepare road

      A.  Use plant-owned grader to minimize
            ruts and low spots
      8.  Rent contractor grader
      C.  Perform no road preparation

 IV.   Mix Coherex3 = and water in application
        truck

      A.  Load Coherex3= into spray truck
          1.  Pump Coherex3 » from storage tank
                or drums into application truck
      2.  Pour Coherexs = from drums into
            application truck, using fork lift
S4,650/treated mile

54,650/treated mile

$7,040/treated mile
$30,000 capital

$5,400 capital
         -0-

         -0-
$20, $30, $60/treated
  mile

         -0-
$70/treated mile
         -0-
$140/treated mile
$630/actual mile

$l,200/actual mile
         -0-
Tank—0 (included in
  price of storage
  tank)
Drums—$1,000 capital
51,000/treated mile
                                                               (continued)
                                 B-28

-------
                         TABLE B-14.  (continued)
Program implementation alternatives
Cost
      B.   Load water into application truck
          1.   Pump from river or lake
          2.   Take from city water line

  V.  Apply Coherex® = solution via surface
        spraying

      A.   Use plant owned application truck
      B.   Rent contractor application truck
$5,000 capital
$40/treated mile
$70,000 capital+5135/
  treated mile for
  tank or $270/treated
  mile for drums

Tank—$500/treated
  mile
Drums—$l,000/treated
  mile
                                B-29

-------
       TABLE B-15.
IDENTIFICATION AND COST ESTIMATION OF PETRO TAG
      CONTROL ALTERNATIVES
Program implementation alternatives
                                Cost
  I.   Purchase and ship Petro Tac
      A.  Ship in truck tanker (4,000-6,000 gal/
            tanker)
      B.  Ship in drums via truck (55 gal/drum)
 II.   Store Petro Tac
      A.  Store on plant property
          1.  In new storage tank
          2.  In existing storage tank
              a.  Needs refurbishing
              b.  Needs no refurbishing
          3.  In railcar tanker
              a.  Own railcar
              b.  Pay demurrage

          4.  In truck tanker
              a.  Own truck
              b.  Pay demurrage
          5.  In drums
      B.  Store in contractor tanks
III.   Prepare road
      A.  Use plant owned grader to minimize
            ruts and low spots
      B.  Rent contractor grader
      C.  Perform no road preparation
(IV.   Mix Petro Tac and water in application
        truck
      A.  Load Petro Tac into spray truck
          1.  Pump Petro Tac from storage tank
                or drums into application truck
          2.  Pour Petro Tac from drums into
                application truck, generally using
                fork!ift
          Load water into application truck
          1.  Pump from river or lake
          2.  Take from city water line
                                $5,400/treated mile

                                $ll,500/treated mile



                                $30,000 capital

                                $5,400 capital
                                        -0-

                                        -0-
                                $20,  $30,  $60/treated
                                  mile

                                        -0-
                                $70/treated  mile
                                        -0-
                                $140/treated mile


                                S630/actual  mile

                                $l,200/actual mile
                                        -0-
                                Tank - 0 (included in
                                  price of storage
                                  tank)
                                Drums—$1,000 capital
                                51,000/treated mile
                                $5,000 capital
                                $40/treated mile
                                                               (continued)
                                 B-30

-------
                         TABLE B-15.   (continued)
Program implementation alternatives
Cost
  V.  Apply Petro Tac solution via surface
        spraying

      A.  Use plant-owned  application truck
      B.  Rent contractor application  truck
$70,000 capital+5135/
  treated mile for
  tank or $270/treated
  mile for drums
Tank—$500/treated
  mile
Drums—$1,000/treated
  mile
                                B-31

-------
                             APPENDIX C
          SCREENING TECHNIQUES, MODELING  INFORMATION, AND
                      HEALTH RISKS INFORMATION
Screening Analysis for Estimating Maximum Annual Average
Ground-Level Particulate Matter Contaminant Concentrations...   C-l

Estimating Health Effects From Fugitive Treatment, Storage,
and Disposal Facilities (TSDF) Air Emissions	   C-19

-------
Screening Analysis for Estimating Maximum Annual Average Ground-
       Level Particulate Matter Contaminant Concentrations

Introduction

      The purpose of this section is to present a screening
technique for estimating maximum annual average ground-level
contaminant concentrations due to fugitive particulate matter
emissions from treatment, storage, and disposal facilities
(TSDFs).

     The screening technique described in this section is a
simplified procedure sufficiently conservative to allow a
determination whether fugitive particulate matter emissions from
a TSOF would: (1) clearly not result in an air quality threat or
(2) pose a potential threat that should be examined with more
sophisticated modeling techniques or measurements.  If the
screening estimates indicate that ground-level concentrations are
not likely to exceed critical health risk levels, further
analysis of the TSDF air quality impact is not necessary.  If
these screening estimates demonstrate that concentrations may
exceed critical levels, this should not be construed as
indicating a problem.  Rather, it is an indication that more
detailed source impact analyses using refined emissions estimates
and air dispersion models are necessary.

     The screening technique described in this section was
developed expressly for TSDF guidance and should not be construed
as EPA policy for any other source type.  For each TSDF analysis,
the State or U. S. Environmental Protection Agency (EPA) Regional
Modeling contact should be consulted concerning the proper
application of the screening technique and interpretation of
results.  The U. S. EPA Regional Modeling Contacts are identified
in Table 1.
                               C-l

-------
                        Table 1

       Regional Meteorologists/Modeling Contacts
Susan Kulstad
EPA Region I
J.F.K. Federal Building
Boston, HA  02203-2211
(617) 565-3225
Ray Werner
EPA Region II
26 Federal Plaza
New York NY  10278
(212) 264-2517

Alan Cimorelli
EPA Region III
841 Chestnut Building
Philadelphia, PA  19107
(215) 597-6563
Lewis Nagler
EPA Region IV
345 Courtland Street, N.E,
Atlanta, GA  30365
(404) 347-2864

Michael Koerber
EPA Region V
230 South Dearborn Street
Chicago, IL  60604
(312) 886-6061
James Yarbrough
EPA Region VI
First Interstate Bank
Tower
1445 Ross Avenue
Dallas, TX  75202-2733
(214) 655-7214

Richard Daye
EPA Region VII
726 Minnesota Avenue
Kansas City, KS  66101
(913) 236-2896

John Notar
EPA Region VIII
999 18th Street
Denver Place - Suite 500
Denver, CO  80202-2405
(303) 293-1755

John Vimont
EPA Region IX
215 Fremont Street
San Francisco, CA  94105
(415) 974-8223

Robert Wilson
EPA Region X
Environmental Services
Division
1200 Sixth Avenue
Seattle, WA  98101
(206) 442-1531
C
                           -

-------
Limitations and Assumptions

     This screening technique is designed to provide a simplified
approach to obtain upper-bound concentration estimates for
situations that may represent complex release scenarios and
source configurations.  Therefore, the screening methods have
certain limitations such as the following:

     * These methods assume that the particulate matter emissions
behave as gaseous pollutants with negligible deposition
velocities.  Thus, these methods are not applicable for
particulates with appreciable settling or when deposition is of
concern.

     * Steady, continuous emissions for the entire year are
assumed.

     * Complex and elevated terrain effects are not considered.
However, for particulate matter plumes resulting from near
ground-1evel, non-buoyant fugitive emission sources such as those
at TSDFs, maximum ground-level concentrations are likely to occur
at or near the property boundary.  Thus, the assumption of flat
terrain should have no effect on the magnitude of the maximum
concentrations.

     * The screening estimates presume that emissions are
distributed evenly over a square area source representing a
composite of all the fugitive particulate matter emission
sources.  This presumption may oversimplify a complex fugitive
emissions source configuration.
                               C-3

-------
     The TSDF permit writer is strongly encouraged to discuss the
appropriate application of this screening technique with the
appropriate State and/or U. S. EPA Regional Modeling Contact.

Overview of Screening Procedures

     Fugitive particulate matter emissions from TSDFs generally
originate from surface areas such as the following:
     * Roads (paved, unpaved)
     * Open waste piles and staging areas
     * Dry surface impoundments
     * Landfills
     * Land treatment
     * Waste stabilization basins

     Concentration estimates of contaminants contained in the
fugitive particulate matter emissions are needed to estimate
health risk to the surrounding populace.  The purpose of a
screening analysis is to provide permit writers an upper-bound .
estimate of contaminant concentration levels (based on fugitive
particulate matter concentration levels) using simplified
procedures.

     The purpose of this screening analysis is to obtain an
upper-bound estimate of the maximum annual average contaminant
concentration resulting from  fugitive particulate matter
emissions from a TSDF.  This  screening analysis presumes that a
total, combined annual emission rate is known for all applicable
fugitive emission sources.  Annual emission rates can be
estimated using the procedures and emission factors detailed in
this document.  As noted above, the screening technique is based
on the assumption of continuous emissions during the entire year.
                                C-4

-------
Screening Analysis Methodology

     This analysis utilizes Figure 1.which relates normalized
annual average concentration estimates,  CHI/Q (X/Q), to downwind
distances for six generic area source sizes.  The figure contains
a "family-of-curves" - representing the six area source sizes -
which plots normalized concentrations as a function of downwind
distance.*

     The X/Q plots shown in Figure 1 were developed by assuming
unit emission rate per unit area per year (kg/m2/yr) for each
generic area source size.  The procedures to obtain an upper-
bound estimate of the maximum annual average concentration from a
particular TSDF follow.

     Step l:  Obtain an estimate of the total, combined fugitive
particulate matter emission area source size.  This size can be
obtained by first estimating the total area containing each
active source-type (e.g., the total area comprised of unpaved
roads) and then summing the areas from each source-type to obtain
a total combined area source size.

     step 2:  Refer to Figure 1.  From the total, combined area
source size found in Step 1, determine the square area in Figure
       The curves were developed by applying the Industrial
Source Complex (ISC) Long-Term Model based on the meteorological
conditions representative of the general locations of the largest
concentration of TSOF sources in the United States.
     Normalized concentration is defined as concentration divided
by emission density or X/Q.  Alternatively, the normalized
concentration is the concentration that would be predicted if the
emission density were 1 kg/m2/yr.
                               C-5

-------



















er
0
u. en
UJ
UJ M
0 H-l
z en
1- UJ
en u
Si
O
in en
3<
CC UJ
UJ CC
O X
*^^ !•>{
x en





u
c
E £ E E
c O O O O
E o m o o
O *- r- CM K)
m
X X X X
X
c £ £ E E
E o o o o
o o in o o
in i- T- CM K5
i I
i i
i 1


















/
e*
S
/'
^ ^
^"^ ^^^- ".'•-•
i i i i I i i i i I i t i i I i

T o m o
M o r^ m
£
o
o


X

o
o
m
I
!
I


•
/ 1 $
• / ^
I |p

/ | 111
' i ill
i Ml
/i •'*
;J
' : i
/ /;i|

/ / 'it
/ / //a
/ / /^ii
/ ; -: '1
-7 / //if
/ x ;' • f

/ ' / I T

<^yn
^••-•-^U-:-.
\ \ \ \ i- i i i i
A
m | c


^

0)
c_
3
O»
Lu




o
o
CO
0
O


o
o
r—

o
o
0

8
00
o
o
CO

o
o

o
o
CN

^
D















^^
^S.

UJ
o
er

o

4^
LU
CC
U.
0
UJ
a
UJ
o
cc
Lu

UJ
o
en
a




   (6_oix)
C-6

-------
1 that best represents the combined area source size.  For
example, if the total combined area source size from Step 1 were
approximately 10,000 square meters, the square area found in
Figure 1 that would represent this source would be approximately
100 meters by 100 meters.

     Step 3:  Based on knowledge of the TSDF property lay-out
(e.g., plot plan), estimate where the square area determined in
Step 2 would be located within the property boundary in relation
to the active sources used to determine the total combined area
source size in Step 1.
                                                                /
     Step 4!  Determine the total annual contaminant emissions
(kilograms) from all fugitive particulate matter emission sources
that have been combined in this area source.  Convert the total
annual emissions to emissions (kilograms) per square meter using
the area comprised of the square area source determined in Step
2.

     Step 5:  Refer to Figure 1.  Locate the curve that best
matches the square area source size determined in Step 2.  Next,
locate the distance to the nearest property boundary from the
edge of the area source (Refer to Step 3).  Locate the X/Q value
for the appropriate downwind distance.  Multiply the X/Q value by
the fugitive particulate matter emissions per square meter
determined in Step 4.  The resulting concentration represents an
upper-bound estimate of the maximum annual average contaminant
concentration (ug/m3) from the area source.  This concentration
is considered an upper-bound estimate because (1) emissions from
all sources occur within the same area (i.e., are collocated),
(2) emissions occur concurrently, (3) the annual X/Q values in
Figure 1 represent the maximum values obtained from modeling a
range of meteorological conditions associated with the locations
of the largest TSDFs, and (4) the annual distribution of
                               C-7

-------
meteorological conditions is such that the maximum concentrations
are assumed to occur in the direction of the nearest property
boundary from the edge of the fugitive emissions area source.

     6;     Compare the annual average contaminant concentration
obtained in Step 5 with critical health risk levels.  If this
value exceeds the critical annual health risk threshold, more
refined analyses are warranted.  Otherwise, the TSDF facility may
be assumed to pose minimal risk to the health threshold.

     Example Problem:  The following is an example of applying
the screening procedure for a hypothetical TSDF.  Consider the
following hypothetical data:

Fugitive            Annual Contaminant
Source              Emission Rate (Kg)          Source Area (m2)

Vehicular Traffic      0.90                        400

Open Wastepiles        0.50                        400

Dry Surface
Impoundments           0.90                        600

Landfills              0.10                        300

Waste Stabilization
Basin                  0.40                        500
 Total  annual contaminant emissions *  2.80 kg
 Total  fugitive emissions area source  size » 2200 m2
 Thus Q -  2.80 kg/2200 m2 -  .00127 kg/m2/yr
 From Figure 1, the  fugitive area source  size  is best  represented
                                C-3

-------
by the 50m X 50m (2500m2) square area.
Distance from edge of square area source to nearest property
boundary » 100 m.
Using Figure 1 as illustrated, X/Q - 11.8xlO~9 yr/m
Therefore, upper-bound annual average concentration * (X/Q)(Q)
                          - (11.8xlO"9)(.00127)
                          - .OlSxio""9 kg/m3 or .015 ug/m3
Summary

     The  X/Q wfamily-of-curves" shown in Figure 1 were developed
by applying the Industrial Source Complex (ISC) Long-Term Model
for six generic area source sizes.  It is possible that using
other less rigorous screening techniques to estimate maximum
annual average concentrations may in some cases yield more
conservative results than those estimated from Figure 1.  For
example, the U. S. EPA's SCREEN Model4 may be used to obtain
short-term concentrations for area sources.  These concentrations
may then be multiplied by a scaling factor to obtain annual
average concentrations.  These annual average concentrations may
be more conservative than those estimated from Figure 1 for some
distances from the edge of the area source.  However, at this
time, there are no EPA-approved short-term to long-term scaling
factors that may be used for area source screening.

     Additionally, the "family-of-curves" shown in Figure 1 were
developed for square area sources no greater than 500m X 500m.
For larger area sources, it is recommended that the area source
be represented by smaller area sources before screening
techniques are applied.  The State or U. S. EPA Regional Modeling
Contact should be consulted for the proper techniques to be used
in estimating annual average concentrations from area sources
larger than 500m X 500m.
                               C-9

-------
              Detailed Facility Modeling Procedures

Introduction

     The purpose of this section is to present an overview of
detailed facility modeling procedures for estimating maximum
annual average ground-level concentrations due to fugitive
particulate matter emissions from treatment/ storage/ and
disposal facilities (TSDFs)

     If screening results indicate that annual average fugitive
contaminant particulate matter concentrations may exceed critical
levels/ a more detailed modeling analysis of fugitive contaminent
emissions from the TSOF is required.  This detailed analysis is
intended to predict maximum concentration levels from the
fugitive particulate matter sources more accurately than the
screening approach.

     General guidance for conducting dispersion modeling is
contained in U.S. EPA'S Guideline on Air Quality Models
rRevisedl.2  This Guideline may be obtained from:
U. S. EPA                               National Technical
Source Receptor Analysis Branch         Information Service(NTIS)
Techniques Evaluation Section           U.S. Dept. of Commerce
MD-14                              or   Springfield, VA 22161
Research Triangle Park, NC  27711       (703) 487-4650
(919) 541-5685                          NTIS No.: PB86-245248

     This section describes a general approach for conducting a
detailed modeling study of fugitive particulate matter emissions
from a TSDF based upon information contained in the Guideline.
Before proceeding with a detailed modeling study/ the TSOF permit
applicant should ensure that the appropriate U. S. EPA Modeling
Contact  (Table 1) is informed of all facets of the modeling
                               C-10

-------
protocol as it is developed.  This is to ensure the proper
consideration of the latest modeling techniques, procedures and
requirements.

Model Selection

     Fugitive particulate matter emissions from TSDFs are
primarily generated from the following sources:

          1. Roads (paved, unpaved)
          2. Open waste piles and staging areas
          3. Dry surface impoundments
          4. Landfills
          5. Land treatment
          6. Waste stabilization basins

     Because of the complexity in characterizing these emission
sources, the most appropriate model to use for detailed modeling
is the Industrial Source Complex (ISC) model.2  Also, because
long-term (i.e., annual average) concentrations are of concern,
the ISC Long-Term (ISCLT) model is recommended for use in the
detailed study.2  The ISC model is available as part of the
User's Network for Applied Modeling of Air Pollution (UNAMAP)
(Version 6).  The  ISC user's guide3 is available from:

                    Computer Products
                    National Technical Information Service (NTIS)
                    U.S. Department of Commerce
                    Springfield, VA 22161
                    (703) 487-4650
                    NTIS No.: PB88-169487

Additionally, the ISC model is available from various commercial
vendors.
                               C-ll

-------
General Modeling Considerations

Receptor Sites

     Receptor sites for the detailed facility modeling should be
sufficient to estimate the highest annual average concentrations
from the facility at locations to which the public has access.
In designing a receptor network, receptor density and location is
more important than the total number of receptors.  The selection
of receptor sites should be determined case-by-case considering
the topography, climatology, monitoring sites and results of the
initial screening.  However, for near ground-level, non-buoyant
TSDF fugitive particulate matter sources, the highest
concentrations would likely be at or near the facility property
boundary.  Therefore, the greatest density of receptors should be
at or near this boundary.  Receptor and source locations are
specified using a consistent set of polar or Cartesian
coordinates.  If polar coordinates are used/ the origin is
generally located at the facility's geographical centroid.

     The State and U.S. EPA Regional Modeling Contacts should be
consulted concerning the appropriate receptor selection prior to
initiating the modeling study.

Gravitational Settling and Deposition

     The ISC model contains settling and deposition algorithms
that are recommended for use when the particulate matter size
distribution can be quantified.  Additional information
concerning exercising gravitational settling and deposition
algorithms in the ISC model is  described below.
                               C-12

-------
Urban/Rural Classification

     The type of land use in which a TSOF is located is important
to selecting the appropriate dispersion coefficients for use in
the ISCLT model.  The land use in the area of the TSDF must be
classified as either urban or rural.  The Guideline2 details the
procedures to use for determining the land use classification
surrounding the TSOF.

     TSDFs located in an area defined as urban should be modeled
using urban dispersion coefficients, and TSDFs located in a rural
area should use rural dispersion coefficients in the modeling.

ISCLT Program General Description

     Cautionary note - because the proper application of many of
the ISCLT model features requires a fundamental knowledge of the
concepts of atmospheric transport and dispersion, the user should
seek expert advice before using any ISCLT model feature not fully
understood.

     The ISCLT model uses statistical wind summaries to predict
seasonal (i.e., quarterly) and/or annual ground-level
concentrations at receptor sites specified using polar or
Cartesian coordinates.  Additionally, the model is designed to
account for the effects of particulate gravitational settling and
dry deposition on the predicted ambient concentrations.  The
annual concentration predictions are made for a finite number of
wind-speed, wind-direction, and atmospheric turbulence
combinations.  The predicted concentrations are then weighted
according to the observed frequency of occurrence of each
combination.
                               C-13

-------
     The ISCLT model accepts the following source types: stack
(e.g, point sources such as process vents), area and volume.  For
TSDF fugitive particulate matter emission sources, the source
types would likely consist primarily of volume and area sources.
The volume source option may be used to simulate line sources
such as paved or unpaved roads.  The area source option will
likely be applied for sources such as landfills, dry surface
impoundments, open waste piles, etc.  The ISCLT user should seek
expert advice on the proper configuration of the source types
specific to the TSDF.

     The ISCLT model has one rural and three urban modes for
specifying the types of dispersion coefficients to use in the
modeling.  The selection of the proper option is dependent on the
land use classification (rural or urban) in the vicinity of the
TSDF.  As described above, the Guideline2 details the procedures
to use for determining the land use classification.  If the
classification is urban, the Urban Mode 3 option should be
selected in the ISCLT modeling.

Summary of ISCLT Input Data

     The input requirements for the ISCLT program consist of four
main categories:
     * meteorological data
     * source data
     * receptor data
     * program control parameters

Meteorological data

     The ISCLT model utilizes  meteorological data consisting of
seasonal or  annual statistical tabulations of the joint frequency
of  occurrence of wind-speed and wind-direction  categories,
                                C-14

-------
classified according to Pasquill stability categories.  These
"STAR" summaries* are based on National Weather Service (NWS)
hourly meteorological observations.  The Guideline2 provides
additional specifications concerning the appropriate NWS data to
use for the modeling of specific TSDFs.  Typically, the nearest
NWS station that best represents the meteorological conditions at
the TSDF are used.,  The State or U.S. EPA Regional Modeling
Contact should be consulted concerning the appropriate data to
use.

Source Data

     The ISCLT model accepts three source types: stack (or point
sources such as process vents), area, and volume.  For each
source, input data requirements include the source location with
respect to a user specified origin, the source elevation (if
terrain effects are included in the model calculations) and
annual average fugitive particulate matter emission rates.  For
stack or point-type sources (e.g., process vents) additional
source input requirements include the physical stack height,
inner stack diameter, stack gas exit temperature and stack exit
velocity.  Additionally, if the stack or point source is adjacent
to a building and aerodynamic wake effects are a possibility, the
length, width, and height of the building are needed as model
input.  The horizontal dimensions and effective emission height
are required for each area and volume source.  As noted above, it
is anticipated that most sources within a TSDF may be
characterized as area or volume sources.
       The "STAR" summary is a tabulation of the joint frequency
of occurrence of wind-speed and wind-direction categories,
classified according to discrete atmospheric turbulence
categories.
                               C-15

-------
     If gravitational settling and dry deposition are considered
in the modeling, then source inputs must include (1) the mass
fraction of particulates in discrete gravitational settling-
velocity categories, and (2) the surface reflection coefficient
and settling velocity of each settling-velocity category (see
Reference 3, section 2.4.1, pg. 2-25).  (For the modeling of PH10
fugitive participate matter emissions, it may not be necessary to
include gravitational settling and dry deposition in the ISCLT
modeling).

Receptor Data

     The ISCLT model uses either a polar (r, 9) or Cartesian
(X,Y) coordinate system.  The typical polar receptor array
consists of radials (usually one for every 10 degrees azimuth) at
several downwind distances.  However, the user is not restricted
to a 10-degree angular separation.  For example, for the low-
level sources associated with TSOF fugitive particulate matter
emission sources, there may be a need to locate receptors closer
than 10 degrees azimuth at the property boundary.  The polar
receptor array is always centered at X-0, Y=0, (usually at the
facility's geographical centroid).  Receptor locations in the
Cartesian system may be input as Universal Tranverse Hercator
(UTM) coordinates or as X  (east-west) and Y (north-south)
coordinates with respect to a user-specified origin.

     Discrete receptor locations corresponding to other points of
interest  (e.g., population centroids) may be used with either
coordinate system.

Program Control Parameters

     The  ISCLT model allows the user to select from a number of
modeling  options.   Some of these options include:
                               C-16

-------
               * whether to specify concentration or deposition
                 estimates at receptor locations,
               * selection of polar versus Cartesian receptor
                 grid system,
               * specification of discrete receptor locations
                 (e.g., population centroids, monitors),
               * whether seasonal or annual output listings are
                 desired,
               * whether concentration calculations are based on
                 rural or urban mode,
               * output formats options,

     The program control parameters for these and other options
are discussed in detail in the ISC model user's guide3 Section
4.2.3.

     The TSDF permit applicant should consult with a modeling
expert familiar with the ISCLT input and program control
requirements prior to initiating a modeling analysis.
                               C-17

-------
                           REFERENCES

1.  Memorandum from D. Doll to L. Elmore documenting the ISCLT
modeling analyses conducted to prepare Figure 1.

2.  U.S. Environmental Protection Agency, Guideline on Air
Quality Models (Revised"! and Supplement A; EPA-450/2-78-027R,
Office of Air Quality Planning and Standards, Research Triangle
Park, NC, July 1986 and July 1987.  Available from the National
Technical Information Service (NTIS) as PB86-245248 and PB88-
150958, respectively.

3.   U.S. Environmental Protection Agency, Industrial Source
Complex (TSC) Dispersion Model User's Guide — Second Edition
Volume 1 & 2; EPA-450/4-86-005a, Office of Air Quality Planning
and Standards, Research Triangle Park, NC, June, 1986.  Available
from the National Technical Information Service as PB88-169487.

4.   U.S. Environmental Protection Agency, Screening Procedures
for Estimating the Air Quality Impact of Stationary Sources: EPA—
450/4-88-010, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, August, 1988.  Available from the
National Technical Information Service as PB89-159388.
                               C-18

-------
    ESTIMATING HEALTH EFFECTS FROM FUGITIVE TREATMENT, STORAGE,
           AND DISPOSAL FACILITIES (TSDF) AIR EMISSIONS
     Many adverse health effects can result from exposure to
fugitive air emissions from hazardous waste treatment, storage,
and disposal facilities (TSDF).  The major pathways for human
exposure to environmental contaminants are through inhalation,
ingestion, or dermal contact.  Airborne contaminants may be toxic
to the sites of immediate exposure, such as the skin, eyes, and
linings of the respiratory tract.  Toxicants may also cause a
spectrum of systemic effects following absorption and
distribution to various target sites such as the liver, kidneys,
and central nervous system. Generally, the health effects of
greatest concern following intermittent or continuous long-term
exposures are those that cause either irreversible damage and
serious impairment to the normal functioning of the individual,
such as cancer and organ dysfunctions, or death.  Many toxicant
effects can be reversible and disappear with cessation of
exposure.  Health effects sometimes associated with exposure to
toxicants include:  central nervous system effects such as
headaches, drowsiness, and tremors; skin, eye, and respiratory
tract irritation; nausea; reproductive and developmental ef fectaf) v
and olfactory effects such as awareness of unpleasant or
disagreeable odors.
     Exposure to contaminants in air can be acute,  subchronic, or
chronic.  Acute exposure refers to a very short-term (i.e., <24
h), exposure to a contaminant.   Acute exposure to very high
concentrations or to low levels of highly toxic substances can
                              C-19

-------
                                2
cause serious and irreversible tissue damage, and even death.  A
delayed toxic response may also occur following acute exposure to
certain agents.  Chronic exposures are those that occur for long
period of time (from many months to several years).  Subchronic
exposure falls between acute and chronic exposure, and usually
involves exposure for a period of weeks or months.
     The risk associated with exposure to a toxic agent is a
function of many factors, including the physical and chemical
characteristics of the substance, the nature of the toxic
response and the dose required to produce the effect, the
susceptibility of the exposed individual, and the exposure
situation.  In many cases individuals may be concurrently or
sequentially exposed to a mixture of compounds, which may
influence the risk by changing the nature and magnitude of the
toxic response.
1.   CANCER.
    Calculation of cancer risk is based on the hypothesis that
cancer rates in human populations are associated with
individuals' exposure to pollutants present in the ambient air.
In general, the scientific evidence that cancer risks may be
associated with air pollution or specific pollutants in air is of
three main types:  epidemiologic studies of factors associated
with cancer incidence; experimental or laboratory studies of the
carcinogenicity and mutagenicity of substances and mixtures in
the ambient air; and monitoring studies of the presence in the
air of substances known to be carcinogenic.1
                             C-20

-------
                                3
    Cancer characteristically progresses through stages, each of
which may be initiated or accelerated by a number of different
intrinsic and extrinsic risk factors.  Each factor may act at one
or more stages, and different factors may interact in an additive
or a synergistic way.  Because of the long latency period between
initial exposure and manifestation of many cancers (20 years or
more), numerous opportunities exist for multiple exposures to
potentially carcinogenic agents.2
    a.  Estimation of Cancer Potency
        Two pieces of information are needed to assess the cancer
risks of exposure to TSDF air emissions:  (1) an estimate of the
carcinogenic potency, or unit risk estimate, of the pollutants in
TSDF air emissions;  and (2) an estimate of the ambient
concentration of the pollutants from TSDF that an individual or
group of people breathe.
        The unit risk estimate (unit risk factor) is used by the
Environmental Protection Agency (EPA) in its analysis of
carcinogens.  It is defined as the lifetime cancer risk occurring
in a hypothetical population in which all individuals are exposed
throughout their lifetime (assumed to be 70 years) to an average
concentration of 1 ^g/ra of the pollutant in the air they breathe.
Unit risk estimates can be used for two purposes:  (1) to compare
the carcinogenic potency of several agents with one another, and
(2) to give a rough indication of the public health risk that
might be associated with estimated air exposure to these agents.2
                              C-21

-------
                                4
        Unit risk estimates for many of the TSDF pollutants
covered in this report have been developed and reviewed by EPA.
In the development of unit risk factors, EPA assumes that, if
experimental data show that a substance is carcinogenic in
animals, it may also be carcinogenic in humans.  The EPA also
assumes that humans may be as sensitive as the most sensitive
animal species tested and that any exposure to a carcinogenic
substance poses some risk.2
        The EPA uses data from animal bioassay and epidemiologic
studies to predict risk to humans at low exposure to the
substance.  Both the animal and human studies usually reflect
populations exposed to relatively high levels of a given
substance; however, general population exposure to the substance
is usually low.  Because the risks at low-level exposure cannot
be measured directly by animal or epidemiologic studies, a number
of mathematical models have been developed to extrapolate low-
dose risk from the high-dose studies.  There is no single model
which is most appropriate for identifying "true" risk estimates.
EPA (1986) noted that in assessments conducted by the Agency, "in
the absence of adequate information to the contrary, the
linearized multistage model will be employed."  This model has
been used for many, but not all, of the TSDF chemicals.  For more
details of the quantitative assessments the reader is referred to
the Integrated Risk Information System (IRIS)at 513-569-7254.
The Agency has chosen to use  nonthreshold models for dose-
response assessment.  The basis for nonthreshold models is
                             C-22

-------
                                5
primarily theoretical based on how background  incidence  and
heterogenicity in the population effect the shape of population
dose-response curves.   [For more information on this issue the
reader is referred to Hoel et al., 19753; Schneiderman et al.,
19794; EPA, 19845; and Gaylor 19886  .
        The unit risk values predicted from human data are
generally maximum likelihood or "best estimates" of risk.  The
unit risk values predicted from animal data provide a plausible,
upper bound limit on public risk at  lower exposure levels if the
exposure is accurately quantified; i.e., the true risk is
unlikely to be higher than the calculated level and could be
substantially lower.
        For animal data, the mathematical formulation chosen to
describe the linear nonthreshold dose-response relationship at
low doses is applied to original unadjusted animal data.  Risk
estimates produced by this model from the animal data are then
scaled to a human equivalent estimate of risk.  This is done by
multiplying the estimates by several factors to adjust for
experiment duration, species differences, and, if necessary,
route conversion.  The conversion factor for species differences
is presently based on models for equitoxic dose.7  The method
that has been used in most of the EPA's quantitative risk
assessments assumes dose equivalence in units of mg/body weight
2/3 for equal tumor response in rats and humans.  It is assumed
that metabolic rate is roughly proportional to body surface areas
and that surface area is proportional to 2/3 power of body weight
                             C-23

-------
                                6
 (as would be the case for a perfect sphere).  The estimate is
 also adjusted for lifetime exposure to the carcinogen considering
 duration of exposure (either occupational exposure duration or
 duration of the experiment) and animal lifetime.2
        For unit risk estimates for air, studies using exposure
 by inhalation are preferred.  When extrapolating results from the
 inhalation studies from animals to humans, consideration is given
 to the following factors:
            The deposition of the inhaled compound throughout
            the respiratory trace
             Retention half-time of the inhaled particles
             Metabolism of the inhaled compound
             Differences in sites of tumor induction.
        Unit risk estimation from human or animal studies is only
 an approximate indication of the actual risk in populations
 exposed to known concentrations of a carcinogen.  Differences
 between species (life span, body size, metabolism, immunological
 responses, target site susceptibility), as well as differences
 within species (genetic variation, disease state, diet), can
 cause actual risk to be much different.  In human populations,
 variations also occur in genetic constitution, diet, living
 environment, and activity patterns.  Some populations may
 demonstrate a higher susceptibility due to certain metabolic or
 inherent differences in their response to the effects of
 carcinogens.  Also, unit risk estimates are based on exposure to
.a referent adult male.  Exposed individuals are represented by a
                              C-24

-------
                                7
referent male having a standard weight, breathing  rate, etc.  (No
reference is made to factors such as race or state of health.)
There may be an increased risk with exposure by fetuses,
children, or young adults.  Finally, humans are exposed to a
variety of compounds, and the health effects, either synergistic
or antagonistic, of exposure to complex mixtures of chemicals are
not known.2'7
    b.  EPA Unit Risk Estimates
        The EPA has developed unit risk estimates  (URE) for about
72 compounds that are either known or suspect carcinogens and
that could be present at a TSDF.  The EPA has verified about 35
of these.  As shown in Table 1, these URE range in value from 5.8
x 10~7 (ng/m3)'1 for perchloroethylene to 3.3 x 10~5 (pg/m3)"1
for dioxin.  In other words, the URS for perchloroethylene is 5.8
cancer cases per every 10 million people exposed to 1 tig/ra3
perchloroethylene.  The URE for dioxin is 3.3 cancer cases per
every 100 people exposed to l iig/m3 dioxin.  The URE for a given
chemical is multiplied by the estimate of exposure to produce a
risk estimate.  For example, if exposure to perchloroethylene was
estimated to be 10 ng/m3 the risk would be
      10 ng/m3 - 5.8 x 10~7/^g/m3 » 5.8 x 10"6
(or 5.8 cancer cases for every one million people  exposed to 10
jig/m3 perchloroethylene).  The URE are revised upon occasion,
therefore the user is advised to confirm that the  values are the
most correct for the compounds of concern.  To verify, call the
IRIS coordinator at 513-569-7254.
                              C-25

-------
                                8
    c.  Noncancer Health Effects
        Although cancer is of great concern as an adverse health
effect associated with exposure to a chemical or a mixture of
chemicals, many other health effects may be associated with such
exposures.  These effects may range from subtle biochemical,
physiological, or pathological effects to gross effects such as
death.  The effects of greatest concern are the ones that are
irreversible and impair the normal functioning of the individual.
Some of these effects include respiratory toxicity, developmental
and reproductive toxicity, central nervous system effects, and
other systemic effects such as liver and kidney toxicity,
cardiovascular toxicity, and immunotoxicity.
    d.  Health Benchmark Levels
        For chemicals that give rise to toxic endpoints other
than cancer and gene mutations, there is often assumed to be a
level of exposure below which adverse health effects usually do
not occur.  This threshold-of-effect assumption maintains that an
organism can tolerate a range of exposures from zero to some
finite value without risk of experiencing a toxic effect.  Above
this threshold, toxicity is observed as the organism's
homeostatic, compensating, and adaptive mechanisms are overcome.
To provide protection against adverse health effects in even the
most sensitive individuals in a population, regulatory efforts
are generally made to prevent exposures from exceeding a health
"benchmark" level that is hypothetically below the lowest of the
thresholds of the individuals within a population.

                             C-26

-------
                                9
    Benchmark levels, termed reference doses  (RfD), are
operationally derived from an experimentally  obtained no-
observed-effect level or a lowest-observed-effect level by
consistent application of generally order-of-magnitude
uncertainty factors.  The uncertainty factors reflect various
types of data used to estimate RfD.  The RfD  is an estimate (with
uncertainty spanning perhaps an order of magnitude or greater) of
daily exposure to the human population (including sensitive
subpopulations) that is likely to be without an appreciable risk
of deleterious effect.  These benchmark levels are compared to
exposure to qualitatively evaluate risk.  The greater the
exceedance of the RfD the greater the risk.
    The Agency has developed verified oral RfD for a large number
of chemicals, but has only recently established a working group
to begin the process for inhalation RfD.  In the absence of such
inhalation values, the noncancer health effects assessment
utilizes an interim approach for deriving chronic exposure limits
that is similar to that used in the Agency's proposed rule on the
burning of hazardous waste in boilers and industrial furnaces (52
FR 16982).  These interim values will be used until such time as
Agency-verified inhalation RfD become available.
    The Agency is using the following strategy to derive the
interim inhalation benchmark levels:
        (1)  Where a verified oral RfD has been based on an
inhalation study,  the inhalation exposure limit will be
calculated directly from the study.

                              C-27

-------
                                10
        (2)  Where a verified oral RfD has been based on an oral
study, the inhalation exposure limit will be calculated by
converting the daily human oral dose to a corresponding
concentration in air.  Route extrapolation will only be performed
after careful evaluation of the confounding factors affecting the
conversion.  Some of these factors include:  (a) occurrence of
critical toxicological effects at the portal of entry, (b)
differences in systemic effects, (c) first pass effects resulting
in inactivation or activation of the chemical before it reaches
the target organ, (d) effect of route upon dosimetry, and (e)
                                                               A
variations in temporal patterns of target organ concentrations.
        (3)  Where there exist appropriate EPA health documents
containing relevant inhalation toxicity data, those data will be
used in deriving an inhalation exposure limit.  Such documents
include Health Assessment Documents, Health Effects Assessments,
and Health Effects and Environmental Profiles.  Other health
documents (e.g., National Institute of Occupational Safety and
Health criteria documents) or sources of toxicity information
(e.g., data used to support the development of the American
Conference of Governmental Industrial Hygienists (ACGIH)
threshold limit value) will also be considered.  The calculation
of an interim inhalation health benchmark level will be made in
accordance with the Agency's RfD Methodology.
    The methodology for converting oral RfDs to interim
inhalation benchmark levels is as follows:
                             C-28

-------
                                11
Benchmark level (mg/m3) =
                          x Bod  weight x Correction factor
                        mj Air breathed/day
where
          RfD is the oral reference dose
      •    Body weight is assumed to be 70 kg for an adult male
      •    Correction factor for route-to-route extrapolation
          (going from the oral route to the inhalation route) is
          1.0, unless specific data exist that indicate a more
          appropriate value
          Volume of air breathed by an adult male is assumed to
          be 20 mvday.
     Until Agency verified inhalation RfD become available,
alternatives measures might be used.  These alternatives might
include each State's own reference doses, that are usually
derived from Threshold Limit Values .  The Agency's RfDs can be
obtained from the Integrated Risk Information System (IRIS).  The
IRIS user support can be accessed by calling 513-569-7254.
    e.  Chemicals of Concern
        A preliminary list of 179 TSDF chemicals of concern for
the noncancer health assessment is shown in Table 2 along with
the available interim health benchmark levels.  Constituents were
drawn from the Agency's final rule on the identification and
listing of hazardous waste (Appendix VIII of Reference 9) and
from several hazardous waste data bases identified by the
Research Triangle Institute.10  To be selected from these
sources, the chemical must have had either an Agency-verified
oral reference dose (as of September 30,  1987 )8 or a Reference
                             C-29

-------
                                12
Air Concentration (RAC) found in the Agency's proposed rule on
the burning of hazardous waste in boilers and industrial
furnaces.12  Additional chemicals were added to Table 2 based on
knowledge of a high toxicity associated with that substance.
2.  RISK ASSESSMENT
    a.  Maximum Lifetime Risk
        Maximum lifetime risk or individual risk refers to the
person or persons estimated to live in the area of highest
ambient air concentrations of the pollutant(s) as determined by
the detailed facility modeling.  This individual is assumed to
reside at the plant fenceline.  The maximum lifetime risk
reflects the probability of an individual developing cancer as a
result of continuous exposure to the estimated maximum ambient
air concentration for 70 years.  The use of the work "maximum" in
maximum lifetime risk does not mean the greatest possible risk of
cancer to the public.  It is based only on the maximum exposure
estimated by the procedure used 10, and it does not incorporate
uncertainties in the exposure estimate or the risk factor.
        Maximum lifetime risk is calculated by multiplying the
highest annual average concentration at the fenceline by the unit
risk estimate.  The product is the probability of developing
cancer for those individuals assumed to be exposed to this
concentration for their lifetimes.  The following example problem
utilizes the upper-bound annual average concentration for
chromium that was determined in the example problem discussed in
the "Screening Techniques for Estimating Maximum Annual Average

                             C-30

-------
                                13
Ground-Level Concentrations Due to Fugitive Particulate Emissions
from TSDF" located in this appendix.  The unit risk estimate for
chromium was obtained from Table  1 and is 0.012/M-g/m3.
     Maximum lifetime risk =  (highest annual avg. cone.)  (unit
                               risk estimate)
                           =  (15.0 ng/m3) (0.012/ng/m3)
                           -  1.8  x 10'1
Thus, anyone residing at this average concentration for 70 years
would have about a 2 in 10 chance of developing the cancer of
concern.
     b.  Noncancer Health Effects
          (1)  Chronic Exposures
               The assessment of noncancer health effects
associated with chronic exposures to TSDF chemicals of concern is
based on a comparison of the chemical-specific health benchmark
levels (as discussed in Section l.d to estimated ambient
concentrations at various receptor locations around a facility.
Inhalation exposure limit will be compared to the highest annual
average ambient concentrations for each chemical at the selected
facilities.  These annual concentrations represent an estimation
of the highest average daily ambient concentration experienced
over a year.  Ambient concentrations that are less than the RfD
are not likely to be associated with health risks.  The
probability that adverse effects may be observed in a human
population will increase as the frequency of exposures exceeding
the RfD increases and as the size of the excess increases.  The
                             C-31

-------
                                14
results of this assessment will be available after completion of
an ongoing scientific review of the inhalation benchmark levels
discussed in Section l.d.
          (2)  Acute Exposures
               Assessment of the potential for noncancer health
effects associated with short-term (acute) exposure to TSDF
chemicals of concern at selected facilities will be conducted as
a screening effort to provide additional qualitative support to
the overall noncancer health effects analysis.  The extent of
this assessment will be limited by the lack of short-term
exposure limits and by the lack of adequate acute inhalation data
for most of the TSDF chemicals of concern.  The assessment will
be conducted by comparing maximum modeled ambient concentrations
for averaging times of 15 minutes, 1 hour, 8 hours, and 24 hours
to available short-term health data matched to the appropriate
averaging time.  These data will be obtained from various
sources, including EPA reports an documents, data used to support
occupational exposure recommendations and standards (e.g., ACGIH
Documentation of TLVsl.  and other published information.
Determination of the risk of adverse health effects associated
with estimated short-term exposures will be based on a
consideration of the quality of the available health data and the
proximity of the exposure concentration to the health effect
level.  This phase of the noncancer health effects assessment is
incomplete at this time.
                              C-32

-------
                                15
3.  ANALYTICAL UNCERTAINTIES APPLICABLE TO CALCULATIONS OF PUBLIC
    HEALTH RISKS IN THIS APPENDIX
     a.  Unit Risk Estimate
          The procedure generally used to develop unit risk
estimates is fully described in Reference 2.  Nickel was selected
as an example.  The model used and its application to
epidemiological and animal data have been the subjects of
substantial comment by health scientists.  The uncertainties are
too complex to be summarized sensibly in this appendix.  Readers
who wish to go beyond the information presented in the reference
should see the following FEDERAL REGISTER notices:  (1)  EPA's
"Guidelines for Carcinogenic Risk Assessment/1 51 FR 33972
(September 24, 19862), and (2) EPA's "Chemical Carcinogens; A
Review of the Science and its Associated Principles," 50 FR 10372
(March 14, 1985), February 1985.
     Other significant uncertainties associated with the cancer
unit risk factors include:  (1)  selection of dose/response
model, (2) selection of study used to estimate the unit risk
estimate, and (3) presence or absence of a threshold.
     b.  ftiubjent Air Concentrations
          The following are relevant to the estimated ambient air
concentrations used in this analysis:
             Flat terrain was assumed in the dispersion
             model.  Concentrations much higher than those
             estimated would result if emissions impact on
             elevated terrain or tall building near a plant.
                             C-33

-------
                   16
The estimated concentrations do not account for the
additive impact of emissions from plants
located close to one another.
The increase in concentrations that could
result from reentrainment of pollutant-
bearing dust from, for example, city streets,
dirt roads, and vacant lots, is not directly
considered.
With few exceptions, the emission rates are based on
assumptions and on limited emission tests.
                C-34

-------
                                17
REFERENCES

1.  Clement Associates, Inc. Preliminary Health Risk Evaluation
    for Emissions from Hazardous Waste Treatment Storage, and
    Disposal Facilities.  December 1985.

2.  U.S. Environmental Protection Agency.  Health Assessment
    Document for Nickel and Nickel Compounds.  Publication No.
    EPA/600/8-83/012FF.  Office of Health and Governmental
    Assessment, Washington, DC.  1986.

3.  Hoel, D.G.; Gaylor, D.W.; Kirschstein, R.L; Saffoptto. U.;
    and Schneiderman, M.A.  Estimation of Risks of Irreversible,
    Delayed Toxicity.  Journal of Toxicology and Environmental
    Health. 1:133-1511.  1975.

4.  Schneiderman, M.A.; Decoufle, P. and Brown, C.C.  Thresholds
    for Environmental Cancer:  Billogic and Statistical Con-
    siderations.  Annals of New York Academy of Sciences.  Pages
    92-107, 1979.

5.  U.S. Environmental Protection Agency.  Background Information
    Document for Final Rules for Radonuclides.  Volume 1.  EPA/
    520/1-84-022-1.  Office of Air and Radiation Programs,
    Washington, D.C.  1984.

6.  Gaylor, D.W.; Sheehan, D.M.; Young, J.F.; and Mattison, D.E.
    The Threshold Dot Concept in Teratogenesis.  Teratology.
    38:389-391, 1988.

7.  U.S. Environmental Protection Agency.  Health Assessment
    Document for Carbon Tetrachloride.  Publication No. EPA-
    600/8-82-001F.  Environmental Criteria and Assessment Office,
    Cincinnati, OH.  1984.

8.  Pepelko, w. E., and J. R. Withey.  Methods for route-to-route
    extrapolation of dose.  Toxicol Ind Health. 1(4):153:175.
    1985.

9.  U.S. Environmental Protection Agency.  Hazardous Waste
    Management System; Identification and Listing of Hazardous
    Waste; Final rule.  51 FR 28296.  1986.

10. Memorandum from Coy, Dave, RTI, to McDonald, Randy,
    EPA/OAQPS.  May 2, 1986.  Listing of waste constituents
    prioritized by quantity.

11. U.S. Environmental Protection Agency.  Status Report of the
    RfD Work Group.  Environmental Criteria and Assessment
    Office, Cincinnati, OH.  1987.

                              C-35

-------
                                IS
12. U.S. Environmental Protection Agency.  Burning of Hazardous
    Waste in Boilers and Industrial Furnaces; Preamble
    Correction.  52 FR 25612.  1987.
                              C-36

-------
TABLE   1.  TSDF CARCINOGEN LIST

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

Constituent
acetaldehyde
(75-07-0)
acryl amide
(79-06-1)
acrylonitrile
(107-13-1)
aldrin
(309-00-2)
an i 1 i ne
(62-53-3)
arsenic
(7440-38-2)
benz(a)anthracene
(56-55-3)
benzene
(71-43-2)
benzidine
(92-87-5)
benzo(a)p>.-ene
(50-32-8)
beryl 1 ium
(7440-41-7)
bis(chloroethyl)
etner (111-44-4)
bis(chloromethyl
ether (542-88-1)
1,3-butadiene
(106-99-0)
cadmium
(7440-43-9)
Unit risk
estimate.
0*g/nH)-l
2.2x10-6

l.lxlO'4

6.8x10-5

4.9x10-3

7.4x10-6

4.3x10-3

8.9xlO-4

8.0x10-6

6.7x10-2

1.7x10-3

2.4x10-3

3.3x10-4

2.7x10-3

2.8x10-4

1.8x10-3

Basis*
CAG UCR
(class B2)
CAG UCR
(class 62)
CRAVE verified
UCR (class 81)
CRAVE verified
UCR (class 82)
CAG UCR
(class C)
CAG UCR
(class A)
CAG UCR
(class 82)
CAG UCR
(class A)
CRAVE verified
UCR (class A)
CAG UCR
(class 82)
CAG UCR
(class B2)
CRAVE verified
UCR (class 82)
CAG UCR
(class A)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class Bl)
                                 (continued)
              C-37

-------
TABLE   1 (continued)

16.

17.

18.

19.

20.

21.

22.

23.


24.


25.

26.

27.

28.

29.

Constituent
carbon tetra-
chloride (56-23-5)
chlordane
(12789-03-6)
chloroform
(67-66-3)
chloromethane
(74-87-3)
chloromethyl methyl
ether (107-30-2)
chromium VI
(7440-47-3)
DDT
(50-29-3)
dibenz(a.h)
anthracene
(53-70-3)
l,2-dibromo-3-
chloropropane
(96-12-8)
1,2-dichloroethane
(107-06-2)
1, 1-dichloro-
ethylene (75-35-4)
dieldrin
(60-57-1)
diethylstilbestrol
(56-53-1)
2,4-dinitrotoluene
(121-14-2)
Unit risk
estimate.
1.5x10-5

3.7xlO'4

2.3x10-5

3.6x10-6

2.7x10-3

1.2x10-2

3.0x10-3

1.4x10-2


6.3x10-3


2.6x10-5

5.0x10-5
•
4.6x10-3

1.4XLO'1

8.8x10-5

Basis3
CRAVE verified
UCR (class B2)
CRAVE verified
UCR (class 82)
CAG UCR
(class 82)
ECAO UCR
(class C)
CAG UCR
(class A)
CRAVE verified
UCR (class A)
CAG UCR
(class 82)
CAG UCR
(class 82)

CAG UCR
(class 82)

CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class C)
CRAVE verified
UCR (class 82)
CAG UCR
(class A)
CAG UCR
(class 82)
                             (continued)
       C-38

-------
TABLE   1 (continued)

30.
31.
32.
33.
34.
35.
36.
37.
38,
39.
40.
41.
42.
43.
Constituent
1,4-dioxane
(123-91-1)
1 , 2-d i pheny 1 hyd raz i ne
(122-66-7)
epichlorohydrin
(106-89-8)
ethylene dibromide
(106-93-4)
ethylene oxide
(75-21-8)
formaldehyde
(50-00-0)
gasoline
(8006-61-9)
heptachlor
(76-44-8)
heptachlor epoxide
(1024-57-3)
hexachlorobenzene
(118-74-1)
hexachlorobutadiene
(87-68-3)
hexachlorocyclohexane .
(no CAS f)
alpha-hexachloro-
cyclohcxane
(319-84-6)
beta-hexachloro-
cyclohexane
(319-85-7)
Unit risk
estimate.
Ug/mJ)-1
1.4x10-6
2.2xlO-4
1.2x10-6
2.2xlO'4
l.OxlO'4
1.3xlO-5
6.6xlO-7
1.3x10-3
2.6x10-3
4.9xlO'4
2.2x10-5
5.4xlO'4
1.8x10-3
5.3xlO-4
^MHMB^O^MMM^M^M^M^^MH^^^H^BMVM^
Basis3
CAG UCR
(class B2)
CRAVE verified
(class B2)
CRAVE verified
UCR (class B2)
CRAVE verified
UCR (class B2)
CAG UCR
(class B1-B2)
CAG UCR
(class Bl)
CAG UCR
(class B2)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class B2)
CAG UCR
(class 82)
CRAVE verified
UCR (class C)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class 82)
                            (continued)
     C-39

-------
TABLE   1 (continued)

44.


45.



46.

47.

48.

49.


50.

51.

52.

53.

54.

55.


56.


***B™"^ •'•^^•M^^MMMMBM^^M"'*1'
Constituent
gamma-hexach 1 oro-
cyclohexane
(lindane) (58-89-9)
hexachlorodibenzo-
p-dioxin,l:2 mixture
(57653-85-7 or
19408-74-3)
hexachloroethane
(67-72-1)
hydrazine
(302-01-2)
3 -methyl chol anthrene
(56-49-5)
4,4'-methylene-bis
(2-chloroaniline)
(101-14-4)
methylene chloride
(75-09-2)
methyl hydrazine
(60-34-4)
nickel refinery
dust (7440-02-0)
nickel subsulfide
(12035-72-2)
2-nitropropane
(79-46-9)
n-nitrosodi-n-
butylamine
(924-16-3)
n-nitroso-
diethylamine
(55-18-5)
Unit risk
estimate.
3.8xlO'4


1.3x10-6



4.0x10-6

2.9x10-3

2.7x10-3

4.7x10-5


4.7x10-7

3.1xlO-4

2.4xlO-4

4.8xlO-4

2.7x10-3

1.6x10-3


4.3x10-2


M^MBMMM^B^^^^BM^^MMMM^^M^
Basis3
CRAVE verified
UCR (class C)

CRAVE verified
UCR (class 82)


CRAVE verified
UCR (class C)
CAG UCR
(class 82)
CAG UCR
(class 82)
CAG UCR
(class 82)

CAG UCR
UCR (class 82)
ECAO UCR
(class 82)
CRAVE verified
UCR (class A)
CRAVE verified
UCR (class 82)
CAG UCR
(class 82)
CRAVE verified
UCR (class 82)

CRAVE verified
UCR (class 82)

                             (continued)
      C-40

-------
                       TABLE   1  (continued)
      Constituent
                                Unit  risk
                                estimate.
     Basis*
 57.   n-nitroso-                 1.4xlO*2
      dimethyl amine
      (62-75-9)

 58.   n-nitroso-n-               8.6x10-2
      methylurea
      (684-93-5)

 59.  ^n-nitroso-                 6.1xlO-4
     'pyrrolidine
      (930-55-2)

 60.   pplychlorinated            1.2x10-3
      biphenyls
      (1336-36-3)

 61.   pentachlorom'tro-          7.3x10*5
      benzene
      (82-68-8)

 62.   pronamide                  4.6x10*6
      (23950-58-5)

 63.   reserpine                  3.0xl0'3
      (50-55-5)

 64.   2,3,7,8-tetrachloro-       3.3x10*
      dibenzo-p-dioxin
      (1746-01-6)

65.   1,1,2,2-tetra-            5.8x10*5
     chloroethane
      (79-34-5)
CRAVE  verified
UCR  (class  82)
CA6 UCR
(class 82)
CRAVE verified
UCR  (class B2)
CAG UCR
(class 62)
CAG UCR
(class C)
CAG UCR
(class C)

CAG UCR
(class B2)

CAG UCR
(class 82)
CRAVE verified
UCR (class C)
66.

67.

68.

tetrach 1 oroethy 1 ene
(127-18-4)
thiourea
(62-56-6)
toxaphene
(8001-35-2)
5.8xlO-7

S.SxlO'4

3.2x10-3

CRAVE verified
UCR (class C)
CAG UCR
(class B2)
CRAVE verified
UCR (class B2)
                                                  (continued)
                              C-41

-------
                      TABLE   1 (continued)
     Constituent
                               Unit risk
                               estimate.
                       Basis*
69.  1,1,2-trichloro-
     ethane
     (79-00-5)

70.  trichloroethylene
     (79-01-6)

71.  2,4,6-trichloro-
     phenol
     (88-06-2)
1.6xlO-5



1.7xlO"6


5.7x10-6
CRAVE verified
UCR (class C)
CAG UCR
(class B2)

CRAVE verified
UCR (class B2)
72.
vinyl chloride
(75-01-4)
4.1x10-6
CAG UCR
(class A)
aThe inhalation exposure limits are derived from unit cancer
 risk (UCR) estimates,  which were either (1) verified by the
 Carcinogen Risk Assessment Verification Exercise   (CRAVE) work
 group or (2) established by the Carcinogen Assessment Group
 (CAG),  but not yet verified by CRAVE.  The unit risk estimates
 for chloromethane and methyl hydrazine were derived by the
 Environmental Criteria and Assessment Office (ECAO).

Note:  The constituents on this list and the corresponding unit
       risk estimates and exposure limits are subject to change.
                             C-42

-------








m
V
2
ER HEALTH EFFECTS ASSE
*
5
*

5
^
M
a i
;
S
»-

i
t-

















*
L
C ^^

ll
j 1
I"
M




U


^*
k
j
u •

ll
Q ^^
i x •
* "
1









1 ^
01
u
*"
1
J
u






i : i :



-
a

i
S
0)
^ •»
a ^
i
^^ ^j u
1 *-> 1 X*
IA M • •
CM 1 » *»
i e v •
*!-!
l-Z*
9 - 1 f
•3 ? ii
i s • «
o *> -9 —
k 9 • •
x x u u





, , .i
1 1 » •







^
•
! PJ ~* J7
1 • « i
i i «
IA IA S
i * '

•• IA a
I 1 u l» s^
* » ^*
! 0 X X
i • • •
£ -9 _ e
I «D X — O
7< i S
5 ™ *~ ^
5 — c a
O 0| o O





*! •
<- • i




^
i
~~s
ciiMi cyan id* (692-01-6
bon disulfid* (76-16-0
bon onyfluorid*b (363-
— k k
• • •
W U U





s : 5




^^
CM
1
S
<•% 1
f a
s a
1
«••. s?
^^ C '
k CM
*. I ?
TJ *• f^
— X S
k *i P<
0 I ^
— CM
x i u
u — e
X —

V 1 U




1 1 1
1 1 1



IA
1
m

bon t*trachlorid*c (60
ord*n*c (12789-03-6)
orin*b (7782-60-6)
• x x
u u u





1 1 1
1 1 1






^
1
1*1
** T

5 Iff

i <
a u 35
^ • w
^* X 1
— S
«•••««
• *»
u e u
— o •
k r •«




! S




£"

g
oroac*t«ld*hyd*b (107-
hloro-l,3-butadi*n*b
126-99-8)
X 1
U CM





1 1 1
1 1 1






~
1
IA
<-> S
« 1
n s

^•^ p^
CM S U
8~-x"
•
o> "o —
S X k
m o o
•»» o —
u • u
e
k X X
•o _ _




1 . .
1 1 I



1
f^
1
M
orofor*^ (67-66-3)
oro«*than*b>c (74-87-3
h 1 oropr op i on i tr i 1 *b (6
X X 1
u u m





fM 1
1




~
1
m
o- X
i ^
a o
IA —
SO
M

O
9 W
•o —
•~ 1
a i
o x i?
X X I
M?
a o m
e e •
1 8 S




|. : S! : : s ! :





^^
/•* j? i
? j?8 J |
T ^ i j?S ~?
» T 2 *? 5 J? *^
V • ^ *• +* ~* 1 X-
» ^ ^ <^ IA a _§
^ V X 1 X • 1 •* 4
M — •» X • 1 C O
•1 > •>-'««IA^'X
•-— k— e • •« a a
• liooe— oo
xxokkaxxx
uuouuwuuu





1 1 1 M 1 S S 1 1
III 1 CM CM 1 1






1
~ IA
IA IA
V l» r»
CM • 1
^ f» CM 0 Q
S<^ •>—/-. i i ^. 7
f» S 0 CM IA IA
~ i x i  » i «

X v •» -•
— «D • » f« i •B a
•0*9 C » *- » — «-» ^
.. ^ • ^ v e
X - *• X • •
ax Cxx<~'uc
o • a e u — u
e — .- o - i f -9 —
.« »^2^55^K
••CCBC9w*MQ
C g g •-• — •• M M
• liwKki.ee
^••••xxxx



!





^
1
nog*n chlorid* (606-77
•
x
w





i











p»
i
^^
^
t
^
f»
^^
uf
—
X
k
X



I 1
1 1





u

loh**anon* (108-94-1)
D (dichloroph*noxy*c*
cid) (94-76-7)
U V 01
x -
W M





1 «
i e






u
S
m
A •
v x
i «*
CM X
IA &

CM ^
^^ X
- x ^
X — 1*
e x i
f «* «
a • i

I f £
— X



1
1







J?
a
CM
S
^•r
U
t






1
1






1
CM
IA
^

•
•
X

E
b
0
u
^
!
X
e
4*
e
w






































C-43

-------










(continued)
«M !
UJ I
b*nch*ork
. 3\c
S =
: !
2
•
r
e





"•
y
o

ji
b
a j
2i! "^
i
i
s
1
? ' '
" 1 1
»




•a
^* •
*• •
$ r
1 1
s~
— o
« "
Si
j s
•£ U <-«
** — «
• ^ ^
S S 7
x 1 8
f ^ ^
• *.



SiT1
i"il
i i ^
- | . S «
!j - =» I ? ?
'' J Ij
:'« •!
1 ; b "~
•
1 ' e
< ' M
i
1
1 '

i I
( 1
i




^^
ia
i
p«
t ^*. ^^
SM •"
l 1
— » •
p« h. JZ
** 1 1 .
^ IA '
• CD S
V W ^ L
s s V
-i - • f
•1 * • 5
ui e x e
•i 1 4 * 4-

1 *.'!•!!! 2 .' ! 1 1 I S 9 ! « ;
•» •


XK
^
i
N.
C" *
M «^ | f^
_ i m i>. ||
* f i »* 7 -.
• iftaw^J, C1 ^
£• S »T.T?ffir I
* * Nl^eMi>!.Y9M •*
i • t!«§S?iDT,; 7
nli? in^ifflii i{
iii*»i*M 1 inm
imsgJI iinn::!ll
iillifiMliiHllliii
* *• ' «2 i i a ^•>g^^|||jjjj






5 I S : 2 - : 1 - j j j . , j § , 5 ,
•v •




•^ •— /—>
hk CD V
1 ' — » 1 — . •">•
e M m » «7 w
V (^ I m ^^ . 1
. . ^ ^^ " V
* 1 ^ I «y IA ^
8IA m IA i S "^ «O
** 1 1* •» I J? ^ Wl<-s
-'-wC-Sei ? J?7 AS10
rifci4£ ? i?-— s T
imcu. i^isis^s ?^c^
it'S-S-fSci^-- a I 2 ••
SiSl^S.^.^-^ -^^-^222,'
ebJx^SS-^t1*- J^'-eoiTCT?-:
»o55.eb^-B-^iC ewci o»ia*
> a • • 1 a » ^ 2 ? Z T = -5 «va • S S « .

! 8 ! S 85





<*^ -^
•? T
«; i
a 7
Mrcury (7439-97-0)
•wthacry 1 on 1 ir i 1 *b ( 1 20-1
MothcMiyl (107K2-77-6)
Mthonyclort (72-43-6)
•tothyl broaldo (broiMMMtl
(74-83-0)






cu
• 1 1 1 1 1
W i I i i i
4*
U
*^
















1
1

1
1


i
1
!
1 1
« 1
M* S «
*" w ^^ flD
O Ok **. ^. i ^
1- « » •» ~ W 1
O I 1 1 M 1 1
- « CD - l IA





X • ^ ^ r» r*.
u M i i a> —
^* ^ ~ 5 *
? u T 8 ^ 0 '
) U ^% ^« »« • '
e 9 *~ ~ — jt i
> •- e a • i


^
                            C-44

-------











,
5"
9
3
O
CM
111 I
*\
h-


i

























J
j'.
ii
J?
J -
M





8
w





L
j
UX
e >•*
x"*I
X J
± 5
•
1 • —
X •
>
' • •
»• ^
1.
•
*>
e
r f*






i

•
u
j

; w

'


: s







a
•4
1
M 1
^^ ^4
3J
e •

f 5
a b
1 X
MA.
A





!















9
1 10
•" 10
„,* t
! T
o •
"5 5
u Ai
e o
I I,
o —
X •-
X V.
I

•




X
0
•o
i
j
J
°w
•
O
j
"•






1















1
10
*•*
1
o
^
<&
u
e
X
5
f

8







i
•
"""" £•
i
?«
?
« 3
C 3
K S
i —







S !















? _
1 1
15 S
«^ i
*
? £
o
*> u
Jl X
X —
X T)
J 0
X X
X X
1 1










?
CM
1
^

>
—






1












*^
1
•
1


•
5

•
e
X
X
I

I ,







oT
0
t
*«
^ CM
1 A
x a
u •^
 >
^ ^






t f*
1 (t













S
1
oT S
i *
m 10
0 N
CM X
s> *
x T
• *>
e e
• ••»
u •
e —
x
x w
X •
I N

,







M
N
1
M
CM
«
S
J

_
o






•















0s
1
1
S
CM
e
0
x

L.
a
X
x
f

i
i







i
m
m
i
<•)

j
~e
u

__
o






1
1













^
m
a
m
i
0
m

u
x
x
L
U
0
JC
u
e

 x /«. *» i i CM a *s
S« » i • 0 r« CM 10 u> 10 e
010** i i « » i i •
>^ ^ 0 ij • 0 ot i i a •« w
• o 7 ^" a c S a 0 • 0 •
• e CM e o CM S" 0 >«•(». 0 •••• c
N • «. c x it 2 >^««<-'«^f
Sx ^ • a 0 CM *-• • •—• i
*» xo « «n >-> J • • x
f^SSSS? SSSJJfcl'gif
MT 1 X X X X X -W ^ 0 U • X — • 9 i
1 ^ V U ^ U ^ • ^ Al U U C A • ^^


• W»— •* • « • — — - — — — — — I 5T
tt ft • . • • • tXXXXXXXXX






800 CM 90
CMI i » i iS00miM00i i
0I« II 1 1 •" » M • C CM0II
0









~+ «
IA 0

1 10 10 1 1"» •» 0
CM 1 /-> 1 ^-. O) II I
i 10 i a i i *. m «« 9
9 **CMI«CM^ II 010
^ •» ••09910 10CM iv ^
*.»<-, X i 0 i ~^ 1 00 9 9
ir»a «000 0 CM** 10* I
a<— i^-w^^r-00 10 i -o ^ CM
-i mm— 0 •— • c i N* ^ — 10 —
IUVIX0U • f- ^— J .— 10 C 1 1
r««»ll00CM>UN0 * CM~«0I^
10 • CM a i ^^x « c >-• x *> ^\ i — • x ui 0
i099i'» «x« • • u> « y i -
— '-o— «ixc«x — x-> c y i us • 9>^
9a««*eoocM— «T 1*01.2
x x — •-'C-OXBI.CI e <« m •> 0 a x
• L ^ — — cxwiu>«yi0c>«»
•«• o — x»** — xa — — i00« — CM —
e — •Bccuoooo»»3>-'>-'yii u
• »>—«.-«>kkbk0cu y •>
x«»N9*eooo»«ux • f e • —
y L e c o- «J — — — — v— | • c 3 3 -o c
• 0 X X X X X I C .--.-.-•
— — OXUCuyyU— C— OXABfiC
• •— ow3a***o*xa&n«aa
jt jt L u •- — w^4j*)cj:e««««ca
yy^*>e6cece«atooj^oo
— — — — 1 ••••••Xj.XXXOOkl.
e e c c T o a. a a a a E a a. a. a & a a.

i
i





^»
*
i
•J
M
b
U
I
I
x






1
t
















0
7
fNk
^
C
j
X
o
u
a.
e
e
3
|














































C-45

-------














f



1
I



.



^^
1

e
g
u


*N 1

UJ
a
^






































U
j
^
*;
•2
j
b
e
**

























i

^J
j
u
c
•
! •*

5
fj
' •
X

|
»M
L

* ^
e

i







i













U
•<
|

"•
±
















^»
•
u
J
<•»












X

«*
1
>»
i

•
*

4
M














«
«
d

1
!
.c
V.







i ^
i «









*^
T ^
flb **
w 1
1 «
8 J
• w
e ^

a
2 4
*•;
* J
X S
u &
i §
• —
- •
M e
» •













^



|














/•K
' x*. m
t OB i
n i a>
•• to »
C4 1
i l CD
!! f^ Q

*•• ^^
^^
si
b 3
X *O


1
1



















^
1
P4
«
«
^^
_c
•
^
•












o












~
1
N
•
1

2

!
0
^
^
b
o

x

••
b
+>
1
•
•


1
1
















^K
INk
1
s
s
«


***
•
^
1
^
>t












1
1












^
u>
1
8

a

*.
^
•
e
b
o

^

*•
b
i
»
"^


i
i















^,
\
*•
2^
i
p>
U>
^
j

*e
»>
u
e













*
p«
















•
X
1
e
3

^
o
e


Q
b
O
X

b


1
1













^^
A
1
•^
<*>
^^
**
4
2
o.
•
&
u
e








































^^

i
a
e
t
U)

"

















j




^^
7
«
M
M
M
!••
U
• 1
c 1





i
i




,

*












^
^
i
g
i
g

•
X
a.
o
b
o

^

.«
^
«>
i
u>
•
*-
X
e
• <^
o 2
-S
*»oe

U M
e j*
u
b •
— u
e
• 4.
I*
c —
^ ••
• b
b 4»

j|
^
e -9
c
^ £
a
• b
5 J

• e
— •
*
b •
3 t
Ik •
3
• 0
b ^
• U
• M
3 •
^
•o e
e
• a
e
Is
O 3

if
b e
O 0
a.
a> •

e 's
3 b
• •B
b •
j» 2.

0 i.
• a
i — •

9 °
: §.§
' X •»
u
X •
•> b
^ ^
• 0
t u e
x •
• a
C — • o
•:l£ .5
i • • ae u
<*» • b
~*l '
. • £ u












•
a
§
x
u

o
*»
^
0
•«^
^J



•

^
b


U
e
^
A
X
—
^
g

J

b
«»
e



jj
^

•o
e

«i
J
-
n
•^
x

o
•
tu«nt
•~
4^
m
e
o
u


X
bJK

f
I
C-46

-------
           APPENDIX D





SAMPLING AND ANALYSIS PROCEDURES
              D-l

-------
                                  SECTION  1

                                 INTRODUCTION


     As Indicated throughout the preceding text, for a given facility the most
reliable estimates  of contaminated  part1culate emissions are  obtained using
site-specific  information  as developed  through a well-designed  S&A program.
This appendix outlines procedures and general Information to be used in devel-
oping  an  S&A  program.   The  appendix  borrows  extensively from  information
contained in References 1 and 2.

     The overall  goal  of  the  S&A  program  1s  to  collect  and  analyze  (for
physical and  chemical properties)  "representative"  samples  of the  soil  and
surface material  at TSDFs.   The  Information generated  from  the  exercise, in
turn,  feeds  directly  into  the emission factor models discussed  in  the  main
text.  The sampling parameters of concern include the following:

          Silt content

     •    Silt loading (for paved surfaces)

     •    Moisture content

     •    (a) level of contamination

     One of  the  most  critical  decisions  required  in development  of  the S&A
plan  involves specification  of  a  sampling  design.   In practice,  sampling
designs can  be divided into two  broad  categories—those that  are statistical
or  probability based,  and  those that  are  based  on professional  judgment.
Whenever possible, statistically based schemes  are preferred as resulting data
often  can  be reduced  to  provide  measures of  sampling  and analysis reliabil-
ity.   Probably the most  widely applied statistical  schemes  are:   (1) simple
random and (2) systematic random sampling.  Both of these schemes proceed with
reference to  a sample grid like  that shown in  Figure 1-1.   Under the simple
random design, a  predetermined number (n) of individual  samples are collected,
one from each of  n randomly selected grid cells.  Under the systematic design,
the  initial  sampling grid  cell  is  randomly selected; subsequent  samples are
then  collected  at a regular  interval (e.g.,  every  other grid  cell) in one or
more  directions from  the  starting point.  Detailed discussions of statistical
sampling designs  can be found  in References 3 and 4.

      Implicit  in  the  use  of  statistically  based  schemes  is a  substantial
commitment  of  resources  for  both  sampling  and  laboratory  analyses.    If
resources  are very limited,  it may be  desirable to use judgment sampling to
produce  initial  data for emissions estimates.   In  essence,  judgment sampling
 implies  that  the  person  responsible   for   collection   can  choose  sampling
 location(s) that  are broadly  representative of  the source in question.


                                      D-2

-------
CFSRiO

WO TV. j-
 PERCENT


  LENGTH
   T



  _ I
   7
SOS +•
     30S
   -f-70?:
                ICS
                           30f.
                                   50*




J
0
i O
*
O
O
*..... 	
O
O
O
O
O
O
O
O
O
O
0
O
0
c
O
0
O
O
O
-
                  X	X    OUTlcS SP'KE « STAKES AND ?LASriC ?H<-.GiU(5



                   -*-  PROCESS BOUNDARY AND FUGS OR STAKES



                     O    SAMPLING LOCATION  '
                                                           /
       Figure 1-1.   Sample  grid for hypothetical  TSDF source.
                                       D-3

-------
                                  SECTION 2*

        COMPOUNDS MEASURED AND DETECTION  LIMITS OF ANALYTICAL METHODS


     The compounds measured  in assessing the  degree of contamination  of the
soil  fractions  are  metals,  cyanide,  semivolatlle  organics,  oil and  grease
(land  treatment samples),  and  pesticides/PCBs.    The  list of  semivolatile
organics and pesticides  for  which analyses are conducted  and  their  detection
limits as presented were developed from  the Hazardous  Substance List (HSL) in
EPA's  Contract  Laboratory  Program (CLP)  Statement of  Work.s   The CLP was
chosen because  the large number  of samples  that  have  currently been analyzed
through  this EPA  program  provide  the  maximum opportunity  for  technology
transfer studies.    The  CLP also draws  heavily  from the procedures  in EPA
SW-8468 which allows  other data to be utilized.

     The detection limits for pesticides (Table 2-1) and semivolatile organics
(Table 2-2) are based on extracting 30 g  of material as  specified by the low-
level extraction procedure  in  the CLP.   If other organic:  material  is present
in  significant  quantities,  sample cleanup  procedures will  have to  be used.
Ultimately, the  laboratory  may have  to  dilute the  sample extract  to protect
the analytical equipment, and these detection limits may not be achievable.

     The quantifiable  detection  limits listed for the metals  (Table 2-3) are
those that can be obtained for the compounds listed using the analytical meth-
ods  described  in this protocol.   In the case  of chromium, samples  are ini-
tially  analyzed for  total  chromium  content.   If the results  for  any  sample
show  a  relatively significant  concentration of chromium, then another aliquot
of that sample  is analyzed for hexavalent chromium (the most toxic form) using
procedures in SW-846.   Cyanide content is determined colorimetrically follow-
ing EPA Method 335.26 with a detection limit of 0.5 yg/g.

      If the user of this protocol has knowledge of the compounds present or is
only  interested  in a few compounds,  the  quantifiable  detection limits  may be
improved  through the  use  of  more  specific analytical  techniques  and  more
sophisticated sample cleanup procedures.
      The material  in this section  is taken directly from Reference 2.


                                      D-4

-------
           TABLE 2-1.  PESTICIDE DETECTION LIMITS
  1                  ALDRIN                          8
  2                  Alpha - BHC                     8
  3                  Beta - BHC                      8
  4                  Delta - BHC                     8
  5                  Gamma - BHC                     8
  6                  CHLORDANE                      80
  7                  4,4'-ODD                       16
  8                  4,4'-DDE                       16
  9 .                 4.4'-DDT                       16
 10                  DIELDRIN                       16
 11                  ENDOSULFAN I                    8
 12                  ENDOSULFAN II                  16
 13                  ENDOSULFAN SULFATE             16
 14                  ENDRIN                         16
 15                  ENDRIN KETONE                  16
 16                  HEPTACHLOR                      8
 17                  HEPTACHLOR EPOXIDE              8
 18                  TOXAPHENE                      160
 19                   AROCLOR  1016                    80
20                   AROCLOR  1221                    80
                             •
21                   AROCLOR  1232                    80
22                   AROCLOR  1242                    80
23                   AROCLOR  1248                    80
24                   AROCLOR  1254                   160
25                   AROCLOR  1260                   160
                           D-5

-------
TABLE 2-2.  SEMIVOLATILE ORGANIC DETECTION LIMITS
Number
1
2
3
4
S
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Compounds
ACENAFHTHENE
ACENAPHTHYLENE
ANTHRACENE
BENZO (•) ANTHRACENE
BENZOZC ACID
BENZO («) PYRENE
BENZO (0hl) PERTLfiWE
BENZO (b) rLUORANTHENE
BOKO (k) FLUOSANTHENE
BENZYL ALCOHOL
BZS (2-CHLOROETHOXY) METHANE
BIS (2-CHLQftOETHYL) (LTHUt
BIS (2-CHLOROZSOPROPYL) ETHER
BZS (2-fiTHXU£JUfL) PHTHALATE
4-BROMOPHENYL PHENYL ETHEB
BUTYL BENZYL PHTHALATE
4-CHLOROAMZLZNE
4-CKLORO-3-METHYLPHENOL
2-CHLORONAPHTHALENE
2-CHLOROPHENOL
4-CHLOROPHENYL PHENYL ETHER
CHRYSENE
OZBENZO («.h) ANTHRACENE
DZBENZOFUKAN
1.2 OZCHLOROBENZENE
1.3 DZCRLOROBENZENE
1.4 DICHLOROBENZENE
3.3' -OZCHLOROBENZIOINE
2. 4-OICHLOROPHENOL
DZETHYLPHTHALATE
2 . 4-OZMETHYLPHENOL
OZMETHYL PHTHALATE
DZ-M-BUTYLPHTHALATE
2 . 4-OXNZTROPHENOL
2 . 4-OZNZTROTOLUENE
2 . 6-OiflH'KUI'OLUENE
DZ-M-OCTYL PHTHALATE
FLUORANTHENE
FLUORENE
HEZACHLOROBENZENE
BEZACHLOROBUTAOIENE
HEZACHLOROCYCLOPENTAOZENE
BEZACHLOROETHANE
ZNDENO(1.2.3-ed) PYRENE
ZSOPRORONE
2-METRYL-4 . 6-OZNTTROPHENOL
2-METHYLNAPHTHALENE
2-METRYLPHENOL
4-METHYLPRENOL
NAPHTHALENE
2-NZTROANZLZNE
3-NZTROANZLZNE
4-NZTROANZLINE
NITROBENZENE
2-iiL2KOni£NOL
4-NZTROFHEROL
N-NZTROSO-OZ-if-PROPYLAMZNE
N-N1TROSODZPHENTLAMUJE
PENTACHLQROPHENOL
PHENANTHRENE
PHENOL
PYRENE
1 . 2 . 4-TRZCHLOROBENZENE
2.4. 5-TRZCHLOROPHENOL
2.4. 6-TRZCHLOROPHENOL
' Detection
(ug/kg)
330
330
330
330
1600
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
1600
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
1600
1600
16OO
330
330
1600
330
330
1600
330
330
330
330
1600
330
                        D-6

-------
  TABLE 2-3.  METALS, MEASUREMENT METHODS,  AND  QUANTIFIABLE DETECTION LIMITS
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Element
Aluminum (Al)
Antimony (Sb)
Arsenic* (As)
Barium* (Ba)
Beryllium (Be)
Bismuth (Bi)
Cadmium* (Cd)
Chromium* (Cr)
Cobalt (Co)
Copper (Cu)
Iron (Fe)
Lead* (Pb)
Manganese (Mn)
Mercury* (Hg)
Molybdenum (Mo)
Nickel (Ni)
Osmium (Os)
Selenium* (Se)
Silver* (Ag)
Thallium (Tl)
Vanadium (V)
Zinc (Zn)
Measurement Method**
ICAP
GFAA
GFAA
ICAP
ICAP
ICAP
ICAP
ICAP*
ICAP
ICAP
ICAP
ICAP
ICAP
Cold Vapor AA
ICAP
ICAP
ICAP
GFAA
ICAP
GFAA
ICAP
ICAP
Quantifiable
Detection Limi
(ve/g)
40
1.0
1.0
0.7
0.1
10.0
0.4
0.7
0.7 , '
7-3
100
10.0
5.9
0.25
9-0
2.2
4.0
1.0
10
1.0
3-9
0.2
Eight RCHA metals
ICAP » Inductively-Coupled Argon Plasmography
GFAA * Graphite Furnace Atomic Absorption
AA » Atomic Absorption

Other methods are used to measure  hexavalent chromium (Cr IV) ,  if appropriate
(see page E-4) .
                                    D-7

-------
                                  SECTION 3

                          FIELD SAMPLING PROCEDURES
3.1  SAMPLING METHODS
     The procedures for sampling the soil/surface material  at  HW TSDFs can be
divided Into four basic types—sweeping, scooping,  coring, and vacuuming.

     Sweeping refers to the use of  a whisk  broom and dust pan to remove loose
surface material from the underlying hard packed surface.  The material should
be  swept  carefully  so that  the fine  dust  is  not  injected  into the  atmo-
sphere.   The hard road base  below the  loose surface material  should  not be
abraded so  as  to generate more fine material than  exists on the  road in its
natural state.

     Scooping  refers  to   the use  of  a  small  garden   spade   or  comparable
implement to obtain samples  of  near surface material from a  potential dust
emitting source.  Note that Reference 2 calls for  the use of small disposable
scoops that may be appropriate depending upon the nature of the material being
sampled.

     Coring  refers  to  the use of  a simple coring tube  to  extract samples to
nominal depths of 15 cm (6 in).   Two  types  of coring tubes are employed:  one
made of stainless steel (to collect samples for organic analyses) and one made
of  PVC plastic  (to  collect   samples  for  metals  analyses).    The  procedure
involves driving the core tube into the surface (with a mallet or other com-
patible tool),  extracting the core tube from the soil,  and  forcing  the soil
core into the sample container by pushing a wooden dowel through the tube.

     Vacuuming  refers  to  the use  of  a common household  vacuum cleaner to
remove surface material from a paved roadway.  Note that the vacuum bag should
use removable, disposable  paper bags.

     In  the  following table the  sampling  procedures   are  cross-referenced
against potentially applicable source categories.
                                      D-8

-------
            Source Category                    Sampling Procedure

            Paved roads                        Vacuuming
            Unpaved roads                      Broom sweeping
            Waste piles                        Scooping, coring
            Dry surface Impoundments           Sweep1nga
            Landfills                          Scooping, sweeping
            Land treatment                     Coring, scooping
            Stabilization/solidification       Scooping, coring
            a  Assumes wind erosion 1s the only particulate
               generating mechanism.


3.2  UNPAVED ROADS

     For unpaved roads, MRI typically  has  recommended  collection of a minimum
gross sample  of  23 kg (50 lb) for  every  3.8 km (3 mi) of unpaved  road.   The
incremental samples  from unpaved  roads  should be  distributed over  the  road
segment as  shown  in  Figure 3-1.   At  least  four Incremental  samples should be
collected and composited to form the gross sample.  Figure 3-2 presents a data
form to be used for the sampling of unpaved roads.

3.3  PAVED ROADS

     In industrial  facilities,  like  HW  TSDFs, MRI  typically  has recommended
that one gross  sample should  be obtained for  each  road  segment in the facil-
ity.  The gross sample should consist of  at least two separate increments per
travel  lane, or each 0.5 mi length should have a separate sample.

     Figure 3-3 presents a diagram showing the location of incremental samples
for  a  four-lane road.   Each  incremental  sample  should  consist  of  a lateral
strip 0.3 to  3 m  (1  to  10  ft) 1n  width across a travel lane.  The exact width
is  dependent  on the  amount  of  loose  surface  material on the  paved roadway.
For  a  visually dirty road, a width  of 0.3 m (1 ft) is sufficient;  but  for a
visually clean  road,  a width of  3 m (10  ft) is needed to obtain an adequate
sample.

     The above sampling procedure may be considered as the preferred method of
collecting surface dust from paved  roadways.   In  many instances, however, the
collection  of eight   sample increments may  not  be feasible due  to  manpower,
equipment, and traffic/hazard limitations.   As an alternative method, samples
can  be obtained  from a single strip across  all the  travel  lanes.  When  it is
necessary to  resort  to this sampling  strategy, care must be  taken  to select
sits that have dust  loading and traffic characteristics  typical of the entire
roadway segment  of interest.   In  this situation, sampling from  a  strip  3 to
9m  (10 to  30 ft)  in width is suggested.   From this width,  sufficient sample
can  be collected,  and a step toward  representativeness  in  sample acquisition
will be accomplished.
                                      D-9

-------
                                                                    •o
                                                                    10
                                                                    o
                                                                    O)

                                                                    id
                                                                    C
                                                                    0


                                                                    O

                                                                    1/1
                                                                    0>
                                                                    en
                              o
                              -3
 c
_2
cs
                              C
                              u
                             o
                             CN
 o

 c

\
                                                S3
                                                  •



                                                II
CU

U
                                                                    O
rd
U
O
                                                                    I
                                                                   m
                                                                    j-
                                                                    3
                                                                    O>
                               =M
                       D-10

-------
Scmple
No.
Time
^
Laccfion
Surface
Area
Deoth
Gucnf ify
of Scrr.sie
Use cooe given  on plant  rr.ap for segment idennficatTcn  and indicate sample
 location on map.
           Figure 3-2.   Sampling data form  for unpaved roads,
                                     D-ll

-------
..»





CJ
c.
m^
l»
M
3
L.
>•
•
••
W
••
3
•V
•^
m^
•
a:
3



'
t














03






T







•
fx






n








•o






f^
M^HBHBB^-^—
-^««— «^—
                                  O
                                 I
                                  Q.
                                  «
                                  (O
                                  4->
                                  U
                                  C
                                  o
                                  
-------
     Samples are removed from the road surface by vacuuming, preceded by broom
sweeping 1f large aggregate 1s present.   The samples should be taken from the
traveled portion of the lane with the area measured and recorded on the appro-
priate data form.  With a whisk broom and a dust pan, the larger particles are
collected  from  the  sampling  area and  placed  1n  a clean,  labeled container
(plastic jar or bag).  The remaining smaller particles are then swept from the
road with an electric broom-type vacuum sweeper.  The sweeper must be equipped
with a prewelghed, prelabeled, disposable vacuum bag.  Care must be taken when
Installing the bags 1n the sweeper to avoid torn bags which can result 1n loss
of  sample.   After the  sample has been  collected,  the bag  should  be removed
from  the  sweeper,   checked  for  leaks   and  stored  in  a prelabeled,  gummed
envelope for  transport.   Figure 3-4 presents  a data form to  be  used for the
sampling of paved roads.

     Values for  the  dust  loading on only the  traveled portion of the roadway
are needed for inclusion  in the  appropriate  emission factor equation.  Infor-
mation pertaining to dust loading on curb/beam and  parking areas is necessary
in  estimating carry-on potential to determine  the  appropriate industrial  road
augmentation factor.

3.3  LANDFILLS, LAND TREATMENT UNITS,  DRY SURFACE IMPOUNDMENTS

     Reference 2  provides  a  detailed  sampling  protocol  that  is  potentially
applicable to  landfills,  land treatment units,  and  dry surface impoundments.
This  sampling  approach is  outlined below;  note that  the discussion  can be
referenced to Figure 3-5.
     For the  systematic grid  sampling, a  5 x 5 rectangular  grid matrix  is
used.  This grid  is superimposed over  the process  area.   Sample aliquots are
then collected from the process  by  sampling  the center of each of the 25 grid
cells.

     To locate the sampling grid on the process, a rough sketch of the process
can  be made  (see  example  in  Figure  3-5).    A form  such as  that  shown  in
Figure 3-6 may be  used  to  sketch the  process  and sampling  grid  in the field.
Considering the   shape  and  dimensions of  the  process,  the  grid should  be
superimposed over the process so that:

     •    The maximum area  of  the process  (at  least 85%  by visual approxima-
          tion) is Inside the grid.

     •    No  more than  25% (by  visual approximation)  of the  grid  area  is
          outside the process boundaries.

Once they are determined, Figure 3-6 may be used to  record the grid dimensions
and orientation.
     Material taken directly from Reference 2.


                                     D-13

-------
Type of Moterici Sampled:

Site or' Sampling:
Type of Paveme-.r:  Aiehoit/Concrete
  No. of Traffic Lanes
  . Surface Condition _
   f
   t Sample Nc.   IVac.Sao    T:nie
   I
location*
Sanple Area
                                , 3room
                                ' Swept?
    "Jse code ;iven on plant  map  for  segment identification ana indicate sample
     "ocif.or or nac.
          Figure 3-4.   Sampling  data form for paved roads.
                                     D-14

-------
C?5SiO
WOT* i-
 PESCENT
  05 '33i'0
  LENS"*   v_
iCS

   -•
     303
SOS -
                IOS
                           30?.
                                               7*f»
                                               v*«
i
! 0

0
,
1
'
i O
(
O
O

0

O

O
O
O

0

0

O
O
O

O

O

O
c
G

0

O

c
O
,-r
                                                                          PCV -
                                                                         ROV:
                  X	X    gUncS SPIRE « STAKES AND ?LAS~C ?UGGiWJ


                   -•-  PROCESS BOUNOAfiY AND R>SS OR STAKES


                     O    SAi1PL:.NQ LCCA~C,N  '
                                                           /
        Figure  3-5.   Sample grid for hypothetical  TSDF  source.
                                       D-15

-------
   5iT£ foAi^E


   LOCATION
   PROCESS NAME
   SAMPLING TECHNIQUE

          TEAM	
I 'OOS
i !OS
1 T«r
| WWI
! SOX

i
i :
t i
i i
1 70S ! ' '
! 90s : !
                                      PROCESS AND 6RID UYOOT

             (IndictU Proctss Boundvin & Oimtnsiora. Grid Oimmcian & OHmtation. md Sampling Locations.)
     Total Area msiae Grid ILxWi
     "otai Ar«i of Processes losioe >irid (inciuoe Names)


     Total Area Covered Inside Gna (Note Covering/ 	


     T«ai £ft>c:i\* Process Area ..isiae<3nc 	
Figure  3-6.    Example  form for sketching process  and  grid specifications  and
                 recording sampling  information.
                                               D-16

-------
     The following technique  is  recommended for marking the  sampling grid on
the process  itself.    First,  drive  wooden  stakes or  gutter spikes  into the
ground at  three  of the grid  corners.   Then, run  lengths  of  plastic flagging
between the  stakes  or spikes to  define two perpendicular  sides  of the grid.
The table 1n Figure 3-6 may be used  to calculate the distances along the grid
length and width used to locate the sampling points.

     For example,  to  locate  the points on  the  first row of  the  grid,  mark a
point  along  the flagging  at 10%  of  the  grid  length.    A test  team  member
located at that point can then feed  the appropriate length of surveyor's tape
or chain to  another  test  team member  as he  moves  across  the  grid parallel to
the grid  width which  has  been marked with plastic  flagging.    The sampling
points will  be at  10%, 30%,  50%,  70%, and  90% of the grid  width.   As these
points are located  they should be staked or flagged.

     Carefully document the relationship of the grid to process boundaries and
other  landscape features.  This  includes estimating the  square  area of those
areas  within the grid that are  permanently covered preventing  wind erosion.
These  areas  include  standing  water,  rock piles, gravel,  drums,  etc.  The sum
of these  areas must  be subtracted from the total area of  the  grid (length x
width) which  will  be  used in  modeling the fugitive  particulate emissions.
Areas of other processes not  being examined may fall within the grid area such
as the active  face  of  a landfill in  a landfill  area,  or an access  road or
equipment access area; these  areas must also be subtracted from the grid area.

     In some cases, there may be  some  obstacle  to sampling at the center of a
grid cell.  These obstacles may  include:   (1)  standing water, (2) waste drums
or other waste containers,  (3) parked  equipment,  (4)  rocks,  and  (5) signifi-
cant holes at  the  location.   Under  these circumstances,  the  sample should be
taken  at  another  suitable  point  in  the cell as  close to  the center as pos-
sible.   If an  entire cell  is covered  by one or more  of these obstacles, then
that cell  is not sampled and as previously stated,  all  the area covered is
subtracted from the total  area of the grid.

     Some process  configurations  and/or relationships  to  other  processes may
increase the difficulty in choosing  a  rational  grid  location.  One such situ-
ation  is  illustrated in Figure  3-7.   In  this  case,  the  grid  is located to
maximize process coverage.   It covers a portion of another process (the access
road)  and  a  portion  of  a  pond.   Two of the  grid cells  are not  sampled at all
since  their entire area covers a  separate process  and/or an area of permanent
coverage  (the  standing water in  the  pond).    As previously discussed,  the
permanently covered  area within  the grid  (the  pond) must  be subtracted from
the  total  area  of the  grid and  the  area of the access road  (a separate
process) within the grid must also be subtracted.  If these grid cells instead
covered  nonprocess  related land  that  was  not  permanently covered, then  it
would  be appropriate to sample them  (see the next example).

     In another case  illustrated  in  Figure  3-8  (1) a  separate process (active
face of  a landfill)  falls  in the middle  of the  process  being  sampled (the
landfill)  and  (2)  one entire grid  cell falls  outside  the process.   In this
case,  all the grid  cells are  sampled  (see Figure 3-8 for locations).  The grid
is superimposed over the process  so that the active face will  not completely


                                     D-17

-------
Figure 3-7.   Example sketch showing optimal  sampling grid location on process and sampling
                   locations resulting from  process configuration.
                                           D-18

-------
                                                       v>
                                                       vt
                                                       V
                                                       o
                                                       o  c
                                                      1-  O
                                                      4J f-
                                                       « -M
                                                       O  «
                                                       O  J-
                                                      V.  C
                                                      O) O
                                                          u
                                                      •— OJ
                                                      o. u
                                                      E O
                                                      (O J_
                                                      (/) Q.

                                                      O. O)
                                                      o c
                                                      C f—
                                                      •»- 3
                                                      2 en
                                                         (/I
                                                      f C
                                                      u o
                                                      -* id
                                                      
-------
cover any grid cells  while still  maximizing process coverage.   For the cells
where the active face covers the center of the cells, the samples are taken as
close to  the  center as possible.  For  the one cell totally outside  the pro-
cess, the sample 1s taken at the middle since 1t 1s nonprocess  related and Its
area will  be  Included in  that  used  for the  emission  models.    In  situations
such as  the  preceding two examples,  the grid should be  situated to  optimize
the process coverage while Introducing the least bias possible.
     Following this protocol, it 1s recommended that one  soil  aliquot (excep-
tions previously  addressed)  be  taken from  each  grid cell sampled.   The five
allquots from each of the five rows  (refer  to Figure 1-1) are then composited
Into a single sample container.

     Implementation of this protocol  requires considerable equipment.
                    *
     The  following  is an  inventory and  description of  the function of  the
equipment.  Quantities and  physical  specifications for each  item  used 1n  the
referenced EPA study* are  presented  in Table 3-1.   The  tester may substitute
similar equipment that will accomplish the desired goals.

     Surveyors Chain—For measuring process dimensions and laying out sampling
     grids.

     Surveyors Tape—For measuring and laying out sampling grid.

     Wooden Survey  Stakes.  Gutter Spikes,  Survey Flags--For  marking  process
     boundaries, grid axes, and sampling points, as needed.

     Plastic  Flagging—For flagging  gutter  spikes  and  stakes,  and  marking
     process boundaries.

     40 Quart Cooler—For transporting sample jars and coolant.

     Plastic  Sheet  Roll—Ground  cloth on  which  to set  coolers  for  sample
     marking and storage.

     Carboy (20 gal)—Contains distilled water for rinsing and decontamination
     of tools.

     Disposable Scoop—For taking near subsurface soil samples.

     Glass Jars—Contain and transport soil samples, without contamination.

     Cap  Liners—Seal glass jars.

     Plastic Core Tube—For collecting core samples for metals analysis.
     Material taken directly from Reference 2.


                                     D-20

-------
                TABLE 3-1.  SAMPLING EQUIPMENT SPECIFICATIONS
    Description
    Dimension*
Material*
Quantity*
Surveyors Chain
Surveyors Tape
Plastic Flagging
Wooden Survey Stakes
Survey Flags
Gutter Spikes
40 Quart Cooler (s)
200' long. 1/4" wide
100' long, 3/8" wide
1-3/10" x 50 yds
1" x 2" x 18"
4" x 5" x 30"
10" long
Adequate to hold
Steel
Steel
Plastic
Wood
Plastic
Aluminum
Plastic
1
2
1 carton
200
100
50
3
Plastic Sheet Roll


Carboy


Disposable Scoops


Glass Jars


Cap Liners

Plastic Core Tube

Steel Core Tube

Dowel


Surveyors Hammer

Wallpaper Paste
Brushes

Vacuum Sweeper



(continued)
 sample jars

12' x 100' roll


20 gallon


190 mm long x lid ml
capacity

Nominal 475 nl capacity
with wide neck

To fit glass jars

30 cm long x 3.2 cm I.D.

30 cm long x 3.2 ca I.D.

40 cm long x 2.5 cm
diameter

5 Ib x 18" handle

7" handle, 3" bristles,
6" wide

N/A
  5 mil poly-           2
  ethylene

  Nalgene               1
  or glass

  Styrene             300
  Glass with          300
  phenolic cap

  Teflon              300

  PVC                  24

  Stainless steel      24

  Wood                 48


  Steel/wood            1

  Plastic with         25
  nylon bristles

  With nylon bristle    1
  attachment and
  utilizing disposable
  paper dust bags
                                     D-21

-------
                            TABLE 3-1 (continued)
Description
Shovel
Pick-ax
Bucket ;
Dimension
Standard long handle
Standard long handle
12 liter
Material
Steel/wood
Steel/wood
Stainless steel
Quantity
1
1
1
Bottle Brush


Plastic Bags
12" long x 1-1/2"
diameter

Assorted: 2-quart and
20 gallon
Wire with plastic        2
bristles

Polyethylene           50 each
tiarKing rens
Log Book
Compass
Standard 8-1/2" x 11"
Liquid filled,
5° increments
rermanenc :uuc
Hard cover
Plastic/glass
£U
1
1
•These are the dimensions, materials of construction, and quantity of  equipment
used in the referenced EPA study;9 the tester may substitute similar equipment
that will accomplish the desired goals.
                                      D-22

-------
     Stainless Steel  Core  Tube—For  collecting core  samples  for organlcs
     analysis.

     Wooden Dowel—For pressing  cored coll  from the metal  and plastic core
     tubes.

     Surveyors Hammer—For driving core tubes Into soil.

     Wallpaper Paste Brushes—For sweeping  and collecting  road  dust.

     Vacuum Sweeper—For  collecting road  dust from paved surfaces.

     Shovel—For  general  excavation.

     P1ck-ax—For general  excavation.

     Stainless and/or Plastic Steel  Bucket—For washing and decontamination of
     sampling tools.

     Bottle Brush—For cleaning and decontaminating  core tubes.

     Plastic Bags—Contain contaminated equipment prior to decontamination and
     materials for disposal.

     Permanent Marking Pen—For marking sampling scoops/jars.

     Bound Log Book—For  recording field  notes.

     Compass—For orienting  processes on  site  plan  and  laying  out process
     sampling grids.

     Site Description Forms—For recording the  layout and  condition of each
     process site at the  time of  sampling.

     Chain-of-Custody Forms—For  tracing the  possession  of  the samples from
     origin to analysis.

3.4  WASTE PILES

     In  sampling the  surface  of  waste  piles  to  determine   representative
properties for use in the wind erosion equation,  a gross sample  made  up of the
top, middle, and bottom  incremental  samples  should  ideally  be obtained since
the  wind disturbs  the   entire  surface  of  the  pile.    However,  it  may  be
impractical to climb waste piles.

     The most practical  approach  in  sampling from  large  piles  is to minimize
the bias  by  sampling as  near to  the  middle  of  the  pile  as  practical  and by
selecting sampling locations  in  a random fashion.  Incremental  samples should
be obtained along the entire perimeter of  the pile.  The  spacing between the
samples  should  be  such   that  the  entire  pile  perimeter is  traversed  with
approximately equidistant  incremental  samples.  If  small piles are sampled,
incremental samples  should be collected from  the  top, middle, and bottom.


                                     D-23

-------
     An Incremental sample (e.g., one shovelful)  1s  collected  by skimming the
surface of the  pile  in a direction upward along  the face.   Every effort must
be made  by the  person obtaining the sample  not  to purposely  avoid  sampling
larger pieces of raw material.  Figure 3-9 presents a data form to be used for
the sampling of storage piles.

     In obtaining  a  gross  sample for the purpose of characterizing a load-In
load-out process,  Incremental samples should  be taken  from the portion of the
storage pile surface:   (1) which  has  been formed  by the addition of aggregate
material, or (2) from which aggregate material 1s being reclaimed.
                                     D-24

-------
Type or Mcfericl Sc.T.pled:.
Sire of Scrr.ciir.c: _^_____
SAMPLING METHOD
  1. Scrupling device:  poinred shovel
  2. Scrnpling deprh: 10 - 15 cm (-i -6  incr.ej)
  3. Sample ccnfciner: metcl or plcs.Sc bucicet wifh sealed poly liner
  -. Gross simple specincsfions:
     (c)  I scmple of 23kg (50 Ib.) rr.inirnurn for  every  pile sampled
     (b) composife of 10 incremenrs
  5. Minimum portion of jfared mcrericl (cf one jire) fo be jcmpled:  25%
SAMP UNO  DATA
Scmpie
No.












Tin-*









i


Lscrricn (Sefer rs mcs












Ic -
iurrccs
J Arec
1











Oeofh












Quenrir/
of Sc.T!cie












           Figure 3-9.   Sampling data form for storage  piles.
                                      D-25

-------
                                  SECTION 4*

                        SAMPLE HANDLING AND TRANSPORT


     This section  describes the  specific  techniques used to  maintain sample
Integrity.

     To avoid  contamination,  field equipment  that  will  be exposed  to sample
material  should  be transported on-s1te  1n sealed  bags  or coolers.   Collect
scooped samples  using  disposable, Individually wrapped,  sterile, nonreactive
plastic  scoops.    Deposit  the  contents of  the  scoops  directly  into sample
jars.   In the  case of  cored samples,  clean the core tubes before sampling and
pack them in sealed plastic bags.   Deposit core sample aliquots directly into
the  sample  jars  from the core tube.   Collect swept samples using  a new dis-
posable brush  (transported to the site in  sealed plastic bags);  disposable
scoops  are  used to  deposit the  swept  samples  into  the  glass  sample  jars.
Collect vacuumed samples using a new vacuum sweeper bag transported on-site in
a sealed plastic bag.   Transfer the contents  of the sweeper bag directly into
the sample jars.**

     In all cases, fill sample jars  completely,  leaving  no head space.  Label
each sample jar with the facility sampled, date, process description,  and sam-
ple number, seal with electric tape, and store it in a cooler at a temperature
less  than 20°C  to minimize  loss of  volatile components.   Use  a chain-of-
custody form for each  process.    The  form  should  identify  each sample and all
personnel  having custody  of  the  samples  at  all  phases  of  sample handling.
Many agencies  will have their own chain-of-custody forms and  procedures which
should  be used by their  staff  and contractors.   An example  chain-of-custody
form in shown  in Figure 4-1.

     When  all  samples  have been  collected  for  a  particular  site,  move the
sample  storage cooler  to  an uncontaminated area and decontaminate it  prior to
transport.   This  decontamination procedure  should  include washing with soap
and  water and  a  final rinse with distilled water.

     Samples should be  transported from  the  field in sealed coolers.  Surface
transportation or  air  freight can be used to transport the  samples for long
distances;  however,  during transport, sample  jars must  be maintained at tem-
peratures   less  than   20°C  to   prevent  the  evaporation  of  any   volatile
components.   Pack samples  to  be  shipped  by  air  freight in insulated, impact
resistant coolers  and cool  with "blue-ice," an airline-approved coolant.  Make
 *    Material  taken directly from Reference 2.
 **   Disregard this step;  leave vacuum  bag intact until  laboratory analyses,


                                      D-26

-------
  DATE:
                    COOLER NO.
SITE NAME:


LOCATION:
                           CHAIN OF CUSTODY FORM
                    CASE A
                   CASEB
LABELS CHECKED:

SAMPLES COOLED:
SAMPLES INVENTORIED:

COOLER CHLLED:  	
SAMPLE TRANSPORT :

SAMPLE RECEIVED:
             •
SAMPLE ANALYS5S:
JAR LIDS CHECKED AND SEALED:
               CASES SEALED:

               COOLER SEALED:
                     Figure 4-1.  Example chain-of-custody form.
                                        D-27

-------
arrangements with  laboratory  personnel so  that  samples can be  picked up and
transferred to the laboratory as quickly as possible.

     Federal regulations governing the  shipment  of hazardous wastes are found
1n  Title  40 of  the Code of  Regulations,  Part  261.    However,  Section 261.4
exempts samples  of "water,  soil, or  air collected for  the sole  purpose of
testing to determine  characteristics  or composition" when the  samples  are
being  shipped  to  a  laboratory for  analysis.    When  shipping samples  to the
laboratory, the tester should comply with shipper requirements and provide, as
a minimum:

     •    Tester's name, mailing address, and telephone number.

     •    Laboratory's name, mailing address, and telephone number.

     •    Date of shipment.

     •    Quantity of sample.

     •    Description of sample.

     •    Suitable packaging as previously discussed.

     Prior to the initial analyses (moisture, silt, and PM-20 determinations),
keep all  field samples  in  a locked refrigerator or cool area at a temperature
less than 20°C.  During the drying, screening, and sieving operations,  samples
must be handled  using  techniques to prevent contamination  (e.g.,  using clean
gloves).  To prevent dispersal  of contaminated soil in sample handling areas,
conduct screening and  sieving operations  within a closed  system  such  as a
glove  box.   All   equipment  that comes into contact with  soil  samples must be
decontaminated initially  and then decontaminated  after each use,  or disposed
of  in the appropriate manner.

     If the chemical  analyses are to  be  performed by  another laboratory, the
resulting  silt and  PM-20  samples should be placed in small  amber sample vials
for transport  or shipment  for  further analysis.   Use  10-mL vials for metals
samples and 40-mL  vials  for organic  samples.    In  cases where they are not
shipped  or  analyzed  immediately,  store  samples at  or  below  20°C.   For
shipping,  pack samples carefully (in bubble pack  in a Styrofoam cooler), and
use "blue-ice" as  the coolant to  keep  sample temperatures at or below 20°C.

     All  samples should be extracted for metals and organics analysis within
14  days of collection.   If,  for some  reason  they must be  kept  for a longer
period,  they  should  be  stored  at  4°C.   All  samples should  be  completely
analyzed  within  40  days of  extraction.
                                      D-28

-------
                                  SECTION 5

                        LABORATORY ANALYTICAL METHODS


     As Indicated 1n Section A.I, the sampling parameters of concern Include:

          S1lt content.

     •    Moisture content.

     •    S1lt loading.

     •    (a) level of contamination.

The first  three parameters  represent  physical  determinations, the  latter of
course,  involves  analytical chemistry.   The  following discusses  analytical
methods for each of the parameters.   Note  that  there are large differences in
sample quantity depending upon whether the  sampled  source  is a waste pile, or
unpaved road versus a landfill,  land treatment unit,  or dry surface impound-
ment (sampled by  the  protocol  recommended in Reference  2).   For  convenience,
the former is referred to as Case A,  the latter as Case B.

5.1  SAMPLE PREPARATION

     Sample  preparation  Case  A—once  the  gross  sample  is  brought  to  the
laboratory,  it  must  be reduced  to a workable size.   Given  that  the material
may contain  unknown  levels of contaminant,  the most  practical procedure  for
sample reduction probably involves cone and quartering.

     The  procedure   for   coning  and  quartering   is   best  illustrated   in
Figure 5-1.   The following  is a description of  the procedure:   (1)  mix  the
material and shovel it  into  a  neat cone;  (2) flatten the cone by pressing the
top without  further  mixing;  (3) divide  the flat  circular  pile into  equal
quarters by cutting or scraping out two diameters at right angles; (4)  discard
two opposite  quarters;  (5) thoroughly mix the  two  remaining quarters, shovel
them into  a  cone, and  repeat  the quartering and discarding procedures  until
the sample has  been  reduced to 0.9 to 1.8 kg (2 to  4  Ib).   Samples  likely to
be affected  by  moisture or  drying must  be handled  rapidly,  preferably  in an
area  with  a controlled  atmosphere, and  sealed in a  container to  prevent
further changes  during transportation and storage.   Care must be  taken that
the material  is not contaminated by anything on the  floor or that  a portion is
not lost through cracks or holes,  preferably, the coning and quartering oper-
ation  should  be conducted on  a  floor covered  with  clean  paper.   Coning  and
quartering is a simple procedure which is applicable to all  powdered  materials
and to sample sizes ranging from a few grams to several hundred pounds.


                                     D-29

-------
Figure 5-1.  Coning and quartering.
                D-30

-------
     The size  of  the laboratory  sample  1s Important—too  little  sample will
not  be  representative and  too much  sample  will be  unwieldy.   Ideally,  one
would like  to analyze the  entire gross  sample 1n  batches,  but  this  1s  not
practical.   While all  ASTM standards  acknowledge this  1mpract1cal1ty, they
disagree on the exact size, as  Indicated  by  the range of recommended samples,
extending from 0.05 to 27 kg (0.1 to 60 Ib).

     The main principle 1n  sizing the laboratory sample 1s to have sufficient
coarse  and  fine portions to  be  representative  of  the material and  to allow
sufficient mass on each sieve so that the weighing 1s accurate.  A recommended
rule  of thumb  1s  to have  twice as  much coarse  sample  as  fine sample.   A
laboratory sample of  800  to 1,600 g  1s recommended  since that 1s the largest
quantity  that  can  be  handled  by  the  scales normally  available  (1,600-g
capacity).  Also,  more sample  than  this  can produce  screen  blinding for the
8 in diameter screens normally  available.  Note that 1t may be more practical
to perform sample reduction (i.e., cone and  quartering) prior to shipping the
sample back to the laboratory.

     Sample preparation  Case B—under the sampling  protocol,  it  is assumed
that sampling procedures yield  five  samples  (in separate sample jars).  These
individual  samples  are  subjected to physical  analyses and  then ultimately
combined into a single silt composite.  The flowchart for the various analyses
is shown in Figure 5-2.

5.2  SILT CONTENT

     The  basic recommended  procedure for silt analysis  is  mechanical,  dry
sieving after moisture analysis  (see  5.3  below).  A step-by-step procedure is
given in Table 5-1.   The  sample should  be oven-dried for 24 h at 230°F before
sieving.  The sieving time  is  variable;  sieving should be continued until the
net  sample  weight collected in the  pan  increases by  less than  3.0%  of  the
previous net sample weight collected  in the pan.  A minor variation of 3.0% is
allowed since  some sample grinding due to interparticle  abrasion  will occur,
and  consequently,  the  weight   will  continue  to increase.   When the  change
reduces to 3.0%, it  is felt that the natural silt has been passed through the
No. 200 sieve screen  and  that  any additional  increase is due to grinding.  An
example silt analysis form is given in Figure 5-3.

5.3  MOISTURE CONTENT

     For moisture  content determinations, MRI  has typically  recommended  the
procedure  given  in   Table  5-2.   An example   recording  form  is  given  as
Figure 5-4.

     The protocol given in  Reference  2 recommends  a  somewhat  different set of
procedures.  These procedures are given below.
                                     D-31

-------


S SAMPLE JARS OF SOIL
ONE
475 ML
SAMPLE JAR
ONE
475 ML
SAMPLE JAR
ONE ONE
475 Ml 475 ML
SAMPLE JAR SAMPLE JAR

V V 1
ONE
475 ML
SAMPLE JAR

LOO DETERMINATION
AND OIL & GREASE
~* DETERMINATION
10 g FROM EACH JAR

i i
DESICCATION OR OVEN DRYING
CONTENTS
OF JAR
CONTENTS
OF JAR
CONTINTS CONTENTS
OF JAR OF JAR

y u 1
CONTINTS
OF JAR

r ir ir
SCREEN FOR SILT CONTENT. <75 ym
CONTENTS
OF JAR
CONTENTS
OF JAR
CONTENTS CONTENTS
OFJAR Of JAR

I i


CONTENTS
OF JAR

r ^r 7
COMBINE ALL SAMPLES AND MAKE HOMOGENEOUS
NEED 40 g
FOR ANALYSIS
REMAINDER OF SILT
^r ^ v
DIVIDE SAMPLE
30 g VIAL
10 g VIAL
T
v ir
SONIC S.EV.N6 ARCH.VE SAMPLES

75 TO 10 i>m

HOg
LESS THAN 10 \m
OPTIONAL
SEAL
PORTION
IN JAR
OPTIONAL FOR QA/QC PURPOSES
T V
MAKE HOMOGENEOUS AMD DIVIDE SAMPLE
DIVIDE SAMPLE
30 g VIAL
10 g VIAL
75 to 10 jim
DIVIDE SAMPLE
30 g VIAL
10 g VIAL
Less Than 10 \tm
OPTIONAL
7
OPTIONAL
^

ANALYZE SAMPLES FCR
METALS. PESTICIDES. SEMIVOLATILE ORGANICS. CYANIDE

DISCARD
MATERIAL



Figure 5-2.  Flow diagram for samples taken from a process.
                           D-32

-------
                 TABLE 5-1.  SILT ANALYSIS PROCEDURES
1.  Select the appropriate 8-in. diameter, 2-in. deep sieve sizes.  Recom-
    mended U.S. Standard Series sizes are:  3/8 in., No. 4, No. 20,
    No. 40, No. 100, No. 140, No. 200, and a pan.  Comparable Tyler Series
    sizes can also be utilized.  The No. 20 and the No. 200 are mandatory.
    The others can be varied if the recommended sieves are not available
    or if buildup on one particulate sieve during sieving indicates that
    an intermediate sieve should be inserted.

2.  Obtain a mechanical sieving device such as vibratory shaker or a Roto-
    Tap.

3.  Clean the sieves with compressed air and/or a soft brush.  Material
    lodged in the sieve openings or adhering to the sides of the sieve
    should be removed (if possible) without handling the screen roughly.

4.  Obtain a scale (capacity of at least 1,600 g) and record make,
    capacity, smallest division, date of last calibration, and accuracy
    (if available).

5.  Tare sieves and pan.  Check the zero before every weighing.  Record
    weights.

6.  After nesting the sieves in decreasing order with pan at the bottom,
    dump dried laboratory sample (probably immediately after moisture
    analysis) into the top sieve.  Brush fine material adhering to the
    sides of the container into the top sieve and cover the top sieve with
    a special lid normally purchased with the pan.

7.  Place nested sieves into the mechanical device and sieve for 20 min.
    Remove pan containing minus No. 200 and weigh.  Replace pan beneath
    the sieves and sieve for another 10 min.  Remove pan and weigh.  When
    the differences between two successive pan sample weighings (where the
    tare of the pan has been subtracted) is less than 3.0 percent, the
    sieving is complete.

8.  Weigh each sieve and its contents and record the weight.  Check the
    zero before every weighing.

9.  Collect the laboratory sample and place the sample in a separate
    container if further analysis is expected.
                                 D-33

-------
Sample No:
Maferial:
Split Sample Balance:
Make	
Capacirv	
jmanes; Division
Material Weight (after drying)
Pan •*• Maferial: ^_^_^^_^
Pan:
Or/ Sample:
Final Weight:
                                         o/ s;)r »  Net Weight <20Q Mesh
                                                      Total Net Weigr.r
                                X 100
                SIEVING
Time: Start:
Initial (Tare):
20 min:
30 rin:
^0 mir:

Weight (Pan Only)





                                   SIZE DISTRIBUTION
Screen
3/8 in.
•i rr.esh
10 mesh
20 -r.esh
•iQ T.esh
100 mesh
UO mesh
200 rr.es h
Pan
Tare Weight
(Screen)









Final Weight
(Screen •*• Sample)









Net Weight (Sample)









°fC

1
!



|
1


                      Figure 5-3.  Example  silt  analysis form.
                                            D-34

-------
               TABLE 5-2.  MOISTURE ANALYSIS PROCEDURE
1.  Preheat the oven to approximately 110°C (230°F).  Recora oven
    temperature.

2.  Tare the laboratory sample containers which will be placed in the
    oven.  Tare the containers with the lids on if they have lids.  Record
    the tare weight(s).  Check zero before weighing.

3.  Record the make, capacity, smallest division, and accuracy of the
    scale.

4.  Weigh the laboratory sample in the container(s).  Record the combined
    weight(s).  Check zero before weighing.

5.  Place sample in oven and dry overnight.d

6.  Remove sample container from oven and (a) weigh immediately if
    uncovered, being careful of the hot container; or (b) place tight-
    fitting lid on the container and let cool before weighing.  Record the
    combined sample and container weight(s).  Check zero before weighing.

7.  Calculate the moisture as the initial weight of the sample and
    container minus the oven-dried weight of the sample and container
    divided by the initial weight of the sample alone.  Record the value.

8.  Calculate the sample weight to be used in the silt analysis as the
    oven-dried weight of the sample and container .minus the weight of the
    container.  Record the value.

aDry materials composed of hydrated minerals or organic materials like
 coal and certain soils for only 1-1/2 h.  Because of this short drying
 time, material dried for only 1-1/2 h must not be more than 2.5 cm
 (1 in.) deep in the container.
                                D-35

-------
Sample No:
Material:
Split Sample Balance:
  Make	
  Ccsacity	
  Smallest Division
Total Sample Weighf:
(Excl. Container)
Number of Splits:	
Split Sample Weight (before drying)
Pan •*• Sample:	
P=n:
Wet Scmpie:
Oven Temperature:	
Date In	Care Out
Time In
                                                                  Time Our
Drying Time
Marerial Weigh? (crrer drying)
Pan - Marericl: 	
Pan:
Dry Sample:
MOISTURE CONTENT:
  (A) Wet Sample Wr.__
  (8) Dry Sample Wr._
  (C) Difference Wt.
    C X  TOO
                                                                           % Moisture
                Figure  5-4.   Example moisture analysis form.
                                        D-36

-------
5.3.1  Loss-on-Dry1nq (LOP) Determination*

     ASTM Method D2216-71  ("Laboratory Determination  of Moisture  Content of
Soils") 1s used to provide an indirect measure of the moisture content of each
soil sample.   A data sheet similar to the one  shown in Figure 5-5 is used to
record the LOD data.

     For the  LOD  determination, analytically weigh  approximately  10 g of the
"raw" soil sample  from  each sample jar into  a  tared,  5-cm-diameter glass jar
with a tight  fitting lid.   If  there  is  standing water  1n the jar, it should
not be  sampled for  the  LOD  determination.    Remove  the jar  lids  and dry the
samples overnight  (12 to 16 h) in an  oven  at 105°C.   Remove the  LOD samples
from the oven  and  place them 1n a desiccator to cool;  remove the cooled sam-
ples from the desiccator and  replace  the jar  lids.   Reweigh  each dried LOD
sample and determine the percent LOD using the following formula:

     % LQP s Jar and sample wet wt. - jar and sample dry wt. x IQQ^
                              sample wet wt.

5.3.2  Oil and Grease Content Determination**

     Analyze any land treatment samples  for  oil  and  grease  content according
to Method 503 D in  "Standard  Methods for the Examination  of Water and Waste-
water."7    The method   involves  extraction  of  10 g  of  "raw"  sample  with
I,l,2-trich1oro-l,2,2-tr1fluoroethane  (Freon 113)   followed  by  gravimetric
determination of the dried extract.

5.3.3  Sample Drying Procedure***

     Dry the contents of each sample jar taken from a process using one of the
two procedures  described below, depending on  the average percent LOD measured
for that process.  If the LOD is less than 10%,  desiccate the five sample por-
tions over  anhydrous calcium sulfate  until  they show no more than 1% weight
loss upon further  drying.   If  the measured  LOD is greater  than  10%,  dry the
five  sample  portions in an oven  at  105°C until  they  show no more  than 1%
weight loss  upon  further drying.   A data sheet similar to  the one shown in
Figure 5-6  is  used  to   record  and  calculate the actual  sample weight  loss
during desiccation or drying.

     In  the  desiccation  procedure,   clean   the desiccator  by  washing  the
interior with  water which  conforms to  the  specifications  for ASTM,  Type 3
water (referred to herein as D.I.  water).*   Follow this with an acetone rinse
and  a final  methylene   chloride  rinse.    Spread  a 1-in  layer of anhydrous
calcium sulfate over the bottom of the desiccator.  Split the contents of each
sample jar (approximately 1 kg) between two tared,  9-in Pyrex pie plates that
have been previously cleaned with D.I. water,  acetone, and methylene
*    Material taken directly from Reference 2,
**   Material taken directly from Reference 2,
     Material taken directly from Reference 2,
                                     D-37

-------
                  LOSS-ON-DRYING (LOD)  DETERMINATION DATA SHEET

Site
Process
^Analyst .    	;	Date

Reviewed by 	 Date


Sample Jar Number           	  	  	

Jat^Tare Wt. (g)             	  	  	

Jar and Sample Wet Wt. (g)  	  	  	

S_ample Wet Wt. (g)          	  	  	
            Sample Wet Wt. = Jar and Sample Wet Wt. - Jar Tare  Wt.
 Oven Temperature(  C)
 Time In Oven

 Time Out Oven

 Total Drying Time


 Jar and Sample Dry Wt. (g)

 % LOD
    '* iTsTT'" J&r and Sample Wet Wt.  - Jar and Samole Dry Wt.     «nr,*
     * LOD   	Sample Wet Wt.  '	'	  x 100%
             Figure 5-5.  Loss-on-drying  (LOD) determination data sheet.
                                       D-38

-------
SAMPLE DRYING WEIGHT LOSS DATA SHEET                              Oven
                                                           Desiccation

Site	 Analyst 	 Date 	
Process                           Reviewer                  Date
Sample Jar Nuaber 	  Pan Tare Weights
Time            Sample Pan Weights                           Sample Weight

	   	 + 	 f 	 - Pan Tare Wts.  *  	

	   	 + 	 a 	 - Pan Tare Wts.  =  	

	   	 * 	 * 	 - Pan Tare Wts.  «  	

                    *          «            Pan Tare Wts.  =
Sample Jar Number 	  Pan Tare Weights
Time            Sample Pan Weights                           Sample Weight

	   	 + 	 = 	 - Pan Tare Wts.  »  	

	   	 + 	 - 	 - Pan Tare Wts.  =  	

	   	 + 	 = 	 - Pan Tare Wts.  =  	

                    *          =            Pan Tare Wts.  =
Sample Jar Number 	  Pan Tare Weights
Time            Sample Pan Weights                           Sample  Weight

	   	 + 	 = 	 - Pan Tare Wts.  =   	

	   	 * 	 = 	 - Pan Tare Wts.  =   	

	   	 + 	 * 	 - Pan Tare Wts.  =   	

	              *          =            Pan Tare Wts.  =
    Percent     Sample Wet Wt.  - Sample Dry Wt.      _„
    Weight =            Sample  Wet Wt.          *  10°"
    Loss

                        % WEIGHT LOSS AT TIME  INTERVAL

Sample Jar Number  	  	   	

Time 	      	  	   	
Time 	      	  	   ZZZZZZZ
Time 	      	  	   	
Time
               Figure  5-6.   Sample drying weight loss data  sheet.
                                      D-39

-------
chloride.   Determine each pie  plate tare weight  and  the weight of  each pie
plate  with Its  wet  sample  contents.   Place  sample  1n the  desiccator and
desiccate until  1t  shows  no more than  1%  weight loss on further desiccation.
Determine the  weight of each pie  plate with the  sample  portion  when dry and
calculate the percent Ioss-on-desiccat1on using the following formula:


          % weight loss = sample wet wt. - sample dry wt.. x 100%
                                 sample wet wt.


     For  oven-drying, first wipe  the  oven  Interior  clean  with D.I.  water,
followed by acetone and methylene chloride.  Split the contents of each sample
jar between two clean, tared pie plates.  Determine each pie plate tare weight
and the  weight of each pie plate with  Its wet  sample  contents.  Set the oven
temperature at 105°C.  Place sample 1n the oven and dry until 1t shows no more
than 1%  weight loss on further drying  (dry  enough to  be sieved); remove to a
clean  desiccator  to cool.   Determine the  dry weight  of the  sample and calcu-
late the percent loss-on-drying using the formula above.

     Return the desiccated or oven-dried samples to clean, dry  sample jars and
store  at  or below 20°C,* or keep  the samples  in the  desiccator and sieve the
same day.

5.4  SILT LOADING

5.4.1  Sample  Preparation and Analysis for Total Loading

     The gross  sample of  paved  road dust will  arrive at  the laboratory in two
types  of containers:  (1) the broom-swept dust  will  be  in sample containers;
and (2)  the vacuum-swept dust will be in vacuum bags.

     Both the  broom-swept dust and the vacuum-swept dust  are weighed on a beam
balance.   The broom-swept dust is  weighed  in  a tared container.   The vacuum-
swept  dust is weighed  in  the  vacuum bag which was tared and  equilibrated in
the  laboratory before going to  the field.   The vacuum  bag and  its contents
should be equilibrated again in the  laboratory before weighing.

     The total  surface dust loading  on the traveled lanes of the paved road is
then calculated  in  units of kilograms of dust on the traveled lanes per kilom-
eter  of road.   When only  one  strip of  length is taken across  the traveled
lanes, the  total dust loading on the traveled lanes is calculated as follows:
      If  not  extracted  within  14 d  after collection, the samples must be stored
      at  4°C.
                                     0-40

-------
where:    mb = mass of the broom- swept dust, kg

          mv - mass of the vacuum-swept dust, kg

          i  = length of strip as measured along the center line of the road,
               km


     When  several   Incremental   samples  are  collected  on  alternate  roadway
halves, the total surface dust loading 1s calculated as follows:

                              mbl * mvl * mb5 *  mv5 .
                           "Ib2 "*" mvZ  * mb6  "*• mv6
                           mb3 * mv3  * mb5  * mv7
                            mb4 * mv4 "*• mb8  •*" mv8
where:  m^. = mass of broom sweepings for increment i, kg

        mv. = mass of vacuum sweepings for increment i, kg

        i   = length of increment i is measured along the road center! ine, km


5.4.2  Sample Preparation and Analyses for Road Dust Silt Content

     After weighing the sample to  calculate  total  surface dust loading on the
traveled lanes, the broom-swept and vacuum-swept dust is composited.  The com-
posited sample  is  usually small  and requires  no  sample  splitting in prepara-
tion for sieving.  If splitting is necessary to prepare a laboratory sample of
800 to 1,600 g, the cone and quartering technique discussed in Section 5.1 can
be used.  The  laboratory  sample  is then sieved using the techniques described
in Section 5.2.

5.4.3  Sample Packing

     If  any  of the  samples  are  sent  to  another  laboratory  for  chemical
analysis,  pack  in  amber  vials  (40-ml  with  Teflon-lined  septa and  phenolic
caps).  Before  use, clean  and  rinse the vials sequentially with dilute nitric


                                     D-41

-------
add, D.I.  water,  acetone, and pesticide-grade  methylene chloride.   Package
Into the vials 30 g of a sample for organic  analysis  and  10 g of  a sample for
metals analysis.   Label  the vials  as  shown in Figure 5-7.
               Sample No.      AW-10-2        Date:   03/18/86

               Description:   30 g S11t Composite

               Process:       Landfill  Cell  10

               Site:          Acme Waste Disposal Co.



                          Figure 5-7.  Sample label.
                                     D-42

-------
                                  SECTION 6*

                              CHEMICAL ANALYSES
6.1  METALS ANALYSIS
     For analysis of the  metals of Interest  listed  1n Table 6-1, the methods
used  are  from  the  EPA  publication,   "Testing  Methods  for  Evaluating  Solid
Waste," SW-846.   Prepare  samples  for analysis  of all metals  except mercury
(Hg) by add  digestion using  EPA  Method 3050 (SW-846).   Prepare and analyze
the mercury sample by the cold vapor atomic absorption procedure following EPA
Method 7471 (SW-846).  Use the  following two  modifications in the final dilu-
tions of the  digested  samples.  Dilute  the samples  for ICAP determination by
EPA  Method 6010  (SW-846) and  furnace   atomic  absorption  determination  of
antimony (Sb) by EPA Method 7041 (SW-846)  to  achieve a final concentration of
5% hydrochloric acid.  Dilute the digested samples for arsenic (As) determina-
tion  by  EPA  Method 7060  (SW-846),  for  selenium  (Se)  determination by  EPA
Method 7740 (SW-846),  and for thallium  (Tl)  determination by EPA Method 7841
(SW-846)  to achieve a final concentration of 0.5% nitric acid.

     For chromium,  the analysis described  above,  using EPA Method 6010« and
ICAP, should  serve  as  a  screening  technique.  If the  results  for any sample
show a relatively significant  concentration of  chromium,  then another aliquot
of  that  sample should be analyzed for  hexavalent chromium  (Cr  VI)  which is
much more  toxic than  the other forms.   This  further analysis  is conducted
using  EPA   Method  3060    (alkaline   extraction)   in   combination  with  EPA
Method 7195 (coprecipitation/AA), 7196 (colorimetric), or 7197 (chelation/AA).

6.2  CYANIDE ANALYSIS

     Cyanide determinations are  performed by  colorimetric measurement follow-
ing EPA Method 335.2 found in  "Methods for  the  Evaluation of Water and Waste-
water. 7  The method involves distillation of the cyanide, as hydrocyanic acid,
into a sodium hydroxide absorbing  solution.   The cyanide ion in the absorbing
solution is determined colorimetrically.

6.3  SEMIVOLATILE ORGANIC ANALYSIS

     For the  semivolatile  organic  analysis, prepare  the samples by sonication
extraction  (Method  3550,  SW-846)  using  the  procedures  specified  in  the  CLP
Statement of Work for  Organic  Analysis.   Prepare the extracts at the low con-
centration level using 30 g of sample and subject them to adsorption
     Material taken directly from Reference 2.


                                     D-43

-------
                  TABLE 6-1.  METALS  AND MEASUREMENT METHODS
                   Element
Measurement Method**
                Aluminum (Al)
                Antimony (Sb)
                Arsenic* (As)
                Barium* (Ba)
                Beryllium (Be)
                Bismuth (Bi)
                Cadmium* (Gd)
                Chromium* (Cr)
                Cobalt (Co)
                Copper (Cu)
                Iron (Fe)
                Lead* (Pb)
                Manganese (Mn)
                Mercury* (Hgj
                Molybdenum (Mo)
                Nickel (Ni)
                Osmium (Os)
                Selenium* (Se)
                Silver*  (Ag)
                Thallium (Tl)
                Vanadium (V)
                Zinc  (Zn)
        ICAP
        GFAA
        GFAA
        ICAP
        ICAP
        ICAP
        ICAP
        ICAP*
        ICAP
        ICAP
        ICAP
        ICAP
        ICAP
  Cold Vapor AA
        ICAP
        ICAP
        ICAP
        GFAA
        ICAP
        GFAA
        ICAP
        ICAP
  *Eight RCRA metals

**ICAP « Inductively-Coupled Argon Plasmography
  GFAA » Graphite Furnace Atomic Absorption
    AA » Atomic Absorption
 Other methods are used to measure hexavalent chromium  (Cr IV), if appropriate
 (see page E-43  for discussion).
                                     D-44

-------
chromatography  on  Sephadex LH-20.    Concentrate the  extracts  and  determine
their weight.   Take  approximately 200 mg of each  concentrated extract, weigh
accurately, and redlssolve 1n 2 ml of a 1:1 mixture of methylene chloride and
methanol.   Calculate the  dilution  factor  for  the LH-20  procedure  using the
following formula:


     LH-20 dilution factor - weight of concentrated extract (mg)
                               exact weight of 200-mg portion


     Calibrate and monitor the LH-20  system according to the procedure 1n the
CLP for  the gel permeation chromatography  system.   For the LH-20 procedure,
use an eluent solvent system consisting of a 1:1 mixture of methylene chloride
and methanol.   Load  the 200-mg  redIssolved sample extract  directly onto the
column.  Adjust the eluent flow rate to 100 mL/h.  Collect the proper fraction
containing the aromatic compounds, and concentrate the fraction to 1 mL.

     Analyze the extract according  to the  CLP  procedure:   screen them by gas
chromatography  with  a  flame  1on1zat1on  detector  (GC/FID)  to  determine the
proper dilution level.  Minimize the amount of dilution to maintain the detec-
tion level  at  as  low a level as  possible.   Use a  capillary  column  gas chro-
matograph/mass  spectrometer  (GC/MS)   to  analyze  for  the organic  compounds
listed  in  Table 6-2 which were  derived  from  the Hazardous  Substances  List
(HSL) in the CLP.  Use  the internal  standard calibration method in the CLP to
quantify the HSL compounds found in the extracts.

     Samples particularly  high in oil  and  grease (such as those obtained from
land treatment  processes)  may require additional  cleanup or  other  treatment
for analysis.  Additional  cleanup can be  achieved  by repeating the LH-20 pro-
cedure.  Each time the  cleanup procedure  is repeated, the sample(s)  should be
rescreened  by GC/FID  to determine if  the additional  cleanup is continuing to
make progress.   Following  each  cleanup, use  the  quality control  guidelines
described  in the CLP for surrogate recovery to ensure that  excessive loss of
aromatic compounds does not  occur.    In some  instances,  even repeated cleanup
will not yield  samples  which  can be  analyzed  at the low concentration level.
In  these  instances  the analyst may  analyze them  at  the medium concentration
level and/or may consider alternative analytical techniques.   These techniques
may include the methods from EPA SW-846 which utilize other cleanup procedures
with  GC  techniques   and  which   are   only   specific   for  certain  groups  of
compounds.

6.4  PESTICIDES ANALYSIS

     For samples selected  for pesticides analysis, follow the CLP procedures
for  pesticides  and  PCBs.   For  this  analysis,  use a portion  of the sample's
semivolatile organic extract  and subject  the  extract  to  solvent  exchange.
Analyze the solvent-exchanged extract for  the  pesticides and PCBs (Aroclor's)
listed  in  Table 6-3  using  gas  chromatography/electron  capture  detection
(GC/ECD).
                                     D-45

-------
TABLE 6-2.  SEMIVOLATILE ORGANIC COMPOUNDS FOR ANALYSIS
                       ACENAPHTHENE
                       ACENAPHTHTLSNE
                       ANTHRACENE
                       BENZO (a) ANTHRACENE
                       BEHZOZC ACZD
                       BENZO U) PTRENE
                       BENZO (Obi) PERY1EHE
                       BENZO (b) FLUORANTHENE
                       BENZO (k) F100RANTHENE
                       BENZTL ALCOHOL
                       BIS (2-CHLOROETHO3C7) METHANE
                       BIS (2-CHLOROETHYL) titttK
                       BIS <2-CHLOROISOPROPYL> ETHER
                       BIS (2-ETHYHEXTL) PHTHALATE
                       4-BROMOPHENYL PHENYL ETHER
                       BUT7L BENZTL PHTHALATE
                       4-CHLORO-3-METHYLPHENOL
                       2-CHLORONAPHTHALENE
                       2-CHLQROPHENOL
                       4-C3LOROPHEN7L PHENTL ETHER
                       OIBENZO («.h) ANTHRACENE
                       OIBENZOFURAN
                       1.2 OICHLOR08ENZENE
                       1.3 OICHLOROBENZENE
                       1,4 OICHLOROBENZENE
                       3.3' -DZCKLOROBENZIDINE
                       2. 4-DICHLOROPKENOL
                       DIETHYLPHTHALATE
                       2. 4-OIMETHYLPHENOL
                       DIMETHYL PHTHALATE
                       DI-N-BUTYLPHTHALATE
                       2.4-OINITROPHENOL
                       2. 4-OINITROTOLUENE
                       2 . 6-OmZTROTOLUENE
                       DI-H-OCTTL PHTHALATE
                       FLUORANTHENE
                       FLUORENE
                       HEZACHLOROBENZENE
                       HEZACHLOR08UTAOI&NE
                       HEZACHLOROCTCLOPENTAOIENE
                       HEXACHLOROETHANE
                       ODENO(1.2.3-cd) PTEtENE
                       ISOPHORONE
                       2-METRYL-4 . 6-OINITROPHENOL
                       2-METH7LNAPHTHALENE
                       2-METHYLPHENOL
                       4-METHTLPHENOL
                       NAPHTHALENE
                       2-NXTROANILINE
                       3-MTrSO ANILINE
                       4-NITROANILINE
                       NITROBENZENE
                       2-HZTROFHENOL
                       4 -HITRO PHENOL
                       N-NZTROSO-OI-N-PROP7LAMINE
                       N-NITROSOOIPHENYLAMINE
                       PENTACHLOROPHENOL
                       PHENANTHRENE
                       PHENOL
                       PYRENE
                       1.2. 4-TRICHLOROBENZENE
                       2.4. S-TRICHLOROPRENOL
                       2.4. 6-TRICHLOROPHENOL
                                D-46

-------
TABLE 6-3.   PESTICIDES FOR ANALYSIS
           ALDRIN
           Alpha - BHC
           Beta - BHC
           Delta - BHC
           Gamma - BHC
           CHLORDANE
           4,4'-DDD
           4.4'-DDE
           4,4'-DDT
           DIELDRIN
           ENDOSULFAN I
           ENDOSULFAN II
           ENDOSULFAN SULFATE
           ENDRIN
           ENDRIN KETONE
           HEPTACHLOR
           HEPTACHLOR EPOXIDE
           TOXAPHENE
           AROCLOR 1016
           AROCLOR 1221
           AROCLOR 1232
           AROCLOR 1242
           AROCLOR 1248
           AROCLOR 1254
           AROCLOR 1260
               D-47

-------
                                  SECTION 7*

                         QUALITY ASSURANCE PROCEDURES


     The quality  assurance  (QA)/qual1ty control  (QC)  procedures described 1n
this section provide a performance audit, quality control, duplicate analyses,
and Independent analyses.   The quality control procedures  for providing cor-
rective action are the  internal  quality control  procedures  Instituted by each
individual laboratory.

     The internal  QC  procedures instituted by each  laboratory should involve
the use of known  QC samples,  spiked  samples,  duplicate samples, matrix-spiked
samples, duplicate matrix-spiked samples, surrogate-spiked samples, and method
blanks.

     For the  metals analysis,  use National  Bureau  of Standards  (NBS)  water
(1643 B) as  check  samples  for  the  accuracy  of  the instrumentation.   Use a
marine  sediment  reference  material  (MESS-1)**  and  an  NBS  fly  ash  sample
(1633 A)*** as  QC samples  to check the  overall  accuracy  of the digestion and
analysis procedures.  Spike  one  process sample with  the eight elements listed
in  Table 7-1,  and  calculate  their  percent  recoveries  to  assess  matrix
effects.  Prepare  and analyze another sample  in duplicate to demonstrate ana-
lytical precision.

     For the  QC on the analysis of  the semivolatile organics and pesticides,
follow the procedures in the Contract Laboratory Program  (CLP) protocol.  Use
an extra 60 g of a sample for a matrix spike  (MS) and a matrix spike duplicate
(MSD).   Determine the  percent  recoveries  and calculate  the  relative percent
difference  (RPD)  for  the duplicates.   Compare  the results  for the MS and MSD
with the  acceptable percent  recovery  range and the RPD  specified in the CLP
protocol.  Spike  all  samples prior to extraction with surrogate compounds and
determine the percent recoveries  of  these compounds.  Recovery of less than
10%  of any one  surrogate  or recovery  of  two or more  surrogates outside the
recovery  limits stated  in the  CLP require that  the sample extract be reana-
lyzed.  If the  recovery is outside the  limits upon reanalysis, then the sample
must be reextracted and reanalyzed.
 *    Material taken directly from Reference 2.
 **   Available  from  Marine  Analytical   Chemistry  Standards  Program  of  the
     Canadian National  Research Division of Chemistry, Montreal Road,  Ottawa,
     Canada, K1A OR9.
 ***  Available  from  the  National  Bureau  of  Standards,  Office  of Standard
     Reference  Materials,  Room B-311,   Chemistry  Building,  Washington,  DC
     20234.
                                      D-48

-------
         TABLE 7-1.  SPIKING COMPOUNDS:  METALS
                                Solvent: 0.5Z Nitric Acid
Compound
Concentration
  (ug/ml)
Arsenic (As)
Barium (Ba)
Beryllium  (Be)
Cadmium (Cd)
Calcium (Ca)
Copper (Cu)
Lead (Pb)
Magnesium  (Mg)
Manganese  (Mn)
Selenium (Se)
Silver (Ag)
Zinc (Zn)
     100
     100
     100
     100
     100
     100
     100
     100
     100
     100
     100
     100
                          D-49

-------
     Analyze two  blank  samples consisting  of  a purified  solid  matrix spiked
with  surrogate  compounds  and  carried through  extraction  and  concentration.
One blank 1s for  the  samples and  the other blank 1s for the MS and MSD.  Com-
pare the results with both the CLP-specified surrogate recovery limits for the
blanks and  with the CLP  limits on the levels of common  phthalate esters and
Hazardous Substances List (HSL) compounds.

     Conduct a  performance  audit  to assess the  precision and accuracy of the
laboratory  analyses.   This audit  should be conducted  only once  during the
entire testing program at a TSOF.   From a homogeneous composite sample, remove
three 30-g  aliquots for  an  organics analysis  audit  and three 4-g aliquots for
a metals  analysis audit.  Spike one of the 30-g aliquots with  the EPA refer-
ence  materials  whose contents  are listed  1n Tables  7-2 through  7-4,  and if
applicable,  Table 7-5.    The  amount of  spiking  material  should  be  at  least
10 times  the anticipated detection limit 1n a  30-g sample.  Spike one of the
4-g aliquots with the elements listed 1n Table 7-1.   Use a multielement atomic
absorption  standard.   The amount  of  material  to  be added  to a  4-g sample
should yield a  concentration 10 times  the anticipated detection limit for the
metals analysis.

     Have  the  laboratory analyze  the  spiked sample  and two of the unspiked
samples.  Replicate analysis of the spiked sample is recommended.  The dupli-
cate  unspiked  samples analyzed by  the laboratories give  a measure of preci-
sion.   Use the mean  of  the duplicate analysis to  correct  the  spiked sample
results for corresponding compounds present in  the unspiked  sample.   Compare
the corrected results of the spiked sample to the true value to determine the
accuracy  of the analysis.
                                     D-50

-------
          TABLE 7-2.  SPIKING COMPOUNDS:  ACID EXTRACTABLES II
Standard Code: C-090-01
   Solvent:
      Compound
Concentration
    (ug/ml)
      Benzoic acid
      p-Chloro-m-cresol
      2-Chlorophenol
      o-Cresol
      p-Cresol
      2,4-Dichlorophenol
      2,4-Dimethylphenol
      4,6-Dinitro-o-cresol
      2,4-Dinitrophenol
      2-Nitrophenol
      4-Nitrophenol
      Pentachlorophenol
      Phenol
      2,4,5-Trichlorophenol
      2,4,6-Trichlorophenol
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
     2000
                                  D-51

-------
       TABLE 7-3.  SPIKING COMPOUNDS:   NEUTRAL  EXTRACTABLES V
Standard Code: C-040
   Solvent:
Compound
                             Purity
Concentration
Acenaphthene
Anthracene
Benzo(k) fluoranthene
Dibenz ( a, h) anthracene
Dibenzofuran
1 ,2-Dichlorobenzene
1 , 4-Dichlorobenzene
bis ( 2-Ethylhexyl ) phthalate
Fluorene
Hexachlorobenzene
Hexachlo rocy clopen taol ene
Isophorone
Nitrobenzene
N-Ni trosodi -n-propylamine
Pyrene
98*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
2000
                                 D-52

-------
       TABLE 7-4.   SPIKING  COMPOUNDS:  NEUTRAL EXTRACTABLES VI
Standard Code: C-041
 Solvent: CH-C1-
  Compound
                               Purity
Concentration
  (ug/ml)
Benzo(a}pyrene
Benzo ( g , h , i ) pery lene
Benzyl alcohol
4-Bromophenyl phenyl ether
bis (2-Chloroethyl ) ether
2-Chloronaphthalene
4-Chlorophenyl phenyl ether
Chrysene
Diethyl phthalate
Dimethyl phthalate
Di-n-butyl phthalate
Di-n-octyl phthalate
Hexachlorobutadiene
Hexachloroe thane
Naphthalene
98*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
99*
2000
2000
2000
2000
2000
2000
2000
2000
20CO
2000
2000
2000
2000
2000
2000
                                D-53

-------
               TABLE 7-5.  SPIKING COMPOUNDS:   PESTICIDES II
  Standard Code:  C-093-01
Solvent: Toluene/Hexane  (1:1)
         Compound
                                                    Concentration
         Aldrin
         Alpha-BHC
         Beta-BHC
         Delta-BKC
         Gamma-BBC
         4,4'-ODD
         4,4'-DDE
         4,4'-DDT
         Dieldrin
         Endosulfan II
         Endosulfan II
         Endosulfan sulfate
         Endrin
         Endrin aldehyde
         Heptachlor
         Heptachlor epoxide
         Endrin ketone
         p,p'-Methoxychlor
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          2000
          1000
          2000
Concentration corrected for purity.
                                   D-54

-------
REFERENCES TO APPENDIX D

1.   Cowherd, C., G. E. Muleskl,  and  J.  S.  Klnsey.  "Control of Open Fugitive
     Dust Sources."   EPA-450/3-88-008, U.S.  Environmental  Protection Agency,
     Research Triangle Park, NC.  September 1988.

2.   DeWees, W. G., S. S.  Stelnsberger,  and S.  J.  Plalsance.  "Field Sampling
     and Analysis Protocol for Collecting and Characterizing Soil Samples From
     Hazardous   Waste   Treatment,   Storage,   and   Disposal   Facilities."
     EPA-450/3-86-014, U.S. Environmental Protection Agency, Research Triangle
     Park, NC.  October 1986.

3.   Cochran, W.  G.   "Sampling Techniques."   John Wiley  and  Sons, New York,
     New York.  413 pp.  1965.

4.   Mason,  B.  J.   "Preparation  of  Soil  Sampling Protocol:   Techniques  and
     Strategies."  EPA Publication No. 600/4-83-020.  August 1983.

5.   U.S. Environmental Protection  Agency  Contract Laboratory  Program, State-
     ment of Work for Organic Analysis.  July 1985 Revision.

6.   "Test  Methods  for  Evaluating Solid  Waste,  Physical/Chemical  Methods,"
     Second  Edition.   U.S.  Environmental  Protection  Agency, Office  of Solid
     Waste and Emergency Response, Publication No.  SW-846.   July 1982.

7.   "Methods  for Evaluation  of  Water  and  Waste Water."   EPA  Publication
     No. 600/4-79-020.

8.   "ASTM D 1193-74, Standard Specifications for Reagent Water."

9.   "Assessment of Hazardous Waste TSDF Particulate  Emissions."   Draft Final
     Report by MRI prepared for the U.S.  Environmental Protection Agency under
     Contract No. 68-02-3891, Work Assignment No. 5.  May 5, 1986.
                                     D-55

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA 450/3-89-019
                              2.
                                                            3. RECIPIENT'S ACCESSION NO.
'4 TITLE AND SUBTITLE
 Hazardous Waste  TSDF - Fugitive Particulate  Matter Air
 Emissions Guidance  Document
                                         5. REPORT DATE
                                            May  1989
                                         6. PERFORMING ORGANIZATION CODE
            Midwebt  Research Institute—
            425  Volker Boulevard
            Kansas City, Missouri  64110
                                         8. PERFORMING ORGANIZATION REPORT NO.
 ). PERFORMING ORGANIZATION NAME.AND ADDRESS
 U.S. Environmental  Protection Agency
 Office of Air  and  Radiation
 Office of Air  Quality Planning and  Standards
 Research Triangle  Park, NC  27711
                                         10. PROGRAM ELEMENT NO.



                                         11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13. TYPE OF REPORT AND PERIOD COVERED
                                                            14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
         The purpose of  this  document is to provide  regulatory and industrial  personnel

    with sufficient information to identify sources  of  contaminated fugitive PM

    emissions, estimate the magnitude of emissions, select  viable control measures,  and

    estimate the effectiveness  of  those measures in  order  to ensure that high  risks

    from these facilities  do  not occur.  The following  sources are discussed in this

    document:   paved and unpaved roads, open waste piles and staging areas, dry

    surface impoundments,  landfills,  land treatment, and waste stabilization.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b. IDENTIFIERS/OPEN ENDED TERMS  L.  COSATI I icid/Group
 PM10
 Fugitive  dust
 TSDF
 Hazardous  waste
 Roads
 Storage piles
Dry surface  impoundments
Landfills
Land treatment
Waste stabilization
Emission estimates
Control techniques
Risk analysis
               '--EVE'
                                              '•9 SEC..P
                                                            S  This .
 Release  Unlimited
                              ncla^sjlied
                                                 Unclassified

-------