Measurement
   of Toxic and
   Related Air
    Pollutants
       May 7-9,1996
     Sheraton Imperial Hotel
 Research Triangle Park, North Carolina
    A Symposium Sponsored by:
                *•
Air & Waste Management Association
      AIR & WASTE MANAGEMENT
      ASSOCIATION
          SlNCt 1907
U.S. Environmental Protection Agency
 T

II

-------
   Measurement of Toxic and

     Related Air Pollutants
             May 7-9,1996
         Sheraton Imperial Hotel
   Research Triangle Park, North Carolina
          ABSTRACTS
       A symposium sponsored by:
    Air & Waste Management Association
                and
    U.S. Environmental Protection Agency
AIR & WASTE MANAGEMENT
ASSOCIATION
       •
      SlNCt 1907

-------
                                                               Abstracts
                      TABLE OF CONTENTS


                                Session 1
                             Semivolatiles
       Chair: Joachim D. Pleil, U.S EPA, Research Triangle Park, NC  -

Comparison of On-line and OIT-Linc SFE GC/MS Methods for the
Determination of SVOCs from Tcnax-GC Filled Cartridges                      I

Phase Distribution of Formic Acid                                          I

Phase Distribution of PAH Under Different Atmospheric Conditions              2

Fine Paniculate Nitro and N-Nitroso Organic Compounds in
the Atmosphere                                                         3

                                Session 2
                                  VOC
    Cochairs: Larry D. Johnson, U.S. EPA, Research Triangle Park, NC;
        William A. McClenny, U.S. EPA, Research Triangle Park, NC;
        and Robert G. Fuerst, U.S. EPA, Research Triangle Park, NC

New York Slate Department of Environmental Conservation PAMS
Monitoring System Using A 3-Column, Double Identification GC                5

Portative Device for Long Term Integral!vc Personal Sampling
of VOCs and Other Gaseous Air Pollutants                                   5

A New Method  lor Measuring  Vapor Emissions                               6

Overview of Prototype EPA  PAMS Data Analysis Workshop
Endocrine Disrupter Exposure  Research                                      7

Data Validation of PAMS Auto-GC VOC Data: Lessons Learned                 8

Analysis of VOC Data in Support of Pollutant Transport Studies
in Shasta County, California                                               8

VOCs from Marker Pens in Exposure Scenarios                               9

Emission of VOCs and Polynuclcar Aromatic Compounds
from Asphalts and Road Paving Operations                                   10

Monitoring for VOCs In Air - Status Report                                   I I
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               *"

-------
 Abstracts
 Comparison of Acetone Concentrations in Ambient Air Using EPA
 Methods TO-14 and TO-II                                                 II

 Recent Advances in the Analysis of Oxygenated Hydrocarbons                   12

 Analysis of Available Ambient and Stationary Source VOC HAP
 Sampling Results to Verify Emission Rale Estimates for Sources
 in Nogales, Sonora, Mexico                                                 12

 A Study of "Holding Times" for Sulfur Compounds in Restek's
 SilcoCan Canisters                                                        13

 Measurements of Hydrocarbons and Reduced Sulfur
 Compounds Emitted from a Waslewater Treatment Pond                         14
 Extractive FTIR Technology - Applications In Combustion
 Source Measurements                                                      14

                                 Session 3
              Natural Emission of Oxidant Precursors
         Cochairs:  Viney P. Aneja, NC State University, Raleigh, NC;
        and Bruce W. Gay, Jr., U.S.  EPA, Research Triangle Park, NC

 Project NOVA (Natural emissions of Oxidanl precursors: Validation
 ol techniques and Assessment) Overview                                      17

 Natural Emissions of Oxidant Precursors: Validation Of Techniques
 and Assessment (Project NOVA 1995) - Site Characleri/.ation                    18

 Biogenic Soil Emissions of Nitric Oxide and Nitrous Oxide Using a
 Closed Static Chamber During the NOVA-95 Experiment                        19

 Measurements of Nitric Oxide Flux From Intensively Managed
 Agricultural Soil in the Coastal Plain of North Carolina                          20

 Controlled Laboratory Studies of NO Flux Into the Lower Levels
 of the Troposphere Under Widely Varying Soil Conditions                       20

 Eddy Correlation Measurements of NO. NO2, and O3 Fluxes
 Nitric Oxide Fluxes Observed Over A Fertilised Corn Field                      21

 Micrometeorology and Flux-Gradient Relation ol'NO-NO2-O3
 System Near the Surface                                                    22

The Relationship Between the Surface Exchange Fluxes of Trace Species
 Determined by Enclosure and by Micrometeorological Techniques                 22

 Flux Measurements of Volatile Organic Compounds Over a Corn Field             23

iv                Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                   Abstracts
 Measurement of Soil Nitrous Oxide Flux from an Intensively Managed
 Agricultural Row Crop in the Lower Coastal Plain Region of North Carolina       23

 Biogenic Nitric Oxide Source Strengths                                       24

 Atmospheric Pollutant Deposition Over a South Central
 North Carolina Cornfield                                                    24

 Contributions of Meteorology on O/one Behavior in Urban  North Carolina         25

 A Climatology of Regional Ozone: Meteorological Effects on O/onc
 Excccdences in the Southeast U.S.                                            26

 Regional Analysis of Non-Methane Hydrocarbons in the Southeastern
 United States                                                       '       26

                                 Session 4
                         Aerosol Measurements
          Chair: Petros Koutrakis, Harvard University, Harvard, MA

 Attribution of College Park Aerosol Paniculate Components by
 Si/.c and Source                                                            29

 Resolution of Aerosol Constituents by Si/.e and Source in the
 Urban Plume Over Lake Michigan                                            29

 Si/.c Distribution of IR-Tagged  Soot from Diesel Sanitation Trucks                30

 Health Risks and Particle Monitoring: New Technologies to Meet
 Emerging Risk Management Needs                                            30

 Harvard/EPA Ambient Fine Particle Concentrator for Conducting
 Human Exposures                                                          31

Characteristics of Ambient Aerosols in Washington D.C. as  Measured
by Continuous and Time-Integrated Sampling Methods                          32

The Role of Transition Metals in Lung Injury After Air Pollution
Particle Exposure                                                           33

Ascssmcnt of Air Toxic Emissions from a Pilot-Scale Combustion Unit            33

Diurnal Variations in Paniculate Mass and Total Sulfatc Concentrations
at a Site in Metropolitan Philadelphia                                          34

A Field Assessment of the Effect of Volatile Constituents on the
Measurement of Ambient Paniculate Matter Mass Concentration                  35

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                  »

-------
 Abstracts
 Pulmonary Toxicily of Urban Air Paniculate Matter (PM)                       35

 Development ol'ihc New Federal Reference Method for Fine Particles:
 Current Methodology                                                      36

                                  Session 5
                                  Canisters
             Chair: Eric Winegar, Air Toxics Limited, Folsom, CA

 Passive Flow Controllers: Revisited -  Per EPA Method TO-15                   39

 Fourteen-Day Stability of 27 Polar VOCs In Summa® and
 Silocan™ Canisters                                                       39

 Development of an Intelligent Canister/Cartridge Sampler for the
 Collection of O/.one or Air Toxics                                            40

 The Use of Fused Silica Lined Canisters for Difficult Ambient Air Samples        40

 PPBV Determination of Reduced Sulfur Compounds using Glass
 Lined Canisters, Deactivated Tedlar Bags and Sorbent  Tubes:
 A Study of the Critical Parameters                                            41

 A Through-The-Prohe Toxic Audit Study to Determine the  Effect of
 Moisture on  Volatile Organic Compound Recovery Rates in
 Stainless Steel Canisters                                                    41

 Meeting the Technical Acceptance Criteria for TO I5 Methodology
 Using a New Canister Autosampler                                           42

 Improved Methodology for Analysis of Ambient Level Reactive
 Sulfur Compounds                                                         43

 Innovative Flow Controller for Time Integrated Passive Sampling
 using SUMMA Canisters: Design, Testing and Applications                     43

                              Poster Session

 Low-Level Analysis of Cr+6 in Air by Ion Chromatograpliy                      45

The Significance of Hydrocarbons in the C10-C20 Range to Mass
Contribution and O/.one Forming Potential During the COAST Study             45

The Application of High Speed Gas Chromatography to Air Analysis              46

Photochemical Assessment Monitoring: Overview and  Data
Analysis Activities                                                         46

vi        '         Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                   Abstracts
 A Novel Method of Measuring Ambient Aerosol Aspiration Efficiencies
 Using Phase Dopplcr Anemometer                                            47

 Methods Development for Clean Air Act Amendment Ha/.ardous Air
 Pollutants from Stationary Sources: Methanol and Isocyanates                    48

 VOST Charcoal Specification Study                                          49

 Method Evaluation Study: The Application of SemiVOST to the
 Nonhulogcnatcd Semivolatilc Organic Compounds from the Clean
 Air Act  Amendments                                                       50

 Development of a Source Test Method for Measurement of Mercury Species        5 I

 FT-IR Transmission  Spcctroscopy for Quantitation of Ammonium
 Bisulfatc in Fine-Particulate Mailer Collected on Teflon® Filters                  52

 Organic Compound Delivery Device                                          52

 Marine Crustaceans as  Indicators of Esluarine Health - The Atlantic
 Blue Crab (Callinectcs  sapidus)                                              53

 Multimedia Analysis by Pyrolysis/Gas Chromalography/Mass Spcclromelry         53

 Application of Thcrmodynamic Parameters to Gas Chromatographic
 Methods Development                                                       54

 Expanding the Use of Passive Sampling  Devices (PSDs)                         55

 Research and Development of a Stationary Source Method
 for Phosgene                                                              55

 Applications for Using  Exhaled Human Breath as an Exposure
 Assessment Tool  for  Volatile Organic Compounds                               56

 Solid Adsorbent Sampling tor VOC - Application in Ostrava, Czech
 Republic Open  Path Monitoring for Trace Gases                                57

 Monitoring for  VOCs in Air - Status Report                                    58

 Methods for  Measurement of Dislodgcable Pesticide Residues in and
 Around the Home                                                          58

 Source Sampling  Particulate Matter  Research                                   60

Performance Characterizations of Laboratory and Field Technologies for
Quantitation  of Lead  in Paints, Dusts, and Soils                                 60
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
Abstracts
Evaluation of Selected Multiadsorhent Tubes for Monitoring TO-14
Volatile Organic Compounds in Ambient Air                                 62

                                Session 6
                  Atmospheric  Mercury Monitoring
     Cuchairs: Robert K. Stevens, U.S. EPA, Research Triangle Park, NC;
            and Thomas D. Atkeson, Florida DEP, Talahassee, FL

Overview of the South Florida Atmospheric Mercury Monitoring
Pilot Study                                                              63

Preliminary Phase of Vapor Phase Mercury Measurements Made
During the SUAMP Project                                               63

Atmospheric Deposition of Mercury in  South Florida: Preliminary
Results from the SUAMP Project                                           64

Dry Deposition of Mercury in South Florida: Preliminary Results                65

Methods to Measure Atmospheric Mercury in South Florida                     65

Atmospheric Transport, Transformation, and Deposition of
Mercury in Florida                                                       66

Meteorological Flow Regimes Influencing  Atmospheric
Concentrations and Deposition During the  1995 SUAMP                       67

Source Characteri/.ation in Support of Modeling the Transport
of Mercury Emissions in South Florida                                       67

Receptor Modeling of Atmospheric  Mercury in South Florida:
Preliminary Results                                                       68

DOAS Measurements of Elemental Mercury and Sulfur Dioxide
from a Resource Recovery Incinerator                                       69

                                Session  7
                     Ambient Air Measurements
        Chair: Stiresh Santanain,  Sterns and Wheler, Cazenovia, NY

Effect of Acidity on the Sampling and Analysis of Carhonyls Using DNPH
Derivati/.alion Method                                                    71

Airborne Hexavalent Chromium in Southwestern Ontario                       71

Monitoring Reformulated Gasoline in Milwaukee, Wisconsin                    72

VH'I                Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                 Abstracts
Volatilization and Dispersion Measurements tor Airborne Methyl
Bromide Downwind of a Treated Agricultural Field                             72

Ambient Trends of Benzene in California From 1990
Through 1995                                                            73

Air Pollution from Cookstoves in Developing Countries:
Results from A Field Study in China                                         73

Size Spectra and Sources of Elemental Constituents of the
Baltimore Urban Plume                                                    74

                                Session 8
                 Tropospheric Ozone Measurement
            Chair: Carmo Fernandez, Arizona DEQ, Phoenix, AZ

Voluntary Emissions Reduction Efforts During the Summer of
1995 in Wisconsin                                                        77

Paso del Norte Pilot Border Study of O/.one Precursors and Air Toxics             77

Overview of the ISO 14000 Environmental Management Standards                78

                                Session 9
                           Particle Exposure
      Chair: Robert M. Burton, U.S. EPA, Research Triangle Park, NC

Characterization of Episodic Air Pollution Study                               79

Policy and Science on PM in the Netherlands, Bridging the Gap                  HO

Influence of Weather on Aerosol Exposure in Selected, Highly
Populated Metropolitan Areas of the United Stales                              81

The Role of Total Human Exposure Measurements of Paniculate
Matter in Community Epidemiological Studies                                 82

A Synoptic Climatology of Fine Particle Acid Aerosols                          82

Assessment  ofthe Temporal Relationship Between Summertime
Ultra-Fine Paniculate Count Concentration With PM in Washington D.C.          83

The Relationship Among TSP, PM10, PM2.5,  and Inorganic Conslituanls
ol Atmospheric Paniculate Matter at Multiple Canadian Locations                 84
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
Abstracts
Relationships Between Daily CoH. Visihility Measurements and
Fine Mass Concentrations in Philadelphia, PA                                  84

Atmospheric Mass and Melal Si/.e Distributions Measured Around
Lake Michigan                                                             S5

Quality Control and Reference Equivalency of PM Monitoring Systems            85

Application of Near-Field Optical Techniques to Ambient Paniculate
Matter: Physical and Chemical Structures                                      86

Carbon Nanotuhes and Buckyhalls: Structure. Mechanics, and Air
Pollutant Potential        .                                                  87

Development and Evaluation of a Condensational Growth/Virtual
Impaction System for Internal Separation of Ultrafinc Particles                    87

Development of a Continuous Mass Monitor by Measuring Pressure
 Drop Across Filters                                                        88

                                 Session 10
                     Integrated Monitoring Study
       Chair: Paul Solomon, Pacific Gas & Electric Co., San Ramon, CA

1995 Integrated Monitoring Study: Study Objectives
and Design                                                                91

1995 Integrated Monitoring Study: Fog Measurements in the Southern
Valley - Preliminary Results                                                 92

1995 Integrated Monitoring Study: Vertical Fog Measurements in the
Northern Valley - Preliminary Results                                         93

1995 Integrated Monitoring Study: Spatial Variability of PM 10 and Related
Precursors During the Fall Saturation Study - Preliminary Results                 93

1995 Integrated Monitoring Study: Temporal Variability of PM 10
and PM2.5 Mass During the Winter Field Program - Preliminary Results           94

1995 Integrated Monitoring Study: Surface and Upper Air
Meteorological Measurements - Preliminary Results                             94

1995 Integrated Monitoring Study: Low Wind Speed Study -
Preliminary Results                                                         95

1995 Integrated Monitoring Study: Emissions Activity Measurements
During the Fall and Winter Components - Preliminary Results                    96

x                  Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                Abstracts
1995 Integrated Monitoring Study: Quality Assurance Methods and Results        96

                               Session 11
                               Indoor Air
       Chair: Ross Highsmith, U.S. EPA, Research Triangle Park, NC

Pattern and Sources of Polynuclear Aromatic Hydrocarbons and
Their Derivatives in Indoor Air Real-Time Low-Level Pollution
Monitoring for IAQ Applications                                           99

Emission Rales of Formaldehyde and Other Carhonyls from Consumer
and Industrial Products Found in California Homes                            100

Testing of Polyurcthane Products for Emission of Toluene Diisocyanate
to Indoor Air                                                            101

Transfer Efficiency of Pesticides from Carpel to Saliva-Moistened Hands          101

Field-Emplaeeablc Air Samplers for Automated Ambient, Site, and IAQ Use        102

Multi-Media Exposure Study: Rationale and Design                            102

                               Session 12
                   Source Sampling Methodology
    Cochairs: Merrill D. Jackson, U.S. EPA, Research Triangle Park, NC;
     and Raymond G. Merrill, Radian Corp., Research Triangle Park, NC

Instrumentation for Continuous On-Line Monitoring ol Non-Methane
Organic Carbon in Air Emissions                                           105

Modification of Method 25 for Low VOC Sources  with High  CO2
and Moisture Levels                                                       105

A Test Method for the Measurement of Arylamines in Stationary
Source Emissions                                                         106

Method 301 Evaluation of a Technique for Collection and Mcasureinenl
of Ammonia in Stationary Source Emissions                                  106

Development of a Manual  Method for Sampling and Analysis of High
Levels of Phenolic Compounds                                             107

Development of an Optical Method for Sampling and Analysis of
High Levels of Phenolic Compounds                                         108
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                xi

-------
Abstracts
Generic FT1R Sampling Method lor Measuring Ha/ardous Air
Pollulanl Emissions from Industrial Point Sources                              I09

Application of Extractive FTIR (Fourier Transform Infrared) Speclroscopy
for Speciated Measurement of Uncontrolled Emissions at an Asphalt
Coaling Process: Sampling and Analysis Challenges at a Complex
Emission Source                                                          109

Method Evaluation Study: The Application of VOST to the
Nonhalogenated Volatile Organic Compounds from the Clean
Air Act Amendments                                                      110

Method Evaluation Study: The Application of SemiVOST to the
Nonhalogenaled Semivolalile Organic Compounds from the
Clean Air Act Amendments                                                1 I I

Stack Sampling Methods for Halogens and Halogen Acids                       112

Development of Calibration Procedures  for Toxic Metal Aerosols
for Continuous Emission Monitoring Systems                                 112

A Comparison of Preconcentralion Procedures for GC/MS Analysis
of 1990CAAA Volatile Ha/ardous Air Pollutants                              113

Dynamic Spiking Studies Using the DNPH Sampling Train                      I 14

Imaging Mass Spcctromctry of Paniculate-Associated Polynuclcar
Aromatic Hydrocarbons                                                    115

Total Organics Guidance Manual For Stationary Source Emissions                115

                                Session 13
                           Quality Assurance
         Chair: Shri Kulkarni, Research Triangle Institute, RTP, NC

NARSTO-NE Hydrocarbon Laboratory Comparison                 ,           117

Accuracy Considerations of Electrochemical  NOX Analy/ers                     I 17

Further Analyses of Wisconsin's Remote Vehicle Emissions Sensing
(RVES) Feasibility Studies                                                 118

Quality Control Guidance for Sampling and Analysis of Air Toxics               118

Effects of QA/QC Procedure Changes on Method 25 Analysis and
50 ppmC Method Detection Limit                                           119
xii                Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                  Abstracts


Uncertainty Analysis tor Open-Path Remote Sensing of Fugitive Emissions         119

Data Quality Objectives for the Yeast Manufacturing Industry                    1 20

Overview of the ISO 14000 Environmental Management Standards               121

The National Ambient Air Quality Standards and Quality Assurance               121

A Report on the Activities of the Photochemical Assessment Monitoring
Stations (Pains) Quality Assurance Work Group                                122

The National Ambient Air Quality Standards and Quality Assurance               123

Performance Comparison of Field-Deployable Gas Chromatographs with
Canister/TO 14 Analyses                                                    124

                                Session 14
                                  General
            Cochairs: R.K.M. Jayanty, Research Triangle Institute,
                        Research Triangle Park, NC;
                  Donald Adams,  Consultant, Pullman, WA;
             and William Gutknecht, Research Triangle Institute,
                        Research Triangle Park, NC

A Comparison of Dry Deposition Modeled from Si/.e Distribution Dala
and Measured with a Knife-edge Surrogate Surface                             125

Can Selected RADM Simulations be Aggregated to Estimate Annual
Concentrations of Fine Parliculate Matter'.'                                    125

Comparison of Risk Management Dispersion Modeling Methodologies            126

The Analysis of Ambient Air Samples  for Polar Volatile Organic Compounds       127

A Wind Tunnel Investigation of the Flow Structure within a
Dense Gas Plume                                                          127

Performance Characteristics of Adsorbents Ulili/ed in Sample
Preparation Modes of Operation                                              128

Comparative Mutagenicity of a Standard Reference Material in the
Salmonella Assay                                                          129

Mutagenic Activity of Raleigh Air Samples al Three Different Elevations
Before, During and After Hurricane Gordon                                    129
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               xiii

-------
Abstracts
Update ol'EPA's Compendia of Toxic Organic and Inorganic Sampling
and Analytical Methods for Ambient Air                                      130

Estimating the Risk From Exposure to Wind-Borne Surface Soil Emissions         131

Sampling & Analysis of Various Wastewalcr Streams at a Petrochemical
Facility for Compliance with the Ha/.ardous Organic  NESHAPs (HON) Rule        132

Perimeter Air Sampling Strategy for Soil Remediation at a Superfund Site          132

GC/MS Analysis of Volatile Organic Compounds During Site Remediation         133

Evaluation of Sludge Characteristics and Source Emissions from
Municipal Sewage Sludge Incinerators                                        134

                                Session 15
                            Human Exposure
  Cochairs: M. Rodon-Naveira, U.S. EPA, Research Triangle Park, NC; and
  Edo D. Pellizzari, Research Triangle Institute, Research Triangle Park, NC

Using Environmental Exposure Data to Set Environmental Health
Priorities/Interventions in a Community Setting.                                135

Evaluation of Field Methods for Estimating Exposure of Children in
Low-Income Families to Polycyclic Aromatic Hydrocarbons                      135

Measuring and Modeling Indoor-Outdoor Pollutants  in a Residence
Using Real-Time Methods                                                  136

Demonstration of Real-Time Measurements of PAH and CO to Estimate
In-Vehicle Exposure and Identify Sources                                     137

Cooking as a Source of Fine Particles: Experiments in a Townhouse               137

Biases in Personal  Aerosol Air Exposure Sampling                              138

Strategies for the Development and Application of Screening Techniques for
Human Exposure and Environmental Assessment Surveys                        138

The Exposure Potential Matrix: A Dietary Model Using Extant Databases
to Evaluate Exposure to Chemical Residues                                    139

Quantitative  Analysis of Road and Carpel Dust on Shoes                         140

Conducting Child Behavior Studies to Identify and Characteri/.c Pathway
and Route of Exposure Through Direct Observation and Videotaping              141
xiv                Measurements of Toxic and Related Air Pollutants, May 7-9, 7996

-------
                                                                  Abstracts


The National Human Activity Patlern Survey (NHAPS) Applications
for Exposure Assessment                                                    142

Application of Human Activity and Microcnvironrnental Concentration
Data Models in THERdbASE to Exposure Assessment                          143

Applying Results from Ecological and Human Exposure Monitoring Studies
for Reducing Risks and Prioritizing Future Environmental Health Research         143

Evaluating Food Contamination Scenarios for Dietary Exposure Studies           144

Measurement of Street-Level Air Pollutants in Bombay City                      145
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                xv

-------
                                                            Abstracts


                             Session  1
                           Semivolatiles
       Chair: Joachim D. Pleil, U.S EPA, Research Triangle Park, NC

Comparison of On-line and Off-Line SFE GC/MS Methods for
the Determination of SVOCs from Tenax-GC Filled Cartridges
James R. Bowyer
ManTech Environmental Technology, Inc., 2  Triangle Drive, Research Triangle
Park, NC 27709
Joachim D. Pleil
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

    Supercritical fluid extraction (SFE) has become the main alternative to
Soxhlet extraction of analytcs from environmental samples primarily because of
the shorter turnaround time in analysis, reduced solvent exposure and disposal
problems, and comparable extraction efficiencies. SFE has been shown to be
comparable to Soxhlet extraction for extracting analytes from a variety of
substances including carpet, soils, and air sampling sorbents. The objective of
this study was to compare two means of introducing SFE extracts, taken from I-
mL Tcnax-lilled stainless steel cartridges, to a GC/MS interlace and an off-line
SFE GC/MS system for determination of semivolatilc organic compounds
(SVOCs) as both spiked samples and ambient air samples. Ambient air samples
were obtained by using  a 6 L canister as a vacuum source, thereby collecting a
given volume of sample in a very simple and easily  transportable manner.
Results indicated that the methods were equally good at extracting various
compounds from the cartridges, but the on-line method had better recovery for
the more volatile SVOCs. Other advantages for the  on-line method included less
sample handling and indications of lower limits of detection that the off-line
method.
    The information in  this document has been funded wholly or in part by the
United States Environmental Protection Agency under Contract 68-D5-0049 to
ManTech Environmental Technology, Inc. It has  been subjected to Agency
review and approved for publication. Mention of trade names does not constitute
endorsement or recommendation for use.

Phase Distribution of Formic Acid
Daniel Driscoll
Atmospheric Processes  Research Division, National Exposure Research
Laboratory, U.S. EPA, Research Triangle Park, NC 27711

    Formic acid is produced in the aqueous-phase ol the atmosphere, in part,

Measurements ol Toxic and Related Air Pollutants, May 7-9, 1996                1

-------
Abstracts


through ihc photo-oxidation of formaldehyde by hydroxyl radicals. One of the
products of this photorcaction is the formate union. As acidic hydrometcors such
as fog, dew droplets, and clouds evaporate, the formate unions can rccomhinc
with hydrogen ions to produce volatile formic acid. The dispersion of formic
acid at the aqueous/gas interface is controlled by the reverse deposition rate
measured in nanomolcs/cnr/min.
     Experiments arc performed to measure the reverse deposition rale of formic-
acid produced by the air drying of acidified aqueous solutions of formate anion.
Ultra pure air, maintained at constant temperature and relative humidity, Hows
over the liquid surface of acidified formate droplets for varying periods of time.
The solutions are contained in U-shaped FEP Teflon tubes which measure 23
centimeters in length and  1.3 centimeters in diameter. The tubes are connected to
a scries of two glass bubblers containing 0.01  N NaOH. Ion Chromatography
(1C) is used to determine the loss of formate anion occurring in the droplets.
Formic acid is measured by analy/.ing the formate concentration of the 0.01 N
NaOH solution at the conclusion of each experiment. The analytical results  are
used to calculate the amount of formate contained in the applied formate
solution, the droplet residue, and in both bubblers.

Phase Distribution of PAH Under Different Atmospheric
Conditions
By H. Li and R. Otson
Health Canada, EHC, Tunney's Pasture, Ottawa, Ont., Canada, KIA OL2
P. Fellin
BOVAR Environmental, 2 Tippctt Road, Downsvicw, Ont., Canada, M3H 2V2

    The vapour/particle phase distribution and particle si/.c of airborne PAH arc
important  for determination of the potential health ha/.ard posed by exposure to
the PAH. The behaviour of 12 PAH in a dynamic, controlled test  atmosphere
produced by dispersion of fortified woodsmokc particles (> 80 % of mass
median diameter < 3.3 mm) in a special apparatus was examined at different
conditions of temperature (6 - 32°C), relative humidity (23 - 71 %), and particle
concentrations (140 -  1000 mg/m3). Seven different active sampling trains
containing filters followed by porous polymer sorbcnts, and operated at different
air sampling rates (0.7 - 33 L/min), as well as two types of passive samplers were
used to collect the airborne vapour and particle phase PAH. The PAH were
extracted, and the extracts were anuly/cd by GC-MS. Typically, less than 40%  of
naphthalene, fluorenc and phcnanthrcne, and more than 90% of
dibcn/.o(a,h)anthracenc and ben/o(g,h,i)pcrylcnc, was found on the filters. The
influence of humidity, temperature, particle concentration, and sampling rate on
the apparent phase distribution of the PAH, and the relationships between some
of these parameters will be discussed.


2               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


Fine Participate Nitro and N-Nitroso Organic Compounds in
the Atmosphere
Yiming Ding, Milton L. Lee and Delbert J. Eatough
Department of Chemistry and Biochemistry, Brigham Young University,
Provo, Utah 84602

    Photochemistry between gas phase organic compounds, NOx and ozone is
expected to result in the formation of nitrogen oxide containing compounds of
toxicological importance. Classes of compounds which may form include nilro
and N-nitroso substituted organic compounds. Many of these compounds will be
labile, semi-volatile organic compounds in equilibrium between the gas and
paniculate phases in the atmosphere. The phase distribution of these potentially
toxic semi-volatile organic compounds may be determined using diffusion
dcnudcr sampling technology. The total concentration of N-nitroso compounds
in the collected samples is determined using N-nilroso specific denilrosation
reactions followed by detection of the NO with a TEA. Specific nitro and N-
nitroso compounds are detected using supercritical fluid chromatography, SFC,
coupled to the TEA using nitro- and nitroso-spccific chcmiluminescence
detection. These analytical techniques for the sampling and determination of
total N-nilroso material, and the separation  and determination of specific N-
nitroso and nilro compounds have been used for the characterization of these
compounds in samples collected in Provo, Utah. The results  indicate that the
majority ol the N-nilroso and nitro organic compounds present in fine participate
matter in the urban area studied arc semi-volatile organic compounds which are
easily lost from particles during sampling. Furthermore, the concentrations  of
these fine paniculate compounds  are comparable to the concentrations of gas
phase nitro and N-nitroso organic species. Detailed analyses of nitrogen, i.ilro-
and nilroso-organic semi-volatile compounds can be expected to improve our
understanding of the etiology of observed health effects associated with exposure
to ambient line particles.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                           Abstracts
                             Session 2
                               voc
    Cochairs: Larry D. Johnson, U.S. EPA, Research Triangle Park, NC;
       William A. McClenny, U.S. EPA, Research Triangle Park, NC;
       and Robert G. Fuerst, U.S. EPA, Research Triangle Park, NC

New York State Department of Environmental Conservation
PAMS Monitoring System Using A 3-Column, Double
Identification GC
Paul M. Sierzenga, Garry A. Boynton, and Brian J. Aho
New York State Dcpt. of Environmental Conservation, Division of Air
Resources, Bureau of Air Quality Surveillance, 80 Wolf Road, Room 406,
Albany, NY 12233-3256

    New  York State conducted its 1995 PAMS measurements for New York
City in the Bronx (New York Botanical Gardens) using a Perkin-Elmer ATD-
400 Thermal Dcsorhtion Unit with an 8000 Scries Gas Chromatograph. Data
was collected in New York City, transferred to Albany, and processed in the
central ollice. The system was checked on a daily basis from Albany to verify
operation. Quality assurance for the system was provided by collecting a one
hour collocated canister sample every day with analysis  performed in Albany
using a GC-MS system. The  normal configuration for the Perkin-Elmcr system
uses only  two columns, a Plot column to separate low weight hydrocarbons (C2-
C5) and a BP-1  column to separate higher weight hydrocarbons (C6-C10). New
York's custom designed system uses a third RTX-5 column to provide a second
identification and quantitalion of most target hydrocarbons. Analysis of the 1995
data provides insight into the usefulness of the third column in correcting both
misidenlified peaks and over-quantification of compound concentrations found
using the standard configuration alone. The  1995 data set also allows an exami-
nation of the increased  complexity versus the benefits of the system modifica-
tion.

Portative Device for Long  Term Integrative Personal Sampling
of VOCs and Other Gaseous Air Pollutants
Philippe Simon and Jean-Pierre  Farant
Environmental Laboratories, McGill University, 3450 University, rm 23,
Montreal, Quebec,  Canada, H3A 2A7

    Apart from passive dosimeters based on the principles of diffusion on
sorbent materials, VOCs can be collected passively using evacuated sampling
vessels such as SUMMA canisters. The basis of this last methodology was never

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                5

-------
Abstracts


applied for personal sampling because of physical limitations in si/.c and How
controlling capacities. From the development of an innovative mechanical
controller which can extend (lie sampling duration and can theoretically achieve
any How rale, we have recently designed a portative device for passive personal
monitoring of VOCs and others gaseous air pollutants such as CO, CO, or CH4.
Prototypes were assembled and tested in the laboratory. They were made with a
rectangular stainless steel vessel of 150 ml equipped with a pressure gauge and a
valve, and connected to our novel How controller. For industrial hygiene
applications, sampling time was defined at 40 hours. The personal sampler was
able to maintain a constant How rate during 5 consecutive periods of 8 hours
over one week. To verify the integrity of long term passive samples using this
portative device, prototypes were challenged with a 100 ppb(v) gas mixture of
40 air toxics. Analysis was performed on a GC/MS using sorbcnt concentration
and drypurge techniques where 10 ml of samples were withdrawn from the
vessels with a gas light syringe and injected into the system. Results demon-
strated the reliability of the personal passive portative device and its applicability
for workplace monitoring. Compare to passive sorbcnt badges, the performance
of this new device is less affected by  temperature and by humidity. The portative
passive monitor can be applied to sample workers or commuters in outdoor and/
or indoor environments. Considering the possibilities offered by the How
controller which is the key part of the device, sampling time can be adapted for
any situations and mean exposure can be more easily obtained  for the  benefit of
air quality studies.

A New Method for Measuring Vapor Emissions
Kent S. Knaebel and Linda D. Yeoman
Adsorption Research, Inc., 6IK5 Shamrock Court, Suite D, Dublin, Ohio 43016

    A new approach for measuring point-source emissions of volatile organic
compounds, acidic vapors, and other species is presented. The basic principle is
that the amount emitted is determined directly, by mass, and is cumulative rather
than intermittent. As a result, wide fluctuations of concentration and erratic  How
behavior are accommodated without  affecting the accuracy of the measured
emission rate. The vapors are retained by a sorbcnt, such as activated  carbon, ion
exchange resin, or /colitc, or a combination of those.
    Validation  tests have been conducted in which a known quantity of vapor in
a carrier  gas was admitted to the test  unit, and that was compared with the
amount measured. The vapor comprised a single VOC, a mixture of VOCs, or a
mixture of a VOC with water. Conditions studied were: the compound or mixture
of compounds, concentration, carrier gas. How  rate, and adsorbent.  In some tests
the VOC was admitted intermittently. The VOCs included: hcxanc, acetone,
toluene, vinyl acetate, and 1,1,1  trichlorocthanc. The average absolute error of
                 Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
the delivered and measured VOC emission rates was 6.8% and the standard
deviation was 3.4%.

Overview of Prototype EPA PAMS Data Analysis Workshop
Paul T. Roberts, Hilary H. Main, and Marcelo E. Korc
Sonoma Teehnology Inc., 5510 Sky lane Blvd, Suite 101,
Santa Rosa, CA 95403-1083
James B. Hemby and Richard D. Scheffe
Emissions, Monitoring, and Analysis Division (MD-14), U.S. Environmental
Protection Agertcy, Research Triangle Park, NC 27711

    In December 1995, the EPA sponsored a prototype PAMS data analysis
workshop. The objectives of this workshop were to present, explain, and discuss
various methods,  procedures, and tools for use in analy/.ing PAMS and similar
acromctric data; to provide a forum for nationwide communication and informa-
tion transfer on the analysis of PAMS data (and that of supplemental air quality
monitoring campaigns and/or field studies); to assist state and local agencies in
the use of these methods, procedures, and tools in the analysis of PAMS and
similar data sets; and to contribute to the general body of knowledge and
literature on air quality analysis through demonstration of case studies and
examples. The topics discussed in the workshop included: data validation, VOC
spatial and temporal issues, o/.one accumulation potential, biogenic emissions,
use of indicators to assess NO and VOC limitations, comparisons of UAM
results with ambient data, emission inventory evaluations, receptor modeling,
trends analyses, 3-D analyses, and case studies. This paper will provide an
overview of the workshop and provide key examples of featured topics.

Endocrine Disruptor Exposure Research
D. A.  Vallero
U.S. EPA, National Exposure Research Laboratory

    Scientists continue to enhance techniques to estimate exposure of human
populations and ecosystems to environmental contaminants. The U.S. Environ-
mental Protection Agency's scientists are studying the exposure of humans and
ecosystems to chemical substances that are persistent, can bioaccumulate, and
can biomagnify in the food chain. A growing number of studies suggest that
several persistent  organic pollutants (POPs) and some organic forms of heavy
metals appear to mimic or to disrupt hormonal mechanisms in humans and
wildlife. EPA is exploring methods and models to measure and to predict
exposure to these  substances. The presentation will address some of the ap-
proaches the Agency may take to conduct exposure research of endocrine
disrupting compounds within the Agency's risk assessment framework. Persis-
tent organics, such as those listed recently by the Inter-Organizational Program

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                7

-------
Abstracts


for the Sound Management of Chemicals, should he characterized physically and
chemically to estimate hioavailability and their potential for endocrine effects.
Models will need to he developed to estimate the chemical forms, atmospheric
and hydrological transformations after release to the environment, and the fate of
these substances within environmental media and the food chain.

Data Validation of PAMS Auto-GC VOC Data: Lessons
Learned
Hilary H. Main and Jeffrey D. Prouty
Sonoma Technology Inc., 5510 Sky lane Blvd, Suite 101, Santa Rosa,
CA 95403-1083
Alan Lesion and Pete Brunelli
Connecticut Department of Environmental Protection, 79 Elm Street,
Hartford, CT 06106-5127

    We have performed data validation of continuous VOC data collected with
automated GC systems as a part of the PAMS network in Connecticut. Using a
graphical platform from which to display the data, we performed quality control
tasks on the data and began data analyses. The hourly concentrations of the
PAMS target species were displayed in several ways including scatter plots,
"fingerprints", and time series. Using scatter plots, we investigated the relation-
ship between species at one site or at a pair of sites. Fingerprint plots show the
concentration ol each species in a sample (in roughly chromatographic  order)
and help to identify unique characteristics of the samples. Time scries plots show
the concentrations of species in every sample over a specified time period and
arc useful in showing the diurnal behavior of a species. The retention limes of
the species (rather than the concentrations) were also explored using the same
graphical features. Calibration  data, contamination from shelter off-gassing, cold
trap failure, and other problems were identified. We will discuss and provide
examples of our findings; these results will aid others in evaluating their continu-
ous VOC data.

Analysis of VOC Data in Support of Pollutant  Transport
Studies in Shasta County, California
Hilary H. Main, Lyle R. Chinkin, and Jerry A. Anderson
Sonoma Technology Inc., 5510 Skylanc Blvd, Suite 101, Santa Rosa,
CA 95403-1083
Rei Rassinussen
Oregon Graduate Institute,  19600 NW Von Neumann Dr., Bcaverton,
OR 97006-1999
8                Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


Rita Cirulis
Shasta County Dept. of Resource Management, Air Quality Management District
1826 Butte St., Redding, CA 96001

    The Shasta County Air Pollution Control District has funded a study to
investigate the origin and extent of o/.one and o/.one precursor concentrations
that are due to upwind transport. As a part of the study, air quality measurements
were made aloft during Scptemher 1995. Hydrocarhon measurements were made
aloft for use as tracers of opportunity for analy/ing the source of (he air mass.
Total nonmethane hydrocarhon, spcciated hydrocarbons, and Acxotics(s>
including Frcon 12, Freon ll.Frcon 113, chloroform, 1,1,1-trichloroethanc,
carbon tctrachloridc, trichlorcthylenc, and perchlorocthylene were quantified. In
earlier studies of transport in this region, a hydrocarbon and exotic species
signature for the Sacramento metropolitan area was identified. This signature
will he compared to (he signatures developed from samples collected in Shasta
County to determine evidence of transport. The relative age of air parcels
entering the area will also be determined using the ratios of more  reactive species
to less reactive species concentrations; lower ratios correspond to more aged
hydrocarbon mixtures. This paper will discuss the analyses of the hydrocarbon
and exotics data and the implications of the results to transport in the region.

VOCs from Marker Pens  in Exposure Scenarios
Edwin J. Furtaw, Jr., Rowena M. Gonzalez, and Muhilan D. Pandian
Harry Reid Center for Environmental Studies, University of Nevada, Las Vegas,
4505 S. Maryland Parkway, Las Vegas, NV 89154-4009

    Human exposure to volatile organic compound (VOC) vapors emitted by
several types of marker pens, both indelible and erasable, was studied experi-
mentally and by modeling. VOCs emitted by the pens were identified by gas
chromatography-mass spectrometry, and airborne VOC concentrations were
measured both in a room-size controlled-environmenl chamber, and in an actual
classroom. A  variety of VOCs including alcohols, methyl isobutyl ketonc, the
isomeric xylcnes, and ethylben/.ene, were emitted by  different types and brands
of pens.
    In the chamber, vapor concentrations in the user's breathing /.one were
higher than those near the chamber comer. The concentration profiles were fitted
to an indoor air quality model which allowed estimation of VOC emission rates.
The model accounts for the observed elevated concentrations in close proximity
to the source and for adsorplion-desorption "sink" effects. Model parameter
values which  fit the chamber data were subsequently  used to model the data from
a classroom. The results show that with the use of appropriate parameter values,
indoor concentrations of vapors from this source type can be adequately simu-
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
Abstracts
lalcd with the proposed model. VOC exposure concentrations for relatively
prolonged usage of some marker pens can he at the ppm level.

Emission of VOCs and Polynuclear Aromatic Compounds
from Asphalts and Road Paving Operations
Badri N. Badriyha, Massoud Pirbazari, Varadarajan Ravindran,
Abdul M. Kitto, and Costas Syiwlakis
Environmental Engineering Program, University of Southern California,
Los Angeles, CA 90089-2531

    Assessment of air pollutants from fugitive and new sources is crucial for
cities and metropolitan areas such as Los Angeles that do nol meet Federal air
quality standards. Assessment of unknown sources or ones that have nol been
fully characterized for their contribution to background emissions is essential if
reduction is sought. The South Coast Air Quality Management District
(SCAQMD) initialed a study for the characterization, quantification and assess-
ment of organic pollutants emitted from hoi-mix asphalt manufacturing and road
paving operations. The objectives of the study were to evaluate and quantify the
magnitudes of these emissions as well as their contribution to smog in Los
Angeles through photochemical transformation. This study which was partially
supported by the National Science Foundation (NSF) and the Strategic Highway
Research Program (SHRP) investigated several asphalts and bitumen for their
emissions through extensive laboratory testing and field sampling programs. In
order to be able to conduct these programs, a protocol for the sampling and
analysis of these emissions had to he developed due to the complex nature of
organic constituents in the emission matrix. The first phase of these laboratory
investigations focused on the qualitative assessment of emissions. The second
phase of the  laboratory studies was directed toward the quantitative assessment
of polynuclcar aromatic hydrocarbons (PAHs) and volatile organic compounds
(VOCs). Furthermore, in the second phase  of these investigations, a specially
designed solar simulator/exposure cell was used to simulate road paving condi-
tions such as temperature, solar radiation intensity, and humidity. Asphalt
emissions were then quantified under a variety of atmospheric conditions. In the
field sampling phase of the study, samplers were installed in asphalt hot-mix
manufacturing plants to monitor the emissions at various locations. The major
area of contribution in the plants was the loading /.one where the asphalt was
loaded onto trucks. The second area monitored involved fugitive emissions from
trucks during transportation of asphalt to paving sites. The third area was the
paving site on the road where the asphalt was delivered to the paving machine
and laid on the road surface. The fourth operation monitored  was the compaction
of asphalt by steel rollers. Based on the field and laboratory studies, the magni-
tudes of emissions from asphalt paving operations were estimated.


10              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts
Monitoring for VOCs In Air - Status Report
William A. McClenny
Methods Branch, Air Measurements Research Division, National Exposure
Research Laboratory, U.S. EPA, RTF, NC

    Recent methods development efforts for volatile organic compounds in air
have included:-
    I.   The placement of one methods description TO-15 on the the AMTIC
        Bulletin Board maintained by the Office of Air Quality Planning and
        Standards and the preparation of two others, TO-16 and TO-17. TO-15
        describes the use of a canister-based sampling procedure and a GC/MS
        analytical procedure to monitor VOCs. Method TO-16 concerns the use
        of FTIR-based open path monitoring and the Method TO-17 involves
        the use  of solid adsorbents for samp 1 ing of VOCs.
    2.   The publication of several journal article publications in support of
        methods development for the North American Research Strategy for
        Troposphcric Ozone (NARSTO) and the Photochemical Assessment
        Monitoring Stations (PAMS) network. These publications show how to
        monitor VOCs with better sample integrity than in the past, particularly
        with respect to polar VOCs.
    3.   The publication of several journal articles on the use of open path
        monitoring of VOCs in lenceline monitoring applications using FTIR-
        based systems. Emphasis has been placed on understanding and treating
        the generation of background spectra.
    4.   Evaluation of a new automated gas chromatographic system as part of
        the NARSTO contribution to the 1995 Southern Oxidant Study (SOS)
        Nashville Field Study. Diurnal patterns showing hourly updates !.>r
        several  biogenic compounds were seen for the first time.

Comparison  of Acetone Concentrations in Ambient Air  Using
EPA Methods TO-14 and TO-11
Eric Anderson and Steve Mischler
Radian Corporation, P.O. Box 201088, Austin, TX 78720-1088

    The accurate measurement of polar compounds, such as acetone, in the
ambient air has proven to be difficult. In particular, determining low atmospheric
concentrations for these compounds using current sampling and analysis methods
has proved quite challenging. A comparison of two accepted sampling methods
for acetone is currently being performed from measurements taken at a long-term
ambient air monitoring network. Samples were collected at five sites using both
SUMMA polished stainless steel canisters analy/ed by CG/MD (modified TO-
14) and dinitrophenylhydra/.ine (DNPH) coated silica gel cartridges analy/.ed by

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               11

-------
Abstracts
HPLC (modified TO-11). Over the last four years more than 1800 canister and
sorhcnt samples have been collected. In this paper we will show a complete
statistical analysis of the measurements techniques discussed, noting significant
differences or similarities between the two measurement methods. Statistical
analysis for method accuracy and precision at different concentration levels
(several orders of magnitude variance in concentrations) will also be discussed.
The large number of samples will allow a significant evaluation of both tech-
niques.

Recent Advances in the Analysis of Oxygenated Hydrocarbons
R.A. Rasmussen
Oregon Graduate Institute, P.O. Box 91000, Portland, OR 97291-1000
C.W.Lewis
U.S. EPA National Exposure Research Laboratory,
Research Triangle Park, NC 2771

    The historical and conventional method of hydrocarbon speciation has been
to remove water with a Nafion® dryer or some other water management method
before passing the air sample through a frec/.c-out loop or cryofocusing step to
concentrate/enrich the analytes for subsequent chromatographic analysis. This
approach is robust and easily automated. Removal of the water from a whole air
sample prior to the concentration step is essential to prevent the frce/.c-up of the
trap. This approach has been used for hundreds of thousands of analyses over the
past 30 years. Unfortunately, these water removal procedures, while generally
not affecting the quantitative measurement  of the hydrocarbons, do attenuate the
more water-soluble organic compounds. Measurements we  have made over the
years indicate a significant fraction of the VOC in suburban, rural and remote
atmospheres are not measured using the conventional method. In these non-
urban samples, compared to urban samples, an homologous scries of C6-C|(|
aldehydes is  present that constitutes from 10% to as much as 90% or more of the
VOC mass. Accounting  for all of the carbon during the processing of whole air
samples being prepared to measure the radiocarbon (WC) abundance in atmo-
spheric VOC is a crucial issue when determining the relative contribution of
carbon from  biogcnic- vs. fossil-carbon sources.

Analysis of Available Ambient and Stationary  Source VOC
HAP Sampling Results to Verify Emission Rate Estimates for
Sources in Nogales, Sonora, Mexico
Bill Powers
Powers Engineering, 10324  Meadow Glen  Way,  Suite 2D,
Escondido, CA 92026
12              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


Carmo Fernandez
Arizona Department of Environmental Quality, 3033 N. Central Avc., Rm. 5035,
Phoenix, AZ 85012
Bill Oliver and Ron Dickson
Radian Corporation, 10389 Old Placcrville Road, Sacramento, CA 95827

    The Arizona Department of Environmental Quality (ADEQ) is currently
funding a project to develop comprehensive hazardous  air pollutant (HAP)
emissions inventories for the cross-hordcr communities of Nogales, Arizona and
Nogalcs, Sonora, Mexico. Ambient VOC and PAH data were collected in
downtown Nogales, Sonora in July 1994 by the ADEQ. This paper addresses the
detailed analysis of the ambient HAP data performed by the project team. The
ambient HAP data evaluation serves as a vital cross-cheek of the accuracy of the
HAP emission estimates developed for each source type included in the HAP
emissions inventory.
    The data show that benzene, toluene, xylcne, ethylbcnzene (BTXE) and
aldehydes are the dominant VOC HAPs in the ambient air. The quantity of
BTXE in the ambient air, and ratio of these compounds to each other, implies
that mobile sources arc the principal source of BTXE in Nogales, Sonora.
Significant levels ol olefinic VOCs were also detected.
    Ambient test data also indicate ethane, propane, butane, penlane are the
predominant VOC species  in (lie ambient air. Compressed gas cylinders are the
probable source of these VOCs. Almost all residential and commercial cooking
performed in Nogalcs, Sonora is done from compressed gas cylinders containing
a mixture of propane (primarily) and butane. These cylinders are ubiquitous, and
many are equipped with potentially leaky valves. VOC and VOC HAP test
protocols arc currently being developed by the Mexican Institute of Ecology for
these cylinders and associated cooking ranges.
    Periodic open burning of municipal solid waste (MSW) is performed at a
site located on the border in Nogalcs, Sonora. Ambient  VOC HAP data collected
in July 1994 during both MSW burning periods and "no burn" periods is being
evaluated to determine if the contribution of open burning to the ambient HAP
burden can be estimated for both paniculate and VOC HAPs from the ambient
data.

A Study  of "Holding  Times" for Sulfur Compounds  in
Restek's SilcoCan Canisters
Suclia Pannar, Abdul M. Kitto, and Litda Ugarova
Atmospheric Analysis and Consulting, Inc., 4572 telephone Road, Suite 920,
Ventura, CA 93003
David M. Shelow
Restek Corporation, 1 10 Bcnncr Circle, Bellcfontc, PA  16823-8812

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               13

-------
Abstracts


    A simple, cosl-effcctivc method lor sampling and analysis of H,S, COS,
mereaptans, organo sull'ide, and disulfides in vvliole air samples is described.
This method involves sampling sulfur compounds in humidified ambient air
using Rcstek's SilcoCan canisters followed by analysis on a modified gas
chromatograph/flamc photometric detector (GC/FPD). The stability of sulfur
compounds was investigated at four different concentrations for 120 hours. The
results show excellent stability of sulfur compounds in SilcoCan canisters over
the time period studied. It is inferred from this study thai SilcoCan canisters arc
suitable for sampling all VOCs including sulfur compounds.

Measurements of Hydrocarbons and Reduced Sulfur
Compounds Emitted from a Wastewater Treatment Pond
Q. Trail, P. Fellin, C. Ceen, and D. Friel
BOVAR Environmental, 2 Tippetl Road, Toronto, Ontario, M3H, 2V2, Canada

    A flux chamber was deployed on the water surface to monitor the emissions
of hydrocarbons and reduced sulfur compounds from a wastewaler treatment
pond at a refinery site. Air samples were collected in Tcdlar bags and analy/.cd
on-site by means of a gas chromalograph/llamc ioni/.ation detector (GC/FID) for
hydrocarbons  and a gas chromatograph/llamc photometric detector (GC/FPD)
for reduced sulfur compounds.
    The standard deviations of the duplicate samples for hydrocarbons and  for
reduced sulfur compounds were better than 4% and 11 %, respectively. The pond
was monitored during the  daytime and nighttime, under two different weather
conditions where the daytime temperatures were about 22 C and 14 C. The
results showed that the difference between day and night emissions of hydrocar-
bons and reduced sulfur compounds were greater during warm weather com-
pared to those during cold weather.
    Air samples were also collected with charcoal adsorbent tubes and analyzed
by means of a gas chromatograph/mass selective detector (GC/MSD) at the
BOVAR Environmental (BE) laboratory to confirm the GC/FID hydrocarbon
analyses. There was excellent agreement between hydrocarbons identified by
GC/FID and GC/MSD.

Extractive  FTIR Technology - Applications In Combustion
Source Measurements
Gunxeli Sagun Shareef and Jeffrey P. LaCosse
Radian International LLC, P.O. Box 13000, Research Triangle Park, North
Carolina 27709
James M. McCarthy
Gas Research  Institute, 8600 West Bryn Mawr Avenue, Chicago, IL 60631
14             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                Abstracts


    In response to the 1990 Clean Air Act requirements, the Gas Research
Institute (GRI) is investigating emissions from industry equipment including
glycol dchydrators, internal combustion engines, boilers, healers, and incinera-
tors. To characterize emissions from internal combustion engines reliably and
accurately, GRI developed and validated a method based on extractive Fourier
Transform Infrared (FTIR) technology. This paper will describe the FTIR lest
protocol approved by U.S. Environmental Protection Agency, including a brief
discussion on external combustion source (i.e., boilcr/heater/incincralor) FTIR
measurement experience. In addition, the approach used in reference spectra
development and the associated challenges will be described.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996                15

-------
                                                            Abstracts


                             Session 3
         Natural Emission of Oxidant Precursors
        Cochairs: Viney P. Aneja, NC State University, Raleigh, NC;
       and Bruce W. Gay, Jr., U.S. EPA, Research Triangle Park, NC

Project NOVA (Natural emissions of Oxidant precursors:
Validation of techniques and Assessment) Overview
Viney P. Aneja, Wayne P. Robarge, S. Arya, and V. Saxena
North Carolina State University, Raleigh, NC;
C. Geron, T. Pierce, and B. Gay
US Environmental Protection Agency, Research Triangle Park, NC;
B. Hicks, W. Luke, and T. Meyers
NOAA, Silver Spring, MD;
M. Wesely
Argonnc National Laboratory - DOE, Argonne, IL;
A. Delany
National Center for Atmospheric Research, NCAR,  Boulder, CO;
J. Levine
NASA, Langley.VA;
T. Ellestad, J. Clark and P. Finkelstein
US Environmental Protection Agency/NOAA, RTP, NC;
B. Doddridge
University of Maryland, College Park, MD;
J. Pierce
Duke University, Durham, NC

    Tliis project attempts  to determine the surface exchange fluxes of the oxides
of nitrogen using several independent methods. It will then compare the results
from dynamic enclosure techniques to measurements using the micrometeoro-
logical methods. Eddy correlation measurements of  the turbulent fluxes of NO,
NO,  (or NOX), and of O3 together with gradient measurements of the same
species will enable the soil emission of NOX (= NO  + NOJ and the effect flux of
NOX to the free atmosphere to be compared over a range of atmospheric and soil
conditions. A comprehensive characteri/.ation of the soil chemistry and microbi-
ology was undertaken. The results of this study will  enable a characteri/.ation of
the dynamic enclosure technique which will he used to survey nitrogen oxide
emissions of selected soils throughout the southeast  United States.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              17

-------
Abstracts


Natural Emissions of Qxidant Precursors: Validation Of
Techniques and Assessment (Project NOVA 1995) - Site
Characterization
Wayne P. Robarge,  Viney P. Aneja, anil YJwngju Li
Departments of Soil Science and Marine, Earth and Atmospheric Sciences,
North Carolina State University. Raleigh, NC
Wayne P. Robarge
Corresponding Author

    A major objective of Project NOVA 1995 is to intercomparc surface
exchange fluxes of N and other gaseous species using both soil enclosure and
micromcteorological techniques. This presentation provides the results of an
intensive site characteri/ation necessary  for completion of this objective.  The
research site (approximately  136 ha of continuous corn \Zea Ma\s\; width 1067
m, length - 1280 m)  is located in Washington County in the Northeastern  Lower
Coastal Plain region of North Carolina. The site has a SW - NE orientation,
allowing a level fetch of = 1000 m over a continuous corn canopy. The level
topography results in soils with seasonal high water tables. Agricultural soils arc
drained with a series of parallel ditches (I m depth; 50 m apart) connected to a
series of larger ditches which drain into primary canals. Three soil types arc
present at the research site: a Conahy muck, an Arapahoc fine sandy loam, and a
Portsmouth fine sandy loam. Visual observations via soil pits, and chemical
analyses, reveal that the soils differ primarily in organic matter content and
thickness of a distinct black surface hori/on ( hulk density =  1.3 gm cc-3). The
corn crop used for the study was planted 12 April 1995. N application totaled
184 kg N ha'1,  with 9 kg N ha"1  I placed in the  row as starter fertilizer, and 73 kg
N ha"1 broadcast as a 30% N solution (equal parts urea, ammonia, nitrate) added
at planting. The remaining 102  kg N ha"1, in the form of the 30% N solution, was
applied as a concentrated band  (2 cm in width) on the soil surface down the
center of the interrow on 20 May, 1995.  Soil samples (0 - 30 cm depth, I Ocm
increments) were  taken daily to estimate soil water content and cxtractablc N (IN
KCI; NH4-N + NO,-N). Average soil cxlractablc N was = 60 kg N ha"' prior to
20 May,  1995. Following the N sidcdrcss, average cxlractablc N became  much
more variable due to method of application. Above ground whole plant samples
were collected once a week starting on 9 May, 1995. Total dry mass, elemental
N and C content, leaf area per plant, and leaf area index were determined for
each week until harvest. This data confirms that soil NO flux measurements were
made during a period (12 May  to 15 June,  1995) of rapid vegetative growth and
N uptake by the crop. The characterization data will be presented in different
temporal formats  (e.g. change in percent water-filled pore space) to facilitate
comparison to temporal trends  in measured soil NO (lux and extrapolation to
published models relating soil variables to predicted  flux.

18              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
Biogenic Soil Emissions of Nitric Oxide and Nitrous Oxide
Using a Closed Static Chamber During the NOVA-95
Experiment
Joel S. Levine, Donald R. Calwon, Jr., and Wesley R. Cofer, III
Atmospheric Sciences Division, NASA Langlcy Research Center,
Hampton,  VA 23681-0001
EdwartL.  Winstead
GATS/SAIC, 28 Research Drive, Hampton, VA 23666
Wayne P. Robarge
Department of Soil Science, North Carolina Stale University,
Raleigh, NC 27695-8208
Viney P. Aneja
Department of Marine, Earth, and Atmospheric Sciences, North Carolina State
University, Raleigh, NC 27695-8208

    Nitric oxide (NO) and nitrous oxide (N2O) are environmentally significant
gases that play key roles in the chemistry of the troposphere (NO), stratosphere
(N2O), and global climate (N2O). Biogenic soil emissions are a important global
source for  both of these gases. Measurements of biogenic soil emissions of nitric
oxide and nitrous oxide using a closed static chamber were obtained during the
NOVA-95 experiment (May - June, 1995). The simultaneous measurements of
biogenic soil emissions of NO and N2O provide important information on the
utilization  of nitrogen by soil microorganisms in an agricultural soil. These
measurements were obtained from more than a dozen sites within the cornfield-
some fertilized and some unfertilized control sites. Diurnal measurements were
obtained at several of these sites to assess the impact of the diurnal soil tempera-
ture at the same site on  the biogenic emission of these gases. Measurements were
also obtained before and after rainfall on these sites. Nitric oxide emissions
exhibited a range of more than two orders of magnitude during the experiment,
from less than I to almost 200 nanograms of nitrogen meler-2 sec-1 (ng N m-2 s-
I), depending on rainfall and fertilization. Highest soil emissions of nitrous
oxide were obtained following a very strong rainfall, when the soil became
saturated with water. The impact of soil temperature, soil moisture, and soil
nitrogen on biogenic emissions of nitric oxide and nitrous oxide will be dis-
cussed. These measurements will be compared with measurements of biogenic
soil emissions of NO and N2O that we obtained on agricultural sites in Virginia
and Colorado and from natural soils in very diverse ecosystems, including
chaparral in southern California, savanna grasslands in southern Africa and
boreal forests in Canada.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               19

-------
Abstracts


Measurements of Nitric Oxide Flux From Intensively Managed
Agricultural Soil in the Coastal Plain of North Carolina
Paul A. Roelle, James R. O'Connor, and Viney P. Aneja
Department of Marine, Earth, and Atmospheric Sciences, North Carolina State
University, Raleigh, NC 27695-8208
Deug-Soo Kim
Department of Environmental Engineering, Kunsan National University,
Kunsan, Korea-573-700
Wayne P. Robarge
Department of Soil Sciences, North Carolina Stale University,
Raleigh, NC 27695-8208

    Rural  Photochemistry is influenced by Nitric Oxide (NO) emissions from
agriculturally managed soils. The llux of (NO) was measured during a one month
intensive measurement period  in the coastal  plain of North Carolina as part of
Project NOVA during May-June 1995. NO emission rates were determined
using a dynamic chamber system. Preliminary results indicate a range of fluxes
from 4.18 to 253 ng N m-2 s-1 for NO. The fluxes were regressed against soil
temperature, air temperature, water filled pore space, and cxtractablc soil
nitrogen; they showed a semi-logarithmic relationship with soil temperature.
Fluxes of NOY and NO2 were also calculated to ascertain the balance of
nitrogen species being emitted. Ambient (10 meter) concentrations of NO, NOY,
and O3 were also measured along with wind speed and wind direction in an
effort to gain insight regarding the local production and/or long range transport
ol these reactive nitrogen compounds.

Controlled Laboratory Studies of NO Flux Into the Lower
Levels of the Troposphere Under Widely Varying Soil
Conditions
J. Jeffrey Peirce and Banu Ormeci
Department of Civil and Environmental Engineering, Duke University,
Durham, NC

    Controlled laboratory studies of NO emissions from soils arc discussed
which contribute to the understanding of the mechanical, chemical and biological
mechanisms responsible for NO flux  from the soil to the lower levels of the
troposphere. Newly designed and constructed laboratory test chambers with
attendant laboratory procedures arc described. The results of tests conducted
under widely varying soil conditions, including  moisture, pH, temperature, and
the presence of sunlight arc presented. The information  is referenced to data
collected in field studies.
20              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


Eddy Correlation Measurements of NO, NO2, and O3 Fluxes
W. Gao, M.L. Wesely, D.R. Cook, and T.J. Martin
Environmental Research Division, Argonne National Laboratory, Argonne, IL

    The micronietcorological technique of eddy correlation was used lo measure
the vertical (luxes of NO, NO2, and O3 in rural North Carolina during the spring
of 1995 as part of a collaborative effort on the natural emission of oxidant
precursors validation of techniques and assessment (NOVA). The net mass  flux
densities were measured at heights of 5 m and 10 m above an agricultural field
with short corn plants and large amounts of exposed bare soil between the rows.
Large upward eddy fluxes of NO2 were seen, and strong NO emissions from the
soil were measured by collaborators using environmental enclosures on the  soil
surface. Preliminary calculations for midday observations indicated that about
50% of the nitrogen emitted from the soil was converted into NO2 llux at a
height of 5m, about 40% remained as NO flux, and 10%) was returned to the
vegetation and soil. Some divergence of the NO2 and O3 tluxes was detected
between heights of 5 m and 10 m. These results are consistent with the likely net
NO2 production and O3 destruction rates. Such data will be used to help develop
parameteri/.ations on NOx llux into the lower troposphere.

Nitric Oxide Fluxes Observed Over A Fertilized Corn Field
Bruce G. Doddridge, Kevin L. Civerolo, and Russell R. Dickerson
Department of Meteorology, University of Maryland, College Park, Maryland,
USA
Richard A. Valigura and Bruce B. Hicks
National Oceanic and Atmospheric Administration, Air Resources Laboratory
Silver Spring, Maryland, USA

    This group maintains a research site within the University of Maryland  Wye
Research Center, Wye River, Maryland located at 38 deg 54' N, 76 deg 14'  W on
Maryland's Eastern Shore. Continuous measurements at 10 m of meteorology,
UV radiative flux, O3, and CO have been made at this site for extended periods
since 1993, sometimes enhanced with continuous measurements of NO and
NOx. From 13 July lo 3 August 1995, NO emission fluxes using the technique of
eddy correlation, along with additional complementary chemical data (mean CO,
O3, NO, NOx) and  UV radiative llux were measured at this site. Fast wind speed
and temperature values were measured with a sonic anemometer. During this
period measurements were made on a NOAA/ARL sampling tower at 6 m, above
a fertili/.ed corn canopy which was about 2.4 m tall by late July. From 25-30
July, 120 x 30-min. average sensible heat and NO flux values were computed.
These results will be reported  and interpreted in detail, with particular focus on
characteristic behaviour at 24-hour and synoptic time scales. The mean (and
median) NO fluxes  for this time were 7.05 (and 7.34) ng N / m**2 / s, and values

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               21

-------
Abstracts


ranged from -12.33 lo 21.67 ng N /m**2 /s. Six oflhc 30-min. average flux
values were negative; during these six periods, ambient observed NO mixing
ratio was above 1  pphv. This suggests that during these periods, the boundary
layer was adjusting to a new equilibrium by depositing NO to the surface. The
NO llux values compare favorably with similar measurements made over plots of
fertilized soil.

Micrometeorology and Flux-Gradient Relation of NO-NO2-O3
System Near the Surface
Y. Li, S.P. Arya, and V.P. Aneja
Department of Marine, Earth & Atmospheric Sciences, Box 8208, North
Carolina Stale University, Raleigh, NC 27695

    Eddy-correlation technique and dynamic chamber technique were employed
at the same lime and  same location in an experiment in May 1995. In this paper,
diffusivilics of momentum, heal and water vapor as calculated from the mi-
cromctcorological data; arc compared using the framework of the Monin-
Obukhov similarity theory. The applicability of the theory to ozone and nitrogen
dioxide is also investigated. The authors compare and discuss the fluxes between
two different height levels and between two measurement methods with the
objective of better understanding micrometcorological and chemical influences
on the flux-gradient relationship of NO-NO.-O, system in the atmospheric
surface layer.

The Relationship Between the Surface Exchange Fluxes of
Trace Species Determined by Enclosure and by
Micrometeorological Techniques
A.C. Delany and S.P. Oncley
National Center for Atmospheric Research, ATD/SSSF, Boulder, CO

    A major objective of the NOVA program is lo characterize the measurement
of lluxcs by the dynamic enclosure technique. To this end the surface exchange
fluxes of oxides of nitrogen and other species arc determined using both enclo-
sure and micromelcorological techniques. To complement the basic flux mea-
surements, meteorological, micromeleorological, soil, chemical and radiation
measurements are also made lo characterize the tolal measurement environment.
The fluxes measured  by eddy correlation and the (luxes measured using enclo-
sures are compared and an attempt is made to analyze the affect of introducing
additional factors to effect a closure between the two different determinations.
An examination of the olhcr prevailing environmental parameters is made lo
enable us to define those conditions when the two methodologies can be brought
into agreement A simple numerical model relaling the two measurement

22               Measurements of Toxic and Related Air Pollutants, May  7-9, 1996

-------
                                                            Abstracts


methods under different conditions has been devised and is used to compare the
observations by the two techniques brought into agreement.

Flux Measurements of Volatile Organic Compounds Over a
Corn Field
M. Das, V.P. Aneja, and L. Hagerman
Department of Marine, Earth and Atmospheric Sciences, North Carolina State
University, Raleigh, NC  27695
W. Lonneman.
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
D. Cook and M. Wesley
Environmental Research Division, Argonne National Lab,  Argonne, IL 60439

    Vertical concentration gradients of volatile organic compounds (VOCs)
were measured over a corn field, in its early growth period, during May 1995, in
the coaslal plains of North Carolina. These measurements were made as part of
project NOVA (Natural emissions of Oxidant precursors: Validation of tech-
niques and Assessment) to estimate the flux of VOCs. A modified Bowen Ratio
technique was adopted for flux calculations. Average emissions of methanol,
ethylene, a-pincne, b-pinene, and d-limonene were found to be 29.66±2().()6,
20.68±14.2I, 37.58±37.28, 9.37+7.51 and 9.09±5.28 (Jg/m2/hr respectively.
Deposition velocities of certain anthropogenic hydrocarbons were also esti-
mated.

Measurement of Soil  Nitrous Oxide Flux from an Intensively
Managed Agricultural Row Crop in the Lower Coastal Plain
Region of North  Carolina
Zlwngju Li, Wayne P. Robarge, and Viney P. Aneja
Department of Soil Science and Marine, Earth and Atmospheric Sciences,
North Carolina State University, Raleigh, NC

    The ratio of soil nitric oxide (NO) flux to soil nitrous oxide (N2O)  flux is
often used to determine the dominant process (nitrification or dcnilrificalion)
associated with NO production in soils. This study, a part of the Natural emis-
sions of Oxidant precursors: Validation of techniques and Assessment (NOVA)
project - 1995, measured soil N2O flux using PVC venled-cylindrical chambers
(diameter-10.1 cm; height-10 cm) in an intensively managed agricultural row
crop (corn - Zca  Mays) from 19 May to 2 July, 1995. PVC  sampling rings (n =
64; diametcr-10 cm; height-10 cm) with beveled edges were installed to a depth
of 8.5 cm in groups of 16 (n=8; row and interrow positions) at four locations
within the sampling area two weeks prior to the start of the measurement period,
and left in the same position for the duration of the study. Air samples were

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              23

-------
Abstracts


obtained through a septum with a two-way needle into an evacuated glass septa
bottle (125 m.) 25 minutes (10-15 minutes during periods of high N2O using a
gas chromatogrnph equipped with an electron capture detector. From  19 May to
4 June, 1995 (total rainfall < I cm), average soil N2O flux was 15.8 (+/- 297)ng
m-2 s-1 for the row and intcrrow positions, respectively, for the remainder of the
study. Calculated N20 fluxes > 2100 ng m-2 s-1  were observed immediately
following (within 6 hours) rainfall events > I cm during this period. Based on the
calculated N20 flux relatively low, conditions which would favor NO produc-
tion by nitrification. After 5 June, 1995, soil N2O flux increased, suggesting that
dcnilrificalion could have accounted for a substantial portion of the NO flux
emitted during the remainder of the study period.

Biogenic Nitric Oxide Source Strengths
Viney P. Aneja, Paul A. Roelle, and James R. O'Connor
Department of Marine, Earth, and Atmospheric Sciences, North Carolina State
University, Raleigh, NC 27695-8208
Wayne P. Robarge
Department of Soil Sciences, North Carolina State University,
Raleigh, NC 27695-8208

    Nitric Oxide (NO) emission rates were measured during the summer of 1995
at 4 different sites, representing major physiographic regions of the Southeast
United States, in an effort to compare lluxes from agriculturally managed soils.
Emission rates were determined using a dynamic chamber system. The measure-
ment campaign began in the Coastal Plains region of North Carolina and
progressed to the Upper Piedmont region of (he state. Additional  measurements
included ambient (10 m) O3, NO and NOY concentrations, wind speed and
direction. Soil temperature (at 5cm depth) and air temperature were also re-
corded. Additionally, soil samples were taken which were analy/.ed for soil water
content, bulk density, total extractablc nitrogen and percent water filled pore
space. Preliminary result indicate a range of fluxes from  1.21 to 78.57 ng N m-2
s-l for NO. Intcrsite differences in NO flux reflect the differences in the amount
of fcrtili/.cr applied to each of the soils. NOY fluxes and NO2 fluxes were also
calculated in an attempt to ascertain the balance of nitrogen species being
emitted.

Atmospheric Pollutant Deposition Over a South Central North
Carolina Cornfield
V.K. Saxena and V.P. Aneja
Department of Marine, Earth & Atmospheric Sciences, Box 8208, North
Carolina State University, Raleigh,  NC 27695
24              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 T.P. DeFelice
 University of Wisconsin at Milwaukee, Atmospheric Science Group, Lapham
 Hall, Box 413, Milwaukee, WI 53201
 M. Wesely
 Environmental Research Division, Argonne National Laboratory,
 Argonne, IL 60439

    The National Oceanic and Atmospheric Administration Atmospheric
 Turbulence and Diffusion  Division (ATDD) model was used to infer the dry
 deposition velocities of gaseous and particle pollutants, such as O3, S02, NO2,
 and sulfatc and nitrate aerosols onto a cornfield in North Carolina (35.7° N,
 76.8° W) during its growing season. The deposition velocity of ozone is in the
 range of 0 to 4 cm s-1. The deposition velocities of both sulfur dioxide and
 nitrogen dioxide arc in the range of 0 to 0.2 cm s-l with an average of 0.1
 (±0.02) cm s-l. Deposition fluxes are calculated and compared with other
 studies. The analysis of the data set is  in progress.

 Contributions of Meteorology on Ozone Behavior in Urban
 North Carolina
 Regi G. Ooinmen, Viney P. Aneja, anil Allen J. Riordan
 Department of Marine, Earth, and Atmospheric Sciences, North Carolina Slate
 University, Raleigh, NC 27695-8208
 George C. Murray, Jr. and Tom Manuszak
 NC DEHNR, Air Quality Section,  Raleigh, NC 27603

    As part of an effort by the slate of North Carolina to develop a Slate
 Implementation Plan (SIP) for o/.onc control, a network of o/.one stalions was
established to monitor o/.one concentrations across the slate. Approximately
 twenty-five o/.one slalions made continuous measurements surrounding the three
 major Metropolitan Statistical Areas (MSAs): Raleigh-Durham (RDU), Charlotte
(CLT), and Greensboro - Winslon-Salcm - High Poinl (GSO). The relationship
of meteorological parameters on ihe o/.onc behavior were analy/.cd during  the
summers of 1993,  1994, and 1995. A A(O3) analysis (difference between the
daily maximum [O3] measured al an upwind and downwind site, which reflects
the nel  increment of photochemical o/.one added to an air mass as it advecls over
Ihe city) was performed for high o/.one days (|O3|3 0.090 ppbv) in an effort lo
study if the current emission controls have been effective  in urban regions of
North Carolina. Synoptic and mesoscale analysis of key meteorological param-
eters were performed using data  from the National Weather Service, AIR Sonde
launches, and the Charlotte Wind Profiler, and it suggested that synoptic
meteorology and urban photochemistry were important in analy/.ing high o/.one
days.

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996               25

-------
Abstracts


A Climatology of Regional Ozone: Meteorological Effects on
Ozone Exceedences in the Southeast U.S.
James R. O'Connor and Viney P. Aneja
Marine, Earth, and Atmospheric Sciences Department. Box 8208, 1125 Jordan
Hall, Raleigh, NC 27695
Wayne Cornelius and George Murray
North Carolina Department of Environmental Health and Natural Resources, Air
Quality Section, Raleigh, NC  27603

    Implementing a plan for urban areas to meet the EPA's guidelines for o/.onc
attainment continues to he a problem, particularly in the southeastern United
States where vast spans of natural forest areas emit greater concentrations of
biogenic hydrocarbons than in other areas of the country. While the EPA
standards have been met for NOX and VOCs, o/.onc concentrations still exhibit
large interannual variation, which is believed to be caused in part by differences
in meteorological parameters. O/.onc records were studied during the photo-
chcmically active season (April 01 - October 3 1 each year) for a fifteen year
period from 19X0-1994 at the  five major Metropolitan Statistical Areas (MSAs)
in the Southeastern United States (i.e., Atlanta, GA; Charlotte, NC; Nashville,
TN; Raleigh, NC; and Greensboro, NC). From the hourly data, all o/.onc
cxceedenccs (concentration > 0.120 ppm) were recorded, along with a daily
maximum concentration and a daily average concentration during the peak o/.onc
production lime of each  day (10:00 am - 4:00 pm). Meteorological data studied
in an effort to explain the interannual variation of o/.onc concentration include
daily maximum temperature, and daily averages (from 10:00 am to 4:00 pm
only) for wind speed, wind direction, temperature, pressure and relative humid-
ity. Regional high pressure stagnation was also studied as an explanatory
variable.

Regional Analysis of Non-Methane Hydrocarbons in the
Southeastern United States
Lisa M. Hagerman and Viney P. Aneja
Department of Marine, Earth, and Atmospheric Sciences, North Carolina State
University, Raleigh. NC 27695-8208

    Concentration of non-methane hydrocarbons as well as o/.onc, meteorologi-
cal and trace gas data were measured at a network of eight rural sites located
within the southeastern United Stales as a part of the Southern Oxidants Study.
C2-CIO hydrocarbon were sampled once every six days  from September 1992
through October 1994. The measurements were made in an effort to enhance the
understanding of the behavior and trends of o/.onc and other photochemical
26              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


oxidants in this region. Biogenics display a maximum in the summer and
minimum during the winter. Lower.molccular weight alkancs and acetylene show
a seasonal variation with a maximum during the winter.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               27

-------
                                                             Abstracts
                             Session 4
                    Aerosol Measurements
         Chair: Petros Koutrakis, Harvard University, Harvard, MA

Attribution of College Park Aerosol Particulate Components
by Size and Source
F. Divita
Desert Research Institute, P.O. Box 60220, Reno, Nevada 89506
C.B. Kidwell, E: Kempis, andJ.M. Ondov
Department of Chemistry and Biochemistry, University of Maryland,
College  Park, MD 20742

    Si/c-fraclioned aerosol paniculate samples in 6 discrete si/.e intervals,
ranging  from <0.056 to<15 um, were collected with micro-orifice impactors
(MOI) in the summer of 1990, in College Park, MD, a heavily-urban, nonindus-
trial area near Washington, D.C. Samples were collected on the roof the Meteo-
rology building at the University of Maryland for 11 hours  at a How rale of 30
L;imin-l using Gelman "Teflo" filters as impaction substrates and backup filters.
The firsl three stages (D503I .K urn) were coaled with a thin film of
dimethylpolysiloxane to reduce contamination of fine-particle slages with coarse
particles. Selected samples were analy/.ed for 44 elements by instrumental
neutron activation analysis using the NIST 20 MW research reactor with flux of
4.5 x 1013 n-cm'2-s''. Chemical mass balance calculations were made on each of
the stages of 17 MOI samples to resolve the  ambient si/.e distributions of aerosol
mass and various elemental constituents of particles associated with  12 generic
sources.  Sulfur, V, and Zn were nearly exclusively attributed to distant coal-fired
power plants (i.e., "regional sulfate"), oil combustion, and municipal incinera-
tion, respectively. Antimony and Cd were strongly associated with incineration,
but had significant contributions from local coal combustion. Si/.e spectra for
regional  sulfate, oil-combustion, and incinerator components were clearly
different.

Resolution of Aerosol Constituents by Size and Source in the
Urban Plume Over Lake Michigan
P.P. Caffrey, M. Han, A.E. Suarez, and J.M. Ondov
Department of Chemistry and Biochemistry, University of Maryland,
College  Park, MD 20742
R.K. Stevens
U.S. Environmental Protection Agency, Research Triangle  Park, NC 2771 1

    Si/.e segregated atmospheric aerosol samples were collected simultaneously

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              29

-------
Abstracts


at three sites on and around Lake Michigan from July 17 through July 28, 1994:
Chicago, IL, over Lake Michigan on hoard the EPA R/V Lake Guardian, and on
the distant shore at South Haven, MI. Micro-orifice impactors (MOI) used for
sample collection yielded  10 aerosol si/.e fractions per sample ranging from
<.()56 mm to 15 mm, and sampling followed a diurnal cycle of separate day and
night collections. Gclman  "Tcflo" Teflon membrane filters were used as impac-
tion substrates, and the first three stages of all impactor samples were coated
with approximately 2 mg of Dow dimcthylpolysiloxanc  grease to reduce coarse
particle bounce. Twenty seven sets of MOI samples have been analyzed for >45
elements using X-ray lluorcsccnce (XRF) and Instrumental Neutron Activation
Analysis (INAA). Chemical mass balance calculations were made using data for
each of the stages of a select group of these samples to resolve the si/.e distribu-
tions of elemental constituents of particles associated with 12 generic sources.

Size  Distribution of IR-Tagged Soot from Diesel Sanitation
Trucks
K.C. Tliaung
U.S. Environmental Protection Agency, Central Regional Laboratory, 839
Bcstgate Road, Annapolis, MD 21401
A. Suarezand J.M. Ondov
Department of Chemistry, University of Maryland, College Park, MD 20742
D. Poster
National Institute for Standards and Technology, Gaithcrsburg, MD  20899

    A part of the U.S. EPA's great waters study of the influence of the Balti-
more  urban plume on water quality of the chcsapeakc Bay, 100 g of indium (III)
acctylacetonatc was used to tag 170 m3 of dicsel fuel burned by the City of
Baltimore's sanitation truck licet during a 20-day period in August, 1995.
Aerosol emitted from the exhaust  pipe of a single, rear-loading sanitation truck
was diluted and si/.c-fractionatcd into 10 discrete si/c intervals, ranging from
<().()56 to <15 (jm, and collected in a micro-orifice impactor (MOI). Each
sample was collected continuously for up to 5 hours as the truck made its rounds.
Samples were analy/.cd for elemental carbon by correlation with optical density,
mcthylcnc chloride extraclablc organic material (EOM), and for the Ir tracer
material by instrumental neutron activation analysis. Concentration vs si/.e
spectra contained broad peaks, presumably, resulting from changing combustion
conditions during the operation of the truck.

Health Risks and Particle Monitoring: New Technologies to
Meet Emerging Risk Management Needs
Robert A. Michaels
RAM TRAC Corporation, 3100 Rosendalc Road, Schcncctady, NY  12309

30              Measurements  of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                               Abstracts


    EPA is reviewing the airborne particle standard, which is based upon
associations with mortality and morbidity in communities. Particle levels during
the 1952 London fog were 1,200 (jg/M3. Recent studies associate effects with
levels within the standard, and below levels explained lexicologically. EPA must
decide whether to control particles to levels below those of demonstrated
lexicological significance. This study inquires whether 24-hour average levels
might encompass undetected excursions which explain mortality.
    Two monitoring technologies—inertial and bcla attenuation—were found
capable of measuring over short intervals. Both represent economical alterna-
tives to gravimetric measurement. Data revealed 15-minute excursions reaching
2,000 ug/M3.
    Toxicology literature confirmed that particles in this range can harm.
Persistent elevations must be more stressful than bricl excursions observed here.
However, particles within the 10-micron range are cleared inefficiently, espe-
cially by individuals suffering from respiratory disease. Consequently, short-term
excursions of respirablc particles may cause long-term internal dosing, and
cardiopulmonary stress for susceptible individuals.
    This study suggests that known lexicological mechanisms explain associa-
tions of particles with mortality if exposure is scrutini/.ed over short intervals.
However, the standard requires averaging over 24  hours, precluding detection
and control of exposures which seem lexicologically significant. EPA may avoid
strengthening 24-hour limits by imposing a less burdensome one-hour limit. This
strategy appears to be health protective, justified by data, and  feasible.

Harvard/EPA Ambient Fine Particle Concentrator for
Conducting  Human Exposures
Constantinos Sioutas, Petros Koutrakis, and Steven Ferguson
Harvard University, School of Public Health, 665 Huntingdon  Avenue,
Boston, MA 02115
Chong S. Kim
U.S. Environmental Protection Agency, Chapel Hill, NC, 27711
Robert M, Burton
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

    This paper presents the development and evaluation of an ambient particle
concentrator for conducting human inhalation exposure studies. The system
concentrates ambient particles in the si/.e range 0.1-2.5 mm (aerodynamic
diameter, dt) by drawing them through a series of three virtual impactors. Each
impaclor contains the majority of ambient fine mass (d<2.5 mm aerodynamic
diameter) in a bleed flow (minor How) that in the first two stages is 20% of the
total How entering the stage, while  in the third stage is 50% of the total How. A
dilution system  has been  added to condition the temperature and relative
humidity of the  concentrated aerosol prior to supplying it to the exposure

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               31

-------
Abstracts


chamber. The intake sampling How is 5500 lilcrs/min and the system delivers an
aerosol enriched in concentration by a factor of 10 at 200 litcrs/min.
The virtual impactors were characleri/ed using outdoor and indoor air samples
as lest aerosols. Fine mass and sulfate concentrations at the outlet of the concen-
trating system were compared to the ambient fine mass and sull'atc levels,
determined using Harvard-Marplc impactors. The experimental tests have
demonstrated that the overall concentration enrichment achieved by our system
is by a factor of 8-11, depending on the average particle size of the sampled
aerosol. Furthermore, the concentrated aerosol is delivered into the exposure
chamber at a pressure of 0.94 atmospheres, thus making it possible to conduct
human exposures at essentially atmospheric pressure.

Characteristics of Ambient Aerosols in Washington D.C. as
Measured by  Continuous and Time-Integrated  Sampling
Methods
Ctmstantinos Sioittas, Petros Koutrakix, and George Allen
Harvard University, School of Public Health, 665 Huntington Avenue,
Boston. MA 021 15
Robert M. Burton
U.S. Environmental Protection Agency, Research Triangle Park, NC. 27711

    During the summer of 1994, an array of real-time continuous and integrated
measurements of the concentration of participate matter (PM1U and PM,S) were
made at the Harvard/EPA, monitoring station in downtown Washington, D.C.
The Scanning Mobility Particle Si/.er (SMPS) and the Aerodynamic Particle
Sizcr (APS) were used in conjunction to count continuously particles ranging
from 0.02-30.Omm.  In addition, the study included real-time measurements of
SO4: and black carbon, as well as  continuous PM|()and PM,5measurements
using the Tapered Element Oscillating Microbalancc (TEOM). Furthermore, 3-
hour integrated measurements of elemental and organic carbon were conducted
using the Rupprecht and Pataschnick carbon monitor, and 24-hour integrated
measurements of the parliculatc mass, sull'atc, and nitrate size distributions were
made using the Micro-orifice Uniform Deposit Impactor (MOUDI).
    Diurnal concentration and particle size distribution patterns will be summa-
rized for approximately 50 days of this study. The effect of relative humidity on
real-time and integrated concentrations will be discussed. Furthermore, the effect
of the relative humidity and particle concentration on the mass median  diameter
of the fine (0-2.5 mm) aerosol will he presented.                     ;
32              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


The Role of Transition Metals in Lung Injury After Air
Pollution Particle Exposure
A.J. Ohio, Z.H. Meng, G.E. Hatch, and D.L Costa
NHEERL, U.S. EPA, Research Triangle Park, NC 27711

     We tested the hypothesis that a metal initially complexed to the surface of a
particle could catalyze the generation of oxidanls and be associated with an
inflammatory injury of the lung. Polystyrene heads  (0.945 mm; Bangs Laborato-
ries, Curmcl, IN) with surface-COOH groups were treated with NaOH and then
agitated in cither 100 mM ferric ammonium sulfate (Iatcx-Fe'+) or sterile water
(latex) for 30 minutes. In vitro oxidant generation was measured as the ahsor-
hance of thioharbituric-acid-products (AS3,) in reaction mixtures including 1.0
mM H.,0,, 1.0 mM ascorbate, 1.0 mM deoxyribose, and particles. Incubations
with 1000 mg/ml latex- Fe'+ were increased relative to those including either an
identical mass of latex or saline (A,,,  of 1. 146 ± 0.036, 0.372 ± 0.015, and 0.093
± 0.016 respectively; mean standard ± deviation). Similarly, maximal luminol-
cnhanccd chcmilumincsccncc by rat alveolar macrophagcs (1.0 X 10'' cells/ml)
was elevated with exposures to 1000 mg/ml particles with complexed metal (3.0
± 0.1, 0.2 ± 0.1, and 0.0 ± 0.0 V). Finally, 24 hours after inlratrachcal instillation
into 60 day old, male Sprague Dawley rats, inflamalion was significantly
increased after latex- Feu (lavage ncutrophils of 37 ± I 1%,  16 ± 9%, and  2 ± 3%
and lavage proteins concentrations of 0.51 ± 0.05, 0.26 ± 0.05, and 0.1 I ± 0.03
mg/ml). We conclude that metal initially complexed to the surface of a latex
particle can be associated with increments in oxidant generation and lung injury.

Asessment of Air Toxic Emissions from a Pilot-Scale
Combustion Unit
Richard A. Hargis and Henry W. Pennline
U.S. Department of Energy, Pittsburgh Energy Technology Center, P.O. Box
10940, Pittsburgh, PA  15236

    The flue gas cleanup projects in the In-House Research  program at the
Pittsburgh Energy Technology Center (PETC) range from laboratory-scale work
to testing with the combustion products  of coal at a  scale equivalent  to about
0.75 MW of electric power generation. The largest unit is  a 500-pound-per-hour
(227 kg/hr), coal combustor, complete with ductwork, spray dryer, bughouse,
and ancillary equipment. Over the past year, the unit has been upgraded and a
number of shakedown tests have been conducted  to prepare the unit  for investi-
gations of ha/.ardous air pollutants, also known as air toxics.
    These investigations of air toxics will be an integral component of the Air
Toxics/Fine Particulate Control Subprogram administered by the Office of
Project Management at PETC. The overall goal of this subprogram is to develop

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              33

-------
Abstracts


innovative control techniques capable of reducing emissions of air toxics and
fine particulale from sources utili/.ing fossil fuels to levels which result in
negligible or acceptable risks to human health and the environment. A major
effort within this subprogram to date has been the comprehensive assessment of
toxic emissions from selected coal-fired electric utility units. Based on the results
and conclusions from this effort, the study described in this paper Ibcusscd on
trace elements and halogens.
    The operation of a 500-pound-per-hour (PPH), pilot-scale combustion unit
has now been characterized in terms of the formation, distribution, and fate of air
toxics. The coal fired during the air toxics testing was from the same coal batch
that had been fired in a full-scale utility boiler during an earlier assessment of air
toxic emissions. A description of the pilot unit and operating conditions during
the air toxics testing is provided, along  with a summary of the test results. A
comparison ol the concentration of trace elements associated with the vapor
phase, solid/paniculate phase, and  paniculate si/.c fraction is made with the
results from the full-scale utility. Calculation of material balances around the
pilot combustion unit,  the baghousc, and the overall system as well as baghousc
removal efficiencies are also provided.  Based on the results of this air toxics
charactcri/ation effort, the 500 PPH unit has been shown to be a valuable facility
for comparison of sampling methods, development of continuous emissions
monitors, and evaluation of air toxics control systems.

Diurnal Variations in Particulate Mass and Total  Sulfate
Concentrations  at a Site in  Metropolitan Philadelphia
Junfeng (Jim) Zhang and Paul J. Lioy
Environmental and Occupational Health Sciences Institute,
UMDNJ-Robert Wood Johnson Medical School and Rutgers University,
681 Frelinghuyscn Road, Piscataway, NJ 08855
Helen H. Suh and Petros Koutrakis
Harvard University, School of Public Health, 665 Huntington Ave.,
Boston, MA 02115

    Most measurements of participate mass and sulfatc concentrations have
been based on samples collected over periods of 24 hours. An accurate exposure
assessment, however, requires diurnal distributions  of ambient paniculate matter.
As part of the Metropolitan Aerosol Acidity Characlcri/.alion Study, hourly
concentrations of PM 10, PM2.5. and total sullalc have been measured at a site in
Metropolitan  Philadelphia area during the summer of 1992. PMK) and PM2-.5
have been monitored using the Rupprechl and Patashnick model 1400 Tapered
Element Oscillation  Microhanlance (TEOM®). Semi-continuous (10 minute
time resolution) measurements of total sulfatc have  been done by using a method
named the Harvard Continuous Sulfate/Thermal Spcciation (CSTS) system.
    Examining diurnal variations in concentrations of total sulfate, PMIO,

34              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


PM2.5, and coarse fractions of PM10 (i.e., PMK)-PM2.5), we have observed
that total sulfate and PM2.5 had a similar diurnal pattern: the peak concentra-
tions typically occurred in the afternoon unless the previous day had had much
higher peak concentrations which might remain for a few hours up to the next
morning. We have also calculated monthly means (and c.v.) of paniculate mass
and sulfate concentrations for each hour of the day. Based on these monthly
means, we have conducted statistical analyses to study relationships among
different factions of paniculate matter measured, leading to a better understand-
ing of diurnal distribution of paniculate matter  in urban Philadelphia area.

A Field Assessment of the Effect of Volatile Constituents on the
Measurement of Ambient Particulate Matter Mass
Concentration
Michael B. Meyer
Rupprccht & Patashnick Co., Inc., 25 Corporate Circle, Albany, NY  12203
R. Neil Olson
State of Utah DEQ, Salt Lake City, UT

    Recently, airborne paniculate matter has been receiving increased attention
as published epidemiological health studies have heightened the concern over
associated human health-related issues. The current court-ordered review of the
paniculate matter standard in the United States is just one example that has
increased focus on the collection, measurement and identification of ambient
paniculate matter.
    With paniculate matter, unlike criteria gaseous pollutants, a potential exists
for what is defined as paniculate matter to change its mass as a result of ui.stable
residence on the filter due to light volatile substances (e.g. water, secondary
aerosols, aromatic hydrocarbons, organic carbon)  associated with the paniculate
matter and filler media. Some have speculated that paniculate matter retention
will become  an acute issue with the U. S. EPA planning to add a fine paniculate
matter (PM2.5) 24-hour and annual standard to the current PM 10 annual
standard in 1997.
    This paper discusses some of the critical aspects of paniculate matter
measurements. In addition, field data are  presented to identify the role that key
parameters such as sampling and equilibration temperature, relative humidity,
and filler face velocity may play in the c|iiantification of paniculate matter mass
concentration. Supporting measurements ol several paniculate matter chemical
constituents during episodic events are used to quantify their contribution.

Pulmonary Toxicity of Urban Air Particulate Matter (PM)
Kevin Dreher and Daniel Costa
NHEERL, U.S. EPA

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               35

-------
Abstracts


Alan Hoffman
NERL, U.S. EPA
James Banner
NIEHS, Research Triangle Park
A. Ost>rnit>-Vargas
National Cancer Institute, Mexico City, Mexico

    Identification of toxic constitucnt(s) of urban air PM would provide some
insight into the adverse health effects associated with exposure to this component
ol air pollution. We found that similar extent of acute pulmonary inflammation
could he produced in rats exposed to different amounts of various urban air PM
samples containing the same amount of transition metal. Additional in vitro
cellular toxicity studies performed on PM 10 samples obtained from Mexico City
identified (he most cytotoxic PM 10 sample to contain the highest amount of
transition metal, sullale and soluble transition metal. Subsequent in vivo toxicily
studies conducted on si/.e-lraclionatcd ambient air PM found the 
-------
                                                                Abstracts


 integrated realtime volummctric How control; particle mass collected on a filler
 allowing for later chemical analysis; ability to operate under wide range of
 ambient conditions (temperature, humidity, and pressure); a sharp cut at 2.5 mm;
 and mandated quality assurance requirements to insure high  field network
 precision.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               37

-------
                                                             Abstracts
                              Session 5
                             Canisters
           Chair: Eric Winegar, Air Toxics Limited, Folsoin, CA

Passive Flow Controllers: Revisited - Per EPA Method TO-15
R.A. Rasmussen
Oregon Graduate Institute, Department of Environmental Sciences and
Engineering, P.O. Box 91000, Portland, Oregon 97291-1000

    Abstract not available.

Fourteen-Day Stability of 27 Polar  VOCs In Summa® and
Silocan™ Canisters
Michael G. Winslow, Matthew M. Booth, and Dwight F. Roberts
Analytical Services Laboratory, Environmental Science & Engineering, Inc.,
(ESE), P.O. Box  1703, Gainesville, Florida 32602.

    Both SUMMA® and SilcoCan™  passivated stainless-steel canisters can be
used effectively to determine ppb and  sub-ppb concentrations of non-polar
volatile organic compounds (NPVOCs) in ambient air using the sampling and
analysis procedures described in Compendium Method TO-14. SUMMA®
passivation of the interior metal surface of a canister involves clectropolishing
techniques while SilcoCan™ passivation involves coating with a sub-micron
layer of inert fused silica. A comparison study was performed to determine the
stability of 27 polar VOCs (PVOCs) over a 14-day period in both SUMMA®
and SilcoCan™ canisters. PVOCs arc  generally more reactive than NPVOCs and
have a greater potential for adsorption witli active sites on the interior surface of
canisters.
    A gaseous standard containing a mixture of the  27 PVOCs at approximately
100 ppbV each in nitrogen was  prepared by Scott Specialty Gases in a high-
pressure cylinder. Pressuri/.ed canister samples (3 SUMMA® and 3 SilcoCan™)
were prepared at 5 ppbV by dynamic dilution of the  Scott mixture with humidi-
fied ultra-pure grade air. Canisters were analy/ed by GC/MS-SCAN. Sample
moisture management without coremoval of PVOCs was performed using an
Entech prcconccnlrator configured with a multi-stage trapping technique.
Samples were analyzed on Days 0, 3,  10, and 14. The temporal stability of
individual PVOCs is graphically presented by plotting their average relative
response factors against the  day analy/.cd.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              39

-------
Abstracts


Development of an Intelligent Canister/Cartridge Sampler for
the Collection of Ozone or Air Toxics
Mark K. Allen, Edward Miller, and Joseph Leair
Wisconsin Department of Natural Resources, Madison, WI 53707-7921

    In 1993, we designed and huill an intelligent sampler capable of collecting
eight canisters and eight DNPH-carlridges daily, to meet rigorous Enhanced
O/.onc Monitoring requirements lor Southeastern Wisconsin. Working from the
mullicanistcr/multicartidge sampler design, we have developed a smaller
sampling unit to collect samples 1'or o/onc precursors, reformulated fuel addi-
tives, and general air toxics. A site-based computer will track all samples from
the sampler setup, through sampling, to sampler take down. Relevant sampling
data such as initial canister vacuum, final canister pressure, and sampling How
rales are all recorded by the sampler. At the completion of the sampling episode
all sampling data are down loaded to the project database. Software for the
original sampler and new sampler has been revised to include internal and
external Quality Assurance procedures. We will report on the design of the new
sampler, important software revisions, and the sampler performance during an
initial season of sampling.

The  Use of Fused Silica Lined Canisters for Difficult Ambient
Air Samples
David Slielow
Restck Corporation, 110 Bcnncr Circle, Bcllefonte, PA 16823-8812

    Many compounds in ambient air have been difficult to collect and analyze
from SUMMA canisters. These difficult compounds include both polar and
sulfur compounds. Many polar compounds, such as alcohols, kctoncs, and
aldehydes are being considered for EPA Compendium Method TO-15. Sulfur
compounds such as hydrogen sulfidc, methyl mcrcaptan, dimethyl disulfidc, and
COS have been a constant struggle for analysts with SUMMA canisters.
        This paper describes work in which we used fused silica lined canisters,
SilcoCans, for the collection and analysis of polar and sulfur compounds. We
will also discuss the importance of an inert sample pathway for these com-
pounds. We will be discussing the use of a fused silica lined valve to ensure an
inert sample pathway. Data will he used to demonstrate the stability and holding
times of many of these polar and sulfur compounds within SilcoCans and
silcostcelcd valves.
40              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


PPBV Determination of Reduced Sulfur Compounds using
Glass Lined Canisters, Deactivated Tedlar Bags and Sorbent
Tubes: A Study of the Critical Parameters
R.R. Freeman, E.D. Winegar, D.S. Panda and C.C. Cruine
Air Toxics Lid., 180 B Blue Ravine Road, Folsom, CA 95630

    Many reduced sulfur compounds are quite reactive which often makes them
difficult to properly sample, transport and analy/.c. Current methodologies ulili/e
Tedlar bags as a sampling media and analysis based on gas chromatography
(GC) and selective detection; that is, (lame photometric (FPD) or sulfur chcmilu-
minescence (SCI) detectors. Existing methods caution the analyst to frequently
cheek the calibration and take every step to ensure system inertness. While the
overall performance of the analytical system is important, the lack of precision in
the measurement can be  attributed to the instability of many sulfur compounds in
Tedlar bags.
        This work investigates various means of collecting and transporting the
sample prior to analysis. Several new technologies arc available for collecting an
air sample containing reactive compounds. Among these are glass lined stainless
steel canisters, various new carbon based sorbenls and deactivated Tedlar bags.
        This work on sample stability ulili/.es GC/SCD because of the selectiv-
ity, sensitivity, cuqimolar response and linearity of the SCO. Several sampling
containers were investigated including untreated glass lined canisters, deacti-
vated glass lined containers, deactivated  Tedlar bags and newly developed
carbon based sorbents. Response factors arc used  to determine holding times for
various sulfur compounds. The relative influence of environmental factors such
as concentration, relative humidity and hydrocarbon background are included in
the investigation.

A Through-The-Probe Toxic Audit Study to Determine the
Effect of Moisture on Volatile  Organic  Compound Recovery
Rates in Stainless Steel Canisters
Alice Westeriner and Barry B. Re is man
Monitoring and Laboratory Division, California Air Resources Board,  California
Environmental Protection Agency. 1927  13th Street, Sacramento, CA 95814

    Through-lhc-probe audits of volatile organic compounds (VOC) have been a
part of the ambient toxic  program in California since 19K8. By introducing
known quantities of VOCs at the probe inlet, the audits check the integrity of the
field probe, the toxic sampler (leaks, contamination, etc.), the transport system
(sample canister), and the laboratory analysis accuracy of the canister contents.
Until 1993, auditing was performed using dry audit gases in dry, clean  stainless
steel canisters. Recovery rales for several aromutics, in particular, were low

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              41

-------
Abstracts


when compared to audit samples introduced directly into laboratory instruments.
To improve VOC recovery rates, a field study was conducted which compared
dry canister auditing with canisters injected with HPLC water (50 to  100 ul).
This  change improved the recovery of most compounds, yet not adequate for the
needs of an audit program.
    This paper contains a report on work to dale to extend the audit develop-
ment program to increase VOC recovery rates in SUMMA canisters by using in-
line humidification. An in-line bubbler was installed in the through-thc-  probe
toxic audit train to humidify the dry air which dilutes the toxic audit gases. This
approach more closely mimics ambient sampling conditions by humidifying the
entire audit train and sampler probe line, potentially improving VOC recovery
rates over the water-injected canister method. The goal of both studies was to
improve our ability to assess the data accuracy for the ambient toxics and PAMS
programs. Currently, the in-line humidification study employing dry sample
canisters is still  in  progress and yield results are incomplete. This paper  presents
the results of the wet-injected canister study.

Meeting the  Technical Acceptance Criteria for TO15
Methodology Using a New Canister Autosampler
Valerie J, Naughton, Edmund T. Lewis, and Joseph A. Borer
Tckmar-Dohrmann ,7143 E. Kempcr Road, Cincinnati, OH 45242

    Approximately 108 of the Ha/.ardous Air Pollutants (HAPs) specified in the
1990 Clean Air Act Amendments (CAAA) are classified as volatile organic
compounds (VOCs). Of the 108 volatilcs, nearly 35% arc oxygenated or polar
compounds. While more than one sample introduction technique exists for the
analysis of these air toxics, only canister sampling provides a medium for the
complete range  of analyses. A  broad concentration range of polar and non-polar,
and low and high molecular weight species can be analyzed  from canisters.
Additionally, duplicate samples can be taken from a single canister, samples can
be diluted, and storage stability in canisters for many of the compounds  listed
has been demonstrated.
     Until the introduction of Compendium Method TO 15, few quality assurance
/quality control  criteria were specified within air methods. This paper demon-
strates  the technical acceptance ojcctivcs outlined  in the TO 15 methodology
including mass spectral  ion abundance criteria, initial calibration, daily calibra-
tion, blank analysis, laboratory control samples, and method detection limits. A
novel technique for managing  high levels of moisture and carbon dioxide
commonly present in source samples is proven. Components analy/cd include a
representative mixture of polar and non-polar species from the TO 14, CLP
Statement of Work, and TO 15  analyte lists. The analytical system consists of a
Tckmar AutoCan Canister Autosampler interfaced to a Hewlett Packard 6890/
5972 MSD.

42              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


Improved Methodology for Analysis of Ambient Level
Reactive Sulfur Compounds
D.B. Car din
Entcch Instruments, Inc.

    Compounds with the formulas RI-S-R2 and RI-S-S-R2 have been very
difficult to sample, store, and analyze due to their highly reactive properties.
Having the same number of valence electrons as oxygen, sulfur lends to also be
polar thus favoring interactions with active sites on surfaces. The valence
electrons in sulfur atoms, however, arc not held as close to the nucleus and are
therefore much more iniluenced by locali/.ed charges. This decreases the
activation energy necessary for adsorption to surfaces. Shielding the sulfur atom
with bulky alkyl groups can reduce the reactivity of sulfur compounds through an
electron donating effect, and maybe even more importantly by the added steric
hindcrance offered by the adjacent C-H(x) groups  that reduce the acceptable
trajectory (cone of acceptance) for a potential rcaclant or reaction site. The
reduction in the steric hindcrcnce by replacing the  alkyl groups with hydrogens
probably accounts for the relative increase in reactivity in the following series:
    Dialkylsullldes < Mercaptans < Hydrogen Sulfide.
    Sampling, storing, and proconcentrating these compounds  requires the near
elimination of surfaces with ionic or highly polar characteristics (ie - active
sites). Low temperatures must also be maintained to reduce the population of
sulfur compounds with energy levels high enough  for reactions to occur. For this
reason, cold trap dehydration has been reported previously as a preferred method
for eliminating water with minimal sulfur compound loss.
    A modified, whole air approach for sampling and preconccntralion will be
presented that improves the recovery and analysis of ppb level mcrcaplans and
hydrogen sulfide by GC/MS. Three different How systems will  he compared for
their ability to reproducibly transfer sulfur compounds to the GC during
preconcentration, including one comprised to  nearly  100% teflon. Recoveries
will be presented using sorbcnl and cryogenic based prcconccnlration and water
management systems. A modified Cold Trap Dehydration procedure will be
introduced that reduces sulfur compound that arc normally occluded within
forming ice crystals in the cold trap. Stability in specially treated Canisters will
also be investigated.

Innovative Flow Controller for Time Integrated Passive
Sampling using SUMMA Canisters: Design, Testing and
Applications
Philippe Simon and Jean-Pierre Farant
Environmental Laboratories, McGill University, 3450 University, rm 23,
Montreal, Quebec, Canada, H3A 2A7

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               43

-------
Abstracts


Harold Cole
Boeing Analytical Services, Missile & Space Division, 499 Boeing Boulevard.
Hunlsvillc, Alabama 35824
Joseph Krasnec
Scientific Instrument Specialist, P.O. Box 8941, 815 Courtney Street
Moscow, ID 83843

    To restrict the entry ol gas inside evacuated vessels such as SUMMA
canisters, mechanical flow controllers arc used to collect integrated atmospheric
samples. From the passive force given by the pressure gradient, the motion of
gas can be controlled to obtain a constant How rate. Presently, devices based on
the principles of critical orifices arc used and they arc all limited to an upper
integrated sampling lime. We have recently developed a novel flow controller
which can he designed to achieve any desired sampling time when used on
evacuated vessels. It can extend the sampling time for hours, days, weeks or
even months for the benefits of environmental, engineering and loxicological
professionals. The design of our controller is made from computer simulations
done with an original set of equations derived from lluid mechanic and gas
kinetic laws. Sampling and analytical testing was performed, static and dynamic
validation studies were also made in the laboratory. Since now, the experimental
results showed excellent agreement with predictions from the mathematical
model, and validation studies have proved its validity. This new controller has
already found numerous applications. Units able to deliver a constant sampling
rate between vacuum and approximately -10 inches Hg during a continuous
seven days period have been  used with SUMMA canisters of different volumes
(100 ml,  500 ml, I liter and 6 I). From our research work, any combinaisons of
sampling times for any  sampler volumes arc made possible. The innovative How
controller contributed in air quality assessment of ambient air around a sanitary
landfill (indoor/outdoor), and inside domestic wastcwater and pulpmill sludge
treatment facilities. It is now proposed as an alternative methodology for
atmospheric sampling in the russian orbital station Mir. The development of this
device applied in sampling methods can now offer true long term passive
monitoring of selected gaseous air pollutants.
44               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts


                          Poster Session

Low-Level Analysis of Cr+6 in Air by Ion Chromatography
Frederick A. Freed
Lancaster Laboratories, Lancaser, PA

    A major difficulty in the analysis of air samples is in obtaining the sensitiv-
ity necessary to measure the analytc of interest. Many results obtained for the
analysis of air samples arc less than the limit of quantitalion. This is particularly
true for the analysis of hexavalent chromium (Cr+h) in air.
    A Dionex DX 500 ion chromalograph was configured to determine Cr+(' in
air. The  How rate of the mobile phase (0.2 M (NH4),SO4/(). 1 M NH4OH) was IA
mL/minutc and a post column addition of diphenylcarbohydrazidc (0.5 mL/
minute)  was made to complex the Cr+61'or detection at 520 nm.
    A guard  column (Dionex NGI) was installed prior to the separator column
    (Dionex  AS7). To increase sensitivity, a concentrator column (Dionex
AG7), which was loaded with 10 mL of sample prior to injection, was used
instead of a sample loop.
    When using the concentrator column, the limit of quanlitalion I'orCr"' in air
was reduced to 10 ng/L. This is nearly a 100 fold increase in sensitivity as
compared to using a 100 L sample loop. A linear calibration range to 500 ng/L
was demonstrated.

The Significance of Hydrocarbons in the C10-C2() Range to Mass
Contribution and Ozone Forming Potential  During the
COAST Study
Gregory Harslifleld, Barbara Zielinska, and Eric Fujita
Desert Research Institute, P.O. Box 60220, Reno, NV 89506

    In order to improve the technical basis for designing effective o/.one
strategies for the Houston-Galveston and Beaumont-Port Arthur-Orange areas of
southeast Texas, the Texas Natural Resources Conservation Commission
(TNRCC) sponsored the Coastal Oxidant Assessment for Southeast Texas
(COAST) project. Part of this study was to assess the significance of hydrocar-
bons in the C1(1-C1(| range and determine their potential for o/.one formation. Field
measurements of volatile organic compounds (VOC) included the collocated
sampling of canisters (C,-Cp range), and Tenax-TA (CS-C.,0 range) for ambient
and source measurements. To assess the significance of hydrocarbons in the C|n-
C,(1 range, data from collocated canister and Tenax-TA measurements were
merged and composite profiles developed. The profiles consist of data from Cy
C,(), with C, to n-octane obtained from canisters and n-octane to C,0 obtained
from Tenax-TA. The relative o/,one forming potential was developed by

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               45

-------
Abstracts


multiplying individual hydrocarbon weight amounts by the appropriate maximum
incremental reactivity (MIR) factor. These values were then normalized by
dividing through by the base reactivity (sum of all reactivities) to yield percent
contribution values.

The Application of High Speed Gas Chromatography to Air
Analysis
Norman Kirshen, David Coe, YUian Bao, and Elizabeth Almost
Varlan Chromatography Systems, 2700 Mitchell Drive, Walnut Creek,
CA 94598

    The requirement for continuous emissions monitoring for ambient or indoor
air pollutants at  industrial sites and ambient air at hazardous waste sites demands
very short analysis times. While Gas Chromatography is a major monitoring
technique, standard GC methods generally have long cycle times from 15 to 60
minutes. The demand for frequent sampling (2 to 5 minutes) approaching a
continuous mode has generated interest in high speed GC analysis. But con-
straints imposed by the use of narrow  bore (to lOOuM i.d.) columns, namely
sample capacity and instrument incompatibility, have impeded acceptance of this
technique for this application.
    This poster describes an apparatus which produces a minimum injection
bandwidth (10 to 50 msec) which provides the primary means of attaining good
column efficiency while achieving High Speed GC Analysis. This FastGC inlet
which is mounted on a Varian 3400 Gas Chromatograph is described and its
operation is detailed from the cyrofocussing of the sample onto a short Nickel
tube to the 100,000° C/sec dcsorption to ten X Meter x 0.32 mm capillary
column. With the very narrow injection bandwidth, high hydrogen carrier
velocities, and the short capillary column, the FaslCG inlet is shown to provide
analysis times of less than one minute for a number of applications of interest.  In
addition the high compound signal to noise ratios obtained requires sampling
times of only one to two minutes to obtain sensitivities approaching one ppbv.
    Applications described in the poster include chlorinated hydrocarbons, polar
organics, an ozone precursor standard (C2 to C13), and BTEX's.'Data regarding
the recovery, response linearity, and retention time and area count precision arc
also reported.

Photochemical Assessment Monitoring: Overview and Data
Analysis Activities
Nash O. Gerald and Mark Schmidt
U.S. EPA, Emissions, Monitoring and Analysis Division (MD-14), Office of Air
Quality Planning and Standards, Research Triangle Park, NC 27711
46               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts
    In accordance with the Clean Air Act Amendments of 1990, EPA developed
rules for the initiation of Photochemical Assessment Monitoring Stations
(PAMS) by Stale and local air pollution control agencies located in serious,
severe, and extreme ozone nonattainmcnt areas. These PAMS networks monitor
for a target list of volatile organic compounds (VOC) including several carbon-
yls, as well as o/.onc, oxides of nitrogen, and surface/upper air meteorological
measurements;.
    This poster will provide an overview of the monitoring requirements for the
PAMS program and provide a lew illustrations of data analysis techniques for
the substantial and complex data base generated by the array of PAMS located in
22 affected areas across the United Stales.

A Novel Method of Measuring Ambient Aerosol Aspiration
Efficiencies Using Phase Doppler Anemometer
Zhili Liu
National Research Council
Fit-Lin Chen and Russell W. Wiener
National Exposure Research Laboratory, U.S. Environmental Protection Agency
Alfred Eisner
ManTech Environmental Technology, Inc.

    In this study, a novel method has been developed to measure the aspiration
efficiency of a sampling inlet using mass balance principles. Phase Doppler
Anemomctry (PDA) measurements are performed in proximity to a thin-walled
sampling inlet. These measurements are non-invasive, requiring no physical
sampling instrument, thus producing no perturbations in the aerosol How. The
measurement area is described by spherical surface enclosing the inlet. The
velocity vector for each particle entering and leaving the sphere is measured. For
a given sampling condition the overall mass flux is /.ero. The aspirated aerosol
represents the fraction of the aerosol withdrawn from the environment. The
integrated aerosol characteristics such as parliculalc mass, number concentration,
and si/.c distribution remain unchanged during the sampling process.
    The aspiration efficiency of a sampling  inlet is the first potential factor that
can cause changes in aerosol characteristics  during a sampling process and can
contribute lo a non-representative sample. Changes in  the aerosol are caused by
the difference in inertial properties between  aerosol particles and gas molecules.
Particles may not follow the fluid stream lines, thus do not enter a sampling inlet
representatively causing an aspiration error.
    Traditionally, trajectory or comparison methods are used to measure the
aspiration efficiency of a sampling inlet. The trajectory method uses visuali/.a-
lion of fluid streamlines and direct observation of particle deviation to assess
aspiration. It requires laminar How conditions and has only been applied to
studying simple sampler systems. The comparison method requires two sets of

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               47

-------
Abstracts


measurements, first, isokinctic samples where (he fluid is disturbed as little as
possible and second, the sampling method of interest. The isokinctic samples are
assumed to have no aspiration error. The comparison method may be applied
over a wider range of experiments. Its major disadvantages include the confusion
of aspiration and apparent efficiencies  and the difficulty of generating large
amounts of uniformly distributed mono-dispersed particles in a large wind
tunnel.
    An argon-ion Phase Dopplcr Anemometer was uliii/.ed to measure the (lux
ofpolydispcrscd particles generated by a fluidi/cd bed system. The flux is then
converted into the aspiration efficiency by comparing the movement of small
particles, representing fluid streamline motion and the large particles possessing
sufficient inertia to generate an aspiration error. Isokinelic, sub-isokinctic, and
supcr-isokinctic sampling processes have been studied. The results arc in good
agreement with those of Belyaev and Levin (1974).' The major advantage of this
new method is that particles are generated locally and they do not need to he
mono-disperse and uniformly distributed across the lest area of the  wind tunnel.
Phase Dopplcr Anemomctry can distinguish the si/c and  the velocity of each
individual particle. The sampling duration of the experiments are, therefore,
significantly reduced and significant additional information is collected on  the
velocity vector distribution of the particles.

    I.  Belyaev S. P. And  Levin L. M.  (1974), "Techniques for collection of
representative aerosol samplers." J. Aerosol Sci. 5, 325-338.

Methods Development for Clean Air Act Amendment
Hazardous Air Pollutants from Stationary Sources:  Methanol
and Isocyanates
Frank  Wilshire
Air Measurements Research Division,  National Exposure Research Laboratory,
Research Triangle Park, NC 27711

    Passage of the Clean Air Act Amendments of 1990 by the U.S. Congress
delegated responsibility to the U.S. Environmental Protection Agency for the
development of methodology required to monitor and protect the environment
from 189 ha/.ardous air pollutants (HAPs) listed for regulation. Contained in the
list of the 189 HAPs arc mcthanol and  four isocyanatcs (2,4-tolucrie diisocyanatc
[TDI|; hcxamcthylene 1,6-diisocyanatc |HDI|; mcthylcncdiphcnyl diisocyanatc
|MDI|; and methyl isocyanate |MI|). The Methods Branch of the National
Exposure Research Laboratory in  Research Triangle Park, North Carolina has
recently developed the methodology required to monitor these HAPs in station-
ary source emissions. A summary of the methodology developed and projects
currently in progress will be reported at the Methods Branch poster session

48              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts
during the EPA/AWMA Symposium lor the "Measurements of Toxic and
Related Air Pollutants," May 7-10, 1996.    .
    Disclaimer: The information in this document has been funded wholly or in
part by the United States Environmental Protection Agency under contract with
Radian Corporation (Contract No. 68-DI-OOIO) and with Research Triangle
Institute (Contract No. 68-D1-0009). It has been subjected  to Agency review and
approved for publication. Mention of trade names or commercial products docs
not constitute endorsement or recommendation for use.
    Methods  Development tor Clean Air Act Amendment  Hazardous Air
Pollutants from Stationary Sources: Methanol and Isocyanates

VOST Charcoal Specification Study
Robert G. Fuerst
U. S. Environmental Protection Agency, Research Triangle Park, NC 27711
A.L. Foster and J.T. Bursey
Radian Corporation, Research Triangle Park, NC 27709

    The volatile organic sampling train (VOST) methodology incorporates SW-
846 sampling Method 0030 and SW-846 analytical Method 5040 or 5041.
VOST is currently one the leading methodology's available lor the sampling and
analysis of volatile principal organic ha/.ardous constituents (POHCs) and
products ol incomplete combustion (PICs) from stationary  sources at very low
levels. However, revisions to the original method are necessary to maintain
VOST as a viable regulatory tool. Method 0030 states  that  the VOST sampling
tube set must consist of a front tube containing Tenax® (a 2,6-diphcnylene oxide
polymer) and  a rear tube containing sequential bed of Tenax® and SKC Lot 104
petroleum-based charcoal "or equivalent."  However, the method does not
identify a specific equivalent, nor.does the method supply the performance
specifications which would allow determination ol an equivalent. Lot 104
petroleum-based charcoal is no longer commercially available and has not been
available for several years. Laboratories are presently using a wide range of
substitutes, usually coconut-based charcoal, and there is a wide range of perfor-
mance from batch to batch of charcoal in one laboratory and from laboratory to
laboratory. To provide performance specifications and  identify a replacement for
SKC Lot 104 charcoal, a VOST charcoal specification  study was initialed. The
following carbon-based candidate  sorbents were considered: Tenax-GR (a
graphiti/.cd Tenax); a Petroleum-based Charcoal; Ambersorb® XE-340 (hydro-
phobic carboni/.ed resin bead); Anasorb® 747 (beaded active carbon with very
regular pore size); Carbosicvc® S-III (carbon molecular  sieve); and  a Beaded
Activated Charcoal (BAC)(with a very regular pore size).
        The results indicated that Tenax-GR showed significantly poorer
performance than the other candidates in preliminary experimental results.
Ambersorb did not retain the gaseous volatile organic compounds tested as well

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               49

-------
Abstracts


as the others and recovery of vinyl chloride was very low at all levels of spiking.
Carhosievc was eliminated as a candidate replacement because of cost and
handling problems. The petroleum-based charcoal was eliminated because of
difficulties in handling a finely-divided powder. The availability of Anasorb 747
proved to be the deciding factor between it and the BAC. Performance, cost,
case of handling, and plentiful supply make Anasorb® 747 a good choice for
replacement ol'SKC Lot 104.
        The information in this document has been funded wholly or in part by
the United Stales Environmental Protection Agency under contract 68-DI-OOK)
to Radian Corporation. It has been subjected to Agency review and approved for
publication. Mention of trade names or commercial products docs not constitute
endorsement or recommendation for use.

Method Evaluation Study: The Application of SemiVOST to
the Nonhalogenated Semivolatile Organic Compounds from
the Clean Air Act Amendments
Merrill D. Jackson,
National Exposure Research Laboratory, U. S. Environmental Protection
Agency, Research Triangle Park, North Carolina 2771 I
James F. McGaughey, Raymond G. Merrill, and Joan T. Bursey
Radian Corporation. P.O. Box  13000,  Research Triangle Park, NC 27709

    A laboratory study and three field studies  were performed to evaluate the
application of the SemiVOST (EPA sampling  Method 0010, EPA sample
preparation Draft Method 3542, and EPA analytical Method 8270) to the
semivolatile halogenated organic compounds (approximately 80 compounds)
listed in the Clean Air Act Amendments (CAAA) of 1990. In these initial
studies,  PCBs, dioxins, and pesticides  were excluded.  Subsequently, a labora-
tory study was performed to assess the  feasibility of the application of the
SemiVOST methodology to the semivolatile nonhalogcnated organic compounds
listed in the CAAA (approximately 70 compounds). Several of the semivolatile
organic  compounds were eliminated from consideration as SemiVOST analytes
because they could not be analy/.ed successfully by gas chromalography/mass
spectromelry (GC/MS), reacted with other compounds in solution, or were
insoluble in methylenc chloride at the levels required to perform dynamic
spiking  in the field. The remaining CAAA semivolatile organic analytes were
grouped as acid/neutrals and base/neutrals and evaluated in the field  using the
guidance of EPA Method 301 for experimental design and statistical evaluation
of the data. Quadruple SemiVOST trains were run in the field, with dynamic
spiking of the semivolatile analytes from a mclhylcnc chloride solution of cither
acid/neutrals or base/neutrals into two of the sampling trains. The bias and
precision of the overall SemiVOST methodology (sampling, sample  preparation,

50               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


and analysis) applied to each of the semi volatile organic analytcs were evaluated.
    The information in this document has been funded wholly by the United
States Environmental Protection Agency under contract 68-D4-0022 to Radian
Corporation. It has been subjected to Agency review and approved for publica-
tion. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

Development of a  Source Test Method for Measurement of
Mercury Species
Peter M. Grohse and James A. O'Rourke
Research Triangle Institute, Research Triangle Park, NC 27709
Joseph E.  Knoll and Frank W. Wilshire
Methods Branch, National Exposure Research Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

    The lack of a validated spcciation method for stack emitted mercury,
according  to the Draft Report to Congress, is a critical information gap in
methodologies to assess human and ecological exposures (and consequently
risks) associated with such mercury. Because no validated method exists, current
assessments must rely on unvalidated assumptions as to the species distributions
to estimate environmental concentrations and ultimately exposures. The present
study deals with research on the development of a method to spccialc mercury
emissions.
    EPA Method 29 and modified versions were studied to assess their mercury
speciation  capability. It was speculated that the bulk of ionic mercury species
should be collected in the upstream acidified peroxide impingers due to the high
solubility in water of HgCI,, while the relatively insoluble elemental mercury
should pass through  to  the acidified permanganate collectors. The studies were
carried out using simulated Hue gas which included sulfur dioxide (l()()()-2()()()
ppm), hydrochloric acid (100 ppm), water vapor (10%), carbon dioxide (10%)
and nitric oxide. Ionic and elemental mercury vapors were obtained using
permeation tubes.
    Initial tests with Method 29 indicated that small amounts (3%) of Hg" were
collected by the hydrogen peroxide/nitric acid impingers. This percentage
increased with increasing sulfur dioxide concentrations. However, when water
was substituted for the  acidified peroxide reagent in  the first two impingers, the
specialion  was very clean, with greater than 99.9% of Hg" passing through the
water impingers to be collected in the permanganate even when sulfur dioxide
was present. Details  of this study  are given.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               51

-------
Abstracts


FT-IR Transmission Spectroscopy for Quantitation of
Ammonium Bisulfate in Fine-Particulate Matter Collected on
Teflon® Filters
K.J. Krost & W.A. McClcnny
US EPA, Research Triangle Park, NC 27711

    A quantitative measurement method for the determination of fine-particle
hisulfte in ammonium bisullate collected from the ambient air onto Tcllon®
filters is described. Infrared absorbahcc measurements of the Teflon® filters are
made before and after particle collection. Subtraction of the two spectra reveals
the absorbancc spectrum of the panicles. The presence of bisulfate is identified
by characteristic and unique spectral features including prominent absorption
bands at 1050 and 870 cm ' is used for quantilation. The lower limit of detection
for the bisulfate ion is 150 nanomoles. This amount corresponds to the ammo-
nium hisulfale which would be collected from an air volume containing 1.2-
microgram/m' samples for 24 hours at  10.0 L/min. This method provides a
specific, nondestructive, direct measurement of ammonium bisulfate.

Organic Compound Delivery Device
Joel Daughtridge
National Exposure Research Laboratory, AMRD, MD-44,
Research Triangle Park, NC 2771 I

    The design and implementation of a transferring device for delivery of
organic gas phase sample to GC or GC/MASS SPEC instrumentation. The
device is simple in design, economical  to build, with low maintenance require-
ments. The device is easily installed with minimal downtime to instrumentation.
Testing of design is conducted using a specific range of selected organic
compounds. Components of design were assembled from  " off-the-shelf"
purchases of materials. The main  objectives include valve options, primary trap
design, carrier-gas How rates, temperature  parameters, and method development
as appropriate. The end product is specifications and Assembly instructions
coupled with standard operating procedures (SOPs) available upon written
request from EPA, NERL, R.T.P.
    The use of this device could greatly increase the consistency and validity of
analytical data between laboratories, especially multi-lab studies.
    The target audience would be "all" laboratories using GC/MS instrumenta-
tion for sample analysis.
    Disclaimer: The information in this document has been funded wholly by
the United States Environmental Protection Agency. Mention of trade names  or
commercial products docs not constitute endorsement or rccomcndation for use.
52              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


Marine Crustaceans as Indicators of Estuarine Health - The
Atlantic Blue Crab (Callinectes sapidus)
Joseph E. Bumgarner
U.S. EPA, MB/AMRD/NERL, Research Triangle Park, NC 27711

    In 1993 AREAL, later NERL, in conjunction with N.C. Stale University,
School of Veterinary Medicine , N.C. State Government DEHNR/Div. of Marine
Resources, Dr. Marius Browcr Duke University School of the Environment and
Region III, began a multimedia study centering around the Blue Crab and other
estuarine media to develop multimedia analytical methodology and to determine
if selected marine crustaceans could serve as bioindicators of estuarine deteriora-
tion. The Blue Crab was selected because of the association of a bacterial
mediated disease (soft shell disease) and water pollution.  The  questions posed in
the study were; is water pollution affecting the immune systems of marine
Crustacea; are chemical pollutants in the water affecting the development of the
chitinous exoskeleton of crustaceans making them more susceptible to attack by
chitinoclastic  bacteria; what role do trace elements play in this process, and,
would selected marine crustaceans serve as bioindicators  of a deteriorating
estuarine environment.

Multimedia Analysis by Pyrolysis/Gas Chromatography/Mass
Spectrometry
Myriam Medina- Vera
U.S. Environmental Protection Agency, MD-44,
Research Triangle Park, NC 27711

    The analysis of organic species in environmental samples is usually complex
and time consuming. The types of techniques and methods used arc selected
depending on  the matrix and the target compounds being analy/.ed. The develop-
ment of fast, precise, reliable,  flexible and ideally multimedia analytical proce-
dures is desirable. Pyrolysis coupled to  a gas chromalograph/ mass spectrometer
(Py/GC/MS) has these characteristics.
     Pyrolysis can be applied to synthetic polymer chemistry,  natural products,
microbiology, etc. Today platinum (Pi) filament instruments can be easily
coupled to a GC and used to perform classical pyrolysis and/or thermal desorp-
tion of the compounds from a  given matrix. The technique can be applied to the
analysis of soil, sediment, tissue, sewage, and other matrices offering a unified
approach to environmental analysis. Some matrices studied in  the laboratory
include tissue, sediment and particulate matter.
    This work evaluates Py/GC/MS as  a fast economical  screening technique for
polycyclic aromatic hydrocarbons (PAHs) in sediment samples. The effects of
different temperatures, gas Hows and limes, solution port  injections and interface

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996               53

-------
Abstracts


injections, packed quart/, tubes, open boats and ribbon volatili/.ation were
considered. The PAHs (ranging from naphthalene to bcn/.o[a]pyrcne) were
thermally desorbcd from sediment samples without any prctrcatmcnt. The
compounds were identified by retention times, quantitation masses, and mass
spectra fit/purity and reverse lit/purity ratios. The results arc reported at a 95%
confidence level.
    Disclaimer: The information in this document has been funded by the
United States Environmental Protection Agency. The pyrolysis equipment has
been provided by CDS Analytical Inc. Under assistance agreement CR# 0021-
92. Mention of trade names or commercial products docs not constitute endorse-
ment or recommendation for use.

Application of Thermodynamic Parameters to Gas
Chromatographic Methods Development
Lynn H. Wright
AMRD/MB MD-44, National Exposure Research Laboratory,
U. S. Environmental Protection Agency, Research Triangle Park, NC 2771 I

    The use of thermodynamic parameters to predict gas chromatographie (GC)
behaviour has been developed theoretically, but has not been widely applied due
to a number of technical difficulties. In theory, liquid phase and compound
thermodynamic parameters are transportable between columns of different
physical dimensions, calculable from liquid phase and compound physical data,
directly applicable to the prediction of retention times for both isothermal and
temperature programmed GC in normal, multi-dimensional, or multi-chromatog-
raphy applications, and potentially applicable to the verification and/or extrac-
tion of additional information from historical data.
    Work in this laboratory demonstrated accurate prediction  of GC retention
limes in temperature programmed serial GC columns (multi-chromalography),
providing a means by which to "dry lab" develop multiple phase GC methods for
complex environmental samples.
    Work in progress includes the development of an accurate approximation
ufgorithm for high speed calculation of temperature programmed'GC retention
times, the development of an assumption-free method for the experimental
determination of GC column dead volumes and across column comparison of
experimentally determined thermodynamic parameters for several common
liquid phases and a limited set of target compounds.
    The information in this work has been funded wholly by the United States
Environmental Protection Agency. It has been subjected to Agency and ap-
proved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
54              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


Expanding the Use of Passive Sampling Devices (PSDs)
Jim Mullk and Jerry Varns
Methods Branch (MD-44), Air Measurements Research Division, National
Exposure Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711

    Passive sampling devices (PSDs) are small, inexpensive and easy-to-use
collection devices for volatile pollutants that do not require electricity for their
operation. These features have made PSDs very attractive for remote outdoor
site monitoring where no power is available and for personal exposure / indoor
air monitoring situations where the attributes of a small, quiet and unobtrusive
monitor are particularly important. The PSDs have furnished time-weighted
measurements, ranging from an eight to 168 hr duration, with good precision and
accuracy for air pollutants such as o/.one, the oxides of nitrogen, and sulfur
dioxide.
    Despite airshed modeling efforts I'or regulating and controlling iropospheric
ozone (O,) formation, over 70 million people currently live in  U.S. counties
which exceed the National Ambient Air Quality Standard (NAAQS) set for O,.
Under the objectives of an EPA internal grant awarded in FY96, an interdiscipli-
nary learn will supplement an existing high o/.one network with placement of 0,-
PSD sites at strategic gaps in the modeling grid. This PSD application is ex-
pected to provide a longer term, more highly resolved data base of regional 0,
measurements, a scenario that is not currently plausible because of economic
constraints. Importantly, more attention to non-urban phenomena will be
necessary now that the Clean Air Scientific Advisory Committee (CASAC) of
the U.S. EPA recommended an cighl-hr O, exposure standard replace the  current
one-hr standard. Interested parties are invited to concomitanlly test I'or oilier O,-
relaled chemical species during this aggressive two year project.

Research and Development of a Stationary Source Method
for Phosgene
Easter A. Coppedge and Larry D. Johnson
U.S. Environmental Protection Agency, National Exposure Research Laboratory,
Research Triangle Park, NC 2771 1
Joetle L. Steger
Radian Corporation, Post Office Box 13000, Research Triangle Park, NC 27709

    Phosgene is listed as one of the ha/.ardous air pollutants in Title I of the
Clean Air Act. Phosgene is a highly toxic gas at standard temperature and
pressure and has been used in military applications and for a variety of  industrial
uses. Although various methods have been developed for detection of phosgene
in ambient air, no method is directly applicable to stationary source emissions.

Measurements of Toxic and Related Air Pollutants,  May 7-9, 1996               55

-------
Abstracts
The EPA has an on-going research project to develop a field ready protocol for
phosgene from stationary source emissions. The results of the derivation studies,
sampling train experiments and other laboratory work will he shown.
    The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under contract 6S-D4-0022 to
Radian Corporation. It has been subjected to Agency review and approved for
publication. Mention of trade names or commercial products docs not constitute
endorsement or recommendation for use.

Applications for Using Exhaled Human Breath as an  Exposure
Assessment Tool for Volatile Organic Compounds
Joachim Pleil
U.S. EPA, Research Triangle Park, NC

    The presence of specific compounds in human blood can indicate a recent
exposure (or biological response) to a drug or an environmental pollutant, or
indicate a disease stale of the individual. The direct measurement of blood for
various substances and biomarkers is one of the primary medical diagnostic tools
in current use. Because the organic constituents of exhaled human breath arc
representative of their blood-borne concentrations through gas exchange in the
blood/breath interface in the lungs, a breath measurement could conceivably
replace a direct blood measurement. A classic example of this is -the
"brcathaly/er" test for  inebriation from ethanol.
    For volatile organic compounds (VOCs), sampling and analysis of breath is
preferred over direct measurement from blood samples because breath collection
is non-invasive, potentially infectious waste is avoided, and the measurement  of
gas-phase analytcs is much simpler in a gas matrix rather than a complex
biological tissue like blood. To exploit these advantages for the assessment of
exposure to environmental pollutants, we have developed a simple direct
collection method for individual alveolar breath samples and adapted conven-
tional gas chromatography mass spcctrometry analytical methods for'tracc level
VOCs analysis. The sampling method is based upon direct exhalation of alveolar
air into a I liter volume stainless steel canister with an internally passivatcd
surface. Subsequent analysis is performed in the laboratory using gas chromalog-
raphy - mass spcctrometry (GC-MS) methods especially modified to accommo-
date the high concentrations of water and carbon dioxide in breath:
    The methodology  has been applied to a variety of exposure situations.
Specifically, we have explored exposure to environmental tobacco smoke,
chloroform from chlorinated municipal water and swimming pools, vinyl
chloride and 1,2-dichlorocthylcne from contaminated well water, various
solvents and propellants in consumer products, and methyl tertiary butyl ether
(MTBE), ben/ene and  toluene  from refueling of automobiles. We  have measured
the time dependent: elimination of VOCs through  collection of scries of samples

56               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


subsequent to an exposure and have e applied a multi-term exponential decay
model to extract information on comparlmental lime constants and total dose.

Solid Adsorbent Sampling for VOC  - Application in Ostrava,
Czech Republic
Maribel Colon and Thomas A. Hartlage
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Christopher R. Fortune
ManTcch Environmental Technology, Inc. Research Triangle Park, NC

    Seventy-nine samples were collected in an industrial city in the eastern part
of the Czech Republic for VOCs. Samples were collected at two different sites: a
school yard and in a residential area. Both sites were close to industrial plants
with coke emissions. Samples were loaded on multiadsorbent tubes packed with
70 mg of Carboxen® 1000 (60/80 mesh) and 160 mg of Carbolrap® graphiti/.ed
carbon black (20/40 mesh). The 79 samples were collected as singlets, tandems,
duplicates, triplicates, and field blanks. Tubes were connected with a Tellon
Swagelock union to Teflon tubing thai was in turn connected to an Alpha 2 pump
set at 33.0 ml/min I'or 60 min producing a sample volume of 2 L. Samples were
taken at both sites three limes a day. At one site side-by-side samples were taken
in conjunction with the Ostrava Hygiene Laboratory. For shipping, tubes were
wrapped in aluminum foil, scaled in a capped glass container for identification
purposes, and stored in a clean paint can with activated carbon to avoid contami-
nation. For the tube analysis, a Pcrkin-Elmer Automatic Thermal Desorption
System (ATD 400) with a PE AutoSyslcm gas chromatogragh  FID/ECD was
used. As suspected, levels of ben/.ene and toluene were higher than other
pollutants. Results for the duplicate and triplicate sample data show good
correlation between the sampling tubes.

Open Path Monitoring for Trace Gases
William A. McClenny
Methods Branch, Ambient Measurements Research Division, National Exposure
Research Laboratory, U.S. EPA

    There are lour EPA methods development objectives for ITIR-based open
path monitoring systems. These objectives are:
     I.   Development of a guidance document to provide the basis for speclro-
        scopic measurements of trace gases in the infrared.
    2.   Development of a method (TO-16) for the EPA Compendium of
        Methods for the Determination of Toxic Organic Compounds in
        Ambient Air.
    3.   Contributions to an international consensus on acceptable performance
	criteria for the systems used for FTlR-bascd monitoring.	
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              57

-------
Abstracts


    4.  Sponsorship for the Development of" a Standard Spectral Database for
        Infrared Spectra by the NIST.
    Item I is now completed while items 2 & 3 arc in the final stages. The item
4 has  been initiated and is expected to be a topic of long term interest for the
EPA.

Monitoring for VOCs in Air - Status  Report
William A. McClenny
Methods Branch, Air Measurements Research Division, National Exposure
Research Laboratory, U.S. EPA

    Recent methods development efforts for volatile organic compounds in air
have included:
     I.  The placement of one methods description TO-15 on the the AMTIC
        Bulletin Board maintained by the Office of Air Quality Planning and
        Standards and the preparation of two others, TO-16 and TO-17. TO-15
        describes the use of a canister-based sampling procedure and a GC/MS
        analytical procedure to monitor VOCs. Method TO-16 concerns the use
        of FTIR-bascd open path monitoring and the Method TO-17 involves
        the  use of solid adsorbents for samp I ing of VOCs.
    2.  The publication of several journal article publications in support of
        methods development for the North American Research Strategy for
        Tropospheric O/.one (NARSTO) and the Photochemical Assessment
        Monitoring Stations (PAMS) network. These publications show how to
        monitor VOCs with better sample integrity than in the past, particularly
        with respect to polar VOCs.
    3.  The publication of several journal articles on the use open path moni-
        toring of VOCs in lenceline monitoring applications using FTIR-bascd
        systems. Emphasis has been placed on understanding and treating the
        generation of background spectra.
    4.  Evaluation of a new automated gas chromatographic system as part of
        the  NARSTO contribution to the 1995 Southern Oxidant Study (SOS)
        Nashville Field Study. Diurnal patterns showing hourly updates for
        several biogcnic compounds were seen for the first time.

Methods for Measurement of Dislodgeable Pesticide Residues
in and Around the Home
Robert G. Lewis
National Exposure Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711
David E. Camann
Southwest Research  Institute, San Antonio, TX 78228

58              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                Abstracts
Marcia Nishioka
Baltcllc Memorial Institute, Columbus, Ohio 43201
John W.  Roberts
Engineering Plus, Redmond, WA 98053

    Surface residues of pesticides in residential environments present a signifi-
cant exposure risk for infants and toddlers, who frequently crawl or lie on the
floor and mouth their hands and other objects. Guidelines for assessing this
potential  exposure are mandated under Subdivision K of Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA). Both acute exposures to freshly-
applied pesticide residues and chronic exposure to pesticides in house dust arc of
concern,  although the latter may be a more important and primary route of
exposure for small children.
    Several monitoring devices and methods have been developed, tested, and
arc being standardi/.cd.  A high-volume sampler capable of collecting house dust
in an accurate and reproducible manner and in quantities sufficient for complex
chemical analysis has been developed and already adopted as an ASTM stan-
dard. Another device rolls a pliable polyurethane  foam  pad over carpeted floors,
lawns, and other surfaces in a manner that replicates the pressure and movement
of crawling infants and  toddlers in order to estimate the residues that may be
picked up by the skin. Human volunteers have also been used to measure
dislodgeable residues by pressing their hands on both untreated and freshly-
treated carpets. A hand  wipe procedure that is very acceptable to small children
has been  developed and field-tested. Further research and development has been
undertaken in an effort  to find a saliva surrogate that will provide a better means
of determining the potential for residues picked up on the skin to he transferred
to the mouth and ingested.
    Pesticides applied to lawns may also enter the home via track-in on shoes
and clothing, spray drift, and soil resuspension. Once inside, they can become
associated with house dust and accumulate in carpets, upholstered furniture,
draperies, and on household surfaces. Protected from degradation by sunlight,
rain, temperature extremes, soil microbes, and wind dispersion, they may persist
I'or years, as opposed to days or weeks on outdoor surfaces. Several small field
studies have been conducted or arc underway to assess the potential for transfer
of lawn-applied chemical to the indoors. Controlled studies were conducted  to
determine the extent of track-in of lawn herbicides, insecticides, and fungicides
on shoes  at various times after application. Other studies have been carried out at
homes where residents have applied pesticides themselves or employed commer-
cial applicators. Multimedia monitoring was performed to determine the relative
importance of the various routes of transfer (('.,1,'., track-in, spray drift) of lawn-
applied pesticides to the indoors.
    Several components of this research arc covered, monitoring tools de-
scribed, and selected findings presented.

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996               59

-------
Abstracts
Source Sampling Participate Matter Research
Thomas E. Ward
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
S.H. Dawes and W.E. Farthing
Southern Research Institute, Birmingham, AL 35255

    Research to determine procedures for measurement of emissions from
stationary sources of in-slack  particulatc matter of aerodynamic diameter
nominally  2.5 micrometer, 2.5um, and smaller (PM-2.5) was performed. Interest
currently exists in ambient air levels of PM-2.5. If regulation of ambient air PM-
2.5 levels occurs, measurement ol these emissions from stationary sources to
determine  their contribution to the ambient levels may be of interest also to
regulatory agencies. Stationary source PM-2.5 emissions control device research
is an additional potential application of this technology.
    The results of this work describe  a procedure using a small cyclone for
performing particle si/.ing to determine in-slack PM-2.5. The small cyclone can
be attached to existing source  sampling hardware. The dimensions of the small
cyclone are described. A constant sampling rate procedure is described. The
procedures for obtaining the sample including determination of How rates, actual
particle si/,c measured and other necessary measurement parameters arc de-
scribed. Field source sampling research will be necessary for evaluation of the
procedures.
    The information in this document has been funded wholly or in part by the
United Slates Environmental Protection Agency under contract to Entropy
Environmentalists, Inc., and Southern Research Institute.  It  has been subjected to
agency review and approved for publication. Mention of trade names or com-
mercial products does not constitute endorsement or recommendation for use.

Performance Characterizations of Laboratory and Field
Technologies for Quantitation of Lead in Paints, Dusts,
and Soils
Sharon L. Harper
National Exposure Research Laboratory, U.S. EPA, Research Triangle Park, NC
William F. Gutknecht and Robert  W. Vanderpool
Research Triangle Institute, Research Triangle Park, NC

     In support of the Title X  mandate for the federal government to establish
guidelines for lead-based paint ha/.ard evaluation  and reduction, performance
evaluations arc being conducted for established and new technologies.
    A modular approach has been taken for development/evaluation of labora-
tory and field lead (Pb) measurement  procedures, where analysis techniques,
extraction  methods, and collection methods have been evaluated sequentially.

60              Measurements of Toxic and Related Air Pollutants, May 7-9,  1996

-------
                                                                Abstracts


Quantitative extraction techniques have been developed and paired with various
laboratory-based and field-based analysis methods resulting in a wide array of
methods with equivalent performance. Testing protocols are being developed
that will predict field behavior in a laboratory setting using publicly-accessible
quality assurance materials (when available).
    Performance of the analytical methods are being characleri/.cd through
evaluations which include the determination of bias, precision, interferences,
sources of error, working range, and throughput/cost. Methods under multimedia
evaluation include (1) quantitative laboratory-based methods of atomic absorp-
tion and inductively-coupled plasma atomic emission spcctrometrics, (2) semi-
quantilativc/scrcening field methods of portable X-ray fluorescence and lest kits,
and (3) the low-cost quantitative field methods of colorimelry, anodic stripping
voltmctry, and reflectance. Extraction methods have included  hotplate-, micro-
wave-, and ultrasonic-based methods.
    Although Pb has been the primary research focus in the mcdias of paint,
dust, and soil, the methods selected have been developed, when possible, with
multielement capability. Additionally, many of the methods have the ability to
unaly/.c Pb and other trace elements in samples from additional medias, such as
amhicnl air, water and sediments. Performance evaluations arc still needed to
optimi/.e and  fully characteri/.e these methods for multielement and other media
applications.
    Evaluation of the chemical composition and si/.e distribution of individual
house dust particles in a sample is being conducted by embedding scanning
electron microscope (SEM) stubs into floor mats to allow SEM/XRF analysis on
undisturbed houscdust samples. Additionally, to better characleri/.e dust collec-
tion techniques, a dust generation/sampler characlcri/.ation system is under
development that will have the ability to consistently generate test surfaces
containing deposits of known dust characteristics. Five  to ten si/.e classes
covering the si/.c range of approximately I to 250 urn physical diameter will be
used for the production of the test surfaces and characterising  the generation
system's performance. Ultimately, real-world houscdust dispersion is planned.
    Development of Research and Standard Reference  Materials (SRM) for Pb
in paint and houscdust is nearir.g completion. These materials  are being devel-
oped to evaluate the accuracy of measurements being conducted.  Eight materials
are being produced, with the  first material, powdered paint SRM  2582 (0.02%
Pb), having been released for public use in  1994.
    Additionally, protocols for producing method evaluation materials for paints
and housedusts have been completed.
    The information  in this document has been funded  wholly or in part by the
United States Environmental Protection  Agency under contract 68-D5-0040 to
the Research Triangle Institute. It has been subject  to Agency review and
approved for publication. Mention of trade names or commerical  products docs
not constitute endorsement or recommendation for  use.

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996              61

-------
Abstracts


Evaluation of Selected Multiadsorbent Tubes for Monitoring
TO-14 Volatile Organic Compounds in Ambient Air
Gary B. Howe and R.K.M. Jayanty
Research Triangle Institute, P.O. Box 12194. Research Triangle Park, NC 27709
Irene 1). DeGraff and William R. Betz
Supelco, Inc.. Supelco Park Bcllefonte, PA 16823-0048

    US EPA Compendium Method TO-14, which involves collection of whole
air samples in passivatcd canisters followed by gas chromatographic analysis,
continues to he a widely used approach for monitoring volatile organic com-
pounds in ambient air. An alternative to TO-14 has recently been developed
(draft Compendium Method TO-17) by The Pcrkin-Elmcr Corporation in
cooperation with the US EPA. This new method involves pumped sampling onto
sorhcnt lubes and analysis by  thermal dcsorption and capillary gas chromatogra-
phy.
    Under contract  to Supelco, Inc., RTI has evaluated two different
multiudsorhcnl tubes for use in Method TO-1 7. Each lube type was leslcd by
sampling a 39 component mixture of TO-14 compounds at a nominal concentra-
tion of 10 parls-per-billion. Compound recovery and sorbent lube breakthrough
were determined for three different sample volumes and  three different relative
humidities. Sample analysis was performed by thermal dcsorption with capillary
gas chromatography and (lame ioni/.ation detection.
62              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


                             Session 6
             Atmospheric  Mercury Monitoring
    Cochairs: Robert K. Stevens,  U.S. EPA, Research Triangle Park, NC;
           and Thomas D. Atkeson, Florida DEP, Talahassee, FL

Overview of the South Florida  Atmospheric Mercury
Monitoring Pilot  Study
Thomas D. Atkeson
Department of Environmental Protection, 2600 Blair Stone Road,
Tallahasscec, FL

    In 1989, elevated  levels of mercury were discovered in Florida freshwater
fish—resulting in an unprecedented issuance of health advisories to Florida
fishermen. After extensive monitoring, approximately two million acres of
Florida surface waters - over half the fresh waters of the stale - are now subject
to bans or restrictions  on consumption of several species of fish. The most severe
problem is located in south Florida, where the entirety of the Florida Everglades
is covered by consumption bans. To begin to assess the potential  significance of
atmospheric deposition of mercury to Florida's water bodies and  watersheds,  in
1991 the Florida Department of Environmental Protection initiated the Florida
Atmospheric Mercury Study (FAMS) to quantify the temporal and spatial
variability in atmospheric deposition across Florida. At the present time insuffi-
cient data is available  to make responsible conclusions on sources of Hg mea-
sured at the FAMS sites. The FAMS system design has served to assess the
spatial and temporal trends in deposition. The long integration periods used by
FAMS arc not ideal for determining the mechanisms that distribute mercury from
local, regional or global sources. In 1995 the US Environmental Protection
Agency and the Florida Department of Environmental Protection along with
Florida Stale University, the University of Michigan, the Electric Power Re-
search Institute, and Florida Power and  Light designed a field air quality
monitoring and modeling study that was implemented in south Florida in August
and September of 1995. The principle objectives of this pilot study were to
obtain wind How patterns, source and ambient chemical composition data to
evaluate the ability of  source and receptor models to determine the transport
mechanisms of Hg to the Everglades. This presentation will expand on the
rationale for this study and the studies implication to our understanding of the
fate and transport of Hg to Eco-systems.

Preliminary Phase of Vapor Phase Mercury Measurements
Made During the  SUAMP Project
J. Timothy Dvonch, Matthew S. Landis, and Gerald J. Keeler,
The University of Michigan Air Quality Laboratory, Ann Arbor,  Ml 48109-2029
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              63

-------
Abstracts


Danny France
U.S. EPA Region IV Environmental Services Division, Athens, GA 30613
Kenneth P. Larson
Broward County Dept. of Natural Resource Protection,
Fl.Lauderdale, FL3330I
Scott W. Rhoney
North Carolina Dept. of Env., Health and Nat. Resources, Raleigh, NC 27607

    Total vapor phase mercury (Hg) measurements were made continuously at
five sampling sites for a one month period as part of the South Florida Urban
Atmospheric Mercury Project (SUAMP). The primary objectives of the SUAMP
project arc to quantify the atmospheric contribution of the Hg loading to South
Florida (Dade and Broward Counties) and to investigate the importance of Hg
emissions from local urban sources relative to the regional anthropogenic and
global sources. The i'ielcl study portion of this project look place from August 6
to September 6, 1995. Total vapor phase Hg measurements were made at five
sites utili/.ing two collection methods: integrated 12-hour samples  using Au-
coatcd glass bead traps and continuous measurements using automated Hg vapor
analy/.ers. Co-located measurements compared very well, with agreement
between the two methods determined to be less than six percent. Monthly-
integrated mean concentrations at the five sites ranged from 1.3 to 4.0 ng/m\
Although this is a considerable range in the monthly mean concentrations, daily
site to site variations were found to be even greater. In some instances, there
were periods (I -2 hours) during which vapor phase Hg concentrations at
individual sites exceeded 15 ng/m', with maximums reaching  50 ng/m'. A
preliminary analysis of this data set including a limited evaluation  with surface
meteorological data will be presented.

Atmospheric Deposition of Mercury in South Florida:
Preliminary  Results from the SUAMP Project
J. Timothy Dvonch and Gerald J. Keeler
The University of Michigan Air Quality Laboratory, Ann Arbor, MI 48109-2029

    Event precipitation samples were collected concurrently at seventeen
sampling sites for a one month period as part of the South Florida  Urban
Atmospheric Mercury Project (SUAMP). A-primary objective of (he project is to
determine the magnitude of the atmospheric contribution to the Hg loading to
South Florida, specifically to Dadc and Broward Counties. The measurement
campaign took place from August 6 to September 6, 1995. Daily event precipita-
tion samples were collected each morning at each site, with several sites being
visited twice daily (morning and evening). Hg concentrations  ranged from 5  to
110 ng/L in event precipitation samples collected at the seventeen  sites. Volume-
weighted mean concentrations at each of the sites for the one month study period

64               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


ranged from 13 to 31 ng/L. Although these monthly means reflect a significant
range in concentration, a more dramatic spatial pattern was evident in the daily
event concentration field. In addition, there are significant differences between
events at each individual site over the one month study period. This paper
includes a preliminary evaluation of the Hg in precipitation measurements.

Dry  Deposition of Mercury in South Florida: Preliminary
Results
Ganda Glinsorn and Gerald J. Keeler
The University of Michigan Air Quality Laboratory, Ann Arbor, Ml 48109-2029
Gary  Evans
U.S. EPA-NERL, Research Triangle Park, NC 27711

    Measurements of the dry deposition of mercury (Hg) were performed daily
at several sampling sites for a one month  period during summer 1995 as part of
the South Florida Urban Atmospheric Mercury Project (SUAMP). A primary
objective of the SUAMP project is to determine the magnitude of the atmo-
spheric contribution to  the Hg loading to  South Florida (Dade and Broward
Counties). Twelve-hour integrated samples were collected for periods ending
0800 and 2000 EOT each day, using a dual surrogate surface approach. A newly
designed and tested aerodynamic  water surface was utili/.ed in tandem with a
well-characterized symmetric low-speed airfoil (Frishee-shaped). This latter
deposition plate was uniformly coated with Apie/.on L grease and fit tightly into
the lop of the airfoil. The newly designed collector replaced the  greased deposi-
tion plate with a circular water collection  surface. Vapor and paniculate Hg
samples taken concurrently with the dry deposition samples are used to apportion
the dry deposition by phase and species. A significant day-io-day variation in the
Hg deposition was observed at the sites. This paper includes a preliminary
evaluation of the Hg dry deposition measurements. Trace elements measured
concurrently with the Hg on the surrogate surfaces will be ulili/.ed to evaluate the
methods for atmospheric Hg and to aid in diagnosing the sources of the dry
deposition.

Methods  to Measure Atmospheric Mercury in South Florida
William M. Landing
Department of Oceanography, Florida Stale University,
Tallahassee, FL 32306-3048
Gerald Keeler
School of Public Health, University of Michigan, Ann Arbor, MI
Gary A. Gill
Department of Oceanography, Texas A&M University, Galveston TX 77551
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               65

-------
Abstracts
    As part of the South Florida Mercury Monitoring and Modeling Pilot Study
(SOFAMS), we collected replicate samples at two sites using all of the various
sampling methodologies which have hcen recently employed to study gaseous
Hg, aerosol Hg, and rainfall Hg. Sampling protocols developed hy the research
group conducting the Florida Atmospheric Mercury Study (FAMS) were used to
collect 24- hour integrated rainfall, aerosol, and total gaseous Hg (TGM)
sampling at the Davie/UF-IFAS station and the FAMS Andytown tower site
approx.  I mile southwest of Highway 27—Griffin Road intersection. In addition,
48-hour integrated Reactive Gaseous Mercury (RGM) samples were collected at
both sites using the ion- exchange filter pack technique being developed by
Nicolas  Bloom at Frontier Gcoscicnces, Seattle. The University of Michigan
group collected 12-hour integrated samples at the Davic site, and 24-hour
integrated samples at the Andytown site using similar sampling techniques for
rainfall, aerosols, and TGM. In addition. TGM was measured using aTckran
continuous TGM analy/.er at the Davie site, and a "surrogate surface" technique
was used to collect RGM at the Davic site. The goal of this work was to provide
direct intercomparison between the sampling and analytical techniques used by
the FAMS and Michigan groups, as well as to link the long-term FAMS data set
to the data generated during the SOFAMS experiment, and future extensions of
this experiment in south Florida.

Atmospheric Transport, Transformation, and Deposition of
Mercury in Florida
Jane L.  Guentiel and  William M. Landing
Dcpt. of Oceanography, Florida Stale University, Tallahassee, FL 32306-3048
Gary A. Gill
Texas A&M University, Galveston, TX 77551
Curtis D. Pollman
KBN Engineering and  Applied Sciences, Gainesville, FL

    A five year study focusing on the atmospheric  deposition of Hg in north,
central,  and south Florida has been operational since May,  1992. One of the
primary objectives of the Florida Atmospheric Mercury Study (FAMS) is to
quantify the seasonal and geographical variations in Hg deposition. Precipitation,
aerosol, and gaseous Hg samples have been collected at nine sites throughout
Florida  for periods ranging from 3-36 months. The  summertime wet season in
south Florida accounts for 80-90% of the annual rainfall Hg deposition. Deposi-
tion rates in south Florida are 30-50% higher than those from north and central
Florida. These rales are dependent upon  increases in rainfall Hg concentration
during the summer as well as increases in summertime precipitation. We are
currently evaluating the marine background Hg contribution to Florida, the
possibility ol  transport from other regions, and the spcciation and transformation
of the Hg found in precipitation.

66              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts


Meteorological Flow Regimes Influencing Atmospheric
Concentrations and Deposition During the 1995 SUAMP
Frank J. Marsik, Khalid I. A I-Wall, and Gerald J. Keeler,
The University of Michigan Air Quality Laboratory, Ann Arbor, MI 48109-2029
Russell Bullock
U.S. EPA NERL, Research Triangle Park, NC 27701
Bruce Albrecht
RSMAS, University of Miami, Miami, FL 32323

    The problem of atmospheric mercury source attribution for the Everglades
and the rest of South Florida is complicated by the complex wind flow and
precipitation patterns observed across this region. As a result, one goal of the
1995 South Florida Urban Atmospheric Mercury Project (SUAMP) has  been to
adequately define the actual three-dimensional wind How and precipitation
patterns that occurred during the project period. Definition of these fields will
allow  for the establishment of empirical relationships with observed mercury
concentration and deposition patterns in an effort to promote a scientifically
credible source attribution of atmospheric mercury deposition through the use of
hybrid-reccptor-hased modeling.
    In support of these efforts, (lie meteorological observations routinely made
by the National Weather Service (NWS) were supplemented with additional
surface and upper air meteorological measurements across the study domain.
Analysis of this data shows that during the 1995 Field Study, three dominant
meteorological How regimes were observed: local land-/sca-brec/.c circulations,
southeasterly synoptic How related to the semi-permanent Bermuda High ;• id
southwesterly synoptic How in the wake ol Tropical  Storm Jerry. This paper
presents the results of this analysis and discusses preliminary efforts to model
these Hows and their potential impacts on atmospheric mercury concentrations
and deposition patterns across South Florida.

Source Characterization in Support of Modeling the Transport
of Mercury Emissions in South Florida
Robert K. Stevens and Roy Zweidinger
U.S. Environmental Protection Agency, RTP, NC
Eric Edgerton and William Mayliew
ES&E Environmental Engineering, Gainesville, FL
Robert Kellog
ManTcch Environmental Technology, RTP,  NC
Gerald Keeler
University of Michigan, Ann Arbor, MI

    Source characterization is an essential component used to model the impact

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              67

-------
Abstracts


of stationary source emissions on selected receptor sites. In the SOFMMS,
source profiles and emission rates for Hg, trace metals, carbon, and SO2 were
obtained from a resource recovery incinerator, a medical waste incinerator and a
coal fired cement kiln operation. The source profiles contain the concentrations
of paniculate mass, trace metal compositions including Hg and S02 present in
the emissions.  Emission rates for SO2 and particulatc matter were also obtained
from these three sources during the period between August 5 and September I,
1995. A recently developed Teflon coated dilution sampling system was used to
obtain the source profiles from the three sources. The dilution sampler was
designed to remove 1  liter/minute from the slack emissions and dilute the sample
with clean nitrogen. Residence time in the dilution system prior to collecting
samples was approximately 20 seconds. This was sufficient time for the hot
gaseous emissions to condense and or coagulate to particles which were col-
lected for subsequent  analysis. A differential optical absorption spectrometer was
also used at the resource recovery incinerator to continuously measure the HgO
and SO2 concentrations in the slack emissions. This  instrument measures these
emissions across an open path in the stack. EPA Method 29 was used lo measure
HgO and water soluble Hg (ie. ionic Hg++) emissions for all three sources. This
presentation will describe the equipment and methodology used to characterize
the source emissions.  The ratios of HgO to total Hg emissions for these three
types of stationary sources was also determined with EPA Method 29 and with
gold coated glass beads coupled to the dilution sampling system. The variability
ol the stack gas emissions and source profiles will be presented.
    The information in this document has been funded wholly or in part by the
United States Environmental  Protection Agency. It has been subjected to Agency
review and approved  for publication. Mention of trade names or commercial
products docs not constitute endorsement or recommendation for use.

Receptor Modeling of Atmospheric Mercury  in South Florida:
Preliminary Results
Roy B. Zweidinger and Robert K. Stevens
U.S. Environmental Protection Agency, RTP, NC 27711
Gerald Keeler
University of Michigan, Ann  Arbor, MI

    The South Florida Atmospheric Mercury Measurement and Modeling Pilot
Study was conducted  in August 1995. An objective of the study was to determine
whether receptor modeling methods could be used to apportion atmospheric
mercury among three  known point sources; a municipal waste incinerator, a
medical waste  incinerator and a coal fired cement kiln. Source signatures were
obtained at each facility over  several days while integrated 12 hour ambient
samples were continuously collected during  the entire month at four receptor
sites located in Dadc and Broward counties.  A fifth background site was .located

68               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


on Adams Key off Florida's cast coast. Teflon and quartz filter packs were used
to collect fine fraction particulatc samples at all source and ambient sites while
annular dcnuders were used to collect inorganic gases. Measurements for SO2,
HCI, HN03, mass, elemental and volatile carbon and trace elements were
obtained. Gold coaled beads were used for collection of vapor phase mercury
and analyzed by cold vapor atomic fluorescence. EPA Method 29 was also
employed at the sources to determine Hg() and Hg+2 emissions. Variations of the
source emissions observed for different days provided information on uncertain-
ties associated with these signatures. Initial receptor modeling results using
chemical mass balance techniques (CMB7) will be presented for several days
with varying meteorological conditions.
    The information in this document has been funded wholly or in part by the
United Stales  Environmental Protection Agency. It has been subjected lo Agency
review and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.

DOAS Measurements of Elemental Mercury  and Sulfur
Dioxide from a Resource Recovery Incinerator
Carl Kainme
OPS IS, Inc., Lund. Sweden

    An OPSIS, Inc. Differential Optical  Absorption Spectrometer (DOAS) was
installed on one of the two emission stacks of a resource recovery municipal
incinerator in  South Florida. The setup included an ultraviolet light-source and a
receiver, monitoring across the 7'  slack. The DOAS spcctromelcr was located in
an air conditioned shelter at the bottom of the stack. The optical signal was
guided through a 160' quart/, fiber lo the  instrumented shelter. The DOAS
instrument was set to monitor sulfur dioxide and mercury  sequentially with a 2-
minutc cycle lime. Sulfur dioxide (SOJ and elemental Hg" were measured for a
period of 30 days between  August and September 1995. These measurements
were made as  part of the South Florida Mercury Atmospheric Monitoring
Modeling Study (SOFAMMS). The objectives of the SOFAMMS were to obtain
source and ambient chemical data lo evaluate source and receptor models to
assess the impact ol stationary source emissions on downwind receptor sites. The
DOAS measurements were used lo assisi in (lie specialion of total Hg emissions
being  emitted by incinerators. This specialion was performed by comparing the
DOAS Hg" measurements with measurements made from  impinger samples
(EPA  Method 29) collected during the same lime period. The measured results
for elemental mercury were in the range of 0-60 ug/m\ and averaged 13 ug/m'
during the measurement period. The measured data for sulfur dioxide was in the
0-600 mg/m1 range, and  averaged I 30 mg/m\ This paper will present details of
DOAS measurement methodology and describe the stalistical comparisons of
Hg" data obtained with the DOAS and Method 29.

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996               69

-------
       '	Abstracts


                             Session 7
                 Ambient Air Measurements
        Chair: Suresh Santanam, Sterns and Wheler, Cazenovia, NY

Effect of Acidity on the Sampling and Analysis of Carbonyls
Using DNPH Derivatization Method
Sudia Parmar, Abdul M. Kitto, and Luda Ugarova
Atmospheric Analysis and Consulting, 4572 Telephone Road, Suite 920,
Ventura, California, 93003

    Determination of carbonyl eompunds in amhient air as well as in industrial
staeks receiving increased attention because of the critical role these compounds
play in tropospheric organic chemistry. Aldehydes and kctones are stampled
using DNPH dcrivatization methods (EPA 001, TO-11, and CARB 430)
followed by HPLC/UV analysis. The acid cataly/.ed chemistry associated with
the reversible dcrivatization reaction is complex and may lead to ambiguous
results. In this investigation we present sampling efficiencies, stability of
sampled carhonyl hydra/ones and recoveries at different pH conditions. Opti-
mum acidic conditions required for carhonyl monitoring using various methods
are discussed.

Airborne Hexavalent Chromium in Southwestern Ontario
Ronald W. Bell and Jet-old C. Hipfner
Science and Technology Branch and The Laboratory Services Branch,
Ontario Ministry of Environment and Energy, Toronto, Ontario, M4V IL5

    This study reports the first field measurements ol airborne hexavalent
chromium (Cr(vi)) in southwestern Ontario. Hcxavalenl chromium has been
identified as an inhalation carcinogen, but questions such as how much of the
routinely monitored airborne chromium (Cr) is in this oxidation state; how much
Cr(vi) arc we being exposed to; and how much is in the aerodynamic si/.e
fraction necessary for it to be classified as an inhalable purticulate still remain to
he answered. The sampling and analytical methodology lor delecting airborne
Cr(vi) at concentrations less than I hg/m1 (nanograms per cubic metre) was
developed in the mid 1980s. Hexavalent chromium was identified as an air toxic
ol concern for the 1991/93 Windsor (Ontario) Air Quality Study and a small
pilot study addressing this contaminant was conducted during the summer of
1992. The results of this study indicated that approximately 20% of the routinely
monitored airborne chromium was in the hexavalent form. In addition, the risk
assessment undertaken on these data found that the range of carcinogenic health
risks attributable to airborne Cr(vi) were between 1.4x10"'' and 3.0xlO~4 for

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               71

-------
Abstracts


people living in the Windsor area. During the next summer, analyses of concur-
rent indoor and outdoor 24-hour air quality samples taken at 33 different
residences in Hamilton (Ontario) resulted in geometric mean Cr(vi) concentra-
tions of 0.20 hg/nv1 and 0.55 hg/m1 respectively. Again, the airborne concentra-
tions ot'Cr(vi) were found to he 20 to 25% of the  total Cr. During the summer of
1994, an airborne Cr(vi) si/.e-fractionation study was conducted in Hamilton.
From analyses of these data, the majority of the Cr(vi) was found to he less than
10 mm (microns) in aerodynamic diameter. Collectively, these findings indicate
that outdoor airborne concentrations  of Cr(vi) arc approximately 20 to 25% of
the total airborne Cr; that indoor airborne concentrations of Cr(vi) bear little or
no relationship to and arc less  than half of those measured outdoors; and that the
majority of the airborne Cr(vi) measured by this new methodology is  in the
inhalahle fraction.

Monitoring Reformulated (Jasoline  in  Milwaukee, Wisconsin
Mark K. Allen, David Grande, and Theresa Foley
Wisconsin Department of Natural Resources,  Madison, WI 53707-7921

    Beginning in January 1995, the 1990 Clean Air Act Amendments required
the introduction of refnrnuiliited ffiixoliiiex (RFC) in six counties in southeastern
Wisconsin. The  reformulated gasoline was to contain an increased percentage of
chemical oxygen (oxygenates); to contain reduced toxic components; and to
provide lower evaporative emissions. With introduction of the oxygenated
gasoline blends, the Wisconsin Department of Natural Resources (DNR) began
receiving numerous citi/cn complaints. Complaints expressed concerns about
damage to internal combustion engines, reduced automotive performance, and
health effects of the reformulated gasoline. Because of questions about health
effects, along with a lack of information on exposure to urban populations, the
DNR was asked to immediately initiate an air monitoring study. This paper will
report on how the monitoring study was conducted and the results of the study.

Volatilization and Dispersion Measurements for Airborne
Methyl Bromide  Downwind of a Treated Agricultural Field
James N. Seiber, James E. Woodrow,  Puttanna  Honaganahalli,
James S. LeNoir, and Kathyrn C. Dowling
Center for Environmental Sciences and Engineering and Department  of
Environmental Resource Sciences, Mail Stop  199, Reno, Nevada, 89557-0187

    A field study was conducted of methyl bromide volatilization flux, disper-
sion, and atmospheric fate. Air concentrations of  methyl bromide  were measured
above a flumigatcd field and downwind from the  field with the objective of
comparing vertical flux with hori/ontal flux based upon measured methyl

72              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


bromide concentration and meterological data. Another objective was to
compare downwind air concentration data to concentrations predicted by a
dispersion model (Industrial Source Complex-Short Term II [ISC-STII] model)
to begin the process of identifying potential methyl bromide sinks. Flux, vertical
profile, and single-height air concentrations were measured at four different
locations extending from the center of the treated field to nearly 0.8 km down-
wind. These concentrations were used to determine air concentration profiles and
downwind dispersion concentrations. The dispersion values were compared with
those generated from the ISC-STII Model. This study showed (1) Hori/.ontal flux
near the downwind edge was 102% of mid-field vcrticle flux; (2) Vertical air
concentration and flux  profiles farther downwind showed depletion of methyl
bromide near the surface of an adjoining mature strawberry field. Absorbed
methyl bromide was found in the soil surface of the same field; and (3) Methyl
bromide air concentrations measured downwind of the treated field source were
in agreement with those predicted by ISC-STII model, allowing the construction
of downwind isopleths (breach day of sampling. Downwind concentration
modeling of this type will be useful for estimating human exposure and also for
selling buffer /.ones for treated fields.

Ambient Trends of Benzene  in California  From  1990
Through 1995
Donald Hammond
California Environmental Protection Agency, Air Resources Board, Monitoring
and Laboratory Division, P.O. Box 2815 Sacramento, CA 98512

    The Air Resources Board (ARB) has for the past  six years maintained a
routine, canister based  toxics monitoring network in urban areas of California.
The record is now extensive enough  to evaluate for trends in toxic pollutants
within the State. This paper evaluates the trend in ben/.ene at over 20 different
sites in major population centers in California between 1990 and 1995. Includ-
ing measurement variability, a significant downward trend has been observed
throughout the State  with reductions frequently approaching fifty percent. The
paper also presents emissions inventory ol ben/.ene lor the same period whose
reductions can be attributed to the vehicle control and clean fuel programs.

Air Pollution  from Cookstoves in Developing Countries:
Results from A Field Study in China
Junfeng (Jim) Zhang
Environmental and Occupational Health Sciences Institute, 681 Frelinghuyscn
Road, Piscataway, NJ 08855
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               73

-------
Abstracts


Kirk R. Smith
Center Tor Occupational and Environmental Health, University of
California, Berkeley, CA
Yiiqing Ma
Institute IbrTcchno-Economics and Energy System Analysis,
Tsinghua University, Beijing, China

    Cookstovcs in developing countries are individually small, hut so numerous
that they could influence global inventories of greenhouse gases. As part of a
USEPA/Easl-West Center cooperative project, which has been primarily aimed
at studying greenhouse gases from widely used small-scale combustion devices,
emissions from 28 types of fuel/stove combinations commonly used in China
were measured for carbon dioxide,  products of incomplete combustion (PICs,
e.g.. carbon monoxide, methane, hydrocarbons, paniculate matter), NOx, SO2,
etc. Since  many of emitted pollutants have adverse health effects, and they could
cause elevated concentrations in homes, results from these measurements also
enable us to assess health risks from these cookstovcs.
    It was found that burning biomass fuels generated greater emission  factors
of PICs than burning fossil fuels, and that increasing stove efficiencies did not
necessarily lead to a reduction in emissions of PICs. PICs can be considered as
an indicator of health risk. In addition, because of their great relative indices of
greenhouse effect, contributions from PICs were found to be significant, some-
times close to or even greater than those from CO2. This suggests that biomass
fuel cycles are not necessarily green-house neutral even when the fuel is har-
vested on a complete renewable basis.


Size Spectra and Sources of Elemental Constituents of the
Baltimore Urban Plume
A.E. Suarez, J.M. Ondov, P.P. Caffrey, C. Wu,  P.V. Borgoul, and C.B. Kidwell
University of Maryland, Department of Chemistry and Biochemistry,
College Park, Maryland, 20742
K.C. Thaung
U.S. Environmental Protection Agency, Central Regional Lab, 839 Bcstgatc
Road, Annapolis, Maryland, 21401
T. Holsen
Pril/.ker Department of Environmental Engineering, Illinois Institute of
Technology, Chicago, Illinois, 60616

    Micro-orifice and rotary impaclors were used to collect si/,c-scgrcgatcd
aerosol samples spanning the range <0.05 mm to > 150 mm at five sites near
Baltimore, M'D, during a 20 day period during August, 1995, as part of the U.S.
EPA Great Waters Program. Samples were collected on the roof of (he Eastern

74              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                               Abstracts


 Avenue Fire Station in downtown Baltimore, on a pier at Hart-Miller Island (15
 km south-west ol Baltimore), onboard the research vessel Anderson (located on
 the Chesapeake Bay), and at the Still Pond Coast Guard Station on Maryland's
 eastern shore. Background samples were collected upwind from Baltimore at the
 University ol Maryland's Baltimore Campus located 15km west of Baltimore
 city. Samples were analy/.ed for up to 45 elements, including V, S, Zn, Sh, As,
 Br. Mg, Mn, K, Na, I, Cd, Cu, Nd, Sm, Yh, U, by x-ray fluorescence (XRF) and
 instrumental neutron activation analysis (1NAA), for total carbon by light
 transmission, and for partculate mass by gravimetry. Used in conjunction with a
 numerical model, the complete aerosol composition vs. si/.c distributions were
 used to characterize dry deposition to Bay waters both by si/.e and by source.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               75

-------
                                                            Abstracts


                             Session 8
             Tropospheric  Ozone  Measurement
           Chair: Carmo Fernandez, Arizona DEQ, Phoenix, AT,

Voluntary Emissions Reduction Efforts During the Summer of
1995 in Wisconsin
Craig S. Rendahl
Remote Sensing Technologies. Inc., 2002 North Forbes Blvd.,
Tucson, AZ 85745-1446
Tina Bryson
Wisconsin Department of Natural Resources, P.O. Box 12436,
Milwaukee, WI 53212-0436

    Many areas of the United States which do not meet the USIiPA standard set
for tropospheric (ground-level) o/.one are implementing voluntary emissions
reduction efforts to reduce emissions which lead to the formation of ground-level
o/.one. The hope is that, through voluntary efforts, enough o/.one-forming
emissions will he forgone to reduce the peak o/.one concentration of a forecasted
high o/.onc day. Aside from reaching attainment of the o/.one standard in a
regulatory vein, public health and welfare would benefit from improved air
quality. The states of the Lake  Michigan basin have developed a common effort
known as O/.onc Action! Day (OAD). This paper will document the OAD
experiences of Wisconsin in its inaugural season of participation in this regional
effort that occurred in the summer of 1995. These experiences included an
official kickoff day prior to the start of summer, regional forecasting efforts,
media notification, development of a network of employers who desired indi-
vidual notification, and other quasi-governmental participation. Specific sugges-
tions for public participation are also included in this paper, along with results of
surveys of public acceptance of the OAD message and efforts.

Paso del Norte Pilot Border Study of Ozone Precursors and
Air Toxics
Barbara Zielinska, Laurence Sheetz, Gregory Harshfield, and John Sagebiel
Desert Research Institute, P.O. Box 60220, Reno, NV  89506
Edward L. Michel
Texas Natural Resources Conservation Commission, P.O. Box 13087,
Austin, TX 7X711-3087

    The Paso del Norte area (El Paso, Texas, Ciudad Juare/., Chihuahua, and
Sunland Park, New Mexico) currently experiences o/.onc concentrations that
exceed the National Ambient Air Quality Standard for o/,one. In addition,

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996              77

-------
Abstracts


concerns have been raised about the nature and quantity of toxic air pollutants in
the area. A comprehensive monitoring program for ozone precursors and air
toxics in the Paso del Norle border area is planned by the U.S. EPA for the
summer of 1996. A pilot study was carried out in October 1995 in this area to
test the appropriateness of proposed sampling and analysis methods and to
provide preliminary data to be used in planning for the Summer 1996 study. Two
monitoring sites were selected, one in Ciudad Juare/., and the other in the El Paso
area. The sampling was carried out every second day from October 21 to
October 29, from 0300 to 0900 hr. The air samples were collected using stainless
steel canisters (for VOC in the C,-C|:, range), Tcnax-TA solid adsorbent car-
tridges (for CK - C.,() hydrocarbons), DNPH impregnated C||( Sep-Pack Cartridges
(for carbonyl compounds) and Tellon impregnated glass fiber filters followed by
PUF/XAD/PUF "sandwich" cartridges (for SVOC). The data set obtained from
the analyses  of these samples will be presented and the implications for the
upcoming 1996 study will be discussed.

Overview of the ISO 14000 Environmental Management
Standards
Gary L. Johnson
U.S. Environmental Protection Agency, Quality Assurance Division (MD-75).
Research Triangle Park, North Carolina

    In 1996, the International Organi/.ation for Slandardi/.alion (ISO) will
publish the first of a series of international standards addressing environmental
management. The emerging ISO 14000 environmental management standards
pose a potentially greater impact on the world community than the ISO 9000/
1 ()()()() standards thai preceded them. Like quality, environmental management
cuts across all business sectors; however, unlike quality, environmental issues
include regulatory aspects, diverse public interest groups, and an arguably
broader scope of applications.
    Initially,  18 standards have been proposed to encompass environmental
management systems (EMS), environmental auditing (EA), environmental
labeling (EL), environmental  performance evaluation (EPE), and life-cycle
assessment (LCA). Five standards have reached the stage of draft international
standard (DIS) and two of these are the proposed EMS standards, ISO 14001
and ISO 14004.
    This paper provides a general overview of the ISO 14000 standards,
including their contents and their role in the overall ISO 14000 series. This paper
will also briefly contrast these standards with  the ISO 9000 quality management
standards currently in use.
78               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                               Abstracts
                               Session 9
                         Particle  Exposure
       Chair: Robert M. Burton, U.S. EPA, Research Triangle Park, NC

 Characterization of Episodic Air Pollution Study
 Henk J. 77;. Bloemen
 Laboratory of Air Research, National Institute of Public Health and
 Environmental  Protection (RIVM), P.O. Box 1,3720 DA, Bilthoven, The
 Netherlands

     In recent epidcmiological studies adverse health effects have been associ-
 ated with increased levels of air pollution. The causal relation between the
 airborne strcssor and the health effect endpoinl is still a matter of debate. The
 assessment of exposure to possible stressors is currently lacking from a suffi-
 ciently detailed phenomcnological description of air pollutants in cities.
     A descriptive study was carried out to belter understand (he behavior and
 sources of a number of air pollutants, in particular line suspended paniculate
 matter. At an urban site, located down-town Amsterdam, and a rural site,
 Biddinghud/.cn 50 km north-east of Amsterdam pollution  levels were monitored
 during the winter of 1993-1994.
     Increased levels of paniculate matter were observed under the following
 conditions: New Year's Eve, with very high but short-lived concentrations; high
 wind speeds, causing rcsuspcnsion of settled dust; and stagnant weather condi-
 tions, with low rates of pollutant transport.
     During winter-smog episodes the contribution of submicron particles (
-------
 Abstracts


 hoping that such episodes will occur for the sake of research. However, until
 now only the New Year's Eve episode has proven to he punctual.

 Policy and Science on PM in the Netherlands, Bridging the
 Gap
 Peter Rombout, Eltlu Burltigh, and Sandra de Loos
 Task Force Man and Environment, National Institute of Public Health and
 Environmental  Protection (RIVM),  P.O. Box I, 3720 BA, Billhovcn,
 The Netherlands

     In quite a number of countries with different climates and cultures cpidcmio-
 logical studies have reported associations between daily concentrations of
 particles and adverse health effects. In terms of protecting public health it would
 be imprudent not to regard the demonstrated associations as causal, however
 there still remain questions on the toxicological mechanisms and the fraction of
 the air pollution that may be held responsible for the effects. In this twilight /one
 of science and policy a project is described, which is set up to give the Dutch
 policy makeers  some guidance on the risk management of PM.
     As health problems can be demonstrated at levels below the current PM
 standards and a threshold which is to be deemed safe cannot yet be established, a
 simple lightening of existing standards doesn't seem to offer the solution, even if
 such a measure  would be economically viable. A second complicating factor is
 that PM is not just one simple component but it is a mixture ol'a number of
 different substances. For the moment even the conjecture that PM  may be a
 proxy for the total air pollution mixture cannot be dismissed on the basis of the
 epidemiological data. For these reasons a different approach is chosen in the
 project to bridge the gap between policy and science.
    This approach is based on a further disenlaglement of the main components
 of PM. In order to discover the sources of these main components  of paniculate
 matter and  to find out which processes determine the ambient concentrations of
 these components, the air quality is one of the (rails that is explored. The other
 trail is formed by the quantitative assessment of health risks, in order to assess
 the extent of the health risks of these components of PM. When this information
 from both of the trails is combined with the results lor the different policy
 scenarios it will be possible to offer the policy makers a choice of different
 alternatives and the health risks that may be associated with these alternatives.
The choices a government has to make for the risk management  of the PM
 problem may then be taken with some scientific guidance.
    As a matter of simplification PM  will be classified in five broad categories
based on the particle sixes and chemical speciation: PMIO, PM2.5, PMO.I,
carbonaceous and secondary aerosol. The emissions and phenomenology of
these five parts of the pentagon of PM will be established for the Netherlands.

 80              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts
 With an adecquatc dispersion model the concentrations for different policy
 options and abatement strategies will he calculated for an averaging period of
 one year and for the situation of a typical air pollution episode in the Nether-
 lands. The different  results of these alternatives may then he used hy policy
 makers as a basis for their decision making.
    The  important gain from such an approach is that it enables you to sec in
 advance if for instance a strategy of abatement of sources of PM has a similar or
 different  result for the levels of e.g. PMO. 1 or carbonaceous aerosol. In this
 period when the exact culprits are not yet known, following a non regret policy
 wich  has effects on a number of PM components at the same lime may be a cost
 effective way of progressing.

 Influence of Weather on  Aerosol Exposure in Selected,  Highly
 Populated Metropolitan  Areas of the United States
 Robert M. Burton
 Exposure Assessment Branch, National Exposure Research Laboratory, U.S.
 EPA, Research Triangle Park, NC 27711
 John J. Stretcher
 Applied Modeling Research Branch, National Exposure Research Laboratory,
 U.S. EPA, Research Triangle Park, NC 27711

    Particle research conducted under the EPA/Harvard Five City Exposure
 Study has included both realtime continuous and integrated 24-hour measure-
 ments of paniculate and meteorological components. Analysis of data indicates
 both direct and indirect associations between temperature, humidity and the
 physical and chemical make-up of particles. Field studies conducted in Philadel-
 phia,  Pittsburgh, Washington, DC, and Nashville suggest correlation between
 ambient fine particles (PM2.5) concentrations and delta T, the absolute value of
 the difference between the average daily temperature and the average indoor
 temperature of 68 F. Conversely, eoarsc particle concentrations (PM2.5 - PMIO)
 do not correlate well with temperature components but a re related more to wind
 speed and local activity.
    Other particle measurements suggest that the high summer temperatures
 simultaneously drive more of the aloft, regionally transported sullates down to
 ground level, while at the same time use the increased  vertical mixing to Hush
 out a significant amount of the ground level, locally generated components such
 as soot and organics.
    Continuous I'inc particle mass, determined simultaneously from realtime
 particle si/.ing measurements and from heated filter continuous mass measure-
ments, reveal significant differences when the ambient relative humidity is above
the  fifty percent level. These differences indicate that the hygroscopic mixed-salt
aerosol is constantly  absorbing or releasing water as the humidity varies above

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               81

-------
Abstracts
the dclinqucscencc point for the aerosol.
    These dynamic propertied which have a profound effect on the characteris-
tics of the aerosol in which populations of large metropolitan areas are exposed.
Choosing the proper aerosol that components for use in health effects exposure
studies should he carefully considered.

The  Role of Total Human Exposure Measurements of
Particulate Matter in Community Epidemiological Studies
William W. Wilson
MD-52, NCEA-RTP, U.S. EPA, Research Triangle Park, NC 27711.

    Measurements of total human exposure are possible for cohort studies
covering a few individuals for a short time. However, many epidemiological
studies of paniculate matter (PM) use community-based rather than cohort data,
e.g. deaths in a large city or emissions to a hospital serving a large metropolitan
area over periods of one to several years. For such studies an indicator of
average, daily community exposure must be used. Typically this is obtained form
measurements of PM concentration at one central site. It will be shown that such
epidemiological studies relate health outcomes to exposure to ambient particles
not to exposure to total particles. Exposure lo ambient particles occur both when
people are outdoors and when people are indoors or in other mieroenvironmenls.
Epidemiological studies using community exposure data do not provide informa-
tion on the potential health effects of particles generated indoors or particles due
lo a person's presence or activities (personal cloud). Therefore, measurements of
total human exposure are not useful for community epidemiological studies
(unless the individuals constitute the cohort and their individual total human
exposure is measure for the duration of the study). However, total human
exposure studies may be useful if they help us understand what fraction ol
ambient particles are found indoors and what parameters influence that fraction.
Examples of how concepts from total human exposure research can be used to
improve estimates ol community exposure to ambient PM, as well as further
research needed in this area, will he discussed. The importance of distinguishing
fine mode from coarse mode PM and for measuring specific species in both the
fine and coarse modes will be emphasi/.ed.

A Synoptic Climatology of Fine Particle Acid Aerosols
Michael P. Zelenka
Atmospheric Sciences Modeling Division, Air Resources Laboratory (MD-56),
National Oceanic and Atmospheric Administration,
Research Triangle Park, NC 2771 1  U.S.A.

    High concentrations of ambient, strong aerosol acidity (H+) were measured

82              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 during the summer of 1990 at Uniontown, Pennsylvania. The synoptic climatol-
 ogy of the region was investigated for clues to the meteorological conditions that
 were associated with the temporal variability of the FT concentrations, with
 particular emphasis on the conditions conducive to high concentrations of acid
 aerosols. Since H* is a secondary pollutant, formed primarily through oxidation
 of sulfur dioxide (SO,), transport of pollutant precursors from their source
 regions is an important factor in determining the magnitude of the H+ concentra-
 tions. Local meteorological factors such as temperature, wind, and the height of
 the mixed layer were also important for determining the variability in H*
 concentrations. This research shows that both local meteorology and regional
 circulation contributed to episodic events of ambient H*. Analyses of the
 meteorological parameters affecting ambient concentrations of acid  aerosols
 should at least include the lempeiature, dew point (or some other indicator of
 atmospheric moisture), mixing height, and the wind speed and direction at the
 surface and for an upper level of the atmosphere that facilitates the regional
 transport ol' pollutants.

 Assessment of the Temporal Relationship Between
 Summertime Ultra-Fine Participate  Count Concentration With
 PM in Washington D.C.
 George Allen, Eileen Abt, and Petrtis Koutrakis
 Harvard School of Public Health, Boston, MA 02115

    Recent cpidemiological evidence has shown a significant relationship
 between ambient daily particulate mass concentrations and human health effects
 as measured by eardio-pulmonary morbidity and mortality. Most of the research
 aimed at determining causal agents focuses on  the combustion-related compo-
 nent of PMK). Some studies have suggested that ultra-fine aerosols may he
 important factor, cither by their count  concentration or by other mechanisms.
 The temporal relationship between mass (PM2.5 and PM 10) and ultra-fine
 particles (mass and number concentration) was charactcri/.ed by making daily
 and hourly measurements of size classified particles between  .02 and 10.0 m,
 PM2.5, and PM 10 during the summer of 1994  in Washington, DC. Particles
 smaller than 0.7 m aerodynamic diameter were measured with a TSI model
 3934L scanning mobility particle si/.er, and a TSI model 33IOA laser-Dopplcr
 vclocimeler was used for particles between 0.7 and 10 m. Two TEOM continu-
 ous mass analyzers were used to measure hourly PM2.5 and PM 10 in conjunc-
 tion with integrated 24 hour low-volume PM samplers. The temporal relationship
 between hourly and daily means of particle count,  surface area, and volume
concentration in different si/.e ranges and collocated measurements of PM2.5
and PMK) is presented. Preliminary analysis shows no correlation (r=.()0)
between daily (mean 24 hour) particle number concentration or ultra-fine mass

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              83

-------
Abstracts


and PM2.5 or PM10. This is consistent with the lack of temporal variability (on
a daily basis), local sources, and short (+1 day) lifetimes ofultrafinc particles
compared to the sources (regional transport) and lifetimes (many days) of the
larger particles (0.4 to  I m) that dominate the fine PM concentration. If the lack
of an association between count and mass concentration holds over other seasons
and locations, this may make it more difficult to postulate theories of mecha-
nisms for PM health effects which rely on a temporal association between
particle counts or ultra-fine mass and measures of PM2.5 or PM 10.

The Relationship Among TSP, PM10, PM2.5, and Inorganic
Constituants  of Atmospheric Particulate Matter at Multiple
Canadian Locations
J. Brook and T. Dann
Environment Canada, Downsview, Ontario, Canada
R.T. Nurnell, Health Canada, Ottawa, Ontario, Canada

    Abstract not available.

Relationships Between Daily CoH, Visibility  Measurements
and Fine Mass Concentrations in Philadelphia, PA
George D. Tlnirstoii and Kazuhiko Ito
Dept. of Environmental Medicine, NYU School of Medicine,
Tuxedo, NY 10987
Petros  Kotitrakis
Dept. of Environ. Health, Harvard School of Public Health, Boston, MA  02115

    A major limitation in the evaluation of the health effects of fine paniculate
mailer has been the fact that  line particle mass (PM2.5) measurements are not
historically available. However, the recent Harvard Metropolitan Acid Aerosol
Characleri/.ation Study (MAACS) provides comprehensive PM data at multiple
sites in Philadelphia, PA that are useful in determining quantitative relationships
between PM2.5 and the historically available Coefficient of Ha/.e (CoH) and
visibility records, which have been collected daily for decades in that city. These
latter two measures are useful metrics of submicron primary and secondary
aerosols,  respectively. Thus, these two measures can provide a basis for predict-
ing historical daily PM2.5 concentrations. In this work, we present predictive
models for PM2.5 based upon daily measurements of both CoH and visibility
(corrected for RH) made in Philadelphia, PA during the years 1992-1993, when
the comprehensive MAACS PM2.5 are also available. These predictive equa-
tions will be especially useful in that they will allow time-series evaluations
assessing the role of PM2.5 in historical TSP-mortality relationships previously
reported for Philadelphia, PA.

84               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


 Atmospheric Mass and Metal Size Distributions Measured
 Around Lake Michigan
 Thomas M. Holsen, Seiing-Miik Yi, Xiaiif; Xlm, Rajcndra D. Paode,
 and Kenneth E. Noll
 Department of Chemical and Environmental Engineering, Institute of
 Technology, IIT Center, Chicago, Illinois 60616

     Numerous atmospheric mass and metal si/.c distributions (0.15-100 um)
 were measured in Chicago, IL, in South Haven, MI, in Sleeping Bear Dunes, MI,
 and over Lake Michigan since 1989. Cascade and Moudi impactor were used to
 determine the si/,c distribution in the line particles (<2.5 pm), while Noll Rotary
 Impactor (NRI) was used to collect coarse particles (>2.5 (.mi). The mass si/.c
 distributions generally followed a bi-modal pattern, with peaks in the 0.1 -1 (fine
 particle mode) and 20-30 (Jin range (coarse particle mode). The anthropogenic
 metals generally followed a bi-modal distribution with peaks in the 0.1-1 (jm
 range,  and the 8-10 jjm range. Calcium and sodium peaked around 8-10 (jm,
 while magnesium generally peaked in the 10-20 (am size range. Si/.e distributions
 of iron demonstrated considerable variation between samples. The iron mass
 peak was generally in the 1-12 mn range.

 Quality Control and Reference Equivalency of PM Monitoring
 Systems
A. van  der Meulen
 National Institute of Public Health and Environmental Protection (R1VM),
 PO Box 1 - 3720 BA Bilthoven, The Netherlands

    Systematic air pollution measurements are  performed for two purposes, vi/
 to check for compliance with air quality directives, and to support rule making.
 An appropriate quality of the systematic measurements can be ensured as
 follows.
    Firstly, the measurements will have be performed  following the general
requirements laid down in EN4500I (general criteria for the operation of testing
laboratories). Basically, a certified quality control system according to EN45001
aims at appropriate product quality,  fixed modes of operation, traccbility, and
rcproducibility.
    Secondly, an appropriate reference method will have to be available to
harmonize the monitoring methodology.
    The European Union (EU) has adopted the  following strategy to improve the
socallcd PM 10 monitoring methodology:
    •   An appropriate reference method has been designated by convention to
        harmonize the SPM monitoring methodology in the EU for the thoracic
        particles (ISO7708, 1995).

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              85

-------
Abstracts
    •   EU Member Stales will have to use either the reference method under
        routine conditions, or methods which arc equivalent to the reference
        method (equivalence methods), for the survey of ambient air quality.
    •   EU Member Stales will have to lake the necessary steps to ensure that
        equivalence methods are tested by an appropriate reference equivalence
        procedure.
    Following these recommendations, the EU has supported the work done by
CEN (European Committee for Standardi/.alion) to harmoni/.c the performance
of sampling instruments for compliance monitoring:
    •   Drafting a field test procedure to demonstrate equivalence of sampling
        methods for thoracic particles, with a reference sampling method for the
        thoracic fraction of SPM;
    •   Validating and assessing the practicability of the designated reference
        equivalence procedure by trying it out in demonstrative field tests at
        characteristic measurement locations in the EU.
    This presentation aims to elucidate the CEN/EU reference equivalence
procedure for PM 10, and the pertinent EN4500I quality systems approach.

Application of Near-Field Optical Techniques to Ambient
Participate Matter: Physical and Chemical Structures
Has Hale
North Carolina Slate University Department of Physics, Raleigh, NC 2769-8202

    In the past few years, new frontiers in materials research have been opened
through the development of the  near-field scanning optical microscope, or
NSOM. The NSOM modifies standard optical techniques by increasing spatial
resolution through the use of geometry to confine the optical field. The applica-
tion of NSOM to paniculate analysis will be discussed. Though many techniques
are able to look at a distribution of samples, i.e. many particulars at a time,
NSOM allows the study of this distribution one paniculate at a time, including its
internal structure, without special sample preparation beyond impaction onto a
substrate. By operating at length scales less than the wavelength of light, NSOM
has the power of optical microscopic characteri/.ation for chemical and phase
information with nanoscale spatial resolution. Optical measurements can be
performed simultaneously with and independently of topographic measurements
which yield a three dimensional image of the particular's physical structure. We
will discuss spectroscopy, including how speelroseopie selection rules, modified
due to the ability of NSOM to couple  with evanescent modes, give additional
information, and our initial data from  ambient paniculate samples.
86               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 Carbon Nanotubes and Buckyballs: Structure, Mechanics, and
 Air Pollutant Potential
 B. I. Yakobson
 Department of Physics, North Carolina State University, Raleigh, NC 27695

    Ubiquitous carbon particulars contain a broad variety of particles, not
 always of well defined structure.1 Recent discoveries of fullerene huckyballs and
 carbon nanotubcs brings a new insight in peculiar properties and potential hu/urd
 of ultrafinc carbon particles. We will discuss their atomic structure, unique
 capillary (straws) and mechanical properties. The results of simulations of
 nanotubc behavior under different mechanical circumstances will be presented in
 some detail.2 The applicability of continuum modeling will be demonstrated and
 its use for predicting nanoparticlc properties. In particular their ability to damage
 soft tissues,1 to absorb/encapsulate and carry metals is of great importance for
 understanding the hazard potential. Finally, the C-C vibrational mode and their
 spectroseopic activity will be discussed in relation with potential high-resolution
 optical imaging (NSOM).4
 1 D. Stcincr and H. Burtschcr, Water, Air and Soil Pollution 68, 159 (1993).
 2 B.I. Yakobson, C.J. Brabec, and J. Bernholc, Pliyx. Rev. Lett. 76, 2511 (1996).
 3 R.S. Ruoff and B.I. Yakobson (unpublished). 4 B.I. Yakobson and
 M.A.Pacslcr, Ultramicmxcopy 57, 204 (1995), ibid. 57, 241 (1995);  sec also
 H.D. Hallen, thix Symposium.

 Development and Evaluation of a Condensational Growth/
 Virtual Impaction System for Internal Separation of Ultrafine
 Particles
 Constantinos Sioutas and Petrost Koutrakis
 Harvard University, School of Public Health, 665 Hunlington Avenue,
 Boston, MA 02115.

    Atmospheric particles smaller than approximately 0.1 mil in diameter are
 known as ultrafinc particles, or Aitkcn nuclei. To-date there is no cpidcmiologi-
 cal evidence linking ultrafinc particles to adverse health effects of human
 populations, partly because health effects have always been associated with the
 mass concentrations of a pollutant. This does not however exclude the possibility
 of ultrafinc particles causing adverse health effects  because increased pulmonary
 toxicity of ultrafinc particles has been demonstrated in several laboratory studies.
To elucidate the toxicity of ambient ultrafine particles, inhalation loxicologists
 need to conduct controlled exposures to increased concentration levels of such
particles by separating them from the majority of the surrounding air, while
keeping them airborne.

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               87

-------
 Abstracts
    This paper presents a method using the virtual impactor technique to
separate ultrafinc particles without subjecting them to a high vacuum. Our novel
approach to the problem of separating and concentrating ultrafine particles using
virtual impaction is to first grow the particles by condensing water vapor on them
to a si/,c thai can be easily separated in a virtual impactor. After separalion from
Ihe surrounding air the particles can return to their initial size by being brought
back to their original  thermodynamic state, using already available technologies
such as diffusion dryers. The aerosol first passes over a pool of warm water (60°
C) lo achieve saturation. Subsequently, the air is drawn through a cooling lube
(10° C) that allows ihe ullrafine panicles to grow to  micron-size. The grown
particles are then drawn through a virtual impactor to be separated from the
surrounding air. The effect of parameters such as sampling How rale, ultrafine
particle chemical composition, and aerosol concentration on the collection
efficiency and losses  of our system will be presenlcd.

Development of a Continuous Mass Monitor by Measuring
Pressure Drop Across Filters
Petros Koittrakis, Peng- You Wang, Constantiiws Sioutas, and
Mikhail WolJ'son
Harvard University, School ol Public Health, 665 Huntington Avenue,
Boston, MA 02II5.

    We have developed a continuous mass measurement method based on
monitoring the pressure drop across a porous membrane (Nuclcpore) filler over a
period  of time. Aerosol samples are firsl drawn through a diffusion dryer that
reduces Ihe sample relative humidity to 40% or less, and subsequently through a
series of porous membrane fillers. The increase in the pressure drop with mass
loading has been studied for different sampling flow  rates, pore si/.e, and particle
density, for particles in the si/.e range 0.1-2.0 um. Filters of pore si/.e from 1 lo 8
um were lesled for sampling Hows varying from 0.7 to 16 LPM. The lest
aerosols used in the experiments included PSL latex,  ammonium sullate, sodium
chloride, and indoor air particles. Aerosol mass concentrations ranged from 8.0
to 132.0ug/m'.
    Our results showed that by approprielly choosing the face velocity and pore
si/.e of the filler, interception becomes ihe dominant mechanism for particle
deposition. This makes the the pressure drop independent of particle si/.e and
only dependent on the mass concentration of ihe sampled aerosol. The pressure
drop per unit time and particle concentration averages to 0.0088 (± 0.0005)
inches  HnO/hr/(ug/m•') for particles in ihe range 0.1 -2.0 um. Our continuous
mass monitoring method can detect as little as 5 ug/m1 in about 1 hour.
    We have also developed a theory that explains the mechanisms for increas-
ing the pressure drop across Nuclepore filters with mass loading. Our sampler

88              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 was also validated in field icsls by comparing its performance to collocated
 continuous (Tapered Element Oscillating Microhalance, TEOM) and timc-
 integralcd (Harvard  Impactor. HI) samplers. Excellent agreement was found
 between our continuous sampler and HI for 3-hour sampling intervals, whereas
 HI and the continuous mass monitor were systematically measuring higher PM,,.
 concentrations than those determined with the TGOM, presumably due to
 volatili/.ation losses  of the latter method.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               89

-------
                                                             Abstracts


                             Session 10
                 Integrated Monitoring Study
      Chair: Paul Solomon, Pacific Can & Electric Co., San Ramon, CA

 1995 Integrated Monitoring Study: Study Objectives
 and Design
 Paul A. Solomon and Richard H. Thuillier
 Pacific Gas & Electric Company, 2303 Camino Ramon, Suite 200,
 San Ramon, CA 94583
 Karen L. Magliano, Andrew J. Ranzieri, Richard Hackney, and Elizabeth
 Niccum
 California Air Resources Board, Sacramento, CA 95S 14
 John G. Watson
 Desert Research Institute, Reno, NV X9506
 Phil M. Roth
 Envair, San Ansel mo, CA 94960
 Fred Lurmann
 Sonoma Technology, Inc, Santa Rosa. CA
 Donald Lehrman
 Technical and Business Systems, Santa Rosa, CA
 Glen Cass
 California Institute of Technology, Pasadena, CA
 Steven Ziman
 Chevron, Richmond, CA
 D. Alan Hansen
 Electric Power Research Institute, Palo Alto, CA

    The California Regional PM10 Air Quality Study (CRPAQS) is a multi-
 year, three-phase program of planning, ambient monitoring, modeling, and data
 analysis with the following objectives to: I) obtain an improved understanding of
 the causes of high particulate matter concentrations (PM 10 and PM2.5) in the
 San Joaquin Valley of California during seasons when the state and Federal PM
 standards arc most often exceeded and 2) develop tools (i.e., air quality model-
 ing and data analyses) that can be used by decision makers in formulating and
comparing candidate emissions management practices for obtaining current and
 future slate and Federal PM standards.
    The 1995 Integrated Monitoring Study (IMS95) is part of the  Phase 1
planning effort for  CRPAQS, and thus, the overall objectives of IMS95 arc to fill
information gaps needed for planning an effective field program later this
decade; to develop an improved conceptual model for pollution buildup (PM:
PM 10, PM2.5, and aerosol precursors) in the San Joaquin  Valley;  to develop a
uniform air quality, meteorological, and emissions data base that can be used to

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               91

-------
Abstracts


perform initial evaluations of aerosol and log air quality models; and to provide
early products that can he used to help with the development of the upcoming
State Implementation Plan for PMIO. Consideration of potential new PM
standards were also included in the planning and design of IMS95.
     Preliminary data analysis, using existing data, indicated that the Federal PM
standard was most often exceeded in the southern San Joaquin Valley during the
fall and winter months, therefore, IMS95 efforts are concentrated in the southern
valley during those lime periods. Components of IMS95 include a two week fall
study during early November (IMS95-F), a four week winter study during
Dcecmher and early January (IMS95-W), and a two week study ol vertical fog
dynamics in mid-January near Sacramento (IMS95-T). Overall nearly  100 air
quality and meteorological monitoring sites were deployed during 1MS95
obtaining detailed information on PM, PM precursors, fog, surface and upper air
meteorology, and emissions activity data. The final data base will include
supplemental data from existing sites in and near the study domain, many of
which were augmented with higher frequency measurements to support IMS95.
This paper will present the specific objectives of IMS95, define the study
components in detail, and present rationale  for sites based on the study objec-
tives for IMS95.

1995 Integrated  Monitoring Study: Fog Measurements  in the
Southern Valley -  Preliminary Results
Jeff Collett, Aaron Bator, Eli Sherman, Oliver Graf, Xin Rao, and Kate Hoag
Colorado Stale University, Department of Atmospheric Sciences,
Fort Collins, CO 80523
Will Richards
Somoma Technology, Inc., Santa Rosa, CA
Karen Magliano
California Air Resources Board, Sacramento, CA
Paul Solomon
Pacific Gas and Electric Company,  San Ramon, CA

    One objective of the  1995 Integrated Monitoring Study is to obtain an
improved understanding between the relationship of I'ogs and paniculate matter
(PM: PM|0, PM-,., and aerosol precursors). During the winter portion of the
study (IMS95-W), surface measurements of fog were obtained during December
and early January at IMS95-W Core sites located in Bakcrsfield, Fresno, and
near the Kern Wildlife Refuge. Fog measurements, including si/.e and lull
chemistry were obtained wilh the other gaseous, aerosol, and meteorological
measurements collected simultaneously at the three sites. This paper will
describe the methods and measurements employed and present preliminary
results of those fog measurements.


92               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts


 1995 Integrated Monitoring Study: Vertical Fog
 Measurements in the Northern Valley - Preliminary Results
 JeJfCollelt, Aaron Bator, Eli Sherman, Oliver Graf, Xin Rao, and Kate Hoag
 Colorado Stale University, Department of Atmospheric Sciences,
 Fort Coll ins, CO 80523
 Will Richards
 Somoma Technology, Inc., Santa Rosa, CA
 Karen Magliano
 California Air Resources Board, Sacramento, CA
 Paul Solomon
 Pacific Gas and Electric  Company, San Ramon, CA

    A second objective of the 1995 Integrated Monitoring Study related to fog is
 to obtain an improved understanding of the vertical dynamics of log and the
 relationship between fogs and parliculate matter (I'M: PM1(), PM,5, and aerosol
 precursors) above the surface. During the vertical fog or Tower study (IMS95-
 T), surface and aloft measurements of fog were obtained during two weeks in
 mid-January at  a 1650 fool lower located just south of Sacramento in Walnut
 Grove, California.  Fog measurements,  including si/.e and full chemistry were
 obtained with relevant gaseous, aerosol, and meteorological measurements
 collected simultaneously at the site. This paper will describe the methods and
 measurements employed and present preliminary results of those fog measure-
 ments.

 1995 Integrated Monitoring Study: Spatial Variability of PM1()
 and Related Precursors During the Fall Saturation Study -
 Preliminary Results
Judy Chow, John Watson, Jack Gillies, and Dick Egami
 Desert Research Institute, 5625 Fox Ave., Reno, NV 89506
Paul Solomon
 Pacific Gas and Electric Company, San Ramon, CA
Karn Magllano
California Air Resources Board, Sacramento, CA
Paul Roberts, David Wright, and Dale Crow
 Sonoma Technology, Inc., Santa Rosa, CA
Joe Newman and Chuck Berkowitz
Metrotech, Reno, NV

   Twenty-two saturation or portable  PM|(| monitoring sites were located
surrounding the core air quality monitoring site in Corcoran, California during
the fall portion of the 1995 Integrated Monitoring Study (IMS95-F).  Sites were
placed to evaluate the representativeness of the core site and the sphere of

Measurements of Toxic and Related Air Pollutants, May 7-9,  1996              93

-------
 Abstracts


 influence of sources within the saturation study area. PM|(| mass and NR, were
 obtained at all sites every day for 14 days and full chemistry (ions, EC/OC, and
 elements by XRF) and HNC^ were determined at ten of those sites during 9 of
 the 14 days. This paper will present preliminary results on the spatial variability
 of PM mass from this 22 station saturation monitoring network. Limited results
, of precursor gas and aerosol chemistry information also will be presented if
 available.

 1995 Integrated Monitoring  Study:  Temporal Variability of
 PM10 and PM2.5 Mass During the Winter Field Program -
 Preliminary Results
 Judy Chow, John Watson, Jack Gillies, and Dick Egami
 Desert Research Institute, 5625 Fox  Ave., Reno, NV 89506
 Paul Solomon
 Pacific Gas and Electric Company, San Ramon, CA
 Karn Magliano
 California Air Resources Board, Sacramento, CA
 Paul Roberts, David Wright, and Dale Crow
  Sonoma Technology, Inc., Santa Rosa, CA
 Joe Newman and Chuck lierkowitz
 Metrolech, Reno, NV

     During the winter program of the 1995 Integrated Monitoring Study
 (IMS95-W), PM|() and PM,5 mass and chemistry, HNO,, and NH, were col-
 lected using a three-hour sampling frequency at the four core air quality monitor-
 ing sites located at Fresno, Bakersfield, near the Kern Wildlife Refuge, and
 southwest of Chowchilla. These aerosol measurements were obtained simulta-
 neously with other gaseous, log, and meteorological measurements obtained at
 the core sites. The information obtained will be used  to determine improved
 strategies for characterising diurnal variations in PM  mass and chemistry and
 PM precursors. This paper will present preliminary results of the diurnal nature
 of PM mass during wintertime conditions in the San Joaquin Valley; chemistry
 data will not be available until laic in 1996.

 1995 Integrated Monitoring Study: Surface and Upper Air
 Meteorological Measurements - Preliminary Results
 Lin Lindsey, Bastion Schoell, and Scott Ray
 Sonoma Technology, Inc., 5510 Skylane Blvd.. Suite 101,
 Santa Rosa, CA 95403
 Don Lehrman
 Technical and Business Systems, Santa Rosa, CA
 94              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts
 Reggie Smith, Rich Hackney, and Karen Magliano
 California Air Resources Board, Sacramento, CA
 Richard Tliuillier and Paul Solomon
 Pacific Gas and Electric Company, San Ramon, CA

     An extensive surface and upper air meteorological monitoring network was
 deployed during the winter program of the 1995 Integrated Monitoring Study
 (IMS95-W). Surface sites included over 100 existing sites, the addition of 5
 standard meteorological sites on 10 in towers and 5 sites employing sonic
 anemometers at 2 m in height. Upper air measurements included existing data
 from the National Weather Service (NWS) rawinsonde sites in Oakland and
 Vandcnherg Air Force Base, rawinsondes launched from Lamore Air Force
 Base, and a NEXRAD (NWS) near Hanford. Sites installed for the study
 included 4 radar profilers with RASS, two rawinsonde (4 per day) sites, a
 doppler acoustic sounder, and an acoustic sounder. These data will he used to
 charactcri/.e the three-dimensional meteorology in during the study and to
 provide data for model evaluation  and application. This paper will present
 preliminary results from this network describing the meteorological conditions
 observed during the study and specific  conditions observed leading up to and
 during a PM episode.

 1995  Integrated Monitoring Study:  Low Wind Speed Study -
 Preliminary Results
 H. Andrew Gray and Edward L. Carr
 Systems Applications International, 101 Lucas Valley Road,
 San Rafael, CA 94903
David Pankratz
 AcroVironmcnt, Inc.,  Monrovia, CA
 Chris Biltoft
 U. S. Army, Dugway Proving Grounds, CA
Jim Pedemon
California Air Resources Board, Sacramento, CA

    The 1995 Integrated Monitoring Study (IMS95) is part of the Phase I
planning effort for the California Regional Paniculate  Matter Air Quality  Study
(CRcPAQS). As part of IMS95, a four-week field program was conducted
(IMS95-LW) during the winter of 1995-96 to characterize mixing and diffusion
under low wind speed  stable conditions. An array of sonic anemometers was
located in a rural field and used to collect high temporal resolution (10 h/.) wind
data. Tracer material (SF6) was released at the center of the field and fifteen
minute average tracer concentrations were measured at 36 locations in three
rings (100m, 200m and 400m radius) surrounding the release point. The sonic


Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               95

-------
Abstracts


anemometer data and tracer experiment data will he analyzed to better under-
stand mixing and dispersion under low wind speed stable conditions. This paper
will describe the measurements and methods and present preliminary results of
the experiment.

1995 Integrated Monitoring Study: Emissions Activity
Measurements During the Fall and Winter Components -
Preliminary Results
Dana Coe, Lyle Chinkin, and Chris Jones
Sonoma Technology, Inc., 5510 Skylane Blvd., Suite 101,
Santa Rosa, CA 95403
Karen Magliano
California Air Resources Board, Sacramento, CA
Paul Solomon
Pacific Gas and Electric Company, San Ramon, CA

    To collect unrecoverable emissions activity data during the fall and winter
programs of the 1995 Integrated Monitoring Study (IMS95-F, -W), an extensive
emissions activity data collection effort was employed. This consisted of
obtaining land use maps in electronic format to set up a baseline land use
inventory, followed by the daily collection of activity data by specially trained
operators in the field using aerial photographs, land use maps, video cameras,
and data sheets. Traffic counts, using automated counters,  and road silt measure-
ments using vacuum techniques were also obtained at up to 30 sites during
IMS95-F and -W programs as part of the emission data collection activities. A
Microsoft Access database is being designed that will allow the emissions
activity data to be easily queried through a query dialog box. This paper will
describe in detail, the methods used, present preliminary results of the findings,
and will evaluate  the success of ihis unique approach to emissions data collec-
tion.

1995 Integrated  Monitoring Study: Quality Assurance
Methods and  Results
Alex Barnetl
AeroVironment, 222 Ii Hunlinglon Dr., Suite 200. Monrovia, CA 91096
David Bush
AeroVironment, Placerville, CA
Michael Miguel,  Charles Cowell, anil Karen Magliano
California Air Resources Board, Sacramento, CA
Paul Solomon
Pacific Gas and Electric Company, San Ramon, CA
96               Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                                Abstracts


     An integral part of the fall and winter programs of ihe 1995 Integrated
 Monitoring Study (IMS95-F, -W) was to provide extensive external and inde-
 pendent quality assurance (EQA) for field monitoring and laboratory analyses to
 ensure that data of know accuracy, precision, and validity would he obtained. An
 external contractor and the California Air Resources  Board provided the external
 quality assurance effort, which included in-field system and performance audits
 prior to and during the measurement campaign as well as laboratory audits
 during the program. Essentially all measurement methods were tested, including
 upper air meteorological systems and special methods deployed as part of the
 methods evaluation. The final data base will include supplemental data from
 existing sites in and near the study domain, however, only data with a docu-
 mented quality control/quality assurance methods and records provided by the
 data sponsor will  he included in the final achieve. This paper will describe QA
 methods and measurements conducted and  present a summary of the results from
 the quality assurance effort.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               97

-------
                                                            Abstracts


                             Session 11
                            Indoor Air
       Chair: Ross Highsinilli, U.S. EPA, Research Triangle I'ark, NC

 Pattern and Sources of Polynuclear Aromatic Hydrocarbons
 and Their Derivatives in Indoor Air
 Somentha Mitra
 Department of Chemical Engineering, Chemistry and Environmental Science,
 New Jersey Institute of Technology, Newark, New Jersey, 07102
 Bonnie Ray
 Department of Mathematics. New Jersey Institute of Technology, Newark,
 New Jersey, 07102  .

    Select polycyclic aromatic hydrocarbons (PAHs) and their derivatives were
 measured in several homes in Columbia (OH), by researchers at Butlcllc,
 Columbus. These homes had different indoor PAH sources such as tobacco
 smoke, gas/heal cooking, and electrical stoves. A combination of correlation
 analysis and Factor Analysis/multiple regression analysis were used to identify
 and apportion the different sources of PAHs. As expected, environmental
 tobacco smoke had the greatest impact on indoor PAH concentrations. Back
 round PAH were the largest contributor to total  PAH in non-smokers' homes.

 Real-Time Low-Level Pollution Monitoring for IAQ
 Applications
 Oleg P. Kishkovich and Michael A. Joffe
 Extraction Systems Inc., P.O. Box  1329, Woonsocket, RI 02895-0839

    Quantitative air analysis provides valuable information for indoor air quality
 (IAQ) investigations in airports, hospitals, museums, etc. Currently, grab
 sampling is the most frequently used monitoring technique. However, for certain
 applications real-time monitors (RTMs) arc most suitable. RTMs allow solutions
 of a number of problems unattainable with traditional collecting devices. Using
 RTMs is especially beneficial when fast time response and quick analysis
 turnaround are required. RTMs prove indispensable in monitoring  short concen-
 tration spikes, accurate measuring of low pollution levels, and applications
 requiring in.situ  statistical validation.
    We present our results of applying RTMs to IAQ problems. We have
delected o/onc, NOx, SO2, formaldehyde, and volatile organic compounds
 (VOCs) with the sensitivity in the low parts-pcr-billion range and time response
of a few seconds. Using RTMs we have determined enlrainmenl paths of
combustion engine exhaust fumes into building air handling systems, have

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996               99

-------
Abstracts


carried out on-sitc evaluation of the remaining chemical filter capacity, have
performed last and comprehensive IAQ surveys in a 3,000,000 ft2 museum
storage facility. We present examples when using RTMs enables better under-
standing of indoor air chemistry.
    We also discuss future applications of RTMs and give recomendations  for
IAQ practitioners.

Emission Rates of Formaldehyde and Other Carbonyls from
Consumer and Industrial Products Found in California Homes
Thomas J. Kelly, Jan R. Satola, and Deborah L. Smith
Ballellc, 505 King Avenue, Columbus, Ohio 43201-2693

    This program determined the emission rates of five carbonyl compounds
from a variety of products, by testing each product in two environmental
chambers maintained at realistic but  different temperatures and air exchange
rates. The products tested were obtained in California, and include bare wood
products, coated wood products used as furniture components, cosmetics, paints,
paper goods, decorative laminates, permanent press fabrics, fiberglass insulation,
and industrially applied  wood finishes. The five target compounds (formalde-
hyde, acetaldchyde, propionaldeliyde, methyl ethyl kelone, and methyl isobulyl
kelone) were determined by DNPH derivuli/.alion and HPLC analysis. In
addition,  formaldehyde was determined using a Baltelle-developed continuous
monitor. The monitor was used both  to confirm the stabilization of chamber
concentrations, for steady-slate emission sources, and to follow the evolution and
decay of emissions, for wet products that undergo a drying or curing phase after
application. Excellent agreement was obtained between the DNPH and continu-
ous monitor results for formaldehyde, and good reproducibility was established
in duplicate tests of several  products. Formaldehyde was the only target com-
pound emitted in significant amounts from most products. The emission rates of
formaldehyde ranged from less than  I ug/nr/hour to about 1.25 x I0(< ug/nr/
hour. The highest emissions were found from a commercially-applied finish for
wood floors. The emission rales of finished or coated wood products were a
small fraction ol those of the corresponding bare wood products, and emission
rates of some consumer  products were comparable to those of the wood prod-
ucts. A single washing reduced the formaldehyde emission from new permanent
press fabrics by about 60 percent. The results of this study can be used to assess
the importance of diverse product types as sources of indoor formaldehyde
exposures.
100             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts
 Testing of Polyurethane Products for Emission of Toluene
 Diisocyanate to Indoor Air
 Thomas J.  Kelly, Jeffrey Myers, and Michael VV. Holdreit
 Batldlc. 505 King Avenue, Columbus. Ohio 43201-2693

    Polyurcthunc products obtuincd in the state of California were subjected to
 chamber testing to determine whether such products release isomers of toluene
 diisocyanate (TDI) to the indoor air in homes. The products tested included
 polyurethanc foam carpet padding, furniture cushions, and foam mattress
 material; polyurethanc coatings and varnishes; and caulks and sealants. Products
 were screened by testing groups of similar products in a 9 liter glass chamber at
 elevated temperature (I2()"F) under nearly stagnant conditions. TDI was
 determined by two methods: a commercial continuous monitor that responds to
 total TDI, and a derivati/.ation/HPLC method thai distinguishes  the 2,4- and 2,6-
 TDI isomers. These two methods showed good agreement in performance checks
 using a 2,4-TDI permeation source. In this paper the presence or absence of TDI
 emission from the various product types will he  reported, along with quantitative
 emission rates where feasible.

 Transfer Efficiency of Pesticides from Carpet to Saliva-
 Moistened Hands
David E. Camann, Tapan K. Majumdar, and H. Jac Harding
 Southwest Research Institute, San Antonio. TX  78228-0510, U.S.A.
 William D.  Ellenson
 ManTcch Environmental Technology
Robert G. Lewis
 U.S. Environmental Protection Agency, Research Triangle Park,
NC2771I

    Through frequent mouthing behavior, the young child is apt to transfer
pesticide residues from recently treated carpet to his/her saliva-moistened hands
and mouth.  An experiment was conducted to determine and compare the
efficiencies of press transfer of dry residues of formulated chlorpyrifos, pyrc-
thrins, and pipcronyl butoxidc from treated carpet to the palm of the volunteer's
hand,  which had been  moistened with human saliva, an artificial saliva, or the
surfactant dioctyl sulfosuccinate (DSS).  Mean transfer efficiencies from plush
carpet to hands moistened with human saliva, artificial saliva, and DSS were
similar, with ranges of 0.7% - 1.3% for chlorpyrifos, 2.9% - 4.8% for pyrcthrin I,
and 1.5% - 2.8%) for piperonyl hutoxidc. Saliva-moistened hands transferred
about  100 times  more  dried pesticide residue from treated carpet than dry hands
did. For determining moistened hand transfers, both artificial saliva and DSS
appear to be suitable substitutes for human saliva.

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             101

-------
Abstracts
Field-Emplaceable Air Samplers for Automated Ambient, Site,
and IAQ Use
Harvey F. Padden
BIOS International Corporation, 230 West Parkway - Unit 1,
Pompton Plains, NJ 07444

    Air samples taken on collection media lor laboratory and field analysis
remain a highly sensitive and cost-effective means of assessment. However,
collection of samples can he difficult using non-automated samplers, particularly
at remote locations. This paper describes a number of automated air samplers for
sorhent-tube and filter use. The samplers arc battery-powered, portable and
weatherproof. They are suitable for:
    Sequential sampling of 24 sorbent tubes over a period up to 1 month. This
technology is suitable for long-term ambient and superfund site  monitoring. A
meteorological station can be integrated.
    Simultaneous sampling of 4 separate media at different Hows during
multiple events over a one month maximum period. This technology is useful for
ha/.ard identification and location.
    Very rapid sampling of dual sorbent lubes for survey use. The  sampler is
very small and lightweight for ease of field use.
    The design of the sampling systems will be discussed, with attention paid to
the sampling train and How calibration methodology.
    One case history of interest to the environmental community is the use of a
variant of the 24 lube sequential samplers by the United Nations for field
verification of  Iraq's agreement to refrain from manufacture  of chemical
weapons. The units are deployed near chemical facilities and  operate unattended
for I month periods in a very severe environment. This application and the
author's field trip to Iraq will be discussed.

Multi-Media Exposure Study: Rationale and Design
R. Otson
Health Canada, HHC, HPB, Tunney's Pasture. Ottawa, Ont., Canada,
KIAOL2
P. Fellin
BOVAR Environmental, 2 Tipped Kd.. Downsview, Out.. Canada,
M3H 2V2
C. Chan
NOVAMANN (Ontario) Inc., 5540 McAdam RD., Mississauga, Ont,
Canada, L4Z IP I

    Substances which may be harmful to the environment or  constitute a danger

102             Measurements of Toxic and Related Air Pollutants, May 7-9,  1996

-------
                                                               Abstracts


 to human health require a risk assessment under the Canadian Environmental
 Protection Act. Since insufficient information exists for many of the substances
 to allow accurate assessment of their health risk, an approach for obtaining
 representative information on typical human exposures to selected compounds
 present in food, beverages, potable water and air was investigated in a multi-
 media pilot study. Fifty volunteers for the multi-media study were randomly
 selected from residents of 28 enumeration areas which were randomly selected
 and evenly distributed within 7 census sub-divisions (CSD) from the Greater
 Toronto Area. The CSD had been randomly selected from among 6009 CSD
 distributed across Canada in a previous study. Personal exposures to airborne
 target compounds over 24 h periods were measured with a passive sampler, and
 during the same period representative samples of food, beverages and potable
 water consumed by the same individuals were obtained. The target compounds
 were extracted from the samples, and the extracts were analy/.ed with GC-MS.
 The study design, procedures and the preliminary results will be described. The
 experience from this study will he used to assess the feasibility of a larger study
 across Canada.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              103

-------
                                                            Abstracts


                            Session  12
               Source Sampling Methodology
    Cuchairs: Merrill D. Jackson, U.S. EPA, Research Triangle Park, NC;
     and Raymond G. Merrill, Radian Corp., Research Triangle Park, NC

 Instrumentation for Continuous On-Line Monitoring of Non-
 Methane Organic Carbon in Air Emissions
 Somentha Mitra, Yung H. Xu, and Wenjun Chen
 Department of Chemical Engineering, Chemistry and Environmental Science,
 New Jersey Institute of Technology. Newark. New Jersey, 07102
 Gary McAllister
 Office of Air Quality Planning and  Standards. U.S. Environmental Protection
 Agency, Durham. North Carolina. 27709

    Non-methane organic carbon (NMOC) is a measure of total organic carbon
 in an air emission except that from methane. EPA Method 25 is used for quanti-
 fication of NMOC in emission sources and in ambient air. This method involves
 laboratory analysis ol collected air samples, and can not be used for real-time
 measurements. Permanent gases such as CO2, CH4 interfere in this analysis and
 so does moisture, in this research, a novel method for continuous, on-line
 monitoring of NMOC in air emissions and ambient air is presented. Detection
 limits were at ppb levels, and interference ol'permanent gases have been
 eliminated.

 Modification of Method 25 for Low VOC Sources with, High
 CO2 and Moisture Levels
 Kochy Fung
 AtmAA, Inc. 23917 Craftsman Road, Calabasas, CA 91302

    The standard Method 25 has difficulties in testing for low levels of VOC in
 sources with high CO2 and moisture content in the lest stream. Common
 problems, among others, include poor method precision, high blank relative to
 the sample, and blockage of trap by ice during sample collection. A modification
 to the method was made to replace the standard trap with a small ice water
 impinger to drop out the moisture in the test stream, collect the polar VOCs, and
control the humidity of the sample in the tank to improve recovery of non-polar
VOCs during analysis. The VOCs determined as CH4 from the water trap and
the tank (SUMMA can) provide the total non-methane hydrocarbons emitted by
the source. The method was evaluated in the laboratory and is being field-tested.
The results will be presented.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             105

-------
Abstracts
A Test Method for the Measurement of Arylamines in
Stationary Source Emissions
B.A. Pate, M.R. Peterson, and R.K.M. Jayanty
Research Triangle Institute, Research Triangle Park, North Carolina, 27709
J.E. Knoll t
U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, 27709

     Eighteen arylaminc compounds are listed under Title III of the Clean Air
Act Amendments of 1990 as pollutants to he regulated. Many of these com-
pounds arc suspected or known carcinogens, toxins, and irritants; and some
decompose when exposed to air.or light. The U.S. EPA and Research Triangle
Institute, through a cooperative agreement, are developing a lest method for the
collection and measurement ofarylamines from stationary source emissions. The
proposed method of collection of arylamincs is based on EPA Reference Method
5 (modified method 5 sampling train). The primary change consists of adding a
strong cation exchange resin after the condenser to collect arylamines. The
arylamines are analy/.ed using high performance liquid chromatography (HPLC)
and a pholodiode array (PDA) detector.
    The large number of compounds, some  of which are carcinogenic, made it
impractical to study all eighteen arylamines. A subset of eight arylamines,
representing characteristics of all the compounds, was used in the laboratory
study. The analytical limit of quanlilalion (LOQ) ranged from less than 100 ppb
for chloramben to about 15 ppm for 2,4-diaminololucne. Considerable effort was
required to identify a resin that could satisfactorily collect the arylamines and
permit their removal for analysis. The LC-SCX resin, a strong cation exchange
resin consisting of a sulfonic acid bonded silica, collected over 95 percent of
each compound tested and retained less than 5 percent when washed with an
eluanl. All compounds studied were found to be stable on  the LC-SCX resin.
    DISCLAIMER: The information in this document has been funded wholly
or in part by the U.S. Environmental Protection Agency through a cooperative
agreement with Research Triangle Institute.  It has been subjected to Agency
review and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.

Method 301 Evaluation of a Technique for Collection and
Measurement of Ammonia in Stationary Source Emissions
W. C. Eaton, E.  E.  Rickman, and R. K. M. Jayanty
Research Triangle  Institute, Research Triangle Park, NC 27709
S. A. Shanklin and D. P. Holzxchuli, Jr.
Entropy, Incorporated, Research Triangle Park, NC 27709
106             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


 R. N. Dishakjian and D. P. Hotzschuli
 U.S. EPA, Research Triangle Park. NC 27711

     Injection of ammonia or urea to the combustion /.one of a boiler reduces the
 formation (and thus emissions) of oxides of nitrogen. Under certain conditions a
 portion of the ammonia may "slip" and may be emitted to the atmosphere. State
 air quality organi/.ations have expressed interested in monitoring this and other
 sources of ammonia. Users of the injection technology need emissions data to
 assist in process optimization. In response to these needs, the EPA Office of Air
 Quality Planning and Standards supported laboratory development and field
 testing of an EPA Method 17-bascd procedure  for quantitaling ammonia at
 stationary sources.
    This paper describes the ammonia collection and measurement method and
 gives results from a field test in which samples were collected at a coal-fired
 power plant that uses urea injection technology. Samples are taken isokinetically.
 Stack gases arc pulled through an in-stack filler to remove paniculate matter and
 ammonium salts, and are then bubbled through a 0.1 N sulfuric acid impingcr
 solution to trap ammonia as ammonium ion. Ammonium in  the impingcr solution
 is assayed by ion chromatography (1C) and compared to field assay with an
 ammonia specific electrode. From the 1C results and the sample volume, the
 stack gas ammonia concentration is calculated. A quadruple train assembly with
 dynamic  spiking, as specified by EPA Method  301, was used to arrive at a
 precision estimate of 19.2% RSD. The bias estimate was 0.96 ppmV for a mean
 spiked sample value of 6.43 ppmV. The procedure met Method 301 criteria if a
 correction factor of 0.87 is applied.
    This work was  funded by the U.S. Environmental Protection Agency under
 contract 6S-D2-OI63 and conducted by Research Triangle Institute and Entropy,
 Incorporated. This paper has been subjected to  Agency review and approved for
 publication. Mention of trade names or commercial products docs not constitute
 endorsement or recommendation for use.

 Development of a Manual Method for Sampling and  Analysis
 of High Levels of Phenolic Compounds
Joseph E. Knoll
 National Exposure Research Laboratory,  U. S. EPA,
 Research Triangle Park, NC 27709
Joan T. Bursey, James F. McGaughey, Joette L. Stcgcr,
and Raymond G. Merrill
 Radian Corporation, P. O. Box 13000, Research Triangle Park, NC 2771 1

    A method has been developed to sample and analy/.c high levels of phenol
and crcsols by collecting phenolic compounds in 2N basic impingcrs. Sampling
and analysis for high levels (20 ppm or above) of phenol and the crcsols present

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              107

-------
Abstracts


special challenges to the test methodology. Because of the boiling points of these
compounds, a logical choice of test methodology would be the Modified Method
5 sampling train using Draft EPA Method 3542 for preparation of the samples
and gas chromatography/mass speclrometry (GC/MS) for analysis. However, at
these high levels. Modified Method 5 sampling trains, even using basic
impingers after the sorbent to ensure collection of the phenolic compounds, arc
ineffective because the standard pH-adjusted extraction techniques recover only
40-50% of the.phenolic compounds from the aqueous phase condensate and the
wet sorbent. A sampling method was developed to collect phenolic compounds
in 2N basic impingers.  Because of the high concentrations of phenolic com-
pounds, direct analysis of the basic impinger contents by high performance liquid
chromatography (HPLC) alter adjustment ol'pH is a very straightforward
analysis. For lower concentrations of phenolic compounds, the aqueous basic
impinger contents can be concentrated. Laboratory studies and a field test using
the criteria of EPA Method 301 have been performed. A written protocol for
sampling and analysis of high levels of phenol and cresols is being prepared on
the basis of the results of the field evaluation study.
    The information in this document has been funded wholly by the United  '
Stales Environmental Protection Agency under contract 68-D4-0022 to Radian
Corporation. It has been subjected to Agency review and approved for publica-
tion. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

Development of an Optical Method for Sampling and Analysis
of High  Levels of Phenolic Compounds
Joseph E.  Knoll,
National Exposure Research Laboratory, U. S. EPA,
Research Triangle Park, NC 27709
Joan T. Burxey, James F. McGaugliey, Jeffrey LaCosse, Thomas Selegue,
and Raymond G. Merrill
Radian Corporation,  P. O.  Box 13000, Research Triangle Park, NC 27711

    A field evaluation study designed according to the guidelines of EPA
Method 301 has been performed to assess the performance of Fourier Transform
Infrared Spectroscopy (FTIR) in the measurement of high source levels of
phenol and the cresols.  Because of the availability of data immediately in the
field, optical monitoring methods such as FTIR are increasingly being applied to
measurement of organic compounds in stationary sources. To allow the operation
of FTIR in both spiked and unspiked modes, the system was  operated cxtrac-
tively, using a sampling manifold to allow the introduction of approximately 40
ppm  ol phenol and the cresols. The FTIR system was operated simultaneously
with  a manual method for comparison to the manual method as a reference

108              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                           Abstracts


 method. EPA Method 301 was applied to provide both an independent evalua-
 tion of the FTIR method and a comparative evaluation ol'FTIR to the manual
 method.
    The information in this document has been funded wholly by the United
 Stales Environmental Protection Agency under contract 68-D4-0022 to Radian
 Corporation. It has been subjected to Agency review and approved for publica-
 tion. Mention of trade names or commercial products does not constitute
 endorsement or recommendation for use.

 Generic FTIR Sampling Method for Measuring Hazardous Air
 Pollutant Emissions from  Industrial Point Sources
 Thomas J. Geyer
 Midwest Research Institute, Suite 350. 401 Harrison Oaks Boulevard.
 Gary, NC, 27513

    In recent years EPA has worked to develop Fourier transform infrared
 (FTIR) spectroscopy for measuring ha/.ardous air pollutants. Part of this effort
 was the preparation of a Protocol with the analytical procedures used to interpret
 infrared spectra for quantitative pollutant measurements. The FflR Protocol has
 been available for public review for several years and has been successfully used
 in a preparation and validation of source, and compound, specific FTIR sam-
 pling methods to accompany the Protocol. These methods were necessary to
 address important sampling issues encountered in measuring HAPs: The
 chemical and physical properties of some  HAPs make them difficult to handle in
 a conventional sampling system. An FTIR system is ideally suited for verifying
 the  integrity of a sampling system for transporting target compounds. Recently,
 MRI has been preparing a Generic FTIR Sampling  Method for EPA.  The
 Generic Method contains procedures that use FTIR techniques to perform HAP
 measurements at a varicrty of sources. This paper presents an overview of the
 Generic Method and explains the reasoning behind its procedures.

 Application of Extractive FTIR (Fourier Transform Infrared)
 Spectroscopy for Speciated Measurement of Uncontrolled
 Emissions at an Asphalt Coating Process: Sampling and
 Analysis Challenges at a Complex Emission Source
 Thomas A. Dander, Thomas Schuster, and Craig D. Stone
 Entropy, Inc., P.O. Box 12291, Research Triangle Park, NC 27709-2291

    Air emission measurements on uncontrolled emissions at an asphalt coating
operation were performed using Extractive FTIR spectroscopy. Significant
sampling and spectral analysis challenges were encountered due to the complex

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             109

-------
Abstracts


effluent composition that included several HAPs and numerous volatile and
semi-volatile organics. The semi-volatile species included high molecular weight
aliphatic hydrocarbons that could condense in the heated sampling system and
continue to he detected at low levels for an extended period. Sampling system
contamination was miniini/.cd by limiting sampling time and purging between
test runs. On-sitc QA/QC procedures included local, system, and dynamic matrix
spiking calibrations with Mclhanol, 1,3-Butadienc, and p-Xylenc. Dynamic
matrix spiking was performed by injecting the three calibration compounds (with
an SF6 tracer) at the sampling probe. Acceptable calibrations and spike recover-
ies demonstrated the sampling system was capable of transporting these types of
compounds. FTIR spectra were quantitatively analy/.ed for 20 compounds, with
maximum possible concentrations determined lor over 100 non-detected HAPs.
Numerous other VOC compounds were positively or tentatively identified, but
reference spectra were not available for quantitative analysis. FTIR analysis
verified that most of the aliphatic hydrocarbon emissions consisted of n-Octanc
and n-Nonane, with only minor emissions of aliphatic HAPs (n-Hexane and
2,2,4-Trimethyl Penlane).

Method Evaluation Study: The Application of  VOST
to the Nonhalogenated Volatile Organic Compounds
from the Clean Air Act Amendments
Merrill D. Jackson
National Exposure Research Laboratory, U. S. Environmental Protection
Agency, Research Triangle  Park, NC 27711
James F. McGaughey, Raymond G. Merrill, and Joan T. Bursey
Radian Corporation, P. O. Box 13000, Research Triangle Park, NC 27709

    A laboratory study and  two field studies were performed to evaluate the
application of the Volatile Organic Sampling Train (VOST, EPA  Method 0030)
to the volatile halogenated organic compounds (approximately 35 compounds)
listed in the Clean Air Act Amendments (CAAA) of 1990. Subsequently, a
laboratory study was performed to assess the feasibility of the application of the
VOST methodology to the volatile non-halogenated organic compounds listed in
the CAAA (approximately 20 compounds). Several ol the volatile organic
compounds were eliminated from consideration as VOSTanalytes because they
could not be analy/.cd successfully by gas chromalography/mass speclrometry
(GC/MS), and others were eliminated because of poor analytical system re-
sponse. The remaining CAAA volatile organic analyles (ben/.ene, toluene,
carbon disulfide, 2,2,4-trimethylpentane, and liexane) were tested in the field
using the guidance of EPA Method 301 for experimental design and statistical
evaluation of the data. Quadruple VOST trains were run in the field, with
dynamic spiking of the volatile analytes from a certified cylinder into two of the

110              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


 sampling trains. The bias and precision of the VOST sampling and analytical
 methodology applied to each of the volatile organic unalytcs were evaluated.
    The information in this document has been funded wholly by the United
 Slates Environmental Protection Agency under contract 68-D4-0022 to Radian
 Corporation. It has been subjected to Agency review and approved lor publica-
 tion. Mention of trade names or commercial products docs not constitute
 endorsement or recommendation for use.

 Method Evaluation Study:  The Application of SemiVOST
 to the  Nonhalogenated Semivolatile Organic Compounds
 from the Clean Air Act Amendments
 Merrill D. Jackson
 National Exposure Research Laboratory, U. S. Environmental Protection
 Agency, Research Triangle Park, North Carolina 27711
 James F. McGaugliey, Raymond G. Merrill, and Joan T. linrsey
 Radian Corporation, P. O. Box  13000. Research Triangle Park, NC 27709

    A laboratory study and three field studies were performed to evaluate the
 application of the SemiVOST (EPA sampling Method 0010, EPA sample
 preparation Draft Method 3542, and EPA analytical Method 8270) to the
 scmivolutilc halogcnatcd  organic compounds (approximately 80 compounds)
 listed in the Clean Air Act Amendments (CAAA) of 1990. In these initial
 studies,  PCBs, dioxins, and pesticides were excluded. Subsequently, a labora-
 tory study was performed to assess the feasibility of the application of the
 SemiVOST methodology to the semivolatile nonhalogenated organic compounds
 listed in the CAAA (approximately 70 compounds). Several of the semivolatile
 organic compounds were  eliminated from consideration as SemiVOST unalylcs
 because they could not be analy/cd successfully by gas chromatography/mass
 spcctromctry (GC/MS), reacted with other compounds in solution, or were
 insoluble in mcthylcne chloride at the levels required to perform dynamic
 spiking in the field. The remaining CAAA semivolatile organic analytcs were
 grouped as acid/neutrals and base/neutrals and evaluated  in the field using the
 guidance of EPA Method 301 for experimental design and statistical evaluation
 of the data. Quadruple SemiVOST trains were run in the field, with dynamic
 spiking of the semivolatile analytes from a nicthylcne chloride solution of cither
 acid/neutrals or base/neutrals into two of the sampling trains. The bias and
 precision of the overall SemiVOST methodology (sampling, sample preparation,
 and analysis) applied to each of the semivolatile organic analyles were evaluated.
   The information in this document has been  funded wholly by the United
 States Environmental Protection Agency under contract 68-D4-0022 to Radian
Corporation. It has been subjected to Agency review and approved for publica-
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             111

-------
 Abstracts
 tion. Mention of trade names or commercial products does not constitute
 endorsement or recommendation for use.

 Stack Sampling Methods for Halogens and Halogen Acids
 Larry D. Johnson
 Methods Branch, Air Measurements Research Division, National Exposure
 Research Laboratory, U.S. EPA, Research Triangle Park, NC 27711

    Sampling and quantilalion of stack emissions from ha/.ardous waste com-
 hustors and from boilers and industrial furnaces co-firing ha/.ardous waste are
 required as part of the Resource Conservation and Recovery Act permitting
 process. Hydrochloric acid and chlorine are currently regulated, and consider-
 ation is being given to setting requirements for the similar bromine compounds.
 Because research information about the methods has not been published in one
 convenient summary, much of the technical community is unaware of its exist-
 ence.
    This paper will provide some of the historical and scientific background for
 the EPA sampling methods in use today, along with some of their strengths and
 limitations. The primary evaluation studies will be summari/.ed, and publication
 references will be given. The SW-846 Methods Manual versions of the proce-
 dures will be compared with the versions from CFR40 part 60. Relatively new
 research work will be summari/.ed, along with recent changes in the methods,
 and critical operating factors.
    NOTICE: This is an abstract of an oral presentation to be made at the EPA/
 AWMA International Symposium on Measurement of Toxic and Related
 Pollutants at RTP, NC, May 7-10, 19%. The  information in this document has
 been funded wholly or in part by the U.S. Environmental  Protection Agency. It
 has been subjected to Agency review and approved for publication. Mention of
 trade names or commercial products does not constitute endorsement or recom-
 mendation lor use.

 Development of Calibration Procedures for Toxic Metal
 Aerosols for Continuous Emission Monitoring Systems
 Thomas J. Logan
 National Exposure Research Laboratory, U. S. EPA,
 Research Triangle Park, NC 27711
James  F. McGanghey and Raymond (1. Merrill
 Radian Corporation, P. O. Box  13000, Research Triangle Park, NC 27709

    An aerosol generation system has been developed to calibrate the perfor-
 mance  of continuous emissions monitoring systems (CEMS) for toxic metal
 aerosols. The U. S. Environmental Protection Agency Office of Solid Waste has

 112             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 proposed the use of metals monitoring as an approach to better assess the risk of
 incinerator emissions. Prior to the development of metals CEMS regulations, it is
 necessary to determine a method to evaluate the expected performance of
 CEMS. To measure the performance of CEMS, a dynamic  calibration approach
 which can he used to  evaluate monitor performance must be developed and
 tested. This calibration approach requires stable aerosol concentrations of the 16
 toxic metals on a system that has a How rate of 20 liters per minute for the air
 stream containing the metals aerosol. The concentration of the metals in the
 aerosol must be able to vary over the range of 10 to 1000 micrograms per cubic
 meter, on an aerosol generating system that can operate in a stable mode for at
 least 30 minutes. The aerosol generation system that has been developed meets
 the above requirements, and has been tested in a laboratory environment.
    The information  in  this document  has been funded wholly by the United
 States Environmental Protection Agency under contract 6X-D4-0022 to Radian
 Corporation. It has  been subjected to Agency review and approved for publica-
 tion. Mention of trade names or commercial products does  not constitute
 endorsement or recommendation.for use.

 A Comparison  of Preconcentration Procedures for GC/MS
 Analysis of 1990  CAAA Volatile Hazardous Air Pollutants
 D.B. Cardin
 Entcch Instruments, Inc.

    The 1990 Clean Air Act Amendment specified a list of 189 Ha/.ard Air
 Pollutants which require monitoring and control. Analysis of these compounds
 down to 0.1 ppbv by GC/MS necessitates the prcconcenlration of large sample
 volumes with subsequent water removal before injection into the GC/MS  to
 avoid interferences  during both the chromatographic separation and the mass
 spectral detection. Preconcentration can be accomplished cryogenically using
 glass bead traps, or  cryogcnlcssly using sorbent based traps. When using
 cryogen, water management techniques that have successfully demonstrated
 utility for this application arc based on cither low temperature dcsorptions or
 cold trap dehydration. Water elimination using sorbenl traps can be accom-
 plished using cither a  dry purging technique or by using a rapid desorplion water
 knockout trap (moisture control /.one).  It has been difficult to compare the
 relative efficiencies or effectiveness of each of these approaches as recoveries
 and sampling precision is a function of not only the trapping and water manage-
 ment technique, but of the entire instrument design and control strategy. No
 single instrument has  been capable of performing all four of these
preconccntration procedures to permit a direct comparison of their advantages
and disadvantages.
    An automated preconccntration system with modular traps is presented that
allows direct comparison of trapping and water management procedures for both

 Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              113

-------
Abstracts


cryogenic and cryogcnlcss prcconccntration procedures. The preconcentrator can
ulili/.e 3-stage cryogenic trapping and focusing to perform both Microscale
Purge & Trap and Cold Trap Dehydration. Alternatively, trapping of the sample
on a 1/K" sorhenl trap followed by transfer to a 1/16" sorbent trap can accom-
plish both a preconcenlralion and sample focusing without cryogen, in which
case water is  removed either by dry purging or by a water knockout /.one
between the two traps. Precision and 5 point calibrations will be compared for
many of the polar and non-polar HAP's compounds, as well as some other
thermally labile VOC's. O/,one will also be added to each trapping system after
loading the analyles to try to reproduce any negative bias or artifact formation
that may occur during the analysis of actual air samples.

Dynamic Spiking Studies Using the DNPH Sampling Train
Joette L. Steger
Radian Corporation, P.  O.  Box 13000, Research Triangle Park, NC 27709
Joseph E. Knoll
National Exposure Rescarcli Laboratory, U. S. EPA,
Research Triangle Park, N.C. 2771 1

    The proposed aldehyde and ketone sampling method using aqueous 2,4-
dinilrophenylhydra/.ine (DNPH) was evaluated in the laboratory and in the field.
The sampling trains studied were based on the train described in SW-846
Method 0011. Nine compounds were evaluated: formaldehyde, acetaldchydc,
quinonc, acrolein, propionaldchyde, methyl isobutyl ketone, mcthy ethyl ketone,
acctophenone, and isophorone. In the laboratory, the trains were spiked both
statically and  dynamically. Laboratory studies indicated  that formaldehyde and
isophorone are efficiently recovered from the first impingcr. Laboratory studies
also investigated potential interferences to the method. Based on their potential
to hydroly/.e in acidic solution to form formaldehyde, dimethylolurea, suligcnin,
s-trioxane, hexamethlycnetelraminc, and paraformaldehyde were investigated.
Dimethylolurea, hexamethyleneletramine, and paraformaldehyde all interfered
with formaldehyde analysis. The sampling train containing 200 inL of reagent in
the first impinger followed by two impingers containing  100 mL of reagent was
then evaluated at a plywood veneer manufacturing plant. Ten runs were per-
formed using  quadruplicate sampling trains. Two of the I'our trains were dynami-
cally spiked with the nine aldehydes and  kelones. The lest results were evaluated
using the EPA Method 301 criteria for method precision (< ± 50% relative
standard deviation) and bias (correction factor of 1.00 ±  0.30). Formaldehyde,
acelaldehyde, propionaldchyde, and acctophenone passed the requirements for
accuracy and  precision.
    The information in  this document has been funded wholly by the United
States Environmental Protection Agency under contract 68-D4-0022 to Radian
Corporation. It has been subjected to Agency review and approved for publica-

114              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
 lion. Mention of trade names or commercial products docs not constitute
 endorsement or recommendation lor use.

 Imaging Mass Spectrometry of Particulate-Associated
 Polynuclear Aromatic Hydrocarbons
 Michael J. Holland and Richard W. Linton
 Department of Chemistry, University of North Carolina, Chapel Hill, NC 27599
 Joette L. Steger and Raymond G. Merrill,
 Radian Corporation, P. O. Box 13000, Research Triangle Park, NC 27709
 Greg S. Strossman and T.F. Fister
 Charles Evans and Associates. Redwood City, CA 94063

    Timc-of-llight secondary ion mass spectrometry (TOF-SIMS) was employed
 for the in situ investigation of PAHs on individual environmental particulars.
 TOF-SIMS is a microprohe mass spcctromclry technique capable of providing
 surface elemental and chemical information with lateral spatial resolution 
-------
Abstracts


Robert F. Martz, Joan T. Bursey, and Raymond G. Merrill
Radian Corporation, P.O. Box 13000, Research Triangle Park, NC 27709

    Under contract to the EPA, a guidance manual was written to address the
methods used to measure and report total organics data from stationary source
emissions. Writers of air quality permit applications for waste combustion units
require total organics data for their assessments. The guidance manual identifies
specific techniques to determine the total organics sampled from stationary
sources. To generate a value for total organics, organics data from three specific
boiling point/vapor pressure classes are combined: light hydrocarbons and
volatile organics (measured using Field GC and Purge and Trap GC),
semivolatile organics (measured using total chromatographic organics [TCO]),
and non-volatile organic  compounds (measured using gravimetric analysis
[GRAY]). The methods for measuring and reporting the individual parameters
are discussed in detail in  the guidance manual.
    The information in this document has been funded wholly or in part by the
United Stales Environmental Protection Agency under contract 68-D4-0022 to
Radian Corporation. It has been subjected to Agency review and approved for
publication. Mention of trade names or commercial products docs not constitute
endorsement or recommendation for use.
116              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
                             Session 13
                        Quality Assurance
         Chair: Sliri Kulkarni, Research Triangle Institute, RTI', NC

 NARSTO-NE Hydrocarbon Laboratory Comparison
 Eric M. Fujita
 Energy and Environmental Engineering Center, Desert Research Institute,
 P.O. Box 60220, Reno, NV 89506
 William A. Lonneman
 U.S. Environmental Protection Agency, Atmospheric Research and Exposure
 Assessment Laboratory, MD 84, Research Triangle Park, NC 2771 1
 Alan  Van Arsdale and Allen Oi
 U.S. Environmental Protection Agency, Region I - New England, 60 Westvicw
 Street, Lexington, MA 02173-3185

    A laboratory comparison study was conducted as part of an external quality
 assessment program for the NARSTO-NE study. The objectives of the labora-
 tory comparisons were to document significant systematic biases that may exist
 between laboratories, and to assess probable causes for the observed differences.
 Participating laboratories included Biospheric Research Corporation, State
 University of New York at Albany, state-operated Photochemical Assessment
 Monitoring Station (PAMS) networks in the NARSTO-NE study area (Connecti-
 cut, Massachusetts, Maine, Delaware, Maryland, Philadelphia Department of
 Public Health, Virginia, New Jersey State, and New York), USEPA Region I
 (regional rcl'crencc laboratory), USEPA -  AREAL (referee laboratory) and
 Desert Research Institute (data compilation and analysis). The laboratory
 comparisons consisted of a review of standard operating procedures (SOPs) used
 by each measurement group and performance audits using both synthetic
 hydrocarbon mixtures and ambient samples. This paper describes the protocol
 for the laboratory comparison study, summari/.e.s the finding of the study, and
 examines some of the remaining technical issues facing the PAMS program and
 their implications.

 Accuracy Considerations of Electrochemical NOX Analyzers
 C. Capetanopoulus
 Energy Efficiency Systems, Wcstbury, New York
B. Hobbs
City Technology Ltd., Portsmouth, England

    Two key components contributing to measurement errors of electrochemical
analy/.crs arc discussed. These arc the sample conditioning system and the
electrochemical nitric oxide and nitrogen dioxide sensors. The problems

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              117

-------
Abstracts
associated with various types of conditioning systems are discussed and some
experimental results are presented using analyte spiking methods. Permeation"
drier based systems are shown to cause the smallest loss of Ihe analyte. Two
major problems of the NO and NO2 sensors are examined. The first problem
deals with the significant effect of temperature on the sensor and its associated
interference rejection filter. The requirement for maintaining sensor and filter
temperature below 30 deg. C is demonstrated. The second deals with the
saturation and drift considerations caused by over exposure to the gas. The
significance of capillary si/,e to minimi/.c drift for diffusion sensors is discussed.
Experimental results are presented and discussed with a view to the recently
published EPA CTM-022 Method.

Further Analyses of Wisconsin's Remote Vehicle Emissions
Sensing (RVES) Feasibility Studies
Craig S. Rendahl, E.I.T.
Remote Sensing Technologies, Inc., 2002 N. Forbes Blvd.,
Tucson, AZ 85745-1446

    Previous papers have addressed quality assurance efforts with regard to
collecting emissions data of  known quality, data validation, and preliminary
analysis of Wisconsin's Remote Vehicle  Emissions Sensing (RVES) project
conducted in 1993 and 1994. This paper will analy/e in greater detail  the field
data collected over the two years ol'studies. This analysis includes making
comparisons of mass emissions of carbon monoxide and total hydrocarbons with
respect to vehicle model year and total contribution to tropospheric o/.onc-
forming emissions in southeastern Wisconsin. A simple analysis of errors of
commission and errors of omission as a function of varying RVES cut points will
be reviewed. And finally, potential emissions reductions gained from the use of
remote vehicle  sensing will also be explored.

Quality Control Guidance for Sampling and Analysis of Air
Toxics
Patrick Farrell, Ann Rosecrance, and Raj Naran
Core Laboratories - Gulf Stales Analytical, 6310 Rolhway Drive,
Houston,  TX 77040

    The inclusion of effective and comprehensive quality control procedures is
critical for the performance of accurate air toxics measurements. Lack of quality
control can add significant error to environmental measurements. The  U.S. EPA
has estimated that 95% of the total error in environmental measurements is due
to sample collection and handling (85% from sample collection and 10% from
subsampling in the laboratory) and that 5% occurs during laboratory analysis.

118             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 This paper provides quality control guidance for performing accurate air toxics
 measurements of ambient air and source emissions samples for volatile organic
 chemicals. The areas covered are equipment handling, field sampling and
 laboratory analysis. Quality control guidance for equipment handling includes
 checks on How controllers, vacuum gauges, pressure gauges, manometers and
 canisters. Quality control guidance for field sampling includes sampling equip-
 ment checks, sample handling and field quality control. Quality control guidance
 for laboratory analysis includes certified standards, laboratory control samples,
 duplicates, blanks, internal standards, surrogate standards, instrument calibration
 and data evaluation. The recommended frequency and acceptance criteria for
 quality control procedures are discussed.

 Effects of QA/QC Procedure Changes on Method 25 Analysis
 and 50 ppmC Method  Detection Limit
 W. Stollings, J.  Venezia, and L. Taylor
 Triangle Environmental Services. Inc., P.O. Box 13294, Research Triangle Park,
 North Carolina

    A comparison was made on samples collected (including ambient air,
 hydrocarbon free air, and certified master gas standards) and analy/.ed over a
 period of months. These sample sets were repealled utili/.ing increasing stricter
 QA/QC procedures. The average range of each of the sample types and the
 difference between  them were compared for consistency. Parameters leading to
 potential bias such as combined small backround concentrations in the recovery
 system, the volume sampled, clcaniness of the sampling equipment, etc. were
 studied.
    Examination of the results suggests that by the removal or deduction of the
 background values the Method Detection Limit could be lowered from the
 current value of 50ppmC. This  view is supported by the additional comparison
 of field sample results below using Method 25, the results demonstrate the
 possiblity os increased applicability of this method at lower levels.

 Uncertainty Analysis for Open-Path Remote Sensing of
 Fugitive Emissions
James Flanagan
Center for Environmental Monitoring and Quality Assurance, Research Triangle
 Institute, Research Triangle Park, NC 27709
Richard Shores
Air Pollution Prevention and Control Division, U.S. EPA,
Research Triangle Park, NC 27711

    Open-path remote sensing techniques such as Fourier transform infrared

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              119

-------
Abstracts


(FTIR) spcctrometry offer a powerful approach for determining emission rales
from line, area, and volume air pollution sources. Typically, a photon beam
downwind of the source intersects the pollutant plume, and a characteristic such
as optical absorption is measured. The path-integraled concentration is calcu-
lated from this measurement. The emission rale can then be estimated using a
tracer gas reference or by dispersion modeling.
    Open-path monitoring has important advantages over conventional methods
for measuring fugitive emissions. However, a different set of design and quality
assurance (QA) considerations must be addressed in developing the measure-
ment and data analysis protocols. Failure to consider the method's unique
characteristics can impair the accuracy and precision of the fugitive emissions
calculation. This paper will use a hypothetical area source monitored by FTIR to
illustrate how the following factors can affect overall uncertainty:
    •    length and location ol the optical  path,
    •    release and measurement of tracer gas,
    •    meteorological measurements,
    •    selection of a dispersion model, and
    •    estimation ol  dispersion coefficients and other model parameters.

Data Quality Objectives for the Yeast Manufacturing Industry
Lara P. An try
Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711
Michael J. Messner
Center for Environmental Measurements and Quality Assurance, Research
Triangle Institute, Washington, D.C., 20036-3209

    EPA (Environmental Protection Agency) must set a MACT (Maximum
Achievable Control Technology) standard  for acetaldehydc emissions from the
manufacturing ol yeast. Once established, the standard  will apply to all existing
and new yeast manufacturing facilities.
    Ordinarily, the DQO (Data Quality Objectives) process would  be used in the
up-front planning of a data collection effort. In this case, however, the data arc
already in hand. Two yeast manufacturers are cooperating with EPA and have
provided data on acetaldehyde emissions and yeasl production. The DQO
process was used to determine how these data will  be used in setting the new
standard.
    This presentation describes the selected DQOs through Step 6 of the  seven-
step DQO process. Step 7, "Optimi/.e the design." was not applicable, as data
generation had already been completed  by  the two manufacturers. Because the
manufacturers used different measurement methods, comparability became an
issue. Measurement performance information will be presented and discussed.
120             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
 To (he extent possible, actual data will he used to illustrate how the DQOs were
 used to derive the proposal standard.

 Overview of the ISO 14000 Environmental Management
 Standards
 Gary L. Johnson
 U.S. Environmental Protection Agency, Quality Assurance Division (MD-75),
 Research Triangle Park, NC

    In 1996, the International Organi/ation for Slandardi/.alion (ISO) will
 publish the first of a scries of international standards addressing environmental
 management. The emerging ISO 14000 environmental management standards
 pose a potentially greater impact on the world community than the ISO 9000/
 I()()()() standards that preceded them. Like quality, environmental management
 cuts across all business sectors; however, unlike quality, environmental issues
 include regulatory aspects, diverse public interest groups, and an arguably
 broader scope of applications.
    Initially, 1S standards have been proposed to encompass environmental
 management systems (EMS), environmental auditing (EA), environmental
 labeling (EL), environmental performance evaluation (EPE), and life-cycle
 assessment (LCA). Five standards have reached the stage of draft international
 standard (DIS)  and two of these are  the proposed EMS standards. ISO 14001
 and ISO 14004.
    This paper provides a general overview of the ISO 14000 standards,
 including their contents and their role in the overall ISO 14000 series. This paper
 will also briefly contrast these standards with the ISO 9000 quality management
 standards currently in use.

 The National Ambient  Air Quality Standards and Quality
 Assurance
Joseph B.  Elkins, Jr.
 U.S. EPA, Office of Air Quality Planning and Standards (MD-14),
 Research Triangle Park, NC 27711
Stan Sleva
TRC Environmental Corporation , 6340 Quadrangle Drive, Suite 200,
Chapel Hill, NC 27514
Randall G. Waite, David Mustek, and Michael Papp

    The U.S. Environmental  Protection Agency (EPA) has established National
Ambient Air Quality Standards (NAAQS) for 6 pollutants. These pollutants arc
carbon monoxide, lead, nitrogen dioxide, o/.one, paniculate matter, and sulfur
dioxide. The EPA's ambient air monitoring program is carried out by State and

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             121

-------
Abstracts


local air agencies and consists of four major categories of monitoring stations:
State and local air monitoring stations (SLAMS); national air monitoring stations
(NAMS); special purpose monitoring stations (SPMS); and photochemical
assessment monitoring stations (PAMS). These networks are composed of
approximately 4835 monitors across the nation. The gathering of the data from
these networks is required by the Clean Air Act and 40 CFR 58. The data from
these networks are used fora wide variety ol  purposes including: NAAQS
attainmcnl/nonulluinmcnt decisions; determining the effectiveness of air pollu-
tion control programs; evaluating the effects of air pollution levels on puhlic
health; tracking progress of State Implementation Plans (SIPS); providing
dispersion modeling support; developing responsible, cost-effective control
strategies; reconciling emission inventories; and developing air quality  trends.
Henceforth, the importance of quality data from these networks can not be
overstated. The Quality Assurance/Quality Control (QA/QC) of the  national
ambient air monitoring program is critical. This paper discusses several major
QA/QC components including: the data quality objective (DQO) process, the
reference and equivalent methods program, the precision and accuracy  (P&A) of
the collected data, EPA's National Performance Audit Program (NPAP), systems
audits, and network reviews.

A Report on the Activities  of the  Photochemical Assessment
Monitoring Stations (Pams)  Quality Assurance Work Group
Randall G. Waite
US Environmental Protection Agency, Office of Air Quality Planning and
Standards (MD-14), Research Triangle Park, NC 27711

    The Quality Assurance Work Group is a team of federal and state experts,
formed under the auspices of the PAMS program. The purpose of the Work
Group is to cooperatively explore new procedures, techniques, and strategies to
ensure the appropriate level of confidence in the data collected. Four main areas
of focus are:
    I.   Quality Assurance and Quality Control Guidance. The present guidance
        is being updated to be more specific and clear, and to include data
        validation recommendations, minimum data acceptance criteria, and
        instrument specific procedures.
    2.   Data Validation and Translation  Software. Available software is being
        evaluated and additional software is being developed as needed to
        ensure thai data can he processed as quickly and accurately as  possible.
    3.   Calibration Standards. Proficiency Studies, and Audits. A long term
        strategy for providing calibration and retention lime standards  and for
        determining the appropriate frequency and type  of audits and profi-
        ciency studies for PAMS sites is being developed.


122              Measurements of Toxic and Related Air Pollutants,  May 7-9, 1996

-------
                                                               Abstracts


     4.   Outreach, Training, and QA Reports. A communications and technical
         assistance plan is being developed.
     The results of the Quality Assurance Work Group's efforts to date will he
 discussed.

 The National Ambient Air Quality Standards and Quality
 Assurance
 Joseph B. Elkins, Jr.
 U.S. EPA, Office of Air Quality Planning and Standards (MD-14),
 Research Triangle  Park, NC 27711
 Stan Sleva
 TRC Environmental Corporation . 6340 Quadrangle Drive, Suite 200,
 Chapel Hill, NC 27514
 Randall G. Waite,  David Mustek, and Michael Papp

    The  U.S. Environmental Protection Agency (EPA) has established National
 Ambient Air Quality Standards (NAAQS) for 6 pollutants. These pollutants arc
 carbon monoxide, lead, nitrogen dioxide, o/.one, particulate matter, and sulfur
 dioxide. The EPA's ambient air monitoring program is carried out by State and
 local air agencies and consists of four major categories of monitoring stations:
 Stale and local air monitoring stations (SLAMS);  national air monitoring stations
 (NAMS); special purpose monitoring stations (SPMS); and photochemical
 assessment monitoring stations (PAMS). These networks are composed of
 approximately 4835 monitors across the nation. The gathering of the data from
 these networks is required by the Clean Air Act and 40 CFR 58. The data from
 these networks arc  used fora wide variety of purposes including:  NAAQS
 attainmcnt/nonattainment decisions; determining the effectiveness of air pollu-
 tion control programs; evaluating the  effects of air pollution levels on public
 health; tracking progress of State Implementation  Plans (SIPS); providing
 dispersion modeling support; developing responsible, cost-effective control
 strategics; reconciling emission inventories; and developing air quality trends.
 Henceforth, the importance of quality data from these networks can not be
 overstated. The Quality  Assurance/Quality Control (QA/QC) of the national
 ambient air monitoring program is critical. This paper discusses several major
 QA/QC components including: the data quality objective (DQO) process, the
 reference and equivalent methods program, the precision and accuracy (P&A) of
 the collected data, EPA's National Performance Audit  Program (NPAP), systems
 audits, and network reviews.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              123

-------
Abstracts


Performance Comparison of Field-Deployable Gas
Chromatographs with Canister/TO 14 Analyses
Richard E. Berkley
U.S. EPA, Research Triangle Park, NC 27711
Mike Davis
U.S. EPA, Region Vii, Kansas City, KS 66115
Mark Hansen
U.S. EPA, Region VI, Dallas, TX 75202-2733
Dennis Lane
University of Kansas, Lawrence, KS

    Field-deployahlc and portable gas chromalographs (PGC) provide immedi-
ate data without storage of samples pending analysis, hut PGC data arc often
trusted less than laboratory data because of the vieissistudcs of field operation
and because of design compromises intended to enhance the portability and
sample througput. If the capabilities ol'PGCs were better established, they might
be used  to pilot collection of data for laboratory analyses, to produce large
volumes of replicate data from simultaneous samples later analyzed  in the
laboratory.
    A comparison field study was conducted during which four commercially-
available PGCs were operated downwind of a source generator. Concurrent
eolocaled PGC and canister samples were collected. The PGC samples were
analy/ed immediately, and the canister samples were analy/.cd later by Method
TO-14. Each instrument also analy/.ed audit standards of TO-14 target com-
pounds kept in pressuri/.ed canisters at five different concentrations in the low
ppd range. Adverse weather typical of field operating conditions complicated the
study. Samples were collected, when possible, during periods of stable wind
velocity and minimal precipitation. A  mobile tripod-quadrupole mass spectrom-
eter (TAGA) operating downwind of the PGCs ensured that the plume was
established before each run. Agreement between PGCs and the reference method
was imperfect for plume analysis data because of weather conditions. Better
performance was observed when audit standards were analyzed.
124              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts
                             Session 14
                              General
           Cochairs: R.K.M. Jayanty, Research Triangle Institute,
                      Research Triangle Park, NC;
                Donald Adams, Consultant,  Pullman, WA;
            and William Gutknecht, Research Triangle Institute,
                       Research Triangle Park, NC

 A Comparison of Dry Deposition Modeled from Size
 Distribution Data and Measured with a Knife-edge Surrogate
 Surface
 Kenneth E. Noll, Thomas M. Holsen, Seung-Muk Yi, Xiang 7.hu, and
 Rajendra D. Paode
 Department ol'Chemical and Environmental Engineering, Illinois Institute of
 Technology, IIT Center, Chicago. IL 60616

    Numerous mass and elemental dry deposition fluxes were measured in
 Chicago, IL, in South Haven, MI, in Sleeping Bear Dunes, MI, and over Lake
 Michigan with a knife-edge surrogate surface. In general the average measured
 fluxes in Chicago were higher and more variable than the fluxes measured away
 from Chicago. In addition the most northern site (Sleeping Bear Dunes, MI) had
 consistently lower fluxes than the other locations. The llux of primarily anthro-
 pogenic metals (Cd, Cr, Cu, Mn, Ph, V, and Zn) was on average 1-2 orders of
 magnitude lower than the flux of primarily crustal elements. Atmospheric
 concentrations were generally highest in Chicago and lowest in the northern part
 of the lake. A modeling procedure thai used measured atmospheric particle si/c
 distributions and modeled deposition velocities was used to calculate the dry
 deposition llux for comparison to the measured flux data. The average ratio of
 calculated/measured llux varied between 0.2 and 2. In general (luxes were
 slightly underestimated probably due to an underestimation of coarse particle
 deposition velocities. Modeling results indicate that the majority of the llux (>98
 %) was due to particles >6.5(Jin in si/.c. A comparison of simultaneously
 measured dry dcpositional llux and the concentration of airborne paniculate
 matter <•!() pm in aerodynamic diameter (PM 10) indicates that PM10 concentra-
 tions are not a good measurement from which to estimate dry deposition llux.

 Can Selected RADM Simulations be Aggregated to Estimate
 Annual Concentrations of Fine Particulate Matter?
Brian K. Ederand Sharon K. LeDuc
 Atmospheric Sciences Modeling Division, ARL/NOAA,
 Research Triangle Park, NC 27711

 Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             125

-------
Abstracts


    The benefits analyses proposed for the Clean Air Act Amendments of 1990
require annual timescales. Unfortunately, most Eulerian models, like EPA's
Regional Acid Deposition Model (RADM), utili/.cd in this study, challenge the
practical limits of current computer resources as well as our ability to collect the
pertinent input data. As a result, application of such models to determine Ihe
long-term relationship between changing emissions patterns and ambient air
concentrations is limited. To circumvent this problem, an aggregation method
(initially developed for acid-deposition applications) has been modified and
applied  to a limited number of RADM simulations in order to provide estimates
of long-term (annual) ambient air concentrations of line particulate matter.
    The aggregation method is based on the premise that at any given location,
ambient air concentrations of fine particulale matter are governed by a finite
number of different, though recurring meteorological regimes. If a scries of
concentration patterns representative of these different meteorological regimes
can be identified, they can be aggregated to produce reasonable estimates of
annual averages.
    This paper describes how the simulation periods were selected and com-
bined to establish long-term estimates and how representative these estimates
were.
    This aggregation method performed belter than a random selection process.

Comparison of Risk Management Dispersion  Modeling
Methodologies
Kevin Eldridge and Krixten Harvey
Roy F. Weston,  Inc.

    Section  I 12 (r) of the Clean  Air Act Amendments of 1990 requires that
facilities that process, use, store,  or otherwise handle substances in quantities that
exceed threshold amounts develop risk management plans. The final  Risk
Management Programs for Chemical Accidental Release Prevention  rules are
due to be issued in early  1996. Facilities subject to these rules must comply
within three years of rule promulgation. The basic provisions of the rules may
require affected facilities to: register, conduct a ha/.ards assessment, develop an
accident prevention program, develop an emergency management program,
develop a  risk management plan, and provide recordkeeping. As part of the
ha/.ards  analysis FPA is proposing that facilities conduct an off-site impact
analysis. This paper provides the results of a comparison analysis between
various  dispersion models and their application to various accidental release
scenarios. ALOHA,  HGSYSTEMS, TSCREEN/SCREEN3, ARCHIE, and
SLAB are compared for a range ol' release scenarios. The analysis will compare
modeling results lor gaseous, liquid, dense gas, neutrality bouyant, instantaneous
and continuous releases.  In addition, an evaluation will be made of various


126             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts


 release heights and meteorological conditions. The results will demonstrate
 which model(s) arc more appropriate for the various release scenarios evaluated.

 The Analysis of Ambient Air Samples for Polar Volatile
 Organic Compounds
 Jeffrey Shire, Gregory Harshfield, and Barbara Zielinska
 Desert Research Institute, P.O. Box 60220, Reno,  NV 89506
 Randy Pasek
 Research Division, California Air Resources Board, P.O. Box 2815,
 Sacramento, CA 95812

    Three separate methods for analy/.ing polar volatile organic compounds
 (PVOCs) in ambient air samples are compared and the appropriateness of each
 method in quantifying specific PVOCs is assessed. In the first two methods,
 PVOCs in ambient air samples are trapped onto mulliabsorbenl beds, trapped
 water is purged  from the sorbent bed with inert gas, then the samples are
 analy/cd by cither GC/FTIR/MS or high resolution gas chromatography with
 Manic ionization detection. The third method included in the comparison consists
 of EPA Method TO-14, in which sample air is cryogenically trapped from
 polishcd-intcrior-surfacc stainless steel canisters and analy/.cd via high resolu-
 tion gas chromatograph with flame ioni/.ation detection. The confounding effects
 of water present in samples are addressed for each  of the methods in this study.
 Specific compounds analy/.cd for include melhyl-t-bulyl ether, methanol,
 elhanol, and other PVOCs of importance.

 A Wind Tunnel  Investigation of the Flow Structure
 within a Dense Gas Plume
 Guwei Zhu and S. Pal Arya
 Department of Marine, Earth and Atmospheric Sciences, North Carolina State
 University, Raleigh, NC 27695

    A wind tunnel study was performed to investigate how a dense gas plume
 could affect the mean How and turbulence structure in the boundary layer. For
 this, a neutral atmospheric boundary layer with an acrodynamically rough
 surface was generated in the EPA Meteorological Wind Tunnel. The dense gas
 plume was simulated by releasing C02 through a ground-level, small circular
 area source. Large roughness elements were used to ascertain that the dense
 plume was turbulent. The initial Froudc number of the dense plume was varied
by changing the  frccstream wind speed and keeping the flow rate of the dense
gas (CO,) constant.
    Special consideration was given to make accurate velocity measurements
within the dense-gas plume due to the sensitivity of hot-wire and hot-film

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              127

-------
Abstracts


anemometers to the CO2 concentration. For accurate concentration measure-
ments, fast response ilamc ioni/ation detectors (FID's) were used. Flow visual-
ization and concentration measurements indicated thai buoyancy had significant
effects on mean How and turbulence structure in the dense gas plume, as well as
plume dispersion.
    Vertical velocity profiles were measured at the same locations as the vertical
concentration profiles, both in the neutral (air) and the dense (CO.,) gas situa-
tions, so that the effects of density stratification could be discovered. In  the
dense plume, the mean velocity profiles were changed significantly in shape near
the surface at low wind speeds. These changes were not only related to the
magnitude of the CO, concentration, but also to the plume height; as the wind
speed increased, the differences of the  profiles between the dense and neutral
plumes diminished. Both the longitudinal and vertical turbulent intensities were
greatly reduced at low wind speeds. Shear stress and  friction velocity were also
considerably reduced. Appropriate scales and dimensionless parameters will be
used, so that the results may  be generali/.ed. The gradient Richardson number is
shown to he the most appropriate parameter for describing the changes in the
mean flow and turbulence structure.

Performance Characteristics of Adsorbents Utilized in Sample
Preparation Modes  of Operation
W.R. Betz, M.J. Keeler, and LI). Degrajf
Supeleo Inc., Supeleo Park, Bellefonte, PA I6H23

    Characterization studies of both porous and non-porous adsorbents  have
been performed using dynamic challenge techniques. The focus of these studies
was to determine the relative adsorption strengths of all commercially available
adsorbents using a system which can be constructed in most analytical laborato-
ries using commercially available equipment.
    For the past several years, we have studied  the adsorption characteristics of
adsorbents/adsorbent lubes using both  a constant challenge inlet approach and a
single elulion point approach. The data obtained from both of these gas-solid
chromatographic (GSC) approaches provide effective insight into the perfor-
mance characteristics of the adsorbents/adsorbent tubes, and assisted  in normal-
izing the data generated over the past decade in  this lab as well as other labs
involved in adsorbent characterization  efforts.
    Also,  a second on-going program in our lab is the preparation of sub-micron
adsorbent  particles which can be adhered to surfaces such as glass and stainless
steel. The  subsequent adsorplion/desorption properties of the resultant surface
arc currently being investigated for their performance characteristics in sample
preparation modes of operation.
128             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
 Comparative Mutagenicity of a Standard Reference Material
 in the Salmonella Assay
 Thomas Hughes, Larry Claxlon, and Joelleii Lcwtax
 U.S. EPA, NHEERL, MD68, Research Triangle Park, NC 27711.

    The purpose of this research was lo determine accurate and reproducible
 slope values (S) in revcrtants/uG Tor a diesel extract in the Salmonella mutagc-
 nicity assay. The diesel extract was a standard reference material (SRM  1975)
 that was Soxhlet extracted with dichloromethanc from 500 Ihs of paniculate
 material obtained from filters in operating forklilts. The SRM 1975 will  he
 available from the NIST in 1996, along with the mutagenicity data generated
 here and a comprehensive analytical  chemistry analysis. Mutagenicity param-
 eters were: TA98. TA100, TA9SNR. TAIOONR (nitroreductase (NR) gene
 deficient), YG 1021, YG 1026 (NR gene addition). YG1024. YGI029
 (acetyltransfcrasc (AT) gene addition), and YG1041 and YG1042 (contain both
 NR and AT genes); 10 dose levels in the linear portion of the dose response
 curve; duplicate plates per dose; S9 at 1.1 nig of protein/plate; plate incorpora-
 tion assay for all strains and Kado microsuspcnsion assay lbrTA98, TA100,
 YG 1021, YG 1026.  Results suggested: (I) mutagenic activity without S9
 correlated with the presence or absence ol  the NR or AT genes; slope values
 (rcvertanls/uG) were TA98NR(I28). TA98(445), YG 1021 (605),
 YG 1024(2096), YGI()4I( 11,443); (2) the Kado microsuspension assay  was 2-
 fold more sensitive  than the plate incorporation assay (e.g., YGI02I, -S9, S =
 605 vs.  1317); (3) the addition ol'S9 significantly reduced the mutagenic activity
 of SRM  1975 (e.g., YG1024, plate, S = 2096 vs. 268); (4) the addition of the NR
 and AT  genes increased the mutagcnic activity of the SRM 1975 89-fold from
 TA98NR(S=I28) to YG1041(S=11,443); (5) the TA98 scries listed above was
 4-20 fold more sensitive than the TAIOO series in measuring the mutagcnic
 activity of the diesel extract; (6) YG104I was sensitive lo nitroarenes at the 10
 nanogram level. These data can be useful for ranking other diesel or air samples
 for mutagcnic activity, QA of data generated in other labs, QC within a lab, and
 as positive control values for future air and automobile exhaust studies.
    This abstract does not necessarily rellect EPA policy.

 Mutagenic Activity of Raleigh Air Samples  at Three Different
Elevations  Before, During and After Hurricane  Gordon
Fred Stratton and Viney P. Aneja
Department of Marine, Earth, and Atmospheric Sciences, North Carolina Slate
University, Raleigh, NC 27695-8208
Robert C. Smart
Department of Toxicology, North Carolina State University,
Raleigh, NC 27695-8208

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             129

-------
Abstracts


Thomas Hughes and Larry D. Claxtun
NHEERL, EPA, Research Triangle Park, NC 27711

     Diurnal o/.one and VOCs were collected and measured at the Raleigh
WRAL lower at three different levels: Surface, Mid (240 m), and Top (460 m) in
July and August of 1993 and 1994. Measurements revealed high ox.onc concen-
trations at Mid and Top. Of the VOCs measured, only arene concentrations were
significantly elevated at Mid and Top and correlated with wind direction from
Raleigh. Due to arene stability, the combined presence of o/.one, arenes and N0x
suggested possible arene nitration. Consequently to collect nitroarcncs, samples
were collected with XAD-filled canisters using high volume air samplers at the
same three levels on lower days before, during, and after hurricane Gordon.
Collected  samples were Soxhlet extracted and analy/.cd with S. typhinuirimn YG
1021 and YG 1026 (+/-S9). Due to sample si/.e limitations of collected air
extracts, the microsuspcnsion assay was used. Mutagcnicily in post-Gordon
stagnant air mass samples with YG 1026 w as observed and the slopes (rcvertant/
ug) were: Top (5.3) and Mid (3.3) with S9, and Top (1.7) and Mid (3.9) without
S9. The Surface level,  Blank, and pre- and during Gordon samples were not
mutagenic. These preliminary data suggest possibility of long-term transport of
mutagenic nitroarenes.
     This abstract does not necessarily reflect IIP A policy.

Update of EPA's Compendia of Toxic Organic and Inorganic
Sampling and  Analytical Methods for Ambient Air
William T. "Jerry" Winberry, Jr.
Midwest Research Institute, Gary, NC
Justice Manning
U.S. Environmental Protection Agency, Cincinnati, OH
Frank McElroy
U.S. Environmental Prolecion Agency, Research Triangle Park, NC

    Air pollution has been a subject of interest and concern for many years.
With the passage of the Clean Air Act of 1970, the U.S. Environmental Protec-
tion Agency (EPA) devloped Federal  Reference Methods (FRMs) for sampling
and analysis of criteria pollutants (i.e.. SO2, NOx. O3, CO, and total suspended
particulars) so that regulatory decisions could be based on precise and accurate
data. With the passage  of the Clean Air Act Amendments of 1990, the Agency
now faces  new challenges of regulating ha/.ardous air pollutants (HAPs). Title III
addresses the control (source/fugitive) of a list of 189 designated HAPs of which
88 percent are organic chemicals or mixtures of organic chemicals, with the
remaining  21 percent occurring as particle-associated phase constituents. At this
time, no uniform national program for measuring HAPs exists; rather, a number
of individual, unrelated, local or State networks or special studies exist in which

130             Measurements of Toxic and Related Air Pollutants, May 7-9,  1996

-------
                                                              Abstracts


 a variety of monitoring and analytical techniques arc used to obtain results.
 Consequently, the results vary widely, as does the data quality. The absence ol
 standardized procedures raises serious concerns about the comaplihility of the
 data collected with its ultimate use.
     As awareness of this problem increased it became necessary for EPA to
 provide uniform, systematic, and reliable methodology to characleri/.e HAPs in
 the ambient air. Through EPA's Center for Environmental Research Information
 (CERI) and EPA's Atmospheric Research and Exposure Assessment Laboratory
 (AREAL), two compendia are being developed. To address the need for moni-
 toring methodology for organic HAPs, a revised Compendium of Methods for
 the Determiniation of Toxic Compounds in Ambient Air has been published. For
 inorganic HAPs, the EPA published a Compendium of Methods lor the Determi-
 nation of Inorganic Compounds in Ambient Air. The objective of this paper is to
 discuss  these two new Compendia and their applicability to charactcri/.ing HAPs
 in ambient air through the use of uniform, systematic, and reliable methodology.
 The presentation will focus on specific methodology for the following classes of
 HAPs.
         • Atmospheric polar organic compounds;
         • Atmospheric acidic and basic gases;
         • Atmospheric mercury vapor and particles;
         • Semi-volatile organics; and
         • Sampling and analysis for atmospheric metals.
    The presentation will discuss organi/ation of the Compendia and methods
 format and validation. Specifically, discussions will explain how the methods arc
 based upon performance criteria. Historically, the Agency has dictated required
 methodology in the Federal Register eliminating flcxibilly of implementation.
 Tcncnts for a performance- base approach will he discussed, including data
 quality objectives, calibration procedures, performance validation, and auditing
 process.

 Estimating the  Risk From Exposure to Wind-Borne Surface
 Soil Emissions
Rhonda S. Berger and Don Marti
Dragun Corp., Farmington Hills, MI

    Inhalation  ol chemicals present in wind-borne surface soil emissions can be
an important route of exposure when developing a risk assessment for a site
containing both erodible and nonerodible areas of exposed surface soil. The U.S.
EPA-apprqvcd Industrial Source Complex Short Term 3 (ISCST3) air dispersion
model was used to estimate the concentration  of potential PM10 surface soil
emissions in air. Model input parameters and receptor locations were designed
specifically for several different potential exposure scenarios being considered in

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              131

-------
Abstracts
the risk assessment. A sensitivity analysis was performed to assess the effect of
breathing /.one height. Exposure point chemical concentrations for wind-borne
surface soil emissions were calculated and evaluated lor exposure to both the
maximum concentration in air at a receptor point and the 95 pcrcentile upper
confidence limit for receptor points located in the area of concern for a particular
exposure scenario.

Sampling & Analysis of Various Wastewater Streams at a
Petrochemical Facility for Compliance with the Hazardous
Organic NESHAPs (HON) Rule
Alston Sykes, Roy Gorman, and Dana Carey
Quanterra Env. Scrv.-RTP, NC
Paul Pilarczyk
Naismith Engineering

    This paper will discuss the sampling & analysis methods 25D and 305
which are referenced in the HON Rule to determine compliance. The HON Rule
requires the owner or operator of a Synthetic Organic Chemical  Manufacturing
Industry facility to determine if they will be required to control ha/.ardous air
pollutants (HAPs) from several points. Waslewater streams are considered to be
a potential air emission point at these facilities, and EPA has established
Methods 25D and 305 as reference methods for this determination. Samples
were collected at  a petrochemical facility to determine if their emissions of
volatile organics would classify their wastcwaler streams as a Group I (requiring
no control devices) or as Group II (requiring MACT). As with any new sampling
and analysis procedure, errors will occur initially in understanding the applicabil-
ity, in interpreting the results, and in following the methods as intended by the
EPA. The methods may nol have addressed all of the possible problems that
could occur. The  authors have been using these methods for several months,  and
have encountered many issues. These issues were brought to the attention of
EPA for clarity and recommendations. Some of the issues thai will be discussed
arc:
    I.   alternative methods;
    2.   sampling protocols;
    3.   QA/QC;
    4.   analytical procedures; and
    5.    industry concerns.

Perimeter Air Sampling Strategy for Soil Remediation at a
Superfund Site
Sandra L. Potter
The Forrester Group Inc.

132              Measurements of Toxic and Related Air Pollutants, May 7-9,  1996

-------
                                                              Abstracts
 William M. Vaughan
 Environmental Solutions
 Jean A. Mesccr
 McKesson Corporation

     The Arkwood Supcrfund Site is a former wood treating facility located in
 northwest Arkansas. Soil at the Site was affected hy pcnlachlorophenol (PCP),
 carcinogenic polynuclcar aromatic hydrocarbons (c-PNAs), and dioxin. The
 remedial action consisted of soil prc-treatment hy dry sieving and oil-Site
 incineration.
     During the remedial action, perimeter and off-Site air monitoring were
 performed to ensure protection of public health and the environment. Monitoring
 was conducted using high volume samplers for total suspended particulars and
 PS-1 PUF samplers for chemical analyses. Total dust data at each downwind
 station was evaluated on a one to twelve work day cycle, dependent on the
 project schedule. The PUF sample at the station with the highest total dust
 reading for the cycle was submitted for PCP and c-PNA analysis. If the results
 exceeded the Site action levels, the sample was then analy/.ed for dioxin.
     This sampling strategy allowed for analysis of the PUF sample reasonably
 suspected  to contain the highest level of contaminants, thereby reducing the
 number of PUFs requiring analysis. Approximately 987 PUFs were collected.
 However,  using the above sampling strategy, only 190 PUFs were analy/.cd. This
 strategy reduced the analytical costs by 80 percent.

 GC/MS  Analysis of Volatile Organic Compounds During Site
 Remediation
Matthew M. Booth, Michael G. Window, and Dwight F. Roberts
GC/MS Department, Environmental Science & Engineering, P.O. Box  1703,
Gainesville, Florida 32605

    Laboratory analysis of ambient air samples was performed in support of the
remediation of a urban  industrial facility located in the Northeast United States.
Site remediation required demolition of the buildings on the site, installation of a
RCRA cap and treatment of contaminated groundwaler. The schedule for this
work was very aggressive, with new construction over the site expected to begin
8 months alter first demolition.
    Since  this work entailed the remediation of a site containing RCRA and
TSCA regulated wastes, the methods used for the analysis of the air samples had
to approach those used  for the analysis of ha/.ardous waste samples at similar
sites. For this work, USEPA method TO 14 was used with the QC requirements
taken from the USEPA  Contract Lab Program Volatile Organics Analysis
methods with slight modifications.

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              133

-------
Abstracts


    Results for all of the QC obtained during the analysis of the air samples
associated with this project shall he presented, with special emphasis placed on
the use of surrogates added to the SUMMA Passivated canisters prior lo sam-
pling.

Evaluation of Sludge Characteristics and Source Emissions
from Municipal Sewage Sludge Incinerators
Robert S. Hart
Stearns & Whclcr Environmental Engineers & Scientists, One Remington Park
Drive, Ca/.cnovia, NY 13035

    The United States Environmental Protection Agency (USEPA) has promul-
gated regulations affecting the disposal of municipal sewage sludge under Title
40 of Code of Federal Regulations (CFR) Part 503. The paper will address
emissions requirements for sewage sludge incineration under 40 CFR Part 503
Subparl E. The paper will focus on I'actors which influence sewage sludge
characteristics, sludge metals feed rates, and the corresponding metals emission
rales.
    Test programs were conducted at three municipal sewage sludge incinera-
tors lo measure metals emissions rates. Concurrent to source testing, metals
content in the influent sludge feed to the incinerator were also measured. Metals
emissions were determined using USIiPA Method 29, and sludge metals
determination was based on USEPA Method SW846. The results from these
programs will he analy/.ed to identify the factors that may influence the metals
emission rates. Particularly, emission rales of toxic metals such as mercury,
cadmium, and arsenic will be evaluated in detail. Metals removal efficiencies by
the installed air pollution control equipment will be quantified. Additionally,
unique sampling and analytical issues encountered in these test programs will
also be addressed.
134              Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                           Abstracts


                            Session 15
                        Human Exposure
  Cochairs: M. Rodon-Naveira, U.S. EPA, Research Triangle Park, NC; and
  Edo 1). Pellizzari, Research Triangle Institute, Research Triangle Park, NC

 Using Environmental Exposure Data to Set Environmental
 Health Priorities/Interventions in a Community Setting.
 Harold Zenick
 National Health and Environmental Effects Research Laboratory, Office of
 Research and Development, U.S. EPA, Research Triangle Park, NC 27711
 Gerry Akland
 National Exposure Research Laboratory, Office of Research and Development,
 Office of Research and Development. U.S. EPA, Research Triangle Park.
 NC 2771 I

    An expanded role for environmental/human exposure data is emerging
 relative to the identification and prioriti/.ation of potential health risks in
 communities and defining the appropriate research and intervention strategics.
 This cxposurc-to-discasc approach can be integrated with traditional environ-
 mental monitoring and public health models to produce a more appropriate
 environmental health paradigm. Such a model will be described in this presenta-
 tion as well as the additional research and data bases needed to strengthen its
 application.
    This is an abstract of a  proposed presentation and docs not necessarily  '
 reflect EPA policy.

 Evaluation of Field Methods for Estimating Exposure of
 Children in Low-Income Families to Polycyclic Aromatic
 Hydrocarbons
 Nancy K. Wilson
 U.S. EPA, Research Triangle Park, NC 27711
Jane C. Chuang
 Batlellc, 505 King Avenue,  Columbus, OH 43201
 Christopher Lyu
 Survey Research Associates, lOOCapitola Drive, Durham, NC 27713

   Children in low-income families arc thought to have higher exposures to
 polycyclic aromatic hydrocarbons (PAH) and related compounds than children
 in higher-income families. These higher exposures could result from the location
of their homes, nearer to industrial sites and traffic; from poorer diet; from
environmental tobacco smoke; or other causes. This study was designed to

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996            135

-------
Abstracts


evaluate methods and estimate total exposures of low-income children to PAH
through various pathways. Nonsmoking participants with preschool children,
incomes at or hclow the official US poverty level, and space heating in their
homes were recruited. The PAH concentrations were measured in the household
indoor and outdoor air, house dust, yard soil, and diet, and urinary metabolites of
PAH were measured for both an adult and a preschool child living in the home.
A preliminary study in two homes and an additional study of eight homes, half
urban and half rural, during the heating season have been completed. The
problems and successes encountered  in the recruitment process and the results of
the heating season measurements will be discussed.
     This abstract has been reviewed in accordance with the U. S.  Environmental
Protection Agency's peer and administrative review policies and approved for
publication.  Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

Measuring and Modeling Indoor-Outdoor Pollutants in a
Residence Using Real-Time Methods
W. Ott, M. Zelenka, A. Huber, and W. Nelson
U.S. EPA, National Research Laboratory, Research Triangle Park, NC 2771 I
P. Switzer
Department of .Statistics, Stanford University, Stanford, CA 94305
N. Kelpeis
Lockheed Environmental Systems and Technologies, Las Vegas, NV 89119

    State-of-the-art, real-time monitors were used to measure carbon monoxide
(CO), line particles, particle-bound polycyclic aromatic hydrocarbons (PAH),
and o/.one (O3) indoors and outdoors at 5-minute intervals in a California
residence over a period of 1-1/2 years. The 18-month study period allowed
seasonal changes to be examined in this Residential Exposure Project, and the
high time resolution of the monitors allowed individual outdoor sources (motor
vehicle, wood smoke) to be identified. The house — a 4 bedroom, occupied
home — was free of indoor sources during most of the monitoring period,
allowing a mathematical indoor-outdoor model to be developed and evaluated.
The mathematical model was used to  predict the indoor concentration time series
from the outdoor time series, anil the model included parameters for the ventila-
tory air exchange rale, the decay rale due to deposition on indoor surfaces  and
removal processes. By examining an outdoor ambient episode for CO and
PAH, ,, it was possible to determine all the decay rate parameters  for the model.
This analysis showed that the penetration factor included in prior indoor air
quality models actually was unity and could he eliminated from the equations.
The predicted indoor concentration lime series for all four pollutants was
calculated from the model using (he observed outdoor time scries as ihe input.

136            Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                             Abstracts
 and the resulting predielions were compared with the observed indoor time
 scries. The effect of varying air exchange rales on the model predictions was
 examined. Regression of the predicted indoor lime series on the observed indoor
 time scries shows that the model can accurately determine indoor concentrations
 of the pollutants, given accurate values of the individual parameters.

 Demonstration of Real-Time Measurements of PAH and CO to
 Estimate In-Vehicle Exposure and Identify Sources
 Timothy J. Buckley and Wayne R. Ott
 U.S. EPA, National Exposure Research Lab (MD-56),
 Research Triangle Park, NC 27711

    Significant inhalation exposures to polycyclic aromatic hydrocarbons (PAH)
 and carbon monoxide (CO) may occur in motor vehicles due to vehicular
 emissions of these contaminants, the proximity of the occupant to the source, and
 the amount of time spent commuting. Portable measurement instrumentation has
 recently become available so that real-lime particle-bound PAH can be measured
 simultaneous with real-time CO. These new instruments were coupled and used
 in a commuting study to evaluate the concentrations and variability ol'CO and
 PAH inside motor vehicles in traffic. Measurements were made on  I I trips in
 each of two cars (22 total) over prescribed routes in two locations (North
 Carolina and California). This paper describes the measurement methodology
 and demonstrates the utility of the measurements in assessing short-term tran-
 sient exposures and in distinguishing exposures from dicsel or gasoline fueled
 vehicles.
    This abstract has been  reviewed in accordance with the U.S. Environmental
 Protection Agency's peer and administrative review policies and approved for
 publication. Mention of trade names or commercial products docs not constitute
 endorsement or recommendation for use.

 Cooking as a Source of Fine Particles: Experiments in a
 Townhouse
Lance A.  Wallace
 U.S. EPA, National Exposure Research Branch,
 Research Triangle Park, NC 2771 I

    Experiments were carried out in an occupied townhouse to determine a) the
effects  of cooking on fine (PM,5 )particlc concentrations indoors; and b) the
decay rate on interior surfaces  of aerosols created by healing cooking oils. It was
 found that cooking with oil could elevate fine particles by amounts ranging
between 10 and 100 g/m\ and a few seconds with the oil hot enough lo emit
visible  smoke were sufficient lo elevate levels  to between 150 and 1000  g/m\

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              137

-------
Abstracts


Seven experiments lo determine the decay rate for k of PM   particles on interior
surfaces resulted in a value for k of 0.33 ±0.08 h'1. This value was in agreement
with the value for PM   particles of 0.39 ±0.18 h'1 estimated from statistical
analyses of 178  Riverside, CA, homes.

Biases in Personal Aerosol  Air Exposure Sampling
Charles E. Rodes, Philip A. Lawless, and Thomas M. Peters
Research Triangle Institute, Research Triangle Park, NC
Lance A. Wallace
U.S. EPA, National Exposure Research Lahoratory, Research Triangle Park, NC

    Recent examination of various aspects of personal aerosol sampling
suggests that phenomena such as the "personal cloud" may significantly bias the
relationship hetween a monitor worn on the body and one in a fixed location in
common microenvironmenls. Particles greater than about Ijjm were shown in
laboratory and residential settings  to he resuspcnded from certain indoor surfaces
while walking, increasing significantly the average room concentrations. Levels
oi'PM25 increased only slightly in  residential settings, while PMm levels in-
creased substantially by over 50|jg/nv'.
    The presence or absence of a "bluff body" simulating the physique of a
person during fixed location sampling can also influence the aerosol mass
concentration. Controlled tests using an Aerodynamic Particle Si/.er were
conducted to determine the sampled si/.e distributions penetrating personal
sampler inlets. These preliminary tests suggest that fixed-location aerosol
sampling for PM|() without a form  simulating the human torso (behind the inlet),
significantly overestimates the exposure lo 2.5 to lOum particles that would be
made with the same inlet worn on  the body. Based on limited data, the levels of
computed mass concentration average 15 to 20% greater without a bluff body
behind the inlet  for both PM,, and PM|(| si/.e fractions.

Strategies for the Development and  Application of Screening
Techniques for Human Exposure and Environmental
Assessment Surveys
J. M. VanEmitn, J. C. Johnson, K. R. Rogers, O. A. Sadik,
and J. J. Quackenhoss
U.S. EPA, National Exposure Research Laboratory, Characleri/.ation Research
Divison, Las Vegas.  NV 891 19

    Exposure assessment studies arc frequently hampered by the lack ol suitable
low-cost analytical methodologies. The EPA NERL CRD-LV is addressing this
issue by developing and applying technologies for implementation into monitor-
ing studies. The CRD-LV immunochcmistry program has investigated the use of

738             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                              Abstracts


 immunoassays for the detection of pesticides in foods and dislodgeahlc residues.
 Supercritical fluid extraction has been coupled successfully with immunoassays
 to analy/c heterogeneous food matrices. Methods for several widely used
 pesticides (e.g. atra/inc, carharyl, metolachlor. etc.) Are heing refined for use in
 exposure assessment surveys. An immunoassay forchlorpyrifos is hei'ng adapted
 to a wide range of food crops (e.g. peach, cucumber, nectarine) to rapidly
 measure dislodgcablc residues in the field to determine sale re-entry. Due to
 their widespread distribution and acute toxicily characteristics, phenolic com-
 pounds show a potential for human and ecosystem exposure. Because these
 compounds arc difficult to measure using standard methods, even under labora-
 tory conditions, methods for screening these compounds in the field could be of
 considerable value. Results will be presented describing an en/.yme-electrode  for
 detection of phenolics. These results will also include the characteri/.ation of this
 biosensor with respect to 20 environmentally significant phenolics (i.e. com-
 pounds listed as ATSDR Priority Ha/ardous Substances.) The CRD-LV is also
 investigating the teaming ol conductive elcclroactivc polymers with antibodies to
 develop a reusable immunoscnsor. These clectroimmunochemical methods show
 promise for the continuous monitoring of waste streams and ground water.
 Methods for pesticides and the polychlorinatcd biphenyls will be discussed.

 The Exposure Potential Matrix: A Dietary Model Using Extant
 Databases to  Evaluate Exposure to Chemical Residues
 Maurice R. Berry
 U.S. EPA, National Exposure Research Laboratory,  Cincinnati, OH
 J. Robert Tomerlin
 Technical Assessment Systems, Inc., Washington, DC.

    Dietary modeling is used in USEPA/NERL's monitoring programs to
 identify the relative importance of diet to other pathways of exposure and to
 indicate when more comprehensive assessments or measurements are warranted.
 Dietary exposure modeling will indicate the potential for high exposure of the
 US and regional populations, and lor certain sub-populations caused by high
 consumption of a particular food, or a highly contaminated food, or unique
 dietary characteristics, such as regional and ethnic influences.
    In many cases the information needed to characteri/e the dietary intake  of an
 environmental contaminant may he obtained from existing databases on food and
drinking water consumption and contaminant residue levels. An inventory of
 such dietary data was followed by the creation of a personal computer-based
dietary exposure model and database system, termed the Exposure Potential
 Matrix (EPM), that correlates food consumption and contaminant residue
 information in a format for exposure modeling. Currently, consumption informa-
tion is based on USDA National Food Consumption Surveys (NFCS) conducted
from 1977-92. The contaminant residue database available to EPM includes

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             139

-------
Abstracts
	1	

government-sponsored monitoring programs, such as the FDA Total Diet Study
(TDS), the Pesticide Residue Information System (PRIS) and the California
Pesticide Monitoring Program. Residue information from the national databases
are extracted, limits of detection are determined from methods codes or follow-
up with laboratory analysts, and custom residue databases arc created for the
EPM. Consumption is based on 10 food groups containing approximately 800
exposure core food types, established from over 6500 common food items, thai
provide a logical format for linking the contaminant residue information to  the
consumption databases through recipes developed for exposure analysis.
    Though not intended for risk analysis, the EPM will be useful for human
exposure measurement programs (o identify highly consumed and highly
contaminated food  items that potentially account for the majority of dietary
exposure to a specific contaminant. Ongoing efforts are directed at testing the
EPM and expanding it through the inclusion ol additional, relevant and  more
recent consumption and residue databases. A current version of the EPM is
available for demonstration and participant review.

Quantitative Analysis of Road and Carpet Dust  on Shoes
John W. Roberts
Engineering  Plus, 818 207m Ave. NE, Redmond, WA 98053
Ernest R. Crutcher, 3rd, and Earnest R. Crutcher, 4th
Microlabs Northwest, 7609 14()th PI.  NE, Redmond, WA 98052
Gregory Glass
Environmental Consultant, 8315 B 5th Ave. NE, Seattle, WA
Thomas M. Spittler
U.S. Environmental Protection Agency, 60 Weslview St., Lexington, MA

    The goal  of this preliminary study was to test the hypothesis that there is
preferential track-in of small particles by measuring the quantity and si/.e
distribution ol' dust  on a smooth vibram shoe sole after a walk over I) standard
Ari/.ona road dust, 2) a carpet embedded with such dust, 3) a clean carpet
following a dust walk, and 4) a suburban road with unpavcd shoulders. The
concentrations of lead in the road and carpet dust were also compared.
    The dust was placed in a box and walked on with a shoe to determine the
number of steps on  the dust required to achieve a maximum shoe dust load. A
similar experiment was done to determine the minimum number ol steps by a
dusty shoe on a carpel required to produce a minimum shoe dust load. Samples
were collected from shoes after walking on:
    I.   dust;
    2.   a carpet embedded with dust;
    3.   a clean carpet after walking on dust;
    4.   and  a suburban road.
    3M transparent Magic tape was used to lift the particles off the shoe sole

140             Measurements of Toxic and Related Air Pollutants, May 7-9, 7996

-------
                                                              Abstracts


 alter each tcsl lor automated image analysis. Each experiment was repeated six
 times. Finally, the concentration of lead in the dust and in a carpel was compared
 alter extensive tracking from the dust to a clean carpet. The dust in the carpet
 was sampled with the High Volume Small Surface Sampler (HVS3).

 Conducting Child Behavior Studies  to Identify and
 Characterize Pathway and Route of Exposure Through Direct
 Observation and Videotaping
 C.C. Dary and S.C. Horn
 U.S. EPA, National Exposure Research Laboratory, Las Vegas, NV
 D. Feliciano and J. Dawson
 Versar. Inc., Springfield, VA
 P. A. Craig
 Lockheed Environmental Systems and Technologies, Las Vegas, NV
 D. A. Eckerman
 University of North Carolina, Chapel Hill, NC

    The Office of Pesticide Programs (OPP) and the Office of Pollution
 Prevention and Toxics (OPPT) have rccogni/.ed the study of human activity
 patterns in certain environments as an important issue in the regulatory oversight
 of chemicals, particularly, pesticides and other consumer products, in residential
 environments. In response, the National Exposure Research Laboratory (NERL)
 in Las Vegas initiated a program to identify human activities of sensitive
 populations, particularly children, that caused them to contact and transfer
 surface residues in residential environments. Infants and children were chosen as
 the sensitive population focus group because, as identified by the Select Com-
 mittee on Children, Youth and Families, (Environmental Toxins and Children:
 Exploring the Risks, Part I, 1991), children have greater skin surface area to
 body weight, they perform daily activities in close proximity to contaminated
 surfaces, and they arc undergoing rapid development of tissues, organs, and
 organ systems. Moreover, little is know about the activities of children that lead
 to random or incidental exposure to chemicals used in the home.
    As a first step, NERL Las Vegas needed to draw together behavioral,
 technical, and exposure assessment information into a focused protocol  for
 studying non-occupational, or more precisely, incidental exposure of children in
 residential environments. To accomplish this first step, a program was designed
 to explore the pediatric and behavioral science literature for activities such as
 "hand-to-mouth" activity that could be used to explain exposures such as
geophagia and pica, the eating of soil and extraneous matter. The activities of
children were compiled into various demographic groups by age, sex, ethnicity
and parental guidance. Statistical prior distributions of activities were established
based on demographic characteristics. In the next step, the prior distributions

Measurements of Toxic and Related Air Pollutants, May 7-9, 1996              141

-------
Abstracts


were then tested against data obtained from a field study as a prototypical day
care center where the behavior of children were directly observed by experts and
by videotaping. What emerged from these studies was a data base of distribu-
tions for certain patterns of behavior that when examined against the biomechan-
ics of contact and transfer of surface residues through video analysis could be
used to predict exposure to children in residential environments without their
involvement in human subject testing.

The National Human Activity Pattern  Survey (NHAPS)
Applications for Exposure Assessment
William C. Nelson and Miriam Rodon-Naveira
National Exposure Research  Laboratory, Air Exposure Research Division,
MD-56, Research Triangle Park, NC 27711

    The National Human Activity Pattern  Survey (NHAPS), a two-year prob-
ability-based telephone interview survey of 9,386 individuals, was conducted to
fill a need for updated activity information on a nationwide scale. Recent
exposure monitoring studies have shown thai human activities play a critical role
in explaining variation in human exposure  because they impact the frequency,
duration, and intensity of exposure to pollutants. Activity pattern data bases with
adequate potential pollutant exposure information have been available only fora
few cities  (e. g., Cincinnati. Denver, Washington, DC) or a state (California), and
only for limited months of the year. The NHAPS  was performed for the USEPA
from October, 1992 through September, 1994, by the Survey Research Center at
the University of Maryland, College Park,  to ascertain the time, location, and
oilier characteristics of activities which are most relevant to estimating pollutant
exposure.  Respondents provided recall of sequential events from the preceding
24-hour day as well as answers to follow-up questions for activity and location
of exposure-related occurrences. Time-activity  data can be effectively used in
several types of statistical analyses, including descriptive, relational, temporal,
and exposure modeling. Also, the analyses can  be defined by various levels of
characteristics. For example,  the micro-environmental components can be
summarized by temporal factors (lime of day, day of week, season), by spatial
items (region, stale), by demographic elements  (age, gender, ethnicity, years of
education), and by frequency and duration  of occurrence. The NHAPS survey
design, data base configuration, and statistical analyses results will be presented
and discussed. Examples of how the NHAPS lime-activity data base can be used
to refine current human exposure models will he described.
142             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                            Abstracts


Application of Human Activity and Microenvironmental
Concentration Data Models in THERdbASE to Exposure
Assessment
M.D. Pandian and E.J. Furtaw
University of Nevada, Las Vegas, NV
JJ. Quackenboss
U. S. EPA, Las Vegas, NV

    Abstract not available.

Applying Results from Ecological and Human Exposure
Monitoring Studies for Reducing Risks and Prioritizing Future
Environmental Health Research
Gerry Akland and Miriam Rodun-Navcira
National Exposure Research Laboratory, Air Exposure Research Division,
MD-56, Research Triangle Park, NC 27711

    The need for total environmental exposure information in assessing risks
and developing risk management strategies is only a recent phenomenon. Despite
its importance, assessment of total human and ecosystem exposure has been
relatively neglected as an area ol research. This may be a result of media-specific
legislation, such as the Clean Air Act, Clean Water Act, and Safe Drinking
Water Act. Another obvious reason is due to the fact that the U.S. Environmental
Protection Agency (EPA) has been organi/.ed along media lines, for example,
Office of Water and Office of Air and Radiation. As a result, the total environ-
mental health research field which incorporates exposure information from all
the media relative to a human and/or ecological receptor is relatively new with a
very limited data base from which to draw upon when developing risk manage-
ment strategics and developing risk-based research priorities.
    The objectives of this presentation arc to illustrate by way of recent total
exposure monitoring studies that (1) monitoring results will reduce the uncertain-
ties associated in the risk assessment process; (2) total exposure information can
be used to develop cost-effective risk reduction strategies; and (3) results from
these studies provide  useful information to  priorili/.e research projects and to
develop efficient exposure study designs. Advantages and disadvantages
associated with use of common and newly developed environmental and
biological indicators in relationship to the above mentioned objectives will also
be discussed.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             143

-------
Abstracts
Evaluating Food Contamination Scenarios for Dietary
Exposure Studies
Linda S. Sheldon
Research Triangle Institute, Research Triangle Park, NC 27709
Maurice Berry
U.S. Environmental Protection Agency, Cincinnati, OH 45268

    Of the three major exposure pathways (inhalation, adsorption, ingeslion),
ingcstion may account for significant exposures to many hazardous environmen-
tal contaminants. Dietary exposure research is now considered an integral part of
the U.S. Environmental Protection Agency's (EPA) Human Exposure Research
Program. For dietary exposure to occur, foods contaminated with ha/.ardous
chemicals must he available and consumed. Major sources of contamination
include commercial and noncommercial (i.e., home gardening) food supplies.
Alternatively, unconlaminated food supplies may be contaminated by incidental
contact in a polluted microcnvironment. Over the years, regulation and monitor-
ing of commercial food supplies has been the responsibility of the Food and
Drug Administration (FDA). EPA's research, on the other hand, should provide
information on important pollutant exposures through unregulated sources of
contaminated food.
    The  purpose of (his research is to test the hypothesis that "food stored,
prepared, or eaten in a contaminated microenvironmenl" will serve as a vehicle
of dietary exposure to chemicals in that microenvironment. Throughout this
project, testing has been conducted to determine the extent to which VOCs,
pesticides, and lead will partition from air, water, and exposed surfaces into
various foods using controlled sorption studies, to generate partition parameters
that can be used to predict food contamination with these pollutants under actual
environmental conditions, to measure pollutant uptake by food during cooking
with contaminated water, and to  measure actual contamination of lead, pesti-
cides, and VOCs through a series of field monitoring tests in contaminated
microcnvironmenls including homes and gas stations.
    Results have shown substantial  contamination of foods for all pollutants and
all contamination scenarios (air,  surfaces, and water). As an example, foods
boiled in lead-contaminated water will absorb more than 80% of the lead from
the water. Similarity, foods serve as very strong sinks for VOCs in air; equilib-
rium partition experiments have shown rapid and  quantitative adsorption of
VOCs onto foods. Details will be provided on the experimental approach and the
resulting data. Implications of these  results for designing dietary exposure
studies will also be discussed.
    Although this research was conducted as part of EPA Cooperative Agree-
ment CR 822070-01-D, it has not been subjected  to the required peer and
administrative review and does not necessarily reflect the views of the Agency,
and no official endorsement should be inferred.

144             Measurements of Toxic and Related Air Pollutants, May 7-9, 1996

-------
                                                         Abstracts


 Measurement of Street-Level Air Pollutants in Bombay City
 M.V.R. Kotweswara Reo
 Consultant. Hyclerbad, India

    Abstract not available.
Measurements of Toxic and Related Air Pollutants, May 7-9, 1996             145

-------