Measurement of
Toxic and
Related
Air Pollutants
Durham, North Carolina
May 1994
Cosponsored by
U.S. Environmental Protection Agency's
Atmospheric Research and Exposure Assessment Laboratory
and
Air & Waste Management Association
-------
Measurement of Toxic
and Related Air Pollutants
An international symposium and course
cosponsored by
AIR & WASTE MANAGEMENT
ASSOCIATION
•
SINCE 1907
U.S. Environmental Protection Agency's
Atmospheric Research and Exposure Assessment Laboratory
-------
Contents
Session 1: Predicting VOC Emission
Rates, Air Quality Models, Deposition
Frank Binkowski, Chair
Model Predicted VOC Emission Rates Based on Measured
Concentrations and Meteorological Data Ray E. Carter, Jr., Dennis
D. Lane, Glen A. Marotz, et al.
Approximation of the Maximum Impact on Air Quality from Area
Source Emissions Norman A. Huey and Joseph F. Froechtenigt
Aircraft Measurements of the Transfer Velocities of Gases and
Aerosols Across Lake Michigan Surface During the Summer A J.
Alkezweeny and J.L. Stith
Aerosol Dispersion Simulation Using OMEGA/ATDM TJ. Dunn,
D.P. Bacon, Z. Boybeyi, et al.
Session 2: VOC Portable Instruments/Methods
Richard Berkley, Chair
New Portable Micro Gas Chromatograph for Environmental Analysis
E.B. Overton, K.R. Carney, H.P. Dharmasena, et al.
Triple Sorbent Thermal Desorption/Gas Chromatography/Mass
Spectrometry Determination of Vapor Phase Organic Contaminants
C.Y. Ma, J.T. Sheen, A.B. Dindal, et al.
Remote Monitoring of Sub Ppb Levels of Vinyl Chloride,
Dichloroethylene and Trichloroethylene Via Modem Operated
Automated GC Amos Linenberg and Neil J. Lander
Field Portable Environmental Sample Concentrator for VOCs in Air
Mark W. Burns and Kent Hammarstrand
Session 3: Acid Aerosols
Robert Burton, Chair
Spatial Variation in Acidic Sulfate Concentrations within Metropolitan
Philadelphia R.M. Burton, HJJ. Suh, GA. Allen, et al.
The Effects of Meteorology on Concentrations of Acid Aerosols
Michael P. Zelenka and Helen H. Suh
Elevational Gradients of Atmospheric Sulfur and Nitrogen Species
Measured at Two Experimental Watersheds in the Eastern U.S. JJ.
Bowser, E.S. Edgerton, and S. Nolan
-------
Gas-Particle Distribution, Neutralization and Size of Sulfate and 10
Nitrate in the Southwestern Desert Aerosol B.J. Turpin, P. Saxena,
P.H. McMurry, et al.
The Influence of Transport on Aerosol Strong Acidity A.B.M. Jeuken, 11
JR. Brook, andJ.D. Spengler
Clean Air Status and Trends Network Cloud Water Monitoring 11
Program /./. Bowser, E.S. Edgerton, and VA. Mohnen
Session 4: Carbonyl Sampling and Analysis Techniques
Robert Arnes, Chair
fa.H,€«tl i j» ()lnvest;igation of Aldehyde Chemistry in Sampling Canisters Thomas 13
^"^V .Kelly and Michael W .Holdren
A New Technique for Carbonyl Sampling and Analysis Sucha 14
S. Parmar, Luda Ugarova, Carmo Fernandez, et al.
Determination of Carbonyl Compounds by HPLC/UV Analysis in the 14
CASTNet Air Toxics Monitoring Program (CATMP) Bradley
A. Weichert, Robert D. Baker, Benjamin T. Johnson, et al.
A Sequential Cartridge Sampler for Measuring Carbonyls and Toxic 15
Organic Compounds DwightA. Landis
Measurement of Carbonyl Compounds in an Industrialized Area James 15
R. Ellis, David F. Gregorski, Michael J. Murphy, et al.
Session 5: Enhanced Ozone
Monitoring Status and Development
Larry Purdue and Nash O. Gerald, Cochairs
Photochemical Assessment Monitoring: Overview and Current Status 17
Nash O. Gerald and Geraldine Dorosz-Stargardt
VOC Measurements
William A. Lonneman, Cochair
Overview of VOC Technology in the PAMS Program William. 17
A. Lonneman
New Developments for the Monitoring of VOCs in PAMS William 18
A. McClenny
Validating Data from Automated, Continuous PAMS GC Systems 18
Joann Rice
A QA Program for the Photochemical Assessment Monitoring Network 19
Jack C. Suggs, Howard Crist, William J. Mitchell, et al.
-------
InterRegional and Interstate Quality Assurance for PAMS: Central 19
Regional Laboratory Comparisons Avraham Teitz and Marcus Kantz
RTF Auto-GC Demonstration Experiment Jimmy C. Pan, Ron Drago, 20
James Reagan, et al.
Connecticut Auto GC Experiment A. Leston 21
Analysis of Ozone Precursor Compounds in Houston, Texas Using 22
Automated, Continuous Gas Chromatographs John Gibich, Larry
Ogle, and Paul Radenheimer
Laboratory Intercomparison for Analysis of Hydrocarbons During the 23
Coastal Oxidant Assessment for Southeast Texas (COAST) Project
Eric M. Fujita, Douglas R. Gqff, Douglas R. Lawson, et al.
A Comparison of the Results of a Total Non-Methane Organic 24
Compound Analyzer with those from a Canister-Based Monitoring
Program Michael Poore and Natividad Lapurga
A Fully Remote Control Cryogenless Ozone Precursor System with 24
Improved Sensitivity D£. Cardin and J.T. Deschenes
The Determination of Ozone Precursors with a Built-in Preconcentrator 25
and Capillary GC/Dual FID Norman Kirshen and David Coe
AutoGC Systems for Enhanced Ozone Monitoring Network Ian 26
Seeley
Enhanced Monitoring of Ozone Precursors - Speciated and Total 26
NMOC Sharon Reiss and Richard Jesser
Carbonyl Measurements
R. Zweidinger, Cochair
Parameters Influencing the Precision and Accuracy of Carbonyl 27
Measurements Using DNPH Cartridges Kochy Fung
The Design and Use of DNPH Coated Silica Cartridges for the 27
Analysis of Carbonyls in Outdoor Air R. Crowley, J. Arsenault,
E. Bouvier, et al.
Measuring Ambient Carbonyls Using DNPH-Coated Substrates: 28
Comparison of CIS and Silica Gel with and without Ozone Scrubbers
Daniel Grosjean and Eric Grosjean
An Evaluation of Commercial Devices for Automated Sequential 28
Sampling of Carbonyl Compounds in Air Thomas J. Kelly, Deborah
L. Smith, Richard N. Smith, et al.
-------
Performance Measurements of Cj-C3 Carbonyl Compounds Using 29
DNPH-Coated Silica Gel and C18 Cartridges TE. Kleindienst, E.W.
Corse, F.T. Blanchard, et al.
Evaluation of an Intelligent Multi-Canister/Multi-Cartridge Sampler for 30
the Collection of Ozone Precursors Mark K. Allen, Edward Miller,
and Joseph Leair
The Development of a Carbonyl Sampler Matthias J. Yoong, Michael 30
A. Pardee, and Ralph D. Herdman
Supelco™ Carbonyl Ambient Air Sampler for EPA - 600/4-84-041, 31
Method TO-11 Dwight Patterson
Oxides of Nitrogen - NO
Joseph Sickles, Cochair
Measurement of Total Reactive Odd-Nitrogen (NOy) in the Rural and 31
Non-Rural Troposphere Scott T. Sandholm and John D. Bradshaw
Recent Advances in the Design of NO2 Photolytic Converters John 32
Bradshaw, Scott Smyth, Rongpo Bai, et al.
Development of NO/NOX/NO Monitors for Urban Air Sampling 32
Thomas J. Kelly, Chester W. Spicer, and Gerald F Ward
A Commercial Approach to NO Measurement Dieter Kita 33
Temporal Variation of Nitrogen Oxide Fluxes from Agricultural Soils 33
in Piedmont North Carolina Benny D. Holbrook, Viney P. Aneja, and
Wayne P. Robarge
Meteorology
Gennaro H. Crescenti, Cochair
Meteorological Instrumentation Requirements for the Photochemical 34
Assessment Monitoring Station Gennaro H. Crescenti
SODAR, RADAR Profiler and RASS Operating Principles and PAMS 35
Applications Gary S. Ziegler and Charles E. Riese
Siting Guidance for Boundary Layer Meteorological Profilers John 35
E. Gaynor
Quality Assurance for PAMS Upper Air Monitoring Sites Brian 36
Templeman
-------
Collecting and Interpreting Upper Air Meteorological Data for the 37
PAMS Network Using RADAR Profilers and RASS Charles G. (Lin)
Lindsey and Timothy S. Dyes
Data Assessment & Interpretation
B. Parzygnat, Cochair
Photochemical Assessment Monitoring: Program Objectives and Data 38
Uses Nash O. Gerald and Barbara A3. Parzygnat
Receptor Modeling of VOC Data Charles W. Lewis, Teri L. Conner, 38
Ronald C. Henry, et al.
Motor Vehicle Volatile Hydrocarbon Source Profiles for Chemical 39
Mass Balance Receptor Modeling Teri L. Conner, William
A. Lonneman, and Robert L. Seila
PAMS Data Uses for Modeling and Control Strategy 39
Development/Assessment Richard D. Scheffe
Results of Diurnal Non-Methane Hydrocarbon Monitoring at Fresno, 40
California Michael Poore, Michelle Dunlop, and Jacquelyn Milliron
Preliminary PAMS Data Analyses Terence Fitz-Simons and James 40
Hemby
Panel Discussion
William F. Hunt, Jr., Moderator
Photochemical Assessment Monitoring Implementation Issues William 41
F. Hunt, Jr.
Session 6: Source Sampling
Merrill Jackson and Raymond Merrill, Cochairs
Sampling and Analysis Information Aids for Stationary Source 42
Personnel Merrill D. Jackson and Larry D. Johnson
Hexavalent Chromium Emissions from Aerospace Operations — A 42
Case Study Ashok Chaurushia and Charles Bajza
The Use of Canisters/GC-MS and a Portable Gas Chromatoqraph to 43
Characterize Emissions from an Air Stripper Chris Figueroa and Jon
Bennett
Improvements in Preparation of Samples Generated by SW-846 44
Method 0010 Merrill D. Jackson, Larry D. Johnson, James
F. McGaughey, et al.
-------
Evaluation of GC Detectors for Total Gaseous Nonmethane Organic 45
Compounds Stephanie B. Philipp, Dave-Paul Dayton, Raymond
G. Merrill, et al.
Sampling of Volatile Organic Compounds from Combustion Sources 46
Using Tedlar® Bags and Analysis Using GC/MS Rohini Kanniganti,
Richard L. Moreno, Joan T. Bursey, et al.
Field Evaluation of a Modified VOST Sampling Method Merrill 47
D. Jackson, Larry D. Johnson, Robert G. Fuerst, et al.
Comparison of Sampling and Analytical Methods for the Collection 47
and Determination of Methylene Diphenyl Diisocyanate (MDI) from
Oriented Strand Board (OSB) Sources Mark D. Baker, William
J. Karoly, and Michael F. Adams
Pen-Based Computer System for Performing Source Test Calculations 48
FrankR. Clay
Session 7: Global Climate Change, Mount Mitchell
V.K. Saxena, Chair
Monitoring of Regional Chemical Climate Change at Mount Mitchell, 49
North Carolina VK. Saxena
Air Mass History Versus Cloud Water Acidity: Observations and 50
Model Results From a Remote Rural Site J.C. Ulman, V.K. Saxena,
K.L. Burns, et al.
The Effect of Anthropogenic Pollution on Cloud Microstructure, 50
Chemical Composition and Albedo K.L. Burns and V.K. Saxena
Aerosols as a Natural Tracer of Air Masses C.K.DeiningerandV.K. 51
Saxena
Greenhouse Warming, Acidic Deposition, and the Dilemma of Climate 51
Change JD. Grovenstein and VjK. Saxena
SAGE II Based Column Surface Area Measurements of the Mt. 52
Pinatubo Aerosol Over the Eastern United States John Anderson and
V.K. Saxena
Session 8: Measurement Methods
James Mulik, Chair
A Study of Interferences in Ozone UV and Chemiluminescence 53
Monitors E.E. Hudgens, T.E. Kleindienst, F.F. McElroy, et al.
Real Time Electrochemical Measurement of Ozone in the Presence of 54
Nitrogen Oxides William R. Penrose, Li Pan, and Will M. Ollison
-------
The Development of an Active Personal Ozone Sampler Using a 54
Diffusion Denuder A.S. Geyh, J.M. Wolfson, P. Koutrakis, et al.
Evaluation of Passive Samplers for Field Measurements of Ambient 55
Ozone in the National Parks John D. Ray and Miguel Flares
Development of a New Semi-Volatile Organic Compound Sampler C. 56
Sioutas, P. Koutrakis, andRM. Burton
Outdoor Air NOX Speciation by a Selective Denuder Collection System 56
Robert S. Bratnan and M. Stacey Thomson
Field Evaluation of a New Glass Honeycomb Denuder/Filter Pack 57
Sampler P. Koutrakis, C. Sioutas, J.M. Wolfson, et al.
The Use of PM10 Anion-Cation Difference as an Index of Historical 58
Aerosol Acidity George D. Thurston, J. Currie, D.He.et al.
Passive Samplers for Ambient Ozone, Formaldehyde and Sulfur 58
Dioxide: Indoor, Outdoor and Personal Exposure Applications Daniel
Grosjean and Eric Grosjean
Session 9: NC O3 State Implementation Plan,
Measuring and Modeling Study
Viney Aneja, Chair
Overview of the North Carolina UAM Project Brock M. Nicholson 60
The Sensitivity of Meteorological and Emissions Uncertainties on 60
Urban Airshed Model Ozone Concentration Results in North Carolina
Brian S. Timin, Janice Godfrey, and Ji-Cheng Carey Jang
Updated Land Use/Land Cover Data - Its Effects on Estimates of 61
Biogenic Emissions in the North Carolina Urban Airshed Modeling
Effort William W. Cure
Enhancements to the Emissions Inventory Inputs for the North Carolina 61
UAM Project Sheila C. Holman
Mobile Emission Calculations for the North Carolina UAM Project 62
BehshadM. Norowzi
Use of Link-Base Data to Add Definition to Highway Mobile 62
Emissions for the UAM Anne S. Galamb
Quality Assurance of the North Carolina Precursor of Ozone 63
Inventories, Emissions Preprocessor System and the Urban Airshed
Model Output Laura Boothe and Victoria Chandler
An Environmentalist's Perspective on Ozone Measurement: The Role 63
of Scientific and "Citizen-Scientific" Methods Brian J. Morton
-------
Auburn Tower Ozone Study George C. Murray, Jr., Thomas ^
L. Manuszak, Robert S. Graves, et al.
Vertical Distributions of Carbonyls for Ozone Control in Urban North 65
Carolina Viney P. Aneja, Jay H. Lawrirnore, Mita Das, et al.
Session 10: Quality Assurance
Chri Kulkarni, Chair
A Statistical Analysis of Audit Results from Stationary Source 66
Compliance Tests Ellen W. Streib, Jack C. Suggs, and William
J. Mitchell
EPA's QA Program on the Suppliers of Protocol Gases Avis P. Hines 66
and William J. Mitchell
Preparation and Evaluation Of Representative Compounds In Small 67
High Pressure Cylinders For Use As Audit Materials William
Mitchell, Jack Suggs, Howard Crist, et al.
Data Handling Issues and Techniques Associated with Data Collected 67
from Automated GC Systems Used for Ozone Precursor Analysis
Larry D. Ogle, Margaret A. Underwood, Pamela R. Chen, et al.
A Computer Controlled Dynamic Dilution System for Improved 68
Accuracy and QA/QC in TOW Standard Preparation D.B. Cardin and
J.T. Deschenes
Importance of Minimum Detection Limits in Air Toxics Data Bases N. 69
Adams
Stability Evaluation of Multicomponent EPA Protocol Gases Richard 70
C. Shores, Michael J. Messner, Robert W. Murdoch, et al.
Session 11: FTIR Studies
George Russwurm, Chair
Open-Path FTTR Air Quality Measurements at a Petrochemical 71
Complex in Brazil Robert H. Kagann, Neuza Neves, and Felipe Villas
Boas
Open Path FTLR: Tomographic and I(o) Strategies and Results S.P 71
Levine, D.Y. Park, S. Giese-Bogdan, et al.
Development of Quality Assurance Procedures in Open-Path FT-IR 72
Monitoring Edgar L. Thompson, Jr., Jeffrey W. Childers, and George
M. Russwurm
Adaptation of a Military FTS to Civilian Air Toxics Measurements 72
James R. Engel and Rick Dorval
-------
Joint Observations of the ETL Ozone Lidar and MPS System During 73
the Los Angeles Free Radical Study Yanzeng Zhao, John Gaynor, and
Daniel Wolfe
Fourier Transform Microwave Spectroscopy a Potential New 73
Analytical Tool for Gas Phase Species Richard Suenram, Frank
Lovas, and Robert Sams
Monitoring Air Pollutants by Molecular Beam Microwave 74
Spectroscopy U. Andresen, U. Kretschmer, C. Thomsen, et al.
Effect of Temperature on the Ability to Collect Data: MDA Scientific 75
Open-path Fourier Transform Infrared Spectrometer with Stirling
Cooler Judith O. Zwicker and William M. Vaughan
Photolysis Assisted Pollution Analysis (PAPA) Philip L. Hanst 75
FTTR Transmission Spectroscopy for Quantitation of Ammonium 76
Bisulfate in Fine Paniculate Matter Collected on Teflon Filters Kenneth
H. Krost and William A. McClenny
United States Environmental Protection Agency Fourier Transform 77
Infrared Spectroscopy Test Program for Emissions Measurement Lori
Lay
Session 12: New Methods for VOCs
William McClenny, Chair
The Concentration and Measurement of Air Pollutants by GC/MS; A 78
Comparison of Sorbent Versus Cryo Trapping Elizabeth Almost and
Norman Kirshen
A Real Time Sorbent-Based Continuous Monitoring System Frank 78
G. Lattin and Donald G. Paul
The Perkin-Elmer ATD-400 System for Monitoring of Ambient VOC 79
Ozone Precursors Paul Radenheimer, John Gibich, and Larry Ogle
System Operation: Continuous Volatile Organic Compound Air 80
Monitoring of 56 Ozone Precursors with the Perkin-Elmer 8700 GC
and Automatic Thermal Desorption System Paul Radenheimer, John
Gibich, and Larry Ogle
A System for the Determination of Trace-Level Polar and Non-Polar 81
Toxic Organic Compounds in Ambient Air A. Tipler, R. Dang, and
H. Hoberecht
Water Management in Capillary Gas Chromatographic Air Monitoring 81
Systems A. Tipler
-------
oo
Identification of Ambient Air Sampling and Analysis Methods for the
189 Title HI Air Toxics R. Mukund, Sydney M. Gordon, and Thomas
J. Kelly
R9
Direct Trace Analysis of Volatile Organic Compounds in Air using °^
Filtered Noise Field Ion Trap Mass Spectrometry Sydney M. Gordon,
Patricks. Callahan, and Donald V. Kenny
Session 13: Applying Total Human Exposure
Methodologies to Address Environmental
Health Issues Along the U.S.-Mexico Border
Gerald Akland and Timothy Buckley, Cochairs
Applying Human Exposure Science to Public Health Concerns: The 84
Lower Rio Grande Valley Environmental Monitoring Study Evolution
and Design G. Akland, O. Ramirez, and M. Schwab
Balancing Participant Burden with Data Needs: The Human Aspect of 85
Exposure Studies Mar go Schwab and Gerry Akland
Pollutant Concentrations in Air, Soil and Housedust Samples in the 86
Lower Rio Grande Valley Environmental Study 7?.^. Stevens,
G. Akland, S. Mukerjee, et al.
Methods for Semivolatile Organics in Air and Housedust Used in the 87
Spring 1993 Lower Rio Grande Valley Environmental Scoping Study:
Performance and Data Comparison R.G. Lewis, S. Mukerjee, R.K.
Stevens, et al.
Residential-Based Human Exposure to Metals, Pesticides, PAH, and 88
VOCs: Results from the Lower Rio Grande Valley Environmental
Pilot Study T.J. Buckley, R.G. Lewis, G.G. Akland, et al.
Session 14: Particles Studies
Petros Koutrakis, Chair
The Role of Size-Dependent Dry Deposition of Sulfate Aerosol in a 89
Three-Dimensional Eulerian Air Quality Model Francis S. Binkowski
and Uma Shankar
Size Distributions Parameters and Hygroscopic Growth of Aerosol 89
Particles Bearing V J.M. Ondov, F. Divita, and T.L. Quinn
Measurement and Speciation of Gas and Paniculate Phase Organic 90
Acids in an Urban Environment Joy Lawrence and Petros Koutrakis
Results of Philadelphia Diesel Paniculate Monitoring Study 9 j
October - November 1993 Breda Phillips, Tom Lumpkin, and Mike
Pleasant
-------
Diurnal and Elevational Variations in Ozone and Aerosols in New 91
Hampshire's Class-I Airsheds L. Bruce Hill and George A. Allen
Temporal and Spatial Characteristics of Particle Mass in Metropolitan 92
Philadelphia GA. Allen, P. Koutralds, HM. Suh, et at
Session 15: Stainless Steel Canister Sampling and Analysis
R.K.M. Jayanty, Chair
SUMMA Canisters - Do They Need to be Cleaned for TO-14 93
Analysis? Rei A. Rasmussen
Beyond Canister Cleaning...What About the Surface Chemistry? R Jt. 93
Freeman, C.C. Crume, and ED. Winegar
Development and Validation of a Heated Canister-Based Source 94
Sampling Method Robert J. Crawford and David L. Elam
Recovery of Oxygenated Organics from SUMMA Canister Rei 95
A. Rasmussen
Analysis of Selected Polar Volatile Organic Compounds via TO-14 and 95
Modified TO-14 Method H.Wang
Why is it so Difficult to Measure Terpenes in Ambient Air Rei 95
A. Rasmussen
Certification of Canister Samplers for VOC Measurements at the Waste 96
Isolation Pilot Plant L. Frank-Supka, C.F. Wu, G.C. Meiners, et al.
A Fully Automated SUMMA Canister Cleaning System for Method 97
TOW DB.Cardin
Networking an Entire TO 14 Laboratory into a Single Windows Based 97
Control System D.B. Cardin, J.T. Deschenes, and EA. Galoustian
Advanced Surface Treatment and Cleaning Techniques for the U.S. 98
EPA Method TO-14 Grab Sampling Containers Joseph Krasnec
Session 16: Ambient Air Measurements of VOCs
Jimmie Hodgeson, Chair
Determination of Volatile Organic Compounds in Ambient Air with 99
Gas Chromatograph-Flame lonization and Ion Trap Detection Shili
Liu, Robert J. Carley, Jianshi Kang, et al.
Exposure to Evaporative Gasoline Emissions Clifford P. Weisel and 99
Krishnan R. Mohan
-------
Hydrocarbons in the C8-C20 Range Measured During COAST Study 10°
in Texas B. Zielinska, J. Sagebiel, G. Harshfield, et al.
The Determination of Hazardous Air Pollutants with a Built-in ^i
Preconcentrator and Capillary GC Norman Kirshen and Elizabeth
Almasi
An Integrated Approach to Parts-Per-Trillion Measurement of Volatile 101
Organic Compounds in Air Eric D. Wine gar, RJR. Freeman, and C.C.
Crume
VOC Quality Control Measurements in the CASTNet Air Toxics 102
Monitoring Program (CATMP) Michael G. Winslow, Matthew
M. Booth, and Dwight F. Roberts
An Automated GC Technique for Enhanced Detection of Organic 103
Compounds Monitored Using Passive Organic Vapor Badges Kochy
Fung
Background Monitoring of Air Toxics at the Waste Isolation Pilot Plant 103
L. Frank-Supka, C.F. Wu, R.H. Lopez, et al.
Experience in the Operation of a PAMS Type 2 Ihab H. Farag, 104
Chunming Qi, Paul Sanborn, et al.
Comparison Studies of Ozone Precursors in Phoenix, Arizona Carmo 104
Fernandez, Jim Guyton, Cheng Peter Lee, et al.
Implementation of a Risk-Based Air Monitoring Program Using 105
Integrated and Continuous Air Monitors Wen-Whai Li, Stephen
T. Washburn, and Mary E. Greenhalgh
A Two-Channel, 16-Position Canister Field Sampler for Improved 105
Performance and Quality Assurance D.B. Cardin and J.T. Deschenes
Session 17: Air Pollutants in General
Petros Koutrakis, Chair
Temporal Variation of Fine Particle Mass at Two Sites in Mexico City 107
Paulina Serrano, George Allen, Margarita Castillejos, et al.
Ability of Fixed Monitoring Stations to Represent Commuter's 108
Exposure to CO Revisited: The Case of Mexico City Adrian
Ferndndez-Bremauntz
Investigation of Spatial and Temporal Pattern of Ozone Concentration 108
within a Metropolitan Area Using Ozone Passive Sampler L.-J. Sally
Liu, Petros Koutrakis, and Irvine Broder
-------
A Comparison of Ozone and Acid Aerosol Exposure Patterns in a 109
Summertime Study of Metropolitan Philadelphia Jed M. Waldman,
Chris S-K Liang, Petros Koutrakis, et al.
Diurnal and Spatial Variation in Fine and Coarse Particle 110
Concentrations in Metropolitan Philadelphia H.H. Suh, GA. Allen,
P. Koutrakis, et al.
Indoor Air Chemistry: Formation of Organic Acids and Aldehydes 111
Junfeng Zhang, Paul J. Lioy, and William E. Wilson
Ozone Reactive Chemistry in Indoor Residential Environments 111
Richard Reiss, P. Barry Ryan, Petros Koutrakis, et al.
CASTNet Air Toxics Monitoring Program (CATMP): VOCand 112
Carbonyl Data for July, 1993 through March, 1994 David P. Harlos
and Eric S. Edgerton
Sources and Factors Influencing Personal and Indoor Exposures to 112
PAHs and PHTHALATES Haluk Ozkaynak, Jianping Xue, and John
D. Spengler
Session 18: Indoor Air Pollution
Edo Pellizan, Chair
Gas and Paniculate Phase Acids, Organic Compounds and Oxidants in 114
a Sick Room Kris Wardrup, Laura Lewis, and Delbert J. Eatough
Tracer Gas Measurement of Indoor-Outdoor Air Exchange Rates Kevin 115
N. Gunn, Zhishi Guo, and Bruce A. Tichenor
Measurement of Airborne Particle Counts and Mass in a Healthy 115
Building During a One-Year Cleaning Effectiveness Study K.E.
Leese, B.C. Cole, R.M. Hall, et al.
Effects of Activated Charcoal Filtration and Ozonation on Hydrocarbon 116
and Aldehyde Levels of Ambient Air Used in Controlled-Exposure
Chamber Studies of Air Pollutant Human Health Effects Beverly
E. Tilton, Joseph J. Bufalini, Sarah AMeeks, et al.
The U.S. EPA/ORD Large Building Indoor Air Quality Study - Results 116
of the Initial Building Characterization Roy Fortmann, Russ Clayton,
V. Ross Highsmith, et al.
Large Building Characterization Marc Y. Menetrez, David C. Sanchez, 117
Bobby Pyle, et al.
A Comparison of Sorbent Sample Cartridges for the Collection and 118
Analysis of Volatile Organic Compounds Collected in Large Office
Buildings Jeffrey T. Keever and Linda Sheldon
-------
118
Managing Residential Sources of Indoor Air Pollution Bruce
A. Tichenor and Leslie E. Sparks
Relationship Among Drag Sled, PUF Roller, and Hand Press Transfer 119
of Pesticide Residue from Floors David E. Camann, H. Jac Harding,
Paul W. Geno, et al.
Comparison of PM2 5 and Open-Face Inlets for Sampling Aerosolized 120
Pesticides of Filtered Polyurethane Foam David E. Camann, H. Jac
Harding, Charles L. Stone, et al.
Session 19: Sources and Fate of Atmospheric VOCs
Joseph Pinto, Chair
Methods to Determine the Biogenic Contributions to Ambient 121
Concentrations of Volatile Organic Compounds R.K. Stevens, C.W.
Lewis, WA. Lonneman, et al.
Radiocarbon Measurements Of Wintertime Atmospheric Carbon 122
Monoxide in Albuquerque, NM: Contributions Of Residential Wood
Combustion George A. Klouda and Michael V. Connolly
Progress Towards Validating the Separation Of Atmospheric Volatile 123
Organic Carbon from Air for 14C Measurements George A. Klouda,
George C. Rhoderick, Robert L. Sams, et al.
Comparison of Ambient Ratios of NMHCs and CO to NOX with 124
Emission Inventory Values for Atlanta Jf. Pinto andM. Somerville
Atmospheric Chemistry of Unsaturated Oxygenated: Alcohols, 124
Aldehydes, Ketones and Esters Daniel Grosjean and Eric Grosjean
Atmospheric Chemistry and Fate of C2-C5 Peroxyacyl Nitrates Daniel 125
Grosjean, Eric Grosjean, and Edwin L. Williams HI
Session 20: Russian Air Pollution Studies
Emma Bezuglaya, Chair
Examination of Atmospheric Diffusion Calculating Schemes Under 127
Experimental Data Serge A. Gromov and Veronica A.Ginzburg
Background Pollution of the Atmosphere: The Multi-Year Observation 127
in Russia F.Rovinsky
Methods of Studying Sources in the Boundary Atmospheric Layer of 128
the Background Areas V. Egorov
Applicability of Trajectory Analysis for Air Background Monitoring J29
Network Optimization Serge G. Paramonov
-------
Long-Range Model for Atmospheric Pollution Analysis of Background 129
Territories Serge A. Gromov
Session 21: Environmental Tobacco Smoke
Delbert Eatough, Chair
Real-Time Monitoring of Polycyclic Aromatic Hydrocarbons (PAH) 131
and Respirable Suspended Particles (RSP) from Environmental
Tobacco Smoke (ETS) in a Home Wayne On, Nancy Wilson, Neil
Klepeis, et al.
Measurement of Environmental Tobacco Smoke Frank E. Jones 132
A Comparison of Smoking and Non-Smoking Areas: Private Homes 132
and Bingo Halls Ronald W. Bell, Robert E. Chapman, Brian
D. Kruschel, et al.
Session 22: Analysis of Polar Volatile Organic Compounds
Joachim Pleil, Chair
GC-MS Analysis of the Exhales Breath Matrix Joachim D. Pleil and 133
Andrew B. Lindstrom
Deactivating SUMMA Canisters for Collection and Analysis of Polar 133
Volatile Organic Compounds in Air David Shelow, Paul Silvis,
Andrew Schuyler, et al.
Cleanliness of Common Air Sampling Sorbents for Application to 134
Phenolic Compounds Measurement Using Supercritical Fluid
Extraction James R. Bowyer and Joachim D. Pleil
Collection of Ambient Air Phenols Using an Anion Exchange 134
Membrane Marcia Nishioka, Hazel Burkholder, Scott Reynolds, et al.
Controlled Desorption Trap (CDT): A Water Management Technique 135
for Quantitative Analysis of Polar VOCs in Ambient Air Sharon Reiss
and Dick Jesser
A Comparison of Concentration Techniques for the Analysis of Polar 136
Compounds in Canister Samples D.B. Cardin and J.T. Deschenes
Session 23: SemiVolatile Organic Compounds
Nancy Wilson, Chair
Method Validation Program for the Long Duration Sampling of 137
PCDDs/PCDFs in Ambient Air Bruce E. Maisel, Gary T. Hunt,
Marilyn P. Hoyt, et al.
-------
1 TO
Effect of Combustion Temperature on the Atmospheric Stability of
Polychlorinated Dibenzo-p-Dioxin and Dibenzofurans David
M. Pennise and Richard M. Kamens
Modeling the Mass Transfer of Semi-Volatile Organics in Combustion 138
Aerosols Jay R. Odum and Richard M. Kamens
Experimental Observations of Non-Equilibrium Gas-Particle
Partitioning of PAHs in an Outdoor Smog Chamber Dana L. Coe and
Richard M. Kamens
An Analytical Method for Determining Low Level Airborne Exposure 140
Levels to Lewisite Frank G. Lattin and Donald G. Paul
Use of the Phenanthrene to Benzo(e)pyrene Ambient Air Ratio as an 141
Indicator for the Source of Polycyclic Aromatic Hydrocarbons Andre
Germain, SoniaRinguette, and Jean Tremblay
Polycyclic Aromatic Hydrocarbons in House Dust and Track-In Soil in 141
an Eight-Home Study Jane C. Chuang, Patrick J. Callahan, Vanessa
Katona, et al.
Session 24: General Papers
W.F. Gutknecht, Chair
Future Research Directions for the Great Waters Program Melissa 143
W. McCullough
Scientific Findings and Regulatory Recommendations of the 1993 143
Great Waters Report to Congress Amy Vasu
Detecting Ambient Oxides of Nitrogen Using Laser-Induced 144
Photofragmentation/Photoionization Spectromerry J.B. Simeonsson,
G.W. Lemire, andR.C. Sausa
Measurement of the Effects of Moisture Distribution on the Transport 144
Properties of Radon and Other Soil Contaminants in EPA's Soil
Chamber Ronald B. Mosley
Comparison of Soil Permeability Measurements Using Probes of 145
Different Sizes and Geometries Ronald B. Mosley
Test Methods for Evaluating Reformulated Fuels Michael 145
C. Croudace
Peculiarity of Toxic Metals Emission Measurements at Wastewater 146
Treatment Plants Vladimir Kogan and Edward Torres
An Assessment of Low Emission Sewer Systems for Industry Reese 147
H. Howie, Charles J. Zukor, and Parag Biria
-------
An Odor Control Study at Bissell Point Wastewater Treatment Plant 148
Jon F. Bergenthal, Robert T. Jorgen, and John R. Gibbons
Poster Session
Air Pollution Monitoring in the Republic of Latvia I. Lyulko and 149
R. Dubrovskaya
Evaluation of the Effects of Humidity on the Transfer of C2-C10 150
Hydrocarbons from Cylinders Ron Bousquet and Ron Brande
Comparison of Break-Thru Volume on Adsorbent vs. Cryogenic Trap 150
Wendy Bollard
Evaluation of Modifications to the Tekmar™ 5010 for the Analysis of 151
Indoor Air Pollutants on Multisorbent Tubes John W. Duncan
Evaluation of Storage Conditions for Indoor Air Pollutants on Solid 151
Sorbents in UHP Helium Purged Mylar™ Packs John W. Duncan and
Frederic J. Mixson
Environmental and Occupational Exposures to PAH in the Czech 152
Republic: Personal Exposure Monitoring Coupled with HPLC/Time-
Programmed Fluorescence Detection R. Williams, K. Hattaway,
R. Watts, et al.
Evaluation of the Transfer Efficiency of High Molecular Weight 153
Hydrocarbons Using Various Types of Regulators Ron Bousquet
Simplified Preparation of TO 14 and Title HI Air Toxic Standards 153
Using a Windows Software Package Supporting Static and Dynamic
Dilution Schemes DJ3. Cardin andEA. Galoustian
Results from the August 1993 DOAS Evaluation in Baytown, Texas 154
Charles P. Conner, Lee Ann B. Byrd, Frank F. McElroy, et al.
Effects of Using Nation® Dryer on Hydrocarbons Analyzed from 155
Canisters by Modified EPA Method TO-14 J.C. Sagebiel and
B. Zielinska
Statistical Distributions of Airborne PCB and Pesticide Concentrations 155
Measured at Regional Sites on the Great Lakes Donald F. Gatz, Clyde
W. Sweet, Ilora Basu, et al.
Stability of Reduced Sulphur Compounds in Whole Air Samplers 156
Quang Tran and You-Zhi Tang
A New Vapor and Gas Test Atmosphere Generator with Broad 157
Concentration and How Output Capabilities J.K. McGee, PA.
Evansky, D. Terrell, et al.
-------
1 ^8
Air Monitoring at Alert in the High Arctic: Results of One Year of
Monitoring of Organochlorine Compounds and PAH D. loom,
L. Barrie, P. Fellin, et al.
Laboratory Preparation of DNPH Derivatives of Carbonyl Compounds 158
in SEP-PAK® Cartridges for Quality Assurance Purposes Rita
M. Harrell
Experimental Studies of an Ethanol-Air Flow Subjected to UV Light 159
Timothy A. Spaeder
Determination of Test Methods for Interior Architectural Coatings 160
Donald A. Whitaker, Linda S. Sheldon, Jeffrey T. Keever, et al.
An Improved DNPH Cartridge Aldehyde Sampler for 3-Day 160
Unattended Sampling D.B. Cardin and E.A. Galoustian
-------
Tuesday, May 3,1994
Session 1: Predicting VOC Emission Rates, Air
Quality Models, Deposition
Tuesday Morning
MODEL PREDICTED VOC EMISSION RATES BASED ON MEASURED
CONCENTRATIONS AND METEOROLOGICAL DATA
Ray E. Carter, Jr., Dennis D. Lane, and Glen A. Marotz
Department of Civil Engineering
4002 Learned Hall
University of Kansas
Lawrence, KS 66045
Michael F. Davis
U.S. EPA, Region VII
25 Funston Road
Kansas City, KS 66115
Charles T. Chaffin, TimL. Marshall, Robert M. Hammaker, and
William G. Fateley
Department of Chemistry
Kansas State University
Willard Hall
Manhattan, KS 66506
Estimated single-point-source VOC emission rates were produced from measured
path-integrated VOC concentrations and meteorological data using a form of the
Gaussian dispersion equation. VOC concentration data were collected at a flat, grass-
covered site in Lawrence, KS, using whole-air canister and FTIR methods.
Measurements were made at four downwind distances (50,100,150, and 225 meters) and
under a variety of atmospheric stability regimes (Classes A, B, C, and D). Estimated
VOC emission rates were compared to accurately measured rates. The estimation
methods use meteorological data to in an attempt to overcome problems associated with
direct application of the Gaussian equation: Model 1 uses means of wind speed and
direction for an entire test period; Model 2 uses one-minute means of those variables
summed over the test period. The direct application of the Gaussian equation provides
less accuracy as conditions become more unstable. Model 1 provides an increase in mean
estimation accuracy under all stability conditions, but especially under unstable
conditions. Model 2 provides a significant additional increase in accuracy, the difference
being more pronounced under unstable conditions. Estimation accuracy was better at 100
and 150 meters than at the other distances. Model 2 provides a significant increase in
accuracy at the three longer distances. Based on data from only those tests during which
Gaussian assumptions were violated, Model 1 provides a significant increase in
estimation accuracy; Model 2 provides a significant additional increase. Thus, emission
-------
rate models described in this paper allow the integrated Gaussian equation to be applied
more accurately in cases where underlying assumptions are violated.
APPROXIMATION OF THE MAXIMUM IMPACT ON AIR QUALITY FROM
AREA SOURCE EMISSIONS
Norman A. Huey
U.S. EPA, Region VIII
Denver, Colorado
Joseph F. Froechtenigt
AARP-Senior Environmental Employment Program, Region VIII
Denver, Colorado
Atmospheric dispersion models for the estimation of ambient air quality impacts as a
function of radial distance and azimuthal direction require representative meteorological
data. Maximum radial concentrations (direction independent) are somewhat insensitive to
meteorological data. To simplify the dispersion modeling process, screening models
which do not require representative meteorological data have been developed to estimate
maximum radial concentrations. No such screening model is available for the estimation
of maximum annual concentrations resulting from area source emissions.
The cost of representative meteorological data and the cost of dispersion modeling
cannot be justified for many of the activities at Superfund sites. Many Superfund air
pathway analyses can be satisfied with direction independent maximum concentrations.
Examination of modeling (ISCST2) results which utilized 27 different
meteorological data sets for various sizes of area sources suggested that an empirical
equation could be derived from the data. This empirically derived equation could then be
used to approximate maximum annual concentrations resulting from area source
emissions. Development and testing of that equation is the subject matter of this paper.
AIRCRAFT MEASUREMENTS OF THE TRANSFER VELOCITIES OF GASES
AND AEROSOLS ACROSS LAKE MICHIGAN SURFACE DURING THE
SUMMER
A.J. Alkezweeny andJ.L. Stith
Department of Atmospheric Sciences
University of North Dakota
Grand Forks, ND 58202
On June 18,1992 our instrumented aircraft made a constant altitude flight over lake
Michigan near the Chicago shoreline and about 50 km downwind of it. We use the eddy
correlation method to calculate the fluxes of C02,03, water vapor and aerosols in the
diameter range of 0.1 to 3.0 (im. The fluxes near the shoreline were found to be
significantly higher than those in the middle of the lake. The turbulence intensities, as
measured by E1/3, were 2.92 ± 0.75 cm^V1 and 1.34 ± 0.39 ernes'1 near the shoreline
and mid-lake respectively. We feel that fluxes we measured near the shoreline were
-------
representative of those at the surface of the lake, because of the strong turbulence we
experienced during our measurement. However, we cannot assume the same for the
fluxes measured at mid-lake. The fluxes near the shoreline for O3 and aerosols were
directed toward the surface and corresponding to transfer (deposition) velocities of
0.15 cm s"1 and 0.86 cm s"1, respectively. For CO2, and water vapor, the fluxes were
directed upward and corresponding to transfer velocities of 0.04 cm s"1 and 0.54 cm s"1,
respectively.
Our west to east constant altitude flight, starting from Chicago, over the lake showed
that turbulence, as measured by e1/3, decayed slowly along the flight track. The O3
concentration steadily increased from 39 ppb to about 52 ppb as the air moved away from
the shoreline.
The entire flight was made between about 1:30 and 3:30 pm. The wind was westerly
at about 12 m s"1.
AEROSOL DISPERSION SIMULATION USING OMEGA/ATDM
T.J. Dunn, D.P. Bacon, Z. Boybeyi, Y-L. Ho, MD. McCorcle, SE.
Peckham, andRA. Sarma
Science Applications International Corporation
1710 Goodridge Drive
McLean, V A 22102
/. Zack
MESO Inc.
Rensselaer Tech Park
185 Jordan Road
Troy, NY 12180
The Operational Multiscale Environment model with Grid Adaptability (OMEGA) is
a new atmospheric simulation system that merges state-of-the-art computational fluid
dynamics techniques with a comprehensive non-hydrostatic equation set. OMEGA is
based on an unstructured triangular prism grid that permits a horizontal grid resolution
ranging from 100 km down to 1 km and a vertical resolution from a few tens of meters in
the boundary layer to 1 km in the free troposphere. OMEGA contains an embedded
Aerosol Transport and Diffusion Model (ATDM) that permits the high resolution
simulation of the advection and diffusion of either grid-based aerosols or of Lagrangian
parcels.
OMEGA is naturally scale spanning and its unstructured grid permits the addition or
subtraction of grid elements at any point in space and time. OMEGA can readily adapt
its grid to fixed surface or terrain features, land/water boundaries, or dynamic features in
the evolving weather. The implications for the ATDM are that enhanced grid resolution
can be provided in localized regions such as emission sources or in the vicinity of a dust,
smoke, or chemical cloud.
In this paper we present the atmospheric simulation and aerosol dispersion
capabilities of OMEGA/ATDM.
-------
This work was supported by the United States Defense Nuclear Agency under
Contract DNA001-92-C-0076.
-------
Tuesday, May 3,1994
Session 2: VOC Portable Instruments/Methods
Tuesday Morning
NEW PORTABLE MICRO GAS CHROMATOGRAPH FOR ENVIRONMENTAL
ANALYSIS
E.B. Overton, K.R. Carney, Hf. Dharmasena,AM. Mainga, and
U. Ehrmann
Institute for Environmental Studies
42 Atkinson Hall
Louisiana State University
Baton Rouge, Louisiana 70803-5705
Efforts directed at developing a truly portable method for the analysis of semivolatile
compounds have led to the construction and testing of a new generation of micro-GC
instrumentation. Building on the successful application of microbore GC columns for
in-field analysis of volatile organic compounds, the instrument development group at the
LSU-Institute for Environmental Studies has developed a hand portable GC capable of
analyzing samples containing compounds with retention indices (100%
dimethylpolysiloxane column) up to at least 2000 in less than 4 minutes, using less than
50 watts (peak) of electrical power. In addition to the ability to analyze semivolatile
compounds, the chromatograph is capable of analyzing volatile organics competitively
with the most sophisticated of die current commercial portable GCs. The presentation
will evaluate chromatographic performance of the instrument at it latest stage of
development and demonstrate some applications to environmental analysis using the
prototype instrument.
TRIPLE SORBENT THERMAL DESORPTION/GAS
CHROMATOGRAPHY/MASS SPECTROMETRY DETERMINATION OF
VAPOR PHASE ORGANIC CONTAMINANTS*
C.Y. Ma, J.T. Skeen, A£. Dindal, andRA. Jenkins
Chemical and Analytical Sciences Division
Oak Ridge National Laboratory*
Oak Ridge, Tennessee, 37831-6120
A thermal desorption/gas chromatography/mass spectrometry (TD/GC/MS) method
has been evaluated for the determination of volatile organic compounds (VOCs) in vapor
phase samples using Carbosieve S-IH/Carbotrap/Carbotrap C triple sorbent traps (TST),
similar to those available from a commercial source. The analysis was carried out with a
Hewlett-Packard 5985A or 5995 GC/MS system with a modified injector to adopt an
in-house manufactured short path desorber for transferring desorbate directly onto a
cryofocusing loop for subsequent GC/MS analysis. Vapor phase standards generated
-------
from twenty target compounds were used for method validation, including alkanes, alkyl
alcohols, alkyl ketones, and alkyl nitriles, a group of representative compounds that have
previously been identified in the target airborne matrix. The method was validated based
on the satisfactory results in terms of repeatability, recovery rate, stability, and linearity.
A relative standard deviation of 3.1 to 17.4% was obtained for the generation of gas
phase standards and 2.1 to 17.6% for spiking the standards on and desorbing from TSTs.
Linear correlation coefficients for the calibration curves as determined ranged from 0.88
to 0.99 for TSTs spiked with 20 to 500 ng of gas standards. For a majority of target
compounds, recoveries of greater than 60% were observed. For four selected target
compounds spiked on TSTs, minimal loss (10 to 22%) was observed after storing the
spiked traps in a 4° C refrigerator for 30 days. The only chromatographable artifact
observed was a 5% conversion of isopropanol to acetone. The validated method has been
successfully applied to the determination of VOCs collected from various emission
sources in a diversified concentration range.
*Research sponsored by the U.S. Department of Energy, Office of Technology
Development, under contract DE-AC05-84OR21400 with Martin Marietta Energy
Systems, Inc.
The submitted manuscript has been authored by a contractor of the U.S.
Government under contract No. DE-AC05-84OR21400. Accordingly, the U.S.
Government retains a nonexclusive, royalty-free license to publish or reproduce the
published form of this contribution, or allow others to do so, for U.S. Government
purposes.
REMOTE MONITORING OF SUB PPB LEVELS OF VINYL CHLORIDE,
DICHLOROETHYLENE AND TRICHLOROETHYLENE VIA MODEM
OPERATED AUTOMATED GC
Amos Linenberg and Neil J. Lander
Sentex Systems Inc.
553 Broad Avenue
Ridgefield, New Jersey 07657
The need for remote monitoring of certain compounds led to the development of a
portable gas chromatograph remotely operating through a modem. The compounds being
monitored were Vinyl Chloride, Cis 1,2 Dichlorethylene and Trichloroethylene. The
requirements of the project were to detect these compounds at the ppb level in air.
The GC was controlled by a lap-top computer. The operation of the instrument was
as follows:
(1) An internal pump pulled the air in and onto a preconcentration device.
(2) The VOCs were then thermally desorbed from the preconcentrator.
(3) Carrier gas swept the VOCs on to the column for separation and detection.
(4) All data was stored automatically to the hard drive of the computer.
-------
The operating parameters were as follows:
Sample Time: 250 seconds, Columns: 4 foot 1% SP1000 on Carbopak B, Oven
Temperature: 90 degrees C, Carrier Gas Flow: 20 cc/mn, Detector: Argon
lonization Detector, Preconcentrator was packed with 60/80 Carbosieve G.
The instrument was calibrated periodically with a certified mixture supplied by Scott
Specialty Gases. The actual calibration concentration of the analytes in the calibration
mixture was 1.15 pb VC, 1.10 ppb DCE and 1.09 ppb TCE.
The remote instrument was located in the midwest portion of the U.S. It was
periodically contacted by the main computer, located in northern New Jersey, for review
and correction of any unexpected variation in instrument responses. The data was
uploaded weekly from the remote instrument to the main computer in New Jersey and
then reformatted for distribution to the client
FIELD PORTABLE ENVIRONMENTAL SAMPLE CONCENTRATOR FOR
VOCS IN AIR
Mark W. Burns and Kent Hammarstrand
MTI Analytical Instruments, Microsensor Technology, Inc.
41762 Christy Street
Fremont, CA 94538
Field screening for low level of volatile organic compounds (VOCs) will become
increasingly valuable as enactment of environmental regulations continues worldwide.
These regulations press the need for rapid, rugged, reliable, high sensitivity
instrumentation which can be utilized in the field for screening and monitoring purposes.
A low cost, microprocessor-controlled, high throughput, portable sample
concentrator has been developed for on-site concentration of VOC contamination in air.
It operates with a rechargeable 12V battery and houses a refillable internal supply of high
purity purge gas providing 6 hours of field operation. Air samples are drawn through and
concentrated on a three bed carbon-based adsorbent trap at ambient temperatures using an
internal pump. The trap is then rapidly heated and the concentrated sample backflushed
with He into a syringe. The contents of the syringe can be introduced to a micro gas
chromatograph (or any standard, portable GC) for rapid on-site field analysis.
Concentration enhancements of 750-fold are achievable. Automated pre- and post-desorb
purge events minimize sample carryover from run to run and improve the accuracy of
analysis.
Data will be presented for the EPA Method TO-14 compounds and will highlight
recovery efficiency, concentration enhancement, analytical reproducibility, sample
carryover, and real world applications.
-------
994
Session 3: Acid Aerosols
Tuesday Morning
SPATIAL VARIATION IN ACIDIC SULFATE CONCENTRATIONS WITHIN
METROPOLITAN PHILADELPHIA
R.M. Burton
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
H.H. Suh, GA. Allen, and P. Koutrakis
Harvard University
School of Public Health
665 Huntington Avenue
Boston, MA 02115
Acidic sulfate concentrations were measured in metropolitan Philadelphia during the
summer of 1992, as part of a continuing effort to characterize acid aerosol and gas
concentrations in urban environments. Sampling was performed simultaneously at six
sites located within metropolitan Philadelphia, a site located in Camden, New Jersey, and
at a rural site approximately 18 miles from the city center. Sites were selected based on
their population density and on their relative locations within Philadelphia Acid aerosol
and gas sampling was performed on alternate days, with all samples collected for 12-hour
periods beginning at SAM. All samples were collected using modified Harvard-EPA
Annular Denuder Systems (HEADS).
In this paper, we examine the spatial variation in acidic sulfate and ammonia
concentrations within metropolitan Philadelphia, We also examine factors that may
influence their variation. Outdoor sulfate concentrations were found to be uniform across
metropolitan Philadelphia; however, considerable spatial variation in both aerosol strong
acidity (H+) and ammonia (NH3) concentrations was observed. This variation was found
to be independent of wind direction and other regional factors, but related to local factors,
such as population density and distance from the center of the city. These results suggest
that models must consider local factors in order to estimate outdoor H+ concentrations
accurately for urban environments.
-------
THE EFFECTS OF METEOROLOGY ON CONCENTRATIONS OF ACID
AEROSOLS
Michael P. Zelenka
Atmospheric Sciences Modeling Division
Air Resources Laboratory
National Oceanic and Atmospheric Administration
MD-56
Research Triangle Park, NC 27711
Helen H. Suh
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
Ambient air samples of strong acid aerosol (H+) sulfate (SOj), ammonia (NH3), and
ammonium (NHJ) were collected in Pittsburgh, State College, and Uniontown,
Pennsylvania during the summer of 1990. High correlations were reported for both H+
and SO^ concentrations between Pittsburgh and the semi-rural sites in State College and
Uniontown. The implication is that acid aerosols and their precursors are regionally
transported.
To test this hypothesis, air samples from the three sites were statistically analyzed for
the meteorological relationships that resulted in episodic events of elevated ambient H+
levels. Standard measures atmospheric conditions included temperature, pressure, relative
humidity, and precipitation. Back trajectory analysis and information on upper air
parameters were also used. The analysis yields important information on the synoptic
scale and mesoscale meteorological parameters which coincided with acid aerosol
episodes. This knowledge can then be used to predict when high concentrations of acid
aerosols would be expected for other cities in the Northeast.
ELEVATIONAL GRADIENTS OF ATMOSPHERIC SULFUR AND NITROGEN
SPECIES MEASURED AT TWO EXPERIMENTAL WATERSHEDS IN THE
EASTERN U.S.
JJ, Bowser and E.S. Edgerton
Environmental Science and Engineering, Inc.
5. Nolan
Institute of Ecosystem Studies
In order to understand the effects of terrain on measurements of concentrations and
fluxes of atmospheric sulfur and nitrogen species, two Clean Air Status and Trends
Network (CASTNet) deposition monitoring sites were located in experimental
watersheds (Hubbard Brook, NH and Coweeta, NC).
At each location, two complete monitoring stations have been operational since the
spring of 1991. Meteorological parameters being measured include; wind speed and
direction, standard deviation of wind direction (sigma theta), temperature, relative
humidity, solar radiation, precipitation, and canopy wetness. Weekly averaged SO2,
-------
HNO3, and particle SO4' and NCy are measured using a three stage filter pack. Ozone is
monitored continuously (hourly averaged) with a UV Photometer.
The sites located within the Hubbard Brook watershed are at elevations of 258 and
600 meters. The Coweeta stations are at elevations of 686 and 1,015 meters. Instruments
are currently being installed at ridge top elevations of 1,600 (Coweeta) and 850 meters
(Hubbard Brook) in order to characterize the entire watersheds.
Preliminary data from the CASTNet deposition monitoring network show that spatial
variability of concentrations and deposition are relatively high in and around the
Appalachian Mountains, especially for the reactive gases SO2 and HNO3. This variability
is of concern because it complicates estimation of deposition to sensitive and potentially
sensitive ecosystems, as well as to mountainous areas in general.
Weekly concentration data for SO4" exhibit a strong correlation between ridge and
valley sites, with a definite tendency for higher concentrations at the ridge site.
Concentration data for SO2 and HNO3 show substantial ridge/valley differences. For
these species, ridge top concentrations are frequently 2 to 3 times higher than valley
concentrations. Ratios of ridge/valley concentrations of SO4", SO2 and HNO3 over a 6
month period of the study at Coweeta were 1.3,2.4, and 2.6 respectively and, in general,
increase in order of calculated deposition velocity Vd. However, the similarity between
SO2 and HNO3 suggests that the two may be depleted at the valley site at similar rates.
This is inconsistent with the Big Leaf Model, which indicates that Vd for HNO3 should
be at least twice that for SO2. This behavior may indicate that other factors, such as
increased frequency of plume detection at ridgetop, contribute to ridge/valley differences.
Estimated fluxes for SO2", SO2, and HNO3 (at Coweeta) exhibit ridge/valley ratios
of 1.4, 2.9. and 4.4, respectively, indicating that strong deposition gradients occur over
short distances in areas of complex terrain.
Information gathered at these locations will be used to evaluate and potentially adjust
modeled flux estimates at other dry deposition monitoring locations in areas of complex
terrain.
GAS-PARTICLE DISTRIBUTION, NEUTRALIZATION AND SIZE OF
SULFATE AND NITRATE IN THE SOUTHWESTERN DESERT AEROSOL
B.J. Turpin, P. Saxena, P.M. McMurry, and G. Allen
Observations made on 47 days during summer 1992 in Meadview, Arizona were
used to characterize inorganic ionic and gaseous concentrations near the Grand Canyon.
A sampler equipped with denuders to remove alkaline and acidic gases measured gaseous
ammonia, sulfur dioxide and nitric acid and fine paniculate ammonium, total extractable
acidity (TEA), sulfate and nitrate. In addition, a set of impactors measured size-resolved
sulfate, nitrate and ammonium.
The median 6-hr average fine paniculate sulfate concentration was 1.6 |J.g/m3 with
about two-thirds of the values below 2. Sulfur dioxide concentrations were comparable to
sulfate concentrations. Two-thirds of the particulate-phase nitrate concentrations were
less than 0.25 (J.g/m3, and the maximum value was 0.5. However, substantial nitric acid
was present in the gas phase (median concentration of 2 |ig/m3). In contrast to prior
10
-------
conclusions that sulfate in this region is generally completely neutralized by ammonia,
the median ammonium to sulfate ratio in this study was 1.4. Furthermore, sulfate and
ammonium concentrations were highly correlated. With one exception sulfate mass
median diameters were 0.15-0.27 [im. More than half of the TEA observations were less
than 8 nanomoles/m3, and the maximum value was 30.
Tuesday Afternoon
THE INFLUENCE OF TRANSPORT ON AEROSOL STRONG ACIDITY
A.B.M. Jeuken, J.R. Brook, and J.D. Spengler
Atmospheric Environment Service
4905 Dufferin Street
Downsview, Ontario M3H 5T4
The degree of neutralization of the sulfuric acid portion of fine atmospheric aerosols
(acid aerosols, H+) varies considerably in time and space. This is due to variations in the
amount of ammonia the acid aerosols come in contact with prior to collection. It is
hypothesized that differences in the predominant transport direction and speed between
collected samples will influence ammonia (NH3) exposure and thus, the observed degree
of neutralization. To study the influence of transport one year of every second day 24
hour acid aerosol measurements from four locations (collected by The Harvard School of
Public Health and Health and Welfare Canada as part of the "24-Community Study")
were combined with back-trajectories. Six different transport patterns were determined
for each location by performing a cluster analysis on the trajectories associated with each
aerosol measurement. The measured ratio of H+ to SO42" was also included in the
clustering in an attempt to further distinguish how transport relates to aerosol
neutralization. As expected, there were significant differences in the ratios between
clusters. There were also differences, some of them significant, between the ambient
levels of H+, SO42", NH3 and SO2. The clusters associated with the higher H+ and SO42'
concentrations were generally associated with transport from a southerly direction or
slow transport (i.e., anticyclones). While there is overlap in the chemical characteristics
of each cluster, the transport patterns associated with the clusters with higher ratios of H+
to SO4 appear to be different than the patterns associated with the other clusters. In
multiple regression models relating H+ to cluster, SO42" and season, knowledge of cluster
was found to explain, at a minimum, an additional 2-6% of the variance depending on
location.
CLEAN AIR STATUS AND TRENDS NETWORK CLOUD WATER
MONITORING PROGRAM
JJ. Bowser, E.S. Edgerton, and VA. Mohnen
The principal objectives of the cloud water sampling program within the Clean Air
Status and Trends Network (CASTNET) are:
ll
-------
Characterize total deposition of sulfur and nitrogen compounds to sensitive
aquatic and terrestrial ecosystems at high elevation, i.e., above cloud base;
Characterize spatial pattern of total deposition and quantify cloud deposition on
a regional basis; and
Provide identification of trend in cloud water chemistry and total deposition by
1997 and quantify a trend, if any, by the year 2001.
High elevation ecosystems can experience increased deposition of sulfur and
nitrogen compounds due to their frequent exposure to clouds, fog, dew or related
deposition processes, and the higher wind speeds aloft causing higher collection
efficiencies by the vegetation canopy.
The CASTNET cloud water deposition monitoring effort will focus on:
The design and implementation of an automated cloud collection system in
combination with the concurrent continuous measurement of cloud liquid water
content (LWC) and meteorological parameters relevant to the cloud deposition
process and trend analysis
The development of regional cloud deposition estimates to high elevation
ecosystems in the Eastern United States
The development of a single database for cloud water chemical measurements
and relevant meteorological parameters by combining the EPA Mountain Cloud
Chemistry Project (MCCP) with the newly initiated CASTNET measurements.
This combined database will be subjected to statistical analysis for trend
identification and assessment.
Four CASTNET high elevation monitoring sites have been selected for cloudwater
collection. These are: Whiteface Mountain, NY; Belleayre Mountain, NY; Whitetop
Mountain, VA; and Clingman's Dome, TN. Two or more additional sites will be
evaluated for future measurements.
The selected sites have a history of atmospheric and ecological research records
available. Furthermore, the sites selected for this long-term monitoring study represent a
range of conditions not only in terms of climate, air quality, soils and vegetation, but also
in atmospheric input of acidifying substances. Therefore, the efforts under this program
either compliment or build on previous research at these sites and intensify their effort at
these sites.
The quality assurance (QA) and quality control (QC) procedures for cloud
collection/LWC measurement will not only ensure the comparability of data throughout
the current CASTNET but will also allow incorporation of all earlier quality assured data
of the MCCP into one single coherent data set. The CCMP QA plan has been developed
far monitoring activities performed during the warm months of the year (approximately
May through October) i.e., in the absence of supercooled cloud water. The collection of
cloud water during winter months will be evaluated and may be initiated in the future
12
-------
Tuesday, May 3,1994
Session 4: Carbonyl Sampling and Analysis
Techniques
Tuesday Morning
INVESTIGATION OF ALDEHYDE CHEMISTRY IN SAMPLING CANISTERS
Thomas J. Kelly and Michael W. Holdren
Battelle
505 King Avenue
Columbus, Ohio 43201-2693
During the summer of 1991, Battelle conducted analyses of air samples from surface
and airborne platforms as part of the Lake Michigan Oxidant Study (LMOS).
Formaldehyde (HCHO) was determined both by the routine analysis of DNPH cartridge
samples, and by direct analysis of air collected in canisters for VOC measurement. The
latter novel approach made use of Battelle's continuous HCHO monitor, and was
motivated by the simplification of sampling that would result if both VOC and HCHO
analyses could be done using canister samples. Comparison of HCHO results from the
cartridge and canister analyses indicates that in most cases agreement is reasonable, but
in about one-third of the results the canister HCHO data are considerably higher than the
DNPH data. This difference has not been traced to any artifact from the canister analysis
procedure for HCHO. Should the discrepancy be due to the formation of HCHO in the
canisters under some conditions, presumably through reactive loss of ozone-precursor
hydrocarbons, then these results would have implications for VOC measurements by
canisters. The present paper reports on a review of the entire LMOS data set, intended to
disclose any chemical causes for the discrepancy in HCHO results. The sample
parameters reviewed for each DNPH/canister pair include: ozone content and total
hydrocarbon content of the air at the time of sampling; residual content of individual
reactive hydrocarbons (e.g., ethylene); location of sample (urban, rural, airborne); altitude
of airborne samples; temperature and humidity of air sampled; and the absolute and
relative HCHO levels indicated by the two analyses. The conclusions and
recommendations drawn from this review will be discussed, in terms of both the LMOS
data set, and the chemistry of air in canisters.
13
-------
A NEW TECHNIQUE FOR CARBONYL SAMPLING AND ANALYSIS
Sucha S. Parmar and Luda Ugarova
Atmospheric Analysis and Consulting
4572 Telephone Road, #920
Ventura, CA 93003
Carmo Fernandas, Peter Lee, and Jim Guyton
Arizona Department of Environmental Quality
3033 N. Central Ave.
Phoenix, Arizona 85012
Ambient carbonyl compounds (aldehydes and ketones) are normally sampled using
2,4-dinitrophenylhydrazine in solution phase or coated on C18 cartridges (solid phase).
The trapping of carbonyls is achieved by the addition of DNPH onto the carbonyl double
bond followed by the elimination of a water molecule. This reaction is almost 100%
efficient provided ambient humidity is above 40% and ambient ozone concentration is
low. Ozone has been found to interfere in carbonyl sampling as it attacks all molecules
containing double bonds.
We have recently conducted a preliminary investigation into the alternative methods
of sampling and detection of carbonyls. In this investigation laboratory and field studies
were carried out to trap carbonyls on activated charcoal tubes followed by extraction with
methylene chloride, derivatization with DNPH and HPLC/UV analysis. In this way
carbonyl collection efficiency is not dependant on atmospheric humidity or ambient
ozone concentration. Results from charcoal and DNPH sampling methods in the presence
of interferents are compared and presented.
DETERMINATION OF CARBONYL COMPOUNDS BY HPLC/UV ANALYSIS
IN THE CASTNET AIR TOXICS MONITORING PROGRAM (CATMP)
Bradley A. Weichert, Robert D. Baker, Benjamin T. Johnson,
and Michael G. Winslow
Environmental Science & Engineering, Inc. (ESE)
P.O. Box 1703
Gainesville, Florida 32602
The determination of carbonyl compounds in ambient air is one of the analytical
requirements of the Environmental Protection Agency's (EPA's) Clean Air Status and
Trends Network (CASTNet) Air Toxics Monitoring Program (CATMP). The CATMP
was established in 1993 by EPA to reactivate and operate the Urban Air Toxics
Monitoring Program (UATMP). The purpose of the program is to establish baseline
toxics concentrations, develop air emission inventories, and to identify air toxic sources
using chemical mass balance modeling techniques.
1993 results for the CATMP analysis of carbonyl compounds by method TO-11 are
presented. Samples were collected using Waters' dinitrophenyl hydrazine (DNPN)
impregnated cartridges and analyzed using high performance liquid chromatography
(HPLC) with ultraviolet (UV) detection. A regulatory program approach to quality
14
-------
control was taken to insure that all aspects of shipping, sampling and analysis contained
sufficient controls in order to produce data of acceptable and consistent precision and
accuracy. Database presentations will include the overall quality assurance approach,
quality control measurements, and sample results from eight sites. An examination of
background interferences detected in the various control samples will also be discussed.
A SEQUENTIAL CARTRIDGE SAMPLER FOR MEASURING CARBONYLS
AND TOXIC ORGANIC COMPOUNDS
DwightA. Landis
Atmospheric Technology
P.O. Box 8062
Calabasas, CA 91372-8062
A multi-channel sequential sampler has been developed to collect carbonyl
compounds with silica gel or C-18 cartridges using EPA Method TO-11. The sampler
automatically draws ambient air through the cartridges over a predetermined time
sequence. The sample flow rate is continuously controlled and monitored using mass
flow controllers. An additional channel is provided so that duplicate samples can be
simultaneously obtained. The sampler can be interfaced with a micro-computer to
provide custom timing sequences, acquire and store flow rate data, and perform quality
assurance procedures. A field test was conducted in Claremont, California, using AtmAA
C-18 cartridges to samples formaldehyde. The results show good agreement with values
obtained with a Tunable Diode Laser Absorption Spectrometer. Additional testing has
also been performed to demonstrate long term reliability of the sampler.
MEASUREMENT OF CARBONYL COMPOUNDS IN AN INDUSTRIALIZED
AREA
James R. Ellis, David F. Gregorski, Michael J. Murphy, and
Louis J. Scarfo
State of Connecticut Department of Environmental Protection
Bureau of Air Management, Air Toxics Monitoring Division
79 Elm Street, 6th Floor
Hartford, CT 01606-5127
Connecticut's Department of Environmental Protection (CT/DEP) has compiled a
history of health related complaints in the town of Wallingford, CT, especially from those
people living in areas adjacent to the heavily industrialized Route 5 corridor. In addition,
the Agency for Toxic Substances and Disease Registry (ATSDR) was petitioned to assess
the environmentally related health complaints and concluded that there was insufficient
air quality data from Wallingford to clarify and/or quantify the possible existence of a
public health hazard. A search of Connecticut's SARA Title HI database suggested that
there were several major sources of toxic air emissions in the vicinity of Route 5. In order
to address these complaints and needs, the DEP's Air Toxics Monitoring Group
15
-------
performed a year long study to monitor for selected VOC and carbonyl compounds. This
paper reports on the measurements of selected carbonyl compounds, from several sites
adjacent to the Route 5 corridor, between October 1992 and October 1993.
The carbonyl samples were collected in conjunction with other selected VOC
samples and relevant meteorological parameters. Sampling was performed using
dinitrophenylhydrazine (DNPH)-coated cartridges. The samples were collected for 24
hours per day, on a six-day schedule, at several sites. These sites were selected to
maximize sample collection during periods of localized prevailing winds. These sample
tubes were then analyzed by high performance liquid chromatography (HPLC). An
analytical protocol has been developed by CT/DEP's analytical laboratory. ERI, for
analyzing 15 carbonyl compounds. This protocol is based on EPA Method TO-11 and
ASTM Method D5197-92. The analytical data show that there have been several
instances where reported values of formaldehyde have exceeded Connecticut's
Hazardous Limiting Values (HLVs) guidelines.
16
-------
Tuesday, May 3,1994
Session 5: Enhanced Ozone Monitoring Status and
Development
Tuesday Afternoon
PHOTOCHEMICAL ASSESSMENT MONITORING: OVERVIEW AND
CURRENT STATUS
Nash O. Gerald and Geraldine Dorosz-Stargardt
U.S. EPA
Technical Support Division (MD-14)
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
The Clean Air Act Amendments of 1990, established requirements for the
Environmental Protection Agency (EPA) to develop rules for the establishment of
enhanced ozone monitoring networks or Photochemical Assessment Monitoring Stations
(PAMS) in ozone nonattainment areas designated as serious, severe, and extreme. The
subsequent rules require these stations to collect ambient air measurements for a target
list of volatile organic compounds including several carbonyls, oxides of nitrogen, ozone,
and meteorological measurements, both surface and upper air.
Twenty-two areas in the United States are obligated to install and operate PAMS
stations to aid in the identification, development, and implementation of effective ozone
control strategies. This paper will examine the specific requirements of the PAMS rules
and will provide information regarding the current status of the networks and overall
implementation issues.
VOC Measurements
OVERVIEW OF VOC TECHNOLOGY IN THE PAMS PROGRAM
William A. Lonneman
U.S. EPA
Research Triangle Park, NC 27711
The PAMS program began in early 1993 with an Implementation Teleconferencing
Workshop for States and EPA Regional Offices to discuss the methodologies and
programs available for assistance to support the enhanced monitoring regulations in the
1990 Clean Air Act Amendments. The teleconference was arranged by the Office of Air
Quality Planning and Standards (OAQPS) with technical discussions from members of
the Atmospheric Research and Exposure Assessment Laboratory (AREAL). Since then
several States have commenced VOC measurement procedures. A number of issues have
17
-------
developed during the first year, including the target VOCs, measurement procedures and
data base development aspects. In this paper, an overview of the current status of the
VOC measurements in the PAMS program will be discussed.
NEW DEVELOPMENTS FOR THE MONITORING OF VOCS IN PAMS
William A. McClenny
U.S. EPA
Research Triangle Park, NC 27711
New developments to improve existing PAMS instrumentation for ozone precursors
include: (a) the use of solid adsorbents in a combined concentration and water
management system; (b) the use of closed cycle coolers for refocusing of concentrated
ozone precursors prior to analysis; and (c) the development of additional information on
storage stability of ozone precursors. The use of solid adsorbents in a multisorbent
packing to adsorb VOCs when sampling directly from the air manifold or when sampling
from a canister, allows for an alternate water management system based on dry purging
after sample collection. Water vapor is preferentially lost during the dry purge and the
trapped ozone precursor hydrocarbons are then thermally desorbed, refocused, and then
separated and detected. Commercial prototype closed cycle coolers have been tested for
refocusing hydrocarbon precursors after desorption from multisorbent packings. The
closed cycle coolers take the place of controller release of liquid nitrogen to refocus
compounds in a small volume before release onto a GC column. For canister, information
on additional storage stability work is now available for samples stored in SUMMA-
poh'shed canisters. The work consists of both model development for prediction of the
storage stability in canisters and experimental tests using many of the ozone precursor
compounds.
VALIDATING DATA FROM AUTOMATED, CONTINUOUS PAMS GC
SYSTEMS
Joann Rice
Radian Corporation
Research Triangle Park, NC
Automated, continuous gas chromatographic (GC) techniques are currently the only
cost effective way to directly meet the rigorous speciated VOC sampling frequency
required for Photochemical Assessment Monitoring Stations (PAMS) ambient air quality
surveillance monitoring networks. Validating data from these automated systems is the
final and most critical part of the chromatographic measurement process. The data that is
ultimately entered into the computer-based Aerometric Information and Retrieval System
(AJRS)-Air Quality Subsystem (AQS) data base is essential to the tracking and reduction
of precursors to tropospheric ozone, therefore; the data must be consistent, reliable, and
accurate. The EPA has defined data validation as a systematic process consisting of data
editing, screening, checking, auditing, verification, certification, and review, for
18
-------
comparing a body of data to an established set of criteria to provide assurance that the
data are adequate for their intended use. Prior to validating data, it is imperative that
pre-measurement GC system setup and characterization, and Quality Assurance (QA)
procedures for validating the GC data be performed to set the stage for successful
validation of the generated data. The final data validation step, prior to ATRS-AQS entry,
includes validating the chromatography result report, implementing a secondary
processing/peak identification software, and qualitative or comparison review of a
summary concentration ASCII report file. This paper discusses the procedures necessary
to develop and implement an approach for validating speciated VOC data from
automated, continuous GC PAMS systems.
A QA PROGRAM FOR THE PHOTOCHEMICAL ASSESSMENT
MONITORING NETWORK
Jack C. Suggs, Howard Crist, and William J. Mitchell
U.S. EPA/AREAL (MD-77B)
Research Triangle Park, NC 27711
Ronald Bousquet and Ronald Brande
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
The results obtained to date from a multi-purpose, cooperative effort between the
USEPA and 22 state and local agencies operating PAMS stations are presented. The
purposes of this cooperative effort are: to provide training to the 22 agencies who are
measuring PAMS compounds as directed by the Clean Air Act Amendments of 1990, to
provide the USEPA data on the precision and accuracy that can be expected from PAMS
monitoring network measurement systems, to provide the USEPA EPA data on the
performance that can be expected from a newly developed auditing system for PAMS
compounds, and to lay the foundation fora nationwide audit program for PAMS
measurement systems which will be initiated in 1995.
DISCLAIMER: The information in this document has been funded wholly by the
United States Environmental Protection Agency. It has been subjected to Agency review
and approved for publication.
INTERREGIONAL AND INTERSTATE QUALITY ASSURANCE FOR PAMS:
CENTRAL REGIONAL LABORATORY COMPARISONS
Avraham Teitz and Marcus Kantz
U.S. EPA, Region II
Edison, New Jersey 08837
To ensure data comparability between the various State agencies conducting PAMS
monitoring in EPA Regions I, n, and HI, a central laboratory (CRL) was designated for
each of these three EPA Regions. Three round robin samples were analyzed by each of
the CRLs to determine the inter-laboratory variability.
19
-------
Each of the round robin samples contained an ambient air sample, which was
collected in passivated stainless steel canisters. Each sample has pollutant concentrations
ranging from 20 ppbC to <1 ppbC. Percent differences between the three CRLs were less
than 30% for most of the PAMS compounds, and typical results were within 10 to 20%.
Compounds which were present in higher concentrations in the sample showed the
closest agreement between laboratories. Compound identification was consistent across
all three laboratories with the exception of two compounds, beta-pinene and
1,2,4-trimethyl benzene.
The results of these round robins show great improvement over similar studies
conducted 5 years ago in Staten Island, where inter-organization data comparability was
typically ± 50%.
RTF AUTO-GC DEMONSTRATION EXPERIMENT
Jimmy C. Pan, Ron Drago, James Reagan, and Larry Purdue
U.S. EPA
Research Triangle Park, NC
An Auto-GC demonstration was conducted at the AREAL test facility located in the
ERC Annex at RTP, North Carolina throughout a four week period from August 31 to
September 28, 1992. Five potential commercial vendors were invited to participate in this
demonstration experiment.
The primary purpose of this experiment was to demonstrate the capabilities of the
commercial automated gas chromatographic systems which had been developed up to
that point, in measuring volatile hydrocarbon species in accordance with the provisions of
the proposed requirements for monitoring ozone precursors. These proposed
requirements and associated guideline documents were provided to each of the
participant vendors.
Each of the five automated GC systems was set-up for hourly collection and analysis
of the species targeted in the requirements and was operated from a common ambient air
manifold under prescribed sampling schedules throughout the experiment. EPA Regional
Offices, State and local air pollution agency representatives were invited to observe the
operations of these systems, and technical information exchange was made between the
EPA researchers and representatives of each of the participating manufacturers.
The experiment successfully demonstrated the operation of each of these emerging
systems. Ninety-seven hours of operation provided sufficient data for inter-comparison of
all five systems. The results of this comparison for 12 hydrocarbon species are
summarized here.
20
-------
Wednesday Morning
CONNECTICUT AUTO GC EXPERIMENT
A. Leston
State of Connecticut
Department of Environmental Protection
79 Elm Street
Hartford, CT 06106-5127
An evaluation has been completed of the relative performance of two commercial
gas chromatographic systems for ambient air analysis. Neither cryogenic cooling of
adsorption trap material nor low temperature (-180° C) refocussing, prior to sample
injection, were found to be necessary for satisfactory separation of the C2-C9 compounds
listed in EPA's enhanced monitoring requirements for Photochemical Assessment
Monitoring Stations (PAMS). However, the permeation-style dryer utilized by both
systems removes many polar and/or oxygenated species of interest. Despite this
drawback, both systems have proven their ability to quantify most of the targeted
compounds at or below the requisite 1.0 ppbc (part per billion carbon) detection level.
In addition to obtaining the technical expertise to operate auto GC systems,
organizations involved in the PAMS program need to reevaluate their data handling
resources. Raw data from a continuously operating, remote GC must be backed up,
telemetered and quickly assessed in order to efficiently utilize human resources and
minimize lost data. Because existing data logging and telemetering systems are structured
for relatively low data flows, the large quantity of VOC-related raw data from a single
GC (>1.5 Mbytes/day) can easily overwhelm such systems. Data retrieval and assessment
may well prove to be the most labor intensive activity associated with the PAMS
program.
21
-------
ANALYSIS OF OZONE PRECURSOR COMPOUNDS IN HOUSTON, TEXAS
USING AUTOMATED, CONTINUOUS GAS CHROMATOGRAPHS
John Gibich
Texas Natural Resource Conservation Commission
P.O. Box 13087
Austin, TX 78711-3078
Larry Ogle
Radian Corporation
P.O. Box 201088
Austin, TX 78720-1088
Paul Radenheimer
Consolidated Sciences, Inc.
1416 Southmore
Pasadena. TX 77502
In the summer of 1993 the Texas Natural Resource Conservation Commission
(TNRCC) sponsored the Coastal Oxidant Assessment for Southeast Texas (COAST)
study that had as its objective improvement of the technical basis for designing effective
and cost-effective ozone control strategies for the upper gulf cost area of Texas. As a part
of the COAST study, two automated, continuous gas chromatographic (GC) systems
from the Perkin-Elmer Corporation were sited in Houston, Texas, to analyze for the 55
U.S. Environmental Protection Agency designated ozone precursor compounds in
ambient air on an hourly basis. The Perkin-Elmer systems were selected because they
appeared to offer two distinct advantages for this work. Firstly, the preconcentration
(trapping) of the compounds of interest used a sorbent trap cooled by a Peltier device,
thus eliminating the need for cryogenic liquids on-site. Secondly, a Dean's pressure
switch permitted the utilization of two capillary columns for the more effective
chromatographic separation of the C2 through C9 compounds. The Radian Corporation in
Austin provided data and overall management for this project and Consolidated Sciences,
Inc., a local firm in Houston, handled all on-site activities. TNRCC oversight of the
installation, optimization and operation of these systems proved to be a valuable source
of information for the future establishment of these systems at Photochemical
Assessment Monitoring Station (PAMS) #2 sites (maximum ozone precursor emissions
impact sites) in Texas. Data collection at the two sites began on June 15 and continued
through November 30, 1993. Simultaneously, a rigorous quality assurance/quality
control program based on the PAMS QA requirements promulgated in 40 CFR Part 58 on
February 12,1993, was implemented to ensure that the measurement data would be of the
highest quality. In general, both systems performed very well with the data collection
averaging above 90% for the entire time. While a number of significant problems were
discovered during both the startup and the operation, most of these were eventually
resolved satisfactorily. Discussion will focus on issues dealing with system operation and
evaluation. Estimates of data precision and accuracy will be given. Estimates of
manpower requirements for all phases of the operation, including data management, will
also be addressed.
22
-------
LABORATORY INTERCOMPARISON FOR ANALYSIS OF HYDROCARBONS
DURING THE COASTAL OXIDANT ASSESSMENT FOR SOUTHEAST TEXAS
(COAST) PROJECT
Eric M. Fujita, Douglas R. Goff, and Douglas R. Lawson
Energy and Environmental Engineering Center
Desert Research Institute
P.O. Box 60220
Reno, NV 89506
(702) 677-3311 and -3316 (FAX)
James Price
Texas Natural Resource Conservation Commission
P.O. Box 13087
Austin, TX 78711-3087
William A. Lonneman
U.S. EPA
Atmospheric Research and Exposure Assessment Laboratory, MD 84
Research Triangle Park, NC 27711
Steve D. Hoyt
Environmental Analytical Services, Inc.
3421 Empresa Suite A
San Luis Obispo, CA 93401
R.A. Rasmussen
Oregon Graduate Institute of Science and Technology
19600 N.W. Von Neumann Drive
Beaverton, OR 97006-1999
Walter L. Crow and Larry D. Ogle
Radian Corporation
8501 Mopac Blvd. North
Austin, TX 78759
In the summer of 1993, a field measurement program was carried out in the Houston-
Port Arthur as part of the Coastal Oxidant Assessment for Southeast Texas (COAST)
project. The central focus of the COAST project is the acquisition of speciated
hydrocarbon data at several ground-level monitoring sites, accompanied by aircraft
measurement of hydrocarbons on forecasted intensive study days. Besides collection of
canister samples, two continuous volatile organic compound monitors were operated at
two sites throughout the July 15 to August 30 study period. Because speciated
hydrocarbon data were acquired by three different contractors, an important aspect of the
measurement program involved the interlaboratory comparisons among participating
laboratories. The intercomparison study consisted of two reference samples, an EPA
standard mixture and an archived ambient sample from Atlanta, and three ambient
samples collected during the field study by each of the three contractors at the same
location and periods. The collected ambient samples were analyzed by the contracting
laboratory and by the Environmental Protection Agency Atmospheric Research and
Exposure Assessment Laboratory, which served as the referee laboratory. This paper
23
-------
describes the study protocol, and reports the results with emphasis on differences among
laboratories in compound identification, fraction identified, and concentrations for
individual and total hydrocarbons.
A COMPARISON OF THE RESULTS OF A TOTAL NON-METHANE
ORGANIC COMPOUND ANALYZER WITH THOSE FROM A
CANISTER-BASED MONITORING PROGRAM
Michael Poore and Natividad Lapurga
The need for frequent measurements of total non-methane organic compounds
(TNMOC) in ambient air has been emphasized by data analysis of both the Southern
California Air Quality Study (1987) and the Atlanta 1990 Ozone and Ozone Precursor
Study. Although the TNMOC concentrations may vary considerably both temporally and
spatially within a given area, it has been shown that, in California at least, the NMOC
composition does not vary appreciably within that area As a result, the staff of the
California Air Resources Board have worked closely with a commercial instrument
manufacturer to produce a prototype TNMOC analyzer. This instrument was field tested
during the 1993 ozone season in parallel with a canister collection program for PDFID
and speciated hydrocarbons. A comparison of the results from the TNMOC analyzer, a
specially modified Combustion Engineering Model 8202a CH4/THC analyzer, and the
canister-based program will be presented.
A FULLY REMOTE CONTROL CRYOGENLESS OZONE PRECURSOR
SYSTEM WITH IMPROVED SENSITIVITY
D.B. Cardin andJ.T. Deschenes
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
In compliance with Title I of the 1990 Clean Air Act Amendment (1990 CAAA),
hydrocarbons in the C2-C10 molecular weight range will be monitored to assess their
contribution to ozone formation in approximately 35 cities which exceed current ozone
standards during summer months. Speciation of these compounds in urban air requires
preconcentration of a 300 cc sample size in order to reach the required 0.1-0.2 ppbc
detection limits by GC/FID. Analyzing the C2-C10 hydrocarbon fraction can either be
performed real time by placing the analytical instrumentation at the sampling site, or by
collecting the ambient air in separate SUMMA passivated canisters every 3 hours. In
addition, sample analysis can be performed either by using cryogen to preconcentrate the
sample and cool the GC for single column GC/FID, or by using sorbent traps and two
columns in the GC which eliminates the need for liquid cryogen. Once implemented,
on-site analysis using cryogenless operation requires me least amount of manpower to
operate providing from weeks to months of uninterrupted data generation.
24
-------
A cryogenless C2-C10 "AUTOGC" system is presented that exceeds the monitoring
requirements specified in the technical assistance document for ozone precursor
monitoring. Samples can be analyzed hourly or once every 3,8, or 24 hours to provide as
much resolution as necessary. Sampling frequency can be changed via modem to
accommodate periods of high and low ozone concentrations. A benchtop 16-position
manifold makes it possible to analyze other gas streams at programmed intervals, such as
propane and hexane standards, retention time standards, and system blanks. Integration of
the ambient air sample first into a canister allows standards and blanks to be run while
continuing to analyze ambient air 24 hours per day. Access to the GC/FID and
preconcentrator are possible via modem using a Windows™ interface for data retrieval,
verification of proper operation, and adjustment of method parameters. Data will be
presented showing performance in determining ozone precursor concentrations in Los
Angeles, California,
THE DETERMINATION OF OZONE PRECURSORS WITH A BUILT-IN
PRECONCENTRATOR AND CAPILLARY GC/DUAL FID
Norman Kirshen and David Coe
Varian Chromatography Systems
2700 Mitchell Drive
Walnut Creek, CA 94598
510-945-2230/FAX 510-945-2335
Tide I of the Clean Air Act Amendments of 1990 requires cities or air quality
districts that are non-compliant with the Ozone National Ambient Air Quality Standard of
120 ppbv to make continuous measurements of targeted ozone precursors. Air samples
must be obtained on a one to three hour basis with the requirement that a high percentage
of the time interval will be spent in the sample collection process. The preconcentration
of the target analytes from these samples by adsorbent trapping rather than the traditional
use of cryogenic preconcentration on glass beads with liquid nitrogen has been utilized
over the past few years for this application.
An Ozone Precursor GC system has been configured with a new built-in sample
preconcentration trap (SPT) and associated valving capable of preconcentrating a 200 mL
air sample. The sample which may be drawn either from a canister or directly from the
ambient air first passes through a stripper column for moisture removal and then through
an adsorbent trap. The adsorbent trap is cooled to initial conditions with a very small
amount of liquid carbon dioxide providing many runs from one cylinder. Following
trapping, the precursors are quickly desorbed with the fast heating (40° C/sec) SPT to
parallel columns of Alumina Plot and DB-1 for chromatographic separation and
quantitation of the trapped analytes. Dual FIDs provide detection.
Optimization of the trapping and chromatographic parameters has been performed
using a 60 component EPA precursor standard. The results of these studies and a
description of the integrated system will be presented.
25
-------
AUTOGC SYSTEMS FOR ENHANCED OZONE MONITORING NETWORK
Ian Seeley
The Perkin-Elmer Corporation
Norwalk,CT 06859
Since September 1992, when the first "AutoGC" systems were evaluated at the EPA
"shoot-out" in AREAL, a number of systems have been installed and have collected data
during the summer of 1993. Data capture rates in excess of 95% have been achieved at
some sites, during the period beginning June 1st through November 1st. Numerous units
based on "cryogen-free" technology are now operational, with tens of thousands of hours
of data collected.
On this basis it is indeed feasible to locate these units in the field for remote
monitoring of the majority of the compounds of concern under Title I of the Clean Air
Act of 1990.
Improvements are being made to these systems, particularly with respect to remote
communications since the systems are being considered for relatively inaccessible
locations. This paper will discuss the criteria required for an autonomous AutoGC,
including details of the systems that have been used in 1993, and some of the
enhancements that are being considered for improving them prior to the start of the 1994
monitoring season.
ENHANCED MONITORING OF OZONE PRECURSORS - SPECIATED AND
TOTAL NMOC
Sharon Reiss and Richard Jesser
Graseby/Nutech
For the past two years, Graseby/Nutech has supported an automated cryogenic
system for the concentration and subsequent speciation of NMOCs that contribute to the
formation of ozone in ambient air. This system consists of a cryogenic concentrator
interfaced with a GC/FID with pressure programming capabilities. The appeal of this
approach is twofold:
a. high quality data associated with cryotrapping;
b. single column/single detector for a simple hardware scheme generating no
more than one data file for each sample.
Although this cryogenic application may be the best choice in a stationary
laboratory, the use of liquid nitrogen in a remote or temporary location is not always
practical. For this reason, Graseby/Nutech is developing a non-cryo continuous
monitoring system that is simple in design without being complicated in operation or
maintenance.
26
-------
Wednesday Afternoon
Carbonyl Measurements
PARAMETERS INFLUENCING THE PRECISION AND ACCURACY OF
CARBONYL MEASUREMENTS USING DNPH CARTRIDGES
KochyFung
AtmAAInc.
23917 Craftsman Rd.
Calabasas.CA 91302
While DNPH cartridges have been used by the author to measure ambient air
carbonyl compounds in all major air quality studies throughout California and other parts
of the United States over the past ten years, there are still remaining issues with the
method that need to be addressed in order to produce unbiased results. At typical
ambient levels, the method precision is in the range of 10 to 25%. Automated sequential
samplers were employed in these studies. They have been very reliable and produced
data with'good precision when memory-protected timers and flow controllers were used.
Samplers account for up to approximately 30% of the error. The precision of the HPLC
analytical technique with an autosampler or using an internal standard has a precision of
± 2% or less. DNPH cartridge blank variability is the largest factor that controls the
method's lower quantifiable limit, which is approximately 0.5 ppbv for 60 liters of air
sampled. The cartridge substrate (CIS vs silica) used for DNPH impregnation appeared
to have varying extent of ozone interference. While it was thought that DNPH cartridges
made with CIS bonded silica have less problems with ozone than plain silica, more
recent experimental data showed that when both types were impregnated under identical
conditions, silica cartridges performed equally or better, suggesting that the composition
of DNPH reagent was also important in determining the extent of the ozone artifact seen.
THE DESIGN AND USE OF DNPH COATED SILICA CARTRIDGES FOR THE
ANALYSIS OF CARBONYLS IN OUTDOOR Affi
R. Crowley, J. Arsenault, E. Bouvier, S. Harris, M. Hopkins, P.
Iraneta, G. Knowles, D. Martin, and T. Walter
Millipore Corporation
Walter Chromatography Division
34 Maple Street
Milford,MA 01757
The determination of formaldehyde, acetaldehyde and acetone concentrations in
ambient air is a required specification of the PAMS regulation. Section 5.0 of the
Technical Assistance Document defines the testing methodology for measuring carbonyls
in ambient air. This testing methodology is based on the use of DNPH coated silica
cartridges. The user is given the choice of either preparing their own cartridges or
27
-------
purchasing them from a commercial source.
Waters introduced the DNPH Silica Sep-Pak Cartridge into the marketplace in 1991.
This cartridge is now in widespread use throughout the world in numerous carbonyl air
monitoring applications. A key component of our success has been the transferal of our
product and application expertise to our customers.
This presentation will focus the keys for the successful utilization of the precoated
DNPH cartridges for carbonyl monitoring in ambient air. Product design and product
performance issues such as lot-to-lot consistency, shelflife, sample stability and potential
analyte interferences will be addressed. Examples of actual analysis problems, their
symptoms and solutions will also be presented.
MEASURING AMBIENT CARBONYLS USING DNPH-C GATED SUBSTRATES:
COMPARISON OF CIS AND SILICA GEL WITH AND WITHOUT OZONE
SCRUBBERS
Daniel Grosjean and Eric Grosjean
DGA, Inc.
4526 Telephone Road, Suite 205
Ventura, CA 93003
Carbonyls in ambient air play a critical role in atmospheric chemistry, including
ozone formation on urban and regional scales. Recently mandated requirements to report
ambient levels of formaldehyde and other carbonyls as part of the nationwide PAMS
monitoring program underline the importance of obtaining reliable, interference-free data
when measuring ppb levels of carbonyls in urban air. The DNPH-liquid chromatography
method, initially described by Kuwata et al. in Japan and by Grosjean and Fung in the
U.S., has since been endorsed by ASTM, EPA (e.g., TO 11) and other agencies; this
method has been suggested by EPA as the method of choice for PAMS monitoring of
ambient levels of carbonyls measured using co-located CIS and silica gel substrates with
and without scrubbers to remove ambient ozone. As part of this study, the performance
of several types of ozone scrubbers has been evaluated. The results will be discussed
with special emphasis on PAMS requirements for monitoring of ambient carbonyls by
state and local air pollution control agencies.
AN EVALUATION OF COMMERCIAL DEVICES FOR AUTOMATED
SEQUENTIAL SAMPLING OF CARBONYL COMPOUNDS IN AIR
Thomas J. Kelly, Deborah L. Smith, Richard N. Smith, and Philip
M. Schumacher
Battelle
505 King Avenue
Columbus, Ohio 43201-2693
This paper reports on a laboratory evaluation of four commercial instruments
designed for unattended sampling of carbonyl compounds in air using DNPH sampling
28
-------
cartridges. These instruments are intended to meet the requirements for automated
sequential sampling of carbonyl compounds in air, as described in the U.S. EPA's
"Technical Assistance Document for Sampling and Analysis of Ozone Precursors". The
four instruments were loaned for this study by the respective manufacturers (i.e.,
Xontech, Xitech, SIS Inc., and Graseby-Andersen), and represented the current state of
development of each sampler as of late 1993. This evaluation was limited to the
engineering aspects of each design, including the extent to which the design met the
requirements stated in the Technical Assistance Document. No field sampling or
intercomparison was done. The instruments were remarkable diverse in design, and
exhibited a wide variety of advantages and disadvantages in meeting the requirements for
automated sequential ambient sampling. This paper will describe each instrument, note
the advantages and limitations of each design, and present recommendations for further
development of such samplers.
PERFORMANCE MEASUREMENTS OF CrC3 CARBONYL COMPOUNDS
USING DNPH-COATED SILICA GEL AND C18 CARTRIDGES
T.E. Kleindienst, E.W. Corse, andF.T. Blanchard
ManTech Environmental Technology, Inc.
P.O. Box 12313
Research Triangle Park, NC 27709
WA. Lonneman
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Measurements have been conducted to examine the performance of formaldehyde,
acetaldehyde, ad acetone using silica and C18 cartridges coated with
2,4-dinitrophenylhydrazine. Laboratory measurements for formaldehyde were conducted
using a paraformaldehyde generator to produce reproducible and constant concentrations
of the compound. For acetaldehyde and acetone, known concentrations were generated
in Teflon chambers. The compounds were routed into a sampling manifold where
simultaneous measurements could be made with multiple cartridges. Typical
concentrations employed in the study were as follows. HCHO: 0.5-25 ppbv; CH3CHO;
0.5-10 ppbv; CH3C(O)CH3: 0.5-10 ppbv. Additional measurements were conducted for
these compounds in the presence of potentially interfering compounds such as ozone and
water vapor. Serial cartridge collections were periodically used to investigate
breakthrough the carbonyl compounds.
29
-------
EVALUATION OF AN INTELLIGENT
MULTI-CANISTER/MULTI-CARTRIDGE SAMPLER FOR THE COLLECTION
OF OZONE PRECURSORS
Mark K. Allen, Edward Miller* and Joseph Leair*
Wisconsin Department of Natural Resources
Madison, Wisconsin 53707-7921
Air monitoring requirements for Enhanced Ozone Monitoring (EOM) set down in 40
CFR Part 58 require the collection 8 three hour canister/cartridge samples each sampling
day. To meet EOM monitoring requirements for Southeastern Wisconsin, we have
designed and built a sampler capable of collecting multiple canisters (for hydrocarbons)
and multiple cartridges (for carbonyls). The sampler is comprised of three modular units,
the canister collection system, the cartridge collection system, and a personal computer
which acts as an intelligent controller for the sampling system. The intelligent controller
will track all samples from the sampler setup, through sampling, to sampler take down.
Relevant sampling flow rates are all recorded by the sampler. At the completion of the
sampling episode all sampling data is down loaded to initial testing of the sampler, and
the sampler performance during an initial season of sampling.
*Mr. Miller and Mr. Leair are stationed at the Wisconsin Department of Natural
Resources Southeast District Headquarters in Milwaukee.
THE DEVELOPMENT OF A CARBONYL SAMPLER
Matthias J. Yoong, Michael A. Pardee, and Ralph D. Herdman
XonTech, Inc.
Environmental Systems Group
7027 Hay venhurst Avenue
VanNuys,CA 91406
In accordance with the Clean Air Act Amendments of 1990, rules were developed
for the establishment of the Enhanced Ozone Monitoring Network and Photochemical
Assessment Monitoring Stations (PAMS) in ozone nonattainment areas, these rules
require the stations to collect ambient air measurements for a target list of meteorological
measurements of surface and upper air. Included in the VOCs are several aldehydes. In
the past, canister sampling and analysis for C2-C10 hydrocarbons have been the subject of
much discussion and implementation. Presently carbonyl compounds, especially low
molecular weight aldehydes and ketones are receiving increased attention as toxic air
pollutants or as intermediates in many photochemical reactions. Because these
compounds occur in trace concentrations ranging from sub-parts per billion to higher
parts per billion in the atmosphere, sampling for carbonyl compounds poses some
challenging problems. XonTech, Inc. has designed a carbonyl sampler suitable for
Ozone Precursors Monitoring Program or the PAMS requirement of short time
monitoring around the clock. The design, features, and operation of this sampler will be
presented.
30
-------
THE SUPELCO™ CARBONYL AMBIENT AIR SAMPLER FOR
EPA-600/4-84-041, METHOD TO-11
Dwight Patterson
Xitech Instruments, Inc.
300C Industrial Park Loop
Rio Rancho, NM 87124
TM
Supelco is proud to announce the introduction of its New Sequential Air Samplers
for PAMS. This new technology satisfies all requirements of Section 182 of the 1990
Clean Air Act, EPA 40 CFR Part 58, and ASTM Standard 5197-92 for sampling of
Formaldehyde and other Carbonyl compounds. Our choices cf sampling equipment
provide great flexibility at the right price. Due to the vigorous demand of non-stop
sampling for several months, Supelco has developed a totally independent modular
sampling concept to insure no sampling time is lost due to maintenance. This technology
is capable of operating unattended for up to four complete days. Some of our unique
features include: electronic timing sequencer for eight samples; ninth collocation sample
for 24 hour reality check; seven day event timer for activating up to five sampler
modules; a common heated inlet manifold with built-in ozone scrubber and check valves
for sample isolation; electronic mass flow controls up to 2 liters per minute for long term
flowrate stability; and inboard optional 2-channel flowrate data logger.
Thursday Morning
Oxides of Nitrogen - NOy
MEASUREMENT OF TOTAL REACTIVE ODD-NITROGEN (NOY) IN THE
RURAL AND NON-RURAL TROPOSPHERE
Scott T. Sandholm and John D. Braashaw
School of Earth and Atmospheric Sciences
Georgia Institute of Technology
Atlanta, GA 30332
Measurements of total reactive odd-nitrogen (NO ) have become increasingly
important to the understanding of the atmospheric transport and transformation processes
affecting the burden of reactive oxides of nitrogen (in particular NOX = NO + NO2) and
their influence on controlling atmospheric oxidant levels (specially O3 and OH) in many
rural and non-urban environments. NO measurements have become useful indicators of
the total abundance of the various reservoir forms of NOX as urban air parcels become
progressively processed in the lower atmosphere. The accurate measurement of NO
under these varying conditions represent a unique set of measurement challenges, as NO
goes from tens of ppbv levels in predominately NOx-rich mixtures of odd-nitrogen to
ppbv and sub-ppbv levels in NOx-poor mixtures. The instrument issues affecting
31
-------
measurements of NO using an Au-catalytic converter system will be addressed for
conditions relevant to sampling in the rural and non-urban atmosphere.
RECENT ADVANCES IN THE DESIGN OF NO2 PHOTOLYTIC CONVERTORS
John Bradshaw, Scott Smyth, Rongpo Bai, and Scott Sandholm
School of Earth and Atmospheric Sciences
Georgia Institute of Technology
Atlanta, GA 30332
Several photolytic converter designs have evolved from the quantitative conversions
of NO2 into the more readily and conveniently measurable species NO (via
NO2 + ho -> NO + O). To date the most popular of these designs utilizes the high
spectral brightness of a well collimated short-arc lamp to produce a relatively compact
and efficient (i.e., 40-80%) system for converting ambient NO2 into NO. Although
reasonably reliable, these converters can be prone to subtle nuances that can lead to
measurement artifacts under differing sampling conditions. To circumvent these
potential problems, as well as the problem of short lamp lifetime, several new photolytic
converter designs have been evaluated. This paper will discuss the design criteria and
test results for these new converter designs.
DEVELOPMENT OF NO/NOX/NOY MONITORS FOR URBAN AIR SAMPLING
Thomas J. Kelly, Chester W. Spicer, and Gerald F. Ward
Battelle
505 King Avenue
Columbus, Ohio 43201-2693
The aim of this study was to devise a method for accurately speciated nitrogen
oxides measurements in urban monitoring networks. The study had two objectives: 1) to
achieve unambiguous separate measurement of NO, NOX (= NO + NO2), and NOy (= NO
+ NO2 + HNO3 + HNO2 + PAN + organic nitrates ...) by use of appropriate converters,
and 2) to implement those measurements with a monitor capable of a detection limit of
0.1 ppbv or less. Two different instrument packages were developed and tested relative
to these objectives. Package A consisted of a Thermo Environmental Model 425 high
sensitivity commercial monitor with heated molybdenum converter for NO and NO ,
coupled with a Scintrex LMA-3 luminol instrument for NO2 detection. Package B
consisted of an LMA-3 for NO2 detection, with a CrO3 converter for NO . The two
instrument packages were tested for sensitivity, linearity, conversion efficiency, and
interferences, using NO, NO2, PAN, nitric acid, nitrous acid, nitrobenzene, acetonitrile,
and isonpropyl nitrate, at concentrations from 10 to 80 ppbv. Both packages can meet the
0.1 ppbv detection limit objective, and linearity and conversion efficiencies were found to
be good. Interferences were found in the luminol NO2 and NOX measurements, primarily
from PAN. Nitro- compounds (e.g., nitrobenzene) were found not to be converted to NO
by the heated molybdenum converter, implying that such compounds are not included in
32
-------
most previous NOy data. The advantages and disadvantages of the two approaches, and
recommendations for further development, will be discussed.
A COMMERCIAL APPROACH TO NOy MEASUREMENT
Dieter Kita
Thermo Environmental Instruments, Inc.
Franklin, Massachusetts 02038
The accurate measurement of total reactive nitrogen compounds, commonly termed
NOy, is considered vital to a better understanding of tropospheric oxidation chemistry,
acid transport and deposition modeling, and as critical component of overall atmospheric
stability and chemistry. Virtually all measurements have relied upon the exclusive use
and operation of instrumentation developed for NO sensitivity and selectivity by the
research community. This paper describes the potential for commercial instrumentation
to provide, either directly or with minimal modification, accurate NO concentration
data.
First, a description of high sensitivity NO chemiluminescence instrumentation is
given. This includes plumbing configuration, modifications made to standard compliance
monitoring analyzers for improved sensitivity, selectivity, and stability, and both general
and specific calibration and operation procedures.
Second, the selection of commercially available NO converters is presented.
Individual advantages and disadvantages are discussed.
Third, modifications to the currently available high sensitivity commercial
instrumentation in order to obtain more reliable NO and NO data are suggested. The
modifications emphasize cost effectiveness, simplicity, and user operation.
Finally, a discussion of field measurements taken with suitably modified commercial
instrumentation is given. Difficulties in operation, instrument reliability, and data
interpretation are discussed.
TEMPORAL VARIATION OF NITROGEN OXIDE FLUXES FROM
AGRICULTURAL SOILS IN PIEDMONT NORTH CAROLINA
Benny D. Holbrook, Viney P. Aneja, and Wayne P. Robarge
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh, NC 27695-8208
Reactive gaseous oxides of nitrogen (NOX=NO+NO2) participate in the production
and/or consumption of atmospheric oxidants. Agricultural biogenic sources may be equal
in magnitude to those from known anthropogenic sources. Soil flux measurements of
NOX and NO were made during August-September 1993 and January-February 1994 at
an agriculturally-managed location in the central piedmont of North Carolina to assess
the variability of these fluxes. The soil type is representative of nearly a third of NC.
Samples were taken from 6AM to 6PM for about one month time period in each season.
33
-------
Additionally, samples were taken over at least one 24-36 hour period during each of these
sampling periods to obtain a diurnal profile. These measurements were accomplished
using a well-stirred dynamic chamber technique. The chamber used in this study
(diameter - 27 cm, height ~ 42 cm, volume ~ 25 liter, all internal surfaces made of
Teflon) was held in place over the soil sample by a stainless steel frame driven into the
ground to a depth of ~ 10 cm. Ambient air was pumped through the chamber at a 9 liter
per minute flow rate to reach steady-stage (- 30 minutes) and then collected in Teflon
bags (volume ~ 10 liter). The collection period was typically - 5 minutes. The samples
in the bags were then immediately analyzed for NO, NO2 and NO on-site. These
subsequent analyses of the nitrogen species concentrations were carried out using an
LMA-3 Luminol-based NO2 analyzer (Scintrex Ltd.) and a TECO 42S chemiluminescent
high sensitivity NO analyzer (Thermo Environmental Instruments Inc.). The instruments
were calibrated periodically using multiple dilution with a mixture of 0.019 ppmv NO in
N2 and a mixture of 0.131 ppmv of NO2 in N2. These fluxes will be correlated to soil/air
temperatures and soil physical/chemical parameters.
Thursday Afternoon
Meteorology
METEOROLOGICAL INSTRUMENTATION REQUIREMENTS FOR THE
PHOTOCHEMICAL ASSESSMENT MONITORING STATION
Gennaro H. Crescenti
U.S. EPA
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
The Photochemical Assessment Monitoring Station (PAMS) requires the
incorporation of surface and upper air meteorological instrumentation. The platform for
the surface instrumentation is generally a 10m tower. The variables collected include
wind speed, wind direction, air temperature, relative humidity, solar radiation and
barometric pressure. Upper air data may be acquired using a variety of methods.
Traditionally, profiles of wind speed, wind direction, air temperature, relative humidity
and barometric pressure have been acquired by using balloon based sonde packages,
either free-floating or tethered. These types of systems can be very labor intensive. As
an alternative, ground-based remoted profilers may be used to obtain profiles of wind
velocity and air temperature. A sodar utilizes acoustic pulses -1-5 KHz) of energy to
determine wind velocity while a radar uses electromagnetic (EM) pulses (-915 MHz).
A radio acoustic sounding system (RASS) uses both acoustic and EM pulses to construct
profiles of virtual air temperature.
This paper will discuss what types of meteorological sensors will satisfy PAMS
requirements. Frequently, resolution, and what types of upper air data acquisition will be
addressed.
34
-------
SODAR, RADAR PROFILER AM) RASS OPERATING PRINCIPLES AND
PAMS APPLICATIONS
Gary S. Zeigler and Charles E. Riese
Radian Corporation
P.O. Box 201088
Austin, TX 78720-1088
Meteorological remote sensing is most commonly thought of as weather satellites
taking the cloud pictures we see on television. Less widely known, but equally well
developed and routinely fielded in recent years, are ground-based systems that look
upward into the lower atmosphere to provide wind and temperature measurements in
vertical profile. All such "profilers" operate on the interaction of their transmitted pulse
with the atmosphere and provide measurements based upon interpretation of the signal
reflected back to the system's antenna Such systems include the "SODAR", using
strictly acoustic signals, the "RADAR" profilers", using electromagnetic transmissions,
and the "RASS", incorporating acoustic and electromagnetic interactions.
This paper provides an overview of SODAR, RADAR Profiler, and RASS
technology, with emphasis on operating principles and potential Photochemical
Assessment Monitoring Station (PAMS) applications. A description of the capabilities
and limitations of each system is included, based upon experience in its operational use
supporting ozone-related field measurement programs in recent years. The combined use
of the three types os systems for upper ar meteorological monitoring at the first PAMS
site, in New Jersey, is highlighted.
SITING GUIDANCE FOR BOUNDARY LAYER METEOROLOGICAL
PROFILERS
John E. Gaynor
NOAA/Environmental Technology Laboratory
325 Broadway
Boulder, Colorado 80303
The recent advent of the operational use of ground-based remote meteorological
profiling necessitates an evaluation of siting criteria for these instruments. The
operational profilers are, in particular, sodars, 915 MHz radar wind profilers and radio
acoustic sounding systems (RASS). Siting involves two scales, local and regional. Local
siting criteria is somewhat different for each one of the remote sensors listed. However,
ground clutter is a problem common for each. At its worst, ground clutter can invalidate
several of the lowest gates of each instrument. Ground clutter does not only depend on
the characteristics of the site, but also on the beam-pattern characteristics of the acoustic
antennas (sodar and RASS) or radar antennas (radar wind profiler and RASS). The beam
pattern depends on both the antenna itself and the acoustic absorber surrounding the
antenna (sodar) or the clutter fence (radar). Given these and other considerations,
guidelines for local siting are presented.
35
-------
For regional siting, consideration is given to the difficult problem of spatial
representativeness. This is of particular importance in the atmospheric boundary layer in
complex terrain regions for which spatial variability of the winds and temperature can be
most pronounced. A spatial correlation technique is offered as a method quantitatively
determining, prior to instrument placement, the spatial representativeness. This
suggested method applies to most types of measurements above the surface layer.
QUALITY ASSURANCE FOR PAMS UPPER AIR MONITORING SITES
Brian Templeman
U.S. EPA
Research Triangle Park, NC 27511
Surface and upper air meteorology play a vital role in the formation and transport of
O3. Consequently, meteorology has an impact on population exposure to O3. In order to
support monitoring objectives associated with model inputs and performance evaluations,
meteorological monitoring is required for each PAMS. Surface meteorological
measurements should begin within the first year of network operation. Upper air
meteorological data for determining mixing heights should be collected for
approximately 10 to 20 key days per year corresponding to specific model input
requirements.
Ground level meteorological variables to be measured as part of enhanced O3
network monitoring are as follows: wind direction; wind speed; ambient temperature;
barometric pressure; relative humidity; and solar radiation. All of these meteorological
variables are to be monitored from a height of 10 meters above the ground level. Specific
guidance on siting and quality assurance may be found in the "On-Site Meteorological
Program Guidance for Regulatory Modeling Applications", EPA^50/4-87-013, and the
"Quality assurance handbook for Air Pollution Measurement Systems: Volume IV-
Meteorological Measurements:, EPA-600/4-90-0003.
The mixing height is the maximum depth of the atmosphere from the surface up to a
vertical height below which thorough mixing of pollutants can occur. Mixing height
estimates are required for PAMS. There are numerous strategies for measuring mking
heights, which are outlines in the PAMS Technical Assistance Document. During this
presentation we will discuss quality assurance and control requirements for upper air
monitoring systems.
The SODAR uses acoustic waves to measure vertical profiles of wind. This device
provides a direct method for estimating winds and mixing heights. SODARS are usually
configured to obtain the most reliable data set possible for a given field site.
Configuration of a profiler may include modification of the profiler output signal
frequency, out signal power, averaging intervals and averaging techniques. The overall
accuracy of an acquired data base is dependent, in part, on the surrounding terrain, nearby
buildings, atmospheric stability, noise sources, insects and birds. Therefore, when
compiling a set of specifications for the purchase of a remote sensing device, it is
important to determine site specific information that will aid the manufacturer in
configuring the device to fit the user's needs. The following sections describe site
36
-------
specific parameters which need to be identified and provide some initial estimates of
expected accuracy, precision and data capture efficiency.
COLLECTING AND INTERPRETING UPPER AIR METEOROLOGICAL
DATA FOR THE PAMS NETWORK USING RADAR PROFILERS AND RASS
Charles G. (Lin) Lindsey and Timothy S. Dyes
Sonoma Technology, Inc.
5510 Skylane Blvd., Suite 101
Santa Rosa, CA 95403-1083
Remote sensors like radar wind profilers equipped with Radio Acoustic Sounding
Systems (RASS) are likely candidates for collecting the upper air meteorological data
required for the PAMS network. Upper air winds and temperatures collected for PAMS
will be used to analyze and model meteorological processes that accompany periods of
high ozone concentrations; to initialize and evaluate the performance of air quality
models; and to support analyses of emission control strategies. Profilers offer several
advantages for collecting continuously and unmanned, providing improved temporal
resolution at lower cost; data are available in near-real time, simplifying quality control
(QC) activities; and profilers measure vertical velocity (w), which is an important
parameter for diagnosing and accurately modeling many meteorological processes.
Wind profilers measure wind speed, wind direction, and vertical velocity from
approximately 100 m agl to altitudes as high as 3-5 km with a vertical resolutions of
60-100 m; RASS measures temperature to altitudes of 1-2 km with the same vertical
resolution. Profilers also produce lower-level information that is proving extremely
useful for identifying and analyzing key atmospheric processes and features that
accompany periods of poor air quality, such as mixing depth and turbulence information.
Using a number of examples of the types of data provided by profilers, we describe uses
of profiler data in recent air quality studies and discuss issues related to data
management, quality control, and data interpretation.
37
-------
Friday
Data Assessment & Interpretation
PHOTOCHEMICAL ASSESSMENT MONITORING: PROGRAM OBJECTIVES
AND DATA USES
Nash O. Gerald and Barbara A.B. Parzygnat
U.S. EPA
Technical Support Division (MD-14)
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
In order to design the Photochemical Assessment Monitoring Stations (PAMS)
program requirements, EPA considered a wealth of program objectives; additionally, the
Agency recognized the vast potential for a myriad of uses for the data. The Agency
anticipates that the measurements will be valuable for verifying emission inventories and
corroborating area-wide emissions reductions. The data is expected to be used to
evaluate, adjust, and provide input to the photochemical grid models utilized by the
States to develop and demonstrate the success of their control strategies. The PAMS will
provide constructive information for the evaluation of population exposure and the
development of ambient ozone and ozone precursor trends.
This paper will examine the development of the PAMS program objectives and the
potential role of the extensive PAMS data base for resolving ozone nonattainment.
RECEPTOR MODELING OF VOC DATA
Charles W. Lewis and Teri L. Conner
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Ronald C. Henry and John F. Collins
Civil Engineering Department
University of Southern California
Los Angeles, CA 90089
Receptor modeling refers to a set of procedures for identifying and quantifying the
sources of ambient air pollution impacting a monitoring site (receptor) primarily on the
basis of chemical species concentration measurements at the receptor. In its purest form
receptor modeling requires neither emissions inventory information nor meteorological
data for its implementation. The very large quantities of volatile organic compound
(VOC) ambient data that are beginning to be generated in the Photochemical Assessment
Monitoring Stations (PAMS) network offer an unusual opportunity for receptor modeling
applications. We will discuss primarily the Chemical Mass Balance (CMB) method of
receptor modeling, illustrating it with recent results from analysis of the EPA 1990
38
-------
Atlanta Ozone Precursor Study. The design of that study has many similarities (species
measured, number of stations, etc.) to what is being implemented in PAMS, and thus
provides a first assessment of the receptor modeling possibilities with PAMS data.
MOTOR VEHICLE VOLATILE HYDROCARBON SOURCE PROFILES FOR
CHEMICAL MASS BALANCE RECEPTOR MODELING
Teri L. Conner, William A. Lonneman, and Robert L. Seila
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
The Chemical Mass Balance (CMB) receptor model can be used to estimate source
contributions of volatile hydrocarbons. The model requires chemical profiles of sources
as well as a ambient data as input. Researchers often rely on "off-the-shelf" profiles
which are not necessarily representative of the airshed they are studying. A method for
obtaining hydrocarbon profiles of motor vehicle and related emissions for the airshed of
interest is described. This work was conducted as part of the 1990 "Atlanta Ozone
Precursor Monitoring Study". Motor vehicle emissions were sampled in canisters beside
a roadway in a tunnel-like underpass during the morning rush-hour. Three octane grades
of gasoline were purchased from 6 major vendors in Atlanta. Canister samples were
prepared using these fuels to approximate the whole gasoline and gasoline vapor
composition of the fuels in use during the study. All samples were analyzed by GC/FID
for their hydrocarbon content. Profiles were developed from these samples to represent
the hydrocarbon composition of emissions from a roadway, composite headspace
gasoline and composite whole gasoline. The roadway profile is compared with similar
profiles in the literature and recommendations are made regarding its use in the CMB
model. The roadway and fuel profiles are discussed in the context of the MOBILES
model outputs. The measured headspace gasoline vapor profile is compared with a
headspace gasoline vapor profile calculated from the measured whole gasoline profile by
means of Raoult's Law.
PAMS DATA USES FOR MODELING AND CONTROL STRATEGY
DEVELOPMENT/ASSESSMENT
Richard D. Scheffe
Office of Air Quality and Standards
U.S. EPA
Research Triangle Park, NC 27711
Data from the Photochemical Assessment Measurement Stations (PAMS) are
intended to serve multiple objectives, including photochemical modeling support,
developing and assessing success of control programs, and tracking air quality and
emission trends. This paper discusses various aspects of potential uses of PAMS data in
developing ozone precursor control strategies.
39
-------
PAMS data will support model applications by providing boundary conditions to
drive the applications and an observational base to diagnose model behavior and evaluate
performance. Within this context, the PAMS data are utilized implicitly, via model
support, in the control strategy development process. PAMS data will be interpreted more
explicitly given the current emphasis on the use of observational data as an independent
means for developing directionally accurate precursor control strategies. This paper will
discuss the relative strengths and gaps in the PAMS program toward supporting
development of sound precursor control strategies.
RESULTS OF DIURNAL NON-METHANE HYDROCARBON MONITORING
AT FRESNO, CALIFORNIA
Michael Poore, Michelle Dunlop, and Jacquelyn Milliron
The California Air Resources Board has been performing non-methane hydrocarbon
(NMHC) monitoring in California since 1988. The program has consisted of canister
samples collected in the morning (maximum emissions) and in the afternoon (maximum
photochemical activity), with the subsequent analysis for total and speciated NMHC. The
results of this program have indicated that, although the total NMHC concentrations may
vary considerably, the speciated NMHC patterns are relatively stable, both temporally
and geographically. As a result of the recent Photochemical Assessment Monitoring
Stations Regulation, the need for around-the-clock monitoring has been identified. To
meet this need, the 1993 summer monitoring program in Fresno, California, was
expanded to a four three-hour sampling schedule, with the sampling periods spaced
evenly throughout the 24-hour period. The data resulting from this program are presented
and compared with the data from previous years.
PRELIMINARY PAMS DATA ANALYSES
Terence Fitz-Simons and James Hemby
Office of Air Quality Planning and Standards
U.S. EPA
Research Triangle Park, NC 27711
The Photochemical Assessment Monitoring Stations (PAMS) program is designed to
provide enhanced monitoring of ozone and its precursors in twenty-two areas throughout
the United States (serious, severe, and extreme ozone non-attainment areas). The PAMS
program will generate a vast data set valuable to air quality modelers, developers of
emissions inventories, regulators, and policy makers. This paper focuses on empirical
statistical analyses that could be used to analyze and interpret this data. The analyses
follow the PAMS objectives for the data: determination of NAAQS attainment status,
control strategy development, SIP control and strategy evaluation, emissions tracking,
and ambient trends. The structure of the data, present availability, and status will be
presented along with example analyses.
40
-------
Panel Discussion
PHOTOCHEMICAL ASSESSMENT MONITORING IMPLEMENTATION
ISSUES
This session is organized as a panel discussion by representatives of the
Environmental Protection Agency's Headquarters and Regions, various State air
pollution control agencies, and representatives of State consortiums who plan, provide
policy for, and implement the Photochemical Assessment Monitoring Stations (PAMS)
rules. The panel is expected to address current issues regarding problems and successes
with instrumentation, siting and network design anomalies, new advances in data
handling and reporting, and funding questions, etc. Within time constraints, topics may
be suggested by the conference attendees.
41
-------
94
Session 6: Source Sampling
Tuesday Afternoon
SAMPLING AND ANALYSIS INFORMATION AIDS FOR STATIONARY
SOURCE PERSONNEL
Merrill D. Jackson and Larry D. Johnson
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
The Environmental Protection Agency in developing and evaluating sampling and
analysis methodology for stationary sources, has compiled information on the sampling
and analytical methods available and when they are applicable. Information also has been
summarized on the use of the mass spectrometer as the analytical finish. This information
has been put into two informational documents. The first is the "Stationary Source
Sampling and Analysis Directory, Version 2" (SSSADIR) and the second is the
"Handbook of GC/MS Data and Information for Selected Clean Air Act Compounds"
(Handbook). The SSSADIR has information on which methods to use for related organic
compounds and the status of the methods validation. The Handbook provides information
on the mass spectra, primary mass ions, relative retention times and compatibility of the
organic compounds in mixtures. The information in this document has been partially
funded by the United States Environmental Protection Agency under contract 68-D1-0010
to Radian Corporation and contract 68-02-4442 to Southern Research 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.
HEXAVALENT CHROMIUM EMISSIONS FROM AEROSPACE
OPERATIONS — A CASE STUDY
Ashok Chaurushia and Charles Bajza
Northrop Aircraft Division
One Northrop Avenue, M/S 5785/18
Hawthorne, CA 90250
Northrop Aircraft Division (NAD) complies with several air toxic regulations such
as EPA SARA Title HI, California Assembly Bill 2588, and Proposition 65, and is a
voluntary participant in air toxic emissions reduction programs such as the EPA 33/50
Program. In reporting emissions, NAD initially followed regulatory guidelines which
recommend that emission inventories of air toxics be based on engineering assumptions
and conservative emission factors in the absence of specific source test data. NAD was
concerned that Chromium VI emissions from NAD's spray coating and chemical tank
line operations were over reported due to these techniques. More recently, NAD has
42
-------
relied upon more accurate information from its ongoing source test program to determine
actual emission rates of Chromium VI. Based on these source test results, NAD revised
emission calculations and conducted air dispersion modeling to assess the potential long
term health risk to the surrounding community. NAD has been successful in
demonstrating the significant difference between the actual emissions calculated utilizing
the source test results and the emissions based on the traditional mass balance methods
using agency suggested data.
The paper will discuss the Chromium VI emission reduction process at NAD based
on the improved emission information and the application of control technologies on
major sources.
THE USE OF CANISTERS/GC-MS AND A PORTABLE GAS
CHROMATOQRAPH TO CHARACTERIZE EMISSIONS FROM AN AIR
STRIPPER
Chris Figueroa and Jon Bennett
Washington State Dept. of Ecology Air Quality Program
P.O. Box 47600
Lacey,WA 98504-7600
The State Washington's Toxic Cleanup Program is conducting one of its largest
cleanup operations of chromium-contaminated groundwater that threatens the City of
Vancouver's drinking water aquifer. The chromium plume originated from an
electroplating operation. Water analysis revealed volatile organic compounds (VOCs),
originating from a neighboring specialty gas facility, at levels which the city's
wastewater treatment plant would not accept. It became necessary to use an air stripper to
remove the VOCs present in the chromium-treated stream.
An air permit was needed to operate the air stripper. A carbon adsorption system was
determined to be best available control technology. The Toxics Cleanup Program
requested assistance from the Air Quality Program in gathering process air data from the
air stripper and carbon adsorption system to obtain that permit.
Trichloroethylene (TCE) was present in water samples at elevated concentrations
(160-180 parts per billion). TCE, a carcinogen, has one of the lowest acceptable source
impact levels (ASIL) in air (130 parts per trillion), therefore it was flagged as the main
VOC of concern from this site.
The Air Quality Program gathered TCE and VOC data through GC/MS analysis of
whole air samples obtained in evacuated Summa passivated canisters, and with a portable
gas chromatograph. This paper contains a comparison of the data obtained from both
methodologies, and of the costs involved.
43
-------
IMPROVEMENTS IN PREPARATION OF SAMPLES GENERATED BY SW-846
METHOD 0010
Merrill D. Jackson and Larry D. Johnson
Methods Research and Development Division
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
James F. McGaughey, Denny E. Wagoner, Joan T. Bursey and Raymond
G. Merrill
Radian Corporation
P.O. Box 13000
Research Triangle Park, NC 27709
A field evaluation study for SW-846 Method 0010 was conducted at a stationary
source emission site with a high moisture content. The recovery of the dynamically
spiked analytes and the spikes added before the laboratory preparation of the XAD-2®
resin were low. The sampling train media had been prepared according to Method 0010
protocol, so the laboratory procedures were examined in detail for sources of error. The
XAD-2, wet because of the source matrix, was difficult to fully remove from the glass
trap using only methylene chloride. Because Method 0010 does not specify the solvent to
use for trap rinsing, and since the probe was washed in the field with a 50:50 mixture of
methylene chloride and methanol, this mixture was used to transfer the XAD-2 from the
trap to the Soxhlet extractor. The low recoveries were attributed to the presence of the
methanol in the final extract. (The methanol is removed from the probe rinses during the
cleanup process.) A proposed method will be presented that permits the physical removal
of wet XAD-2 from the sampling trap without the use of methanol and an alternate
procedure is provided for removal of methanol from the trap rinse before sample extract
concentration.
The information in this document has been funded wholly by the United States
Environmental Protection Agency under contract 68-D1-0010 to Radian Corporation. 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.
44
-------
EVALUATION OF GC DETECTORS FOR TOTAL GASEOUS NONMETHANE
ORGANIC COMPOUNDS
Stephanie B. Phllipp, Dave-Paul Dayton, and Raymond G. Merrill
Radian Corporation
P.O. Box 13000
Research Triangle Park, NC 27709
Merrill D. Jackson
Methods Research and Development Division
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
The development of an emissions monitoring prototype instrument that could
provide continuous or semi-continuous quantitative measurement of total gaseous
nonmethane organic carbon (TGNMOC) emissions from stationary sources, would allow
better control of compounds under the Clean Air Act Amendments of 1990, Title in. This
system would have a GC with a column that could separate methane from the other
organic compounds. To meet this goal a search was initiated to identify detection systems
for TGNMOC that are both simple to use and accurate. The measurement of oxygenated
compounds were of particular interest since emissions of interest may have a large
proportion of them. Work towards this objective was accomplished by conducting a
search of detector manufacturers' literature, and talking with the manufacturers' technical
personnel. Several systems that were marketed as appropriate for TGNMOC
measurement were identified and evaluated in the laboratory. The detectors evaluated
were a Catalyzed Flame lonization Detector and a Thermionic lonization Detector, both
from DETector and Engineering Technology, Inc. and an Oxygen-Flame lonization
Detector and an Elemental Analyzer, both from Fisons Corporation.
None of the detectors evaluated during the course of this project met the primary
performance goal of measuring organic carbon uniformly, regardless of the chemical
structure of compounds in the sample. That primary goal required that the detector yield
equal response for all organic compounds in a mixture including oxygenated compounds
based on carbon number. While some success was realized for many classes of organic
compounds, oxygenated compounds presented a challenge that none of the detectors
could master, that of measuring them as organic carbon when compared to a hydrocarbon
standard.
The information in this document has been funded wholly by the United States
Environmental Protection Agency under contract 68-D1-0010 to Radian Corporation. 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.
45
-------
SAMPLING OF VOLATILE ORGANIC COMPOUNDS FROM COMBUSTION
SOURCES USING TEDLAR® BAGS AND ANALYSIS USING GC/MS
Rohini Kanniganti, Richard L. Moreno, Joan T. Bursey, and Raymond G.
Merrill
Radian Corporation
P.O. Box 13000
Research Triangle Park, NC 27709
Robert G. Fuerst and Larry D. Johnson
Atmospheric Research and Exposure Assessment Laboratory
Methods Research and Development Division
Source Methods Research Branch
U.S. EPA
Research Triangle Park, NC 27711
Tedlar® bags are best suited for sampling volatile organic compounds at
concentrations above approximately Ippm in stack gas. The purpose of draft Method
0040 is to establish standardized test conditions and sample handling procedures for the
collection and analysis of volatile organic compounds from gaseous effluent sources such
as hazardous waste incinerators. Compounds for which this method is applicable must be
volatile organic compounds with boiling points less than 121° C. Applicable compounds
must exhibit a concentration loss in a Tedlar bag of less than 20% over a 72-hour storage
time. This method uses either constant or proportional sampling depending upon the
variability in the emission flow rate. Gas chromatography/mass spectrometry (GC/MS) is
the analytical method of choice.
To develop an accurate and reproducible analytical method, a fixed loop interface to
the GC/MS was designed. Sample was introduced from a Tedlar bag onto the loop and
then loaded onto a GC column by means of a pneumatically controlled valve. Calibration
and instrument detection limits were determined for selected analytes. Instrument
detection limits for vinyl chloride and benzene were 0.2 and 0.4 ppm, respectively.
Accurate and reproducible preparation of gaseous standards in Tedlar bags was
demonstrated. Accurate and reproducible dynamic spiking into the sampling train was
demonstrated by spiking in replicate using two different levels.
The field sampling and analytical protocols were both thoroughly reviewed. The
method is now ready for validation and testing in the field.
The information in this document was funded wholly by the United States
Environmental Protection Agency under Contract No. 68-D1-0010 to Radian
Corporation. It was subjected to Agency review and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or
recommendation for use.
46
-------
FIELD EVALUATION OF A MODIFIED VOST SAMPLING METHOD
Merrill D. Jackson, Larry D. Johnson, and Robert G. Fuerst
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
James F. McGaughey, Joan T. Bursey, and Raymond G. Merrill, Jr.
Radian Corporation
P.O. Box 13000
Research Triangle Park, NC 27709
The VOST (SW-846 Method 0030) specifies the use of Tenax® and a particular
petroleum-based charcoal (SKC Lot 104, or its equivalent). This particular lot of
charcoal has long been depleted. In field evaluation studies of the VOST methodology, a
replacement petroleum-based charcoal has been used. Since SKC Lot 104 charcoal is no
longer commercially available, candidate replacement sorbents for charcoal were studied,
and Anasorb® 747, a carbon-based sorbent, was selected as a sorbent for field testing.
The sampling train was modified to use only Anasorb in the back tube and the use of two
Tenax front tubes to avoid analytical difficulties associated with the analysis of the
mixed-bed back tube used in the standard VOST train. The Method 0030 VOST train was
modified to incorporate this three-sorbent-tube configuration.
The standard (SW-846 Method 0030) and the modified VOST methods were
evaluated at a chemical manufacturing facility using a quadruple probe system with
quadruple trains. In this field test, known concentrations of the halogenated volatile
organic compounds from the Clean Air Act Amendments 1990, Tide in were introduced
into the VOST train and the modified VOST train, using the same certified gas cylinder
as a source of test compounds. Statistical tests of comparability of methods were
performed on a compound-by-compound basis, and for most compounds, the VOST and
modified VOST methods are statistically equivalent.
The information in this document has been funded wholly by the United States
Environmental Protection Agency under contract 68-D1-0010 to Radian Corporation. 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.
COMPARISON OF SAMPLING AND ANALYTICAL METHODS FOR THE
COLLECTION AND DETERMINATION OF METHYLENE DIPHENYL
DHSOCYANATE (MDI) FROM ORIENTED STRAND BOARD (OSB) SOURCES
MarkD. Baker, William J. Karoly, and Michael F.Adams
ICI Polyurethanes Group
286 Mantua Grove Rd.
West Deptford, NJ 08066
The ICI Polyurethanes Group conducted a multi-variant study of sampling and
analytical methods for the collection and determination of MDI from 3 different OSB
47
-------
sources. The study evaluated 3 different sampling trains and derivatizing reagents at 2
probe temperatures as follows:
1. A 4-Nitro-N-propylbenzylamine (Nitro reagent) filter alone at ambient
probe temperature,
2. A l-(2-Methoxyphenyl) piperazine (12MP) filter followed by two
impingers containing the same reagent in Toluene at ambient probe
temperature,
3. A dual 12MP impinger train followed by a 12MP treated filter also at
ambient probe temperatures,
4. A dual 12MP impinger train followed by a 12MP treated filter also at a
probe temperature of 250 ° F,
5. A dual l-(2-pyridyl)piperazine (12PP) impinger train also at a probe
temperature of 250 ° F.
The analytical portion of the study compared results using Normal and Reversed
Phase HPLC determination of the Nitro reagent samples and Reversed Phase HPLC
determination of the 12MP and 12PP samples. Additional LC-MS studies were
conducted to confirm and identify the cause of a positive bias in the Normal Phase Nitro
reagent analytical results.
The analytical and sampling results will be presented.
PEN-BASED COMPUTER SYSTEM FOR PERFORMING SOURCE TEST
CALCULATIONS
FrankR. Clay
Emission Measurement Branch
Technical Support Division, OAQPS
U.S. EPA
Research Triangle Park, NC 27711
The Emission Measurement Branch has developed open-based computer system for
Methods 1 through 5 (40 CFR 60, appendix A) which is designed to perform on-site
emission source test calculations and for data reduction after the test effort has concluded.
The system uses EXCEL for Windows and is suitable for use by anyone who performs
emission tests, observes emission tests, or who needs to reduce or verify emission testing
data. The system is designed for people who have minimal computer skills, and data can
be handwritten upon a computer screen using an electronic stylus or "pen"; keyboard and
"spin button" data entry options are also available. When sufficient data have been
entered, the system performs the calculations and displays the results. Future plans are to
adapt the pen-based system to as many Environmental Protection Agency methods as
applicable; and the programs will be available on the Emission Measurement Technical
Information Center Computer Bulletin Board as well as upon request.
48
-------
Tuesday, May 3,1994
Session 7: Global Climate Change, Mount Mitchell
Tuesday Afternoon
MONITORING OF REGIONAL CHEMICAL CLIMATE CHANGE AT MOUNT
MITCHELL, NORTH CAROLINA
V.K. Saxena
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh, NC 27695-8208
One of the objectives of the National Global Change Program is to seek plausible
answers to the following question: What is the radiative forcing from greenhouse gases,
aerosols, and clouds on the Earth's climate system, and what are the feedbacks and
processes that regulate the net radiative balance of the earth's atmosphere? A critical
survey of the existing evidence reveals that the trends in the average global temperature
of the Earth-Troposphere system and in global precipitation patterns are mitigated by
several natural and anthropogenic differences between the northern versus southern
hemisphere. In contrast, climatic changes on a regional scale have been more convincing
during the last two decades. A case in point is the frequency of droughts experienced in
the Southeast. The objectives of our study are addressed to understand the cloud-climate
feedback mechanisms on a regional scale, with particular emphasis on the climate of the
Southeastern U.S. We aim to study the impact of the natural and anthropogenic aerosols
on the regional cloud albedo. This is done by observing the microphysico-chemical
characteristics of clouds that form at Mt. Mitchell, NC (highest peak, 6,684 ft or 2,017 m
MSL in the eastern U.S.; Mt. Mitchell State Park is a designated United Nations
Biosphere Reserve). The cloud reflectivity is simultaneously monitored by the satellite-
based Advanced Very High Resolution Radiometer (AVHRR). Clouds with contrasting
microphysical and radiative characteristics are formed at the site when air masses of
marine, continental, or highly polluted origins arrive. Analysis of the First Year Data
Base suggests that our project offers some very promising insights into the cloud-climate
feedback mechanisms. Clouds formed by polluted air masses had pH as low as 2.4 and by
marine air masses as high as 4.75. The latter were found least reflective.
49
-------
AIR MASS HISTORY VERSUS CLOUD WATER ACIDITY: OBSERVATIONS
AND MODEL RESULTS FROM A REMOTE RURAL SITE
J.C. Ulman, V.K. Saxena, K.L. Burns, andJ.D. Grovenstein
Department of Marine, Earth and Atmospheric Science
North Carolina State University
Raleigh, NC 27695-8208
Cloud water acidity, as measured at a remote rural observing site, is related to the
pathway, or trajectory, taken by air parcels passing over areas with varying levels of
pollutant emissions. Areas of the country that are high emitters of pollutants, such as SOX
and NOX, will affect cloud water acidity by reducing it, if the clouds passing over these
regions are intercepted at the observing site. Field studies at Mount Mitchell State Park
(35° 44'05"N, 82° 17' 15"W) during June and August of 1993 have resulted in
observations of cloud water acidity for 37 cloud events of varying lengths, by taking
measurements with a passive cloud water collector mounted atop a 16.5 meter
meteorological tower. Calculations of the backward trajectories for these cloud events are
accomplished by utilizing a hybrid Eulerian-Lagrangian computer model called Hybrid
Single-Particle Lagrangian Integrated Trajectories, or HY-SPLIT. In this study,
comparisons of cloud events from the 1993 field season and cloud events from a field
study undertaken under the auspices of the Mountain Cloud Chemistry Project (MCCP),
during 1986-1987, will be made by investigating the inter-relationships of a number of
meteorological criteria Ultimately, long-term changes in the acidity characteristics of
clouds impacting Mount Mitchell should be observable with the large datasets now
available for analysis.
THE EFFECT OF ANTHROPOGENIC POLLUTION ON CLOUD
MICROSTRUCTURE, CHEMICAL COMPOSITION AND ALBEDO
K.L. Burns and V.K. Saxena
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh, NC 27695-8208
In situ cloud measurements were taken during 39 individual cloud events between
June and October 1993, in Mount Mitchell State Park, North Carolina Cloud droplet
size spectra were obtained and used to determine total droplet concentrations, average
droplet radii and cloud liquid water contents. Cloud water samples were collected and
analyzed for pH and chemical composition. Meteorological data was recorded and used
to verify air mass histories through back trajectory analysis. A total of 136 hourly cases
were available with coincident droplet size spectra, pH measurements, chemical analysis
and meteorological data
A strong positive correlation (coefficient = +0.69) was detected between pH and
average droplet radius. Also, there was a strong negative correlation (coefficient = -0.60)
between pH and total droplet concentration. The data was then sorted into three
populations based on pH: pH < 3.1 (n = 33), 3.1 < pH < 3.7(n = 71), and
50
-------
pH > 3.7 (n = 32). It was observed that low pH values were associated, on average, with
higher number concentrations and lower average radii, while high pH values were
associated, on average, with lower number concentrations and higher average radii.
Cloud albedos were calculated for each population and shown to vary with pH.
AEROSOLS AS A NATURAL TRACER OF AIR MASSES
C.K. Deininger and V.K. Saxena
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh NC 27695-8208
The physical and chemical characteristics of clean marine, continental, and highly
polluted air masses were compared using aerosol and cloud water data collected during
field studies at a site located in Mt. Mitchell (2038 m MSA -35° 44'05"N, 32° 17'15"W)
State Park, NC. The site is particularly advantageous for studying the effects of air
pollution, since it intercepts air masses arriving from the East (marine), the West
(continental), and the Northwest (polluted). Particles collected during cloud free periods
were measured using a field emission microscope. Analysis of the size distribution and
mass concentration of aerosols present in different air masses shows that they are good
indicators of the history of the air mass. Principal component analysis (PCA) was used to
study the relationship between the ionic constituents of the cloud water collected and the
type of air mass which formed the cloud. By applying PCA to the cloud water chemistry,
a set of highly intercorrelated variables was replaced with a set of uncorrelated principal
components. PCA seems to be most effective for studying highly polluted air masses,
where the variability among the ions is the greatest The dominant source of variance in
continental air is the contrast between marine ions, Na and Cl, and other ions present. For
clean marine air variance among the different ionic constituents is minor. History of the
air masses was confirmed by using back trajectory analysis.
GREENHOUSE WARMING, ACIDIC DEPOSITION, AND THE DILEMMA OF
CLIMATE CHANGE
JD. Grovenstein and V.K. Saxena
North Carolina State University
Department of Marine, Earth and Atmospheric Sciences
Raleigh, NC 27695-8208
Tropospheric aerosols produce climatic perturbations through direct and indirect
contributions to radiative forcing. Sulfate aerosols produce a cooling effect by elevating
concentrations of cloud condensation nuclei (CCN) which enhance cloud-mediated
albedo. Results from an ongoing field study at Mt. Mitchell, NC (35° 45'5" N, 82°
17' 15" W, 2038 m or 6,684 ft MSL) are presented. In response to the serious
environmental problems due to acidic deposition, the Clean Air Act was amended
mandating the reduction in fossil fuel emissions. Diagnostic modeling was used to
51
-------
simulate the effect of an emissions control policy on greenhouse warming. Three policies,
suggested by the Intergovernmental Panel on Climate Change (IPCC), were used. They
are: growth of emissions by 2% per year, a reduction of 2% per year, and a "business as
usual" policy of 0% per year. A 2% per year reduction in fossil fuel emissions indicates a
30% reduction in sulfate-derived cooling (as compared to the present) in the next 50
years. Two percent growth per year for 50 years will cause a 40% increase in sulfate-
derived cooling. It is stressed that the impact of the enhanced CO2 concentration on the
temperature of the earth-troposphere system cannot be evaluated unless the impact of
SO2 on the same system is evaluated.
SAGE H BASED COLUMN SURFACE AREA MEASUREMENTS OF THE MT.
PINATUBO AEROSOL OVER THE EASTERN UNITED STATES
John Anderson and V.K. Saxena
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh, NC 27695-8208
The role of ozone (O3) in the atmosphere takes on 2 different identities. In the
troposphere, the presence of ozone is considered pollution and therefore a hindrance
whereas in the stratosphere, the presence of ozone is needed in that ozone absorbs the
sun's harmful ultraviolet (UV) radiation that can cause skin cancer and agricultural
damage. Numerous studies have shown that volcanic aerosols may possess the effective
surface area needed for heterogeneous reactions to occur that frees active chlorine
products to destroy ozone in the stratosphere in northern mid-latitudes. Since the increase
in anthropogenic chlorine is estimated to be about 40% in the last 12 years, the eruption
of Mt. Pinatubo (15.1° N, 120.4° E) during June 12-14, 1991 has the potential to possess
the surface area needed for the destruction of ozone on a very large scale since it is the
largest volcanic event recorded in recent history. This study focuses on the temporal
variations of the resulting surface areas of the Mt. Pinatubo aerosol in the eastern half of
the United States since the eruption. The Pinatubo aerosol characteristics in a unit column
between 15-30 km such as surface area S(|j.m2cm"2), number density N(cnT2), mass
loading M(|j.gm~2), and effective radius ReQjm) are inferred from the Stratospheric
Aerosol and Gas Experiment (SAGE) II measurements using a Randomized
Minimization Search Technique (RMST) in the radii range of 0.1|om to 0.8|jm in 0.1 mm
increments. Preliminary results for 1991 indicate that the surface areas observed were
between 1-20 |jm2cm"3 at individual layers between 18-26 km with total column
abundances of up to 140 |im2cnr2 in November 1991. The Pinatubo surface areas were
within the range of 1-50 |U.m2cnr3 found in the El Chichon aerosol months after the
eruption which also coincided with record lows of ozone in 1983.
52
-------
Tuesday, May 3,1994
SessionS: Measurement Methods
Tuesday Afternoon
A STUDY OF INTERFERENCES IN OZONE UV AND
CHEMILUMEMESCENCE MONITORS
E.E. Hudgens and T.E. Kleindienst
ManTech Environmental Technology, Inc.
P.O. Box 12313
Research Triangle Park, NC 27709
F.F. McElroy
U.S. EPA
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
W.M. Ollison
American Petroleum Institute
1220 L Street, NW
Washington, DC 20005
A study was conducted to examine measurement interferences in ultraviolet ozone
monitors. Previous results had shown that there was no direct interference with these
monitors due to the presence of water at non-condensing concentrations. The present
study continues this effort, examining both potential positive and negative effects of
moisture on other interferences for these monitors. Aromatic compounds and their
oxidation products could potentially show a positive interference, and test measurements
were made with aromatics such as toluene, benzaldehyde and nitrocresol to determine
their possible retention in the ozone scrubber and their absorption in the cell as a function
of the humidity. A detailed examination of the scrubbers used in these systems has also
been undertaken. Ozone scrubbers that have shown anomalous behavior in the field have
been studied hi various failure modes under controlled laboratory conditions. Conditions
under which new scrubbers fail were investigated, and procedures for determining
defective scrubbers were evaluated.
53
-------
REAL TIME ELECTROCHEMICAL MEASUREMENT OF OZONE IN THE
PRESENCE OF NITROGEN OXIDES
William R. Penrose and Li Pan
Transducer Research, Inc.
Naperville, IL 60540
WHIM. Ollison
American Petroleum Institute
Washington, DC 20005
Current regulatory models (USEPA/pNEM-03) estimate population ozone exposures
within multiple microenvironments averaged over clock hour time periods. Validation of
model algorithms requires measurement of personal ozone exposures of free-ranging
individuals over hourly time frames within changing microenvironmental air quality.
Ozone can be measured with electrochemical sensors at sensitivity limits of 5 ppb if the
sensor is periodically corrected for baseline drift. The electrochemical sensor responds
comparably to O3, NO2 and HONO. Initial attempts to measure ozone passively in the
presence of nitrogen oxides by periodic ozone-scrubbed baseline drift corrections were
unsuccessful due to the rapid changes in NOX levels within several microenvironments.
We have constructed a prototype active, battery-powered, data-logging personal monitor
using two sensors in tandem with an ozone removal filter in between. The difference in
the sensor signals is due to ozone; thus, corrections for NOX interferences are made
on-line in real time without sacrifice of sensitivity.
THE DEVELOPMENT OF AN ACTIVE PERSONAL OZONE SAMPLER USING
A DIFFUSION DENUDER
A.S. Geyh, J.M. Wolfson, and P. Koutrakis
Harvard University
School of Public Health
665 Huntington Avenue
Boston, Massachusetts 02115
J. Mulik
U.S. EPA
Research Triangle Park, NC 27711
Personal, microenvironmental and indoor ozone monitoring is currently carried out
using a passive sampling device which is both light-weight and inexpensive. However,
the collection properties of these samplers have been found to be sensitive to wind effects
and sampler placement, thus limiting their potential use for personal monitoring. In
addition, because of their relative insensitivity, these samplers cannot used for short-term
monitoring of ozone at low concentrations. In response to these problems we are
developing a light-weight active ozone sampler which uses a single tube diffusion
denuder for sample collection.
The new single tube diffusion denuder (STTD) consists of a 1.4 cm (ID) x 10 cm
etched Pyrex tube attached to a very small, low-flow, relatively low-cost personal pump.
54
-------
Tube diameter and length were chosen to maximize collection efficiency at a sampling
rate of 65 mL/min. The tube is coated with a nitrite reagent which has been successfully
used in the passive ozone samplers.
Variations in relative humidity, ozone concentration, and total ozone exposure have
relatively small affects on the accuracy and precision of the STTD. In addition, the low
limit of detection (LOD) of approximately 20 ppb-hrs gives a 10-fold increase in
sensitivity over the passive samplers for which a 200 ppb-hr LOD has been determined.
This new active sampler thus makes possible both short-term personal and
microenvironmental monitoring.
EVALUATION OF PASSIVE SAMPLERS FOR FIELD MEASUREMENTS OF
AMBIENT OZONE IN THE NATIONAL PARKS
John D. Ray and Miguel Flores
U.S. National Park Service
Denver, CO 80225-0287
The National Park Service has conducted field studies to evaluate passive samplers
as a means of getting integrated ozone exposures at remote locations. Initial tests were at
8 different national parks for exposure periods of 1 to 12 weeks. The results from longer
exposure times were unreliable and there was more variation in the ozone measured than
expected. In a follow-up study, a 23 factorial design experiment was used to verify
consistency between samplers and between analysis laboratories. Significant differences
in measured ozone were introduced by the polypropylene rain shields that had been used
in the first trials. PVC plastic rain shields were used subsequently. Repeatability and
agreement by two different laboratories that did the ion chromatographic analysis was
about 1%.
In the summer of 1993, five sites in two different parks were used to compare one
week passive sampler ozone measurements to average hourly exposures determined with
uv absorption method ozone analyzers. Repeated passive sampler measurements agreed
well for each site with relative standard deviations of <2%. However, ozone collection
rates for the samplers varied somewhat by site; the average was 23 ± 2 cm sec"1. Possible
factors effecting the collection rates, such as elevation, wind speed, and humidity, will be
discussed. A plan for how these samplers might be used in studies of tree exposures to
ozone will be presented.
55
-------
DEVELOPMENT OF A NEW SEMI-VOLATILE ORGANIC COMPOUND
SAMPLER
C. Sioutas and P. Koutrakis
Harvard University
School of Public Health
Boston, MA 02115
R.M. Burton
U.S. EPA
Research Triangle Park, NC 27711
A new sampler has been developed to sample semi-volatile organic compounds. The
sampler utilizes the principle of virtual impactor to efficiently separate the paniculate
from the gas phases of organic compounds. The virtual impactor consists of a slit-shaped
nozzle where the aerosol is accelerated, and another slit-shaped nozzle that collects the
paniculate phase of organics (plus a small and known fraction of the gas phase). The
acceleration slit is 0.023 cm wide, the collection slit is 0.035 cm wide, and both slits are
11 cm long.
The virtual impactor's 50% cutpoint has been determined experimentally to be
0.12 (j.m. In addition, interstage losses have been determined (in all configurations tested,
particle losses ranged from 5-15%). The impactor's sampling flow rate is 284
liters/minute, with a corresponding pressure drop of 100 inches H2O. Higher or lower
sampling flow rates can be achieved by increasing or decreasing the length of the slits.
Tests for volatilization losses have been conducted by generating organic aerosols of
known volatility, and comparing the impactor's collection to that of a filter pack
sampling in parallel. The experiments demonstrated negligible volatilization losses
(<5%) for the compounds tried.
Particles are collected on a filter connected to the minor flow of the impactor,
followed by a sorbent bed to collect material that volatilized from the particles. The
organic gas phases is collected on a sorbent bed, connected to the major flow of the
impactor.
OUTDOOR AIR NOX SPECIATION BY A SELECTIVE DENUDER
COLLECTION SYSTEM
Robert S. Braman and M. Stacey Thomson
Department of Chemistry
University of South Florida
Tampa, FL 33620
A series of interior coated hollow tubes has been studied for the collection and
analysis of ambient air for gas phase nitric acid, nitrous acid, nitrogen dioxide,
nitrosamines, and nitric oxide. The tube surface sequence consists of, in order, tungstic
oxide, potassium-iron oxide, copper iodide, carbon, and cobalt oxides. Collected analytes
are removed by heating the collection tubes with chemiluminescent detection for the
56
-------
analysis. Detection limits are in the fractional nanogram per sample range. Prior work
has verified high collection efficiency and specificity of the sequential collection.
Results of the surface analysis of the coatings will be presented. Current work
includes a study of interferences and application of the method. Possible interferences
from carbon dioxide, carbon monoxide, ozone, formaldehyde, nitrous oxide were studied.
Temperature and humidity effects were also studied. The analysis system has been
applied to analysis of outside air in a producing orange grove area at which some impact
from nitrogen oxides was suspected to have a serious impact. A contrast of indoor versus
outdoor air will also be presented.
FIELD EVALUATION OF A NEW GLASS HONEYCOMB DENUDER/FILTER
PACK SAMPLER
P. Koutrakis, C. Sioutas, andJM. Wolfson
Harvard University
School of Public Health
Boston, MA 02115
J.D. Mulik
U.S. EPA
Research Triangle Park, NC 27711
We have developed a compact diffusion denuder/filter pack sampler to sample
ambient gas and paniculate pollutants. The main feature of this system is the denuder
component, which has a large number of small hexagonal glass tubes sealed inside an
outer glass tube. The collection efficiency of the honeycomb denuder was compared to
those of the annular denuder for HNO3 and NH3 for experiments simulating atmospheric
conditions and were found identical. In addition, inlet losses of reactive gases in both
systems were found to be insignificant (<5%). The capacity of the honeycomb denuders
under simulated atmospheric conditions was found significantly higher than the capacity
of annular denuders for tests using HNO3 and NH3. Furthermore, laboratory and field
experiments indicated that fine particle losses (dp<2.5 (Am) were less than 5%.
In this study we conduct a field evaluation of the new honeycomb denuder/filter pack
sampler by comparing its performance to the performance of a previously developed
annular denuder/filter pack system. Colocated sampling was conducted using both
samplers, and the concentrations of nitric acid and ammonia, as well as the
concentrations of fine particle mass (particles smaller than 2.5 |jm in diameter), sulfate,
and nitrate obtained from both samplers were compared. The results of this comparison
will be presented.
57
-------
THE USE OF PM10 ANTON-CATION DIFFERENCE AS AN INDEX OF
HISTORICAL AEROSOL ACIDITY
George D. Thurston, J. Currie, D. He, andJE. Gorczynski, Jr.
Nelson Institute of Environmental Medicine
New York University Medical Center
Tuxedo, NY 10987
The composition of particles in the air may have a significant influence on the health
implications of inhaling these particles. Paniculate matter less than 10 \im in
aerodynamic diameter (PM10) has been routinely sampled throughout the U.S. and
elsewhere in the world for years, but these samples are usually only weighed for total
mass and summarily stored. However, these samples can be (and sometimes are)
analyzed for their ionic composition (i.e of sulfates, SO4=; nitrates, NO3-; and
ammonium, NH4+) via ion chromatography. Furthermore, it is hypothesized that the
imbalance of these easily measured ions (i.e. anions-cations) may provide a useful
estimate of the remaining unmeasured major particulate cation: aerosol strong acidity,
H+. Conventional methods for directly measuring H+ entail great care to protect the
collected acid from neutralization by ambient ammonia, basic particles, or alkaline filter
media, in order that all the H+ show up via a pH determination. We have conducted such
direct H+ measurements side-by-side with state-run PM10 samplers in Albany, NY and in
New York City, NY in order to test whether the PM10 ion difference estimates concur
with the directly measured H+. Despite the lack of neutralization protection of the PM10
samples, it was found that the ion difference method yields H+ highly correlated with,
and not significant different from, the directly measured H+. It is thought that the ion
difference is maintained on these samples by the rapid neutralization of ambient H+ by
large basic particles coexisting on the PM10 filters, as well as by the weakly alkaline
quartz filters themselves. This would preempt the conversion of particulate acids (e.g.,
H2SO4) to ammonium sulfates (e.g., [NH4]2SO4) by ammonia gas, thereby maintaining
the original sulfate-ammonium imbalance associated with the ambient H+. Archived
PM10 samples therefore represent a potentially valuable resource regarding the nature of
acid aerosol exposures throughout the U.S. and elsewhere.
PASSIVE SAMPLERS FOR AMBIENT OZONE, FORMALDEHYDE AND
SULFUR DIOXIDE: INDOOR, OUTDOOR AND PERSONAL EXPOSURE
APPLICATIONS
Daniel Grosjean and Eric Grosjean
DGA, Inc.
4526 Telephone Road, Suite 205
Ventura, CA 93003
Time-integrated measurements of air pollutants have many applications in the
context of regulations pertaining to indoor air quality, outdoor (ambient) monitoring, and
personal exposure assessment. For several years, the passive samplers developed at DGA
have been applied to cost-effective measurements of parts per billion levels of ozone,
58
-------
formaldehyde and sulfur dioxide. Examples of applications will be described. These
include (a) formaldehyde measurements in indoor settings including museums, public
buildings and personal exposure; (b) ozone measurements indoor (museums, cultural
heritage buildings) and outdoor (Class I Wilderness areas; air quality surveys in Europe,
Canada and Latin America, long-term monitoring of ozone exposure in forests) and (c)
surveys of ambient levels of sulfur dioxide in several eastern European countries.
59
-------
1994
Session 9: NC O3 State Implementation Plan,
Measuring and Modeling Study
Wednesday Morning
OVERVIEW OF THE NORTH CAROLINA UAM PROJECT
Brock M. Nicholson
NCDEHNR/DEM/Air Quality Section
15 N. West Street
Raleigh, NC 27603
In 1992 North Carolina committed to perform photochemical gridded dispersion
modeling to support the nonattainment state implementation plan (SIP) process. In
particular, UAM results would be used to demonstrate attainment by 1996 for the three
moderate nonattainment areas (Charlotte/Gastonia, Raleigh/Durham, and
Greensboro/Winston-Salem/High Point). However, all areas measured clean air quality
for ozone by 1992. The UAM project then was directed at developing a maintenance plan
for the Charlotte/Gastonia area to support a redesignation request. This paper details the
overall scope, organization, current status, and future plans for the North Carolina UAM
project.
THE SENSITIVITY OF METEOROLOGICAL AND EMISSIONS
UNCERTAINTIES ON URBAN AIRSHED MODEL OZONE CONCENTRATION
RESULTS IN NORTH CAROLINA
Brian S. Titnin and Janice Godfrey
North Carolina Division of Environmental Management
P.O. Box 29535
Raleigh, NC 27626
Ji-Cheng Carey Jang
North Carolina Supercomputing Center
P.O. Box 12889
Research Triangle Park, NC 27709
The State of North Carolina- Air Quality Section has modeled three multi-day ozone
episodes with the Urban Airshed Model (UAM-IV). There are significant uncertainties
associated with each of the inputs to the model. Sensitivity analyses were conducted to
quantitatively and qualitatively examine the impact of meteorological and emissions
changes on modeled ozone concentrations.
The three episodes each represent a different meteorological regime and thus display
differences in the magnitude of sensitivity to changes. The first episode represents a case
with strong northerly transport. The second episode is a stagnant period with very little
60
-------
movement of pollutants. The third episode represents a mixture of moderate northerly
and southerly transport of pollutants.
The identification of parameters that exhibit a strong sensitivity in the model will
help improve future modeling by identifying areas to concentrate future research and
monitoring studies. Parameters having a strong influence on modeled ozone
concentrations will be put in the forefront of model improvement activities.
UPDATED LAND USE/LAND COVER DATA -ITS EFFECTS ON ESTIMATES
OF BIOGENIC EMISSIONS IN THE NORTH CAROLINA URBAN AIRSHED
MODELING EFFORT
William W. Cure
NC DEHNR/DEM
Air Quality Annex
15 North West St.
Raleigh, NC 27603
919-733-1806
919-733-1812 (FAX)
In North Carolina, emissions from vegetation comprise the bulk of the volatile
organic compound (VOC) inventory. Estimation of these emissions will thus play a
substantial role in the regulatory strategy suggested by modeling. The Urban Airshed
Model (UAM), as available from the EPA, presently estimates biogenic emissions from
land use/land cover data more than 20 years old. Recent data obtained from the U.S.
Forest Service's periodic Forest Inventory and Assessment surveys have been gridded to
the modeling domain. Emission estimates based upon these data will be compared with
those from the older land use/land cover data set.
ENHANCEMENTS TO THE EMISSIONS INVENTORY INPUTS FOR THE
NORTH CAROLINA UAM PROJECT
Sheila C. Holman
NCDEHNR/DEM/Air Quality Section
15 N. West Street
Raleigh, NC 27603
In an effort to better define and allocate both base year and projected emissions
inventories, North Carolina Division of Environmental Management, Air Quality Section
implemented a rigorous exploration of certain aspects of emissions inventory calculation,
projection, and temporal allocation for use in an UAM analysis. In particular, a variety of
growth factors were examined, several VMT projection methodologies were scrutinized,
and default temporal allocation profiles for certain emissions categories were studied to
determine if local data existed to develop alternative profiles. This paper presents the
findings of each of the above mentioned studies.
61
-------
MOBILE EMISSION CALCULATIONS FOR THE NORTH CAROLINA UAM
PROJECT
Behshad M. Norowzi
NC DEHNR/DEM
Air Quality Annex
15 North West Street
Raleigh, NC 27603
In development of highway mobile emissions for use in the North Carolina Urban
Airshed Model (UAM) project, the Division of Environmental Management (DEM), Air
Quality Section focused on the following aspects of the highway mobile emission
calculations.
1- A method to estimate the inspection and maintenance (I/M) program influence in
any of the North Carolina counties. The concept of 1/M fractions was developed and
implemented using North Carolina Accident Data from 1988-1992.
2- Use of the daily rather than hourly emission factors in calculation of highway
mobile portion of UAM emissions inventories based on a comparative analysis of daily
and hourly emission factors for a test case.
3- North Carolina specific vehicle age distribution developed to be used in MobileSa
runs to improve accuracy of highway mobile emission estimates.
This paper details the rationale for exploring these parameters as well as the
justification for the final decision on the mobile emission calculations.
USE OF LINK-BASE DATA TO ADD DEFINITION TO HIGHWAY MOBILE
EMISSIONS FOR THE UAM
Anne S. Galamb
NC DEHNR/DEM
Air Quality Annex
15 North West Street
Raleigh, NC 27603
In developing highway mobile emissions for the Urban Airshed Model (UAM), the
North Carolina Division of Environmental Management (DEM), Air Quality Section
utilized link-based data, in addition to highway performance monitoring system (HPMS)
data, in order to add definition to the mobile source emissions. HPMS data provided
vehicle miles travelled by functional class per county. Link-base data allowed emissions
to be calculated for each road segment, or link, for each of the top three functional
classes.
The link-based mobile emission estimate method was selected for North Carolina
because of the nature of the state. The majority of North Carolina's population is in three
62
-------
city clusters (Raleigh/Durham, Greensboro/Winston-Salem/High Point and
Charlotte/Gastonia). These clusters are composed of medium-sized cities with an average
of sixty miles between the clusters. With a concentration of population in three
widespread city centers, the link-based emissions were thought to give a more accurate
estimate of emissions and modeling results than simply allocating the emissions over
each county on a population density basis without respect to the location of roads. This
paper details the structure of the project, as well as the quality assurance process.
QUALITY ASSURANCE OF THE NORTH CAROLINA PRECURSOR OF
OZONE INVENTORIES, EMISSIONS PREPROCESSOR SYSTEM AND THE
URBAN AIRSHED MODEL OUTPUT
Laura Boothe and Victoria Chandler
NC DEHNR/DEM
Air Quality Annex
15 North West Street
Raleigh, NC 27603
The Urban Airshed Model (UAM) is being utilized in North Carolina as a tool for
developing regulatory strategies. In order to have confidence in the results of the UAM
output, the modeling inventories and the Emissions Preprocessor System (EPS) outputs
needed to be rigorously quality assured. North Carolina Division of Environmental
Management (DEM), Air Quality Section developed quality assurance strategies to
ensure data integrity at all phases of inventory development and preprocessing for input
into the UAM. This paper outlines the quality assurance strategies developed and
implemented for the emissions inventories that were used in the UAM modeling effort.
Wednesday Afternoon
AN ENVIRONMENTALIST'S PERSPECTIVE ON OZONE MEASUREMENT:
THE ROLE OF SCIENTIFIC AND "CITIZEN-SCIENTIFIC" METHODS
Brian J. Morton
North Carolina Environmental Defense Fund
128 East Hargett Street, Suite 202
Raleigh, North Carolina 27601
From an environmentalist's perspective, measurement of tropospheric ozone is most
effective when conducted according to a systems approach that combines scientific and
social technologies. The traditionally scientific measurement of ozone involves
sophisticated instruments and techniques which, when used by skilled operators, produce
precise and reliable estimates of ozone concentrations in the atmosphere. These
measurements are used, among other ways, for air quality trends analysis and
attainment/nonattainment determinations. The new component in the measurement
system advocated here is inclusion of trained laypersons — ''citizen scientists" — who
63
-------
examine gardens, farms, and forests for ozone-induced injury to plants and trees.
Citizens monitor bio-indicators of ozone for symptoms of exposure and evaluate severity
of the symptomatology. A prototypical program applicable to forest ecosystems is
described. An advantage of citizen-based effects monitoring compare to technology-
based air quality monitoring (including related reports) is that the former makes ozone
pollution problems tangible. Thus an important but complex environmental problem is
made more understandable to citizens. Citizen-based ozone effects monitoring facilitates
public involvement in policy discussions, moving our nation closer to the important goal
of using consultation and dialogue to create strategies for solving environmental
problems. Further, increased public understanding may lead to public pressure which is
critical to new regulatory action and which increases the likelihood of voluntary emission
reduction efforts.
AUBURN TOWER OZONE STUDY
George C. Murray, Jr., Thomas L. Manuszak, Robert S. Graves, and
M. Jeffrey Gobel
North Carolina Department of Environment, Health, and Natural Resouces
Division of Environmental Management
The Raleigh-Durham area has been designated as a moderate non-attainment area for
ozone because of measured excessive ozone concentrations. Redesignation proceedings
are in progress for this area. The Auburn Tower, a 2000 foot broadcasting tower located
about 10 miles southeast of Raleigh, provided the opportunity to perform multiple
elevation atmospheric sampling. A study was designed to measure the ozone
concentrations at three elevations. Three ozone monitors, hydrocarbon samplers and
carbonyl samplers began sampling on July 26,1993. Organic compounds were collected
by contract laboratories who changed canisters and DNPH reagent cartridges and
analyzed the samples. The organic sampling results are discussed in other related papers.
Sampling continued until September 3,1993 when all ozone equipment was audited and
disconnected. Ozone was monitored continuously, 23 hours a day with one hour set aside
for nightly automatic zero/span checks. Long sampling lines and probes were attached to
the tower from each elevation down to the air conditioned room used for the ozone
monitoring equipment. Heated lines and water traps were attached inside this air
conditioned room along with ah1 monitors, calibrators, dataloggers and chart recorders. A
summary of the ozone concentrations measured is presented as well as a graphical
representation of typical data. The normal diurnal pattern seen at ground level monitoring
was not seen at 820 foot and 1420 foot elevation. Daily averages and maximums were
higher at elevated levels. The average ozone concentration at ground level for August
was .034 ppm and at 1420 foot level the average was .061 ppm. The maximums at
ground level for August was .094 ppm versus the maximums for August at 1420 foot
level was .105 ppm. The study will continue in 1994.
64
-------
VERTICAL DISTRIBUTIONS OF CARBONYLS FOR OZONE CONTROL IN
URBAN NORTH CAROLINA
Viney P. Aneja, Jay H. Lawrimore, Mita Das, and Fred Stratton
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University
Raleigh, North Carolina 27695-8208
Thomas P. Murray
Department of Chemistry
University of North Alabama
Florence, Alabama 35632
William G. Lonneman
AREAL, U.S. EPA
Research Triangle Park, NC
George C. Murray
North Carolina Department of Environment, Health, and Natural Resources
Raleigh, North Carolina 27626
Ambient levels of five carbonyl compounds (formaldehyde, acetaldehyde, acetone,
propanal, and 2-butanone) were measured near Raleigh, NC from a 610 meter television
tower. Clg resin cartridges were used to collect three hour integrated samples from 5-8
AM weekday mornings at the surface, 250 meters, and 426 meters. Boundary layer wind
and temperature profiles were determined by balloon soundings on selected days and
hourly ozone concentrations were measured at each level. Formaldehyde concentration
was greatest at all three levels followed by acetaldehyde and acetone. Concentrations
were generally lowest at the surface with maximum values varying between the two
upper levels. Formaldehyde averaged 0.8 ppbv at the surface, 2.63 ppbv at 250 meters,
and 2.54 ppbv at 426 meters. Acetaldyde was 0.08 ppbv at the surface, 0.13 ppbv at 250
meters, and 0.14 ppbv at 426 meters.
65
-------
Session 10: Quality Assurance
Wednesday Morning
A STATISTICAL ANALYSIS OF AUDIT RESULTS FROM STATIONARY
SOURCE COMPLIANCE TESTS
Ellen W. Streib, Jack C. Suggs, and William J. Mitchell
U.S. EPA/AREAL (MD-77B)
Research Triangle Park, NC 27711
Certain EPA compliance test methods for stationary sources specify accuracy limits
that must be achieved on EPA provided audit materials for the compliance test to be
valid. This paper presents the results from an in-depth statistical analysis on 6,280 audit
results that examined the frequency with which the EPA limits were being met. The
results of this statistical analysis will be helpful to those responsible for determining the
validity of compliance test results as well as to those concerned with improving the
accuracy of stationary source test methods.
DISCLAIMER: The information in this document has been funded wholly by the
United States Environmental Protection Agency. It has been subjected to Agency review
and approved for publication.
EPA'S QA PROGRAM ON THE SUPPLIERS OF PROTOCOL GASES
Avis P. Hines and William J. Mitchell
U.S. EPA/AREAL (MD-77B)
Research Triangle Park, NC 27711
In 1992, EPA's Atmospheric Research and Exposure Assessment Laboratory
initiated a nationwide QA program on the suppliers of EPA Protocol Gases. The
program has three goals: to increase the acceptance and use of Protocol Gases by the air
monitoring community, to provide a QA check for the suppliers of these gases, and to
help the users of these gases identify suppliers who can consistently provide accurately-
certified Protocol Gases. In this QA program which operates continuously, Protocol
Gases are procured by EPA and the supplier's certification of the pollutant
concentration(s) is verified by EPA. The results are published on the EPA Technology
Transfer Network's electronic bulletin board. The results obtained to date are presented
along with an assessment of the quality of Protocol Gases today.
DISCLAIMER: The information in this document has been funded wholly by the
United States Environmental Protection Agency. It has been subjected to Agency review
and approved for publication.
66
-------
PREPARATION AND EVALUATION OF REPRESENTATIVE COMPOUNDS IN
SMALL HIGH PRESSURE CYLINDERS FOR USE AS AUDIT MATERIALS
William Mitchell, Jack Suggs, and Howard Crist
U.S. EPA/AREAL (MD-77B)
Research Triangle Park, NC 27709
Ron Bousquet, Ron Brande, John Holland, and John Duncan
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
Complex mixtures of VOCs have been prepared in Summa™ polished canisters for
use as audit materials. Quite often there is not enough material at the limited pressures,
imposed by the canister, to be useful. This low pressure limitation also prevents the
preparation of low concentration samples from a high concentration master cylinder. We
have chosen a representative group of compounds that includes halocarbons, aromatics,
oxygenates, and alkanes at high and low molecular weights to be evaluated as audit
materials in small high pressure cylinders. Master cylinders containing these compounds
at high ppm levels are mixed and diluted using a high pressure manifold system.
Duplicate audit cylinders are prepared at the desired concentration range and pressure.
Operation of the system modifications made, and estimations of the utility of this method
are discussed. Evaluation of the stability of the representative compounds at various
pressures and concentrations is described.
The information in this abstract has been funded wholly by the United States
Environmental Protection Agency. It has been subjected to Agency review and approved
for submission to the AWMA.
DATA HANDLING ISSUES AND TECHNIQUES ASSOCIATED WITH DATA
COLLECTED FROM AUTOMATED GC SYSTEMS USED FOR OZONE
PRECURSOR ANALYSIS
Larry D. Ogle, Margaret A. Underwood, Pamela R. Chen, and Pat G.
Edwards
Radian Corporation
P.O. Box 201088
Austin, TX 78720-1088
Jim Price
Texas Natural Resources Conservation Commission
P.O. Box 13078
Austin, TX 78711-3078
Paul Radenheimer
Consolidated Sciences, Incorporated
1416 Southmore
Pasadena TX 77502
The U.S. EPA-designated 55 ozone precursor compounds were monitored in the
ambient air in Houston, Texas on a continuous basis using two Perkin-Elmer automated
67
-------
continuous gas cinematographic systems. This project was sponsored by the Texas
Natural Resources Conservation Commission as a portion of the Coastal Oxidant
Assessment for Southeast Texas (COAST) study. Samples were collected over a 40
minute time period during each hour and analyzed hourly. A total of 22 ambient samples
was collected each day with a blank run and a calibration check sample in the remaining
two hours. On some days 23 samples were analyzed daily with an ambient sample in the
place of the blank.
The Perkin-Elmer analyzer utilizes a two column system to separate the ozone
precursor compounds. Therefore, 24 files are collected from each of the two columns
daily for a total of 48 data files per site. The result is 96 data files per day from the two
sites. In order to process the data, method files, sequence files and an electronic logbook
file were also transferred from each site daily. A total of two to three megabytes of data
was received daily for processing; validation and inclusion in a database for reporting in
AIRS format. This paper will describe data transfer, handling, data
verification/reprocessing techniques, validation and reporting procedures. In addition, a
number of tools will be described which were developed to aid handling and processing
the data.
A COMPUTER CONTROLLED DYNAMIC DILUTION SYSTEM FOR
IMPROVED ACCURACY AND QA/QC IN TO14 STANDARD PREPARATION
D.B. Cardin andJ.T. Deschenes
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
Dynamic Dilution is the method of choice for preparing low level TOM standards in
canisters. One or more cylinders containing NIST certified standards can be blended
together with a diluent gas under mass flow control to produce very consistent standards
at ppb levels. Blending manifolds can be designed to maintain constant temperatures,
pressures, and flow rates during the entire canister filling operation to insure that a proper
mass balance is maintained.
One major source of error can exist when using Dynamic Dilution that can be
significant in some cases. Manufacturers of mass flow controllers usually specify an
absolute error of not more than +-2% of the full scale flow rate. This means that MFC's
rated up to 100 seem (standard cc per minute) could be off as much as 2 cc. At 100% of
full scale, this results in an unfortunate but tolerable error of +-2%. However, at 10% of
full scale, a 2cc error corresponds to a 20% relative error which is not acceptable.
Available dynamic dilution systems have been unable to perform automatic calibrations
of mass flow controllers to account and correct for these inaccuracies.
A Dynamic Dilution system is presented that interfaces to a Windows™ based
operating system allowing implementation of sophisticated flow calculations and
feedback control. The dilution system is capable of using temperature compensated
vacuum reservoir pressurization to calibrate up to 6 MFC channels unattended. Multiple
calibration events can be set up to run sequentially with Means and %RSDs given for the
-------
data obtained for improved reliability. Calibrations over a several day period show the
stability of mass flow controllers and can indicate the presence of unusually large drifts.
These calibration factors are then used by the dilution system to make corrections in flow
signals to provide very accurate ppb level standards. The preparation of multiple
standards in different canisters without user intervention is also supported for making 2,
5,10,20, and 50 ppb level standards for TOM instrument calibration.
IMPORTANCE OF MINIMUM DETECTION LIMITS IN AIR TOXICS DATA
BASES
N. Adanis
U.S. EPA
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Environmental measurement data often produce many "less than" values or
minimum detection limits (MDLs) for pollutant concentrations. The EPA defines the
MDL as the smallest concentration of an analyte that has a 99% probability of not being
zero. MDL values may be used in the calculation of a mean concentration from a source
or locality, and these mean values may be used in determining compliance with a
regulatory limit, in the determination of an emission factor (typical concentration emitted
by a given type of source), or in various other modelling efforts that feed into various air
toxics data bases.
The determination of the MDL is therefore an important concern in planning an
environmental measurement program. The measurement instrument manufacturer's
stated MDL is frequently used as the estimated MDL, but this MDL is probably too low.
The manufacturer's MDL is often obtained experimentally under ideal conditions in a
matrix with minimal interferences. A more accurate MDL can be obtained by using a
"real world" matrix that approximates the measured matrix and by performing multiple
determinations using the same laboratory personnel and instrumentation that will be used
in the measurement project
After an appropriate MDL has been determined experimentally, there are still
problems in using the MDL in calculating a mean value. When experimentally
determined MDL values are used in the calculation of an average concentration, they may
introduce a positive bias; this approach yields the most conservative values to use in
regulation or modelling. The use of zero in place of the MDL may create a negative bias.
The range of means using zero and using the MDL may be calculated. The use of half of
the MDL has also been suggested as a reasonable compromise. However, if there have
been a sufficient number of measurements made to determine the distribution
characteristics, statistical methods may be used to obtain a more accurate estimate of the
mean value. These statistical methods involve the use of "look-up" tables to estimate the
values that would have been obtained if the measurement instrument could detect all
values, including those near zero.
69
-------
There are many approaches to dealing with the MDL problem. This meeting of
persons knowledgeable in air pollutant measurement and regulation may serve as a forum
for consideration of standards for the appropriate use of MDL values.
STABILITY EVALUATION OF MULTICOMPONENT EPA PROTOCOL
GASES
Richard C. Shores, Michael J. Messner, and Robert W. Murdoch
Center for Environmental Measurements and Quality Assurance
Research Triangle Institute
Research Triangle Park, NC 27709
Easter A. Coppedge, Thomas J. Logan, and MR. Midgett
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
As assessment of Specialty Gas Manufacturers' gases was conducted by Research
Triangle Institute (RTI) for the U.S. Environmental Protection Agency (EPA)
Atmospheric Research and Exposure Assessment Laboratory (AREAL) in 1991 to
evaluate the accuracy of the manufacturers' reported concentrations for multicomponent
gases in two concentration ranges. Because this was the eigth assessment of protocol gas
accuracy, this work was referred to as Cylinder Audit No. 8. The results of Cylinder
Audit 8 were presented at the Air and Waste Management Association (AWMA)
conference in May 1992. These same cylinders remained in the custody of RTI after the
audit and have been reanalyzed to evaluate the stability of multicomponent protocol gases
over a two-year period.
The cylinders evaluated during this study were purchased from nine different
manufacturers and contained both SO2 and NO with a balance gas of nitrogen. Half of
the cylinders contained SO2 at 1500 ppm and NO 900 ppm and the remaining cylinders
contained SO2 at 300 ppm and NO at 400 ppm.
The stability of multicomponent gases is important because the EPA traceability
protocol for certification of calibration standards specifies recertification intervals for
gaseous standards. This paper presents the results of evaluating the extent to which the
concentrations of multicomponent cylinder gases change over a two-year period would
be required to detect a change with statistical confidence.
DISCLAIMER: The information in this document has been funded wholly or in part
by the U.S. Environmental Protection Agency under Contract no. 68-D1-0009 to
Research Triangle Institute. It has been subjected to Agency review and approval for
publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
70
-------
Wednesday, May 4,1994
Session 11: FTIR Studies
Wednesday Morning
OPEN-PATH FTIR AIR QUALITY MEASUREMENTS AT A
PETROCHEMICAL COMPLEX IN BRAZIL
Robert H. Kagann
2895 Bordeaux Blvd
Gumming, Georgia 30131
Neuza Neves and Felipe Villas Boas
CETREL S.A.
CX. Postal 01 1 CEP 42810
Camacari, Bahia, Brazil
An open-path FTIR sensor was used to determine the characteristic air pollutants at
ten different locations in a large petrochemical complex in Bahia, Brazil. These sets of
measurements were part of an initial survey in preparation for a measurement program
(using both open path FTTR and GC/MS analysis of collected air samples) to completely
characterize the air quality within the complex and to characterize and obtain emission
rates of the individual sources. In the survey, a total 17 different compounds were
measured with the FTIR sensor. These includes everal polar species, including ammonia
and acrylonitrile.
OPEN PATH FTIR: TOMOGRAPHIC AND I(O) STRATEGIES AND RESULTS
SJP. Levins, D.Y. Park, 5. Giese-Bogdan, M. Simonds,
and H.K.Xiao
University of Michigan
School of Public Health
Ann Arbor, MI 48109-2029
R.A. Spear, M. Yost, A. Drescher, and Y. Zhou
University of California
School of Public Health
Berkeley, CA 94720
This will be a summary presentation of the progress of our research group in the two
areas. First, progress will be reported in the application of OP-FTIR to the tomographic
reconstruction of concentration fields for air contaminants. In this experiment, a fast-
aiming beam system was used to scan 55 mirror and corner cube positions in less than 4
minutes. Using the points of intersection of the beams as key elements in the tomographic
reconstruction strategy yielded results that were enhanced when compared to
tomographic methods using parallel and radial beam geometries. Second, progress will be
reported in the use of the differential method for generation of a spectrum without the use
71
-------
of a clean air background. The differential method has been applied and compared to the
iterative backfit method for field data from an acrylonitrile-butadiene-styrene plant.
DEVELOPMENT OF QUALITY ASSURANCE PROCEDURES IN OPEN-PATH
FT-IR MONITORING
Edgar L. Thompson, Jr., Jeffrey W. Childers, and George M. Russwurm
ManTech Environmental Technology, Inc.
P.O. Box 12313
Research Triangle Park, NC 27709
There is currently no consensus regarding the appropriate quality assurance (QA)
procedures to be used to validate open-path FT-ER data. To address this issue, routine
open-path FT-ER. data were collected over a six-month period. A semi-permanent
monitoring site with an FT-ER. system in the monostatic configuration was established
over a grassy field with a total path length of 512 m. A daily protocol that included
measuring the instrument's electronic noise, the single-beam intensity of the return
signal, and the baseline noise over spectral regions of interest was followed. Ancillary
measurements, including relative humidity, ambient temperature, and wind direction,
were also made. Spectral measurements were typically made by acquiring 5-min co-
averaged spectra at 15-min intervals over 7 to 8 hrs starting in the morning and
continuing through the afternoon. The ambient concentrations of CO, methane, and N20
were determined from these spectra to assess the stability of the instrument and to
investigate the feasibility of using ambient gas concentrations for QA purposes. Extended
studies were also conducted continuously over a 36 to 72 hr period. We also investigated
the use of a QA cell with surrogate standards, such as SF6, n-butane, ethylene, methane,
and CO, to determine instrument stability. In addition to the noise and concentration
measurements, the repeatability of the position and full width at half height of selected IR
absorption bands was also determined.
ADAPTATION OF A MILITARY FTS TO CIVILIAN AIR TOXICS
MEASUREMENTS
James R. Engel and Rick Dorval
OPTRA, Inc.
461 Boston Street
Topsfield, MA 01983
Phone: (508) 887-6600
FAX: (508) 887-0022
In many ways, the military problem of chemical agent detection is similar to the
civilian problem of toxic and related air pollutions detection. A recent program to design
a next generation Fourier transform spectrometer based chemical agent detection system
has been funded by the U.S. Army. This program has resulted in a Fourier transform
spectrometer system that has a number of characteristics that make it suitable for
72
-------
applications to the civilian measurement problem. Low power, low weight, and small
size lead to low installation, operating, and maintenance costs. Innovative use of diode
lasers in place of HeNe reference sources leads to long lifetime and high reliability.
Absolute scan position servos allows for highly efficient offset scanning. This paper will
discuss these characteristics and relate the performance of this system to present air
monitoring requirements.
JOINT OBSERVATIONS OF THE ETL OZONE LJDAR AND MPS SYSTEM
DURING THE LOS ANGELES FREE RADICAL STUDY
Yanzeng Zhao
Cooperative Institute for Research in Environmental Sciences
University of Colorado
Boulder, CO 80309-0449
John Gaynor and Daniel Wolfe
NOAA/ERL/Environmental Technology Laboratory
325 Broadway
Boulder, CO 80303-3328
Lidar systems measuring ozone concentration and aerosol profiles in the lower
troposphere can make significant contribution to the understanding of chemical and
transport processes for regional air quality assessment and control. The Environmental
Technology Laboratory (ETL) of the National Oceanic and Atmospheric Administration
has developed a very compact, efficient, transportable lower troposphere ozone lidar
which is capable of continuously remote sensing ozone concentration profiles from near
the surface to about 3 km with high range resolution and measuring aerosol backscatter
profiles from near the surface to about 10 km. The ozone lidar was deployed in two field
experiments in California in 1993. The second field experiment discussed here was the
Free Radical Study in the Los Angeles basin during September 1993. The ETL Mobile
Profiling System (MPS), co-located with the lidar during this study, continuously
measured surface and profiles of meteorological parameters using remote and in-situ
techniques. The meteorological information is combined with the ozone lidar
measurements providing valuable insight into the complexity of the ozone profiles.
FOURIER TRANSFORM MICROWAVE SPECTROSCOPY A POTENTIAL
NEW ANALYTICAL TOOL FOR GAS PHASE SPECIES
Richard Suenram, Frank Lovas, and Robert Sams
National Institute of Standards and Technology,
Gaithersburg, Maryland 20899
A number of recent improvements in Fourier Transform Microwave Spectroscopy
(FTMS) have increased the detectability limits for gas phase species down to the 10-100
ppb range. This high sensitivity coupled with 100% species selectivity, millisecond
response times and complete automation of the instrument provides analytical chemists
73
-------
with a new technique for the determination of trace gas species that are present in
industrial chemical processes. The technique can be used either in-lieu-of or as a
complementary detection technique to gas chromatography mass spectrometric
techniques (GCMS) for the detection of gas phase species.
Sensitivity tests of the new spectrometer were carried out using 800 ppm samples of
acrolein (CH2=CHCHO and propionaldehyde (CH3CH2 CHO). Using these samples and
the known isotopic abundances of 13C = 1.1%, 18O = 0.2%, and H = 0.015%, single shot
(one gas pulse from the pulsed nozzle) detection limits of approximately 300 ppb were
attained. Averaging for 100 nozzle pulses yielded a signal-to-noise ratio of 15/1 on
transitions of the mono deuterated isotopes in natural abundance. This corresponds to
detection limits of 10 to 100 ppb. Additional improvements in the instrument's detection
system and software modifications should yield another order of magnitude improvement
in sensitivity.
Advantages of the new technique in monitoring trace gas constituents include the
100% certainty of the identity of the species being monitored. Using a pulsed molecular
beam nozzle in conjunction with a specially designed flow nozzle which operates at up to
30 Hz repetition rates allows rapid response to changes in industrial process flow streams.
The cooling provided by the pulsed molecular beam permits larger molecular species to
be monitored with the same ease as smaller species since only the lowest rotational
energy levels will of any species will be populated at the 1 Kelvin temperature of the
molecular beam.
MONITORING AIR POLLUTANTS BY MOLECULAR BEAM MICROWAVE
SPECTROSCOPY
U. Andresen, U. Kretschmer, C. Thomsen, and H. Dreizler
Abteilung Chemische Physik im Institut fiir Physikalische Chemie der Universitat Kiel
24098 Kiel, Germany
The molecular beam Fourier transform microwave spectroscopy has been proved to
be a very powerful tool to determine molecular parameters in the gas phase with high
precision and high resolution: It will be shown that this method is also suitable for
quantitative analysis, this means the evaluation of concentration of single components in
complex gas mixtures like pollutants in air.
Microwave spectroscopy is mainly restricted to the investigation of polar molecules.
Many industrial solvents, chlorofluoromethanes or other air pollutants belong to this
group of substances. Presently about 2000 molecules have been analyzed and their
parameters published. Many of these gases may be monitored with only one instrument.
To switch from one substance to another takes about one minute.
Due to the high resolution of the apparatus there is nearly no cross sensitivity. The
usable range of concentration is from percent to the lower ppm at least, but may be lower
depending on the gas. The total measuring time ranges from seconds to minutes
depending on concentration and substance.
The method and the apparatus will be presented and some features of instrument will
be discussed.
74
-------
Wednesday Afternoon
EFFECT OF TEMPERATURE ON THE ABILITY TO COLLECT DATA: MDA
SCIENTIFIC OPEN-PATH FOURIER TRANSFORM INFRARED
SPECTROMETER WITH STIRLING COOLER
Judith 0. Zwicker and William M. Vaughan
Remote Sensing =Air, Inc.
8147 Delmar, Suite 219
St. Louis, Missouri 63130
(314) 863-9801
(314)7210759
The experience of several users of the MDA Scientific, Inc. open-path Fourier
Transform Infrared spectrometer has shown that the system has difficult collecting data at
low temperatures, especially when the instrument is subjected to long periods at low
temperatures when the instrument is not operating. It was decided that it would be useful
to document how long the instrument needed to warm up before data could be collected.
It was also decided to try different insulating and heating arrangements to determine the
most advantageous for various monitoring conditions. The instrument was set up in an
unheated garage and the evening temperature, the low temperature, and the morning
temperature in the garage recorded. The instrument was turned on in the morning and the
time recorded. The continuous monitor program was turned on at the same time. The
continuous monitor program was set to collect data every 5 minutes. The resulting data
saved indicated the time at which the instrument began collecting data and when the
intensity was sufficient to obtain reasonable data. The experiment was carried on for
several days with the instrument unheated to determine the time scales under those
conditions. Then the instrument was fitted with a heating jacket. The same experiment
was conducted with the heating jacket turned on at the same time as the computer and
instrument as would be needed if generator power were required. In the field and with the
jacket left on over night and with the jacket turned on with a timer one hour before the
instrument was turned on. Data are presented and recommendations for heating for
various temperature scenarios.
PHOTOLYSIS ASSISTED POLLUTION ANALYSIS (PAPA)
Philip L. Hanst
Infrared Analysis, Inc.
11629 Deborah Drive
Potomac, MD 20854
Photolysis Assisted Pollution Analysis (PAPA) is a method of utilizing the infrared
spectrum to measure gases at low concentrations. PAPA goes beyond standard infrared
absorption techniques by using a compound conversion process that selects reactive gases
from among the non-reactive and reveals their spectra At the same time, the technique
75
-------
renders invisible the spectra of water and carbon dioxide. Interference removal is the
main benefit of PAPA, but a second benefit is the conversion of some difficult-to-
measure compounds (such as hydrogen sulfide) to more easily measured forms (such as
sulfur dioxide). A PAPA system couples a Fourier transform spectrometer to a multiple-
pass absorption cell that contains an internal quartz-mercury ultraviolet lamp. When the
lamp is turned on for just a few seconds, the ultraviolet radiation initiates reactions that
convert most pollutant gases to other forms. The radiation has no effect on water, carbon
dioxide and other fully-oxidized molecules. A spectrum is recorded before and after the
irradiation. One spectrum is then divided by the other and the result is put into
absorbance form. The infrared bands of compounds transformed appear downwards in
the absorbance spectrum, and the infrared bands of compounds produced appear
upwards. Bands of water and carbon dioxide do not appear at all. The amount of reactants
and products are measured with aid of digitized quantitative reference spectra. The PAPA
method can measure pollutant gases down to parts-per-billion concentration levels. No
calibration gases or other chemicals are required for the analysis. A measurement of
many pollutant gases can be carried out in just a few minutes time. Applications
envisioned for the PAPA system include measurements of: (1) nitrogen oxides and other
nitrogen-containing pollutants in air and in combustion effluents; (2) organic pollutant
gases in air and in combustion effluents; (3) isoprene, pinenes and other reactive natural
hydrocarbons in air; (4) hydrocarbons and other impurities in ground water; (5)
hydrocarbons and other impurities in soil; (6) sulfur-containing pollutants in air,
combustion effluents, soil and water; (7) impurities in carbon dioxide, including sulfur-
containing compounds.
FTIR TRANSMISSION SPECTROSCOPY FOR QUANTITATION OF
AMMONIUM BISULFATE IN FINE PARTICULATE MATTER COLLECTED
ON TEFLON FILTERS
Kenneth H. Krost and William A. McClenny
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
A quantitative measurement method for fine particle bisulfate in ammonium
bisulfate collected from the ambient air onto Teflon filters is described. Infrared
absorbance measurements of the Teflon filters are made before and after particle
collection. Subtraction of the two spectra reveals the absorbance spectrum of the
particles. The presence of bisulfate is identified by characteristic and unique spectral
features including prominent absorption bands at 1050 cm"1 and 870 cm"1. The
integrated band absorbance at 870 cm"1 is used for quantitation. Wet chemical
measurements of sulfate and hydrogen in synthetically-generated ammonium bisulfate
standards are used in calibration. The lower limit of detection for the bisulfate ion is 150
nanomoles. This corresponds to the ammonium bisulfate which would be collected from
an air volume containing 1.2 micrograms/m3 samples for 24 hours at 10.0 L/min. This
method provides a specific, nondestructive, direct measurement of ammonium bisulfate.
76
-------
As such, the method has distinct advantages over indirect methods involving ion balance
in solution.
DISCLAIMER: The research described in this article has been funded by the United
States Environmental Protection Agency. It has been subjected to Agency review and
cleared for publication. Mention of trade names of commercial products does not
constitute endorsement or recommendation for use.
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY FOURIER
TRANSFORM INFRARED SPECTROSCOPY TEST PROGRAM FOR
EMISSIONS MEASUREMENT
Lori Lay
Emission Measurement Branch
Technical Support Division, OAQPS
U.S. EPA
Research Triangle Park, NC 27711
The U.S. Environmental Protection Agency (EPA) published amendments to the
Clean Air Act (CAA) November 15,1991. Title III of the CAA amendments included a
list of 189 hazardous air pollutants (HAPs) for which test procedures must be established.
An extractive stack sampling test method, using Fourier Transform Infrared (FTIR)
spectroscopy, is being developed for measuring HAP compounds. The FT1R procedure
has the potential to detect over 100 of the listed compounds, plus additional compounds
such as criteria pollutants. This procedure has the ability to detect multiple compounds
simultaneously and will provide near real time data.
Since the development of this procedure, many source categories have been screened
for HAP emissions. Modifications to the techniques have been made and validation
testing has been performed. Currently, this technique is being used to collect data for
Maximum Achievable Control Technology standard development. We plan to adapt this
technique to be used in continuous emissions monitoring for facilities burning hazardous
waste.
77
-------
•,1994
Session 12: New Methods for VOCs
Wednesday Afternoon
THE CONCENTRATION AND MEASUREMENT OF AIR POLLUTANTS BY
GC/MS; A COMPARISON OF SORBENT VERSUS CRYO TRAPPING
Elizabeth Almasi and Norman Kirshen
Varian Chromatography Systems
2700 Mitchell Drive
Walnut Creek, CA 94598
The measurement of toxic air contaminants and ozone precursors in ambient air is
gaining more importance as the 1990 Clean Air Act Amendments (CM) take effect.
Simple, reliable and low maintenance instrumentation is needed to detect the more than
100 volatile organic compounds (VOCs) in ambient air at low or sub ppb levels as
specified in the CAA.
A Saturn GC/MS ion trap system equipped with an inboard concentrator (SPT) was
used to study the advantages and disadvantages of cryo and sorbent trapping of the
pollutants. The cryo trapping was carried out on glass beads at -150C or -180C, using
liquid nitrogen as coolant. The sorbent trapping took place at or around ambient
temperatures employing custom selected sorbent traps and liquid CO2 for cooling.
Sample volumes of 60ml and higher were concentrated, and after separation on a
capillary column, full scan detection was performed.
System precision, detection levels and linearity data will be presented for different
compound groups and results of ambient air samples will be shown.
A REAL TIME SORBENT-BASED CONTINUOUS MONITORING SYSTEM
Frank G. Lattin
U.S. Army Edgewood Research, Development and Engineering Center
Operations Directorate
Chemical Support Division
Aberdeen Proving Ground, MD 21010
Donald G. Paul
Dynatherm Analytical Instruments, Inc.
P.O. Box 159
Kelton, PA 19346
A real time sorbent-based continuous monitoring system is described that provides
quantitative analysis of airborne levels of several volatile cyano and chlorinated
compounds of military interest: hydrogen cyanide, cyanogen chloride, phosgene, and
chloropicrin. All are listed in the Schedule 3 class of dual use chemicals produced in high
78
-------
commercial volume that will be monitored under terms of the recently signed multi-
lateral Chemical Arms Treaty.
The monitoring system is characterized by its sensitivity (ability to measure at low
concentrations), selectivity (ability to filter out interferences), dynamic range and
linearity, real time mode (versus methods requiring extensive sample prep procedures),
and ability to interface with complimentary GC detectors.
A saved portion of sample from each sampling cycle is available for confirmatory
analyses to verify positive responses, resolving a problem encountered in traditional
sorbent-based sample collection and thermal desorption methodologies. Splitting a
portion of each sample to a second sorbent tube provides a reservoir of sample from
which multiple analyses can be made.
Included in the presentation is a description of quality assurance objectives in the
monitoring system, and an assessment of method accuracy, precision and detection
levels.
THE PERKEV-ELMER ATD-400 SYSTEM FOR MONITORING OF AMBIENT
VOC OZONE PRECURSORS
Paul Radenheimer
Consolidated Sciences Corporation
1416 Southmore
Pasadena, TX 77502
John Gibich
Texas Natural Resource Conservation Commission
P.O. Box 13087
Austin, TX 78711-3078
Larry Ogle
Radian Corporation
P.O. Box 201088
Austin, TX 78720-1088
The systems described were used in the Texas Natural Resource Conservation
Commission (TNRCC) Coastal Oxidant Assessment for Southeast Texas (COAST)
program. Continuous VOC monitoring programs were implemented in June through
November of this year in Houston. This Perkin-Elmer developed monitoring system
included a dual capillary column chromatographic application in an 8700 GC with a
modified ATD-400 sampling system. The resulting separations are monitored and
quantified using Turbochrpm (ver. 3.2) chromatographic data handling software. The key
element to this system is the thermal desorption device which is capable of concentrating
VOCs from ambient air then desorbing them directly to the gas chromatograph on a
continuous cycle. The ATD-400, unlike other similar devices, contains an activated
carbon trap which is electrically cooled during collection via Peltier technology. This
eliminates the need for cryogenic liquids or compressed air (vortex) for sample collection
and makes unattended field operation more feasible than previous systems. The entire
system is integrated to provide a completely automatic sample collection cycle, gas
79
-------
chromatographic analysis and data collection, reduction and report generation. Reliability
and repeatability data suggest that the system is robust. Good correlation between
systems and quality audits confirms data from the manufacturer. Modifications which are
being implemented to the chromatographic system, system sites, and operator behavior to
enhance the system performance will also be presented.
SYSTEM OPERATION: CONTINUOUS VOLATILE ORGANIC COMPOUND
AIR MONITORING OF 56 OZONE PRECURSORS WITH THE
PERKIN-ELMER 8700 GC AND AUTOMATIC THERMAL DESORPTION
SYSTEM
Paul Radenheimer
Consolidated Sciences Corporation
1416 Southmore
Pasadena, TX 77502
John Gibich
Texas Natural Resource Conservation Commission
P.O. Box 13087
Austin, TX 78711-3078
Larry Ogle
Radian Corporation
P.O. Box 201088
Austin, TX 78720-1088
As part of the Coastal Oxidant Assessment for Southeast Texas (COAST) program,
two sites were chosen by the Texas Natural Resource Conservation Commission
(TRNCC) and equipped with a Perkin-Elmer VOC system composed of the 8700 Gas
Chromatograph, ATD-400 Automatic Thermal Desorption and Turbochrom HI Data
system on DEC computers. The systems were equipped with a dual capillary column
application capable of resolving 56 distinct target ozone precursors. These components
were separated and quantified on an hourly basis 24 hours each day. Each system
generated 96 data files and approximately 30 documentation files each day totaling nearly
3 megabytes of information. The system was fully automated and monitored rigorously
via high-speed modem communication. The modem communication proved to be
essential in the handling of the large volume of data generated each day. A fully
automated data transfer system was developed to allow unattended file archiving thus
eliminating many problems associated with manual handling of files and facilitating the
rapid evaluation of the data. This paper will identify the major issues in operation and
maintenance of these systems (not including the chromatographic application). Problems
which were encountered can be subdivided into two categories, a) hardware system
problems such as power failures, equipment malfunction and temperature/humidity
fluctuations, and b) software issues: capability/incompatibility, bugs, communication
problems and a plethora of computer or computer-related issues (confusion).
80
-------
A SYSTEM FOR THE DETERMINATION OF TRACE-LEVEL POLAR AND
NON-POLAR TOXIC ORGANIC COMPOUNDS IN AMBIENT AIR
A. Tipler, R. Dang, and H. Hoberecht
GC Applications Group
Perkin Elmer Corp
50 Danbury Road
Wilton, CT 06897-0280
Tel. (203) 761 5183
This paper describes the principle and application of a gas chromatographic system
for the determination of volatile toxic organic compounds at sub-ppb levels in ambient
air. The system is compatible with both on-line (direct) and off-line (passivated canisters
and adsorbent tubes) sampling strategies.
Central to the system is a multi-bed adsorbent trap which is held at an near-ambient
temperature while the sample is drawn through it. Analytes of interest are trapped and
concentrated on the adsorbent whereas most of the water vapor in the sample is
unretained. Next, a volume of dry helium is passed through the trap to reduce the level of
the residual water. The trap is then rapidly heated and a flow of carrier gas backflushes
the desorbed analytes, without splitting, into a very low phase-ratio a polar capillary gas
chromatographic column. The thick film of stationary phase enables the separation of
even the most volatile components without the need for sub-ambient GC operation.
System performance is demonstrated using the various sampling techniques and
classical and mass spectrometric detectors for the target analytes specified in the USEPA
Compendium Method TOW and the polar analytes being proposed in Method TO 15.
WATER MANAGEMENT IN CAPILLARY GAS CHROMATOGRAPHIC AIR
MONITORING SYSTEMS
A. Tipler
GC Applications Group
Perkin Elmer Corp
50 Danbury Road
Wilton, CT 06897-0280
Tel. (203) 761 5183
Capillary gas chromatography is an excellent technique for the speciated quantitation
of low-level volatile organic compounds (VOCs) in ambient air.
Although GC detectors have excellent sensitivity, some sample pre-concentration
will be necessary to enable detection of VOCs at sub-ppb levels. This process normally
employs a cooled and/or adsorbent trap to retain the analytes from a large volume of
sample air. For very volatile VOCs, a very retentive trap is used and this will also retain
any water present as vapor in the sample.
This trapped water causes significant problems with the chromatography and
detector operation and methods must be sought to remove it or eliminate its effects. This
paper investigates the magnitude of the problem and examines the various alternatives for
81
-------
managing the trapped water. The application of some of these techniques is
demonstrated in a method for the determination of volatile polar and non-polar toxic
organic compounds in ambient air.
IDENTIFICATION OF AMBIENT AIR SAMPLING AND ANALYSIS
METHODS FOR THE 189 TITLE HI AIR TOXICS
R. Mukund, Sydney M. Gordon, and Thomas J. Kelly
Atmospheric Science and Applied Technology
Battelle
505 King Avenue
Columbus, OH 43201
A systematic approach has been applied to identify sampling and analysis methods
for ambient levels of the 189 Hazardous Air Pollutants (HAPs) listed in the 1990 Clean
Air Act Amendments (CAAA). A comprehensive data base of available physico-
chemical data on these pollutants has been assembled, with corrections for erroneous and
conflicting literature data. The data base includes boiling point, vapor pressure, electronic
polarizability, solubility in water, overall half-lives due to reaction or deposition in the
atmosphere, and aqueous reactivity. The data base has been used to classify the
compounds into several categories and sub-categories, such as very volatile and volatile
organic and inorganic compounds, semi-volatile organic compounds, and non-volatile
inorganic compounds. Within each sub-category of compounds, available sampling and
analysis methods have been identified for each compound. Methods include established
U.S. EPA techniques, workplace and source emissions monitoring methods, and other
techniques reported in the literature. Using a tabular format, available methods for each
compound are denoted as either demonstrated, likely, or potential based on pre-
established assignment criteria. For each demonstrated method, Limit of Detection
(LOD) values or ranges are provided. For likely or potential methods, LOD ranges are
estimated where necessary. A discussion of the approach used and examples of
compound entries in the tabulation will be presented.
DIRECT TRACE ANALYSIS OF VOLATILE ORGANIC COMPOUNDS IN AIR
USING FILTERED NOISE FIELD ION TRAP MASS SPECTROMETRY
Sydney M. Gordon, Patrick J. Callahan, and Donald V. Kenny
Atmospheric Science and Applied Technology
Battelle Memorial Institute
505 King Avenue
Columbus, OH 43201
There is increasing interest in the development of field portable mass spectrometers
to monitor environmental pollutants in real time. A direct air sampling filtered noise field
(FNF) ion trap mass spectrometer has been evaluated in the laboratory under controlled
conditions with an environmental test chamber serving as the source of the target
82
-------
compound mixtures at known concentrations. The FNF ion trap technology developed by
Teledyne has been used with direct sampling interfaces (semipermeable membrane; glow
discharge ionization source) to measure nonpolar and polar VOCs at trace levels. This ion
trap is capable of true selective ion monitoring and, when operated in the MS/MS mode,
provides a unique means of simultaneously isolating individual target compounds in
complex mixtures with high sensitivity and specificity. The device is small and light-
weight, and can be easily deployed in the laboratory or the field. Using the combination
direct sampling/ion trap system, experiments have been carried out to evaluate the
specificity, sensitivity, response time, and effects of relative humidity on the detection of
44 nonpolar and 15 polar VOCs of environmental interest.
83
-------
1994
Session 13: Applying Total Human Exposure
Methodologies to Address Environmental
Health Issues Along the U.S.-Mexico
Border
Wednesday Afternoon
APPLYING HUMAN EXPOSURE SCIENCE TO PUBLIC HEALTH
CONCERNS: THE LOWER RIO GRANDE VALLEY ENVIRONMENTAL
MONITORING STUDY EVOLUTION AND DESIGN
G. Akland and O. Ramirez
U.S. EPA
M. Schwab
ManTech Environmental Technology, Inc.
The Environmental Protection Agency, the Centers for Disease Control, and the
State of Texas initiated an environmental monitoring investigation in the Lower Rio
Grande Valley in Texas in response to Valley residents' concerns about the potential rink
between their health and pollution. According to community leaders, potential sources of
contamination include industrial emissions into the air and water from Mexican industry,
extensive agricultural pesticide use, and inadequate infrastructure. But without existing
environmental data, the impact of these concerns on the public health of the community
has not been documented. A pilot field measurement program was conducted during the
Spring and Summer of 1993 to provide preliminary data about the levels, sources, and
pathways of actual human exposure in the Valley. Specifically, samples of indoor and
outdoor air, house dust, soil, food, drinking water, and biological samples from each of
the study participants were collected and analyzed for several compound classes. This
paper explains the community partnership used to design the study, the concepts
underlying the design, the perceived utility and value of the approach chosen, the field
implementation methods, and the implications of the experiences gained through this
process.
84
-------
BALANCING PARTICIPANT BURDEN WITH DATA NEEDS: THE HUMAN
ASPECT OF EXPOSURE STUDIES
Mar go Schwab
ManTech Environmental
2 Triangle Drive
Box 12313
Research Triangle Park, NC 27709
Gerry Akland
U.S. EPA
During the last 15 years, our ability to estimate personal exposure has evolved
substantially. Specifically, assessment techniques currently focus on total exposure; field
studies are now being designed to quantify as many pathways and routes of exposure as
possible. Recent methods have encompassed dermal and ingestion routes of exposure, as
well as inhalation. Quantification of these exposure routes requires estimation of the
concentrations of environmental contaminants in all media with which people come into
contact (i.e., dust and soil around the home and work place, drinking water, food, indoor
and outdoor air) and the extent of contact with these media (i.e., their activity patterns).
Collecting such detailed information at a personal level, however, puts a large burden on
field study participants in terms of both time commitment and invasion of privacy. This
paper uses the experiences gained through the Lower Rio Grande Environmental
Monitoring Study to illustrate how scientists must balance the value of information with
the demands that the data collection process places on the participants. It is recommended
that, in order to facilitate participant cooperation and obtain quality data, the study design
and implementation should explicitly address the participant's needs and the community
context.
85
-------
POLLUTANT CONCENTRATIONS IN AIR, SOIL AND HOUSEDUST
SAMPLES IN THE LOWER RIO GRANDE VALLEY ENVIRONMENTAL
STUDY
R.K. Stevens, G. Akland, S. Mukerjee, R.W. Lewis, and T. Buckley
U.S. EPA
Research Triangle Park, NC
WD. Ellenson, R. Kellogg, and R.W. Willis
ManTech Environmental
Research Triangle Park, NC
L. Shelton
Research Triangle Institute
Research Triangle Park, NC
R. Rasmussen
Oregon Graduate Center
Beaverton, OR
D. Camann
Southwest Research Institute
San Antonio, TX
The study objectives of the Lower Rio Grande Valley Environmental Scoping Study
(LRGVESS) includes the testing and evaluation of sampling and analysis methods to be
considered for full scale population-based human exposure study. As part of this program
a 22-day multi-media monitoring study was conducted during the spring of 1993 in and
around Brownsville, TX. This report describes the results of the sampling and analysis of
air, soil and housedust samples collected at 9 residential locations and at a central air
monitoring station on the campus of the University of Texas at Brownsville. At the
central monitoring site, twenty-two consecutive 24-hour (8am-8pm) samples of aerosols
were collected and analyzed for mass loadings of trace metals, sulfate, pesticides and
polynuclear aromatic hydrocarbons (PAHs). Volatile organic species were also measured
at the central site and at the residences. Housedust samples and aerosol samples collected
at the residences were analyzed for pesticides, trace metals and PAHs. Soil samples
collected near the residences were analyzed for trace metals and content.
Intercomparison of data from aerosol sampling systems designed to collect airborne
pesticides and other air toxic species were an integrated part of this study. Indoor,
outdoor and central site pollutant concentrations were also compared. The results of these
measurements and methods evaluations derived from this study are discussed in
relationship to the development of data quality objectives for the design of future full-
scale human exposure studies in the Lower Rio Grande Valley.
86
-------
METHODS FOR SEMIVOLATILE ORGANICS IN AIR AND HOUSEDUST
USED IN THE SPRING 1993 LOWER RIO GRANDE VALLEY
ENVIRONMENTAL SCOPING STUDY: PERFORMANCE AND DATA
COMPARISONS
R.G. Lewis, S. Mukerjee, andR.K. Stevens
U.S. EPA
Research Triangle Park, NC
D.S. Shadwick, M.C. Somerville, and W.D. Ellenson
ManTech Environmental Technology Inc.
Research Triangle Park, NC
D.E. Camann
Southwest Research Institute
San Antonio, TX
L.S. Sheldon ,
Research Triangle Institute
Research Triangle Park, NC
An essential component of the Lower Rio Grande Valley Environmental Scoping
Study (LRGVESS) was the evaluation and intercomparison of candidate methods for
potential use in a future full-scale study. For this purpose, three methods were compared
for semivolatile organic chemicals (S VOCs), specifically polycyclic aromatic
hydrocarbons (PAHs) and pesticides. They were the high-volume PS-1 sampler (modified
Method TO-04), the low-volume PUF sampler (Method TO-10, ASTM D4861), and the
modified Versatile Air Pollution Sampler (VAPS). All three samplers used quartz fiber
particle filters backed up with polyurethane foam (PUF) vapor traps. Collocated samplers
of the same and different types were operated simultaneously for 24 hours at either the
central site or a residential sites, or both. In addition housedust was collected at each
residence with the HVS3 sampler (ASTM D5438) and analyzed for PAHs and pesticides.
The primary laboratory for PAHs was Research Triangle Institute (RTF) and the
primary laboratory of pesticides was Southwest Research Institute (SwRI). However,
both laboratories analyzed most of the samples collected for both groups of chemicals.
ManTech Environmental Technology, Inc. (METI) operated the samplers at the central
site, while RTI did the residential monitoring.
The data obtained will be statistically compared with respect to reproducibility of a
given method, variance between methods, and interlaboratory differences.
87
-------
RESIDENTIAL-BASED HUMAN EXPOSURE TO METALS, PESTICIDES, PAH,
AND VOCS: RESULTS FROM THE LOWER RIO GRANDE VALLEY
ENVIRONMENTAL PILOT STUDY
T.J. Buckley and E.G. Lewis
U.S. EPA
Atmospheric Research and Exposure Assessment Laboratory (MD-56)
Research Triangle Park, NC 27711
G.G.Akland
U.S. EPA
Office of Modeling, Monitoring Systems, and Quality Assurance (MD-75)
Research Triangle Park, NC 27711
MR. Berry and L. Johnson
U.S. EPA
Environmental Monitoring Systems Laboratory
Cincinnati, OH 45268
M. Schwab
ManTech Environmental
Research Triangle Park, NC
/. Liddle
Centers for Disease Control
Atlanta, GA 30341
D. Kendall
U.S. Food and Drug Administration
Lenexa, KS 66214
An interagency task force lead by the EPA conducted a residential based human
environmental exposure scoping study in the Lower Rio Grande Valley near Brownsville,
TX during the Spring of 1993. As a scoping study, the effort was designed to provide an
indication of the range of multi-media human exposure for multiple classes of pollutants.
In addition, the study was to test measurement methods to support the effective design of
a large-scale human exposure study planned for the region. Monitoring was conducted
sequentially over 24-h periods in nine homes for metals, pesticides, PAH, and VOCs in
the relevant media of drinking water, food, air, soil, and house dust. Based on the
monitoring results, and questionnaire and time-activity information, route-specific
estimates of human exposure are derived. These data are related to body-burden
measurements made in blood and urine. The study design, measurement methods, and a
preliminary human exposure assessment is presented. Human exposure is
phenomelogically explored by examining source information and the covariate analysis
of exposure by media and pollutant class. The significance of these results with respect to
the design of a large population-based human exposure study is discussed.
-------
Wednesday, May 4,1994
Session 14: Particles Studies
Wednesday Afternoon
THE ROLE OF SIZE-DEPENDENT DRY DEPOSITION OF SULFATE
AEROSOL IN A THREE-DIMENSIONAL EULERIAN AIR QUALITY MODEL
Francis S. Binkowski*
Atmospheric Sciences Modeling Division
Air Resources Laboratory, NOAA
Mail Drop 80
Research Triangle Park, NC 27711
Uma Shankar
MCNC
P.O. Box 12889
Research Triangle Park, NC 27709-2889
The Regional Paniculate Model, a three-dimensional Eulerian air quality model, was
developed to investigate aerosol particle issues of importance to the U.S. EPA and to
meet the demands of the Clean Air Act Amendments of 1990. In addition to aerosol
dynamics such as growth and coagulation, the model includes photochemistry, transport,
and deposition. A new formulation of dry deposition as a function of the aerosol size
distribution has been incorporated into the model. This formulation allows for the
representation of dry deposition of total particle number and total particle mass by
deposition velocities specifically formulated for these two quantities as a function of
particle size. Results for the dry deposition of sulfate mass from the new model will be
compared with those from the Tagged Species Engineering Model (McHenry, et al.,
1992) for a variety of local conditions. The behavior of the aerosol size distribution
responding to the new formulation will also be discussed.
McHenry, J.N., Binkowski, F.S., Dennis, R.L., et al. (1992) Atmos. Environ.,
26A:1427-1443.
'On assignment to the Atmospheric Research and Exposure Assessment Laboratory,
U.S. EPA.
SIZE DISTRIBUTIONS PARAMETERS AND HYGROSCOPIC GROWTH OF
AEROSOL PARTICLES BEARING V
J.M. Ondov, F. Divita, and T.L. Quinn
Department of Chemistry and Biochemistry
University of Maryland
College Park, MD 20742
Size-segregated submicrometer aerosol particles were collected with microorifice
impactors (MOI) at three sites in the heavily urban, but nonindustrialized Washington,
89
-------
DC. metropolitan area, during a 40-day period in August and September of 1990, when
atmospheric V was principally derived from commercial and utility oil combustion.
Results for 34 MOI samples, analyzed for V by instrumental neutron activation analysis,
were fit with a least-squares technique which used impactor calibration data to determine
log-normal distribution parameters, i.e., mass median aerodynamic diameter (mmad) and
geometric standard deviation (o ) for fine-particles bearing V. The median mmad for 19
College Park (CP) samples was 0.361 ± 0.006 pirn. At this site mmads for samples
collected in the absence of rain and with V concentrations >0.61 ng/m3 increased
continuously with increasing RH from 56 to 79% according to the equation
d 3 = 0.02168 ± 0.00325 ln(aw) - 0.0130. Mmads for samples collected at the other
sites were characteristically smaller than those determined at CP at comparable RH,
possibly, due to the influence of nearby oil-fired boilers. Vanadium aerosol data for rural
Chesapeake Bay sites will also be presented.
MEASUREMENT AND SPECIATION OF GAS AND PARTICIPATE PHASE
ORGANIC ACIDS IN AN URBAN ENVIRONMENT
Joy Lawrence andPetros Koutrakis
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
Organic acids are important contributors to ambient acidity, in both gas and
paniculate phase. Particulate phase organic acids represent an important fraction of
organic paniculate matter. This paper presents the results of a field study conducted in
Philadelphia, PA, during the summer of 1992, to measure the concentrations of gas and
paniculate phase organic acids. Formic acid was found to be the most abundant gas phase
organic acid, with acetic and propionic acids detected at lower concentrations. Organic
acids constituted approximately 5% of the paniculate fine mass, whereas sulfate and
ammonium constituted 40% and 15%, respectively. Dicarboxylic acids and even-carbon
monocarboxylic acids were found to account for a large fraction of paniculate weak
acidity; odd-carbon monocarboxylic acids accounted for a very small fraction. The
pronounced even carbon preference of the monocarboxylic acid distribution suggests a
biogenic origin; the dicarboxylic acid distribution may suggest that primary emission is
more important than photochemical production. This paper discusses the measurement
and analytical techniques used in this study and the chemistry and origins of organic
acids.
90
-------
RESULTS OF PHILADELPHIA DIESEL PARTICULATE MONITORING
STUDY OCTOBER - NOVEMBER 1993
Breda Phillips
OAQPS,U.S.EPA
MD-15
Research Triangle Park, NC 27711
Tom Lumpkin
AREAL/ORD, U.S. EPA
MD-76
Research Triangle Park, NC 27711
Mike Pleasant
ManTech Environmental Technology, Inc.
P.O. Box 12313
Research Triangle Park, NC 27711
Because of the nature of their use, diesel trucks and buses are predominately found in
urban areas with large populations. A monitoring study was conducted in Philadelphia to
learn more about the contribution of diesel paniculate emissions to PM-10 levels in urban
areas. The study was conducted for 30 days using portable PM-10 saturation monitors at
locations within two bus corridors of the center city. The saturation monitors were
operated daily and covered both morning and evening commuting hours. A dichotomous
sampler was also collocated near one bus corridor and operated on a 24-hour daily
schedule. Hourly bus counts were taken at various locations within the corridors on
different days.
The paper will describe the daily operation of the study, including any problems that
were encountered. PM-10 results will be presented as particulate weight by date, type of
sampler, and site number. Bus count results, meteorologic data, and QA results will also
be discussed. The discussion will conclude by presenting practical applications for what
was learned from the study.
DIURNAL AND ELEVATIONAL VARIATIONS IN OZONE AND AEROSOLS
IN NEW HAMPSHIRE'S CLASS-I AIRSHEDS
L. Bruce Hill
Appalachian Mountain Club
P.O. Box 298
Gorham,NH 03581
George A. Allen
Harvard School of Public Health
665 Huntington Ave.
Boston, MA 02115
Ozone and acid aerosols demonstrate systematic diurnal and elevational patterns in
New Hampshire's Presidential/Dry River and Great Gulf Class-I Airsheds. The AMC and
the Harvard School of Public Health have operated EPA-protocol ozone and Harvard-
91
-------
protocol aerosol monitors at high and low elevation sites on Mount Washington (1917m)
since 1987. Concentrations were provided for the base ozone site (477m) by the
University of Massachusetts.
The 0.12 ppm NAAQS was exceeded three times at the summit and but never at the
base. Weak hourly diurnal variation is recorded at the summit site with maxima near
midnight and minima near mid-day, while the base site exhibits a strong diurnal cycle
peaking near mid-day and with minima just before sunrise. Vertical transsects using a
portable ozone analyzer detected vertical stratification at night. A high elevation
boundary about one kilometer above MSL suggests the Planetary Boundary Layer (PBL).
A low-elevation valley layer is attributed to radiational cooling. Daytime transects
generally demonstrate vertical homogeneity due to thermal mixing.
Daytime fine-mass aerosol concentrations from high and low-elevation sites
correlate well, however the higher elevation sites show consistently lower concentrations.
Relationships between aerosols and ozone suggest diurnal and elevational differences are
due to a combination of photochemical, regional transport, and vertical mixing processes.
TEMPORAL AND SPATIAL CHARACTERISTICS OF PARTICLE MASS IN
METROPOLITAN PHILADELPHIA
G.A. Allen, P. Koutrakis, andH.H. Suh
Harvard University
School of Public Health
665 Huntington Avenue
Boston, MA 02115
WE. Wilson andRM. Burton
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC
Particle mass concentrations (PM2 5 and PM10) were measured daily or every other
day at seven residentially oriented sites in Philadelphia, PA during the summers of 1992
and 1993. Samples were collected for 24-h periods using 10 L- min'1 Harvard Impactors
for mass. Coarse mass concentrations (2.5
-------
Thursday, May 5,1994
Session 15: Stainless Steel Canister Sampling and
Analysis
Thursday Morning
SUMMA CANISTERS-DO THEY NEED TO BE CLEANED FOR TO-14
ANALYSIS?
Rei A. Rasmussen
Oregon Graduate Institute
19600 N.W. von Neumann Drive
Beaverton, OR 97006
Once properly cleaned, should Summa canisters undergo the riggers of cleaning once
again when used to collect ambient air samples? The importance of thorough and proper
cleaning of stainless steel canisters is known for the proper collection and analysis of
ambient air samples. The need for continual cleaning of each canister before use for
collection of ambient air samples when used previously for ambient air will be discussed.
BEYOND CANISTER CLEANING...WHAT ABOUT THE SURFACE
CHEMISTRY?
RJR. Freeman, C.C. Crume, andE.D. Winegar
Air Toxics Ltd.
180 Blue Ravine Road, Suite B
Folsom.CA 95630-4719
There continues to be considerable interest and, in fact, some controversy about how
to clean and certify SUMMA canisters. It has been demonstrated by Rasmussen and
others that the procedures outlined in Section of EPA Method TO-14 will effectively
remove both VOCs and condensed water from a canister if the matrix is ambient air. The
fact is, however, that many projects attempt to extend TO-14 and the use of SUMMA
canisters to matrices other than ambient air. In these instances there is the real possibility
that the sample may be quite acidic and often have a fair amount of non-volatile
constituents. Either will damage the SUMMA passivation and/or leave a chemically
reactive residue on the inner surface of the canister. The once inert "SUMMA" surface is
no longer.
The problem is compounded by the fact that the two most common methods of
"certifying" canisters (TO-14 and TO-12) do not measure the inertness of the SUMMA
surface. They simply measure the residual level of volatiles remaining after the
"cleaning" process. Some laboratories recognize the dangers of using damaged canisters
for ambient air determinations and segregate their canisters into "ambient" and "other".
This approach is not only expensive but often impractical given the degree of
93
-------
sophistication required of the canister tracking program and the fact that "ambient" is a
relative term and is often misapplied to a given set of samples.
There exists a need for a rapid, definitive test for measuring the degree of surface
activity of a canister. The test must meet four criteria: (1) the test solution must be stable
in and of itself, (2) the test must be rapid (i.e., less than 10 minutes), (3) the test must be,
by definition, a GC test and (4) the test must discriminate between "inert, questionable
and active" surfaces.
Results will be presented using three different test solutions and a number of new
canisters from two different manufacturers. Test probe recovery as a function of time
demonstrates the variability in the surface of the new canisters and also allows the
calculation of a statistically derived "inert" surface data base. The tests performed on a
variety of older canisters clearly demonstrates the ability to identify inert, questionable
and active canister surfaces.
This chemical test provides the laboratories an additional means of demonstrating the
inertness of a canister to prospective data users. It adds validity to the data set.
DEVELOPMENT AND VALIDATION OF A HEATED CANISTER-BASED
SOURCE SAMPLING METHOD
Robert J. Crawford
NCASI
David L. Elam
Roy F. Weston, Inc.
Southern Regional Center
P.O. Box 141020
Gainesville, FL 32614-1020
In response to the Clean Air Act Amendments of 1990, the United States pulp and
paper industry through the American Forest and Paper Association (AFPA) has instituted
a program to characterize hazardous air pollutant (HAP) emissions from a variety of
sources at approximately 15 facilities. To meet some of the specific needs of this
program, a method has been developed, based on EPA Method 18, that uses a heated
sampling system to transfer source gas samples to a heated stainless steel SUMMA
polished canister. After sampling, the canister is kept hot in an insulated box and
transferred to an on-site mobile laboratory. All of the analyte system components are also
heated so that the moisture is not allowed to condense in the sample before it is analyzed.
An initial mill screening study, laboratory evaluation/validation, and an EPA Method 301
validation on pulp mill sources have all been completed with acceptable results. This
method is being used to quantitate 26 VOCs, e.g., methanol, acetone, methylene chloride,
chloroform, benzene, 1,2,4-trichlorobenzene, and acrolein.
94
-------
RECOVERY OF OXYGENATED ORGANICS FROM SUMMA CANISTERS
Rei A. Rasmussen
Oregon Graduate Institute
19600 N.W. von Neumann Drive
Beaverton, OR 97006
It is well known that the atmospheric reaction products of many anthropogenic and
biogenic hydrocarbons are oxygenated organics. Many analytical techniques have been
used to monitor and determine ambient air concentrations of these species with limited
success. Collection of oxygenated organics and recovery in various sampling resignes
have been used but because of the sticky nature of these compounds to adhere to surfaces,
analytical results have been questioned. The use of Summa canisters to collect
oxygenated organics and their recovery for analysis will be discussed.
ANALYSIS OF SELECTED POLAR VOLATILE ORGANIC COMPOUNDS VIA
TO-14 AND MODIFIED TO-14 METHOD
H. Wang
ENSR Consulting and Engineering
33 Industrial Way
Wilmingtin, MA 01887
SUMMA canister based sampling and analytical system (TO-14 method) has gained
wide acceptance for the collection and analysis of integrated whole ambient air samples
containing volatile organic compounds (VOCs). This method has been very well
developed for most non-polar volatile halogenated hydrocarbons and hydrocarbons.
The studies are concentrated on suitability of TO-14 method for several polar VOCs,
including ketones, ethers, sulfur contained compounds, etc. Studies on alternate technical
approach and method modification are also conducted. The results on humidity effect,
interaction between parameters, time dependent stabilities, and comparison of different
modifications are presented.
WHY IS IT SO DIFFICULT TO MEASURE TERPENES IN AMBIENT AIR
Rei A. Rasmussen
Oregon Graduate Institute
19600 N.W. von Neumann Drive
Beaverton, OR 97006
Biogenic compounds and in general Terpenes make up a large fraction of the volatile
organic compounds emitted into the ambient atmosphere. The measurement of Terpenes
is of importance in knowing the biogenic contribution to the atmospheric loading of
organic compounds. The ambient concentrations of Terpenes are usually low in the
ambient atmosphere which may in itself result in difficulties in their analysis. The
chemical and physical nature of the Terpenes many also lead to difficulties in their
95
-------
analysis. Ambient air samples collected in Summa canisters under conditions in which
Terpenes should be present, and results of experiments with Terpenes in Summa canisters
will be discussed.
CERTIFICATION OF CANISTER SAMPLERS FOR VOC MEASUREMENTS
AT THE WASTE ISOLATION PILOT PLANT
L.Frank-SupkaandC.F Wu
Westinghouse Electric Corporation
Waste Isolation Division
P.O. Box 2078
Carlsbad, New Mexico 88220
G.C. Meiners
International Technology Corporation
11499 Chester Road
Cincinnati, Ohio 45246
RA. Zimmer
Harding Lawson Associates
2400 ARCO Tower
707 Seventeenth Street
Denver, Colorado 80202
The Waste Isolation Pilot Plant (WIPP) site is designed to demonstrate safe disposal
of transuranic mixed-waste. An air monitoring program has been established at the WIPP
site to verify that volatile organic compounds (VOCs) do not migrate out of the disposal
unit. In this air monitoring program, modified, commercially available, two canister
samplers are used to collect air samples for VOC analysis. Sampler certification, sample
collection, and sample analysis are performed based on the procedures contained in U.S.
EPA's Compendium Method TO-14. The canister samplers are certified for cleanliness
by passing humid zero air through the entire sampling system and collecting a sample in
a canister over a 24-hour period. In addition, each canister sampler is certified for target
compound recovery efficiency by passing a humid calibration gas standard through the
entire sampling system and collecting a sample in a canister over a 24-hour period.
This paper discusses the techniques developed for meeting the stringent certification
requirements of the monitoring program and presents a statistical evaluation of the
certification data obtained over the past two years.
96
-------
Thursday Afternoon
A FULLY AUTOMATED SUMMA CANISTER CLEANING SYSTEM FOR
METHOD TO14
D.B. Cardin
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
Cleaning SUMMA passivated canisters used in EPA method TOM is required to
prepare them for reuse as a field sampling container. This process requires several
cleanup and quality assurances steps, including initial leak checking of the manifold after
canister attachment, preliminary evacuation of the manifold with valves closed to insure
leaktight valves, cycling between filling and evacuation to flush out contaminants, high
vacuum pump down to finish contaminant removal, and standby monitoring to verify the
absence of leaks in the canister welds. This procedure can require a lot of technician
attention and is subject to human error when cleaning large numbers of canisters.
A system is presented that automatically performs the functions described above
while recording all pressures and vacuums to a QA report. The only interaction required
by the operator is to open the canister valves after the first round of valve leakchecking is
completed. The canister cleaner uses a molecular drag pump to perform evacuation of 8
to 12 canisters simultaneously down to 10-20 mtorr. Flexible, all stainless tubing is used
to connect each canister to the 3/8" diameter manifold allowing canisters of virtually any
size and shape to be cleaned. Canister heating and humidification of flush gas are
performed to assist in displacing heavy or polar VOCs from canister surfaces.
Discussions will be focused around the improvements in quality assurance and cleaning
consistencies relative to manually operated systems.
NETWORKING AN ENTIRE TO14 LABORATORY INTO A SINGLE
WINDOWS BASED CONTROL SYSTEM
D.B. Cardin, J.T. Deschenes, andEA. Galoustian
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
EPA Compendium Method TOM specifies the analysis of Volatile Organic
Compounds (VOCs) in ambient air as collected in SUMMA passivated stainless steel
canisters. In order to perform Method TOM, provisions must be made for field collection
of samples in canisters, preconcentration and analysis of the VOC fraction in the canister,
cleaning of the canisters for reuse, and preparation of analytical standards for method
validation and GC/MS response factor determination. The temporary or long term loss of
any of these capabilities within a laboratory will cause a bottleneck and ultimately delay
operations until alternatives can be found. Even when successfully completing each of
97
-------
these tasks, a laboratories ultimate efficiency and profitability will be determined by
sample throughput which will be hindered by the task that is the least automated.
A novel approach to instrument communications is presented that makes it possible
to monitor and control field sampling, canister cleaning, standard preparation, and
preconcentration and analysis systems in a 7014 laboratory using a single IBM
compatible PC operating under Windows™. The network permits communication with
microprocessor based devices, such as 32-position aldehyde/sorbent tube samplers and
16-position canister field samplers, for transfer of new program information and recovery
of field collected data. Canister cleaning operations are improved by providing feedback
and control of leak-checking operations, mass dilution canister cleaning, and final high
vacuum evacuation without user intervention. Generation of standards by dynamically
diluting an N1ST traceable ppm level standard to ppb levels is supported with provisions
for autocalibration of mass flow controller channels for ultra-accurate VOC blends. Full
communication and control of a 3-stage TO14 preconcentrator is also supported with
extended QA/QC protocols and diagnostics for improved uptime and performance. An
overview of the system will be presented with primary attention paid to enhancing TOW
laboratory productivity.
ADVANCED SURFACE TREATMENT AND CLEANING TECHNIQUES FOR
THE U.S. EPA METHOD TO-14 GRAB SAMPLING CONTAINERS
Joseph Krasnec
Scientific Instrumentation Specialists
P.O. Box 8941
Moscow, ID 83843
Increased demands and an expanding list of toxic organic compounds, including
oxygenated organics and other special groups, e.g., organic sulfur compounds, require
modifications and improvements to the existing sampling and measurement hardware.
One critical area is the passivated stainless steel surface of the grab sampling containers.
The established and proven electropolishing (SUMMA passivation) works well for the
sampling of hundreds of organic and inorganic volatile compounds. However, there are
instances (i.e., some oxygenated organics) where the normal passivation falls short of the
required stability and storability requirements. Recent R&D efforts show some promising
avenues of improvement for the surface treatment and cleaning of the sampling
containers. The passivated surface can be coated with several inorganic materials to
enhance its performance. The initial work shows performance improvements for some
groups of organics, but not an across-the-board enhancement. The effect of surface
saturation with water vapor and other materials has also been investigated. Some novel
surface cleaning techniques have been explored with encouraging results. This paper will
attempt to bring the audience up to date on some of the above discussed efforts.
-------
Thursday, May 5,1994
Session 16: Ambient Air Measurements of VOCs
Thursday Morning
DETERMINATION OF VOLATILE ORGANIC COMPOUNDS IN AMBIENT
AIR WITH GAS CHROMATOGRAPH-FLAMEIONIZATION AND ION TRAP
DETECTION
Skill Liu, Robert J. Carley, Jianshi Kang, Jianping Chen,
and James D. Stuart
Environmental Research Institute
University of Connecticut
Rte.44
Longley Building
Storrs, CT 06269-3210
Two new techniques are utilized to integrate the following equipment: an Entech
2000 automated air concentrator, a Hewlett Packard gas chromatograph (GC) with flame
ionization detector (FID) and an ion trap mass spectrometer detector (ITD). This
combined analytical system is used to determine low ppbv level VOCs in ambient air.
The first technique is to reconfigure the inlet system of the GC, so that the pressure
regulated flow control system of the GC injection port is used to control the flow of both
the desorb gas of the air concentrator and the carrier gas of the GC column. The injection
port still can be used to inject gas and liquid samples directly.
The second technique is to split the effluent of GC column at 1:1 ratio to the ITD and
the FID. In this way, both FID and ITD data can be obtained for each analysis. For
ambient air non-methane hydrocarbons monitoring, the FID detector is widely used.
Oxygen containing and halogenated organic compounds can not be differentiated by FID
detector and would be quantified as coeluting hydrocarbons. However, VOCs other than
target hydrocarbons can be identified by ITD. This analytical system is a very valuable
research tool for non-methane hydrocarbons and urban toxic monitoring. The
performance of this developed system will be presented.
EXPOSURE TO EVAPORATIVE GASOLINE EMISSIONS
Clifford P. Weisel and Krishnan R. Mohan
EOHSI
Robert Wood Johnson Medical School
UMDNJ
681 Frelinhuysen Road
Piscataway, NJ 08854
Exposure to evaporative gasoline emissions can occur in microenvironments or
during specific activities and can contribute significantly to the daily exposure individual
99
-------
receive. Three microenvironments influenced by evaporative gasoline emissions were
evaluated, residential garages, a public parking garage and activities surrounding
refueling of an automobile. In addition exhaust emissions contribute to public parking
garage air concentrations. Measurements of the benzene and toluene levels were made
within these microenvironments using a portable gas chromatograph (GC) to evaluate
temporal changes and/or by collecting a sample using an adsorbent trap. The latter was
analyzed by thermal desorption/gas chromatography/mass spectrometry to provide
integrated air concentration measurement. The GC/MS analysis also provided a quality
control check on the peak identification and concentration measurement of the portable
GC. The benzene and toluene air concentrations measured within the garages were
elevated compared to ambient levels. The degree of elevation increased with temperature
and use of an older automobile entering the a residential garage. The benzene and toluene
concentrations within the public parking garage were also greatest during the summer,
when the temperatures were elevated, with the benzene concentration ranging between 25
Hg/m3 and 250 |J.g/m3. The benzene exposure to individuals within service stations and
within an automobile while refueling were also elevated compared to other automotive
related activities. An individual refueling his/her own vehicle had five minute integrated
exposures of 200 (J.g/m3
HYDROCARBONS IN THE C8-C20 RANGE MEASURED DURING COAST
STUDY IN TEXAS
B. Zielinska, J. Sagebiel, G. Harshfield, andL.H. Sheetz
Desert Research Institute
P.O. Box 60220
Reno, NV 89506
WJ. Hauze
North American Weather Consultants
1293 West 2200 South
Salt Lake City, Utah 84119
J.H. Price
Texas Natural Resource Conservation Commission
12124 Park 35 Circle
Austin, TX 78753
Hydrocarbons in the C8-C20 range were collected using Tenax-TA solid adsorbent
cartridges during the Costal Oxidant Assessment for Southern Texas (COAST) field
study, carried out in the Houston-Galveston and Beaumont-Port Arthur areas. Ambient
samples were collected at the Galleria, Clinton and Port Arthur sampling sites, from
August 26 to August 30,1993. In addition, a number of Tenax samples were collected in
the vicinity of the main industrial complexes (such as Exxon, Amoco Chemical, Union
Carbide, Dow Chemical) and in an urban tunnel. All samples were analyzed with high-
resolution capillary-column gas chromatography with flame ionization detection
(GC/FID) for quantification and selected duplicate samples were analyzed with combined
gas chromatography/Fourier transform infrared/mass spectrometry (GC/FTTR/MS) for
100
-------
identification of individual compounds. The results of these analyses will be presented.
THE DETERMINATION OF HAZARDOUS AIR POLLUTANTS WITH A
BUILT-IN PRECONCENTRATOR AND CAPILLARY GC
Norman Kirshen and Elizabeth Almost
Varian Chromatography Systems
2700 Mitchell Drive
Walnut Creek, CA 94598
510-945-2230
FAX 510-945-2335
Under the 1990 Clean Air Act Amendments industries identified as major sources of
hazardous air pollutants must meet Maximum Achievable Control Technology standards
within a certain time frame. The resultant reduction of emissions is enforced under the
Title V permit program. This reduction may require monitoring of the source as well as at
the fenceline of the facility precursors. Multiple location monitoring is required during
the remediation of certain hazardous waste sites to track the air pathways ensuring that
pollutants are not transferred outside the site. Gas Chromatography and Gas
Chromatography/Mass Spectrometry in combination with air sample reconcentration is
the primary analytical technique used for these monitoring requirements since it provides
the best sensitivity and quantitative reliability. One problem with this approach is the
size as well as the complexity of these preconcentrator/GC systems.
A hazardous air pollutant GC system has been configured with a new built-in sample
preconcentration trap (SPT) and associated varying capable of preconcentrating air
samples of variable volumes. Since the preconcentrator is built into the GC, the system
has a small footprint, simple interfacing, and control from one PC. The air sample is
drawn either from a canister or directly from the ambient air through SPT adsorbent trap.
The adsorbent trap is cooled to initial conditions using the option of a cryo or non-cryo
technique. Following trapping, the air toxics are quickly desorbed with the fast heating
(40° C/sec) SPT to a wide bore capillary column with detection by FID, PID, ECD, or a
combination of detectors. No column oven cryogenics are required.
Optimization of the trapping and chromatographic parameters has been performed
using a TO-14 air standard. The results of these studies and a description of the integrated
system will be presented.
AN INTEGRATED APPROACH TO PARTS-PER-TRILLION MEASUREMENT
OF VOLATILE ORGANIC COMPOUNDS IN AIR
Eric D. Winegar, RJl. Freeman, and C.C. Crume
Air Toxics Limited
180 Blue Ravine Road, Suite B
Folsom,CA 95630-4719
The quest for ever lower detection limits continues, with part-per-trillion (pptv)
101
-------
measurements of volatile organic compounds becoming the latest "Holy Grail" in air
quality measurements. In particular, the 0.1 ppbv level is a recent defacto benchmark in
analytical performance. A unique combination of factors must be considered in any
measurement made at this level. These factors include canister cleaning and certification,
advanced analytical technologies, and adequate quality assurance. Assuming a 0.1 ppbv
detection limit, the canister and sampling system must be cleaned to at least that level or
below. The analytical system must have correctly derived detection limits, and the usual
quality assurance requirements (calibration using accurate standards, precision and
accuracy, etc.) must be met for this level while taking into account the usual limitations
such as inherent imprecision associated with each step in the process. Each of these
factors alone will not assure success; it is the combination that provides the requisite
analytical accuracy and precision.
High quality ambient air sampling and analysis is the obvious recipient of such
performance goals, especially in light of the current unit risk factors in use for risk
assessment. This paper will examine some of the factors that must be considered in order
to generate pptv data. Specific measurements of low level concentrations of volatile
organic compounds will be presented using this method, including ambient and indoor
air, and emissions from various consumer products such as textile floor coverings,
computers, and misc. other materials in an environmental chamber. An integrated
experimental approach to the sampling and analysis of pptv concentration values will be
presented.
VOC QUALITY CONTROL MEASUREMENTS IN THE CASTNET AIR
TOXICS MONITORING PROGRAM (CATMP)
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
The determination of volatile organic compounds (VOCs) in ambient air with
acceptable precision and accuracy at concentration levels as low as 0.1 ppbV is the
primary analytical requirement of the Environmental Protection Agency's (EPA's) Clean
Air Status and Trends Network (CASTNet) Air Toxics Monitoring Program (CATMP).
The CATMP was established in 1993 by EPA to reactivate and operate the Urban Air
Toxics Monitoring Program (UATMP). The purpose of the program is to establish
baseline toxics concentrations, develop air emission inventories, and to identify air toxic
sources using chemical mass balance modeling techniques.
Samples are collected from 15 urban sites in passivated stainless steel canisters every
12 days and analyzed with a gas chromatograph/mass spectrometer (GC/MS) operated in
the electron impact full-scan mode. 53 target VOCs are determined. Analytical
procedures and instrument operating parameters are described in detail. The paper
focuses on the analytical quality control (QC) measurements which accompany every
batch of samples analyzed. These include the analysis of instrument performance check
102
-------
standards, calibration standards, MIST reference samples, canister blanks, replicates, and
internal standards/surrogates. Results of QC measurements performed during 1993 are
presented, including precision and accuracy determinations. Data for selected VOCs
determined in network samples are also presented.
AN AUTOMATED GC TECHNIQUE FOR ENHANCED DETECTION OF
ORGANIC COMPOUNDS MONITORED USING PASSIVE ORGANIC VAPOR
BADGES
KochyFung
AtmAA, Inc.
23917 Craftsman Road
Calabasas, CA 91302
The principle use of organic vapor badges is for monitoring of exposure to organic
compounds in the workplace, where the concentrations of target species are typically in
ppmv to sub-ppmv range. When this passive device is used for monitoring of organic
compounds in ambient air, the analytical method used must have good sensitivity and
precision to allow the small amount of compounds collected to be measured. A Hewlett
Packard 5840A gas chromatograph with an autosampler was configured for two-
dimensional gas chromatography with intermediate cryo-trapping to measure species like
benzene, toluene, PERC and MEK. The autosampler gave high injection precision. The
primary column was used for separating the large solvent peak from target compounds.
Heart-cutting and cryo-focusing allowed target species to be analyzed with capillary
column with good sensitivity. The technique was used in several monitoring programs
and showed very good precision. Findings will be presented and discussed.
BACKGROUND MONITORING OF AIR TOXICS AT THE WASTE
ISOLATION PILOT PLANT
L. Frank-Supka, C.F Wu, andRM. Lopez
Westinghouse Electric Corporation
Waste Isolation Division
P.O. Box 2078
Carlsbad, New Mexico, 88220
R.A. Zimmer
Harding Lawson Associates
2400 ARCO Tower
707 Seventeenth Street
Denver, Colorado, 80202
The Waste Isolation Pilot Plant (WIPP), located southeast of Carlsbad, New Mexico,
has been constructed to demonstrate safe disposal of containerized solid or solidified
transuranic (TRU) mixed-waste. The respository is constructed in a massive salt bed
formation, 2,150 feet below the surface. An air monitoring program for detecting
103
-------
airborne release of volatile organic compounds (VOCs) is required to verify that there
will be no-migration of toxic constituents from the site boundary to the environment.
Periodic background monitoring for VOCs has been performed at the WIPP since
1992. The monitoring program routinely quantifies airborne concentrations of five target
VOCs in the ventilation airstream of this underground facility, using the U.S.
Environmental Protection Agency's Compendium Method TO-14. This paper describes
the monitoring program in-place at the facility and presents a summary of the monitoring
results for the last two years.
EXPERIENCE IN THE OPERATION OF A PAMS TYPE 2
lhab H. Farag and Chunming Qi
University of New Hampshire (UNH)
Chemical Engineering Department, Kingsbury Hall
Durham, NH 03824-3591
Paul Sanborn, Tom Noel, and Dennis Lunderville
New Hampshire Department of Environmental Services (DBS)
New Hampshire Air Resources Division
Concord, NH 03302-2033
The 1990 Clean Air Amendments (CAAA) mandate the U.S. EPA to work with the
states to achieve cleaner air. Ozone, a primary ingredient of smog, is not emitted directly
into the air. It is created photochemically through the action of sunlight on the precursor
pollutants nitrogen oxides (NOX and volatile organic compounds (VOCs).
In the Seacoast area of New Hampshire, the ground level ozone is classified as
serious. In response to the CAAA, an enhanced ozone monitoring station is to be
installed. UNH and DBS are working together on the installation and operation of a
photochemical assessment monitoring station (PAMS) in the Portsmouth, NH area.
This paper addresses our experience in installing and operating the hydrocarbon
portion of the monitoring. A continuous Perkin Elmer gas chromatograph (GC) is being
installed and tested, and is to be situated downwind of Portsmouth (Type 2 site). The
paper will address several issues, such as the equipment used, sample probe and
manifold, VOC sampling, GC standards, support gases, gas consumption, data handling,
and the telemetry and remote retrieval of data.
COMPARISON STUDIES OF OZONE PRECURSORS IN PHOENIX, ARIZONA
Carmo Fernandez, Jim Guyton, and Cheng Peter Lee
Arizona Department of Environmental Quality
Phoenix, AZ 85012
Sucha Parmar
Atmospheric Analysis & Consulting Company
Ventura, CA 93003
This paper will present the comparison of the ozone precursors monitoring program
104
-------
for Phoenix, Arizona during 1992 and 1993. Specific details and methodologies will be
presented involving collection of air samples and analysis of speciated measurements for
reactive VOC and carbonyl precursors responsible for ozone formation. Quality control
and quality assurance techniques will also be discussed.
IMPLEMENTATION OF A RISK-BASED AIR MONITORING PROGRAM
USING INTEGRATED AND CONTINUOUS AIR MONITORS
Wen-Whai Li, Stephen T. Washburn, and Mary E. Greenhalgh
ENVIRON Corporation
210 Carnegie Center, Suite 201
Princeton, NJ 08504
A risk-based air monitoring program was conducted during limited excavation
activities at a Superfund site. The objectives of the air monitoring program were to: (1)
ensure that the air emissions during excavation activities would not adversely impact the
health and safety of workers and the surrounding community; (2) evaluate the reliability
of continuous VOC and paniculate air monitoring equipment; (3) evaluate the basis and
feasibility of preliminary risk-based action levels developed for continuous monitoring;
and (4) provide a basis for revising air monitoring instrumentation and procedures, as
necessary.
Personal air samples were collected within the exclusion zone using integrated and
continuous air samplers. Ambient air monitoring was conducted at seven stations using
nine integrated air samplers and three continuous air monitors. Three mobile stations
were positioned at the perimeter of the work area, and four permanent stations were
installed along the site boundary. Integrated air samples collected each day were
evaluated on the basis of the site activities, results from the concurrent continuous air
monitoring, and the previously developed health-based action levels to determine if
further laboratory analysis was necessary.
A TWO CHANNEL, 16-POSITION CANISTER FIELD SAMPLER FOR
IMPROVED PERFORMANCE AND QUALITY ASSURANCE
D.B. Cardin and J.T. Deschenes
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
Current automated field samplers used to sequentially collect whole air into multiple
SUMMA canisters contain many deficiencies that add suspecion to data quality.
Mechanical mass flow controllers used have been shown to have temperature dependent
flow rates resulting in different flow rates during day and night operation. With the
absence of dynamic, on-line pressure measuring devices, canisters that have leaked since
being evacuated in the laboratory would be analyzed with data reported without question.
In addition, the lack of data showing whether sample flow into the canister dropped off
105
-------
before the end of the sampling period due to leak prepressurization of the canister or
inconsistent sampling flow rates can result in a sample that may or may not be
representative of the local VOC concentration during the entire requested sampling time.
A new programmable field sampler is presented that records sampling information
for later retrieval by a Windows based PC. Collected data can include flow rate into each
canister during the entire sampling event, beginning and ending pressures in the canisters,
feedback verification that the correct canister was being filled at the correct date and
time, and even the optional flow rate of surrogate into the canisters during sampling.
Flow rates are controlled using electronic mass flow controllers that are unaffected by
changing temperatures and pressures, with the added luxury of being able to record actual
rather than theoretical flow rates. The canister sampler can be set up for collecting
samples literally once a minute to once every week, depending on the mass flow
controllers installed and the size of the canisters used. Samplers can be chained together
so that the next sampler will commence as soon as the previous sampler has completed
all 16 positions. Monitoring and control of the sampler can be achieve via modem link to
improve implementation efficiency. Data collected during field sampling operations will
be presented as well as a new procedure for unattended dilution sampling of high
concentration samples (stack gas, landfill gas, etc.) which keeps target analytes well
below saturation points for improved stability and recovery from canisters without
heating.
106
-------
Thursday, May 5,1994
Session 17: Air Pollutants in General
Thursday Morning
TEMPORAL VARIATION OF FINE PARTICLE MASS AT TWO SITES IN
MEXICO CITY
Paulina Serrano and George Allen
Harvard School of Public Health
Boston, MA
Margarita Castillejos
Universidad Autdnoma Metropolitana
Xochimilco, D.F., Mdxico
Diane Gold and Frank Speizer
Brigham and Women's Hospital
Boston, MA
Mauricio Hernandez
Institute Nacional de Salud Publica
Cuernavaca, Mor., Mexico
Carl Hayes and William McDonnell
US EPA
Research Triangle Park, NC
Simultaneous sampling of fine mass (PM2 5, using an integrated 24 hour gravimetric
method) and the particle scattering extinction coefficient (b , using a heated integrating
nephelometer) were used to estimate continuous fine particle concentration at two sites in
Mexico City. Linear regression analysis of the 24 h averages of b and the PM2 5
integrated samples was done on a seasonal basis. The coefficients of determination (R2)
between these methods ranged from 0.84 to 0.90 for the different sampling periods.
These data are the first attempt to describe the diurnal variation of fine mass in
Mexico city. Distinct and different diurnal patterns were observed for both sites. For the
site located near an industrialized area, a sharp peak occurred between 0700 and 0900
hours and a second smaller but broader peak occurred late at night. This site is
characterized by the presence of primary pollutants, with PM10 annual mean
concentrations exceeding 150 ng-m"3.
The second site is located in a residential area down wind of the industrialized area,
and is characterized by the presence of secondary pollutants with much lower PM10
concentrations (annual mean of under 50 jig-m"3). The diurnal fine mass pattern at this
site had a broad peak between 0900 and 1200 hours. On individual days, fine mass was
sometimes highly correlated with ozone.
107
-------
ABILITY OF FIXED MONITORING STATIONS TO REPRESENT
COMMUTER'S EXPOSURE TO CO REVISITED: THE CASE OF MEXICO
CITY
Adridn F erndndez-Bremauntz
Harvard School of Public Health
The aims of this study were to determine in-vehicle CO levels in major commuting
modes in Mexico City, to identify the main factors affecting variation in these CO
concentrations and to compare in-vehicle measurements with concurrent CO
concentrations measured at fixed-site monitoring (FSM) stations.
CO concentrations inside private and public transport vehicles were consistently
higher than those measured at FSM stations and indeed much higher that those reported
in previous commuter studies in U.S. cities. Relationships between in-vehicle and FSM
measurements were analyzed through ratios and regression models. There were
significant differences in CO concentrations and travel speeds between transport modes.
The highest CO concentrations were found inside autos and minibuses, while metro
trains, trolley-buses and buses had lower concentrations. The time at which the trips were
made (AM or PM) and the route of travel also affected in-vehicle CO concentrations.
INVESTIGATION OF SPATIAL AND TEMPORAL PATTERN OF OZONE
CONCENTRATION WITHIN A METROPOLITAN AREA USING OZONE
PASSIVE SAMPLER
L.-J. Sally Liu andPetrosKoutrakis
Harvard School of Public Health
Department of Environmental Health
665 Huntington Avenue
Boston, MA 02115
Irvine Broder
The Gage Research Institute
223 College Street
Toronto, Canada M5T 1R4
The Canadian Research on Exposure Assessment Modeling (CREAM) pilot studies
were conducted during the winter and summer of 1992 in Toronto, Ontario, Canada. In
the pilot studies, personal, indoor, and outdoor samples were collected using passive
ozone (03) samplers. Personal 03 samples were taken from 89 participants from 50
different households. Indoor 03 samples were taken from participants' homes and work
places and a variety of non-residential buildings. The indoor samplers were located on a
"sampling tree" that included a fan to maintain a constant air flow. Outdoor O3 samples
were collected by placing passive samplers outside participants' homes. Outdoor O3
concentrations also were monitored simultaneously by continuous monitors at the 21
stationary ambient monitoring sites operated by the Ministry of Environment throughout
Metropolitan Toronto, Central Ontario, and Hamilton. Approximately 20 duplicate
passive samples per week were taken for quality control and quality assurance purposes.
108
-------
Air exchange rate of the homes were taken weekly using perfluorocarbon tracer gas
method. Participants were asked to complete time-activity diaries throughout the
monitoring period.
The performance of the passive sampler is evaluated for various weather conditions
and various applications. Spatial variation of outdoor O3 concentrations is examined
using standardized scores and analysis of variance (ANOVA) techniques. Factors
affecting O3 outdoor spatial variation, indoor O3 concentrations, and personal O3
exposures are examined. A personal O3 exposure model is developed by using home
outdoor, home indoor, work place, and other microenvironmental O3 concentrations.
A COMPARISON OF OZONE AND ACID AEROSOL EXPOSURE PATTERNS
IN A SUMMERTIME STUDY OF METROPOLITAN PHILADELPHIA
Jed M. Waldman and Chris S-K Liang
Environmental & Occupational Health Sciences Institute (Room 301)
Division of Exposure Measurement & Assessment
681 Frelinghuysen Road, Piscataway, NJ 08855
Petros Koutrakis, Helen Suh, and George Allen
Harvard School of Public Health (Bldg 1 - Gil)
Environmental Science and Engineering Program
665 Huntington Avenue
Boston, MA 02115
Robert Burton and William E. Wilson
U.S. EPA (MD-56)
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
A study of acid aerosol chemistry and exposure patterns was conducted in
metropolitan Philadelphia between June 1992 and September 1993. Included in the study
design were components of daily monitoring at a network of 7 sites. The resultant dataset
includes acid aerosol species measured with HEADS (Harvard/EPA Annular Denuder
Sampler), such as paniculate strong acidity (H+) and sulfate (SO4=), as well as ammonia
(NH3), nitric acid (HNO3)and sulfur dioxide (SO2). At the same time, hourly data for
ozone (O3) are available. Previous analysis of the HEADS data have shown fairly
homogeneous concentrations for SO4= aerosol among the 7 sites. Simultaneously, aerosol
H+ levels were observed to decrease in proportion to proximity to the city center. The
current paper will discuss patterns of the other pollutant components measured, chiefly
03. The impact of oxidant levels on the acid aerosol patterns will be evaluated. Also, the
temporal patterns of the two lung irritants in concert (H+ and O3) will be considered with
respect to the combined exposure and potential for synergistic impacts.
109
-------
DIURNAL AND SPATIAL VARIATION IN FINE AND COARSE PARTICLE
CONCENTRATIONS IN METROPOLITAN PHILADELPHIA
H.H. Suh, GA. Allen, P. Koutrakis
Harvard University
School of Public Health
665 Huntington Avenue
Boston, MA 02115
W.E. Wilson, R.M. Burton
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Particle mass (PM2 5 and PM10) concentrations were measured in metropolitan
Philadelphia during the summer of 1992, as part of a larger effort to characterize acid
aerosol concentrations within urban environments. Sampling was performed
simultaneously at seven sites located with metropolitan Philadelphia and at a rural site
approximately 18 miles from the city center. Sites were selected based on their
population density and on their relative locations within Philadelphia. Particle sampling
at the eight sites was performed on alternate days, with sampling conducted over 24-h
periods beginning at Sam. All samples were collected using 10 L-miiT1 Harvard
Impactors. At one of the metropolitan sites, additional particle mass measurements were
made daily using continuous methods.
In this paper, we examine and compare the temporal and spatial variation in fine
(da<2.5 |j.m) and coarse (2.5
-------
INDOOR AIR CHEMISTRY: FORMATION OF ORGANIC ACIDS AND
ALDEHYDES
Junfeng Zhang and Paul J. Lioy
Environmental and Occupational Health Sciences Institute (EOHSI)
UMDNJ - Robert Wood Johnson Medical School and Rutgers University
681 Frelinghuysen Road
P.O. Box 1179
Piscataway, NJ 08855-11792
William E. Wilson
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA, MD-75
Research Triangle Park, NC 27711
Laying emphasis on the formation of aldehydes and organic acids, the study has
examined the gas-phase reactions of ozone with unsaturated VOCs. The formation of
formaldehyde and formic acid was observed for all the three selected unsaturated VOCs:
styrene, limonene, and 4-vinylcyclohexene. In addition, benzaldehyde was detected in
the styrene - ozone - air reaction system, and acetic acid was also found in limonene -
ozone -air system. The study has also examined the gas-phase reactions among
formaldehyde, ozone, and nitrogen dioxide and found the formation of formic acid. The
nitrate radical was suggested to play an important role in converting formaldehyde into
formic acid. Experiments for all the reactions were conducted by using a 4.3 m3 Teflon
chamber. Since the conditions for the reactions were similar to those for indoor
environments, the results from the study can be implicated to real indoor situations and
can be employed to support the findings and suggestions from the previous studies:
certain aldehydes and organic acids could be generated by indoor chemistry.
OZONE REACTIVE CHEMISTRY IN INDOOR RESIDENTIAL
ENVIRONMENTS
Richard Reiss, P. Barry Ryan, Petros Koutrakis, and Sarah Bamford
Harvard University
School of Public Health
Recent laboratory studies have shown that ozone can react in indoor environments to
form several organic species. Our field study was designed to verify these laboratory
experiments with field data. Ozone, organic acids, aldehydes and ketones were measured
over an approximately 24 hour period in 4 residences during the winter and in 9
residences during the summer. All residences were in the Boston, Massachusetts area.
Additionally, several other variables that may affect indoor pollutant concentrations were
measured, including the temperature and relative humidity (both indoor and outdoor) and
the air exchange rate. The correlations between these variables and the ozone removal
make the analysis of the ozone reactions difficult Nonetheless, plausible mechanisms for
VOC formation via ozone reactions are presented, and it is shown that the production of
these VOCs is correlated with the ozone removal.
ill
-------
CASTNET AIR TOXICS MONITORING PROGRAM (CATMP): VOC AND
CARBONYL DATA FOR JULY, 1993 THROUGH MARCH, 1994
David P. Harlos and Eric S. Edgerton
Environmental Science & Engineering, Inc.
P.O. Box 1703
Gainesville, FL 32602-1703
The U.S. EPA has, under the auspices of the CASTNet program (Clean Air Status
and Trends Network), initiated the CASTNet Air Toxics Monitoring Program (CATMP).
Volatile Organic Compounds (VOC) and carbonyls and metals are sampled for 24-hour
periods on a 12-day schedule using TO-14 samplers (SUMMA canisters) and
dinitrophenylhydrazine-coated (dmph) sorbent cartridges and high volume particle
samplers. Sampling was begun at most sites in July of 1993. The sites are operated by
state and local air pollution control programs and all analysis is performed by
Environmental Science & Engineering (ESE) in Gainesville, Florida. The network
currently supports 15 VOC sites, of which 7 also sample carbonyls. Three sites sample
metals only in Pinellas County, Florida. Analytical methods for the network are
discussed in two other papers in this symposium (Winslow, 1994; Prentice, 1994). The
limits of detection of 0.05 ppb for VOCs allow routine tracking of a wide range of
pollutants including several greenhouse gases, transportation pollutants and
photochemically-derived compounds. The sites range from major urban areas (Chicago,
St. Louis) to a rural village (Waterbury, Vermont). Results of the first three quarters of
VOC and carbonyl data collection are summarized in this presentation.
SOURCES AND FACTORS INFLUENCING PERSONAL AND INDOOR
EXPOSURES TO PAHS AND PHTHALATES
Haluk Ozkaynak, Jianping Xue, and John D. Spengler
Department of Environmental Health
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
During the fall of 1990, a large-scale field monitoring program for personal exposure
to PM10 was conducted in Riverside, California by Research Triangle Institute, Harvard
School of Public Health and the Accurex Corporation. The pilot PTEAM (Particle Total
Exposure Assessment Methodology) study, co-sponsored by the EPA and the California
Air Resources Board, collected personal exposure data on PM10 and elemental mass for
175 residents of Riverside. During this study, indoor and outdoor concentrations of
PM10, PM2 5, and elements were also collected at participants' homes; PAH and
phthalate data were collected in a subset of 125 homes. Twelve-hour recall time-activity
diaries and questionnaires regarding exposure to ETS and other sources of particles,
PAHs and phthalates were also obtained and later uses in exposure modeling.
Measurements showed that most PAHs and phthalates had high detection rates:
60-100%. Aside from smoking, analysis did not indicate cooking, spraying, house
112
-------
cleaning activities or proximity to busy roadway as possible sources of PAHs or
phthalates. Both physical and empirical statistical models were used to estimate the
contributions of outdoor sources, cigarette smoking, and other indoor sources. Results
from modeling showed that: (1) The physical models fit the PAH data well. Coefficients
between model predicted concentrations and observed concentrations averaged about 0.7.
(2) Penetration factors for most PAHs were found to be very close to one. (3) The
estimated average decay rates for PAHs ranged from 0.4 to 1.6 per hour with sizable
variation. (4) Smoking contributed 20-40% of the total concentrations of eight PAHs in
homes reporting smoking. (5) Smoking was not estimate to be an indoor source of
phthalates. (6) In the entire set of homes, outdoor air contributed more than half the total
concentrations of six PAHs, mostly the less volatile ones, and "other" (unidentified)
indoor sources contributed more than half of three volatile PAHs.
113
-------
Session 18: Indoor Air Pollution
Thursday Afternoon
GAS AND PARTICULATE PHASE ACIDS, ORGANIC COMPOUNDS AND
OXIDANTS IN A SICK ROOM
Kris Wardrup, Laura Lewis, and Delbert J. Eatough
Department of Chemistry
Brigham Young University
Provo, UT 84602
Employees in a single large room of the administration building at Brigham Young
University began experiencing respiratory problems in the Spring of 1993. The cause of
the symptoms was not apparent from a review of the area by campus safety personnel.
Accordingly we undertook a study of the atmospheric environment in the building to see
if we could identify any causes for the appearance of a "sick room" problem.
The concentrations of fine paniculate phase mass, sulfate, nitrate, ammonium ion
and acidity were determined using diffusion denuder sampling techniques. The
corresponding concentrations of gas phase SO2, HNO3 and HNO2 were also determined
from the denuder sampling results. Concentrations of NO, NO2,03 and formaldehyde
were determined using appropriate Dra'ger absorption tubes. The concentrations of total
volatile organic material was determined using a charcoal sorbent filter. The
concentrations of each of these species were determined in the "sick room", at two
control locations in the building (a copy center and an administrative office) and in the
outdoor environment as a function of time of day for six different days. Subsequently,
temperature and humidity were also monitored in the various study areas.
The symptoms experienced by the personnel in the room appear to be associated
with slightly elevated temperature and humidity in the study area, a build-up of oxidants,
nitrogen oxides, and possibly gas phase1 organic species associated with changes in air
recirculation in the room during the evening and the generation of emissions during the
production of materials for large mailings during the evening. The results of the study
will be presented and the factors which contribute to elevated concentrations of measured
species in the "sickroom", as compared to the control rooms, will be discussed.
114
-------
TRACER GAS MEASUREMENT OF INDOOR-OUTDOOR AIR EXCHANGE
RATES
Kevin N. Gunn and Zhishi Guo
Acurex Environmental Corp.
P.O. Box 13109
Research Triangle Park, NC 27709
Bruce A. Tichenor
Indoor Air Branch
Air and Energy Engineering Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Indoor air quality studies often require the determination of the indoor-outdoor air
exchange rate. The air exchange rate is defined as the volume of outdoor air that enters
the indoor environment in 1 hour divided by the volume of the indoor space. Air
exchange rates are usually determined by tracer gas techniques. Researchers at EPA's Air
and Energy Engineering Research Laboratory recently completed a study that compared
two tracers and three measurement systems, namely: SF6 measured by a photoacoustic
infrared detector, SF6 measured by gas chromatography with an electron capture detector,
and CO measured by anon-dispersive infrared detector. Controlled tests, using the tracer
gas decay method, were conducted in a small (~ 170 liter) environmental chamber.
Uncontrolled experiments were run in a test house. Data describing the accuracy and
precision of each method are presented.
MEASUREMENT OF AIRBORNE PARTICLE COUNTS AND MASS IN A
HEALTHY BUILDING DURING A ONE-YEAR CLEANING EFFECTIVENESS
STUDY
KE. Leese, B.C. Cole, andRM. Hall
Research Triangle Institute
Research Triangle Park, NC 27709
MA. Berry
U.S. EPA
Research Triangle Park, NC 27711
The Research Triangle Institute, under a cooperative agreement with the U.S. EPA's
Environmental Criteria and Assessment Office conducted a year-long field study to
understand and assess the effectiveness of an improved cleaning program to better indoor
air quality. A four-story healthy building was monitored with a Laser Particle Counter
(LPC), which is routinely used in clean rooms, to characterize airborne particle size
ranges of >0.5 to > 15 |jm. Monitoring was also done with a Fine Particle Sampler (FPS)
that measures airborne dust mass concentrations, which are routinely used to assess dust
exposure and potential health effects. LPC and FPS measurements were taken once
monthly on each floor. For five months, building housekeeping was performed as usual.
Professional cleaners then deep cleaned the building to include walls, windows, furniture,
115
-------
light fixtures, bathrooms, and tiled and carpeted floors, after which, an improved cleaning
regime was instituted and monitoring continued for seven months.
While LPC total particle counts correlated well with those outdoors, they were of
limited value, providing particle size distributions which did not correlate with total
airborne particle mass. The second floor, a child care facility which had the greatest
activity levels and access to an outdoor sand moat along the side of the building, also had
the highest concentration of total airborne dust. On the second floor, LPC large particle
counts (5, 10,15 (im) decreased significantly, while small particles (0.5, lijm) increased
immediately after cleaning. In comparing the two modes of measurement in a healthy
building environment, FPS total airborne dust mass gave the most meaningful data and
showed a 48% improvement with improved cleaning.
EFFECTS OF ACTIVATED CHARCOAL FILTRATION AND OZONATION ON
HYDROCARBON AND ALDEHYDE LEVELS OF AMBIENT AIR USED IN
CONTROLLED-EXPOSURE CHAMBER STUDIES OF AIR POLLUTANT
HUMAN HEALTH EFFECTS
Beverly E. Tilton, Joseph J. Bufalini, Sarah A.Meeks, and Bruce W. Gay
Air sampling experiments were done in 1985,1987, and in 1993-1994 at the
facilities of U.S. EPA Health Effects Research Laboratory in Chapel Hill NC.
Measurements of VOCs by GC-FID and aldehyde measurements by the DNPH silica gel
cartridge method were made, comparing levels at the outside air intake to levels in the
human controlled-exposure chamber. Ambient air passed through activated-charcoal
filtration can contain varying residual amounts of water vapor and gaseous pollutants
such as hydrocarbons, oxygenates, and other organic and inorganic species. When the
charcoal-filtered air stream is exposed to high-intensity radiation to generate ozone for
use in human controlled-exposure experiments, there may be formation of compounds
other than ozone. Results of measurements of some compounds of interest are discussed.
THE U.S. EPA/ORD LARGE BUILDING INDOOR AIR QUALITY
STUDY - RESULTS OF THE INITIAL BUILDING CHARACTERIZATION
Roy Fortmann andRuss Clayton
Acurex Environmental Corporation
P.O. Box 13109
Research Triangle Park, NC 27709
V. Ross Highsmith and C.J. Nelson
Atmospheric Research and Exposure Assessment Laboratory
Office of Research and Development
U.S. EPA
Research Triangle Park, NC 27711
The U.S EPA's Office of Research and Development Atmospheric Research and
Exposure Assessment Laboratory (EPA/ORD/AREAL) has undertaken a research
116
-------
program to collect fundamental information on the key parameters and factors that affect
occupant perception of indoor air quality and comfort in large office buildings. The
program will involve characterization of the building and test space; measurements of
key physical and chemical parameters; characterization of the heating, ventilating, and air
conditioning (HVAC) system(s); and collection of human response data during different
seasons in 40 buildings.
Forty Federal government office buildings were randomly selected using a
probability proportional to the number of occupants. Following initial review of the
sample and contact with the building managers, each building was visited to verify its
eligibility and to collect information on the building and potential test spaces. The
buildings range in age from two to 80 years old. Building size ranges from 3172 m2
(34,140 ft2) to 198,775 m2 (2,140,000 ft2). The number of occupants in the buildings
ranges from 46 to 8000. The buildings are highly diverse with respect to design, floor
plans, office layout, furnishings, density of workstations, and ventilation systems.
Although most of the buildings are mechanically ventilated, some are naturally ventilated
including one building that is 33,340 m2 in size.
This presentation provides an overview of the EPA Large Buildings study and
objectives, including the study design, measurement parameters, and measurement
protocol. The results of the initial survey are presented to show the diversity of randomly
selected study buildings.
LARGE BUILDING CHARACTERIZATION
Marc Y. Menetrez and David C. Sanchez
U.S. EPA
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Bobby Pyle, Ashley Williamson, and Susan McDonough
Southern Research Institute
P.O. Box 55305
Birmingham, AL 35255-5305
Andrew Persily
U.S. Dept. of Commerce
National Institute of Standards and Technology
Gaithersburq, MD 20899-0001
Buildings are characterized in this project by examining radon concentrations and
indoor air quality levels as affected by building ventilation dynamics. Indoor air quality
data collection stations (IAQDS), for monitoring and data logging, remote switches
(pressure and sail switches) and a weather station were installed. Measurements of
indoor radon, carbon dioxide and particle concentrations; temperature; humidity; indoor
to outdoor or sub-slab pressure differential; ambient and sub-slab radon concentrations;
and outdoor air intake flow rates were collected. The outdoor air intake is adjusted in
intervals of 5, from 0 to 20 cfm per person and exhaust fans are controlled while tracer
gas measurements were taken in all zones and IAQDS data is processed. Ventilation,
117
-------
infiltration, mixing rates, radon entry, pressure/temperature convective driving forces,
carbon dioxide generation/decay concentrations and IAQ levels were defined. These
dynamic interacting processes characterize the behavior of this and similar large
buildings. The techniques incorporated into the experimental plan are discussed with
project rationale.
A COMPARISON OF SORBENT SAMPLE CARTRIDGES FOR THE
COLLECTION AND ANALYSIS OF VOLATILE ORGANIC COMPOUNDS
COLLECTED IN LARGE OFFICE BUILDINGS
Jeffrey T. Keever and Linda Sheldon
Research Triangle Institute
Research Triangle Park, NC 27709
The use of sorbent materials for the collection of volatile organic compounds
(VOCs) in ambient air has been utilized for decades. Both polymeric (i.e., Tenax) and
carbonaceous(i.e., Ambersorb, charcoal) sorbents have been used for the collection of
VOCs. Each material has a unique affinity for various volatile compounds and,
subsequently, are often selectively employed for the collection of specific classes of
chemicals. In an effort to increase the number of VOCs which can be collected and
analyzed using a single sample cartridge, draft BASE protocols for the collection and
analysis of VOCs using both multi-sorbent sample cartridges have been developed.
The purpose of this study was to compare the performance of Tenax GC and
multisorbent cartridges. A sample collection of VOCs on both Tenax and multisorbent
sample cartridges was conducted in three large office buildings. Both sample types were
analyzed by thermal desorption-gas chromatography/mass spectrometry. Data for
instrument calibration, method blanks, method controls, and estimated method
quantitation limits are presented. The results of the sample analysis for the co-located
cartridges are compared. The agreement between the Tenax and multisorbent methods is
discussed.
Although the research described was funded by the U.S. EPA (Contract 68-
D2-0131), it has not been subjected to the required peer and administrative review and
does not reflect the views of the Agency, and no official endorsement should be inferred.
MANAGING RESIDENTIAL SOURCES OF INDOOR AIR POLLUTION
Bruce A. Tichenor and Leslie E. Sparks
Air and Energy Engineering Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Sources of indoor air pollutants in residential environments can be managed to
reduce occupant exposures. Techniques for managing indoor air pollution sources
include: source elimination, substitution, modification, pre-treatment, and altering the
amount, location or time of use. Intelligent source management requires knowledge of the
118
-------
source's emission characteristics, including chemical composition, emission rates, and
decay rates. In addition, knowledge of outdoor air exchange rates, heating/air
conditioning duct flow rates, and kitchen/bath exhaust fan flow rates is needed to
determine pollutant concentrations. Finally, indoor air quality (IAQ) models use this
information and occupant activity patterns to determine instantaneous and/or cumulative
individual exposure. This paper describes a number of residential scenarios for various
indoor air pollution sources, several air flow conditions, and typical occupant activity
patterns. IAQ model predictions of occupant exposures for these scenarios are given for
selected source management options.
RELATIONSHIP AMONG DRAG SLED, PUF ROLLER, AND HAND PRESS
TRANSFER OF PESTICIDE RESIDUES FROM FLOORS
David E. Camann, H. Jac Harding, and Paul W. Geno
Southwest Research Institute
San Antonio, Texas 78228-0510
Robert G. Lewis
U.S. EPA
Research Triangle Park, NC 27711
Comparisons were made of the transfer of formulated pesticide residues from treated
carpets and vinyl flooring by two dislodgeable residue methods and by human skin. Both
the Dow drag sled and Southwest Research Institute polyurethane foam (PUF) roller were
found to be acceptable dislodgeable residue methods on the basis of these studies. The
transfer efficiency of the drag sled consistently exceeded the transfer efficiency of the
PUF roller, which consistently exceeded the transfer efficiency of human hand presses.
The ratio of dislodgeable residue transfer to simultaneous hand press transfer from
adjacent sections of the treated surface was quite stable for the four pesticides evaluated
over three orders of magnitude of loadings by two application methods on two surfaces.
Across n=17 sets of observations, the mean and standard deviation of the transfer ratios
were x= 7.4, s = 2.8 for the drag sled/hand press ratio and x= 3.3, s = 2.1 for the PUF
roller/hand press ratio. Crude estimates of dermal exposure to pesticide residues on floors
can be derived from drag sled or PUF roller measurements using these ratios.
119
-------
COMPARISON OF PM2 s AND OPEN-FACE INLETS FOR SAMPLING
AEROSOLIZED PESTICIDES ON FILTERED POLYURETHANE FOAM
David E. Camann and H. Jac Harding
Southwest Research Institute
San Antonio, Texas 78228-0510
Charles L. Stone
University Research Glassware
Carrboro, North Carolina 27510
Robert G. Lewis
U.S. EPA
Research Triangle Park, NC 27711
In sampling the particles and vapor of an aerosolized pesticide through an open-face
inlet, a filtered polyurethane foam (PUF) trap will collect some particles which are too
large to be respired. A personal PUF sampler with 2.5 jam cutpoint inlet (University
Research Glassware) was compared to the standard open-face sampler. Ten pairs of air
concentrations of an aerosolized pesticide formulation were obtained in a wind tunnel
using collocated PM2 5 and open-face inlets. The open-face measurement usually
exceeded the paired PM2 5 measurement. Significantly larger air concentrations of
chlorpyrifos, 2,4-D, and mecoprop were determined over the ten pairs with the open-face
inlet. The PM2 5 inlet sampler appears to provide a more accurate estimate of personal
inhalation exposure to pesticides.
120
-------
Thursday, May 5,1994
Session 19: Sources and Fate of Atmospheric VOCs
Thursday Afternoon
METHODS TO DETERMINE THE BIOGENIC CONTRIBUTIONS TO
AMBIENT CONCENTRATIONS OF VOLATILE ORGANIC COMPOUNDS
R.K. Stevens, C.W. Lewis, and WA. Lonneman
U.S. EPA
Research Triangle Park, NC
R.A. Rasmussen
Oregon Graduate Institute
Beaverton, OR
G.A. Klouda
National Institute of Standards and Technology
Gaithersburg, MD
W. Ellenson
ManTech Environmental
Research Triangle Park, NC
SL. Dattner
Texas Air Control Board
Austin, TX
Vegetative emissions of photochemicaUy reactive volatile organic compounds
(VOCs) (e.g., isoprene) are considered important sources of ozone precursors. The U.S.
EPA is currently evaluating a number of procedures to obtain reliable biogenic emission
inventories. This report describes ambient measurement procedures to collect, process
and analyze,VOC samples to measure the relative percent of the total VOCs that
originates from vegetative emissions. In 1992, VOC were collected in 32L canisters at a
downtown location in Atlanta as part of the summer Southern Oxidant Study. These
canister samples were to be submitted to the National Institute of Standards and
Technology to measure their 14C content. Vegetative (also referred to as biogenic) VOCs
are enriched in 14C while emissions from fossil fuels (e.g., gasoline) are practically
devoid of this isotope. Thus, the ratio of 14C to 12C in ambient VOC samples can provide
a direct determination of the fraction of carbon whose origin is biogenic. To date, the
cryogenic procedure being developed at NIST to isolate the total VOC fraction from
whole (humid) air effectively separates all VOCs from CO and CH4 and only the less
volatile VOCs, defined by their equilibrium vapor pressure at -80° C, from CO2.
Separation of the more volatile VOCs from CO2 remains problematic until the necessary
preparative gas chromatographic/combustion technique is developed. In the mean time, a
new pre-cryogenic procedure has been developed to reduce the CO2 concentration by a
factor of 104 (e.g., from 360 ppm to 30 ppb) in whole air samples through the use of a
LiOH trapping system. These samples would then be submitted to NIST for subsequent
121
-------
cryogenic separation of VOCs from CO and CH4, followed by oxidation to CO2
measurement processing. The identification and quantification of VOC species lost
during the CO2 removal step has been incorporated into the methodology. For example,
as part of the procedure an aliquot of whole air sample, prior to removing the CO2, is
analyzed by identical gas chromatographic (GC) systems, one equipped with a flame
ionization detector (FID) and the other a mass selective (MS) detector. The GC/FID
analyzer quantifies each of the chromatographic peaks and the GC/MS system identifies
each species eluted from the GC column. After the CO2 has been removed from the
whole VOC air sample, it is reanalyzed by the GC/FID and GC/MS analyzers. In this
way, losses of VOCs (especially oxygenates) may be quantified and identified. Those
species which have been identified as biogenic or biogenic reaction products are
incorporated into estimates of the fraction of VOCs that may originate from biogenic
sources. The report describes the status of these studies and presents preliminary results
of the gas chromatographic analysis and 14C measurements.
RADIOCARBON MEASUREMENTS OF WINTERTIME ATMOSPHERIC
CARBON MONOXIDE IN ALBUQUERQUE, NM: CONTRIBUTIONS OF
RESIDENTIAL WOOD COMBUSTION
George A. Klouda
National Institute of Standards and Technology
Gaithersburg, MD 20899
Michael V. Connolly
Albuquerque Environmental Health Department
Albuquerque, NM 87103
In the past during winter months, Albuquerque, NM, has occasionally exceeded the 9
HL/L (part-per-million by volume, ppmv) - 8 hour National Ambient Air Quality
Standard (NAAQS) for carbon monoxide (CO), thereby, requiring action towards
attainment. Previous results of CO concentrations and radiocarbon (14C) measurements
from a wintertime 1984-1985 study of Albuquerque suggested that, during this time of
year, residential wood combustion (RWC) and motor vehicle emissions were the primary
sources of ground-level CO concentrations [Einfeld et al. (1988), Klouda et al. (1986)
and (1988)].
The major objective of this study was to reevaluate these source contributions in
light of the more recent implementation of no-burn days based on the meteorological
forecast 24 hours in advance. Whole air samples were collected during the Winter of
1989-1990 in Albuquerque, NM, for CO concentration and 14C measurements. Since
resources for this study were limited, a 23-factorial design with limited replication was
used to obtain optimal source information given a constraint of 10 ambient samples. The
three-factor two-level sampling design included the following: 1) sampling location;
residential vs traffic sites, 2) time-of-day; 0630 to 1430 vs 1630 to 0030, and 3)
forecasted meteorology; dynamic vs stagnant air mass, to effectively cover time-space.
On three occasions, samples were collected for target periods of maximum wood burning,
as predicted by the forecast meteorology that suggested times of cold stagnant air which
122
-------
would likely result in temperature inversions with little mixing. Radiocarbon results,
assuming that a measured 14CO background component applies, indicated that the
contribution of RWC ranged from 0% to 32% for all samples analyzed. For samples
collected when conditions were most favorable for high CO but designated as no-burn,
the RWC contribution ranged from 0% to 18%. The data suggested that the no-burn
strategy has been effective and that further controls on motor vehicles may be necessary
especially during unique periods, e.g. over Christmas holiday, when the NAAQS is more
likely to be exceeded. Additionally, the relation of fossil CO and benzene was explored in
light of a possible reduction in an occupational health and safety standard from 1 ppmv to
0.1 ppmv.
Einfeld et al. (1988), Sandia Report, SAND88-0121.
Klouda et al. (1986), Radiocarbon, 28, 625.
Klouda et al. (1988), J. Radioanal. Nucl. Chem., 123, No. 1,191.
PROGRESS TOWARDS VALIDATING THE SEPARATION OF
ATMOSPHERIC VOLATILE ORGANIC CARBON FROM AIR FOR 14C
MEASUREMENTS
George A. Klouda, George C. Rhoderick, and Robert L. Sams
National Institute of Standards and Technology
Gaithersburg, MD 20899
Charles W. Lewis and Robert K. Stevens
U.S. EPA
Research Triangle Park, NC 27711
Rei A. Rasmussen
Oregon Graduate Institute
Beaverton, Oregon 97006
It is well known that atmospheric volatile organic compounds (VOCs) from fossil
and biomass sources contribute to the production of ozone (O3) during summer months
via photochemical pathways. The degree to which each source participates in the
chemical process is important to help design effective strategies to control O3 levels in
urban atmospheres. An accurate measure of source contributions can be obtained from
radiocarbon (14C) measurements made directly on the VOC fraction. With the sensitivity
of accelerator mass spectrometry 14C/12C measurements now at 10 |o.g modem carbon (1
|0.g blank) and 2% precision (Currie et al., 1994), the ability to obtain quality
(interpretable) VOC-14C measurements relies on other factors: 1) the recovery of VOCs
from air, 2) the bounds for bias (chemical and isotopic) due to inefficiencies in the
chemical separation process, and 3) minimal chemical and isotopic contamination
(blanks).
This report will include more recent steps taken to validate techniques developed at
NIST for separating VOCs from either untreated (humid) air (Klouda et al., 1993) or air
preprocessed with LiOH (Stevens et al., this proceedings) to remove H2O and CO2. A
hydrocarbon (gasoline) reference material, diluted gravimetrically to - 2 |j.mol C/mol
with hydrocarbon-free air containing ambient CO2 concentration, and ambient samples
123
-------
from Atlanta and Houston are used to evaluate recovery and isotopic blank. Results of
these experiments will be presented along with a discussion of other chemical and
isotopic information necessary for reliable interpretation of atmospheric VOC-14C
measurements.
Currie et al. (1994), 6th International Symposium on
Accelerator Mass Spectrometry, Nucl. Inst. and Meth., in press.
Klouda et al. (1993), proceedings of the AWMA Meeting
on Measurement of Toxic and Related Air Pollutants.
Stevens et al. (1994), this proceedings of the AWMA
Meeting on Measurement of Toxic and Related Air Pollutants.
COMPARISON OF AMBIENT RATIOS OF NMHCS AND CO TO NOX WITH
EMISSION INVENTORY VALUES FOR ATLANTA
J.P. Pinto
U.S. EPA
Research Triangle Park, NC 27711
M. Somerville
MET!
Research Triangle Park, NC 27709
Ratios of NMHCs and CO to NOX obtained during the 1990 Atlanta Ozone
Precursor Study are compared to values predicted by emissions inventories. Ambient data
and emissions inventory values for the early morning rash hour are used following
methods originally adapted for the Los Angeles Air Shed by Fujita et al. (JAWMA, 42,
pp. 264-276, 1992.) A number of factors which could affect the conclusions of ambient
data-emissions inventory reconciliation studies such as the photochemical processing of
emissions, transport from background areas and the effectiveness of pollution control
devices are discussed. Results from these analyses are qualitatively consistent with other
studies from the Los Angeles air basin and the Lake Michigan air quality region,
indicating an underpredicu'on of these ratios by the emissions inventories compared to
ambient data.
ATMOSPHERIC CHEMISTRY OF UNSATURATED OXYGENATES:
ALCOHOLS, ALDEHYDES, KETONES AND ESTERS
Daniel Grosjean and Eric Grosjean
DGA, Inc.
4526 Telephone Road, Suite 205
Ventura, CA 93003
It has been known for many years that biogenic emissions include unsaturated
oxygenates such as alcohols, aldehydes, ketones and esters. Yet, while the atmospheric
chemistry of other important biogenic hydrocarbons (isoprene and terpenes) has received
much attention, little is known regarding the atmospheric persistence and fate of
unsaturated oxygenates. As olefins, unsaturated oxygenates are expected to be oxidized
124
-------
in the atmosphere in pathways initiated by their reactions with ozone and with the
hydroxyl radical. These reactions produce carbonyls, carboxylic acids, and, in the
presence of oxides of nitrogen, peroxyacyl nitrates. Thus, information on the atmospheric
chemistry of unsaturated oxygenates is important in the context of assessing the role and
impact of biogenic hydrocarbon emissions.
We have measured ozone reaction rate constants for a number of unsaturated
alcohols, esters and carbonyls. Using these rate constants together with structure-
reactivity relationships, rate constants have been estimated for the reaction of the
hydroxyl radical with the unsaturated biogenic oxygenates of interest. From these kinetic
data, estimates could be made of the atmospheric lifetimes for a number of unsaturated
alcohols, esters and carbonyls. We have also investigated, in laboratory experiments
carried out under conditions that are relevant to the atmosphere, the oxidation of the
unsaturated alcohols cis-3-hexen-l-ol, 3-buten-l-ol, and allyl alcohol. Two types of
experiments have been performed, one involving sunlight irradiations of unsaturated
alcohol-NO mixtures and the other involving the reaction of ozone with the unsaturated
alcohol in the dark. Carbonyl and peroxyacyl nitrate products of the alcohol-NO reaction
in sunlight have been identified and their concentrations measured. Peroxypropionyl
nitrate (PPN, CH3CH2C(O)OONO2) was a major product of cis-3-hexen-l-ol and
accounted for 14-20% of the initial NO. Atmospheric persistence of unsaturated alcohols
and implications for the formation of propanal and PPN from biogenic emissions of
cis-3-hexen-l-ol will be discussed.
ATMOSPHERIC CHEMISTRY AND FATE OF C2-C5 PEROXYACYL
NITRATES.
Daniel Grosjean, Eric Grosjean, and Edwin L. Williams III
DGA, Inc.
4526 Telephone Road, Suite 205
Ventura, CA 93003
Peroxyacyl nitrates, RC(O)OONO2 (PANS), are of major importance in atmospheric
chemistry. They have also received attention as toxic air contaminants: the several PANs
studied so far are eye irritants, mutagenic and phytotoxic. While the simplest compound,
PAN (CH3C(O)OONO2), has been studied in detail, little is known regarding the ambient
concentrations, atmospheric persistence and atmospheric fate of higher molecular weight
PANs.
In this study, we have investigated twelve saturated and unsaturated aliphatic PANs,
R = CH3(PAN),C2H5,C3H7 (2 isomers), C^ (4 isomers), CI^ = CH-, CJ^ = C(CH3)-;
CH2 = CXCy^)- and CH3CH = CH-. These compounds have been synthesized in the
liquid phase, prepared in-situ by sunlight irradiation of NO-carbonyl mixtures (e.g.,
CH2 = C(C2H5)C(O)OONO2 from NO and 2-ethylacrolein), measured at ppb levels by
electron capture gas chromatography, and characterized using a number of chemical and
physical diagnostic tests. Thermal decomposition, a major loss process for all PANs in
the atmosphere (RCO3NO2 -» RCO3 + NO2) has been studied at ambient temperature
and 1 atm. of air. Other removal processes studied include reactions with OH, and, for the
125
-------
unsaturated PANs, reaction with ozone. These removal processes will be discussed with
respect to the persistence of PANs in the atmosphere.
126
-------
Thursday, May 5,1994
Session 20: Russian Air Pollution Studies
Thursday Afternoon
EXAMINATION OF ATMOSPHERIC DIFFUSION CALCULATING SCHEMES
UNDER EXPERIMENTAL DATA
Serge A. Gromov
Institute of Global Climate and Ecology
Veronica A.Ginzburg
Moscow State University
Application of atmospheric diffusion models to calculate the pollution level of air
surface layer is immediately related to satisfying of received data. Evaluation of obtained
results by field and regular measurements gives the possibility to postulate that relative
difference of values decreases if averaging period grows. But there areany cases that
short-term concentration needs for determining the dangerous level of source's emission.
Moreover, the better convergence of calculated and measured short-term data will
obviously allow to have more realistic long-term values.
We proceeded the first investigating phase of air pollution diffusion model which
used in studies of background monitoring data analysis and evaluation as well as of
environmental pollution loads from local sources. Basic model formula realizes analytical
solution in Gaussian approach with dispersion parameters by Brigg's approximation.
Calculating scheme and programs were modified according to experiment conditions.
The wind-tunnel experimental measurements of diffusion were used for study in the
case of floor roughness surface. The problems of calculating and experimental data
harmonization were solved. The comparison shows a good correspondence of results. In
addition the model of another type (K-model) which is widespread in Russia was used on
same data set
BACKGROUND POLLUTION OF THE ATMOSPHERE: THE MULTI-YEAR
OBSERVATION IN RUSSIA
F. Rovinsky
Institute of Global Climate and Ecology
The observations on the integrated background monitoring programme at a network
of stations in biosphere reserves embracing a considerable part of the Eurasian continent
began late in the seventies in the former USSR. The atmospheric part of the programme
envisages daily evaluation of sulfur and nitrogen compounds, ozone, heavy metals (Pb,
Cd, Hg, Cu, Zn, etc.), organochloride and polyaromatic compounds. The same
compositions of the components are evaluated in precipitation samples.
The spatial distribution of background concentrations for the most of pollutants is
irregular a decrease could be observed eastwards and northwards, and the concentrations
127
-------
are significantly lower in the mountain areas also. This regularity is mostly distinct for
pollutants of fuel-power cycle - sulfur and nitrogen dioxides, benzpyrene, Pb; as for DDT
and HCCH the regularity is weaker.
The time-related daily and seasonal variations are present. These variations are
connected both with power source volumes and atmospheric processes.
Multi-year trends demonstrate that during the last decade the mean annual sulfur
dioxide, benzpyrene and Pb concentrations are 2-4 times lower in the western and central
parts of the European territory of Russia and they practically do not change over the
southern Russia and Siberia, and the DDT concentrations monotonously decrease with
time.
The assessment of the anthropogenic heavy metals contribution (Pb, Cd, As, Cu, V)
in background pollution of the atmosphere was obtained with the help of the geochemical
relations methods. This value changes from 20-50% for the continental regions up to
70-98% in the western part of the former USSR (more urbanized).
The multi-year statistical data on background pollution of the precipitation and their
acidity demonstrate that noted irregularities for the atmosphere in the whole are
characteristic for the precipitation, as well.
As it is shown by the multi-year observation data, the background pollution of the
atmosphere endure the anthropogenic influence over large-scale territories.
METHODS OF STUDYING SOURCES IN THE BOUNDARY ATMOPSPHERIC
LAYER OF THE BACKGROUND AREAS
V. Egorov
Institute of Global Climate and Ecology
The data on major sources required for assessment and prediction of ozone variation
in the boundary atmospheric layer and their effects on environmental objects. The most
important sources of ozone input in the lower atmospheric layers are photochemical
reactions and transport from upper atmospheric layers, advection of ozone and its
precursors.
The method of the simultaneous measurements of ozone in the chamber volume
filling by ambient air and in the outdoor atmosphere was applied to study the effect of
photochemical ozone production in the atmosphere of the background areas. The material
used as chamber shell was inert with respect to heterogeneous ozone sink and transparent
for solar radiation.
The results of carried-out observations indicate of ozone concentration exceeding in
chamber from 10 to 100 percent in the Berezinsky reserve area (Belarus) in the daytime
(July) caused by its photochemical production in comparison with ozone content level in
the outdoor air.
Used technique allows to estimate the upper boundary of ozone content in the air is
caused its photochemical sources in the observation area.
The effect of air transport and possible photochemical ozone production on its
content level in boundary atmospheric layer was investigated with help of ozone gradient
measurements method and meteorological parameters (air temperature, speed and
-------
direction of wind). For this purpose was using captive balloon (v=800 m3)with the basket
containing measuring instruments.
The results of the measurements carried out in the background area of Kurakaya
district, their comparison with existing data in world literature on type of vertical ozone
distribution in the lower atmospheric layers indicate of dominant effect of ozone
convective input in the near surface layer for the upper atmospheric layers during
observation period (June - August).
APPLICABILITY OF TRAJECTORY ANALYSIS FOR AIR BACKGROUND
MONITORING NETWORK OPTIMIZATION
Serge G. Paramonov
Institute of Global Climate and Ecology
Russia
There are five working stations of air background monitoring over the European part
of Russia nowadays. These stations proceed measurements of some pollutant
concentrations in the surface layer troposphere. The level of observed concentrations
depends on the lifetime of the pollutant and emission source density in the way of
"source-receptor".
With the help of a back-trajectory model, the areas with emissions influencing the
concentrations in the receptor site were estimated. The model is based on using 5-days of
back trajectories calculated according to ageostrophical relations between fields of wind
and pressure.
The trajectory calculations for background monitoring stations started in July 1989
and since this time have continued daily. A data base of 5.5 thousand trajectories has
been collected. All trajectories are classified and statistics of long range transport are
estimated. On this data the areas of polluted air portions ways were determined for each
station for the period from one to three days. The areas have an elliptical shape and are
situated in the prevailing air transport direction. There are territories where the areas from
different stations intersect. At the same time there are vast territories where pollution
transport isn't controlled by any existing monitoring network, among them areas of
northeast European Russia, Kola peninsula, east Ural. The named regions need air
monitoring stations organized.
LONG-RANGE MODEL FOR ATMOSPHERIC POLLUTION ANALYSIS OF
BACKGROUND TERRITORIES
Serge A. Gromov
Institute of Global Climate and Ecology
Background atmospheric pollution is mainly determined by large scale transport.
Goals of measurement data analysis appear to be the evaluation of the inputs of regional
sources into long-term pollution levels and the investigation of its seasonal or multi-year
changes.
129
-------
Lagrangian model is supposed to study atmospheric pollutant transport. It is based
on 5-days of back trajectories calculated from ageostrophical wind-pressure relations
using meteorological data from the Moscow World Meteorological Center.
The two-layer homotropic model of BPL is included to estimate such parameters as
average transporting wind, turbulence coefficient and exchange with upper layers.
Pollutant advection is simulated as moving of separate "puffs". Lateral diffusion
seems as Gaussian approach. Vertical diffusion equation is solved numerically taking into
account the BPL's results and dry deposition depended by seasonal features. It uses
time-depending decrease of concentrations by first order chemical reactions when
transformation of pollutants exists. Wet deposition is simulated by the same manner
when cloudiness or precipitation takes place.
The results are concentrations and depositions at each point for the chosen time
interval.
The model has been used for analytical evaluations of long-range environmental
background pollution over the former USSR's European territory.
130
-------
Thursday, May 5,1994
Session 21: Environmental Tobacco Smoke
Thursday Afternoon
REAL-TIME MONITORING OF POLYCYCLIC AROMATIC
HYDROCARBONS (PAH) AND RESPIRABLE SUSPENDED PARTICLES (RSP)
FROM ENVIRONMENTAL TOBACCO SMOKE (ETS) IN A HOME
Wayne Ott
U.S. EPA and Department of Statistics
Sequoia Hall 126
Stanford University
Stanford, CA 94305
Nancy Wilson
Human Exposure Research Branch (MD-56)
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Neil Klepeis and Paul Switzer
Department of Statistics
Stanford University
Stanford, CA 94305
Real-time measurement of polycyclic aromatic hydrocarbons (PAH) on fine particles
was evaluated in a home with environmental tobacco smoke (ETS) as a source. The PAH
PAS lOOOi monitor (EcoChem Technologies, Inc., West Hills, CA) is based on
photoelectric ionization of surface PAH, loss of the photoelectrons, and subsequent
measurement of the remaining positively charged particles in a filter electrometer.
Cigarettes were smoked in different rooms of a test house, with the time series of PAH
concentrations logged using a Langan DataBear data logger. Respirable suspended
particles (RSP) were monitored using the Model 8510 Piezobalance (TSI, Inc., St. Paul,
Mn), which measures fine particle concentrations with 2-min averaging periods.
Comparison of PAH and RSP concentrations from these experiments suggests: (1)
concentrations from the current version of the real-time PAH monitor are linearly related
to RSP concentrations; (2) exponential decay rates for RSP and PAH concentrations are
similar; (3) the PAH concentration time series for two different cigarettes — a Marlboro
Filter and a Kentucky Research Cigarette 2R1 — was found to be similar; (4) the RSP
concentration time series for the same two cigarettes was found to be different The
real-time PAH monitor appears to be useful tool for developing mathematical models to
predict concentrations in indoor microenvironments.
131
-------
MEASUREMENT OF ENVIRONMENTAL TOBACCO SMOKE
FrankE. Jones
Independent Consultant
32 Orchard Way South
Potomac, MD 20854
The subject of environmental tobacco smoke, and particularly the health
consequences of exposure of nonsmokers to environmental tobacco smoke, is of intense
interest. Since one of the prevalent sources of air pollution of particular concern indoors
in the workplace, in the home, in restaurants, in public buildings, in buses, trains, etc. is
smoke from cigarettes, cigars, and pipes, this paper concentrates on environmental
tobacco smoke and organic gases therefrom. Environmental tobacco smoke and its
constituents, mainstream smoke and sidestream smoke, are defined and discussed.
Among other subjects discussed are human exposure to environmental tobacco smoke;
proxies, surrogates, tracers, and markers; benzene in smoking; and nicotine. Perhaps one
of the most impressive and significant of findings of studies reviewed is that "smoking is
by far the largest anthropogenic source of background exposure to benzene."
A COMPARISON OF SMOKING AND NON-SMOKING AREAS: PRIVATE
HOMES AND BINGO HALLS
Ronald W. Bell, Robert E. Chapman, Brian D. Kruschel,
and Michael J. Spencer
Science and Technology Branch
Ontario Ministry of Environment and Energy
Toronto, Ontario
Personal exposure studies were undertaken in Windsor and Hamilton as part of the
Cities of the Nineties program at the Ontario Ministry of Environment and Energy.
Volatile organic compounds (VOCs), carbonyls and trace metals were the main
compound classes of interest. Environmental tobacco smoke (ETS) was an important
source of many compounds in each of those classes, so a separation of results into
smoking/nonsmoking areas was a logical one. Private residences were a primary target
for the Windsor study and were the only target for Phase I of the Hamilton study. A
questionnaire was given to one occupant of each dwelling, and the answers allowed a
correlation of the results with several factors, including smoking. Another situation that
allows potentially large exposures to ETS is the bingo hall, where patrons often spend
many hours each day, several days a week. One bingo hall had a separate no smoking
room, with a separate ventilation system as well, while a second hall had no separation
between the smoking and no smoking areas. Both halls had "smoke-eater" systems for
control of particulates, but had no controls for VOCs other than ventilation. Differences
between smoking and no smoking areas were more evident in the bingo halls than in the
private homes, but isoprene was clearly seen in both situations as a primary component of
ETS.
132
-------
Friday, May 6,1994
Session 22: Analysis of Polar Volatile Organic
Compounds
Friday
GC-MS ANALYSIS OF THE EXHALED BREATH MATRIX
Joachim D. Pleil and Andrew B. Lindstrom
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC
The organic components of the exhaled breath matrix are considered a non-invasive
'"window" into the blood gases. These compounds reflect normal biogenic function, direct
exposure to pollutants, and metabolites of such exposure. Given appropriate analysis and
interpretation of the data, the organic fingerprint of the exhaled breath can become an
important exposure and health assessment tool. Standard gas chromatography-mass
spectrometry (GC-MS) methods for measurement of volatile organic compounds in
ambient air are not well suited to the breath matrix due to the high carbon dioxide and
moisture content and the wide variety and concentration ranges of potential analytes. This
paper presents a set of three complementary analytical methods for GC-MS analysis
specifically designed for measuring Co2 and VOCs in the exhaled breath matrix.
Example chromatograms and the performance of the methods will be discussed.
DEACTIVATING SUMMA CANISTERS FOR COLLECTION AND ANALYSIS
OF POLAR VOLATILE ORGANIC COMPOUNDS IN AIR
David Shelow, Paul Silvis, Andrew Schuyler, and Joe Stauffer
Restek Corporation
Bellefonte, PA
Joachim Pleil
AREAL (MD-44)
U.S. EPA
Research Triangle Park, NC
Michael Holdren
Battelle Memorial Institute
Columbus, OH
SUMMA canisters are commonly used as the collection medium for whole air
samples for the measurement of a variety of volatile organic compounds (VOCs). The
interior electropolished surface is inert for most non-polar compounds even at trace
levels, however, certain polar species (PVOCs) exhibit reduced recoveries after storage.
A surface deactivation process, referred to as Silcosteel, has been developed wherein the
133
-------
interior of the canister is coated with a thin layer of fused silica. In addition to the silica
layer, a variety of chemical deactivations can be used to increase the inertness of the
surface for specific classes of VOCs. Canisters with the standard electropolished interior
surfaces were tested and compared to those treated with the Silcosteel process and
various deactivation layers. This paper presents an explanation of the process and the
results of the comparisons with particular emphasis on the storage stability and minimum
detectible limits for PVOCs.
CLEANLINESS OF COMMON AIR SAMPLING SORBENTS FOR
APPLICATION TO PHENOLIC COMPOUNDS MEASUREMENT USING
SUPERCRITICAL FLUID EXTRACTION
James R. Bowyer
ManTech Environmental Technology, Inc.
Research Triangle Park, NC
Joachim D. Pleil
AREAL (MD-44)
U.S. EPA
Research Triangle Park, NC
The trace-level measurement of phenolic compounds in the ambient air is
complicated by the acidic and polar nature of the compounds especially during recovery
from the sampling medium. Recently, supercritical fluid extraction (SFE) has been
proposed as an alternative extraction method to Soxhlet extraction or thermal desorption
to achieve more efficient recoveries. For such methodology to become practical, the
candidate sorbents must first be tested for stability and cleanliness under SFE conditions.
This paper describes exploratory research results of background contamination tests and
cleanup properties of some common air sampling sorbent media with respect to
application to phenolic compounds monitoring.
COLLECTION OF AMBIENT AIR PHENOLS USING AN ANION EXCHANGE
MEMBRANE
Mar da Nishioka and Hazel Burkholder
Battelle Memorial Institute
Columbus, OH 43201-2693
Scott Reynolds and Nydia Burdick
SC Department of Health and Environmental Control
Columbia, SC
Joachim Pleil
U.S. EPA
Research Triangle Park, NC
We have previously demonstrated the feasibility of collecting vapor phase ambient
air phenols by reversible chemical reaction with a solid sorbent. This sorbent, similar to
134
-------
XAD-2, consists of a styrene divinylbenzene polymeric backbone with chemically bound
quaternary amine functional groups and exchangeable anion. The smallest mesh size used
(200-400 mesh, 37-75 (am diameter) provided the highest collection efficiency for diverse
phenols at sampling rates of 100 mL/min. Detection limits were about 1 ppbv
We report here enhanced detection limits for ambient phenols using an anion
exchange membrane that allows high collection efficiency at 10 L/min sampling rate. The
membrane consists of 5 |im particles of the anion exchange resin enmeshed in a Teflon
microfibril matrix. This membrane is similar to "Empore" membranes, with the addition
of the anion exchange capacity. Sampling is accomplished using a 10.5 cm (diameter)
membrane and a General Metal Works PS-1 sampler. A Teflon-coated glass fiber filter,
spiked with deutered phenols, and placed ahead of the membrane, is used to deliver these
surrogate recovery standards to the membrane during the sampling. Following sampling,
membranes are shaken gently in an acidified mixture of methanol and dichloromethane.
The extract is derivatized with BSTFA and analyzed using either GC/FID or El GC/MS.
Analytical methodology allows detection at the 0.02 ppbv level for 12 hrs of sampling
CONTROLLED DESORPTION TRAP (CDT): A WATER MANAGEMENT
TECHNIQUE FOR QUANTITATIVE ANALYSIS OF POLAR VOCS IN
AMBIENT AIR
Sharon Reiss and Dick lesser
Graseby/Nutech
Durham, NC 27703-9000
Compendium Method TO-14 details the analysis of Volatile Organic Compounds
(VOCs) in ambient air using cryogenic preconcentration for sample enrichment. These
trace level organics require preconcentration of volumes greater than 50ml to provide a
sufficient sample mass to an analytical detector.
When working with high humidity sample volumes of 50ml and greater, water
management becomes critical. Compendium Method TO-14 specifies the NafionR semi-
permeable membrane to reduce water vapor in the sample stream during cryogenic
concentration. The same mechanism responsible for this effective dehydration of the
sample compromises recoveries of polar species. Consequently, alternate water
management techniques are necessary when polar compounds are analyzed.
The CDT is effective as a water management techniques since it allows volatile
organics, collected on a primary cryotrap, to vaporize at a temperature that limits the
partial pressure of water vapor. Temperature control of the cryotrap and the total volume
of the desorption gas are essential parameters in optimizing CDT. A secondary
cryotrapping step is necessary to refocus the slowly desorbed sample prior to injection to
a GC or GC/MS.
The analytical system consists of the Graseby/Nutech 3550A Cryogenic
Concentrator, the 354A Cryofocusing Accessory and the HP 5971 GC/MS. Some of the
135
-------
classes of compounds investigated using this application include alcohols, ketones and
ethers.
The effectiveness of the CDT will be compared with analyses using (a) no water
management and (b) the Nafion dryer.
A COMPARISON OF CONCENTRATION TECHNIQUES FOR THE ANALYSIS
OF POLAR COMPOUNDS IN CANISTER SAMPLES
D.B. Cardin andJ.T. Deschenes
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
The analysis of Polar Volatile Organic Compounds (PVOCs) in ambient air by
GC/MS requires sample preconcentration to achieve 0.1 ppb detection limits. Necessary
sample volumes can exceed 300 cc resulting in the co-collection of approximately 3-6 ul
of water, depending on the humidity of the sample. This much water will degrade column
performance and will cause signal attenuation in benchtop mass spectrometers making
quantification of target analytes difficult. Although most of the water vapor can be
eliminated by passing the sample through a Nafion Dryer, such an approach usually
results in loss of the polar VOC fraction in the sample due to passive and/or active
interaction with the dryer.
Two 3-Stage preconcentraton techniques have been previously reported which allow
both polar and non-polar VOCs to be quantified by eliminating most of the water before
injection into a GC/MS. These techniques, namely Automated 2-Dimensional
Chromatography (A2DC) and Microscale Purge & Trap (MP&T), both eliminate water
and CO2 in the sample before GC/MS injection to improve detection limits and
quantitation accuracy.
A concentration system utilizing yet a third water management technique called Cold
Trap Dehydration (CTD) will be presented. Using this technique, water can be
subtantially eliminated without loss of polar VOCs of interest. CO2 is also eliminated
before GC/MS injection resulting in superior chromatographic performance and a more
consistent GC/MS response for the extremely light VOCs. The preconcentrator uses the
same hardware trapping configuration for Cold Trap Dehydration as it does for
Automated 2-Dimensional Chromatography and Microscale Purge & Trap, and can select
any one of the three applications under software control. To determine which approach is
best for TOM and CAAA Title III compounds, all three water management procedures
will be examined and compared. Data will be presented showing detection limits and
%RSD's from the analysis of PVOCs in canisters using the 3-stage Entech 2000/2016CM
Automated preconcentration system and an HP 5972 GC/MS.
136
-------
Friday, May 6,1994
Session 23: Semi Volatile Organic Compounds
Friday
METHOD VALIDATION PROGRAM FOR THE LONG DURATION
SAMPLING OF PCDDS/PCDFS IN AMBIENT AIR
Bruce E. Maisel, Gary T. Hunt, and Marilyn P. Hoyt
ENSR Consulting and Engineering
Acton, MA 01720
Newt Rowe and Louis Scarfo
Connecticut Department of Environmental Protection
Hartford, CT
A method validation program was completed to assess the technical viability of
extended, long duration sampling periods (15- and 30-day) for the collection of
PCDDs/PCDFs in ambient air in lieu of the 48-hour sampling periods typically
employed. This long duration approach, if successful, would provide measurements data
more representative of average ambient PCDDs/PCDFs levels on an annual basis, and
hence provide enhanced support of the 1.0 pg/m3 annual ambient standard for
PCDDs/PCDFs (expressed as 1987 EPA toxic equivalents) required by Connecticut
regulation.
The method validation program utilized nine collocated "PUF" samplers which were
operated for 15-day and 30-day periods during each of two seasonal monitoring
campaigns (winter and summer). Each "PUF" sampler was outfitted with a Teflon coated
glass fiber filter and polyurethane foam (PUF) cartridge for the collection of paniculate-
associated and vapor phase PCDDs/PCDFs, respectively.
Samples were analyzed using high resolution gas chromatography/high resolution
mass spectrometry (HRGC/HRMS) based on EPA Method 8290. Each "PUF" cartridge
consisted of two foam halves; the top half PUF and filter were analyzed as a single
sample separately from the bottom half PUF section. This approach provided an
assessment of analyte breakthrough using the sampling system for large sample volumes
of approximately 4,000 m3 and 8,000 m3 for the 15-day and 30-day sampling periods,
respectively.
137
-------
EFFECT OF COMBUSTION TEMPERATURE ON THE ATMOSPHERIC
STABILITY OF POLYCHLORINATED DIBENZO-P-DIOXINS AND
DIBENZOFURANS
David M. Pennise and Richard M. Kamens
Department of Environmental Sciences and Engineering
University of North Carolina
Chapel Hill, NC 27599-7400
Atmospheric emissions of polychlorinated dibenzo-p-dioxins and dibenzofurans
(PCDDs and PCDFs) are likely to increase in the future due to an increase in municipal
and hazardous waste incineration. There is little information regarding the atmospheric
stability of PCDDs and PCDFs. In this study PCDDs and PCDFs were generated from
the combustion of a mixture of pentachlorophenol, polyvinyl chloride pipe shavings, and
wood chips treated with pentachlorophenol. These emissions were injected into outdoor
Teflon film chambers and aged in sunlight under typical atmospheric conditions.
Incineration experiments were conducted using "low temperature" combustion
(400-470° C range) and "high temperature" combustion (670-800° C range).
Concentrations of PCDDs and PCDFs were determined over time by collecting both
paniculate and vapor phase samples. These compounds were found to exist primarily in
the paniculate phase. Based on previous results with polybrominated dioxins and furans,
we expect paniculate phase PCDDs and PCDFs to slowly degrade over periods of hours
in the "low temperature" experiments. However, in "high temperature" experiments, we
expect paniculate phase PCDD and PCDF emissions to be stable. Differences in the
morphology and chemical composition of the combustion particles generated can explain
the differences in the Atmospheric stability of particle associated organics produced from
the "low temperature" and "high temperature" experiments. Previous results show that
incinerators operating near 450° C can generate paniculate phase brominated dioxins and
furans and PAH emissions with atmospheric half-lives of 1-2 hours. These observations
also indicate that the corresponding emissions from incinerators operating near 800° C
will have much longer half-lives, allowing for the possibility of long distance transport. If
the same observations are made for PCDDs and PCDFs, this dependence on combustion
temperature must be considered when evaluating the potential exposure to toxic organic
compounds emitted from waste incinerators.
MODELING THE MASS TRANSFER OF SEMI-VOLATILE ORGANICS IN
COMBUSTION AEROSOLS
Jay R. Odum and Richard M. Kamens
Department of Environmental Sciences and Engineering
University of North Carolina
Chapel Hill, NC 27599-7400
The atmospheric transport and fate of airborne organic compounds are highly
dependent upon which phase or phases (i.e., gas or particle or gas/particle) the compound
exists. Semi-volatile organics partition into both gas and particle phases and this
138
-------
partitioning is a function of the compound's vapor pressure, the amount of available
surface area, and the ambient temperature. Over the last 10 years, efforts to predict
atmospheric semi-volatile partitioning have been primarily based on equilibrium theory
(June 1977; Pankow 1987,1991,1992; Bidelman 1986,1988). However, recent
discoveries in this area suggest that full partitioning equilibrium may rarely be achieved
in the atmosphere. Therefore dynamic mass transfer models, rather than equilibrium
models, may be better suited to predict semi-volatile partitioning.
Recently Rounds and Pankow (ES&T 1990,1993) developed a radial pore-diffusion
model to simulate the mass transfer of semi-volatile organics in and out of combustion
aerosols. Preliminary results from their model and other recent discoveries suggest that
many types of combustion aerosols may be coated with a liquid organic layer and that
diffusion of semi-volatile organics through this layer impedes rapid mass transfer of these
compounds. Therefore a radial diffusion model was developed to describe the mass
transfer of semi-volatile organics into and out of combustion aerosols. The model
combustion aerosol consists of a solid carbon core that is surrounded by a viscous,
liquid-like, organic layer. Diffusion takes place only within the organic layer and is
controlled by mass transfer at the particle surface. Modeling of semi-volatiles requires the
tuning of two separate parameters: a diffusion coefficient and a surface mass transfer
coefficient. Preliminary testing of the model on the uptake of deuterated pyrene by diesel
exhaust aerosol at 25° C suggests that diffusion coefficients for PAH are on the order of
10"15 cm2/sec and that surface mass transfer coefficients for pyrene are on the order of
10"9 cm/sec.
EXPERIMENTAL OBSERVATIONS OF NON-EQUILIBRIUM GAS-PARTICLE
PARTITIONING OF PAHS IN AN OUTDOOR SMOG CHAMBER
DanaL. Coe and Richard M. Kamens
Department of Environmental Sciences and Engineering
University of North Carolina
Chapel Hill, North Carolina 27599-7400
To study non-equilibrium gas-particle partitioning of various PAHs, three specially
designed smog chamber experiments were conducted (October 1993, January 1994, and
February 1994). Automobile diesel exhaust was injected for five minutes into a 190 m3
Teflon film chamber and allowed to age during the night at temperatures below 15
degrees Celsius. A large denuder system was utilized during the injection period in order
to remove PAH vapors from the injection stream. Thus, PAH-laden particles were
observed to off-gas in the near absence of vapor phase PAHs during the initial stages of
the 8-hour experiments. The large denuder was designed as a parallel plate system, made
of activated charcoal impregnated filters. It was characterized to remove greater than
90% of PAH vapors from the diesel injection system. During the experiments, air
samples were collected in the chamber at twenty-minute intervals for the first two hours,
and hourly thereafter. The sampling system consisted of an XAD-4 coated annular
denuder, followed by a quartz-fiber filter, which is then followed by a second annular
denuder. Sample extracts were analyzed on Hewlett-Packard GC/MS. Results from these
139
-------
experiments are compared to output from a radial diffusion computer model, detailed in
another paper ("Modeling the Mass Transfer of Semi-Volatile Organics in Combustion
Aerosols" by Jay R. Odum and Richard M. Kamens).
AN ANALYTICAL METHOD FOR DETERMINING LOW LEVEL AIRBORNE
EXPOSURE LEVELS TO LEWISITE
Frank G. Lattin
U.S. Army Edgewood Research
Development and Engineering Center
Operations Directorate, Chemical Support Division
Aberdeen Proving Ground, MD 21010
Donald G. Paul
Dynatherm Analytical Instruments, Inc.
P.O. Box 159
Kelton, PA 19346
An analytical method for determining low level airborne exposure levels to lewisite,
an arsenical compound of military interest, is described. The method consists of a
screening technique and a confirmation technique using a single sample. A solid sorbent
tube is used for collection of airborne vapors and is also used to derivatize the lewisite to
facilitate analysis. The derivatizing agent, 1,4 thioxane, produces a stable compound that
responds well on several GC detectors with high selectivity and high sensitivity. During
thermal desorption, the derivatized lewisite sample is split between a screening GC
equipped with an electron capture detector (BCD) and a second sorbent tube. If a positive
response occurs on the screening GC, the saved portion of the sample is transported to a
confirming GC equipped with an atomic emission detector (AED). If the AED responds
to arsenic within the proper GC retention time window, the positive response is
confirmed. Precision and accuracy studies have been completed on both the BCD and
AED responses to the thioxane derivatized lewisite. Responses from both detectors are
linear within the required range, 0.25-1.50 TWA (.003 mg/m3), 1.25-7.50 ng on column,
with sample sizes from 0.5-3.0 liters. Field tests of the method are described.
140
-------
USE OF THE PHENANTHRENE TO BENZO(E)PYRENE AMBIENT AIR
RATIO AS AN INDICATOR FOR THE SOURCE OF POLYCYCLIC
AROMATIC HYDROCARBONS
Andre Germain, Sonia Ringuette, and Jean Tremblay
Environment Canada
Environmental Protection Service
Toxic Substances and Air Pollution Control Division
1179 deBleury Street
Montreal (Quebec) Canada, H3B 3H9
Polycyclic Aromatic Hydrocarbons (PAHs) are emitted by many industrial, domestic
and natural sources. In 1990, the principal sources of PAHs for the Province of Quebec
were the primary Aluminum smelters (8581), residential heating with wood (1621),
forest fires (148 t) and transportation (33 t). A sampling program was developed to
measure PAH levels in ambient air at different locations influenced by these sources. The
highest concentrations of PAHs (470 ng/m3 geometric mean) were measured in the
ambient air near primary aluminum smelters using So'derberg technologies. Areas
influenced by wood heating (157 ng/m3 winter geometric mean) and transportation (80
ng/m3 geometric mean) had lower total PAH concentrations. Ratios of ambient air
concentration for phenanthrene/benzo(e)pyrene were lower in samples collected in the
surroundings of the primary aluminum smelters, whereas high ratios were observed for
residential heating with wood (20-45) and transportation (20-45). The use of this ratio
was found to be a good indicator of the PAH origin for primary aluminum smelters.
POLYCYCLIC AROMATIC HYDROCARBONS IN HOUSE DUST AND
TRACK-IN SOIL IN AN EIGHT-HOME STUDY
Jane C. Chuang, Patrick J. Callahan, Vanessa Katona, and Sydney M.
Gordon
Battelle
505 King Avenue
Columbus, Ohio 43201-2693
Robert G. Lewis and Nancy K. Wilson
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC 27711
The analytical method to determine polycyclic aromatic hydrocarbons (PAH) in
house dust and soil samples was validated. The method consists of sonication with 10 mL
of hexane (C6) for two 30-minute extractions, and analysis of the C6 extract by gas
chromatography/mass spectrometry (GC/MS). An eight-home pilot field study was
conducted in Columbus Ohio before and after the 1992/1993 heating season to obtain
concentration profiles of PAH in house dust and track-in soil, and to determine whether
track-in of outdoor soil residues is an important source of PAH in house dust, A total of
19 PAH, ranging from 2-ring naphthalene to 7-ring coronene, was monitored. The sum of
141
-------
the concentrations of all target PAH in the house dust samples evaluated in this study
ranged from 41 to 580 ppm and from 25 to 310 ppm in the samples collected during
October 1992 and April 1993. Higher concentrations were observed in entryway soil
samples and the sums of the concentrations of target PAH ranged from 68 to 4000 ppm
and 58 to 5500 ppm in samples collected before (October 1992) and after (April 1993)
the winter heating season, respectively. The sum of the concentrations of PAH in the
pathway soil samples varied from 3.0 to 1200 ppm in samples collected before the
heating season. The sum of the concentrations of PAH ranged from 0.58 to 610 ppm and
from 0.63 to 63 ppm in pathway soil and foundation soil samples collected after the
heating season, respectively. The concentrations of most 4- to 6-ring PAH, the sum of all
target PAH, and the sum of PAH that are probable carcinogens in house dust correlated
well with the corresponding levels in the entryway soil. However, there was no
correlation between the PAH concentrations in house dust, and in pathway soil, nor was
any relationship found between house dust and foundation soil.
142
-------
Friday, May 6,1994
Session 24: General Papers
Friday
FUTURE RESEARCH DIRECTIONS FOR THE GREAT WATERS PROGRAM
Melissa W. McCullough
OAQPS
U.S. EPA
Research Triangle Park, NC 27711
The 1990 Report to Congress contained Section 112(m), requiring the assessment of
the deposition of air pollutants to the Great Lakes, Lake Champlain, Chesapeake Bay and
coastal waters. This program, called the Great Waters program, is required to have a
report to Congress in three years and every two years thereafter. The first Report to
Congress is to be released this spring (paper by Amy Vasu is describing the report).
With the completion of the first Great Waters report and the ensuing discussions
within the Agency, it is now appropriate to assess the future needs and direction of the
program. We know now where we stand in terms of the state of the knowledge, and what
kinds of efforts are needed to provide a comprehensive picture of atmospheric deposition
of toxics to aquatic ecosystems. Given that the problem is a vastly complex one, and that
research in this area is extremely expensive, the Agency must now determine where
efforts are best spent to collect the most important information to meet the mandate of
Section 112(m) of the Clean Air Act.
The Agency is working on a program strategy to target those most-effective efforts.
This paper will describe the strategy and the rationale for its design.
SCIENTIFIC FINDINGS AND REGULATORY RECOMMENDATIONS OF THE
1993 GREAT WATERS REPORT TO CONGRESS
Amy Vasu
U.S. EPA
OAQPS
Research Triangle Park, NC 27711
As a requirement of Section 112(m) of the 1990 Clean Air Act, the Environmental
Protection Agency (EPA) must submit a report to Congress on the deposition of air
pollutants to the Great Lakes, Lake Champlain, Chesapeake Bay, and coastal waters (i.e.,
"the Great Waters") by November 15,1993 and every two years thereafter. The 1993
report to Congress includes scientific findings on the following: (1) human health and
environmental effects associated with deposited air pollutants, (3) atmospheric loadings
of pollutants to the Great Waters, and (3) sources of the pollutants being deposited to the
Great Waters. Findings of the report indicate that significant adverse effects on human
health and wildlife have been caused by exposure, especially through fish consumption,
143
-------
to persistent chemicals that bioaccumulate. Atmospheric deposition is shown to be a
major contributor of mercury, PCBs, and other persistent chemicals that bioaccumulate.
Emissions from local as well as distant sources may contribute to pollutant loadings to
water bodies. The scientific findings provide support for the regulatory recommendations
of the report. These recommendations include taking actions under the Clean Air Act
(e.g., early completion of emission standards for sources of Great Waters pollutants) and
under other Federal authorities (e.g., the Federal Insecticide, Fungicide, and Rodenticide
Act), and continuing research efforts in areas where critical data gaps exist.
DETECTING AMBIENT OXIDES OF NITROGEN USING LASER-INDUCED
PHOTOFRAGMENTATION/PHOTOIONIZATIONSPECTROMETRY
J£. Simeonsson, G.W. Lemire, andR.C. Sausa
U.S. Army Research Laboratory
AMSRL-WT-PC
Aberdeen Proving Ground
Maryland 21005-5066
Phone (410) 278-9991
FAX (410) 278-9734
Recently we have reported a novel laser-induced photofragmentation/photoionization
(PF/PI) spectrometric method for detecting nitrocompounds under atmospheric
conditions. The method employs a molecular beam, time-of-flight mass spectrometer,
and a single laser operating at 226 nm or 193 nm to both photolytically generate the
characteristic NO fragment from the precursor molecule and photoionize it via its via its
A2T+ < -X2rc, B27r < -X27i and/or D2Z < -X27i- bands. In this study, the utility of a simple
jL 1 1 1
measurement cell maintained at near atmospheric pressures has been investigated. Ion
detection is accomplished directly using miniature stainless steel electrodes. The
analytical selectivity is determined strictly by the spectral selectivity of the
photoexcitation/photoionization processes at 226 nm. Alternate PF/PI wavelength
schemes have been explored. Studies have been performed to assess the analytical
capabilities of the technique using nonselective detection. Limits of detection have been
determined and will be reported for NO, NO2, HNO3 and CH3NO2.
MEASUREMENT OF THE EFFECTS OF MOISTURE DISTRIBUTION ON THE
TRANSPORT PROPERTIES OF RADON AND OTHER SOIL CONTAMINANTS
IN EPA'S SOIL CHAMBER
Ronald B. Mosley
U.S. EPA
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Significant health risks are associated with radon and other soil-gasborne
contaminants that enter the indoor environment. In an effort to develop better methods to
144
-------
reduce these risks, EPA is studying the physical mechanisms by which soil gas
contaminants migrate through the soil and enter buildings. It is widely accepted that
advective flow is the dominant mechanism by which radon and other soil gas
contaminants enter buildings. Consequently, it is important to understand how the
properties of soils influence these processes. The present measurements were performed
in EPA's soil chamber. Both diffusive and advective transport of SF6 was studied to
investigate the effects of moisture distribution on the transport rates. SF6, used as a tracer,
served as a surrogate for a contaminant gas. These measurements are important because
both the diffusion coefficient and the permeability for gaseous migration in soils depend
strongly on the distribution of moisture in the soil. These measurements along with
measured moisture profiles are used to infer the degree of uniformity of the soil
properties in the horizontal direction within the chamber. Packing density and radon
distribution measurements will also be discussed.
COMPARISON OF SOIL PERMEABILITY MEASUREMENTS USING PROBES
OF DIFFERENT SIZES AND GEOMETRIES
Ronald B. Mosley
U.S. EPA
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Recently, a number of questions have arisen concerning permeability measurements
associated with advective transport of radon in soils. While some of these questions relate
to distributed macroscopic flow paths in soils, others relate to the methods of analyzing
the traditional measurements. The traditional measurement method has used highly
localized probes of varying size and geometry to collect and measure the total flow
passing through the probe over a range of applied pressures. The permeability is typically
extracted from the measured flow/pressure relationship using a derived solution to an
idealized geometry that frequently does not match the reality of the probe. Rigorous scale
factors to correct for the differences in geometry are seldom used to compare
measurements from different probes. This study compares side-by-side measurements of
soil permeability for a number of probes with different geometries and relative sizes. The
results are discussed in terms of appropriate scale factors based on geometrical
differences.
TEST METHODS FOR EVALUATING REFORMULATED FUELS
Michael C. Croudace
PetroSpec, Inc.
60 Wells Avenue
Newton, MA 02159
The U.S. Environmental Protection Agency (EPA) introduced regulations in the
1989 Clean Air Act Amendment governing the reformulation of gasoline and diesel fuels
145
-------
to improve air quality. These statutes drove the need for a fast and accurate method for
analyzing product composition, especially aromatic and oxygenate content. The current
method, gas chromatography, is slow, expensive, non portable, and requires a trained
chemist to perform the analysis. The new mid-infrared spectroscopic method uses right to
identify and quantify the different components in fuels. Each individual fuel component
absorbs a specific wavelength of light depending on the molecule's unique chemical
structure. The quantity of light absorbed is proportional to the concentration of that fuel
component in the mixture. The mid-infrared instrument has significant advantages; it is
easy to use, rugged, portable, fully automated and cost effective. It can be used to
measure multiple oxygenate or aromatic components in unknown fuel mixtures.
Regulatory agencies have begun using this method in field compliance testing; petroleum
refiners and marketers use it to monitor compliance, product quality and blending
accuracy.
PECULIARITY OF TOXIC METALS EMISSION MEASUREMENTS AT
WASTEWATER TREATMENT PLANTS
Vladimir Kogan and Edward Torres
County Sanitation Districts of Orange County, California (CSDOC)
10844 Ellis Ave.
Fountain Valley, CA 92728-8127
Toxic metals emission can have a profound effect on the results of a facility's health
risk assessment, because of the extremely high unit cancer risks or hazard index values of
many metals.
The testing of metals emission at Publicly Owned Treatment Works (POTWs) offer
significant challenges due to the specifics of emissions from wastewater treatment
processes. Among these specifics are low concentrations of compounds of interest, high
air flow rates, and the fact that most of the metals considered as toxics may be present in
the plant's influent wastewater. It often makes the results of the testing very difficult for
proper evaluation.
CSDOC conducted extensive air- and liquid-phase source testing to characterize
toxic metals emission from the treatment processes at our two large wastewater treatment
plants. The presented paper describes the methods of sampling used, results of the testing
and their affect on the facilities' health risk assessments. A significant part of the paper is
devoted to a discussion concerning the sources of uncertainties in testing results,
including the effect of metals concentration in ambient air. Methods for the handling and
evaluation of results below the detection limit is also discussed.
146
-------
AN ASSESSMENT OF LOW EMISSION SEWER SYSTEMS FOR INDUSTRY
Reese H. Howie, Charles J. Zukor, and Parag Biria
Alpha-Gamma Technologies, Inc.
900 Ridgefield
Raleigh, NC 27609
The U.S. EPA is currently developing air emission standards to control the emissions
of hazardous air pollutants (HAPs) from industrial wastewater sources. These new
emission standards will require maximum achievable control technology (MACT) which
represents the most stringent level of control that is practically achieved. A variety of
emission control techniques are capable of meeting MACT criteria, therefore, these
standards provide some flexibility in selecting a strategy to reduce HAP emissions from
wastewater sources.
Emission control techniques that allow facilities to maximize the use of existing
capital equipment in sewer and pretreatment systems may offer a more cost effective
approach in complying with MACT standards. Water seals and covers can be installed on
existing sewer and pretreatment systems and are capable of achieving MACT emission
control standards. The purpose of this assessment was to:
1) identify emission suppression techniques that minimize the release of volatile
organic HAPs and other organic pollutants from industrial sewers and pretreatment
systems;
2) Characterize the applicability, design criteria, and technical issues associated with
the use of each suppression technique identified; and
3) Relate the significance of these emission suppression techniques to current and
future federal air emission standards for wastewater sources.
The assessment indicated that industrial facilities are currently using a variety of
techniques for minimizing emissions from sewer and pretreatment systems. In addition,
safety concerns have been expressed with the use of emission suppression techniques.
Some techniques allow volatile organic emissions to accumulate within an enclosed
space and create potential fire or explosion hazards. These hazards are minimized when
current safety practices and safeguards are implemented and maintained.
Common suppression techniques applied to sewer systems include: installing above
ground hard pipe sewers; installing water seals on process drains and sewer ventilation
pipes; creating surcharged sewer lines; and creating negative pressure sewer lines. An
increasing trend of using above ground hard pipe sewers at facilities was noted. This
technique is favored because it allows the development of an integrated environmental
strategy to comply with multimedia standards.
Common suppression techniques applied to pretreatment systems include: using
purge gases for fixed roof structures; installing floating roofs for storage tanks and oil-
water separators; and installing floating membranes for large surface impoundments. No
trends of favoring the use of any one emission suppression technique were observed for
pretreatment systems.
147
-------
Based on this assessment, sufficient information on individual control techniques is
available. However, little guidance is available for selecting and applying alternative
emission control techniques to reduce air emissions from an entire sewer and
pretreatment system.
AN ODOR CONTROL STUDY AT BISSELL POINT WASTEWATER
TREATMENT PLANT
Jon F. Bergenthal, Robert T. Jorgen, and John R. Gibbons
Sverdrup Civil, Inc.
Environmental Division
13723 Riverport Dr.
Maryland Heights, MO 63043
The Metropolitan St. Louis Sewer District's Bissell Point Treatment Plant has been
in operation since 1970. Currently, it receives an average of 111 mgd of wastewater from
a 57,000 acre service area covering the northern and eastern parts of the City of St. Louis,
and portions of North St. Louis County. Over half of the treatment plant's influent BOD5
and TSS loadings result from industrial sources. Odor complaints from the vicinity of the
plant have occurred yearly, primarily from June to September. The number of
complaints has increased since the new secondary treatment facilities were brought on-
line in October 1992. The purpose of this study was to identify the sources of odors,
determine the odor-causing constituents and their emission rates, and recommend a
phased plan that leads to the abatement of objectionable odors from the Bissell Point
Treatment Plant In addition, the recommended plan addresses Missouri Department of
Natural Resource's (MoDNR) rules relating to odor emission, U.S. EPA's pending
Hazardous Air Pollutant standards, and MoDNR's pending plan on achieving VOC
reductions mandated by the Clean Air Act Amendments of 1990.
148
-------
Friday, May 6,1994
Poster Session
Friday
AIR POLLUTION MONITORING IN THE REPUBLIC OF LATVIA
/. Lyulko andR. Dubrovskaya
Environment Pollution Observation Centre
Latvian Hydrometeorological Agency
19, K. Valdemara Street
LV-1010 Riga Latvia
First investigations of the air pollution in Latvia were started by the
Hydrometeorological Agency in 1968 in Riga. The surveys and preliminary results,
information available on the pollution sources, knowledge of climatic and meteorological
peculiarities of the Republic's regions formed the basis for the foundation in 1972 of the
National air pollution observation service.
Currently the system of observations covers 9 cities (20 posts) with emission
maximum concentrations from industrial sources and roads, as well as recreation zones
(Jurmala) and an area for the studies of transboundary pollutants transfer (Rucava). The
annual data base averages about 70,000 units.
The key principles of observations made in the Republic are:
• regularity of observations and unity of terms;
• complexity of observations (meteorological parameters included);
• unity of sampling and analysis methods.
The list of substances under observation is specified taking into account the amount
and composition of emissions into the atmosphere. Simultaneously with the air
samplings, meteorological variables (wind direction and speed, humidity and temperature
of the air, atmospheric phenomena) are measured at the posts. The most widespread
admixtures encompass solid substances, sulfur dioxide, nitrogen dioxide, carbon oxide,
phenol, formaldehyde, ammonia, aromatic hydrocarbons, hydrogen chloride, hydrogen
sulfide and metals (zinc, cadmium, lead, copper). From the observation data,
formaldehyde, nitrogen dioxide, phenol and ammonia highly contribute to the air
pollution of the Republic's towns.
To estimate the air pollution level, indirect methods are also run, including
precipitation chemical composition and snow pollutant content definitions.
The results of the analyses of precipitation chsmical compositions are used to
measure the contributions of local emitting sources and to define admixtures transfer by
air masses. Monitoring of the chemical composition of precipitation, gases and aerosols
is carried out at the station of EMEP's Rucava
149
-------
Prospects in the development of the atmospheric pollution observation service
envisage control of the ozone concentrations and greenhouse gases (carbon dioxide,
methane, nitrous oxide, ETS) and membership with GSA of WMO (GAW).
EVALUATION OF THE EFFECTS OF HUMIDITY ON THE TRANSFER OF
C2-C10 HYDROCARBONS FROM CYLINDERS
Ron Bousquet and Ron Brande
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
We prepare complex mixtures of VOCs in 1.5 liter high pressure cylinders. These
cylinders are used as proficiency test samples. Audit materials of this type are usually
prepared in humidified SUMMA™ polished canisters. We compared the recovery of high
molecular weight compounds in humidified canisters and high pressure 1.5 liter
cylinders. A high pressure 1.5 liter cylinder containing C2-C10 hydrocarbons at the 5
ppbv level was prepared by diluting a master cylinder containing these compounds at the
50 ppbv level. The contents of this small cylinder were used to prepare a humidified
SUMMA polished canister. The relative humidity of the canister was approximately 50
percent and the final pressure was about 45 psia. Both the 1.5 liter cylinder and the
canister were analyzed and compared to a calibration curve prepared using the master
cylinder. Recoveries for high molecular weight compounds in the humidified canister
were greater, and closer to the theoretical values than in the 1.5 liter cylinder used to
prepare the canister. This may happen because the contents were properly delivered from
the small cylinder to the canister but that humidity was needed to deliver the higher
molecular weight compounds to the detector.
The information is this abstract has been funded wholly by the U.S. EPA. It has
been subjected to Agency review and approved for submission to the A&WMA.
COMPARISON OF BREAK-THRU VOLUME ON ADSORBENT VS.
CRYOGENIC TRAP
Wendy Ballard
Graseby Nutech
4022 Stirrup Creek Drive, Suite 325
Durham NC 27703-9000
Many air analysis techniques require concentrating VOCs from a sample stream onto
a trap for subsequent detection. The aim is to convert a large volume at low concentration
to a small volume at higher concentration which is more suitable for available detectors
such as GC/MS. The trapping technique must therefore retain the compounds throughout
the concentration event without significant loss due to migration through the trap. As
regulatory limits for compounds are lowered, the ability to trap larger sample volumes in
analysis has increased importance, particularly in whole air analysis.
150
-------
Sample concentration has been performed using adsorbent and cryogenic trapping
for many years. When a targeted compound migrates completely through the trap prior to
the completion of the concentration period it is referred to as break-thru. Variables which
affect sample break-thru include trapping temperature, rate and volume of collection,
sample concentration, and trapping medium. In addition, properties of target compounds
such as size, molecular weight, and boiling point affect the break-thru limitations. The
purpose here is to examine the break-thru point as a function of sample volume for a
number of compounds particularly some of the lighter volatiles. Although there is data on
the general subject, data for specific types of trap will be very useful. Data presented will
help to define limiting factors and provide comparative information for two types of trap.
EVALUATION OF MODIFICATIONS TO THE TEKMAR™ 5010 FOR THE
ANALYSIS OF INDOOR AIR POLLUTANTS ON MULTISORBENT TUBES
John W, Duncan
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
Solid sorbents, such as Tenax™, may be used for the collection of complex chemical
pollutants. The pollutants are desorbed from the tenax at the head of a GC column and
analyzed by various detectors. Tenax and other sorbents may be packed into tubes of
various sizes and shapes. Because we provide quality assurance support to a wide variety
of projects, our equipment must be adaptable to may sample types. The Tekmar [™] 5010
automatic desorber has furnace sizes capable of desorbing seven inch tubes of 5/8 inch
and 1/4 inch diameter. Desorbing other, multisorbent bed tubes that are eight inches long
requires a modification to the furnace cap. We have designed and built a replacement cap
for the 1/4 inch furnace that allows the Tekmar 5010 to desorb the eight inch
multisorbent bed tubes. The modification is simple enough to be performed by small
machine shops. This modification increases the flexibility of existing equipment. Details
of the modifications are provided along with an evaluation of its performance.
The information in this abstract has been funded wholly by the U.S. EPA. It has
been subjected to Agency review and approved for submission to the A&WMA.
EVALUATION OF STORAGE CONDITIONS FOR INDOOR AIR
POLLUTANTS ON SOLD3 SORBENTS IN UHP HELIUM PURGED MYLAR™
PACKS
John W. Duncan and Frederic J. Mixson
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
Solid sorbents, such as Tenax™, may be used for the collection of complex chemical
pollutants. The pollutants are desorbed from the tenax at the head of a GC column and
analyzed by various detectors. Many laboratories pack Tenax into glass tubes of various
sizes and shapes. Shipping and storage of both clean and used Tenax cartridges are
151
-------
problematical due to the fragility of the glass tubes. This laboratory provides audit
materials for a wide variety of projects, and an optimum shipping and storage systenvwill
maintain sample integrity and protect against breakage. Sealing cartridges in a Mylar
package purged and slightly pressurized with ultra high purity helium provides an inert
environment, and helps to maintain sample integrity. The sealed package also provides a
protective bubble-like environment perfect for shipping. The preparation method is
described and the results of stability studies reported for important indoor air pollutants
when compared to traditional storage methods.
The information in this abstract has been funded wholly by the U.S. EPA. It has
been subjected to Agency review and approved for submission to the A&WMA.
ENVIRONMENTAL AND OCCUPATIONAL EXPOSURES TO PAH IN THE
CZECH REPUBLIC: PERSONAL EXPOSURE MONITORING COUPLED
WITH HPLC/TIME-PROGRAMMED FLUORESCENCE DETECTION
R. Williams andK. Hattaway
Integrated Laboratory Systems
P.O. Box 13501
Research Triangle Park, NC 27709
R. Watts and]. Lewtas
U.S. EPA
HERL, MD-68A
Research Triangle Park, NC 27709
The U.S. EPA has collaborated with health researchers in the Czech Republic to
determine PAH exposures for populations in high pollution environments and in various
occupations. These investigations used 24 h personal exposure monitors that were
developed to allow separate and simultaneous collection of fine particles, vapor phase
nicotine and vapor phase organics. Samples were extracted and analyzed for 16 priority
PAHs by optimized HPLC coupled with time-programmed fluorescence detection.
Personal exposure monitoring periods of up to 24 continuous hours were conducted for:
1) Teplice policemen, who spent a major portion of their day outdoors; 2) open-pit coal
miners; 3) health researchers working in a laboratory; and 4) topside coke oven workers
(larry car operators) working outdoors. Total particle-bound PAHs ranged from 1.5
Hg/m3 for the health researchers to 52 |ig/m3 for the topside coke oven workers. Vapor
phase PAH concentrations also varied greatly depending on occupation and ranged from
0.6 |J.g/m3 for city policemen to 261 |ag/m3 for the coke oven workers. Carcinogenic
PAHs, which were predominantly found associated with particles (>90%), typically
included benzo(a)anthracene, chrysene and benzo(a)pyrene. Cigarette smoking did not
appear to significantly affect PAH exposures in these environments.
This is an abstract of a proposed presentation and does not necessarily reflect EPA
policy.
152
-------
EVALUATION OF THE TRANSFER EFFICIENCY OF HIGH MOLECULAR
WEIGHT HYDROCARBONS USING VARIOUS TYPES OF REGULATORS
Ron Bousquet
ManTech Environmental Technology Incorporated
Research Triangle Park, NC 27709
We prepare complex mixtures of VOCs in 1.5 liter high pressure cylinders. These
cylinders are used as proficiency test samples. We also provide a regulator with those
samples supplied to participants in the PAMS network. Different regulator types from
various suppliers were evaluated for cleanliness, recovery efficiency and our ability to
reclean them for use with compounds in the parts per billion range. All regulators were
alternately attached to the same cylinder containing UHP nitrogen. For each regulator,
about 500 millimeters of nitrogen that had passed through it were cold trapped and
analyzed by GC/MS. The number of chromatographic peaks and total area was recorded.
All regulators were then made dirty by attaching to a 20 ppmv source of VOCs and
purging the regulator. The regulators were then cleaned by purging with UHP nitrogen.
This nitrogen was analyzed as before and the number of peaks and area was noted. All
cylinders were attached to a source of 50 ppbv C2-C10 hydrocarbons and evaluated for
relative recovery. Low dead volume, highly polished, single stage regulators proved to
be the regulator of choice for our purposes.
The information in this abstract has been funded wholly by the U.S. EPA. It has
been subjected to Agency review and approved for submission to the A&WMA.
SIMPLIFIED PREPARATION OF TOW AND TITLE HI AIR TOXIC
STANDARDS USING A WINDOWS SOFTWARE PACKAGE SUPPORTING
STATIC AND DYNAMIC DILUTION SCHEMES
D.B. Cardin andEA. Galoustian
Entech Laboratory Automation
950 Enchanted Way #101
Simi ValleY, CA 93065
The preparation of Air Toxic standards in the laboratory can be performed using
several methods. These include injection of purge and trap standards, static dilution from
pure compounds, and dynamic dilution from NIST traceable standards. Static dilution
and dynamic dilution are the preferred choices as they keep high methanol concentrations
from being introduced into the TO14 SUMMA passivated sample containers (methanol is
a TITLE III target compound). Using static and dynamic dilution, standards can be
created accurately with the flexibility to add new compounds to a laboratory's list of
target analytes at a very low cost.
Using static and dynamic dilution in the laboratory to prepare standards containing
40 to 80 target analytes requires performing complex, time consuming calculations. To
simplify manually performed calculations, corrections for room temperatures and
barometric pressures are sometimes ignored, which can add to the error in performing
these calculations. In addition, mass flow controllers are assumed to be correct rather
153
-------
than trying to manually determine and implement correction factors for each flow
channel using multipoint calibrations.
A software package running under Windows has been developed that makes
calculating dilution parameters for even complex mixtures fast and simple. Compound
parameters such are name, molecular weight, boiling point, and density are saved in a
data base for later access. Gas and liquid mixtures can be easily defined and saved as an
inventory item, with preparation screens that calculate appropriate transfer volumes of
each analyte. These mixtures can be utilized by both the static and dynamic dilution
analysis windows to calculate proper flow rates and injection volumes for obtaining
requested concentrations. A particularly useful approach for making accurate polar VOC
standards will be presented.
RESULTS FROM THE AUGUST 1993 DOAS EVALUATION IN BAYTOWN,
TEXAS
Charles P. Conner
ManTech Environmental Technology Inc.
Research Triangle Park, NC
Lee Ann B. Byrd
Office of Air Quality, Planning, and Standards
U.S. EPA
Research Triangle Park, NC
Frank F. McElroy and Robert K. Stevens
Atmospheric Research and Exposure Assessment Laboratory
U.S. EPA
Research Triangle Park, NC
An EPA-sponsored field study to further evaluate DOAS technology took place in
Baytown, Texas in August, 1993. The goals of the study were to evaluate calibration and
audit procedures and to further evaluate DOAS performance in an area with potentially
high pollution levels. The gases being measured were ozone, nitrogen dioxide, and sulfur
dioxide. Two separate DOAS systems were operated simultaneously using nearly
identical air paths. Two sets of conventional point measurement monitors were operated
at the site to provide EPA-approved reference concentration measurements. Hourly-
average concentration measurements were recorded from all measurement systems.
Intercomparisons of the various data sets have been carried out Excellent agreement
between the two DOAS systems and between the two point measurement systems was
seen. The agreement between the DOAS and point measurement systems varied from
excellent to poor, depending upon the gas being measured. The quality of the correlations
between the two fundamentally different techniques was dependent upon the
homogeneity of the air mass. When nearby sources were present, the pollutant plume at
the monitoring site was narrow enough to differentially affect the two types of monitors.
Thus the different techniques measured different concentrations and the data correlations
were degraded.
154
-------
EFFECTS OF USING NAFION® DRYER ON HYDROCARBONS ANALYZED
FROM CANISTERS BY MODIFIED EPA METHOD TO-14
J.C. Sagebiel andB. Zielinska
Desert Research Institute
P.O. Box 60220
Reno, Nevada 89506
Hydrocarbons in the range of C2 to C12 were analyzed from SUMMA®-polished
stainless steel canisters by cryogenic preconcentration, desorption with boiling water, and
high-resolution capillary-column chromatography with flame ionization detection.
Eleven samples taken inside a highway tunnel were analyzed with and without a Nation®
dryer (Perma Pure Products, Inc.). The membrane successfully removed water and polar
species like methanol and ethanol, however, it also removed a significant amount of the
olefins and aromatics and lowered the total measured ppbC by 10-15%. The paraffins
seemed to be less influenced by the membrane. Several other peaks were also noted to
appear when using the membrane. We therefore do not recommend using this type of
water removing device when performing speciated hydrocarbon analyses.
STATISTICAL DISTRIBUTIONS OF AIRBORNE PCB AND PESTICIDE
CONCENTRATIONS MEASURED AT REGIONAL SITES ON THE GREAT
LAKES
Donald F. Gatz, Clyde W. Sweet, Ilora Basil, and Karen S. Harlin
Illinois State Water Survey
2204 Griffith Dr.
Champaign, EL 61820
Concentrations of total PCBs and eleven pesticides have been measured in air and
precipitation over periods of 20-33 months at regionally-representative locations on four
Great Lakes. The purpose of this paper is to report results of testing the observed
distributions for goodness of fit to normal, lognormal, and other common statistical
distributions. Twenty-four hour samples of airborne particles and vapor were collected at
12-day intervals on glass fiber filters and vapor traps of polyurethane foam (PUF) or
XAD-2 resin, using modified high volume samplers. Twenty-eight day precipitation
samples were collected using wet-only samplers with stainless steel sampling surfaces
and a heated enclosure containing an XAD-2 absorption column. The XAD and PUF
adsorbents and the glass fiber filters were extracted for 24 hours in a 1:1 hexane/acetone
mixture, concentrated by rotary evaporation, and cleaned using silica chromatography.
Fractions eluting from the silica column were analyzed for 103 PCB congeners and 11
pesticides using GC/ECD. Measured concentrations in air and precipitation, and
logarithms of the concentrations were tested for normality by the nonparametric Lilliefors
test using SYSTAT statistical software. After applying the Bonferroni procedure for
simultaneous inference, only two of the 48 measured vapor-phase distributions were
found to be statistically different from lognormal at the 5% level. Both of these involved
gamma-HCH (Lindane), at the Sleeping Bear and Sturgeon Point sites, and local sources
155
-------
are suspected. By the same method, the hypothesis that concentrations were distributed
normally was rejected in all but 12 (four of them were hexachlorobenzene) of 48 cases.
Among the precipitation samples, none of the 44 observed distributions was found to be
significantly different from the lognormal distribution by the same method, while 17 of
44 were found to be significantly different from normal. Thus it appears that the
lognormal distribution may be used to describe total PCB and pesticide concentrations in
air and precipitation in the absence of local sources.
STABILITY OF REDUCED SULPHUR COMPOUNDS IN WHOLE AIR
SAMPLERS
Quang Tran and You-Zhi Tang
Concord Environmental Corporation
2 Tippett Road
Toronto, Ontario, Canada
M3H 2V2
Reduced sulphur compounds are important constituents in the sulphur cycle between
hydrosphere, lithosphere, biosphere and atmosphere. Some of them, such as hydrogen
sulphide (H2S), mercaptans (e.g., CH3SH and C2H5SH), dimethyl sulphide (DMS) and
dimethyl disulphide (DMDS), can cause odour nuisance problems associated with kraft
mill and sewage treatment operations. Accurate and reliable determination of reduced
sulphur compounds is often required, but it is a challenging task due to the reactivity of
reduced sulphur species thus difficulties in collection and storage of air samples.
A variety of sampling methods, in association with different measurement
techniques, have been used for determination of reduced sulphur compounds. They all
have their pros and cons. Several whole air samplers were evaluated for collection of
reduced sulphur compounds at concentrations of 100 ppbv and 1 ppmv. Severe losses of
H2S and mercaptans were found in samples collected in electro-polished stainless steel
canisters, although these canisters have been proven suitable for many volatile organic
compounds. The losses of H2S and mercaptans were less severe in a Teflon-coated
stainless steel cylinder, and glass and silanized glass bottles with Teflon-lined septum
caps. Apparent losses of carbon disulphide (CS2) were only observed in these glass
bottles. In general, carbonyl sulphide (COS), CS2. DMS and DMDS were more stable
than H2S and mercaptans and all tested reduced sulphur compounds were more stable in
Tedlar bag than In other tested samplers.
156
-------
A NEW VAPOR AND GAS TEST ATMOSPHERE GENERATOR WITH BROAD
CONCENTRATION AND FLOW OUTPUT CAPABILITIES
JK. McGee, PA. Evansky, D. Terrell, andL.C. Walsh
Mantech International Corporation
P.O. Box 12313
Research Triangle Park, NC 27709
D.W. Davies andDL. Costa
U.S. EPA
Research Triangle Park, NC 27709
Labs which analyze air directly or perform environmental chamber studies need to
be able to prepare a wide range of chemical atmospheres for calibration or test. Multiple
generation methods are frequently required. We describe a new dynamic flow vapor
generator which can deliver virtually any desired concentration of a single compound or
mixture, from parts-per-billion to percent, within the limiting constraint of the
compound(s) saturated vapor pressure(s). Total test atmosphere output flows range from
two L/min to over 100 L/min. The generator consists of a pump-supplied liquid
evaporator which is coupled to a unique output control section. The output control section
allows either all or a fraction of the vapor to be directed to a dilution air stream.
Metering of a fraction of the generator's vapor output is accomplished using a mass flow
controller which requires very little head pressure to actuate over its entire delivery range.
In either mode the output control section is operated, vapor concentration is based
directly on primary standard calibrations of liquid feed rates and gas feed rates through
the generator, mass flow controller, and dilution gas stream. By a simple modification the
generator can also be converted into a gas blending system, by metering the desired gas
through the output control section mass flow controller and mixing with dilution air.
(This abstract does not necessarily reflect EPA policy.)
157
-------
AIR MONITORING AT ALERT IN THE HIGH ARCTIC: RESULTS OF ONE
YEAR OF MONITORING OF ORGANOCHLORINE COMPOUNDS AND PAH
D. Toom and L. Barrie
Atmospheric Environment Service
4905 Dufferin Street
Downsview, Ontario M3H 5T4
P. Fellin andD. Dougherty
Concord Environmental
2 Tippet Road
Toronto, Ontario M3H 2V2
D. Muir, B. Grift, L. Lockhart, andB. Billick
Freshwater Institute
Dept. of Fisheries and Oceans
501 University Crescent
Winnipeg, Manitoba R3T 2N6
In January 1992, an air toxics sampler was set up at Alert, on Northern Ellesmere
Island in the Canadian Arctic (82.5° N, 62.3° W) as part of an arctic air toxics monitoring
and assessment program. Since then, three more sites were added to the network: Tagish,
Yukon Territory near Whitehorse, Dunai Island in the former Soviet Union, and Cape
Dorset on Baffin Island. Organochlorines (OCs) and PAHs were sampled weekly to
determine the types, concentrations and vapour-particle relationships. High volume
samplers with a 10 micrometre size selective inlet were used with a collection cartridge
consisting of glass fibre filter(s) for particulates followed by two 20 cm diameter 5 cm
thick polyurethane foam plugs for vapours, with a weekly sample volume of
approximately 11,000 m3. Samples were analyzed by gas chromatography using electron
capture detection and confirmation by GC/MS.
This paper will focus on selected compounds for 1992 at Alert: a volatile OC
hexachlorobenzene, a semi-volatile OC DDT, and the PAHs pyrene, benzo(e) and
benzo(a) pyrene. We will present annual average concentrations as well as weekly
integrated values to look for potential seasonal variation through temporal profiles and
their distribution between the particle and gas phases.
LABORATORY PREPARATION OF DNPH DERIVATIVES OF CARBONYL
COMPOUNDS ON SEP-PAK® CARTRIDGES FOR QUALITY ASSURANCE
PURPOSES
Rita M. Harrell
ManTech Environmental Technology Incorporated
P.O. Box 12313
2 Triangle Drive
Research Triangle Park, NC 27709
Aldehydes and ketones are receiving increased attention both as hazardous
substances and as promoters in the photochemical formation of ozone in the atmosphere.
158
-------
They enter the atmosphere in the exhaust of motor vehicles and other equipment using
hydrocarbon and alcohol fuels. Formaldehyde, the most prevalent aldehyde, is widely
used as a preservative, a textile-treatment agent, and an intermediate in the manufacture
of urea-formaldehyde and phenol-formaldehyde resins. Waters Sep-Pak® DNPH-Silica
cartridges are convenient, reproducible sampling devices for quantifying aldehydes and
ketones in gases, including air. These cartridges trap the compounds by reacting them
with the DNPH, 2,4-Dinitrophenylhydrazine, in the cartridge to form stable hydrazone
derivatives. This derivatization reaction takes place during sample collection. Derivatives
are later eluted and analyzed by HPLC. Cartridges spiked in the laboratory are used for
quality assurance and instrument performance verification. This poster describes how
Mantech Environmental Technology Incorporated prepare such cartridges and the
proficiency involved.
EXPERIMENTAL STUDIES OF AN ETHANOL-AIR FLOW SUBJECTED TO
UV LIGHT
Timothy A. Spaeder
Applied Research Laboratory
The Pennsylvania State University
P.O. Box 30
State College, PA 16804
The photolytic destruction of ethanol in air streams illuminated with high purity
quartz low-pressure mercury lamps was investigated. A gas tight composite 316 stainless
steel and acrylic reactor loop was constructed which allowed for doping ambient air with
ethanol, and circulating it past the lamps. Ethanol concentration measurements were
recorded when only one 40 watt lamp was illuminated, and again when three 40 watt
lamps were illuminated. It was determined that by increasing the actinic flux from 40
watts to 120 watts the overall rate constant for ethanol increased from 8.25E-5 s"1 to
1.06E-3 s"1 when corrected for wall loss. Organic chemical intermediates were formed
by this process, and depended upon the actinic flux and the residence time in the reactor
loop. Finally, it was determined the 7.2mW/CFM of 185 nm radiation increased the
kinetic rate constant by over two orders of magnitude.
159
-------
DETERMINATION OF TEST METHODS FOR INTERIOR ARCHITECTURAL
COATINGS
Donald A. Whitaker, Linda S. Sheldon, and Jeffrey T. Keever
Research Triangle Institute
Research Triangle Park, NC 27709
Niren L. Nagda
ICE Incorporated
Fairfax, VA 22031
Pauline Johnston
Office of Air and Radiation
U.S. EPA
Washington, DC 20460
Numerous building and consumer products used in indoor environments emit VOCs.
The U.S. EPA has undertaken a project to analyze VOC emissions from various types of
indoor products to determine which types of products result in the greatest overall
exposures to pollutants indoors. Critical to this project are appropriate techniques to
evaluate emissions of total and specific VOCs from these sources. During 1992, the
Research Triangle Institute (RTI) evaluated seven methods for determining the VOC
emissions from interior architectural coatings-specifically interior alkyd and latex paints.
Current work has involved a more detailed evaluation of three of the seven methods with
the final result being the development of test methods for the determination of volatile
and semi-volatile organic compounds and aldehydes from interior latex and alkyd paints.
The methods evaluated during this study were bulk product analysis by direct
injection onto a GC/MS, emissions testing using 52 L small stainless steel chambers and
ASTM standard methods for total volatiles and water content. Twelve interior paints
made by two major paint manufacturers were chosen for testing. The paints represented
three floss types (flat, semi-gloss, and gloss) and six color groups. This paper discusses
the analytical test methods and results.
Although the research described was funded by the U.S. EPA (Contract No. 68-
D2-0131), it has not been subjected to the required peer review and does not necessarily
reflect the views of the Agency and no official endorsement should be inferred.
AN IMPROVED DNPH CARTRIDGE ALDEHYDE SAMPLER FOR 3-DAY
UNATTENDED SAMPLING
D.B. Cardin andEA. Galoustian
Entech Laboratory Automation
950 Enchanted Way #101
Simi Valley, CA 93065
Title I of the 1990 Clean Air Act Amendment requires monitoring of all compounds
suspected of contributing to ozone formation in urban atmospheres. Included in the list
of suspected agents are aldehydes and ketones, with formaldehyde being of particular
interest due to its toxicity and normally higher concentrations. Monitoring of the
160
-------
aldehydes and ketones during high ozone periods will be performed by collecting eight
3-hour samples daily, seven days a week using samplers operating unattended. The
aldehydes and ketones will be sampled by passing ambient air through
Dinitrophenylhydrazine (DNPH) cartridges allowing later recovery and analysis in a
laboratory.
A new aldehyde/ketone sampler is presented that conforms to the requirements and
QA/QC set forth by Federal, State, and Local monitoring agencies. The 32-position
sampler allows up to 4 days of unattended sampling if spikes and duplicates are not
utilized. To improve QA, 3-days of sampling can be scheduled allowing one duplicate
and one spike to be obtained every 24 hours in addition to the eight ambient air samples.
Two sampler positions left unused can them be analyzed as field blanks. To provide for
the simultaneous collection of samples and spikes or duplicates, two 2000 seem electron
mass flow controllers are used to control the 11/min required flow rates. An onboard
microcomputer both controls the MFC's and records their actual flow rates for later
downloading to a Windows based software package for reporting. Programming and data
retrieval can be performed using a local PC system running under Window™ or by a
remote system using a modem. Provisions are made for continuously adding a surrogate
standard for added quality assurance for both field sampling and laboratory extraction
and analysis. An ozone scrubber is utilized which reduces sample losses on the cartridge
due to oxidation of the carbonyl-DNPH complex. Default sampling protocols are
provided, although a flexible Windows based editor allows considerable user
modification to the sample collection scheme. Field data and software will be presented
showing precision and pratical advantages.
161
------- |