Evaluation of Septa Quality for Automatic SPME–GC–MS Trace Analysis
Journal of Chromatographic Science
Evaluation of Septa Quality for Automatic SPME - GC - MS Trace Analysis
Agnieszka Ulanowska 0
Tomasz Ligor 0
Anton Amann 1 2
Bogus-law Buszewski 0
0 Chair of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, Nicolaus Copernicus University Gagarin Street 7, 87-100 Torun ? , Poland
1 Univ.-Clinic of Anesthesia, Innsbruck Medical University , Anichstrasse 35, 6020 Innsbruck , Austria
2 Breath Research Institute, Austrian Academy of Sciences , Dammstrasse 22, A-6845 Dornbirn
*Author to whom correspondence should be addressed: email . The vials used for the preparation of breath samples for automated solid-phase microextraction-gas chromatography-mass spectrometry analysis are crimped with septa. These septa often emit specific volatile organic compounds (VOCs) confounding the measurement results of breath samples. In the current paper, 14 different brands of magnetic caps with silicone-polytetrafluoroethylene (PTFE), butyl-PTFE, or butyl rubber septa were tested. The total emission of septa over a 4 h period was also evaluated. The tested septa emitted 39 different compounds, which are mainly hydrocarbons, alcohols, and ketones. Acetone and toluene are the most abundant out-gassing products. The concentration of acetone was in the range from 55 to 694 ppb for butyl -PTFE septum (brand 14) and butyl rubber (brand 10), respectively. The measured toluene amount was 69-1323 ppb for the septum brand 14 and brand 8 (silicone-PTFE), respectively. Generally, the butyl rubber septa released higher amounts of contaminants in comparison to the silicone ones.
The determination of trace gas components in samples with
complex and varying matrix compositions is demanding. The
final result of trace gas analysis is influenced by many factors,
such as sampling, storage, sample pretreatment, separation of
constituents, detection, and quantification. Negligence with
respect to one of the mentioned steps may become a source of
errors that makes the measured data useless (
). The biggest
error can be made during sampling and sample preparation
(60%), measurement and calibration (30%), and during data
Volatile organic compounds (VOCs) identified in natural
samples (e.g., exhaled air, headspace of bacteria, and tissue) are
present in trace concentrations ( ppbv/pptv level) (
2 ? 4
Therefore, for detection of the analytes, different methods for
pre-concentration and enrichment are usually applied. Thermal
desorption (TD) and solid-phase microextraction (SPME) are
the most often used techniques for this purpose (
Gaseous samples for automatic SPME process are usually
stored in glass halogenated solvent (HS) vials closed with
crimp or screw caps equipped with a septum. Septa might
release different substances that are detected during sample
analysis. Additionally, the emission of contaminations can
interfere with the investigated or potentially important compounds.
Septa are typically made of butyl rubber or silicone. Butyl
rubber is airtight and seals well (
). This polyisobutylene
polymer consists of 98% of isobutylene with approximately 2%
of isoprene. Therefore, butylene-derivatives might be emitted
by those septa. They also release COS and CS2 (
intended for covering of analytical samples are coated by
polytetrafluoroethylene (PTFE, Teflon). Teflon coating is applied to
reduce negative factors such as: permeation through the
septum, absorption of the analytes, and the off-gassing process.
Additionally, PTFE is a chemically inert material that does not
react with analytes (
). They can be monomers,
depolymerization, or ageing, and organic solvents applied during the
technological process. The silicone ? PTFE septa are widely used
because this material is temperature resistant (the maximum
service temperature is 2258C) and soft and easy to puncture by
needle even if tapered. In general, silicone septa display less
off-gassing product (
10 ? 12
). This kind of septum was used for
sample preparation before SPME ? GC ? MS analysis of explosives
in soils (
). Therefore, it is often preferred. On the other
hand, butyl rubber ? PTFE septa are mechanically hard and
more resistant to the absorption of the constituents of a
sample, and it is stable up to 1308C (
). Kolb et al. showed
that in using silicone ? PTFE rubber septa, 69% of acetone and
89% of dichloromethane were lost, while in butyl ? PTFE rubber
septa, the losses were less than 3%. This outcome indicated
that butyl rubber is denser and less permeable than silicone
In the present work, 12 different brands of magnetic crimp
caps with silicone ? PTFE, butyl rubber ? PTFE septum, and two
different types of magnetic screw caps with silicone ? PTFE
membrane were tested. The evaluation of septum emission
over a 4 h period was also examined. The goal of this
investigation was the identification of VOCs emission from different
kinds of HS septa and the quantitation of the most abundant
The GC ? MS analysis was performed on an Agilent 5975 Inert
XL MSD, coupled with 6890 N gas chromatograph (Agilent
Technologies, Waldbronn, Germany) with a split/splitless
injector. The temperature of the injector was 2008C, and
injections/SPME desorption were made in splitless mode. Helium
was used as a carrier gas with a linear velocity of 40 cm/s. The
MS analyses were carried out in a full scan mode, with a scan
range of 15 ? 220 amu. The scan rate was 3.46 scan per s. The
electron impact ionization was used at energy 70 eV. The
temperature of the ion source and the transfer line was 1908C and
1508C, respectively. The acquisition of the chromatographic
data was performed by means of Chemstation Software
(Agilent). The 25 m 0.25 mm 3 mm capillary column, CPQ
(Varian Inc., Middelburg, The Netherlands) was used. The oven
temperature program was as follows: initially, 408C held for
2 min, then ramped 108C/min to 1408C, next ramped 58C/min
into 2708C and held for 3 min.
The auto sampler for the GC ? MS system equipped with
SPME Carboxen-PDMS coated (75 mm) fiber was purchased
from Gerstel (Gerstel, GmbH & Co. KG, Germany). For the
sample preparation, 20-mL HS glass vials (Perkin Elmer,
Shelton, CT) were used. Gas tight syringes were bought from
Hamilton (Hamilton, Reno, NV). The 1 dm3 Tedlar bags (SKC,
Eighty Four, PA) were used for the breath samples collection.
Twelve types of crimp caps with butyl rubber ? PTFE or the
silicone ? PTFE septum and two types of screw caps with
silicone ? PTFE septum were tested. The details on the tested
caps are presented in Table I.
The manual SPME holder with Carboxen-PDMS coated
(75 mm) fiber was purchased from Supelco (Supelco Park
Bellefonte, PA) and was used for the pre-concentration of
VOCs in breath directly from a Tedlar bag.
Preparation of vials prior to SPME ? GC ? MS analysis
Before using the glass vials, they were flushed with argon and
heated at 608C for a few hours to remove any contaminants.
The vials were purged for 3 min in a stream of pure nitrogen
(5.0) and then crimped. Ten minutes after sealing, an SPME
fiber was introduced to the vial for pre-concentration and to
enrich the gases emitted by the septum. This extraction stage
was performed for 10 min at 308C. Afterwards, a GC ? MS
analysis was performed. The analysis of outgassing from each
septum was repeated 3 times.
Breath sample collection
Before human breath collection, the Tedlar bags were
thoroughly cleaned by flushing with gas, and then filled with
nitrogen and heated at 608C for several hours to remove
residual contaminants. The automatic breath sampler was
used for the collection of alveolar breath and ambient air
(Medical University of Innsbruck, Austria). This device
enables exhaled breath sample collection in a CO2-controlled
manner. The first exhalation is used as a reference value and
is never collected. During subsequent exhalations, the Tedlar
bag was filled with alveolar breath. Afterwards, 10 mL of the
breath sample was transferred from the Tedlar bag to a
sealed glass vial.
Results and Discussion
The VOCs emitted (outgassing products) from various HS septa
were tested. Altogether, 39 different compounds were
identified. Emitted compounds belong to various chemical classes,
such as alcohols, ketones, aldehydes, and hydrocarbons. Their
contributions in total concentration were in the range of
2 ? 10%, 5 ? 35%, 1 ? 5%, and 10 ? 65%, respectively. Such
emission can be attributed to the production process of septa
material, such as the solvents, monomers, and additives. These
compounds could also be formed during the processing of the
septa. When two commercially available materials are
compared, the butyl rubber septa releases higher amounts of
contaminants than the silicone ones. The concentrations of the
main VOCs found in septa emissions is presented in Table II.
Septum brand 7 (silicone/PTFE) gives the lowest emission of
VOCs, and only 18 compounds were found. In contrast, the
highest outgassing level was observed for brand 10 (butyl
rubber). These septa present incredibly high emission of
various compounds (i.e., the concentration of acetone exceeds
600 ppb). It shows that a PTFE covering decreases the VOCs
emitted from the rubber.
Acetone and toluene are the most abundant outgassing
products of commercially available headspace septa. In the case of
the remaining septa, (brand 1 ? 5, 9, and 13) the amount of
toluene was in the range of 45 ? 65% of the total emission. The
observed concentrations of toluene ranged from 1323 ppb to
69 ppb for brand 8 (silicone ? PTFE) and 14 (butyl ? PTFE),
respectively. The concentration of acetone was in the range
of 65 ? 694 ppb.
The emission of other compounds, such as ethanol,
2-methylpentane, and hexane was in the range 32 ? 87 ppb,
16 ? 93 ppb, and 17 ? 36 ppb, respectively. In general, aldehydes,
except etanal and propanal, were not found in outgassing
products. However, the amount of acetaldehyde exceeds 200 ppb
in a few cases. Among all released compounds,
2-methyl1,3-butadiene (isoprene), which is one of the main metabolism
products, was not found. All compounds identified in the
emission from the tested septa are presented in Table III.
Evaluation of Septa Quality for Automatic SPME?GC?MS Trace Analysis 11
The VOCs concentrations in two samples of breath that
were measured in two different fashions were compared.
The first one was manually SPME extracted directly from the
bag containing the breath. In the second case, the sample was
transferred into an HS vial and measured. Figure 1 shows the
two overlapped chromatograms of exhaled air. The samples
measured in the vials display an elevated concentration of the
selected compounds (e.g., butane, pentane, 2-methylpentane,
3-methylpentane, methylcyclopentane, hexane, and toluene)
and false positive results for the other. Table IV presents data
obtained for in bag and in vial sampling. In the first case, the
extraction process was performed in a Tedlar bag. In the
second, the sampling was done in vials, using septum brand 1
(butyl ? PTFE).
12 Ulanowska et al.
A tree model, performed in Statistica 7.1 Data Miner
(Statsoft, Polska) software running on a Windows XP platform,
showed that septa brand 1 ? 5 and 13 reveal similar off-gassing
products. Additionally, these septa are characterized by a low
emission of VOCs. The received dendrogram indicates also a
high similarity of VOCs emitted for the membranes brand 11,
12, and 6, 7. A cluster analysis confirmed that septa brand 9
and 7 are quite different.
Time-dependant emission has been investigated for septum
butyl ? PTFE (brand 1), and the results are presented in
Figure 2. The first sample was measured 5 min after crimping.
The second vial was used 1 h later, and the last three were
sequentially injected in 1 h intervals. Each sample was
equilibrated during 5 min and 1, 2, 3, and 4 h respectively, prior to
analysis. The compounds, such as acetone and pentane, were
on a similar level over a 4 h period. Only the concentration of
methylcyclopentane, ethyl acetate, and 3-methylpentane
slightly increased. Decreasing concentrations of hexane
between time point 1 and 2, or toluene between time points
1 ? 3, might be a result from a poorly established
physiochemical equilibrium in the vial. In the case of most
determined substances, after 1 h, the concentration of the released
VOCs was stable.
Baking septa in an incubator under a vacuum was tried in
order to eliminate the VOCs emission. However, heating the
septa at 908C for 30 days (without opening the incubator) did
not decrease off-gassing. In addition, an increasing emission
(20 ? 25%) of carbon disulfide for butyl rubber septa was
In view of these results, the application of such septa is
limited to brand 7. In the case of the remaining brands, even a
short period of storage gives a remarkable contamination
of the samples. Additionally, if the outgassing level is similar to
the concentration in the breath, the common correction for
the background concentrations by subtracting it from the
exhaled breath concentrations is not suitable.
The selection of a proper, low emission septa for the crimping
of HS vials containing breath samples is often essential for a
high-throughput SPME ? GC ? MS analyses. A chromatographic
technique, making a blank of pure nitrogen and a control of
the laboratory air, allowed for the determination of a group of
compounds emitted by commonly applied septa. In the
emission from the septa, 39 substances were identified, mainly
hydrocarbons. In many cases, volatiles released by membranes
are significantly higher than a typical concentration of selected
VOCs in a breath. Additionally, outgassing products might
interfere with some compounds.
To eliminate this problem, septum baking was not
appropriate. Septum made from butyl rubber with a PTFE covering and
heated over a 30 day period did not decrease the emission.
Evaluation of Septa Quality for Automatic SPME?GC?MS Trace Analysis 13
Therefore, carefully checking every kind of septa, as well as a
new batch of received, up to now is suitable for control of
emission. However, high quality septa without any emission are
1. Namie s?nik, J.; Szefer, P. Preparing samples for analysis - the key to analytical success . Ecol. Chem. Eng . 2008 , 15 , 167 - 244 .
2. Buszewski , B. ; Ulanowska , A. ; Ligor , T. ; Jackowski , M. ; Kl-odzi n?ska , E.; Szeliga , J. Identification of volatile organic compounds secreted from cancer tissues and bacterial cultures . J. Chromatogr. B 2008 , 868 , 88 - 94 .
3. Buszewski , B. ; Ulanowska , A. ; Ligor , T.; etc. Analysis of exhaled breath from smokers, passive smokers and non-smokers by solidphase microextraction gas chromatography/mass spectrometry . Biomed. Chromatogr . 2009 , 23 , 551 - 556 .
4. Arnold , J.W. ; Senter, S.D. Use of digital aroma technology and SPME GC-MS to compare volatile compounds produced by bacteria isolated from processed poultry . J. Sci. Food . Agric. 1998 , 78 , 343 - 348 .
5. Deng , C. ; Zhang, J.; Yu , X. ; Zhang, W. Determination of acetone in human breath by gas chromatography - mass spectrometry and solid-phase microextraction with on-fiber derivatization . J. Chromatogr. B 2004 , 810 , 269 - 275 .
6. Jones , A.W. ; Lagesson , V. ; Tagesson , C. Determination of isoprene in human breath by thermal desorption gas chromatography with ultraviolet detection . J. Chromatogr. B 1995 , 672 , 1 - 6 .
7. Hungate , R.E.; Smith , W. ; Clarke, R.T.J. Suitability of butyl rubber stoppers for closing anaerobic roll culture tubes . J. Bacteriol . 1966 , 91 , 908 - 909 .
8. Mochalski , P. ; Wzorek , B. ; Sliwka , I. ; Amann , A. Improved preconcentration and detection methods for volatile sulphur breath constituents . J. Chromatogr. B . 2009 , 877 , 1856 - 1866 .
9. Huang , R. ; Hsu, P.S. ; Kuo , C.Y.; Chen , S.C. ; etc. Paste extrusion control and its influence on pore size properties of PTFE membranes . Adv. Polym. Tech . 2007 , 26 , 163 - 172 .
10. Hyspler , R.; Crhova, S. ; Gasparic , J. ; Zadak , Z. ; etc. Determination of isoprene in human expired breath using solid-phase microextraction and gas chromatography - mass spectrometry . J. Chromatogr. B 2000 , 739 , 183 - 190 .
11. Forrest , M.J. Rubber analysis-polymers compounds, products . Rapa Rev. Rep . 2001 , 12 , 55 .
12. Doig , S.D.; Boam , A.T. ; Livingston , A.G. ; Stuckey , D.C. Mass transfer of hydrophobic solutes in solvent swollen silicone rubber membranes . J. Membr. Sci . 1999 , 154 , 127 - 140 .
13. Mayfield , H.T.; Burr , E. ; Cantell , M. Analysis of explosives in soil using solid phase microextraction and gas chromatography . Anal. Let . 2006 , 39 , 1463 - 1474 .
14. Liu , J.; Hara , K. ; Kashimura , S. ; etc. Headspace solid-phase microextraction and gas chromatographic - mass spectrometric screening for volatile hydrocarbons in blood . J. Chromatogr. B 2000 , 748 , 401 - 406 .
15. Kolb , B. ; Ettre , L.S. Statistic Headspace - Gas Chromatograpgy: Theory and Practice , 2nd ed. A.J. Wiley and Sons, Inc., Hoboken , New Jersey, p. 64 , 2006 .