Determination of anti-inflammatory drugs in water samples,

by in situ derivatization, solid phase microextraction and

gas chromatography mass spectrometry
Lilia Araujo, Johana Wild, Noreiva Villa, Nuris Camargo,
Dalia Cubillan, Avismelsi Prieto*
Laboratory of Analytical Chemistry and Electrochemistry, Faculty of Engineering,
University of Zulia, P O Box 4011-A-526, Maracaibo, Venezuela.

Abstract
A new analytical method is presented for the determination of non-steroidal acidic antiinflammatory drugs in water samples. These compounds are used as anti-inflammatory,
antipyretic and analgesic drugs in human health care and in veterinary applications. The
analytical procedure involves in situ derivatization of analytes to their methyl esters
with dimethyl sulphate, headspace sampling using solid-phase microextraction
(SPME), and gas chromatography-mass spectrometry (GC-MS). Optimization of
derivatization, pH, ionic strength, extraction time, SPME fibre and extraction
temperature are described. Methyl esters were extracted with a fused-silica fibre coated
with 100 m polydimethylsiloxane. Response was linealy dependent on concentration
in the range 0.1 10.0 ng/mL. Detection limits are achieved at the level of 0.3 2.9
ng/L. Derivatization SPME/GC-MS analysis yielded good precision (RSD between
7.9 and 14.3 %). The method was validated by analysis of spiked matrix samples.
Key words: non-steroidal acidic anti-inflammatory drugs, derivatization, solid-phase
microextraction, gas chromatography-mass spectrometry.

*Corresponding author. Fax: 0058-261-7598736; e-mail: aprieto@luz.edu.ve

1. Introduction
Pharmaceuticals used for humans and animals have been identified as emerging
environmental pollutants [1-2]. Non-steroidal acidic anti-inflammatory drugs (NSAIDs)
are among the group of pharmaceutical compounds most often used to treat human and
animal illnesses. These compounds are excreted unchanged or as active metabolites in
high percentages and continuously discharged into domestic wastewaters. Incomplete
removal during wastewaters biological treatments

has caused their presence into

surface waters at concentrations from ng/L to g/L level [ 3-5]. The complete effects on
aquatic organisms and humans are not well understood. Due to their biological activity,
their determination in the environment is important [6].
Suitable methods exist for a quantitative measurement of NSAIDs in water
samples, using gas chromatography-mass spectrometry (GC-MS), high performance
liquid chromatography-mass spectrometry (HPLC-MS), micellar electrokinetic capillary
chromatography (MEKC) and capillary electrophoresis [7-12], but in general the
preconcentration step is extensive, requires large sample volumes

(between 0.5 and

2.0 L), sample clean-up or the use of organic solvents. There is a need for fast and
simple methods to supply these analysis.
Solid phase microextraction (SPME) is a two-step process conducive to the
simultaneous extraction and preconcentration of organic compounds. Since its
introduction, SPME has gained popularity as a simple, solvent-free, reliable and flexible
tool for the sampling of a variety of volatile and semivolatile compounds. SPME
requires less sample volume than solid phase extraction or liquid liquid extraction.
SPME is suitable for analysis of drugs residues in water. On the other hand GC analysis
of acidic drugs is complicated because a derivatization step is necessary prior to
analysis. Generally, prior to the determination of acidic drugs in water, the analytes are

transferred to an organic matrix where derivatization takes place. SPME allows the
derivatization of the analytes to take place in the sample matrix, in the SPME fibre
coating or in the GC injector port [13-16]. However few studies consider the option of
performing the derivatization directly in the water matrix. To date, NSAIDs have not
been investigated by direct derivatization-SPME procedure. By in situ methylation, the
acidic drugs are converted to less polar methyl esters, which may improve the extraction
into the SPME fibre.
In this paper, a new method for analysis of six non steroidal acidic antiinflammatory drugs (ibuprofen, flufenamic acid, naproxen, mefenamic acid, tolfenamic
acid, and meclofenamic acid) in water samples using headspace SPME/GC-MS is
proposed. The analytical procedure involves in situ derivatization of acidic drugs to
their methyl esters with dimethyl sulfate.
2. Experimental
2.1. Materials and reagents
All reagents were of analytical-reagent grade unless stated otherwise. Water was
purified with a Milli-Q plus system (Millipore). Ibuprofen, flufenamic acid, naproxen,
mefenamic acid, tolfenamic acid and meclofenamic acid were supplied by Sigma (St.
Louis, MO, USA). A stock standard soultion of 1000 g/mL of each compound was
prepared in basic Milli-Q water. Working solutions were obtained by appropriated
dilutions with Milli-Q water. The derivatization reagent dimethyl sulfate (DMS) was
purchased from Riedel-de Haen. Tetrabutylammonium hydrogen sulfate (TBA-HSO4) was
obtained from Fluka. A manual fibre holder for SPME was purchased from Supelco
(www.sigma-aldrich.com). Two types of fibre, 100 m polydimethylsiloxane and 85
m polyacrylate were obtained from the same manufacturer and conditioned before use

as recommended by the manufacturer. The Statgraphics software package was used for
the regression analysis (linear model).
2.2. Instrumentation
A Agilent Technologies 6890N GC system equipped with a split-splitless injector
for the HP-5MS fused silica capillary column (30 m x 0.25 mm i.d., 0.25 m film
thickness), and mass spectrometer detector model Agilent Technologies 5973 was used.
A silanized narrow-bore injector liner (0.75 mm id) for the SPME injections was
installed and the fiber was inserted into this injector using the splitless mode with the
split closed for 3 min. The injector temperature was set at 250 C and the transfer line
temperature was 280 C. The oven temperature was held at 50 C for 3 min, then heated
to 250 C at a heating rate of 30 C/min. Temperature was held at 250 C for 4.5 min.
The carrier gas was helium (purity 99.999 %) at a flow rate of 1 mL/min. The mass
spectrometer detector was tuned by maximum sensitivity autotune. The following m/z
values were acquired in the electron impact ionization mode by single ion monitoring
and used for quantification of the analytes: 161- 220 for ibuprofen, 263- 295 for
flufenamic acid, 185-244 for naproxen, 223-255 for mefenamic acid, 208-275 for
tolfenamic acid and 242-309 for meclofenamic acid.
2.3 Procedure
6.0 mL of standard solution or sample was placed in a 14 mL screw-cap glass
vial. Phosphate buffer solution (pH 6.0, 3.0 mol/L, 2.0 mL) and Na2SO4 (2.88 g) were
then added and the sample was agitated with a 1.8 cm long PTFE-coated stir bar. After
addition of the ion-pairing reagent (TBA-HSO4, 7.5x10-4 M, 60 L), the vial was
closed. 15 L of derivatization reagent (DMS) was then injected through the septum
and the vial was immersed in a temperature-controlled oil bath. After 10 min at 65 2

C, the PDMS fibre was exposed to the headspace above the aqueous solution for 45
min. The samples were agitated with a magnetic stirring bar at 500 rpm during both
derivatization and extraction. After the extraction the fibre was directly exposed to the
hot injector of the GC for 5 min and the chromatogram was registered.
Calibration graphs in Milli-Q water were thus constructed using solutions of the
analytes of known concentrations.
3. Results and discussion
3.1. Optimization of the microextraction with in situ derivatization
It has been reported that DMS reacts with haloacetic acids and chlorophenoxy
acid herbicides

in water to produce the corresponding methyl ester [17-18]. The

addition of ion-pairing agents, which activate the analytes during derivatization,

increases esterification yields and thus improves the sensitivity of the procedure. In this
work, experiments were performed in which DMS and the ion-pair agent TBA-HSO 4
was added to aqueous solutions containing each of the six test compounds. After the
reaction, the solution was extracted with headspace SPME and analysed by means of
GC-MS. Detectable yields of methyl esters were achieved for the analytes and identified
on the basis of their mass spectra. Next, preliminary experiments were carried out to
optimise the main parameters affecting both derivatization and SPME of the analytes
investigated. In these studies Milli-Q water samples spiked with the appropriate amount
of the standard solution were used.
The more adequate fibre was found by comparing the extraction behavior on two
commercial SPME fibres, 100 m polydimethylsiloxane
polydimethylsiloxane

and 85 m polyacrylate.

fibre exhibited the highest extraction performance and was

chosen for the rest of experiments.

The optimum pH was determined in the range between 3 and 10. As shown in
the Fig. 1, no significant effect was observed in the range of 3 - 7. At higher pH, a
decrease was observed in the response. From these results it was decided to adjust the
pH of water samples to 6.0. A 0.75 M concentration of the phosphate buffer (pH 6.0)
was selected to obtain an adequate buffering capacity.
Figure 1

The effect of temperature was monitored by extracting samples of 10 ng/mL of

NSAIDs in the range of 40 - 80 C. Fig. 2 shows a clear increase in the amount of
analytes adsorbed when the temperature increases. However, when the temperature
exceeded 70 C, there was a decrease in the amount of NSAIDs extracted except for
mefenamic acid, tolfenamic acid and meclofenamic acid. As a compromise, 65 C was
chosen as the optimum temperature for the derivatization and SPME extraction of all
the methyl derivatives of NSAIDs.
Figure 2

In order to set an optimal volume of the derivatization reagent necessary for

complete esterification of NSAIDs, a range of 5-100 L of DMS volumes were tested
at 65 C. Increasing the volume of DMS in the range 5-60 L improved slight the yield
of methyl esters while the response decreased at higher amounts of DMS. However, 100
m PDMS fibre was too fragile at DMS volume higher than 15 L and it could only be
used for a limited number of experiments. For this reason 15 L was used for the rest of
experiments.
We investigated the effect of the TBA-HSO4 volume on the amount of NSAIDs
extracted fron the sample onto the PDMS fibre. The TBA-HSO 4 volume profile was

studied by monitoring the GC area counts as a function of TBA-HSO 4 volume. The

effect of the ion-pairing agent on the yield of the methylation is shown by the increase
in the peak area obtained with the use of TBA-HSO 4 (Fig. 3). The addition of 60 L of
TBA-HSO4 ensured the maximum response for NSAIDs methyl esters.

Figure 3

Extraction time profiles were studied extracting samples of 10 ng/mL of

NSAIDs and monitoring the GC area counts as a function of exposure time. Equilibrium
was not attained even after 90 min. For quantitative analysis it is not necessary for the
analytes to have reached equilibrium, but only for sufficient loading onto the fibre and
reproducible extraction time [19]. An extraction time of 45 min was selected as a
compromise between analyte response and time analysis.
The role of the ionic strength of the matrix was investigated using sodium
chloride and sodium sulfate. For many organic analytes, aqueous solubility decreases
with increasing ionic strength, and thus, the partitioning from the aqueous solution to
the headspace is improved. The amounts tested for a sample volume of 8 mL varied
from 0 to 2.88 g of NaCl or Na 2SO4. The results obtained show an strong increment in
the signal for all analytes when the NaCl or Na 2SO4 amount increases. The highest peak
area response was achieved using Na2SO4. Thus, 2.88 g of Na2SO4 was selected to
obtain an adequate salting-out effect.
To study the carryover effect, blanks were run after extractions of 50 ng/
mL of NSAIDs. No signals were obtained when a 5 min desorption time was chosen,
which ensured a complete desorption of methyl esters.

3.2 Application and Validation of the Proposed Method

Calibration graphs for Milli-Q water samples treated according to the procedure
described previously, monitored using SIM mode, are linear for the concentration range
0.1 - 10.0 ng/mL. This range agreed with environmental levels in water currently
reported in the literature for these compounds [20-22]. Two replicates were used for
each of six prepared standards to obtain the calibration graphs. The results for the
intercept (a), slope (b), regression coeficient (r) and linearity [1 - R.S.D(b)] (%) are
summarised in Table 1. The precision was measured by performing 8 independent
determinations. Precisions ranged from 7.9 to 14.3 % R.S.D. (Table 1), which should be
satisfactory for determining the NSAIDs in water matrix.
The detection limit was calculated by comparing the signalto-noise ratio (S/N)
of the lowest detectable concentration to a S/N = 3. A S/N of 10 was applied for the
calculation of the quantification limit. The detection limits found were between 0.3 and
2.9 ng/L ( Table 1). The detection limits are in accordance with other methods based on
SPE-MEKC,

SPE-derivatization-GC/MS,

SPME-derivatization-GC/MS

or

SPE-

HPLC/MS [11, 14, 23-24].

Table 1

The optimum headspace SPME sampling conditions for Milli-Q water were
applied to the tap water and wastewater matrices.
Absolute response was lower with wastewater samples than with Milli-Q water.
This might be due to the presence of additional organic and inorganic compounds that
may interfer with diffusion and volatilization of methyl esters decreasing the

performance extraction by headspace SPME. To improve the extraction yield, samples

were diluted 3:1 with buffer solution.
We tried to find NSAIDs in tap water and wastewater samples. We did not find
NSAIDs above our detection limit. Samples were fortified with different levels of
NSAIDs. External calibration was used in the evaluation of NSAIDs in tap water and
standard addition calibration was used in wastewater because matrix effects were
observed. Table 2 shows results and a representative chromatogram of wastewater
samples spiked with 1.0 ng/mL of each analyte is depicted in Fig. 4. The recoveries
were greater than 76.0 % and no interfering peaks were observed for the blank samples.

Figure 4

4. Conclusions
A simple and practical GC-MS method in combination with in situ derivatization
headspace SPME for the determination of the non steroidal acidic anti-inflammatory
drugs ibuprofen, flufenamic acid, naproxen, mefenamic acid, tolfenamic acid, and
meclofenamic acid in water samples is presented. Sensitive responses were obtained
using 15 L of DMS, 60 L of TBA-HSO4, a 100 m polydimetilsiloxane fibre, 2.88 g
Na2SO4, 45 min extraction time, pH 6.0 and 65C in combination. Non-equilibrium
conditions were adopted in order to reduce the extraction time. Matrix effects can be
avoided with the use of the standard addition method for quantitation. In view of it
simplicity and sensitivity, the present method is applicable for the quantification of
residues of NSAIDSs studied in tap water and wastewater samples.
Acknowledgements
The authors thank the CONDES-LUZ for providing support for this work.

References

[1] E. Zuccato, S. Castiglioni, R. Fanelli, G. Reitano, R. Bagnati, C. Chiabrando, F.

Pomati, C. Rossetti, D. Calamari, Environ. Sci. & Pollut. Res. 13 (2006) 15.
[2] J. Debska, A. Kot-Wasik, J. Namiesnik, Crit. Rev. Anal. Chem. 34 (2004) 51.
[3] Z. Moldovan, Chemosphere 64 (2006) 1808.
[4] N. Nakada, T. Tanishima, H. Shinohara, K. Kiri, H. Takada, Water Res. 40 (2006)
3297.
[5] A. Tauxe-Wuersch, L. De Alencastro, D. Grandjean, J. Tarradellas, Water Res. 39
(2005) 1761.
[6]

E. Emmanuel, G. Keck, J. Blanchard, P. Vermande, Y. Perrodin, Environ. Int. 30

(2004) 891.
[7]

Z. Yu, S. Peldszus, P. Huck, J. Chromatogr. A 1148 (2007) 65.

[8] W.C. Lin, H.C. Chen, W.H. Ding, J. Chromatogr. A 1065 (2005) 279.
[9] M. Gros, M. Petrovic, D. Barcel, Talanta 70 (2006) 678.
[10] A. Macia, F. Borrull, C. Aguilar, M. Calull, Electrophoresis 25 (2004) 428.
[11] A. Macia, F. Borrull, M. Callul, C. Aguilar, J. Chromatogr. A 1117 (2006) 234.
[12] S. Weigel, R. Kallenborn, H. Huhnerfuss, J. Chromatogr. A 1023 (2004) 183.
[13] T. Henriksen, B. Svensmark, B. Lindhardt, R. Juhler, Chemosphere 44 (2001)
1531.
[14] I. Rodrguez, J. Carpinteiro, J.B. Quintana, A.M. Carro, R.A. Lorenzo, R. Cela, J.
Chromatogr. A 1024 (2004) 1.
[15] M. del Olmo, A. Zafra, B. Suarez, A. Gonzalez-Casado, J. Taoufiki, J.L. Vilchez, J.
Chromatogr. B 817 (2005) 167.
[16] M. Lee, R. Lee, Y. Lin, C. Chen, B. Hwang, Anal. Chem. 70 (1998) 1963.

10

[17] M.I. Catalina , J. Dalluge, R. Vreuls, U. Brinkman, J. Chromatogr. A 877 (2000)

153.
[18] M.N. Sarrin, F.J. Santos, M.T. Galceran, Anal. Chem. 72 (2000) 4865.
[19] J. Ai, Anal. Chem. 68 (1997) 3260.
[20] B. Soulet, A. Tauxe, J. Tarradellas, Intern. J. Environ. Anal. Chem. 82 (2002)
659.
[21] H. Lee, T. Peart, M. Svoboda, J. Chromatogr. A 1094 (2005) 122.
[22] D. Bendz, N. Paxeus, T. Ginn, F. Loge, J. Hazardous Materials 122 (2005) 195.
[23] S. Ollers, H. Singer, P. Fassler, S. Muller, J. Chromatogr. A 911 (2001) 225.
[24] M. La Farr, I. Ferrer, A. Ginebreda, M. Figueras, L. Olivella, L. Tirapu, M.
Vilanova, D. Barcel, J. Chromatogr. A 938 (2001) 187.

11

Table 1
Analytical parameters
IBU

FLU

Analytes*
NAP
MEF

TOL

MEC

296,390

-345,102

Intercept (a)

-152,486 102,452 -117,012 215,416

Slope (b)

2.9x106

2.5x106

1.2x106

1.8x106

1.4x106

1.0x106

Correlation coefficient

0.999

0.996

0.998

0.991

0.997

0.994

Linearity
[1 - R.S.D(b)] (%)
Detection limit (ng/L)

99.3

97.6

98.7

97.9

98.6

98.1

0.3

0.4

2.7

1.3

1.6

2.9

Quantification limit
(ng/L)
Repetibility
(R.S.D.,%), n = 8
0.1 ng/mL

1.0

1.4

9.0

4.4

5.4

9.7

13.2

13.9

14.3

14.2

14.1

11.8

1.3 ng/mL

11.4

10.6

11.3

10.0

12.7

10.9

2.5 ng/mL

11.5

10.7

11.4

10.2

12.6

11.2

3.0 ng/mL

7.9

8.9

9.8

9.6

10.2

11.4

*IBU: ibuprofen, FLU: flufenamic acid, NAP: naproxen, MEF: mefenamic acid, TOL:
tolfenamic acid, MEC: meclofenamic acid.

12

Table 2
Results of assays to check the accuracy of the proposed method for NSAIDs in spiked
top water and wastewater samples.
Analyte
Sample
Spiked
Found*
% Recovery
(ng/mL)
(ng/mL)
Ibuprofen
Tap Water
0.09
0.08 0.01
88.9
3.0
2.83 0.33
94.3
Wastewater
0.5
0.38 0.05
76.0
4.0
3.47 0.48
86.8
Flufenamic
Tap Water
0.09
0.12 0.02
133.3
acid
3.0
2.75 0.37
91.6
Wastewater
0.5
0.57 0.09
114.0
4.0
3.46 0.55
86.5
Naproxen
Tap Water
0.09
0.09 0.01
100.0
3.0
3.08 0.41
102.7
Wastewater
0.5
0.48 0.06
96.0
4.0
3.69 0.70
92.3
Mefenamic
Tap Water
0.09
0.07 0.01
77.8
acid
3.0
2.76 0.40
92.0
Wastewater
0.5
0.58 0.11
116.0
4.0
3.63 0.67
90.8
Tolfenamic
Tap Water
0.09
0.08 0.02
88.8
acid
3.0
2.97 0.39
99.0
Wastewater
0.5
0.56 0.12
112.0
4.0
3.46 0.56
86.5
Meclofena
Tap Water
0.09
0.10 0.02
107.5
mic
acid
3.0
2.76 0.32
92.0
Wastewater
0.5
0.41 0.10
82.0
4.0
3.14 0.59
78.5
* Average value standard deviation of five determinations.

"% Recovery" refers to the NSAIDs concentrations determined rather than the actual
percent of analytes extracted by the SPME analysis.

Figures captions

Figure 1. Effect of sample pH for in situ methylation head space SPME of the six
NSAIDs.

13

Figure 2. Effect of temperature on the extraction of 10 ng/mL of NSAIDs with

polydimetilsiloxane fibre. For ibuprofen, the scale is shown on the right. Concentration
of ibuprofen is 5 ng/mL.
Figure 3. Influence of amount of 7.5x10-4 M TBA-HSO4 added on NSAIDs reaction
yields to the corresponding methyl esters. For ibuprofen, the scale is shown on the right.
Figure 4. Typical chromatogram obtained in the SIM mode of a wastewater sample
spiked with 1.0 ng/mL of each analyte. 1: ibuprofen, 2: flufenamic acid, 3: naproxen, 4:
mefenamic acid, 5: tolfenamic acid, 6: meclofenamic acid.

14

Ibuprofen
Naproxen
Tolfenamic acid

Flufenamic acid
Mefenamic acid
Meclofenamic acid

6,0E+07

Area

4,0E+07

2,0E+07

0,0E+00
3

11

pH

Figure 1

15

Figure 2

16

Flufenamic acid
Mefenamic acid
Meclofenamic acid

1,2E+08

Naproxen
Tolfenamic acid
Ibuprofen

1,5E+07

1,0E+07

4,0E+07

5,0E+06

0,0E+00

0,0E+00
300

Area

8,0E+07

50

100

150

200

250

L of TBA-HSO4 added

Figure 3

17

Figure 4

18