Talanta 58 (2002) 899
/907

www.elsevier.com/locate/talanta

Determination of As(III), As(V), monomethylarsonic acid,

dimethylarsinic acid and arsenobetaine by HPLC
ICP
MS: / /

analysis of reference materials, fish tissues and urine

Kazimierz Wrobel 1, Katarzyna Wrobel 2, Bryan Parker, Sasi
S. Kannamkumarath, Joseph A. Caruso *
Department of Chemistry, University of Cincinnati, Cincinnati, OH 45221-0172, USA

Received 17 May 2002; received in revised form 2 August 2002; accepted 5 August 2002

Abstract

In this study, a rapid and sensitive high performance liquid chromatography
/inductively coupled plasma
/mass
spectrometry (HPLC
/ICP
/MS) determination of primary As species in fish tissues and urine is reported. The
separation was achieved on an Altima C18 column with a mobile phase containing citric acid and hexanesulfonic acid
(pH 4.5). As(V), monomethylarsonic acid (MMA), As(III), dimethylarsinic acid (DMA) and arsenobetaine (AsB) were
separated in less than 4 min with retention times of 83, 99, 130, 166 and 208 s, respectively. This separation of five
species in less than 4 min should be attractive to those interested in As speciation. The quantification limits were 44, 56,
94, 64, 66 ng l 1 and the relative standard deviations (R.S.D.) for day-to-day injections of As at 2 mg l 1 were 2.0, 3.1,
2.4, 3.8 and 4.0%. The procedure was tested using two reference materials (DORM-2 dogfish muscle tissue, NIST SRM
2670 Freeze-dried Urine, normal level) and then applied to real-world samples. The results obtained demonstrate the
suitability of the procedure for screening and quantification at physiological levels of primary As species in biological
samples.
# 2002 Elsevier Science B.V. All rights reserved.

Keywords: Arsenic; Speciation; Biological samples; HPLC; ICP
/MS

1. Introduction

Arsenic is one of the most studied elements in

* Corresponding author. Fax: /1-513-556-9239 speciation analysis. Depending on the sample type,
E-mail address: joseph.caruso@uc.edu (J.A. Caruso). it occurs in a variety of chemical forms, including
1
At the University of Cincinnati while on the leave from two oxidation states, different methylated species,
Instituto de Investigaciones Cientificas, Universidad de arseno-sugars, arsenolipids, etc. [1
/4]. As the
Guanajuato, L. de Retana No. 5, 36000 Guanajuato, Mexico.
2
At the University of Cincinnati while on the leave from
toxicity of arsenic is species dependent [5], specia-
Instituto de Investigaciones Cientificas, Universidad de tion information is mandatory for wide-ranging
Guanajuato, L. de Retana No. 5, 36000 Guanajuato, Mexico. studies including, environmental pollution, ele-
0039-9140/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 3 9 - 9 1 4 0 ( 0 2 ) 0 0 4 0 4 - 6
900 K. Wrobel et al. / Talanta 58 (2002) 899
/907
ment mobility and biotransformations as well as
on its metabolism in living organisms. A number
of analytical methods, capable of separating and
quantifying several arsenic species have been
reported, all of them based on high performance
liquid chromatography (HPLC) coupling to ele-
ment specific detectors [6
/9]. However, in many
applications there is no need for high selectivity,
because different sample types usually contain a
limited number of arsenic species at the concentra-
tion levels enabling their detection. Thus, arseno-
betaine (AsB) is the main species found in fish
tissues with significantly lower contribution of
dimethylarsinic acid (DMA) and trace amounts
of inorganic arsenic (As(V)), arsenocholine (AsC)
and/or tetramethylarsonium (TETRA) [10
/13]. In
human urine, the two metabolites (monomethy-
larsonic acid (MMA), DMA) together with traces
of inorganic forms were commonly observed
[9,14
/16] and, additionally the presence of AsB
was reported after fish consumption [17
/19]. The
NIST reference material SRM 2670 (freezed-dried
Urine, normal level) have often been analyzed for
arsenic species and, as reviewed by Wei et al. [20]
only these five species (As(III), As(V), MMA,
DMA, AsB) were reported.
The vast majority of arsenic speciation studies
with applications to the above mentioned materi-
als were focused on the separation/quantification
of As(III), As(V), MMA, DMA and AsB. The
separations were achieved using HPLC coupled to
the element selective detectors such as inductively
coupled plasma
/mass spectrometry (ICP
/MS),
hydride generation atomic absorption (HG-AAS)
or atomic fluorescence (HG-AFS) spectrometries
[9,14,21]. Depending on the pH conditions, these
species can be cationic, anionic or uncharged
(pKa1, respectively, 9.3, 2.3, 2.6, 6.2 and 2.2 [22])
that makes possible the separation by anion-
exchange, ion-pairing and, to a lesser extent, by
cation-exchange chromatographies. In the applica-
tions of anion exchange HPLC, good resolution
was achieved at pH ranges from 5 to 11. The most
common elution order is AsB, As(III), DMA,
MMA, As(V) but this is sometimes different,
depending on the pH, gradient conditions, etc.
The time analysis reported varied from about 7 to
30 min [9,10,16,22
/26]. The cation-exchange
HPLC, requiring lower pH values (below 4) was
preferred for the analysis of mixtures/samples
containing AsC, TETRA and/or trimethylarsine-
oxide (TMAO) [12,14,21,25]. The known advan-
tage of ion-pairing chromatography is its potential
to separate charged and neutral solutes; moreover,
hydrophobicity of the ion-pairs can be controlled
by careful selection of ion-pairing reagent thereby,
enhancing further the selectivity. In studies on
arsenic speciation, tetrabutylammonium ion (pH
6
/11) and a number of alkylsulfonates (pH 2.7
/
4.5) were used in reversed phase columns offering
efficient species separation within 6
/20 min
[7,9,12,27
/30]. A dianionic ion-pairing reagent
(BDSA-benzene-1,2-disulfonic acid) was also
used; the separation was achieved on anion
exchange columns in a gradient of nitric acid
(pH from 3.4 to 1.8 or to 1.3) [7,31].
The previous studies carried out in our labora-
tory were focused on the evaluation of different
pretreatment procedures suitable for arsenic spe-
ciation in biological materials (fish and lobster
tissues) [10,32,33]. The analytical system used was
anion exchange chromatography coupled with
ICP
/MS detection that allowed the detection
limits of few mg kg1 but with a run time of
about 27 min. The goal of this work was to
develop a rapid (less than 5 min) HPLC
/ICP
/
MS procedure for screening and quantifying five
arsenic primary species in biological materials. The
sulfonate containing ion interaction reagent was
used and the separation was achieved at low pH.
Further acidification of the sample prior to its
injection onto the column was proposed in order
to obtain a resolution of As(III), As(V), MMA,
DMA, AsB at short times. Both, the reference
materials and the real-world samples were ana-
lyzed.
2. Experimental
2.1. Apparatus
The high performance liquid chromatographic
system (HPLC) used was an Agilent (Agilent
Technologies, Palo Alto, CA) series 1100 equipped
with an autosampler, a diode array detector and
 K. Wrobel et al. / Talanta 58 (2002) 899
/907
Chemstation. The chromatographic column was:
C18 Altima (150 /4.6 mm, 5 mm).
Agilent 7500s inductively coupled plasma-mass
spectrometer (ICP
/MS) connected with con-
centric nebulizer and Scott type double pass spray
chamber (cooled to 2 8C) was used for arsenic
specific detection. A nickel cone and a platinum
skimmer were used. The solution eluting from the
column was introduced on-line to ICP
/MS.
The chromatographic and instrumental operat-
ing conditions are given in Table 1.
2.2. Reagents and samples
Doubly de-ionized (18.2 MV cm) water, pre-
pared by passing de-ionized water through a
NanoPure treatment system (Barnstead, Boston,
MA, USA), was used. Analytical reagent-grade
chemicals, HPLC grade chloroform and methanol
(Fisher Scientific, Pittsburgh, PA, USA) were
used.
Inorganic arsenic standards, 1000 mg ml 1
As(III) and 1000 mg l 1 As(V) were kindly
provided by the US Food and Drug Administra-
tion. AsB was acquired from the Department of
Chemistry, University of British Columbia (Van-
couver, Canada), DMA and disodium methyl
arsenate (MMA) were purchased from Chem
Service (West Chester, PA, USA). All stock
standards were made based on arsenic and their
Table 1
Instrumental operation conditions for HPLC
/ICP
/MS system
HPLC parameters
Column C18 Altima, 150/4.6 mm, 5 mm
Mobile phase 5 mmol l 1 citric acid/NaOH, 5 mmol l 1
hexanesulfonic acid, pH 4.5
Flow 0.9 ml min 1
Injected volume 20 ml
ICP
/MS parameters
Forward power 1300 W
Nebulizer gas flow
Carrier gas 0.75 l min1
Make-up gas 0.40 l min1
Sample introduc- Meinhard nebulizer
tion
Channels moni- 75, 77, 78
tored
901
concentrations were validated using NIST 1643c
[10].
A mobile phase was 5 mmol l 1 hexanesulfonic
acid in 5 mmol l1 citric acid (pH 4.5 adjusted
with sodium hydroxide), prepared from Fisher and
Sigma reagents, respectively.
Solutions of following Sigma reagents were
used: phosphoric acid, sodium hydroxide and
sodium chloride.
DORM-2 (dogfish muscle tissue) was a refer-
ence material purchased from National Research
Council (NRCC, Ottawa, Ont., Canada). Fish
samples analyzed were the freeze-dried salmon,
white ocean fish and shark tissues [10]. NIST SRM
2670 Freeze-dried Urine (normal level) with a
recommended total arsenic concentration in the
reconstituted solution 60 mg l 1 was used as the
reference sample (NIST, Gaithersburg, MD,
USA). Two urine samples from volunteers were
analyzed.
2.3. Procedures
2.3.1. Urine samples
The reconstituted urine sample (20 ml deionized
water) was stored at 4 8C and used within a week.
The urine samples from volunteers were filtered
(0.45 mm) and 2
/5-fold diluted (10 mmol l1
phosphoric acid) prior to injection.
2.3.2. Extraction of arsenic species from fish
samples
The method employed by Beauchemin et al. [34]
was adopted. To do so, 0.5 g of the freeze-dried
fish tissue (DORM-1, DORM-2, salmon, white
ocean fish and shark) were weighed in the plastic
tubes (50 ml), 10 ml of methanol and 10 ml of
chloroform were added and the mixtures were
placed in the ultrasonic bath (30 min). After
centrifugation (10 min, 5000 rpm), the
chloroform
/methanol fractions were collected,
the procedure was repeated and, for each sample,
the extracts were combined in the separatory
funnels (125 ml). Then, 20 ml of water and 20 ml
of chloroform were added, the mixtures were
shaken vigorously, the water
/methanol fractions
were collected and evaporated in the nitrogen
stream. The residues were dissolved in 10 mmol
902 K. Wrobel et al. / Talanta 58 (2002) 899
/907

l 1 phosphoric acid and introduced to the chro-

matographic system.

3. Results and discussion

3.1. Method development

In the separation of arsenic species both the

electrical charge and possible hydrophobic char-
acter of the analytes have been considered [9].
However, the exact mechanisms of chromato-
graphic separation are not always clear, because
the assessment of actual charge distribution in
complex solutions is not straightforward (the
dissociation equilibria of arsenic species are af-
fected by such parameters as ionic strength, the
presence of ion-pairing reagents, etc). Generally, in
acidic media the contribution of uncharged As(V),
DMA, MMA molecules becomes higher; AsB can
be present as neutral zwitterion or cationic species
and As(III) is uncharged or cationic [35]. Although
anion exchange or the use of cationic ion-pairing
reagents at pH values from around 6 to 11 have
been preferred for the separation of these five
species, the alkylsulfonate ion-pairing reagents
requiring lower pH were also successfully applied
with the advantage of short time analysis [30]. In
the preliminary experiments carried out on a Fig. 1. HPLC
/ICP
/MS chromatograms of arsenic species’
reversed phase column with 5 mmol l 1 citrate standards (8 mgAs l 1) in the presence and in the absence of
buffer (pH 4.5) as the mobile phase, the three acid modifier (instrumental conditions given in Table 1). The
species (As(III), As(V) and MMA) eluted in form elution order: As(V), MMA, As(III), DMA and AsB. (a)
Without phosphoric acid; (b) 10 mmol l 1 phosphoric acid;
of one, broad peak, close to the void volume. The (c) 80 mmol l 1 phosphoric acid.
other two species (DMA and AsB) were separated
with retention times around 132 and 186 s. When
uncharged DMA with the stationary phase. No
hexanesulfonic acid was added to the mobile
phase, the elution of AsB was delayed and a broad improvement in MMA and As(III) separation was
chromatographic peak split: As(V) eluted in a first achieved by lowering the pH of the mobile phase
peak and As(III) coeluted with MMA. With down to 3.0, while the retention time of DMA and
increasing concentration of ion interaction re- AsB increased. In a similar chromatographic
agent, the peak corresponding to As(III) and experiment, but with the higher concentration of
MMA co-elution became distorted, which indi- hexanesulfonic acid, at lower pH (pH 3.5) and in
cated stronger retention of As(III). However, the presence of 0.1% methanol, Le et al. [30]
separation of these two species was not achieved reported separation of five species with generally
(Fig. 1a). Similar retention of DMA was observed higher retention times (As(V) tret /144 s, As(III)
in the presence and in the absence of hexanesul- tret /186 s, MMA tret /228 s, DMA tret /294 s
fonic acid suggesting that the retention mechanism and AsB tret /360 s). In order to keep the total
was primarily due to hydrophobic interactions of separation time as short as possible, in this study
 K. Wrobel et al. / Talanta 58 (2002) 899
/907
pH 4.5 and 5 mmol l 1 hexanesulfonic acid were
used. On the other hand, a pH gradient elution
(nitric acid pH 3.4
/1.3) was reported for control-
ling the charge of arsenic species during separation
[7]. Thus, it was assumed that the resolution of
MMA and As(III) could be achieved by lowering
the pH value during their elution. For this
purpose, the acidification of the solution prior to
its introduction onto the column was suggested.
Phosphoric acid was used as a common compo-
nent of the mobile phases for arsenic speciation
(no interaction with arsenic species, nor with
mobile phase, not an oxidizing agent). In Fig. 1,
typical chromatograms of five arsenic species are
presented that were obtained in the absence and in
the presence of acid modifier. It can be observed
that in the acidified solutions, the elution of
As(III) was delayed (tret /130 s) with respect to
MMA (tret /99 s). Moreover, at phosphoric acid
of 10
/20 mmol l 1, the elution of other species
was not modified significantly (retention times for
As(V), DMA and AsB, respectively, 83, 166, 208
s), while in the presence of higher acid concentra-
tions DMA and AsB were retained more strongly
and the asymmetric As(V) peak appeared. This
indicates that phosphoric acid at 10
/20 mmol l 1
had a capacity for lowering pH during the elution
of first species (enhanced interaction As(III)
/
hexanesulfonic acid), but was further neutralized
in the column causing only minimum modification
in the elution of DMA and AsB. Thus, the
important advantage of sample acidification ver-
sus lowering pH of the mobile phase is the short
time analysis (five species resolved in 240 s)
assuring lower peak broadening and better quan-
tification limits. The calibration range for arsenic
species was from 0.5 up to 200.0 mgAs l 1 (Fig. 2).
In Table 2, the analytical figures of merit are
presented showing that the proposed procedure
enables fast quantification of five arsenic species at
ultra trace levels.
Before application to the analysis of real sam-
ples, possible chloride interferences were investi-
gated. To do so, a chromatogram of sodium
chloride solution was obtained with ICP
/MS
detection at two channels: m /z 35 for chloride
and 75 to monitor possible 40Ar35Cl  signal. A co-
elution of chloride with As(V) was observed (m /z
903
Fig. 2. Typical chromatograms of calibration solutions: (---)
blank, ( */) 0.5 mg l 1 and (


) 2 mg l 1 of arsenic in each
species (instrumental conditions detailed in Table 1).
35) and, for 100 mmol l 1 sodium chloride, the
signal due to polyatomic interferences (m /z 75)
corresponded to an As(V) signal at the concentra-
tion 0.8 mg l 1. This interference was corrected
using the interference equation based on the
counts monitored at three channels (m /z 75, 77
and 78) [36]. On the other hand, the presence of
salt may affect the chromatographic separation of
arsenic species. The retention times for five species
as well as the resolution of As(V) and MMA were
evaluated in the absence and in the presence of
sodium chloride (up to 200 mmol l 1 */approx-
imate level in urine). The obtained results are
presented in Table 3, where it can be observed that
the salt affected only the elution of As(V). It
should be stressed that the eluted peak became
asymmetric. It can also be observed that a good
resolution of As(V) and MMA was achieved in the
presence of sodium chloride up to a concentration
50 mmol l 1.
3.2. Analysis of fish tissues
The proposed procedure was tested using the
reference material DORM-2 (dogfish muscle tis-
sue). The assignment of chromatographic peaks
was accomplished by matching the retention times
with those of the commercial standards and by
spiking experiments. The extraction of arsenic
species with a chloroform
/methanol mixture was
adopted in order to assure complete extraction of
904 K. Wrobel et al. / Talanta 58 (2002) 899
/907

Table 2
Analytical figures of merit for the proposed speciation procedure

Parameter As(V) MMA As(III) DMA AsB

tret9/S.D. (s) 83.09/1.2 99.09/0.6 130.09/0.6 166.09/1.2 2089/6

r2 0.9995 0.9999 0.9999 0.9999 0.9990
s 14308.4 14186.2 13928.8 14066.8 14536.0
S.D.s 137 35.0 15.6 47.5 174
QL (ng l 1) 44 56 94 64 66
R.S.D. (%) (2.0 mg l 1) 2.0 3.1 2.4 3.8 4.0
R.S.D. (%) (100 mg l 1) 0.7 2.3 2.1 2.0 3.8

tret, Retention time; column void time, tv 82.0 s; r , regression coefficient; s , slope of the linear regression calibration (sensitivity, (cps
s)/(mg l 1)); S.D.s, for the calibration slope; QL, quantitation limit based on 6/S.D. of the noise level; R.S.D., between-days
precision: R.S.D. for five replicates of standard solution.

AsB [32]. The mean result for this species was
16.19/0.6 mg g1 As as dry mass (three replicates),
in agreement with the certified value (16.19/1.1 mg
g1). The results obtained for other species are
given in Table 4 together with the values reported
previously. It can be observed that for DMA our
results agreed with those obtained after water
/
methanol extraction and using ion-exchange chro-
matography [11] or ion-pairing reagent in an anion
exchange column [7]. The recently published data,
however, suggest higher DMA content [31]. The
levels of As(V) observed in DORM-2 were lower
with respect to the primary species AsB as well as
with respect to DMA and the obtained results
were generally similar to those reported by other
authors [7,11,31]. The procedure was then applied
to analysis of arsenic speciation in the samples of
tissue from different fish. In Fig. 3 the chromato-
grams obtained for shark, white ocean fish and
salmon tissues are presented and the quantitative
results given in Table 4. AsB was the main species
in each of the samples, the levels of DMA were
much lower, while As(V) and MMA were detected
only in the shark tissue. In the study by McKier-
nan et al. [10], these same samples were analyzed
showing similar relative arsenic distribution
among different species. Finally it should be
mentioned that in the analysis of real samples,
m /z 35 was monitored for chloride. In the insert
on Fig. 3a, a co-elution of chloride and As(V) as
well as the minimum contribution of polyatomic
interference for As(V) can be observed. For the
quantitative approach the three channels (m /z 75,
77, 78) required for the interference equation [34]
were used.
3.3. Analysis of urine
The reference material most frequently used for
arsenic speciation in urine has been NIST SRM

Table 3
Effect of sodium chloride on the chromatographic separation of arsenic species (the column void time at 0.9 ml min 1 flow rate was
82.0 s)

NaCl (mmol l 1) Retention times (mins) Resolution of As(V) and MMA

As(V) MMA As(III) DMA AsB

0 83 99 130 166 208 0.27

25 84 99 130 166 207 0.24
50 88 100 129 166 206 0.20
100 91 101 129 166 207 0.17
200 97 102 130 167 210 0.14

Resolution R /2 [tret(MMA)
/tret(As(V))]/[wH(MMA)/wH(As(V))], where wH is the peak width at half height.
 K. Wrobel et al. / Talanta 58 (2002) 899
/907 905

Table 4
Quantitative results obtained in the analysis of reference materials, fish tissue and urine samples (for fish samples results in mg g 1 As
dry mass, for urine mg l 1 As in undiluted sample)

Sample As(V) MMA As(III) DMA AsB

DORM-2 (this work) 0.059/0.01 nf nf 0.299/0.02 16.19/0.6

DORM-2 [7] 0.4 nf 0.1 0.3 13.5
DORM-2 [11] B/0.03 B/0.03 B/0.03 0.289/0.01 16.09/0.7
DORM-2 [31] 0.059/0.02 0.149/0.02 0.059/0.01 0.499/0.03 16.19/0.7
Shark 0.019/0.01 0.039/0.01 nf 0.099/0.02 12.89/0.6
White ocean fish nf nf nf 0.049/0.01 6.169/0.38
Salmon nf nf nf 0.059/0.01 1.739/0.12
SRM 2670 (this work) 1.59/0.2 9.69/0.4 nf 47.29/1.4 12.49/1.8
SRM 2670 [20] 1.39/0.2 9.89/0.3 nf 48.29/2.3 17.89/1.1
Urine, volunteer 1 0.39/0.1 nf nf 1.59/0.2 1.79/0.2

As reviewed by Wei et al. [20], the results reported

Fig. 3. Typical HPLC
/ICP
/MS chromatograms of the ex-
tracts from the fish tissues: (a) salmon: (


) m /z 35 and ( */)
m /z 75 (b) (


) shark and ( */) white ocean fish, both at m /z 75.
2670 freezed-dried urine (normal level) with re-
commended total arsenic concentration 60 mg l1.
in the speciation literature ranged from 60 up to
109 mg l1 and these discrepancies were ascribed
to different type of interferences occurring both in
the separation and detection steps. In the cited
work, the arsenic species were separated on an
anion exchange column and the authors addressed
the problem of chloride interferences and of poor
resolution of early eluted species (AsB, As(III) and
DMA). In Fig. 4a, a chromatogram of reference
urine is presented that was obtained using the
experimental conditions listed in Table 1 (the two
plots correspond to the sample 5-fold diluted with
10 mmol l 1 phosphoric acid and the acidified,
spiked sample). In Table 4, the quantitative results
of arsenic speciation are presented and compared
with the results reported by Wei et al. The total As
concentration was closer to the recommended
value in this work (70.7 vs. 77.1 mg l 1 [20]) and
a good agreement between the results obtained in
the two studies can be observed for As(V), MMA
and DMA. The AsB concentration found in this
work was lower with respect to the values obtained
while using anion exchange chromatography [20].
However, AsB is not retained in anion exchange
columns and the signal seems to be more suscep-
tible to possible interferences as compared with
ion-pairing chromatography used in this work. In
Fig. 4b and in Table 4 the results of arsenic
speciation in the urine of volunteers are presented
(quantitative results evaluated by the method of
906 K. Wrobel et al. / Talanta 58 (2002) 899
/907
Fig. 4. Typical HPLC
/ICP
/MS chromatograms of urine
samples: (a) NIST SRM 2670 five fold diluted with phosphoric
acid 10 mmol l 1 ( */) and after spiking with standards of
arsenic species (


). The concentration of arsenic standards
spikes in the solution injected to the column corresponded to
2.0 mg As l 1. (b) Urine sample from volunteer 1 (twice diluted
and acidified) with the ICP
/MS detection at m /z 35 (


) and
75 ( */).
standard addition). Due to low physiological levels
of arsenic in urine, the real samples were only
diluted 1:2. As can be observed in Fig. 4b and c, a
large chloride peak co-eluted with As(V), but the
contribution of 40Ar35Cl  interference was mini-
mum (correction equation was always applied).
4. Conclusions
The separation of five arsenic species important
in the analysis of fish and urine samples was
carried out using HPLC
/ICP
/MS with a mobile
phase containing 5 mmol l 1 hexanesulfonic acid
and 5 mmol l1 citrate buffer (pH 4.5). It was
demonstrated that the acidification of sample
prior to its injection onto the column could be
used to modify the retention of early eluting
species (As(III)). The advantage of sample acid-
ification versus lowering pH of the mobile phase is
the short time analysis (five species resolved in 240
s) that results in lower peak broadening and better
quantification limits. Possible chloride interfer-
ences were investigated and the conditions were
established to correct the signal for As(V) (dilution
of urine sample and the use of interference
equation). The analysis of two reference materials
(DORM-2 and SRM 2670) showed similar pattern
(profile) of arsenic species in these two samples
with respect to the data presented by other
authors. In conclusion, the procedure is suitable
for rapid HPLC determination of As(V), MMA,
As(III), DMA and AsB at low trace levels in real-
world biological samples such as fish tissue and
urine.
Acknowledgements
The partial support from NIEHS, grant
#ES04908 is gratefully acknowledged. K. Wrobel
and K. Wrobel sabbatical research stays were
supported by the University of Guanajuato (Mex-
ico), by CONACyT (Mexico) and the University
of Cincinnati (USA).
References
[1] R. Cornelis, F. Borguet, J. De Kimpe, Anal. Chim. Acta
283 (1993) 183.
[2] A. Chatterjee, Y. Shibata, J. Yoshinaga, M. Morita, Anal.
Chem. 72 (2000) 4402.
[3] Z. Mester, J. Pawliszyn, J. Chromatogr. A 873 (2000) 129.
[4] S.A. Pergantis, S. Wangkarn, K.A. Francesconi, J.E.
Thomas-Oates, Anal. Chem. 72 (2000) 357.
[5] H. Yamauchi, B.A. Fowler, in: J.O. Nriagu (Ed.), Arsenic
in the Environment. Part II: Human Health and Ecosys-
tem Effects, Wiley, New York, 1994, pp. 35
/43.
[6] Z. Slejkovec, J.T. van Elteren, A.R. Byrne, Talanta 49
(1999) 619.
[7] S. Londesborough, J. Mattusch, R. Wennrich, Fresenius J.
Anal. Chem. 363 (1999) 577.
[8] K.A. Francesconi, J.S. Edmonds, Rapid Commun. Mass
Spectrom. 15 (2001) 1641.
 K. Wrobel et al. / Talanta 58 (2002) 899
/907
[9] T. Guerin, A. Astruc, M. Astruc, Talanta 50 (1999) 1.
[10] J.W. McKiernan, J.T. Creed, C.A. Brockhoff, J.A. Car-
uso, R.M. Lorenzana, J. Anal. Atom. Spectrom. 14 (1999)
607.
[11] W. Goessler, D. Kuehnelt, C. Schlagenhaufen, Z. Slejko-
vec, K.J. Igorlic, J. Anal. Atom. Spectrom. 13 (1998) 183.
[12] J.J. Corr, J. Anal. Atom. Spectrom. 12 (1997) 537.
[13] D. Velez, N. Ybanez, R. Montoro, J. Anal. Atom.
Spectrom. 12 (1997) 91.
[14] M. Burguera, J.L. Burguera, Talanta 44 (1997) 1581.
[15] B. Do, P. Alet, D. Pradeau, J. Poupon, M. Guilley-Gaillot,
F. Guyon, J. Chromatogr. B: Biomed. Sci. Appl. 740
(2000) 179.
[16] G. Samanta, U.K. Chowdhury, B.K. Mandal, D. Chakra-
borti, N.C. Sekaran, H. Tokunaga, M. Ando, Microchem.
J. 65 (2000) 113.
[17] J. Lintschinger, P. Schramel, A. Hatalak-Rauscher, I.
Wendler, B. Michalke, Fresenius J. Anal. Chem. 362
(1998) 313
/318.
[18] R. Ritsema, L. Dukan, T.R. Navarro, W. van Leeuwen, N.
Oliveira, P. Wolfs, E. Lebert, Appl. Organomet. Chem. 12
(1998) 591.
[19] D.L. Tsalev, M. Sperling, B. Welz, Analyst 123 (1998)
1703.
[20] X. Wei, C.A. Brockhoff_Schwegel, J.T. Creed, J. Anal.
Atom. Spectrom. 16 (2001) 12.
[21] M.A. Suner, V. Devesa, I. Rivas, D. Velez, R. Montoro, J.
Anal. Atom. Spectrom. 15 (2000) 1501.
907
[22] E. Vassileva, A. Becker, J.A.C. Broekaert, Anal. Chim.
Acta 441 (2001) 135.
[23] T. Dagnac, A. Padro, R. Rubio, G. Rauret, Anal. Chim.
Acta 364 (1998) 19.
[24] R. Pongratz, Sci. Total Environ. 224 (1998) 133.
[25] J. Wu, Z. Mester, J. Pawliszyn, Anal. Chim. Acta 424
(2000) 211.
[26] M. Pardo-Martinez, P. Vinas, A. Fisher, S.J. Hill, Anal.
Chim. Acta 441 (2001) 29.
[27] M.B. Amran, F. Lagarde, M.J.F. Leroy, Microchim. Acta
127 (1997) 195.
[28] I. Ipolyi, P. Fodor, Anal. Chim. Acta 413 (2000) 13.
[29] B. Do, S. Robinet, D. Pradeau, F. Guyon, J. Chromatogr.
A 918 (2001) 87.
[30] X.C. Le, X.F. Li, V. Lai, M. Ma, S. Yalcin, J. Feldman,
Spectrochim. Acta Part B 53 (1998) 899.
[31] U. Kohlmeyer, J. Kuballa, E. Jantzen, Rapid Commun.
Mass Spectrom. 16 (2002) 965.
[32] J.A. Brisbin, C. B’Hymer, J.A. Caruso, Talanta 58 (2002)
133.
[33] C. B’Hymer, J.A. Caruso, J. Liquid Chromatogr. R.T. 25
(2002) 639.
[34] D. Beauchemin, M.E. Bednas, S.S. Berman, J.W. McLa-
ren, K.W.M. Siu, R.E. Sturgeon, Anal. Chem. 60 (1988)
2209.
[35] E.G. Jaeger, K. Schoene, G. Werner, ‘‘Elektrolytgleichge-
wichte und Elektrochemie’’, DVG Leipzig, 1989.
[36] Agilent 7500 ICP
/MS ChemStation Operator’s Manal,
Agilent Technologies Inc., 2000.