United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Research Triangle Park NC 2771
Research and Development
EPA/600/S4-87/033 Dec. 1987
SERA Project Summary
Supercritical Fluid Extraction-
Gas Chromatography of
Volatile Organic Compounds
(VOC) from Tenax Devices
Bob W. Wright, Andrew J. Kopriva, and Richard D. Smith
This report describes the develop-
ment and evaluation of on-line super-
critical fluid extraction-gas chromatog-
raphy instrumentation and method-
ology for the analysis of volatile organic
compounds (VOC) from adsorbent
sampling devices. Supercritical fluid
extraction offers potential advantages
for the removal and transport of organic
components from adsorbent matrices
including rapid and efficient extraction
at mild temperatures. Extraction at
mild temperatures eliminates potential
problems such as analyte decomposi-
tion that can be encountered with the
high temperatures needed for thermal
desorption analysis. Since a major
objective of this study was to develop
viable instrumentation and methodol-
ogy, a relatively detailed description of
the instrumentation design require-
ments and present limitations are
discussed. The results of several series
of methodology validation studies are
also presented. These studies included
recovery studies of standard VOC
spiked on three types of Tenax®*
sampling devices including authentic
actively pumped (volatile organic sam-
pling train, VOST) and passive (EPA)
devices (PSD). Replicate devices
spiked in an exposure chamber were
also subjected to parallel analyses using
this new methodology and traditional
•Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
thermal desorption gas chromatog-
raphy.
This Project Summary was devel-
oped by EPA's Environmental Monitor-
ing Systems Laboratory, Research
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
The monitoring of hazardous organic
vapors in ambient air is an important
concern. A widely utilized approach for
these determinations incorporates selec-
tive collection and concentration on
adsorbent traps followed by either
thermal desorption or solvent extraction.
For determining volatile organic com-
pounds (VOC), sample collection on
Tenax-GC® followed by thermal desorp-
tion and gas chromatographic analysis
is generally employed. Although some
studies suggest that decomposition of
analytes and artifact formation can occur
during collection, these methods have
been widely used and have provided
generally satisfactory performance.
However, it has become clear as more
experience has been gained that serious
problems such as analyte decomposition,
analyte reactivity, artifact formation, and
nonquantitative recoveries can be
encountered during the thermal desorp-
tion step. These problems can be espe-
cially acute for analytes requiring high
desorption temperatures (>300°C) and
actually limits the volatility range of
-------
components that can be successfully
desorbed from Tenax®. For instance,
1,1,1-trichloroethane (methyl chloro-
form) is known to decompose at higher
temperatures and undergo dehydroha-
logenation during thermal desorption
from Tenax®.
Supercritical fluid extraction provides
an alternative approach for the removal
and transport of organic compounds from
adsorbent materials that can be both
efficient and thermally mild. The poten-
tial advantages of supercritical fluid
extraction accrue from the properties of
a solvent at temperatures and pressures
above its critical point. At elevated
pressure this single phase will have
properties that are intermediate between
those of the gas and liquid phases. The
compressibility of supercritical fluids is
large and densities typically 102 to 103
times greater than the gas are obtained.
Consequently, molecular interactions
increase due to shorter intermolecular
distances and solvating characteristics
approaching those of the liquid are
achieved. However, high diffusion coef-
ficients and low viscosities similar to
those of the gas are retained which
provide the potential for rapid and
efficient extraction rates. By choosing a
fluid with a low critical temeprature (Tc),
such as carbon dioxide (Tc = 31 °C),
extractions can be carried out at mild
temperatures to alleviate any thermally
induced problems. Supercritical fluid
extraction also provides the potential for
removal of less volatile compounds
which would require excessively high
temperatures for thermal desorption. The
combination of on-line supercritical fluid
extraction with capillary gas chromato-
graphy instrumentally links the extrac-
tion and analysis steps and allows the
extraction effluent to be directly ana-
lyzed. Since this process allows the
collection and concentration of trace
level analytes at the head of a capillary
column, high resolution and high sen-
sitivity analyses can be obtained.
Procedure
Various studies were conducted to
expedite the development and initial
evaluation of on-line supercritical fluid
extraction-gas chromatography (SFE-
GC) methodology for the analysis of VOC
from adsorbent sampling devices. This
was the first investigation conducted to
specifically address such an approach for
VOC analysis and essentially no previous
studies of reports exist relevant to such
an analytical approach. This project
included the development of appropriate
instrumentation and methodology, eva-
luation and quantitative validation of the
methodology using small sampling devi-
ces spiked with standard VOC, and
application of the developed methodol-
ogy to actual passive air sampling devices
(PSD) and actively pumped devices
(VOST). Tenax® was primarily used as
the adsorbent in these studies, but some
initial evaluations utilizing Spherocarb®
were also conducted.
The on-line supercritical fluid
extraction-gas chromatography instru-
mentation consisted of three primary
sections which included a high pressure
syringe pump and appropriate extraction
cell, an interface region for depressur-
ization of the extraction effluent, and a
gas chromatograph with a flame ioniza-
tion detector. A schematic diagram of the
final design of the instrumentation
developed in this investigation is shown
in Figure 1. The instrumentation was
evaluated by analyzing various sampling
devices that were spiked with standard
VOC mixtures. Devices that had been
spiked in an exposure chamber were also
subjected to analysis and compared to
the results obtained from replicate
devices analyzed by traditional thermal
desorption gas chromatography.
Results
Supercritical carbon dioxide at temper-
atures of 50° to 75°C and a pressure of
125 bar provided adequate solvating
power to extract the range of VOC utilized
in these studies from the Tenax® sam-
pling devices. Relatively clean blank SFE-
GC analyses could be obtained from
Tenax® sampling devices, but high levels
of contamination from Spherocarb®
sampling devices persisted even after
prolonged conditioning with supercritical
fluid extraction. Consequently, only
Tenax® sampling devices were subjected
to detailed evaluation.
Quantitative extraction of the analytes
from the sampling devices required
several fluid volumes of the device to be
used. Thus, large volumes of gas from
the depressurized carbon dioxide(severa)
liters) were required to be passed
through the chromatographic column.
This requirement imposed the following
specific instrumental constraints:
1. Minimum dead-volume construc-
tion of the sampling device extrac-
tion cells was required to minimize
extraction fluid volumes.
2. Ultrapure carbon dioxide was
necessary to prevent concentration
of impurities.
3. A wide-bore (0.53 mm i.d.) chro
matographic column was neede
to provide maximum gas flow an
to maintain high chromatographi
separation efficiency for the sam
pling devices examined in thi
work.
The large gas volume and high flow rat
impaired efficient analyte concentratioi
which required extreme focusin
methods of whole column cryotrappin
at -50°C with a thick-filmed (5 //m
stationary phase and a 15 m Ion
retention gap.
Good quantitative agreement betweei
the spike and recovery levels of the VOI
were obtained for the smaller volunv
adsorbent devices. The larger volum
sampling devices (e.g., actively pumpec
presented greater challenges ani
required extraction times in excess of 3(
minutes to obtain complete removal o
the analytes. Longer extraction time;
impaired efficient chromatographic sep
arations and less reliable quantitativi
recovery levels (and greater departure:
from the spike levels) were obtained. Thi
broadened peaks from less efficien
separations generally corresponded t<
greater than quantitative recoveries. Thi
quantitative SFE-GC analyses of the PSC
and VOST devices spiked in the exposun
chamber were ambiguous due to poor!'
resolved compounds and coeluting con
taminants that were extracted from thi
Tenax® matrix. However, peaks match
ing the appropriate retention times of thi
spiked VOC could be distinguised. Pre
liminary investigations utilizing varioui
PCS (polychlorinated biphenyl) congen
ers suggested that extension of the SFE
GC methodology to the analysis of semi
volatile materials from Tenax® samplin<
devices would also be feasible.
Conclusions and
Recommendations
The general concept of combined on
line supercritical fluid extraction-gas
chromatography proved viable for th«
quantitative extraction, recovery, con
centration, and analysis of standarc
volatile organic compounds that were
spiked on Tenax® sampling devices. Ni
apparent physical changes or deleterious
effects on the sorption behavior of thi
Tenax® were observed after numerou;
(>20) supercritical fluid extractioi
cycles. However, some minor caking o
the adsorbent was observed after exten
sive exposure and supercritical fluii
extraction cycles.
-------
Heated
Valve
1/16" Swage/ok
Tee
Syringe
Pump
Purified
Carrier
Gas
Heated Injection Valve
Sample Wgste
Inlet
Output
Gass-Lined s.s
Expansion Tube
(1/16"o.d.x
0.020" i.d.)
Heated
Block
Retention Gap
500 fil Loop
f
i
i
i
i
Carrier
Gas
FID
15 m x530 ym i.d.
Retention Gap
30 m x530 (im i.d.
Analytical Column
Gas Chromatographic Oven
Figure 1. Schematic diagram of the final design of the on-line SFE-GC instrumentation.
A number of recommendations for
additional studies can be made. Improved
analyte trapping methods need to be
developed to increase the separation
efficiency for the more volatile com-
pounds and to allow faster extraction
rates without compromising the chroma-
tographic separation quality. This could
be done by using longer analytical
columns, longer retention gaps (e.g., 50
m) and/or incorporation of an integral
baffle zone consisting of a short length
of column packed with particles of
deactivated Chromosorb W® or glass
beads. Smaller sampling devices would
also improve overall performance, but
the impact upon detection limits would
require study. With the development of
improved trapping methods, faster
extraction rates should be developed and
the limitations on analysis speed defined.
For smaller volume devices, extraction
times of five minutes or less should be
conceivable. Sampling devices optimized
for the SFE-GC methodology should also
be designed and evaluated.
Additional quantitative studies should
be done using rigorously prepared stand-
ard mixtures for adsorbent exposure and
instrument calibration. Improved quan-
titative reliability could also be obtained
by utilization of a mass selection detector
(MSD) to eliminate potential interferen-
ces from coeluting contaminants. These
studies should also include standard
VOC that present problems for thermal
desorption analyses.
The extraction cells for the sampling
devices should be redesigned to use
stainless steel 0-rings rather than the
present polymer 0-rings to eliminate the
possibility of organic contaminants being
leached from the 0-rings.
A rigorous investigation for the exten-
sion of SFE-GC methodology to the
analysis of semi-volatile materials from
adsorbent sampling devices should be
conducted and the range of compounds
that could be analyzed from a single
analysis defined (including detection
utilizing an MSD). In addition to Tenax®,
such a study could include the evaluation
of other adsorbent materials. For
instance, Spherocarb®, or another car-
bon molecular sieve material that was
properly purified (perhaps using super-
critical fluid extraction with a highly
solvating fluid) could be utilized.
A related study on the effects of
supercritical fluid extraction on the
sorbent properties of the utilized adsor-
bents (e.g., Tenax® and perhaps Sphe-
rocarb®) should be conducted. Such a
study would include rigorous evaluation
of the retention volumes of selected
standard VOC on adsorbents exposed to
various supercritical fluid extraction
conditions (e.g., fluids, densities, expo-
sure times).
-------
Bob W. Wright, Andrew J. Kopriva. and Richard D. Smith are with Battelle.
Pacific Northwest Laboratory. Richland, WA 99352.
Nancy K. Wilson is the EPA Project Officer (see below).
The complete report, entitled "Supercritical Fluid Extraction-Gas
Chromatography of Volatile Organic Compounds (VOC) from Tenax Devices,"
(Order No. PB 88-124 300/AS; Cost: $14.95; 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:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Center for Environmental Research
J.S.OKaO&AtEMAt
Environmental Protection Information
Agency Cincinnati OH 45268
Official Susmess
Penalty for Private Use $300
EPA/600/S4-87/033
/^"••"-?\ POSTA
• •• — ' /(J ">\ PENALTY
1 JAN-6'88 UpfiyflER
-—• "— I /USE S300
,_, \, p j_j ^(}, X '- ,d M^ E T E R
vllTNo G-35
? 0 .2 ? =
—
00003?9 PS
------- |