&EPA
         United States
         Environmental Protection
         Agency
          Industrial Environmental Research
          Laboratory
          Research Triangle Park NC 27711
EPA-600/7-79-116
May 1979
Development of
Measurement Techniques
for Fugitive Emissions
from Process and
Effluent Streams

Interagency
Energy/Environment
R&D Program Report

-------
                  RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional  grouping  was consciously
planned to  foster technology transfer and a maximum interface in related fields.
The nine series are:

    1.  Environmental Health Effects Research

    2.  Environmental Protection Technology

    3.  Ecological Research

    4.  Environmental Monitoring

    5.  Socioeconomic Environmental Studies

    6.  Scientific and Technical Assessment Reports  (STAR)

    7.  Interagency Energy-Environment Research and Development

    8.  "Special" Reports

    9.  Miscellaneous Reports

This report  has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort  funded  under the 17-agency  Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems.  The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments  of, and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
                       EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not signify that the contents necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                         EPA-600/7-79-116

                                                  May 1979
Development of Measurement Techniques
    for Fugitive Emissions from  Process
              and  Effluent  Streams
                             by

                        Henry J. Kolnsberg

                TRC—The Research Corporation of New England
                      125 Silas Deane Highway
                    Wethersfield, Connecticut 06109
                      Contract No. 68-02-2113
                     Program Element No. EHE624A
                   EPA Project Officer: D. Bruce Harris

                 Industrial Environmental Research Laboratory
                  Office of Energy, Minerals, and Industry
                    Research Triangle Park, NC 27711
                          Prepared for

                 U.S. ENVIRONMENTAL PROTECTION AGENCY
                   Office of Research and Development
                       Washington, DC 20460

-------
                                FOREWORD
     This  report  summarizes  the  work  completed  by TRC  - THE  RESEARCH
CORPORATION  of  New  England   under   EPA  Contract  Number  68-02-2133,
"Development of Measurement  Techniques  for Fugitive Emissions from Process
and Effluent Streams."

     The results of  three major research and  development  tasks  assigned
under  the  task-level-of-effort  contract  are   presented  in  detailed
summaries.  Two additional major research  and development tasks, for which
separate reports have been published, are  summarized by abstracts of their
reports.    Twenty  two  assigned  service  area  and  documentation  and
publication tasks are briefly summarized.

     The effort was  sponsored  by  the Process  Measurements  Branch  of the
Industrial Environmental  Research  Laboratory at  Research  Triangle Park,
North Carolina.  Dr.  Robert M.  Statnick and D. Bruce Harris directed the
effort as Project Officers.  Henry J. Kolnsberg  was the Project Manager for
TRC.  Gordon T. Brookman, Dr.   Robert  E.  Kenson,  and Roland L. Severance
served as Task Managers.
                                     ii

-------
                                   CONTENTS


Foreword	ii

     1.   Introduction 	    1

     2.   Research and Development Task Descriptions 	    5

               Evaluation of Non-Point Source Waterborne Emissions .  .    5

               Study of Remote Sensing Methods Applicability to
                 Fugitive Emissions Measurement  	    5

               Evaluation of Surface Runoff at Integrated Iron and
                 Steel Plants	17

               Design of a Fugitive Assessment Sampling Train  ....   17

               Preliminary Investigation of Techniques for the
                 Determination of Hood Capture Efficiency  	   22

     3.   Service Task Descriptions	33

     4.   Documentation and Publication Task Descriptions  	   39
                                     iii

-------
                                SECTION 1

                               INTRODUCTION

     Fugitive emissions  are  defined as those  pollutants  introduced into
the  atmosphere  or into  ground or  surface  waters without  first  passing
through some  stack,  duct,  pipe or channel designed  to  direct or  control
their flow.  They are generated from a wide variety of industrial  process
sources, ranging  from small,  single-point  sources such  as process equip-
ment leaks to large, area sources  such  as material  storage piles.  Their
physical and chemical characteristics are varied, including airborne par-
ticulate matter,  gases  and vapors;  and waterborne suspended  solids  and
dissolved  gases,  liquids,  and solids.  Their  one common  characteristic,
their lack of containment, makes both the assessment of the environmental
impact and the control of the fugitive emissions considerably more diffi-
cult than the assessment and control of point-source emissions.

     The initial  step in establishing  controls for fugitive emissions is
the determination of  the  emission strength of the various sources and their
impact on  the environment.   Determinations  of  specific  fugitive emission
concentrations,  distributions  and  physical  or  chemical  characteristics
usually cannot be made with the recognized standard sampling and measure-
ment techniques presently in use.  Special measurement techniques  must be
developed and verified to obtain data that will be universally accepted as
baseline information for the establishment of  control systems and limita-
tions for industrial fugitive emissions.

     Three general measurement  techniques,  related  to the size or nature
of the fugitive  emissions source,  have  been developed for airborne emis-
sions.  The quasi-stack  technique utilizes  a  temporary hood  to capture the
emissions from a small source and transport  them to a duct for measurement
using standard  stack-sampling methods.   The roof monitor technique uses
the structure enclosing  sources  as a capture hood and measures  the  flow and
concentration of  the  emissions from all sources  within  the structure as
they pass through a roof monitor  or  similar  vent to  the atmosphere.  The
upwind-downwind technique measures the emission concentrations in the am-
bient atmosphere, utilizing recognized diffusion equations in mathematical
models to  back-calculate  source strength  from  the differences in  concen-
trations.  These  techniques are very general and  require modification and
adaptation for almost every specific site/source  combination.  Additional
techniques  for   waterborne  emissions,  both general  methods  and  site-
specific modifications,  are also needed.

     This report summarizes the work completed  by TRC - THE RESEARCH CORPO-
RATION of New England under EPA Contract 68-02-2133, "Development  of Mea-
surement  Techniques   for Fugitive  Emissions  from  Process  and  Effluent
Streams" toward the fulfillment of these needs.   The contractual effort was
conducted from December  1975  to December 1978 for  the Process Measurements
Branch  of  the  EPA's  Industrial  Environmental  Research  Laboratory  at
                                   -1-

-------
Research Triangle Park, North  Carolina.   Dr.  Robert M.  Statnick was the
EPA Project Officer  for the first half of the contract  and D.  Bruce Harris
was the EPA Project Officer  for  the  concluding half.   Henry J. Kolnsberg
was the Project Manager for TRC.

     The objective  of  the contract was  to  provide  measurement methodol-
ogies for airborne and waterborne fugitive emissions that would be appli-
cable  to  environmental  assessment  and  control  technology  development
projects related  to the  stationary source energy and  industrial  process
programs of  the  Agency.   The  contract provided a  continuing  program of
evaluation, development,  testing  and  field adaptation of existing and pro-
posed measurement techniques  and  programs  in a 20,000 man-hour task-level-
of-effort format.

     The contract effort  consisted of 27 individual  tasks,  divided into
three general work  areas.   Five Research and  Development  tasks required
about 13,700 man hours of effort,  12  Service  tasks  about 1,400 man-hours
and 10 Documentation and  Publication tasks   about  1,900 man-hours.   An
additional  1,400 man-hours were  expended in the technical  management of
the contract.

     The Research and  Development tasks included:
     Task 02

     Task 03


     Task 04


     Task 05

     Task 07
Evaluation of Non-Point Source Waterborne Emissions

Study of Remote Sensing Methods Applicability to Fugi-
tive Emissions Measurement

Evaluation of  Surface Runoff  at  Integrated Iron and
Steel Plants

Design of a Fugitive Assessment Sampling Train

Preliminary Investigation of Techniques for  the Deter-
mination of Hood Capture Efficiency
     The Service tasks  were:

     Task 20A  -
     Task 20D  -
     Task 20E  -
     Task 20H  -
Evaluation of a proposed no-duct quasi stack measure-
ment technique

Preparation of alternative measurement approaches for
emissions from a coke quench tower

Preparation of a  sampling program for fugitive emis-
sions from a coal cleaning facility

Assistance in  the  preparation  of a guideline on par-
ticulate fugitive emissions
                                   -2-

-------
Task 201  -
Task 20J  -
Task 20L  -
Task 20M  -
Task 20N  -
Task 20P  -
Task 20R  -
Task 20U  -
Critical review of a feasibility study  of source iden-
tification from ambient air monitoring

Preparation of  a  measurement  program for particulate
matter emissions from iron ore loading operations

Review of a test  plan  for ambient monitoring of flu-
idized bed coal combustor emissions

Review of a sampling plan for fugitive emissions from
petroleum refineries

Critique of  a  proposed  plan  for  the  measurement  of
mining particulate matter emissions

Preparation of  a  conceptual  roof monitor measurement
approach for an open hearth shop

Critique of exposure profiling technique for particu-
late matter fugitive emissions

Consultation and  assistance  to an  EPA contractor  in
sampling refinery hydrocarbon fugitive emissions
The Documentation and Publication tasks included:
Task 20B  -


Task 20C  -


Task 20F  -

Task 20G  -


Task 20K  -


Task 20Q  -


Task 20S  -
Preparation of a guideline for fugitive emissions mea-
surements

Preparation of  a summary of  airborne  fugitive emis-
sions measurement techniques

Preparation of papers for symposium presentations

Preparation of  a summary of  fugitive  emissions work
completed under IERL contracts

Preparation of a bibliography of research program ab-
stracts

Study to determine the  need  for  a technical manual for
the measurement of industrial process water flows

Preparation of  a chapter on  fugitive  emissions mea-
surements for a Level 1 environmental assessment pro-
cedures manual
Task 20T  -    Preparation and  presentation of a  paper  on fugitive
               emissions measurement for an EPA Symposium

Task 20V  -    Preparation and presentation of a paper on the devel-
               opment of the fugitive assessment  sampling train for
               an EPA Symposium
                              -3-

-------
     Task 06   -    Organization of the Second Symposium on Fugitive Emis-
                    sions:  Measurement and Control.

     The work completed in each of these tasks is described in this report
along with  the  conclusions and  recommendations  resulting from  the  com-
pleted efforts.  Only abstracts of the final reports for Tasks 02 and 04,
which were published as separate documents, are included.
                                   -4-

-------
                                SECTION 2

               RESEARCH AND DEVELOPMENT TASK DESCRIPTIONS

TASK 02 - EVALUATION OF NON-POINT SOURCE WATERBORNE EMISSIONS

     A full description of  this  task is contained in Sampling and Modeling
of Non-Point Sources at a Coal-Fired Utility,  EPA-600/2-77-199, September
1977.  The report abstract is  quoted as follows:

          The report gives  results  of a measurement and modeling pro-
     gram  for  non-point  sources (NFS)  from two coal-fired  utility
     plants, and  the  impact of  NFS on receiving waters.   The field
     measurement  survey,  performed   at  two  utility   plants  in
     Pennsylvania, included measurement  of  overland  runoff from NPS
     and river sampling upstream and downstream of each  plant site.
     NPS sampled were stormwater runoff and leachate from coal stor-
     age piles and runoff from impervious areas such as parking lots
     and roofs which were covered with  dust  fallout from coal and ash
     handling  operations.   A mathematical model  was developed  to
     simulate both the quantity and quality of industrial NPS pollu-
     tion  and  its impact on receiving  waters.   Field data indicated
     that  NPS  pollution from  utilities had  little  impact on the two
     rivers, compared  to  the  impact from sources  upstream  of each
     site.  Modeled results compared to  field measurements within a
     factor of 4 for  both the  quantity  and  quality  of stormwater
     runoff and  its  impact on the quality  of  the  receiving waters.
     Field  survey results  indicate that, for a cost-effective pro-
     gram,  sampling must be supplemented with modeling (the modeling
     results indicate  that the developed model can be  used with a
     minimum of  field data to  successfully simulate industrial NPS
     pollution and  its impact  on  receiving waters  for  the utility
     industry).

TASK  03 - STUDY OF  REMOTE  SENSING  METHODS  APPLICABILITY  TO FUGITIVE
EMISSIONS  MEASUREMENT - JUNE  1976

     Large  scale  programs  to  assess  the  environmental impact of fugitive
air emissions from industrial  sources,  using basically  research  type fugi-
tive  emissions  measurement   strategies,  have  proven to  be  costly  and
limited  in accuracy.   In  many cases,  the cost  of  measuring the fugitive
emissions  from only  one major  source in a complex, multiple-emissions-
source process can exceed  the cost of all sampling required  to  define the
point source emissions from the process.

     The complexity of many proposed fugitive  emission measurement methods
means that the time consumed  in equipment preparation, setup and calibra-
tion is  substantial and precludes  their application  for  survey-type fugi-
tive emissions testing.   Some  quick,  accurate,  and potentially low-cost
                                   -5-

-------
method is needed.   Remote  sensing of the fugitive emissions would be ideal,
since most remote  systems are self-contained and  do  not necessarily re-
quire access to the emissions source.

     Four remote  sensing  techniques have  been  identified  as potentially
useful for fugitive emissions measurement.  They are:

     1.   Tunable Laser Method
     2.   Lidar Method
     3.   Long Path Transmissometer Method
     4.   Time-Lapse Photography Method

     The objective  of  this task was to  review  the capabilities of these
four sensing methods.  Each method was reviewed separately with regard to
its principle of operation, limitations,  and  application  to fugitive emis-
sion measurement.

Remote Sensing of Fugitive Gaseous  Emissions  Using  Laser  Infrared Spectro-
scopy

     The use  of laser  systems for  gaseous  air pollution  detection and
monitoring has advanced greatly over the past few years due to improvements
in both laser technology and signal processing techniques.  Because almost
every gaseous air pollutant has characteristic absorption bands at middle
infrared frequencies (2.5-25um), tunable infrared  laser  systems have been
successfully  used   for  pollutant  monitoring  in several  recent studies.
Such devices have  found application in point sampling,  in in situ source
monitoring, and in  ambient air measurement of CO and  NO.

     Among those lasers which are  capable of emitting infrared radiation
are parametric oscillators, spin-flip Raman  lasers, gas  lasers, and semi-
conductor diode  lasers.   The use  of a combination of these lasers makes
possible complete coverage of the infrared  absorption  band regions between
0.63 and  34|jm.  Compared to conventional  infrared  instrumentation,  the
laser system usually has better sensitivity and  specificity because of the
high intensity, narrow spectral bandwidth,  coherence,  small beam diameter,
and the small directional divergence offered by the laser.

     Three fundamental monitoring techniques are available with laser sys-
tems:   remote  heterodyne  detection,  resonance  fluorescence,  and  direct
absorption.  Remote heterodyne detection is a completely passive, single-
ended technique  in which characteristic  infrared  emission lines  from a
gaseous pollutant  are detected by heterodyning them with  tunable laser
radiation of the same wavelength.   The  technique of resonance fluorescence
is an active one making use  of absorption and  subsequent re-radiation of
laser light.   This  method, however, appears most  useful for only  a few
atoms and  molecules.   Direct absorption  of laser radiation  offers  the
greatest versatility of the  three  schemes.   This  technique was therefore
considered for  the quantitative remote  measurement  of  gaseous fugitive
emissions from industrial  sources.
                                   -6-

-------
oc'
 O
Principle of Operation—
     The measurement  of air  pollutant concentration  by attenuation  of
laser infrared lines depends upon finding coincidences between laser lines
and pollutant  absorption  lines.   When  infrared radiation of  a  resonant
frequency  of  a  molecule  impinges  on  the  molecule,  radiant  energy  is
absorbed by that species.  The  intensity of the absorption is proportional
to the concentration of the absorbing species and  to  the path  length,  as
described by the Lambert-Beer Law:

     P.   =  P   exp (-«'  CL)
      t      o    r    o


where

     P    is transmitted power

     P    is initial power

      c'   is the absorption coefficient of gas

     C    is gas concentration

     L    is path length

     Thus, measurement  of the  transmitted  power will provide the concen-
tration of the absorbing species if the other parameters are known.

     Instrumentation used for  infrared  absorption  techniques must include
a source of radiation and  a detector together with  associated optics.  One
possible arrangement consists of the source and  detector  virtually side by
side and requires  the  use of a reflector to return  the  radiation to the
detector.  A  second  possibility  puts  the detector and source at opposite
ends of the measurement path.  Both are considered long-path techniques.

Limitations—
     Although long path laser infrared systems  look promising for measure-
ment of fugitive emissions,  there are  several critical  problems which need
to be considered:

     Interferences—One very  serious  problem  in  infrared  spectroscopic
studies  is  that of  interferences  from atmospheric C02  and water vapor.
This  is  an ever-present difficulty,  for each  contains strong absorption
bands in the infrared region of the spectrum.  Water  completely blanks out
the far infrared and absorbs  strongly  in the middle region between 5.5 and
7.5ym.   Fortunately, most of  the major pollutants  contain at least one
strong absorption line  or band from 3-15um  so  that it may be possible to
work around the  C02 and water  vapor lines and bands using high resolution
instrumentation.
                               -7-

-------
     Temperature Effects—Because semiconductor diode lasers can be tuned
by varying their temperature,  they are sensitive to changes in temperature
and must be cooled to liquid nitrogen temperatures when in use.  This can
be accomplished by use of  a closed-cycle cooler and liquid-nitrogen-cooled
detectors.  The detector  must  be refilled  with liquid nitrogen every ten
hours.

     Distance  Effects—Long  optical  paths  are generally  needed to bring
bands  of  low-concentration pollutants  to  detectable  levels.    The path
length  required  depends  on pollutant concentration,  the  strength of the
absorption bands,  and system  sensitivity.   Successful work has been done
over paths less than 1 km.

Quantitative/Qualitative  Considerations—
     Long path remote sensing techniques are  capable  of measuring only the
average concentration  of  a pollutant  over the optical path  between the
transmitter and  the  receiver.  Pollutant  concentrations  that  can be de-
tected  over a  1 km path,  assuming  that  a 5% signal change is detectable,
range  as low as 0.02 ppm  for CzHn with a COz laser.

     Although  qualitative applications of  this  technique have not been
reported in the literature, it seems that  such applications could be suc-
cessful.   Although the entire spectrum cannot be scanned  with a  single
radiation  source  (because no  laser  is  yet  tunable over the whole range),
several lasers set on  the most probable pollutant absorption wavelengths
lould  be used  over the path to be measured.  A slgnal change, which would
indicate  the  presence  of  an  absorbing  pollutant, would  only  need to be
calibrated to  make the measurement quantitative as well.

Application to Fugitive Emissions—
     No reports of the  use of  infrared laser techniques to  measure  fugitive
emissions have been made to date.  Using the absorption technique described
above,  CO, NO, S02 and ethylene have been  successfully measured from open
sources,  in  ambient  air  and  in in  situ  situations.   Considering these
results, it appears that the application of such techniques in the measure-
ment of fugitive  emissions would be  successful provided that:

      (a)  The  emitted plume is large enough to  permit the use of long paths
          if the  pollutant concentration is  low,

      (b)  It is possible to install a retroreflector or detector on the far
          side of  the plume,

      (c)  The  pollutant   to  be measured has an  infrared  absorption band
          outside  a COz or water absorption  band,

      (d)  Any  aerosol backscatter is accounted for.

     In addition  to  these limitations, several  considerations should be
kept in mind:
                                   -8-

-------
     (a)  The fugitive  source that is being measured must be well defined.

     (b)  Only the average concentration over the path can be determined.

     (c)  The uncontrolled  environment  may cause  optical  scintillations
          which could reduce sensitivity compared to controlled laboratory
          situations.

     (d)  Vibrations and misalignment could cause problems.

Remote Sensing of Fugitive Particulate Emissions Using Lidar

     Lidar has  been used  successfully for air pollution  detection since
its development in 1963 for meteorological use.  Its utility stems from its
ability to remotely detect  particulate  matter in  the  atmosphere  at dis-
tances up to  30 km.  The  lidar  technique also has the advantage of being
independent of external sources  of electromagnetic radiation because it is
an active system.   Lidar systems have been used in the delineation of cloud
layers, determination  of  upper atmosphere  density and temperature pro-
files, determination of smoke plume opacity and in the remote measurement
of transport, diffusion and relative particulate concentration profiles of
clouds of particulates.

Principle of Operation—
     The principle of  lidar operation is  very similar to  that of conven-
tional  radar.    In both  systems  a  pulse of  energy  is  transmitted  and
scattered by  objects in the atmosphere.   A small fraction of this pulsed
energy is scattered in the backward direction, detected and measured as a
function of the range at which the scattering occurred.

     The difference between the two systems is the type of energy source.
Lidar systems utilize  laser radiation of wavelengths somewhere between the
far  infrared and  the  near ultra-violet  portion of  the  electromagnetic
spectrum, operating at much shorter wavelengths  than conventional radar
systems.  This characteristic  gives  lidar the  capability of detecting very
small  objects  such as atmospheric particles.   Conventional radars would
totally overlook objects of this size.

     The  laser  energy associated with  lidar  systems  is  highly monochro-
matic,  essentially coherent  and concentrated  in  very short high-powered
pulses.  This energy is directed by refracting or reflecting lens systems
in a beam signal.   A suitable receiver lens system then collects  the  energy
back-scattered  by the molecules  and particulates within the  beam.   An
energy sensitive transducer (normally a  photomultiplier tube) detects this
back-scattered  energy  which is  then displayed  on an oscilloscope as a
function of  range.

     Return  signals may be photographed or magnetically recorded.   The use
of  magnetic  video disc  memory provides  an input to more  sophisticated
                                   -9-

-------
analysis procedures such as automatic data input for computer processing.
This is a key feature of sophisticated lidar systems.

     The general lidar equation of the received signal is:
          P  Cr/3180
     Pr =             exp
where

     P         =  power received from range r

     P         =  transmitted power

     C         =  velocity of light

     T         =  pulse duration

     3180      =  volume backscattering coefficient of the atmosphere at
                  range r

     A         =  effective receiver aperture

     a         =  extinction (attenuation) coefficient at range r

     All the values of this equation are known values of the system except
the volume  backscattering coefficient (3180) and  the  extinction coeffi-
cient (0).

     The  volume  backscattering coefficient  (3180) of  the  atmosphere at
range r is defined as the ratio  of  the amount of energy backscattered from
a defined unit volume to the  amount initially received.  It is  a measure of
the effectiveness of  the  scatterers  as a  target.   The extinction coeffi-
cient (a) is defined as the attenuation of a pulse traveling to a point r
and back  to the  receiver.  It accounts   for all  the  interactions  that
extract energy from the laser beam.

     The magnitude of 8180 and 0 depend upon the  wavelength of the incident
energy and the  number, size, shape and refractive properties of  the illumi-
nated particles  per unit  volume.    These  particles can  consist  of  true
particulate components or an aerosol.

Limitations on Method—
     Although lidar systems look promising  for the measurement  of fugitive
particulate emissions, there  are several critical parameters which need to
be considered:
                                  -10-

-------
     Particle Size—Lidar  systems can  detect scattering  from particles
larger than 1/10 of the wavelength of the light source.  For example, the
ruby laser, commonly used in laser radar  systems, operates at a wavelength
of 0.694um and can theoretically detect scattering particles greater than
0.07ym in  diameter.    This  theory however  applies only for  "clear" and
stable atmospheric conditions.  Small particles such as those in aerosols
can therefore in reality be uncounted because of background  effects not due
to the aerosol itself.

     Distance Effects—There are short and long range distance limitations
with the use of lidar systems.  Short range limitations are a function of
the unit's transmitter/receiver optical system geometry.  The intersection
of the transmitter beam  and the receiver field of view can create a  false
peak in the receiver system, thereby masking valid scattering return sig-
nals.  The  short  range  limitation for the lidar system  used  in a recent
study was approximately  300 meters.  Closer than this, false signals pre-
dominated over scattering return  signals.

     Long  range  distance limitations are a  function  of  the power of the
laser source and the composition of  the scattering medium.  The more  power
a  laser source  has,  the farther the signal will travel before it is com-
pletely attenuated.   Conversely,  the larger the scattering particles and
the denser the scattering medium the  shorter distance the laser signal will
travel before it is completely attenuated.  Typical  lidar systems used for
meteorological  observation  have long range distance limitations of  about
40 kilometers.

     Time Limitations—The  time  required to make one  complete  observation
of a dispersing fugitive cloud depends upon the number of  cross sections of
the emission cloud that are made and the number of vertical lidar shots per
cross section.  These two factors  depend on the size of the  fugitive cloud.
The  Stanford Research  Institute  lidar study  group  has  stated that each
vertical  scan of  20-30  lidar  shots  requires 2-5 minutes.  A  set of  three
vertical  scans  or cross  sections, therefore,  requires  10-15 minutes.

     Each plume  cross section observation must be  calibrated  for  the fol-
lowing  three factors in order to obtain  contours of  relative  particulate
concentrations:

      1.    Range effect
      2.    Log-amplifier transfer function
      3.    Pulse-to-pulse variation  in transmitted  power

      Location Effects—Another limitation to  the use of lidar  systems  is
the  testing location.   The direction  of the transmitted beam  should  be
perpendicular to the  centerline of the plume or fugitive cloud in order  to
obtain  maximum return  power.   This is  an important  factor because  lidar
systems  detect backscattered signals which may be  relatively  small.  Any
                                    -11-

-------
plan to  measure fugitive  emissions  with a  lidar system  should include
several test sites  that are located perpendicular to the most frequent wind
directions.

Qualitative/Quantitative Considerations—
     Lidar systems have primarily been used for qualitative applications.
There are two factors which limit the use of lidar systems for quantitative
applications; reducing  the lidar  equation to one  unknown  and accounting
for multiple scattering effects.

     The lidar equation defines the received  power of the lidar systems as
a function of range in terms of two variables; the backscattering coeffi-
cient and the attenuation coefficient.   The lidar equation must be reduced
to one variable if the return signal is to generate quantitative informa-
tion.  This may be done by independently determining the value of one of
the variables or by  assuming  that  a specific relationship exists between
the two variables.

     Independent determination  of  one  of the variables requires specific
measurements or qualifying assumptions.   The measurements  required may
include  particle shape,  refractive index,  density and particle size dis-
tribution.  Independent  measurements,  qualifying assumptions and assump-
tions  relating backscatter  coefficient  to  attenuation  coefficient all
place limitations on the accuracy of any quantitative results attained.

     Multiple scattering effects must  also be considered when attempting
to  obtain  quantitative  information  from the  lidar  equation.   Multiple
scattering  can  produce responses  in  the lidar  receiver  system  that are
greater  than expected  and  indicate misleading high quantitative amounts.
Multiple scattering effects occur to the greatest extent during turbulent
meteorological  conditions.  Lidar testing is normally done through stable
atmospheres where  multiple scattering effects are considered negligible.
The degree to which this  assumption is valid limits the accuracy of quanti-
tative results.

Application to  Fugitive Emissions—
     Lidar as a remote sensor of fugitive emissions would find best appli-
cation where particulate emissions or aerosols would normally be measured
by upwind-downwind networks.  Lidar  shows  promise of being more accurate
than the usual  upwind-downwind techniques although no application to fugi-
tive emission measurements has been documented.

     Proven Fugitive  Emission Applications—The Stanford Research  Insti-
tute conducted a program in May  and October of 1968 which utilized a lidar
system to remotely track the dispersion of the effluent  from tall stacks at
a generating station.  Although emissions from a stack cannot be considered
a fugitive source,  the  technique used to measure them  can be almost direct-
ly applied to the measurement of fugitive emissions.
                                  -12-

-------
     The program consisted  of several studies.  One study remotely tracked
the stack  plumes for distances  up to  20 kilometers.   The most  common
distance reported was 3  kilometers.  Each  observation was made at a single
point located  as close  to perpendicular to  the  plume  as possible  and
consisted of 20-30 vertical lidar shots  for  each of as many as 3 azimuth
angles.

     The study also attempted to obtain  quantitative  data  by  making cor-
rections  for  attenuation,  computing  return  signals  on   the  basis  of
scattering theory and by making independent particle size measurements.  A
short series of measurements showed  a close agreement with calculated val-
ues from power plant data and wind speed.

     In obtaining this type of quantitative data,  several limiting factors
were apparent.  First, the  size distribution  and  the optical properties of
the particulates were assumed to be spatially uniform.  Second, multiple
scattering  was considered  to  be negligible.   Third,  considerable effort
was necessary  to correct for attenuation.

Remote  Sensing of Fugitive Particulate Emissions Using Transmissometera

     In the measurement of stack  emissions,  transmissometers  have been
utilized  as continuous  monitors  of plume opacity near the stack outlet.
Response  curves  of  particulate emission rate versus opacity have made it
possible  to continuously monitor particulate  emissions  as & function of
the measured opacity.   If  a roof monitor is  considered as a large cross-
sectional  area stack, a  transmissometer ought  to  be applicable  to quanti-
tatively monitoring  the  roof monitor fugitive  particulate  emissions.

Principle  of Operation—
     Transmissometers are  devices that measure the attenuation  of visible
light across a fixed path  in  relation to some characteristic of the path
that  tends to change light transmittance.   The most common example is  a
transmissometer  used  to measure the opacity  of suspended particulate mat-
ter emissions  from  a stack.

     While the principal  application  of such devices  is  to measure  and
record  stack plume  opacity, they can be  used as  indicators of  particulate
mass  concentrations  in  the stack.   The  application of interest  here  is to
measure particulate mass  concentration  across the relatively  long  paths
normally required to encompass fugitive  emissions.

      The transmittance  (T) of a particulate cloud  is related to opacity  (0)
as:

                               T   -  1-0

 The optical density (D) of a  cloud is  related to transmittance as:

                              D   =  log 1/T
                                   -13-

-------
     In general, for any given type  and thickness of participate cloud the
concentration of particulates  is  directly related to optical  density or
the log of 1/T.  The  relative optical density of a cloud of particulates at
a given concentration  and  optical path length  is  strongly  influenced by
particle size and index  of  refraction, acceptance  angle  of  the transmis-
someter and other variables.   Much  work  has been done over  the  years in
developing  relationships between the  transmittance  (or  opacity)  of an
aerosol cloud and  its  particle concentration.   The most  useful  equation
relating particle concentration and several particle variables to opacity
is:

                           cm  -  fE   log 1/T

where

     c    -  particle mass concentration,
      m
     K    *  specific particulate volume divided by the aerosol attenua-
             tion coefficient,

     p    -  particle density,

     L    «  path length through aerosol cloud,

     T    -  transmittance through  aerosol cloud.

Limitations—
     Although transmissometers can  be considered  for the measurement of
fugitive particulate emissions, there are several problem areas:

     Uniformity of  Particulate Composition—Because of the  strong depen-
dence of response of the transmissometer to particle size and refractive
index, there can be  serious problems in quantifying roof monitor fugitive
emissions.   If  the emission  source or  sources within the  building have
variable  particle  size and/or refractive  index emissions,  the transmis-
someter readings may not give valid mass emission results unless special
calibration  procedures  can  be established  for the  specific  conditions
involved.

     Path Length  of Light Beam—Obtaining  sufficient signal strength at
the receiver to register the full range of opacity may impose limitations
on the path length over which  the  light beam can be transmitted.  This must
be experimentally determined using an emission source  similar, in terms of
particle size and refractive  index, to  that being measured.   The maximum
path length can then be  calculated  from  opacity versus  light path length
test results.
                                   -14-

-------
     Uniformity and Representativeness  of  Roof Monitor Emissions—If the
emissions flow out  of  the roof monitor  is  not uniform, problems  may be
encountered in using a fixed  path  and  geometry transmissometer to estab-
lish the average  emissions from the roof monitor in any given time period.
A second set of transmissometers mounted perpendicular to the primary set
might be required in such cases to establish  the  relative  uniformity of
emission flows from the roof monitor.

     Mechanical Vibration—Since alignment  of the source  and the sensor of
transmissometers is critical  to the  determination of  the opacity of the
emissions, vibration can  be a serious  problem in  their use  on roof moni-
tors.  Severe vibrations  can be experienced  in  roof monitors as the result
of craneway movements or  thermal plumes from high temperature operations.

Application to Fugitive Emissions—
     No published reports have been located describing the application of
transmissometers to fugitive emissions measurement.

     For a  transmissometer  to be successfully applied to the measurement
of  particulate  fugitive   emissions, data must be  obtained  regarding the
physical characteristics of the particles and the constancy of  their char-
acteristics  over  the  measurement  period.    In general,  transmissometers
could  best  be used  on fugitive  emissions  from processes  that generate
particulates of constant  physical  properties.

     Transmissometers might also be used in an upwind-downwind configura-
tion  if particle characteristics  remain constant  and it is  possible to
obtain  a reliable relationship between  particle concentration  and  the ex-
tinction coefficient.  Several transmissometer beams  could be  arranged to
intercept  the fugitive  particulate  plume  at various elevations.  Wind
velocity measurements  at each elevation would permit computation of the
particulate material flux at  several levels.

Remote   Sensing  of  Fugitive  Particulate  Emissions  Using  Time-Lapse
Photography

     Time-lapse  photography has  been used  in  a wide variety of studies to
identify and monitor pollution resulting from natural  and man-made  causes.
The main advantages of time-lapse photographic measurement are:  Action is
stopped and a permanent  record provided; changes in  emissions generation
and  flow direction  over relatively long  time periods  are  detected and
discretely  identified.

Principle of Operation—
      In time-lapse  photography a series of  discrete still photos  are  made
at  a  predetermined  time  interval on motion picture film and projected at
standard  speeds  to  reduce the time span of  the action.  Events  occurring
over  extended periods  of time  can  then be  viewed   in  short  periods or
stopped entirely for analysis.
                                   -15-

-------
     Time-lapse photography can effectively be employed to trace the path
of a dispersed plume, indicate changes in the siee of a plume, and deter-
mine if visible emissions are emitted from a  specific site over a specific
period of time, without long-term visual observation. It is also theoreti-
cally possible to  relate the optical density of  the  filmed image to the
opacity of a fugitive cloud.   The opacity  can in theory also be related to
a mass emission rate. To date this has not been accomplished successfully.
Image density measurements can be made at  this time by commercially avail-
able  densitometers,  but  no  technique  has been developed  to  relate such
density measurements to plume opacity or  particulate concentration.

Limitations—
     Time-lapse  photography  techniques have certain  limitations  for the
measurement  of  fugitive  emissions.  They are only  applicable to visible
emissions.   This mandates that they  be  used only to measure particulate
emissions of high  opacity.   They can ordinarily be used only during day-
light hours.  They cannot be  used during adverse weather conditions such as
rain, snow,  fog, etc., because of  visibility and contrast problems.

Application  to Fugitive Emissions-
      Purdue  University  is currently  conducting a  study for the National
Science Foundation to measure trace metals in the environment.  One portion
of  the  program is involved  in  a  preliminary study of fugitive emissions
from  coke pushing  operations at  a  steel plant along Lake Michigan   This
study was  conducted during  the summer  and  fall of 1975 and attempted to
quantify emissions primarily through the  use of hi-yolume  samplers  in the
fugitive emission  plume.   Photographs of the emission plume were used to
determine the time periods in which the sampler was within the  plume  and to
define  the  shape  of the  plume.   Stop-motion  analysis of the  film was
utilized,

Conclusions  and  Recommendations

      While each  of the four methods reviewed showed some promise of adapt-
ability  as  a useful tool  for  the  measurement  of airborne fugitive emis-
sions, no single method may  be used as a general technique  for the assess-
ment  of all types  of  emissions  to the atmosphere.   Each  could, with
considerable developmental effort, be used  to  measure either gaseous or
particulate  emissions, but not both.

      Interviews  with  users  and   manufacturers  of  the various  devices
studied  indicated  that no significant efforts were being made or planned
for their development as remote fugitive emissions monitors. Most  of those
interviewed  could  not provide even a  rough idea of  the amount of develop-
mental effort that might be required  to adapt their devices to  the intended
purpose or offer any estimates as to the probability of  the  success of such
efforts.

      In light of these facts, it was  recommended that no  further effort be
expended on  the  investigation or on  the  development  of a remote sensing
                                   -16-

-------
technique at  this  time under this  task.   It was suggested  that  TRC and
IERL/RTP maintain an awareness of  further  developments  in remote  sensing
techniques and devices  and  resume the effort if such developments indicate
a reasonable probability of successfully adapting a technique to the mea-
surement of fugitive emissions.

TASK 04 - EVALUATION OF SURFACE RUNOFF AT INTEGRATED IRON AND STEEL PLANTS

     A full description of  this  task is contained in Assessment of Surface
Runoff from  Iron  and  Steel Mills,  EPA-600/2-79-046, February  1979.   The
report abstract is quoted as follows:

          The  report presents the  results  of a program whose objec-
     tive was  to  assess whether surface runoff  from  iron and steel
     mills is an environmental problem.   Included is a compilation of
     existing data available before this  program, results  of informa-
     tion gathered from plant tours, and results  of a field survey at
     two  fully integrated  mills on tidal rivers.   Data collected at
     the  two sites indicate that the coal and coke storage piles and
     the  coal  and coke handling areas have the highest potential for
     contaminating stormwater.  The data also indicate that TSS run-
     off  concentrations  are typical of  urban runoff concentrations
     while TDS values  are  approximately one to two times  the typical
     urban runoff concentrations.  From plant tours  it was found that
     stormwater  controls  which presently exist  within  the  steel
     industry  are limited.  The only system specifically designed  for
     stormwater  control exists at  Armco's Houston Works where coal
     piles have been diked as a control measure for both fugitive  air
     emissions and stormwater runoff.  Some mills collect stormwater
     runoff  with process  wastewater  for subsequent  treatment at  a
     terminal  plant.   Those  methods which  are applicable  to  the
     industry  include  rainfall detention ponding  rings  for flat
     roofs,  swirl degritters, and retention basis or   sedimentation
     ponds.

          TRC  concluded that, with the  exception of runoff from coal
     and coke  storage  areas,  stormwater  runoff is not a problem when
     compared  to  point source control.

 TASK 05  - DESIGN  OF A  FUGITIVE  ASSESSMENT  SAMPLING TRAIN (FAST)

     This task, which  could not be  completed within the time limitations of
 this contract, has  been transferred to a new contract,  number  68-02-3115,
 for continuation.  A detailed report on the design, fabrication and testing
 of the sampling system will be prepared under the new  contract.   A brief
 summary of  the work completed on  the  task  is presented  below.

     Discussions  between  TRC  personnel  and  the  EPA  Project  Officers
 resulted in a target  design specification  for an  ideal sampling  train as
 the development starting point.  This ideal sampler was described as being
                                   — 17—

-------
able  to  obtain, from  the ambient air  in the vicinity  of  an industrial
fugitive emissions source, a 500-milligram sample of suspended particulate
matter and a similar-sized sample of organic vapors in an eight-hour sam-
pling period.  The particulate matter sample would be separated into respi-
rable  (smaller  than  3  micrometer)  and  non-respirable  (larger  than 3
micrometer) fractions.  The sample sixes were selected  to  correspond to  the
then-considered  minimum for complete  analysis including bioassay.    The
sampler was also to be self-contained and portable; it would require mini-
mum power and, using commercially available components wherever possible,
cost  less than $10,000 to  fabricate  in  the prototype version.

      An extensive computerized literature  search and review was conducted
in the hope of obtaining  sufficient  information on ambient concentrations
of industrial  fugitive emissions as  particulate matter and organic vapors
to prepare a realistic system design specification for the system, desig-
nated the Fugitive Assessment Sampling Train (FAST).  While this search  and
review revealed almost no data on ambient concentrations,  it did provide a
wealth of information on  emission rates from a wide variety of industrial
processes.  A  series of calculations based on recognized  atmospheric dif-
fusion equations  for  a  range of atmospheric,  topological and wind condi-
tions were then performed to relate the published emission rates to ambient
concentrations.   These  calculations indicated that an ambient concentra-
tion  of  200 micrograms  per cubic meter  can  be found within  100  to  500
meters of sources emitting between 0.6  and 23 kilograms  per hour—a range
covering  about 90 percent of  the industrial  sources for which  data ia
available.

      This 200  microgram per cubic meter concentration was used to deter-
mine  the  sampling rate required  to  obtain a 500 milligram  sample in an
eight-hour period of  5.2  cubic meters  per minute (184 CFM) as the initial
system  design parameter.   A Roots  lobe-type vacuum blower,  capable of
moving the required volume of air against a pressure drop of about 10 cm
Hg, was  selected as the  particulate sampling  prime mover.   A system of
drive belts  and  pulleys was utilized to operate the  blower at  the required
3800  RPM from  a three horsepower  drive motor.   The drive  system also
provides enough flexibility to adjust the speed and the sampling rate up to
about 20% if required.

      To provide  for the separation of  the particulate matter sample into
respirable and non-respirable fractions, an Air Correction Design 6UP San-
itary Cyclone  Separator was selected.   Its  design  capacity of 6.3 cubic
meters per  minute (222 CFM)  provides a Dso at  about  2  micrometers at a
pressure drop  of about 0.6 cm Hg.  The  cyclone was selected as preferable
to filter- or impactor-type collectors since the sample is removed from  the
sampling streams and minimizes the degradation in sampling rate or effec-
tiveness  caused  by the deposition of  particulate matter on flow-through
filters or impactor plates.

      Consultations  with  Mr.  Kenneth  Gushing of  the Southern  Research
Institute, under contract to the Process Measurements Branch in the area of
                                   -18-

-------
particulate matter sampling,  indicated that Reeves-Angel 934AH glass fiber
filter material would be  about  99.95%  effective in collecting the fraction
of  the  particulate matter sample  down to  about 0.3 micrometers  passed
through the cyclone.  A circular format was selected for the filter mate-
rial  to  provide the most  even distribution of the sample  on  the  filter
surface and minimize the pressure drop buildup.  A circular filter holder
was designed to acconmodate a 929 square centimeter (1  square foot) filter,
limiting the  pressure  drop across  the unloaded  filter  to  3.7  cm Hg.   A
louvered inlet  section was also  designed  to reject particles larger than
100 micrometers to complete the particulate matter  sampling section of the
train.

     To provide stable samples  of airborne organic  vapor emissions, it was
decided to utilize an adsorbent resin in a  removable canister that could be
easily transported  from  the  sampling  site to a laboratory for extraction
and analysis of the  sample.  Dr. Philip Levins of Arthur D. Little, Inc.,
under contract to the Process Measurements Branch in the organics sampling
area, provided consultation to  TRC on the resin.   The best available resin,
XAD-2, which  is almost  100%  effective in  retaining organic vapors C« and
higher, was determined to require a  canister  containing about 75 kilograms
to  provide  a  500-milligram sample.  This  was prohibitive from the stand-
points of size and cost, and the design criterion was revised to obtain the
minimum  sample  required  for  a  Level 1 assessment of  14 milligrams.  This
sample size requires only  2.1  kilograms of resin  and a  sampling rate of
only 0.14 cubic meters per minute (5 CFM).  A canister was designed and an
oil-less Cast  vacuum pump  selected  to draw  the  sampling  stream from the
main stream after  the particulate matter  is removed.

     The system design was reviewed and approved and  procurement and fab-
rication efforts  started.  At this time, the EPA's  Health Effects Research
Laboratory  suggested that  an additional size fraction of the particulate
matter sample  be  included to help in the assessment of the inhalable (less
than  15  micrometer) portion of  the emissions.   It was decided  to add  a
battery  of  six single stage  Sierra  Instrument impactors to the system to
effect this additional  fractionation  betweeen the inlet and the cyclone.
These impactors were designed to provide a Das for 15  micrometer particles
at  the system sampling rate with a  pressure drop of only 0.05 cm Hg, and
could therefore be added without affecting the system design.

      The final  system design is shown schematically in Figure 1.  Design
flow rates and pressure drops for each system element are shown enclosed in
brackets.   Samples retained  by each element  are  shown in parenthesis.
                                   -19-

-------
   INLET
  [184  CFM]
IK
•
ILE
V
^~
T
l\
—».
— •
T.C.
i
--S
                                       ,,.,r«
                                       FILTER
                                                    VACUUM
                                                     GA6E
                                                                   EXHAUST
                                                                    EXHAUST
                          [0.6  CM Ht]
                                                                  FLOW
                                                                  METER
                                                                  [6  CFM]
                     Figure  1.  Fugitive  assessment sampling
                               train design  operating condition
     The  procured and  fabricated elements  of  the prototype  system were
then loosely packaged onto  a  space  frame about  75 cm (2.5 feet) square by
183 cm  (6 feet) high to  allow easy  access to the elements during develop-
ment testing.  The main sampling blower and the organic vapor sampling pump
were separately mounted to improve the system's portability and permit the
location of the blower and  pump exhausts away from the sampling inlet.

     After a  successful  operational test  and a few modifications  to the
system were completed at  TRC,  the FAST was shipped to the Southern Research
Institute's laboratory for calibration testing of the particulate sampling
section.  Tests were run by Southern using monodisperse ammonium fluores-
cein aerosols provided by their vibrating orifice aerosol generator at
3, 10  and 15 micrometers.   The  test  results  for  the cyclone,  shown in
Figure 2 as points plotted on the manufacturer's design curve, are in very
good agreement. The test results for the impact or, also shown in Figure 2,
indicate good agreement with the design curve for smaller particles but are
considerably lower than expected  for the 15  micrometer particles of major
concern.
                                   -20-

-------
         100


         II


         II


         70

      M
      5  ••
      IM
      3  co
      *Z
      ik
      IM
      «  *°
      o
      u   SO
      Ml
      -J
      _l
      S   20


          10
              oust Finn
              SUIITIATII*
     IMPACT!*	
             ORCAOtO riATISO

     CYCLONE	-*
                                  I  T
   GYCIONE OISION CURVE
                                     1 '  IMPACTOR OESIIN
                                               CURVi    -I
                                     I
.3  .4  .5 •« .7.0.01.0       2     3   4  B  0  7  0 010

            PARTICLE DIAMETER.  MICROMETERS
                                                                   10
        Figure 2.   Collection  efficiency as  e function  of  particle
                    diameter for  the  FAST  impactor and cyclone
     Since it was felt that this  discrepancy was caused by bouncing of the
larger particles off  the  glass  fiber substrate used  in  the impactors,  a
test was run using a grease substrate in an attempt to reduce the bounce.
This resulted in a slight  improvement in  performance, but  not to a level
considered satisfactory for further development.  A joint  effort by TRC and
Southern Research has been initiated  to design and fabricate an elutriator
to replace the impactor as the 15 micrometer fractionator.  It is expected
that the  elutriator,  envisioned  as a battery of  parallel  settling cham-
bers, will replace and perform the  functions of both the  inlet louvers and
the impactors.

     A  field  test of  the  FAST has been planned at a coke oven battery,
where the system will be tested  simultaneously with two Battelle mega-vol
samplers and a standard hi-vol sampler in the measurement  of emissions from
coke pushing operations. This test, to be run  prior to the  fabrication and
                                   -21-

-------
 installation of the elutriator,  should provide sufficient data  for  con-
 tinued development of the cyclone and filter sections as well as an initial
 indication of the  effectiveness  of the  adsorbent canister train design.

      It  is expected that the  PAST will require some modifications and  a
 repeat of  the calibration and field testing cycle after the addition of the
 elutriator.  Verification tests of the total system will be performed after
 the modifications  and  additions  have been completed and  an operating  pro-
 cedures  manual will be prepared  and published for the  final version.

      The efforts to date in  the development of  this Fugitive  Assessment
 Sampling Train have  been quite successful and encouraging.  The completion
 of the  planned effort wil  provide a  useful tool for rapid,  reasonable
 assessments of fugitive particulate matter and organic vapors  from a  wide
 variety  of industrial  sources.

 TASK 07  -  PRELIMINARY INVESTIGATION OF TECHNIQUES FOR THE DETERMINATION OP
 HOOD CAPTURE EFFICIENCY

      Hooding has been used for many years as an effective means of collect-
 ing open source emissions such as fumes,  gases,  dust and  particulates, «o
 that they  may be transported to  a  control device  for  proper  treatment to
 reduce the amount  of pollutants  discharged into  the atmosphere.   Most of
 the previous work with hoods  and hooding  systems has been directed toward
 the control of small industrial  process emissions.   The  objective of  this
 task was to perform a  preliminary investigation  of methods to measure the
 capture  efficiency  of  hoods  for  large process emission source  control.

      The eventual goal  of this effort and subsequent efforts  indicated by
 the preliminary investigation is to produce  a  technical manual on  hood
 capture  efficiency measurements that will describe the techniques that may
 be employed to determine hood  efficiencies.

      A comprehensive  literature  search  was  conducted to determine  the
 state-of-the-art in hood capture  efficiency measurements.   Through discus-
 sions  with  individuals concerned with hood capture efficiency measurements
 in the Environmental Protection Agency, the National Institute of  Occupa-
 tional Safety and  Health,  industry, and research organizations, a number
 of processes  where  the measurement of hood efficiency could be important
 were  identified.  The  techniques currently being  used  to determine  the
 efficiencies  of  capture hoods for  some of  these  industrial processes were
 investigated  in  an attempt to formulate  a basic set  of specifications  that
 could  be applied to a general  technique.   New  techniques identified during
 the study were also analyzed on this basis.

Literature  Search

     An exhaustive  literature search was performed using the System Devel-
opment Corporation  (SDC) computerized data banks,  along  with literature
searches  performed  by the University of Connecticut Health Center and the
                                  -22-

-------
EPA library service.  Data baaea covered included NTIS, Pollution, Compen-
dex, APTIC, Toxicology, MEDLARS and the Library of Congress.

     Before  scanning  the data  bases, a  search strategy was  formulated
whereby a list of key word  identifiers, with  major and minor categories,
was generated.  These  identifiers  were  fed  into each data base, starting
with very general categories and working down to specific identifiers.  The
computer then searched its files and printed the titles of those articles
which contained those key words in the title  or abstract.  Typical key word
identifiers  used  in  this  search  included:   HOOD,  CAPTURE,  EMISSIONS,
EXHAUST, DESIGN, EFFICIENCY, ENCLOSURE, AIR POLLUTION, VENTILATION, FUME,
COLLECTION,  TESTING, AND PERFORMANCE  EVALUATION.   By using these identi-
fiers individually and  in various  combinations and permutations, as well
as  keying in certain  categories  of  articles not to  be  accessed, major
articles and reports on hood capture efficiency measurements were located.
The search  into  each  data base was usually considered complete when the
same  articles  or  group  of  articles  was accessed  by using  two  or three
different  groupings of key words.   Subsequent to the literature search,
phone calls  were made to the authors  of  the   more  pertinent articles to
attempt to trace secondary references  and to discuss their knowledge of the
state-of-the-art of hood capture efficiency measurements.

     Thirteen  primary  references were found in the open literature which
could apply to measuring hood capture efficiency.  The bulk of the material
found in  the literature  dealt with a variety of emission measurement tech-
niques, but  contained  little information on hooded processes or  the effi-
ciencies  of  hoods.

ADDITIONAL INVESTIGATIONS

      Since  the information obtained through  the literature  search  was so
sparse, additional information was obtained by directly  contacting knowl-
edgeable  individuals identified in the  search or  in  previous professional
contacts  as having an  interest  in, or concern with, hood capture  efficiency
measurements.

      Telephone contacts  and personal interviews with ten individuals,  four
from  academic  or research  organizations,  four from industry, and two  from
the EPA;  resulted  in the  identification of  nine measurement  techniques
potentially  applicable  to the problem and a number of industrial processes
that  could require capture  efficiency measurements as part of a  program to
improve  the  control of their emissions.

      These techniques  and processes were reviewed in detail and  reduced to
a list  of four techniques  that  could  provide  measurements with  acceptable
error limits and  five  processes representative of the problems  that might
be encountered in the  general application of  the techniques.  The measure-
ment  techniques  are:
                                   -23-

-------
      1 - Tracer techniques
      2 - Multi-point probes
      3 - Physical modeling
      4 - Optical measurements

 The processes ares

      1.   Metallurgical coke production
      2.   High alloy electric furnace steel production
      3.   Electrolytic smelting of aluminum
      4.   EOF steel production
      5.   Rubber tire production

      The techniques and their possible  applications  to the processes are
 described in the section following.

 Measurement Methods Applications

 Tracer Methods—
      Tracer studies  have  already  been  conducted for  industrial  hygiene
 tests of hood and ventilation system capture efficiency.  Nitrous oxide has
 been used  as  a gas tracer  to determine the air change efficiency  of an
 operating room ventilation  system, determined by measuring the  rate  of
 decay of nitrous oxide  in the operating room air.   Uranine dye released
 through a nebulizer has been used  as a  particulate tracer  to measure the
 effectiveness  of an auxiliary air  hood  in  a chemical use  laboratory.  in
 this application, auxiliary air is used to increase the air dilution inside
 a vent hood to more rapidly  dissipate toxic  fumes.  Any fumes not captured
 by the hood could  leave  the hood  through   the hood  face  area where  the
 operator stands.   Air samples were  taken in  the hood face with and without
 auxiliary air.  The hood  face  sample was  assumed to  be the  uncaptured
 emissions,  and  the  hood capture efficiency was thereby calculated from the
 weight  of uranine dye released in the hood and the hood face  sample  weight
 This  hood  capture efficiency method appears  attractive  for  application to
 larger  air  pollution emission  sources.

      Scale-up of the  tracer  technique to large emission sources would re-
 quire  that  uncaptured  emissions  be  sampled at several different  locations
 around  the hood face.  The  larger area within which the escaped tracer must
 be sampled can increase the errors in the hood capture efficiency measure-
ment  unless  a  very complex  sampling network is  set  up.  Costs and  error
 limit constraints would  have to  be examined on  a case-by-case basis  to
calculate the most  cost-effective configuration  for  a hood capture  effi-
ciency  test.    Another limitation  is  the  temperature  stability  of the
tracer.  For high temperature processes,  Freon or sulfur hexafluoride gas
and  uranine  dye or  dioctyl  phthalate particulate  tracers  would  not be
suitable.   High temperature  tests  would be  limited  to the  use  of more
stable materials, such  as  helium, titanium  chloride  and  zinc sulfide as
tracers.
                                  -24-

-------
     Tracers offer some distinct advantages over other hood capture effi-
ciency methods in that background  concentrations of the actual air contam-
inant  do  not interfere  with the efficiency  measurements.  Most  of the
tracers used for such  tests are detectable at far lower concentrations than
most air contaminants.  This can increase the  accuracy of the hood capture
efficiency  measurements  even where  tracer concentrations  are  low  as  a
result of  the  distance from the hood  to the  sampling point or the large
face area of the hood  where  they are released.  The tracers become diluted
with tracer-free air  as they  travel  further  from the hood or  when they
encounter large volumes of  outside air  at the  face of a large hood.  Since
most tracers  can be  accurately measured in  the  parts  per billion range
while most pollutants  can only be  measured in  the parts per million range,
it  is  possible  to measure  a tracer added  at a  concentration  as much as
three orders of magnitude lower  than that of the pollutant emitted and have
the same measurement  accuracy.

     A basic procedure for  a tracer test  would be as follows:

     1.   Background  measurements are taken  of  the tracer concentration
          before any  tracer gas is released in the area.

     2.   The  tracer   is introduced  uniformly into  the emission  source.
          The time period and rate of  addition are measured.

     3.   A sample  of the   air  captured  by the hood is taken in the duct
           leading from the hood at a point where  flow is generally  uniform
           in  the duct.  Tracer concentration and total gas flow are mea-
           sured for a definite  time  interval.

     4.   Over   the same time  interval,  the  hood  face concentration of
           uncaptured  tracer and the flow of  air away  from the hood are
          measured.

     5.    Background  measurements of tracer concentrations are taken after
           release of  the tracer gas  has  stopped.

     6.    The mass  flow rates  of tracer in  the duct and  around the hood
           face  are calculated.  This data is used to determine hood  capture
           efficiency.

      7.    The test  is repeated as required to determine the  accuracy  and
           precision of results.

     Application of  tracers to sources of interest  for hood capture effi-
 ciency measurements would  include:
                                    -25-

-------
         Source
                              Characteristics
                                     Tracer
1.  Metallurgical Coke
    Production
    High Alloy Electric
    Furnace Steel
    Electrloytic Alum-
    inum Smelting
4.
BOF Steel
Production
5.  Rubber Tire
    Production
Hot, short tine interval high
volume flow through large hood
area.  Fume, particulates &
gases released.

Hot, short time interval high
volume flow through large hood
area.  Fume, particulates &
gases released.

Hot, continuous low volume
flow through large hood area.
Fume, particulatea & gases
released.

Hot, short time interval high
volume flow through large hood
area.  Fume, particulates
released.

Warm, short or long time
interval low volume flow
through small hoods.  Fume,
particulates & gases released.
                                                        Helium or
                                                        titanium
                                                        chloride
                                                        Helium or
                                                        titanium
                                                        chloride
                                                        Helium or
                                                        zinc sulfide
Helium or
titanium
chloride
                                                           ,  Freons,
                                                        uranine dye or
                                                        dioctyl phthalate
The tracer method is given  the  highest  ranking of the four possible hood
capture efficiency measurement methods.

Multipoint Probe Methods—
     Multipoint probe methods have already been applied with some success
to the determination of hood capture efficiency for large scale processes.
Recent EPA studies on the capture of emissions from coke pushing operations
using coke-side sheds have shown that there are deficiencies in the pushing
system design that allow some particulate and fume emissions to escape at
the ends of the sheds.  Tests were  performed using EPA method 5 particulate
test trains traversing the shed leakage points during pushing operations.
Tests at two coke ovens showed the utility of  the method.  At the same time
as these  tests  were conducted,  the particulate and  fume  captured by the
shed were measured using an  EPA method 5 train traversing the duct leading
from the  shed  to  a  stack or control device.   The success  of these tests
demonstrates  that  this method  can be  applied  to  significant  emission
sources  where  hood  capture efficiency measurements  would be needed to
define the potential degree of control of the source in question.

     In comparison to the tracer-method, which can use a similar sampling
method and network,  the multipoint probe method suffers from two disadvan-
tages.  The first  is that a  direct measurement of the  pollutant in question
                                   -26-

-------
can be less sensitive and  accurate  than  that of a tracer.  The second is
that the background levels  of  pollutant within the plant atmosphere can be
significant enough to cause appreciable errors in direct measurement.

     The multipoint probe method is  a direct measure of hood capture effi-
ciency under actual operating  conditions, and in that sense can be termed a
more dependable  method  of  determining hood  capture efficiency  than the
tracer method.   As with  tracers,  the  sampling network can be complex and
the cost-effectiveness of the  testing method must be  assessed on a case-by-
case basis.

     The  basic  test  procedure would be  identical to  that  described for
tracers,  except  that  the actual pollutant  rather than  a  tracer  would be
measured.

     Application  of  multipoint  probes  to  sources  of  interest  for hood
capture efficiency measurements would include:
        Source

1.  Metallurgical
    Coke Production
2.  High Alloy
    Electric Furnace
    Steel
3.  Electrolytic
    Aluminum Smelting
4.  BOF Steel
    Production
 5.  Rubber Tire
    Production
         Characteristics

Hot, short time interval high
volume flow through large hood
area.  Fume, particulates &
gases released.

Hot, short time interval high
volume flow through large hood
area.  Fume, particulates &
gases released.

Hot, continuous low volume flow
through large hood area.  Fume,
particulates & gases released.

Hot, short time interval high
volume flow through large hood
area.  Fume and particulates
released.

Warm, short or long time
interval low volume flow
through small hoods.  Fume,
particulates & gases
released.
   Pollutant
   Measured

Particulates
by EPA Method
5
Particulates
by EPA Method
5 (Modified)
Particulates
by EPA Method
5

Particulates
by EPA Method
5 (Modified)
Hydrocarbons
by FID.
Particulates
by EPA
Method 5
     The multipoint probe method is given the second highest ranking of  the
 four possible hood capture  efficiency measurement methods.
                                   -27-

-------
Physical Models—
     Physical models have been applied to the study of hood capture effi-
ciency measurements,  especially for  industrial  hygiene purposes.   Com-
pleted studies  include an investigation of  a  modeled mining ventilation
system using  methane  as  a quantitative  tool  and  smoke  for  visual  and
quantitative measurements of hood capture efficiency.  A scale model, low-
volume, high-velocity hood system was studied using  dioctyl phthalate as a
tracer, its concentration being measured  by a light  scattering photometer.
For hot processes, laboratory heating devices and titanium chloride smoke
was used to follow the travel path of heated fumes from their source to a
nearby capture  hood.   Temperature measurements  were used  to determine
pollutant flux.

     The physical modeling technique  serves  best  as a screening test for
full scale measurements rather than an end in  itself.  There are several
dangers in using the  model  data  to  scale up to full scale hoods.  The model
system may have different characteristics unless all required similitudes
are carefully  examined.   Even  a  single  key element out of  scale could
distort the test results substantially.  In small scale systems, the mea-
surement  techniques   can  physically  perturb the  system by  introducing
obstructions to flow  or by diverting  the flow path.  Models  can be very
helpful in evaluating the design parameters for  full scale systems.  It is
much more cost-effective to work out  design  changes to optimize emission
collection on  a small  scale  before  constructing  a  full scale  hood  and
attempting to measure its emission capture efficiency.

     A basic procedure for a physical model  test would be as follows:

     1.   A scale model system is constructed which retains the similitude
          of the key variables of a full  scale system.

     2.   An appropriate gas,  particulate or  aerosol is selected to repre-
          sent the pollutant of interest  in  the system.

     3.   A system is installed  to  present the pollutant to the hood in the
          same manner as in a full scale  system.

     4.   Background readings of the scale model pollutants  are taken.

     5.   A test is run using a  probe system to capture the pollutant  in
          the hood and on the hood  face.    The  time  interval and  total
          pollutant release are measured.

     6.   Background  pollutant readings are  taken after pollutant release
          has been terminated.

     7.   If possible, all of the  pollutant  captured  by  the  hood is col-
          lected in a control device as is all  of  the pollutant  not cap-
          tured by the hood.
                                  -28-

-------
     8.   The concentration, mass flow rate and/or total mass of collected
          and uncollected  pollutant are  measured and used  to determine
          hood capture efficiency.

     Application of physical models  to sources  of interest for hood cap-
ture efficiency measurements would include:
         Source

1.  Metallurgical Coke
    Production
4.
5.
    High Alloy Electric
    Furnace Steel
    Electrolytic
    Aluminum Smelting
BOF Steel
Production
Rubber Tire
Production
      Characteristics

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases
released.

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases
released.

Hot, continuous low volume
flow through large hood
area.  Fume, particulates
& gases released.

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases
released.

Warm, short or long in-
terval low volume flow
through small hoods.
Fume, particulate & gases
released.
                                                        Physical Model

                                                    Rectangular opening
                                                    with infrared heater.
                                                    Titanium chloride
                                                    smoke or helium in
                                                    test gas.

                                                    Cylinder with gas
                                                    flame at bottom.
                                                    Titanium chloride
                                                    smoke or helium in
                                                    test gas.

                                                    Scale model probably
                                                    not possible.
Cylinder with gas
flame at bottom.
Titanium chloride
smoke or helium
as test gas.

Scale model of
machine & infrared
heater.  SFe, Freons,
zinc sulfide, dioctyl
phthalate, uranine
dye or titanium
chloride tracer
     The physical model method is considered the first  step in the applica-
tion of other hood capture efficiency measurement methods,  and not an end
in itself.

Optical Methods—
     Optical methods have not yet been applied to the measurement of hood
capture efficiency, but they have been applied to fugitive emissions mea-
surement.   The  extrapolation from that one use  to  the  other can be made
quite  reasonably.   The  fugitive emissions from coke oven  pushing were
studied using  two movie cameras  located  on horizontal axes  through the
                                   -29-

-------
plume perpendicular to each other to define the plume size and  shape and to
verify auxiliary  in-plurae  particulate measurements.   Fluoride emissions
from gypsum ponds have been measured using a  long  path infrared  transmis-
someter  with  auxiliary  wet  chemical  fluoride   measurements.    Optical
methods are therefore applicable to both gases and particulates and would
be well suited to hood capture efficiency measurement.

     Optical  methods  require  calibration  by an  independent  measurement
method and are therefore  no more  reliable or  accurate than the calibration
method.  In addition,  for specific sources, the backscatter of  light radia-
tion by the emissions  can be so severe that no meaningful measurements can
be taken.  This can be true whether the optical method measures backscatter
of light from the emissions or  forward transmission through the emissions.

     Optical methods do have the advantage of not requiring complex sam-
pling networks and can be operated at a safe  distance from hazardous areas
near the source of  emissions.  Optical  methods can also isolate specific
pollutant in the source if  all emissions  are gaseous,  and can follow the
travel path of uncaptured emissions for considerable distances.

     Considerable effort would be required to extend the state-of-the-art
of optical methods to include their use in hood capture efficiency tests.
The basic methods  are applicable as  they  are now employed  for fugitive
emissions measurement, but modifications to  effective  ranges and sensi-
tivities and extensive calibration efforts are required before efficiency
tests can be undertaken.

     A basic procedure for an optical method  test would be as follows:

     1.   The optical monitor  is  set up  to scan  the  area around the hood
          face.  Velocity measurement devices are also set up here.

     2.   A calibration run is  done on the optical  monitor by introducing a
          known amount of  the  pollutant to  be measured  into  its  optical
          path  and  a background  reading   done  for  the  pollutant  of
          interest.

     3.   The pollutant is  released  in  the vicinity  of  the  hood,  and the
          sampling time of the optical system determined.

     4.   The optical monitor scans the  hood  face area and a sample taken
          from the hood duct.

     5.   The  calibration  run is  repeated  and  a background  reading is
          taken.

     6.   The mass  flow rate  in the duct  and around  the  hood  face is
          determined from the calibrated optical  measurements and the ve-
          locity traverses.
                                   -30-

-------
     7.   Hood capture efficiency  is  then directly calculated  from mass
          flow rate data.

     Application of optical methods to sources of  interest  for  hood cap-
ture efficiency measurements would include:
         Source

1.  Metallurgical Coke
    Production
2.
High Alloy Electric
Furnace Steel
3.
Electrolytic
Aluminum Smelting
4.  EOF Steel Production
5.  Rubber Tire
    Production
      Characteristics

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases re-
leased.

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases
released.

Hot, continuous low volume
flow through large hood
area.  Fume, particulates
& gases released.

Hot, short time interval
high volume flow through
large hood area.  Fume,
particulates & gases
released.

Warm, short or long in-
terval low volume flow
through small hoods.
Fume, particulates &
gases released.
                                                     Optical Methods

                                                    Photographic plus
                                                    EPA Method 5, or
                                                    calibrated trans-
                                                    mis someter.
Photographic plus
EPA Method 5, or
calibrated trans-
mi s some ter .
Infrared long
path or calibrated
transmissometer.
                                                    Photographic plus
                                                    EPA Method 5, or
                                                    calibrated trans-
                                                    missometer.
                                                    Infrared long path
                                                    or calibrated
                                                    transmissometer.
     The optical monitor can be used for  hood capture efficiency  measure-
ment  if development work  is done  to  adapt it  from its  present  use in
fugitive emission measurement.

RECOMMENDATIONS FOR FURTHER  INVESTIGATION

     The  information gathered  in  the  literature search  and interviews
conducted  during  this  task, while  sufficient  to provide descriptions of
the applicability of a few possible measurement  techniques for the deter-
mination  of large hood  capture efficiencies, clearly  indicates that no
really  definitive  work has  been  completed in this  area.    There  is not
enough  information to identify any single method  as universally applicable
or to formulate a meaningful specification  for such a method.  To correct
                                   -31-

-------
these deficiencies  and  provide inputs to a  technical  manual,  TRC recom-
mends that a program be  implemented to test the proposed methods to deter-
mine  the  most applicable  method  for  general  use and  the modifications
required  to  its  application to specific source  and  hood configurations.
The method and its  modifications could then be  evaluated  in a series of
field trials  at  actual  process sites  and  the  results  used  to  prepare a
technical manual describing the planning and performance of broad capture
efficiency measurement programs.
                                  -32-

-------
                                SECTION 3

                        SERVICE TASK DESCRIPTIONS

TASK  20A -  EVALUATION OF  A  PROPOSED NO-DUCT  QUASI STACK  MEASUREMENT
TECHNIQUE

     An evaluation was made  of a measurement method proposed by another EPA
contractor to  determine  the emission strength from an  electric furnace.
The method proposed  to determine  source  strength by simultaneously moni-
toring  the temperature and  pollutant concentration at points  within the
furnace plume  and relating  the measurements  to source strength through a
hypothetical relationship developed in  a series of laboratory-scale exper-
iments .

     The evaluation resulted in a recommendation that additional analysis
of furnace plume  dynamics  be performed to more  fully define  the circum-
stances  in  which the  proposed test method  or a  similar  approach would
greatly simplify emissions quantification.

TASK 20D - PREPARATION OF ALTERNATIVE MEASUREMENT APPROACHES FOR EMISSIONS
FROM A COKE QUENCH TOWER

     A test methodology and  plan for the measurement of emissions from coke
quenching towers was prepared.  The test program, to provide data for the
establishment  of  specific emission  factors,  was  to be  carried  out  by a
contractor to another office within the EPA.

     Methodologies employed  in earlier  attempts  to  measure  coke quench
tower emissions were reviewed and found basically ineffective.  The prob-
lems involved in making measurements were defined:

       o  Emission period less than 3 minutes long.

       o  Emission velocity,  temperature,  and moisture content vary with
          time.

       o  Emission concentration likely to vary across tower opening.

       o  High water vapor and liquid droplet proportion in emissions.

       o  Emission exit flow likely to be skewed by tower baffles.

       o  Area of emission exit flow large.

       o  Particulates present in  wide  size range  and as  dry solids,
          suspended solids, and solids condensed from vapors.
                                   -33-

-------
     The test methodology suggested utilized a modified EPA Method 5 sam-
pling train/Aerothera HVSS  4 CFM stack sampler with  a probe designed to
traverse  the tower opening.   A  resin  bed was added  to  collect organic
species emissions.  Standard sample analyses were suggested for the prob-
able pollutants.  Analyses of the various quench waters used were suggested
to  determine the effect  of water  composition on  the character  of  the
emissions.

     The test  plan  specified the  necessary pretest survey information to
be  obtained;  the mechanical operation, manner and  frequency of sampling
required; and  the data  to be  recorded.   It also  included an estimate of
manpower and time requirements and total program costs.

TASK 20E - PREPARATION OF A SAMPLING PROGRAM FOR FUGITIVE EMISSIONS FROM A
COAL CLEANING FACILITY

     A sampling program for  the measurement of fugitive emissions at a coal
cleaning facility proposed to be built in the state of Pennsylvania was the
subject of a report prepared as the  initial  phase of this task.  The report
specified the  allowable emissions from such a  facility under applicable
Federal and State Agency regulations for both atmospheric contaminants and
waterborne effluents.   Principal  potential  sources  of fugitive emissions
at  the proposed facility were identified,  along with their typical air and
water pollutants.  Brief upwind-downwind air sampling and receiving body-
runoff water sampling programs were described for a typical coal cleaning
facility and rough cost-manpower estimates developed.

     The final phase of this task was a review of a preliminary test plan
prepared by  an EPA  contractor  for the  proposed Homer  City  coal cleaning
facility.  The plan was judged to be quite effective  for meeting most of
its stated objectives.   An improved  program  for obtaining specific ambient
measurements was  suggested.  It was  also  suggested that the test plan be
revised to include measurements of opacity from specific plant sources as
indicated in the proposed EPA standards cited in the Phase I report.

TASK 20H -  ASSISTANCE  IN THE  PREPARATION  OF A GUIDELINE ON PARTICULATE
FUGITIVE EMISSIONS

     A number of  service  task-type  reports  on  particulate fugitive emis-
sions from this  and previous EPA contracts were supplied to another  EPA
contractor preparing a  particulate  guideline document.   A review of that
contractor's work plan  resulted in  a  number of suggestions  to facilitate
the proposed effort or to readily broaden its scope.

TASK 201 - CRITICAL REVIEW OF A FEASIBILITY STUDY OF SOURCE IDENTIFICATION
FROM AMBIENT AIR MONITORING

     A critique of a very lengthy feasibility study of source identifica-
tion from ambient  air measurements, prepared by another  EPA contractor,
resulted in  a  suggestion for modifications to the  report.   A format  was
                                   -34-

-------
outlined, and a section-by-section  list of  actions  suggested  to make the
study more readable and useful.

TASK 20J  -  PREPARATION OF  A MEASUREMENT  PROGRAM FOR  PARTICULATE  MATTER
EMISSIONS FROM IRON ORE LOADING OPERATIONS

     This task, to prepare  a proposed  test  method for the measurement of
particulate matter as  fugitive emissions  from an iron ore mining opera-
tion,  involved  a  visit to  the U.S. Steel Mintac mine in Mountain Iron,
Minnesota, to observe the  operations generating the  emissions.  The opera-
tion of  principal  concern was the  loading  of iron  ore into rail cars by
electric  shovels, which generates  a puff  of suspended particulate matter
with each shovel dumping.   The puffs were  observed to diffuse rapidly into
the ambient atmosphere, becoming invisible after  traveling no more than 10
meters even  on a day  with low prevailing wind  speeds (estimated  at 3-6
km/hr).

     A method was  proposed  to obtain particulate matter samples from the
puff as  it  passes  a  point  immediately downwind  of  the loading operation
utilizing isokinetic  samplers with 37 millimeter filters.  The samplers
were to  be  suspended  from poles that could be manually located near the
centerline of the puff and controlled to operate  only when actually in the
puff.  Microscopic  examination of  the filters would  provide approximate
concentration  levels  and  particle  size distributions.   Two synchronized,
electrically operated  cameras would be utilized  to  obtain  pictures of the
puff in short intervals as it passed the sampling point, one camera aligned
along the puff's path and  the other normal  to the  path.   The pictures would
be analyzed to determine  the  size and shape  of the puff.  These  size data,
combined  with  the  concentration data obtained from the filter analyses,
would  provide  a good  estimate of the total  particle flux  at the sampling
point, assuming that the particle density  is constant and  that the distri-
bution within the puff is reasonably close  to normal.

TASK 20L  - REVIEW OF A TEST PLAN FOR AMBIENT MONITORING OF FLUIDIZED BED
COAL COMBUSTOR EMISSIONS

     A plan prepared  by another EPA contractor  for  an ambient monitoring
program designed to determine the impact of  operating a fluidized bed coal
combustor on  the ambient  atmosphere in the  Georgetown, D.C. area was re-
viewed under this task.

     The  plan was found to  be generally well conceived.   Suggestions were
made to  add an additional SOa sampling site  and  to reschedule the proposed
program  to  shorten the required  time duration and reduce  the overall pro-
gram cost.
                                   -35-

-------
TASK 20M - REVIEW OF A SAMPLING PLAN FOR FUGITIVE EMISSIONS FROM PETROLEUM
REFINERIES

     The objective  of this  task  was  to review and  evaluate  a work plan
prepared by another EPA  contractor for  the measurement of fugitive emis-
sions at petroleum refineries.  The evaluation included a visit to a Gulf
Coast refinery in the company of  the contractor's personnel who were then
making preparations to conduct the measurement program.  The task report,
based on  the  observations made  during  that visit,  discussions  with the
contractor's personnel and our prior experience  in measuring similar fugi-
tive emissions, included a summary section quoted in part as follows:

          TRC believes the plan to be generally sound in its concep-
     tion and intent.   Specific recommendations  are discussed in some
     detail in ... this Report.   Those recommendations include:

     1.   Removal  of  refinery  location,  size, and  age  as  "choice
          parameters" and the consequent restriction of choice param-
          eters  to  so-called micro factors.   We believe  the  focus
          should initially be on specific pieces of  equipment cur-
          rently in use at U.S. refineries.

     2.   Revision  of the statistical  experimental  design  so that
          within a  factored  "micro-type"  the sampling  be conducted
          randomly.

     3.   Revision of the measurement method for baggable sources to
          eliminate potential  and,  in our  judgement,  likely oppor-
          tunities for improper  bag sealing around  pieces of equip-
          ment.  Use  of  a  quasi-stack method  appears  to be a better
          alternative.

     4.   Revision of the measurement method for certain non-baggable
          sources, such as the uncovered API Separators at Gulf Coast
          Refinery A.   The emphasis here is that direct measurement,
          even with admitted imperfections,  is  to  be  preferred  to
          indirect measurements   combined  with  theoretical  assump-
          tions implicit in  the mass  transfer equations  proposed in
          the Plan.

TASK 20N  - CRITIQUE  OF  A PROPOSED PLAN FOR  THE  MEASUREMENT OF  MINING
PARTICULATE MATTER EMISSIONS

     This  task,  essentially  a  follow-on  effort to  Task 20J  described
above,  consisted of a  critique of  the measurement plan prepared by another
EPA con  tractor for  the ore-loading operation  particle  sampling.   The
critique was delivered as a verbal discussion with the EPA project officer
and was not  formally  documented.   Its basic  content  is  reflected  in the
critique prepared under Task 20R, described below.
                                  -36-

-------
TASK 20P -  PREPARATION  OF A CONCEPTUAL ROOF MONITOR MEASUREMENT APPROACH
FOR AN OPEN HEARTH SHOP

     This task,  to develop a conceptual approach for an open hearth shop
roof monitor  fugitive  emissions  measurement  program,  was designed  to
assist another contractor to EPA Region III  to  quantify the  fugitive emis-
sions from the U.S.  Steel Corporation's Fairless Works.

     A visit to the site resulted in the preparation of an extensive report
that outlined  the  variables affecting the  generation and  measurement of
the fugitive emissions  and proposed  a measurement approach with detailed
descriptions of the monitoring system principles  and design, the necessary
meteorological monitoring system and its associated data transmittal sub-
systems, and the methods proposed  for data acquisition,  processing,  and
presentation.   Cost and schedule  estimates  were  also  included  in  the
report.

TASK 20R -  CRITIQUE OF EXPOSURE PROFILING TECHNIQUE FOR PARTIUCLATE MATTFW
FUGITIVE EMISSIONS                                                 "Ana*

     This task was a critique of the  exposure profiling technique for the
quantification of participate fugitive emissions  as developed  and tested
by  another  contractor   for  lERL's  Metallurgical  Processes Branch    The
technique is basically the same as that used  in  the measurement of the iron
ore loading operations  referenced  in Task 20N,  above.   A  summary of the
critique is quoted as follows:

          The exposure  profiling technique described [in the] report
     'A Study of  Fugitive Emissions from Metallurgical Process (inte-
     grated Iron and  Steel Plants)1 appears  to  be a viable method for
     estimating  source   strengths  or  emission  factors  from  open
     sources  of  particulate  fugitive emissions.    Its  level  of
     accuracy,  as  indicated  by the  test  results in the report,  is
     probably commensurate with the requirements  of a level 1 assess-
     ment,  within a factor of 2 or  3  of the actual emissions.

          Both the vertical  grid array of  samplers,  used  primarily
     for line or large-area  sources,   and  the  two-dimensional  array
     of samplers, used  for small-areas or virtual point sources  are
     essentially small  versions of  the more generally applicable up-
     wind-downwind technique's sampling arrays or network.   Designed
     to be used much closer to emission sources than upwind-downwind
     samplers,  the exposure  profiling technique  can obtain particu-
     late samples  with  a much  lower  ratio of background  to  source
     emission  concentrations,  enabling  the   estimation of  source
     strength  from a specific  source  that would  be difficult  or
     impossible to obtain using other techniques. Exposure profiling
     is, in general, less  expensive  and faster than upwind-downwind
     sampling in situations where both are applicable.
                                -37-

-------
          The technique is not  as  universally applicable, however, as
     upwind-downwind sampling.   It requires site conditions that per-
     mit  close  access  to  the  source  and  is  limited  to winds  of
     moderate speed and fairly constant direction.  Its proximity to
     the sources means that  the samples will  contain a high propor-
     tion of particulates larger than those normally considered sus-
     pended,  thus  increasing   the   difficulty  in  quantifying  the
     suspended fraction.

          The  review  of  the  exposure  profiling technique  leaves
     little  doubt  that  the method  is  reasonable effective  within
     limits.  Additional  testing is required to determine what those
     limits are and to help  refine the method so that it will produce
     more consistent results.

TASK 20U -  CONSULTATION  AND ASSISTANCE TO AN EPA CONTRACTOR  IN SAMPLING
REFINERY HYDROCARBON FUGITIVE EMISSIONS

     A meeting at RTF  with  three  other  EPA contractors and their Project
Officers was held to define and discuss the problems  being encountered in
the sampling of hydrocarbon  fugitive  emissions in  programs being conducted
at a number of petroleum refineries.  The meeting resulted in a subsequent
visit to one of the refineries  to  observe the situation.  As the result of
this visit,  TRC  prepared  a  procedure for the sampling of  Cs  and lighter
hydrocarbons using Tedlar bags and evacuated plexiglass  containers.   TRC
also prepared recommendations to  improve  sampling methods  to  ensure iso-
kinetic sampling of multiphase  sources.
                                  -38-

-------
                                SECTION 4

             DOCUMENTATION AND PUBLICATION TASK DESCRIPTIONS

TASK 20B - PREPARATION OF A GUIDELINE FOR FUGITIVE EMISSIONS MEASUREMENTS

     The preparation of a guideline for the selection of the moat effective
program for the measurement of fugitive emissions from industrial sources
was the objective  of this  task.  A  draft of the guideline  was  prepared
according to the original directive to  include definitions and categoriza-
tions  of  airborne  industrial  fugitive  emissions;  descriptions  of  the
existing techniques for their measurement;  criteria  relevant  to the selec-
tion of measurement  techniques; and  baseline  estimates  of  manpower,  time
requirements, and costs for typical  measurement programs.

     Prior to the release of this draft  for EPA approval, it was decided by
the EPA Project  Officer to broaden the scope of the guideline to include
the same  information  for waterborne fugitive  emissions.   The  task  was
therefore deferred until the completion of the water-related Tasks 02 and
04 which were expected  to provide much of the  information  required.   The
task will be  resumed at a later date  and  will  result  in  a guideline for
measurement programs for all types of industrial fugitive emissions,  both
air and waterborne.

TASK  20C  -  PREPARATION  OF  A SUMMARY  OF  AIRBORNE  FUGITIVE  EMISSIONS
MEASUREMENT TECHNIQUES

     The initial phase of this task was a review of a chapter on fugitive
emissions sampling in a Level 1 Environmental Assessment Procedures Manual
prepared by  another contractor for  IERL/RTP.   A meeting  with the other
contractor and their EPA Project Officer  to  discuss the comments made in
the review led to  a directive to  prepare  another version of the chapter.
This version  included  summaries  of  measurement  techniques, measurement
equipment, sampling  system  designs,  analysis methods,  and data reduction
for both  air  and waterborne  fugitive  emissions in a  Level 1  or survey
assessment.   Estimates of manpower requirements  for  the designated assump-
tions of preliminary  survey  completion and a one-week field program were
prepared.

     The TRC-prepared  chapter  was integrated with some of the originally
reviewed material by the other contractor to produce  a chapter in the draft
version of the document that concentrated  on an  ambient  air  assessment.  A
separate chapter addressed  the problem of sampling liquids but  made no
specific reference to waterborne fugitives.

TASK 20F - PREPARATION OF PAPERS FOR SYMPOSIUM  PRESENTATIONS

     This task included the  preparation and  presentation of  two technical
papers relative  to the measurement of fugitive emissions for the Symposium
                                   -39-

-------
on  Fugitive  Emissions;  Measurement  and  Control,  held  in  Hartford,
Connecticut,  in  May of  1976.   The  Symposium was  organized  by  TRC for
IERL/RTP  under a  previous  contract.    The papers  and  their  published
abstracts were:

     Fugitive Emissions Problems in Perspective - J.E. Yocom

          Primary emphasis on control of air and water pollutants has been
     placed on point sources.   Fugitive emissions,  or those emitted from
     non-point sources, are important  contributors to environmental degra-
     dation  in many areas  and  in relation to many  types  of industrial
     operations.  Because of the potentially high cost of controlling such
     emissions  it  is  important  that  their  significance  be  accurately
     assessed.

          This paper  discusses many types  of  fugitive  emissions  and the
     methods  for  assessing them.  Fugitive emissions are  extremely site
     specific in respect to their measurement and control, and this paper
     presents examples of measurement  programs  that put fugitive emissions
     in  proper perspective  in  relation  to other source  categories and
     other environmental impacts.
     A Guideline for the Measurement of Air-borne Fugitive Emissions from
     Industrial Sources - H.J. Kolnsberg

          The paper presents a guide for the selection of the most effec-
     tive program for the measurement of airborne fugitive emissions from
     an  industrial  source.   The  quasi-stack,  roof monitor  and upwind-
     downwind techniques presently utilized for  sampling  a wide variety of
     air-borne pollutants are described.  General criteria for the selec-
     tion of  the most effective  sampling  program,  relative to character-
     istics of  the  site,  process and emissions  are  discussed.   Baseline
     estimates of manpower,  time and cost requirements  for  typical mea-
     surements programs for each technique are provided.

     The publication of the proceedings  of  this  symposium and the prepara-
tion of  the  symposium review article that was  published in the November
1977 issue of  the  Journal  of the Air Pollution Control  Association were
also included in this task.

     The  task was  completed  with  the publication  of three  Technical
Manuals on the measurement of fugitive emissions by the Quasi-Stack, Roof
Monitor and Upwind-Downwind sampling methods.  These Manuals were prepared
under previous contracts for  IERL/RTP and  are  designed to  guide environ-
mental engineers in the  utilization of the methods.  Each  presents criteria
for the selection of the best  applicable method and describes the sampling
strategies and equipment,  sampling system design, sampling techniques, and
data reduction.  Estimates  of  manpower and  time requirements are presented
for typical overall and specific measurements programs.
                                   -40-

-------
TASK  20G  - PREPARATION OF A SUMMARY OF FUGITIVE EMISSIONS WORK COMPLETED
UNDER IERL CONTRACTS

      This  task  involved the  preparation of a brief historical summary of
the  fugitive emissions work carried  out by  TRC for  IERL/RTP  under two
previous contracts and  early work on the current contract.

      The summary  described the  results  of  the tasks performed under Con-
tract Number 68-02-1815, "Development of Sampling Procedures to Determine
the Emission Rate and Chemical and Physical  Characteristics  of Fugitive
Emissions" and Contract Number  68-02-2110,  "Development of Sampling Proce-
dures - Fugitive  Emissions (Particulate and Gaseous)."

     The tasks  under  Contract  Number  68-02-1815  included the identifica-
tion of major industrial sources of fugitive emissions, the evaluation of
samping  strategies  applicable  to  the measurement   of  the  identified
sources, the preparation of  a  field test procedure utilizing one  of the
evaluated   strategies   (the   quasi-stack   method   was   selected),   the
performance of the field program, and the preparation of a technical manual
for the selected strategy.

     The tasks  under  Contract Number  68-02-2110  included the preparation
of technical manuals  for the roof  monitor and  upwind-downwind  sampling
strategies; the review and  evaluation of fugitive  emissions  data  and/or
proposed sampling plans and methodologies  for coal  storage  piles,  copper
smelters,   cattle  feedlots,   sand  and gravel  transport,  abestos  storage
piles, and steel furnace operations; and the organization of the Symposium
on Fugitive Emissions: Measurement and Control.

TASK 20K - PREPARATION OF A BIBLIOGRAPHY OF RESEARCH PROGRAM ABSTRACTS

     A bibliography of research program abstracts related to the measure-
ment or control of  fugitive  emissions was prepared  from the most recent
published compilation of such abstracts.  This was EPA 600/7-76-025, pub-
lished in November 1976 and listing 377 programs started in FY 1975 under
the  auspices of  9  different   government  agencies.    The  bibliography,
intended by the project officer as  an  internal IERL reference,  listed 74
fugitive emissions related programs and briefly summarized their fugitive
emissions relevance.

TASK  20Q  - STUDY TO  DETERMINE  THE NEED FOR  A TECHNICAL MANUAL  FOR THE
MEASUREMENT OF INDUSTRIAL PROCESS WATER FLOWS

     An extensive  literature survey into the  state-of-the-art of process
water flow measurement indicated that no definitive document or collection
of literature exists that could be used  as a reference  in designing a flow
measurement program.  Since such measurements  are an important part of the
assessment of the effects of  process-water-borne emissions,  it was  recom-
mended that a program be undertaken to develop a technical manual on flow
measurement  and sampling of process  water streams.   A program  for the
                                   -41-

-------
manual development was proposed which included a tentative outline of the
manual format.

TASK 20S - PREPARATION OF A CHAPTER ON FUGITIVE EMISSIONS MEASUREMENTS FOR
A LEVEL 1 ENVIRONMENTAL ASSESSMENT PROCEDURES MANUAL

     A  section on fugitive  emissions  measurements was  prepared for the
revised version of the Level 1 environmental assessment manual being pre-
pared by another EPA contractor.  The section expanded the section of the
original version, prepared by  a  third  contractor  using the TRC inputs of
Task 20C, to  include  applications  of the FAST system and measurements of
waterborne fugitives.

TASK 20T - PREPARATION AND PRESENTATION OF A PAPER ON FUGITIVE EMISSIONS
MEASUREMENTS FOR AN EPA SYMPOSIUM

     A paper entitled "Environmental Assessment Measurement Techniques for
Fugitive Emissions" was prepared and presented at the lERL-Sponsored Sym-
posium on Process Measurements for Environmental  Assessments in Atlanta,
Georgia in February 1978.  The paper was published in the symposium pro-
ceedings.  The abstract of the paper is cited below as published.

          The paper describes the sampling and measurement techniques cur-
     rently being employed or developed to determine the impact of indus-
     trial fugitive emissions on the environment.   Three general sampling
     techniques  for  airborne  fugitive  emissions  and one  for waterborne
     fugitive emissions as stormwater runoff  are  presented  and evaluated
     with respect  to their  inherent accuracies  and  limitations.   Site-
     specific  modifications  of  the general  techniques  used in  recent
     studies at  a variety of industrial  locations are  described and the
     results of  the  measurement  programs reviewed.   Efforts  toward the
     development of a fugitive ambient sampling train for the measurement
     of airborne particulate and organic emissions are summarized.

TASK 20V - PREPARATION AND PRESENTATION  OF  A PAPER ON THE DEVELOPMENT OF
THE FUGITIVE ASSESSMENT SAMPLING TRAIN FOR AN EPA SYMPOSIUM

     A paper entitled "Development  of a Fugitive Assessment Sampling Train
for Particulate and Organic  Emissions" was  prepared  and presented at the
lERL-sponsored Symposium  on  the  Transfer and Utilization  of Particulate
Control Technology in Denver, Colorado in July 1978.   The abstract of the
paper is cited below as published in the symposium program.

          The measurement  of  fugitive emissions from sources that preclude
     the capture  of  emissions  before their  diffusion into  the ambient
     atmosphere poses some unique  problems.  Devices designed  to obtain
     ambient samples for area studies generally do not provide samples of
     particulate matter  large  enough  for meaningful quantification and
     qualification   analyses  in  a  reasonable  sampling period,  and are
     usually completely non-directional in their sampling.
                                  -42-

-------
          The paper describes a program for the development of  a prototype
     portable Fugitive Ambient  Sampling Train  (FAST) designed  to obtain a
     500-milligram particulate  matter sample in an 8-hour  sampling period
     downwind of most  industrial  sources.   The development of the design
     criteria, establishment of operating parameters, selection and design
     of hardware components,  fabrication  and  initial testing of the FAST
     are  described  in detail,  and  the   program  for  the qualification
     testing of the completed unit is outlined.

TASK 06  - ORGANIZATION  OF THE SECOND  SYMPOSIUM ON  FUGITIVE  EMISSIONS:
MEASUREMENT AND CONTROL

     The  organization  of this  symposium  as  a forum  for  the  exchange of
information relative to  fugitive  emissions among  the industrial,  govern-
mental and academic communities  included the selection of a site, arrange-
ment for facilities,  solicitation  of  speakers and session chairmen, prepa-
ration of published  and  mail  announcements,  processing of registrations,
management of the meeting, and publication of proceedings.

     The symposium was conducted  at  the Greenway  Plaza Hotel  in Houston,
Texas on May 23-25, 1977.  It was  attended by  about  200 individuals from
consulting and  academic   (25%), governmental  (30%)  and  industrial  (45%)
organizations.   Eighteen technical papers  were presented and published in
!«S^ -,-fW1™  °n Fugitive Emissions;   Measurement and  Control,  EPA-
600/7-77-148, December 1977.                "	~~	
                                   -43-

-------
                               TECHNICAL REPORT DATA
                         (Please reed Instructions on the reverse before completing)
1. REPORT NO.
 EPA-600/7-79-116
                                                     3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Development of Measurement Techniques for Fugitive
 Emissions from Process and Effluent Streams
                                6. REPORT DATE
                                Mav 1979
                                5. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                     8. PERFORMING ORGANIZATION REPORT NO.
Henry J.  Kolnsberg
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TRC--The Research Corporation of New England
125 Silas Deane Highway
Wethersfield, Connecticut 06109
                                                     10. PROGRAM ELEMENT NO.
                                EHE624A
                                11. CONTRACT/GRANT NO.
                                68-02-2113
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                13. TYPE OF REPORT AND PERIOD COVERED
                                Final; 12/75 -12/78	
                                14. SPONSORING AGENCY CODE
                                 EPA/600/13
is. SUPPLEMENTARY NOTES JERL-RTP project officer is D. Bruce Harris, MD-62, 919/541-
2557.
16. ABSTRACT
              repOr|. summarizes work completed in this continuing program of eval-
uation, development, testing, and adaptation of existing and proposed measurement
techniques for air and waterborne industrial process fugitive emissions.  Results  of
five major research and development tasks are presented: (1) a measurement and
modeling program for waterborne emissions from coal-fired utility boilers; (2) a
study of the applicability of remote sensing methods to the measurement  of airborne
fugitive emissions; (3) a survey and measurement program to assess the environ-
mental problems caused by runoff from integrated iron and steel mills; (4) the design
and development of a fugitive assessment sampling train for particulate and organic
vapor emissions; and (5) an evaluation of techniques for the determination of large
hood captive efficiencies. Summaries of service area tasks related to fugitive emis-
sions measurements include: methodology developments for emissions measurement
programs at coke quenching towers , coal cleaning plants ,  iron ore loading opera-
tions,  and an open hearth shop's roof monitors; and reviews of proposed  sampling
programs and methods for fugitive emissions from mining operations , petroleum
refineries, and  coal cleaning operations. Abstracts of three papers on fugitive emis-
sions topics presented at EPA symposia are included.
17.
                            KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                    b.lDENTIFIERS/OPEN ENDED TERMS
                        c. COSATI Field/Group
Pollution
Measurement
Industrial processes
Leakage
Mathematical Models
Coal
Combustion
Iron and Steel Indus-
 try
Dust
Electric Power Plants Organic Compounds
Petroleum Refining
Pollution Control
Stationary Sources
Fugitive Emissions
Particulate
Coal Cleaning
13B
14B
13H

12A
10B
21D
21B

11F

07C
13. DISTRIBUTION STATEMENT
 Unlimited
                                         19. SECURITY CLASS (This Report)
                                          Unclassified
                                            21. NO. OF PAGES
                                                47
                    20. SECURITY CLASS (This page)
                    Unclassified
                                            22. PRICE
EPA Form 2220-1 (9-73)

-------