United States
Environmental Protection
Agency
Atmospheric Research and
Exposure Assessment Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-89/070 Feb. 1990
&EPA Project Summary
Analysis of Ambient Polar
Volatile Organic Compounds
Using Chemical lonization-lon
Trap Detector
Sydney M. Gordon and Michael Miller
The current approach to measuring
trace levels of volatile organic
compounds (VOCs) in ambient air
requires cryogenic trapping of the
analytes, followed by thermal desorp-
tion and low-temperature refocussing
onto a column for analysis by
capillary gas chromatography/mass
spectrometry (GC/MS). This approach
has been successfully applied to
nonpolar VOCs, but its use for more
polar species has been complicated
by the problems associated with the
ambient water vapor collected with
the VOCs.
A promising technique for meas-
uring polar VOCs is chemical ion-
ization GC/MS (CI-GC/MS) in the
quadrupole ion trap. This approach
allows whole air samples to be taken
since the water present in the air is
used as the Cl reagent gas. Water Cl
leads to appreciable intensities for
the proton transfer agent H3O*,
which produces intense pseudomo-
lecular ions and class-specific frag-
mentation patterns for various low
molecular weight polar compounds.
Using water Cl enhances sensitivity
by a factor of about 5, as is apparent
from a comparison of Cl and electron
impact intensities obtained with the
Finnigan MAT 800 Ion Trap Detector™
(ITD) in full scan mode.
Standard mixtures of polar species
at low concentrations in humidified
zero air were analyzed without a
membrane dryer, using a cryogenic
trap and GC/CI-ITD with the ion trap
detector in the full scan mode. Water
appears to be an effective Cl reagent
gas in the ion trap, and the GC/CI-ITD
system can exploit the water vapor
normally present in ambient air to
circumvent the problems usually en-
countered when analyzing humid
ambient air for these compounds.
The water vapor present in the
samples does, however, have a dele-
terious effect on the quality of the
chromatography achieved. Future
work will focus on optimizing the
analytical technique and evaluating
its suitability for various polar com-
pound classes.
This Project Summary was devel-
oped by EPA's Atmospheric Research
Exposure Assessment Laboratory, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
Increasing public concern over polar
volatile organic compounds (VOCs) of
low molecular weight has spurred the
current interest in developing better
measurement techniques for these
compounds. Some polar VOCs are often
the subject of nuisance complaints from
the public (e.g., alkylthiols, amines, alde-
hydes, etc.), while others are cited as
potentially toxic compounds from indus-
trial emissions (e.g., ethylene oxide,
propylene oxide, acrolem, etc.). Analysis
of trace levels of polar VOCs in air,
however, poses many problems, and at
present, the nature and distribution of
these compounds in the atmosphere
cannot be adequately characterized.
-------
By contrast, the method for deter-
mining a large number of nonpolar VOCs
in air is well established. Stainless steel
polished sampling canisters are used
with a combination of cryogenic trapping
and high resolution gas chromatography
(GC) coupled to a suitable detector, such
as flame ionization, electron capture, or
mass spectrometry (GC/MS). Cryogenic
trapping preconcentrates the analytes by
first passing the gaseous sample through
a cold tube in which the organic com-
ponents condense on the cold surface;
the condensate is then desorbed and
analyzed.
In general, attempts to sample polar
VOCs at trace levels have been inef-
fective due to their chemical reactivity,
affinity for surfaces, and tendency to
undergo polymerization. The analytical
problem has been exacerbated, espe-
cially for GC/MS measurements, by co-
collected water vapor, which occurs at
concentrations that are several orders of
magnitude greater than that of the
analytes.
The moisture problem has been over-
come for many nonpolar compounds by
incorporating a drying step before GC
analysis. The air sample passes through
a Nafion membrane dryer tube before
being cryogenically concentrated; water
vapor in the air stream permeates the
dryer walls while most of the air
pollutants of interest pass through the
tube. This technique is limited, however,
to relatively nonpolar compounds since
small polar molecules also permeate the
walls of the drying tube.
A promising technique for measuring
volatile polar compounds in whole air
samples is chemical ionization GC/MS
(GC/CI-MS) in the quadrupole ion trap,
using the water present in the air as the
chemical ionization reagent gas. Water Cl
has been shown to produce intense
pseudomolecular ions and class-specific
fragmentation patterns for several polar
compound classes in a conventional
quadrupole GC/MS system. To date, the
chemical ionization capabilities of the
Finnigan MAT Ion Trap Detector have
been investigated in detail for the reagent
gases ammonia, methane, and isobutane.
A recent study has shown that chemical
ionization in the trap can improve the
lower detection limit for some com-
pounds by as much as a factor of 10 over
that for electron impact (El) ionization.
The relatively low cost and exceptional
sensitivity of the ITD make it an ideal
candidate detector for the analysis of
polar VOCs of environmental interest.
The objectives of this study were to
evaluate the conditions under which
water chemical ionization can be used for
the analysis of selected polar VOCs in
whole air samples, using GC/CI-ITD in
the full cyclic scan mode. The study
consisted of generating and comparing
electron impact and water Cl mass
spectra for several representative polar
compounds, and evaluating the effects of
humidified air and dry air on the CI-ITD
analysis of these compounds.
Procedures
Sampling System
The system used for sampling and
analyzing volatile organic compounds in
zero-grade air streams consisted of a
semi-automated cryogenic trapping and
desorption unit (Nutech Model 3538-02)
and a Fmnigan MAT 800 Ion Trap
Detector. Normally, whole air samples
collected in stainless steel canisters are
preconcentrated by passing them
through the cryogenic trapping system
prior to analysis. To investigate the
effects of ambient water vapor on the
analysis of polar VOCs, a special sample
delivery system was constructed. This
system, which was used in lieu of
canisters for whole-air samples, allowed
us to generate sample mixtures of polar
compounds in dry zero air or humidified
zero air. Samples were injected into a dry
zero-grade air stream through a septum.
Humidified samples were prepared by
first passing the air through a flask
partially filled with water, while dry
samples were generated by bypassing
the humidifier. Experiments were also
carried out, using a Nafion membrane
dryer tube, to demonstrate the effects of
removing water from a humidified gas
stream on the analysis of polar VOCs.
Instrumentation
A Varian 3400 GC, equipped with a 60
m x 0.32 mm id fused silica capillary
column (J&W DB-1701) with a film
thickness of 0.2 jim, was attached directly
via a heated transfer line to a Finnigan
MAT 800 Ion Trap Detector (ITD). The
column was temperature programmed
from -10°C at 10°C/min to 250°C. To
avoid problems with humidified samples
that caused ice to form in the column and
block it, a 1m x 0.53 i.d. fused silica pre-
column was connected to the analytical
column inlet.
The Finnigan MAT 800 ITD is a fully
integrated benchtop mass spectrometer
with high sensitivity in both the electron
impact (El) and the chemical ionization
(Cl) mode. It is also capable of full cyclic
scanning and selected ion monitoring. AJ
opposed to Cl in a conventional mass
spectrometer ion source, Cl in the ITC
requires only 10-6 to 10'4 torr of reagem
gas. This low pressure is sufficient tc
convert sample molecules to ions with
high efficiency because of the long
reaction times (milliseconds) in the trap
The ion trap operates in a pulsed mode
a special sequence of rf voltages anc
time intervals are used to select the
reagent ions, then react them with the
sample to form analyte ions before the rl
voltage is scanned for mass analysis.
EI/CI Mass Spectra Study
For the water Cl experiments, an exter-
nally mounted water source was used tc
maintain a constant reagent gas pressure
in the ion trap. The water was introduced
into the ion trap from a vial attached tc
the reagent gas inlet port. Several repre-
sentative polar compounds were injected
into the instrument to obtain their El and
water Cl mass spectra.
Trapping and Recovery Studies
Cryogenic trapping and recovery
studies of several polar VOCs of interest
(and some nonpolar compounds, for
comparison) were carried out using the
sample delivery system and the cryoi
genie trapping unit described above.
Using a gas-tight syringe, a fixed volume
of a standard mixture was injected into
the humidified zero-grade air stream in
the sample delivery system. The VOCs
were concentrated in the cryogenically
cooled trap. To desorb the trapped
sample, the temperature of the trap was
raised rapidly and the contents of the trap
were flushed onto the capillary column
for GC-ITD analysis. Several experiments
were carried out under dry conditions
(i.e., no water in the humidifier), and
some runs were performed using
humidified air with the Perma-Pure dryer
between the sample delivery system and
the trapping unit.
Results and Discussion
Comparison of El and Water Cl
Mass Spectra
Some of the water Cl and El mass
spectra obtained for selected polar VOCs
using fixed operating conditions in the ion
trap are summarized in Table 1. The
reduction in molecular ion fragmentation
achieved by using water Cl instead of El
is striking. All of the water Cl spectra,
except for amyl acetate, have thd
(M + H)* pseudomolecular ion as one of
-------
fable 1. Ion Trap Cl and El Mass Spectra of
Selected Polar VOCS
% Pel. Abundance
Compound
3-Octanone
(MW 128)
Benzophenone
(MW 182)
Acroleirt
(MW 56)
Amyl Acetate
(MW 130)
Heptanal
(MW114)
m/z
129
99
73
72
71
57
55
184
183
182
181
105
77
51
50
57
55
56
71
70
61
55
43
42
41
40
39
115
97
81
71
70
69
68
67
57
55
53
45
Water Cl
100
10
13
100
47
100
15
100
35
23
100
25
17
15
46
El
35
16
34
65
100
16
41
10
100
86
67
29
100
32
11
15
14
100
20
20
12
15
97
36
18
62
17
20
18
50
100
10
22
the major ions. Heptanal is the only
compound listed in which the (M + H)*
ion does not occur as the base peak.
Benzophenone and naphthalene were
used to compare the sensitivity of the
ITD in the full cyclic scan mode under
both El and water Cl conditions. For
naphthalene, which undergoes very little
fragmentation by either El or water Cl,
the estimated limit of detection was about
0.05 ng. For benzophenone, which frag-
ments to a much larger extent by El, the
smallest amount of sample required to
give good quality full scan mass spectra
(mass range m/z 45-300) was about 0.03
ng by water Cl and 0.15 ng by El This
^enhancement in sensitivity using Cl
(irises because the ion current is usually
"concentrated in only a few fragments in
Cl, as opposed to the large number of
fragments involved in the El case.
Furthermore, the increased reaction time
in the CI-ITD mode results in a greater
accumulation of sample ions for detec-
tion.
Water CI-ITD Analysis of
Humidified and Dry Air Samples
A standard mixture containing selected
polar and nonpolar VOCs was analyzed
by water Cl in the ion trap, to evaluate
the effects of humidified zero-grade air,
dry zero-grade air, and Nafion dried air
on simulated whole-air samples. The
chromatographic behavior of the com-
pounds listed was generally quite poor.
This was largely due to the fact that the
water vapor had a significant effect upon
the chromatography of the compounds,
and that neither the GC column nor the
operating conditions were optimized for
this work. Nevertheless, the experiments
carried out were able to demonstrate the
value of the water Cl technique for the
analysis of polar VOCs.
Despite the effects of the water vapor
on the chromatography, relatively little
retention time variability was observed in
these experiments. Table 2 shows the
average retention times obtained for the
test mixture that was analyzed under
humidified, dry and Perma-Pure dried
conditions. The increased fluctuations
noted with shorter retention times in the
table has been observed before for non-
polar compounds, and occurs when
cryogenically trapped compounds in-
cluding water are transferred from the
trap to a capillary column held initially at
a reduced temperature. Despite the use
of a precolumn and a slow initial temper-
ature program, the column is probably
blocked temporarily by ice until the
column temperature rises above the
freezing point of the water. Since this
condition is difficult to control due to the
erratic character of the blocking process,
the retention time fluctuations for the
early eluting compounds are not unex-
pected
The effect of the Perma-Pure dryer on
some of the compounds in Table 2 is
striking. All of the polar compounds, with
three exceptions, are removed from the
gas stream along with the water vapor.
The exceptions are acetone (very small
peak), acrylonitrile, and 1-methyl-2-pro-
panethiol. Both acetone and acryloni-
trileare soluble in water. Thus, the small
amount of acetone detected is most
likely due to the fact that acetone was
present at a very high level in the original
mixture. The disappearance of limonene
from this sample is equally surprising,
since it is insoluble in water and should
have passed through the dryer into the
GC column, in much the same way as
styrene or m-dichloro- benzene. The
reasons for this aberrant behavior in the
case of acrylonitrile and limonene are not
understood at this time.
Conclusions and
Recommendations
The laboratory tests that have been
conducted to evaluate the effects of
humid air matrixes on the analysis of
polar VOCs using water Cl with the
Finnigan MAT Ion Trap Detector, indicate
the following:
-------
Table 2. Retention Times for Several Polar and Nonpolar Compounds
Cryogenically Collected and Analyzed by Water CI-ITD
Retention Time, minutes
Compound
Acetone
Acetonitrile
Acrylonitrile
Tetrahydrofuran
Ethyl aceiate
2-Butanone
1 -Methyl-2-propanethiol
Acrolein
Styrene
Amyl acetate
LJmonene
m-Dichlorobenzene
Nitrobenzene
Humidified*
8.66
10.11
10.73
12.89
12.97
13.20
16.68
18.39
23.49
24.12
25.77
26.21
29.43
Dry
8.22
968
10.55
12.68
12.83
13.00
16.57
18.15
23.45
24.10
25.75
26.18
29.38
Perma-Pure
Dried
9.05
ND*>
10.83
NO
ND
ND
16.70
ND
23.22
ND
ND
26.25
29.50
"Average of two runs
bND = not detected
• Water appears to be an effective Cl
reagent gas in the ion trap. Its proton
affinity is low enough to allow it to react
with a wide range of polar VOCs.
• The ITD permits the use of the very long
reaction times necessary to detect low
sample concentrations with high
sensitivity.
• The GC/CI-ITD system can exploit the
water vapor present in the air, thus
circumventing the problems normally
encountered when analyzing humid air.
However, the presence of water vapor in
the air has a deleterious effect on the
quality of the chromatography achieved.
The exceptional sensitivity of the Ion
Trap Detector, together with water chemical
ionization, may lead to the use of
significantly smaller whole air samples to
further reduce the deleterious effects of
humidified air on the trace-level analysis of
polar VOCs.
Sydney M. Gordon and Michael Miller are with IIT Research Institute, Chicago, IL
60616-3799.
Joachim Pleil is the EPA Project Officer (see below).
The complete report, entitled "Analysis of Ambient Polar Volatile Organic
Compounds Using Chemical lonization-lon Trap Detector," (Order No. PB 90-106
451 /AS; Cost: $15.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Research and Exposure Assessment Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
X '6NALTV
US.OFFICIAL MAIL
U.S.POSTAG1
= 0 .2 5 H
Official Business
Penalty for Private Use $300
EPA/600/S3-89/070
------- |