Kusch, Knupp

539

Peter Kusch,
Gerd Knupp

Fachhochschule Bonn-Rhein-
Sieg, University of Applied
Sciences, Fachbereich
Biologie, Chemie und
Werkstofftechnik, von-Liebig-
Str. 20, D-53359 Rheinbach,
Germany

Analysis of residual styrene monomer and other
volatile organic compounds in expanded
polystyrene by headspace solid-phase
microextraction followed by gas chromatography
and gas chromatography/mass spectrometry

A method for determination of residual styrene monomer and other volatile organic
compounds in expanded polystyrene (EPS) was developed using HS SPME and gas
chromatography with FID. The extraction products were identified by GC/MS. Good
reproducibility of the measurements with RSD values between 3.2 – 3.6% was achiev-
ed by extraction using a 75 lm Carboxen-polydimethylsiloxane fiber at 608C with
15 min sample sonication. The contents of residual styrene monomer in two samples
of EPS were 153.2 and 65.7 mg/kg, respectively. Other compounds identified in EPS
were pentane, benzene, toluene, ethylbenzene, isomers of xylene, n-propylbenzene,
1,2,4-trimethylbenzene, o-methylstyrene, benzaldehyde, benzyl alcohol, and aceto-
phenone.

Key Words: Styrene; Expanded polystyrene (EPS); Headspace SPME; Carboxen-
poly(dimethylsiloxane); GC/MS

;

Received: October 4, 2001; revised: February 6, 2002; accepted: February 20, 2002

1 Introduction

EPS is an expanded polystyrene, produced by polymeriz-
ing styrene monomer and adding pentane as a blowing
agent. It is used for food packaging and for protection of
products against damage during transport and storage.
EPS is also used in the building industry for insulation of
exterior walls and foundations.

It has been known for a long time that EPS emits residual
styrene monomer and other volatile organic compounds
(VOCs) at ambient temperature. Styrene is harmful when
inhaled, is irritating to eyes, nose, throat, skin, and acts as
a depressant on the central nervous system, causing neu-
rological impairment [1].

VOCs have to be determined in mixtures in many areas of
analytical endeavor, such as environmental, food, foren-
sic, fragrance, oil, pharmaceutical, and polymer/copoly-
mer analysis [2]. The method of choice for many of these
polymer/copolymer analyses is static headspace sam-
pling [3 – 4], dynamic headspace sampling [5 – 8], pyroly-
sis [9, 10], and more recently solid-phase microextraction
(SPME) [11 – 15] followed by GC and GC/MS determina-
tion.

Several headspace-GC or -GC/MS methods have been
used to determine the level of styrene residues emitted
from polystyrene [3, 16]. However, no previous work on
the analysis of residual styrene monomer and other vola-
tile organic compounds in EPS by SPME-GC or SPME-
GC/MS has been reported in the literature.

In this work, headspace SPME followed by GC and GC/
MS has been applied to the analysis of residual styrene
and other volatile organic compounds in EPS. The poly(di-
methylsiloxane) (PDMS) SPME coating is one of the most
widely used coatings for extracting volatile analytes from
environmental samples via absorption [17]. In contrast,
the sensitivity of adsorptive solid SPME coatings, such as
PDMS/divinylbenzene

(PDMS/DVB)

and

Carboxen/

PDMS (CAR/PDMS), was reported to be much higher
compared to PDMS for extracting VOCs [18 – 19]. We
have selected the CAR/PDMS fiber for our investigations.
The fiber works very well in the headspace extraction
mode. Carboxen has a similar surface to DVB. The major
difference is the much higher content of micropores of the
Carboxen/PDMS fiber, making it the overwhelming choice
as material for the extraction of volatile and low molecular
analytes at trace levels. Carboxens differ from other por-
ous carbons because the pores are not sealed but pass
entirely through the particle and allow analytes to desorb
more efficiently than in the case of the sealed pores com-
monly found in charcoals and many carbon molecular
sieves.

J. Sep. Sci. 2002, 25, 539–542

Correspondence: Dr. Peter Kusch, Fachhochschule Bonn-
Rhein-Sieg, von-Liebig-Str. 20, D-53359 Rheinbach, Ger-
many.
E-mail: [e-mail]
Fax: [phone]

Short

Communication

i

WILEY-VCH Verlag GmbH, 69469 Weinheim 2002

1615-9306/2002/0806–0539$17.50+.50/0

2 Experimental

2.1 Materials

The fused silica capillary columns tested in this investiga-
tions were obtained as follows: 30 m60.32 mm ID, film
thickness 4.0 lm, SPB-1 Sulfur from Supelco (Bellefonte,
PA, USA); 60 m60.25 mm ID, film thickness 0.25 lm,
Optima d-3 from Macherey-Nagel (Düren, Germany);
60 m 6 0.25 mm ID, film thickness 0.25 lm, DB-5ms
from J & W Scientific (Folsom, CA, USA); 60 m60.25 mm
ID, film thickness 0.25 lm, BPX-50 from SGE (Mel-
bourne, Australia); 60 m60.32 mm ID, film thickness
0.50 lm, Rtx-225 from Restek (Bellefonte, PA, USA);
50 m60.32 mm ID, film thickness 1.0 lm, Permaphase
CPMS/225 from Perkin-Elmer (Norwalk, CT, USA);
30 m60.25 mm ID, film thickness 0.25 lm, SolGel WAX
from SGE and 60 m60.32 mm ID, film thickness
0.20 lm, Rtx-2330 from Restek.

The SPME fiber holder for manual use and the 75 lm Car-
boxen-Polydimethylsiloxane (CAR/PDMS) fibers were
obtained from Supelco (Bellefonte, PA, USA). The fiber
was conditioned at 2808C for 30 min prior to use.

20-mL headspace glass vials with an aluminum-coated
silicone rubber septum and pressure released aluminum
seal, obtained from LABC Labortechnik (Hennef, Ger-
many), were used.

A Sonorex Super RK 31H compact ultrasonic bath from
Bandelin electronic (Berlin, Germany) was employed for
sample agitation.

2.2 Chemicals

The chemicals used were BTEX standard solution N8 722
372 obtained from Macherey-Nagel (Düren, Germany),
styrene (stabilized) for synthesis obtained from Merck-
Schuchardt (Hohenbrunn, Germany), ethyl methyl ketone
(MEK, internal standard), and N,N-dimethylformamide
(DMF), both of analytical grade, supplied by Merck (Darm-
stadt, Germany).

2.3 Samples

Samples of commercially available expanded polystyrene
(Styropor) from Germany (sample A) and from Australia
(sample B) were used in our investigations.

2.4 Instrumentation

GC analyses were performed on a Perkin-Elmer (Nor-
walk,

CT,

USA)

AutoSystem

gas

chromatograph

equipped with split/splitless injector at 2508C and a flame-
ionization detector (FID) operated at 2708C. Helium 5.0
grade (Westfalen AG, Muenster, Germany) was used as

a carrier gas. The helium inlet pressure was 140 kPa and
the split flow was 50 mL/min. A SPME glass liner for split/
splitless injector was used (Supelco).

The oven temperature was programmed from 808C (1 min
hold) at 58/min to 1008C and then 88/min to 2208C (10 min
hold) or from 608C (1 min hold) at 28C to 1008C and then
88/min to 2008C (10 min hold). Chromatographic data were
processed with TurboChrom 4.0 software (Perkin-Elmer).

GC/MS measurements were made using an Thermo-
Quest Trace 200 gas chromatograph (ThermoQuest CE
Instruments, Milan, Italy) interfaced to a ThermoQuest/
Finnigan Voyager quadrupole mass spectrometer (Ther-
moQuest/Finnigan MassLab Group, Manchester, UK)
with an ThermoQuest Xcalibur data system, the NIST 98
spectra library, and a CombiPAL autosampler (CTC Ana-
lytics AG, Zwingen, Switzerland).

The oven was programmed from 608C (1 min hold) at 58/
min to 1008C and then 108/min to 2208C (10 min hold).
Helium 6.0 grade (Westfalen) was used as a carrier gas.
A constant pressure of 70 kPa helium was used during the
whole analysis. The temperature of the split/splitless
injector was 2508C and the split flow was 10 mL/min. The
transfer line temperature was 2708C. The ion source tem-
perature was kept at 2008C. Ionization was induced by
impacting electrons with a kinetic energy of 70 eV. The
detector voltage was 350 V. Mass spectra and recon-
structed chromatograms (total ion current, TIC) were
obtained by automatic scanning in the mass range m/z
46 – 204. Compound identification was carried out by com-
parison of retention times and mass spectra of standards,
study of the mass spectra, and comparison with data in
the NIST 98 spectra library.

2.5 Procedure

A standard solution of MEK (internal standard) was pre-
pared by making up 100 mg of the substance with DMF to
100 mL. A 30 lL aliquot of this solution (containing 30 lg
of the internal standard) was added from a 100 lL syringe
to 50 mg of the crumbled EPS sample sealed in a 20 mL
headspace glass vial with aluminum-coated silicone rub-
ber septum. The septum of the vial was pierced with the
needle of the SPME device and the fiber was exposed
approximately 10 mm above the sample. Afterwards the
glass vial with the SPME injector was placed in an ultraso-
nic bath and agitated by sonication for 1 to 30 min at ambi-
ent temperature, at 408C and at 608C. Furthermore, EPS
samples were heated in the heating block of the Combi-
PAL autosampler equilibrated at 808C without agitation.

After a sorption time of 1 to 30 min in the headspace
above the sample, the fiber was retracted into the protec-
tive sheath and removed from the vial. It was transferred
without delay into the injection port of the gas chromato-

540

Kusch, Knupp

J. Sep. Sci. 2002, 25, 539–542

graph or the GC/MS. The fiber was thermally desorbed in
the injector at 2508C for 3 min and the run was started.
The samples for GC/MS identification were analyzed with-
out using an internal standard solution.

For the determination of the HS SPME response factor
50 lL (0.0530 g) of styrene and 50 lL (0.0394 g) of MEK
were placed a 10 mL flask and the flask was filled up with
DMF. 1 mL of the solution obtained was then diluted 1 : 10
with DMF. 2 mL of the diluted solution was transferred to
the headspace vial for adsorption onto the CAR/PDMS
fiber and than analyzed by GC after thermal desorption of
the fiber in the injector. The HS SPME conditions and the
analytical conditions were the same as in the determina-
tion of EPS samples.

3 Results and discussion

The studies were conducted to establish the optimum
conditions for extraction, desorption, chromatographic
separation, and identification of the VOCs from the
expanded polystyrene using headspace SPME followed
by GC and GC/MS. Extraction temperature and time,
sample agitation technique, and desorption temperature
and time are very important parameters in the optimiza-
tion of the SPME. In general, the highest possible tem-
perature should be used. In headspace SPME, an
increase in extraction temperature leads to an increase of
analyte concentration in the headspace, and helps to facil-
itate faster extraction [17]. However, at high temperature,
coating headspace partition decreases and the fiber coat-
ing begins to lose its ability to adsorb analytes. Thus the
temperature effect between 25 – 808C was studied.

The most efficient agitation method evaluated to date for
SPME applications is direct sample sonication, which can
provide very short extraction times [17]. In this work we
have used sonication as sample agitation method.

The choice of the desorption temperature of the GC injec-
tor is also a critical point. The results obtained show that
after 3 min at 2508C a complete desorption of all com-
pounds was achieved.

For the chromatographic separation of residual VOCs
from samples of the expanded polystyrene, several fused
silica capillary columns with stationary phases of different
polarity were examined. The best results were achieved
using the Rtx-225 (60 m, 0.32 mm, 0.5 lm) or Perma-
phase-CPMS/225 (50 m, 0.32 mm, 1.0 lm) columns with
the (50%-cyanopropenyl)-dimethylpolysiloxane station-
ary phase. Figure 1 shows the GC/FID chromatograms of
both the investigated EPS samples handled by SPME
with the 75 lm CAR/PDMS fiber in the headspace mode
at ambient temperature and agitated by sonication for
15 min. The peaks were identified by GC/MS and by using
the mixture of BTEX standards.

In order to optimize the extraction temperature, the con-
tent of styrene monomer in EPS was determined at four
different temperatures. The internal standard method was
used for the quantitative determination according to
Eq. (1):

c

i

= A

i

N

f

i

N

W

st

N

10

6

/A

st

N

W

p

(1)

where A

i

refers to the peak area of styrene, and A

st

refers

to the peak area of the internal standard (MEK), W

st

is the

mass of MEK, W

p

is the mass of the sample (EPS), and f

i

is the HS SPME response correction factor of styrene
(see Section 2.5). The HS SPME response correction fac-
tor of styrene was calculated according to Eq. (2):

J. Sep. Sci. 2002, 25, 539–542

Analysis of styrene and other VOCs in EPS

541

Figure 1. SPME/GC/FID chromatograms of a headspace of
EPS at ambient temperature, agitated for 15 min by sonica-
tion. Top chromatogram: sample A, bottom chromatogram:
sample

B.

Fused

silica

capillary

column:

Rtx-225,

60 m60.32 mm ID, film thickness 0.5 lm. Column tempera-
ture: programmed from 808C (1 min hold) at 58/min to 1008C
and then 88/min to 2208C (10 min hold). Split/splitless injec-
tor: 2508C, split flow 50 mL/min, FID: 2708C. Peak identifica-
tion:

1 = pentane

(3.95 min);

2 = non-aromatics

(5.88 –

6.32 min); 3 = benzene (6.63 min); 4 = toluene (7.05 min);
5 = ethylbenzene

(7.60 min);

6 = m/p-xylene

(7.81 min);

7 = o-xylene (8.48 min); 8 = styrene (9.10 min); 9 = n-propyl-
benzene (9.29 min); 10 = 1,2,4-trimethylbenzene (9.51 min);
11 = o-methylstyrene

(12.19 min);

12 = benzaldehyde

(13.66 min); 13 = acetophenone (16.15 min); 14 = benzyl
alcohol (16.80 min).

f

i

= c

i

N

A

st

/c

st

N

A

i

(2)

where c

st

and c

i

are the content of MEK and styrene in the

mixture, respectively, and A

st

and A

i

are the peak area of

MEK and styrene, respectively. The value obtained was
0.2 l 0.01 (n = 20).

The results obtained and precision (repeatability) of the
extraction for both the EPS samples are summarized in
Table 1. As can be seen from Table 1, six replicate SPME
extractions at 608C with 15 min sonication followed by GC
measurements showed RSD within 3.6% for EPS sample
A and within 3.2% for EPS sample B. The repeatability of
the extraction at ambient temperature and at 408C was
not

satisfactory

(RSD = 7.7% – 11.7%).

The

results

obtained by extraction at 808C without sonication indicate
a degradation process of the EPS (retropolymerization).
Therefore, the extraction temperature of 608C was cho-
sen as the optimum. In order to determine the equilibration
time, different extraction times between 1 and 30 min at

608C with sonication were studied. Figure 2 shows the
effect of the extraction time at 608C on the peak area of
styrene determined in EPS, sample A. The extraction time
of 15 min was chosen for subsequent analyses.

4 Conclusions

The proposed HS-SPME-GC procedure is appropriate for
determination of residual styrene monomer and other
VOCs in expanded polystyrene. The rapid, simple, and
cost-effective method with good repeatability can be
applied to the process control of EPS production and to
measure and screen styrene and VOCs in the environ-
ment.

References

[1] H. Schmitt, Baubedarf Manager 1985, June, 1 – 6.

[2] M. E. Miller, J. Stuart, Anal. Chem. 1999, 71, 23 – 27.

[3] B. Kolb, L.S. Ettre, Static Headspace-Gas Chromatogra-

phy: Theory and Practice, Wiley-VCH, New York 1997.

[4] W. Tashkov, V. Getchev, Chromatographia 1993, 37,

546 – 548.

[5] S. Jacobsson, J. High Resol. Chromatogr. 1984, 4,

185 – 190.

[6] D.W. Wright, K.O. Mahler, J. Davis, J. Chrom. Sci. 1992,

30, 291 – 295.

[7] M. Kaljurand, H.C. Smit, Chromatographia 1994, 39,

210 – 215.

[8] S.C. Hodgson, R.J. Casey, J.D. Orbell, S.W. Bigger, J.

Chem. Educ. 2000, 77, 1631 – 1633.

[9] P. Kusch, Chem. Anal. 1996, 41, 241 – 252.

[10] P.B. Smith, A.J. Pasztor, Jr., M. McKelvy, D.M. Meunier,

S.W. Froelicher, F.C.-Y. Wang, Anal. Chem. 1997, 69,
95R – 121R.

[11] M. Hakkarainen, A.-C. Albetsson, S. Karlsson, J.

Environ. Polym. Degrad. 1997, 5, 67 – 73.

[12] H. C

ˇ izˇkova, M. Voldrˇich, J. Dobiasˇ, Czech J. Food Sci.

1998, 16, 81 – 84.

[13] D.C. Hinz, W. Kwarteng-Acheampong, B.W. Wencla-

wiak, Fresenius J. Anal. Chem. 1999, 364, 641 – 642.

[14] Y. Takao, K. Arizono, Shigen Kankyo Taisaku 1999,

35(5), 447 – 451.

[15] R. Charvet, C. Cun, P. Leroy, Analusis 2000, 28, 980 –

987.

[16] S. Abrantes, J. High Resol. Chromatogr. 1993, 16, 113 –

115, and references cited therein.

[17] J. Pawliszyn, Solid-Phase Microextraction: Theory and

Practice, Wiley-VCH, New York, 1997.

[18] T. Go´recki, X. Yu, J. Pawliszyn, Analyst 1999, 124,

643 – 648.

[19] J. Kozieł, M. Jia, J. Pawliszyn, Anal. Chem. 2000, 72,

5178 – 5186.

[JSS 1116]

542

Kusch, Knupp

J. Sep. Sci. 2002, 25, 539–542

Table 1. Results of the GC/FID determination of residual
styrene in expanded polystyrene (EPS) after headspace
SPME for 15 min at different temperatures. Samples at tem-
peratures 25, 40, and 608C were agitated by sonication.

Styrene, mg/kg

EPS (A)

EPS (B)

i

258C

408C

608C

808C

258C

608C

1

125

137

154

210

43

69

2

111

128

157

194

41

66

3

127

143

161

192

34

67

4

141

112

145

221

38

66

5

131

134

147

199

47

63

6

139

117

155

198

48

63

n

6

6

6

6

6

6

Mean

129.0

128.5

153.2

203.3

41.8

65.75

RSD,%

7.7

8.5

3.6

5.0

11.7

3.2

Figure 2. Effect of the extraction time on the peak area of
styrene determined in EPS, sample A at 608C, agitated by
sonication for 15 min. Analytical conditions: fused silica capil-
lary column Permaphase-CPMS/225 (50 m60.32 mm ID,
film thickness 1.0 lm), column temperature programmed
from 608C (1 min hold) at 28/min to 1008C and then 88/min to
2008C (10 min hold), split/splitless injector at 2508C, split
flow 50 mL/min, FID at 2708C.