vvEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-91-008
January 1991
Air
1990 NONMETHANE
ORGANIC COMPOUND
AND
THREE-HOUR
AIR TOXICS MONITORING
PROGRAM
-------
EPA-450/4-91-008
1990 NONMETHANE
ORGANIC COMPOUND
AND
THREE-HOUR
AIR Toxics MONITORING
PROGRAM
By
Robert A. McAllister
Phyllis L. O'Hara
Dave-Paul Dayton
John E. Robbins
Robert F. Jongleux
Raymond G. Merrill, Jr.
Joann Rice
Emily G. Bowles
Radian Corporation
Research Triangle Park, NC 27709
EPA Contract No. 68D80014
EPA Contract Officer: Neil J. Berg, Jr.
Office Of Air Quality Planning And Standards
Office Of Air And Radiation
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
January 1991 y s Env?rorimenta( proM,,ion ,r,nc>,
Region 5, Library (PL-V^. " '
77 West Jackson BCL;, , , i:-Lh
Chicago, IL 60604-35* J ' "" '
-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U. S. Environmental
Protection Agency, and has been approved for publication as received from the contractor. Approval does
not signify that the contents necessarily reflect the views and policies of the Agency, neither does mention
of trade names or commercial products constitute endorsement or recommendation for use.
EPA-450/4-91-008
-------
TABLE OF CONTENTS
Section . Page
LIST OF FIGURES vii
LIST OF TABLES x
SYMBOLS AND ABBREVIATIONS xiii
1.0 SUMMARY AND CONCLUSIONS 1-1
1.1 NMOC MONITORING PROGRAM 1-2
1.1.1 Introduction and Data Summary 1-2
1.1.2 Calibration and Drift 1-5
1.1.3 Precision 1-5
1.1.4 Accuracy 1-5
1.1.5 Other Quality Assurance Measurements 1-6
1.2 THREE-HOUR AIR TOXICS MONITORING PROGRAM 1-16
1.2.1 Overall Data Summary 1-16
1.2.2 Site Results 1-16
1.2.3 Gas Chromatography/Mass Spectrometry
Confirmation Results 1-17
1.2.4 Precision 1-17
1.2.5- External Audit : . . 1-17
2.0 NMOC DATA SUMMARY 2-1
3.0 NMOC TECHNICAL NOTES 3-1
3.1 NMOC FIELD SAMPLING EQUIPMENT 3-1
3.1.1 Installation 3-1
3.1.2 Operation 3-3
3.1.3 Troubleshooting Instructions 3-4
3.1.4 Sampler Performance for 1990 3-6
3.1.5 Field Documentation 3-7
3.2 NMOC ANALYSIS 3-7
3.2.1 Instrumentation 3-7
3.2.2 Hewlett-Packard Model 5880 Gas Chromatograph
Operating Conditions 3-7
3.2.3 NMOC Analytical Technique 3-10
3.3 CANISTER CLEANUP SYSTEM 3-10
3.3.1 Canister Cleanup Equipment 3-12
3.3.2 Canister Cleanup Procedures 3-14
cah.!97f 11
-------
TABLE OF CONTENTS (Continued)
Section Page
4.0 NMOC QUALITY ASSURANCE/QUALITY CONTROL PROCEDURES 4-1
4.1 INTRODUCTION AND CONCLUSIONS 4-1
4.2 CALIBRATION AND INSTRUMENT PERFORMANCE 4-2
4.2.1 Performance Assessment 4-2
4.2.2 Calibration Zero, Span, and Drift 4-2
4.2.3 Calibration Drift 4-8
4.3 IN-HOUSE QC SAMPLES 4-18
4.4 REPEATED ANALYSES 4-24
4.4.1 Site Sample Results 4-27
4.4.2 Local Ambient Samples 4-31
4.5 DUPLICATE SAMPLE RESULTS 4-37
4.5.1 Analytical Precision 4-44
4.5.2 Components of Variance 4-44
4.6 CANISTER PRESSURE RESULTS 4-52
4.7 CANISTER CLEANUP RESULTS : 4-52
4.8 EXTERNAL AUDIT RESULTS 4-55
4.9 DATA VALIDATION 4-55
4.10 NMOC MONITORING PROGRAM RECORDS 4-62
4.10.1 Archives 4-63
4.10.2 Magnetic Disks 4-63
5.0 NMOC DATA ANALYSIS AND CHARACTERIZATION 5-1
5.1 OVERALL CHARACTERIZATION 5-1
5.2 MONTHLY VARIATIONS, 1984 - 1990 5-3
cah.!97f
-------
TABLE OF CONTENTS (Continued)
Section Page
6.0 RECOMMENDATIONS, NMOC MONITORING PROGRAM 6-1
6.1 SITING CRITERIA 6-1
6.2 OPERATING PROCEDURE CHANGES 6-1
6.3 VERTICAL STRATIFICATION STUDY . . . 6-1
6.4 SEASONAL NMOC STUDIES 6-2
6.5 DIURNAL STUDIES 6-2
6.6 CANISTER CLEANUP STUDIES 6-2
6.7 COORDINATED SAMPLING AT NMOC SITES 6-3
6.8 FIELD AUDIT 6-3
6.9 DUPLICATE SAMPLE AND REPLICATE ANALYSIS 6-4
7.0 THREE-HOUR AIR TOXICS DATA SUMMARY 7-1
7.1 OVERALL RESULTS 7-1
7.2 SITE RESULTS 7-4
8.0 THREE-HOUR AIR TOXICS TECHNICAL NOTES 8-1
8.1 S.AMPLING EQUIPMENT AND INTERFACE 8-1
8.2 THREE-HOUR AIR TOXICS SAMPLING CERTIFICATION .... 8-1
8.2.1 Sampler Certification Blanks - Humidified
Zero Air 8-3
8.2.2 Sampler Certification Challenge - Selected
Target Compound 8-3
cah.!97f IV
-------
TABLE OF CONTENTS (Continued)
Section Page
8.3 STANDARDS GENERATION 8-3
8.4 CALIBRATION ZERO AND SPAN 8-8
8.5 GAS CHROMATOGRAPH/MULTIDETECTOR ANALYSIS AND
COMPOUND IDENTIFICATION 8-8
8.6 GAS CHROMATOGRAPH/MASS SPECTROMETER ANALYSIS AND
COMPOUND IDENTIFICATION CONFIRMATION 8-8
8.7 QA/QC DATA • 8-10
8.7.1 GC/MD and GC/MS Method Detection Limits . . . 8-10
8.7.2 Repeated Analyses 8-10
8.7.3 Duplicate Sample Results 8-13
8.7.4 GC/MS Confirmation Results 8-13
8.7.5 External Audits 8-13
8.8 DATA RECORDS 8-19
9.0 RECOMMENDATIONS, THREE-HOUR AIR TOXICS PROGRAM 9-1
9.1 COMPOUND STABILITY STUDIES 9-1
9.2 CANISTER CLEANUP STUDIES 9-1
9.3 CARBONYL STUDIES 9-2
10.0 CARBONYL SAMPLING, ANALYSIS, AND QUALITY
ASSURANCE PROCEDURES 10-1
10.1 SAMPLING EQUIPMENT AND PROCEDURES 10-1
10.2 ANALYTICAL PROCEDURES 10-3
10.3 QUALITY ASSURANCE PROCEDURES 10-4
10.4 CALIBRATION PROCEDURES 10-4
10.4.1 Daily Quality Control Procedures 10-4
10.4.2 Duplicate Samples 10-4
10.4.3 Trip Blanks 10-6
10.4.4 Precision 10-6
10.4.5 Practical Quantitation Limit 10-6
cah.!97f
-------
TABLE OF CONTENTS (Continued)
Section Page
10.5 RESULTS 10-6
10.5.1 Formaldehyde Control Standards 10-11
10.5.2 Recoveries of Spikes 10-11
10.5.3 Blanks 10-15
11.0 REFERENCES 11-1
APPENDICES
APPENDIX A: SAMPLING SITES FOR 1990 NMOC MONITORING PROGRAM
APPENDIX B: CRYOGENIC PRECONCENTRATION AND DIRECT FLAME
IONIZATION DETECTION (PDFID) METHOD
APPENDIX C: 1990 NMOC MONITORING PROGRAM SITE DATA
APPENDIX D: 1990 NMOC MONITORING PROGRAM INVALIDATED AND MISSING
SAMPLES
APPENDIX E: PDFID INTEGRATOR PROGRAMMING INSTRUCTIONS
APPENDIX F: 1990 NMOC DAILY CALIBRATION DATA
APPENDIX G: 1990 NMOC IN-HOUSE QUALITY CONTROL SAMPLES
APPENDIX H: MULTIPLE DETECTOR SPECIATED THREE-HOUR SITE DATA SUMMARIES
cah.!97f VI
-------
LIST OF FIGURES
Number Page
1-1 In-house propane QC results Channel A 1-7
1-2 In-House propane QC results Channel B 1-8
1-3 In-house propane QC results Channel C 1-9
1-4 In-House propane QC results Channel D 1-10
1-5 Orthogonal regression comparing QAD with Radian NMOC
analyses 1-14
3-1 Sampling system for collecting 3-hour integrated
ambient air samples 3-2
3-2 NMOC sampling field data form 3-5
3-3 NMOC invalid sample form 3-8
3-4 NMOC analytical equipment 3-11
3-5 Canister cleanup apparatus 3-13
4-1 NMOC performance results, Channel A 4-4
4-2 NMOC performance results, Channel B 4-5
4-3 NMOC performance results, Channel C 4-6
4-4 NMOC performance results, Channel D 4-7
4-5 Daily calibration zero, Channel A 4-9
• 4-6 Daily calibration zero, Channel B 4-10
4-7 Daily calibration zero, Channel C 4-11
4-8 Daily calibration zero, Channel D 4-12
4-9 Daily calibration span, Channel A 4-13
cah.!97f VII
-------
LIST OF FIGURES (Continued)
Number Page
4-10 Daily calibration span, Channel B 4-14
4-11 Daily calibration span, Channel C 4-15
4-12 Daily calibration span, Channel D 4-16
4-13 In-house quality control results, Channel A 4-19
4-14 In-house quality control results, Channel B 4-20
4-15 In-house quality control results, Channel C 4-21
4-16 In-house quality control results, Channel D 4-22
4-17 Orthogonal regression comparing QAD with Radian
NMOC analyses 4-28
4-18 95% Confidence intervals for mean NMOC difference .... 4-39
4-19 Audit bias, Radian Channel A vs. EPA-QAD 4-57
4-20 Audit bias, Radian Channel B vs. EPA-QAD 4-58
4-21 Audit bias, Radian Channel C vs. EPA-QAD 4-59
4-22 Audit bias, Radian Channel D vs. EPA-QAD 4-60
5-1 Stem-and-leaf plot of the 1990 NMOC data 5-2
5-2 Stem-and-leaf plot for the In(NMOC) data 5-4
5-3 Cumulative frequency distribution for 1990 NMOC data . . 5-5
5-4 Cumulative frequency distribution for 1990 In(NMOC)
data 5-6
5-5 Stem-and-leaf plot of the NMOC data for
June, 1990 5-9
5-6 Stem-and-leaf plot of the NMOC data for
July, 1990 5-10
cah.!97f
-------
LIST OF FIGURES (Continued)
Number Page
5-7 Stem-and-leaf plot of the NMOC data for
August, 1990 5-11
5-8 Stem-and-leaf plot of the NMOC data for
September, 1990 5-12
5-9 Monthly mean NMOC emissions for 1985 through 1990 5-13
8-1 Gas chromatographic multidetector system 8-2
8-2 Dynamic flow dilution apparatus 8-7
8-3 Air toxics multiple detector system 8-9
10-1 3-Hour carbonyl sampling subsystem 10-2
10-2 Field data and custody form 10-5
cah.!97f IX
-------
LIST OF TABLES
Number Page
1-1
1-2
1-3
1-4
1-5
1-6
2-1
2-2
2-3
3-1
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
1990 NMOC COMPLETENESS RESULTS
1990 NMOC SITE STATISTICS
LINEAR REGRESSION PARAMETERS FOR IN-HOUSE QUALITY
CONTROL DATA
AUDIT SAMPLE RESULTS, PERCENT BIAS
AUDIT SAMPLE RESULTS, ABSOLUTE PERCENT BIAS
ORTHOGONAL REGRESSION PARAMETERS FOR COMPARATIVE
ANALYSES OF SITE SAMPLES
1990 NMOC COMPLETENESS RESULTS
1990 NMOC SITE STATISTICS
1990 NMOC LOGNORMAL STATISTICS
SUPPORT GAS OPERATING CONDITIONS
1990 PERFORMANCE ASSESSMENT SUMMARY, RADIAN CHANNELS . .
SUMMARY NMOC CALIBRATION FACTOR DRIFT RESULTS
LINEAR REGRESSION PARAMETERS FOR IN-HOUSE QUALITY
CONTROL DATA
IN-HOUSE QUALITY CONTROL STATISTICS, BY RADIAN
CHANNEL
OVERALL IN-HOUSE QUALITY CONTROL STATISTICS
ORTHOGONAL REGRESSION PARAMETERS FOR REPEATED
ANALYSES OF SITE SAMPLES
SUMMARY STATISTICS OF COMPARATIVE ANALYSES FOR
RADIAN VS. QAD CHANNELS
SUMMARY STATISTICS OF COMPARATIVE ANALYSES FOR RADIAN
VS. QAD CHANNELS, BY RADIAN CHANNELS
1-3
1-4
1-11
1-12
1-13
1-15
2-2
2-4
2-5
3-9
4-3
4-17
4-23
4-25
4-26
4-29
4-30
4-32
cah.!97f
-------
LIST OF TABLES (Continued)
Number Page
4-9 OVERALL STATISTICS FOR LOCAL AMBIENT SAMPLES 4-33
4-10 SUMMARY STATISTICS LOCAL AMBIENT SAMPLES,
BY CHANNEL PAIR 4-34
4-11 LOCAL AMBIENT SAMPLES CONFIDENCE INTERVALS 4-38
4-12 DUPLICATE SAMPLES FOR 1990 4-40
4-13 REPLICATE ANALYSES FOR THE 1990 NMOC MONITORING PROGRAM . . 4-45
4-14 ANOVA FOR DUPLICATE SAMPLES 1 THROUGH 15 4-48
4-15 ANOVA FOR DUPLICATE SAMPLES 16 THROUGH 59 4-49
4-16 COMPONENTS OF VARIANCE, NMOC SAMPLING AND ANALYSIS 4-51
4-17 NMOC PRESSURE STATISTICS 4-53
4-18 PRESSURE DISTRIBUTION OF NMOC AMBIENT AIR SAMPLES .... 4-54
4-19 1990 NMOC AUDIT SAMPLE RESULTS 4-56
4-20 AUDIT SAMPLES, RELATIVE TO EPA-QUALITY ASSURANCE
DIVISION (QAD) RESULTS 4-61
5-1 SUMMARY STATISTICS FOR 1990 NMOC SITES,
BY MONTH 5-7
7-1 THREE-HOUR AMBIENT AIR SAMPLES AND ANALYSES 7-2
7-2 COMPOUND IDENTIFICATION WITH GC/MD FOR ALL
3-HOUR SITES . . . 7-3
f
7-3 FREQUENCY OF OCCURRENCE OF TARGET COMPOUNDS IN
3-HOUR AMBIENT AIR SAMPLES 7-5
7-4 COMPOUND IDENTIFICATIONS WITH GC/MD BY SITE CODE 7-6
cah.!97f XI
-------
LIST OF TABLES (Continued)
Number Page
8-1 SAMPLER CERTIFICATION ZERO RESULTS 8-4
8-2 SAMPLER CERTIFICATION CHALLENGE RESULTS 8-5
8-3 METHOD DETECTION LIMITS FOR 3-HOUR AIR TOXICS
COMPOUNDS 8-11
8-4 3-HOUR AIR TOXICS REPLICATE ANALYSES BY GC/MD 8-12
8-5 SINGLE COMPOUND IDENTIFICATIONS OF GC/MD
REPLICATE SAMPLE ANALYSES 8-14
8-6 THREE-HOUR AIR TOXICS DUPLICATE SAMPLE ANALYSES BY GC/MD . 8-15
8-7 GC/MD 3-HOUR AIR TOXICS DUPLICATE PRECISION BY COMPOUND . 8-16
8-8 SINGLE COMPOUND IDENTIFICATIONS OF GC/MD DUPLICATE
SAMPLE ANALYSES 8-17
8-9 COMPOUND IDENTIFICATION CONFIRMATION 8-18
10-1 PRACTICAL CARBONYL QUANTITATION LIMIT 10-7
10-2 CARBONYL RESULTS FOR BATON ROUGE, LA (BRLA) 10-8
10-3 CARBONYL RESULTS FOR NEWARK, NJ (NWNJ) 10-9
10-4 CARBONYL RESULTS FOR PLAINFIELD, NJ (PLNJ) 10-10
10-5 CARBONYL LABORATORY REPLICATES 10-12
10-6 ANALYSIS OF QUALITY CONTROL STANDARDS 10-13
10-7 FORMALDEHYDE LABORATORY SPIKES 10-14
10-8 FIELD SPIKE RECOVERIES OF FORMALDEHYDE 10-16
10-9 CARBONYL SPIKE RECOVERIES 10-17
10-10 CARBONYL LABORATORY BLANKS 10-18
10-11 CARBONYL FIELD BLANKS 10-19
cah.!97f XII
-------
SYMBOLS AND ABBREVIATIONS
AC, or
A.C. area counts, generated from a gas chromatograph
ADELTA absolute value of DELTA
ADIF absolute value of DIP
ADIFF absolute value of DIFF
AIRS Aerometric Information Retrieval System
a.m. ante meridiem
APDIFF absolute value of PDIFF
APDIF absolute value of PDIF
APR April
AREAL Atmospheric Research and Exposure Assessment Laboratory
Aug August
Bldg.
BMTX
building
Beaumont, TX - AIRS No. 48-245-0009
Cal., or
Calib.
cm
CRM
calibration
centimeter
Certified Reference Material
DELTA
DIF
DIFF
Dup.
Radian NMOC concentration - QAD NMOC concentration, ppmC;
Radian NMOC concentration - ASRL concentration, ppmC; or
AREAL NMOC concentration - QAD NMOC concentration, ppmC
(NMOC concentration for the second channel) - (NMOC
concentration for the first channel
measured NMOC concentration - calculated NMOC concentration
ppmC for in-house quality control samples
duplicate
e
ECD
EPA
base of natural logarithm, 2.71828...
electron capture detector
United States Environmental Protection Agency
F
FID
Friday
flame ionization detector
(Continued)
cah.!97f
xi n
-------
SYMBOLS AND ABBREVIATIONS (continued)
GC/ECD gas chromatography electron capture detection
GC/FID gas chromatography flame ionization detection
GC/MD gas chromatography multidetection
GC/MS gas chromatography mass spectrometry
H Thursday
Hg mercury
HTCT Hartford, CT - AIRS No. 09-003-1003
i .d. inside diameter
ID identification
INST. instrument
Jul July
Jun June
L liter
LINY Hempstead, NY (Long Island) - AIRS No. 36-059-0005
Lpm liters per minute
m meter
M Monday
MAX maximum
/*g microgram
MID multiple ion detection
MIN minimum
min. minute
mL milliliter
mm millimeter
MNY New York, NY - AIRS No. 36-061-0010
MU mean of In(NMOC)
NC North Carolina
NIST National Institute of Standards and Technology
NMOC Nonmethane organic compound
NOX oxides of nitrogen
NWNJ Newark, NJ - AIRS No. 34-013-0011
(Continued)
cah.!97f XIV
-------
SYMBOLS AND ABBREVIATIONS (continued)
Oct October
o.d. outside diameter
Off. Office
PCDIFF percent difference = DIFF/calculated NMOC concentration x 100,
for in-house QC samples
PDELTA DELTA x 100;
[(Radian NMOC concentration + QAD NMOC concentration)/2]
DELTA x 100;
[(Radian NMOC concentration + AREAL NMOC concentration)/2]
or,
DELTA x 100
[(AREAL NMOC concentration + QAD NMOC concentration)/2]
PDFID preconcentration, direct flame ionization detection
PDIF DIF/([(NMOC concentration, 1st channel) + (NMOC concentration,
2nd channel)]/2) x 100
PLNJ Plainfield, NJ - AIRS No. 34-035-1001
p.m. post meridiem
ppb parts per billion
ppbv parts per billion by volume
ppm parts per million
ppmC parts per million by volume as carbon
ppmv parts per million by volume
psi pounds (force) per square inch
psig pounds (force) per square inch gauge
QA quality assurance
QAD Quality Assurance Division (EPA)
QAPP Quality Assurance Project Plan
QC quality control
RAO Radian analysis order: RAO = 1 for the local ambient duplicate
sample analyzed first by Radian; RAO = 2 for the local ambient
duplicate sample analyzed first by EPA
RT retention time
RTP Research Triangle Park
(Continued)
cah.!97f XV
-------
SYMBOLS AND ABBREVIATIONS (continued)
SAROAD
Sep
SOP
SOx
SRM
SIGMA
STD. DEV.
SD
Storage and Retrieval of Aerometric Data
September
standard operating procedure
oxides of sulfur
Standard Reference Material
standard deviation of In(NMOC)
standard deviation
Tuesday
UATMP Urban Air Toxics Monitoring Program
ug microgram
U.S. United States
UTM Universal Transverse Mercator
Wednesday
°C
°F
%CV
degrees Celsius
degrees Fahrenheit
percent coefficient
of variation
cah.!97f
XVI
-------
1.0 SUMMARY AND CONCLUSIONS
In certain areas of the country where the National Ambient Air Quality
Standard (NAAQS) for ozone is being exceeded, additional measurements of
ambient nonmethane organic compounds (NMOC) are needed to assist the affected
states in developing revised ozone control strategies. Because of previous
difficulty in obtaining accurate NMOC measurements, the U.S. Environmental
Protection Agency (EPA) has provided monitoring and analytical assistance to
these states through Radian Corporation. This assistance began in 1984 and
continues through the 1990 NMOC Monitoring Program.
Between June 4 and September 28, 1990, Radian analyzed 593 ambient air
samples, including 59 duplicate samples, collected in SUMMA® polished
stainless steel canisters at 7 sites. These NMOC analyses were performed by
the cryogenic preconcentration, direct flame ionization detection (PDFID)
method.1 Based on the 1984 through 1989 studies, the method was shown to be
precise, accurate, and cost effective relative to the capillary column gas
chromatographic, flame ionization detection (GC/FID) method (see Appendix B).
The 1990 study confirmed these findings and supported the conclusion that the
PDFID method is the method of choice to measure NMOC concentration in ambient
air.
In 1986 specific toxic compounds, primarily aromatics and halocarbons,
were also determined in the ambient air samples used for the NMOC analyses.
In 1987 Radian Corporation developed a gas chromatographic multidetector
(GC/MD) method to determine the concentration -of 38 selected toxic organic
compounds in ambient air. In 1988, air toxic analyses were conducted by GC/MD
on ambient air samples taken at 13 of the 45 sites at which NMOC samples were
taken. In 1989, air toxic analyses were conducted on ambient air samples
taken at seven of the 23 sites at which NMOC samples were taken. In 1990, air
toxic analyses were conducted on ambient air samples taken at three sites at
which NMOC samples were taken. These samples were called 3-hour air toxics
samples because the sampling period was three hours, from 6:00 a.m. to
9:00 a.m. A related monitoring program, the 1990 Urban Air Toxics Monitoring
Program (UATMP), began sampling in March 1990 at urban sites and extended
through February 1991. The samples from the latter program were 24-hour
integrated ambient air samples and are referred to as UATMP samples throughout
-------
this report. The final report for the 1990 UATMP will be presented under
separate cover.
The Final Report for the 1990 Nonmethane Organic Compound and
Three-Hour Air Toxics Monitoring program are included in Sections 1.0 through
12.0. Sections 1.0 through 6.0 report the data, procedures, and assessment of
the NMOC portion of the monitoring program. Sections 7.0 through 10.0 report
the data, procedures, and assessment of the 3-hour air toxics portion of the
monitoring program. Section 11.0 lists references.
The sampling sites for the 1990 NMOC Monitoring Program are listed in
Appendix A. Appendix A also gives the EPA Regions for each site, the Radian
Site Code, the Aerometric Information Retrieval System (AIRS) site code and
site information, and whether or not 3-hour air toxics analyses were performed
on selected ambient air samples from the site.
Appendix B contains the detailed instructions on the Cryogenic
Preconcentration and Direct Flame lonization Detection (PDFID) method.
Appendix C lists the 1990 NMOC Monitoring Program Site Data. Appendix D lists
the 1990 NMOC Monitoring Program Invalidated and Missing Samples information.
Appendix E gives PDFID Integrator Programming Instructions. Appendix F gives
1990 NMOC Daily Calibration Data. Appendix G gives 1990 In-House Quality
Control Samples, and Appendix H gives Multiple Detector Speciated Three-Hour
Site Data Summaries.
1.1 NMOC MONITORING PROGRAM
1.1.1 Introduction and Data Summary
The sampling schedule is given in the 1990 NMOC Quality Assurance
Project Plan.2 For all the sites in the 1990 NMOC Monitoring Program,
sampling occurred from 6:00 a.m. to 9:00 a.m., Monday through Friday from
June 4, 1990 through September 28, 1990. Table 1-1 gives details of the
sample completeness results. Percent completeness, a quality measure is shown
in Table 1-1. Completeness, which ratios the number of valid samples to the
number of scheduled samples, averaged 95.8% in 1990 compared to 95.5% in 1989,
93.4% in 1988, 95.0% in 1987, 96.8% in 1986, 95.8% in 1985, and 90.6% in 1984.
Percent completeness for 1990 ranged from 92.2 at HTCT to 100.00 for PLNJ.
Statistics for the NMOC concentrations in parts per million carbon (ppmC) y
volume are listed in Table 1-2. All sites collected samples from 6:00 a.r. to
9:00 a.m. Statistics in Table 1-2 include all duplicate sample
concentrations.
cah.!97f 1-2
-------
TABLE 1-1. 1990 NMOC COMPLETENESS RESULTS
Radian Scheduled
Site Sampling
Code Days
BMTX
BRLA
HTCT
LINY
MNY
NWNJ
PLNJ
OVERALL
84
74
84
83
77
84
84
570
Total
Scheduled
Duplicate
Samples
8
7
8
8
8
8
8
55
Total
Scheduled
Samples
92
81
92
91
85
92
91
625
Total
Valid
Duplicate
Samples
8
9
8
8
6
11
9
59
Total
Valid
Samples
83
77
85
86
81
90
il
593
Percent
Complete
96.74
95.06
92.24
94.51
95.29
97.83
100.00
95.84
cah.!97f
1-3
-------
TABLE 1-2. 1990 NMOC SITE STATISTICS
Radian
Site
Code
Sampling
BMTX
BRLA
HTCT
LINY
MNY
NWNJ
PLNJ
Overall
NMOC, DDmC
No. of
Samples Minimum
Median
6:00 to 9:00 a.m., local
83
77
85
86
81
90
91
593
0.558
0.055
0.066
0.060
0.186
0.110
0.012
0.012
1.440
0.738
0.205
0.298
0.502
0.381
0.319
0.726
Standard
Mean Maximum Deviation
time
1.717
1.062
0.257
0.409
0.803
0.464
0.478
0.727
4.283
14.255
0.738
1.867
6.200
2.040
2.374
14.255
0.901
1.742
0.145
0.353
0.888
0.284
0.441
0.940
Skewness
1.213
6.560
1.253
2.087
4.012
2.418
2.044
6.837
Kurtosis
0.733
46.953
1.239
4.915
19.113
10.155
4.743
82.333
cah.!97f
1-4
-------
1.1.2 Calibration and Drift
Each Radian PDFID channel was calibrated daily, using propane
standards referenced to the National Institute of Science and Technology
(NIST) Reference Material No. 1666B propane. Daily, before zero and
calibration checks were performed, the analytical systems were purged with
cleaned, dried air that had been humidified. Zero readings were determined
with cleaned, dried air. Daily percent drift of the calibration factor ranged
from -13.4% to +5.2 percent. The absolute value of the percent drift of the
daily calibration factors ranged from zero to 13.4 percent.
1.1.3 NMOC Precision
Analytical precision was determined by repeated analyses of 22 site
samples. Percent differences between the second and the first analysis
averaged -1.36 percent. The average of the absolute values of the percent
difference was 7.6% with a standard deviation of 10.3 ppmC. The analytical
precision includes the variability between Radian channels and within Radian
channels. The data quality objective for this measurement as published in the
1990 Quality Assurance Project Plan (QAPP)2 was 9.8%, based on previous NMOC
program experience3 with this measurement.
Overall precision, including sampling and analysis variability, was
determined by analysis of 59 duplicate site samples, simultaneously collected
in two canisters from a common sampling system. Percent difference for
Radian's analyses of the duplicates averaged 6.33 percent. The average
absolute percent difference was 12.3% with a standard deviation of 14.4 ppmC.
The data quality objective for this measurement was 12.2%, based on previous
experience.2
1.1.4 Accuracy
Because the NMOC measurements encompass a range of mixtures of unknown
compounds, it was not possible to define absolute accuracy. Instead, accuracy
was determined relative to propane standards with internal and external audit
samples.
Accuracy was monitored internally throughout the program by the use of
in-house propane standards. Periodically an in-house propane quality control
cah.!97f 1-5
-------
(QC) sample was prepared with a flow dilution apparatus and analyzed by the
PDFID method. The propane used to prepare the in-house QC standards was
certified by the EPA Quality Assurance Division (QAD) and was referenced to
NIST propane Certified Reference Material (CRM) No. 1666B.
Figures 1-1 through 1-4 show the in-house quality control results for
Radian Channels A, B, C, and D. Measured propane values are plotted against
calculated propane standards. Table 1-3 shows the linear regression
parameters for the Radian in-house quality control data. Quality control
samples of propane were mixed from a propane standard certified by EPA-QAD and
referenced to NIST propane Certified Reference Material (CRM) 1666B. The
regression used the propane concentration calculated from the mixing operation
as the independent variable and concentration measured by each Radian channel
as the dependent variable. The concentration range of the in-house quality
control samples was 0.159 to 1.464 ppmC. Table 1-3 indicates excellent
quality control for each channel since, as expected, the intercepts are all
near zero, and the slopes and coefficients of correlation are all near 1.0.
External propane audit samples were provided by EPA-QAD. The propane
samples were referenced to NIST propane Certified Reference Material
(CRM) 666B. Table 1-4 summarizes the percent bias of the Radian channels
relative to the EPA-QAD channel. The audit samples were given Radian ID
Numbers upon receipt. The average percent bias for the Radian channels was
1.07%, ranging from -3.18% for Channel D to +6.23% for Channel A. Absolute
percent biases are listed in Table 1-5 and range from 2.19% for Channel C to
6.96% for Channel D, averaging 5.04% overall for the Radian channels.
1.1.5 Other Quality Assurance Measurements
The results of other quality assurance measurements are discussed
below. Canister cleanup studies established that there was little carryover
of NMOC from one sample to the next, using the canister cleanup apparatus and
procedure developed for this study. In over 100 separate determinations,
percent cleanup averaged 99.747 percent. Cleanup was defined in terms of the
percent of the NMOC concentration that was removed in the cleanup cycle.
Figure 1-5 shows a between-laboratory comparison of site sample analyses
involving Radian channels and EPA-QAD channel for the PDFID method. Table 1-6
gives the orthogonal regression parameters, assuming a linear relationship,
cah.!97f 1-6
-------
CO
CO
LJJ
DC
O
O
LU<
<
Q_
DC
CL
LU
CO
O
O
q
c\i
o o o
03 CO
o
CM
o
O
o
00
o
co
•
o
o
<3-
•
o
o
CM
•
O
o
q
c\i
o
LOO
-CD
°0
: E
Q.
s Q.
O
CM
O
.0
0
CO
O
CO
•
O
o
o
o
-CM
O
O
.q
OO
o
O
C
O
o
T3
0
JO
O
73
o
0
CO
6
05
CO
0)
O
o
CO
g-
o
CO
o
I
I
2
•5
D)
Li.
(Oiudd) uoiiBJiuaouoQ OOIAIN pejnsB9|/\|
1-7
-------
CO
CO
LU
DC
O
O
LLJcQ
CL
DC
CL
LU
CO
O
O
q
c\i
o
CO
i
O
CD
O
CM
O
O
O
00
•
O
O
CD
•
O
O
^r
•
o
O
CM
(QOidd) UOI1184U90UOO OOIAIN
O
O
c\i
o
Uoo
o
UCD
c
o
KM
O
o c
o o
T-: O
0
CO
• OJ
o TJ
CO CD
• •+-•
O _03
o o
o o
o
.q
00
o
•
o
OQ
15
c
CO
6
CO
CO
o
o
O £
03
CT
CD
CO
cvj
I
CD
D
O)
Li.
1-8
-------
w^
•••^^
ID
CO
LJJ
CL
O
O
LJJO
^^ /IN
^S^^^M ^^«i
o
o
-CM
o
o
-0
DO
,
o
"CD
CC
jf^m
o
CO
CO
CD
"c
o
^!
"cc
o-
0
CO
13
1
—
co
CD
2J
0)
LJL
OOIAIN
1-9
-------
C/)
CO
LLJ
CC
O
O
LLJQ
Q_
Oo
DC
CL
LLJ
CO
O
O
q
cvi
o
oo
o
CO
o
CM
O
O
O
CO
•
O
o
CO
•
o
o
^r
•
o
o
CM
O
q
cvi
o
1-00
-CO
°0
e
Q.
C
f- .O
o S?
cvj ~
T- 0
O
0 £
o o
o
kco
•
o
o
I-co
•
o _
O
-------
TABLE 1-3. LINEAR REGRESSION PARAMETERS FOR
IN-HOUSE QUALITY CONTROL DATA
Radian
Channel
Cases
Intercept
Slope
Coefficient of
Correlation
A
B
C
D
5
5
5
5
0.008606
0.025200
-0.014900
-0.017800
1.060236
1.030042
1.048497
1.053398
0.997253
0.999370
0.995747
0.995685
cah.!97f
1-11
-------
TABLE 1-4. AUDIT SAMPLE RESULTS, PERCENT BIAS3
Channels
Radian
ID
Number
1053
1054
1073
1074
Average
Std. Dev.
A
Percent
Bias
2.19
7.16
10.51
5.06
6.23
3.50
B
Percent
Bias
-1.36
4.06
4.79
-8.93
-0.36
6.34
C
Percent
Bias
-0.39
2.78
4.79
-0.82
1.59
2.67
D
Percent
Bias
-4.65
4.17
3.40
-15.62
-3.18
9.21
Radian
Percent
Bias
1.07
6.01
"Percent Bias = [(Measured NMOC - QAD NMOC) / QAD NMOC] x 100.
cah.!97f
1-12
-------
TABLE 1-5. AUDIT SAMPLE RESULTS, ABSOLUTE PERCENT BIAS3
Radian
ID
Number
1053
1054
1073
1074
A
Percent
Bias
2.19
7.16
10.51
5.06
Channel
B
Percent
Bias
1.36
4.06
4.79
8.93
s
C
Percent
Bias
0.39
2.78
4.79
0.82
D
Percent
Bias
4.65
4.17
3.40
15.62
Radian
Absolute
Percent
Bias
Average 6.23 4.78 2.19 6.96 5.04
Std. Dev. 3.50 3.13 2.02 5.80 3.87
'Absolute'Percent Bias = ABS[[(Measured NMOC - QAD NMOC) / QAD NMOC] * 100]
cah.!97f
1-13
-------
CD
CO
03
CO
CD
0
O
D)
O
0)°
o
05
CO
QC
Q
<
O)
.
£
3
CO
CO
2
o>
CO
o
D)
(OUjdd) uQ[i?Jiueouoo OOIAIN
IT
1-14
-------
TABLE 1-6. ORTHOGONAL REGRESSION PARAMETERS FOR COMPARATIVE
ANALYSES OF SITE SAMPLES
Channel
Pair Coefficient of
(X-Y) Cases Intercept Slope Correlation
QAD-Radian 202 -0.038190 1.025019 0.995887
Radian-QAD 202 0.037260 0.975590 0.995887
cah.!97f 1-15
-------
for Figure 1-5 and the other possible comparison. The type of orthogonal
regression used here refers to a linear regression in which the Sum of Squares
of the perpendicular distances between the data points and the regression line
are minimized.
The results show good agreement because the intercepts are very close
to zero, the slopes are within 3% of unity, and the coefficients of
correlation are within 0.4% of unity. One hundred percent (100%) of the NMOC
data base was validated by checking data transcriptions from original data
sheets for 36 entries per sample. The errors found equal a data base error
rate of 0.956 percent. The data validation included 100% of the reported NMOC
concentration values. All errors that were found were corrected.
1.2 THREE-HOUR AIR TOXICS MONITORING PROGRAM
At three sites, Baton Rouge, LA (BRLA); Newark, NJ (NWNJ); and
Plainfield, NJ (PLNJ), 3-hour NMOC samples were speciated by a GC/MD
analytical system for 38 UATMP target compounds for a total of 25 NMOC ambient
air samples. After NMOC analysis, the NMOC sample canisters were bled to
atmospheric pressure, stored at least 18 hours for equilibration, and then
analyzed by GC/MD. Duplicate samples were collected at all three of the sites
simultaneously and analyzed individually by GC/MD. Replicate analyses were
performed on one duplicate sample per site. A total of 31 GC/MD analyses were
performed, including duplicate samples and replicate analyses.
1.2.1 Overall Data Summary
Twenty-six target compounds were identified in the 31 analyses.
Benzene, m/p-xylene, toluene, ethylbenzene, styrene/o-xylene, carbon
tetrachloride, and 1,1,1-trichloroethane were the most frequently identified
compounds. Concentrations of the target compounds identified ranged from
0.004 ppbv for n-octane to 20.60 ppbv for propylene. The overall average
concentration of the target compounds identified was 1.63 ppbv, averaged over
all sites and target compounds. The air toxics data are tabulated by site
code in Section 7 (Table 7-4) showing numbers of cases identified, minima,
maxima, and means for all target compounds.
1.2.2 Site Results
Overall site mean concentrations were 2.34 ppbv for BRLA, 1.34 ppbv
for NWNJ, and 1.12 ppbv for PLNJ averaged over all target compounds
identified. These air toxic data are presented in Section 7.0.
cah.!97f 1-16
-------
1.2.3 Gas Chromatoqraphy/Mass Spectrometrv Confirmation Results
Three 3-hour air toxics ambient air samples representing 10% of the
total samples were analyzed by Gas Chromatography/Mass Spectrometry (GC/MS)
for compound identification confirmation of the GC/MD analyses. The GC/MS
analyses were performed after the GC/MD analyses. The GC/MS analyses
confirmed 88.57% of the GC/MD analyses.
For the 3-hour air toxics samples the negative GC/MD-positive GC/MS
analyses were 9.52 percent. The positive GC/MD-negative GC/MS analyses were
1.90 percent. Comparisons labeled "negative GC/MD-positive GC/MS" refer to
specific samples in which a compound was not identified by GC/MD but
positively identified by GC/MS analysis. Comparisons labeled "positive GC/MD-
negative GC/MS" indicate specific samples in which a compound was positively
identified by GC/MD but not identified by GC/MS analysis.
1.2.4 Precision
Sampling and analytical precision of 3-hour air toxics samples was
estimated by analyzing duplicate samples. In terms of overall average
absolute percent difference, the sampling and analysis precision was
38.40 percent.
Analytical precision was estimated by repeated analyses of three
duplicate samples. The analytical precision measured by the overall average
absolute percent difference was 3.40 percent. Both the sampling and
analytical precision results are excellent in view of the concentration range
found in this study.
The data analyses showed that both for the duplicate and replicate
results, the imprecision was significantly higher at concentrations less than
2 ppbv. Both the duplicate sample and repeated analyses results are discussed
in Section 8.0.
1.2.5 External Audit
The external audit for the 3-hour air toxics compounds is conducted
bimonthly on the Urban Air Toxics Program and the results will be reported in
the 1990 UATMP Final Report. The audit samples that are used are furnished by
the Quality Assurance Division of the U.S. EPA.
cah.!97f 1-17
-------
2.0 NMOC DATA SUMMARY
This section presents the data summary for the 1990 NMOC Monitoring
Program conducted during June, July, August, and September. Daily NMOC
concentrations and other pertinent monitoring data are given by site in
Appendix C. The majority of the data presented in this section summarize the
NMOC concentrations measured for samples collected at seven sites throughout
the continental United States. Sites were selected in urban and/or industrial
locations; they are described in Appendix A. The site codes for the 1990 NMOC
Monitoring Program are listed in Appendix A and are used throughout the report
to identify the sites. Samples were collected in 6-liter (L) stainless steel
canisters by local site operators trained by Radian Corporation personnel.
The sampling procedure was described in detailed written instructions and
given to the site operators. The sampling procedure instructions also appear
in Section 3.1.2. Analytical concentration measurements of NMOC were made in
the Radian Corporation Research Triangle Park (North Carolina) laboratory
according to the PDFID method TO-12.1 The complete procedure is described in
Appendix B. . -
The concentration of oxides of nitrogen (NOJ, site temperature,
barometric pressure, wind direction, and weather conditions were provided on
the field sampling forms by site personnel at the time of sampling. These
data were recorded in the 1990 NMOC data base, but are not presented in this
report because they were not measured by Radian equipment or personnel, nor
were the data subjected to project quality assurance procedures.
Table 2-1 lists the NMOC Monitoring Program completeness results by
site code. The scheduling of sample days and the scheduling of duplicate
analyses is given in the QAPP.2 For the 1990 NMOC sites, completeness was
over 90%, and generally very near to 100 percent. A complete listing of
invalid samples and the reasons for the invalidation are given in Appendix D.
Overall completeness figures for the 1990 NMOC Program show 95.8%
complete. This compares with 95.5% in 1989, 93.4% in 1988, 95.0% complete in
1987, 96.8% complete in 1986, 95.8% complete in 1985 and 90.6% complete in
I 984 _ 3.4.5,6.7.8
-------
TABLE 2-1. 1990 NMOC COMPLETENESS RESULTS
Total
Radian Scheduled Scheduled
Site Sampling Duplicate
Code Days Samples
BMTX
BRLA
HTCT
LINY
MNY
NWNJ
PLNJ
OVERALL
84
74
84
83
77
84
84
570
8
7
8
8
8
8
8
55
Total
Scheduled
Samples
92
81
92
91
85
92
21
625
Total
Valid
Dupl icate
Samples
8
9
8
8
6
11
9
59
Total
Valid
Samples
83
77
85
86
81
90
91
593
Percent
Complete
96.74
95.06
92.24
94.51
95.29
97.83
100.00
95.84
cah.!97f
2-2
-------
Completeness was defined as the percentage of samples, scheduled in the QAPP,2
that were collected and analyzed as valid samples, beginning with the first
valid sample and ending with the last scheduled sample, with the exception of
MNY. An unexpected site situation forced MNY to collect its last sample on
September 20, 1990.
Table 2-2 summarizes statistics by sites. All sites collected an
integrated sample from 6:00 a.m. to 9:00 a.m. The overall average of the NMOC
concentration is seen to be 0.716 ppmC. The averages pertain only to the
sites for the 1990 Monitoring Program.
In Table 2-2, the means are the arithmetic averages of the NMOC
concentrations at each site. The numbers given for standard deviation,
skewness, and kurtosis are the second, third, and fourth moments, respectively
about the arithmetic means. A skewness value greater than zero applies to
distributions having a longer tail to the right. A distribution that is
normally distributed would have a kurtosis of 3.0. A distribution more peaked
(or pointed) than a normal distribution, having the same variance, would have
a kurtosis greater than 3.0. All the kurtosis figures listed in this report
are zero centered, which means that 3.0 has been subtracted from the fourth
moment to give a reported kurtosis of 0.0 for a symmetrical distribution.
NMOC monitoring data can be better characterized by a lognormal
distribution than by a normal distribution, following the findings of previous
years.3'4'5'6-7'8 Table 2-3 summarizes the 1990 NMOC data using the definitions
that characterize a lognormal distribution overall and for each site. MU and
SIGMA are the mean and standard deviation, respectively, of the logarithm of
NMOC to the Napierian base e. The geometric mean is e raised to the power MU;
the geometric standard deviation is e raised to the power SIGMA. The mode is
the most frequently occurring logarithm of NMOC value for a continuous
probability distribution function.
Information listed in Appendix A includes the location of the site,
street address as well as the Universal Transverse Mercator (UTM) coordinates
for the site, the site code used throughout this report, the Aerometric
Information Retrieval System (AIRS) Number. Appendix A gives the AIRS
printouts for all the sites that are in the system for 1990.
cah.!97f 2-3
-------
TABLE 2-2. 1990 NMOC SITE STATISTICS
NMOC
Radian
Site Code
Minimum
Sampling 6:00 to 9
BMTX
BRLA
HTCT
LINY
MNY
NWNJ
PLNJ
Overall
0
0
0
0
0
0
0
0
Median
:00 a.m. ,
.558
.055
.066
.060
.186
.110
.012
.012
1
0
0
0
0
0
0
0
local
.440
.738
.205
.298
.502
.381
.319
.726
Mean
time
1.717
1.062
0.257
0.409
0.803
0.464
0.478
0.727
. DornC
Standard
Maximum Deviation
4
14
0
1
6
2
2
14
.283
.255
.738
.867
.200
.040
.374
.255
0.901
1.742
0.145
0.353
0.888
0.284
0.441
0.940
Skewness
1
6
1
2
4
2
2
6
.213
.560
.253
.087
.012
.418
.044
.837
Kurtosis
0.733
46.953
1.239
4.915
19.113
10.155
4.743
82.333
cah.!97f 2-4
-------
TABLE 2-3. 1990 NMOC LOGNORMAL STATISTICS
Parameters of the Lognormal
Distribution of NMOC, DDmC
Radian
Site Code
Minimum
Mode Median
Sampling 6:00 to 9:00 a.m., local
BMTX
BRLA
HTCT
LINY
MNY
NWNJ
PLNJ
Overall
0.558
0.055
0.066
0.060
0.186
0.110
0.012
0.012
1.199
0.343
0.169
0.172
0.370
0.296
0.152
0.206
time
1.440
0.738
0.205
0.298
0.502
0,381
0.319
0.726
Mean3
1.715
0.995
0.256
0.407
0.761
0.463
0.497
0.711
Maximum
4.283
14.255
0.738
1.867
6.200
2.040
2.374
14.255
Parameters of
Logarithmic
Transformation
of NMOC
Concentrations
MUb SIGMAC
0.420
-0.360
-1.500
-1.185
-0.514
-0.919
-1.095
-0.754
0.489
0.843
0.528
0.758
0.694
0.547
0.889
0.909
aMean = exp (MU + SIGMA2/2).
bMU is the mean of In(NMOC). eMU is the geometric mean.
CSIGMA is the standard deviation of In (NMOC). es'GMA is called the
geometric standard deviation.
cah.!97f
2-5
-------
Appendix C gives the daily NMOC concentration data listed
chronologically for the entire sampling season. In addition, figures are
given for each site in which NMOC concentrations in ppmC are plotted versus
the 1990 Julian date on which the sample was taken. Data tables for each site
include the following:
• calendar date sampled;
• Julian date samples;
• weekday sample (M, T, W, H, F);
• sample ID number, assigned consecutively upon receipt of the
sample;
• sample canister number;
• Radian analysis channel;
• NMOC concentration in ppmC, determined by Radian; and
• NMOC concentration in ppmC, determined by U.S. EPA, Quality
Assurance Division.
Appendix D lists invalidated or missing samples. Table D-l lists
these data chronologically, while Table D-2 groups the listings by site code.
For each sample, the tables list the site code, the date of the missing or
invalid sample, a brief description of the possible cause of the invalid or
missing sample, and the assigned cause for the failure.
cah.!97f 2-6
-------
3.0 NMOC TECHNICAL NOTES
This section summarizes descriptions of the installation and operation
of the field sampling equipment, a summary of the analytical equipment and
procedures for NMOC measurement, and a description of the canister cleanup
equipment and procedures.
3.1 NMOC FIELD SAMPLING EQUIPMENT
The field sampling equipment used to collect ambient air samples for
NMOC measurement is relatively simple to operate. Ambient air is drawn
through a sintered stainless steel filter (2 micron) and critical orifice by a
Metal Bellows® pump and delivered to a SUMMA® canister. The sampler
components are made of stainless steel. Figure 3-1 is a schematic diagram of
the NMOC sampling system.
3.1.1 Installation
NMOC sampler installation configurations were site dependent. All
field sites were installed by or under the direction of Radian personnel.
Installation requirements included a temperature-controlled environment (70°
to 86°F), close proximity to the atmosphere to be sampled, and
noncontaminating sampler connections. Glass tubing or gas-chromatographic-
grade stainless steel tubing and stainless steel fittings are the preferred
materials of construction for all connections contacting the sample. Typical
sampler installations involved three configurations including direct
connections to a ventilated glass manifold, a slipstream connection prior to
the station NOX analyzer with a bypass pump, or collocated NMOC and NOX sample
inlet lines. For sites where the distance between the sample inlet and the
stainless steel post was greater than eight feet, an auxiliary pump, as shown
in Figure 3-1, was used. The auxiliary pump helps ensure that the air in the
sample line is representative of the ambient air. The critical orifice was
sized to maintain a constant flow rate and to fill a 6-L stainless steel
canister from the 0.5 mm Hg vacuum to about 15 psig in three hours. When
duplicate samples were taken, the critical orifice used for single sample
collection was replaced with an orifice sized to fill two canisters during the
3-hour sampling period.
-------
CO
03
cn
CO
ct
^
C
03
3
O
r—
ob
C5
O
_Q3
~0
a
c
OJ
CO
O5
C
CO
CO
CO
CD
i_
3
C3
U_
crt
cah.!97f
3-2
-------
3.1.2 Operation
Presampling
The following instructions pertain to the sampling operation prior to
collection of the field sample.
1. Verify timer program (see timer instructions). Set to MANUAL
position to leak check sampling system. Once the system passes
the leak check, turn timer to AUTO position.
2. With no canisters connected to the sampling system, turn the
timer switch to the MANUAL position.
3. Disconnect the sample inlet from the top of the orifice/filter
assembly mounted on the pump inlet. Connect the rotameter to
the top of the orifice/filter assembly. Tighten Swagelok*
(1/4") fitting securely with a wrench. Do not overtighten.
4. Turn timer switch ON. Do not turn the power off and on rapidly.
Wait 20 seconds between cycles to prevent premature
timer/solenoid failure. The pump should run and the latching
valve should open (audible click with 2 to 5 seconds delay).
Verify that the rotameter reading is approximately the same
(±15%) as the reading obtained during installation as
recommended on the orifice tag. If the rotameter reading is not
correct, see the troubleshooting instructions,
5. Allow the pump to run for at least 20 seconds, then press the
timer OFF button.
6. Connect a cleaned, evacuated canister to the sampling system.
If duplicate samples are to be collected, remove the plug from
the second port of the tee and connect a second canister to the
sampling system. Remove the orifice assembly marked with an
"S," denoting a single orifice. Install the orifice assembly
marked with a "D," denoting a double orifice. Replace the
filter holder on the "D" orifice. After obtaining scheduled
duplicate samples, replace the plug and the "S" orifice assembly
to return to single sample collection status.
7. With the pump off, open completely the valve on the canister (or
on one of the canisters if two are connected) and verify that no
flow is registered on the rotameter. If any flow is detected by
the rotameter, immediately close the canister valve and see the
troubleshooting instructions.
8. If no flow is observed, disconnect the rotameter and reconnect
the inlet sample line to the filter assembly. If two canisters
are connected, completely open the valve on the second canister.
cah.!97f 3-3
-------
9. Reverify that the canister valve(s) is (are) completely open and
the timer is properly set for sampling from 6 a.m. to 9 a.m. the
next weekday. Set timer to AUTO mode.
10. Reset the elapsed time counter.
Postsamplino.
The instructions that follow outline the NMOC postsampling operation
procedures in the field.
1. Close the canister valve(s) firmly. Disconnect the canister(s)
from the sampling system.
2. Connect the pressure gauge to the canister inlet and open the
canister valve. Record the canister pressure on the field
sampling data form. Close the canister valve and remove the
pressure gauge. Repeat pressure measurement for second canister
if collecting a duplicate sample. If the pressure reading is
not at least 11 psig, see the troubleshooting instructions.
3. Fill in the required information on the NMOC SAMPLING FIELD DATA
FORM. PLEASE PRESS HARD AND WRITE WITH A BALLPOINT PEN; YOU ARE
MAKING THREE COPIES, (see Figure 3-2).
4. Verify elapsed time counter reading equals 3 hours.
5. Verify that the timer shows the correct time setting. If not,
note that fact on the sample form along with any information
pertaining to the possible cause. Reset the timer to the
correct time, if necessary.
6. Verify that the canister valves are closed firmly. Do not
overtighten them. Put the protective cap(s) on the valve(s) and
prepare the canister(s) for shipment to Radian, RTP.
3.1.3 Troubleshooting Instructions
A list of troubleshooting instructions was given to each field site
during the site installation and operator training. Typical problems
encountered with the field sampling apparatus included: loose fittings,
misprogrammed timer, or clogged orifices. To minimize downtime, field site
operators were encouraged to relay sampling problems to the Radian laboratory
daily, by telephone. Most sampling problems were addressed promptly through
these telephone discussions.
cah.!97f 3-4
-------
CORPORATION
NMOC SAMPLING FIELD DATA FORM
Site Code: SAROAD # : • •
Site Location : City:
Sample Collection Date :
Operator :
State:
Sampling Period :
Elapsed Time :
Final Canister Pressure (psig) :
Sample Canister Number:
Side :
Sample Duplicate for this Date : YesD NoD
If yes, Duplicate Canister Number :
NOx Analyzer Operating? YesD NoD
If yes, Average Reading (ppmv as NOx) :
Average Wind Speed :
Rotameter Indicated Flow Rate :
Average Wind Direction : _
Orifice Number :
Average Barometric Pressure (mm Hg or inches Hg) :
Ambient Temperature (°F) : Relative Humidity :
THC Model (if available) : Average THC : _
Sky/Weather Conditions :
Site Conditions/Remarks :.
Canister Number :
Initial Canister Vacuum
Received By :
Date :
Sample Validity : _
If Invalid, Reason :
Figure 3-2. NMOC Sampling field data form.
cah.!97f
3-5
-------
3.1.4 Sampler Performance for 1990
The NMOC sampler was modified in 1989 to improve performance. This
modification involved replacing the mechanical timer previously used with an
electronic version. The electronic timer improves sample integration. An
elapsed time counter was added to the sampler to verify sample duration. This
modified system was used during the 1990 program. In addition, all sampler
orifice(s) and canisters were subjected to a preseason QC check to ensure
field performance. All orifices were checked against the rotameter enclosed
in each sampling kit, and referenced to a transfer standard (bubble
flowmeter). Prior to field installation, all samplers were operated in the
laboratory to establish an expected final pressure range for the canister
samples. Two single orifices and one double orifice were tested for each
sampler kit.
Due to the preseason checks and modifications, the NMOC sampler
performance was improved for the 1990 sampling season. This assessment is
based on the consistency of the final sample pressures on a site-specific
basis (see Section 4.6). The sampler performance in terms of successful
sample collection (i.e., completeness) was comparable to previous years.
Overall completeness from all sites averaged 94.9 percent. The site-specific
completeness ranged from 92.2% for BRLA to 100.0% for PLNJ.
*->
Invalidated samples were primarily due to operator error and equipment
problems. Completeness can be improved at all sites through greater attention
to sampling procedure, and by ensuring that trained site personnel are
available.
A total of 26 invalidated/missing samples were recorded in the 1990
NMOC Monitoring Program. Appendix D lists the invalidated/missing samples in
chronological order. Also, the reason for invalidation is presented.
Avoidable operator error accounts for 69% and equipment problems account for
27% of the invalidated samples. The remaining 4% reflects one missed s~mple
collection.
A further improvement in completeness may be possible as site
operators gain familiarity with the electronic timer. Revised sampler
operating instructions will focus additional attention on timer programrm g
cah.!97f 3-6
-------
and operation, and will include a daily checklist to eliminate common operator
errors.
3.1.5 Field Documentation
The field sample collection information was documented by the site
operator on printed forms. Figure 3-2 is an example NMOC Sampling Field Data
Form. Each canister sent to the field was accompanied by this form. The
field data form is a multiple part unit. A copy of the field data form was
retained by the site operator for the site notebook. Figure 3-3 is the
Invalid Sample Form. This form was completed by the site operator to document
the reasons for a missed or invalid field sample collections.
3.2 NMOC ANALYSIS
The NMOC analysis equipment and analysis procedure are described in
greater detail in Appendix B. A brief description of the equipment and
operating procedure used in this study follows.
3.2.1 Instrumentation
Two gas chromatographs were used by Radian. Each was a dual-channel
Hewlett-Packard Model 5880 (HP-5880) using flame ionization detection (FID).
NMOC instrument Channels A and B refer to the two FIDs on one HP-5880 unit,
and Channels C and D refer to the two FIDs on the other HP-5880 unit. These
chromatographs were modified to be similar to the prototype unit (EPA-QAD
•%
instrument), which is described in Appendix B. The EPA-QAD instrument was
used as a reference during this program.
3.2.2 Hewlett-Packard, Model 5880. Gas Chromatograph Operating Conditions
The sample trap consisted of 30 cm of 1/8-inch outside diameter (o.d.)
stainless steel tubing, packed with 60/80 mesh glass beads.
Three support gases were used in this analysis: helium, hydrogen, and
hydrocarbon-free air. Details of their use are given in Table 3-1.
The operating temperatures of the HP-5880 were controlled for the NMOC
analysis. The FID and auxiliary area were controlled at 250°C and 90°C,
respectively. The oven temperature was programmed from 30"C to 90"C at a rate
of 30°C per minute for 4 minutes, holding at 90"C for the fourth minute. Oven
and integration parameters were controlled by HP Level 4 programmable
integrators. A complete listing of the integrator programming sequence for
NMOC measurement by the PDFID method is given in Appendix E.
cah.!97f 3-7
-------
Tl OH
V-
NMOC INVALID SAMPLE FORM
Site Code: SAROAD*: •
City: State:
Sample Collection Date: Operator:
Sample Canister Number:
Sample Duplicate for this Date : YesQ NoQ
If Yes, Duplicate Canister Number:
Reason for Invalid or Missed Sample:
Average NOx Analyzer Reading for this Collection Date:
Wind Speed: Wind Direction : __
Average Barometric Pressure (mm Hg or inches Hg):
Ambient Temperature. (*F): ] Relative Humidity
Sky/Weather Conditions:
Received By:
Date:
Action Taken:
Resolution : _ .
K
_ s
Field Invalid or In-house Invalid §
Figure 3-3. NMOC Invalid sample form.
cah.!97f
-------
TABLE 3-1. SUPPORT GAS OPERATING CONDITIONS
Purpose
Carrier Gas
FID Air
FID Fuel
Cylinder
Composition
Hel ium
Hydrocarbon-
free air
Hydrogen
Pressure
30 psig
30 psig
32 psig
Mean
Flow Rate3
29.5 mL/min
300.7 mL/min
29.0 mL/min
"Flow rates corrected to standard conditions (1 atmosphere pressure, 20°C)
cah.!97f 3-9
-------
3.2.3 NMOC Analytical Technique
The modified HP-5880, dual-FID chromatographs were operated during the
1990 study according to a project specific Standard Operating Procedure (SOP).
Further description is given below to help explain the analytical apparatus
and procedure.
The six-port valve shown in Figure 3-4 was installed in the auxiliary
heated zone of the HP-5880 and was pneumatically actuated using
chromatographic valve control signals to apply either compressed air or vacuum
to the valve. The sample trap itself was located inside the chromatograph's
column oven. A section of 1/16-inch o.d. stainless steel tubing was sized to
a length that prevented pressure and flow surges from extinguishing the FID
flame. This length was determined experimentally and differs for each
chromatograph and for each channel within chromatographs. Although the length
of tubing effectively substitutes for the pressure restriction provided by a
column, it does not perform the separation function of a column.
During sample trapping, a slight excess of sample gas flow was main-
tained. A pressure change of 80 mm Hg in a 1.7-L vacuum reservoir was used to
gauge and control the volume of sample gas cryogenically trapped. After the
trapping cycle was complete, the HP-5880 program shown in Appendix B was
initiated. When the program triggered a horn emitting an audible beep, the
cryogen was removed from the trap and the oven door was closed. The
chromatographic program then assumed control of raising the oven temperature,
at the preset rate, to release the trapped sample to the FID, and set up the
integration parameters.
3.3 CANISTER CLEANUP SYSTEM
A cleanup cycle consisted of first pulling a vacuum of 0.5 mm Hg
absolute pressure in the canister, followed by pressurizing the canister to 20
psig with cleaned, dried air that had been humidified. This cycle was
repeated two more times during the canister cleanup procedure. The cleanness
of the canister was qualified by PDFID analysis. Upon meeting the cleanness
criterion, the canister was evacuated to 0.5 mm Hg absolute pressure a fourth
time, in preparation for shipment to the site.
cah.!97f 3-10
-------
c59CO/S9
C
05
CO
C
co
u
O
CO
0)
u.
C3
cah.!97f
3-11
-------
3.3.1 Canister Cleanup Equipment
A canister cleanup system was developed and used to prepare sample
canisters for reuse after analysis. A diagram of the system is shown in
Figure 3-5. An oil-free compressor with a 12-gallon reservoir provided source
air for the system. The oil-free compressor was chosen to minimize
hydrocarbon contamination. The compressor reservoir was drained of condensed
water each morning. A coalescing filter provided water mist and particulate
matter removal down to a particle size of one micron. Permeation dryers
removed water vapor from the compressor source air. These permeation dryers
were followed by moisture indicators to show detectable moisture in the air
leaving the dryer. The moisture indicators never showed any water, indicating
that the permeation dryers effectively removed all of the water vapor.
Air was then passed through catalytic oxidizers to destroy residual
hydrocarbons. The oxidizers were followed by inline filters for secondary
particulate matter removal, and by a cryogenic trap to condense any water
formed in the catalytic oxidizers and any organic compound not destroyed by
the catalytic oxidizer. A single-stage regulator controlled the final air
pressure in the canisters and a metering valve was used to control the flow
rate at which the canisters were filled during the cleanup cycle. The flow
was indicated with a rotameter installed in the clean, dried air line. There
was a shutoff valve between the rotameters and the humidifier system. The
humidifier system consisted of a SUMMA* treated 6-L canister partially filled
with high performance liquid chromatographic-grade (HPLC-grade) water. One
flowmeter and flow-control valve routed the cleaned, dried air into the 6-L
canister where it was bubbled through the HPLC-grade water. A second flow-
control valve and flowmeter allowed air to bypass the canister/bubbler. By
setting the flow-control valves separately, the downstream relative humidity
was regulated. For the 1990 study, 80% relative humidity was used for
canister cleaning. There was another shutoff valve between the humidifier and
the 8-port manifold where the canisters were connected for cleanup.
The vacuum system consisted of a Precision Model DD-310 turbomolecular
vacuum pump, a cryogenic trap, an absolute pressure gauge, and a bellows valve
connected as shown in Figure 3-5. The cryogenic trap prevented the sample
cah.!97f 3-12
-------
atoiwz
*
i
a
s
< 03 O Q
cah.!97f
3-13
-------
canisters from being contaminated by back diffusion of hydrocarbons from the
vacuum pump into the cleanup system. There are no oil-free high vacuum pumps
currently available at a competitive cost. The bellows valves enabled
isolation of the vacuum pump from the system without shutting off the vacuum
pump.
3.3.2 Canister Cleanup Procedures
After NMOC analyses were completed, a bank of eight canisters was
connected to each manifold shown in Figure 3-5. The valve on each canister
was opened, with the shutoff valves and the bellows valves closed. The vacuum
pump was started and one of the bellows valves was opened, drawing a vacuum on
the canisters connected to the corresponding manifold. After reaching
0.5 mm Hg absolute pressure as indicated by the absolute pressure gauge, the
vacuum was maintained for 30 minutes on the eight canisters connected to the
manifold. The bellows valve was then closed and the cleaned, dried air that
had been humidified was introduced into the evacuated canisters until the
pressure reached 20 psig. The canisters were filled from the clean air system
at the rate of 7.0 L/min. This flow rate was recommended by the manufacturer
as the highest flow rate at which the catalytic oxidizers could handle
elimination of hydrocarbons with a minimum 99.7% efficiency.
When the first manifold had completed the evacuation phase and was
being pressurized, the second manifold was then subjected to vacuum by opening
its bellows valve. After 30 minutes, the second manifold was isolated from
the vacuum and connected to the clean, dried air that had been humidified.
The first manifold of canisters was then taken through a second cycle of
evacuation and pressurization. Each manifold bank of eight canisters was
subjected to three cleanup cycles.
During the third cleanup cycle, the canisters were pressurized to
20 psig with clean, dried air that had been humidified. For each bank of
eight canisters, the canister having the highest precleanup NMOC concentration
was selected for NMOC analysis to determine potential hydrocarbon residues.
If the analysis measured less than 0.020 ppmC, then the eight canisters on the
manifold were considered to be clean. Finally the canisters were again
evacuated to 0.5 mm Hg pressure absolute; they were capped under vacuum and
then packed in the containers used for shipping to the field sites.
cah.!97f 3-14
-------
4.0 NMOC QUALITY ASSURANCE/QUALITY CONTROL PROCEDURES
This section details the steps taken in the 1990 NMOC Monitoring
Program to ensure that the data taken were of known quality and were well
documented. Analysis results are given in terms of precision, completeness,
and accuracy. Repeated analyses provided analytical precision. Duplicate
samples provided sampling and analysis precision. Completeness was measured
in terms of percent of scheduled samples that resulted in valid samples,
beginning with the first valid site-specific sample collected and ending with
the last scheduled site-specific sample. Accuracy of NMOC concentrations was
reported as percent bias of audit samples referenced to an NIST SRM propane by
EPA-QAD.
4.1 INTRODUCTION AND CONCLUSIONS
Completeness for the 1990 NMOC study was 94.9 percent. This value
indicates that good communication and planning were maintained between the
site personnel and the laboratory personnel. Precision for the 1990 NMOC
study averaged 7.6% absolute percent difference of repeated analysis and
compared to 14.2% for the 1989 study, 10.1% for the 1988 study, 9.61% for the
1987 study, 9.01% for the 1986 study, and 10% for the 1985 study. The
absolute percent difference in 1990 was higher than in previous years and
probably related to the fact that the overall average NMOC concentration for
1990 was lower than in previous years. For smaller values of NMOC
concentrations, imprecision increases.
Bias of the Radian channels for the 1990 audit results ranged from
-3.2% to +6.2 percent. In 1989 the accuracy determined from the external
audit samples ranged from +1.3% to +4.5%, from 1.3% to 4.5% in 1988, and from
-2.9% to -0.06% in 1987. In 1986 bias ranged from -0.52% to -3.3% and in 1985
bias ranged from -2.3% to +5.2 percent.
An initial multipoint performance evaluation was done with propane
responses for each Radian channel. Daily calibration checks and in-house
propane QC samples monitored instrument and operator performance. Duplicate
site samples showed good overall sampling and analysis precision.
Data validation was performed on 100% of the 1990 NMOC data base, as
described later in this section.
-------
Calibration and drift determinations showed that the instrumentation
was stable and that the calibration procedures were consistent. Canister
cleanup results showed there was negligible carryover from one sample to the
next. In-house QC samples of propane demonstrated that the analytical systems
were in control.
Precision, accuracy, and completeness results for 1990 are comparable
to results from previous years and indicate that the data quality are good and
meet all of the data quality objectives of the QAPP.2
4.2 CALIBRATION AND INSTRUMENT PERFORMANCE
Initial performance assessments for NMOC were conducted with propane.
Daily calibrations were checked with about 3.0 ppmC propane for the NMOC
measurements.
4.2.1 Performance Assessment
An initial performance assessment was done on each Radian channel,
using propane certified by EPA-QAD. EPA-QAD referenced the certified propane
to an NIST propane CRM No. 1666B. The concentration of the propane used in
the performance assessment ranged from 2.95 to 19.43 ppmC. The "zero" value
was determined using cleaned, dried air from the canister cleanup system
described in Section 3.0. Table 4-1 summarizes the performance assessments
below. The FID responses for propane were linear, having coefficients of
correlation from 0.999521 to 0.999992. Figures 4-1 through 4-4 show plots of
the NMOC performance results for Radian Channels A, B, C, and D, respectively.
The plots show the regression line.
4.2.2 Calibration Zero. Span, and Drift
Radian PDFID channels were tested daily for zero and span. Zero
readings were measured using cleaned, dried air. The zero air was supplied by
the same system that cleans air for the canister cleanup system. Span
readings used a mixture of about 3.0 ppmC propane in dry air. Calibration
factors were calculated from the span and zero readings for each Radian
channel. Initial calibration factors were determined in the morning before
any site samples were analyzed and final calibration factors were determined
in the afternoon on randomly selected days after all the ambient air samples
had been analyzed. Percent calibration factor drifts were determined based on
cah.!97f 4-2
-------
TABLE 4-1. 1990 PERFORMANCE ASSESSMENT SUMMARY, RADIAN CHANNELS
Radian
Channel
A
B
C
D
Linear
Cases Intercept
5 -9.123
5 20.263
5 -157.444
5 260.600
Rearession
Slope
3424.023
3311.255
3185.909
3122.973
Results3
Coefficient of
Correlation
0.999991
0.999992
0.999521
0.999927
aFigures 4-1 through 4-4 plot propane area counts vs. concentration in ppmC.
cah.!97f 4-3
-------
o
CO
6
CO
CO
2
0
O
03
0
Q.
O
O
0
3
D)
Li.
asuodsay
4-4
-------
OQ
0
03
6
(0
I
CO
0
0
O
c.
CO
0
Q.
O
o
CM
£
g>
ii.
(spuesnoMi)
B9JV) euedoJd JQi esuodsay
4-5
-------
o
03
6
~ CO
CO
I.
o
a
05
&
O
O
CO
4
2>
g>
L.
(spuBsnoqjj
osuodssH
4-6
-------
o
<0
CD
CO
6
of
CO
£
CD
O
CO
-------
the initial calibration factor. The data for zeros, calibration factors, and
calibration factor drifts are given in Appendix F for each Radian channel and
each calendar day of the analysis season. Figures 4-5 through 4-8 show plots
for daily calibration zeros for Radian Channels A, B, C, and D. Figures 4-9
through 4-12 show the daily calibration span data as a function of the 1990
Julian date. Inspection of the percent drift figures shows that the maximum
percent drift was 5.16. The average absolute % drift ranged from 0.643 for
Channel B to 1.937 for Channel A.
4.2.3 Calibration Drift
Summary calibration factor drift data are given in Table 4-2. The
table presents calibration factor drift, percent calibration factor drift, and
absolute percent calibration factor drift. Calibration factors were
calculated from an analysis of a propane-air mixture whose concentration was
known and was referenced by the EPA-QAD to an NIST propane CRM No. 1666B
reference standard as follows:
calibration = concentration of propane standard (ppm) x 3 ppmC/ppm
factor (propane standard response (area counts) - zero response
(area counts))
Daily calibration factors ranged from 0.000288 ppmC/area count to
0.000313 ppmC/area count, depending on the channel. Maxima, minima, and mean
values are given in Table 4-2 for calibration factor drift and percent
calibration factor drift. If drift and percent drift are random variables and
normally distributed, the mean values would be expected to be zero. The means
shown in Table 4-2 for the drift and percent drift are approximately zero,
showing little bias overall, or for any channel. The overall mean values
shown in Table 4-2 were weighted according to the number of calibration drift
data for each channel. The last two columns of Table 4-2 show the means and
standard deviations of the absolute percent calibration factor drifts. The
fact that the standard deviations are the same order of magnitude as the means
indicates that the mean calibration factor drifts are not significantly
different from zero.
cah.!97f 4-8
-------
o
a:
LU
N
^^^^^^^
j^r ^^
O o
1— c
<' 05
/=
DC°
CD £
I T3
"~Tj 03
-------
o
DC
LJJ
N
1
CD
— !
I— C
L^ tf
£s
CDS
1 T3
!^ ^
^^ CT
^v, LJ-
O
w
>-
Q
C
C
/^Sai
7i%
J
TTC
\*'
/x
\L^?
1
)l
X.
\ri
7«
^
•
•
:
t
i
i>
>
^
/
^
5
•
O C\J T- O i
cd
*N^
Q
o i
-IO Tt
•r"~
£
3
O)
LL
O
-O
O'1"-
o o o o o o o
Oiudd '
QOIAIN
4-10
-------
O
o
DC
LLJ
N
1
0
CD
••••• C""
|_ c
*~. 03
-**^^ >w
<£ .c
rr O
LJ_ w
CDS
1 T3
^2 «
^i rr"
^< LL
O
>-
r
—
<
Q
i
1
\j
\
/
-^i
^
\i/I
M/X"2T?
/i^(/ *^*
^
. . x
$
y&
^&7
7^,
:
|
i
I
i
•
[
*
\
|
i
«
9
i
\
O
-LO
CO
O
-0
CO Q
c
03
^00
O o)
-10 O5 0
rvi u-
txj •»— o
- N
0 r-
•4-^ ^—
CO .2
Q 15
c £
8 = "3
"°- ^
'03
Q
r^
0 4
-10 ^
T- O
i—
13
O)
LJ_
o
-o
CO CM T- 0 T- CM CO^
p p p o o o o
CD C) CD CD Q C) C)
Oiudd '
OOI/MN
4-11
-------
o
cc
LU
N
ZQ
T^
Q)
r~ =
t §
^S
CCO
rA c
CD cd
1 73
•^•M /r^
<£
O
Q
l
^
w,
3
2
^
\
2^1
/!>/'
^i
M^i
/iv\
i
7^3
f
T^C;
\4
V1/-I
f
^
i
!
i
t
(
i
i
k
t
i
O
-in
CO
o
-o
CO Q
"o
c
c
CO
o 5
2 O5 0
-in o) ii
CM t— ^
. N
/IN /—
vJJ ^-
^5 a
Q «
*— _Q
^^ «ita4
O CO "cO
-o = o
CM = >,
^ rzz
CO
Q
m
VAJ
o 4
i ^>
"2 0
* L-«
0)
LL.
0
-O
CO
-------
z
MIMM
<
CL
CO
1
^
-------
z
mmmm
<
CL
CO
I
•
••^•p
Zm
O 0
^^^^ ^"*
HC
tem
«•<
<£
nro
l^M
CD S
U-l CD
I T5
^ "5
.
r
^-4
<
\ .
^ '
X '
1^
1
<
7
J'
^
^
^
\VN
71
— 'l '
N^r
K
0
-in
CO
-8 m
CO -0
c
CO
f"
o
o
o °^ S
-in OD S
r\i t— £2-
^ ^ CO
® S
•e o
/TT •— ~
TO +->
Q 2
__ /•>
c .-==
2 cc "co
-o -^= o
w^ >•
^ ^s
CO
Q
0 2
-in ,
T- Tt
CD
u.
O)
LL
0
-o
QomoLnoinom o^-
in^-^cococMCsJi-T-
ooooooooo
ooooooooo
ooooooooo
• ••••••(•
ooooooooo
junoo
/ Otudd
esuodsey QQIAIN
4-14
-------
o
z
mimmmm
CL
CO
1
Zo
O^_1
/IN
VA^
— - C
1 ^MB
i""" c
^* r—
CCO
r^^ d
^*J rrt
W^
^•^* -5-1
_l ^
^x 03
— ' — o
C\J D
~~3 ^
^^^3
03
Q
-% T
O
^^
13
.^
j i
I_L_
O
^•r—
^1M
ooooooooo
ooooooooo
ppppppooo
OCDCJCDOCDCDOCD
lunoo eejv / Qiudd
ssuodsey OOIAIN
4-15
-------
o
lf^
z
Ai^M
<
Q_
CO
1
Q
O 0
•^^M f™
f- |
<£
2-6
me
03
^ ^0
i ^
^ °*
g^r /Y*
^< CL
o
s_
^
— '
<
^'
\5 '
J
~/fS£
1
\i
%
i
~^a
w*
/ra
i
/
"
t,
' 1 J
*
•
k
3 4/
J
X i
A C
\1 s
J r
1 5
s
•*J
CO
-o Q
00 1
CO
f"
^s O
o
O O) C
-m OD co
*•*-_&
>1% .^^
0 c:
cS .2
NS^ •^••J
Q co
c -^
§ .5 «
-vj uz: o
CM 13 "
-J ^
CO
Q
•
o £!
-m 7
^ ^t
0
LM
O)
%x*
ir
o
-0
QOLOOLDOLOOlO O'1"
m^^-COCOCMCMT-T-
ooooooooo
ooooooooo
ooooooooo
• •*••••••
ooooooooo
iunoo B8jy / Qiudd 'JOIOBJ esuodsey OOIAIN
4-16
-------
OO
f
13
00
LU
ce
h-
Ll_
3
a
a:
o
to
e
a>
o .,
^ "
Cu '*"
01 =
**"* 5_
^ 0
1-2
*- S-
S- j_
<2s
o
ro
u_
S- 0
ro •>-
-a -M
c re
re -i-
•4^ ^
OO CD
a
c
re
0)
X
ro
s:
c
rO
a>
z:
c
-r*
2C
•*->
^4« to
•— O
a ^
X
O -M
0 3
13 O
U. (_)
C ro
O 0)
•r- S.
ro ^x.
XJ'E
•i- Q
i — a
ro
U
X
re
z:
c
ro
0)
3tT
c
•^
z:
VI
CD
CO
ro
C CD
(TJ C
••- C
^3 ro
re -£=
ce o
to ^* to
en o co
• I •
CO i— « O
1 — CO CSJ
CO «^ tO
en vo to
^H 0 O
i— < r^ oo
to csj CO
i— i *r csj
to o csj
co P^ en
to to i— i
o o o
en to p*.
CO CO csj
CO O •— *
CO ««• Csj
i— t i i
i
o o o
to •—* r*»
"*
0 O O
^^ CM t— 1
i i t
o o o
• • •
!-. CM IO
CO <- « i
i i
CO CO f
<£ CO 0
CO
CO
CO
•
,«^
O
1—4
^
0
to
en
CO
CM
CO
o
i
i— i
o
en
to
* i
o
00*
o
r™H
1
O
•
CO
1
^
to
o
o
CO
CO
•
T— 1
IO
OO
en
o
to
I— t
to
to
CSJ
O
en
00
co
co
I-H
1
O
to
•-1
CO
1— 4
1
O
f^.
CO
1
^te
CSJ
CSJ
,_
rO
i.
0)
0
4-17
-------
Calibration factor drift was defined as final calibration factor for
the day, minus initial calibration factor. Percent calibration factor drift
was defined as the calibration factor drift divided by the initial calibration
factor, expressed as a percentage. The absolute percent calibration factor
drift is a measure of the calibration drift variability and averaged 0.9854%
overall. The mean absolute percent calibration drift ranged from 0.643% for
Radian Channel B to 1.937% for Radian Channel A.
4.3 IN-HOUSE QC SAMPLES
In-house quality control samples were prepared by Radian personnel.
Local ambient sample results are presented and discussed in Section 4.4.4.
In-house quality control samples were prepared by diluting dry propane with
cleaned, dried air using calibrated flowmeters. The propane used for the in-
house quality control samples was certified by the EPA-QAD against an NIST
Reference Standard. The concentration of the in-house standard ranged from
about 0.159 ppmC to 1.464 ppmC, but was set to average near the concentration
levels that were being analyzed. The analyst did not know the concentration
of the in-house standard prior to analysis.
The daily in-house QC data for each Radian channel are given in
Appendix G, and include:
• Calendar date analyzed;
• Julian date for 1990;
• Radian ID Number;
• Calculated NMOC concentration in ppmC;
• Measured NMOC concentration in ppmC;
• Bias (measured NMOC-calculated NMOC); and
• % Bias (Bias * 100 / calculated NMOC).
Measured versus calculated NMOC concentrations in Figures 4-13 through
4-16 show excellent agreement. Table 4-3 summarizes the results of the linear
regressions for the Radian in-house quality control data, showing regre sion
intercepts near zero, and slopes and coefficients of correlation all nea 1.0.
cah.!97f 4-18
-------
CO
CO
LU
DC
O
O
or
o_
LJJ
CO
O
O
q
c\i
o o o
CO .CO
o
CM
o
O
o
00
*
o
o
CO
o
•
o
I
o
(Otudd) uoiiBJiueouoo OOIAIN
o
q
c\i
o
i-CO
Q.
o 2
UCM. ^
T- Q)
o
0 S
o o
T-: O
O
0
Ueo
O T3
-CO
LJ.
4-19
-------
CO
CO
LLJ
or
o
o
LLJ CD
Q_
DC
Q_
LL1
CO
O
O
O
c\i
o o
00 CO
o
CM
o
O
T
o
00
o
CO
(Oiudd)
o o
OOI/MN
o
^f
•
o
o
CM
•
o
o
o
c\i
o
00
Q.
o 2
Ucxi *
T- 0
O
0 S
o o
T-: O
O
o
00
dz
O "O
.CO 0
O .CO
D
,0
CO
o o
o
.CM
•
o
o
.0
OC)
o
CD
03
6
CO
CO
o
o
CD
CO
o
I
4
CD
O)
4-20
-------
CO
CO
LJJ
o:
o
O
LUO
Q_ S
oc
CL
LJJ
CO
O
o
O
oooo
COCQ^CVJ
ooo
OCOCO
oo
o
q
CM
o
Uco
o
CO
Q.
3.
o
.8
o
Uoo
o
Uco
CD
o
d
o
d
o
°.
od
o
8
§
O
o
o
15
o
(Oiudd) UOI1B4U90UOO OOIAIN pejnsB8|/\|
O
"o
c
6
co"
CO
CD
8
3.
o-
"D
CD
-*^ J_
CO £
^ 10
4
o
O)
LL
4-21
-------
CO
CO
LLJ
DC
O
O
LJJQ
DC
0.
LLJ
CO
O
X
O
O
c\i
o
oo
0
CD
. c
T- O
o 25
CN| •£
T- 0
o
o c
o o
0
O T3
Ucq 0
o _cc
o o
^ -a
o o
U?,
c
.0
OQQQQQOOOQ QO
Q
"0
CO
O
CO
CO
o
o
o
CO
C7
CD
CO
O
I
C
CO
4
CD
O)
b_
(Ouidd) uoiiBJiueouoQ QOWN pejnsi39|/M
4-22
-------
TABLE 4-3. LINEAR REGRESSION PARAMETERS FOR IN-HOUSE
QUALITY CONTROL DATA
Radian Coefficient of
Channel Cases Intercept Slope Correlation
A 5 0.0086 1.060236 0.997253
B 5 0.0252 1.030042 0.999370
C 5 -0.0149 1.048497 0.995747
D 5 -0.0178 1.053398 0.995685
cah.!97f 4-23
-------
Tables 4-4 and 4-5 give statistics for in-house quality control
measurements. DIFF is the ppmC difference between the measured and the
calculated NMOC concentrations, and PCDIFF is the percentage of the difference
relative to the calculated value. Both DIFF and PCDIFF may oe considered to
be bias terms, assuming that the calculated value is the correct NMOC
concentration for the in-house QC sample. Overall, PCDIFF shows a mean bias
of -1.43%, and ranges from -11.47% for Channel C to +8.45% for Channel A.
ADIFF and APCDIFF, absolute values of DIFF and PCDIFF, respectively, were used
as measures of precision. The absolute percent difference ranged from 8.08%
for Channel B to 11.47% for Channel C and averaged 9.71 percent. These
figures show excellent agreement and consistency for the in-house quality
control data and include variability not only in the instrumental analysis but
also in the apparatus and method used to generate the QC samples.
4.4 REPEATED ANALYSES
Two types of repeated analyses were conducted in this study. The
first type of repeated analysis was conducted primarily to establish
precision, and to determine if significant differences in precision existed
among Radian (PDFID) channels and the EPA-QAD (PDFID) channel. Two samples
were selected weekly from the received site samples for a second analysis on a
Radian channel on the following workday. The second replicate analysis was
performed on the day after the first analysis to allow time for the ambient
air sample in the canister to equilibrate between analyses. At the beginning
of the first analysis, the pressure in the canister is typically about 15
psig. At the beginning of the second analysis, the canister pressure is
typically 9 psig.
The EPA-QAD channel randomly repeated analyses of the site samples
already analyzed once by Radian. Shortly after the beginning of the 1988 NMOC
Monitoring Program, the decision was made by Radian and the EPA to do repeated
analyses only on duplicate samples, and to have the second analysis, not only
on the day after the first analysis, but also on the same Radian channel as
the first analysis. The purpose of the latter specification on replicate
cah.!97f 4-24
-------
TABLE 4-4. IN-HOUSE QUALITY CONTROL STATISTICS, BY RADIAN CHANNEL
Variables
Channel
A
B
C
D -
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
DIFFa
5
0.012000
0.136000
0.053880
0.051574
0.023065
0.834878
-0.720667
5
0.009000
0.074900
0.047780
0.024624
0.011012
-0.696930
-0.606814
5
-0.574000
-0.011000
-0.147800
0.242566
0.108479
-1.377939
0.062944
5
-0.705000
0.001000
-0.149600
0.310689
0.138944
-1.494924
0.243176
PCDIFF"
5
1.119134
16.877637
8.452601
6.339540
2.835129
0.115945
-1.320966
5
2.549575
17.721519
8.083983
6.033084
2.698077
0.826474
-0.664283
5
-41.897810
-1.136364
-11.473710
17.171850
7.679485
-1.430692
0.153121
5
-46.968688
0.177620
-10.771777
20.382069
9.115138
-1.448239
0.174921
ADIFFC
5
0.012000
0.136000
0.053880
0.051574
0.023065
0.834878
-0.720667
5
0.009000
0.074900
0.047780
0.024624
0.011012
-0.696930
-0.606814
5
0.011000
0.574000
0.147800
0.242566
0.108479
1.377939
0.062944
5
0.001000
0.705000
0.150000
0.310448
1.495343
1.495343
0.243736
APCDIFFd
5
1.119134
16.877637
8.452601
6.339540
2.835129
0.115945
-1.320966
. 5
2.549575
17.721519
8.083983
6.033084
2.698077
0.826474
-0.664283
5
1.136364
41.897810
11.473710
17.171850
7.679485
1.430692
0.153121
5
0.104384
46.968688
10.842825
20.334924
9.094055
1.450727
0.178461
aDIFF = Measured NMOC concentration - Calculated NMOC concentration, ppmC.
bPCDIFF = Absolute value of DIFF.
CADIFF = DIFF/calculated NMOC concentration x 100.
"APCDIFF = Absolute value of PCDIFF.
cah.!97f
4-25
-------
TABLE 4-5. OVERALL IN-HOUSE QUALITY CONTROL STATISTICS
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Standard Error
Skewness
Kurtosis
DIFFa
20
-0.705000
0.136000
-0.048935
0.209472
0.046839
-2.425102
4.577324
PCDIFFb
20
-46.968688
17.721519
-1.427226
16.269424
3.637954
-1.784011
2.768662
ADIFFC APCDIFFd
20
0.001000
0.705000
0.099865
0.189476
0.042368
2.546347
4.963889
20
0.104384
46.968688
9.713280
12.943140
2.894174
2.013069
3.018044
aDIFF = Measured NMOC concentration - Calculated NMOC concentration, ppmC.
"PCDIFF = Absolute value of DIFF.
CADIFF = DIFF/calculated NMOC concentration x 100.
dAPCDIFF = Absolute value of PCDIFF.
cah.!97f
4-26
-------
analyses was to avoid any bias that may be caused by a different analysis
channel. None of the site samples selected for repeated analyses by Radian
channels was analyzed a third time by the EPA-QAD channel.
All replicate analyses were performed on duplicate samples, but not
all the analyses on duplicate samples were replicated. Each analysis
consisted of two or three consecutive injections from a canister that was
connected to the GC. After the first analysis, the canister valve was closed,
the canister was disconnected from the GC, and the canister was stored at
laboratory temperature overnight. The second replicate analysis on the sample
in the canister was performed on the next day, or the following Monday if the
first analysis was on Friday. Replicate analyses were performed on the same
analytical channel, i.e., Radian Channel A, B, C, or 0, for a given duplicate
sample. By conducting repeated analyses of the duplicate samples it was
possible to investigate the relative magnitude of the duplicate sampling
precision and the analytical precision. The results for this investigation
are given in Section 4.5.
The second type of comparative analysis was done on local ambient
samples collected by EPA-QAD personnel in Raleigh, in Research Triangle Park,
or near Research Triangle Park, North Carolina. These samples were taken once
weekly in duplicate at an initial pressure of about 35 psig. Each local
ambient sample., called a round-robin sample, was analyzed by all four Radian
channels and the EPA-QAD channel. The purposes of these studies were:
• to compare the precision of all the channels; and
• to compare the results among Radian channels.
4.4.1 Site Sample Results
Figure 4-17 compares the EPA-QAD analyses with Radian analyses of the
same site samples. Orthogonal regression parameters for the three data sets
are summarized in Table 4-6.
Summary statistics of the comparative analyses for Radian channels
versus the EPA-QAD channel are given in Table 4-7. The table gives DIFF, the
difference between the Radian NMOC concentration and the QAD NMOC
cah.!97f 4-27
-------
£
8
co
0
Q-
Og
o
o>
o
O
O
O
O
O
O
OO
OO
O
O
O
O
o
o
CO
o
o
80
0 <-
°. o
o -4=
T- 03
-«—»
g o
.q o
00 §
O
•§8
CD 2
CO
o
CO
^
03
03
O
O
03
••s
03
oc
Q- £
+-»
>
Q
<
o
O)
03
Q.
O
O
c
g
"co
CO
O)
£
"03
o
O)
o
O
4
uo!JBJiU90UOO OOI/MN
4-28
D)
Li.
-------
TABLE 4-6. ORTHOGONAL REGRESSION PARAMETERS FOR
REPEATED ANALYSES OF SITE SAMPLES
Channel
Pair
(X-Y)
Cases
Intercept
Slope
Coefficient of
Correlation
QAD-Radian
Radian-QAD
203
203
-0.01883
0.01943
0.969127
1.031856
0.99763
0.99763
cah.!97f
4-29
-------
TABLE 4-7. SUMMARY STATISTICS OF COMPARITIVE ANALYSES
FOR RADIAN vs. QAD CHANNELS
Variables
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
DIFF
203
-0.658000
0.245000
-0.041298'
0.084773
0.005950
-2.785000
19.670167
DIFF = Radian NMOC concentration
ADIFF = Absolute value of DIFF.
PDIFF = DIFF/((Radian
APDIFF = Absolute value
NMOC cone. +
of PDIFF.
ADIFF PDIFF
203
0.000000
0.658000
0.058322
0.074042
0.005197
4.567531
29.306331
APDIFF
203 203
-92.411467
47.584973
-11.649193
16.276369
1.142377
-1.447560
5.137470
0.000000
92,411467
13.645930
14.634290
1.027126
2.219136
6.636771
- QAD NMOC cocentration, ppmC:
QAD NMOC cone. )/ 2)
x 100.
cah.!97f
4-30
-------
concentration in ppmC; and PDIFF, the percent difference relative to the mean
of the Radian and QAD analyses. ADIFF and APDIFF are the absolute values of
DIFF and PDIFF, respectively. The mean percent difference shows Radian NMOC
concentrations to average 11.65% lower than the QAD NMOC concentration. This
is an average bias figure for the Radian analyses relative to a mean NMOC
concentration. The average absolute percent difference is 13.64, which is a
measure of the precision.
In 1985, the mean percent difference showed Radian NMOC concentrations
to average 0.49% higher than QAD, and 3.77 lower in 1986. In 1987, the mean
percent differences showed Radian concentrations to average 4.48% lower than
the QAD NMOC concentration. In 1988, the percent difference was shown to be
1.674 percent. In 1989, the percent difference was shown to be 11.11 percent.
The average absolute percent difference was 10.5% in 1985, 14.8% in 1986,
14.07% in 1987, 11.76% in 1988, and 13.92% in 1989. The agreement among the
precision results is good, and shows that the instruments and operating
procedures were consistent for those years.
Summary statistics are given for the same data in Table 4-8 by Radian
channel. The data show a mean absolute percent difference ranging from
8.05% for Channel A to 17.00% for Channel B. The mean percent differences
range from -15.70% for Channel C to -4.49% for Channel A.
Of NMOC concentration measurements, the comparison between Radian and
the EPA-QAD channel represents between-laboratory comparisons for the PDFID
method.
4.4.2 Local Ambient Samples
Table 4-9 presents the overall statistics for local ambient samples.
These data include comparisons among Radian channels and EPA channels. The
mean differences and the mean percent differences are both relatively small,
which indicates that they are random variables. The overall mean absolute
percent difference (APDIFF) is 10.79, which is slightly higher than the
precision for repeated analyses (7.59 percent).
Table 4-10 presents the same information comparing each Radian channel
to the QAD results and to other Radian channels. Note from the definition of
percent difference, PCDIFF, in this table that the Radian-QAD comparisons are
cah.!97f 4-31
-------
TABLE 4-8. SUMMARY STATISTICS OF COMPARITIVE ANALYSES
FOR RADIAN vs. QAD CHANNELS, BY RADIAN CHANNELS
Channel
A
B
C
D
DIFF
ADIFF
PDIFF
APDIFF
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
DIFF
40
-0.094000
0.245000
0.000493
0.057657
0.009116
1.883789
6.310887
39
-0.548000
0.038000
-0.045762
0.095242
0.015251
-3.999361
18.166997
59
-0.658000
0.227000
-0.061363
0.104481
0.013601
-2.920211
16.994923
65
-0.287000
0.160000
-0.046123
0.062285
0.007726
-0.457410
5.272974
= Radian NMOC concentration - QAD
= Absolute value of
= DIFF/ ((Radian NMOC
= Absolute value of
DIFF.
Variables
ADIFF
40
0.000000
0.245000
0.038608
0.042377
0.006700
3.026966
12.127198
39
0.004000
0.548000
0.052890
0.091374
0.014632
4.362992
20.505008
59
0.002000
0.658000
0.074990
0.095006
0.012369
4.134173
22.256604
65
0.001000
0.287000
0.058585
0.050540
0.006269
2.321424
6.759744
concentration,
cone. + QAD NMOC conc.)/2)
PDIFF.
PCDIFF APCDIFF
40
-38.532110
12.522361
-4.492402
10.406435
1.645402
-1.025228
1.312598
39
-92.411467
4.679245
-10.603389
17.656426
2.827291
-3.033304
10.415898
59
-84.782609
15.802297
-15.697656
17.529110
2.282096
-1.259254
2.611505
65
-66.277940
47.584973
-13.006094
16.029352
1.988196
-0.198840
3.791189
ppmC.
x 100.
40
0.000000
38.532110
8.054578
7.902742
1.249533
1.816715
3.887183
39
0,472255
92.411467
11.458382
17.099425
2.738099
3.250218
11.519870
59
0.395517
84.782609
16.996967
16.249825
2.115547
1.608231
3.585038
65
0.085800
66.277940
15.357580
13.755043
1.706103
1.591507
2.585024
cah.!97f
4-32
-------
TABLE 4-9. OVERALL STATISTICS FOR LOCAL AMBIENT SAMPLES
Statistics
Cases
Minimum
Maximum
Mean
Standard
•
Dev.
Std. Error
Skewness
Kurtosis
DIFF =
ADIFF =
PDIFF -
APDIFF =
DIFF
285
-0
0
0
0
0
-0
3
.128500
.146000
.004720
.036019
.002134
.233562
.958082
NMOC concentration on Channel
Channel X, ppmC.
Absolute value of DIFF.
DIFF/((NMOC concentration on
Channel X)/2) x 100.
Absolute value of PDIFF.
ADIFF
285
0
0
0
0
0
2
4
Variables
PDIFF
.000000
.146000
.023733
.027468
.001627
.163849
.970242
285
-53
63
3
15
0
0
2
APDIFF
.953500
.478260
.098817
.598875
.923998
.003183
.680728
Y - NMOC concentration on
Channel Y + NMOC concentration
285
0
63
10
11
0
1
3
on
.000000
.478260
.792590
.664957
.690972
.834720
.244192
cah.!97f
4-33
-------
TABLE 4-10. STATISTICS FOR LOCAL AMBIENT SAMPLES, BY CHANNEL PAIR
Channel
Pair
(X-Y)
A-QAD
B-A
B-QAD
C-A
Variables
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
DIFF
28
-0.009000
0.101000
0.020946
0.026247
0.004960
1.336522
1.557195
30
-0.068500
0.135000
0.009117
0.037609
0.006866
1.544978
4.174428
, 28
-0.054000
0.146000
0.029536
0.044426
0.008396
1.285584
1.434881
29
-0.128500
0.041000
-0.010397
0.035428
0.006579
-1.864971
3.620013
ADIFF
28
0.000000
0.101000
0.022804
0.024589
0.004647
1.572985
2.177765
30
0.001000
0.135000
0.021483
0.031984
0.005839
2.341493
4.916535
28
0.001000
0.146000
0.034679
0.040388
0.007633
1.718092
1.742346
29
0.002000
0.128500
0.021431
0.029857
0.005544
2.388067
5.154058
PCDIFF
28
-33.962300
43.983400
8.568138
15.214099
2.875194
0.169040
1.554233
30
-18.357500
46.481880
5.607768
13.489326
2.462803
1.580053
2.810512
28
-24.000000
63.478260
14.177094
19.134648
3.616109
0.843743
0.582839
29
-29.894700
30.943400
-1.766509
11.991603
2.226785
0.344951
0.982810
APCDIFF
28
0.000000
43.983400
11.482086
13.072088
2.470392
1.202195
0.145295
30
0.214592
46.481880
8.805819
11.587796
2.115632
2.166382
3.967172
28
0.547945
63.478260
16.620156
16.973759
3.207739
1.303612
0.784650
29
0.209644
30.943400
8.771303
8.206534
1.523915
1.218734
1.030121
(Continued)
cah.!97f
4-34
-------
TABLE 4-10. CONTINUED
Channel
Pair
(X-Y)
C-B
C-D
C-QAD
D-A
Variables
Statistics
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
Cases
Minimum
Maximum
Mean
Standard Dev.
Std. Error
Skewness
Kurtosis
DIFF
29
-0.118000
0.039000
-0.020103
0.040070
0.007441
-1.309112
0.947050
29
-0.069000
0.025000
-0.015486
0.022088
0.004102
-0.565194
-0.019772
27
-0.104000
0.062000
0.010370
0.031466
0.006056
-1.948687
5.027928
29
-0.059500
0.037000
0.005090
0.021945
0.004075
-1.126078
1.388990
ADIFF
29
0.001000
0.118000
0.027828
0.034959
0.006492
1.649387
1.531250
29
0.000000
0.069000
0.019762
0.018214
0.003382
1.090211
0.482245
27
0.002000
0.104000
0.024222
0.022192
0.004271
1.877617
4.344986
29
0.000000
0.059500
0.016841
0.014651
0.002721
1.014888
0.605716
PCDIFF
29
-53.953500
29.657790
-7.585464
18.012517
-0.943821
-0.943821
1.112597
29
-25.000000
20.080320
-4.467421
8.652038
1.606643
0.663244
1.705929
27
-48.275900
43.356640
6.708410
16.645594
3.203446
-0.883131
3.154209
29
-20.095700
19.548870
2.703663
8.589304
1.594994
-0.745355
0.704756
APCDIFF
29
0.470588
53.953490
12.445828
14.953731
1.574649
1.574649
1.300876
29
0.000000
25.000000
7.792632
5.712668
1.060816
1.199562
1.605557
27
0.479616
48.275860
12.749633
12.450063
2.396016
1.464378
1.520432
29
0.000000
20.095690
7.131970
5.354010
0.994215
0.877557
0.198807
(Continued)
cah.!97f
4-35
-------
TABLE 4-10. CONTINUED
Channel
Pair
(X-Y)
D-B
D-QAD
DIFF
ADIFF
PDIFF
APDIFF
Variables
Statistics DIFF
Cases 29
Minimum -0.110000
Maximum 0.032000
Mean -0.004617
Standard Dev. 0.034177
Std. Error 0.006346
Skewness -1.970547
Kurtosis 2.926870
Cases 27
Minimum -0.043900
Maximum 0.066000
Mean 0.025856
Standard Dev. 0.022241
Std. Error 0.004280
Skewness -0.917508
Kurtosis 2.049126
= NMOC concentration on Channel Y -
Channel X, ppmC.
= Absolute value of DIFF.
ADIFF
29
0.000000
0.110000
0.019452
0.028254
0.005247
2.075796
3.249982
27
0.002000
0.066000
0.029700
0.016519
0.003179
0.321019
-0.313723
PCDIFF
29
-40.609100
9.722222
-3.163752
14.370359
2.668509
-1.793970
1.863676
27
-24.000000
30.534350
11.520043
12.254973
2.358471
-0.717170
0.932706
APCDIF
29
0.000000
40.609140
8.525078
11.899649
2.209709
1.938092
2.285483
27
2.113606
30.534350
14.269380
8.749483
1.683839
0.450707
-1.206561
NMOC concentration on
= DIFF/((NMOC concentration on Channel X + NMOC
concentration on Channel Y)/2) x
= Absolute value of PDIFF.
100.
cah.!97f
4-36
-------
--i of bias, using QAD as the reference. The
:a PCDIFF in Table 4-10 is the average of the Radian
•reas for a bias term, the QAD NMOC is used to
"3 95% confidence intervals for the local ambient
Tiean values of DIFF (from Table 4-10). Figure 4-18
ole 4-11 graphically.
^ ESULTS
-.icate samples were taken during the 1990 NMOC
:ate samples were taken approximately every two
••jolicate samples, two canisters were connected to
-- of a Metal Bellows® Pump, MB151, as shown in
:rifice was gauged to deliver the appropriate
— -snt air samples simultaneously into two canisters.
: statistics for the duplicate samples, listing sample
- ^.Tioled, Julian date sampled, sample identification
---antrations for up to three injections, the canister
arcent differences between the mean canister
- iDsolute percent differences.
— samples give the results for duplicate samples whose
^- The results of these 15 samples' analyses were used
~r of samples, duplicate samples, and replicate
:2rs are followed by "R" for the replicate analyses.
i.~ at least two injections from the canister to the
:casionally, three injections were performed in
~~col detailed in the QAPP2. Replicate analyses were
"~ the initial analysis of a given canister. The
~-n in ppmC, given in Column 9, is the arithmetic
~:>ns for a given canister. These results enabled the
, or imprecision, from sampling error (or
ted as Duplicate Percent Difference (% Diff) was
:ifference between the average canister
4-37
-------
TABLE 4-11. LOCAL AMBIENT SAMPLES CONFIDENCE INTERVALS
Channel
Pair
(X-Y)
A-QAD
B-QAD
C-QAD
D-QAD
B-A
C-A
C-B
C-D
D-A
D-B
Mean
Difference
(ppmC)
0.
0.
0.
0.
0.
-0.
-0.
-0.
0.
-0.
02095
02954
01037
02586
00912
01040
02010
01549
00509
00462
Standard
Deviation
(ppmC)
0
0
0
0
0
0
0
0
0
0
.02625
.04443
.03147
.02224
.03761
.03543
.04007
.02209
.02195
.03418
Cases
28
28
27
27
30
29
29
29
29
29
T-0.975.n-1
2
2
2
2
2
2
2
2
2
2
.052
.052
.056
.056
.045
.048
.048
.048
.048
.048
95% Ccnf-idence
Intervals
Upper Lower
0
0
0
0
0
0
-0
-0
0
0
.03112
.04676
.02282
.03466
.02316
.00308
.00486
.00709
.01344
.00838
0
0
-0
0
-0
-0
-0
-0
-0
-0
.01077
.01231
.00208
.01706
.00492
.02387
.03534
.02389
.00326
.01761
to.975.n-i = Student's t-statistic for 95% confidence interval,
where n = the number of cases in mean DIFF.
cah.!97f
4-38
-------
Local Ambient Samples
i
i -^
I->.i. i
O 1
1 ."> 1
^ i
v i
i
i
i . 1
A 1
1
9
d
Q
9
o
~ o
— 1
o
2
O
1
9
Q
-0
6
Q
&
0
-9
I 1 1 1 1 1 1 1
Scopjr-o^'cvicn^io
ooooooooo
oododdodddo
Figure 4-18. 95% Confidence intervals for mean NMOC difference
(QUJdd)
4-39
-------
TABLE 4-12. DUPLICATE SAMPLES FOR 1990
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Site
Code
LINY
LINY
LINY
LINY
LINY
LINY
LINY
LINY
HTCT
HTCT
HTCT
HTCT
PLNJ
PLNJ
PLNJ
PLNJ
HTCT
HTCT
HTCT
HTCT
BMTX
BMTX
BMTX
BMTX
BRLA
BRLA
BRLA
BRLA
MNY
MNY
MNY
MNY
BMTX
BMTX
BMTX
BMTX
BRLA
BRLA
BRLA
BRLA
LINY
LINY
LINY
LINY
HTCT
HTCT
HTCT
HTCT
BMTX
BMTX
BMTX
BMTX
BRLA
Julian
Date Date
Sampled Sampled
13-Jun-90
13-Jun-90
13-Jun-90
13-Jun-90
15-Jun-90
15-Jun-90
26-Jun-90
26-Jun-90
26-Jun-90
26-Jun-90
28-Jun-90
28-Jun-90
28-Jun-90
28-Jun-90
29-Jun-90
29-Jun-90
29-Jun-90
29-Jun-90
16-Jul-90
16-Jul-90
26-Jul-90
26-Jul-90
27-Jul-90
27-Jul-90
27-Jul-90
27-Jul-90
27-Jul-90
27-Jul-90
27-Aug-90
27-Aug-90
27-Aug-90
27-Aug-90
28-Aug-90
28-Aug-90
28-Aug-90
28-Aug-90
04-Sep-90
04-Sep-90
04-Sep-90
04-Sep-90
05-Sep-90
05-Sep-90
10-Sep-90
10-Sep-90
ll-Sep-90
ll-Sep-90
ll-Sep-90
ll-Sep-90
12-Sep-90
12-Sep-90
12-Sep-90
12-Sep-90
18-Sep-90
164
164
164
164
177
177
177
177
179
179
179
179
180
180
180
180
208
208
208
208
239
239
239
239
240
240
240
240
247
247
247
247
254
254
254
254
255
255
255
255
261
261
261
261
263
263
263
263
268
268
268
268
271
Sample
ID
Number
1046
1046R
1045
1045R
1132
1132R
1133
1133R
1124
1124R
1125
1125R
1146
1146R
1145
1145R
1281
1281R
1282
1282R
1453
1453R
1452
1452R
1467
1467R
1466
1466R
1495
1495R
1494
1494R
1535
1535R
1534
1534R
1544
1544R
1543
1543R
1572
1572R
1571
1571R
1595
1595R
1594
1594R
1611
1611R
1610
1610R
1632
Inj 1 Inj 2
NMOC NMOC
(ppmC) (ppmC)
0.409
0.341
0.415
0.381
0378
0.378
0.395
0.380
0.198
0.203
0.245
0.238
0.523
0.572
0.506
0.518
0.167
0.152
0.156
0.165
1.429
1320
1394
1.539
0.783
0.812
0.826
0.821
0.473
0.583
0.554
0.484
1.581
0.953
1.578
1.644
0.283
0.265
0332
0.431
0.234
0301
0.150
0352
0.214
0.203
0.194
0.191
1195
2.246
2.242
2.223
1.679
0.411
0.338
0.407
0.383
0385
0379
0.383
0376
0.199
0.197
0.263
0.266
0.509
0.524
0.526
0.535
0.188
0.179
0.191
0.167
1.510
1.126
1377
1.513
0.796
0.816
0.827
0.788
0.468
0.459
0.541
0.523
1.595
0.849
1.688
1.619
0.272
0.277
0351
0.448
0.258
0.294
0.352
0.301
0.202
0.203
0.175
0.205
2.256
2.244
2.000
1.732
1.658
Duplicate
Canister
Mean %
(ppmC) Diff.
0.3748
03965 5.640
03800
0.3835 0.917
0.1993
0.2529 23.721
0.5320
0.5213 -2.041
0.1715
0.1698 -1.026
1.4019
1.4556 3.770
0.8017
0.8155 1.703
0.4956
0.5090 2.669
1.2446
1.6324 26.953
0.2741
0.3906 35.035
0.2717
0.2971 8.859
0.2055
0.1912 -7.183
2.2519
1.9606 -13.917
Abs%
Diff.
5.640
0.917
23.721
2.041
1.026
3.770
1.703
2.669
26.953
35.035
8.859
7.183
13.917
4-40
-------
TABLE 4-12. DUPLICATE SAMPLES FOR 1990
Sample
No.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
-
31
32
33
34
35
36
37
38
Site
Code
BRLA
BRLA
BRLA
PLNJ
PLNJ
PLNJ
PLNJ
MNY
MNY
HTCT
HTCT
NWNJ
NWNJ
PLNJ
PLNJ
NWNJ
NWNJ
BMTX
BMTX
BRLA
BRLA
BMTX
• BMTX
MNY
MNY
LINY
LINY
HTCT
HTCT
NWNJ
NWNJ
PLNJ
PLNJ
BMTX
BMTX
BRLA
BRLA
MNY
MNY
LINY
LINY
NWNJ
NWNJ
PLNJ
PLNJ
BMTX
BMTX
BRLA
BRLA
LINY
LINY
NWNJ
NWNJ
Julian
Date Date
Sampled Sampled
18-Sep-90
18-Sep-90
18-Sep-90
20-Sep-90
20-Sep-90
20-Sep-90
20-Scp-90
ll-Jun-90
ll-Jun-90
13-Jun-90
13-Jun-90
14-Jun-90
14-Jun-90
15-Jun-90
15-Jun-90
28-Jun-90
28-Jun-90
02-Jul-90
02-Jul-90
05-Jul-90
05-Jul-90
09-Jul-90
09-Jul-90
09-JuI-90
09-Jul-90
10-Jul-90
10-Jul-90
ll-Jul-90
ll-Jul-90
12-Jul-90
12-Jul-90
13-Jul-90
13-Jul-90
16-Jul-90
16-JuI-90
17-Jul-90
17-Jul-90
23-Jul-90
23-Jul-90
24-Jul-90
24-Jul-90
26-Jul-90
26-Jul-90
27-Jul-90
27-Jul-90
30-Jul-90
30-Jul-90
31-Jul-90
31-Jul-90
07-Aug-90
07-Aug-90
09-Aug-90
09-Aug-90
271
271
271
271
271
271
271
162
162
164
164
165
165
166
166
179
179
183
183
186
186
190
190
190
190
191
191
192
192
193
193
194
194
197
197
198
198
204
204
205
205
207
207
208
208
211
211
212
212
219
219
221
221
Sample
ID
Number
1632R
1631
1631R
1642
1642R
1641
1641R
1027
1026
1047
1048
1062
1061
1066
1065
1128
1129
1140
1141
1160
1159
1171
1163
1180
1178
1188
1187
1190
1189
1205
1204
1219
1220
1214
1215
1224
1223
1259
1258
1271
1270
1285
1286
1293
1292
1302
1301
1310
1309
1364
1359
1385
1370
Injl
NMOC
(ppmC)
1.764
1.798
1.736
2.363
2.415
2.354
2.364
0.316
0.384
0.290
0.220
1.188
1.034
0.419
0.365
0.269
0.239
1.398
1.510
0.309
0.331
1130
2.090
0.345
0.387
0.336
0.356
0.275
0.256
0.415
0.487
0.206
0.176
0.947
0.952
0.278
0377
0.417
0.378
0.345
0.287
0.170
0.234
0.279
0.268
1.990
Z128
0.740
0.745
0.252
0.342
0.220
0.200
Inj2
NMOC
(ppmC)
1.737
1.682
1.786
2.364
2,404
2.373
2.355
0.338
0.380
0.262
0.251
1.192
1.034
0.378
0.392
0.276
0.248
1.336
1.600
0.317
0.309
1080
2.140
0.356
0.370
0.305
0.353
0.276
0.253
0.454
0.524
0.182
0.175
0.925
1302
0.279
0.277
0.395
0.442
0.331
0.282
0.169
0.204
0.294
0311
2.071
2.204
0.739
0.787
0.233
0.297
0.224
0.203
Canister
Mean
(ppmC)
1.7094
1.7526
2.3865
23613
0.3270
0.3820
0.2760
0.2355
1.1900
1.0340
03985
0.3777
0.2725
0.2435
1.3670
1.5550
03130
03200
Z1050
2.1150
03505
0.3785
03205
03545
0.2755
0.2545
0.4345
0.5055
0.1940
0.1755
0.9360
1.1270
0.2785
03270
0.4060
0.4417
0.3380
0.2845
0.1695
0.2190
0.2865
0.2895
2.0305
2.1660
0.7395
0.7660
0.2425
03195
0.2220
0.2016
Duplicate
%
Diff.
2.493
-1.059
15.515
-15.836
-14.029
-5.397
-11240
12.868
2.212
0.474
7.682
10.074
-7.925
15.106
-10.014
18.517
16.020
8.345
-17.189
25.483
1.042
6.458
3.520
27.402
-9.609
Abs%
Diff.
2.493
1.059
15.515
15.836
14.029
5.397
11.240
12.868
2.212
0.474
7.682
10.074
7.925
15.106
10.014
18.517
16.020
8.345
17.189
25.483
1.042
6.458
3.520
27.402
9.609
4-41
-------
TABLE 4-12. DUPLICATE SAMPLES FOR 1990
Sample
No.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
Site
Code
PLNJ
PLNJ
BRLA
BRLA
BMTX
BMTX
MNY
MNY
HTCT
HTCT
NWNJ
NWNJ
PLNJ
PLNJ
LINY
LINY
LINY
LINY
HTCT
HTCT
NWNJ
NWNJ
PLNJ
PLNJ
MNY
MNY
BRLA
BRLA
BRLA
BRLA
NWNJ
NWNJ
NWNJ
NWNJ
HTCT
HTCT
NWNJ
NWNJ
PLNJ
PLNJ
NWNJ
NWNJ
Julian
Date Date
Sampled Sampled
10-Aug-90
10-Aug-90
14-Aug-90
14-Aug-90
14-Aug-90
14-Aug-90
20-Aug-90
20-Aug-90
22-Aug-90
22-Aug-90
23-Aug-90
23-Aug-90
24-Aug-90
24-Aug-90
21-Aug-90
21-Aug-90
05-Sep-90
05-Sep-90
06-Sep-90
06-Sep-90
07-Sep-90
07-Sep-90
10-Sep-90
10-Sep-90
18-Sep-90
18-Sep-90
25-Sep-90
25-Sep-90
26-Sep-90
26-Sep-90
21-Sep-90
21-Sep-90
26-Sep-90
26-Sep-90
27-Sep-90
27-Sep-90
27-Sep-90
27-Sep-90
27-Sep-90
27-Sep-90
28-Sep-90
28-Sep-90
222
222
226
226
226
226
232
232
234
234
235
235
236
236
233
233
248
248
249
249
250
250
253
253
261
261
268
268
269
269
264
264
269
269
270
270
270
270
270
270
271
271
Sample
ID
Number
1377
1375
1386
1387
1388
1389
1418
1417
1428
1424
1435
1434
1451
1450
1456
1426
1504
1503
1512
1511
1522
1521
1524
1523
1580
1579
1616
1613
1621
1620
1601
1602
1629
1630
1634
1633
1622
1623
1624
1625
1643
1644
Injl
NMOC
(ppmC)
0.448
0.475
0.512
0.504
1.366
1.507
0.250
0.333
0.249
0.172
0.389
0.404
0.153
0.287
0.091
0.074
1.352
1.355
0.434
0.498
0.605
0.520
0.228
0.261
0.420
0.419
0.280
0.430
1385
1.288
0.651
0.681
0.224
0.559
0.400
0.484
0.753
0.807
1.193
0.991
1.940
2.041
Inj2
NMOC
(ppmC)
0.460
0.453
0.501
0.504
1.406
1.549
0.250
0.302
0.179
0.165
0.255
0.419
0.175
0.307
0.084
0.045
1.433
1.387
0.444
0.518
0.532
0.541
0.200
0.243
0.434
0.404
0.238
0.438
1.341
1304
0.672
0.672
0.197
0.533
0.382
0.490
0.6%
0.809
1.231
1.109
2.130
2.050
Canister
Mean
(ppmC)
0.4540
0.4640
0.5065
0.5040
1.3860
1.5280
0.2503
0.3175
0.2140
0.1685
0.3200
0.4114
0.1641
0.2970
0.0875
0.0595
1.3924
1.3712
0.4390
0.5084
0.5685
0.5305
0.2140
0.2520
0.4270
0.4113
0.2589
0.4343
13628
1.2958
0.6615
0.6765
0.2106
0.5460
03910
0.4870
0.7275
0.8079
1.1891
1.0924
2.0350
2.0455
Duplicate
%
Diff.
-
2.179
-0.495
9.746
23.690
-23.791
25.644
57.617
-38.095
-1.530
14.656
-6.915
16.309
-3.746
50.603
-5.040
2.242
88.649
21.868
10.584
-8.472
0.515
Abs%
Diff.
2.179
0.495
9.746
23.690
23.791
25.644
57.617
38.095
1.530
14.656
6.915
16.309
3.746
50.603
5.040
2.242
88.649
21.868
10.584
8.472
0.515
4-42
-------
concentrations (NMOC2 - NMOC.,), dividing by the average of the two canister
concentrations, and multiplying by 100. The choice of which duplicate was
designated No.l and which was designated No. 2 was arbitrary, and thus the
overall average percent difference was near zero and equaled 6.82 percent.
Duplicate percent difference ranged from -37.91 to 88.65 percent. The
duplicate percent difference range and overall average compare favorably with
the percent difference range and average in the 1989 Final Report3 of from
-80.51 to 71.54, averaging 4.24 percent. Absolute percent difference
disregarded the sign of the percent difference previously calculated and
averaged 13.75%, which compares favorably with results from previous years in
the NMOC Monitoring Program.3 The range for duplicate differeces in 1989
was -80.52 to +71.55 percent. Absolute percent difference in 1989 averaged
14.19 percent.
Measured NMOC concentrations for duplicate canister samples provide the
data used to estimate an overall sampling and analytical error. Sampling
error includes differences caused by any residual organic compounds in the
canisters after cleaning and before sampling. Sampling error also involves
differences caused by drawing the ambient air sample into the canister, and
differences between the analysis of the two canisters caused by removing the
sample from the canister when it is introduced into the analytical device.
Analytical error is measured from replicate analyses from the same canister.
In the protocol used in the 1990 NMOC Monitoring Program, replicate analyses
are performed as follows.
Each analysis involved at least two aliquots taken from a canister. The
first analysis was performed on the same day the canister was received from
the site. The second, or replicate analysis, also involved at least two
aliquots taken from the same canister, but the second analysis was performed
at least one day after the first analysis. The time interval between the
first and the second analysis allowed the contents of the canister to re-
establish equilibrium between the sample and the solid surfaces of the
canister. For each analysis, if the standard deviation between the first and
the second injection was greater than 0.030 ppmC, then a third injection was
performed.
cah.!97f 4-43
-------
In the following section, analytical error, or precision, is estimated
from replicate analyses. In subsequent sections, components of the error (or
precision) are examined to determine what fraction of the total precision (or
imprecision) is attributable to sampling, and what fraction results from
analytical imprecision.
4.5.1 Analytical Precision
Analytical precision is estimated from replicate analyses. Table 4-13
summarizes the replicate analysis results for the 1990 NMOC Monitoring
Program. The first 15 samples listed in Table 4-13 show results in which the
analyses of duplicate samples were replicated. As shown in the table, the
overall average precision in terms of average percent difference was -1.36. In
terms of average absolute percent difference, the analytical precision was
7.59. These figures compare with the duplicate precisions discussed in the
previous section of 6.34% difference, and 12.33 absolute percent difference.
The duplicate imprecision statistics are expected to be greater than the
analytical imprecision statistics, because the duplicate precision includes
both sampling and analytical error. The range for the analytical precision
was from -55.17 to 25.11 with a standard deviation for percent difference
equal to 12.79. For the replicate analyses, percent difference was calculated
from the following:
Percent difference = (NMOC2- NMOC,) / ((NMOC2 + NMOCJ / 2), (1)
where:
NMOC-, is the mean concentration in ppmC from the first analysis, and
NMOC2 is the mean concentration in ppmC of the second, or replicate
analysis.
4.5.2 Components of Variance
All of the duplicate samples (see Table 4-12) were subjected to an
analysis of variance (ANOVA) to separate the sampling variability from the
analyttcal variability. Equation 2 gives the model for this ANOVA.
NMOCl]k = M + S, + D1(l) + tm, (2)
cah.!97f 4-44
-------
CJ3
O
CC
Q.
Q-:
O
o
o
o
cn
oo
LU
oo
a.
UJ
oc
CQ
•f.
u
I
c.
u
a:
^ gu
i| I
c323
•
M z U
•— -55 B
rt
•~S B
""
.2
Si1
cS
m T »>
oq Jn c;
od K o
00
ad
(sjoONOcsNO^rON
f;
o
^^ ^^ ON "^
O O ON "T
^^ n
r- Ov
p vo
rn M
»-^ (N
P 2
vi r~ —. t~ ON
r-" wn r^ QO Ol
M p — O (^1
O wn un o (N
2 §
o wS w-i
r- r- ON
op|ool'^(*iONprr)»OQOONNpf*.wijo'-iin.-<«*^:aoaqT<'5!r-i^ocie>oc>G>ooooc>ooooO'"'--'--odcioc>c>do-^d" — ooo
t~ >t
m o
irt *r
<-i O
doooooooddodoodoooc>d^-i-'--c>dooc>odo—id — — ooo
«OG>dOOPOC>POG>C>OodoOOOi-<--'-«OOG>dOC>OGi«O-"^OC>O
i|i|glsl5|2|l!|!||||s|||||l|Sli|||l|2|2
c c c e e c e c c e e e c e s s s e-g-s-s-i-i-i-i-i-i-i y*?yi?!?yi?°fG-G-a.
4-45
-------
CO
CQ
t*
a
*B
e CJ
i 8 a
•1*1
a"~5
ooooooooor^r^^^^^ — ^
-------
where:
NMOCijk = NMOC concentration from sample injection ijk;
M = overall mean concentration in ppmC;
Sj = effect of sample on NMOC,, i = 1, 2,...,a;
DJ(I) = pooled duplicate effect, j = l,..b; and
ek(ij) = residual, taken to be the analytical error.
The duplicate effects are nested within samples, and DJ(I) represents the pooled
duplicate effect of all the duplicates. As shown in Table 4-12, 59 pairs of
duplicate samples were included in the 1990 NMOC Monitoring Program. It was
necessary to break the duplicate samples into two subgroups to process the
results. These results are shown in Tables 4-14 and 4-15. For each case, the
nested duplicate effect is highly significant, as testified by the value of P
in the ANOVA table for Dj(k) being less than 0.010. It is therefore possible to
separate the sampling variance from the analytical variance. In each ANOVA
table , the figure under the column titled Mean Square and in the row for D1(0
is the pooled variance for Dj(l). Equation 3 shows how this variance is
apportioned between the sampling variance a* and the analytical variance ae2.
Pooled variance for DJ(I) = ae2 + a(b-l)crd2, ' (3)
where:
a£2 is shown in the ANOVA table in the row containing «k(lj) and listed
under the Mean Square column. For example for the ANOVA in Table 4-14, and
using Equation 3,
0.039591 = 0.010940 + 15 ad2, (4)
from which <7d2 = 0.001910. (5)
The total variance, or variability, for the PDFID method of measuring NMOC is
o* = °* + °?> (6)
cah.!97f 4-47
-------
TABLE 4-14. ANOVA FOR DUPLICATE SAMPLES 1 THROUGH 15
Source
Sum of
Squares
df Mean Square F Ratio
D,
id)
-K(ij)
0.687146
0.593862
1.061139
14 4.908183
15 0.039591
97 0.010940
0.448663
3.618921
0.999201E-15
0.000056
cah.!97f
4-48
-------
TABLE 4-15. ANOVA FOR DUPLICATE SAMPLES 16 THROUGH 59
Source
Sum of
Squares
df Mean Square F Ratio
0.513211
0.391607
0.165758
43 1.193515
44 0.008900
95 0.001745
0.684027E+03 0.999201E-15
5.100857 0.159790E-10
cah.!97f
4-49
-------
Thus for the data included in Table 4-14, the total variance, sampling plus
analytical error, may be estimated
a2 = 0.010940 + 0.001910, or (7)
a2 = 0.012850. (8)
The percent of the total variability due to analysis is:
a€2/a2.100 = (0.010940 / 0.01285)100 (9)
Analytical variability= 85.14%. (10)
Similarly sampling variability is calculated as follows:
Sampling variability = (0.001910 / 0.012850)100 (11)
Sampling variability = 14.86%. (12)
Standard deviations (or precisions) for analysis, s£, sampling, sd, and total
error, s, may be calculated from the positive square roots of the
corresponding variances:
se = 0.1046 ppmC, a pooled standard deviation for analysis;
sd = 0.0437 ppmC, a pooled standard deviation for sampling; and
s = 0.1134 ppmC, a pooled standard deviation for the total NMOC
measurement process.
These statistics, derived for the duplicate pairs analyzed in Table 4-14, are
summarized in Table 4-16, which also shows the mean NMOC concentration for
each subset, and estimates of the percent coefficients of variation.
As seen from Table 4-16, the analytical component of the ~iability
ranges from about 85% to 91%, leaving the sampling variability to nge from
15% to about 9 percent.
cah.!97f 4-50
-------
00
t-^
oo
Q-
.
2
«t
V
u
c
1
2
0
s
a
8
tn
in
>,
S
c
§
d
^~
O)
o
o
d
m
^
^-
OJ
m
in
f-*.
OJ
oi
OJ
OJ
l*^
S
f^.
in
d
00437
8
o
d
CO
r^
s
8
d
f*>
Y—
o
8
d
LO
<•
pv.
§
d
-------
4.6 CANISTER PRESSURE RESULTS
Canister pressure results for the NMOC Monitoring Program are
important to be sure that the ambient air samples obtained are representative.
The NMOC sampling systems are designed to obtain an integrated ambient air
sample between 6:00 a.m. and 9:00 a.m., or at other programmed intervals.
Canister pressures are being measured to obtain a better understanding of the
range and magnitude of pressures being generated by the NMOC sampling systems.
Canister pressure data are given in Tables 4-17 and 4-18 for both single
canister samples and duplicate samples. The pressures reported in Tables 4-17
and 4-18 are the canister sampling pressures measured immediately before
analysis in the laboratory. A significant decrease between the field sampling
pressure and the laboratory value might indicate a leak. The canister was
leak tested when this occurred.
Table 4-17 gives statistics for single and duplicate samples. All
sample canisters averaged 15.3 psig, while duplicate samples averaged
16.4 psig. The column entitled "All Samples" includes pressures from both
single samples and duplicate samples. Standard deviations were 3.4 and
3.0 psig, respectively.
4.7 CANISTER CLEANUP RESULTS
Prior to the start of the 1990 NMOC Sampling and Analysis Program all
of the canisters were cleaned and analyzed for their NMOC content to establish
canister initial conditions. The resulting analysis with cleaned, dried air
that had been humidified averaged 3.2 area counts (0.0010 ppmC), ranging
fromzero to 40.63 area counts (0.012 ppmC). Any canisters that produced more
than 0.025 ppmC were recleaned.
Continual monitoring of the cleanup was important to ensure that there
was negligible carryover from one site sample to the next. The daily canister
cleanup procedure is described in detail in Section 3.4. The NMOC content was
below 0.020 ppmC and cleanup was considered to be satisfactory.
Average percent recoveries, or average percent cleanup, in 1990
averaged 99.747% (99.742% in 1989, 99.689% in 1988, 99.374% in 1987, 99.c-l%
in 1986, and 99.898% in 1985), ranging from 92.12% to 100 percent. The
reported average percent recovery is based on average nMOC concentration and
average cleanup concentration. The reported percent cleanup figures should be
considered minimum values. The actual percent cleanup was greater than the
cah.!97f 4-52
-------
TABLE 4-17. NMOC PRESSURE STATISTICS
Statistics
All
Samples
Duplicate
Sample
Canisters
Number of Cases
Minimum Pressure, psig
Maximum Pressure, psig
Mean Pressure, psig
Median Pressure, psig
Standard Deviation, psig
Skewness
Kurtosis
536
6.0
36.0
15.3
14.5
3.4
1.47
5.67
114
10.0
30.0
16.4
17.0
3.0
1.43
6.14
cah.!97f
4-53
-------
TABLE 4-18. PRESSURE DISTRIBUTION OF NMOC AMBIENT AIR SAMPLES
Pressure
Range, psig
Blank3
6.0 to 6.9
7.0 to 7.0
8.0 to 8.9
9.0 to 9.9
10.0 to 10.9
11.0 to 11.9
12.0 to 12.9
13.0 to 13.9
14.0 to 14.9
15.0 to 15.9
16.0 to 16.9
17.0 to 17.9
18.0 to 18.9
19.0 to 19.9
20.0 to 20.9
21.0 to 21.9
22.0 to 22.9
23.0 to 23.9
24.0 to 24.9
25.0 to 25.9
26.0 to*26.9
27.0 to 27.9
28.0 to 28.9
29.0 to 29.9
30.0 to 36.0
Total
Single
Sample
Cases
8
1
2
2
3
4
52
39
61
89
29
26
64
63
14
7
1
2
0
0
1
1
1
2
0
5
477
Dupl icate
Sample
Canister Casesb
0
0
0
0
0
4
2
0
10
16
12
12
24
20
10
0
0
0
2
0
0
0
0
0
0
2
114
aBlank indicates no pressure reading given for sample.
"Equals 57 duplicate samples.
cah.!97f 4-54
-------
reported values because, after the percent cleanup was measured, the canister
was evacuated a third time before being shipped to the site.
4.8 EXTERNAL AUDIT RESULTS
Primary measures of accuracy were calculated from the results of the
analysis of audit samples provided by EPA-QAD. Results are reported in terms
of percent bias, relative to the EPA standards.
Audit samples of propane provided by EPA-QAD were referenced to NIST
propane CRM No. 1668B. Each Radian channel analyzed each audit sample. The
results of these analyses are given in Table 4-19. Audit sample bias,
percent bias, and absolute percent bias are shown in Table 4-20. In
Table 4-20, all bias measurements are relative to the QAD results. Overall
Radian average bias was 1.07%, indicating Radian channels averaged 1.07%
higher than the EPA-QAD reference values. Radian mean bias ranged from -3.18%
for Channel D to 6.23% for Channel A. The overall average absolute percent
bias for the Radian channels was 5.04 percent. These accuracy measurements
show excellent agreement with the reference values, and lend absolute percent
bias for the Radian channels was 5.04 percent. These accuracy measurements
show excellent agreement with the reference values, and lend confidence to the
1990 NMOC concentration results determined on all the Radian channels.
Figures 4-19, 4-20, 4-21, and 4-22 show the audit bias results for the
Radian channels versus the reference values provided by EPA-QAD.
4.9 DATA VALIDATION
The secondary backup disks were updated daily on 20 megabyte hard
disks. At the completion of the sampling and analysis phase 100% of the data
base was checked to verify its validity. Items checked included original data
sheets, checks of all the calculations, and data transfers. In making the
calculations for the final report and other reports, corrections were made to
the data base as errors or omissions were encountered.
A total of 780 NMOC concentration measurements were performed by
Radian in June through September 1990. For the regular 1990 NMOC Monitoring
Program, there were 613 NMOC concentration measurements which included 593
sample analyses, 37 repeated analyses, 26 local ambient samples (x 4 analyses
each with the exception of two samples for which only two analyses each were
performed), and 4 audit samples (x 4 analyses each). The remaining 12
analyses included analyses from the 1990 Portland Monitoring Program.
cah.!97f 4-55
-------
TABLE 4-19. 1990 NMOC AUDIT SAMPLE RESULTS
Channel
Analyzed Radian A B C D QAD
Julian ID NMOC NMOC NMOC NMOC NMOC
Date Date Number ppmC ppmC ppmC ppmC ppmC
6/20/90 171 1053 2.330 2.249 2.271 2.174 2.280
6/20/90 171 1054 1.003 0.974 0.962 0.975 0.936
6/20/90 171 1073 0.715 0.678 0.678 0.669 0.647
6/20/90 171 1074 1.930 1.673 1.822 1.550 1.837
cah.!97f 4-56
-------
Q
5S
o
n
3OOOOOOOOO OO
DCOCQ'srcXJOCXJ^CQCOO
• *•••••••••
r-OOOOOOOOOr-
or
<
CL
LU
CO
>
O
0 S-
LO ^Q.
*~ Q
O
o
0
•
CD
c
CO
6
c
CO
••a
CO
DC
co"
CO
2
4
O)
Li.
(QlJUdd)
4-57
-------
Q
i
DOOOOOOOOO 00
DcocQ^cvjocNj^cqcqq
p^odddooood-r^
^- ^
o
E
Q .2
o
Q
<
O
<
CL
LLI
•
CO
>
en
c
c
CO
6
CO
co
DC
co"
S5
la
o
CN
4
2
O)
LL
(Qiudd)
4-58
-------
o
r^i
Q
^£
O
p ^N ^"^
\j j *
^* to
^^ >
^r o
LJLJ —
0
^^^B ^~
5S
Z)°
•"i c
^ .CO
03
cc
^
1
L
vk
i
i
7K
1
— -^i^^-
CO
Q
0 r*
.in V
C\J <
a.
UJ
•
o >
-q o
CVJ —
Q
f*
o §
E -^
o Q. 0
-LO Q. C
• ^-~ ' 03
< «
§ 5
-8 3
T— ^
<
•
-S- *f
0 xf
<3>
LIB*
=3
.O)
O
o
0000
• • • •
o o o o
OOOO
o o o o o o o
I I I I I
4-59
-------
Q
<
8?
55
(D
Qg
cc
• ^HH
•a
CO
cc
\i/
7K
I
1
I
\y
/T\
^
o
-0
CO
o
-in
c\i
o
-O
c\i
o
-in
o
-°.
o
-in
CD
o
a
DOOOOOOOOO OO
DCOCQ^CNJOCM^fCDCOO
m m m m •••• • • •
r-OOOOOOOOOT-
a
a.
LJJ
•
CO
Q
6s
E
o a.
in QL
^ Q
a
03
JZ
O
03
'•a
03
DC
CO
03
(N
CM
4
£
D)
u_
(Oiudd)
4-60
-------
oo
UJ
C£
a
O-
^T
o
HI
OO
HI
>
HI
a
SSURANCE
J3 O 03
CJ ex
U *-
c o .2
^ *- rn
J3 4J W
U ex
c § 3
OS i_ m
f-» *} ^LJ
U cx
< e
C y .2
j= u CQ
CJ (X
§U
H E
J a.
U3 Cu
Q
U
f"* ^^
k^ P
N—J r*Yi
PJ a,
Q
en
H E
J a,
W p.
Q
<
H E
J a,
UJ a.
Q
—
.2 § g
•a o .2
3 u «
a cx
CN P~ 1-1 O
O\ so O co
T^ sO O C^
so 1-1 T^ \o
Tt Tf CO WO
1
Tf 00 rj- WO
Tf f^ r— 1 \O
ON r~- ON i— i
co r^ t^ oq
d r4 Tt d
wo co Tt o
so 00 CO so
wo wo ON CM
CO O t*» ON
i— i Tf Tf 00
i i
00 tS wo O
ON 1-1 O so
ss 1
^Pi 0
CO CO jfl
0 so ,5;
ON co
O Tl-
co O
SO CO
8 S
co O
00 WO
O »— (
o d
wo O
00 CO
O fS
o o
d d
0 C^
wo so
1-1 co
O O
d d
i
0 *•«
wo t~-
ON r*-
o o
d d
O >
2?Q
%%
CJ _,
d '^^ **^
« 5 § ^
'•6 o u .2
5 03 CJ CQ
* < ^
CN r~ 1-1 o
11 so co co
ON so O co
T^ ^^^ ^— ^ ^^i
so 1-1 T^ so
Tt -rf rO WO
•Tj- 00 T}- WO
t~ C^ co wo
co r~^ c~- oo
d r4 Tt d
wo co TJ- c
so 00 co so
ON ON 1-1 P-
wo wo ON (N
co O P~ ON
^H -^ Tf 00
oo r^ wo c
ON 1-1 ^D so
(N 00 O CN
ON wo 1-1 so
i-( 1-1 WO O
8
^m^
*
Q
<;
OU
• £
^ o
"3
p
II UJ
CJ ;**«
E T
0. "
^ ca
^ 3
H c
.. _J CJ
•^* pj J^
— Q u
U U fx,
C
c
os
U
1
CN P~ O WO
1-1
-------
One hundred percent of the data base was validated according to the
procedure outlined below.
A. Calibration factors were checked.
1. The area count from the strip chart that was used to
determine the calibration factor was examined to verify
that the data had been properly transferred to the
calibration form.
2. The calibration form was examined to verify that the
calculations had been correctly made.
3. Each datum on the disk was compared to the corresponding
datum on the calibration sheet for.accuracy.
B. Analysis data were checked.
1. Area counts were verified from the appropriate strip
chart.
2. Calculations were reverified on the analysis forms.
3. Each datum on the disk was compared to the corresponding
item on the analysis form.
C. Field data sheet was checked.
1. Each datum on the disk was compared to the corresponding
datum on the field data sheet.
The error rate was calculated in terms of the number of items transferred from
the original data sources. For each NMOC value in the 1990 data set, 36 items
were transferred from original sources to the magnetic disks. In the data
validation study each item on the disk was compared with the corresponding
value on the original source of data. Two hundred and four errors were found
(and corrected) for an expected error percentage of 0.956.
Each time the data file was opened and a suspected error found, the
error was checked against the original archived documents, and corrected where
appropriate.
4.10 NMOC MONITORING PROGRAM RECORDS
The quality assurance records developed by Radian for this projec- are
extensive and will be preserved as archives. One of the most important
objectives of the study was to develop a data base that is well planned and
documented and contains NMOC data of known and verifiable quality. Achieving
cah.!97f 4-62
-------
that objective has involved keeping and preserving a number of records that
trace the project from planning through reporting.
4.10.1 Archives
In order to keep detailed records that document the quality of the
measurements made, Radian developed the following original material:
• Quality Assurance Project Plan (QAPP);
• Notebooks;
• Field Data Sheets;
• Laboratory Calibration Sheets;
• Laboratory Analysis Sheets;
• Chromatographic Strip Charts;
EPA-QAD NMOC Results;
• Bi-weekly, Monthly Reports to EPA;
• Memoranda and Correspondence; and
• Final Report.
In addition to the above items, several papers to be presented at
technical meetings and symposia and published in technical journals will be
added to the archives.
The QAPP2 was the Quality Assurance Project Plan and the workplan.
The QAPP was designed according to the EPA Quality Assurance Guidelines, and
set the pattern of steps necessary to document and control the quality of the
data obtained throughout the study.
Several notebooks were necessary to maintain day-to-day records of the
project. Field and laboratory data sheets were designed in advance, so that
the data recorded appeared in a logical sequence and filled in blanks on the
sheet. Additional space was provided for other comments. Each NMOC analysis
was assigned a unique Radian Identification Number. Field data sheets and
shipping records accompanied the canisters in transit.
4.10.2 Magnetic Disks
In order to manage the data base for report generation and data
analysis, pertinent data from the various data sheets and notebooks were
transferred to 20 megabyte magnetic disks. The following software was used in
the construction of the data base: Paradox 3®, Lotus 1-2-3®, and PC File+®.
cah.!97f 4-63
-------
Statistical calculations were performed using SYSTAT® software. The data
access is rapid and in a convenient form. The primary 20 megabyte magnetic
disk has three backup disks.
cah.!97f 4-64
-------
5.0 NMOC DATA ANALYSIS AND CHARACTERIZATION
The purpose of this section is to characterize the NMOC data
qualitatively as well as quantitatively. The NMOC data are shown to fit a
two-parameter lognormal distribution better than a normal Gaussian
distribution. The summary NMOC data for the sites of the 1990 study are given
in Appendix E.
5.1 OVERALL CHARACTERIZATION
Figure 5-1 gives a stem-and-leaf plot of the 1990 Morning Site NMOC
data along with statistics for NMOC. The stem-and-leaf plots show the actual
NMOC concentrations truncated to two or three decimal points. The digits to
the left of the vertical open space are called stems and the digits to the
right of the open space are the leaves. The data are sorted from the smallest
at the top of the graph to the largest at the bottom of the graph. The
minimum NMOC value measured was 0.012 ppmC and is shown as "0 1" on the
first row at the top of the plot. The maximum NMOC concentration measured was
14.254, shown as "142 50" in the bottom row of the chart. The plot shows
536 leaves, one for each NMOC Site datum in the 1990 program. The H's in the
open vertical space locate the stem and leaf for the upper and lower hinges,
and the M locates the stem and leaf for the median. The median separates the
sorted NMOC concentrations into two equal halves; the hinges (or quartiles)
separate each half into quarters. The "H spread" or inter-quartile range is
the difference between the NMOC values of the two hinges.
Statistics shown for NMOC are number of cases, minimum, maximum, mean,
median, standard deviation, standard error, skewness, kurtosis, and the two
hinges. Each NMOC determination is the average of two or three injections of
the site samples. Where duplicates were collected, the NMOC determination is
the average of the two canister content concentrations. In the case of
replicates, each NMOC determination is the average of the original and
repeated analysis concentrations.
The standard error is the standard deviation divided by the square
root of the number of cases. Positive skewness is a third moment about the
mean value, and characterizes a tail to the right of the mean value. A normal
Gaussian distribution has a skewness of zero. The skewness of 6.840 for the
-------
0 1
0 55566 66788 89
1 00000 00111 11112 22223 33333 34444 444
1 55555 55555 56666 66666 67777 77777 78888 88888 89999 99999 9999
2 H 00000 01111 11122 22222 23333 33344 44444 44
2 55555 55666 67777 77778 88888 88888 89999 9999
3 000000011111111111112222333333334444
3 55555 66666 66677 77788 88888 88888 9999
4 M 00000 01111 11111 22223 33334 4444
4 55556 66667 77777 78888 88999 99
5 00000 01111 2244
5 55555 66667 7889
6 - 01111 1222333344
6 55556 66666 77789 9
7 001222333444
7 55555 66778 9
8 001233333334
8 H 5566778999
9 00012 2334
9 56788 S99
10 0334
10 678888
11 12444
11 55666778899
12 01233
12 6
13 1222224
13 678999
14 1112234
14 558
15 122223444
15
16 1
16 9
17 011333
17 78
18 1
18 46
19 338
20 34447 9
21 00689
22 9
23 337
25 8
27 7
28 9
30 9
31 2
32 16
33 2
34 236
38 3
40 04
41 2
42 8
62 0
142 5
NMOC, ppmC
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
534
0.012
14.255
0.728
0.941
0.041
6.830
82.131
0.245
0.439
0.883
o
in
o
Figure 5-1. Stem-and-leaf plot of the 1990 NMOC data.
5-2
-------
1990 NMOC data suggests a lognormal frequency distribution; that is supported
by the fact that for the logarithm of the NMOC value (In(NMOC)) (see
Figure 5-2), skewness equals 0.16, which is close to zero. Kurtosis is the
fourth moment about the mean and relates to the pointedness of the
distribution. A distribution more pointed than a normal distribution, having
the same standard deviation, has a kurtosis greater than 3.0.
Figure 5-2 is a stem-and-leaf plot of the 1990 In(NMOC) data. The
plot shows an approximately symmetrical distribution (skewness = 0.16). The
kurtosis equal to 0.204 indicates the In(NMOC) distribution to be less pointed
than a normal distribution.
The shape of the stem-and-leaf plots suggests a lognormal
distribution. Figures 5-3 and 5-4 support the lognormal distribution
hypothesis for NMOC. The vertical scales in Figures 5-3 and 5-4 are arranged
so that if the cumulative frequency of occurrence of NMOC were normally
distributed, the numbers would plot into a straight line. The line in
Figure 5-3 has a noticeable concave downward trend, indicating that the data
do not fit a normal distribution well. Figure 5-4 plots the logarithm of NMOC
on the same vertical scale. The fact that the digits on the graph plot into
approximately a straight line supports the hypothesis that the NMOC data are
approximately lognormally distributed. An asterisk on the graph indicates the
location of a single datum. Integers, such as 2, 6, or 9, show the location
of the corresponding number of data points. The number 999 shows the
approximate location of either 27 data points or 99 + 9 data points. The
results, although qualitative, show a dramatic difference between the normal
and lognormal hypotheses, and suggest that the latter more nearly describes
the NMOC data. Figure 5-4 is labeled a "Normal Probability Plot," but since
the independent variable is the logarithm (to the base e) of NMOC, if the
relation between the EXPECTED VALUE and In(NMOC) is linear, a lognormal
distribution is indicated.
5.2 MONTHLY VARIATIONS, 1984 - 1990
Table 5-1 partitions the NMOC data for the summer of 1990 into groups
which correspond to monthly intervals.
cah.!98f 5-3
-------
-2 988
-2 77766
2 444
-2 33222 2222
-2 11111 1000000000000
-1 99999 99888 88888 88888 88888 8
-1 77777 77777 77666 66666 66666 66666 66666 6
-1 H 55555 55555 55554 44444 44444 44444 44444
-1 33333 33333 33333 22222 22222 22222 22222 22222 2
-1 11111 11111 11111 11111 11111 10000000000000000000000
-0 M 99999999999999999999999998888888888888888888888888
-0 77777 77777 77777 77777 77777 76666 66666 66666
-0 55555 55555 55555 44444 44444 44444 44444 44444
-0 33333 33333 33333 22222 22222 22222 22
-0 H 1111111111111111111110000000000000000
0 00000000001111111111111111111
0 22222 22222 33333 33333 33333 3
0 44444 44444 55555 55555
0 66666 77777 77777 7
0 88889
1 00111 1
1 22223 33
1 44
1 8
2 6
In(NMOC)
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
534
-4.443
2.657
-0.751
0.908
0.039
0.161
0.211
-1.409
-0.823
-0.124
Figure 5-2. Stem-and-leaf plot of the In(NMOC) data.
5-4
-------
tdfCOSOl*
OJ
O
O
O
O)
O)
C
CO
o
c
o
3
CT
0)
-X-
3
3
O
in
in
I
O)
CM
CD
O> CM
O)
O
O>
O) O)
OJ O> O) 0)
O» O» O>
C\J
en|BA pepedxg
5-5
-------
aerosoi*
CO
in
eo
co
K
en
eo
OJ
OJ
OJ
05
O3
OJ
O)
OJ
OJ
O>
O)
o>
OJ
O)
O)
0)
OJ
o>
m
CO
04 CM
-CM
CO
ra
o
o
O)
o>
o
'•3
.9
to
I
CD
O)
o
LO
CD
D)
CM
CM
enpsA pepectg
5-6
-------
TABLE 5-1. SUMMARY STATISTICS FOR 1990 NMOC SITES, BY MONTH
Sample Minimum Median
Month NMOC NMOC Standard
1990 ppmC ppmC Mean Maximum Deviation Cases
June 0.069 0.444 0.592 6.200 0.646 'l!8
July 0.089 0.365 0.562 3.099 0.533 143
August 0.012 0.439 0.745 14.254 1.257 149
September 0.057 0.659 1.016 4.283 1.031 126
cah,198f 5-7
-------
For the summer of 1990, the monthly means and medians of the NMOC
sites for June, July, August, and September parallel one another. That is,
the NMOC concentrations mean and median for July 1990 are less than the mean
and median for June 1990. Means and median for August and September show
dramatic increases compared to July 1990. Figure 5-9, in which monthly means
for NMOC emissions are plotted for the years 1985 through 1990, the results
show that September means are higher than August means for each year. This
oservation is particular interesting because the number and location of the
NMOC sites changes from year to year, and the average concentrations shown in
Figure 5-9 are for all of the sites for a particular year. Arithmetic means
are used in Table 5-1 in spite of the observations given in Section 5.1 which
conclude that the frequency distribution of NMOC concentrations in ambient air
are logarithmic normal distributed. Comparison of Tables 2-2 and 2-3
containing site average concentrations for the Gaussian and lognormal
distributions, respectively, emphasize that the lognormal means may be less
than, equal to, or greater than the respective arithmetic means. In all cases
the means are within 10% of one another. Either the arithmetic means, or the
mean of the lognormal distribution may be used as a measure of central
tendency of the data. Table 5-1 also gives monthly minima, medians, and
maxima. These latter three statistics are independent of the probability
distribution from which they derive.
Figures 5-5 through 5-8 give the stem-and-leaf plots of the NMOC data
for June, July, August, and September 1990, respectively. All the plots show
the general shape of lognormal distribution. The data for June, July, August,
and September may be considered typical of the sites tested during the
indicated time period. Monthly mean NMOC emissions are plotted in Figure 5-9
for 1985, 1986, 1987, 1988, 1989 and 1990. No general trends are evident for
the years shown. For all six years, September means are higher than August
means, and for five of the six years, July means are less than June • a.ns. At
present, however, it must be concluded that random behavior is respot ble for
apparent month-to-month changes.
During the seven years of the NMOC Monitoring Program, one sit
participated in th.e program for all seven years. Two sites have been -. the
cah.!98f 5-8
-------
69
00112 35555 66889 999
2 H 00012234555788988
3 0011233456788888
4 M 0001144557778889
5 12456
6 11233345566
7 M 345589
8 13356789
9 0279
10 038
11 1568
12 01
13 9
15 2
17 11
20 7
62 0
NMOC, ppmC
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
118
0.069
6.200
0.592
0.646
0.059
5.820
46.482
0.256
0.444
0.756
8
UJ
o
Figure 5-5. Stem-and-leaf plot of the NMOC data for June, 1990.
5-9
-------
0 3
1 00011 1124555666677777888899
2 H 0011223333444444555677838999
3 M 011111123334556667788899
4 11122233446677789
5 00112668
6 H 069
7 012345
8 39
9 06
10 368
11 66
12 26
13 12
13 4
14 1128
15 12
16 1
17 37
19 33
20 9
21 0
23 3
30 9
NMOC, ppmC
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
143
0.089
3.099
0.565
0.534
0.045
1.982
4.087
0.231
0.365
0.677
cc
o
IN
O
Figure 5-6. Stem-and-leaf plot of the NMOC data for July, 1990.
5-10
-------
0
1
2 H
3
4 M
5
6
7
8 H
9
10
11
12
13
14
15
16
17
18
19
20
23
25
31
142
155666783
00223 33444 45555 66788 99
01122 33456 66777 8888
00001 1123455666889
00011 334568899
0145789
11223456789
026
03334 6789
058
47
245778
3
22679 9
1245
234
038
4
8
344
3
3
2
5
NMOC, ppmC
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
148
0.012
14.255
0.750
1.260
0.104
8.508
87.647
0.239
0.442
0.904
-------
0 01111 11111 11111 11111 1
0 H 22222222222333333333333
0 44444 44445 55555 55
0 M 66666677777777
0 88889 99999 99
1 00111 1
1 H 2333
1 44555
1 67
1 38
2 01111
2 23
2
2 7
2 8
3 22344 48
4 0012
NMOC, ppmC
Cases
Minimum
Maximum
Mean
Standard Deviation
Standard Error
Skewness
Kurtosis
Lower Hinge(H)
Median(M)
Upper Hinge(H)
126
0.057
4.283
1.018
1.034
0.093
1.608
1.757
0.298
0.658
1.325
GC
co
8
Figure 5-8. Stem-and-leaf plot of the NMOC data for September, 1990.
5-12
-------
o
o
o
2
O
O C5 CO Is- CO lO
O CO CO CO CO CO
0505050005
-f- O < X >
O
o
o
05
O
co
I
o
o
o
tr>
CO
cr>
c
o
s_
-»->
c
O)
c
o
fTJ
O)
1=
o
cr»
t
in
(OUidd) OOWM
cah.!98f
5-13
-------
program 6 years; 4 sites for 5 years; 1 site for 4 years; 3 sites for 3 years;
12 sites for 2 years; and 73 sites for only 1 year. In all cases the sites
were urban sites, but it is difficult to draw conclusions from year to year
because of the difference in yearly site participation.
The April and May NMOC monitoring data for 1988 were from only four
Florida sites, MIFL, M2FL, T1FL, and T2FL. The remainder of the points
located on the 1988 trend line included data from 45 NMOC Monitoring Program
sites.
cah.!98f 5-14
-------
6.0 RECOMMENDATIONS, NMOC MONITORING PROGRAM
Based on the experiences and results of past NMOC Monitoring Studies,
certain recommendations can be made with respect to equipment design and
validation procedures.
6.1 SITING CRITERIA
At urban centers experiencing extreme or severe ozone problems,
additional NMOC monitoring sites should be located at strategic points in the
urban center. A representative number of sites per urban center should be no
less than four. At these sites,'meteorological data, ozone data, and NOX data
should also be gathered.
6.2 OPERATING PROCEDURE CHANGES
Current operating procedures call for the use of dry propane standards
and external audit samples, yet all the ambient air samples have water vapor
in them. The effect of humidity on propane calibration (and audit) results is
currently unknown and should be determined. The experimental design
recommended to study this conundrum would cover the present NMOC span of 0 to
9 ppmC, and at least 3 levels of humidity: zero, low (-10%), and medium (-30%)
relative humidity.
6.3 VERTICAL STRATIFICATION STUDY
In 1987, 1988, and 1989 ambient air samples were taken at ground level
(3 to 10 meters) and at the 1197-foot (364.9 meter) level at one site. In
1988, an additional site was located on top of the World Trade Center in New
York, a height of over 1000 ft. It is recommended that further study be
performed at these sampling heights and that at least one more level (at
100 meters or some other appropriate height above ground level) be sampled at
the same location. At the same time, barometric pressure and wind velocity
and direction data should be obtained at each sampling level. These samples
should be analyzed for NMOC content as well as for the air toxics compound
concentrations. It is also recommended that ozone concentrations and NOX
concentrations be monitored at the same locations and altitudes. The
information gained from such a study would be useful in validating various
atmospheric model predictions.
-------
6.4 SEASONAL NMOC STUDIES
Data derived in a study qualifying NMOC and NOX in seasons other than
summer could be useful in understanding the relationship of NMOC to NOX and
meteorological conditions. Currently a year-round study for 24-hour air
toxics ambient air samples is being conducted. No study is currently
progress to determine seasonal NMOC concentration changes.
6.5 DIURNAL STUDIES
It is proposed that a diurnal study be made at an appropriate
monitoring site to measure NMOC concentrations by the PDFID and CB-4
measurement techniques, 24-hours per day, seven days per week, for at least
four weeks. An appropriate site for such a study would be one at which the
NMOC concentration averaged 0.800 ppmC or greater, and one at which
meterological as well as NOX data were available. Sampling plans could
include both continuous NMOC measurement, and collection of integrated samples
at various times through the day.
6.6 CANISTER CLEANUP STUDIES
The present canister cleanup procedure appears to be adequate for the
NMOC program, since the concentrations of interest are at the ppmC level.
However, the 3-Hour Air Toxics and UATMP, the concentration levels are at the
ppbv levels, i.e., 0.01 to 50 ppbv, and the present canister cleanup procedure
may not be sufficient to prevent significant carryover of target compounds
from one sample to the next.
Additional cleanup studies are proposed to determine more specifically
the carryover of organic material after cleaning, and to determine how storage
of cleaned, evacuated canisters affects NMOC concentration of a sample.
Storage effects up to three months under vacuum and under pressure should be
included in the study.
Additional studies are proposed to compare cleanup procedures at room
temperature with cleanup procedures involving heating of the canir' -rs.
Radian has proposed7 initiation of several studies to detet ie
whether the present canister cleanup procedure is adequate to preven
significant carryover of organic compounds from one canister to the r
These studies are needed since equilibration in a canister may take a ek or
longer.
cah.!98f 6-2
-------
The effect of sample pressure on the measured NMOC concentration is
not clear. Ambient air samples are sufficiently humid so that at 15 psig,
liquid water condenses inside the canister. Migration of liquid water to the
canister walls affects the adsorption equilibrium, and at the same time,
provides a medium for further-depletion of the vapor phase organic compounds
because of the solubility of organics in water. Equilibration under these
conditions would take longer, perhaps 30 days or more, and the effect on the
measured air sample NMOC (and UATMP target compound) concentration has not
been determined. These effects, however, are probably not significant for the
NMOC measurements, but could affect 3-hour air toxics measurements.
6.7 COORDINATED SAMPLING AT NMOC SITES
It is recommended that where possible the following sampling take
place at NMOC sites for the 1991 monitoring programs:
• NMOC samples;
• CB-4 samples;
• Carbonyl samples;
• 3-hour air toxics compounds; and
• UATMP sampling (at least 38 target compounds).
This kind of program would effect some economy in setting up and monitoring
the sampling program, and also provide some opportunity for cross-correlation
of the results. It is recommended that meteorological data, temperature,
barometric pressure, NOX concentration, and possibly radiation intensity are
continuously monitored at the sites.
Coordinated sampling would be most meaningful at sites where NMOC,
CB-4, and/or UATMP monitoring occurred the previous year (or years).
6.8 FIELD AUDIT
It is recommended that a field audit be designed and conducted at
several NMOC sites during the 1991 Monitoring Program. It is suggested that
one field audit per month be performed at an NMOC site during June, July,
August, and September 1991. The field audit should use at least one standard
of known NMOC concentration and should collect duplicate samples plus a zero-
air blank for each site. The audit samples should use both dry and humid
standards.
cah.!98f 6-3
-------
6.9 DUPLICATE SAMPLE AND REPLICATE ANALYSIS
During the 1991 NMOC Monitoring Program records should be kept of
(1) the NMOC concentration in a duplicate canister before cleanup, and (2) the
zero-air NMOC concentration at the time of the third pressurization with
clean, humidified zero air. The duplicate samples should be scheduled so that
the same amount of time elapses between sampling and analysis for all
duplicate samples.
cah.!98f 6-4
-------
7.0 THREE-HOUR AIR TOXICS DATA SUMMARY
The 1990 NMOC Program included three-hour air toxics samples at three
NMOC urban sites (See Table 7-1) located in the contiguous United States.
Overall concentration results are reported in parts per billion by volume
(ppbv) in Section 7.1, and site-specific results are
-------
TABLE 7-1. THREE-HOUR AMBIENT AIR SAMPLES AND ANALYSES
Site
Code
BRLA
NWNJ
PLNJ
Total
No.
9
8
_8
25
Duplicate
Pairs
GC/MD Analyses
Replicate
1
1
1
3
Total
11
10
10
31
GC/MS
Analy es
cah.!98f
7-2
-------
TABLE 7-2. COMPOUND IDENTIFICATION WITH GC/MD FOR ALL 3-HOUR SITES
Compounds
1,3-Butadiene
Chloromethane
Chloroethane
Methyl ene chloride
trans -1,2-Dichloroethylene
Chloroprene
Chloroform
1,1, 1 -Tri chl oroethane
Carbon tetrachloride
Benzene
Trichloroethylene
1,2-Dichloropropane
Bromodi chl oromethane
Toluene
n-Octane
1,1, 2 -Tri chl oroethane
Tetrachl oroethyl ene
Chlorobenzene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
m-Dichlorobenzene
p-Dichlorobenzene
o-Dichlorobenzene
Propylene
trans-l,3-Dichloropropylene
Cases
%
58
3
6
6
10
68
13
100
100
100
29
48
3
100
19
16
90
13
100
100
100
16
16
6
100
6
No.
18
1
2
2
3
21
4
31
31
31
9
15
1
31
6
5
28
4
31
31
31
5
5
2
31
2
Minimum
ppbv
0.11
0.06
0.08
0.92
0.31
0.03
0.16
0.19
0.05
0.27
0.09
0.07
0.12
0.78
0.004
0.14
0.05
0.01
0.06
0.32
0.16
0.02
0.30
0.22
0.58
0.73
Maximum
ppbv
6.83
0.06
0.31
2.69
0.72
1.78
0.72
12.32
0.30
5.92
0.97
1.51
0.12
13.34
0.46
5.06
3.73
0.08
1.07
6.24
2.72
0.05
1.68
9.27
20.60
2.69
Mean
ppbv
1.98
0.06
0.20
1.80
0.48
0.46
0.39
1.15
0.15
1.65
0.39
0.74
0.12
4.08
0.26
3.41
0.72
0.06
0.40
1.94
1.02
0.03
0.90
4.74
5.64
1.71
cah.!98f
7-3
-------
compound was identified, the minimum, maximum, and mean (arithmetic average)
concentrations of the compound in ppbv. In cases where duplicate samples were
taken, or replicate analyses were performed, the results of all the analyses
were averages for each sample. The mean refers to the daily sample averages,
not the averages of all the analyses. The target compounds identified fall
into at least four categories: (1) those occurring in more than 70% of the
samples tested, (2) those occurring in from 40% to 69% of the samples,
(3) those occurring in less than 30% of the samples, and (4) those not
identified in any of the 3-hour air samples at concentrations above their
method detection limits. These results are summarized in Table 7-3.
Overall concentrations ranged from 0.004 ppbv for n-octane to
20.60 ppbv for propylene.
7.2 SITE RESULTS
Table 7-4 gives 3-hour ambient air concentrations by site code for the
38 target air toxics compounds. The overall site means range from 1.12 ppbv
for PLNJ to 2.34 for BRLA. Appendix H tabulates the complete analytical
results and includes the NMOC concentrations for each of the 3-hour air toxics
samples.
cah.l98f 7-4
-------
TABLE 7-3. FREQUENCY OF OCCURRENCE OF TARGET COMPOUNDS
IN 3-HOUR AMBIENT AIR SAMPLES
Range for
Frequency of
Occurrence
Target Compounds
100% to 70%
69% to 40%
39% to >0%
Zero
1,1,1-Trichloroethane
Benzene
Tetrachloroethylene
m/p-Xylene
Propylene
1,3-Butadiene
1,2-Dichloropropane
Chloromethane
Methylene chloride
Chloroform
Bromodi chloromethane
1,1,2-Tri chloroethane
m-Dichlorobenzene
o-Dichlorobenzene
Acetylene
Bromomethane
Bromochloromethane
cis-l,3-Dichloropropylene
Bromoform
Carbon tetrachloride
Toluene
Ethyl benzene
Styrene/o-Xylene
Chloroprene
Chloroethane
trans-1,2-Dichloroethylene
Trichloroethylene
n-Octane
Chlorobenzene
p-Dichlorobenzene
trans-l,3-Dichloropropylene
Vinyl chloride
1,1-Dichloroethane
1,2-Dichloroethane
Di bromochloromethane
1,1,2,2-Tetrachloroethane
cah.!98f
7-5
-------
TABLE 7-4. COMPOUND IDENTIFICATIONS WITH GC/MD BY SITE CODE
Minimum Maximum
Site Compound Cases ppbv ppbv
BRLA 1,3-Butadiene
trans- 1 , 2-Di chl oroethyl ene
Chloroprene
Chloroform
1 , 1 , 1 -Tri chl oroethane
Carbon tetrachloride
Benzene
1,2-Dichloropropane
Toluene
n-Octane
1,1,2-Tri chl oroethane
Tetrachl oroethyl ene
Chlorobenzene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
m-Dichlorobenzene
p-Dichlorobenzene
Propylene
trans-l,3-Dichloropropylene
NWNJ 1,3-Butadiene
Chl oroethane
Chloroprene
Chloroform
1 , 1 , 1 -Tri chl oroethane
Carbon tetrachloride
Benzene
Trichloroethylene
1,2-Dichloropropane
Bromodichloromethane
Toluene
n-Octane
Tetrachl oroethyl ene
Chlorobenzene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
p-Dichlorobenzene
o-Dichlorobenzene
Propylene
8
3
7
2
11
11
11
6
11
2
3
10
1
11
11
11
3
3
11
2
6
2
5
1
10
10
10
6
5
1
10
2
10
1
10
10
10
2
1
10
0.44
0.31
0.26
0.37
0.29
0.14
1.00
0.91
1.76
0.34
4.23
0.05
0.08
0.25
0.86
0.57
0.03
0.30
2.63
0.73
0.11
0.08
0.03
0.31
0.70
0.13
0.50
0.10
0.10
0.12
2.01
0.33
0.14
0.07
0.19
0.81
0.52
0.50
9.27
0.92
6.83
0.72
1.78
0.72
0.52
0.30
5.90
1.46
8.18
0.46
5.06
1.68
0.08
0.98
4.81
2.61
0.05
1.52
20.60
2.69
0.46
0.31
0.64
0.31
2.16
0.18
2.24
0.97
1.51
0.12
11.49
0.34
3.73
0.07
1.07
6.24
2.72
1.68
9.27
8.10
Mean
ppbv
4.15
0.48
0.87
0.55
0.41
0.19
2.86
1.24
4.38
0.40
4.66
0.34
0.08
0.48
2.22
1.22
0.04
0.77
10.90
1..71
0.21
0.20
0.36
0.31
1.28
0.14
0.88
0.48
0.53
0.12
4.32
0.34
0.85
0.07
0.41
2.10
1.06
1.09
9.27
3.08
(Continued)
cah.!98f
7-6
-------
TABLE 7-4. (Continued)
Minimum Maximum
Site Compound Cases ppbv ppbv
PLNJ 1,3-Butadiene
Chloromethane
Methyl ene chloride
Chloroprene
Chloroform
1 , 1 , 1-Trichloroethane
Carbon tetrachloride
Benzene
Trichloroethylene
1,2-Dichloropropane
Toluene
n-Octane
1,1,2-Trichloroethane
Tetrachl oroethyl ene
Chlorobenzene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
m-Dichlorobenzene
o-Dichlorobenzene
Propylene
4
1
2
9
1
10
10
10
3
4
10
2
2
8
2
10
10
10
2
1
10
0.18
0.06
0.92
0.06
0.16
0.19
0.05
0.27
0.09
0.07
0.78
0.00
0.14
0.16
0.01
0.06
0.32
0.16
0.02
0.22
0.58
0.47
0.06
2.69
0.35
0.16
12.32
0.15
3.00
0.34
0.62
13.34
0.07
2.91
3.13
0.08
1.01
4.84
2.40
0.03
0.22
8.96
Mean
ppbv
0.28
0.06
1.81
0.19
0.16
1.82
0.13
1.09
0.22
0.26
3.49
0.04
1.53
1.04
0.05
0.31
1.48
0.77
0.03
0.22
2.40
cah.!98f
7-7
-------
8.0 THREE-HOUR AIR TOXICS TECHNICAL NOTES
This section describes the equipment used to sample and analyze the
3-hour air toxics samples. Also described are sample handling procedures,
sampler certification procedures, standards generation and instrument
calibration procedures, compound identification procedures, GC/MS compound
identification confirmation, quality assurance/quality control procedures, and
data records for the 3-hour air toxics compounds.
8.1 SAMPLING EQUIPMENT AND INTERFACE
The sampling equipment for the 3-hour air toxics samples was the NMOC
Monitoring Program sampling equipment described in Section 3.1. The original
sample was collected as an integrated ambient air sample from 6:00 a.m. to
9:00 a.m. with a final sample pressure of about 15 psig. As stated above,
after NMOC analysis the canister was bled to atmospheric pressure and allowed
to stand at least 18 hours before being analyzed by GC/MD.
An interface system was designed and built by Radian Corporation to
take a sample from the canister and inject it into the gas chromatograph for
analysis.
Figure 8-1 shows the GC/MD system including the Sample Interface
System, Analytical System, and Data System. The sample interface takes a
250-mL sample approximately from the canister, draws it through Trap
Assembly 1 and condenses all the water and organic compounds, with the
exception of methane, in the air sampl'e drawn from the canister. Trap
Assembly 1 is a cryogenic, liquid argon trap packed with glass beads. The
cryogen is removed, and an electrical heater quickly heats Trap Assembly 1,
vaporizing the water and organic compounds condensed from the canister sample.
8.2 THREE-HOUR AIR TOXICS SAMPLING SYSTEMS CERTIFICATION
The sampling systems used to collect 3-hour air toxics samples were
certified for use per the specifications described in U.S. EPA Compendium of
Methods TO-14.12
-------
Jl
oT
aj
Q
jt
>.
03
"c5
o
«i
c
<
,-fU
"i
3 a
S2
•=l
io
> .
j
1
e
c =
< 3
. :
ll
i «
^iO
3i
il
ilP-
c S o a J S
'•r«« 3
Z^UO>K
1
• !&;
.Mi1
>. « I g j ^ •
!i*-ija fy.li
E
o>
In
>.
03
0)
u
fl
03
03
cah.lSSf
"co
>»
en
o
"o
03
"3
£
g
a.
2
O)
2
"co
o
£
o
CO
CO
O
CO
2
5
g>
u_
-------
8.2.1 Sampler Certification Blanks - Humidified Zero Air
Zero certification consisted of purging the sampler with cleaned,
humidified air, followed by collecting a sample of the cleaned, dried air that
had been humidified through the purged NMOC samplers for GC/MD analysis. The
purpose of the wet purge was to help remove any adherent contaminants from the
sampler. The chromatograms from these certification sample analyses were
archived for each sampler. Results presented in Table 8-1 showed a range of
0.002 ppmC to 0.005 ppmC of NMOC, with an average of 0.004 ppmC. The sampling
systems were determined to be very clean and showed no characteristics of
additive bias.
8.2.2 Sampler Certification Challenge - Selected Target Compound
Following the NMOC sampler blank certification, a challenge gas
containing five selected target compounds was passed through the samplers.
The average concentration of the compounds in the challenge gas was
18.5 ppbv/species. Table 8-2 shows the average system percent bias calculated
with the analysis of the challenge gas being used as a reference
concentration.
System percent bias ranged from 1.5% to 11.9% with an overall average
of 8.2 percent. The systems showed acceptable subtractive bias
characteristics.
8.3 STANDARDS GENERATION
The GC/MD analytical equipment was calibrated daily with a gas mixture
that averaged 5 ppbv of each of the 38 target compounds at 70% relative
humidity.
The standard gas mixtures were generated by dynamic flow dilutions of
Scott certified gas mixtures with cleaned, dried air that had been humidified
with HPLC-grade water. The Scott gas mixtures were dry and contained in
cylinders under pressure at a concentration of 500 ppbv for all of the
standards. The concentration for each target gas in the 500 ppbv Scott
cylinder was certified to within ±10 percent.
cah.!98f . 8-3
-------
TABLE 8-1. SAMPLER CERTIFICATION ZERO RESULTS
Sampler Sampler
Canister Blank Sample Sampler Zero Zero
Canister Blank Concentration System Collection Sample D erence
Number Date (ppmC) Number Date (ppmC) ppmC)
052
087
101
6-05-90
3-16-90
3-16-90
0.003
0.003
0.001
22
3
6
6-06-90
3-19-90
3-19-90
Average
0.005
0.002
0.004
0.004
-0.002
-0.001
0.003
cah.!98f
8-4
-------
TABLE 8-2. SAMPLER CERTIFICATION CHALLENGE RESULTS
Percent
System Average Compound System
Number Recovery Bias
22 101.5 . 1.5
3 111.9 11.9
6 113.3 ' 11.3
Average 108.9 8.2
cah.!98f 8-5
-------
Figure 8-2 diagrams the dynamic flow dilution apparatus. One Scott
cylinder contains 18 of the air toxics target compounds, a second Scott
cylinder contains eight target compounds, a third Scott cylinder contains
eleven target compounds, while the fourth cylinder contains only one target
compound. The four Scott cylinders were connected to Channels 2, 3, 5, i 6
of the flow dilution apparatus. The fourth channel was connected to a
cylinder of zero-grade air by way of a catalytic oxidizer that oxidizer ill of
the hydrocarbon material in the zero-grade air. The five mass flow
controllers were set to flow rates that would give the desired final
concentration of the diluted gas.
The cleaned zero-grade air was partially humidified by bubbling part
of the air stream through HPLC-grade water contained in a stainless steel
canister. The wet and dry rotameters and all the mass flow controllers were
calibrated with a bubble flowmeter before being connected to the flow dilution
apparatus. All of the flow controllers, the connecting lines, and the mixing
flask were heat traced to reduce adsorption of the target compounds. The
temperature controller that regulated electrical current flow to the heat
tracing was set for 100'C.
To generate a standard, the following procedure is used. The canister
into which the standard is being mixed is connected to the flow dilution
apparatus at the bellows valve shown in Figure 8-2. The temperature -
controller for the heat tracing is activated. The mass flow controllers are
then set for the appropriate flow rate to obtain the desired dilution and the
humidifier, lines, and mixing flask are purged for at least 10 minutes. The
isolation valve is closed and the vacuum pump turned on. The tubing, the
canister, and the absolute pressure gauge are all evacuated initially to about
0.5 mm Hg absolute pressure. The vacuum pump is then isolated from the system
and the isolation valve is opened to the diluted gas mixture. The standard
mixture fills the canister at a controlled rate until atmospheric pressure is
reached. The canister with the diluted standard is disconnected from the flow
dilution apparatus and allowed to equilibrate before use. The barometric
pressure and room temperature are also recorded.
In order to calculate the exact concentration of each target compound
in the standard mixture, a correction is made for the residual gas in the
cah.!98f 8-6
-------
u
03 03
5
33
to
O
tr
C
>
T5
5
S
E
2
o
cc
•^•••••B
03
3^
=3
3
_o
u.
1
iZ
2
0
u
"5
c
o
0
03
_>
as
03
>
fO
nj
01
00
_i
03
CN
CO
3
+-*
03
^.
CO
a
a.
co
C
_g
^—«
_3
^
$
_g
y
E
CO
C
>
Q
CN
cab
(U
u.
3
O3
cah.!98f
8-7
-------
standard canister before the filling with the diluted gas is begun. A
correction also is made for the water vapor added to the dilution air.
8.4 CALIBRATION ZERO AND SPAN
Most of the compound quantitation is performed with the calibrated
response of the FID detector. For purposes of compound identification and
quantitation (when there may be interference on the FID detector) it is also
necessary to calibrate the PID and ECD responses. Initial calibration curves
for each compound were generated on all three detectors with calibration
standards at 1, 5, and 10 ppbv. In addition to the usual response (area
counts) versus concentration curves, response times and response ratios for
PID/FID and ECD/FID were determined for each target compound.
8.5 GAS CHROMATOGRAPH/MULTIDETECTOR ANALYSIS AND COMPOUND IDENTIFICATION
A Varian® 3700 gas chromatograph, configured with a PID in series with
an FID and an ECD operating in parallel, performed the air toxics analyses.
Fused silica was used for the detector-to-detector connections. The Air
Toxics Multiple Detector System is shown in Figure 8-3 and diagrams the
effluent splitter and multidetectors connected to the end of the Megabore®
DB-624 capillary column.
The entire GC/MD system is shown in Figure 8-1, including the sample
interface, the gas chromatograph/multidetector analytical system, and the data
handling system. Sample volumes for the GC/MD analyses were about 250-mL.
Compound identification was performed using measured retention times
and ratios of PID/FID and ECD/FID responses. The analyst's skill and
experience was also needed in making a judgment about the presence or absence
of a target compound because of the variability of retention times, and the •
presence of interfering compounds.
8.6 GAS CHROMATOGRAPH/MASS SPECTROMETER ANALYSIS AND COMPOUND
IDENTIFICATION CONFIRMATION
Three of the 3-hour air toxics samples were analyzed by GC/MS for
compound identification confirmation following completion of the GC/MD
analyses. So that the sensitivity of the GC/MS compared favorably with that
of the GC/MD, the GC/MS was operated in the multiple ion detection (MID) mode,
and the sample volume was about 500-mL (compared to 250-mL for the GC/MD
analyses).
cah.!98f 8-8
-------
V
ll
*3
.— *
e « S
25-
** ** 2;
o aS
^ co «
UjUQ
£ B jj 1 1 0
(Q N 3) O |5 Q}
E 'E » £ '£ «
00 "-Jo
1 i
u !<>
2 "8 is I
*!* L
c 5
« C u
O CO CD
<5 .. „„. in -» ==
' C) O £
Z x Eg
E «
u
o
U)
S
E
c a
^•«C9 !
^* IV (
zS c
T
1
: P
) a
• en
C/3
i §
C a
s 2 e 3
a> o c o
ft fc- IT) 2.
D) CO • '*3
S '§• x E
^wro o
g <
: «
• m
: co
3 2
f §5
' LL
<
•1
5
^
cah.!98f
8-9
-------
No comparison of the quantitative results for GC/MS and GC/MD was
made, because the purpose of the GC/MS analyses was compound identification
confirmation only. This comparison is discussed below in Section 8.7.4.
8.7 QA/QC DATA
Quality assurance and quality control in the 3-hour air toxics data
included a determination of method detection limits (MDL) for both the GC/MD
and the GC/MS analytical methods.
One of the objectives of the UATMP was to make the MDLs as low as
possible, recognizing that the lower MDLs may increase the number of false
positive or false negative identifications. Other quality measures reported
here involved analytical precision results from repeated analyses, and
sampling and analysis precision from duplicate samples. Accuracy was assessed
for both the GC/MD and GC/MS using external audits supplied by the EPA-QAD.
8.7.1 GC/MD and GC/MS Minimum Detection Limits
MDLs for the GC/MD and GC/MS analytical systems used in this study are
given in Table 8-3. MDLs for the GC/MD analytical system are estimated from
the minimum area count that reflects approximately three times noise for every
compound and are based on a sample approximately 250-mL in volume. The sample
volume for the GC/MS system was about 500-mL. The GC/MS was operated in the
MID mode, which detected specific ions representative of the 38 air toxics
target compounds.
8.7.2 Repeated Analyses
Repeated analyses were performed on three site samples by GC/MD. The
analyses were performed on consecutive days with at least 24 hours between
removing samples from the canister. From these analyses there were 31 cases
in which a concentration for a target compound was found in both replicate
analyses. Statistics for these data are summarized in Table 8-4, showing the
overall minima, maxima, and means of the mean concentrations, standard
deviations, coefficients of variation, and absolute percent differences for
the replicate pairs. The absolute percent difference averages 26.430%, which
is excellent agreement.
cah.!98f 8-10
-------
TABLE 8-3. METHOD DETECTION LIMITS FOR 3-HOUR AIR TOXICS COMPOUNDS
Compound
Acetylene
Propylene
Chloromethane
Vinyl chloride
1,3-Butadiene
Bromomethane
Chloroethane
Methylene chloride
trans-l,2-Dichloroethylene
1,1-Dichloroethane
Chloroprene
Bromochloromethane
Chloroform
1,1,1-Tri chloroethane
Carbon tetrachloride
1,2-Dichloroethane
Benzene
Trichloroethylene
1,2-Dichloropropane
Bromodi chl oromethane
trans-1 ,3-Dichloropropylene
Toluene
n-Octane
cis-l,3-Dichloropropylene
1,1, 2 -Tri chloroethane
Tetrachl oroethyl ene
Di bromochl oromethane
Chlorobenzene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
Styrene
o-Xylene
Bromoform
1 , 1 , 2 , 2-Tetrachl oroethane
m-Di chl orobenzene
p-Dichlorobenzene
o-Dichlorobenzene
GC/MD
MDL
ppbv
1.00
0.10
0.20
0.20
0.10
0.20
0.10
0.11
0.04
0.04
0.06
0.003
0.006
0.001
0.001
0.04
0.04
0.004
0.04
0.001
0.04
0.02
0.03
0.04
0.02
0.07
0.001
0.02
0.02
0.04
0.02
-
-
0.001
0.002
0.02
0.09
0.02
GC/MS
MDL
ppbv
a
0.40
0.56
0.44
0.57
0.25
0.38
0.31
0.39
0.53
0.57
0.48
0.27
0.70
0.37
0.59
0.34
0.37
0.44
0.22
0.50
0.50
0.29
0.81
0.22
0.32
0.25
0.57
0.72
0.46
-
0.46
0.39
0.27
0.37
0.28
0.56
0.37
aBelow mass spectrometry range.
cah.!98f
8-11
-------
TABLE 8-4. 3-HOUR AIR TOXICS REPLICATE ANALYSES BY GC/MD
Statistics
Minimum
Maximum
Overall
Mea-
Mean Concentration, ppbv
Standard Deviation, ppbv
Percent Coefficient of Variation
Absolute Percent Difference
0.075 9.435
0.000 0.587
0.000 69.305
0.000 100.000
1.512
0.072
6.942
26.430
cah.!98f
8-12
-------
Table 8-5 lists the cases in which a target compound was found in only
one of the replicate analyses. Although the list of single compound
identifications is not long, no pattern to the behavior emerges.
8.7.3 Duplicate Sample Results
Six duplicate 3-hour ambient air samples were analyzed by GC/MD for
the 38 target compounds. Summary precision results are given in Table 8-6 in
terms of mean concentration and concentration range in ppbv. Other precision
statistics are given in terms of standard deviation, percent coefficients of
variation, and absolute percent difference. The data in Table 8-6 are
accumulated over all compounds and site locations.
The percent coefficients of variation ranged from 1.418% to 93.274%,
averaging 15.272 percent.
Table 8-7 also shows that the imprecision is also compound specific.
The compound absolute percent difference means ranged from 3.636 to
89.878 percent. The precision for the 3-hour air toxics compounds is good
with an overall average absolute percent difference of 21.6 percent.
Table 8-8 lists the compounds that were identified in only one of the
duplicate sample analyses. Again, no pattern to the behavior emerges.
8.7.4 GC/MS Confirmation Results
Based on three GC/MS analyses of the 3-hour air toxics samples, one
from each site location, the following results were obtaine'd. The GC/MS
analyses confirmed 88.57% of the GC/MD analyses. The results are summarized
in Table 8-9, showing 27.62% positive GC/MD-positive GC/MS confirmation, 1.90%
positive GC/MD-negative GC/MS confirmation, 9.52% negative GC/MD-positive
GC/MS comparisons, and 60.95% negative GC/MD-negative GC/MS comparisons.
8.7.5 External Audits
The external audit for the 3-hour air toxics compounds is conducted
bimonthly on the Urban Air Toxics Program and the results will be reported in
the 1990 UATMP Final Report. The audit samples that are used are furnished by
the Quality Assurance Division of the U.S. EPA.
cah.!98f 8-13
-------
TABLE 8-5. SINGLE COMPOUND IDENTIFICATIONS OF GC/MD
REPLICATE SAMPLE ANALYSES
Compound
Concentration
ppbv
Radia
ID
Chloroprene
Trichloroethylene
1,2-Dichloropropane
1,2-Dichloropropane
1,1,2-Trichloroethane
Chlorobenzene
p-Dichlorobenzene
0.51
0.22
1.51
0.16
5.06
0.08
1.52
13,. 31
1293 R2
1285 Rl
1293 Rl
1310 Rl
1310 R2
1310 R2
cah.!98f
8-14
-------
TABLE 8-6. THREE-HOUR AIR TOXICS DUPLICATE SAMPLE ANALYSES BY GC/MD3
Statistics
Mean Concentration, ppbv
Standard Deviation, ppbv
Percent Coefficient of Variation
Absolute Percent Difference
Overall
Minimun Maximum Mean
0.070 10.080 1.384
0.003 1.633 0.186
1.418 93.274 15.272
2.005 131.910 38.399
aBased on first analysis of contents of each duplicate canister.
cah.!98f
8-15
-------
TABLE 8-7. GC/MD 3-HOUR AIR TOXICS DUPLICATE PRECISION BY COMPOUND
Compound
1,3-Butadiene
Chloroprene
1,1,1-Trichloroethane
Carbon tetrachloride
Benzene
Trichloroethylene
1,2-Dichloropropane
Toluene
Tetrachl oroethyl ene
Ethyl benzene
m/p-Xylene
Styrene/o-Xylene
Propylene
Cases
1
2
3
3
3
1
2
3
3
3
3
3
3
Mean
SD
0.156
0.166
0.124
0.006
0.148
0.038
0.028
0.316
0.816
0.038
0.111
0.045
0.255
Mean
% CV
2.571
29.258
11.114
4.060
12.372
10.490
5.643
11.018
63.553
11.472
8.938
5.814
12.034
Mean
Absolute
% Diff
3.636
41.377
15.717
5.742
17.496
14.835
7.980
15.582
89.878
16.225
12.641
8.223
17.019
-cah.!98f
8-16
-------
TABLE 8-8. SINGLE COMPOUND IDENTIFICATIONS OF GC/MD DUPLICATE SAMPLE ANALYSES
Compound
1,3-Butadiene
Chloroprene
Trichloroethylene
1,2-Dichloropropane
Bromodi chl oromethane
n-Octane
n-Octane
1 , 1 , 2-Tri chl oroethane
m-Dichlorobenzene
Concentration
ppbv
0.18
0.46
0.09
1.51
0.12
0.07
0.46
5.06
0.04
Radian
ID
1292
1286
1292
1285
1286
1292
1309
1310
1309
cah.!98f • 8-17
-------
TABLE 8-9. COMPOUND IDENTIFICATION CONFIRMATION
GC/MD versus GC/MS comparison
Positive GC/MD
Positive GC/MD
Negative GC/MD
Negative GC/MD
Total compound
- Positive
- Negative
- Positive
- Negative
GC/MS
GC/MS
GC/MS
GC/MS
Total
identification confirmation =
Cases
29
2
10
64
105
27.62% + 60.95% =
Percentage
27.62
1.90
9.52
60.95
99 . 99%
88.57%
cah.!98f - 8-18
-------
8.8 DATA RECORDS
Data records for the 3-hour air toxics samples include:
• NMOC concentration of the sample;
• Copies of the gas chromatographic trace for FID, PID, and ECD;
• Response data on Bernoulli disk;
• Retention time for each compound; and
• Area counts for each detector.
In addition, daily calibration response factors are recorded on
magnetic disk along with the retention time and area counts for each compound
in the standard.
cah.!98f 8-19
-------
9.0 RECOMMENDATIONS, THREE-HOUR AIR TOXICS PROGRAM
The following recommendations derive from the 3-hour Air Toxics
Monitoring Program. The studies (Sections 9.1 and 9.2) are directed toward
areas in which additional information is needed to validate further the air
toxics results.
9.1 COMPOUND STABILITY STUDIES
Compound stability in this context refers to whether the apparent
concentration of a compound in a sample taken from a canister is changing over
time. The apparent change in concentration may result from a chemical
reaction of the compound while it is in the canister, or result from a change
in the gas phase concentration caused by adsorption of the compound on the
interior canister surfaces.
A study needed to investigate this phenomenon would take several
canisters--at least three from each initial concentration — ranging in target
compound concentration from zero to 20 ppbv. The canisters would be analyzed
24 hours after mixing, 72 hours after mixing, 30 days after mixing, and
60 days after mixing to determine any concentration changes. It is also
recommended that the same concentrations be mixed in canisters, but that
equilibration times of 7 days and 30 days be assigned before the first samples
are drawn from the canisters to determine the effect of equilibration time on
the concentration samples withdrawn from the canisters.
9.2 CANISTER CLEANUP STUDIES
The present canister cleanup procedure has not been studied in
sufficient detail to determine the amount of carryover for each of the air
toxics compounds. Experience has shown4'10'11 that the present cleanup procedure
is satisfactory so long as a period less than a week elapses between sampling
and analysis.
A study needs to be conducted to determine the effects of:
• Additional pressurization/vacuum cycles on cleanup;
• Heating the canisters during cleanup;
• Vacuum holding time during cleanup; and
• Holding time between cleanup and sampling
on the carryover for each air toxics target compound.
-------
The present canister cleanup procedure is described in Section 3.3.2
and consists of three vacuum/pressurization cycles with cleaned, dried air
that has been humidified. These cycles are followed by a final vacuum step to
5 mm Hg vacuum. Preliminary measurements11 have indicated that after this
cleaning procedure has been completed, there may be sufficient organic
compounds still adsorbed on the canister interior surfaces to be desorbed and
measured in the 0.05 to 0.50 ppbv range, especially for holding times of
7 days, 14 days, and 28 days.
9.3 CARBONYL STUDIES
Recommendations for carbonyl samples are already referred to in
Section 6.7. Sections 6.1, 6.2, 6.3, 6.4, and 6.5 would also apply to
carbonyl sampling.
cah.!98f 9-2
-------
10.0 CARBONYL SAMPLING, ANALYSIS, AND QUALITY ASSURANCE PROCEDURES
Sampling and analysis procedures for the carbonyl samples, along with
the quality assurance procedures used to quantify data quality are described
in this section. The site operator's task involved recognizing problems with
sampling equipment and procedures, and notifying Radian personnel at Research
Triangle Park so that appropriate corrective action might be taken.
10.1 SAMPLING EQUIPMENT AND PROCEDURES
A schematic diagram of a typical carbonyl sampler is shown in
Figure 10-1. The 3-hour carbonyl sample subsystem collects a discrete sample
through the use of a control system that is common with the NMOC system.
Ambient air was drawn through the carbonyl sampler cartridges from a glass
manifold. The ambient air was introduced through a short section of
chromatographic-grade stainless steel tubing into an ozone scrubber column.
The scrubber column was maintained at 200"F to prevent moisture condensation.
Before entering the carbonyl cartridges, the ambient air samples flowed
through an ozone scrubber, or denuder. The carbonyl cartridges are mounted in
parallel, so that the carbonyl samples are collected in duplicate during each
sample collection period. The carbonyl cartridge is a commercially available
(Waters Co.) silica gel Sep-Pak® cartridge which is coated with
2,4-dinitrophenol hydrazine (DNPH). The cartridges are prepared in batches by
the laboratory and stored under refrigeration until shipped to the field.
The carbonyl cartridges are installed in the sample line one day prior
to scheduled sample collection. A 3-hour sample collection period was
utilized for both the canister and cartridge samples. In addition to the
carbonyl cartridges installed in the sample line, a third cartridge is sent
along in the shipment box as a trip blank or spare cartridge.
The flow rate through each sample cartridge was measured before and
after each collection period by the site operator. The flow rate was measured
with a calibrated rotameter and recorded on a preformatted data sheet. The
volume of ambient air sampled through each cartridge was calculated in the
laboratory based on the field-recorded flow rate measurements.
-------
To NMOC
Collection System
Sample Manifold
— 1 -^"^
^ ^~
By-Pasa V J
Pump r-~^
Programmable
Electronic
Timer
Temperature Controlled
Ozone Scrubber
110 VAC
LJneSupply
Chassis/Housing
Vent
Figure 10-1. 3-Hour carbonyl sampling subsystem.
cah.!98f
10-2
-------
10.2 ANALYTICAL PROCEDURES
The analytical procedures for aldehydes are given below. Sample
preparation and analyses are performed at the Radian PPK laboratory. A sample
cartridge is removed from its shipping vial and attached to the end of a 10-mL
polypropylene* syringe. Four milliliters of acetonitrile are added to the
syringe and allowed to drain through the cartridge into the graduated
centrifuge tube. After the drainage has stopped the volume of extract is made
up to 4-mL with acetonitrile and mixed. This solution is transferred to a
4-mL sample vial fitted with a Teflon*-lined self-sealing septum and stored in
a refrigerator until used. TO-11 high pressure liquid chromatography (HPLC)
column and elution solvents used for analysis were-modified to decrease
analysis time.
The separation is done using a 25 cm x 4.6 mm CIS analytical column
with 5-micron particle size. Typically 50-microliter samples are injected
with an automatic sample injector. The following gradient elution is carried
out at 0.9 mL/min:
Time (Min.) % Water % Acetonitrile % Methanol
0 40 20 40
5 25 5 70
15 15 5 80
21 15 5 80
23 40 20 40
33 40 20 40
Detector signals from a multiwavelength detector are collected for 30 minutes
at 360 nanometers (nm). Any residual sample is eluted from the column in
two minutes with a'50:50 mixture of water and acetonitrile.
The relevant chromatographic peaks determined by acceptable retention
times are integrated and the concentrations calculated using standard curves.
Detection limits are determined by a one-sided tolerance interval around the
repeatability of the lowest standard. The analysis optimized sample
throughput for analysis of formaldehyde, acetaldehyde, acrolein, acetone,
methyl ethylketone, butyraldehyde, isobutyraldehyde, propionaldehyde,
benzaldehyde, tolualdehyde, and dimethylbenzaldehyde. Some other species may
cah.!98f 10-3
-------
coelute with these target compounds using this HPLC column and gradient
elution approach.
Aldehyde species reported are formaldehyde and acetaldehyde. Acetone,
a ketone, is also reported. All results were reported in parts per billion by
volume (ppbv). All Radian reported analyses were identified by the unique
tube numbers which were recorded on the preformatted field data sheets.
10.3 QUALITY ASSURANCE PROCEDURES
Quality assurance procedures relative to calibration data for acetone,
formaldehyde, and acetaldehyde are discussed below. Daily quality control
procedures are also discussed. Sampling and analysis precision was determined
from the analysis of duplicate field samples and duplicate laboratory
analyses. Sample custody records were maintained throughout the program.
Figure 10-2 shows the multipage field data and custody sheet used for the
carbonyl cartridges.
10.4 CALIBRATION PROCEDURES
The calibration procedures used for this study followed Radian
standard operating procedures.
10.4.1 Daily Quality Control Procedures
Daily calibration checks were used to assure that the analytical
procedures were in control. About 100 tubes were analyzed for aldehydes.
Daily QC checks were performed each day analyses were conducted.
10.4.2 Duplicate Samples
Duplicate field cartridges were installed in parallel in the sample
probe for each sampling episode, as shown in Figure 10-1. The paired average
of the analyses of these duplicate cartridges was used as the average sample
concentration in this report. If one of the field duplicates was broken or
otherwise declared invalid, the analysis of the remaining cartridge was used
as the reported sample concentration. Likewise if one of the duplicates
showed a concentration above the detection limit and the other duplicate
reported a concentration below the detection limit, the concentration above
the detection limit was reported as the sample concentration.
cah.!98f 10-4
-------
COR I»OK AT
URBAN TOXICS MONITORING PROGRAM
Aldehyde Data Sheet
City
Sample Date
SAROAD No. A05 Sampler No.
Cartridge
Tube No.
Lot No.
Scrubbed
R G
Unscrubbed
R G
Blank
Rotameter No.
Rotameter Readinq1 /
Rotameter Readinq1 /
Samplinq Time/Duration
Sampiinq Volume3
(before)
(after)
(hours)
(liters)*.
Flow Rate2
Before
After
Average
Average Ambient Temperature
Average Barometric Pressure _
Site Operator
Comments/Remarks
(C° or F°)
(mm Hg)
LPM
Q.
O
o
C
O
o
JS
Q.
CO
0)
1 Rotameter reading center of black ball.
2 Calculated from calibration curve by the laboratory.
3 Calculated by laboratory.
Figure 10-2. Field data and custody form.
tr
m
3
n
o
8
cah.!98f
10-5
-------
10.4.3 Trio Blanks
For each pair of aldehyde samples, a trip-blank cartridge was included
in the field site shipment. The blank consisted of a regular DNPH-cartridge
with caps, identical with the sample cartridges. Each cartridge had a unique
serial number for identification purposes. The blank cartridge accompanies
the duplicate sample cartridges on the trip to and from the site, without
removing the caps on each end of the cartridge. The trip-blank cartridge is
not exposed to air at any time during the shipment or sampling periods. The
trip-blank cartridges are analyzed for aldehydes at the same time the sample
cartridges are analyzed. The purpose of the trip blank was to assess the
potential for field trip contamination. The results are not blank corrected.
10.4.4 Precision
Precision was measured as the average standard deviation of the paired
duplicate samples. The relation between precision and mean concentration, and
between absolute percent difference and mean concentration were investigated.
Percent differences for formaldehyde gives a measure of the precision of the
sampling and analysis from the pairs of carbonyl cartridges samples with each
ambient air samples. Sampling and analytical error ranged from -61.4% to
103.0, averaging 0.35% overall, with a standard deviation of the percent
differences equal to 46.5 ppbv.
10.4.5 Practical Quantitation Limit
Approximate values for the Practical Quantitation limits (PQLs) are
given in Table 10-1 for the carbonyls of interest in this study. PQLs for
formaldehyde, acetaldehyde, valeraldehyde, hexaldehyde, and benzaldehyde were
estimated experimentally. PQLs for the other compounds in Table 10-1 were
estimated from the DNPH derivative of formaldehyde.
10.5 RESULTS
Analytical results of ambient air samples for carbonyl compounds at
Baton Rouge, LA (BRLA); Newark, NJ (NWNJ); and Plainfield, NJ (PLNJ) are given
in Tables 10-2 through 10-4. The tables give the sampling date, the Radian
sample ID, sample tube serial number, the formaldehyde and acetaldehyde
concentrations in ppbv, and the percent difference for the pair of carbor ;
cartridges taken with each ambient air sample (Radian ID Number). No carbonyl
cah.!98f 10-6
-------
TABLE 10-1. PRACTICAL CARBONYL QUANTITATION LIMIT3
PQL
Compound (ppbv)
Formaldehyde 0.81
Acetaldehyde 0.71
Acrolein 0.71
Acetone 0.71
Propionaldhyde 0.71
Butyraldehyde 0.67
Isobutyraldehyde 0.67
Methyl ethylketone 0.67
Valeraldehyde 0.60
Isovaleraldehyde 0.36
Hexaldehyde 0.57
Benzaldehyde 0.57
o-Tolualdehyde 0.57
m-Tolualdehyde 0.57
p-Tolualdehyde 0.57
Dimethylbenzaldehyde 0.55
Practical Quantitation Limit (PQL) assumes at least a 140-L gas (air) sample.
cah.!98f 10-7
-------
TABLE 10-2. CARBONYL RESULTS FOR BATON ROUGE, LA (BRLA)
Sampling
Date
08/10/90
08/10/90
08/14/90
08/14/90
08/16/90
08/16/90
08/20/90
08/20/90
08/22/90
08/22/90
08/24/90
08/24/90
08/28/90
08/28/90
08/30/90
08/30/90
09/05/90
09/05/90
09/07/90
09/07/90
09/11/90
09/11/90
Radian
Sample
ID
1384
1384
1386
1386
1406
1406
1416
1416
1430
1430
1437
1437
1466
1466
1478
1478
1500
1500
1519
1519
1533
1533
Tube
Number
26E
47M
Average
51D
53F
Average
133M
152J
Average
111M
127E
Average
18J
39E
Average
44J
52E
Average
19K
38D
Average
37C
50C
Average
. 24C
25D
Average
11C
20L
Average
15G
49B
Average
Concentration
Formaldehyde
(ppbv)
7.9
10.5
9.2
4.7
10.4
7.6
2.2
b
2.2
b
2.2
2.2
12.8
12.0
12.4
4.4
2.5
3.5
4.6
3.3
3.9
2.1
1.8
1.9
9.7
12.1
10.9
13.2
12.8
13.0
3.2
4.4
3.8
Acetaldehyde
(ppbv)
a
a
a
1.48
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
Formaldehyde
Percent
Difference
27.4
74.7
-6.5
-53.4
-31.1
-13.5
22.3
-3.6
31.7
Average = 5.3
Std. Dev. = 38.1
a Below detection limit of 0.71 ppbv.
b Quality assurance cartridge used to determine percent recovery for formaldehyde spike.
10-8
-------
TABLE 10-3. CARBONYL RESULTS FOR NEWARK, NJ (NWNJ)
Sampling
Date
08/06/90
08/06/90
08/08/90
08/08/90
08/10/90
08/10/90
08/14/90
08/14/90
08/16/90
08/16/90
08/20/90
08/20/90
08/22/90
08/22/90
08/24/90
08/24/90
08/28/90
08/28/90
08/30/90
08/30/90
Radian
Sample
ID
1648
1648
1371
1371
1380
1380
1391
1391
1412
1412
1415
1415
1429
1429
1446
1446
1461
1461
1484
1484
Tube
Number
67G
63C
Average
66F
70L
Average
69K
65E
Average
107G
105E
Average
77F
108J
Average
149E
170D
Average
174J
176L
Average
103C
110L
Average
169C
167 A
Average
172F
173A
Average
Concentration
Formaldehyde
(Ppbv)
3.3
3.1
3.2
9.8
9.1
9.5
4.3
3.1
3.7
1.8
5.6
3.7
b
4.5
4.5
b
1.3
1.3
0.9
0.5
0.7
1.2
2.3
1.7
1.3
0.8
1.0
1.3
1.3
1.3
Acetaldehyde
(ppbv)
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
Formaldehyde
Percent
Difference
-3.8
-7.3
-32.4
103.0
-19.0
63.6
-55.1
0.8
Average = 7.5
Std. Dev. = 57.4
a Below detection limit of 0.71 ppbv.
b Quality assurance cartridge used to determine percent recovery of formaldehyde spike.
10-9
-------
TABLE 10-4. CARBONYL RESULTS FOR PLAINFIELD, NJ (PLNJ)
Sampling
Date
08/06/90
08/06/90
08/08/90
08/08/90
08/10/90
08/10/90
08/14/90
08/14/90
08/16/90
08/16/90
08/20/90
08/20/90
08/22/90
08/22/90
08/24/90
08/24/90
08/28/90
08/28/90
08/30/90
08/30/90
Radian
Sample
ID
1350
1350
1361
1361
1377
1377
1397
1397
1403
1403
1442
1442
1444
1444
1450
1450
1460
1460
1491
1491
Tube
Number
22A
23B
Average
83B
83A
Average
90J
89G
Average
13E
14F
Average
113B
118G
Average
130J
131K
Average
171E
125C
Average
122M
112A
Average
79J
76E
Average
117F
120K
Average
Concentration
Formaldehyde
(ppbv)
a
1.4
1.4
a
2.9
2.9
1.7
0.9
1.3
a
1.5
1.5
a
a
a
c
a
a
c
2.1
2.1
0.9
a
0.9
1.7
a
1.7
a
a
a
Acetaldehyde
(ppbv)
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
Formaldehyde
Percent
Difference
-61 .40
Average = -61 .40
a Below detection limit of 0.81 ppbv.
b Below detection limit of 0.71 ppbv.
c Quality assurance cartridge used to determine percent recovery of formaldehyde spike.
10-10
-------
was detected in any of the samples other than formaldehyde and acetaldehyde,
as shown.
The results from the field sites are presented as individual and mean
concentrations for the 3-hour sample period. The concentration variability
and sample population preclude the use of a site mean concentration. The
average percent difference in formaldehyde for BRLA and NWNJ are within the
expected range of results based on historical data. The higher percent
difference (-61%) is due to limited statistical input which result from lower
than expected formaldehyde concentrations.
Measures of analytical precision for formaldehyde are given in
Table 10-5. For the samples shown in Table 10-5, the analytical precision
ranged from -57.7% to 47.0%, averaging -5.4 percent difference. The replicate
analysis performed in the laboratory was performed on 10% of the field sample
population, covering the range of formaldehyde concentrations. The quality
control check of analytical precision indicates that the analysis was in
control during the program.
10.5.1 Formaldehyde Control Standards
As a quality control (QC) procedure on the analytical results for
formaldehyde, a solution of known-concentration was formulated. Throughout
the period of time that analyses were performed, QC samples were analyzed on
the dates indicated in Table 10-6. As shown in the table, the percent of
formaldehyde recovered is shown to range from 85.3 to 114.3, averaging
100.3 percent. These results shown that the analytical technique was
consistently good.
10.5.2 Recoveries of Spikes
Three types of spikes were used as quality control procedures:
(1) Known amounts of formaldehyde were spiked onto DNPH cartridges
and then analyzed. These were called laboratory spikes because
all of the procedures were performed in the laboratory.
Laboratory spike percent recoveries are tabulated in Table 10-7.
(2) Known amounts of formaldehyde were spiked on DNPH cartridges.
The cartridges were taken to the field and used as one of the
pair of cartridges during a regular sample collection. The pair
of cartridges were returned to the laboratory for analysis. The
amount of spike recovered from the spiked cartridges was
calculated as a percent recovery using the formaldehyde found on
the unspiked cartridge to correct the total formaldehyde found
cah.!98f 10-11
-------
TABLE 10-5. CARBONYL LABORATORY REPLICATES
Site
Code
NWNJ
NWNJ
BRLA
BRLA
NWNJ
NWNJ
Lab
Lab
NWNJ
NWNJ
BRLA
BRLA
NWNJ
NWNJ
NWNJ
NWNJ
Radian
ID
1412
1412
1478
1478
1391
1391
Blank
Blank
1461
1461
1406
1406
1429
1429
1648
1648
Tube
Number
108J
50C
102B
139F
169C
133M
176L
67G
Formaldehyde
Concentration
(ppbv)
4.5
2.5
1.8
1.5
1.4 b
2.2 b
2.0 b
1.1 b
1.3
1.6
2.2
2.4
0.5
0.5
3.3
4.4
Percent
Difference
(%) a
-55.9
-16.8
47.0
-57.7
17.3
10.5
-18.2
30.5
Average = -5.4
Acetaldehyde
Concentration
(ppbv)
<0.71
<0.71
<0.71
<0.71
< 1.0 b
< 1.0 b
< 1.0 b
< 1.0 b
<0.71
<0.71
<0.71
<0.71
<0.71
<0.71
<0.71
<0.71
a Percent difference = ((ConcentrationZ - Concentrationl)/((Concentration2 + Concentrationl)/2)* 100.
b Blank ug presented because no volume for blanks.
10-12
-------
TABLE 10-6. ANALYSIS OF QUALITY CONTROL STANDARDS
Analysis
Date
09/28/90
09/28/90
10/15/90
10/15/90
10/16/90
10/16/90
10/16/90
10/27/90
10/27/90
11/12/90
11/27/90
12/03/90
Formaldehyde
Concentration
(ng/uL)
5.19
• 4.93
5.16
4.17
4.99
4.42
5.30
5.16
4.88
4.66
5.61
4.45
Percent
Of
Target a
106.1
100.9
105.6
85.3
102.0
90.4
108.5
105.6
99.8
94.9
114.3
90.6
Average = 100.3
a Known concentration is 4.9 ng/uL.
10-13
-------
TABLE 10-7. FORMALDEHYDE LABORATORY SPIKES
Analysis
Date
09/29/90
10/15/90
10/15/90
10/16/90
10/16/90
10/16/90
10/16/90
10/26/90
10/26/90
10/27/90
10/27/90
10/28/90
10/28/90
11/12/90
11/13/90
11/14/90
11/14/90
11/14/90
11/27/90
Tube
Number
138E
144M
155M
129G
134A
140C
153K
151G
237D
135B
146B
136C
214Q
249R
143L
145A
154L
242K
150F
Formaldehyde
Concentration
(ug)
25.13
17.11
22.37
21.30
25.21
20.09
19.29
23.94
24.10
22.80
19.56
20.77
22.78
23.21
23.07
16.08
17.86
23.36
19.62
Percent
of
Target
(%)
123.2 a
83.9
109.6
104.4
123.6 a
98.5
94.5
111 A
118.1
111.8
95.9
101.8
111.7
113.8
113.1
78.8 a
87.6
114.5
96.2
Average = 105.2
a Outside of desired +/- 20% window.
10-14
-------
on the spiked cartridge. Three spiked sample pairs were termed
field spikes. The results of percent recovery for field spikes
is shown in Table 10-8.
(3) Carbonyl spikes were formulated by the following procedure.
Four silica gel cartridges, previously coated with DNPH, were
spiked in the laboratory with formaldehyde, acetaldehyde, and
acetone at the following levels: Formaldehyde 20.5 ^g;
Acetaldehyde 20.5 /^g; and Acetone 15.1 ^9- The spiked
cartridges were then processed through the regular extraction -
analytical procedure as the other samples. Percent recoveries
were calculated and are shown in Table 10-9.
Recoveries of the carbonyl spikes ranged from 88.2% to 110% for
formaldehyde; 87.5% to 94.1% for acetaldehyde; and 121% to 134% for acetone.
The recoveries for acetone were noticeably higher than the recoveries for
formaldehyde and acetaldehyde. Such a result can be explained because of the
high probability of contamination from acetone in the laboratory environment.
Average recovery of formaldehyde laboratory spikes was 105.2 percent.
Average recovery of formaldehyde field spikes was 115 percent. Both these
recovery statistics reflect excellent sampling and analysis procedures.
10.5.3 Blanks
Laboratory blank analysis are summarized in Table 10-10, and field
blank results are summarized in Table 10-11. Four out of seven laboratory
blanks showed formaldehyde values greater than 1.00 ^g, while 19 out of 31
field blanks showed concentrations of formaldehyde greater than 1.00 ^g. An
acetaldehyde concentration greater than 1.00 /*g was indicated in only one
field blank from NWNJ. Acetone was detected in one field blank from PLNJ.
The reported field concentration were not blank corrected. The significance
of measurable formaldehyde deviations in some of the field and laboratory
blanks required further investigation.
cah.!98f 10-15
-------
TABLE 10-8. FIELD SPIKE RECOVERIES OF FORMALDEHYDE
Site
Code
BRLA
BRLA
BRLA
BRLA
NWNJ
NWNJ
NWNJ
NWNJ
PLNJ
PLNJ
PLNJ
PLNJ
Sample
Date
08/16/90
08/16/90
08/20/90
08/20/90
08/16/90
08/16/90
08/20/90
08/20/90
08/20/90
08/20/90
08/22/90
08/22/90
Sample
ID
1406
1406
1416
1416
1412
1412
1415
1415
1442
1442
1444
1444
Tube
Number
152J
133M
HIM
127E
77F
108J
149E
170D
130J
131K
171E
125C
Percent
Recovery
(%) a
121.4
157.6
94.5
108.3
130.3
92.9
Average 117.5
a Percent Recovery = (Recovered ug from Spiked Cartridge - Recovered ug from Unspiked Cartridge)/20.46* 100
10-16
-------
TABLE 10-9. CARBONYL SPIKE RECOVERIES
Cartridge Formaldehyde Acetaldehyde Acetone
Number Recovery (%) Recovery (%) Recovery (%)
1 110.6 87.5 121.6
2 110.0 94.0 134.3
3 88.2 89.2 126.8
4 91.8 94.1 131.4
cah.!98f 10-17
-------
TABLE 10-10. CARBONYL LABORATORY BLANKS
Analysis
Date
10/04/90
10/13/90
10/13/90
10/16/90
10/26/90
11/11/90
11/27/90
Tube
Number
31J
137D
36B
329
234A
142K
139F
Concentration
Formaldehyde
(ug)
< 1.0
< 1.0
1.08
< 1.0
4.52
1.56
2.01
Acetaldehyde
(ug)
<1.0
< 1.0
< 1.0
< 1.0
< 1.0
< 1.0
< 1.0
10-18
-------
TABLE 10-11. CARBONYL FIELD BLANKS
Site
Code
BRLA
^
NWNJ
PLNJ
Sampling
Date
08/10/90
08/14/90
08/16/90
08/20/90
08/22/90
08/24/90
08/28/90
08/30/90
09/05/90
09/07/90
09/11/90
08/06/90
08/08/90
08/10/90
08/10/90
08/14/90
08/20/90
08/22/90
08/24/90
08/28/90
08/30/90
08/06/90
08/08/90
08/10/90
08/14/90
08/16/90
08/20/90
08/22/90
08/24/90
08/28/90
08/30/90
Sample
ID
1384
1386
1406
1416
1430
1437
1466
1478
1500
1519
1533
1648
1371
1380
1412
1391
1415
1429
1446
1461
1484
1350
1361
1377
1397
1403
1442
1444
1450
1460
1491
Tube
Number
59L
52K
128F
132L
10B
45K
35A
41G
27F
12D
46L
61A
64D
62B
109K
102B
168B
114C
106F
104D
175K
33L
86D
85C
21M
121L
123A
124B
115D
73B
116E
Formaldehyde
Derivative
Concentration
(ug)
2.97
2.27
2.14
3.01
<1.00
<1.00
1.80
1.73
2.65
2.18
1.13
1.99
<1.00
4.46
5.21
1.35
<1.00
<1.00
<1.00
3.76
<1.00
<1.00
2.33
1.49
1.83
1.69
<1.00
1.80
<1.00
<1.00
<1.00
Acetaldehyde
Derivative
Concentration
(ug)
<1.00
<1.00
. <1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
1.75
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
Acetone
Derivative
Concentration
(ug)
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
8.70
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
<1.00
10-19
-------
11.0 REFERENCES
1. Compendium Method TO-12, "Determination of Non-Methane Organic
Compounds (NMOC) in Ambient Air Using Cryogenic Pre-Concentration and
Direct Flame lonization Detection (PDFID)," Quality Assurance
Division, Environmental Monitoring Systems Laboratory, U.S.
Environmental Protection Agency, Research Triangle Park, NC, 27711,
May 1988.
2. Radian Corporation. 1990 Nonmethane Organic Compound Monitoring and
Three-Hour Urban Air Toxics Monitoring Programs, Final Work Plan and
Quality Assurance Project Plan. DCN No. 90-262-045-23. Prepared for
the U.S. Environmental Protection Agency, Research Triangle Park, NC.
EPA Contract No. 68D80014.
3. Radian Corporation. 1989 Nonmethane Organic Compound and Three-Hour
Air Toxics Monitoring Program. Final Report. Prepared for U. S.
Environmental Protection Agency, Research Triangle Park, NC, 27711,
EPA-450/4-90-011. May 1990.
4. Radian Corporation. 1988 Nonmethane Organic Compound and Urban Air
Toxics Monitoring Program. Final Report. Volume I. U. S.
Environmental Protection Agency, Research Triangle Park, NC, 27711,
EPA-450/4-89-003. December 1988.
5. Radian Corporation, 1987 Nonmethane Organic Compound and Air Toxics
Monitoring Programs. Final Report Volume 1 - Hydrocarbons, U. S.
Environmental Protection Agency, Research Triangle Park, NC.
EPA-450/4-88-011. August 19, 1988.
6. McAllister, R. A., R. F. Jongleux, D-P Dayton, P. L. O'Hara, and
D. E. Wagoner (Radian Corporation). Nonmethane Organic Compound
Monitoring. Final Report. Prepared for U.S. Environmental Protection
Agency, Research Triangle Park, NC. EPA Contract No. 68-02-3889,
July 1987.
7. McAllister, R. A., D-P, Dayton and D. E. Wagoner (Radian Corporation).
Nonmethane Organic Compound Monitoring. Final Project Report.
Prepared for U.S. Environmental Protection Agency, Research Triangle
Park, NC. EPA Contract No. No. 68-02-3889, January 1986.
8. Radian Corporation. Nonmethane Organic Compounds Monitoring
Assistance for Certain States in EPA Regions III, IV, V, VI, and VII,
Phase II. Final Project Report. Prepared for the U.S. Environmental
Protection Agency, Research Triangle Park, NC. EPA Contract No. 38-
02-3513, February 1985.
9. Radian Corporation. Proposed Diurnal Nonmethane Organic Compound
Sampling Plan. DCN No. 88-262-045-11. Prepared for U.S.
Environmental Protection Agency, Research Triangle Park, NC. EPA
Contract No. 68D80014. September 30, 1988.
-------
10. McAllister, R. A., Radian Corporation. Letter and Proposal, entitled
"Wet Zero Study", to Frank F. McElroy, Quality Assurance Division,
Environmental Systems Monitoring Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, NC. November 15, 1988.
11. McAllister, Robert A., Memorandum to Vinson L. Thompson, Frank
F. McElroy, U.S. Environmental Protection Agency, AREAL,
"Vince Thompson Canister Cleanup Study Results," dated July 10, 1989.
12. Compendium Method TO-14, "The Determination of Volatile Organic
Compounds (VOCs) in Ambient Air Using SUMMA® Passivated Canister
Sampling and Gas Chromatographic Analysis," Quality Assurance
Division, Environmental Monitoring Systems Laboratory, U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711,
May 1988.
cah.!98f 11-2
-------
APPENDICES
APPENDIX A: SAMPLING SITES FOR 1990 NMOC MONITORING PROGRAM
APPENDIX B: CRYOGENIC PRECONCENTRATION AND DIRECT FLAME
IONIZATION DETECTION (PDFID) METHOD
APPENDIX C: 1990 NMOC MONITORING PROGRAM SITE DATA
APPENDIX D: 1990 NMOC MONITORING PROGRAM INVALIDATED AND MISSING
SAMPLES
APPENDIX E: PDFID INTEGRATOR PROGRAMMING INSTRUCTIONS
APPENDIX F: 1990 NMOC DAILY CALIBRATION DATA
APPENDIX G: 1990 NMOC IN-HOUSE QUALITY CONTROL SAMPLES
APPENDIX H: MULTIPLE DETECTOR SPECIATED THREE-HOUR SITE DATA SUMMARIES
-------
APPENDIX A
SAMPLING SITES FOR 1990 NMOC MONITORING PROGRAM
-------
CO
UJ
CO
oo
o
4- -r-
3C S- X
I •,- O
00 <£. I—
S-
-Q
CO
a:
o
co
o
I—I
X
o
o:
o
UJ
a:
Q
oj
••- d) 0)
"O +J T3
cu
z >->->-
Lf> O •—I i—I
O ^H ^ O
o o o o
O O O —i
I I I I
CTV .—I CO LO
in 10 t—i co
o o o o
till
^ VO ^ ^*
CO CO CO CO
00
O 0)
z >-
o o
O CT>
o o
LO CO
- <-3 <~3
•—> Z 3 _J
—I S Z Cu
X <
t— I
s a:
00 CO
S-
D.
s-
(U
C
Q)
)
c
(Q 00
Q) Q)
Q t—
S-
i— ro
O) -C
'+-> "-3
co z
-t— --^ -a
fO
oo
fO
-a
(O
a>
-M
oo
a.
E
O)
i.
OJ
X -
i— a>
en
—< j<: aj +-> o
01 o -*: M- o
c >- i- c EC
o «-i- 3 o
—i 2 2 to f^ Q) ro
Z Z CX. CQ CO
-—i CsJ
co
LO
vo
CO
ai
ro
O
-------
IP IM
(f< £ CM
X ^v •
•O \r> •*»
•v
>. T-
^ 7) UJ -U
»tf •» ZI 4 :O
»- T t: _j 3 -«
tfi (J ** UJ CO »—
< — c z « «
UJ • «X 2 «tC-i
U
.r
3 U* I »— 3 C U- I »—
i •- •- i 3 i — •- r
i z a: z
—• O 'J -
i « 2 — — -3
r«. -
•n
a: nui •—
^— >- 2» O
j M ij ^ — < i
jj i 2j'jJ^j3_a"v
— » ii—jus
r. — ic .9 < u i ui
i :*2C,x>-*
i. ij"^t>-»luJtJ<
J] _J _J — I UJ (-
_i — ia. — wni
< a -ra>-3-
i. jj uj C _i "v
i i — _i J 2
j^r<-oiu.
^ z a x — x
< z i z —
c o <\i — ;;
— i->3
SH-U
-U"-!
— u r
»— ^
ara:-<
•« 3 r
ru alC
I I t- I
^ I uj
_i I r u
I •- I oj
-i t-uzx <«ija:3e
-'
I » »9 X C X •
*»'fM(M>«Vv»."»l»ia. "*
oi/»r-—z>-.-.i:3 o
c f^ o 3 a: — s. 'N
<« « 'j 2: x
r < .j
i- .3 O Z
•< £ •* 'J I I
»- < _i -r «- z
n _i i c <-i 3
I I i- I uj z
z: •>* j — u.
—1 >-UJ i U. U.
*i —J —J 3 u^ 3
•- « j a. i
i- T 3 W C >•
< 4 U X Z 3
3
-o
o
a
O
O-
j a i-t -
S Z 2 Z
u. UJUI UJUI uJ UJ
z uj t- z uj H- s: a. jj < i a.aJ«z a. AJ < z
i >- — 2.03 >• i- a o 3 *- •- a o 3
X X r-<3IJ X — * 3 • J X >-«3IJ
uj oj i^xri >u i o r r i -u IQTCI
>- •- o a 3
»-<3l'J
— a. Q. UJ «- < XZ3 '-u Z u. UJ ol C
>- xuziu.u.z
-• «-• O •< «C « X -3 O «i
»u>« oo o 3 aua.
»- xt»->-x
uj 3 u. uj uj ^ J D u. uj J _;
< p O <- uj uj 3
a. o. x z at x z
a. a. z z x x
t «
« > »
\ *
-------
Jj
<
i
c t —
K\ XX
i> r- -o
~- ru o
J3 xx
~ 3 -I O
to
rv
«T
i
—' — z « n
<">-3XtO«»>-
.u • o — a.
w\ ••• ox cJ O U
z u i i i r
•a m o -3 J I
? I — z 2: jj
r - u. i —
C I — r- I.
t»5 »»J U -« i— O
I _» .2 1 1 -H
»- t —I (J J Lu
2. U X. I II
_J
o:
I J 2 1 X «
^ < -< LJ u vt;
^ u i i i z
LJ ~> — • 1
t. f! £ ~
Z, C OJ I
S 'a.' -S — •—
I _J 2 i X «
T « 3 i- •- 3
2 s: f « « Q.
I 1 r T>
- jj r — — i
~ s r a> ft a
z * -z >- —
r uj o —• —
~ T Z 05 03
XI
•x.
>.
-V
a: «o -
>— >•
J 3)
-i X 3 «-
•_) «I UJ w
— »— CO
C- 9)
UJ
S
O
•x.
I
•o
-7 in
• M
-\J 3
3 I
3 "M
•• I
3 •»;
3 O
O 'J
» Q. —
3 C; aj
? o»
3 Z J3
^ fM UJ
2 f~ UJ
3-1
.z o
3
cn
J 3 .*J 35 UJ
il 3 3 •« X
I I (- II U 3
•J J «* U t UJ
2 I 3 X — ^
•< -i I J LJ «.
_1 _l »- I JJ >-
i a. -* »? -5 i
r c o >- 3 —
« «I —> «r -r _>
23 OJ
< uj ^ n 4
JO U 93 UJ
Lil 3 i « t
Z '.U '-L! 3 _I X
I !>-_)•->?
tr a < u i uj
2 3 3 X — j;
— « I X1 (J <
_l _l •- I J —
X 1. — »> -> I
Z Z 3 " 3 —
T «i n « M _j
tn -o 3 c a. -t
J 3
UJ -O
« UJ 3 TJ
• 3 3 UJ
•O ^J U.' 3 _I X
I I h- _1 J ^
J J * U> I U.
33 3 x *- «:
— -« I f i-> •*
_l J — I -i( —
0. £L -• 95 ~J I
£ Z 3 >- 3 •-
•» - 3 —
< -9 3 < i O
>• —1 3
3 3 UJ < - i—i — <
3-»»si — a.-7>
•jj < uJ 3 -j ? —. a
• ~?'j^:3 XTOJJ
•— T •_> t- _i
„ — « — ^ _, 7) ,3
_,-,-,-,(_ jj-vic
— o -^ 3 «a
>n -^ —• jj 2 o« 'jj
l«1 05 3 •» X
— >- 3 x < f
X X
£t 3
r T < X < -9 I
• - i— 1 _i I t—
•O "VI CO _l t O - I -L.
• X Z *4 'J •- j_
3 rv —i >- jj x u.
j — * _i _i 3 jj
• •- < _i a. I
< X < 'J I
i- «r _) a: »-
31 J I 3 U
I I t- I UJ
r ^ -> •- n-
9} U.
•M
*- 3
X -
* < J\
*- •? 3 UJ T)
< < (j a: i
<- T ^
« I « 'J I
•- •« _I CC «-
« J I O U
I I I- I i.
~ -^ J •— *.
— >- jj JC U,
•« _l J O -k.
»- «t _j a i
— 2 3 UJ 3
< < U I Z
UJ
t-
<
e
»J _J I O LJ
I I >- I uJ
in
• xx:
' O — ••>
Ola
•^ o I
. o — — •
x
in
32 AjauJO Ztt ••
3OZuJ-< Z1T3
z a uj i
•~t«a'«^»a-3ia«
X JJ < £
>-»- O.2 3
r — « o i (j
UJ I 3 T Z I
•- a: i »- t- TC
uj o u. uj uj O
Z •->•. T T *-
-C -H UJ I I -•
a: u z i u. u. z
•< a o •— aj uj a
a. a. x z at at z
3. UJ « Z
f- •- a o 3
a. xi 4 r
>- >- a a 3
X t- < 3 I (J X
aj i a x r i jj
r- X t t- t- X —
jj 3 U. UJ UJ 3 -1
f f- U. I I H- T
< — UJ I I —• < I
X O 2 I U-U, 2 X O
-x
3U.U1UJ3
UJ UJ
« z.
>• H- o a 3
x —«3io
uj i a £ x i
-j a u. -jj ui 3
«t >-- uj ua 3
a. a. s. z x x z:
-------
G Z c-j CNJ
-o x X
in a o cr-
a m -i o
m xx
m G* o
a to oj 01
> Q. cr
33-10
a 41—
4 t- > LJ
,1 UJ UJ
•- Z 4 01
z a —j cj 4
u M LJ 01 p-
• 4 cr 4 a
u- > a t- Q.
U. UJ LJ 01 Z P-
a u EE «j u Z
W1
J) CM
— x a:
— O* 21
c a K> a
— x o >
--. o o
T S3 Z
o o in
-1 Z CM CSJ
M X X
•s, m o -T
3 -i -i O
-T XX
03 eo z z
X
X
X
X
X
X
-i a
11 UJ
Z >- '
4
uj a
> a cr
a 3 -i o
a < »-
4 i- > •* O£
a z z (/> in a
£3 uj i i-
3
O
~5 >- 4
a cc Q
O UJ c I—
~ U C£ I
4 M O O
CO UJ O t~l *^ (J
i _i Z K e i-i
i— 4 1-1 U U UJ
2 U C I . =
- o
a uj a M « Q:
^^ (/>*-» i** « ta
= 4 ~ v- i- o
a uj o >-. — cr
3 Z Z tfl (rt a.
3
+T
>• Z
1,1
il
4 tr
> z o
X VI UJ 2:
— 3- > a
jj 10 Z >
C 3 u —
— a oo
Z —J uj
-» ac a —
-t 4
O 4 UJ •—
— J —
-U Q LJj i—
~ "7 ^ -
2
••*
hH
U
UJ
^*
^
o
i — LJ
: > 4
:Js
>
!•*
u
CE
—
> 0
3
.. ^fl
4
4
> 03 Pa
cr r» o
UJ X r*
ii in
Z x a.
3 M 01
CM *
— O >
4 01 UJ
3
x x
-a
3
X X
•0
S3
z a
•j o
LU Z
a LU
'J o
— «-i
x
a in
UJ tO
a 4
UJ 3 —
— -J U
4 U 1
a x p-
1 UJ O
X
a
01
4
UJ
cr
x
UJ
*
4
86/00/05
/ /
/ /
z a
4 UJ
CD Q
UJ Z
ca uj uj
CD U 4
z z a
a i
'i.1 •
a 4
^ ^ _
-J U
U 1
>: >-
LU U
X
•4
>
— a
a.
UJ 01
i 4
4
86/06/05
/ /
Z Q
4 UJ
'j a
uj Z
a uj
CD C3
Z Z
X
a
"«.
uj 3
4 u
a x
' UJ
X
01
.. .. <
SEE
01 UJ
01 4 *—
'.1 UJ 01
4 CE P"
.— X Z
• jj U
P- 1L
a a
>• z
_ 4
>• -J Z
— Z 4
o u >- «j cr
-j ». 3
»; i: 01 i—
= — cr z -
Z M S3
a x x
21 O f
•> X
Z O er
u. O O
O -•
— o cr
_^ — oj cr a. -o in
3 — p* — ujop-axx
3 *O v- — <,-, j- 01 3 in
333 :J 4 UJ —
UI ~ .- 5 • = 4 — O
- >• - _ X l- — — -
— — uj cr ol a
^ o
CX CX PH 01 1 I O
z z a 3- o i- z
4 4 3 4 a; u 3
01 01 4 o a. 4 a
CX
CD
3 co x cr
= 3 3
ul =
3 4 LJ
— a a z
4 CO 4 O I
— 4 -j cr i— Z
01 -j .1 a u
UJ — ^
— a o 4 a 4 i — -oj
ULJ4~'__='aPJ01
~ "i a
N- a
— Q.
4 Z
a 4
' - • 'JO 1 CE
— -. in x x p- >• •• — u
•3 Ps ^ — ^ t.1 Ul
3 o in z •* • Z
1 --xxzucin
-SPsgj 4Z3OO
— •>. O- —
a —, — u o
ZZ^OUJ3Z~- ••
— — oi>-4Lnoo1a-
— — — 4Z — 4>-Z
4 4 _ Z Z PI
334Z = ujZ --cr
— — p-aacrpioiuj
— C 3011-Q4— — O
— — JJ CjCXZr-
o-i u 4 a a 01 3 01 u 01
Z Ni o P- u.
— >- _. cr u. u.
4 a uj a
— 4 _. CX I
— Z O UJ O >-
4 4 u CE ce a
3
01
O
X
•a
3
X
ex a. w O i- >-
4 4 3 4 CE U 01
01 01 4 O a 4 —I
4
u
X
3 03 X 4
= 3 01
a
2
uj a
01 —a.
3 4 z
— a a a
4 03 4 CD I U
— 4 _i cr p-
oi -i i a oj u
Z N u l- u. Z
»* >• iu cr u. 4
•a -; —i a LU CD
— 4 _i cx i cr
— z o LJ a a
4 4 <_> CE CC
LU
CX CX PI 01
Z — a > o >— cr
4 4 3 4 cr u oj
01 01 4 a EX 4 u.
= a
uj C
01 — i.
3 4 Z
— a a a
4 CO 4 CD I -
— 4 _i d >—
01 —i » a u cj
Z fc ^ U P- u- Z
a I w § UJ 3
— •• w a. i =1
i- .; a uj a c
4 4 U IT Z
a
— k- Z 4 "" 01
— r a -j o 4
4 p. a K 01 4
ui • 4 er 4 o
— > O I— Q.
Z U. UJ LJ 01 Z P-
— •« -j uj 4 a uj
_ a LU cc ^ u —.
"^ - ^
— CD UJ P-
_l =3 > O
-J a UJ r— JJ 3
• Z Z Z ol 4
a — — cj p- u 01
.-33 = = = = g
4 a
3 § S
OJ > UJ CE
a c.
U 3
-i a
4 —
_u C3 O
a i-
u Z
in
a
X
a
3
-r IM KI co
Z in z
' 3
Z X X Z
a cv -o 3
Z 3 3 Z
-i xx
•X U
c z z z
O a z PI
>• z a z -
4 a a 01
Z — — a
>- UJ Z > — —
— z 4 z z «
— o > a LJ a cx
— u >• -j cr z -j
LU 4 3 — 21
^ ^ 01 M a M
:r 21 01 u •• 3 u.
caujcr— — z a
.- >- ~> uj P* Z 3
"~ "~ — P- "Z. 4 *J
— — ^ — "~ ^ ?F °~ ~ °
3JD^T — 01401 303
3OO _^Z4UJ =
-1 01 3 • = 4 P" O
— >• 3 p-i «- — — Q 01
— — ui cr 01 — —
— OLjSat-^zz*:
__ Z 5- CX PH
S a rs. -o ui cr Z •-". a
- - o 3 — cx
-i — in x x •:>-•• z
x x o iu o
- — >• O 01
>- — a 3 o
L- — a a a
st — 4 a i oj
' a = = i
cr i — — CE
£. UJ < —
5- p- c a o
cr »•— 4 a ' oj
— cr T — — cr
— — •— < a — 4>z
4 4 -J Z Z i-l
— -I O P-. •• 4 CD i
-• 4 h*
CTUZlu,-.Z —UZ
-^ 4 —
4 a a *•• iu uj a
ooh-ujuja 4 a a — u* uj a
o.zz:re:= =.=. = _; =i = =
'.iu4aaoi3oiojoi
-------
0-0
M O
~ N.
l O Ul
_ O
-N S
> a. =:
C 3 -i O
a < i—
C3 < *— > u
Z -1 UJ UJ
— — -S. < '.1
r- — O -J CT <
(.1 U •-> -u ol 1-
< — a E 01 <
uj • < e: < a
a. > O t- Q.
C u. uj u 01 Z >—
*— — —i uj < O UJ
Z a uj o: —. u Z
r>. a-
rt *f\
.. -.
^1 05
3
c:
-1
\
N.
LU — *l *•« tj U UJ
z u a: i • 3
- — a:
2 z Z ;i 01 a.
LJ - w «
5 5 uj i i
.1 UJ
h. ~ — Z Z uj
—i 01 1-4 « •* CD
Z < Z i Q
c uj a ** »-• (x
c C Z LI 01 a
U — N- ««
tx - z a
— < i-» u o -j
3 to a u CD T
z i >- z z -j
— 01 >-> — — a, s =
3 i- > a —
~ ^ 1 UJ UJ Z
— Z 3 — CT < 5
.1 a 11 -i i/i i- i
•* — a a: ci « >—
-j • a t- c- -i
Z — ^ooiZr-o:
a
« •« Q.
H 3 a
o
\ N \
;^^
u
LU
UJ
U
a
—
O (.1
U
UJ
2
—
•"
g
•—
UrERIA?
36-X/25/a7
U N
" N
— in
£
|
-
™
<
O
a
UJ LU
=L ^
^ ^
"^ 13
01 <
a
X
UJ
01
>
•
a L
a. <
0
01
j
> z a
( 4 UJ
j a o
! a uj LK
j a u S
: z Z c
C n ** i
Q 2. G- •-»
Z O Li 01 5
EXCIUIHO :
.CT-CIAS.S :
rAi/cn^DCA<~. nrl-
O >•
4 Si '•,
a a. <
j
j o
£ O
-. a.
J Ol
i
_ ^
J VI
z
£
5
z a o
- O »~
<3 < Z iu — UJ2I-O ~" 2 2 rZ ^ — — C "~ 3 2
— ^_r^*j1*S(/l^*^^ *3C3^CJt <— — C,1C^« — * uj Z ik— ' _u -J . i u— i ^j
— ^ Z*^Z*~ -^N. Z^JO^~u_ Z*sJLJ^*^ Z Z ^-J O *** "-
— •— •/) GI >• LU wi •" >• uj d u_ v. ^">-ujCCu« - — zcuao G ^-ZCLJO
LJLJ4^^ -~UJCQ*< .-•- <-* —
ZZ-iQ^CZO ••
CC = -J -CuJ«O
o
3
r F-t K> OD
r CM fo ao
— _. C3 ^^ •• •< tJ
E£-SES£^£ - -22l;
— 3 C -" •*=••* - - - - ™* —"*""••.
_ = - LJ O
zz_ai- — z-j
— — — «*:;— *5 —
— —
i « =. Z
e (.1 Z L,-, u
-------
HI ««
cs as cr a
3 a u i »-
o i i- >- z
1.1 LU O >1 « O
>- 4 1-4 u U UJ
Z U 0£ ' I Z
t 01 O U OS I
Z I H- 2 Z UJ
-1 00 M M « 03
z < z i- i- a
O uj o >-« 1-4 o:
a ~ Z oi 1/1 i
LU
LU 4i—
u ~t K 4
a: a z a
3 C UJ I 1—
T i — »- X
wl LU U M M O
' -i z cr ex IH
— 4 »-t OJ OJ UJ
Z u a: i is
4 00 O O CS I
Z I H- Z Z UJ
-* oo 1-4 i-> <-4 a)
~ 4 r i- *- o
a u o p-> 1-1 <
a: a a: o
a o u i t-
a ' <- *- z
CO LU CS >•* >-« O
i _ Z a o: 1-1
*- 4 *H OJ OJ UJ
— u a; i i z
4 oo o cs cs i
2 1 >- Z Z LU
-* V) ••* ••* I-* CQ
Z 4 2 i— •— O
O LU O 1-1 i- cc.
Q Z Z 01 V) Q.
Z CM
X X
Nl O
,o~SS
a co
x
X
LU a
> a.
a 3
a
•t >—
01
= 12
U 1-1
I- Q
• 4 cr
u. > o
Lv UJ OJ
2 a uj 01
z z
ex
-1 O
•a t- X
> o t-
LU LU
01
cy -" "-* OS
Q S Z 01 01 0.
in ••
~i ts
3
4
LU
CO
xxx
4
3
XXX
on
03
z a
o
c^
xxx
X *0
a
xxx
on
a
a
01 Z Q
X
5
00 ftj
/ /
/ /
BI/90/S0
Z Q
x
CJ
00
o cr
01 LU
3 = X
>- O Q
•* Kl Z
— -i 3 3 = o =
_J i_i •• CO
4 z a o
> 4 Z —
UJ I— l-t
01 4
»-» CJ 01
= a = Z
a
a
xxx
^o
a
xxx
0}
a)
uj oo Z -1
Si to l-t
=: <» a: r
C 2 U —
— an
Z -J uj
— a: a —
—• <
CJ 4 IU —
a
'J « -J 01 U.
uj Z a 4 cr
a uj uj 2 — x
' i *- -J u Z
c; u «t u i uj
z z a x i- *
— — I Ul U <
—. .J *- I UJ »—
s. o. M co -) i o
r r 3 > a t- s
1 1 3 < S U 3
-,i oi « a a. « o
COI
4 ou at/)'
>i C2 LU 00 LU OJ
LU z a 4 cc o
03 UJ UJ 3 —I X CX
> ' i- -J u Z
O CS 4 U I UJ >
Z Z a x >- * 4
—• >-4 I UJ U 4
-i -j i- i LU H- cr
CL Q. M 00 n I UJ
z z a > a t- i-
4 4 3 4 CT U 00
01 oo 4 a a. 4 1-1
u
X
UJ C 01
u a LU oo LU co
u z cs < o; w
a uj uj 3 -J x Z
i i *— -J cj Z <
CJ C3 < U i uj CJ
Z Z O X I- *
** *< I UJ LJ < ^*
_ ^ H- I LU *- Z
c. a. 1-1 oo ~? i ou
Z Z a :- o ^ OE
o < 3 « te u LU
01 co < a a. «. u.
Z o
< LU
a a
3
~
X
CO X
a
a*
LU
—
4
O
cr
O
rjj
<
z
3
c*
co
o
0-
X
X
LU
LU
4
01
a
5
o
X
3 S3 X
~ a
ou
— «f
— a a _
4 CO 4 CS I '
— 4 _j c: t- z
oi -i i a u o
I h— I UJ """
Z N U u. u. ~
<—< >- UJ Ol LU U.
4 _> _, a LU a
— 4 -J CL I
— Z O LU O 5-
4 4 oj cc cr a
LU 3
i- a. oo
a a a
4 en 4 cs i u
— 4 -j cr. »_
01 -j i a u u
Z N OJ >- ui z
— >• uj cr u. 4
4 -i -J O OU (J
>— 4 —i a. i cr
i— z o LU o a
4 4 u cr cr
LU
O
>— a.
.3 4 Z
i- a a o
4 CO 4 CS I U
i- 4 _i a: i-
01 -J ' O OJ OJ
O
oo
4 Z
** 4
oo a
03 S
OJ O -J 03
-143
Z —i OO
O Z V-
oo cr z ••
a ou LU —
LU .= a CM
ot 3 01 ~
i LU a LU cs x
i T 01 cr z ^
o on >o «-4 uj z
a j- • -
O OJ '
'.-, O
i— 4 _j cu i a:
i- z o LU o o
4 4 u cr cr
o
— 5- _
U» CQ
>- 2 CS Q 4
*- 3 OJ Z OJ •
— O O 4 O Z
OJ U 4 — -J 4
O
_)
CL <%t
01 O
a
x
X
X
o
•T
X
o
3
X
ul
CO
O M 03
O O 03
o — in _
ui eo rt x x 2
j- -• — LU Z
. uj O
CO O Z
- o
a
a 01 i'
UJ 00 UJ
a < cr
CD uj uj 3 —J x
i ' t— -j u Z
U CS < OJ ' LU
z z a x K *
!•* m I LU OJ <
-J ^ *— I UJ ^
a a. P-I oi ~i i a
Z = c > o i- Z
<« « 3 "* C£ OJ 3
oo i/i < o a. < a
^
o
LU «
3 4 LU
>- ca a z
4 CO 4 CS I
i— 4 j a; *- z
01 _j i a oj o
' I k— ' LU Z
Z M U h- -.
—" ^ LU a u_ _
4 ^ .j a LU C
— Z 3 UJ C3 >-
4 4 u cr or a
a
x
CL UJ 4 Z
>- — a a o
N. 4 o i u cr
c: i i
OU.
o
U
CE
O
= u
•x a
-__ ^
LU LU
ou — Z'
CL UJ 4 Z
cr ^ 4 o i oj
—i ' a z z i
u z ^
. a. Z :
a 'j
Z Z -j a cr
O O Z LU O
l-l ** OO ^ LU
4 4 — Z
3 3 4 Z ic
CL a. ^- cr a
a a o 01 LU a
*H GL CL LU ^ '
uj ^ cr ou uj
k— t— OJ *- — •—
-4 i_i c? 4 4 >->
oi u 4 a a 01
co LU
a a
~ — a a o
cr ^- «a c i oj
Z O uj
4 CL Z .
co a. z
cc 3 O
2 01 u
_ — cr
UJ G U. UJ LU O
4 i- uj T T Z
ac u Z i u. u. z
4 Q O — LU LU O
-------
O rti M«
9 X Ml*
9- -V N.
-N. >,
* >-
39 a 2 2
I •-
«. I
3 _ia
z •» ut tu
-» — Z « 91
~ « 3 _i a «
31 i3 "• Jj « •-
< r- e z a <
-u «4 ac « a
j. > C »- a.
3 a iu J UJ >-
z «-
— SIX a •*
< 3 — — 3 — 'J»>
1 Z
» I 3
_ ^1 •- <
— — ^ z
IB——
- >- — 3 —
- - •- 3 C
i — 1. ~ I » *J
t _l — M < I 'J
r « r M x <
- W
J2_»_iX
j. «c a: u •• —
r vu x ~ -
r : uj — j
T *• ^- T ^* *
- ft 3 r •-
« « 3 3 J r
- •» 3 i •£ -j>-
— a. < E-> — •- o •—•• — -o
I! 3. < ->—a«<—«
-------
APPENDIX B
CRYOGENIC PRECONCENTRATION AND DIRECT FLAME
IONIZATION DETECTION (PDFID) METHOD
-------
COMPENDIUM METHOD TO-12
DETERMINATION OF NON-METHANE- ORGANIC
COMPOUNDS (NMOC) IN AMBIENT AIR USING
CRYOGENIC PRE-CONCENTRATION AND
DIRECT FLAME IONIZATION DETECTION
(PDFID)
QUALITY ASSURANCE DIVISION
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
MAY, 1988
-------
Revi sion 1.0
June, 1987
METHOD T012
METHOD FOR THE DETERMINATION OF NON-METHANE ORGANIC COMPOUNDS (NMOC)
IN AMBIENT AIR USING CRYOGENIC PRECONCENTRATION AND DIRECT FLAME
. IONIZATION DETECTION (PDFID)
1. Scope
1.1 In recent years, the relationship between ambient concentrations
of precursor organic compounds and subsequent downwind concentra-
tions of ozone has been described by a variety of. photochemical
dispersion models. The most important application of sucn models
is to determine the degree of control of precursor organic com-
pounds that is necessary in an urban area to achieve compliance
with applicable ambient air quality standards for ozone (1,2).
1.2 The more elaborate theoretical models generally require detailed
organic species data obtained by multicomponent gas chromatography (:
The Empirical Kinetic Modeling Approach (EKMA), however, requires
only the total non-methane organic compound (NMOC) concentration
data; specifically, the average total NMOC concentration from 6
a.m. to 9 a.m. daily at the sampling location. The use of total
NMOC concentration data in the EKMA substantially reduces the
cost and complexity of the sampling and analysis system by not
requiring qualitative and quantitative species identification.
1.3 Method T01, "Method for The Determination of Volatile Organic
Compounds in Ambient Air Using Tenax* Adsorption and Gas
Chromatography/Mass Spectrometry (GC/MS)", employs collection
of certain volatile organic compounds on Tenax* GC with subse-
quent analysis by thermal desorption/cryogenic preconcentration
and GC/MS identification. This method (T012) combines the same
type of cryogenic concentration technique used in Method T01
for high sensitivity with the simple flame ionization detector
(FID) of the GC for total NMOC measurements, without the GC
columns and complex procedures necessary for species separation.
-------
T012-2
?
1.4 In a flame ionization detector, the sample is injected into a
hydrogen-rich flame where the organic vaoors burn producing
ionized molecular fragments. The resulting ion fragments are
then collected and detected. The FID is nearly a univer ,
detector. However, the detector response varies with t species
of [functional group in] the organic compound in an oxy 1 atmos-
phere. Because this method employs a helium or argon carrier
gas, the detector response is nearly one for all compounds.
Thus, the historical short-coming of the FID involving varying
detector response to different organic functional groups is
minimized.
1.5 The method can be used either for direct, in situ ambient
measurements or (more commonly) for analysis of integrated
samples collected in specially treated stainless steel canisters.
EKMA models generally require 3-hour integrated NMOC measurements
over the 6 a.m. to 9 a.m. period and are used by State or local
agencias to prepare Sta^e Implementation Plans (SIPs) for ozone
control to achieve compliance with the National Ambient Air
Quality Standards (NAAQS) for ozone. For direct, in situ ambient
measurements, the analyst must be presen* iring the 6 a.m. to
9 a.m. period, and repeat measurements (a oximately six per
hour) must be taken to obtain the 6 a.m. to 9 a.m. average
NMOC -oncentration. The use of sample canisters allows the
col" 'ion of integrated air samples over the 6 a.m. to 9 a.m.
period by unattended, automated samplers. This method has
incorporated both sampling approaches.
Applicable Documents
2.1 ASTM Standards
D1356 - Definition of Terms Related to Atmospheric
Sampling and Analysis
E260 - Recommended Practice for General Gas Chromato-
graphy Procedures
• E355 - Practice for Gas Chromatography Terms and
Relationships
-------
T012-3
2.2 Other Documents
U. S. Environmental Protection Agency Technical Assistance
Documents (4,5)
Laboratory and Ambient Air Studies (6-10)
3. Summary of Method
3.1 A whole air sample is either extracted directly from the ambient
air and analyzed on site by the GC system or collected into a
precleaned sample canister and analyzed off site.
3.2 The analysis requires drawing a fixed-volume portion of the
sample air at a low flow rate through a glass-bead filled trap
that is cooled to approximately -186°C with liquid argon. The
cryogenic trap simultaneously collects and concentrates the
NMOC (either via condensation or adsorption) while allowing
the methane, nitrogen, oxygen, etc. to pass through the trap
without retention. The system is dynamically calibrated so
that the volume of sample passing through the trap does not
have to be quantitatively measured, but must be precisely
repeatable between the calibration and the analytical phases.
3.3 After the fixed-volume air sample has been drawn through the
trap, a helium carrier gas flow is diverted to pass through
the trap, in the opposite direction to the sample flow, and
into an FID. When the residual air and methane have been
flushed from the trap and the FID baseline restabilizes,
the cryogen is removed and the temperature of the trap is
raised to approximately 90°C.
•
3.4 The organic compounds previously collected in the trap revol-
atilize due to the increase in temperature and are carried into
the FID, resulting in a response peak or peaks from the FID.
The area of the peak or peaks is integrated, and the integrated
value is translated to concentration units via a previously-
obtained calibration curve relating integrated peak areas with
known concentrations of propane.
3.5 By convention, concentrations of NMOC are reported in units of
parts per million carbon (ppmC), which, for a specific compound,
is the concentration by volume (ppmV) multiplied by the number
of carbon atoms in the compound.
-------
T012-4
3.6 The cryogenic trap simultaneously concentrates the NMOC wnile
separating and removing the methane from air samples. The
technique is thus direct reading for NMOC and, because of
the concentration step, is ^ire sensitive than conventional
continuous NMOC analyzers.
Significance
4.1 .Accurate measurements of ambient concentrations of NMOC
are important for the control of photochemical smog because
these organic compounds are primary precursors of atmospheric
ozone and other oxidants. Achieving and maintaining compliance
with the NAAQS for ozone thus depends largely on control of
ambient levels of NMOC.
4.2 The NMOC concentrations typically found at urban sites may
range up to 5-7 ppmC or higher. In order to determine transport
of precursors into an area, measurement of NMOC upwind of the
area may be necessary. Upwind N..JC concentrations are likely
to be less than a few tenths of 1 ppra.
4.3 Conventional methods that depenj on gas chromatography and
qualitative and quantitative species evaluation are excessively
difficult and expensive to operate and maintain when speciated
measurements are not needed. The method described here involves
a simple, cryogenic preconcentration procedure with subsequent,
direct, flame ionization detection. The method is sensitive and
provides accurate measurements of ambient NMOC concentrations
where speciated data are not required a.s applicable to the
EKMA.
Definitions
[Note: Definitions used in this document and in any user-prepared
Standard Operating Procedures (SOPs) should be consistent with ASTM
Methods D1356 and E355. All abbreviations and symbols are defined
within this document at point of use.]
-------
T012-5
5.1 Absolute pressure - Pressure measured with reference to absolute
zero pressure (as opposed to atmospheric pressure), usually ex-
pressed as pounds-force per square inch absolute (psia).
5.2 Cryogen - A substance used to obtain very low trap temperatures
in the NMOC analysis system. Typical cryogens are liquid argon
(bp -185.7) and liquid oxygen (bp-183.0).
5.3 Dynamic calibration - Calibration of an analytical system with
pollutant concentrations that are generated in a dynamic, flow-
ing system, such as by quantitative, flow-rate dilution of a
high concentration gas standard with zero gas.
5.4 EKMA - Empirical Kinetics Modeling Approach; an empirical model
that attempts to relate morning ambient concentrations of non-
methane organic compounds (NMOC) and NOX with subsequent peak,
downwind ambient ozone concentrations; used by pollution control
agencies to estimate the degree of hydrocarbon emission reduction
needed to achieve compliance with national ambient air quality
standards for ozone.
5.5 Gauge pressure - Pressure measured with reference to atmospheric
pressure (as opposed to absolute pressure). Zero gauge pressure
(0 psig) is equal to atmospheric pressure, or 14.7 psia (101 kPa).
5.6 in situ - In place; in situ measurements are obtained by direct,
on-the-spot analysis, as opposed to subsequent, remote analysis
of a collected sample.
5.7 Integrated sample - A sample obtained uniformly over a specified
time period and representative of the average levels of pollutants
during the time period.
5.8 NMOC - Nonmethane organic compounds; total organic compounds as
measured by a flame ionization detector, excluding methane.
5.9 ppmC - Concentration unit of parts per million carbon; for a spe-
cific compound, ppmC is equivalent to parts per million by volume
(ppmv) multiplied by the number of carbon atoms in the compound.
5.10 Sampling - The process of withdrawing or isolating a representative
portion of an ambient atmosphere, with or without the simultaneous
isolation of selected components for subsequent analysis.
-------
T012-6
6. Interferences
6.1 In field and laboratory evaluation, water was found to cause a
positive shift in the FID baseline. The effect of this shift
is minimized by carefully selecting the integration srmination
point and adjusted baseline used for calculating the area of
the NMOC peak(s)
6.2 When using heliui. ^s a carr is, FID response is quite
uniform for most hydrocarbon ^..npounds, but the resoonse can
vary considerably for other types of organic compounds.
7. Apparatus
7.1 Di.act Air Sampling (Figure 1)
7.1.1 Sample manifold or sample inlet line - to bring
sample air into the analytical system.
7.1.2 Vacuum pumo or blower - to draw sample air through a
sample r ifold or long inlet line to reduce inlet
residence time. Maximum residence time should De no
greater than 1 minute.
7.2 Remote Sample Collection in Pressurized Canisters (Figure 2)
7.2.1 Sample can-1 *.er(s) - stainless steel, Summa«-polished
vessel(s) 01 4-6 I capacity (Scientific Instrumentation
Specialists, Inc., P.O. Box 8941, Moscow, ID 83843), used
for automatic collection of 3-hour integrated field
air samples. Each canister should have a unique identi-
fication number stamped on its frame.
7.2.2 Sample pump - stain" r.ss steel, metal bellows type
(Model MB-151, Metal Bellows Corp., 1075 Providence
Highway, Sharon, MA 02067) capable of 2 atmospheres
minimum output pressure. Pump must be free of leaks,
clean, and uncontaminated by oil or organic compounds.
7.2.3 Pressure gauge - 0-30 psig (0-240 kPa).
7.2.4 Solenoid valve - special electrically-operated, bistable
solenoid valve (Skinner Magnelatch Valve, New Britain,
-------
T012-7
CT), to control sample flow to the canister with negligi-
ble temperature rise (Figure 3). The use of the Skinner
Magnelatch valve avoids any substantial temperature rise
that would occur with a conventional, normally closed
solenoid valve, which would have to be_ energized during
the entire sample period. This temperature rise in the
valve could cause outgasing of organics from the Viton
valve seat material. The Skinner Magnelatch valve
requires only a brief electrical pulse to open or close
at the appropriate start and stop times and therefore
experiences no temperature increase. The pulses may
be obtained with an electronic timer that can be pro-
grammed for short (5 to 60 seconds) ON periods or with
a conventional mechanical timer and a special pulse
circuit. Figure 3 [a] illustrates a simple electrical
pulse circuit for operating the Skinner Magnelatch
solenoid valve with a conventional mechanical timer.
However, with this simple circuit, the valve may
operate unpredictably during brief power interruptions
or if the timer is manually switched on and off too
fast. A better circuit incorporating a time-delay
relay to provide more reliable valve operation is
shown in Figure 3[b].
7.2.5 Stainless steel orifice (or short capillary) - capable
of maintaining a substantially constant flow over the
sampling period (see Figure 4).
7.2.6 Particulate matter filter - 2 micron stainless steel
sintered in-line type (see Figure 4).
7.2.7 Timer - used for unattended sample collection. Capable
of controlling pump(s) and solenoid valve.
7.3 Sample Canister Cleaning (Figure 5)
7.3.1 Vacuum pump - capable of evacuating sample canister(s)
to an absolute pressure of <5 mm Hg.
7.3.2 Manifold - stainless steel manifold with connections
for simultaneously cleaning several canisters.
7.3.3 Shut off valve(s) - seven required.
7.3.4 Vacuum gauge - capable of measuring vacuum in the manifold
to an absolute pressure of 5 mm Hg or less.
-------
T012-8
7.3.5 Cryogenic trap (2 required) - U-shaped open tubular trap
cooled with liquid nitrogen or argon used to prevent con-
tamination from back diffusion of oil from vacuum pump,
and to provide clean, zero air to sample canister(s).
7.3.6 Pressure gauge - 0-50 psig (0-345 kPa),-to monitor
zero ai r pressure.
7.3.7 Flow control valve - to regulate flow of zero air into
nister(s).
7.3.8 Humidifier - water bubbler or other system capable of
providing moisture to the zero air supply.
7.4 Analytical System (Figure 1)
7.4.1 FID detector system - including flow controls for the
FID fuel and air, temperature control for the FID, and
signal processing electronics. The FID burner air,
hydrogen, and helium carrier flow rates should be set
according to the manufacturer's instructions to obtain an
adequate FID response while maintaining as stable a flame
as possible through t all phases of the analyt- al cycle.
7.4.2 Chart recorder - c itible the output ignal,
to record FID respo..ie.
7.4.3 Electronic integrator - capable of integrating the area
of one or more FID response peaks and calculating peak
area corrected for baseline drift. If a separate inte-
grator and chart recorder are used, care must be exer-
cised to be sure the . these components do not interfere
with each other electrically. Range selector controls
on both the integrator and the FID analyzer may not pro-
vide accurate range ratios, so individual calibration
curves should be prepared for each -ange to be used.
The integrator should be capable of .narking the beginning
and ending of peaks, constructing the appropriate base-
line between the start and end of the integration period,
and calculating the peak ared.
-------
T012-9
Note: The FID (7.4.1), chart recorder (7.4.2), inte-
grator (7.4.3), valve heater (7.4.5), and a trap heat-
ing system are conveniently provided by a standard lab-
oratory chromatograph and associated integrator. EPA
has adapted two such systems for the MJFID method: a
Hewlett-Packard model 5880 (Hewlett-Packard Corp., Avon-
dale, PA) and a Shimadzu model GC8APF (Shimadzu Scientific
Instruments Inc., Columbia, MD; see Reference 5). Other
similar systems may also be applicable.
7.4.4 Trap - the trap should be carefully constructed from a
single piece of chromatographic-grade stainless steel
tubing (0.32 cm O.D, 0.21 cm I.D.) as shown in Figure 6.
The central portion of the trap (7-10 cm) is packed with
60/80 mesh glass beads, with small glass wool (dimethyldi-
chlorosilane-treated) plugs to retain the beads. The
trap must fit conveniently into the Dewar flask (7.4.9),
and the arms must be of an appropriate length to allow
the beaded portion of the trap to be submerged below
the level of liquid cryogen in the Dewar. The trap should
connect directly to the six-port valve, if possible,
- to minimize line length between the trap and the FID. The
trap must be mounted to allow the Dewar to be slipped
conveniently on and off the trap and also to facilitate
heating of the trap (see 7.4.13).
7.4.5 Six-port chromatographic valve - Seiscor Model VIII
(Seismograph Service Corp., Tulsa, OK), Valco Model 9110
(Valco Instruments Co., Houston, TX), or equivalent.
The six-port valve and as much of the interconnecting
tubing as practical should be located inside an oven or
otherwise heated to 80 - 90°C to minimize wall losses
or adsorption/desorption in the connecting tubing. All
lines should be as short as practical.
7.4.6 Multistage pressure regulators - standard two-stage,
stainless steel diaphram regulators with pressure gauges,
for helium, air, and hydrogen cylinders.
7.4.7 Pressure regulators - optional single stage, stainless
steel, with pressure gauge, if needed, to maintain
constant helium carrier and hydrogen flow rates.
-------
T012-10
7.4.8 Fine needle valve - to adjust sample flow rate through
trap.
7.4.9 awar flask - to tr- liquid cryogen to cool the trap,
sized to contain suomerged portion of^trap.
7.4.10 Absolute pressure gauge - 0-ACQ mm Hg,(2 mm Hg "scale
divisions indicating units]), o monitor repeatanle
volumes of sample air through cryogenic trap (Wallace
and Tiernan, Model 61C-ID-0410, 25 Main Street, Belle-
ville, NO).
7.4.11 Vacuum reservoir - 1-2 L capacity, typically 1 L.
7.4.12 Gas purifiers - gas scrubbers containing Drierite« or
silica gel and 5A molecular sieve to remove moisture
and organic impurities in the helium, air, and hydrogen
gas flows (Alltech Associates, Deerfield, IL). Note:
Check Mty of gas purifiers prior to use by passing
zero-o.. through the unit and analyzing according to
Section 11.4. Gas purifiers are clean if produce
[contain] less than 0.02 ppmC hydrocarbons.
7.4.13 Trap heating system - chromatoc .phic oven, hot water,
or other means to heat the trap to 80° to 90°C. A simple
heating source for the trap is a beaker or Dewar filled
with water maintained at 80-90°C. More repeata^.e types
of heat sources are recommended, including a temperature-
programmed chromatograph oven, electrical heating of
the trap itself, or any type of heater that brings the
temperature of the trap up to 80-90°C in 1-2 minutes.
7.4.14 Toggle shut-off valves (2) - leak free, for vacuum valve
and sample valve.
7.4.15 Vacuum pump - general purpose laboratory pump capable
of evacuating the vacuum reservoir to an appropriate
vacuum that allows the desired sample volume to be
drawn through the trap.
7.4.16 Vent - to keep the trap at atmospheric pressure during
trapping when using pressurized canisters.
7.4.17 Rotameter - to verify vent flow.
-------
T012-11
7.4.18 Fine needle valve (optional) - to adjust flow rate of
sample from canister during analysis.
7.4.19 Chromatographic-grade stainless steel tubing (Alltech
Applied Science, 2051 Waukegan Road, Deerfield, IL, 60015,
(312) 948-8600) and stainless steel plumbing fittings -
for interconnections. All such materials in contact
with the sample, analyte, or support gases prior to
analysis should be stainless steel or other inert
metal. Do not use plastic or Teflon* tubing or fittings.
7.5 Commercially Available PDFID System (5)
7.5.1 A convenient and cost-effective modular PDFID system suit-
able for use with a conventional laboratory chromatograph
is commercially available (NuTech Corporation, Model 8548,
2806 Cheek Road, Durham, NC, 27704, (919) 682-0402).
7.5.2 This modular system contains almost all of the apparatus
items needed to convert the chromatograph into a PDFID
analytical system and has been designed to be readily
available and easy to assemble.
Reagents and Materials
8.1 Gas cylinders of helium and hydrogen - ultrahigh purity grade.
8.2 Combustion air - cylinder containing less than 0.02 ppm hydro-
carbons, or equivalent air source.
8.3 Propane calibration standard - cylinder containing 1-100 ppm
(3-300 ppmC) propane in air. The cylinder assay should be
traceable to a National Bureau of Standards (NBS) Standard Refer-
ence Material (SRM) or to a NBS/EPA-approved Certified Reference
Material (CRM).
8.4 Zero air - cylinder containing less than 0.02 ppmC hydrocar-
bons. Zero air may be obtained from a cylinder of zero-grade
compressed air scrubbed with Drierite* or silica gel and 5A
molecular sieve or activated charcoal, or by catalytic cleanup
-------
T012-12
of ambient air. All zero air should be passed througn a liquid
argon cold trap for final cleanup, then passed througn a hyrdo-
carbon-free water bubbler (or other device) for humidification.
8.5 Liquid cryogen - liquid argon (bp -185.7°C) or, liquid oxygen,
(bp -183°C) may be used as the cryogen. Experiments have shown
no differences in trapping efficiency between liquid argon and
liquid oxygen. However, appropriate safety precautions must be
taken if liquid oxygen is used. Liquid nitrogen (bp -195°C)
should not be used because it causes condensation of oxygen and
methane in the trap.
9. Direct Sampling
9.1 For direct ambient air sampling, the cryogenic trapping system
draws the air sample directly from a pump-ventilated distribution
manifold or sampie line (see Figure 1). The connecting line should
be of small diameter (1/8" 0.0.) stainless steel tubing and as
short as possible to minimize its dead volume.
9.2 Multiple analyses over the sampling period must be made to estab-
". sh hourly or 3-hour NMOC concentration averages.
10. Sample Collection in Pressurized Canister(s)
For integrated pressurized canister sampling, ambient air is sampled
by a metal bellows pump through a critical orifice (to maintain
constant flow), and pressurized into a clean, evacuated, Surama*-
polished sample canister. The critical orifice size is chosen so
that the canister is pressurized to approximately one atmosphere above
ambient pressure, at a constant flow rate over the desired sample
period. Two canisters are connected in parallel for duplicate samples.
The canister(s) are then returned to the laboratory for analysis,
using the PDFID analytical system. Collection of ambient air samples
in pressurized canisters provides the following advantages:
o Convenient integration of ambient samples over a specific
time period
o Capability of remote sampling with subsequent central
laboratory analysis
o Ability to ship and store samples, if necessary
-------
T012-13
o Unattended sample collection
o Analysis of samples from multiple sites with one analytical
system
o Collection of replicate samples for assessment of measurement
precision
With canister sampling, however, great care must be exercised in
selecting, cleaning, and handling the sample canister(s) and sampling
apparatus to avoid losses or contamination of the samples.
10.1 Canister Cleanup and Preparation
10.1.1 All canisters must be clean and free of any contaminants
before sample collection.
10.1.2 Leak test all canisters by pressurizing them to approxi-
mately 30 psig [200 kPa (gauge)] with zero air. The
use of the canister cleaning system (see Figure 5) may
be adequate for this task. Measure the final pressure -
close the canister valve, then check the pressure after
24 hours. If leak tight, the pressure should not vary
more than +_ 2 psig over the 24-hour period. Note leak
check result on sampling data sheet, Figure 7.
10.1.3 Assemble a canister cleaning system, as illustrated in
Figure 5. Add cryogen to both the vacuum pump and zero
air supply traps. Connect the canister(s) to the mani-
fold. Open the vent shut off valve and the canister
valve(s) to release any remaining pressure in the canis-
ter. Now close the vent shut off valve and open the
vacuum shut off valve. Start the vacuum pump and evacuate
the canister(s) to <_ 5.0 mm Hg (for at least one hour).
[Note: On a daily basis or more often if necessary, blow-
out the cryogenic traps with zero air to remove any
trapped water from previous canister cleaning cycles.]
10.1.4 Close the vacuum and vacuum gauge shut off valves and
open the zero air shut off valve to pressurize the canis-
ter(s) with moist zero air to approximately 30 psig [200
kPa (gauge)]. If a zero gas generator system is used,
-------
T012-14
the flow rate may need to be limited to maintain the
zero air quality.
10.1.5 Close the zero ;hut off valve and allow canisten's) to
vent down to atmospheric pressure through the vent shut
off valve. Close the vent shut off valve. Rep - -teps
10.1.3 through 10.1.5 two additional times for a total of
three (3) evacuation/pressurization cycles for each set of
canisters.
10.1.6 As a "blank" check of the canister(s) and cleanup proce-
dure, analyze the final zero-air fill of 100% of the
canisters until the cleanup system and canisters are
proven reliable. The check can then be reduced to a
lower percentage of canisters. Any lister that does
not test clean (compared to direct . jlysis of humidified
zero air of less than 0.02 ppmC) should not be utilized.
10.1.7 The canister is then re-evacuated to <_ 5.0 mm Hg, using
the canister cleaning system, and remains in this con-
dition until use. Close the canister valve, remove the
canister from the canister cleaning system and ca.
canister connection with a stainless steel fitting. The
canister is now ready for collection of an air sample.
Attach an identification tag to the neck of each
canister for field notes and chain-of-custody purposes.
10.2 Collection of Integrated Whole-Air Samples
10.2.1 Assemble the sampling apparatus as shown in Figure 2.
The connecti -g lines between the sample pump and the
canister(s) should be as short as possible to minimize
their volume. A second canister is used when a duplicate
sample is desired for quality assurance (QA) purposes
(see Section 12.2.4). The small auxiliary vacuum pump
purges the inlet manifold or lines with a flow of
several L/min to minimize the sample residence time.
The larger metal bellows pump takes a small portion of
this sample to fill and pressurize the sample camster(s)
Both pumps should be shock-mounted to minimize vibration.
Prior to field use, each sampling system should be leak
-------
T012-15
tested. The outlet side of the metal bellows pump can
be checked for leaks by attaching the 0-30 psi g pressure
gauge to the canister(s) inlet via connecting tubing and
pressurizing to 2 atmospheres or approximately 29.4 psig.
If pump and connecting lines are leak free pressure should
remain at ^2 psig for 15 minutes. To check the inlet
side, plug the sample inlet and insure that there is no
flow at the outlet of the pump.
10.2.2 Calculate the flow rate needed so that the canister(s)
are pressurized to approximately one atmosphere above
ambient pressure (2 atmospheres absolute pressure)
over the desired sample period, utilizing the following
equation:
(T)(60)
where:
F = flow rate (cm^/min)
P * final canister pressure (atmospheres absolute)
- (Pg/Pa) + 1
V = volume of the canister (cm^)
N = number of canisters connected together for
simultaneous sample collection
T - sample period (hours)
Pg = gauge pressure in canister, psig (kPa)
Pa = standard atmospheric pressure, 14.7 psig (101 kPa)
For example, if one 6-L canister is to be filled to 2
atmospheres absolute pressure (14.7 psig) in 3 hours,
the flow rate would be calculated as follows:
F * 2 x 6000 x 1 = 67 cm3/min
3 x 60
10.2.3 Select a critical orifice or hypodermic needle suitable
to maintain a substantially constant flow at the cal-
culated flow rate into the canister(s) over the desired
sample period. A 30-gauge hypodermic needle, 2.5 cm
-------
T012-16
long, provides a flow of approximately 65 cm^/min with
the Metal Bellows Model MBV-151 pump (see Figure 4).
Such a needle will maintain approximately constant flow
up to a canister pressure of about 10-psig (71 kPa),
after which the flow drops with increasing pressure.
At 14.7 psig (2 atmospheres absolute pressure), the
flow is about 10% below the original flow.
"10.2.4 Assemble the 2.0 micron stainless steel in-line particu-
late filter and position it in front of the critical
orifice. A suggested filter-hypodermic needle assembly
can be fabricated as illustrated in Figure 4.
10.2.5 Check the sampling system for contamination by filling
two evacuated, cleaned canister(s) (See Section 10.1)
with humidified zero air through the sampling system.
Analyze the canisters according to Section 11.4. The
sampling system is free of contamination if the canisters
contain less than 0.02 ppmC hydrocarbons, similar to
that of humidified zero air.
10.2.6 During the system contamination check procedure, check
the critical orifice flow rate on the sampling system
to insure that sample flow rate remains relatively con-
stant (+10%) up to about 2 atmospheres absolute pressure
(101 kPa). Note: A drop in the flow rate may occur
near the end of the sampling period as the canister
pressure approaches two atmospheres.
10.2.7 Reassemble the sampling system. If the inlet sample line
is longer than 3 meters, install an auxiliary pump to
ventilate the sample line, as illustrated in Figure 2.
10.2.8 Verify that the timer, pump(s) and solenoid valve are
connected and operating properly.
10.2.9 Verify that the timer is correctly set for the desired
sample period, and that the solenoid valve is closed.
10.2.10 Connect a cleaned, evacuated canister(s) (Section 10.1)
to the non-contaminated sampling system, by way of the
solenoid valve, for sample collection.
-------
T012-17
10.2.11 Make sure the solenoid valve is closed. Open the
canister valve(s). Temporarily connect a small rotameter
to the sample inlet to verify that there is no flow.
Note: Flow detection would indicate a leaking (or open)
solenoid valve. Remove the rotameter after leak de-
tection procedure.
10.2.12 Fill out the necessary information on the Field Data
Sheet (Figure 7).
10.2.13 Set the automatic timer to start and stop the pump
or pumps to open and close the solenoid valve at the
appropriate time for the intended sample period.
Sampling will begin at the pre-determined time.
10.2.14 After the sample period, close the canister valve(s) and
disconnect the canister(s) from the sampling system.
Connect a pressure gauge to the canister(s) and briefly
open and close the canister valve. Note the canister
pressure on the Field Data Sheet (see Figure 7). The
canister pressure should be approximately 2 atmospheres
absolute [1 atmosphere or 101 kPa (gauge)]. Note: If
the canister pressure is not approximately 2 atmospheres
absolute (14.7 psig), determine and correct the cause be-
fore next sample. Re-cap canister valve.
10.2.15 Fill out the identification tag on the sample canister(s)
and complete the Field Data Sheet as necessary. Note
any activities or special conditions in the area (rain,
smoke, etc.) that may affect the sample contents on the
sampling data sheet.
10.2.16 Return the canister(s) to the analytical system for
analysis.
11. Sample Analysis
11.1 Analytical System Leak Check
11.1.1 Before sample analysis, the analytical system is assemblec
(see Figure 1) and leak checked.
-------
T012-18
11.1.2 To leak checic the analytical system, place the six-port
gas valve in the trapping position. Disco ~ect and cap
the absolute pressure gauge. Insert a prr ,ure gauge
capable of recording up to 60 psig at th ac i valve
outlet.
11.1.3 Attach a valve and a zero air supply + che sample
inlet port. Pressurize the system t .Dout 50 psig
(350 kPa) and close the valve.
11.1.4 Wait approximately 3 hrs. and re-check pressure. If
the pressure did not vary more than +_ 2 psig, the
system is con dered leak fight.
11.1.5 If the system is leak free, de-pressurize and reconnect
absolute pressure gauge.
11.1.6 The analytical system leak check procedure needs to
be performec luring the system checkou , during a series
of analysis or if leaks are suspected. This should b
part of the user-prepared SOP manual (see Section 12.-
11.2 Sample Volume Determination
11.2.1 The vacuum reservoir and absolute pressure gauge are
used to meter a precisely repeatable volume of sample
air through the cryogemcally- -/oled - -ap, as follows:
With the sample valve closed and the vacuum valve open,
the reservoir is first evacuated with the vacuum pump
to a predetermined pressure (e.g., 100 mm Hg). Then
the vacuum valve is closed and the sample valve is
opened to allow sample air to be drawn through the
cryogenic trap and into the evacuated reservoir until
a second predetermined reservoir pressure is reached
(e.g., 300 mm Hg). The (fixed) volume of air thus
sampled is determined by the pressure rise in the
vacuum reservoir (difference between the predetermined
pressures) as measured by the absolute pressure gauge
(see Section 12.2.1).
-------
T012-19
11.2.2 The sample volume can be calculated by:
V
s (Ps)
where:
Vs = volume of air sampled (standard cm3)
AP = pressure difference measured by gauge (mm Hg)
Vr = volume of vacuum reservoir (cm3)
usually 1 L
Ps = standard pressure (760 mm Hg)
For example, with a vacuum reservoir of 1000 cm3 and a
pressure change of 200 mm Hg (100 to 300 mm Hg), the volume
sampled would be 263 cm3. [Note: Typical sample volume
using this procedure is between 200-300 cm3.]
11.2.3 The sample volume determination need only be performed once
during the system check-out and shall be part of the
user-prepared SOP Manual (see Section 12.1).
11.3 Analytical System Dynamic Calibration
11.3.1 Before sample analysis, a complete dynamic calibration
of the analytical system should be carried out at five or
more concentrations on each range to define the calibra-
tion curve. This should be carried out initially and
periodically thereafter [may be done only once during
a series of analyses]. This should be part of the
user-prepared SOP Manual (See Section 12.1). The
calibration should be verified with two or three-point
calibration checks (including zero) each day the analyt-
ical system is used to analyze samples.
11.3.2 Concentration standards of propane are used to calibrate
the analytical system. Propane calibration standards
may be obtained directly from low concentration cylinder
standards or by dilution of high concentration cylinder
-------
T012-20
standards with zero air (see Section 8.3). Dilution
flow rates must be measured accurately, and the combined
gas stream must be mixed thoroughly for successful cali-
bration of the analyzer. Calibration sxanaaras should
be sampled directly from a vented manifold or tee. Note:
Remember that a propane NMOC concentration in ppmC is
three times the volumetric concentration in ppm.
11.3.3 Select one or more combinations of the foil ing parameters
to provide the desired range or ranges (e.g., 0-1.Q ppmC
or 0-5.0 ppmC): F 0 attsnuator setting, output voltage
setting, integrator resolution (if applicable), and sample
volume. Each individual range should be calibrated sep-
arately and should have a separate calibration curve.
Note: Modern GC integrators may provide automatic '.nging
such that sev -?1 decades of concentration .y be wered
in a single rai.je. The user-prepared SOP manual should
address variations applicable to a specific system design
(see Section 12.1).
11.3.4 Analyze each calibration standard three times according
to the procedure in Section 11.4. Insure that flow
rates, pressure gauge art ' stop readings, initial
cryogen liquid level in the . ar, timing, heating, inte-
grator settings, and other variables are the same as
those that will be used during analysis of ambient
samples. Typical flow rates for the gases are: hydrogen,
30 cm3/minute; helium carrier, 30 cm3/minute; burner
air, 400 cm3/minute.
11.3.5 Average the three analyses for each concentration standard
and plot the calibration curve(s) as average integrated pea*
area reading versus concentration in ppmC. The relative
standard deviation for the three analyses should be less
-------
T012-21
than 3% (except for zero concentration). Linearity should
be expected; points that appear to deviate abnormally
should be repeated. Response has been shown to be linear
over a wide range (0-10,000 ppbC). If nonlinearity is
observed, an effort should be made to identify and correct
the problem. If the problem cannot be corrected, addi-
tional points in the nonlinear region may be needed to
define the calibration curve adequately.
11.4 Analysis Procedure
11.4.1 Insure the analytical system has been assembled properly,
leaked checked, and properly calibrated through a dynamic
standard calibration. Light the FID detector and allow to
stabilize.
11.4.2 Check and adjust the helium carrier pressure to provide the
correct carrier flow rate for the system. Helium is used
to purge residual air and methane from the trap at the
end of the sampling phase and to carry the re-volatilized
NMOC from the trap into the FID. A single-stage auxiliary
regulator between the cylinder and the analyzer may not
be necessary, but is recommended to regulate the helium
pressure better than the multistage cylinder regulator.
When an auxiliary regulator is used, the secondary stage
of the two-stage regulator must be set at a pressure
higher than the pressure setting of the single-stage
regulator. Also check the FID hydrogen and burner air
flow rates (see 11.3.4).
11.4.3 Close the sample valve and open the vacuum valve to
evacuate the vacuum reservoir to a specific predetermined
value (e.g., 100 mm Hg).
11.4.4 With the trap at room temperature, place the six-port
valve in the inject position.
11.4.5 Open the sample valve and adjust the sample flow rate
needle valve for an appropriate trap flow of 50-100
cm-Vmin. Note: The flow will be lower later, when the
trap is cold.
-------
T012-22
11.4,6 Check the sample canister pressure before attaching it
to the analytical system and record on Field Data
Sheet (see Figure 7). Connect the sample canister or
direct sample inlet to the six-port valve, as shown in
Figure 1. For a canister, either the canister valve
or an optional fine needle valve installed between the
canister and the vent is used to adjust the canister
flow rate to a value slightly higher than the trap
flow rate set by the sample flow rate needle valve.
The excess flow exhausts through the vent, which
assures that the sample air flowing through the trap
is at atmospheric pressure. The vent is connected to
a flow indicator such as a rotameter as an indication of
vent flow to assist in adjusting the flow control
valve. Open the canister valve and adjust the canister
valve or the sample flow needle valve to obtain a
moderate vent flow as indic.ateu by the rotsr.eter. The
sample flow rate will be lower (and hence the vent
flow rate will be higher) when the the trap is cold.
11.4.7 . Close the sample valve and open the vacuum valve (if
t a1 ~!y op - to evacuate the vacuum reservoir.
-.h t -.ix-p., . vane in the inject position and the
dcuum valve open, open the sampie valve for 2-3 minutes
[with both valves open, the pressure reading won't
change] to flush and condition the inlet lines.
11.4.8 Close the sample valve and evacuate the reservoir to
the predetermined sample starting pressure (typically
100 mm Hg) as indicated by the absolute pressure gauge.
11.4.9 Switch the six-port valve to the sample pos-'ion.
11.4.10 Submerge the trap in the cryogen. Allow a f • minutes
for the trap to cool completely (indicated wh the
cryogen stops boiling). Add cryogen to the i ial
level used during system dynamic calibration, "he level
of the cryogenic liquid should remain constant with
respect to the trap and should completely cover the
beaded portion of the trap.
-------
T012-23
11.4.11 Open the sample valve and observe the increasing pressure
on the pressure gauge. When it reaches the specific pre-
determined pressure (typically 300 mm Hg) representative
of the desired sample volume (Section 11.2), close the
sample valve.
11.4.12 Add a little cryogen or elevate the Dewar to raise the
liquid le'vel to a point slightly higher (3-15 mm) than
the initial level at the beginning of the trapping.
Note: This insures that organics do not bleed from the
trap and are counted as part of the NMOC peak(s).
11.4.13 Switch the 6-port valve to the inject position, keeping
the cryogenic liquid on the trap until the methane and
upset peaks have deminished (10-20 seconds). Now close
the canister valve to conserve the remaining sample in
the canister.
11.4.14 Start the integrator and remove the Dewar flask containing
the cryogenic liquid from the trap.
11.4.15 Close the GC oven door and allow the GC oven (or alter-
nate trap heating system) to heat the trap at a predeter-
mined rate (typically, 30°C/min) to 90°. Heating the trap
volatilizes the concentrated NMOC such that the FID pro-
duces integrated peaks. A uniform trap temperature rise
rate (above 0°C) helps to reduce variability and facili-
tates more accurate correction for the moisture-shifted
baseline. With a chromatograph oven to heat the trap,
the following parameters have been found to be acceptable:
initial temperature, 30°C; initial time, 0.20 minutes
(following start of the integrator); heat rate, 30°/minute;
final temperature, 90°C.
11.4.16 Use the same heating process and temperatures for both
calibration and sample analysis. Heating the trap too
quickly may cause an initial negative response that
could hamper accurate integration. Some initial exper-
imentation may be necessary to determine the optimal
heating procedure for each system. Once established,
the procedure should be consistent for each analysis
as outlined in the user-prepared SOP Manual.
-------
T012-24
11.4.17 Continue the integration (generally, in the range of
1-2 minutes is adequate) only long enough to include
all of the organic compound peaks and to establish the
end point FID baseline, as illustrated In Figure 8.
The integrator should be capable of marking the begin-
ning and ending of peaks, constructing the appropriate
operational baseline between the start and end of the
integration period, and calculating the resulting
corrected peak area. This ability is necessary because
the moisture in the sample, which is also concentrated
in the trap, will cause a slight positive baseline
shift. This baseline shift starts as the trap warms
and continues until all of the moisture is swept from
the trap, at which time the baseline sturns to its
normal level. The shift always continues longer than
the ambient organic peak(s). The integrator should be
programmed to correct ~ir this shifted baseline by
ending the integrati it a pcint after the last NMOC
peak and prior to the return of the shifted baseline to
normal (see Figure 8) so that the calculated operational
baseline effectively compensates for the water-shifted
baseline. Electronic integrators either do this auto-
matically or they should be programmed to make this cor-
rection. Alternatively, analyses of humidified zero air
prior to sample analyses should be performed to determine
the water envelope and the proper blank value for
correcting the ambient air concentration measurements
accordingly. Heating and flushing of the trap should
continue after the integration period has ended to
insure all water has been removed to prevent buildup of
water in the trap. Therefore, be sure tnat t!~ 5-port
valve remains in the inject position until all .oisture
has purged from the trap (3 minutes or longer).
-------
T012-25
11.4.18 Use the dynamic calibration curve (see Section 11.3)
to convert the integrated peak area reading into
concentration units (ppmC). Note that the NMOC peak
shape may not be precisely reproducible due to vari-
ations in heating the trap, but the total NMOC peak
area should be reproducible..
11.4.19 Analyze each canister sample at least twice and report
the average NMOC concentration. Problems during an
analysis occasionally will cause erratic or incon-
sistent results. If the first two analyses do not
agree within +_ 6% relative standard deviation (RSD),
additional analyses should be made to identify in-
accurate measurements and produce a more accurate
average (see also Section 12.2.).
12. Performance Criteria and Quality Assurance
This section summarizes required quality assurance measures and pro-
vides guidance concerning performance criteria that should be achieved
within each laboratory.
12.1 Standard Operating Procedures (SOPs)
12.1.1 Users should generate SOPs describing and documenting
the following activities in their laboratory: (1)
assembly, calibration, leak check, and operation of the
specific sampling system and equipment used; (2) prepara-
tion, storage, shipment, and handling of samples; (3)
assembly, leak-check, calibration, and operation of the
analytical system, addressing the specific equipment used;
(4) canister storage and cleaning; and (5) all aspects of
of data recording and processing, including lists of
computer hardware and software used.
12.1.2 SOPs should provide specific stepwise instructions and
should be readily available to, and understood by, the
laboratory personnel conducting the work.
-------
T012-26
12.2 Method Sensitivity, Accuracy, r-ecision and Linearity
12.2.1 The sensitivity and precision of the method is proportional
to the sample volume. uowever. ice formation in the
trap may reduce or s~ ;he sv;ple ~ ~.n "during trapping
if the sample volume aeds -0 cm-. Sample volumes
below about 100-150 u.j may cause increased measurement
variability due to dead volume in lines and valves. For
most typica ambient NMOC concentrations, sample vol -.as
in the range of 200-400 cm^ appear to be appropriate.
If a response peak obtained with a 400 cm^ sample is
off scale or exceeds the calibration range, a second
analysis can be cam ' out with a smaller volume. The
actual sample volume .ad need not be accurately known
1f it is precisely repeatable during both calibration
and analysis. Similarly, the actual volume of the
vacuum reservoir need not be accurately known. But the
reservoir volume should be matched to the pressure
range and resolution of the absolute pressure gauge so
that the measurement of the pressure change in the reser-
voir, hence the sample volume, is repeatable within 1%.
A 1000 cm^ vacuum reservoir and a pressure change of
200 mm Hg, measured with the specified pressure gauge,
have provided a sampling precision of +_ 1.31 crn^. A
smaller volume reservoir may be used with a greater
pressure change to accommodate absolute pressure gauges
with lower resolution, and vice versa.
12.2.2 Some FID detector systems associated with laboratory
chromatographs may have autoranging. Others may
provide attenuator control and internal full-scale
output voltage selectors. An appropriate combination
should be chosen so that an adequate output level for
accurate integration is obtained down to the detection
limit; however, the electrometer or integrator must not
be driven into saturation at the upper end of the
calibration. Saturation of the electrometer may be
indicated by flattening of the calibration curve at
-------
T012-27
high concentrations. Additional adjustments of range
and sensitivity can be provided by adjusting the sample
volume used, as discussed in Section 12.2.1.
12.2.3 System linearity has been documented (6) from 0 to 10,000
ppbC.
12.2.4 Some organic compounds contained in ambient air are
"sticky" and may require repeated analyses before they
fully appear in the FID output. Also, some adjustment
may have to be made in the integrator off time setting
to accommodate compounds that reach the FID late in the
analysis cycle. Similarly, "sticky" compounds from
ambient samples or from contaminated propane standards
may temporarily contaminate the analytical system and
can affect subsequent analyses. Such temporary contam-
ination can usually be removed by repeated analyses of
humidified zero air.
12.2.5 Simultaneous collection of duplicate samples decreases
the possibility of lost measurement data from samples
. lost due to leakage or contamination in either of the
canisters. Two (or more) canisters can be filled simul-
taneously by connecting them in parallel (see Figure 2(a))
and selecting an appropriate flow rate to accommodate
the number of canisters (Section 10.2.2). Duplicate (or
replicate) samples also allow assessment of measurement
precision based on the differences between duplicate samples
(or the standard deviations among replicate samples).
13. Method Modification
13.1 Sample Metering System
13.1.1 Although the vacuum reservoir and absolute pressure gauge
technique for metering the sample volume during analysis is
efficient and convenient, other techniques should work also.
13.1.2 A constant sample flow could be established with a vacuum
pump and a critical orifice, with the six-port valve being
switched to the sample position for a measured time period.
-------
T012-28
A gas volume meter, such as a wet test meter, could
also be used to measure the total volume of sample air
drawn through the trap. These alternative techniques
should be tested and evaluated as part of- a user-prepared
SOP manual.
13.2 FID Detector System
13.2.1 A variety of FID detector systems should be adaptable to
the method.
13.2.2 The specific flow rates and necessary modifications for
the helium carrier for any alternative FID instrument
should be evaluated prior to use as part of the user-
prepared SOP manual.
13.3 Range
13.3.1 It may be possible to increase the sensitivity of the
method by increasing the sample volume. However,
limitations may arise such as plugging of the trap by ice.
13.3.2 Any attempt to increase sensitivity should be evaluated
as part of the user-prepared SOP manual.
13.4 Sub-Atmospheric Pressure Canister Sampling
13.4.1 Collection and analysis of canister air samples at sub-
atmospheric pressure is also possible with minor modifi-
cations to the sampling and analytical procedures.
13.4.2 Method TO-14, "Integrated Canister Sampling for Selective
Organics: Pressurized and Sub-atmospheric Collection
Mechanism," addresses sub-atmospheric pressure canister
sampling. Additional information can be found in the
literature (11-17),
-------
T012-29
1. Uses, Limitations, and Technical Basis of Procedures for Quantifying
Relatiqnsnips Between Photochemical Oxidants and Precursors.. EPA-
450/2-77-21a, U.S. Environmental Protection Agency, Research Triangle
Park, NC, Novemoer 1977.
2. Guidance for Collection of Ambient Non-Methane Organic Compound
(NMOC) Data for Use in 1982 Ozone SIP Development, EPA-450/4-80-Q11.
U.S. Environmental Protection Agency, Research Triangle Park, NC,
June 1980.
3. H. B. Singh, Guidance for the Collection and Use of Ambient Hydrocarbons
Species Data 'in Development of Ozone Control Strategies. EPA-45Q/480-Q08.
U.S. Environmental Protection Agency, Research Triangle Park, NC,
April 1980.
4. R. M. Riggin, Technical Assistance Document for Sampling and Analysis
of Toxic Organic Compounds in Ambient Air, EPA-600/483-027, U.S.
Environmental Protection Agency, Research Triangle Park, NC, 1983.
•
5. M. J. Jackson, et^^l_., Technical Assistance Document for Assembly and
Operation of the Suggested Preconcentration Direct Flame lonization
Detection (PDFID) Analytical System, publication scheduled for late
1987; currently available in draft form from the Qualilty Assurance
Division, MD-77, U.S. Environmental Protection Agency, Research
Triangle Park, NC 27711.
6. R. K. M. Jayanty, et.il't Laboratory Evaluation of Non-Methane Organic
Carbon Determination in Ambient Air by Cryogenic Preconcentration and
Flame lonization Detection. EPA-600/54-82-019, U.S. Evironmental Protec-
tion Agency, Research Triangle Park, NC, July 1982.
7. R. 0. Cox, et_jil_., "Determination of Low Levels of Total Non-Methane
Hydrocarbon Content in Ambient Air", Environ. Sci. Techno!., 16_ (1):57,
1982.
8. F. F. McElroy, et_ *}_., A Cryogenic Preconcentration - Direct FID (PDFID)
Method for Measurement of NMOC in the Ambient Air, EPA-600/4-85-063,
U.S. Environmental Protection Agency, Research Triangle Park, NC,
August 1985.
9. F. W. Sexton, et_ aj_., A Comparative Evaluation of Seven Automated
Ambient Non-Methane Organic Compound Analyzers, EPA-600/5482-Q46,
U.S. Environmental Protection Agency, Research Triangle Park, NC,
August 1982.
10. H. G, Richter, Analysis of Organic Compound Data Gathered During 1980
in Northeast Corridor Cities, EPA-450/4-83-017, U.S. Environmental
Protection Agency, Research Triangle Park, NC, April 1983.
-------
T012-30
11. Cox, R. D. "Sample Collection and Analytical Techniques for Volatile
Organics in Air," presented at APCA Speciality Conference, Chicago, II,
March 22-24, 1983.
12. Rasmussen, R. A. and Khalil, M.A.K. " Atmosp- ric Halocarbons:
Measurements and Analyses of Selected Trace cases," Prsc. NATO ASI on
Atmospheric Ozone, 1980, 209-231.
13. Oliver, K. D., Pleil J.D. and McClenny, W.A. "Sample Intergrity of
Trace Level Volatile Or- TIC Compounds in Amb;-^nt Air Stored in
"SUMMA*" Polished •' nir s," accepted for put i cat ion in Atmospheric
Environment as of .nua 1986. Draft available from W. A. McClenny,
MD-44, EMSL, EPA, Researcn Triangle Park, NC 27711.
14. McClenny, W. A. Pleil J.D. Holdren, J.W.; and Smith, R.N.; 1984.
" Automated Cryogenic Preconcentration and Gas Chromatographic
Determination of Volatile Organic Compounds," Anal. Chem. 56:2947.
15. Pleil; J. D. and Oliver, K. D., 1985, "Evaluation of Various Config-
uratio" 3f Nafion Or •: Water Remova1 from Air Samples Prior to
Gas C roatographic S-. ysis". EPA Contract No. 68-02-4035.
16. Olive: ;. D.; Pleil, .id McClenny, W. A.; 1986. "Sample Integrity
of Trace Level Volatile Organic Compounds in Ambient Air Stored in
Summa* Polished Canisters*" Atmospheric Environ^ 20:1403.
17,. Oliver, K. D. Pleil, 0. D., 1985, "Automated Cryogenic Sampling and
Gas Chromatographic Analysis of Ambient Vapor-Phase Organic Compounds:
Procedures and Comparison Tests," EPA Contract No. 68-02-4035,
Research Triangle Park, NC, Northrop Services, Inc. - Environmental
Sciences.
-------
T012-31
PRESSURE
REGULATOR
ABSOLUTE
PRESSURE GAUGE
VACUUM
VALVE
VACUUM
PUMP
PRESSURIZED (EXCESS)
CANISTER
SAMPLE
DIRECT AIR SAMPLING
x^--—»
CANISTER
VALVE
CANSfTER
ROTAWETER
(OPTIONAL FINE
NEEDLE VALVE)
DEWAR
FLASK
GLASS
BEADS
CRYOGENIC
TRAP COOLER
(LIQUID ARGON)
PRESSURE
GAS REGULATOR
PURIFIER
INTEGRATOR
RECORDER
FIGURE 1. SCHEMATIC OF ANALYTICAL SYSTEM FOR
NMOC—TWO SAMPLING MODES
-------
T012-32
SAMPLE
IN
CRITICAL
ORIFICE
AUXILIARY
VACUUM
PUMP
TIMER
SOLENOID
VALVE
METAL
BELLOWS
PUMP
PRESSURE
GAUGE
CANISTER(S)
FIGURE 2. SAMPLE SYSTEM FOR AUTOMATIC COLLECTION
OF 3-HOUR INTEGRATED AIR SAMPLES
-------
T012-33
100K
TIMER
SWITCH
^
o— cr
115 VAC
O— T
PUMP\_
"1
1 Cl I
!^" ! I ™" 1
40Lild, 450 V
R2 100K
^WVV
L c,lh~
RED
fcl
DC
Oi
BLACK
Id .-
^ 1
J 40Litd, 450 V DC 07
I
WHITE
—
MAGNELATCH
SOLENOID
VALVE
n*mtmi MI MM M7 ' OS ^ •
/I °
17K /
H2 f^\
Ci z ( )RELAT
~w ^ X-X
L> 200 «rt TICK
- 200 VON COIL
C
vt
1
^
w
2
^
^
2
1C
\\
BED
BLACK
WHrTE
MAGNELATCH
VALVE
B"dg« Mcww . 200 PflV. 1J A (XCA. SK SI OS or
Owe* Oi ind 0] • 1000 P*V. U A (RCA. SK 3011 or
CWKMOT Ci - MO ut 2SO VOC (Spngu* Atom* TVA 11M or
T C} • 20 ut 400 VOC Men-^oHrUod (Sonou* Atom* TVAN 1«SJ or
. 10.000 onm eeu. IS nw (AMF PMMT ond kuMkdd. KCP S, or
OJ w»R. «M
20 u(
400 Volt
NON.POLARIZED
FIGURE 3[b]. IMPROVED CIRCUIT DESIGNED TO HANDLE POWER INTERRUPTIONS
FIGURE 3. ELECTRICAL PULSE CIRCUITS FOR DRIVING
SKINNER MAGNELATCH SOLENOID VALVE
WITH A MECHANICAL TIMER
-------
T012-34
T SERIES COMPACT, INLINE FILTER
W/2 tan SS SINTERED ELEMENT
r?i
LTT
n
FEMALE CONNECTOR. 0.25 in O.D. TUBE TO
0.25 in FEMALE NPT
HEX NIPPLE. 0.25 in MALE NPT BOTH ENDS
30 GAUGE x 1.0 in LONG HYPODERMIC
NEEDLE (ORIFICE)
FEMALE CONNECTOR, 0 " in O.D. TUBE T
0.25 in FEMALE NfT
THERMOGREEN LBI 6 mm (0.25 in)
SEPTUM (LOW BLEED)
0.25 in PORT CONNECTOR W/TWO 0.25 in NUTS
FIGURE 4. FILTER AND HYPODERMIC NEEDLE
ASSEMBLY FOR SAMPLE INLE" FLOW
CONTROL
-------
T012-35
3-PORT
GAS
VALVE
VACUUM VACUUM PUMP
PUMP SHUT OFF VALVE VENT VALVE
ZERO AIR
SUPPLY
VENT SHUT OFF
VALVE
X
CRYOGENIC
TFIAP
VACUUM SHUT OFF
VALVE
VENT
-•—fcifr
ZERO AIR
SUPPLY
CRYOGENIC
/TRAP
VACUUM
GAUGE
VACUUM GAUGE
SHUT OFF VALVE
VENT SHUT OFF
VALVE
HUMIDIFIER
PRESSURE
GAUGE
ZERO SHUT OFF
VALVE
FLOW
CONTROL
VALVE
VENT SHUT OFF
VALVE
'MANIFOLD
A A A
CANISTER VALVE
SAMPLE CANISTERS
FIGURE 5. CANISTER CLEANING SYSTEM
-------
T012-36
TUBE LENGTH: -30 cm
O.D.: 0.32 cm
I.D.: 021 cm
CRYOGENIC LIQUID LEVEL
60/80 MESH GLASS BEADS
(TO FIT DEWAR)
FIGURE 6. CRYOGENIC SAMPLE TRAP DIMENSIONS
-------
TQ12-37
«c
t—
-• CD
CS LU 3 •— ^
LU — O _J «£
c- a: _i <£ LU
o o u_ o —i
a:
LU
Z
SO
: LU
£3
• Z
O =3
• -> O.
o i— <_j z z:
K •—• o a =3
a. to _i 2: o-
Q. -O
(0 3
OO Z
O)
1/1
-------
T012-2-
NMOC
PEAK
CO
c
CO
LJJ
cc
Q
START
INTEGRATION
END
INTEGRATION
CONTINUED HEATING
OF TRAP
WATER-SHIFTED
BASELINE
1
t
OPERATIONAL BASELINE
CONSTRUCTED BY INTEGRATOR
TO DETERMINE CORRECTED AREA
NORMAL BASELINE
TIME (MINUTES)
FIGURE 8. CONSTRUCTION OF OPERATIONAL BASELINE
AND CORRESPONDING CORRECTION OF
PEAK 1EA
-------
APPENDIX C
1990 NMOC MONITORING PROGRAM SITE DATA
-------
APPENDIX C -- LIST OF FIGURES
Figure Page
C-l Plot of NMOC concentration for Beaumont, TX C-l
C-2 Plot of NMOC concentration for Baton Rouge, LA C-4
C-3 Plot of NMOC concentration for Hartford, CT . . . . - C-8
C-4 Plot of NMOC concentration for Long Island, NY C-ll
C-5 Plot of NMOC concentration for Manhattan, NY C-14
C-6 Plot of NMOC concentration for Newark, NJ C-18
C-7 Plot of NMOC concentration for Plainfield, NJ C-21
cah.!98f
-------
APPENDIX C -- LIST OF TABLES
Table Page
C-l SUMMARY OF THE 1989 NMOC DATA FOR BEAUMONT, TX (BMTX) C-2
C-2 SUMMARY OF THE 1989 NMOC DATA FOR BATON ROUGE, LA (BRLA) ... C-6
C-3 SUMMARY OF THE 1989 NMOC DATA FOR HARTFORD, CT (HTCT) C-9
C-4 SUMMARY OF THE 1989 NMOC DATA FOR LONG ISLAND, NY (LINY) . . . C-12
C-5 SUMMARY OF THE 1989 NMOC DATA FOR MANHATTAN, NY (MNY) C-16
C-6 SUMMARY OF THE 1989 NMOC DATA FOR NEWARK, NJ (NWNJ) C-19
C-7 SUMMARY OF THE 1989 NMOC DATA FOR PLAINFIELD, NJ (PLNJ) .... C-22
cah.!98f
-------
X
-E
co
OS
Q_
O
CO0
CQ
o
O)
o o
p
in
oooooo
o
LO
O1
o
•
o
Oiudd 'OOIAIN
G-l
-------
TABLE C-l. SUMMARY OF THE 1990 NMOC DATA FOR BEAUMONT, TX (BMTX)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
04-Jun-90
05-Jun-90
06-Jun-90
07-Jun-90
08-Jun-90
ll-Jun-90
12-Jun-90
13-Jun-90
14-Jun-90
15-Jun-90
20-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
27-Jun-90
29-Jun-90
02-Jul-90
02-Jul-90
03-Jul-90
04-Jul-90
nS-Tul-90
06-Jul-90
09-Jul-90
09-Jul-90
10-Jul-90
ll-Jul-90
12-Jul-90
13-Jul-90
16-Jul-90
16-Jul-90
17-Jul-90
18-Jul-90
19-Jul-90
20-Jul-90
23-Jul-90
24-Jul-90
25-Jul-90
26-Jul-90
27-Jul-90
30-Jul-90
30-Jul-90
31-Jul-90
Ol-Aug-90
03-Aug-90
06-Aug-90
07-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
14-Aug-90
14-Aug-90
lS-Aug-90
16-Aug-90
21-Aug-90
Julian
Date
Sampled
155
156
157
158
159
162
163
164
165
166
171
173
176
177
178
180
183
183
184
185
186
187
190
190
191
192
193
194
197
197
198
199
200
201
204
205
206
207
208
211
211
212
213
215
218
219
220
221
222
226
226
227
228
233
Sample
ID
Number
1006
1005
1012
1023
1024
1033
1039
1056
1051
1057
1086
1100
1105
1114
1115
1135
1141
1140
1150
1153
1155
1173
1171.
1163
1181
1192
1199
1202
1215
1214
1222
1235
1243
1256
1262
1269
1280
1284
1299
1302
1301
1320
1321
1332
1337
1345
1356
1360
1646
1388
1389
1401
1402
1422
Sample
Canister
Number
659
838
772
690
766
783
676
146
400
685
853
48
878*
658
928
657
630
632
74
885
656
649
838
897
149
618
666
77
648
193
60
762
800
690
360
131
853
663
36
60
717
630
766
649
634
854
651
765
783
711
626
400
875
776
Sample
Pressure
(psif?)
11.8
11.2
11.5
11.5
11.2
11.5
11.5
11.5
11.5
12.0
11.0
10.8
11.2
11.5
1U
113
13.0
13.0
22^
11.1
110
12.1
13.0
13.0
11.5
11.5
11.5
11.0
14.0
16.0
12.0
14.0
11.5
11.0
11.5
11.0
11.0
11 .5
11.0
13.0
13.0
11.5
11 J
20.0
11.5
8.0
11.5
10.0
9.5
11.0
11.0
7.5
7.5
17.0
Analysis
Pressure
(psif!)
13.0
11.0
13.0
11.0
12.0
13.0
13.0
12.0
1ZO
14.3
12.0
9.0
12.0
110
12.0
13.0
14.0
16.0
24.0
12.0
^^0
12.0
16.0
16.0
12.0
12.0
14.0
14.0
13.0
13.0
1ZO
14.0
13.0
13.0
12.0
12.0
12.0
12.0
12.0
14.0
14.0
14.0
12.0
22.0
13.0
9.0
12.0
10.0
10.0
11.0
11.0
8.0
8.0
18.0
Radian
Channel
C
D
B
B
A
C
D
B
C
A
D
D
C
D
B
A
C
C
B
C
C
A
A
A
A
D
D
D
B
A
D
C
C
C
C
D
C
D
C
C
D
C
D
C
D
D
D
D
C
A
B
C
D
D
Mean
NMOC
ppmC
1.002
1.719
1.083
0.558
0.754
1.187
0.883
0.908
1.150
0.999
1.165
0.739
1.520
1.396
0.895
0.613
1.550
1.367
1.938
1.930
1 <7J1
1.610
2.100
2.079
1.770
1.423
1.160
0.901
1.130
0.936
0.961
0.662
2.335
1.514
1.220
1.310
1.415
0.691
1.349
2.030
2.166
1.737
1.523
0.647
1.390
1.415
0.871
1.079
1.810
1.386
1.528
2330
2.580
1372
QAD
NMOC
ppmC
1.166
1.236
0.868
1.323
0.963
1.59
0.669
0.989
C-2
-------
TABLE C-l. SUMMARY OF THE 1990 NMOC DATA FOR BEAUMONT, TX (BMTX)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
22-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
27-Aug-90
28-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
04-Sep-90
05-Sep-90
06-Sep-90
07-Sep-90
10-Sep-90
ll-Sep-90
ll-Sep-90
12-Sep-90
14-Sep-90
17-Sep-90
18-Scp-90
19-Sep-90
20-Sep-90
21-Sep-90
24-Sep-90
25-Sep-90
25-Sep-90
26-Sep-90
27-Sep-90
28-Sep-90
Juli&n
Date
Sampled
234
235
236
239
239
240
241
242
243
247
248
249
250
253
254
254
255
257
260
261
262
263
. 264
267
268
268
269
270
271
Sample
ID
Number
1431
1440
1448
1453
1452
1455
1472
1476
1475
1493
1502
1509
1613
1525
1534
1535
1538
1550
1559
1573
1588
1587
15%
1598
1611
1610
1609
1615
1626
Sample
Canister
Number
91
131
676
607
872
137
684
400
113
776
618
870
780
80
306
685
885
151
833
191
868
164
854
624
111
176
813
833
178
Sample
Pressure
(psig)
16.5
13.0
13.5
18.0
18.0
16.0
17.5
16.0
16.0
18.0
17.5
16.0
17.5
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
16.0
21.0
14.0
14.0
17.0
16.0
16.0
Analysis
Pressure
(PS'g)
17.0
19.0
18.0
19.5
19.0
18.0
18.0
18.0
17.0
18.0
18.0
17.0
18.0
18.0
18.0
18.0
18.0
17.0
17.0
17.0
17.0
16.0
16.0
17.0
15.0
15.0
18.0
17.0
16.0
Radian
Channel
C
B
A
D
C
C
B
D
D
B
D
A
A
C
C
D
D
D
C
B
C
D
D
D
D
A
D
A
C
Mean
NMOC
ppmC
1.981
0.909
1.737
1.469
1386
1391
1320
1.532
1326
2.299
4.283
3.423
2.161
1.523
1.633
1.588
1.145
Z777
4.040
3323
3.466
3.220
3.260
2.195
2.258
1.946
3.836
3.435
4.008
QAD
NMOC
ppmC
3.238
C-3
-------
*=sr*
OOOOOOOOOOOOOOO"1
ooooooooooooooo
Oiudd 'OOIAIN
C-4
-------
O
O
Ln
OO
in
O
OOOOO
Lninin
O1
-------
TABLE C-l SUMMARY OF THE 1990 NMOC DATA FOR BATON ROUGE, LA (BRLA)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
27-Jun-90
28-Jun-90
02-Jul-90
03-Jul-90
05-Jul-90
05-Jul-90
06-Jul-90
10-Jul-90
ll-Jul-90
12-Jul-90
13-Jul-90
16-Jul-90
17-Jui-90
17-Jul-90
18-Jul-90
19-JuI-90
20-JuI-90
23-Jul-90
24-Jui-90
25-Jul-90
26-Jul-90
27-Jul-90
30-Jul-90
31-Jul-90
31-JuI-90
Ol-Aug-90
02-Aug-90
03-Aug-90
06-Aug-90
07-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
13-Aug-90
14-Aug-90
14-Aug-90
15-Aug-90
16-Aug-90
17-Aug-90
20-Aug-90
22-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
28-Aug-90
28-Aug-90
29-Aug-90
Julian
Date
Sampled
169
170
171
172
173
176
177
178
179
183
184
186
186
187
191
192
193
194
197
198
198
199
200
201
204
205
206
207
208
211
212
212
213
214
215
218
219
220
221
222
225
226
226
227
228
229
232
234
235
236
239
240
240
241
Sample
ID
Number
1069
1064
1078
1089
1094
1103
1111
1110
1127
1149
1148
1160
1159
1174
1184
1197
1198
1203
1218
1224
1223
1230
1241
1244
1255
1267
1274
1277
1283
1300
1310
1309
1322
1327
1333
1351
1352
1355
1363
1384
1374
1387
1386
1395
1406
1405
1416
1430
1433
1437
1441
1467
1466
1469
Sample
Canister
Number
762
20
17
796
775
667
607
666
854
678
775
104
20
780
860
676
711
197
651
833
137
780
18
681
675
92
48
874
719
77
762
897
628
838
707
692
792
777
193
789
630
80
833
52
658
674
149
671
656
793
789
626
624
37
Sample
Pressure
(psig)
14.0
13.0
13.0
14.0
14.0
14.5
14.0
15.0
13.0
14.0
14.0
15.0
17.0
15.0
14.0
15.0
14.0
14.0
15.0
17.0
15.0
15.0
12.0
15.0
14.0
14.0
13.0
14.0
15.0
15.0
17.0
17.0
15.0
14.0
15.0
15.0
15.0
30.0
14.0
15.0
36.0
30.0
30.0
14.0
14.0
14.0
14.0
14.0
14.0
30.0
15.0
18.0
18.0
14.0
Analysis
Pressure
fpsig)
115
110
110
13.0
13.0
13.0
13.0
12.0
15.0
13.0
13.0
15.0
15.0
15.0
12.0
14.0
13.0
12.0
14.0
15.0
110
14.0
10.0
13.0
14.0
13.0
14.0
15.0
16.0
16.0
16.0
16.0
13.0
14.0
14.0
15.0
16.0
30.0
15.0
14.0
310
30.0
30.0
15.0
15.0
13.0
16.0
14.0
14.0
30.0
15.0
18.0
18.0
14.0
Radian
Channel
D
D
D
A
D
C
C
D
C
D
C
C
C
B
A
C
C
C
B
D
C
D
C
D
D
D
A
C
D
D
A
B
D
B
D
C
D
C
C
B
D
D
C
D
D
C
D
D
D
B
A
B
A
A
Mean
NMOC
ppmC
0.651
0.838
0.346
0.515
0.444
1.030
1.713
0.495
0.456
0.398
0.382
0313
0320
0.838
U20
3.099
1.031
0.474
0.319
0.279
0327
0.243
0315
0.318
0.207
0.432
1.262
0.367
0383
1.416
0.740
0.766
0.487
0.679
0.989
0.954
0.860
0.154
0.723
1040
0.055
0.504
0.507
0.618
1046
0.663
1.140
1.156
1.785
1.172
0.318
0.789
0.826
1.844
QAD
NMOC
ppmC
0.213
1.767
0.441
0.506
0.867
0.313
0.285
0.275
0.479
0.402
1147
0.537
C-6
-------
TABLE C-2. SUMMARY OF THE 1990 NMOC DATA FOR BATON ROUGE, LA (BRLA)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
30-Aug-90
31-Aug-90
04-Sep-90
05-Sep-90
07-Sep-90
ll-Sep-90
12-Sep-90
12-Sep-90
13-Sep-90
14-Sep-90
17-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
21-Sep-90
24-Sep-90
25-Sep-90
25-Sep-90
26-Sep-90
26-Sep-90
27-S--P-90
28-Sep-90
28-Sep-90
Julian
Date
Sampled
242
243
247
248
250
254
255
255
256
257
260
261
262
263
264
267
268
268
269
269
270
271
271
Sample
ID
Number
1478
1488
1492
1500
1519
1533
1544
1543
1554
1560
1565
1577
1583
1590
1597
1605
1616
1613
1621
1620
1fi27
1632
1631
Sample
Canister
Number
823
766
659
874
131
400
813
928
607
TJf,
842
137
828
162
50
60
762
618
137
17
409
697
30
Sample
Pressure
(PS'R)
14.0
14.0
15.0
14.0
13.0
14.0
17.0
17.0
14.0
18.0
14.0
13.0
14.0
14.0
14.0
14.0
18.0
17.0
17.0
17.0
14.0
17.0
17.0
Analysis
Pressure
(psij?)
14.0
13.0
15.0
14.0
13.0
14.0
17.0
17.0
14.0
14.0
13.0
14.0
12.0
13.0
13.0
14.0
18.0
18.0
18.0
18.0
16.0
17.0
17.0
Radian
Channel
C
B
A
D
D
A
B
A
D
D
D
C
D
D
C
C
B
C
C
D
D
B
A
Mean
NMOC
ppmC
14.255
0.832
2.890
0.778
1.455
0.836
0.277
0342
0.281
0.920
1.191
0.801
0.562
0.983
0.666
0.130
0.299
0.434
1.363
1.296
2.1SS
1.669
1.744
QAD
NMOC
ppmC
14.912
0.78
C-7
-------
o
OE
05
P85
o
O
0>
o o
o in
o o o o o o
o in o m o LO
;
o
o
-CX)
QOidd 'OOIAIN
O
•o
-------
TABLE C-3. SUMMARY OF THE 1990 NMOC DATA FOR HARTFORD, CT (HTCT)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
04-Jun-90
05-Jun-90
06-Jun-90
07-Jun-90
ll-Jun-90
12-Jun-90
13-Jun-90
13-Jun-90
14-Jun-90
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
27-Jun-90
28-Jun-90
28-Jun-90
29-Jun-90
02-Jul-90
03-Jul-90
05-Jul-90
06-Jul-90
09-Jul-90
10-Jul-90
ll-Jul-90
ll-Jul-90
13-Jul-90
16-Jul-90
17-Jul-90
18-Jul-90
19-Jul-90
20-Jul-90
23-Jul-90
24-Jul-90
25-Jul-90
26-Jul-90
27-Jul-90
27-Jul-90
31-JuI-90
Ol-Aug-90
02-Aug-90
03-Aug-90
06-Aug-90
07-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
13-Aug-90
14-Aug-90
15-Aug-90
16-Aug-90
17-Aug-90
Julian
Date
Sampled
155
156
157
158
162
163
164
164
165
169
170
171
172
173
176
177
178
179
179
180
183
184
186
187
190
191
192
192
194
197
198
199
200
201
204
205
206
207
208
208
212
213
214
215
218
219
220
221
222
225
226
227
228
229
Sample
ID
Number
1032
1038
1037
1034
1036
1044
1048
1047
1060
1067
1084
1102
1093
1099
1108
1137
1120
1125
1124
1138
1151
1152
1175
1170
1191
11%
1190
1189
1206
1221
1227
1226
1245
1246
1266
1260
1294
1287
1282
1281
1311
1325
1330
1348
1344
1347
1368
1369
1383
1378
1382
1409
1407
1410
Sample
Canister
Number
635
854
638
878
885
803
193
800
74
719
872
634
925
927 .
815
648
681
36
777
663
651
690
20
684
899
48
686
775
815
813
875
671
807
837
178
789
793
638
666
868
74
636
924
837
49
17
118
684
624
131
60
660
17
823
Sample
Pressure
(psig)
19.0
19.0
18.4
11.5
16.6
19.0
19.0
19.0
18.0
19.0
17.8
25.0
18.0
19.0
18.0
18.0
27.0
17.0
18.0
18.2
18.0
18.0
17.5
18.0
26.0
18.2
18.2
18.2
17.8
18.0
17.8
18.2
18.0
17.7
18.0
18.0
17.8
18.0
18.0
18.0
17.5
18.0
17.5
18.0
17.5
30.0
30.0
18.0
20.0
17.8
18.0
20.0
18.0
17.5
Analysis
Pressure
(psig)
17.0
17.0
17.5
12.0
16.5
18.5
18.0
18.0
17.0
18.0
16.0
23.0
18.0
18.0
18.0
18.0
28.0
17.0
17.0
18.0
19.0
18.0
16.0
18.0
26.0
17.0
17.0
18.0
15.0
18.0
16.0
18.0
17.0
18.0
16.0
18.0
19.0
18.0
18.0
18.0
16.0
19.0
18.0
19.0
18.0
30.0
30.0
19.0
19.0
18.0
17.0
21.0
18.0
19.0
Radian
Channel
D
C
D
D
C
A
B
A
B
C
D
D
C
C
D
C
B
C
D
A
A
C
B
A
C
D
C
D
C
C
C
D
C
D
C
C
C
C
C
D
B
A
A
C
C
D
D
D
C
D
D
C
C
D
Mean
NMOC
ppmC
0.135
0.205
0.293
0.233
0.104
0.097
0.235
0.276
0.184
0.116
0.168
0.226
0.159
0.305
0.272
0.298
0.193
0.254
0.198
0.125
0.089
0.241
0.239
0.119
0.718
0.215
0.275
0.254
0.124
0.152
0.270
0.231
0.443
0.371
0.298
0.109
0.421
0.165
0.173
0.177
0.177
0.243
0.146
0.518
0.133
0.066
0.157
0.147
0.361
0,382
0.106
0.394
0.160
0.445
QAD
NMOC
ppmC
0.207
0.224
0.256
0.272
0.489
0.410
0.446
C-9
-------
TABLE C-3. SUMMARY OF THE 1990 NMOC DATA FOR HARTFORD, CT (HTCT)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
20-Aug-90
22-Aug-90
22-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
04-Sep-90
05-Sep-90
06-Sep-90
06-Sep-90
07-Sep-90
10-Sep-90
ll-Sep-90
12-Sep-90
13-Sep-90
14-Sep-90
17-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
20-Sep-90
21-Sep-90
24-Sep-90
25-Sep-90
26-Sep-90
27-Sep-90
27-Sep-90
28-Sep-90
Julian
Date
Sampled
232
234
234
235
236
239
241
242
243
247
248
249
249
250
253
254
255
256
257
260
261
262
263
263
264'
267
268
269
270
270
271
Sample
ID
Number
1408
1424
1428
1432
1439
1454
1473
1471
1474
1497
1496
1511
1512
1514 .
1541
1536
1551
1558
1557
1586
1585
1591
1594
1595
1612
1617
1636
1635
1633
1634
1637
Sample
Canister
Number
842
20
813
667
618
674
800
193
638
658
20
17
868
137
649
854
658
878
648
780
872
711
171
309
192
626
302
692
628
80
20
Sample
Pressure
(PS'g)
17.8
17.0
17.0
18.0
18.0
18.0
18.0
17.0
18.0
18.0
17.0
17.0
17.0
16.0
18.0
18.0
18.0
17.0
17.0
18.0
18.0
18.0
16.0
16.0
17.0
18.0
17.0
18.0
19.0
19.0
19.0
Analysis
Pressure
(PS'g)
18.0
17.0
18.0
19.0
18.0
18.0
18.0
18.0
18.0
19.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
19.0
18.0
18.0
18.0
18.0
17.0
17.0
18.0
18.0
17.0
19.0
16.0
20.0
20.0
Radian
Channel
D
D
D
C
A
A
A
A
C
C
D
C
D
C
C
B
B
D
C
B
A
C
£
D
A
C
B
D
C
D
C
Mean
NMOC
ppmC
0.408
0.169
0.214
0.141
0.185
0.495
0.357
0.265
0.439
0.180
0.189
0.508
0.439
0.129
0.183
0.506
0.193
0.272
0.116
0.134
0.176
0.509
0.185
0.208
0.738
0.171
0.315
0.140
0.487
0.391
0.558
QAD
NMOC
ppmC
0.192
0.318
0.172
010
-------
"O
0>
O
q
in
o
in
o
o
o
in
CO
o
CD
co
o
in
OUjdd 'OOIAIN
c-ii
-------
TABLE C-4. SUMMARY OF THE 1990 NMOC DATA FOR LONG ISLAND, NY (LINY)
SamplePenod: 6:00a.m. to9:00a.m.
Date
Sampled
05-Jun-90
06-Jun-90
07-Jun-90
08-Jun-90
ll-Jun-90
13-Jun-90
13-Jun-90
14-Jun-90
15-Jun-90
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
26-Jun-90
27-Jun-90
28-Jun-90
29-Jun-90
02-Jul-90
03-Jul-90
OS-Jul-90
06-Jul-90
09-Jul-90
10-Jul-90
10-Jul-90
ll-JUl-90
12-Jul-90
13-Jul-90
16-Jul-90
17-Jui-90
18-Jul-90
19-JuI-90
20-Jul-90
23-Jul-90
24-M-90
24-Jul-90
25-Jul-90
26-Jui-90
27-Jul-90
30-Jul-90
31-Jul-90
Ol-Aug-90
02-Aug-90
03-Aug-90
06-Aug-90
07-Aug-90
07-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
14-Aug-90
17-Aug-90
Julian
Date
Sampled
156
157
158
159
162
164
164
165
166
169
170
171
172
173
176
177
177
178
179
180
183
184
186
187
190
191
191
192
193
194
197
198
199
200
201
204
205
205
206
207
208
211
212
213
214
215
218
219
219
220
221
222
226
229
Sample
ID
Number
1002
1020
1019
1021
1041
1045
1046
1058
1055
1085
1081
1083
1097
1101
1119
1132
1133
1116
1139
1136
1154
1158
1156
1185
1183
1187
1188
1212
1211
1210
1231
1232
1234
1254
1257
1253
1270
1271
1276
1296
1298
1324
1323
1318
1335
1340
1341
1359
1364
1365
1647
1394
1398
1419
Sample
Canister
Number
48
711
927
618
764
813
77
924
91
842
131
897
723
675
783
660
800
118
711
400
765
803
30
719
80
878
626
628
927
697
928
657
607
629
890
20
899
722
765
400
878
667
22
927
678
780
80
776
28
899
890
874
137
800
Sample
Pressure
d»>s)
123
12.0
12.0
12.0
110
17.0
17.0
14.0
11.8
11.5
11.5
11.4
11.8
11.8
11.9
17.0
17.0
11.7
11.9
11.8
11.6
11.8
110
12.0
11.6
16.4
16.2
1ZO
11.6
1Z3
12.2
12.2
8.0
12.1
11.6
115
10.0
10.1
10.9
11.8
11.6
110
11.4
110
110
28.5
1U
17.0
16.5
113
14.0
13.0
13.0
14.0
Analysis
Pressure
(J»ig)
13.5
13.5
11.0
14.0
14.0
16.5
16.5
14.0
13.0
115
110
110
13.0
10.0
110
17.0
17.0
110
13.0
110
13.0
110
110
13.0
110
16.0
16.0
13.0
13.0
14.0
13.0
14.0
8.0
14.0
110
110
9.0
9.0
14.0
110
13.0
14.0
110
110
14.0
28.0
110
17.0
16.5
110
15.0
13.0
13.0
15.0
Radian
Channel
C
A
B
A
B
A
B
B
A
C
D
D
C
C
A
B
A
A
B
B
C
D
D
A
B
B
B
D
C
D
C
D
D
C
D
D
A
B
B
A
A
D
C
C
C
D
D
C
D
C
D
C
C
A
Mean
NMOC
ppmC
0.796
0.660
0333
0.401
0.159
0.411
0.410
0.189
0.644
0.159
0.402
0.198
0.494
0.851
0.651
0382
0389
0.389
0.289
0.207
0.119
0.419
0.359
0.116
0.216
0355
0320
0.150
0.171
0.112
0.476
0.584
0.700
0.410
0.506
0.230
0.285
0338
0.104
0.224
0.171
0.100
0.609
0.071
0302
0.157
0.088
0319
0.242
0.193
0.210
0.135
0.223
0.598
QAD
NMOC
ppmC
0.745
0.420
0.495
0.252
0.968
0317
0.267
0.160
0.615
0.534
0.195
0.162
0.132
C-12
-------
TABLE C-4. SUMMARY OF THE 1990 NMOC DATA FOR LONG ISLAND, NY (LINY)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
ZO-Aug-90
21-Aug-90
21-Aug-90
22-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
28-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
04-Sep-90
05-Sep-90
05-Sep-90
06-Sep-90
07-Sep-90
10-Sep-90
ll-Sep-90
12-Sep-90
13-Sep-90
U-Seo-90
17-Sep-90
18-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
21-Sep-90
24-Scp-90
25-Sep-90
26-Sep-90
27-Sep-90
28-Sep-90
Julian
Date
Sampled
232
233
233
234
235
236
239
240
241
242
243
247
248
248
249
250
253
254
255
256
257
260
261
261
262
263
264
267
268
269
270
271
Sample
ID
Number
1420
1426
1456
1465
1423
1443
1447
1483
1480
1482
1481
1505
1503
1504
1510
1539
1542
1540
1574
1568
1570
1582
1571
1572
1607
1608
1606
1603
1645
1638
1639
1640
Sample
Canister
Number
878
853
780
172
36
870
649
28
651
854
833
60
671
629
842
635
77
783
28
638
178
91
690
899
198
400
66
106
671
635
674
112
Sample
Pressure
(ps'l?)
13.8
17.0
17.0
143
13.0
17.0
14.0
22.0
13.8
13.5
13.2
14.0
23.5
23.0
16.0
13.5
28.0
14.0
13.0
14.0
14.0
1ZO
19.0
19.0
13.5
13.5
14.5
13.0
15.0
14.0
14.0
14.0
Analysis
Pressure
(f»'g)
11.0
18.0
16.0
14.0
14.0
18.0
14.0
22.0
14.0
14.0
14.0
14.0
24.0
16.0
16.0
14.0
26.0
15.0
14.0
14.0
14.0
12.0
19.0
20.0
13.0
14.0
14.0
13.0
15.0
14.0
12.0
14.0
Radian
Channel
B
D
B
B
C
D
B
C
D
D
C
D
D
C
A
D
D
C
C
B
A
C
D
C
D
C
C
D
D
C
D
C
Mean
NMOC
ppmC
0.128
0.060
0.088
0.848
0.153
0.170
0.551
1.710
0.236
0.190
1.190
0.133
1.371
1.392
0.743
0.162
0.298
1.200
0.427
0398
0.245
0.460
0.228
0.246
0.575
0.190
0.465
0.281
0.916
0.195
0.457
1.867
QAD
NMOC
ppmC
1.308
C-13
-------
"D
0
05
Q.
X
05
05
05
2
G)
o
o
o
OOOOOOOOOOOOOO'
iiininininL
Oiudd 'OOIAIN
C-14
-------
c
CO
i
eof
o
O)
in
CM CM i-
Qtudd 'OOIAIN
C-15
-------
TABLE C-5. SUMMARY OF THE 1990 NMOC DATA FOR MANHATTAN, NY (MNY)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
04-Jun-90
05-Jun-90
06-Jun-90
07-Jun-90
08-Jun-90
ll-Jun-90
ll-Jun-90
13-Jun-90
14-Jun-90
15-Jun-90
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
26-Jun-90
27-Jun-90
28-Jun-90
29-Jun-90
02-Jul-90
03-Jul-90
05-Jul-90
06-Jul-90
09-Jul-90
09-Jul-90
10-Jul-90
ll-Jul-90
12-Jul-90
13-Jul-90
16-Jul-90
17-Jul-90
18-Jul-90
19-Jul-90
20-M-90
23-Jul-90
23-Jul-90
24-Jul-90
25-Jul-90
26-Jul-90
27-Jul-90
30-Jul-90
31-Jul-90
Ol-Aug-90
02-Aug-90
03-Aug-90
07-Aug-90
06-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
13-Aug-90
14-Aug-90
15-Aug-90
16-Aug-90
Julian
Date
Sampled
155
156
157
158
159
162
162
164
165
166
169
170
171
172
173
177
178
179
180
183
184
186
187
190
190
191
192
193
194
197
198
199
200
201
204
204
205
206
207
208
211
212
213
214
215
217
218
220
221
222
225
226
227
228
Sample
ID
Number
1003
1072
1013
1015
1022
1026
1027
1043
1049
1059
1071
1077
1082
1087
1098
1113
1117
1126
1134
1147
1167
1186
1177
1180
1178
1182
1201
1200
1209
1228
1233
1236
1239
1247
1259
1258
1264
1272
1288
1291
1303
1314
1319
1331
1334
1346
1336
1649
1366
1372
1373
1392
1396
1404
Sample
Canister
Number
872
672
853
724
651
875
677
30
780
36
178
807
921
638
49
697
875
924
624
764
723
17
677
872
717
500
624
722
868
7%
704
636
775
723
660
146
711
624
684
91
813
928
875
626
638
674
146
671
878
870
648
607
766
637
Sample
Pressure
(psig)
13.0
13.0
13.0
14.0
14.0
17.0
17.0
13.0
13.0
12.0
14.0
13.0
12.5
4 13.0
12.0
14.0
14.0
14.0
14.0
14.0
16.0
20.0
13.0
15.0
15.0
12J
14.0
14.0
13.0
14.0
14.0
13.0
13.0
13.0
16.0
16.0
9.0
12.0
12.0
10.0
11.0
9.0
12.3
13.0
13.0
13.0
15.0
14.0
13.0
14.0
14.0
13.0
13.0
13.0
Analysis
Pressure
(psig)
12.0
12.0
11.5
12J
12.0
15.0
15.0
12.0
11.5
13.0
12J
1Z5
12.0
12.5
11.0
12.0
16.0
13.0
13.0
13.0
15.0
18.0
110
16.0
14.0
1ZO
14.0
13.0
13.0
13.0
10.0
13.0
13.0
15.0
14.0
14.0
9.0
9.0
13.0
10.0
12.0
8.0
15.0
14.0
14.0
14.0
14.0
14.0
12.0
14.0
14.0
14.0
15.0
110
Radian
Channel
B
C
A
C
B
C
D
B
C
A
C
A
C
B
D
C
B
D
A
D
B
A
B
C
C
B
D
C
C
D
C
D
C
C
D
C
C
A
D
B
D
B
D
B
C
C
D
D
D
B
C
C
C
D
Mean
NMOC
ppmC
0.634
0.202
0.549
0.756
0.457
0.382
0.327
0.331
0.664
0.979
6.200
0.474
0.294
0.476
1.203
0.488
0.443
1.210
0.418
0.248
0.890
0.469.
0.194
0.350
0.379
0.225
0.338
0.358
0.186
0.340
0.514
1.084
0.485
0.720
0.416
0.441
0.396
0.284
0.291
0.322
0.387
0.565
0.225
0.418
0.700
0.486
0.278
0.623
0.692
0.882
1.230
0.838
0.767
3.120
QAD
NMOC
ppmC
0.403
0.518
0.496
1.421
0.435
0.314
0.490
0.384
0.362
0.645
0.449
0.348
0.700
0.832
3.130
C-16
-------
TABLE C-5. SUMMARY OF THE 1990 NMOC DATA FOR MANHATTAN, NY (MNY)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
17-Aug-90
20-Aug-90
20-Aug-90
21-Aug-90
22-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
28-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
04-Sep-90
04-Sep-90
05-Sep-90
06-Scp-90
07-Sep-90
10-Sep-90
ll-Sep-90
12-Sep-90
13-S*p-90
14-Sep-90
17-Sep-90
18-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
Julian
Date
Sampled
229
232
232
233
234
235
236
239
240
241
242
243
247
247
248
249
250
253
254
255
256
257
260
261
261
262
263
Sample
ID
Number
1414
1417
1418
1421
1427
1462
1438
1445
1464
1470
1479
1477
1494
1495
1501
1508
1520
1527
1537
1545
1552
1561
1564
1580
1579
1589
1593
Sample
Canister
Number
685
666
193
783
690
704
711
717
678
878
685
54
. 885
91
648
690
684
113
828
111
618
60
629
783
649
684
77
Sample
Pressure
(PS'g)
13.0
15.0
15.0
13.0
13.0
14.0
13.0
13.0
13.0
12.0
13.0
13.0
15.0
15.0
14.0
14.0
13.0
13.0
13.0
13.0
13.0
12.0
6.0
15.0
15.0
13.0
14.0
Analysis
Pressure
(psi*)
15.0
15.0
15.0
12.0
15.0
14.0
13.0
14.0
13.0
110
14.0
13.0
16.0
16.0
14.0
25.0
13.0
13.0
13.0
13.0
13.0
12.0
13.0
15.0
15.0
14.0
12.0
Radian
Channel
B
C
D
C
C
A
A
A
C
B
D
C
C
D
C
C
C
B
C
C
A
C
C
B
A
C
A
Mean
NMOC
ppmC
1.120
0318
0.250
0346
0.657
0.418
0.403
1367
1.545
1.175
0.499
2.033
0.547
0.470
0.611
1.239
0.948
1.698
1.542
0.729
0.857
0.936
4.125
0.427
0.411
0.934
0335
QAD
NMOC
ppmC
0.862
-------
2
D)
_ o
CO CL
0
o
O)
o o
o in
o o
o to
o o
O LO
CO
.0
"cO
CD
C
O
O
QL
CD
6
D)
LL
OUidd 'OOIAIN
C-18
-------
TABLE C-6. SUMMARY OF THE 1990 NMOC DATA FOR NEWARK, NJ (NWNJ)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
04-Jun-90
OS-Jun-90
06-Jun-90
07-Jun-90
08-Jun-90
ll-Jun-90
12-Jun-90
13-Jun-90
U-Jun-90
14-Jun-90
lS-Jun-90
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
28-Jun-90
28-Jun-90
29-Jun-90
02-Jul-90
03-Jul-90
05-Jul-90
06-Jul-90
09-Jul-90
10-Jul-90
ll-Jul-90
12-Jul-90
12-Jul-90
13-Jul-90
16-Jui-90
19-Jul-90
20-Jul-90
23-Jul-90
24-Jul-90
25Jul-90
26-Jul-90
, 26-Jul-90
27-Jul-90
30-Jul-90
31Jul-90
Ol-Aug-90
02-Aug-90
03-Aug-90
06-Aug-90
OS-Aug-90
09-Aug-90
09-Aug-90
10-Aug-90
13-Aug-90
14-Aug-90
15-Aug-90
16-Aug-90
Julian
Date
Sampled
155
156
157
158
159
162
163
164
165
165
166
169
170
171
172
173
176
177
179
179
180
183
184
186
187
190
191
192
193
193
194
197
200
201
204
205
206
207
207
208
211
212
213
214
215
218
220
221
221
222
225
226
227
228
Sample
ID
Number
1010
1008
1007
1014
1025
1028
1040
1052
1061
\062
1070
1075
1076
1091
1107
1090
1106
1109
1128
1129
1144
1143
1157
1176
1164
1169
1195
1194
1204
1205
1225
1229
1275
1248
1263
1240
1273
1285
1286
1305
1295
1312
1326
1338
1339
1648
1371
1370
1385
1380
1390
1391
1400
1412
Sample
Canister
Number
717
666
899
658
667
686
649
792
22
149
870
707
830
838
104
626
137
686
113
146
628
500
921
.796
762
178
131
635
853
638
692
632
707
191
864
777
815
764
672
676
925
30
803
697
704
815
864
681
690
807
928
657
629
659
Sample
Pressure
(psig)
14.1
15.1
15.0
143
15.0
14.4
15.0
14.5
183
18.3
14.0
15.9
13.6
14.0
14.5
15.0
15.4
15.0
13.2
13.0
14.0
19.5
13.0
17.8
15.0
13.9
14.2
16.0
18.1
18.1
12.4
14.0
15.5
14.0
14.1
15.5
14.4
14.0
14.0
15.2
17.0
14.0
13.4
14.0
20.2
17.0
14.0
19.0
19.0
143
14.8
17.0
16.2
14.5
Analysis
Pressure
(psiR)
16.5
18.0
17.0
16.5
17.0
17.0
17.5
17.0
20.0
20.0
12.0
18.5
16.0
13.0
17.5
14.0
18.0
18.0
15.0
15.0
16.0
22.0
14.0
12.0
17.0
14.0
16.0
15.0
17.0
17.0
15.0
15.0
16.0
14.0
16.0
18.0
16.0
14.0
14.0
18.0
19.0
15.0
16.0
17.0
21.0
17.0
19.0
21.0
21.0
15.0
19.0
20.0
16.2
17.0
Radian
Channel
B
B
D
D
D
C
A
D
C
D
D
A
B
A
A
B
D
C
D
D
D
C
C
A
B
B
C
D
C
D
C
C
B
D
D
D
B
C
C
C
D
B
B
C
C
C
A
B
A
D
C
D
B
D
Mean
NMOC
ppmC
0.409
0.283
0.634
0.871
0.785
0.163
0.253
0.110
1.034
1.190
0.812
0.621
0.199
0.216
0.486
0.929
0.326
0.617
0.273
0.244
0376
0.273
0318
0.257'
0.231
0.243
0.410
0370
0.505
0.435
0.428
0.243
0.148
0.527
0.560
0.516
0.492
0.169
0.220
0.287
0.184
0.737
0.131
0330
0.573
0329
0.432
0.202
0.222
0.585
0.614
0.262
0312
0.627
QAD
NMOC
ppmC
0.635
0.211
0304
0.195
0.392
0.591
0375
C-L9
-------
TABLE C-6. SUMMARY OF THE 1990 NMOC DATA FOR NEWARK, NJ (NWNJ)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
17-Aug-90
20-Aug-90
21-Aug-90
22-Aug-90
23-Aug-90
23-Aug-90
24-Aug-90
27-Aug-90
28-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
04-Sep-90
05-Sep-90
06-Sep-90
07-Sep-90
07-Sep-90
10-Sep-90
ll-Scp-90
12-Sep-90
13-Sen-QO
14-Sep-90
17-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
21-Sep-90
21-Sep-90
24-Sep-90
25-Sep-90
26-Sej>-90
26-Sep-90
27-Sep-90
27-Sep-90
28-Sep-90
28-Sep-90
Julian
Date
Sampled
229
232
233
234
235
235
236
239
240
241
242
243
247
248
249
250
250
253
254
255
256
257
260
261
262
263
264
264
267
268
269
269
270
270
271
271
Sample
ID
Number
1413
1415
1425
1429
1434
1435
1446
1463
1461
1457
1484
1487
1499
1516
1515
1521
1522
1529
1549
1556
1555
1567
1566
1569
1575
1584
1601
1602
1618
1619
1629
1630
1622
1623
1643
1644
Sample
Canister
Number
145
146
890
924
681
899
663
630
722
660
777
36
924
22
704
80
674
624
20
35
172
93
193
301
676
885
678
54
695
776
707
7%
629
28
104
632
Sample
Pressure
(psig)
11.2
12.4
14.3
14.1
18.5
1&5
13.0
123
12.0
12.0
12.0
13.5
8.1
114
16.0
19.0
19.0
0.1
14.9
13.2
14.4
13.9
15.9
16.3
17.2
15.7
18.0
18.0
17.0
17.3
18.6
18.6
16.0
187
18.0
18.0
Analysis
Pressure
(psig)
12.0
14.0
15.0
15.0
22.0
22.0
16.0
110
110
14.0
12.0
16.0
110
16.0
19.0
210
210
14.0
18.0
16.0
18.0
16.0
19.0
19.0
20.0
18.0
18.0
18.0
20.0
16.0
21.0
20.0
18.0
21.0
18.0
18.0
Radian
Channel
A
C
C
C
C
D
D
D
C
D
C
A
C
C
D
D
C
C
A
D
C
C
A
D
D
C
D
C
B
A
B
C
A
B
C
D
Mean
NMOC
ppmC
0.544
0.302
0.835
0.273
0.411
0.320
0.315
0.366
0.464
0.266
0.283
0.355
0.171
0.289
0.758
0.531
0.569
0.658
0.385
0.988
OJ550
0.489
0.193
0.315
0.670
0.295
0.660
0.677
0.413
0.753
0.211
0.541
0.728
0.808
1033
1046
QAD
NMOC
ppmC
0.405
0.557
0.592
C-20
-------
O)
II
= O
o)
O)
o o
O LO
o o o o
o LO o in
o o o o
o in o in
Qtudd 'OOIAIN
C-21
-------
TABLE C-7. SUMMARY OF THE 1990 NMOC DATA FOR PLAINFIELD, NJ (PLNJ)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
04-Jun-90
05-Jun-90
06-Jun-90
07-Jun-90
08-Jun-90
ll-Jun-90
12-Jun-90
13-Jun-90
14-Jun-90
15-Jun-90
15-Jun-90
18-Jun-90
19-Jun-90
20-Jun-90
21-Jun-90
22-Jun-90
25-Jun-90
26-Jun-90
27-Jun-90
28-Jun-90
29-Jun-90
29-Jun-90
02-Jul-90
03-Jul-90
05-Jui-90
06-Jul-90
09-JuI-90
10-Jul-90
ll-Jul-90
12-Jul-90
13-Jul-90
13-Jul-90
16-Jul-90
17-Jul-90
18-Jul-90
19-Jul-90
20-Jul-90
23-Jul-90
24-Jul-90
25-Jul-90
26-Jul-90
27-Jui-90
27-Jul-90
30-Jul-90
31-Jul-90
Ol-Aug-90
02-Aug-90
03-Aug-90
06-Aug-90
07-Aug-90
08-Aug-90
09-Aug-90
10-Aug-90
10-Aug-90
Julian
Date
Sampled
155
156
157
158
159
162
163
164
165
166
166
169
170
171
172
173
176
177
178
179
180
180
183
184
186
187
190
191
192
193
194
194
197
198
199
200
201
204
205
206
207
208
208
211
212
213
214
215
218
219
220
221
222
222
Sample
ID
Number
1011
1009
1004
1018
1030
1029
1035
1050
1042
1065
1066
1068
1063
1079
1080
1088
1104
1112
1122
1123
1145
1146
1142
1172
1168
1165
1166
1179
1193
1216
1219
1220
1217
1238
1242
1237
1261
1268
1278
1279
1297
1292
1293
1304
1308
1367
1379
1349
1350
1362
1361
1376
1375
1377
Sample
Canister
Number
617
837
634
897
657
883
113
80
630
624
656
672
659
899
54
690
677
676
91
671
193
724
722
49
659
766
672
925
400
667
634
783
630
724
28
49
854
104
674
84
686
883
921
632
7%
618
113
149
724
104
675
872
500
74
Sample
Pressure
(PS'g)
18.2
18.5
19.0
18.0
19.2
18.0
18.2
18.2
18.6
15.9
15.9
18.0
18.0
17.6
17.7
18.0
18.2
18.2
17.9
18.3
14.8
14.8
183
183
19.0
17.2
17.6
17.8
18.0
18.0
14.4
14.4
17.9
18.0
18.3
17.4
17.8
17.2
16.8
17.0
18.0
14.0
14.0
17.9
17.4
17.4
20.0
17.2
19.8
14.4
17.5
173
13.9
13.8
Analysis
Pressure
(f»'g)
16.0
17.5
17.0
16.0
17.0
16.0
17.0
17.0
15.5
17.0
17.0
17.0
16.5
15.0
16.0
16.5
17.0
18.0
17.0
17.0
15.0
15.0
17.0
17.0
19.0
17.5
17.0
17.0
18.0
18.0
15.0
18.0
16.0
18.0
16.0
18.0
18.0
18.0
18.0
16.0
18.0
110
8.0
16.0
16.0
19.0
21.0
18.0
20.0
17.0
19.0
18.0
14.0
14.0
Radian
Channel
A
A
C
C
C
D
C
D
A
A
B
C
C
C
C
B
D
D
C
D
R
A
D
B
A
B
A
D
C
A
A
B
B
D
D
C
D
C
D
D
D
D
C
B
A
A
C
D
D
D
C
D
C
C
Mean
NMOC
ppmC
0381
0.746
Z074
0.476
0.836
0.069
0.106
0.636
0387
0.378
0.399
0.196
0.241
0.159
0317
0.862
0313
0.561
0309
0360
0.516
0.516
0.161
0.285
0.173
0.244
0.208
0.250
0.188
0.433
0.194
0.175
0.156
0.443
0.742
0.466
0.502
0363
0.163
0.162
0.256
0.289
0.286
0339
1.165
0.143
0.061
0.899
0.066
0.102
0301
0.308
0.464
0.454
QAD
NMOC
ppmC
0.481
0.137
0.439
0.393
0.409
0312
0.248
0.216
0.292
0.106
C-22
-------
TABLE C-7. SUMMARY OF THE 1990 NMOC DATA FOR PLAINFIELD, NJ (PLNJ)
Sample Period: 6:00 a.m. to 9:00 a.m.
Date
Sampled
13-Aug-90
14-Aug-90
15-Aug-90
16-Aug-90
17-Aug-90
20-Aug-90
21-Aug-90
22-Aug-90
23-Aug-90
24-Aug-90
24-Aug-90
27-Aug-90
28-Aug-90
29-Aug-90
30-Aug-90
31-Aug-90
(M-Sep-90
05-Sep-90
06-S«p-90
07-Sep-90
10-Sep-90
lO-Sep-90
ll-Sep-90
12-Sep-90
13-Sep-90
14-Scp-90
17-Sep-90
18-Sep-90
19-Sep-90
20-Sep-90
21-Sep-90
24-Sep-90
26-Sep-90
27-Sep-90
27-Sep-90
28-Sep-90
28-Sep-90
Julian
Date
Sampled
225
226
227
228
229
232
233
234
235
236
236
239
240
241
242
243
247
248
249
250
253
253
254
255
256
257
260
261
262
263
264
267
269
270
270
271
271
Sample
ID
Number
1381
1397
1399
1403
1411
1442
1449
1444
1436
1450
1451
1459
1460
1458
1491
1490
1489
1506
1507
1526
1523
1524
1546
1548
1547
1553
1581
1576
1578
1592
1604
1614
1628
1624
1625
1641
1642
Sample
Canister
Number
868
860
22
793
54
82
837
765
118
74
500
657
928
87*5
666
853
634
890
717
667
626
54
651
302
114
112
17
870
667
113
800
685
607
927
925
675
807
Sample
Pressure
(psig)
17.0
17.2
20.1
17.6
17.1
17.9
17.8
17.7
17.0
14.1
14.1
18.0
18.0
18.0
17.5
17.1
18.1
17.2
17.4
17.4
10 .5
13.9
17.7
17.7
16.2
16.7
16.8
19.4
19.0
18.9
19.0
19.0
18.1
14.4
14.4
14.0
14.0
Analysts
Pressure
(f»'g)
18.0
18.0
20.0
19.0
18.0
18.0
18.0
20.0
17.0
14.5
14.5
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
1ZO
14.0
18.0
17.0
18.0
17.0
17.0
20.0
19.5
18.0
19.0
19.0
18.0
14.0
14.0
14.0
14.0
Radian
Channel
C
D
A
C
C
C
C
B
C
A
B
C
D
A
A
B
A
C
D
D
A
B
B
C
D
C
D
B
D
B
D
B
A
C
D
D
C
Mean
NMOC
ppmC
1.040
0.400
0.012
1.440
0.682
0.201
0.128
0.210
0.282
0.297
0.164
0.278
0.387
0.170
0.635
0.250
0.141
0.198
0.779
0375
0.252
0.214
1.550
0.346
1.089
0.9%
0.057
0.321
1.089
0.226
0.822
0380
1325
1.189
1.092
2363
2364
QAD
NMOC
ppmC
0.168
C-23
-------
APPENDIX D
1990 NMOC MONITORING PROGRAM
INVALIDATED AND MISSING SAMPLES
-------
APPENDIX D
1990 NMOC PROGRAM
VOID OR INVALIDATED SAMPLES - DATE
*
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Site
HTCT
MNY
LINY
BMTX
BMTX
MNY
NUNJ
BMTX
BRLA
LINY
BRLA
NUNJ
BMTX
NUNJ
BMTX
LINY
LINY
BMTX
BMTX
HTCT
BRLA
HTCT
BRLA
BMTX
PLNJ
HTCT
Date
OS-Jun-90
12-Jun-90
12-Jun-90
18-Jun-90
18-JW-90
25-Jun-90
27-Jun-90
28-Jun-90
29-Jun-90
04-Jul-90
09-Jul-90
17-Jul-90
02-Aug-90
07-Aug-90
10-Aug-90
13-Aug-90
15-Aug-90
17-Aug-90
20-Aug-90
21-Aug-90
21-Aug-90
28-Aug-90
06-Sep-90
13-Sep-90
25-Sep-90
28-Sep-90
Description
Power outage
System leak
Double orifice not installed
Valve open when disconnecting
Valve open when disconnecting
Valve open when disconnecting
Canister leak
Power Outage
Can valve not opened
Valve open when disconnecting
Power outage
No pressure
No sample collected
Can valve not opened
Canister not evacuated
Can valve not opened
Timer mi sprog rammed
Can valve not opened
System leak
Can valve not opened
Power outage
High pressure-double orifice used
Timer misprogranmed
Can valve not opened
Can valve not opened
Valve opened when disconnecting
Assigned
Equipment
Equipment
Operator
Operator
Operator
Operator
Equipment
Equipment
Operator
Operator
Equipment
Operator
Operator
Operator
Radian
Operator
Operator
Operator
Equipment
Operator
Equipment
Operator
Operator
Operator
Operator
Operator
-------
APPENDIX 0
1990 NMOC PROGRAM
VOID OR INVALIDATED SAMPLES - SITE
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Site
BMTX
BMTX
BMTX
BMTX
BMTX
BMTX
BMTX
BMTX
BRLA
BRLA
BRLA
BRLA
HTCT
HTCT
HTCT
HTCT
LINY
LINY
LINY
LINY
MNY
MNY
NUNJ
NWNJ
NWNJ
PLNJ
Date
18-Jun-90
18-Jun-90
28-Jun-90
02-Aug-90
10-Aug-90
17-Aug-90
20-Aug-90
13-Sep-90
29-Jun-90
09-Jul-90
21-Aug-90
06-Sep-90
08-Jun-90
21-Aug-90
28-Aug-90
28-Sep-90
12-Jun-90
04-Jul-90
13-Aug-90
15-Aug-90
12-Jun-90
25-Jun-90
27-Jun-90
17-Jul-90
07-Aug-90
25-Sep-90
Description
Valve open when disconnecting
Valve open when disconnecting
Power Outage
No sample collected
Canister not evacuated
Can valve not opened
System leak
Can valve not opened
Can valve not opened
Power outage
Power outage
Timer misprogrammed
Power outage
Can valve not opened
High pressure-double orifice used
Valve opened when disconnecting
Double orifice not installed
Valve open when disconnecting
Can valve not opened
Timer misprogrammed
System leak
Valve open when disconnecting
Canister leak
No pressure
Can valve not opened
Can valve not opened
Assigned
Operator
Operator
Equipment
Operator
Radian
Operator
Equipment
Operator
Operator
Equipment
Equipment
Operator
Equipment
Operator
Operator
Operator
Operator
Operator
Operator
Operator
Equipment
Operator
Equipment
Operator
Operator
Operator
-------
APPENDIX E
PDFID INTEGRATOR PROGRAMMING INSTRUCTIONS
-------
INTEGRATOR PROGRAMMING INSTRUCTIONS
Instructions for programming the integrators are as foljows.
Be sure to press ENTER after each key sequence.
Control Integrator
Oven Temp 90
Oven Temp Limit 405
Oven Temp ON
Oven Temp OFF
List Oven Temp
(A listing should say OvenTemp X°C Setpt 90°C Limit 405°C)
Oven Temp Initial Time 0.20
Oven Temp Initial Value 90
Oven Temp Pgrm Rate 30.00
Oven Temp Final Value 90.00
Oven Temp Final Time 4.00
Oven Temp Equil Time 1.00
Detector A ON
Signal A
Chart speed 4.00
%0ffset 10
Zero
Attn 2A 4
Run Time Annotation ON
Run Table Annotation ON
Clock Table Annotation OFF
Program Annotation OFF
Oven Temp Annotation OFF
Report Annotation OFF
Slave Integrator
Detector B ON
Signal B
Chart speed 4.00
%0ffset 10
Zero
Attn 2A 4
Run Time Annotation ON
Run Table Annotation ON
Clock Table Annotation OFF
Program Annotation OFF
Oven Temp Annotation OFF
(should say ***Warning***Oven Temp now owned by Chnl 2]
Report Annotation OFF
E-L
-------
INTEGRATOR PROGRAMMING INSTRUCTIONS (Continued)
Control Integrator
Oven Temp Annotation OFF
(should say ***Warning***Oven Temp now owned by Chnl 1
Run Time 0.01 Intg OFF
Run Time 0.01 Valve 5 ON
Run Time 0.01 Page
Run Time 0.02 List Attn2A
Run Time 0.04 Oven Temp ON
Run Time 0.20 Valve 2 ON
Run Time 0.21 Valve 2 OFF
Run Time 0.22 Intg ON
Run Time 0.23 Set BL
Run Time 0.23 List Intg
Rum Time 1.87 Set BL
Run Time 1.88 Intg OFF
Run Time 1.89 List Intg
Run Time 1.90 Chart Spped 1.5
Run Time 3.44 Valve 2 ON
Run Time 3.45 Valve 2 OFF
Run Time 3.46 Valve 2 ON
Run Time 3.47 Valve 2 OFF
Run Time 3.48 Valve 2 ON
Run Time 3.49 Valve 2 OFF
Run Time 3.50 STOP
Slave Integrator
Run Time 0.01 Intg OFF
Run Time 0.01 Page
Run Time 0.02 List Attn2A
Run Time 0.22 Intg ON
Run Time 0.23 Set BL
Run Time 0.23 List Intg
Rum Time 1.87 Set BL
Run Time 1.88 Intg OFF
Run Time 1.89 List Intg
Run Time 1.90 Chart Spped 1.5
Run Time 3.50 STOP
Control Integrator
Det 1 Temp 250
Det 1 Temp Limit 405
Inj 1 Temp 31
Inj 1 Temp Limit 405
Aux 1 Temp 90
Aux 1 Temp Limit 405
Flow A 30
Flow A Limit 500
E-2
-------
INTEGRATOR PROGRAMMING INSTRUCTIONS (Continued)
Slave Integrator
Flow B 30
Flow B Limit 500
Control Integrator
Valve 1 OFF
Valve 2 OFF
Valve 3 OFF
Valve 4 OFF
Valve 5 ON
Valve 6 OFF
Valve 7 OFF
Valve 8 OFF
Valve 9 OFF
Valve 10 OFF
Valve 11 OFF
Valve 12 OFF
Threshold 1
Peak Width 0.04
Slave Integrator
Threshold 1
Peak Width 0.04
Control Integrator
20 Valve 5 OFF
25 List Valve 5
30 Oven Temp Initial Value 30
35 Oven Temp OFF
40 Wait 2
60 Start
70 Oven Temp 90
80 Vale 5 ON
Sync ON
E-3
-------
APPENDIX F
1990 NMOC DAILY CALIBRATION DATA
-------
TABLE F-1. DAILY CALIBRATION DATA SUMMARY (CHANNEL A)
Cal
Date
06/06/89
06/08/89
06/12/89
06/15/89
06/19/89
06/20/89
06/21/89
06/23/89
06/27/89
06/29/89
07/03/89
07/05/89
07/06/89
07/10/89
07/11/89
07/12/89
07/18/89
07/19/89
07/27/89
07/31/89
08/02/89
08/04/89
08/09/89
08/11/89
08/14/89
08/16/89
08/22/89
08/28/89
08/29/89
08/30/89
08/31/89
09/05/89
09/06/89
09/07/89
09/11/89
09/12/89
09/13/89
09/14/89
09/15/89
09/18/89
09/19/89
09/20/89
09/21/89
09/22/89
09/26/89
09/27/89
09/29/89
09/29/89
10/02/89
10/03/89
10/04/89
Julian
Cal
Date
157
159
163
166
170
171
172
174
178
180
184
186
187
191
192
193
199
200
208
212
214
216
221
223
226
228
234
240
241
242
243
248
249
250
254
255
256
257
258
261
262
263
264
265
269
270
272
272
275
276
277
Intial
Zero
A.C.
0.000
0.000
0.000
3.935
7.025
0.000
7.485
0.000
15.560
4.685
5.050
0.000
0.000
6.020
0.000
0.000
0.000
0.000
16.175
0.705
14.725
14.355
21.645
5.065
1.380
3.900
5.060
8.315
5.970
10.470
2.915
2.875
2.605
0.000
7.275
5.385
0.000
4.400
6.655
4.040
4.780
3.240
6.590
4.175
5.455
8.030
0.000
0.000
0.000
0.000
0.000
Final
Zero
A.C.
0.000
0.000
0.000
3.935
7.025
0.000
7.485
0.000
15.560
4.685
5.050
0.000
0.000
6.020
0.000
0.000
0.000
0.000
16.175
0.705
14.725
14.355
21.645
5.065
1.380
3.900
5.060
8.315
4.860
10.470
2.915
4.480
3.425
9.580
0.000
5.385
0.000
4.400
6.655
4.040
4.780
3.240
6.590
4.180
5.455
0.000
0.000
0.000
0.000
0.000
0.000
Initial
Zero
ppmC
0.003226
0.004325
0.001551
0.001489
0.001535
0.002487
0.002309
0.001923
0.001389
0.001364
0.001608
0.001865
0. 001546
0.001742
0.002169
0.001305
0.001332
0.001370
0.001072
0.001157
0.001674
0.001701
0.000230
0.000972
0.000000
0.001154
0.001533
0.000798
0.001325
0.001118
0.000645
0.000808
0.000853
0.001061
0.001485
0.001394
0.001431
0.001084
0.000785
0.000624
0.001048
0.001694
0.001192
0.000990
0.001022
0.000734
0.001010
0.001010
0.001092
0.001011
0.000680
Final
Zero
ppnC
0.003226
0.001844
0.001551
0.001333
0.002268
0.002691
0.002334
0.001923
0.001388
0.000932
0.000000
0.001317
0.001149
0.001759
0.002193
0.000000
0.001286
0.002217
0.001259
0.001090
0.003173
0.000201
0,002524
0.003500
0.002793
0.001647
0.002575
0.000000
0.001434
0.001118
0.000593
0.001783
0.004587
0.000000
0.001485
0.001170
0.001407
0.001084
0.001645
0.000000
0.001694
0.000968
0.000000
0.000990
0.001022
0.000000
0.001010
0.001010
0.001092
0.000000
0.000680
Initial
Cal
Factor
0.000304
0.000305
0.000306
0.000306
0.000303
0.000303
0.000299
0.000300
0.000297
0.000299
0.000299
0.000296
0.000294
0.000294
0.000294
0.000293
0.000293
0.000293
0.000293
0.000298
0.000300
0.000292
0.000293
0.000292
0.000291
0.000294
0.000292
0.000299
0.000292
0.000298
0.000294
0.000295
0.000296
0.000294
0.000294
0.000292
0.000293
0.000293
0.000293
0.000290
0.000290
0.000290
0.000288
0.000288
0.000290
0.000291
0.000291
0.000291
0.000288
0.000290
0.000288
Final
Cal
Factor
0.000304
0.000304
0.000306
0.000307
0.000303
0.000303
0.000302
0.000300
0.000304
0.000302
0.000298
0.000294
0.000294
0.000296
0.000297
0.000295
0.000293
0.000296
0.000297
0.000299
0.000296
0.000291
0.000294
0.000290
0.000293
0.000295
0.000297
0.000295
0.000297
0.000298
0.000298
0.000297
0.000295
0.000296
0.000294
0.000295
0.000295
0.000293
0.000292
0.000292
0.000290
0.000289
0.000292
0.000288
0.000290
0.000295
0.000291
0.000291
0.000288
0.000291
0.000288
Cal
Factor
Drift
0.000000
0.000000
0.000000
-0.000001
-0.000000
-0.000000
-0.000003
0.000000
-0.000007
-0.000003
0.000001
0.000003
0.000001
-0.000002
-0.000003
-0.000002
0.000000
-0.000003
-0.000004
-0.000001
0.000004
0.000002
-o.oooooo
0.000003
-0.000002
-0.000001
-0.000005
0.000004
-0.000004
0.000000
-0.000004
-0.000002
0.000001
-0.000003
0.000000
-0.000003
-0.000001
0.000000
0.000001
-0.000002
-0.000001
0.000001
-0.000004
0.000000
0.000000
-0.000003
0.000000
0.000000
0.000000
-0.000002
0.000000
Cal
Factor
% Drift
0.000000
0.097962
0.000000
-0.329681
-0.129774
-0.080692
-1.110452
0.000000
-2.194403
-0.867026
0.435590
0.846345
0.244776
-0.553429
-1.099212
-0.714895
0.001611
-1.195837
-1.255266
-0.430754
1.273258
0.581712
-0.059249
0.875000
-0.698089
-0.395674
-1.739579
1.416399
-1.516439
0.000000
-1.501712
-0.634905
0.495894
-0.931626
0.000000
-0.981665
-0.475028
0.000000
0.437559
-0.687442
-0.279998
0.381854
-1.402463
0.000000
0.000000
-1.165454
0.000000
0.000000
0.000000
-0;595562
0.000000
-------
TABLE F-2. DAILY CALIBRATION DATA SUMMARY (CHANNEL B)
Julian
Cal Cal
Date Date
06/06/89
06/07/89
06/12/89
06/15/89
06/19/89
06/20/89
06/21/89
06/23/89
06/27/89
06/29/89
07/03/89
07/05/89
07/06/89
07/10/89
07/11/89
07/12/89
07/18/89
07/19/89
07/27/89
07/31/89
08/02/89
08/04/89
08/09/89
08/11/89
08/14/89
08/16/89
08/22/89
08/28/89
08/29/89
08/30/89
08/31/89
09/05/89
09/06/89
09/08/89
09/12/89
09/13/89
09/14/89
09/15/89
09/18/89
09/19/89
09/20/89
09/21/89
09/22/89
09/25/89
09/27/89
09/28/89
09/29/89
10/02/89
10/03/89
10/04/89
157
158
163
166
170
171
172
174
178
180
184
186
187
191
192
193
199
200
208
212
214
216
221
223
226
228
234
240
241
242
243
248
249
251
255
256
257
258
261
262
263
264
265
268
270
271
272
275
276
277
Intial
Zero
A.C.
0.000
0.000
0.000
0.000
1.750
0.000
3.755
0.000
3.235
0.970
1.140
0.000
0.000
3.315
0.000
0.000
0.000
0.000
0.565
0.000
2.540
2.500
6.685
0.600
0.980
1.755
6.160
4.040
5.340
5.915
6.755
1.785
1.515
1.455
2.860
1.350
0.000
4.290
5.455
4.210
0.805
3.200
9.435
5.450
3.245
4.905
0.000
0.000
0.000
0.000
Final
Zero
A.C.
0.000
0.000
0.000
0.000
1.750
0.000
3.755
0.000
3.235
0.970
1.140
0.000
0.000
3.315
0.000
0.000
0.000
0.000
0.565
0.000
2.540
2.500
6.685
0.600
0.980
1.755
6.160
4.040
3.875
5.915
6.755
1.360
0.535
3.785
0.000
1.350
0.000
4.290
5.455
4.210
0.805
3.200
9.435
6.995
3.245
0.000
0.000
0.000
0.000
0.000
Initial
Zero
ppmC
0.000000
0.000000
0.000477
0.000000
0.000188
0.000270
0.000097
0.000397
0.000000
0.000000
0.000546
0.000000
0.000000
0.000000
0.000000
0.000137
0.000000
0.000553
0.000285
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
'0.000190
0.000604
0.000000
0.000000
0.000000
0.000000
0.000000
0.000264
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Final
Zero
ppmC
0.000000
0.000452
0.000477
0.000000
0.000000
0.000681
0.000098
0.000397
0.000000
0.000000
0.000000
0.000000
0.000000
0.001747
0.000000
0.000000
0.000000
0.000987
0.000000
0.000000
0.000827
0.000000
0.000000
0.000734
0.000000
0.000000
0.000571
0.000000
0.000619
0.000000
0.000000
0.000612
0.000779
0.000264
0.000000
0.001624
0.000000
0.000277
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Initial
Cal
Factor
0.000305
0.000305
0.000309
0.000309
0.000313
0.000313
0.000309
0.000309
0.000309
0.000309
0.000307
0.000307
0.000304
0.000304
0.000305
0.000305
0.000303
0.000304
0.000304
0.000307
0.000309
0.000305
0.000301
0.000301
0.000302
0.000301
0.000303
0.000304
0.000305
0.000304
0.000304
0.000305
0.000304
0.000304
0.000304
0.000303
0.000303
0.000303
0.000299
0.000299
0.000298
0.000299
0.000299
0.000298
0.000304
0.000300
0.000300
0.000297
0.000297
0.000299
Final
Cal
Factor
0.000305
0.000305
0.000309
0.000310
0.000312
0.000313
0.000312
0.000309
0.000314
0.000311
0.000308
0.000304
0.000302
0.000304
0.000308
0.000307
0.000303
0.000305
0.000307
0.000309
0.000309
0.000302
0.000304
0.000301
0.000303
0.000302
0.000308
0.000308
0.000305
0.000304
0.000303
0.000306
0.000308
0.000304
0.000306
0.000304
0.000303
0.000306
0.000299
0.000298
0.000301
0.000300
0.000299
0.000298
0.000303
0.000300
0.000300
0.000297
0.000301
0.000299
Cal
Factor
Drift
0.000000
-0.000000
0.000000
-0.000001
0.000001
-0.000001
-0.000003
0.000000
-0.000005
-0.000002
-0.000000
0.000003
0.000002
0.000000
-0.000003
-0.000002
0.000000
-0.000002
-0.000003
-0.000002
-0.000000
0.000003
-0.000003
0.000000
-0.000002
-0.000001
-0.000005
-0.000004
-0.000000
0.000000
0.000001
-0.000001
-0.000004
0.000000
-0.000002
-0.000001
0.000000
-0.000003
0.000000
0.000002
-0.000003
-0.000001
0.000000
0.000000
0.000001
0.000000
0.000000
0.000000
-0.000004
0.000000
Cal
Factor
% Drift
0.000000
-0.032930
0.000000
-0.202326
0.409948
-0.249473
-1.066989
0.000000
-1.520471
-0.636530
-0.146081
0.888594
0.597033
0.059573
-0.828714
-0.726438
0.000000
-0.526010
-0.831039
-0.594243
-0.060981
0.925512
-1.011834
0.102413
-0.646297
-0.365781
-1.499200
-1.190001
-0.061877
0.000000
0.213428
-0.320421
-1.166536
0.000000
-0.648759
-0.384231
0.000000
-1.078605
0.099762
0.503915
-1.115485
-0.302059
0.000000
0.000000
0.294636
0.000000
0.000000
0.000000
-1.253519
0.000000
-------
TABLE F-3. DAILY CALIBRATION DATA SUMMARY (CHANNEL C)
Cal
Date
06/06/89
06/12/89
06/13/89
06/14/89
06/19/89
06/20/89
06/22/89
06/26/89
06/27/89
.06/28/89
07/03/89
07/05/89
07/06/89
07/12/89
07/13/89
07/17/89
07/17/89
07/19/89
07/20/89
07/21/89
07/24/89
07/25/89
07/26/89
07/31/89
08/01/89
08/01/89
08/03/89
08/07/89
08/08/89
08/10/89
08/11/89
08/14/89
08/15/89
08/16/89
08/17/89
08/18/89
08/21/89
08/23/89
08/24/89
08/25/89
08/29/89
08/30/89
09/01/89
09/05/89
09/06/89
09/07/89
09/08/89
09/11/89
09/12/89
09/13/89
09/14/89
09/15/89
09/18/89
09/19/89
09/20/89
Julian
Cal
Date
157
163
164
165
170
171
173
177
178
179
184
186
187
193
194
198
198
200
201
202
205
206
207
212
213
213
215
219
220
222
223
226
227
228
229
230
233
235
236
237
241
242
244
248
249
250
251
254
255
256
257
258
261
262
263
Intial
Zero
A.C.
0.000
6.630
1.430
7.835
6.365
0.000
4.780
2.390
5.965
2.040
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
7.020
3.385
0.000
1.000
0.000
0.000
0.000
10.355
0.000
0.000
0.000
12.545
4.105
1.100
5.525
6.115
2.880
1.970
1.515
4.380
6.515
2.795
1.045
0.000
0.000
0.000
0.000
0.000
0.000
2.210
0.000
0.000
Final
Zero
A.C.
0.000
6.630
1.430
7.835
6.365
0.000
4.780
2.390
5.965
2.040
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
7.020
3.385
0.000
1.000
0.000
0.000
0.000
10.355
0.000
0.000
0.000
12.545
4.105
1.100
5.525
6.115
2.880
0.000
1.515
4.380
6.070
11.635
1.045
6.580
0.000
0.000
0.000
0.000
0.000
2.760
0.000
0.000
Initial
Zero
ppmC
0.000000
0.002016
0.000000
0.000643
0.000224
0.000000
0.000000
0.000581
0.000000
0.002196
0.000096
0.001076
0.000117
0.000222
0.000000
0.000372
0.000372
0.000000
0.000000
0.000377
0.000288
0.000265
0.000426
0.000000
0.000000
0.000000 <
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000117
0.000000
0.001187
0.000000
0.000000
0.000000
0.001056
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Final
Zero
ppmC
0.000000
0.002016
0.000000
0.000000
0.000000
0.001519
0.000549
0.000000
0.000000
0.002196
0.000000
0.001456
0.000000
0.000000
0.000000*
0.001526
0.001526
0.000000
0.000000
0.000000
0.001166
0.001192
0.001256
0.000000
0.000436
0.000436
0.000000
0.000172
0.000691
0.000000
0.000000
0.000875
0.000316
0.000000
0.001721
0.000091
0.000000
0.000335
0.000000
0.001823
0.000000
0.001056
0.000000
0.005563
0.000266
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000400
0.000000
0.000000
0.000000
Initial
Cal
Factor
0.000303
0.000310
0.000309
0.000310
0.000311
0.000311
0.000311
0.000309
0.000307
0.000310
0.000311
0.000308
0.000307
0.000304
0.000303
0.000304
0.000304
0.000306
0.000306
0.000305
0.000303
0.000307
0.000305
0.000310
0.000310
0.000310
0.000310
0.000304
0.000302
0.000303
0.000302
0.000302
0.000302
0.000303
0.000303
0.000295
0.000306
0.000306
0.000311
0.000307
0.000307
0.000306
0.000307
0.000308
0.000303
0.000307
0.000307
0.000304
0.000307
0.000303
0.000303
0.000305
0.000299
0.000302
0.000301
Final
Cal
Factor
0.000303
0.000310
0.000309
0.000311
0.000311
0.000311
0.000311
0.000311
0.000309
0.000310
0.000309
0.000307
0.000307
0.000306
0.000303
0.000307
0.000307
0.000306
0.000307
0.000307
0.000305
0.000306
0.000307
0.000311
0.000307
0.000307
0.000311
0.000304
0.000306
0.000299
0.000303
0.000306
0.000304
0.000302
0.000304
0.000308
0.000312
0.000312
0.000322
0.000317
0.000304
0.000306
0.000307
0.000312
0.000311
0.000308
0.000307
0.000304
0.000307
0.000304
0.000303
0.000304
0.000305
0.000301
0.000302
Cal
Factor
Drift
0.000000
0.000000
-0.000000
-0.000001
0.000001
-0.000001
0.000000
-0.000002
-0.000001
0.000000
0.000002
0.000000
-0.000000
-0.000002
0.000000
-0.000003
-0.000003
-0.000000
-0.000001
-0.000002
-0.000001
0.000001
-0.000001
-0.000001
0.000003
0.000003
-0.000001
0.000000
-0.000004
0.000004
-0.000001
-0.000003
-0.000002
0.000001
-0.000001
-0.000013
-0.000006
-0.000006
-0.000011
-0.000010
0.000003
0.000000
0.000000
-0.000005
-0.000008
-0.000000
0.000000
0.000000
0.000000
-0.000001
0.000000
0.000001
-0.000005
0.000001
-0.000001
Cal
Factor
% Drift
0.000000
0.000000
-0.108095
-0.300575
0.227065
-0.210027
0.014707
-0.666836
-0.470070
0.000000
0.620594
0.137097
-0.037251
-0.614882
0.000000
-1.127941
-1.127941
-0.002155
-0.258112
-0.705133
-0.400412
0.214445
-0.488261
-0.455049
0.889591
0.889591
-0.476143
0.116014
-1.403917
1.279521
-0.337705
-1.106086
-0.753107
0.373168
-0.406379
-4.541017
-1.886510
-1.981042
-3.505237
-3.373133
1.016790
0.000000
0.000000
-1.505648
-2.778766
-0.022856
0.000000
0.000000
0.055230
-0.364337
0.000000
0.202849
-1.751838
0.434796
-0.397107
-------
TABLE F-3. DAILY CALIBRATION DATA SUMMARY (CHANNEL C)
Cal
Date
09/21/89
09/22/89
09/24/89
09/25/89
09/26/89
09/27/89
09/28/89
09/29/89
10/02/89
10/03/89
10/04/89
Julian
Cal
Date
264
265
267
268
269
270
271
272
275
276
277
Intial
Zero
A.C.
1.250
2.385
0.000
0.000
0.000
0.000
0.000
0.000
2.515
0.000
0.000
Final
Zero
A.C.
1.250
2.385
0.570
0.000
0.000
0.000
0.000
0.000
2.515
0.000
0.000
Initial
Zero
ppmC
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Final
Zero
ppnC
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Initial
Cal
Factor
0.000301
0.000300
0.000299
0.000298
0.000301
0.000302
0.000300
0.000302
0.000299
0.000301
0.000300
;;.Tai
Cal
Factor
0.000300
0.000300
0.000299
0.000298
0.000301
0.000305
0.000300
0.000302
0.000299
0.000302
0.000300
Cal
Factor
Drift
0.000001
0.000000
O.OOOOC
O.OOOOf
0.000000
-0.000002
0.000000
0.000000
0.000000
-0.000002
0.000000
ll
:or
•ift
.410038
j. 000000
0.000000
0.000000
0.000000
-0.807703
0.000000
0.000000
0.000000
-0.534528
0.000000
-------
TABLE F-4. DAILY CALIBRATION DATA SUMMARY (CHANNEL D)
Cal
Date
06/06/89
06/07/89
06/08/89
06/09/89
06/12/89
06/13/89
06/14/89
06/15/89
06/16/89
06/19/89
06/22/89
06/23/89
06/26/89
06/29/89
07/03/89
07/05/89
07/06/89
07/10/89
07/11/89
07/12/89
07/13/89
07/14/89
07/17/89
07/18/89
07/20/89
07/21/89
07/25/89
07/28/89
07/31/89
08/02/89
08/03/89
08/04/89
08/06/89
08/07/89
08/08/89
08/09/89
08/10/89
08/14/89
08/15/89
08/16/89
08/18/89
08/21/89
08/23/89
08/24/89
08/25/89
08/28/89
08/29/89
08/30/89
08/31/89
09/01/89
09/05/89
09/06/89
09/07/89
09/08/89
09/11/89
09/12/89
09/13/89
Julian
Cal
Date
157
158
159
160
163
164
165
166
167
170
173
174
177
180
184
186
187
191
192
193
194
195
198
199
201
202
206
209
212
214
215
216
218
219
220
221
222
226
227
228
230
233
235
236
237
240
241
242
243
244
248
249
250
251
254
• 255
256
Intial
Zero
A.C.
0.000
9.260
0.000
0.000
3.510
0.000
6.190
8.135
0.000
1.845
0.370
0.505
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
13.320
o.coc
0.000
0.000
0.000
0.000
0.000
13.285
0.000
0.000
0.000
0.865
0.405
0.000
0.000
5.695
0.000
2.905
0.000
8.345
2.210
8.530
2.380
0.000
0.000
1.370
0.000
0.000
0.000
0.990
0.000
1.135
0.000
0.340
Final
Zero
A.C.
.0.000
9.260
0.000
0.000
3.510
0.000
6.190
8.135
0.000
1.845
0.370
0.505
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
13.320
o.ccc
0.000
0.000
0.000
0.000
0.000
13.285
0.000
0.000
0.000
0.865
0.405
0.000
0.000
5.695
0.000
0.000
0.000
8.345
0.000
0.435
2.380
1.000
0.000
0.000
0.000
0.000
0.000
2.600
0.000
1.135
0.000
0.340
Initial
Zero
ppmC
0.000000
0.001374
0.000000
0.000000
0.002253
0.001342
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000011
0.000000
0.000689
0.001627
0.000000
0.000000
0.000000
0.000066
0.000000
0.000163
0.000000
3.00CCCO
0.000000
0.000000
0.000000
0.000000
0.001788
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000538
0.000000
0.003872
0.000000
0.000109
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Final
Zero
ppmC
0.000000
0.000000
0.000000
0.000135
0.002253
0.000000
0.000000
0.000149
0.000000
0.000000
0.000618
0.000000
0.000000
0.000363
0.000000
0.001542
0.002320
0.000000
0.000000
0.000000
0.000066
0.000000
0.001426
0.000000
O.C3C035 .
0.001186
0.001479
0.000000
0.000000
0.000000
0.000000
0.000925
0.000000
0.000000
0.000697
0.000000
0.000000
0.000000
0.001132
0.000000
0.000000
0.000000
0.000174
0.000000
0.000000
0.000000
0.002642
0.003872
0.000000
0.000109
0.000665
0.000116
0.000000
0.000000
0.000000
0.000000
0.000000
Initial
Cal
Factor
0.000299
0.000313
0.000304
0.000302
0.000308
0.000307
0.000307
0.000307
0.000306
0.000310
0.000307
0.000308
0.000308
0.000306
0.000306
0.000305
0.000304
0.000302
0.000303
0.000302
0.000300
0.000303
0.000302
0.000302
0.000301
0.000301
0.000303
0.000307
0.000307
0.000305
0.000307
0.000300
0.000299
0.000300
0.000299
0.000301
0.000300
0.000300
0.000299
0.000299
0.000306
0.000294
0.000297
0.000303
0.000305
0.000302
0.000303
0.000304
0.000302
0.000303
0.000304
0.000303
0.000303
0.000304
0.000302
0.000303
0.000309
Final
Cal
Factor
0.000299
0.000302
0.000304
0.000300
0.000308
0.000311
0.000308
0.000310
0.000306
0.000309
0.000307
0.000308
0.000307
0.000309
0.000305
0.000304
0.000304
0.000304
0.000303
0.000303
0.000300
0.000303
0.000304
0.000302
0.000302
0.000305
0.000303
0.000307
0.000310
0.000307
0.000308
0.000303
0.000299
0.000300
0.000301
0.000302
0.000303
0.000306
0.000305
0.000302
0.000292
0.000298
0.000302
0.000309
0.000312
0.000304
0.000297
0.000304
0.000308
0.000303
0.000311
0.000310
0.000304
0.000304
0.000302
0.000303
0.000309
Cal
Factor
Drift
0.000000
0.000010
0.000001
0.000002
0.000000
-0.000005
-0.000001
-0.000003
0.000000
0.000000
0.000000
0.000000
0.000001
-0.000003
0.000001
0.000002
0.000000
-0.000002
0.000001
-0.000001
0.000000
0.000000
-0.000001
0.000000
-0.000000
-0.000004
0.000000
0.000000
-0.000003
-0.000002
-0.000001
-0.000004
0.000000
0.000000
-0.000002
-0.000001
-0.000003
-0.000006
-0.000006
-0.000002
0.000014
-0.000004
-0.000004
-0.000006
-0.000006
-0.000002
0.000005
0.000000
-0.000006
0.000000
-0.000007
-0.000006
-0.000001
0.000000
0.000000
-0.000001
0.000000
Cal
Factor
X Drift
0.000000
3.350653
0.272276
0.671377
0.000000
-1.509456
-0.294841
-0.942282
0.000000
0.111115
0.136536
0.000000
0.196775
-1.003048
0.307133
0.540610
0.131631
-0.638951
0.195601
-0.416619
0.000000
0.113669
-0.475004
0.000000
-0.144582
-1.201915
0.083444
0.000000
-0.901830
-0.657493
-0.358708
-1.208061
0.000000
0.029398
-0.616333
-0.336481
-1.107990
-1.855203
-1.921009
-0.779748
4.460500
-1.423636
-1.498292
-2.056354
-2.074276
-0.588713
1 .816768
0.000000
-1.993487
0.000000
-2.357163
-2.121072
-0.327210
0.000000
0.000000
-0.205518
0.023649
-------
TABLE F-4. DAILY CALIBRATION DATA SUMMARY (CHANNEL D)
Cal
Date
09/15/89
09/18/89
09/19/89
09/20/89
09/21/89
09/22/89
09/24/89
09/25/89
09/26/89
09/27/89
09/28/89
09/29/89
10/02/89
10/03/89
10/04/89
Julian
Cat
Date
258
261
262
263
264
265
267
268
269
270
271
272
275
276
277
Intial
Zero
A.C.
0.000
0.000
0.000
7.085
0.000
0.000
8.480
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
Final
Zero
A.C.
0.000
0.000
0.000
0.000
0.000
0.000
8.480
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
Initial
Zero
ppnC
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Final
Zero
ppnC
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
Initial
Cal
Factor
0.000303
0.000299
0.000300
0.000298
0.000298
0.000297
0.000297
0.000296
0.000299
0.000300
0.000297
0.000299
0.000297
0.000296
0.000296
Final
Cal
Factor
0.000303
0.000302
0.000298
0.000301
0.000297
0.000297
0.000297
0.000296
0.000299
0.000303
0.000297
0.000299
0.000297
0.000299
0.000296
Cal
Factor
Drift
-0.000001
-0.000003
0.000002
-0.000003
0.000000
0.000000
0.000000
0.000000
0.000000
-0.000004
0.000000
0.000000
0.000000
-0.000003
0.000000
Cal
Factor
X Drift
-0.185613
-1.101487
0.784490
-1.012285
0.154848
0.000000
0.000000
0.000000
0.000000
-1.190461
0.000000
0.000000
0.000000
-0.970956
0.000000
-------
APPENDIX G
1990 NMOC IN-HOUSE QUALITY CONTROL SAMPLES
-------
TABLE G-1. NMOC INHOUSE QUALITY CONTROL SAMPLES (CHANNEL A)
Date
Analyzed
08-Jun-90
24-Jul-90
02-Aug-90
06-Aug-90
31-Aug-90
Julian
Date
Analyzed
159
205
214
218
243
QC
I.D.
Number
1001
1213
1306
1307
1468
Calculated
NMOC
ppmC
0.5960
1.4640
0.2370
0.3530
1.1080
Measured
NMOC
ppmC
0.6650
1.6000
0.2770
0.3650
1.1204
NMOC
Bias
ppmC
0.0690
0.1360
0.0400
0.0120
0.0124
NMOC
Percent
Bias
11.5772
9.2896
16.8776
3.3994
1.1191
-------
TABLE G-2. NHOC INHOUSE QUALITY CONTROL SAMPLES (CHANNEL B)
Date
Analyzed
08-Jun-90
24-JuL-90
02-Aug-90
06-Aug-90
31-Aug-90
Julian
Date
Analyzed
159
205
214
218
243
QC
I.D.
Number
1001
1213
1306
1307
1468
Calculated
NHOC
ppmC
0.596
1.464
0.237
0.353
1.108
Measured
NHOC
ppmC
0.653
1.520
0.279
0.362
1.183
NMOC
Bias
ppmc
0.057
0.056
0.042
0.009
0.075
NMOC
Percent
Bias
9.564
3.825
17.722
2.550
6.760
-------
TABLE G-3. NHOC INHOUSE QUALITY CONTROL SAMPLES (CHANNEL C)
Date
Analyzed
07-Jun-90
U-Jun-90
24-Jul-90
12-AU8-90
07-Sep-90
Julian
Date
Analyzed
158
165
205
224
250
QC
I.D.
Number
1000C
1031C
1213C
1315C
1498C
Calculated
NMOC
ppmC
0.45
0.968
1.584
0.58
1.37
Measured
NMOC
ppnC
0.432
0.957
1.464
0.564
0.796
NMOC
Bias
ppmC
-0.018
-0.011
-0.120
-0.016
-0.574
NMOC
Percent
Bias
-4.000
-1.136
-7.576
-2.759
-41.898
-------
TABLE G-4. NMOC INHOUSE QUALITY CONTROL SAMPLES (CHANNEL D)
Date
Analyzed
07-Jun-90
U-Jun-90
24- Jut -90
12-Aug-90
07-Sep-90
Julian
Date
Analyzed
158
165
175
224
250
QC
I.D.
Number
1000
1031
1213
1315
1498
Calculated
NMOC
ppmC
0.459
0.958
1.480
0.563
1.501
Measured
NHOC
ppmC
0.432
0.957
1.464
0.564
0.796
NMOC
Bias
ppmC
-0.027
-0.001
-0.016
0.001
-0.705
NMOC
Percent
Bias
-5.882
-0.104
-1.081
0.178
-46.969
-------
APPENDIX H
MULTIPLE DETECTOR SPECIATED THREE-HOUR SITE DATA SUMMARIES
-------
APPENDIX H -- LIST OF TABLES
Table Page
H-l MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR BRLA . H-l
H-2 MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR NWNJ . H-4
H-3 MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR PLNJ . H-6
cah.!98f
-------
TABLE H-1. MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR BRLA
Sample ID
Sample Date
Total NMOC, ppmC
Compound
Acetylene
1,3- Butadiene
Vinyl chloride
Chloromethane
Chkxoethane
Bromomethane
Methyleoe chloride
trans- 1 ,2-Dwhloroethylene
1,1-Dtehkxoethane
Chkxoprene
Bromochloromethane
Chloroform
1.1.1-Trichtoroethane
Carbon tetrachkxide
1,2-Dtchlofoethane
Benzene
Trichloroethylene
1 ,2-Dtchloropfopane
Bromodichloromethane
Toluene
n-Octane
cis-1 ,3-Dichloropropylene
1,1,2-Trichlofoethane
Tetrachloroemylene
Dibromochloromethane
Chlorobenzene
Ethylbenzene
m/p-Xytene
Styrene/o-Xylene
Bromoform
1 , 1 ,2.2-Tetrachkx oethane
nvDJchkxobenzene
p-Dichlorobenzene
o-Dtchlofobenzena
Propylene
trans- 1 ,3-Dichloropropylene
1274
7-25-90
0.148
3.010
0.723
0.415
0.276
2.930
1.140
5.680
0.340
0.156
0.660
3.100
1.750
19.410
(H)
(M)
(L)
(L)
(H)
(U
(H)
(H)
(M)
(L)
(D
(L)
-------
TABLE H-1. BRLA (Continued)
Sample ID
Sample Date
Total NMOC, ppmC
Compound
Acetylene
1,3-Butadiene
vinyl chloride
Chloromethane
Chtoroethane
Bromomethane
Methy tone chloride
trans- 1 ,2-Dtehloroethylene
1,1-Dichloroethane
Chkxoprene
Bromochloromethane
Chloroform
1,1.1-Trichkxoethane
Carbon tetrachkxide
1,2-Dichloroethane
Benzene
Trichtoroethytene
1 ,2-Dichloropropane
Bromodichloromethane
Toluene
n-Octane
cis-1 ,3-DJchkxopropylene
1,1.2-Trichloroethane
Tetrachtoroethytene
Dibromochkxomethane
Chlorobenzene
Ethylbenzene
m/p-Xylene
Styrene/o-Xylene
Bromoform
1 . 1 ,2,2-Tetrachloroetriane
m-Dichkxobenzene
p-Dichlorobenzene
o-Dtchlorobenzene
Propylene
trans-1,3-Dichlorooropylene
131 OR
7-31-90
0.740
1322
8-1-90
0.487
1324
8-2-90
0.100
1333
8-2-90
0.989
1351
8-6-90
0.954
Concentration, ppbv
5.890
0.516
0.174
3.220
1.380
3.610
0.083
0.080
0.300
1.360
0.870
1.520
9.020
(H)
(L)
(L)
(H)
(H)
(M)
(L)
(L)
(H)
(H)
(H)
(D
(U
0.440
0.420
0.392
0.168
2.670
4.640
1.677
0.520
2.660
1.320
0.490
8.240
(H)
(L)
(L)
(L)
(H)
(M)
(L)
(L)
(U
(L)
(L)
0.310
0.419
0.144
2.270
1.140
4.110
4.230
0.445
0.500
3.040
1.570
6.230
(H)
(U
(U
(H)
(H)
(H)
(H)
(M)
(L)
(L)
(L)
(U
1.780
1.780
0.502
0.300
4.100
8.180
0.532
0.980
4.810
2.610
1Z540
(H)
-"
(H)
(L)
(U
(H)
(H)
(U
(H)
(H)
(H)
(L)
3.150
0.370
0.324
0.199
2.300
0.910
4.220
0.130
0.470
1.870
1.100
11.760
0.730
(H)
(H)
(U
(U
(H)
(M)
(M)
(L)
(L)
(L)
(U
(L)
(L)
H High confidence level
0 Duplicate sample
M Medium confidence level
R Replicate sample
L Low confidence level
(Continued)
H-2
-------
TABLE H-1. BRLA (Continued)
Sample ID
Sample Date
1352
8-7-90
Total NMOC. ppmC
0.860
Compound
Concentration, ppbv
Acetylene
1,3- Butadiene
Vinyl chloride
Chkxomethane
Chtoroethane
Bromomethane
Methytene chloride
trans-1,2-Dichloroethylene
1.1-Dtchkxoethane
Chkxoprene
Bromochloromethane
Chloroform
1.1,1-Trichloroethane
Carbon tetrachloride
1,2-Oichloroetnane
Benzene
Trichtoroemytene
1,2-Oichloropropane
Bromodichloromethane
Toluene
n-Octane
ci*-1,3-Dichloropropylene
1,1,2-Trichtofoethane
Tetrachloroethylene
Dibromochloromethane
Chkxobenzene
Ethylbenzene
m/p-Xylene
Styrene/o-Xylene
Bromoform
1,1,2,2-Tetrachkxoethane
m-Dichlorobenzene
p-OJchkxobenzene
o-Dichlorobenzene
Propyteoe
trans-1,3-Dichloropropylene
0.410 (M)
0.260 (M)
0.345 (L)
0.140 (L)
1.000 (H)
1.760 (L)
0.054 (L)
0.250 (L)
0.860 (L)
0.570 (L)
9.270 (L)
H High confidence level
M Medium confidence level
L Low confidence level
H-3
-------
TABLE H-2. MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR NWNJ
Sample ID
Sample Data
1285D
7-26-90
1285R
7-26-90
1286D
7-26-90
1295
7-30-90
1312
7-31-90
Total NMOC, ppmC
0.169
0.169
0.220
0.184
0.737
Compound
Concentration, ppbv
Acetylene
1.3-Butadiene
Vinyl chloride
Chkxomethane
Chkxoethane
Bromomethane
Methyleoe chloride
trana-1.2-Dichloroethylene
1,1-Dichloroethane
Chkxoprene
Bromochloromethane
Chloroform
1,1.1-Trichkxoethane
Carbon tetrachloride
1,2-Dtehloroethane
Benzene
Trichkxoethylene
1,2-Dichloropropane
Bromodichloromethane
Toluene
n-Octane
cis-1,3-Dtchloropropytene
1,1,2-Trichtofoethane
Tetrachtoroethylene
Dibromochloromethane
Chlorobenzene
Ethylbenzene
m/p-Xytene
Styrene/0-Xylene
Bromoform
1.1,2.2-Tetrachkxoethane
m-Dichlorobenzene
p-Dtchlorobenzene
o-Dichkxobenzene
Propylene
trans-1,3-Dtchkxopropylene
0.460 (H)
1.145 (L)
0.126 (L)
0.500 (H)
0.337 (L)
1.510 (H)
2.010 (H)
0.295 (L)
0.270 (M)
0.980 (L)
0.650 (M)
1.260 (L)
1.275 (H)
0.144 (L)
0.560 (H)
0.310 (L)
2.140 (L)
0.862 (L)
0.220 (M)
1.040 (M)
0.590. (M)
1.170 (L)
0.460 (L)
1.590 (H)
0.138 (L)
0.730 (H)
0.391 (L)
0.120 (M)
2.690 (M)
1.438 (L)
0.330 (L)
1.330 (L)
0.730 (L)
0.920 (L)
0.937 (L)
0.126 (L)
0.590 (H)
0.100 (H)
2.270 (H)
0.355 (M)
0.280 (H)
1.260 (H)
0.750 (M)
1.770 (L)
1.622 (H)
0.148 (L)
2.240 (H)
0.769 (H)
0.500 (H)
11.490 (H)
0.340 (H)
0.523 (M)
1.070 . (H)
6.240 (H)
£720 (H)
8.100 (L)
H High confidence level
D Duplicate sample
M Medium confidence level
R Replicate sample
L Low confidence level
(Continued)
H-4
-------
TABLE H-2. NWNJ (Continued)
Sam pie ID
Sample Date
Total NMOC, ppmC
Compound
1323
8-1-90
0.609
1338
S-2-90
0.330
1357
8*90
0.066
Concentration, ppbv
1371
8*90
0.432
1380
8-30-90
0.585
Acetylene
1,3-Butadiene
Vinyl chloride
Chloromethane
Chkxoethane
Bromomethane
Methytene chloride
trans-1,2-Dichloroethylene
1,1-Dtchloroethane
Chlofopfene
Bromochloromethane
Chloroform
. 1,1,1-Trichloroethane
Carbon tetrachtoride
1,2-Dtchtoroethane
Benzene
Trichkxoetnytene
1,2-Dtchkxopropane
Bromodichloromethane
Toluene
n-Octane
cts-t ,3-Oichloropropylene
1,1.2-Trichloroethane
Tetrachloroethytene
Dibromochloromethane
Chlorobenzene
Ethylbenzene
m/p-Xylene
Styrene/o-Xylene
Bromoform
1.1,2,2-Tetrachkxoethane
m-Dichlorobenzene
p-Oichlorooenzene
o-Oichlorobenzene
Propytene
trans-1,3-Oichloropropylene
0.140 (H)
0.310 (L)
0.450 (H)
0.702 (L)
0.149 (L)
0.550 (H)
2.440 (M)
0.381 (L)
0.190 (M)
0.810 (M)
0.520 (M)
1.640 (L)
0.110 (H)
1.021 (H)
0.144 (L)
0.690 (H)
5.010 (H)
0.397 (M)
0.310 (M)
1.680 (M)
0.960 (M)
2.350 (L)
0.210 (H)
0.640 (L)
0.959 (L)
0.136 (L)
0.910 (H)
3.470 (H)
0.135 (H)
0.070 (H)
0.410 (H)
1.820 (H)
0.880 (M)
0.500
2.320 (L)
0.170 (H
0.080
0.200 (M)
1.408 (H)
0.177 (L)
0.960 (H)
0.104 (L)
0.230 (L)
4.520 (H)
0.330 (H)
0.337 (L)
0.490 (M)
0,286 (H)
1.330 (M)
1.680
7.090 (M)
0.180 (H)
0.030 (H)
0.310 (M)
2.162 (H)
0.147 (L)
1.080 (H)
0.974 (H)
0.320 (L)
7.240 (M)
3.728 (L)
0.540 (M)
2.990 (M)
1.440 (M)
9.270 (L)
4.170 (L)
H High confidence level
M Medium confidence level
L Low confidence level
H-5
-------
TABLE H-3. MULTIPLE DETECTOR SPECIATED THREE-HOUR DATA SUMMARY FOR PLNJ
Sample ID
Sample Date
1279
7-25-90
1292D
7-27-90
1293D
7-27-90
1293R
7-27-90
1304
7-30-90
Total NMOC, ppmC
0.162
0.289
0.286
0.266
0.339
Compound
Acetylene
1.3-Butadiene
Vinyl chloride
Chloromethane
Chkxoethane
Bromomethane
Methylene chloride
trans-1,2-Dtchtoroethylene
1,1-Dtchkxoethane
Chkxoprene
Bromochloromethane
Chloroform
1,1,1-Trichloroethane
Carbon tetrachkxide
1,2-Dtehtoroethane
Benzene
Trichtoroetnylene
1,2-Dichloropropane
Bromodichloromethane
Toluene
rv Octane
cis-1,3-Dichloropropylene
1,1,2-Trichloroethane
Tetrachloroethylene
Dibromochloromethane
ChJorobenzene
Ethylbenzene.
m/p-Xyteoe
Styrane/o-Xylene
Bromoibrm
1.1 ,2,2-Tetrachloroethane
nvDichlorobenzene
p-Dtchkxopenzene
o-Dichlorobenzene
Propylene
tran»-1,3-Dichk>ropropylene
0.320 (L)
0.484 (L)
0.122 (L)
0.530 (H)
1.090 (L)
0.387 (H)
0.150 (H)
0.320 (L)
0.240 (H)
0.020 (L)
0.650
0.180 (H)
0.180 (H)
0.230 (M)
0.957 (H)
0.142 (L)
1.290 (H)
0.088 (U
0.180 (H)
3.840 (M)
0.070 (M)
0.691 (M)
0.320 (M)
1.640 (H)
0.800 (H)
t.970
0.300 (L)
0.938 (H)
0.138 (L)
1.260 (H)
0.160 (M)
3.520 (M)
3.001 (L)
0.350 (M)
1.720 (M)
0.880 (M)
1.690
0.350 (L)
1.000 (H)
0.148 (L)
1.140 (M)
0.222 (L)
3.190 (M)
3.130 (L)
0.340 (M)
1.760 (H)
0.880 (M)
1.660 (L)
0.120 (H)
0.695 (H)
0.143 (L)
1.360 (H)
3.130 (M)
0.186 (H)
0.320 (H)
1.630 (H)
0.830 (H)
0.220 (L)
4.050 (L)
H High confidence level
0 Duplicate sample
M Medium confidence level
R Replicate sample
L Low confidence level
(Continued)
H-6
-------
TABLE H-3. PLNJ (Continued)
Sample 10
Sample Date
Total NMOC, ppmC
Compound
1308
7-31-90
1.165
1349
8-3-90
0.899
1350
8-6-90
0.066
1359
8-8-90
0.319
1360
8-7-90
1.079
Acetylene
1.3-Butadiene
Vinyl chloride
Chkxomethane
Chloroethane
Bromomethane
Methytane chloride
tram-1,2-Dtchloroethy tone
1,1-Dichloroethane
Chkxoprene
Bromochloromethane
Chloroform
1,1,1-Trichkxoettiane
Carbon tetrachkxide
1,2-Dfchtoroethane
Benzene
Trichloroethylene
1,2-Dtchloropropane
Bromodichloromethane
Toluene
n-Octane
cw-1,3-Oichloropropylene
1.1,2-Trichtoroethane
Tetrachloroettiylene
Oibromochloromethane
Chkxobenzene
Ethylbenzene
m/p-Xylene
Styrene/o-Xylene
Bromoform
1,1,2,2-Tetrachloroethane
m-Oichlorobenzene
p-Oichloropenzene
o-Dichlorobenzene
Propylene
trans-1.3-0ichloropropylene
0.470 (H)
0.280 (H)
0.080 (I)
1Z323 (L)
0.107 (U
3.000 (H)
0.344 (L)
0.620 (H)
13.340 (H)
0.543 (L)
0.080 (H)
1.010 (H)
4.840 (H)
2.400 (L)
8.960 (L)
2.690 (H)
0.060 (M)
0.163 (M)
0.322 (H)
0.136 (L)
0.310 (H)
0.070 (H)
1.280 (H)
0.140 (H)
0.170 (L)
0.520 (L)
0.300 (L)
1.410 (M)
0.060 (L)
0.920 (H)
0.190 (M)
0.051 (L)
0.270 (H)
0.900 (H)
0.004 (H)
2.910 (H)
0.060 (H)
0.320 (H)
0.160 (H)
0.600 (L)
0.110 (H)
0.961 (L)
0.143 (L)
1.350 (H)
3.850 (M)
0.186 (M)
0.330 (H)
1.710 (H)
0.950 (H)
2.450 (L)
0.180 (M)
0.369 (H)
0.134 (L)
0.420 (H)
0.780 (H)
0.162 (H)
0.010 (H)
0.090 (H)
0.380 (M)
0.210 (H)
0.030 (M)
0.580 (L)
H High confidence level
M Medium confidence level
L Low confidence level
H-7
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1. REPORT NO.
EPA-450/4- 91-008
2.
3. RECIPIENTS ACCESSION NO.
4. TITLE AND SUBTITLE
1990 Nonmethane Organic Compound And Three-hour
Air Toxics Monitoring Program
5. REPORT DATE
January 1991
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Radian Corporation
Research Triangle Park, NC
8. PERFORMING ORGANIZATION REPORT NO.
27709
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68D80014
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
In certain areas of the country where the National Ambient Air Quality Standard
(NAAQS) for ozone is being exceeded, additional measurements of ambient nonmethane
organic compounds (NMOC) are needed to assist the affected States in developing
revised ozone control strategies. Because of previous difficulty in obtaining accurate
NMOC measurements, the U.S. Environmental Protection Agenc.y (EPA) has provided
monitoring and analytical assistance to these States, beginning in 1984 and
continuing through the 1990 NMOC Monitoring Program.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Ozone Control Strategies
National Ambient Air Quality Standards
Nonmethane Organic Compound
Monitoring
Analysis
1990 NMOC Monitoring Program
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report/
21. NO. OF PAGES
279
Unlimited
20. SECURITY CLASS (Tin's page)
22. PRICE
EPA Form 2220-1 (R«v. X-77) PREVIOUS EDIT-ON i _ oesouere
-------
U.S. Envi
R:gion 5,
77 West
Chicago,
Library (PL-]
Jackson EC,
-------
AUSTIN
Corporate Headquarters
8501 Mo-Pac Blvd
PO Box 201088
AusUn. TX 78720-1088
(5121154-4797
FAX 512-454-7129
Summit Park
Austin. TX
(5121244-0100
FAX 512-244-0160
DENVER
1801 Broadway, Suite 1000
Denver. CO 80202
(303| 292-0800
FAX 303-292-5860
HOUSTON
10675 Richmond Ave , Suite 190
Houston, TX 77042
(713)785-9225
FAX 713-785-9390
LOS ANGELES
2250 E. Imperial Hwy, Suite 140
El Segundo. CA 90245
[213)640-0045
FAX 213-640-8940
7 Corporate ParK, Suite 240
Irvine. CA 92714
(714|261-8611
FAX 714-261-6505
MILWAUKEE
5101 West Beloit Rd
Milwaukee. WI 53214
(414)643-270!
FAX 414-643-2699
NASHVILLE
50 Century Blvd . Suite 323
Nashville, TN 37214
(6151885-4281
FAX 615-885-4395
RALEIGH/DURHAM
3200 E Chapel Hill Rd /Nelson Hwv
PO Box 13000
Research Triangle Park, NC 27709
(9191541-9100
FAX 919-541-9013
Perimeter Park
Momsville. NC
(9191481-0212
FAX 919-460-1631
ROCHESTER
155 Corporate Woods, Suite 100
Rochester. NY U623
(716)292-1870
FAX 716-292-1878
SACRAMENTO
10395 Old Placerville Road
Sacramento, CA 95827
(9161362-5332
FAX 916-362-2318
WASHINGTON, D.C.
2455 Horsepen Road Suite 250
Herndon, VA 22071
(7031834-1500
FAX 703-834-1512
HONG KONG
2001 Bond Center
GPO Box 12606
Hong Kong
5-231016
FAX (051 868-1686
LONDON
Radian Limited
Dukes Court
Duke Street
Wokmg
Surrey GL'Jl 5BH
UK
(04831 729307
FAX (04831 725233
TAIWAN
3FL -4 No 392 Tun Hua S Rd
Taipei Taiwan ROC
(0217816161
FAX (021 775-5906
\ comuanv of The Hartford Steam Boiler
Inspection and Insurance Co
01/90
------- |