RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2002; 16: 17661773

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/rcm.790

Direct determination of alkyl phosphates in human urine

by liquid chromatography/electrospray tandem mass
spectrometry
Felix Hernandez*, Juan V. Sancho and Oscar J. Pozo
Analytical Chemistry, Experimental Sciences Department, ESTCE, University Jaume I, E-12071 Castellon, Spain
Received 10 June 2002; Revised 18 July 2002; Accepted 18 July 2002

A novel and rapid procedure based on liquid chromatography/tandem mass spectrometry (LC/
MS/MS) for the determination of dialkyl phosphates (metabolites of organophosphorus pesticides)
in human urine has been developed. After addition of 40 mM tetrabutylammonium acetate, 10 mL of
urine sample were directly injected into the LC/MS/MS system. The method was validated yielding
calibration curves with correlation coefficients greater than 0.997 and repeatability coefficient of
variation (CV) lower than 9%. The accuracy and precision were evaluated by direct injection of
spiked samples at 10 and 100 mg/L obtaining recoveries between 78 and 119% with coefficients of
variation below 12%. Limits of detection of 1 mg/L for diethyl phosphate (DEP), diethylthiophosphate (DETP) and diethyldithiophosphate (DEDTP) and 2 mg/L for dimethyldithiophosphate
(DMDTP) were achieved, all the analytes being detected in negative ion mode. The fragmentation
pathway of dialkyl phosphates allowed us the use of an additional transition for confirmation in
order to improve their identification in real-world samples. The applicability of the LC/MS/MS
method was demonstrated by applying it to the analysis of urine samples of farmers exposed to the
organophosphorus pesticide chlorpyrifos. Good correlation between application of the product in
the field (citrus orchards), concentration levels of dialkyl phosphates and levels of the chlorpyrifosspecific metabolite (1,3,5-trichloro-2-pyridinol) was obtained. Copyright # 2002 John Wiley & Sons,
Ltd.

The presence of pesticide metabolites in human biological

fluids has shown to be an indicator of human exposure to
agrochemical compounds. Due to the great variety and high
activity of human enzymes, which metabolize most xenobiotics, pesticides metabolism is extensive and follows
numerous pathways. For organophosphorus insecticides,
typical metabolism is carried out in two steps, giving the
specific alcohol of each pesticide and alkyl phosphates,
which are common metabolites of this family of compounds.1
Exposure to organophosphorus pesticides is usually
determined by measuring a reduction in the cholinesterase
enzyme activity in blood. However, this method lacks
selectivity, and also sensitivity for low-level exposure, and
needs the establishment of a baseline activity level from nonexposed subjects.
Determination of the specific alcohol is usually chosen for
biological monitoring in order to indicate low-level exposure
to some organophosphorus compounds.25 However, the
non-availability of most alcohol metabolite standards and
*Correspondence to: F. Hernandez, Analytical Chemistry, Experimental Sciences Department, ESTCE, University Jaume I, E12071 Castellon, Spain.
E-mail: hernandf@exp.uji.es

the high specificity of this type of analysis make the

determination of alkyl phosphates a suitable approach to
evaluate general human exposure to organophosphorus
compounds.
O,O-Dimethyl phosphate (DMP), O,O-diethyl phosphate
(DEP), O,O-dimethylthiophosphate (DMTP), O,O-diethylthiophosphate
(DETP),
O,O-dimethyldithiophosphate
(DMDTP) and O,O-diethyldithiophosphate (DEDTP) are
the alkyl phosphates frequently analyzed in order to monitor
exposure to organophosphorus pesticides. Results available
from the literature show that dialkyl phosphorodithioates
are not usually detected in urine samples, either in exposed
or non-exposed populations, while dialkyl phosphorothioates and phosphates are commonly detected.
Although the higher polarity of alkyl phosphates recommends their determination by liquid chromatography (LC),
these compounds are typically determined by methods
based on gas chromatography (GC) with flame photometric
detection612 (FPD), flame ionization detection13,14 (FID),
nitrogen phosphorus detection15,16 (NPD), electron capture
detection17 (ECD) or mass spectrometry14,17,18 (MS). In this
way a derivatization prior to analysis is compulsory.
Pentafluorobenzyl bromide is the most commonly used
derivatization agent,612,1518 although diazomethane13,14 is
also employed for this purpose. The derivatization reaction
Copyright # 2002 John Wiley & Sons, Ltd.

LC/MS/MS determination of phosphates in urine

is frequently followed by a solvent exchange step appropriate to GC determinations. Detection limits between 2 and
100 mg/L are usually achieved in these methods.616 Mass
spectrometric detection is sometimes used for analyte
confirmation.14,16 Recently, GC/MS17 and GC/MS/MS18
methods have been used for the quantitative determination
of dialkyl phosphates in urine.
Obviously, the traditional LC detectors, UV and fluorescence, are not appropriate for the direct determination of
dialkyl phosphates due to the lack of chromophore or
fluorophore moieties in the analyte molecules. The recent
introduction of atmospheric pressure interfaces, mainly
electrospray ionization (ESI), for coupling LC to MS has
popularized the use of LC/ESI-MS for the determination of
many analytes unsuitable for conventional liquid chromatography.19,20 In addition, the use of tandem mass spectrometry (MS/MS), due to its inherent high selectivity,21,22
allows the direct injection of complex matrices into the LC
system.
Nowadays, the potential of LC/MS and LC/MS/MS
techniques in the direct determination of pesticide metabolites in urine has been demonstrated2,23 and some compounds similar to organophosphorus metabolites such as
phosphonates,24 inositol phosphate25 or dibutyl phosphate26
have been determined by LC/MS methods.
In this paper, a new analytical method based on direct
injection into an LC/MS/MS system is developed for the
sensitive determination of dialkyl phosphates in urine
samples. To our knowledge, this paper describes the first
application of LC/MS/MS in the analysis of alkyl phosphates in urine samples. The potential of the developed
procedure for the biomonitoring of chlorpyrifos exposure in
pesticide applicators has been confirmed.

EXPERIMENTAL
Reagents and chemicals

Diethylthiophosphate (DETP) potasium salt, diethyldithiophosphate (DEDTP) potasium salt, dimethyl chlorophosphate (DMClP) and dimethyl chlorothiophosphate
(DMClTP) were purchased from Aldrich (Milwaukee, WI,
USA); diethyl phosphate (DEP) was purchased from Supelco
(Bellfonte, PA, USA). HPLC-grade acetonitrile was purchased from ScharLab (Barcelona, Spain). LC-grade water
was obtained by purifying demineralized water in a
Nanopure II system (Barnstead Newton, MA, USA). Analytical-grade formic acid (HCOOH, content >98%) was
purchased from Fluka (Buchs, Switzerland) and tetrabutylammonium acetate (TBA) was obtained from Sigma (St
Louis, MI, USA).
Standard stock solutions of DEP, DETP and DEDTP were
prepared by dissolving 25 mg powder, accurately weighed,
in 50 mL of HPLC-grade acetonitrile obtaining a final
concentration of 500 mg/mL. Stock solutions of synthesized
DMP and DMTP were diluted with HPLC-grade water to
obtain a concentration of 500 mg/mL. A standard stock
mixture of five dialkyl phosphates was prepared taking 5mL aliquots of each individual solution and diluting to
50 mL with HPLC-grade acetonitrile. For the LC/MS/MS
Copyright # 2002 John Wiley & Sons, Ltd.

1767

Figure 1. Synthetic route scheme for the preparation of DMP

and DMTP.

analysis, this stock solution was diluted with LC-grade

water.

Synthesis of dimethyl alkyl phosphates

The preparation of dimethyl phosphate (DMP) and dimethylthiophosphate (DMTP) was carried out very efficiently by a simple hydrolysis of DMClP and DMClTP, as
shown in Fig. 1. A solution of triethylamine (3 mL) in 40 mL
LC-grade water/HPLC-grade acetonitrile (10:30) was prepared and stirred at 0 C. Then, DMClP (5 g) or DMClTP (5 g)
was carefully and quantitatively dissolved into this solution
and HPLC-grade acetonitrile was added in order to obtain
an exact volume of 200 mL, achieving concentrations of
21,800 and 22,118 mg/L for DMP and DMTP, respectively.

Liquid chromatography

A Waters Alliance 2690 HPLC system (Waters, Milford, MA,

USA) was interfaced to a Quattro LC triple quadrupole mass
spectrometer (Micromass, Manchester, UK). The LC separation was performed using a Discovery C18 5-mm, 50  2 mm
column (Supelco, Bellefonte, PA, USA), at room temperature
and at a flow rate of 300 mL/min with mobile phases
consisting of acetonitrile/1mM TBA in water.

Mass spectrometry

A Quattro LC (quadrupole-hexapole-quadrupole) mass

spectrometer with an orthogonal Z-spray-electrospray interface (Micromass, Manchester, UK) was used. The drying gas
as well as nebulizing gas was nitrogen generated from
pressurized air in a NG-7 nitrogen generator (Aquilo, EttenLeur, NL). The nebulizer gas flow was set to approximately
80 L/h and the desolvation gas flow to 800900 L/h. Infusion
experiments were performed using a model 11 single syringe
pump (Harvard, Holliston USA), directly connected to the
interface.
For operation in the MS/MS mode, the collision gas was
99.995% argon (Carburos Metalicos, Valencia, Spain) with a
pressure of 2  10 3 mbar in the collision cell. A capillary
voltage of 3.0 kV was used in negative ionization mode. The
interface temperature was set to 350 C and the source
temperature to 120 C. Selected precursor and product ions,
cone voltages and collision energies used in the multiple
reaction monitoring (MRM) function are shown in Table 1.
Dwell times of 0.2 s/scan were chosen. A solvent delay of
4 min was chosen in order to have an additional clean-up
Rapid Commun. Mass Spectrom. 2002; 16: 17661773

1768 F. Hernandez, J. V. Sancho and O. J. Pozo

Table 1. Optimized mass spectrometry parameters for the determination of alkyl phosphates
Compound
DMP
DMTP
DEP
DETP
DEDTP
a
b

Precursor ion (m/z)

Cone voltage (V)

Collision energy (eV)

Product ion (m/z)

125a
125b
141a
141b
153a
153b
169a
169b
185

30
30
20
20
20
20
30
30
30

17
25
17
25
12
20
20
20
20

63
79
126
95
125
79
95
141
111

Quantification transition.
Confirmation transition.

using the built-in divert valve controlled by the Masslynx

NT vers. 3.5 software (Micromass, Manchester, UK). This
software was also used to process the quantitative data
obtained from calibration standards and urine samples.

Analytical procedure

Urine samples were supplied from non-exposed and

exposed populations by the Gabinete Tecnico de Seguridad
e Higiene en el Trabajo (Conselleria de Empleo, Industria y
Comercio, Castellon, Spain).
The urine samples were centrifuged in an Angular 6
centrifuge (Selecta, Barcelona, Spain) for 10 min at 3500 rpm.
In order to obtain a TBA concentration of 40 mM, 100 mL of a
solution of 600 mM tetrabutylammonium acetate were
added in a 2-mL vial containing 1.5 mL of the centrifuged
urine or the standard solution.
Analyses were performed under experimental conditions
indicated above, by injecting 10 mL, using as mobile phase
1 mM TBA in water/acetonitrile, where the percentage of
acetonitrile was changed linearly as follows: 0 min, 5%;
3 min, 5%; 7 min, 30%; 11 min, 30%; 12 min, 5%; 15 min, 5%.

Validation study

The precision of the method was evaluated within-day by

determining dialkyl phosphates in two standard solutions
prepared at 10 and 100 mg/L (n = 5). The calibration curve
was obtained by analyzing standard solutions in triplicate at
eight concentrations between 5 and 1000 mg/L (5, 10, 25, 50,
100, 250, 500, 1000 mg/L). The percentage recovery was
evaluated by determination of dialkyl phosphates in two
blank samples spiked at two concentration levels (10 and
100 mg/L), and the precision of the overall analytical
procedure was expressed as coefficient of variation (CV) in
%. In all cases, experiments were performed in quintuplicate
(n = 5).
The limit of quantitation (LOQ) was established as the
lowest concentration assayed that gave acceptable recoveries
(>70%) and precisions (<15%).
The limit of detection (LOD), defined as the lowest
concentration that the analytical process can reliably
differentiate from background levels, was obtained when
the signal was three times the background noise in the
chromatogram at the lowest analyte concentration assayed.
Copyright # 2002 John Wiley & Sons, Ltd.

RESULTS AND DISCUSSION

Infusion experiments

The full-scan mass spectra and the MS/MS spectra of dialkyl

phosphates are shown in Fig. 2. They were obtained from
infusion of 5 mg/mL 50:50 acetonitrile/water solutions of
each compound at 10 mL/min. All compounds were analyzed in negative ion mode due to their inherent negative
charge.
The spectra of the synthesized compounds (DMP and
DMTP) did not show any ions for their chlorophosphate
precursors (m/z 144 and 160, respectively), as shown in Figs
2(d) and 2(e). Therefore, the synthesis was considered fully
quantitative.
Figure 3 shows the proposed fragmentation pathways for
the dialkyl phosphates, which vary with the nature of the
alkyl group. The ethyl group in DEP, DETP, and DEDTP
allows the McLafferty rearrangement yielding m/z 125, 141
and 157, respectively. The subsequent neutral loss of ethanol
from each fragment allows the formation of additional
product ions at m/z 79, 95 and 111, when the collision energy
is increased.
For DMP and DMTP, the presence of a methyl group
prevents the formation of the six-membered ring required in
the McLafferty rearrangement. In this case, losses of methyl
(to m/z 110 and 126) and dimethyl ether (to m/z 79 and 95)
were obtained (see Fig. 3). DMP can also lose two methoxy
groups yielding a fragment at m/z 63. These fragmentations
are less favored than the McLafferty rearrangement and,
therefore, ethyl derivatives were more sensitive than methyl
derivatives.
The chemical similarity of these compounds and their
fragmentation pathways means that the specific DMP
transition (125 79) could be interfered with by the
presence of DEP. Cone fragmentation of the [M H] ion
for DEP (m/z 153) produced a fragment at m/z 125, which
decomposed further in the collision cell to generate a
product ion at m/z 79 (See Fig. 3). The same behavior was
observed for the compounds DMTP/DETP, which shared
the 141 95 transition. In this situation an efficient
chromatographic separation was required, although additional fragmentations could be used as confirmation transitions in order to have less possibility of producing false
positives.
Rapid Commun. Mass Spectrom. 2002; 16: 17661773

LC/MS/MS determination of phosphates in urine

1769

Figure 2. Negative ion electrospray full-scan mass spectra (bottom) and product ion spectra (top) from (a) DEP, (b) DETP, (c) DEDTP,
(d) DMP, and (e) DMTP.

Table 1 shows the mass spectrometric parameters selected

as precursor ion, product ion and optimized cone and
collision energy for selected compounds in the quantification
and confirmation transitions.
Copyright # 2002 John Wiley & Sons, Ltd.

LC optimization

As stated above, the coincidence of some transitions (e.g. 125

79 in the case of DMP and DEP) demanded an adequate
chromatographic separation between analytes. On the other
Rapid Commun. Mass Spectrom. 2002; 16: 17661773

1770 F. Hernandez, J. V. Sancho and O. J. Pozo

high amounts of endogenous substances in urine samples,

which could compete with alkyl phosphates in the formation
of TBA ion pairs. As a result, analytes appeared either as
unretained compounds (the most polar compounds like
DMP or DMTP) or with bad peak shape (like DEP or DETP).
In order to increase the accessibility of TBA for all analytes,
addition of 40 mM TBA to urine samples was assessed. Good
peak shapes were obtained for all analytes, as Fig. 4 shows,
and only DMP remained as an unretained compound.
Therefore, a multiresidual method including DMTP, DEP,
DETP and DEDTP was further validated.
On the other hand, early-eluting analytes showed shorter
retention times (around 1 min lower) in spiked urine
samples than in standard solutions. These variations in
retention were, however, similar for all the urine samples
tested, and were possibly due to the interaction between
TBA and urine interferences, which impeded the formation
of adducts between analytes and TBA. Attempts to solve this
situation included the addition of 80 mM TBA to samples;
however, similar results were obtained.
This undesirable situation regarding the variability of
retention times was greatly compensated for by using the
additional confirmation transitions for each analyte (see
Table 1), therefore producing more reliable results when
dealing with real-world samples.
The first-eluted millilitres of the chromatographic column
were sent to waste by using the built-in divert valve in the
mass spectrometer controlled by the Masslynx software. This
solvent delay (4 min) gave an additional clean-up step and
avoided the overload of the interface with early-eluting
interferences.

Validation study

Figure 3. Proposed fragmentation pathway for dialkyl

phosphates.

hand, the high polarity of alkyl phosphates made the use of

an ion-pairing reagent advisable in order to achieve
sufficient retention time, to allow the separation between
highly polar interferences (salts) and analytes. A volatile ionpairing reagent, TBA acetate, was selected in order to
prevent interface blockage.
First, addition of 1 mM TBA in the mobile phase was
tested. Chromatograms with good peak shape and enough
retention time were obtained for all analytes using standard
solutions. However, when a spiked urine sample was
directly injected under these conditions, only the more
retained compound (DEDTP) was obtained with a good
peak shape. This situation could be due to the presence of
Copyright # 2002 John Wiley & Sons, Ltd.

Analytical characteristics resulting from the validation of

alkyl phosphates study are reported in Table 2. Calibration
curves obtained in the range 51000 mg/L showed excellent
linearity, with correlation coefficients between 0.997 and
0.999. The method was precise for all compounds (relative
standard deviation (RSD) <9%) with limits of detection,
calculated from a 5 mg/L chromatogram, of 1 mg/L for ethyl
derivatives (DEP, DETP, DEDTP) and 2 mg/L for DMTP.
Precision and accuracy of the overall analytical procedure
were evaluated from urine samples spiked at the concentration levels of 10 and 100 mg/L (n = 5 each), and are reported
in Table 3. Considering the analytical difficulties inherent to
the trace level determination of these analytes, the method
was found to be precise (CV <12%) and accurate (recovery
between 78 and 119%).
Typical chromatograms of 10-mg/L standard solutions
and urine samples (blank and 10 mg/L spiked) are shown in
Fig. 4.

Application to real samples

The validated method was successfully applied to evaluate

the exposure to chlorpyrifos ethyl in a group of volunteer
farmers who applied this organophosphorus insecticide in
citrus orchards. Urine samples were collected daily over a
week from exposed and non-exposed growers. Figure 5
gives an overview of the results obtained, and shows the
Rapid Commun. Mass Spectrom. 2002; 16: 17661773

LC/MS/MS determination of phosphates in urine

1771

Figure 4. LC/ESI-MS/MS chromatograms for DMTP, DEP, DETP and DEDTP: (a) 10 mg/L standards, (b)
urine sample blank and (c) urine sample spiked at 10 mg/L.

Table 2. Analytical characteristics of the developed LC/ESMS/MS method

Linearity

DMTP
DEP
DETP
DEDTP

Repeatability (n = 7, CV%)

Range (mg/L)

10 mg/L

100 mg/L

51000
51000
51000
51000

0.999
0.997
0.999
0.999

9
4
3
7

8
7
2
4

Table 3. Validation study of the developed procedure for the

determination of dialkyl phosphates in spiked urine samples (two
samples at two levels, n = 5 each)
Recovery
Compound
DMTP
DEP
DETP

correlation between application of chlorpyrifos and the

concentration levels of dialkyl phosphates in urine samples.
In the exposed workers (samples a and b), the metabolites
DETP and DEP were detected at concentration levels of
around 40 and 80 mg/L, respectively, after several hours of
chlorpyrifos application in the field (application was made
on Wednesday). These levels decreased to 10 and 30 mg/L
after a weekend without applications. After a second
application of chlorpyrifos in the field on Monday, metabolite concentrations of 5070 (DETP) and 90100 mg/L (DEP)
were reached in urine samples from Tuesday and Wednesday. In non-exposed farmers (sample c), levels of DEP and
DETP lower than 10 mg/L were always obtained.
In the case of DMTP, no relation between application of
chlorpyrifos and the presence of this compound was
observed. Nearly all analyzed samples were positives with
an average level of 20 mg/L. These findings, as well as the
results of DEP and DETP in non-exposed workers, were
comparable with those of recent studies,17,18 where average
concentrations of 34 mg/L (DEP), 14 mg/L (DETP) and 10
22 mg/L (DMTP) were obtained in non-exposed subjects.
DEDTP was not detected in any samples in this study, in
agreement with other authors who only found this compound in 2% of urine samples from volunteers (both exposed
and non-exposed to organophosphorus pesticides).17,18
As an additional confirmation of the correlation between
dialkyl phosphates and the application of chlorpyrifos, the
Copyright # 2002 John Wiley & Sons, Ltd.

DEDTP
a
b

10 mg/L
Sample
Sample
Sample
Sample
Sample
Sample
Sample
Sample

1
2
1
2
1
2
1
2

110
119
90
82
78
86
83
82

(9 )
(11)
(10)
(6)
(9)
(5)
(10)
(7)

100 mg/L
119
101
95
92
79
87
90
90

(12)
(12)
(9)
(6)
(6)
(4)
(5)
(5)

Recovery (%).
Coefficient of variation in parentheses (%).

specific metabolite of this pesticide, 1,3,5-trichloro-2-pyridinol (TCP), was also determined in all urine samples. The
method used was based on the direct injection of hydrolyzed
urine samples into an LC/LC/MS/MS system.2 Figure 5
shows good correlation between the behavior of the diethyl
phosphates and that of TCP, although the concentration
levels of TCP were much higher than those of DEP and
DETP. This was possibly due to the higher excretion rates of
very polar compounds like phosphates as they do not have
to suffer a phase II metabolism process in order to be
excreted. However, TCP has to be conjugated before the
excretion and, therefore, its metabolism will probably be
slower. As urine samples were collected early in the morning
after the day of exposure, dialkyl phosphates could be
largely excreted during the previous day, while TCP could
attain a maximum excretion level after some hours.
Figure 6(a) shows chromatograms of alkyl phosphates and
TCP for an exposed worker (sample a, see Fig. 5).
Confirmation as well as quantification transitions are also
shown in Fig. 6(b) for DEP and DETP in order to authenticate
the presence of these analytes.
Rapid Commun. Mass Spectrom. 2002; 16: 17661773

1772 F. Hernandez, J. V. Sancho and O. J. Pozo

Figure 5. Evolution of concentration levels of chlorpyrifos metabolites in urine during one

week of application in citrus orchards obtained for exposed (a and b) and non-exposed
farmers (c).

Figure 6. Chromatograms from urine sample of a farmer exposed to chlorpyrifos (sample a, collected on Thursday): (a) LC/ESIMS/MS chromatograms for alkyl phosphates and LC/LC/ESI-MS/MS chromatogram for TCP; (b) confirmation of the two dialkyl
phosphates derivatives found, (Q) quantitation transition; (q) confirmation transition.
Copyright # 2002 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2002; 16: 17661773

LC/MS/MS determination of phosphates in urine

1773

CONCLUSIONS

REFERENCES

The analytical method described demonstrates that LC/

MS/MS is a suitable technique for the direct determination
of alkyl phosphates in urine samples. The method developed
presents good analytical figures and low detection limits
(around 12 mg/L) without any type of sample treatment.
However, some changes in retention time of early-eluting
compounds take place, a problem that can be minimized by
using an additional confirmation MS/MS transition with the
aim of unequivocally confirming the identity of analytes in
real-world samples. A chromatographic run time of 15 min is
sufficient to obtain acceptable results in the multiresidual
analysis of DMTP, DEP, DETP and DEDTP, making this
method suitable for the biological monitoring of organophosphorus pesticides exposure. The method was successfully applied to urine samples from farmers exposed to
chlorpyrifos, achieving good correlation between concentration levels of DEP and DETP, and the specific metabolite,
1,3,5-trichloro-2-pyridinol.
Further investigation will be required on the analysis of
other urinary metabolites of organophosphorus pesticides,
such as DMP and monoalkyl phosphates. The higher
polarity and the low masses of these molecules suggest that
a derivatization step in order to improve retention time and
mass characteristics would be advisable.

1. Chambers JE, Ma T, Chambers HW. In The Mammalian

Metabolism of Agrochemicals, Hutson DH, Paulson GD (eds).
John Wiley: Chichester, 1995; 135.
2. Sancho JV, Pozo O, Hernandez F. Rapid Commun. Mass
Spectrom. 2000; 14: 1485.
3. Chang MJW, Lin CY, Lo LW, Lin RS. Bull. Environ. Contam.
Toxicol. 1996; 56: 367.
4. Hunter DL, Lassiter TL, Padilla S. Toxicol. Appl. Pharmacol.
1999; 158: 16.
5. Aprea C, Betta A, Catenacci G, Lotti A, Magnaghi S, Barisano
A, Passini V, Pavan I, Sciarra G, Vitalone V, Minoia C. J.
AOAC Int. 1999; 82: 305.
6. Nutley BP, Cocker J. Pestic. Sci. 1993; 38: 315.
7. Aprea C, Sciarra G, Lunghini L. J. Anal. Toxicol. 1996; 20: 559.
8. Aprea C, Sciarra G, Sartorelli P, Desideri E, Amati R,
Sartorelli E. Int. Arch. Occup. Environ. Health. 1994; 66: 333.
9. Takamiya K. Bull. Environ. Contam. Toxicol. 1994; 52: 190.
10. Loewenherz C, Fenske RA, Simcox NJ, Bellamy G, Kalman
D. Environ. Health Perspectives 1997; 105: 1344.
11. Reid SJ, Watts RR. J. Anal. Toxicol. 1981; 5: 126.
12. Moate TF, Lu C, Fenske RA, Hahne RMA, Kalman DA. J.
Anal. Toxicol. 1999; 23: 230.
13. Drevenkar V, Radic Z, Vasilic Z, Reiner E. Arch. Environ.
Contam. Toxicol. 1991; 20: 417.
14. Drevenkar V, Stengl B, Frobe Z. Anal. Chim. Acta 1994; 290:
277.
15. Bardarov V, Mitewa M. J. Chromatogr. 1989; 462: 233.
16. Miki A, Tsuchihashi H, Ueda K, Yamashita M. J. Chromatogr.
A 1995; 718: 383.
17. Hardt J, Angered J. J. Anal. Toxicol. 2000; 24: 678.
18. Oglobine AN, Elimelakh H, Tattam B, Geyer R, O'Donell GE,
Holder G. Analyst 2001; 126: 1037.
19. Ingelse BA, van Dam RCJ, Vreeken RJ, Mol HGJ, Steijger
OM. J. Chromatogr. A 2001; 918: 67.
20. Ahrer W, Buchberger W. J. Chromatogr. A 1999; 854: 275.
21. Kato K, Jingu S, Ogawa N, Higuchi S. Rapid Commun. Mass
Spectrom. 1999; 13: 1626.
22. Casetta B, Romanello M, Moro L. Rapid Commun. Mass
Spectrom. 2000; 14: 2238.
23. Sancho JV, Pozo O, Lopez FJ, Hernandez F. Rapid Commun.
Mass Spectrom. 2002; 16: 639.
24. Black RM, Read RW. J. Chromatogr. A 1998; 794: 233.
25. Buscher BAP, Van der Hoeven RAM, Tjaden UR, Andersson
E, Van der Greef J. J. Chromatogr. A 1995; 712: 235.
26. Lamouroux C, Virelizier H, Moulin C, Tabet JC, Jankowski
CK. Anal. Chem. 2000; 72: 1186.

Acknowledgements

The authors are very grateful to the Gabinete Tecnico de

Seguridad e Higiene en el Trabajo, Generalitat Valenciana,
Castellon, Spain (M. J. Saez and C. Armelles) for providing
urine samples from exposed workers, and to the Serveis
Centrals d'Instrumentacio Cientfica (SCIC) of the University
Jaume I for using the Quattro LC triple quadrupole mass
spectrometer. This work was partially supported by Research Project PB98-1043 (Direccion General de Ensenanza
Superior e Investigacion Cientfica, Ministerio de Educacion
y Cultura, Spain).

Copyright # 2002 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2002; 16: 17661773