xvEPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA-600/2-79-042
February 1979
Research and Development
Methodology for
Collecting and
Analyzing Organic
Air Pollutants
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are
1 Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5 Socio-economic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8 "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental 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 through the National Technical Informa-
tion Service. Springfield. Virginia 22161
-------
EPA-600/2-79-042
February 1979
METHODOLOGY FOR COLLECTING AND
ANALYZING ORGANIC AIR POLLUTANTS
by
Corazon Hastings Vogt
Environmental Trace Substances Research Center
Columbia, Missouri 65201
Grant No. R-801050
Project Officer
Eugene Saw1ck1
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency/ nor does mention of trade
names or commercial products constitute endorsement or recommendation for use.
-------
ABSTRACT
A number of support-bonded liquid phase sorption media were developed
and evaluated 1n model systems for collecting and analyzing organic air
pollutants. Polymers with various functional groups were synthesized and
chemically bonded onto inert supports In thick layers. A media consisting
of a s111 cone liquid bonded to Chromosorb W was used with excellent results.
Retention times of most organic compounds on this liquid are extremely long
at ambient temperatures, and sampling can be carried out for 24 hours at a
rate of 10 liters of air per minute. In contrast, subsequent counter current
liquid extraction takes only a few minutes since retention volumes are very
small. Extracts were analyzed largely by gas chromatography.
111
-------
CONTENTS
Abstract iii
Figures vi
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. Experimental Procedures 6
Construction of a prototype air sampler 6
Extraction apparatus for the loaded cartridge 8
Silicones bonded to diatomaceous earths as
collection media 9
Modified adsorbents based on silica gel 18
GC characterization of hydrothermally and
polymer modified silica gels 22
A gas chromatographic cartridge desorption port 26
Other polymer modified silica gels 32
Humidity and sampling flow rates 34
5. Results and Discussion 47
References 48
Publications 49
-------
FIGURES
Number Page
1. Typical set-up for atmospheric sampling. The deflated balloon
shown is used to roughly estimate the air flow rate for each
sampling experiment 7
2. Set-up for extraction of collected air contaminants. Freshly
distilled solvent enters the cartridge at the bottom and over-
flows back into the flask
3. Initial laboratory set-up for collecting organics on, and
extracting them from, support-bonded silicons phases contained
in the "trap", a, Test compound sampling; b, atmosphere sampling;
c, extraction 10
4. 20 ft. glass column, 3% OV-101 on Chromosorb-W H. P. 80/100 mesh.
Microtek MT-220 E. C. detector att. 32 x 10. Temperature 220°,
N flow 16 ml/min 14
5. Gas chromatograms of car exhaust components collected on two
identical cartridges containing 24% [CisHa78103,2] on Chromo-
sorb A. Sampling time, 60 min at 30 1/min. The injections
represent ca. 0.15 sec of sampling. GC conditions: 3% OV-101
on Chromosorb W-HP, 80/100 mesh, 20 ft. x 2.5 mm I.D. Pyrex;
nitrogen flow optimized, 16 ml/min; flame ionization detector. ... 15
6. 2 microliters injection representing 0.54 seconds of sampling.
20 ft glass column, 3% OV-101 on Chromosorb-W, H. P., 80/100
mesh. MicroTec MT-220, FID att. 8 x 10, EC 64 x 10. Temper-
ature: 80°. N flow 16 ml/min 16
7. (Industrial Area) 1 microliter injection representing 30.6 liters
of air in the industrial area, St. Louis, Missouri. 20 ft. glass
column; 2.5 mm i. d. 3% OV-101 on Chromsorb-W, H. P., 80/100
mesh. MicroTek MT-220, FID att. 1 x 8. Program: 10 min at
50° followed by 8°/min., final hold at 230°. N? flow 16 ml/min. . . 17
(Downtown Area) 1 microliter injection representing 19.6 liters
of air in the downtown area, St. Louis, Missouri. 20 ft. long
column; 2.5 mm i.d. 3% OV-101 on Chromosorb-W, H.P., 80/100 mesh.
MicroTek MT-220, FID att. 1x8. Program: 10 min. at 50° followed
by 8°/min., final hold at 230°. NZ flow 16 ml/min 17
8. Surface area of Davison Silica Gel 62 as a function of hydro-
thermal temperature. (A) Ref. 3 (B) This study, to be pub-
lished 19
VI
-------
9. Retention temperatures of model hydrocarbons on different
surface area silica gels 20
10. Thickness of non-extractable layers of organic polymers on
hydrothermally treated silica gel 21
11. Separation of even-numbered n-alkanes (octane through
octadecane) on bare silica gels. Both columns: Silica Gel 62,
40-60 mesh, acid washed; in 120 cm x 4 mm I.D. Pyrex U-tube,
nitrogen flow-rate 80 ml/min. 6°/rain temperature program, FID.
(A) Silica Gel 62 hydrothermally treated at 280° overnight and
acid washed; (B) untreated silica gel. The n-alkanes are rep-
resented by their number of carbon atoms 23
12. Separation of n-alkanol standards (methanol through pentanol)
on bare silica gels. All columns: Silica Gel 62, 40-60 mesh,
acid washed; in 120 cm x 4 mm I.D. Fyrex U-tube at 140°,
nitrogen flow-rate 80 ml/min, FID. (A) Bare silica gel; (B)
Silica Gel 62 hydrothermally treated at 280° overnight and
acid washed; (C) carrier gas saturated with water at ambient
temperature; (D) water-deactivated and hydrothermally treated
silica gel 24
13. Separation of n-alkanol standards (methanol through pentanol)
on Carbowax-coated silica gels. All columns: Silica Gel 62,
40-60 mesh, acid washed, coated with Carbowax 20 M, heat-
treated and extracted; in 120 cm x 4 mm I.D. Pyrex U-tube at
140°, nitrogen flow-rate 80/mlmin, FID. (A) Silica Gel 62
hydrothermally treated at 280° overnight and acid washed before
coating with Carbowax, carrier gas saturated with water at
ambient temperature; (B) water-deactivated silica gel: (C)
hydrothermally treated silica gel; (D) untreated silica gel 25
14. Separation of n-alkanols (methanol through hexanol). Column:
Silica Gel 62, 40-60 mesh, acid washed HT-2800, coated with
Carbowax 20 M; in 150 cm x 2 mm I.D. Pyrex U-tube, nitrogen
flow-rate 80 ml/min, temperature program 6°/rain, FID 25
15. Collection cartridge, holder and desorption port 27
16. Flow patterns of the valving system with some typical uses 28
17. Chromatographic comparisons of different sample introductions
to the valving system. Conditions: injected, 1 yl of
alcohol-hydrocarbon mixture: initial temperature, 50°; program
rate, 8°/min; temperature FID, 225°; analytical column, 3%
OV-17 on Chromosorb W, 100-120 mesh, 6 ft. x 2 mm I.D. boro-
silicate glass 30
vii
-------
18. Chromatography with the introduction flow pattern as shown in
position A of Fig. 2. 10-min collection in atmosphere enriched
with gasoline vapors. Solid adsorbent is hydrothermally treated
with gasoline vapors. Solid adsorbent is hydrothermally treated
(210°) silica gel 62 modified with non-extractable layers of
OV-101. Temperature: multiport introduction system, 190°;
FID, 225°. Analytical column: 10% FFAP on Chromosorb W, AW
DMCS, 80-100 mesh, 20 ft. x 2 mm I.D. borosilicate loop tube.
Attenuation x 2000 31
19. Gas chromatogram of hydrocarbons on silica gel, hydrothermally
treated at 210°C and modified with non-extractable layers of
OV-101 32
20. 1 microliter alcohol-hydrocarbon mix injected to 28 liters
N~ in teflon bag, contents pumped out in 70 minutes; solid
adsorbent is hydrothermally treated (210°) Silica Gel 62,
modified with non-extractable layers of OV-101; temperature
multiport introduction system 190°; temperature FID 225°;
analytical column is 3% OV-17 on Chromosorb W 100/120; x 1000;
6 ft. x 2 mm I.D. borosilicate U-tube 33
21. Relative degree of collection as functions of humidity and
flow rates 36
22. Relative degree of collection as functions of humidity and
flow rates 37
23. Relative degree of collection as functions of humidity and
flow rates 38
24. Relative degree of collection as functions of humidity and
flow rates 30
25. Relative degree of collection as functions of humidity and
flow rates 40
26. Relative degree of collection as functions of humidity and
flow rates 41
27. (a) Effects of humidity and sampling flow rates on collection
efficiency. Heat desorption introduction system, 190°; FID,
225°; analytical column: 3% OV-17 on Chromosorb W, 100-120
mesh, 6 ft. x 2 mm I.D. borosilicate; program rate, 8 /rain 42
(b) Same as Figure 27(a) 43
28. Five minutes collection behind a car exhaust with the engine
idling; solid adsorbent is hydrothermally treated (208°)
Silica Gel 62 modified with non-extractable layers of OV-101;
collection flow rate, 21/min; heat desorption temperature, 190°;
FID temperature, 225°; analytical column is 10% FFAP on Chromo-
sorb, W, AW, DMCS, 80-100 mesh; x 500; 20 ft. x 2 mm I.D.
borosilicate loop tube. 44
viii
-------
29. Ten minutes collection in an atmosphere of gasoline contained in
a beaker. Conditions same as Fig. 28 45
30. Fifteen minutes collection behind an idling station wagon.
Conditions same as Fig. 28 46
ix
-------
SECTION 1
INTRODUCTION
One usually thinks of air pollution in terms of its most obnoxious consti-
tuents: Sulfur dioxide, peroxyacetylnitrate, particulate matter, and so forth.
The area of organic compounds of higher molecular weight has, in comparison,
been much less investigated. Of course, we are familiar with the role of
hydrocarbons from automobile exhausts in the production of smog, with the
occasional severe agricultural damage done by pesticides translocated by wind,
and with the presence of potent carcinogens originating from combustion pro-
cesses.
The last few years have seen a tremendous increase in our knowledge on the
constitutents of air, not only because of increased public interest, but also
because of suitable, highly sensitive instrumentation which has become available
recently. Two examples in point may be the use of capillary columns in gas
chromatography, and the use of computer-interfaced high-resolution mass spec-
trometers connected via molecular separator to a gas chromatograph. Although
many compounds have been identified, it is safe to assume that a myriad more
exists in air at trace levels—beneficial, neutral, or detrimental to man.
Historically, the collection of air-borne organics (in gaseous, not partic-
ulate form) has often been a much harder task than their subsequent analysis.
Collection is generally performed by
a. condensation in cold traps, usually at the temperature of liquid
nitrogen or dry ice,
b. bubbling or impinging air on such compounds as ethyleneglycol or
hexyleneglycol which are subsequently extracted wtih an organic sol-
vent,
c. trapping on solid absorbents such as activated carbon, or
d. sorption on typical gas chromatographic phases of the liquid-coated
supported or the resin particle type.
These methods suffer from certain drawbacks. Labile compounds can de-
compose or rearrange on the highly catalytic carbon surface. The atmosphere
contains considerable amounts of water and sometimes the oxides of sulfur and
nitrogen in substantial quantities—all of which condense at liquid nitrogen
temperatures. Furthermore, the cryogenic methods need suitable technical equip-
ment and are difficult to operate under field conditions. GLC-type phases must
be heated to re-vaporize the collected materials and generally permit only one
chromatographic run. The literature describing the techniques for, and the
difficulties associated with, the collection of airborne organics is quite ex-
tensive.
-------
This report describes the development of solid media for the collection of
organic pollutants in air. It includes the work accomplished between the sub-
mission of the proposal and the final award of the EPA grant R-801050. In the
following pages, highlights of typical achievements will be presented briefly.
For a complete list of achievements, please refer to the list of publications
supported by this grant.
-------
SECTION 2
CONCLUSIONS
The ideal method for collecting organic air pollutants would be one which
is not affected by water and acidic inorganic compounds in the atmosphere,
which can be easily used at remote locations, minimizes decomposition of the
collected substances, and collects organics in a form suitable for various
types of subsequent analyses. The collection of all organic air pollutants—
except those of very high volatility—with support-bonded liquid phases repre-
sents a superior method compared to other techniques of collection.
A highly loaded, support-bonded silicone liquid phase—ca. 30% to 40%
(Ci8H3703 .2) on Chromosorb-W 30/60—was used with excellent results for the
collection of organics from air under various circumstances. Retention times
of most organic compounds on this phase are extremely long at ambient temper-
ature, and sampling can be carried out for 24 hours at a rate of 10 liters per
minute—with only the most volatile compounds breaking through at the exit of
the small cartridge containing the support-bonded polymer. In contrast to
these long retention times, retention volumes in liquid extraction are ex-
tremely small and the countercurrent extraction of a cartridge is completed in
a few minutes.
For the purpose of liquid extraction, a prototype apparatus which allows
the continuous recycling of low-boiling solvent was constructed. This
approach had become feasible only through the fact that the 40-odd percent of
silicone and other polymers are completely support-bonded and none of it is
extractable, no matter what solvent (including alcohols, water or acids) is
used. The low boiling point of the extracting solvents—preferentially pentane
or dichloromethane—allows easy concentration of the extracts, averts
evaporation losses and minimizes decomposition of heat-labile compounds.
Pentane is an ideal solvent for later use with the alkali flame or electron
capture detectors, dichloromethane is very well suited for the flame ionization
detector.
The organic compounds enrich in solid collection media could be released
not only by liquid extraction but also by heat desorption. The later is pre-
ferred when the collecting media is critically tested for inertness and/or
catalytic decomposition. Details of the collection and desorption ports and
the various simultaneous flow patterns of the system are shown in the report.
Initially, chromatograms for the different modes of sample introduction—direct
injection to GC, direct injection through the hot cartridge, and heat desorp-
tion of cartridge loaded with test mixture from doped air—is compared to
establish the performance of the system. The well-known technique of venting
(solvent or solute) can serve to introduce only desired retention ranges from
-------
a loaded cartridge into the gas chromatograph; loop injection is facilitated by
the two valve system. In actual field sampling, the collection assembly has
the ease of portable handling and the "loaded" cartridge can easily be trans-
ported to the laboratory.
Support-bonded liquid phases were extended from the polysiloxanes to
other polar and non-polar polymers, i.e. Carbowax 20M, AN-600, ON 870, FFAP,
OV101, SE-30 and linear polyethylene. Further, the solid support bonded to
the organic moiety was extended to the naturally occuring diatomaceous earths
to the man-made silica gels. The latter materials were modified by treating
with water under high pressure followed by bonding with organic polymers.
A wide series of different surface areas, large pore, adsorbents have been
prepared from available silica gel. The materials were deactivated to produce
semi-selective phases. The report described the technique used to vary the
surface area of the gel and the deposition of organic surface modifiers. The
potential use of the otherwise too retentive silica gel as support for chroma-
tography is characterized by the GC performance of polar and non-polar com-
pounds. The retention volumes of polar solutes (which are irreversibly adsorbed
on the authentic silica gel) are correlated to the size of the pores, i.e., sur-
face areas, on one hand, and/or the properties of the surface modifier on the
other. Non-polar solutes though affected less significantly, generally become
early eluters as the accessible areas decrease and in the absence of deactiva-
tors. Applications of the adsorbents as solid media to collect organic pollu-
tants in air as well as industrial organic vapors in working atmospheres have
been investigated under a wide range of conditions. The cutoff of trapping
efficiency or the breakthrough, of the vapors, are studied as functions of such
parameters as volatility and structure of the compounds, flow rate of sampling
and humidity of the atmosphere.
The most promising collection media were used for sampling of (a) standards,
including chlorinated hydrocarbon insecticides, (b) automobile exhaust, and
(c) the relatively clean air of the University of Missouri campus at Columbia.
Standards were quantitatively recovered; virtually no heavier organic compounds
came through the cartridge in spite of hugh amounts of water and a heavy load of
hydrocarbons emitted from the exhaust pipe; and finally, even the clean campus
air was shown to contain a multitude of peaks. These results certainly indicate
the broad potential of our appproach to the analysis of air pollutants.
-------
SECTION 3
RECOMMENDATION
Based on the work done under this grant, we can make the following
recommendations:
SOLID COLLECTION MEDIA
Solid collection media are recommended for the collection of volatile
organics in air. These media are far superior to the liquid media used in im-
pinger type samplers. They offer better collection efficiency, greater ease of
handling, more flexibility in manupilation of analysis,and longer life.
SUPPORT BONDED PHASES
The use of solid collection media with bonded liquid phases are recom-
mended over simpler media such as charcoal and untreated silica gel. Solid
support-bonded phases are superior to solid media without a liquid phase
because they minimize decomposition of the pollutants of interest, increase the
collection efficiency and capacity of the media and provide less sensitivity to
changes in humidity and collection flow rates.
LIQUID EXTRACTION
Continuous liquid extraction is recommended for removal of the sample from
the collection media when sample decomposition is a serious concern and when
analysis by more than one technique is required. A continuous extracting ap-
paratus which is designed to the fit the collection cartridge allows each concen-
tration of the extracts, averts evaporation losses and minimizes decomposition
of heat-labile compounds.
HEAT DESORPTION
Heat desorption is recommended for fast routine testing where only a
single GC analysis is needed, where the variable parameters are known and where
sample decomposition is not a serious concern. For this type of analysis, heat
desorption is more cost-effective than liquid extraction. Additional studies
of thermal desorption units are recommended in order to minimize sample de-
composition. Such sample decomposition may be caused by contact with metals or
insulating materials in the desorption unit.
-------
SECTION A
EXPERIMENTAL PROCEDURES
CONSTRUCTION OF A PROTOTYPE AIR SAMPLER
The efficiency of collection depends on several factors. If a certain
volume of air is passed through a cylindrical cartridge packed with solid
media whose surface is modified with organic liquid layers, the amount of air
contaminant collected depends on the partition ratio of the compound at the
prevailing temperature as well as the dimensions of the cartridge and its
pressure drop from inlet to outlet. Given a certain cartridge volume, a
longer cartridge will retain more. However, if a steep pressure differential
is necessary to maintain the desired flow, some of the advantages of the
longer cartridge may be lost.
In an earlier study (1), the enrichment of organic air contaminants was
done by sampling with a single cartridge. To determine which compounds are
partially, and which are completely retained under defined sampling conditions,
a second cartridge can be inserted between the first cartridge and the pump.
A portion of the present study pursued almost exclusively the two-
cartridge approach, mainly because comparison of compounds collected by the
first and second cartridges can give important information on collection
efficiency and characteristics. The two phases may or may not be the same,
depending on the experiment.
Therefore, a two-cartridge sampling apparatus providing 30 1/min air
flow rates and featuring easy cartridge changes was constructed. This is
shown on Fig. 1. The double-cartridge sampling approach was used throughout
the experiments to indicate the breakthrough range (which occurred, in most
cases, around Kovats index 1000, i.e., n-decane). Oil-free pumps were con-
nected with Cajon fittings and flexible stainless-steel tubing to two car-
tridges containing the supported-bonded phases, held in place by stainless-steel
screens and glass wool. The flow rate was estimated by measuring the diameter
of inflated balloons attached to the pump for a measured time. Air entered
the system through a teflon Millipore filter of 5-y pore size, which was
supported by a stainless-steel screen in a modified Millipore filter holder.
The cartridges were Chemical Research Service "Hydrocarbon traps." Teflon is
the preferred material for the packing rings; if silicone rubber is used, it
should be covered by teflon tape to avoid extraneous GLC peaks. The 0-rings
in the fittings are somewhat less prone to cause trouble. All Cajon Fittings
and Flexible tubing were % in. nominal, making for facile set-up, take-down or
interchange, of the system components.
-------
3
O
•a
u
3i
3
.2
G
*
O
M
C
O
•a
u
4-*
rt
5S
u
-
" a
•r E
u a!
O «
e "
-M O
u
H u
-------
EXTRACTION APPARATUS FOR THE LOADED CARTRIDGE
Fig. 2 shows the set-up used for extraction of collected organics from
the traps. The three-necked flask had Clearfit joints, precluding the need
for grease. A high-vacuum jacket surrounded the distillation column to avoid
excessive reflux. Teflon tubing and adapter ferrules (Scientific Research
Service) were used for connections via Swagelok fittings with % in. tube ex-
tensions. Since these extensions match the Cajon fittings on the cartridge,
the latter could be inserted or removed without the use of tools in less than
a minute. A brisk flow of solvent pentane or other liquids of high purity and
volatility—was maintained for approximately 15 min after the cartridge over-
flowed into the flask.
The pentane extract was concentrated, first in a flask and then in a
graduated tube, to ca. 0.3 ml or less by blowing dry, pure nitrogen at the
surface of the liquid. If further concentration was required, the extract was
transferred to a capillary tube and concentrated with nitrogen introduced
through a syringe needle.
Figure 2
Set-up for extraction of collected air contaminants Freshly distilled solvent enters the
cartridge at the bottom and overflows back into the flask
-------
SILICONES BONDED TO DIATOMACEOUS EARTHS AS COLLECTION MEDIA
Support-bonded silicones possess three characteristics for collecting,
and thus concentrating, small amounts of organics from the atmosphere or
natural water systems: They are inert and hydrophobic, they can be produced
in very thick layers and various compositions and, most importantly, they are
non-extractable by any conceivable solvent short of sulfuric acid. Thus, the
atmosphere or water can be passed through a bed of a suitable silicone phase,
and the phase later extracted with an organic solvent to release the collected
organics. This approach has been shown to allow fairly easy analysis of dif-
ficult-to-analyze gaseous air pollutants in the parts per billion range, and
certain water pollutants (chlorinated hydrocarbon insecticides, PCB's, etc.)
in the parts per trillion range. These studies were highly successful.
A method for the collection of organic contaminants from air and other
gases, which makes use of these novel materials is demonstrated. Silicone
polymers are chemically bonded to diatomaceous particles, e.g. 26 weight %
(CieHaySiOa ,2) to Chromosorb W, 30/60 mesh. These materials, packed into two
cartridges connected in series, remove organics from fast-flowing gas streams.
The collected compounds can then be extracted and used to analytical or
preparative ends. This phase was used to sample the atmosphere in St. Louis
and Columbia, collection of organics from car exhaust, and removal of chlor-
inated hydrocarbon insecticides from air. Examples of analysis include dual-
channel EC/FID gas-liquid chromatography and GLC mass spectrometry.
Fig. 3 provides a generalized picture of the apparatus configurations of
the initial experiments. Lumped together in one prototype, the schematic
shows how the cartridge is used for the collection of test compounds (flowpath
a), for the collection of air contaminants (b), and for extraction with pentane
(c). Since the various parts can be easily connected and disconnected, only
pertinent elements of the set-up were used in the following experiments.
Cartridges and Phases
A commercial, charcoal-containing "hydrocarbon trap" (Chemical Research
Services, Addison, 111.) was emptied and the aluminum hull filled with support-
bonded phases held by stainless-steel screens. The end-pieces were drilled
out to accept % in. Cajon fittings. Similar fittings were put on a Millipore
filter holder and an oilless Bell & Gossett pump, so that the whole system
could be connected with % in. o.d. Cajon stainless-steel flexible tubing.
(This tubing rusts profusely and should be treated with care.)
Several types of supports, coated with polymer of various percent loads,
were used to fill the cartridges. Octadecyltrichlorosilane was the monomer
for all liquid phases mentioned in this paper, however, other types of poly-
mers are equally suitalbe. Support bonding and polmerization were carried
out as reported earlier (2). The raw phases were then extracted with benzene
in a soxhlet for at least 30 hours with as short a cycle time as was con-
sidered safe. This procedure removes any non-support-bonded material. A
small amount of the dried phase was weighed, kept in a muffle furnace at 1000°
for three hours, and reweighed to obtain an estimate of liquid phase load (=
"ignition"). The bulk of the dried phase was filled into a cartridge and
-------
air
CaCl2
ether trap,
dry iot bath
oil-free pump
Initial laboratory set-up for collecting organics on, and extracting them from, support-
bonded silicone phases contained in the "trap", a, Test compound sampling; b, atmosphere
sampling; c, extraction.
Figure 3
-------
extracted with pentane in a set-up according to Fig. 3 (c), then produced the
first blank before actual use for sampling.
The various phases serving in this investigation had the following minimum
liquid loads as determined by ignition: 14 and 16% on Chromosorb G 45/60, 24%
on Chromosorb A 20/30, 28 and 31% on Diatom W 20/40, and 26% on Chromosorb W
30/60 mesh. A typical experiment is described below.
Collection
Atmosphere Samples—
The schematic presented in Fig. 3(b) was carried out in field sampling as
follows: The Millipore filter holder was connected with flexible SS tubing to
the first cartridge, the first to the second cartridge, and the second car-
tridge directly to the pump. The approximate air flow through the system was
measured by attaching deflated balloons to the pump exhaust and calculating
the air flow after a certain time interval from the diameter of the inflated
balloon. Sampling was carried out in the downtown section of Columbia, and
downtown and industrial sections of St. Louis.
Car Exhaust Samples—
A similar arrangement was used to sample car exhaust, the filter holder
being connected fo a funnel held, upside down, just above the end of the
exhaust pipe. The warmed-up V-8 was idling during the experiment.
Other Test Samples—
Using the bubbler shown in Fig. 3(a), two types of mixtures were sampled.
First, a small amount of gasoline—the same as used in the car exhaust experi-
ment—and second, a mixture of common chlorinated hydrocarbon insecticides.
In the latter case, a slow stream of nitrogen swept through the heated bubbler
and carried the substances into a much larger stream of laboratory air. This
air was drawn in through an activated carbon filter, doped with sample in the
bubbler as described, and flowed then through the Millipore teflon filter, the
two cartridges, a dry ice trap, and finally, the pump. The use of air rather
than nitrogen was chosen to combine better simulation of atmospheric sampling
with reduced expenditures.
Thermostating of the trap, as shown in Fig. 3, was used only in pre-
liminary tests; all experiments mentioned were run at ambient temperatures.
Extraction
The schematic shown in Fig. 3(c) was followed. The three-necked flask
with Clearfit joints (no grease) was charged with at least 200 ml "chromato-
graphic ally pure" pentane and brought to a rapid boil. When the cartridge had
filled up and began discharging pentane back into the flask, a brisk flow was
maintained for ca. 15 minutes. The solution from the flask (excluding the
solvent left in the cartridge) was concentrated to 0.3 ml in a tube suspended
in water of ambient temperature, by blowing dry nitrogen at the surface of the
solution.
11
-------
The extraction was performed in a countercurrent manner, i.e. the flow of
gas during sampling and the flow of liquid during extraction occurred in
reverse directions. Teflon tubing was used to connect the cartridge with the
reflux condenser and the fask via h o.d., 1/8 i.d. adapters, which fitted
into the Cajon terminals of the cartridge. Assembly and disassembly of this
set-up can thus be achieved in a few seconds and cartridges can be extracted
at a rapid rate.
Analysis by GLC
Several types of analysis based on gas-liquid chromatography were per-
formed, and they are described below in random succession.
All samples were first analyzed on a 20 foot, 2.5 mm i.d. glass column
filled with 3% OV-101 on Chromosorb W, H.P., 80/100 mesh, in a Microtec MT-
220 gas chromatograph equipped with a hydrogen flame detector. Samples and
blanks were usually followed by a mixture of n-hydrocarbons to allow a rough
estimate of Kovats indices under the particular temperature program chosen.
No such program was possible, of course, with the Ni-63 electron capture
detector. This detector was used in two configurations—alone, and in a
micro-capillary valve (Precision Sampling, Baton Route, La.) inside the column
oven to regulate the split ratio.
The pesticide samples were run on the EC detector only, the car exhaust
samples were analyzed with the splitting arrangement. Read on a dual-channel
recorder, the two detectors can yield some information about the polarity of
the eluting compounds.
An air sample from the Columbia campus, which contained a greater number
of FID peaks, was also run with an alkali flame and a flame photometric detec-
tor, but failed to exhibit peaks indicative of phosphorus-containing compounds.
The sulfur channel of the flame photometric detector was used in a temper-
ature-programmed analysis to show the presence of a few sulfur-containing
compounds in gasoline and car exhaust.
A car exhaust sample was further used to demonstrate its analysis by a
combined gas chromatograph-mass spectrometer (Packard-CEC HOB) system with a
Watson-Biemann separator and direct oscillographic recording.
Results and Discussion
All results lumped together, support-bonded silicone phases behaved well
and as expected, in the collection of organic vapors present in air or nitro-
gen. Obviously a range of compound volatility exists, for which this method
is best suited. Depending on a host of parameters—chemical and physical
nature of the phase, temperature, sampling rate and volume, dimension of and
pressure drop in this cartridge, etc.—hydrocarbons of lower molecular weight
will be either insufficiently retarded, or not retarded at all. In the two
cartridge approach, this point is signaled by a breakthrough of volatile com-
pounds into the second cartridge. Compounds of still higher volatility will
12
-------
show similar amounts in both cartridges.
The sampling rates varied between 10 and 30 1/min in different set-ups.
Atmospheric tests usually involved 5 to 30 m3 of air, and the tests using the
car exhaust, gasoline, and chlorinated hydrocarbons made use of 0.4 to
2.2 m3 samples. Judged by gas-liquid chromatography on OV-101, the compounds
in the transition region (where breakthrough from the first to the second
cartridge occurs) have Kovats indices around 1000 (n-decame).
There is no lower limit of volatility in reference to collection, however,
the concentration of organics in vapor form, in the atmosphere, drops off
rapidly with increasing molecular weight. Yet, it is perhaps in this area that
the method can render its most valuable contribution to pollution problems
e.g. when pesticides, certain industrial effluents or polynuclear aromatics,
etc., occur in air. Furthermore, it could conceivably be employed in the
collection of natural emissions into the atmosphere, e.g. terpenoids from
conifers, etc.
Obviously, it is the organics of higher molecular weight which are most
difficult to identify, and the described method seems to be one of the best
approaches yet to collect enough material for several types of analysis. With
so many parameters influencing collection characteristics, the method is
flexible enough to be used for a wide variety of conditions and problems.
Chosen from a greater number of chromatograms, four examples may illustrate
some of the discussed points. Fig. 4 shows an example of insecticide sampling.
The vapor pressure of these chlorinated hydrocarbons is comparatively low;
they were consequently completely retained on the first cartridge. Fig. 5
gives an example of car exhaust sampling which the breakthrough region can
be clearly observed. Fig. 6 represents an isothermal run of car exhaust with
simultaneous use of flame ionization and electron capture detectors, and Fig. 7
compares the profiles of organics sampled in an urban and an industrial area of
St. Louis.
Car exhausts was also run on a GC-MS unit and the major peaks identified
from their mass spectra. Most of them turned out to be aromatics from toluene
to acenaphthene, whose presence in car exhaust, of course is amply documented
in the literature. No effort was made to identify all components—which, in-
cidentally, would have been better done with capillary columns and high-
resolution photoplate readout.
One intriguing possibility inherent in this approach would be to send
cartridges to different locations, where persons interested in the higher
organics, present in their respective atmospheres, would need only a pump to
draw a suitable amount of air through the cartridge and would then send it
back for extraction and analysis. The advantages of one centralized analytical
facility, which could survey at low cost any area where a pump and an electri-
cal outlet can be found, are obvious. However, further work on support-bonded
phases and their use in the field of air contamination, is definitely needed
before plans of these proportions should be seriously entertained.
13
-------
Collection of Chlorinated Hydrocarbons
Lindane
First Cartridge
Heptachlor
Aldrin
Inj. .
20 ft glass column, 3%OV-101on Chromosorb-W H. P. 80/100 mesh
MicroTek MT-220 E. C. detector alt 32 x 10.
Temperature 220, Ny flow 16 ml/mi n.
Figure 4
-------
130°
120°
110° 190"
90-
80«
70*
60°
5p« <0« 30-
Start
prog.
Room
temperature .
4x1
x10
16x1
4x10 Attenuation
Gas chrornatograms of car exhaust components collected on two identical cartridges con-
taining 24 % [C18H37SiO3/2]n on Chromosorb A. Sampling time, 60 min at 30 1/min. The iniections
represent ca. 0.16 sec of sampling. GC conditions: 3 °/0 OY-ioi on Chromosorb \V-HP. 80/100
mesh, 20 ft. x 2.5 mm I.D Pyrex; nitrogen flow optimized. 16 ml/min; flame ionization detector.
Figure 5
15
-------
tAK tXMAUil
Flame lonisation
2 microliters injection representing a 54 seconds of sampling. 20 ft. glass column HOV-101 en
Chromosorb-W. K P. . 80/100 mesh. Micro TekMT-220, FID alt > X ID. EC 64 X la
Temperature: 80*. N, (low 16 ml/min.
Figure 6
16
-------
Industrial Art*
Owntoan Art*
1 1
130* 1701
230°
|
Isothermal »t 230*
SO* 90* 1JO§ 170' 230*
| | | | | I sol her ma I at 230
1 microliltr injedion representing 30. 6 liters of air in the industrial
area SL Louis. Missouri. 20ft. glass column; £5 mm i. d 3*OV-101
on Chromosorb-W. H. P.. 80/100 m«h MicroTek MT-220.FIDatt 1X8
Program: 10 mm. at 50* followed by 8*/min., final hold at 230
N flow 16 ml/mm.
1 microliter injection representing 19. 6 liters of air in the downtown
area, SL Louis, Missouri. 20 ft tang column; 2. 5 mm i. d 3*OV 101
on Chromosorb-W. Ij. P.. 80/100 mesh. MicroTek MT-220. FIDan. 1X8
Program: 10 min. at SO* followed by 8 /mm , final hold at 230
N flow 16 ml /min.
Figure 7
17
-------
MODIFIED ADSORBENTS BASED ON SILICA GEL
Silica gels are the most widely used adsorbents for column chromatography.
This preference prevails because these materials have large surface areas,
diverse surface activities, easily controllable porosity and finally because
they are available in narrow-size range spherical particles.
A systematic study of treating silica gel (60/80 mesh) in the presence of
liquid water at high temperatures (hydrothermal treatment) was done under more
controlled conditions. In this controlled study, an interesting phenomenon
was observed: The plot of surface area versus hydrothermal treatment temper-
ature gave a minimum at the vicinity of 300°C. Although this controlled
procedure is in someway similar to that done earlier (3), the early experi-
mentations did not show a minimum up to 380°C. The difference may be attribu-
ted to the pronounced difference in cooling rate which in turn influence the
redeposition of silica. The early work used a small Parr Bomb (Model 4740)
and under the experimental conditions, a 5-hour cooling time was observed
after removal from an unthermostated muffle furnace; the controlled method in
the present experiments used a big volume Parr Bomb (Model 4652) and had a
1.5-hour of cooling time after removal from a thermostated bath; both after 18
hours of hydrothermal exposure.
The apparent similarity in the two hydrothermal treatment procedures gave
different results. Kiseley, Q al. (4), Unger, e£ al. (5) and Kirkland (6)
described various methods of preparing silica gel. It is now well accepted
that the man-made silica gel commands a wide range of surface areas, pore
volumes, distributions and sizes, all these properties being highly dependent
on the method of preparation.
Results and Discussion
Figure 8(a) and 8(b) show the effect of hydrothermal treatment with
surface areas produced by both methods. Fig. 9 shows the retention temperatures
of octane and hexadecane on different surface area silica gels shown on Fig.
8(b). As expected, the uncoated silica gel shows decrease retention temper-
atures with decrease surface area. In the GC testing of still lower surface
area, uncoated silica gels, (equivalent to greater than 300°C HT), results
seem to indicate an upturn to increasing retention of hydrocarbons and alcohols.
To speculate, this could be a change to a more active surface of the gel.
More important perhaps—this study demonstrated that the silica gels
obtained by both methods around the critical temperature of water, surprisingly,
can bond very thick layers of polymers. Typical examples of this very inter-
esting and potentially useful observation is shown on Fig. 10 with Carbowax
20M and OV101. Thus, as the surface area decreases, the decreasing contribution
to the retention of the absorption process is made up for, in part, by the
increasing contribution of partition in the thickening polymer layer. This is
the reason that coated, low surface-area silica gels do not lose, to any great
degree, their ability to retain organics as blank silica gels do.
18
-------
00
VO
Figure 8
100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 °C
SURFACE AREA OF DAVISON SILICA GEL 62 AS A FUNCTION OF HYDROTHERMAL
TEMPERATURE, (A) REF, 3 (B) THIS STUDY, TO BE PUBLISHED,
-------
ZOO
200
100
SO
10
Surface Area, m /g
I I I 1 I I 1 i 1
I 1 I I I I 1 I I
I I
1 10 100 200 300
RETENTION TEMPERATURES OF MODEL HYDROCARBONS ON DIFFERENT SURFACE AREA SILICA GELS,
Figure 9
-------
o 40
-------
GC CHARACTERIZATION OF HYDROTHERMALLY AND POLYMER MODIFIED SILICA GELS
The silicic supports were further investigated to provide better stability
and increased retention. The most promising supports as seen in the previous
section were modified, wide pore silica gels. The silica gels were subjected
to hydrothermal treatment and then modified with organic polymers. Testing of
the modified silica gels, which were prepared in the early study, were done by
gas chromatography with non-polar and polar substances.
General preparation of the silica gel: Silica Gel Davison 62 (Fisher
Scientific Co., St. Louis, Mo.), 40-60 mesh, was washed in a Soxhlet (Kontes
Model K-585100) with HC1 at boiling point temperatures until no yellow hue
could be detected in a fresh charge of cone. HC1, hot or cold, after several
hours of extraction. The silica gel was then washed with distilled water to
neutrality.
Hydrothermal Treatment
A high-temperature bomb (Model 4740 with Pyrex insert tube, Parr Instru-
ment Co., Moline, 111.) was filled with a slurry of acid-washed silica gel and
distilled water such that their levels were at approximately 2/3 and 3/4 of
the tube, respectively. The closed bomb was lowered into a muffle furnace at
280°, left overnight, and removed after the furnace had been allowed to cool
down in the morning. The silica gel was then again acid-washed as before to
remove any traces of metal originating from the steel bomb, its nickel gasket,
or copper lubricant.
Modification with Carbowax 20M
Dry silica gel was coated with 6% Carbowax 20M, heat-treated under nitro-
gen overnight at 260°, and exhaustively extracted with methanol.
Addition of Water to the Carrier Gas
The carrier gas line of a Microtek-220 gas chromatograph was cut between
flow-control module and injection port, and the nitrogen made to bubble through
distilled water contained in a stainless-steel trap.
Results and Discussion
The hydrothermal treatment at 280°C reduced the surface area of silica
gel considerably (to 141 m2/g. Micromeritics Instrument Corp., Norcross,
Ga.). The non-extractable coating of Carbowax 20M on this support had a
nominal thickness of 2 X (carbon 1.52%, Peninsular Chemical Research).
Acid-washed Silica Gel 62, without the benefit of further modifications,
gave good chromatograms of n-hydrocarbons (Fig. 11). It failed to pass any
alcohols, however, at 140° (Fig. 12, trace A). This separation must be obtained
22
-------
12 10
16
18
14
r
12
10
16
18
14
1C
I
Separation of even-numbered ti-alkanes (octane through octadecane) on bare sUica gels.
Both columns: Silica Gel 62, 40-60 mesh, acid washed; in 120 cm X 4 mm I-D- Pyex U-tube,
nitrogen flow-rate 80 ml/min. 6 "/rnin temperature program. FID. (A) Silica Gel 62 hydrother-
mally treated at 280° overnight and acid washed; (B) untreated silica gel. The n-alkanes are rep-
resented by their number of carbon atoms.
Figure 11
23
-------
HT-treated
H20 -deactivated
Bare Silica Gel
Separation of n-alkanol standards (methano! through pentanol) on bare silica gels. All
columns: Silica Gel 62, 40-60 mesh, acid washed; in 120 cm x 4 mm I.D. Pyrex U-tube at 140",
nitrogen flow-rate 80 ml/min, FID. (A) Bare silica gel; (B) Silica Gel 62 hydrothermally treated at
280° overnight and acid washed; (C) carrier gas saturated with water at ambient temperature;
(D) water-deactivated and hydrothermally treated silica gel.
Figure 12
by either hydrothermal treatment (Fig. 12, trace B), or coating with Carbowax
20M (Fig. 13, trace D), or the presence of water in the carrier gas (Fig. 12,
trace C). Other traces show various combinations of the three treatments, all
tested under the same chromatographic conditions.
GC of lower aliphatic alcohols provides in our opinion a severe test for
the degree of deactivation of the adsorbent surface. Each of the methods did
well on its own—combining all three of them (Fig. 13, trace A) may have been
detrimental (at least in this particular analysis), but was included to allow
a visual comparison of effects accumulating under the same chromatographic
conditions. Fig. 14 represents a more reasonable choice of parameters for the
separations of the alcohols, i.e. a temperature-programmed run without the
benefit of water in the carrier gas.
24
-------
Judging from the chromatographic improvement obtained by the use of these
three simple techniques, it would appear feasible to modify silica gels further
(e.g. by further reduction of the surface area combined with the use of better
monomolecular layers) to a point where they can be used in place of diato-
maceous supports.
HT-(rated
CW-20M coated
H,0-deactivated
Untreated
CW-20M coated
H|0- deactivated
HT-treated
CW-2DM coated
Untreated
CW-2DM coated
175
I
145
I
115
I
15
I
Fig 13 F«g.l4
Fig. 13. Separation of n-alkanol standards (mcthanol through pcntanol) on Carbowax-coated silica
gels. All columns: Silica Gel 62,40-60 mesh, acid washed, coated with Carbowax 20 M. heat-treated
and extracted; in 120 cm X 4 mm I.D. Pyrex U-tubc at 140*. nitrogen flow-rate So ml/min, FID.
(A) Silica Gel 62 hydrothermally treated at 280° overnight and acid washed (before coating with
Carbowax), carrier gas saturated with water at ambient temperature; (13) water-deactivated silica
gel; (C) hydrothermally treated silica gel; (D) untreated silica gel.
Fig.I 4. Separation of n-alkanols (mcthanol through hcxanol). Column: Silica Gel 62, 40-60 mesh,
acid washed. HT-28o°, coated with Carbowax 20 M; in 150 cm X 2 mm I.D. Pyrcx U-tubc, ni-
trogen flow-rate 80 ml/min, temperature program 6°/min, FID.
s 13 & 14
25
-------
A GAS CHROMATOGRAPHIC CARTRIDGE DESORPTION PORT
In the course of studies on the efficiency of solid media for the col-
lection of trace vapors in the atmosphere, it became necessary to dii'ise a
suitable cartridge system for sample acquisition, transport and introduction
into a gas chromatograph for analysis.
The collection media to be tested were solid materials, either bare or
with a polymer layer bonded to their surface. Although collected compounds
could be removed easily by liquid extraction, heat desorption was used for the
purposes of this study: it is faster, more sensitive, and much more prone to
decompose collected compounds. While the latter would be considered a dis-
advantage in actual sampling, it is valuable in a testing program to establish
the catalytic activity of collection phases.
These phases were filled into stainless steel cartridges, "loaded" by
drawing through defined volumes of doped air or other gases, and transferred
to the GC introduction system. The latter was designed for testing several
variants of a finally envisioned procedure, and thus wound up being somewhat
more complex than one would consider necessary for straight heat desorption.
Results and Discussion
Fig. 15 shows some details of the collection and desorption ports and
Fig. 16 a schematic of the flow system. The two valves (Valco Instruments,
Houston, Texas) and the lab-made cartridge desorption port can perform several
functions. A "loaded" cartridge can be swept with carrier gases at a wide
range of temperatures or flow rates and can also be completely closed off;
while the carrier gas stream to the GC column is maintained at all times.
This can serve to introduce only desired retention ranges from a loaded
cartridge into the GC system (position A) and vent other (B); to "clean"
cartridges at higher temperature (b) and then check for the absence of arti-
facts (A); to heat a directly injected sample or an externally loaded car-
tridge in a closed-loop position (C and D) and then release the vaporized
compounds cum decomposition products suddenly into the GC system (A); to study
the influence of various parameters on analytical accuracy; to inject directly
into a cold or hot cartridge or the GC column (C, A, B), and so forth.
Needless to say that the principles of these approaches have been well known
in gas chromatography; with examples too numerous to cite.
Both port and valves are contained in an aluminum box with thick bottom
heated by cartridge heaters and surrounded by Marinite insulation. The
cartridge port is, in addition, independently heated by a cartridge heater, or
cooled by a flow of water or carbon dioxide. It is also thermally insulated.
The whole arrangement is situated close to the regular injection port of the
gas chromatograph, with thermal configurations designed to avoid any cold
spots in connecting lines.
Aside from laboratory testing, the system can be used for straight field
sampling and subsequent cartridge transport to the laboratory. Sample streams
26
-------
ALUMINUM DESORPTION PORT
j BOB!
STAINLESS STEEL COLLECTION CARTRIDGE
[) "T it' ll ruf•••
STAINLESS STEEL CARTRIDGE HOLDER
V It Ttf
CROSS SECTION
A STAINLESS STEEL SPPING
B CONE TO FIT SEAT C
D SAMPLING STREAM, IN OR OUT
E LEAD SEPTUM
Collection cartridge, holder and desorpiion port.
A MARINITE INSULATION
B COOLING COILS
C FOR HEATING CARTRIDGE
D FOR THERMOCOUPLE
E FOR COLLECTION CARTRIDGE
F SEPTUM, LEAD OR SILICONS RUBBER
Figure 15
-------
10
OO
cartridge heater
carrier
-------
can be made to enter the cartridge from either side and, if necessary, at
temperatures different from ambient.
Figure 17 shows some typical gas chromatograms obtained with a text
mixture containing n-hydrocarbons and n-alcohols. Direct injections into the
GC column and through the hot cartridge are contrasted with two runs involving
heat desorption from a cartridge loaded by drawing through a greater volume of
doped air. In this particular case, some decomposition of the higher alcohols
is apparent, with collection cut-offs being quite different for alkanes and
alkanols of comparable GC retention time.
Here as in Fig. 18, which shows a typical run with gasoline vapors in
air, the cartridge contained a surface-modified silica gel. After collection,
it was heated in the cartridge port from ambient to 230° in ca. 2.5 minutes
and left at that temperature for three more minutes. The vaporized compounds
were routed to the GC column (position A in Fig. 16), which was close to
ambient temperature during the heat desorption step and was then temperature
programmed to obtain the chromatograms shown.
When the desorption port was initially installed, the valves had to be
carefully conditioned (i.e. frequent rotation of the valve during initial
heat-up and conditioning at 300° according to the manufacturer's instructions)
to prevent them from getting stuck, and some minor leaks developed on occasion.
Otherwise, the system performed to our satisfaction.
29
-------
V
X
II
c.
U>
O
V
41
•o
•a-*
xo
tic
200
150
100
J
50
200
150
100
50
100
50
200 150
POSITION B POSITIOM A POSITION A
DIRECT INJECTION TO GC DIRECT INJECTION THROUGH HOT CARTRIDGE HEAT DESORPTION OF LOADED CARTRIDGE
Chromatographic comparisons of different sample introductions to the valVing system. Conditions: injected, 1/^1 of alcohol-hydrocarbon mixture:
initial temperature, 50°; program rate, 8°/min; temperature FID, 225°; analytical column, 3% OV-17 on Chromosorb W, 100-120 mesh, 6 ft. x 2
mm l.D. borosilicate glass.
Figure 17
-------
210 165 120 75
Chromalography with the introduction flow pattern as shown in position A of Fig. 2. 10-min
collection in atmosphere enriched with gasoline vapors. Solid adsorbent is hydrothcrmally treated
(210°)silica gel 62 modified with non-cxtractablc layers of OV-101. Temperature: multiporl introduc-
tion system, 190°; FID, 225". Analytical column: 10% FFAP on Chromosorb W, AW DMCS. 80
100 mesh. 20ft. x 2mm I.D. borosilicate loop tube. Attenuation. 2000.
Figure 18
31
-------
OTHER POLYMER MODIFIED SILICA GELS
The success obtained with the modification of hydrothermally treated
silica gels with Carbowax 20M led to the use of various polymers commonly used
as liquid phase in gas chromatography. This included linear polyethylene, AN-
600, ON 870, FFAP and SE-30. These polymers were used with silica gels of
different surface areas. The modification procedure was similar to that
described in the previous section. A total of about 40 packings were made.
Results and Discussion
In the gas chromatographic testing (dry N? as carrier gas) alcohols
(butanol) do not elute on silica gel hydrothermally treated around 200°C or
below even modified with an organic polymer (for example see Figure 19). The
packings were also tested as collection media for a mixture of alcohols and
hydrocarbons pumped out of teflon bag (for example see Figure 20).
o
20CTC
I
Gas chromaotgram of hydrocarbons on silica gel,
hydrothermally treated at 210 C and modified
with non-extractable layers of OV 101.
Figure 19
32
-------
75
210 165 120
1 M)CROLITER ALCOHOL-HYDROCARBON MIX INJECTED TO 28 LITER H, IN
TEFLON BAG, CONTENTS PUMPED OUT IN 70 MINUTES; SOLID ADSOR-
BENT IS HYDROTHERMALLY TREATED (210») SILICA GEL 62, MOD-
IFIED WITH NON-EXTRACTABLE LAYERS OF 0V 101; TEMPERATURE
MULTIPORT INTRODUCTION SYSTEM 190«; TEMPERATURE FID 225°;
ANALYTICAL COLUMN IS 31 0V 17 ON CHROMOSORB W 100/120; XlOOO;
6 FT X 2 MM ID BOROS1L1CATE U-TUBE
Figure 20
33
-------
HUMIDITY AND SAMPLING FLOW RATES
The effects of sampling flow rates and high humidity on the various solid
collection media are well known. Therefore, selected silica gels carrying a
variety of bonded layers were tested for the presence and absence of these
effects. The most promising cartridge packing appeared to be silica gels
hydrothermally treated at 208°C and 280°C (4 and 30 m2/g) modified with thick
layers of non-extractable linear polyethylene and OV101.
Dry Air
A 28-liter teflon bag was filled with dry air and spiked with alcohol-
hydrocarbon mixture. The bag was left under a heat lamp for 20 minutes to
insure homogeneous vapor state, before the bag was pumped out. The experiment
was repeated for each packing at four different flow rates. Flow rates
between 20 and 200 cm/sec (0.4 to 4 liters/min) were tested. Silica gel mesh
size was 25/40.
Moist Air
The compressed air from the tank was slowly bubbled through a water
container then used to fill a 28-liter teflon bag. The bag was spiked with
the alcohol-hydrocarbon mixture and left as before for 20 minutes under a
heat lamp. Four experiments were performed at four different flow rates.
Heat Desorption
The loaded cartridge in each of the experiments described above was
analyzed immediately after collection. The alcohols and hydrocarbons trapped
in the cartridge were released by heat desorption (see previous section) to an
analytical column. The GC conditions were 3% OV17, FID at 230°, inlet 210°,
transfer line of 215° and valves temperature of 250°C.
Results and Discussion
Based on the data, it can be generalized that none of the four systems
tested showed a significant decrease in collection efficiency as a function of
flow rates. It is therefore possible to use a high sampling speed with a
particular cartridge with little regard to humidity, support surface area, or
type of bonded (unpolar) polymer.
Figures 21-26 summarize the effects of flow rates and humidity on the
collection efficiency. Typical GC chromatograms for this type of sampling is
shown on Figures 27(a) and 27(b) from which Figures 21-24 were deduced.
Generally the effects of different levels of humidity were small, and notice-
able only with compounds at the border of the collection range. The retention
of alcohols on polyethylene showed, if anything, a slight improvement in moist
atmosphere. Hydrocarbons under the same conditions did worse. On polys-
iloxane both alcohols and hydrocarbons did somewhat worse under moist condi-
tions, but the effects, as stated above, were quite small.
34
-------
The effect of surface area on retention was as predicted in direction,
but not in magnitude. The lower surface area silica gels did collect less of
the "border-case" compounds—but the effect was quite small. This is most
likely caused by the thicker polymer layers being found on the silica struc-
tures with the widest pores.
The effect of surface area on decomposition rates of higher alcohols
during heat desorption at ca. 250°C was again predictable and small—the low
surface area material inducing somewhat less decomposition. Under conditions
of solvent extraction, of course, one would not expect any decomposition at
all. Even under the trying conditions of heat desorption, however, the extent
of decomposition was generally small.
Figures 28-30 simply show some examples of samplings in doped atmosphere
in 5, 10 and 15 minutes of collection.
35
-------
OJ
cm
12
11
10
8
7
6
in
I 1 i
ADSORBENT: LINEAR POLYETHYLENE ON HT-280
I I
I
L I
I
C14H30 -Dry
C14H3Q - Moist
C12H26
COLLECTION FLOW, 1/min
I I I I I
C12H26 " Mo1st
I
0 0.2 0.4 0.6 0.8 1.0 2.0
Figure 21 RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
3.0
-------
CO
cm
12
11
10
9
8
7
6
5
4
3
2
1
ADSORBENT: LINEAR POLYETHYLENE ON HT-280
\ C1()H21OH - Moist
O CgH17OH - Moist
CgH17OH - Dry
C1QH21OH - Dry
1 I I I I
I
I I 1
COLLECTICN FLOW, 1/min
I I 1 I I I
0.2 0.4 0.6 0.8 1.0 2.0 3.0
RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
Figure 22
-------
00
cm
20
18
16
14
12
10
8
ADSORBENT: LINEAR POLYETHYLENE ON HT-2080
53
fc
e
^-«
_ a
C12H26 - Moist
i i
I I
OOLLECTIC3N FLOW, 1/min
I I I
0 0.2 0.4 0.6 0.8 1.0
2.0
3.0
4.0
RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
Figure 23
-------
to
CO
12
11
10
9
8
7
6
5
4
3
2
1
in
4J
09
ADSORBENT: LINEAR POLYETHYLENE ON HT-2080
CDLLJECTICN FLOW, 1/mn
I I I I I I I L 1 1 1 I 1 I I 1 I I I
0.2 0.4 0.6 0.8 1.0
2.0
3.0
4.0
RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
Figure 24
-------
cm
12
11
10
9
8
7
6
5
4
3
2
ADSORBENT: 0V 101 ON HT-2080
in
4->
rt
C16H34 - Moist
C16H34
OOLLECTICN FLOW, 1/min
I I I I I I I I I i
0 0.2 0.4 0.6 0.8 1.0 2.0 3.0
Figure 25 RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
-------
cm
10
9
8
7
5
4
3
2
1
I/I
4-1
rt
ADSORBENT: 0V 101 ON HT-2080
L llii
COLLECTION FLOW, 1/min
I 1 I I i I 1
0 0.2 0.4 0.6 0.8 1.0
2.0
3.0
RELATIVE DEGREE OF COLLECTION AS FUNCTIONS OF HUMIDITY AND FLOW RATES
Figure 26
-------
ADSORBENT: LINEAR POLYETHYLENE ON HT-280
DRY MOIST
Figure 27(a)
EFFECTS OF HUMIDITY AND SAMPLING FLOW RATES ON COLLECTION EFFICIENCY.
HEAT DESORPTION INTRODUCTION SYSTEM, 190°; FID, 225°; ANALYTICAL
COLUMN: 3% 0V 17 ON CHROMOSORB W,100-120 MESH, 6 FT.X 2MM I.D,
BOROSILICATE; PROGRAM RATE, 8°/ MIN,
42
-------
ADSORBENT: LINEAR POLYETHYLENE ON HT-280*
DRY
MOIST
°C 160 130
Figure 27(b). same as Figure 27(a)
43
-------
ISO. 220
COLLECTION BEHIND A CAR EXHAUST WITH THE ENGINE IDLING; SOLID ADSORBENT IS HYDROTHERMALLY
SILICA GEL 62 MODIFIED WITH NON-EXTRACTABLE LAYERS OF 0V 101; COLLECTION FLOW RATE, 2L/MIN;
FIVE MINUTES
TREATED (208
HEAT DESORPTION TEMPERATURE,
80-100 MESH;X 500; 20 FT X 2 MM I.D. BOROSI
190°; FID TEMP, 225°; ANALYTICAL COLUMN IS 10$ FFAP ON CHROMOSORB W, AW, DMCS,
LI GATE LOOP TUBE;
Figure 28
-------
220 ISO.
TEN MINUTES COLLECTION IN AN ATMOSPHERE OF GASOLINE CONTAINED IN A BEAKER. CONDITIONS SAME AS FIG. 28.
Figure 29
-------
FIFTEEN MINUTES COLLECTION BEHIND AN IDLING STATION WAGON. CONDITIONS SAME AS FIG. 28.
Figure 30
-------
SECTION 5
RESULTS AND DISCUSSION
Since this work is a multi-faceted endeavor, the results and discussion
for each phase of the work are given immediately after each experimental pro-
cedure. This provides a systematized organizational concept to enable the
reader to consult the chromatograms which are an integral part of the results
and discussion section of each phase.
47
-------
REFERENCES
1. Aue, W. A., C. R. Hastings, S. Tsai and P. Teli: "Trace Analysis with
Support-Bonded GLC Phases," Proceedings, Trace Substances in Environmental
Health - IV, D. D. Hemphill, Ed., University of Missouri, Columbia,
Missouri, 386-395 (1970).
2. Aue, W. A. and C. R. Hastings: "Preparation and Chromatographic Uses of
Surface-Bonded Silicones," J. Chromatogr., 42. 319-335 (1969).
3. Kapila, S., W. A. Aue and J. M. Augl: "Chromatographic Characterization
of Surface-Modified Silica Gels," J. Chromatogr., JT7, 35-48 (1973).
4. Kiselev, A. V. and Y. I. Yashin: "Gas Adsorption Chromatography," Plenum
Press, New York (1969).
5. Unger, K., J. Schick-Kalb and K. F. Krebs: "Preparation of Porous Silica
Spheres for Column Liquid Chromatography," J. Chromatogr., 83, 5-9 (1973).
6. Kirkland, J. J.: "Porous Silica Microsphere Column Packings for High
Speed Liquid-Solid Chromatography," J. Chromatogr., 83, 149-167 (1973).
48
-------
PUBLICATIONS
Based on Work Supported by EPA Grant R-801050
PUBLICATIONS
Research which was totally or partially supported by the EPA grant R-
801050.
W. A. Aue, C. R. Hastings, J. M. Augl, M. K. Norr and J. V. Larsen: "The
Distribution of Support-Bonded Silicones on Chromosorbs," ^J. Chromatogr. 56,
295-311 (1971).
C. R. Hastings, W. A. Aue and F. N. Larsen: "In Situ Synthesis of Silicone
Liquid Phases for Chromatography," ^J. Chroma togr. 60, 329-44 (1971).
W. A. Aue and P. M. Teli: "Sampling of Air Pollutants with Support-Bonded
Chromatographic Phases,"^. Chromatogr. 62, 15-27 (1971).
W. A. Aue, P. M. Teli, C. W. Gehrke and C. R. Hastings: "Collection and
Analysis of Organic Air Pollutants," Proceedings, Trace Substances in Environ-
mental Health-V, D. D. Hemphill, Ed., University of Missouri, Columbia,
Missouri, 435-446 (1971).
W. A. Aue, C. R. Hastings and S. Kapila: "Synthesis and Chromatographic
Application of Bonded, Monomolecular Polymer Films on Silicic Supports," Anal.
Chem. 45, 725-728 (1973).
W. A. Aue, S. Kapila and K. 0. Gerhardt: "The Gas Chromatographic Properties
of a Modified, Wide-Pore Silicagel," .J. Chromatogr. ^8, 228-232 (1973).
W. A. Aue, C. R. Hastings and S. Kapila: "On the Unexpected Behavior of a
Common Gas Chromatographic Phase," J_. Chromatogr. 77, 299-307 (1973).
S. Kapila, W. A. Aue and J. M. Augl: "Chromatographic Characterizations of
Surface-Modified Silicagels," J_. Chromatogr. 87, 35-48 (1973).
C. R. Hastings, J. M. Augl, S. Kapila and W. A. Aue: "Non-Extractable Polymer
Coatings (Modified Supports) for Chromatography," _J. Chromatogr. 87, 49-55
(1973).
W. A. Aue and D. R. Younker: "Characteristics of a Novel Gas Chromatographic
Phase," J. Chromatogr. 88, 7-14 (1974).
49
-------
C. R. Hastings and W. A. Aue: "Novel Polymer-Deactivated Adsorbents as Sup-
ports in Gas Chromatography," JL Chromatogr. 89, 369-373 (1974).
W. A. Aue, C. R. Hastings and K. 0. Gerhardt: "Gas Chromatography en Modified
Supports," 2- Chromatogr. 99^, 45-49 (1974).
C. R. Hastings, T. R. Ryan and W. A. Aue: "Products from the Electron Capture
Reactions." Anal. Chem. 47., 1169-1173 (1975).
W. A. Aue, C. R. Hastings Vogt and D. R. Younker: "Gas Chromatographic
Cartridge Desorption Port," .1. Chromatogr. 114, 184-189 (1975).
C. R. Hastings Vogt, C. Y. Ko and T. R. Ryan: "Modification of an Analytical
Procedure for Isocyanates to High Speed Liquid Chromatography," a comprehensive
manuscript prepared for the National Institute for Occupational Safety and
Health, April, 1976 on industrial air pollutants. Available from the National
Technical Information Service (NTIS //PB262-675), U. S. Dept. of Commerce,
Springfield, Va. 22151.
C. R. Hastings Vogt, C. Y. Ko and T. R. Ryan: "Simple Ureas Derived from
Diisocyanates and Their Liquid Chromatography on a 5-cm Column," _J. Chromatogr.
134. 451-458 (1977).
C. R. Vogt, T. R. Ryan and J. S. Baxter: "High Speed Liquid Chromatography on
Cadmium Modified Silica Gel," J_. Chromatogr. 136. 221-231 (1977).
C. R. Vogt and K. R. Leimer: "Chromatographic Profile and Analysis of Organic
Compounds in Selected Waters," to be published in Proceedings, Trace Substances
in Environmental Health-XI, D. D. Hemphill, Ed., University of Missouri, June,
1977.
C. R. Vogt, T. R. Ryan and J. S. Baxter: "Silver on Low Surface Area Silica
Gel and Its Performance In Liquid Chromatography," submitted to £. Chromatogr.
C. R. Vogt and T. R. Ryan: "Surface Characteristics of Metal Loaded Silica
Gel," to be submitted to J_. Chromatogr.
C. R. Vogt and V. P. Kukreja: "A Thermally Stable Pseudo-Picric Acid Chroma-
tographic Phase: Separations of Polynuclear Aromatic Compounds," to be sub-
mitted to J^. Chromatogr.
C. R. Vogt and V. P. Kukreja: "A Chemically Bonded Stationary Phase for
Complexation in Gas Chromatography," to be submitted to J^. Chromatogr.
C. R. Vogt, D. R. Younker and W. A. Aue: "Polymer-Modified Silica Gel as
Chromatographic Phases and Collection Media," manuscript in preparation.
C. R. Vogt, D. R. Younker and W. A. Aue: "Collection of Organic Vapors on
Modified Silica Gel," manuscript in preparation.
50
-------
PRESENTATIONS
Totally or partially supported by the grant and contains relevant in-
formation.
W. A. Aue, P. M. Teli, K. 0. Gerhardt and C. R. Hastings: "Collection and
Analysis of Organic Air Pollutants," 5th Annual Conference on Trace Substances
in Environmental Health, Columbia, Missouri (June, 1971).
W. A. Aue, C. R. Hastings, J. M. Augl, M. K. Norr and J. V. Larsen: "Scanning
Electron Microscopy of Support-Bonded GLC Phases," 7th Midwest Regional ACS
Meeting, St. Louis, Missouri (October, 1971).
"Tragergebundene Fliissigphasen fur die Chromatographie" plenary lecture,
UniversitSt des Saarlands, Saarbriicken, West Germany, July, 1972.
"Choice of Detectors and Columns for the Analysis of Pesticides by GLC,"
plenary lecture, Int'l Symposium on Recent Advances in Analytical Chemistry
of Pollutants: Halifax, N.S., Canada; August, 1972.
W. A. Aue, C. R. Hastings and S. Kapila: "Synthesis of Support-Bonded Polymers
and Their Use in Chromatographie and Trace Analyses," International Congress on
Analytical Chemistry; Kyoto, Japan; April, 1972.
S. Kapila, K. 0. Gerhardt and W. A. Aue: "Modification of Wide-Pore Silicagels
for Chromatography," 8th ACS Midwest Regional Meeting, Columbia, Mo., November,
1972.
C. R. Hastings, S. Kapila and W. A. Aue: "Gas Chromatography on Surface-
Modified Chromosorbs," Ibid.
W. A. Aue, C. R. Hastings and S. Kapila: "On the Unexpected Behavior of a
Common Gas Chromatographie Phase," 165th National ACS Meetings, Dallas, Texas,
(April, 1973).
W. A. Aue, C. R. Hastings and S. Kapila: "Synthesis and Chromatographie Appli-
cation of Bonded Monomolecular Polymer Layers on Silicic Supports," 8th Inter-
national Symposium on Advances in Chromatography, Toronto, Ontario, (April,
1973).
W. A. Aue, C. R. Hastings and S. Kapila: "Synthesis and Chromatographie Appli-
cation of Bonded Monomolecular Polymer Layers on Silicic Supports," 8th Inter-
national Symposium on Advances in Chromatography, Toronto, Ontario, April,
1973.
W. A. Aue, C. R. Hastings and K. 0. Gerhardt: "Gas Chromatography on Modified
Supports," 9th International Symposium Advances in Chromatography, Houston,
Texas, (November, 1974).
C. R. Hastings Vogt: "Industrial Organic Vapors—Sample Handling and Analysis,"
Clean Air - Applied Technology Conference Warrensburg, Missouri 64093 (April,
1975).
51
-------
C. R. Hastings Vogt, D. R. Younker and W. A. Aue: "A Multiport Introduction
System for Gas Chromatography and Its Application to Air Pollution," 2nd Annual
Meeting Federation of Analytical Chemistry and Spectroscopy Societies, Indian-
apolis, Indiana (October, 1975).
C. R. Hastings Vogt, M. B. Jones, D. R. Younker, T. Krob, and W. A. Aue: "Use
of Silicagel-Based Adsorbents for Collection of Organic Air Pollutants and for
Gas Chromatography," First Chemical Congress of the North American Continent,
Mexico City, Mexico (December, 1975).
T. R. Ryan, C. Y. Ko and C. R. Hastings Vogt: "Parameters in the Analysis of
Isocyanates by High Performance Liquid Chromatography," 3rd Annual Meeting
Federation of Analytical Chemistry and Spectroscopy Societies, Philadelphia,
Pennsylvania (November, 1976).
C. R. Vogt and K. R. Leimer: "Chromatographic Profile and Analysis of Organic
Compounds in Selected Waters," llth Annual Conference on Trace Substances in
Environmental Health, Columbia, Missouri, (June, 1977).
C. R. Vogt, J. S. Baxter and T. R. Ryan: "Modified Silica Adsorbents for
Chromatography," to be presented at the 26th International Congress of Pure and
Applied Chemistry, Tokyo, Japan, (September 4-10, 1977).
OTHER PUBLICATIONS
Other publications, supported mainly by other sources and which are not
directly concerned with support-bonded phases, but will have a direct bearing
on future research in the air pollution field, (such as new selective detectors
for gas Chromatography) are listed below:
W. A. Aue and R. F. Moseman: "The Spectral Response of the Alkali Flame
Detector," j;. Chromatogr. 6J,, 35-44 (1971).
W. A. Aue, K. 0. Gerhardt and S. Lakota: "Determination of Hetero-Element
Content in Organics by Alkali Flame-Gas Liquid Chromatography," J Chromatogr.
63, 237-47 (1971).
R. F. Moseman and W. A. Aue: "A Dual-Mode Indium Flame Detector," £.
Chromatogr. 63_, 229-36 (1971).
W. A. Aue and H. H. Hill: "A Tin-Sensitive Hydrogen Flame Detector," J_.
Chromatogr. _70, 158-61 (1972).
H. H. Hill and W. A. Aue: "Selective Detection of Organometallics in Gas
Chromatographic Effluents by Flame Photometry," J_- Chromatogr. 74, 311-18
(1972).
W. A. Aue and H. H. Hill: "A Hydrogen-Rich Flame lonization Detector Sensitive
to Metals," ^ Chromatogr. _74_, 319-24 (1972).
W. A. Aue and H. H. Hill: "Selective Determination of Hetero-Organics by a
Dual-Channel Detector Based on Flame Conductivity and Emission," Anal. Chem.
45, 729-732 (1973).
52
-------
W. A. Aue and S. Kapila: "The Electron Capture Detector—Controversies,
Comments, and Chromatograms," £. Chromatogr. Sci. 11, 255-263 (1973).
W. A. Aue and C. R. Hastings: "Response of a Filter-Less Flame Photometric
Detector to Hetero-Organics," £. Chromatogr. 87, 232-235 (1973).
C. R. Hastings, D. R. Younker and W. A. Aue: "Filter-Less Flame Photometric
Analysis for a Thiophosphate Pesticide in Gas Chromatographic Effluents,"
Proceedings, Trace Substances in Environmental Health-VIII, D. D. Hemphill,
Ed., University of Missouri, 265-271 (1974).
C. R. Hastings, T. R. Ryan and W. A. Aue: "Products from Electron Capture
Reactions," 10th Midwest Regional ACS Meeting, Iowa City, Iowa, November, 1974.
53
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-600/2-79-042
3. RECIPIENT'S ACCESSION-NO.
4 TITLE AND SUBTITLE
METHODOLOGY FOR COLLECTING AND ANALYZING ORGANIC
AIR POLLUTANTS
5 REPORT DATE
February 1979
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8 PERFORMING ORGANIZATION REPORT NO.
Corazon Hastings Vogt
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Trace Substances Research Center
Columbia, Missouri 65201
10. PROGRAM ELEMENT NO.
1AA001 (FY-75)
11. CONTRACT/GRANT NO.
801050
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
Final
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16 ABSTRACT
A number of support-bonded liquid phase sorptlon media were developed and
evaluated in model systems for collecting and analyzing organic air pollutants.
Polymers with various functional groups were synthesized and chemically bonded onto
inert supports in thick layers. A media consisting of a si 11 cone liquid bonded to
Chromosorb W was used with excellent results. Retention times of most organic
compounds on this liquid are extremely long at ambient temperatures, and sampling can
be carried out for 24 hours at a rate of 10 liters of air per minute. In contrast,
subsequent counter current liquid extraction takes only a few minutes since retention
volumes are very small. Extracts were analyzed largely by gas chromatography.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air pollution
Organic compounds
Collecting methods
Sorption
Chemical analysis
Gas chromatography
13B
07C
14B
07D
18. DISTRIBUTION STATEMEN1
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report I
UNCLASSIFIED
21 NO OF PAGES
_64
20 SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
54
------- |