Journal of Chromatography A, 1216 (2009) 470–480

Contents lists available at ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Review

Analysis of musk fragrances in environmental samples

Kai Bester a,b,∗
a
Department for Biotechnology, Chemistry, and Environmental Engineering, Aalborg University, Sohngaardsholmsvej 57, DK-9000 Aalborg, Denmark
b
Institute of Environmental Analytical Chemistry, University Duisburg-Essen, Universitätsstr. 15, 45141 Essen, Germany

a r t i c l e i n f o a b s t r a c t

Article history: The methods for the determination of polycyclic and nitro-aromatic musk compounds in comparison
Available online 2 September 2008 to other fragrances such as OTNE ([1,2,3,4,5,6,7,8-octahydro-2,3,8,8-tetramethylnaphthalen-2yl] ethan-
1-one) as well as those for the respective metabolites are described in this contribution. It covers
Keywords: instrumental aspects, as well as procedures for extraction and clean-up. Protocols for the determina-
Synthetic fragrances tion of musks in water, sludge, biota, and air are summarised and discussed. Extractions by means of solid
Musks
phase extraction (SPE) and liquid–liquid extraction (LLE) in case of water samples are evaluated for the
Enantioselective determination
diverse applications, i.e., wastewater, surface water and seawater. While LLE is preferred for the analysis
of bulk for transport studies and for special process studies SPE might be worth the effort. Considering
sludge, sediment and biota samples, drying and successive accelerated solvent extraction. Soxhlet extrac-
tions as well as cold column extractions are being compared. ASE has proven to be the most exhaustive and
quickest to adopt method. Clean-up by means of size exclusion chromatography and silica sorption chro-
matography with their respective merits and problems are demonstrated. Suggestions for routine and
research analysis are also given. The diverse approaches for enantioselective separations are discussed
in respect to HHCB, AHTN and the metabolite HHCB-lactone. The power of two-dimensional (GC × GC)
approaches is demonstrated considering the various production impurities (isomers) of the two poly-
cyclic musks with the highest usage rates. The usage of tandem mass spectrometry and high resolution
mass spectrometry for the same purpose is also discussed. The identification of an isomer of the HHCB-
transformation product HHCB-lactone from wastewater treatment that has not been described in the
literature before, is presented as well. Additionally some ideas to make the REACh process more efficient
are discussed considering the special experiences from the development of the analysis of musk fragrances
in the environment.
© 2008 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
1.1. Nitro-aromatic musks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
1.2. Polycyclic musks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473
1.3. Macrocyclic musks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
1.4. Other fragrances with environmental concern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2. Methods for determination of synthetic fragrances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.1. Blank issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.2. Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.3. Internal standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.4. Instrumental analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.4.1. Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
2.4.2. Chromatographic approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475
2.5. Sample preparation and clean-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475
2.5.1. Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475
2.5.2. Sludge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477

∗ Correspondence address: Department for Biotechnology, Chemistry, and Environmental Engineering, Aalborg University, Sohngaardsholmsvej 57, DK-9000 Aalborg,
Denmark. Tel.: +45 9940 9939.
E-mail address: kb@bio.aau.dk.

0021-9673/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.chroma.2008.08.093
 K. Bester / J. Chromatogr. A 1216 (2009) 470–480 471

2.5.3. Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478

2.5.4. Biota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
2.5.5. Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
2.5.6. Commodities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
3. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479

1. Introduction close skin contact (e.g., via treated laundry) human exposure is also
high. Up to 190 ng HHCB/g lipid has been reported for human con-
Fragrances have been used since antiquity improve attractive- tamination [22–24]. It was also shown that work hygiene issues
ness of persons and items. However, until the middle of the last might arise from high exposure through air, e.g., in hairdressers as
century biogenic compounds such as floral or animal extracts were an involuntary exposure to these compounds [25]. Additionally a
used. Around 1950 several synthetic fragrances became cheaply long range transport and air water distribution issue has risen on
available [1]. However, it took until the eighties of the last cen- the Great Lakes [14,15,26]. Though, musks share in principle high
tury until a proof was given, that these compounds were present in lipophilicity, it should be taken into account, that octanol water par-
the environment as a consequence to vast usage and incomplete titioning (Kow ), bioconcentration factors (BCF) and organic carbon
removal in wastewater treatment [2–4]. An overview on chem- water partitioning (Koc ) differ for some orders of magnitude for the
ical properties of these synthetic fragrances is given in Table 1. diverse compounds [16,12].
For some reason the musks, such as the nitro-aromatic musks
musk xylene or musk ketone as well as the polycyclic musk such 1.1. Nitro-aromatic musks
as 7-acetyl-1,1,3,4,4,6-hexamethyl-1,2,3,4-tetrahydronaphthalene
(AHTN), trade name, e.g., Tonalide® or 1,3,4,6,7,8-hexahydro- This group of musks all have a nitro-aromatic part (Table 1).
4,6,6,7,8,8-hexamethylcyclopenta-(g)-2-benzopyran (HHCB), trade Discussions on toxicology have risen very early, and it was demon-
name, e.g., Galaxolide® became of mayor concern in this debate strated, that these compounds can be transformed in wastewater
(Table 1). The reason for this not only had scientific and technical treatment as well as in vertebrate physiology into the aniline trans-
aspects, such a the usage volumes, and bioconcentration factors, formation products [27–29] (Fig. 1). Possibly these transformation
but also included elements of chance, such as the ease with which products are even more problematic that the parent compounds.
the respective compounds were identified by means of non-target Though both isomers are possible, in the environment usually
screenings by gas chromatography coupled to mass spectrometry the 4-amino musk xylene predominates [9,28].
with electron impact ionisation. A spectrum of a nitro-aromatic These transformation products are the main reason, why musk
compound was more easily elucidated and identified as stem- xylene has been withdrawn from most European markets, and thus
ming from an anthropogenic compound than one resembling concentrations in wastewater have dropped by about two orders of
to a terpenoid-like structure (OTNE ([1,2,3,4,5,6,7,8-octahydro- magnitude [3,8,9]. Not so many time series for the nitro-musks are
2,3,8,8-tetramethylnaphthalen-2yl] ethan-1-one; brand name e.g., available. However, data gained from blue mussels from the Jade
Iso-E-super). Also, only some of the compounds are represented Bight (North Sea, Germany) revealed a decreasing trend for musk
in mass spectral libraries, while others are not. One polycyclic ketone. In 1988 up to 0.6 ng/g wet weight (ww) were documented in
musk compound, i.e., versalide, proved to be neurotoxic and was samples from the Germen specimen bank, all concentrations were
thus withdrawn from the market earlier [5]. However, there are
other compounds such as OTNE ([1,2,3,4,5,6,7,8-octahydro-2,3,8,8-
tetramethylnaphthalen-2yl] ethan-1-one), which are nowadays
used with higher volumes than polycyclic or nitro-aromatic musks
in Europe. For a comparison: the consumption of OTNE, HHCB,
AHTN, musk xylene and musk ketone in Europe in 1998 were
3000, 1473, 385, 86 and 40 t annually, respectively [1,6]. The nitro-
aromates being removed from all high volume commodities in
Europe, and thus they are only present in some perfumes nowadays.
Also the efforts especially in the detergent industry to reduce poly-
cyclic musks content in washing powders and softeners resulted in
decreasing concentrations of the polycyclic musks in the environ-
ment. Thus, the concentrations of HHCB in bream from river Rhine
(Weil, Germany) have decreased from 2710 ng/g lipid weight in
1995 to 1960 ng/g lipid in 2003 [6]. AHTN concentrations decreased
from 837 to 203 ng/g lipid in this dataset, respectively [6]. The
reader should be aware, that there are some indications for a differ-
ent situation (i.e., higher usage of nitro-musks and polycyclic musk
in comparison to OTNE) in the US and in Asia (e.g., [7,39,44]).
Nowadays it is well known, that these compounds are emitted
into the wastewater [3,7–12], reach freshwaters [8,3,13] and also
huge freshwater reservoirs such as the Great Lakes [14] as well as
the marine environments [15,16]. As most of these compounds are
very lipophilic, they tend to accumulate in sediments, sludges, and
biota such as fish [12,17–22,53]. However, as humans also are in Fig. 1. Transformation of musk xylene to musk xylene amines.
472 K. Bester / J. Chromatogr. A 1216 (2009) 470–480

Table 1
Overview on synthetic fragrances in environmental discussions

Name; abbreviation (trade name) Structural formula CAS no. References

Polycyclic musks

7-Acetyl-1,1,3,4,4,6-hexamethyl-1,2,3,4- 1506-02-1 [21,44]

tetrahydronaphthalene; AHTN (e.g.,
Tonalide® )

6-Acetyl-1,1,2,3,5-hexamethyldihydroindene; AHDI 15323-35-0 [21]

(Phantolide® )

5-Acetyl-1,1,2,6-tetrametyl-3-isopropyl-dihydroindene; ATII 68140-48-7 [21]

(Traesolide® )

1,3,4,6,7,8-Hexahydro-4,6,6,7,8,8-hexamethylcyclopenta-(g)- 1222-05-5 [21,44]

2-benzopyran; HHCB
(Galaxolide® )

1,3,4,6,7, 8-Hexahydro-4,6,6,7,8,8-hexamethylcyclopenta [17]

[g]-2-benzopyran-1-one; HHCB-lactone (Galaxolidone)

Nitro aromatic musks

Musk ketone 81-14-1 [44,24]

 K. Bester / J. Chromatogr. A 1216 (2009) 470–480 473

Table 1 (Continued )

Name; abbreviation (trade name) Structural formula CAS no. References

Musk xylene 81-15-2 [44,24]

Macrocyclic musks

Ethylenbrassylate 105-95-3

Habanolide 34902-57-3

Cyclopentadecanolide 106-02-5

Other fragrances

[1,2,3,4,5,6,7,8-Octahydro-2,3,8,8-tetramethylnaphthalen- 54464-57-2 [44]

2yl] ethan-1-one; OTNE
Iso-E-super

below 0.1 ng/g ww following 1996. The drop in concentration in
surface water may however also be partially due to installation of
denitrification processes in most wastewater treatment plants in
Europe at the same time. In this process the nitro-musks are usu-
ally quantitatively transformed into the amino derivatives. A risk
assessment on musk xylene and musk ketone has been performed
in 1999 by Tas et al. [30].
1.2. Polycyclic musks
Several polycyclic musks have been introduced onto the mar-
ket [1] (Table 1). However, currently the musks HHCB and AHTN
are dominating the market [9]. AHTN seemingly being phased out
drastically in the last decade, while HHCB is still present in rather
high concentrations [5,8]. The transformation of polycyclic musks
has been elaborated in [12]. While the suggested transformation
scheme for AHTN is complex, HHCB, seems to be predominantly
oxidised to the respective HHCB-lactone also nominated Galaxoli-
done [17]. This six-ring lactone can be further hydrolysed to form
the respective acid [12]. However, this reaction is a pH-driven equi-
librium reaction, thus it will be impossible to prove or falsify which
of the two compounds (lactone or acid) is present in aquatic sam-
ples by indirect methods such as GC–MS or other chromatographic
techniques. Maybe direct infrared spectroscopic (IR) or nuclear
magnetic resonance (NMR) based methods give more insight into
this. However, as the equilibrium of this reaction should be reached
474 K. Bester / J. Chromatogr. A 1216 (2009) 470–480
quickly, it will be rather irrelevant which of the two compounds was
present in the beginning.
An interesting aspect for toxicology as well as for analytical
chemistry is, that most of the polycyclic musks are chiral [1,17].
The current industrial synthesis is racemic, though normally only
one enantiomer, e.g., S-AHTN but not R-AHTN is effective [1,31].
Some compounds, such as HHCB even contain two chiral cen-
tres, thus four stereoisomers are possible, but only the two 4S
stereoisomers are active to the receptors in the human nose [31].
These two stereoisomers leave a different sensoric impression,
each. These issues in enantioselectivity might become a topic in
chemical regulation such as REACh [32] as it already is in pesticides
and pharmaceuticals.
A risk assessment on HHCB and AHTN has been performed in
1999 [33]. Currently there is a debate as to which extent especially
polycyclic musks have also estrogenic or anti-androgenic endocrine
disrupting activity [34–37].
1.3. Macrocyclic musks
Macrocyclic musks (Habanolide, Cyclopentadecanolide,
Ethylenbrassylate) are getting more and more available. From
some points of view they do have advantages in comparison to
the polycyclic ones, though they are currently more expensive
than those as calculated per kg. They seem to be smelling more
intensive, thus less mass is needed to gain the same performance
in perfumery, which is of environmental benefit [1,31]. Addition-
ally these compounds seem to be more easily degradable in the
environment [9]. However, their mass spectra resemble very much
those of natural fatty acids or their derivatives, and may thus
often escape attention while analysing environmental samples as
well as commodities. Also their chemical properties are similar
to those of natural products, so separation from these is more
difficult. Currently there are no reports on environmental samples
in connection with macrocyclic musks, though several groups have
included these compounds into their research programmes.
1.4. Other fragrances with environmental concern
A shift in fashion as well as in environmental concerns seem
to have triggered a change away from the more “erotising” musks
to woody or fruity smelling compounds, e.g., OTNE, marketed, e.g.,
as Iso-E-super [1]. First reports have shown, that this compound is
also present in environmental samples. In Europe even in higher
concentrations that those of the musks are reported [9,38,39].
Other fragrance compounds like lyral have up to now rather been
discussed due to their potency to induce allergies than to environ-
mental issues.
Musks have also been discussed as a marker for xenobiotic con-
taminations concerning sewage sludges to be used as fertilisers on
agricultural soils [40,41].
Thus, fragrances in environmental and food issues as well as
in human exposure is currently an important issue. It might be the
right time to evaluate the published methods for determining these
compounds in environmental and food samples. It might also be
time to consider the lessons learnt from the often random research
on synthetic fragrance compounds in the environment for manage-
ment programmes of chemicals such as REACh, i.e., it is important,
that mass spectra of compounds with market relevance compounds
as well as good analytical standards are available.
2. Methods for determination of synthetic fragrances
Analysis of musk compounds and other fragrances require all
usual method validation data, such as determination of recovery
rate for the respective extraction and clean-up procedures, repro-
ducibility of the used procedures as well as determining the linear
or working range. In this paper attention will only be given to those
issues that are of special concern in analysing these compounds
and not to those, which are well known to all organic trace anal-
ysis by chromatographic methods (e.g., cleanliness of injector and
column).
2.1. Blank issues
Analysing fragrances asks some discipline from the laboratory
personnel. Soaps, creams, shower-gel, shampoos, washing soften-
ers, etc. which contain the target compounds have to be avoided
for labcoats, and detergents used in the laboratories as well as in
the personal use of the laboratory personnel. Also contamination
from room cleaning agents, such as floor polishers, window clean-
ers, etc. is to be taken into account. Sunscreen protection creams
and lotions might likewise be critical especially during sampling
campaigns. Indeed in contrary to most other analytes, these com-
pounds might be contained in virtually everything used in daily life.
For an overview on musk compounds in household commodities
the reader might refer to Reiner and Kannan [42]. Concentrations
as high as 5,000,000 ng/g in a perfume was detected, while environ-
mental concentrations for HHCB are, e.g., 100 ng/L in surface water,
or 10,000 ng/g in sludge or biota samples [9,12], while the nitro-
aromatic musks concentrations are normally about two orders of
magnitude lower.
2.2. Standards
Most fragrance analytes have become available as commer-
cial compounds, except HHCB. This compound is commercially
currently only available with ∼50% purity. The impurity being
dibutyl-phthalate. On the other hand, the transformation products
of the respective fragrances become more and more known and
relevant, but none of them are available on the market. While it is
more or less easy to synthesize the respective nitro-musk amines
[27–29], it is not quite so simple to purify them. The synthesis
of HHCB-lactone (the major transformation product of HHCB) is
tedious [17].
2.3. Internal standards
Currently only two reasonable internal standards (IS) for quan-
tifying fragrance compounds are available: Deuterated (D15 ) musk
xylene and deuterated (D3 ) AHTN. However, (D3 ) AHTN has been
discussed [7,43] to undergo proton exchange and thus the deuter-
ation is reversed. Thus, the IS might be lost and additional AHTN
might be produced during sample processing and storage. Addi-
tionally fragrances such as versalide that have long been stopped
to be used in fragrance industry due to toxicological issues [5]
might be used as IS. However, Simonich et al. [44,45] used a vari-
ety of deuterated compounds (D3 -terpeniol, D3 -benzyl acetate,
D3 -␥ methyl jasmonate, D3 -dihydrojasmonate, D3 -OTNE, D4 -acetyl
cedrene, D6 -musk xylene, D3 -AHTN, D7 -musk ketone, that have not
become commercially available.
2.4. Instrumental analysis
2.4.1. Detection
Nitro-musks have successfully been analysed by GC-AIFD or
GC-ECD as the nitro-groups induce signals on both of these detec-
tors with considerable response [27,28]. However, more emphasis
on sample clean-up as well as verification on a second (different)
gas chromatographic column is required to prevent false positive
 K. Bester / J. Chromatogr. A 1216 (2009) 470–480
results. Concerning all other fragrance compounds, neither poly-
cyclic and macrocyclic musks nor OTNE have functional groups
that can be utilised by analytic techniques, thus mass spectrom-
etry with electron impact (EI) is routinely used for detection of
all musk fragrances [2,3,6,7]. Usually three mass fragments can be
detected, so that verification can be performed within the same gas
chromatographic run as the quantification. However, the fragmen-
tation pattern usually is dominated by cleavage of methyl groups,
giving thus not very significant fragments. Quadrupole mass spec-
trometers have been utilized as well as ion traps [6,29]. The latter
ones have preferentially been used in chemically induced dissoci-
ation mode to improve sensitivity and selectivity (MS/MS in time)
[29]. High resolution or tandem mass spectrometric techniques are
rarely used, as the sensitivity of the low resolution mass spectrom-
eters usually is sufficient for the analysis of these compounds. In
the literature there are no indications for compounds that co-elute
with musk fragrances, giving similar mass spectra. However, there
are several production impurities (probably stereoisomers) that can
be detected, e.g., in HHCB. However, as these compounds usually
are isobaric, neither tandem MS nor high resolution mass spec-
trometry will give better (more true) results. It is rather feasible to
perform a better gas chromatographic separation to make sure the
peak is identified correctly.
2.4.2. Chromatographic approaches
Due to their inherent properties, all fragrances have a rea-
sonable vapour pressure. They are stable against temperature
[33,30], and they are lipophilic (Table 1), thus they can eas-
ily be analysed by gas chromatography (GC), as long as lipids
and other macromolecules are eliminated from the sample [46].
All current injectors such as split/splitless (SSL), on-column (OC)
and programmed temperature vapourizers (PTV) have been used
successfully [7,8,17,27,28,44,47,46] as the musk fragrances of envi-
ronmental concern are not thermo-labile. While SSL injections are
usually limited to about 4 ␮L injection volume at their best, 50 ␮L
can be injected to PTV and OC injectors. However, this only makes
sense, if the samples are very clean and the limit of quantification
is dominated by the physics of the respective instrument such as
experienced for the analysis of sea and drinking water samples. OC
and PTV injectors can get easily polluted and then give unreliable
results.
For the separation of the musk compounds often columns
with films containing 5% phenyl in dimethylsilylsiloxane such as
DB-5 are used. However, these columns are often only able to
perform a partial separation or no separation at all of the diastere-
omers of HHCB [7,8,14,47]. For this purpose the RTX-50 (50%
phenyl in dimethylsiloxane) have been used successfully [6,40,41].
An overview of chromatographic approaches is given in Table 2.
Usually the operating speed of the used quadrupole mass spec-
trometers does not allow to decrease peak width below 2 s, thus
the used columns have usually an inner diameter of about 0.2 mm.
Additionally the enantiomers of the polycyclic musks and their
metabolites can be separated by enantioselective gas chromatog-
raphy [8,9,12,17,19,48]. Thus, new insights into bioaccumulation,
transformation and the underlying processes can be gained.
Currently only one enantioselective GC phase (i.e., heptakis-(2,3-
di-O-methyl-6-O-t-butyldimethyl-silyl)-␤-cyclodextrin) has been
used to perform this enantioseparation, thus some problems
considering verification of results arise. It must be taken into con-
sideration, that HHCB, AHTN and several production impurities are
isomers, thus around 30 peaks with similar mass spectra have to
be separated if all stereoisomers and enantiomers are considered.
Sophisticated mass spectrometry (high resolution or tandem mass
spectrometry) [17] does not overcome this obstacle, as all the con-
cerning compounds are isomers, thus isobaric signals are obtained.
475
Fig. 2. Log plot of a comprehensive 2D GC separation of all stereoisomers of
HHCB, AHTN and their production impurities. Primary column: enantioselec-
tive (25 m 0.25 mm i.d., heptakis-(2,3-di-O-methyl-6-O-t-butyldimethyl-silyl)-␤-
cyclodextrin) secondary column: DB-wax, 1 m 0.1 mm i.d.).
To overcome this calamity, comprehensive two-dimensional gas
chromatography (GC × GC) was used to separate all compounds
[48]. In this approach the enantioselective column was the primary
one, while a 1 m DB-wax column with an inner diameter of 0.1 mm
was used as a secondary dimension column with a time of flight
mass spectrometer as a detector. Time of flight mass spectrometers
can give the resolution in time needed for these high speed sec-
ondary column chromatograms. In this set-up it is even beneficial,
that the chromatographic peaks on the enantioselective column are
somewhat broader (15 s) than on classical capillary columns (5 s) so
more time could be spent on the secondary chromatogram (Fig. 2).
2.5. Sample preparation and clean-up
2.5.1. Water
The main issue for analysing musk compounds from water sam-
ples are blanks, as concentrations in the water samples may be
in the ng/L (ppt) range and products used in the laboratory might
contain the analytes in much higher concentrations (see part on
commodities).
2.5.1.1. Wastewater. Wastewater usually contains large amounts of
surfactants and particles that may interfere with the analysis of
organic micro-pollutants. This holds especially true if solid phase
extraction (SPE) is utilised. These obstacles can in principle be over-
come by either filtration, centrifugation or by using larger amount
of extraction materials or extraction devices with higher diameter,
to decrease clogging (e.g., extraction disks, which are nowadays
available with polymeric sorbents or C18-derivatised silica instead
of extraction cartridges). As the filter residue might sorb some of
the analytes from the water passing through, it might be hard to
validate such a filtration procedure properly. Particle separation by
centrifugation before SPE usually works well, though the effort for
performing centrifugation of wastewater for trace analysis under
high throughput is high.
Musks have been extracted using solid phase extraction (SPE)
by C18 Empore disk [44,45], Biobeads SM-2 [43], divinylbenzene
based polymers [49] or by using liquid–liquid extraction (LLE) using
toluene [7,8] or n-hexane [50] from wastewater. LLE has the advan-
tage, that particles and surfactants usually do not influence the
extraction very much. Artola-Garicano et al. [10] used solid phase
micro-extraction (SPME) to determine the dissolved fraction of
 476
Table 2
Overview on methods for solid environmental samples

Extraction Clean-up 1 Clean-up 2 + 3 GC injector GC separation Detector Remark References

column

Soxhlet SEC SX-3 Layered silica and alumina, SSL HP-1 ECD (nitro-musks) Biota and sediment [56]
n-hexane/toluene
EE/Cycl. 1:1 EE/Cycl. 100/98.5 ITD-MS (polycyclic musks)

Cold extraction water/acetone/petroleum ether SEC SX-3 NP-HPLC (silica) n-hexane OC SE 54 AFID Biota [18]
EE/Cycl. SSL DB-5MS ITD-MS (EI)
1/1

K. Bester / J. Chromatogr. A 1216 (2009) 470–480

ASE DCM (in cell clean-up with silica) Sodium sulphate Silica column n-hexane/DCM ng DB 1701 MS (quadrupole) Sludge, soil [55]
column n-hexane
ASE SEC Aminopropyl-Silica ng ng MS quadrupole Biota [49]
DCM

Soxhlet Activated copper Silica column n-hexane HP5MS ITD-MS Sediments [47]
DCM Quadrupole MS

Soxhlet Activated copper SEC ng CPSil 8 MS Quadrupole Sludges [60]

Dichlorormethane Followed by: layered SX-3 HP5MS

silica/alumina
DCM/pentane 1/1 n-hexane/DCM
1/1

ASE n-hexane SEC SX-3 Silica SSL RTX-50 ITD-MS in CID Biota [6]
DCM/n-hexane 1/1 DCM/cyclohexane

Soxhlet SEC Silica ng BPX 5 MS quadrupole Biota [61]

DCM/n-hexane

Soxhlet SEC SX-3 Silica ng DB 5 MS quadrupole Biota including [23]

human tissue
DCM/n-hexane 3/1 DCM/n-hexane 1/1

Soxhlet SEC SX-3 Silica n-hexane MTBE PTV DB 5 MS quadrupole Sludge [7,9]
EE EE/Cycl. 1/1
AFID: alkali flame ionisation detector, ASE: accelerated solvent extraction, EE: ethyl acetate, Cycl.: cyclohexane, ECD: electron capture detector, ITD-MS ion trap detector mass spectrometer, MTBE: methyl tert-butylether, ng:
not given, SSL: split/splitless injector, SEC size exclusion chromatograph.
 K. Bester / J. Chromatogr. A 1216 (2009) 470–480 477

the fragrances in wastewater and wastewater treatment. However,

before this approach is used on a broader scale it should be tested,
whether sorption to the SPME-fiber is faster than desorption from
particles contained in wastewater.

2.5.1.2. Surface water. Surface water usually contains low amounts

of analytes but also less particles than, e.g., raw wastewater. It has
been extracted by SPE, using XAD-2 resin [14], or liquid–liquid
extraction using toluene [8] or n-pentane [16].
While in wastewater or normal surface waters 0.1–1 L sample
intake are utilized, in marine areas or Great Lakes sample intake was
around 2–1600 L [14–16] depending on the respective concentra-
tions and instrumental analytical techniques [14–16]. Additionally
the sensitivity of the method can be improved by large volume
injections (up to 50 ␮L from a 1 mL extract) if the blank problems
are under control. With such methods LOQs in the range of about Fig. 3. Chromatographic characterization of two HHCB-lactones.
10–30 pg/L can be reached [15,16].
Water has also been analysed by simultaneous steam distilla-
tion [20], though this difficult to control and not a lot of sample
throughput can be generated.
While there is no doubt, that all these methods can give
correct results, the author does not see any added value in apply-
ing SPE for lipophilic compounds such as the musk fragrances
from bulk water, which usually gives severe problems in clog-
ging if the particles are not removed from the samples before
extraction, as these compounds are very lipophilic and can thus
be easily extracted by LLE. Sometimes decreased recovery rates
are experienced when using SPE for natural (surface) waters for
very lipophilic compounds in comparison to LLE [51]. Only for
high volumes exceeding 100 L, there is a clear benefit in SPE as
LLE will give handling problems if the sample volume exceeds
100 L. On the other hand centrifugation/SPE can give valuable
information on processes concerning compounds that are particle
bound.

2.5.1.3. Metabolites in wastewater. For the nitro-aromates, it must

be taken into account that in most current day’s wastewater
treatment plants the respective amines are being formed, prob-
ably during denitrification. For AHTN only photo-transformation
products have been described in the literature [52,12]. However,
photooxidation is not relevant in current sewer systems or wastew-
ater treatment. For HHCB an oxidation product (HHCB-lactone) as
well as the respective HHCB-acid is known to be produced dur-
ing wastewater treatment [7,8,15,17,52,12]. It should be taken into
account, though, that six-ring lactones and the open acids usually
are transformed rapidly from one form into the other, thus it is hard
to demonstrate the real presence of on in exclusion of the other.
However, it seems, it has not been realised up to now, that there is
another isomer of HHCB-lactone eluting slightly after the usually
detected one (compare Figs. 3 and 4), It cannot be excluded, that
this compound, detected in an effluent is identical with those dis-
cussed as stemming from production impurities in [12]. However,
at that time neither chromatographic nor mass spectrometric data
were published.

2.5.2. Sludge
Generally sludges are analysed as activated sludge as originating
directly from the wastewater treatment or as dewatered stabilised
sludge as originating from anaerobic sludge stabilisation. Activated
sludge is the active part of the wastewater treatment, contain-
ing more than 95% water being thus liquid. Dewatered sludge is
the final product on most wastewater treatment plants containing
Fig. 4. Mass spectrometric characterization of two HHCB-lactones. (a) HHCB-
60–70% water and is thus semi-solid. Critical aspects in the analysis lactone pure standard. (b) HHCB-lactone main peak; RT 11.75 min. (c) HHCB-lactone
of sludge are: secondary peak; RT 12.00 min.
478 K. Bester / J. Chromatogr. A 1216 (2009) 470–480
(a) to make sure the analyte has been extracted completely. Usually
several methods are being compared during the method devel-
opment to make sure that at least the most effective method
is being used. Spike materials are hard to produce, as the basis
material has to be free from analyte (blank) and the analytes
have to be sorbed in the same way as in real sludge. This is very
hard to perform, as the exact sorption mechanism is generally
unknown. In the author’s laboratory a mixture of manure and
soil (50:50) has been used for some aspects of quality control
[7];
(b) to make sure that all compounds that might be deteriorating the
GC system are being removed prior to instrumental analysis.
Liquid (activated) sludges are often analysed to gain in-depth
knowledge on sludge treatment processes. They are usually han-
dled as water, thus either liquid–liquid extraction with n-hexane is
performed, sometimes salts (NaCl) are used to improve the phase
separation [29,50,54]. Critical part in this approach is to make sure
a good phase separation is achieved. Additionally it should be taken
into account, that some fragrances might not only be sorbed to par-
ticulate material but also be incorporated into larger particles such
as toilet papers, etc.
Accordingly the dewatered sludges are often analysed to gain
mass balances on wastewater treatment plants and to provide infor-
mation for sludge disposal. Dewatered sludge is normally handled
as solids: i.e., the respective samples are dried (e.g., lyophilised)
and extracted with Soxhlet [7] or accelerated solvent extraction
(ASE) [9,55]. Ethyl acetate does easily wet all relevant particles,
but other solvents have been used as well. An overview is given
in Table 2. It is essential to evaluate the respective methods against
losses during the drying processes as well as on completeness of
extraction. In the author’s laboratory several drying and extraction
combinations were assessed. Lyophillisaton overnight, followed by
Soxhlet extraction with ethyl acetate gave higher recovery rates
than a slurry extraction with excessive usage of solvents (no drying)
with toluene, or drying the samples with sodium sulfate followed
by Soxhlet extraction with ethyl acetate [7]. Faster method eval-
uation is possible with ASE, as temperature and pressure can be
modified quickly and the optimal conditions can be found within a
day or two. The reader should keep in mind that the harsher the ASE
extraction conditions the more the matrix is co-extracted and thus
more efforts have to be spent in clean-up of the respective extracts.
After extraction, co-extracted macromolecules as well as sul-
phur are normally removed by size exclusion chromatography (SEC)
[7,8,9,29]. Often a successive step such as a silica micro-column is
utilised to remove polar compounds and particles [7,8,29] (com-
pare Table 2). However, it should be taken into account that the
known metabolites of the nitro-musks, i.e., the anilines as well
as HHCB-lactone are more polar than their parent compounds.
They are usually not eluted from silica clean-up columns with n-
hexane/MTBE or n-hexane/dichloromethane (DCM) mixtures but
need solvents as polar as ethyl acetate to elute from these columns.
With the usual fractionation schemes they are thus collected in
different sub-fractions than the parent compounds. For other ana-
lytes the use of activated copper or silver has been utilised to
eliminate elemental sulphur from the samples. When analysing
nitro-aromates this is not possible as these compounds can be
reduced to the respective amines during this treatment. Other
destructive clean-ups (sulphuric acid, acid treated silica, etc.) may
lead to an oxidation of the polycyclic musks or other compounds.
To the author’s knowledge all these methods will work, if the
methods are developed thoroughly and are validated accordingly.
However, when developing a new method it might be considered,
that one of the crucial points is the wetability of the sample matrix.
This is mainly triggered by the water content, thus drying is impor-
tant, however in case of lyophillisation, drying too far means loosing
analytes. It might thus be wise to dry carefully, and use a solvent
that can mix with water to some extent to provide good wetting of
the sludge particles. However, it should not be too polar, to avoid co-
extraction of too many unwanted compounds. Using ethyl acetate
(or DCM) for the extraction and starting with a size exclusion clean-
up is probably a wise approach.
2.5.3. Sediments
Sediments are usually analysed as dried sediments. These sam-
ples can be Soxhlet extracted, e.g., using dichloromethane, with
a successive clean-up utilizing silica micro-columns [47,56]. An
overview is given in Table 2.
Microwave assisted solvent extraction has been utilized by
Rice and Mitra [57], while Kronimus et al. [58] used sequential
dispersion extraction with acetone followed by n-hexane for an
identification/quantification approach. In this study the extracted
samples were dried and treated with activated copper for sul-
phur removal. Thus, no nitro-musks but only HHCB, AHTN, and
4-oxoisophorone were reported. A silica based clean-up was used
for fractionation. Fromme et al. [56] used a combination of Soxh-
let extraction with ethyl acetate:cyclohexane 1:1 with a successive
size exclusion chromatography utilizing Biobeads SX-3 with cyclo-
hexane: ethyl acetate 98.5:100 as eluent. This step was followed
by a silica micro-column clean-up. Again ensuring good wetting of
the particles is essential during the extraction. Sulphur clean-up is
even more important than for the sludge samples, and all the issues
mentioned in the paragraph above apply for sediment samples as
well.
2.5.4. Biota
Most authors use some kind of excessive lipid extraction with
increased temperature and medium polar solvents (Soxhlet or ASE),
assuming all lipophilic compounds will be co-extracted with the
lipids. In this case it is at least crucial to demonstrate all lipids
indeed have been extracted. Again wetability and number of extrac-
tion cycles be it Soxhlet, ASE or other) are critical for the success of
extraction. In the literature usually only the time for, e.g., Soxhlet
extractions is documented, while of course, in reality the num-
bers of cycles and the geometry of the extraction cartridge matter
more. While the flow through a glass cartridge with a glass frit
is laminar from up to down, the flow in a glass fiber cartridge
also has a flow to the side. The biota samples need to be dried
for most procedures before extraction, however. As the musk com-
pounds are slightly more polar than organo-chlorines such as PCBs,
usage of very inpolar extractants alone such as pure n-hexane
is not recommended. Successive lipid removal from the extracts
by size exclusion chromatography is used. It is occasionally com-
bined with silica clean-up to remove residual lipids is usually
used for clean-up (an overview is given in Table 2). This step is
essential to prevent pollution of the injector and the column of
the GC system [46]. However, the reader should be aware, that
especially the polycyclic musks are rather large (bulky) molecules,
thus they elute in front of the usual pesticide fraction on the
SEC columns. Classical SEC-fractioning should thus be checked
for recovery before in depth method development is performed.
Fromme et al. [56] used Soxhlet, with ethyl acetate:cyclohexane
1:1 with a successive size exclusion chromatography utilizing
Biobeads SX-3 with cyclohexane:ethyl acetate 98.5:100 as elu-
ent. This step was followed by a silica micro-column clean-up
[56].
Gatermann et al. used a cold column extraction with successive
size exclusion chromatography and normal phase HPLC silica clean-
up [18]. These cold extractions need to be thoroughly tested against
hot and pressurised extractions on whether or not the lipid and
 K. Bester / J. Chromatogr. A 1216 (2009) 470–480
the analyte extraction is complete. Concerning clean-up treatment
with concentrated sulphuric acid, acid treated silica or florisil is not
appropriate for analysing polycyclic and macrocyclic musks, as they
will be oxidised under these conditions, the macrocyclic musk will
undergo hydrolysis as well as re-arrangement.
2.5.5. Air
Studies by Aschmann et al. [59] on atmospheric life time of fra-
grances were performed by extracting 100 mL air through a tenax
tube, which was then analysed by thermo-desorption into a DB
1701 GC column. Final analysis was performed by Flame Ionisation
Detection. In contrast to that, Peck and Hornbuckle performed high
volume (∼500 m3 ) air sampling above the Great Lakes on XAD-2,
which was successively extracted by acetone/n-hexane [26]. These
extracts were cleaned up with a silica column. As for all methods
to determine trace analytes from air, it is the crucial step to prove
that the analytes are extracted completely, and are not breaking
through the sampling devices during extraction. This is usually a
point that is neglected although it is by far not trivial. It makes
sense to include extracting high volumes of air with pre-spiked
sorbents in parallel to unspiked ones (sampling as standard addi-
tion). Using just a backup adsorbent is usually not convincing, as
the pressure is decreased in the second packing of adsorbent (air is
usually sucked through the device by vacuum) and thus the losses
would be even higher on the secondary sorption device than for
the primary one. Thus, breakthroughs (low recovery rates) would
remain undetected.
2.5.6. Commodities
Reiner and Kannan [42] demonstrated that the extraction from
liquid and solid materials can be performed by extraction of
0.1–0.5 g material with triplicate shaking extraction with 5 mL
n-hexane for 15 min with a successive centrifugation. D10 phenan-
threne was used as internal standard. Commodities usually contain
large quantities of diverse surfactants or solvents, such as alcohols.
These make phase separation difficult (in case of LLE). However, in
case of SPE, surfactants may change the solubility of the analyte by
micelle formation so much that the analyte is no longer retained in
the SPE material.
3. Conclusions
By now a rather settled literature on the methods on the analysis
of musks have been published. In case of surface waters, especially
marine waters the blank problems might be drastic. However, for
all environmental samples, sound methods have been published
that can be used for nitro-aromatic and polycyclic musks more or
less in routine work now. Challenges for the future include pro-
viding methods, which are able to determine macrocyclic musks
and other fragrances and especially the transformation products
at environmental levels. Additionally it needs to be explored how
enantioselective analysis can be made so rugged that it can be used
more often. Additionally it makes sense to explore to what extent
this might give new knowledge in environmental sciences as well
as in fragrance development.
It has been shown, that those compounds for which library
spectra were available in the commercial spectral libraries were
identified much earlier in environmental samples, than those com-
pounds were no spectra were documented in beforehand. While in
the past these spectra were submitted to the respective libraries for
all kinds of compounds at least in the moment these compounds
were becoming of commercial relevance, in the last decades this
process has become slower. It should be taken care of within the
REACh process, that mass spectral data are being included into
spectral databases, e.g., on basis of the existing NIST library or
479
a comparable one. It should also be mentioned that it would be
wise to verify the predictions within the registration process by
analysing samples of specimen banks for quality control as has
already been demonstrated [6].
Acknowledgements
The author is indebted to various friends in the fragrance indus-
try for the willingness to provide knowledge on the current state
of fragrance development and the willing gifts of standard com-
pounds. Also I am indebted to various friends in environmental
research who were willing to share their expertise and ideas
through various meetings and collaborations. I am also grateful to
the support of LECO, Mönchengladbach, Germany who provided
me with several days of time on a comprehensive 2D GC–MS instru-
ment.
References
[1] M. Gautschi, J.A. Bajgrowicz, P. Kraft, Chimia 55 (2001) 379.
[2] T. Yamagashi, T. Miyazaki, S. Horii, S. Kaneko, Bull. Environ. Contam. Toxicol. 26
(1981) 656.
[3] H.D. Eschke, J. Traud, H.J. Dibowski, UWSF-Z. Umweltchem. Ökotox. 6 (1994)
183.
[4] H.D. Eschke, J. Traud, H.J. Dibowski, Vom Wasser 83 (1994) 373.
[5] P.S. Spencer, A.B. Sterman, D.S. Horoupian, M.M. Foulds, Science 204 (1979)
633.
[6] H. Rüdel, W. Böhmer, C. Schröter-Kermani, J. Environ. Monit. 8 (2006) 812.
[7] K. Bester, Chemosphere 57 (2004) 863.
[8] K. Bester, J. Environ. Monit. 7 (2005) 43.
[9] K. Bester, Personal Care Compounds in the Environment, VCH-Wiley, Wein-
heim, 2007.
[10] E. Artola-Garicano, I. Borkent, J.L.M. Hermens, W.H.J. Vaes, Environ. Sci. Technol.
37 (2003) 3111.
[11] E. Artola-Garicano, J.L.M. Hermens, W.H.J. Vaes, Water Res. 37 (2003)
4377.
[12] G.G. Rimkus (Ed.), Synthetic Musk Fragrances in the Environment, The Hand-
book of Environmental Chemistry, Anthropogenic Compounds, vol. 3, Springer,
Berlin, 2004.
[13] L. Dsikowitzky, J. Schwarzbauer, R. Littke, Org. Geochem. 33 (2002) 1747.
[14] A.M. Peck, K.C. Hornbuckle, Environ. Sci. Technol. 38 (2004) 367.
[15] J.A. Andresen, D. Muir, D. Ueno, C. Darling, N. Theobald, K. Bester, Environ.
Toxicol. Chem. 26 (2007) 1081.
[16] K. Bester, H. Hühnerfuss, W. Lange, G.G. Rimkus, N. Theobald, Water Res. 32
(1998) 1857.
[17] S. Franke, C. Meyer, N. Heinzel, R. Gatermann, H. Hühnerfuss, G. Rimkus, W.A.
Konig, W. Francke, Chirality 11 (1999) 795.
[18] R. Gatermann, S. Biselli, H. Hühnerfuss, G.G. Rimkus, M. Hecker, L. Karbe, Arch.
Environ. Contam. Toxicol. 42 (2002) 437.
[19] R. Gatermann, S. Biselli, H. Hühnerfuss, G.G. Rimkus, S. Franke, M. Hecker, R.
Kallenborn, L. Karbe, W.A. Konig, Arch. Environ. Contam. Toxicol. 42 (2002)
447.
[20] H. Fromme, T. Otto, K. Pilz, Toxicol. Lett. 111 (1999) 37.
[22] G. Rimkus, M. Wolf, Chemosphere 33 (1996) 2033.
[23] K. Kannan, J.L. Reiner, S.H. Yun, E.E. Perrotta, L. Tao, B. Johnson-Restrepo, B.D.
Rodan, Chemosphere 61 (2005) 693.
[24] H.D. Eschke, H.J. Dibowski, J. Traud, Deutsche Lebensmittel Rundschau 91
(1995) 375.
[25] R. Kallenborn, R. Gatermann, S. Planting, G.G. Rimkus, M. Lund, M. Schlabach,
I.C. Burkow, J. Chromatogr. A 846 (1999) 295.
[26] A.M. Peck, K.C. Hornbuckle, Atmos. Environ. 40 (2006) 6101.
[27] G.G. Rimkus, R. Gatermann, H. Hühnerfuss, Toxicol. Lett. 111 (1999) 5.
[28] R. Gatermann, H. Hühnerfuss, G. Rimkus, A. Attar, A. Kettrup, Chemosphere 36
(1998) 2535.
[29] J.D. Berset, P. Bigler, D. Herren, Anal. Chem. 72 (2000) 2124.
[30] J.W. Tas, F. Balk, R.A. Ford, E.J. van de Plassche, Chemosphere 35 (1997) 2973.
[31] P. Kraft, G. Frater, Chirality 13 (2001) 388.
[32] European Commission, Official Journal, L 396 (2006) 1.
[33] F. Balk, R.A. Ford, Toxicol. Lett. 111 (1999) 57.
[34] G. Carlsson, L. Norrgren, Arch. Environ. Contam. Toxicol. 46 (2004) 102.
[35] N. Bitsch, C. Dudas, W. Korner, K. Failing, S. Biselli, G. Rimkus, H. Brunn, Arch.
Environ. Contam. Toxicol. 43 (2002) 257.
[36] R.H.M.M. Schreurs, J. Legler, E. Artola-Garicano, T.L. Sinnige, P.H. Lanser, W.
Seinen, B. van der Burg, Environ. Sci. Technol. 38 (2004) 997.
[37] W. Seinen, J.C. Lemmen, R.H.H. Pieters, E.M.J. Verbruggen, B. van der Burg,
Toxicol. Lett. 111 (1999) 161.
[38] K. Bester, J. Klasmeier, T. Kupper, Chemosphere 71 (2008) 2003.
[39] K. Bester, N. Hüffmeyer, E. Schaub, J. Klasmeier Chemosphere 73 (2008) 1366.
[40] N.T. Litz, W. Böhmer, J. Müller, J. Soil Sediment 7 (2007) 36.
480 K. Bester / J. Chromatogr. A 1216 (2009) 470–480
[41] J. Müller, W. Böhmer, N.T. Litz, J. Soil Sediment 6 (2006) 231.
[42] J.L. Reiner, K. Kannan, Chemosphere 62 (2006) 867.
[43] I.J. Buerge, H.R. Buser, M.D. Müller, T. Poiger, Environ. Sci. Technol. 37 (2003)
5636.
[44] S.L. Simonich, W.M. Begley, G. Debaere, W.S. Eckhoff, Environ. Sci. Technol. 34
(2000) 959.
[45] S.L. Simonich, T.W. Federle, W.S. Eckhoff, A. Rottiers, S. Webb, D. Sabaliunas, W.
De Wolf, Environ. Sci. Technol. 36 (2002) 2839.
[46] G.G. Rimkus, W. Butte, H.J. Geyer, Chemosphere 35 (1997) 1497.
[47] A.M. Peck, E.K. Linebaugh, K.C. Hornbuckle, Environ. Sci. Technol. 40 (2006)
5629.
[48] W. Vetter, K. Bester, Gas Chromatographic Enantioseparation of Chiral
Pollutants—Techniques and Results, Bush and Bush: Chiral Analysis, Elsevier,
2006.
[49] L.I. Osemwengie, S.L. Gerstenberger, J. Environ. Monit. 6 (2004) 533.
[50] J.L. Reiner, J.D. Berset, K. Kannan, Arch. Environ. Contam. Toxicol. 52 (2007)
(2007) 451.
[51] S. Weigel, K. Bester, H. Huehnerfuss, J. Chromatogr. A 912 (2001) 151.
[52] S. Valdersnes, R. Kallenborn, L.K. Sydnes, Intern. J. Environ. Anal. Chem. 86
(2006) 461.
[53] R. Kallenborn, R. Gatermann, T. Nygard, J. Knutzen, M. Schlabach, Fresen. Envi-
ron. Bull. 10 (2001) 832.
[54] T. Kupper, J.D. Berset, R. Etter-Holzer, R. Furrer, J. Tarradellas, Chemosphere 54
(2004) 1111.
[55] A.M. Difrancesco, P.C. Chiu, L.J. Standley, H.E. Allen, D.T. Salvito, Environ. Sci.
Technol. 38 (2004) 194.
[56] H. Fromme, T. Otto, K. Pilz, F. Neugebauer, Chemosphere 39 (1999) 1723.
[57] S.L. Rice, S. Mitra, Anal. Chim. Acta 589 (2007) 125.
[58] A. Kronimus, J. Schwarzbauer, L. Dsikowitzky, S.L. Heim, Water Res. 38 (2004)
3473.
[59] S.M. Aschmann, J. Arey, R. Atkinson, S.L. Simonich, Environ. Sci. Technol. 35
(2001) 3595.
[60] J.L. Stevens, G.L. Northcott, G.A. Stern, G.T. Tomy, K.C. Jones, Environ. Sci. Technol.
37 (2003) 462.
[61] H. Nakata, H. Sasaki, A. Takemura, M. Yoshioka, S. Tanabe, K. Kannan, Environ.
Sci. Technol. 41 (2007) 2216.