EPA-650/2-75-067
May 1975 Environmental Protection Technology Series
DEVELOPMENT
OF AIR SAMPLING
METHODOLOGY
U.S. Environmental Protection Agency
Office of Research and Development
Washington, 0. C. 20460
-------
EPA-650/2-75-067
DEVELOPMENT
OF AIR SAMPLING
METHODOLOGY
by
Bernard E. Sallzman
University of Cincinnati
Cincinnati, Ohio 45267
Grant No. 800869
ROAP No. 26ACX-37
Program Element No. 1A1010
EPA Project Officer: Dr. Eugene Sawicki
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, D.C. 20460
May 1975
-------
EPA REVIEW NOTICE
This report has been reviewed by the National Environmental Research
Center - Research Triangle Park, Office of Research and Development,
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environ-
mental Protection Agency, have been grouped into series. These broad
categories were established to facilitate further development and applica-
tion of environmental technology. Elimination of traditional grouping was
consciously planned to foster technology transfer and maximum interface
in related fields. These series are:
1. ENVIRONMENTAL HEALTH EFFECTS RESEARCH
2 . ENVIRONMENTAL PROTECTION TECHNOLOGY
3. ECOLOGICAL RESEARCH
4. ENVIRONMENTAL MONITORING
5. SOCIOECONOMIC ENVIRONMENTAL STUDIES
6. SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
9. MISCELLANEOUS
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to
develop and demonstrate instrumentation, equipment and methodology
to repair or prevent environmental degradation from point and non-
point sources of pollution. This work provides the new or improved
technology required for the control and treatment of pollution sources
to meet environmental quality standards.
This document is available to the public for sale through the National
Technical Information Service, Springfield, Virginia 22161.
Publication No. EPA-650/2-75-067
11
-------
ABSTRACT
Major efforts were focussed on use of solid absorbents to
provide simple, compact, and inexpensive, yet reliable gas
sampling and analysis systems for three gaseous pollutants:
nitrogen dioxide, "reactive" hydrocarbons, and vinyl chloride.
Time fluctuations of monitoring data by computer analyses
showed alignment of Fourier spectra for common pollutants.
Nitrogen dioxide was sampled with small glass tubes packed
with copper shot. This device could be readily used as a
personal dosimeter, as well as at a fixed station. The tubes
were capped, and could be stored for up to 5 weeks. Nitrogen
compounds released by heating the tube in a stream of hydro-
gen were passed into a catalytic furnace, converted to
ammonia, and reproducibly determined with a conductivity
detector.
The dual system for monitoring hydrocarbons with an FID
detector passed one sample stream through a chromium trioxide-
sulfuric acid absorbent and the other through a bypass. Con-
ditions were adjusted to obtain complete response (taken as
the difference) to olefins and partial response to aromatics,
to parallel photochemical reactivity.
Permeation tubes containing vinyl chloride were evaluated
as standards for calibrating gas chromatographs in analysis
of air for this gas. Permeation rates and their temperature
coefficients were determined for FEP and TFE Teflon for up to
8 months. Best conditions for sampling with charcoal tubes,
iii
-------
desorption with carbon disulfide, and analysis by injection
into a gas chromatograph were determined.
This report was submitted in fulfillment of Project 26 ACX
37, Grant No. R800869, by the Department of Environmental
Health, University of Cincinnati under the partial sponsor-
ship of the Environmental Protection Agency. Work was com-
pleted as of February 28, 1975.
IV
-------
TABLE OF CONTENTS
Pago
List of Figures vi
Acknowledgments viii
Sec 'lions:
[ Conclusions 1
II Recommendations 3
III Introduction 4
IV Solid Absorbent Method for Sampling
and Analysis of Nitrogen Dioxide in
Ambient Air 6
V Continuous Monitoring Instrument for
"Reactive" Hydrocarbons in Ambient Air 12
VI Calibration of a Gas Sampling Valve for
Gas Chromatography 17
VII Use of Permeation Tubes for Calibration
of Vinyl Chloride Analyses 19
VIII Sampling and Analysis of Airborne Vinyl
Chloride 23
IX Fourier Analysis of Air Monitoring Data 27
X List of Publications and Manuscripts 31
XI NT1S Bibliographic Data Sheet 33
v
-------
LIST OF FIGURES
Number Page
1. Sampling train for NO2. 7
2. Desorption and detection system. 9
3. Performance of method for different
sampling conditions. 10
4. Schematic of the hydrocarbon monitor-
ing system. 13
5. Preliminary studies of effects of
temperature and contact time on
the absorbent performance. 15
6. Recording from instrument of ambient
air in Cincinnati on June 2, 1973.
Upper half of each 5-minute cycle
shows total hydrocarbon concentration,
and the decrease to the lower half is
the reactive hydrocarbon concentration. 16
7. Permeation rates of vinyl chloride in
Teflon tubes of 0.250" o.d. x 0.03"
wall thickness. 21
8. Fourier amplitude spectrum of concen-
trations of total.hydrocarbons in
Cincinnati for Oct.-Dec. 1968. 29
VI
-------
9. Fourier amplitude spectrum for con-
centrations of sulfur dioxide in
Cincinnati for October, 1968. 30
VII
-------
ACKNOWLEDGEMENTS
The principal investigator for this project was Bernard
E. Saltzman, Ph.D. William R. Burg, Ph.D., and John E.
Cuddeback, Ph.D. served as co-investigators, with assistance
in some phases from Shelton R. Birch, a graduate student.
This work was supported in part by the Center for the Study
of the Human Environment under U.S. Public Health Service
Grant ES 00159, and in part by the Environmental Protection
Agency under Research Grant R800869.
VI11
-------
SECTION I
CONCLUSIONS
A small glass tube packed with copper shot provided a
simple, compact, convenient, and effective means for sampling
ambient air for low concentrations of nitrogen dioxide, for
periods varying from a few minutes to 24 hours. Interfering
high concentrations of sulfur dioxide were removed by a
filter impregnated with potassium carbonate. The exposed
tubes were sealed with Teflon plugs, and could be stored for
35 days with only 11% loss. The analytical system which in-
cluded heating in a stream of hydrogen, converting released
nitrogen compounds to ammonia in a furnace train, and determ-
ining them with a Coulson Conductivity Detector, yielded a
reproducible stoichiometric factor of 59%. This sampling
tube is small enough for use as a personal sampling device,
and could be reused repeatedly after each analysis.
A monitoring system for total and "reactive" hydrocarbons
was developed by adapting a current commercial total hydro-
carbon analyzer, and was successfully used in Cincinnati.
Response was based on the difference between cyclical measure-
ments on sample air, and on the air passed through an ab-
sorbent. It could be adjusted to parallel photochemical re-
activities of various classes of hydrocarbons. This should
provide more significant measurements than those currently
made of non-methane hydrocarbons, and makes feasible an im-
proved air quality standard.
- 1 -
-------
Sampling and analysis of ambient concentrations of vinyl
chloride were critically examined. Teflon permeation tubes
were found to be accurate means of providing known low con-
centrations for testing carbon tube sampling methods and
calibrating gas chromatographs. Analytical methodology was
optimized to provide 90% recovery even after two weeks of
sample storage.
Computer analysis of air monitoring data for Cincinnati
yielded Fourier spectra for lead, sulfur dioxide, and hydro-
carbons. Similar patterns were obtained for fluctuations
of corresponding time periods. Further work should clarify
atmospheric processes, the relationships between pollutants,
and should produce improved models of concentrations.
- 2 -
-------
SECTION II
RECOMMENDATIONS
It was not possible within the funding made available to
fully develop and field test the methods described. The
scope was limited to exploration, determination of feasibili-
ty, and critical examination of significant parameters.
Efforts were focussed on three important pollutants. We be-
lieve that the results justify continued work for ultimate
routine application of these procedures.
- 3 -
-------
SECTION III
INTRODUCTION
Recent legislation has resulted in a great expansion of
work in air quality monitoring by multiple networks. This
has accentuated the constant need for simplified, more con-
venient, and more accurate analytical methods. Rapid ad-
vances in analytical instrumentation and detectors can be
applied to air pollution with appropriate testing and develop-
ment. The low concentrations of pollutants specified in the
Air Quality Standards require measurements at the limits of
sensitivity and specificity of many available procedures and
instruments. Greater accuracy is needed not only because of
legal and economic significance of the measurements, but also
for scientific studies of health and economic effects. The
dosimetric personal sampling approach to study of health
effects requires reduction in size and weight of sampling
devices.
A broad scope of investigations was originally proposed for
this project. However, limitations of funding that was made
available reduced the work to exploration and initial develop-
ment of devices for three gaseous pollutants: nitrogen dioxide,
reactive hydrocarbons, and vinyl chloride. Computer analysis
of air monitoring data to extract information on temporal
patterns of fluctuations also was explored. The work has been
fully described and references in manuscripts prepared for
appropriate scientific journals. These are submitted in the
-------
Appendices. In order not to preclude their acceptance and
publication, we request that they not be published with this
final report. Major aspects of each paper are given in the
text which follows.
- 5 -
-------
SECTION IV
SOLID ABSORBENT METHOD FOR SAMPLING AND ANALYSIS OF
NITROGEN DIOXIDE IN AMBIENT AIR
A small sampling device was designed to obtain time inte-
grated exposures to nitrogen dioxide in ambient air. One
basic objective of the research effort was to meet the need
for generating data on personal exposures as contrasted with
the presently available data obtained from continuous moni-
toring instruments at fixed locations. The ideal requirements
for a personal dosimeter sampling device were set up as
follows: (1) small convenient size which can be carried
during daily routine; (2) economical to make and analyze;
(3) accurate; (4) stable before and after sampling; and
(5) free of common interferences.
A solid absorbent system was selected as best meeting the
requirements. Some of the more commonly used absorbents were
examined as well as active metals and metal oxides. Copper
metal was chosen for extensive studies since it was compatible
with a number of possible analytical systems. Studies were
focused on the ability of the copper to absorb a reproducible
fraction of the incoming ambient nitrogen dioxide under a
variety of sampling conditions and ambient air conditions.
The final sampling tube design is shown in Figure 1. It is
small enough to be worn with a small sample pump without en-
cumbering the wearer.
Sample tubes were constructed of borosilicate glass and
-------
CRITICAL
ORIFICE \. 65mm
TO P' IMP —- ^ F agg |k _^.^ IH — SAMPLE
j / 20/30 MESH Cu \K INLET
6mm OD. X
8mm O.D. «2C03
FILTER
Figure 1. Sampling train for
-------
packed with copper shot. The copper was purified and acti-
vated by heating under a stream of hydrogen. The cooled
absorption tube could be stored for weeks before use. For
use the tube was opened and connected to a small pump. A
filter impregnated with potassium carbonate was placed on
the inlet of the adsorption tube, to remove sulfur dioxide
which was a negative interference. A sample was then drawn
through the adsorption tube for a period of time up to 24
hours. When small personal pumps are used the critical ori-
fice shown may be omitted, and the flow adjusted with the
pump mechanism. Upon completion of sampling the tube was
sealed and could be analyzed at the laboratory within 5
weeks.
Analysis was performed by catalytic reduction of the nitro-
gen containing compounds absorbed on the copper. A small
furnace as shown in Figure 2 was used to drive the absorbed
nitrogen compounds into the catalytic furnace where all
nitrogen compounds were converted to ammonia. Any acidic
compounds which could interfere were removed by the potassium
carbonate section of the furnace. The ammonia was then
analyzed by a Coulson Conductivity Detector. Calibration of
peak areas was accomplished by introducing a known volume of
standard silver nitrate solution into empty absorption tubes
and following the same heating procedure. This standard was
convenient and the nitrogen content was completely converted
to ammonia.
Parameters of the sampling and analysis system were studied
to ensure specificity and to determine recoveries. A total
of 84 sample runs were made. A reproducible stoichiometric
factor of 58.8% was found. Consistent linear responses were
found under a variety of sampling conditions as shown in
Figure 3. Storage effects also were studied and it was found
-------
TEFLON
DESORPTION 3WAY
FURNACE VALVt
1
^^j-u-^ , . _ mn n —
^*^J~~L-r 350°C H P i'"
c\A/flf;FI HK .-.. .. t
FITTING H2
INLET
600°C
NICKEL
CATALYST
900°C
K2C03
SCRUBBER
l
420° C
r -r*i
/
ABSORBER AND
CONDUCTIVITY
CELL
^
1 I
DISTILLED
WATER
RESERVOIR
ION EXCHANGE
BED
PUMP
COULSON CONDUCTIVITY
DETECTOR
Figure 2. Desorption and detection system.
- 9 -
-------
700 r
10 20 30 40
PEAK AREA
50
Figure 3. Performance of method for
different sampling conditions.
0 - Varying times of sampling 0.164 ppm
at constant flow rate.
A - Varying concentrations of N02 sampled
for a constant time at constant flow rate.
- 10 -
-------
that the tubes could be stored after sampling for up to 35
days and still have an acceptable, although decreased, pre-
cision. Analyses of ambient air samples collected for a
twenty-four period were compared with the data from the Cin-
cinnati air monitoring station and showed a 0.776 correlation
coefficient.
In conclusion the copper absorbent method has shown promise
as a possible personal monitoring device suitable for time-
integrated sample collection. The device is relatively in-
expensive and can be reused. The samples may be shipped to
a central laboratory for analysis by skilled technicians.
The procedural techniques developed may be applied to a number
of other pollutants such as S0~, H_S or HC1 with the proper
selections of absorbents and sample analysis trains.
This work is completely referenced and supporting data are
included in the Appendix.
- 11 -
-------
SECTION V
CONTINUOUS MONITORING INSTRUMENT FOR "REACTIVE"
HYDROCARBONS IN AMBIENT AIR
The objective of this effort was the development of a
monitoring instrument with specific responses to the classes
of hydrocarbons with the greatest potential for forming
photochemical smog. Common instruments are used to monitor
total hydrocarbons or total hydrocarbons minus methane.
Either of these measurements is only partially successful as
a predictor of the severity of the smog that later develops.
Two two main stages of this study were the selection and
characterization of a solid chemical absorbent capable of
removing the olefinic and aromatic hydrocarbons primarily
responsible for smog, and the construction and testing of a
dual monitoring instrument incorporating this absorbent
which determined both total and reactive hydrocarbons.
The instrument, shown schematically in Figure 4, was
assembled from commercially available components: a Beckman
Model 109A Hydrocarbon Analyzer, a thermostated valve oven
(Carle Model 4301) containing the absorbent, an 8-port micro-
valve (Carle Model 2012) turned by a valve actuator (Carle
Model 4201) operated by a timer (Carle Model 4102). Two
equal flows of sample air were passed through the absorbing
column and an empty column. The output from each column was
alternately directed by the valve to the flame ionization
detector or vented in a timed five-minute cycle.
- 12 -
-------
VALVE OVEN
-JU,
FID
SAMPLE
INLET
fil
BECKMAN HYDROCARBON TRAP WATER
ANALYZER BUBBLER
Figure 4. Schematic of the hydrocarbon monitoring system.
- 13 -
-------
Candidate absorbents were screened for their ability to
remove olefins from a sample stream without affecting paraf-
fins. A chromium trioxide-sulfuric acid mixture on 60-80
mesh chromosorb possessed the most desirable absorbent
characteristics and was selected for more detailed study.
The effects of moisture, flow rates and temperature on the
ability of the absorbent to remove various hydrocarbons were
examined. Figure 5 summarizes some exploratory studies with
propane and propene and shows the effects of temperature and
contact time. It was found that to obtain reliable results
it was necessary to humidify the air sample stream by passing
it through an upstream bubbler containing water. Under con-
stant humidity conditions, the responses for important hydro-
carbons were determined. The absorber characteristics can
be adjusted by changing the absorbent column temperature or
dimensions. The optimal temperature for the absorber was
74°C, at which the response characteristics best matched the
photochemical reactivities of various types of hydrocarbons.
Ambient air was monitored for hydrocarbons for more than a
month from a second-story window in the laboratory. Figure 6
shows a typical recording giving the usual pattern for total
and reactive hydrocarbon concentrations for elevated morning
and evening levels. Short-term studies indicated the method
was practical, reliable and accurate and can provide new in-
formation about air quality.
Complete details are given in the Appendix.
- 14 -
-------
§1
I
<0
1.0
.8
.6
1.0
.8
.6
.4
85°C
95° C
I05°C
I20°C
50°C
70°C
0.0 .010 .015 .020 .025 .030
TIME (minutes)
Figure 5. Preliminary studies of effects of temperature
and contact time on the absorbent performance.
- 15 -
-------
8.0
6.0
fc
4.0 1-
I
2,0|-
O1—•-
6PM 8PM 10PM 12PM 2AM 4AM 6AM BAM 10AM
TIME
Figure 6. Recording from instrument of ambient air in Cin-
cinnati on June 2, 1973. Upper half of each 5-
minute cycle shows total hydrocarbon concentration,
and the decrease to the lower half is the reactive
hydrocarbon concentration.
- 16 -
-------
SECTION VI
CALIBRATION OF A GAS SAMPLING VALVE
FOR GAS CHROMATOGRAPHY
The inultiport gas sampling valve with a sample loop of
known volume is in widespread use for chromatographic
analysis of gas mixtures. The volume of sample which is
actually passed into the gas chromatograph usually differs
from the loop volume stated by the manufacturer, because it
also includes the volume of all of the connecting tubing and
dead space in the valve. These vary with custom mounting or
design differences. In many analytical situations the true
sample volume need not be known exactly, since comparisons
are made between identical volumes of standards and samples
in the calibration procedure. However, sometimes, it is
necessary to know the sample volume. In the present case, it
was desired to compare samples of vinyl chloride in air with
standards of vinyl chloride in a solvent introduced by
syringe injections onto the same column. The exact volume
of the loop was therefore determined by a novel application
of the method of standard additions.
The stainless steel loop labelled 2cc was removed from the
instrument, and filled with water from a 10 ml buret. The
tube was dried and the operation repeated twice. The average
volume was determined as 2.43 ml. Three more loops were made
from 1/8-in. copper tubing, and their volumes determined in
the same manner as 0.66, 0.32 and 0.22 ml. Then a fixed
- 17 -
-------
gaseous concentration of vinyl chloride was sampled succes-
sively with each of the loops in place in the gas chromato-
graph, and the corresponding peak heights were determined.
The gas mixture was provided by a permeation tube and flow
dilution system although a plastic bag also could have been
used. It was not necessary to know its exact concentration,
but only that it was constant and within the linear dynamic
range of the detector.
A plot was made of the peak heights as ordinate vs. the
measured sample loop volumes as abscissas. A straight line
fitted by the method of least squares was extrapolated to the
X axis, which it intersected at -1.38 ml. Thus the permanent
volume was determined to be 1.38 ml, and the actual volume
of sample 3.81 ml rather than the nominal 2 cc. This method
was convenient, did not require shutting down the instrument
or removal of plumbing other than the loop, and required only
a buret or balance for the calibration. Complete details are
given in the Appendix.
- 18 -
-------
SECTION VII
USE OF PERMEATION TUBES FOR CALIBRATION
OF VINYL CHLORIDE ANALYSES
Recent recognition of the carcinogenic hazards associated
with vinyl chloride exposure generated the need to monitor
ppm levels in work areas and ambient air. Accurate analysis
is essential. The Environmental Protection Agency recently
published a tentative method of analysis for ambient air con-
centrations similar to that recommended by the National
Institute for Occupational Safety and Health for occupational
areas. Both methods were based on adsorption of vinyl
chloride on charcoal, desorption into carbon disulfide and
injection into a gas chromatograph for determination with a
flame ionization detector. Quantitative results were ob-
tained by comparing the sample responses to those from known
concentrations. The accuracy of the method was thus depen-
dent upon the accuracy of the standards as well as of the
sampling procedures. Standards were prepared by gravimetric
and volumetric procedures. Gravimetric techniques were used
to determine the weight of vinyl chloride absorbed when it
was bubbled through toluene; aliquots of this solution were
diluted with carbon disulfide. Standards were also prepared
volumetrically using measured volumes of the vinyl chloride
gas which were injected directly into carbon disulfide.
Both of these methods were somewhat inconvenient and subject
to errors from volatilization.
- 19 -
-------
Permeation tubes have proven useful in preparing continuous,
low level concentrations of a variety of gaseous compounds.
Their performance with each individual compound, however,
must be determined experimentally since a stable permeation
rate cannot be achieved if the tubing is affected by the
permeant gas or if the liquified contents polymerize, decom-
pose or react with the dilution air. The purpose of this
work was to prepare and evaluate permeation tubes for gener-
ating accurate gaseous vinyl chloride standards.
Permeation tubes were made from 0.250" O.D. x 0.030" wall
thickness tubing of both FEP (fluorinated ethylene propylene
copolymer) and TFE (tetrafluoroethylene polymer) Teflon. A
long-lived tube also was constructed using an 8 ml stainless
steel bottle. After the tubes were filled with liquid vinyl
chloride the ends were plugged with a Teflon rod secured with
Swagelok ferrules.
Permeation rates were determined gravimetrically over periods
up to 8 months, during which the tubes were held at three
temperatures in a cyclical pattern. Several weighings were
made at each temperature and the slopes of plots of weights
versus age were used to determine the rates. Results are
shown in Figure 7, which is an Arrhenius-type plot in which
the abscissa scale is proportional to the reciprocal of ab-
solute temperature, and the ordinate is a logarithmic scale
of permeation rates. Lines were fitted to the data by the
method of least squares. The rates at 25°C for FEP and TFE
Teflon were respectively 0.259 and 1.98 yg/cm min., and from
the slopes the activation energies of permeation were cal-
culated to be 15.0 and 11.9 kcal/g mol.
The monomeric liquid vinyl chloride remained stable in the
permeation tubes which were exposed to flows of dry dilution
air for periods as long as eight months. A stable permeation
- 20 -
-------
5.0
Qj 4.0
.1 3.0
£ 2.5
2.0
s
1.5
1.0
^ .70
^ -50
g-40
2 -30
^ .25
.10
20
J L
'''I 1 1 L
25 30
TEMPERATURE,°C
35
Figure 7. Permeation rates of vinyl chloride in Teflon
tubes of 0.250" o.d. x 0.03" wall thickness.
+ FEP Teflon, 14 cm long
x TFE Teflon, 20 cm long
0 TFE Teflon, 11 cm long, with 8 ml.
stainless steel bottle
- 21 -
-------
rate was maintained when the tube was kept at a constant
temperature. For greatest accuracy, it was found desirable
after a temperature cycle to determine the new permeation
rate. The relative standard deviation of the permeation rate
for a 2-month period was 1.1 percent.
These gaseous standards can give very precise calibrations
when the gas sampling valve is also used for gaseous samples
for injection into the column in an identical manner. Al-
ternatively, the gaseous standards can be used with the
charcoal tube sampling procedure in which a liquid sample is
injected into the instrument.
The vinyl chloride permeation tube was found to be a con-
venient and accurate device for making low level mixtures of
vinyl chloride in air. Application of the permeation tube
system is not limited to generation of standard samples; the
device can also be used for animal exposure studies since
the tube lifetimes are on the order of two-thirds of a year.
The permeation device is also one of the safest ways to
handle a toxic agent such as vinyl chloride in that an acci-
dental rupture of the permeation tube will release on the
order of 50 g of the gas whereas a cylinder leak may release
kg quantities of vinyl chloride. Full details are given in
the Appendix.
- 22 -
-------
SECTION VIII
SAMPLING AND ANALYSIS OF AIRBORNE VINYL CHLORIDE
The Environmental Protection Agency is in the process of
formulating standards for vinyl chloride monomer (VCM) in
ambient air in the vicinity of its processing plants. The
Occupational Safety and Health Administration (OSHA) has
set standards for the working environment of 1.0 ppm VCM
as a time weighted average. Great public concern and debate
brought about because of the implication of VCM as a human
carcinogen has accelerated the standards setting ahead of the
demonstrated capability for its accurate sampling and analysis,
Therefore, the performance of the charcoal tube method was
critically studied.
A flow dilution system was set up using a vinyl chloride
permeation tube (see Section VII), with a dynamic range of
0.5 to 50 ppm VCM. Gaseous samples were drawn from the
manifold through charcoal tubes. Gaseous samples were also
taken by means of a gas sampling valve on a gas chromatograph
both directly from the manifold and downstream from the
charcoal tubes. This experimental design provided semi-con-
tinuous monitoring and accurate direct calibration with a
standard gas source. Commercial tubes (Mine Safety Appliance
Company) made for personal monitoring devices and containing
150 mg of charcoal were used throughout the study. All
analyses for VCM were performed on a gas chromatograph
equipped with a flame ionization detector.
- 23 -
-------
The initial studies were with dry air (less than 10%
relative humidity) containing vinyl chloride concentrations
of 5, 25 and 50 ppm. Breakthrough volumes were determined
at three different sampling rates.
Table 1 represents the results. The 50 ml/min. sampling
rate was suboptimal and showed a reduced retention volume.
Neither of the 100 ml/min. and 150 ml/min. sampling rates
exhibited unequivocally larger retention volumes. However,
100 ml/min. was recommended for reasons of convenience and
sampling time. At this rate a sample volume of approximate-
ly 10 liters was collected during 100 minutes before 10%
breakthrough was reached at 5 ppm VCM.
Recovery of the absorbed VCM from charcoal also was studied.
Because the source and treatment history of the charcoal can
make a difference in both absorption and desorption, the
charcoal used in this study was all from the same batch. It
was found that about 87.5% of the VCM that went into the tube
could be recovered upon analysis within 2 hours of sampling.
Storage of the samples for a period of one week at room tem-
perature decreased the recovery to about 80%. Longer storage
time decreased recovery only of low concentration samples;
thus after two weeks average recovery was as low as 71% for
samples containing 2.5 yg of VCM. Storage time, temperature,
sample concentration, and method of tube closure were all
important variables affecting recovery.
The charcoal in the commercial tubes was packed in two
separate sections, 100 mg in the entrance section, and 50 mg
in the rear. These were made for sampling of many compounds,
so that if the two sections are analysed separately, absence
of significant amounts in the second section demonstrates
that breakthrough has not occurred. For VCM, this technique
was successful when samples were analysed without delay.
- 24 -
-------
CO
w
ffl
D
EH
kH1
O
U
cd
PC;
o
PM
<:
EH
<
Q
S
O
H
EH
S
W
EH
tf
*
rH
0)
rH
X!
(d
EH
*
rH CO
oj w tn
-P (0 3.
0 g
EH
cn (N v£)
• • «
r^inro «^VOCN mrH
CNCNP- OinrH COrH
rHr-iro m^rro CNC-O
rH H rH CN
*
£
o
•H
-P 0) •
C g G
0) -H -H
-P -H g
0)
OS
ocom cocot^- OH
ocTicn mcNro coco
CN rH rH CN rH rH rH
^
c
O dJ en
•H g SH
-p 3 d)
C H -p
0) O -H
4J >rH
CD
«
OCOrO CTlCOin OrH
• •• ••• •*
ocno> r~-cNO crico
H CN CM CN rH
0)
S -P •
U (t) G
> SH -H
g
m >> N.
UJ t> \^
cn o en
rO rH 3.
S 4-1
(Tl CO <&
ror-rH CTiCOCO COCO
VDCNCTI rHrom ror-
OrHrH rO*J3Ch W3CN
rH
0) G
rH 0) -rH
Ci-P g
g fO\
(0 M rH
CO g
OOO OOO OO
inom moin mo
rH rH rH H rH
fll O
U> ^4
rH ti Ol
>^-H
C M •
•H O O
> rH G
-d 0
O O
ininin mmm oo
CN CN CN in m
rH
cn
CO
rH
r-
•
CO
cn
•
QJ
X!
3
-P
HH
O
CO
C
^ O
rH -H
-p
o
Q)
W
-P
CJ
o
M
m
«3 g
rH O
i *
• >-l
cn m
rH
rj
Cn
^
O
M
f—t
4J
rX
o rrj
in 0)
•H M
A
cP
O
rH
i 1
-M
<
*
o
in
- 25 -
-------
However, appreciable migration occurred during storage at
room temperature, and after 2 weeks recovery of 32 yg was
79% from the front, and 21% from the rear section, which was
not far from the 67% - 33% distribution of the weights of
charcoal. Migration of VCM can be retarded by storage under
refrigeration. However, presence in the second section
cannot be regarded as a reliable indication of an invalid
sample because of breakthrough.
Studies of the analytical procedure showed a definite re-
lationship between the final recovery and the desorption
procedure used. Recovery from tubes stored for two weeks
at room temperature depended on whether the two tube sections
were combined in one sample vial for desorption with carbon
disulfide (90%) or separately desorbed in two vials (total
recovery 78%). Thus, if breakthrough was to be checked on
each tube it had to be stored under refrigeration and its
sections analysed separately, with possibly greater losses.
If stored at ambient temperatures, the tube should be
analysed as a unit.
The tubes were convenient and inexpensive but required
recovery studies for each charcoal batch and particular
attention paid to details of analysis. Complete details
are given in the Appendix. There is a need for a small de-
vice to replace the charcoal tube which has better recovery
and is suitable for extended sampling times.
- 26 -
-------
SECTION IX
FOURIER ANALYSIS OF AIR MONITORING DATA
The proliferation of pollutant monitoring activities is now
providing massive amounts of data. A major problem has been
to obtain the "signal" from the data in the presence of large
amounts of "noise" produced by environmental fluctuations.
Computers have been utilized to provide information on the
statistical distributions of the numbers. Tabulations also
have been presented of data averaged by time of day and/or
by season. The purpose of this effort was to explore the
application of another technique, Fourier analysis of data,
which offers the promise of extracting significant new types
of information.
Monitoring data shows somewhat irregular fluctuations with
time of the concentrations of air pollutants. These can be
regarded as the sum of a mean value, and of a series of
fluctuations of differing periods, amplitudes, and phases.
In the analogous mixture of colors of visible light, resolu-
tion into a spectrum can be accomplished by a spectroscope.
Recent developments in computer programs make feasible
Fourier analysis of air monitoring data to obtain spectra
which can serve as equivalent models of the pollutant patterns
and aid in understanding atmospheric processes.
Exploratory calculations were carried out on Cincinnati data
with a Wang computer and plotter, which was conveniently
- 27 -
-------
available. The results were tabulated and plotted. Gome
plots are shown in Figures 8 and 9. Surprisingly there is
a dip in the amplitudes at a period of seven days, although
there are peaks at 3 1/2, 6, and 8 days. Phase information
was also tabulated and plotted by the program. The similar-
ity of patterns for pollutants coming from entirely different
sources is remarkable. This suggests that atmospheric dis-
persion processes exert a major role in determining the levels
of these pollutants.
More complete analysis appears justified by these explora-
tory results. A large computer capable of handling hourly
data for a year is required, and a program is being prepared
for the IBM-370-166. Future work on more data from more
cities will clarify the significance of the results. In-
teresting possibilities for analysing and modelling atmos-
pheric processes are opened up by this technique. Full
details are in the Appendix.
- 28 -
-------
6 12
(hours)
4
(days)
8
PERIOD
16
Figure 8. Fourier amplitude spectrum of concentrations
of total hydrocarbons in Cincinnati for
Oct.-Dec. 1968 (as reported by CAMP).
- 29 -
-------
0.
6 12
(hours)
PERIOD
Figure 9. Fourier amplitude spectrum for concentrations
of sulfur dioxide in Cincinnati for October,
1968 (as reported by CAMP).
- 30 -
-------
SECTION X
LIST OF PUBLICATIONS AND MANUSCRIPTS*
(Appendices)
A. Cuddeback, J.E., B.E. Saltzman, and W.R. Burg,
Solid Absorbent Method for Sampling and Analysis of
Nitrogen Dioxide in Ambient Air. J. Air Poll. Control
Assoc. (in press) (1975).
B. Saltzman, B.E., W.R. Burg, and J.E. Cuddeback.
Continuous Monitoring Instrument for "Reactive" Hydro-
carbons in Ambient Air. (submitted to Analytical
Chemistry).
C. Birch, S.R., W.R. Burg, and J.E. Cuddeback.
Calibration of a Gas Sampling Valve for Gas Chromato-
graphy. Anal. Chem. 47, 355 (1975).
D. Burg, W.R., S.R. Birch, J.E. Cuddeback, and B.E. Saltzman,
Use of Permeation Tubes for Calibration of Vinyl
Chloride Analyses. (to be submitted to Env. Sci. Tech.)
* Prepared with appropriate acknowledgments to grant for sub-
mission to scientific journals. In order not to preclude
their acceptance, these are submitted separately as
Appendices for confidential use until published.
- 31 -
-------
E. Cuddeback, J.E., W.R. Burg, and S.R. Birch.
Sampling and Analysis of Airborne Vinyl Chloride.
(submitted to Env. Sci. Tech.).
F. Saltzman, B.E., Fourier Analysis of Air Monitoring
Data. In Proceedings of the Symposium on Statistical
Aspects of Air Quality Data, p. 11-1 to 11-13. EPA-
650/4-74-038, October 1974, Research Triangle Park,
N.C.
- 32 -
-------
TECHNICAL REPORT DATA
(Please read tHUtuctions on the ret ersc bcjorc completing)
RGPOrtl NO
EPA-650/2-75-067
4 TITLt" AND SUBTITLE
Development of Air Sampling Methodology
3. RECIPIENT'S ACCESSION-NO
5. REPORT DATF
May, 1975
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Bernard E. Saltzman
8. PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Department of Environmental Health
University of Cincinnati
Cincinnati, Ohio 45267
i:> SPONSORING AGENCY NAMF AND ADDRESS
Chemistry £ Physics Laboratory
National Environmental Research Center
Research Triangle Park, N. C. 27711
10 PROGRAM ELEMENT NO.
1A1010 ROAP 26ACX-37
11. CONTRACT/GRANT NO.
800869
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
1b SUPPLEMENTARY NOTES
16. ABSTRACT
Solid absorbent systems were developed for analysis of three gaseous ambient
air pollutants. Nitrogen dioxide was sampled with small glass tubes packed with
copper shot, which could be used as personal samplers, and be stored before
analysis for up to 5 weeks. Reactive hydrocarbons were determined in a dual
analyser as the concentration absorbed on chromium trioxide-sulfuric acid. These
monitoring measurements should correlate better with photochemical reactivity than
those of non-methane hydrocarbons. Permeation tubes for vinyl chloride were
critically examined and successfully used to optimize the carbon tube sampling
method. Time patterns of monitoring data were examined by Fourier analysis on a
computer.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
AIR QUALITY MEASUREMENTS
ANALYTICAL METHODS
CHEMICAL METHODS
MEASUREMENT METHODS
MONITORING
SAMPLING METHODS
GAS SAMPLING
b.lDENTIFIERS/OPEN ENDED TERMS
SOLID ABSORBENTS
NITROGEN DIOXIDE
REACTIVE HYDROCARBON
VINYL CHLORIDE
POLLUTION DATA ANALYSIS
PERSONAL AIR SAMPLING
PERMEATION TUBE
COSATI 1 icld/Gioup
05-RES CRT
16-MEAS. METH.
17-AIR QUAL. MEAS
18-THEO
18-LAB
18 DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21
NO. OF PAGES
40
20 SECURITY CLASS (This page)
Unclassified
__
22 PRICE
EPA Form Z220-1 (9-73)
- 33 -
-------
33
o
:T
D
c
<
33
2. O!
O
-o
-o
O
DO
H
C
O
<
O
OJ
CO
z
m
CO
Z =• oi. O
o 3 :± —i
rt 5' o rn
^^50
r-i tu -j
ilfl
f as
-o
CO
N 2
" o
o 3-
Q- 03
.
CD
03 CD
Q- '-
. _.
en a.
13 °
,-,
D
O. fD
— O
O CD
CD CD
(d g
_CD g
— CD
13 —•
Q_ CO
C ~
Q.
O
-<
°
to H n
ui m m
o m
-H cn
m
2
n
------- |