- Email: [email protected]
In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides
Accepted Manuscript Title: In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides Authors: J.I. Cacho, N. Campillo, P. Vi˜nas, M. Hern´andez-C´ordoba PII: DOI: Reference:
S0021-9673(17)31857-5 https://doi.org/10.1016/j.chroma.2017.12.059 CHROMA 359115
To appear in:
Journal of Chromatography A
Received date: Revised date: Accepted date:
26-9-2017 19-12-2017 21-12-2017
Please cite this article as: J.I.Cacho, N.Campillo, P.Vi˜nas, M.Hern´andezC´ordoba, In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides, Journal of Chromatography A https://doi.org/10.1016/j.chroma.2017.12.059 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides
IP T
J.I. Cacho, N. Campillo, P. Viñas and M. Hernández-Córdoba
Department of Analytical Chemistry, Faculty of Chemistry, Regional Campus of
SC R
International Excellence “Campus Mare Nostrum” University of Murcia, E-30100 Murcia, Spain
U
*Corresponding author:
N
Prof. Manuel Hernández-Córdoba
A
Department of Analytical Chemistry Faculty of Chemistry
M
University of Murcia
SPAIN Tel.: +34 868887406
PT
FAX: +34 868887682
ED
E-30100 Murcia
CC E
e-mail: [email protected]
A
Possible highlights
Ionic liquids have proven to be suitable extractants for OPPs in waters In situ IL formation facilitates its dispersion through an aqueous phase The whole IL recovered extract is analyzed by GC Microvial insert thermal desorption allows the OPPs injection
Nine organophosphorus pesticides (OPPs) were determined in environmental waters from different origins using in situ ionic liquid dispersive liquid liquid 1
microextraction (IL-DLLME). This preconcentration technique was coupled to gas chromatography-mass spectrometry (GC-MS) using microvial insert thermal desorption, an approach that uses a thermal desorption injector as sample introduction system. The parameters affecting both the microextraction and sample injection steps were optimized. The proposed method showed good
IP T
precision, with RSD values ranging from 4.1 to 9.7%, accuracy with recoveries
SC R
in the 97-113% range, and sensitivity with DLs ranging from 5 to 16 ng L-1.
Keywords: In situ ionic liquid dispersive liquid-liquid microextraction as
N
U
chromatography-mass spectrometry Organophosphorus pesticides Waters
M
A
1. Introduction
Ionic liquids (ILs) have attracted great attention in the field of Analytical
ED
Chemistry in recent years, especially in sample preparation research [1]. Several microextraction techniques have been adapted for use with ILs [2–5].
PT
One such technique is in situ ionic liquid dispersive liquid–liquid microextraction
CC E
(IL-DLLME) [6].
To date, such IL-based microextraction techniques have not been
successfully coupled to gas chromatography (GC), since ILs do not meet the
A
volatility requirements for GC samples. Different approaches have been proposed to overcome this limitation, including specific interfaces [7]. One of the simplest ways to achieve this coupling is to use thermal desorption injectors as a sample introduction system [8] in a technique known as microvial insert thermal desorption [9]. 2
In this approach, the IL extracting phase is placed in a glass microvial inside the thermal desorption tube, and the whole assembly is submitted to a temperature programme in the thermal desorption unit (TDU). When the IL is heated, the extracted analytes are vaporized, and a carrier gas impels them to the programmed temperature vaporization injector (PTV), where they are
IP T
focused before entering the chromatographic column. Due to its high boiling point and low vapour pressure, the IL matrix remains in the disposable glass
SC R
microvial after the heating stage [10]. The singular characteristics of ILs have stimulated their application for analytical microextraction purposes. Most of the
procedures reported use the ILs directly while others generated the reagent in
U
situ by mixing their anionic and cationic components.
N
The use of a thermal desorption injector for the coupling of IL-based
A
microextraction techniques and GC has two main advantages. The first, the
M
whole drop resulting from microextraction (about 20 μL) can be submitted to
ED
thermal desorption, since the glass microvials can hold large volumes. Secondly, the shape of the glass microvial ensures that no IL enters the GC
PT
system. Even if some IL vapours are impelled by the carrier flow, they would be retained in the disposable PTV liner, and not reach the GC column [11].
CC E
The microvial insert thermal desorption approach has been used in this
work for the GC-MS determination of organophosphorus pesticides (OPPs)
A
previously preconcentrated by IL-DLLME from water samples. These compounds constitute one of the main classes of pesticides and have been extensively used in agriculture over the last 40 years [12]. There are more than 200 different OPPs commercially available, representing more than 45% of all registered pesticides [13].
3
These species are slightly soluble in water, with moderate values of partition coefficients (log KO/W ≈ 3) and low vapour pressure. One of the main advantages of these chemicals is their ease of environmental degradation, so they are not bio-accumulated in living organisms [12]. Different degradation pathways, including hydrolysis, photolysis and biodegradation have been
IP T
described for these compounds [14].
OPPs are usually esters, amides or thiols of phosphoric or phosphonic
SC R
acids. All of them contain a phosphorus-oxygen bond that, in some cases, is replaced by a phosphorus-sulphur bond. Their toxicity is related to the inhibition of the acetylcholinesterase enzyme [15], which allows the transmission of nerve
U
impulses. Due to their action mechanism, OPPs are not only toxic for their
N
target pest, but in the case of acute exposure cause weakness or muscle
A
paralysis in humans [16], while chronic exposure may affect neurodevelopment
M
and growth [17].
ED
The extensive use of OPPs, the hazards associated with their use, and restrictive legal regulations [18], underline the need for efficient and sensitive
PT
analytical methods to control these pollutants in water sources [19]. The high sensitivity required for OPP determination makes it necessary to include
CC E
preconcentration techniques in the sample preparation step [20]. Different microextraction techniques, such as solid-phase microextraction [21] or liquid-
A
phase microextraction [22] have been evaluated for this purpose. Microextraction techniques based on ILs have been previously used for
the determination of pesticides in bananas [23], table grapes and plums [24], as well as in wastewater [25] and water samples [26-28]. Of these proposed methods, only two were applied to the determination of OPPs [26,27].
4
Moreover, in all cases, separation was carried out by liquid chromatography (LC) coupled to UV detector [23–28]. Despite the relevant advantages involved in the use of ILs for microextraction stage, at the best of our knowledge, no procedures have been up to date reported for the successful coupling with GC for OPPs determination.
IP T
Therefore, the aim of this work was to optimize in situ IL-DLLME using direct microvial insert thermal desorption and GC-MS for the determination of nine
SC R
organophosphorus pesticides, included in the US EPA 8270 method, in
N
U
environmental water samples.
2.1. Reagents
ED
M
A
2. Experimental
PT
Multi-component solutions were obtained from Sigma (St. Louis, MO,
CC E
USA), containing the following pesticides at 2000 mg L−1 each component in dichloromethane: dimethoate (DM), disulfoton (DS), famphur (FM), parathion (PT), parathion-methyl (MPT), phorate (PR), sulfotep (SF), thionazin (TN) and
A
triethyl thiophosphate (TE). Tributylphosphate (TBP), used as internal standard (IS), was obtained of Sigma-Aldrich Chemie (Germany). Working solutions (200 mg L−1) were prepared by diluting the commercial products in methanol, and kept at −18 ºC in darkness. Sodium chloride was purchased from Fluka-Sigma– Aldrich (Steinheim, Germany). 5
The ILs 1-butyl-3-methylimidazolium chloride ([C4MIm]Cl), 1-hexyl-3methylimidazolium chloride ([C6MIm]Cl), 1-octyl-3-methylimidazolium chloride ([C8MIm]Cl), 1-decyl-3-methylimidazolium chloride ([C10MIm]Cl), 1-dodecyl-3methylimidazolium ([C12MIm]Cl) and litium bis(trifluoromethyl)sulfonylimide
employed for the preparation of 1 M solutions of these ILs.
SC R
2.2. Instrumentation
IP T
(Li[NTf2]) were supplied by IOLITEC (Heilbronn, Germany). Ultrapure water was
For sample introduction purposes a combination of a thermal desorption
U
unit (TDU-2) equipped with a multipurpose autosampler (MPS-2) and a
N
programmed temperature vaporization (PTV) cooled injector system (CIS-4)
A
provided by Gerstel (Mullheim an der Ruhr, Germany) was used. The
M
experimental conditions used for this sample introduction system are shown in
ED
Table 1.
GC analyses were carried out on an Agilent 6890N (Agilent, Waldbronn,
PT
Germany) gas chromatograph coupled to an Agilent 5973 quadrupole mass selective spectrometer equipped with an inert ion source. Working under the
CC E
experimental conditions described in Table 1, the analytes eluted with retention times of between 4.4 and 11.6 min, corresponding to TE and FM, respectively
A
(Table 2). The studied compounds were quantified using the selected ion monitoring (SIM) MS mode, monitoring the most significant ions of each compound (Table 2). The identity of the detected species was confirmed by their chromatographic retention time, the presence of the target and the qualifier
6
ions, as well as the qualifier-to-target ion ratios (±20% relative deviation) for each compound. An ultrasonic processor UP 200 H (Dr. Hielscher, Teltow, Germany), with an effective output of 200 W in liquid media equipped with a titanium sonotrode
Tuttlingen, Germany) was used to disrupt the sample emulsions.
SC R
2.3. Samples and analytical procedure
IP T
(7 mm i.d.), was used for emulsion generation. An EBA 20 centrifuge (Hettich,
Sixteen different samples were obtained from different sites, including
U
rivers, irrigation channels and marshes, all located in south-east Spain. The
N
samples were directly collected from the central area of the surface water
A
source into the containers, amber glass lab vessels, without disturbing the
M
bottom sediment. They were immediately placed in a portable cooler, and latter
ED
stored in a freeze at 4 ºC until analysis, which was carried out 72 hours after collection. No other pretreatment was carried out since they had pH values in
PT
the 6-8 range.
For in situ IL extraction, a 10 mL-aliquot of the sample was placed in a 15
CC E
mL conical bottomed glass tube. The ionic strength of the sample was adjusted by addition of 0.5 g of sodium chloride. Samples were spiked with the IS (TBP)
A
at a concentration of 1 ng L-1. Next, 240 μL of a 1 M Li[NTf2] solution and 200 μL of a 1 M [C4MIm]Cl solution were injected into the sample, leading to the formation of a cloudy dispersion. This emulsion consists of very fine droplets of IL dispersed through the sample solution, from which the OPPs are extracted. The emulsion was broken by centrifuging for 2 min at 3000 rpm. The IL phase
7
settled as a drop in the bottom of the conical bottomed glass tube, and it was recovered using a microsyringe. A 20 μL volume of the IL extract was placed in a 150 µL glass microvial and introduced into a TDU glass desorption tube. The desorption tube was submitted to a heating programme in the TDU. For recovery studies, three water samples, in which the analytes were
IP T
confirmed to be absent, were spiked at two different concentration levels (0.2 and 1 μg L-1) for validation purposes. Three replicates were analyzed in each
U
SC R
case.
A
N
3. Results and discussion
M
3.1. Ionic liquid microextraction procedure
ED
The extraction with ILs usually requires relatively long extraction times, enough to ensure that the equilibrium partition between the aqueous sample
PT
and the extracting phase is reached. This extraction process can be speeded
CC E
up by increasing the contact surface area between both phases, which can be achieved through the formation of emulsions. The in situ generation of the IL, via a metathesis reaction between its anionic and cationic components, leads to
A
the formation of water-insoluble IL nanodroplets dispersed through the sample medium. OPP pesticides are extracted into this IL dispersed phase, and the equilibrium between phases is reached after few seconds. The IL containing the analytes is recovered after a centrifugation step.
8
As a previous stage, different ILs, including [C4MIm][NTf2], [C6MIm][NTf2], [C8MIm][NTf2], [C10MIm][NTf2] and [C12MIm][NTf2], were first assayed as the extraction solvents. For this, 100 μL of the IL were added to the sample, and dispersed with the aid of ultrasounds, until a cloudy solution was obtained. The resulting emulsion was broken by centrifugation, and 20 μL of the recovered IL
IP T
drop was submitted to TD. The results obtained are shown in Fig. 1A. Due to
the relative polar nature of the studied analytes, extraction efficiency was
SC R
significantly lower with long alkyl chain ILs. The most suitable was found to be [C4MIm][NTf2].
After selecting the IL to be used as extracting phase, its in situ generation
U
was studied. A cloudy solution was achieved by adding, to the aqueous sample,
N
a solution containing the cationic part of the IL ([C4MIm]Cl), followed by the
A
addition of the same volume of a solution containing the anionic part (Li[NTf2]).
M
Different volumes of these solutions, ranging from 150 to 250 μL were
ED
considered for this purpose. As shown in Fig. 1B, best responses were obtained when 200 μL of both solutions were used.
PT
The ratio between the added volume of the anionic and cationic parts of the IL may modify the obtained response. The influence of this ratio was
CC E
evaluated in the 0.8-1.2 range, covering both an excess of the anionic and cationic parts. For this purpose, the added volume of the cationic part solution
A
was set to 200 μL, while the added volume of the anionic part solution was varied from 160 to 240 μL. Taking into account the average response of the studied compounds (Fig. 1C), the ratio 1.0:1.2 was finally selected. The possible influence of the order of addition to the sample of each of the IL part was also evaluated and no significant differences were observed.
9
An increase in the ionic strength of the medium may reduce the solubility of the OPPs, and shift their extraction equilibrium from the aqueous to the organic phase. On the other hand, an increase in the salt content of the medium may reduce the solubility of the IL extracting phase, leading to higher IL recovered volumes that, due to a dilution effect, may decrease sensitivity. The
IP T
influence of the ionic strength of the medium was evaluated by the addition of
different amounts of sodium chloride, ranging from 0 to 10% (w/v), to the
SC R
sample. The obtained responses are shown in Fig. 2. In general, an increase in the ionic strength increased the analyte responses, although a high salt concentration led to a decrease in the analyte areas. Therefore, and taking into
U
account the obtained average response, a compromise value corresponding to
ED
3.2 Thermal desorption
M
A
N
5.0% (w/v) NaCl was selected.
In order to introduce the previously extracted OPPs into the GC system,
PT
20 µL of the settled IL phase were placed into a glass microvial using a
CC E
microsyringe, and submitted to thermal desorption. This step may be influenced by desorption temperature, desorption time and the gas flow rate. In general, high desorption temperatures facilitate the vaporization of the
A
analytes, increasing the obtained responses. However, despite the thermal stability of ILs, high temperatures (above 290 ºC) lead to their thermal decomposition, which causes a significant increase in the chromatographic background and may damage analytical instrumentation. Thermal desorption step usually takes some minutes due to the relatively large volume of IL placed 10
in the glass microvial (20 µL) and the small surface of the drop. The desorption temperature should be maintained for a short period of time, which should be long enough to ensure complete vaporization. While the analytes are being thermally desorbed, a helium carrier gas flow impels the vaporized compounds from the heated TDU to the cooled PTV.
IP T
These three parameters were carefully optimized using a central
composite design (α = 1.5, 8 cube points, 6 axial points and 2 central points, in
SC R
duplicate), in the following ranges: 230-290 ºC for the desorption temperature,
1-7 min for the desorption time and 40-100 mL min-1 for the gas flow rate. The effects of these factors on the response obtained for each analyte are shown in
U
Fig. 3. Taking into account the composite response, which summarizes the
N
behaviour of all OPPs, the sensitivity of the method was highest using 242 ºC
A
as desorption temperature, 3.8 min as desorption time and 40 mL min -1 as
M
carrier gas flow rate.
ED
Before they entered the chromatographic column, the vaporized compounds were focused and retained in the PTV. The effect of different filling
PT
materials in the liner of PTV (fiberglass, Tenax, Carbotrap and PDMS foam) as well as a deactivated baffled liner were evaluated for this purpose. The best
CC E
results for all compounds were obtained using the baffled liner (Fig. 4A). Lower temperatures in the PTV increase the retention efficiency and
A
minimize losses of the more volatile compounds. A Peltier unit was used for cooling purposes, and temperatures of 10, 15 and 20 ºC were tested. As expected, analytes responses were higher with lower cooling temperatures (Fig. 4B), especially in the case of the more volatile compounds, such as TE.
11
After the vaporization step, the PTV was heated in order to transfer the compounds trapped in the liner to the chromatographic column. Different heating temperatures, from 270 to 290 ºC, were tested, and no significant differences in analytical performance for OPPs were observed. An intermediate value of 280 ºC was selected as heating temperature. Therefore, a program
IP T
temperature increasing from 10 to 280 ºC at 540 ºC min -1, with a hold time of 2
SC R
min, was applied in the PTV.
U
3.3. Method performance
N
Tributylphosphate (TBP), which shows a relatively similar chemical and
A
chromatographic behaviour to the studied compounds, and was not present in
M
any of the analyzed samples, was evaluated for use as internal standard (IS) to minimize any matrix effect. This led to an enhancement of the repeatability of
ED
the proposed method, as was evaluated in water samples (n=10) spiked with the studied compounds (1 μg L-1). The RSD values decreased from 5.9-16.2%
PT
when no IS was added, to 4.1-9.7% depending on the compound, when IS was
CC E
added (Table 3).
Calibration graphs were obtained for analyte concentrations ranging from
0.1 to 2 μg L-1. Linear regression of the obtained areas for each compound
A
divided by the IS area provided good correlation coefficient values (r > 0.99) in the evaluated range. The slopes of different aqueous calibrations were compared using an ANOVA test. No statistically significant differences were observed among calibrations carried out on three different days, with “p” values in the 0.06-0.73 12
range, or with waters obtained from three different origins, with “p” values in the 0.09-0.83 range. These results confirm the good reproducibility for the proposed method. LODs and quantification limits (LOQs) were calculated taking into account two different criteria: the standard deviation of the y-intercepts of the
IP T
calibration graphs and signal-to-noise ratios. Values within the 10-19 ng L-1 and 33-64 ng L-1 ranges were obtained for LOD and LOQ, respectively, using
SC R
standard deviation of the y-intercept, while values within the 5-16 ng L-1 and 1754 ng L-1 ranges were obtained for LOD and LOQ, respectively, using signal-tonoise ratios (Table 3).
U
The sensitivity attained is enough to use the procedure for the control of
N
OPPs level in water, a quality standard of 0.1 μg L-1 maximum level being
A
established for pesticides [29]. Moreover, the described procedure could be
M
adapted without difficulty to meet a reliable control of the lowest established EU
ED
maximum pesticide residues levels in foods for these organophosphorus compounds (0.01-0.05 mg kg-1) [30]. As shown in Table 4, the sensitivity of the
PT
method is in line with the reported sensitivity of other methods using similar preconcentration methodologies based on microextraction using conventional
CC E
solvents and specific phosphorus detectors [31–35].
A
3.4. Analysis of samples and recovery studies
The optimized procedure was applied to the analysis of sixteen different samples obtained from different watercourses in the south-east of Spain. Only one of the studied compounds, MPT, was present in two of the analyzed
13
samples, at concentrations corresponding to 55±5 ng L-1 and 65±16 ng L-1, respectively (n=3). These two samples were obtained from irrigation drain channels of an area in which the presence of this OPP has been previously reported [36]. Fig. 5 shows a typical chromatogram obtained by in situ IL-DLLME direct
IP T
microvial thermal desorption in combination with GC-MS for a spiked water sample. The analysis of the blank samples confirmed the absence of interfering
SC R
peaks at the analyte retention times, the OPPs being identified by comparing their retention time and mass spectra in samples and standard solutions. Recovery assays, fortifying three different water samples at two
U
concentration levels, corresponding to the pesticides MRL in waters [29], 0.1 μg
N
L-1, and ten times this level, 1 μg L-1, provided recovery values in the 85-118%
PT
4. Conclusion
ED
M
A
range for the lowest level, and in the 97-113% for the highest one.
ILs were seen to be suitable solvents for the extraction and
CC E
preconcentration of OPPs by in situ DLLME, avoiding the use of toxic organic solvents. Direct microvial insert thermal desorption allowed the introduction of
A
the IL extracts directly into the GC system, facilitating the operation, while increasing sensitivity. The analytical method, based on the combination of these two approaches, is useful for the quantification of OPPs in water samples, due to its good sensitivity, precision, accuracy and selectivity.
14
Acknowledgements The authors acknowledge the financial support of the Comunidad Autónoma de la Región de Murcia (CARM, Fundación Séneca, Project 19888/GERM/15), the Spanish MINECO (Project CTQ2015-68049-R) and the
IP T
European Commission (FEDER/ERDF). J.I. Cacho also acknowledges a
A
CC E
PT
ED
M
A
N
U
SC R
fellowship from the University of Murcia.
15
References [1]
R. Liu, J. Liu, Y. Yin, X. Hu, G. Jiang, Ionic liquids in sample preparation, Anal. Bioanal. Chem. 393 (2009) 871–883.
[2]
E. Aguilera-Herrador, R. Lucena, S. Cárdenas, M. Valcárcel, The roles of
IP T
ionic liquids in sorptive microextraction techniques, TrAC Trends Anal. Chem. 29 (2010) 602–616.
V. Vičkačkaitė, A. Padarauskas, Ionic liquids in microextraction
SC R
[3]
techniques, Cent. Eur. J. Chem. 10 (2012) 652–674. [4]
A. Spietelun, Ł. Marcinkowski, M. de la Guardia, J. Namieśnik, Green
M.J. Trujillo-Rodríguez, P. Rocío-Bautista, V. Pino, A.M. Afonso, Ionic
M
[5]
A
N
techniques, Talanta 119 (2014) 34–45.
U
aspects, developments and perspectives of liquid phase microextraction
liquids in dispersive liquid-liquid microextraction, TrAC Trends Anal.
[6]
ED
Chem. 51 (2013) 87–106.
J. López-Darias, V. Pino, J.H. Ayala, A.M. Afonso, In-situ ionic liquid-
PT
dispersive liquid-liquid microextraction method to determine endocrine
CC E
disrupting phenols in seawaters and industrial effluents, Microchim. Acta 174 (2011) 213–222.
A
[7]
[8]
E. Aguilera-Herrador, R. Lucena, S. Cardenas, M. Valcárcel, Direct coupling of ionic liquid based single-drop microextraction and GC/MS, Anal. Chem. 80 (2008) 793–800. A. Chisvert, I.P. Román, L. Vidal, A. Canals, Simple and commercial readily-available approach for the direct use of ionic liquid-based singledrop microextraction prior to gas chromatography, J. Chromatogr. A 1216
16
(2009) 1290–1295. [9]
X. Du, M. Qian, Quantification of 2,5-dimethyl-4-hydroxy-3(2H)-furanone using solid-phase extraction and direct microvial insert thermal desorption gas chromatography-mass spectrometry, J. Chromatogr. A 1208 (2008)
IP T
197–201. [10] J.I. Cacho, N. Campillo, P. Viñas, M. Hernández-Córdoba, Improved sensitivity gas chromatography–mass spectrometry determination of
SC R
parabens in waters using ionic liquids, Talanta 146 (2016) 568–574.
[11] J.I. Cacho, N. Campillo, P. Viñas, M. Hernández-Córdoba, Cloud point
U
extraction and gas chromatography with direct microvial insert thermal
N
desorption for the determination of haloanisoles in alcoholic beverages,
A
Talanta 160 (2016) 282–288.
M
[12] S.O. Pehkonen, Q. Zhang, The degradation of organophosphorus
ED
pesticides in natural waters: a critical review, Crit. Rev. Environ. Sci. Technol. 32 (2002) 17–72.
PT
[13] S. Magdic, A. Boyd-Boland, K. Jinno, J.B. Pawliszyn, Analysis of organophosphorus insecticides from environmental samples using solid-
CC E
phase microextraction, J. Chromatogr. A 736 (1996) 219–228.
[14] K.V.
Ragnarsdottir,
Environmental
fate
and
toxicology
of
A
organophosphate pesticides, J. Geol. Soc. London 157 (2000) 859–876.
[15] C.N. Pope, Organophosphorus pesticides: Do they all have the same mechanism of toxicity?, J. Toxicol. Environ. Heal. Part B 2 (1999) 161– 181. [16] L. 17
Karalliedde,
Organophosphorus
poisoning
and
anaesthesia,
Anaesthesia 54 (1999) 1073–1088. [17] B. Eskenazi, A. Bradman, R. Castorina, Exposures of children to organophosphate pesticides and their potential adverse health effects, Environ. Health Perspect. 107 Suppl (1999) 409–419.
IP T
[18] EU Regulation Nº 1007/2009, Off. J. Eur. Union. L309 (2009) 1–50. [19] M. Tankiewicz, J. Fenik, M. Biziuk, Determination of organophosphorus
SC R
and organonitrogen pesticides in water samples, TrAC - Trends Anal. Chem. 29 (2010) 1050–1063.
[20] M. Tankiewicz, J. Fenik, M. Biziuk, Solventless and solvent-minimized
U
sample preparation techniques for determining currently used pesticides
N
in water samples: a review, Talanta 86 (2011) 8–22.
A
[21] J. Beltran, F.J. López, F. Hernández, Solid-phase microextraction in
M
pesticide residue analysis, J. Chromatogr. A 885 (2000) 389–404.
ED
[22] M.I. Pinto, G. Sontag, R.J. Bernardino, J.P. Noronha, Pesticides in water and the performance
of
the
liquid-phase
microextraction based
PT
techniques. A review, Microchem. J. 96 (2010) 225–237.
CC E
[23] L.M. Ravelo-Pérez, J. Hernández-Borges, M. Asensio-Ramos, M.Á. Rodríguez-Delgado,
Ionic
liquid
based
dispersive
liquid-liquid
microextraction for the extraction of pesticides from bananas, J.
A
Chromatogr. A 1216 (2009) 7336–7345.
[24] L.M. Ravelo-Pérez, J. Hernández-Borges, A.V. Herrera-Herrera, M.Á. Rodriguez-Delgado, Pesticide extraction from table grapes and plums using ionic liquid based dispersive liquid-liquid microextraction, Anal. Bioanal. Chem. 395 (2009) 2387–2395. 18
[25] T. Trtić-Petrović, A. Dimitrijević, Vortex-assisted ionic liquid based liquidliquid microextraction of selected pesticides from a manufacturing wastewater sample, Open Chem. 12 (2014) 98–106. [26] B. Tadesse, E. Teju, A. Gure, N. Megersa, Ionic-liquid-based dispersive liquid-liquid microextraction combined with high-performance liquid
IP T
chromatography for the determination of multiclass pesticide residues in water samples, J. Sep. Sci. 38 (2015) 829–835.
SC R
[27] H.M. Albishri, N.A.M. Aldawsari, D.A. El-Hady, Ultrasound-assisted temperature-controlled ionic liquid dispersive liquid-phase microextraction
U
combined with reversed-phase liquid chromatography for determination of
N
organophosphorus pesticides in water samples, Electrophoresis 37
A
(2016) 2462–2469.
M
[28] C. Yu, S. Zhang, J. Zhang, S. Li, W. Zhou, H. Gao, E. Lu, An in situ ionic liquid dispersive liquid-liquid microextraction method for the detection of
ED
pyrethroids by LC-UV in environmental water samples, J. Brazil. Chem.
PT
Soc. 24 (2013) 1034–1040.
[29] Directive 2006/118/EC, Off. J. Eur. Union. L372 (2006) 19–31.
CC E
[30] EU Regulation Nº 396/2005, Off. J. Eur. Union. L70 (2005) 1–16. [31] Y. Abdollahzadeh, Y. Yamini, A. Jabbari, A. Esrafili, M. Rezaee,
A
Application of ultrasound-assisted emulsification microextraction followed by gas chromatography for determination of organophosphorus pesticides in water and soil samples, Anal. Methods 4 (2012) 830.
[32] Q. Xiao, B. Hu, C. Yu, L. Xia, Z. Jiang, Optimization of a single-drop microextraction procedure for the determination of organophosphorus 19
pesticides in water and fruit juice with gas chromatography-flame photometric detection, Talanta 69 (2006) 848–855. [33] M.R. Khalili-Zanjani, Y. Yamini, N. Yazdanfar, S. Shariati, Extraction and determination of organophosphorus pesticides in water samples by a new
detection, Anal. Chim. Acta 606 (2008) 202–208.
IP T
liquid phase microextraction-gas chromatography-flame photometric
[34] F. Ahmadi, Y. Assadi, S.M.R.M. Hosseini, M. Rezaee, Determination of pesticides
in
water
samples
by
single
SC R
organophosphorus
drop
microextraction and gas chromatography-flame photometric detector, J.
U
Chromatogr. A 1101 (2006) 307–312.
liquid–liquid
microextraction
combined
with
gas
A
Dispersive
N
[35] S. Berijani, Y. Assadi, M. Anbia, M.-R. Milani Hosseini, E. Aghaee,
M
chromatography-flame photometric detection, J. Chromatogr. A 1123
ED
(2006) 1–9.
[36] R. Moreno-González, J.A. Campillo, V.M. León, Influence of an intensive
PT
agricultural drainage basin on the seasonal distribution of organic pollutants in seawater from a Mediterranean coastal lagoon (Mar Menor,
A
CC E
SE Spain), Mar. Pollut. Bull. 77 (2013) 400–411.
20
Figure captions
Fig. 1. Influence of IL nature (A), IL volume (B), and ratio of cationic to anionic
A
CC E
PT
ED
M
A
N
U
SC R
IP T
IL components (C) on OPPs extraction efficiency.
21
Fig. 2. Effect of sodium chloride concentration in the aqueous medium on the
N
U
SC R
IP T
analyte responses.
A
Fig. 3. Effects of thermal desorption parameters (temperature, time and gas
A
CC E
PT
ED
M
flow rate) on response plots.
22
IP T SC R U N A M
ED
Fig. 4. Influence of PTV liner filling material (A) and PTV focusing temperature
A
CC E
PT
(B) on OPPs responses.
23
IP T SC R U N A M
ED
Fig. 5. GC-MS elution profile corresponding to a water sample spiked with the
A
CC E
PT
studied OPPs at a concentration of 0.2 μg L-1.
24
Table 1 Experimental conditions of the TD-GC-MS procedure Thermal Desorption Unit Mode
Splitless
Temperature program
75 – 242 ºC at 360 ºC min-1, held 3.8 min
Desorption flow
40 mL min-1
Solvent Venting
Liner
Empty baffled, deactivated, 2 mm i.d.
Temperature programme
10 – 280 ºC (2 min) at 540 ºC min-1
SC R
Mode
IP T
Cooled Injector System
GC-MS
HP-5MS, 5% diphenyl-95% dimethylpolysiloxane
U
Capillary column
(30 m x 0.25 mm, 0.25 μm) Helium (1 mL min-1)
Oven programme
75 ºC held 0.5 min
A
N
Carrier gas
M
75 – 215 ºC at 20 ºC min-1, held 1.25 min 215 – 240 ºC at 25 ºC min-1,
Transfer line temperature
PT
Quadrupole temperature
ED
240 – 280 ºC at 40 ºC min-1, held 2.25 min
CC E
Ion source temperature
A
Ionization mode
25
300 ºC 150 ºC 230 ºC Electron-impact mode (70 eV)
RT, min
Monitorized ions, m/z
TE
4.4
65, 97 (79), 121 (75), 198 (69)
TN
7.4
97, 96 (92), 107 (72), 143 (46)
IS
7.6
99, 155 (11), 211 (6)
SF
7.8
322, 202 (48), 174 (33), 238 (30)
PR
8.0
75, 121 (25), 93 (12), 260 (7)
DM
8.5
87, 93 (67), 125 (45), 229 (4)
DS
8.8
88, 153 (10), 186 (8), 274 (4),
MPT
9.5
109, 125 (74), 263 (46), 79 (37)
PT
10.1
97, 109 (89), 291 (52), 139 (42)
FM
11.6
218, 125 (41), 93 (40), 109 (19)
N
U
SC R
Compound
A
Underlined numbers correspond to m/z of the target ion.
A
CC E
PT
ED
M
Values into brackets correspond to the qualifier-to-target ion ratios (%)
Table 3 Method characteristics
26
IP T
Table 2 Method characteristics
RSDa, LOQb, LOQc, -1 % ng L ng L-1 TE 7.8 26 43 TN 9.7 17 37 SF 9.5 24 64 PR 5.2 28 33 DM 7.5 39 57 DS 6.2 32 41 MPT 6.7 54 51 PT 4.1 36 54 FM 5.9 36 55 a b b n=10. Calculated for S/N=10. Calculated for Sy-intercept.
SC R
IP T
Compound
U
Table 4 Comparison with other methods based on microextraction for the determination of OPPs
A
CC E
PT
ED
M
A
N
Microextraction Detection LOD, Extraction RSD, % Reference technique system ng L-1 time, min USAEME GC-FID 10 - 100 6.9 - 8.7 0.5 [31] SDME GC-FPD 200 - 600 1.7 - 10.0 20 [32] SDME GC-FPD 10 - 40 3.5 - 8.9 20 [33] SDME GC-FPD 1-5 1.1 - 8.6 40 [34] DLLME GC-FPD 3 - 20 1.2 - 5.6 [35] IL-DLLME GC-MS 5 - 16 4.1 - 9.7 This work USAEME: Ultrasound assisted emulsification microextraction, SDME: Single drop microextraction, FID: Flame ionization detector, FPD: Flame photometric detector
27
S0021-9673(17)31857-5 https://doi.org/10.1016/j.chroma.2017.12.059 CHROMA 359115
To appear in:
Journal of Chromatography A
Received date: Revised date: Accepted date:
26-9-2017 19-12-2017 21-12-2017
Please cite this article as: J.I.Cacho, N.Campillo, P.Vi˜nas, M.Hern´andezC´ordoba, In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides, Journal of Chromatography A https://doi.org/10.1016/j.chroma.2017.12.059 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
In situ ionic liquid dispersive liquid-liquid microextraction coupled to gas chromatography-mass spectrometry for the determination of organophosphorus pesticides
IP T
J.I. Cacho, N. Campillo, P. Viñas and M. Hernández-Córdoba
Department of Analytical Chemistry, Faculty of Chemistry, Regional Campus of
SC R
International Excellence “Campus Mare Nostrum” University of Murcia, E-30100 Murcia, Spain
U
*Corresponding author:
N
Prof. Manuel Hernández-Córdoba
A
Department of Analytical Chemistry Faculty of Chemistry
M
University of Murcia
SPAIN Tel.: +34 868887406
PT
FAX: +34 868887682
ED
E-30100 Murcia
CC E
e-mail: [email protected]
A
Possible highlights
Ionic liquids have proven to be suitable extractants for OPPs in waters In situ IL formation facilitates its dispersion through an aqueous phase The whole IL recovered extract is analyzed by GC Microvial insert thermal desorption allows the OPPs injection
Nine organophosphorus pesticides (OPPs) were determined in environmental waters from different origins using in situ ionic liquid dispersive liquid liquid 1
microextraction (IL-DLLME). This preconcentration technique was coupled to gas chromatography-mass spectrometry (GC-MS) using microvial insert thermal desorption, an approach that uses a thermal desorption injector as sample introduction system. The parameters affecting both the microextraction and sample injection steps were optimized. The proposed method showed good
IP T
precision, with RSD values ranging from 4.1 to 9.7%, accuracy with recoveries
SC R
in the 97-113% range, and sensitivity with DLs ranging from 5 to 16 ng L-1.
Keywords: In situ ionic liquid dispersive liquid-liquid microextraction as
N
U
chromatography-mass spectrometry Organophosphorus pesticides Waters
M
A
1. Introduction
Ionic liquids (ILs) have attracted great attention in the field of Analytical
ED
Chemistry in recent years, especially in sample preparation research [1]. Several microextraction techniques have been adapted for use with ILs [2–5].
PT
One such technique is in situ ionic liquid dispersive liquid–liquid microextraction
CC E
(IL-DLLME) [6].
To date, such IL-based microextraction techniques have not been
successfully coupled to gas chromatography (GC), since ILs do not meet the
A
volatility requirements for GC samples. Different approaches have been proposed to overcome this limitation, including specific interfaces [7]. One of the simplest ways to achieve this coupling is to use thermal desorption injectors as a sample introduction system [8] in a technique known as microvial insert thermal desorption [9]. 2
In this approach, the IL extracting phase is placed in a glass microvial inside the thermal desorption tube, and the whole assembly is submitted to a temperature programme in the thermal desorption unit (TDU). When the IL is heated, the extracted analytes are vaporized, and a carrier gas impels them to the programmed temperature vaporization injector (PTV), where they are
IP T
focused before entering the chromatographic column. Due to its high boiling point and low vapour pressure, the IL matrix remains in the disposable glass
SC R
microvial after the heating stage [10]. The singular characteristics of ILs have stimulated their application for analytical microextraction purposes. Most of the
procedures reported use the ILs directly while others generated the reagent in
U
situ by mixing their anionic and cationic components.
N
The use of a thermal desorption injector for the coupling of IL-based
A
microextraction techniques and GC has two main advantages. The first, the
M
whole drop resulting from microextraction (about 20 μL) can be submitted to
ED
thermal desorption, since the glass microvials can hold large volumes. Secondly, the shape of the glass microvial ensures that no IL enters the GC
PT
system. Even if some IL vapours are impelled by the carrier flow, they would be retained in the disposable PTV liner, and not reach the GC column [11].
CC E
The microvial insert thermal desorption approach has been used in this
work for the GC-MS determination of organophosphorus pesticides (OPPs)
A
previously preconcentrated by IL-DLLME from water samples. These compounds constitute one of the main classes of pesticides and have been extensively used in agriculture over the last 40 years [12]. There are more than 200 different OPPs commercially available, representing more than 45% of all registered pesticides [13].
3
These species are slightly soluble in water, with moderate values of partition coefficients (log KO/W ≈ 3) and low vapour pressure. One of the main advantages of these chemicals is their ease of environmental degradation, so they are not bio-accumulated in living organisms [12]. Different degradation pathways, including hydrolysis, photolysis and biodegradation have been
IP T
described for these compounds [14].
OPPs are usually esters, amides or thiols of phosphoric or phosphonic
SC R
acids. All of them contain a phosphorus-oxygen bond that, in some cases, is replaced by a phosphorus-sulphur bond. Their toxicity is related to the inhibition of the acetylcholinesterase enzyme [15], which allows the transmission of nerve
U
impulses. Due to their action mechanism, OPPs are not only toxic for their
N
target pest, but in the case of acute exposure cause weakness or muscle
A
paralysis in humans [16], while chronic exposure may affect neurodevelopment
M
and growth [17].
ED
The extensive use of OPPs, the hazards associated with their use, and restrictive legal regulations [18], underline the need for efficient and sensitive
PT
analytical methods to control these pollutants in water sources [19]. The high sensitivity required for OPP determination makes it necessary to include
CC E
preconcentration techniques in the sample preparation step [20]. Different microextraction techniques, such as solid-phase microextraction [21] or liquid-
A
phase microextraction [22] have been evaluated for this purpose. Microextraction techniques based on ILs have been previously used for
the determination of pesticides in bananas [23], table grapes and plums [24], as well as in wastewater [25] and water samples [26-28]. Of these proposed methods, only two were applied to the determination of OPPs [26,27].
4
Moreover, in all cases, separation was carried out by liquid chromatography (LC) coupled to UV detector [23–28]. Despite the relevant advantages involved in the use of ILs for microextraction stage, at the best of our knowledge, no procedures have been up to date reported for the successful coupling with GC for OPPs determination.
IP T
Therefore, the aim of this work was to optimize in situ IL-DLLME using direct microvial insert thermal desorption and GC-MS for the determination of nine
SC R
organophosphorus pesticides, included in the US EPA 8270 method, in
N
U
environmental water samples.
2.1. Reagents
ED
M
A
2. Experimental
PT
Multi-component solutions were obtained from Sigma (St. Louis, MO,
CC E
USA), containing the following pesticides at 2000 mg L−1 each component in dichloromethane: dimethoate (DM), disulfoton (DS), famphur (FM), parathion (PT), parathion-methyl (MPT), phorate (PR), sulfotep (SF), thionazin (TN) and
A
triethyl thiophosphate (TE). Tributylphosphate (TBP), used as internal standard (IS), was obtained of Sigma-Aldrich Chemie (Germany). Working solutions (200 mg L−1) were prepared by diluting the commercial products in methanol, and kept at −18 ºC in darkness. Sodium chloride was purchased from Fluka-Sigma– Aldrich (Steinheim, Germany). 5
The ILs 1-butyl-3-methylimidazolium chloride ([C4MIm]Cl), 1-hexyl-3methylimidazolium chloride ([C6MIm]Cl), 1-octyl-3-methylimidazolium chloride ([C8MIm]Cl), 1-decyl-3-methylimidazolium chloride ([C10MIm]Cl), 1-dodecyl-3methylimidazolium ([C12MIm]Cl) and litium bis(trifluoromethyl)sulfonylimide
employed for the preparation of 1 M solutions of these ILs.
SC R
2.2. Instrumentation
IP T
(Li[NTf2]) were supplied by IOLITEC (Heilbronn, Germany). Ultrapure water was
For sample introduction purposes a combination of a thermal desorption
U
unit (TDU-2) equipped with a multipurpose autosampler (MPS-2) and a
N
programmed temperature vaporization (PTV) cooled injector system (CIS-4)
A
provided by Gerstel (Mullheim an der Ruhr, Germany) was used. The
M
experimental conditions used for this sample introduction system are shown in
ED
Table 1.
GC analyses were carried out on an Agilent 6890N (Agilent, Waldbronn,
PT
Germany) gas chromatograph coupled to an Agilent 5973 quadrupole mass selective spectrometer equipped with an inert ion source. Working under the
CC E
experimental conditions described in Table 1, the analytes eluted with retention times of between 4.4 and 11.6 min, corresponding to TE and FM, respectively
A
(Table 2). The studied compounds were quantified using the selected ion monitoring (SIM) MS mode, monitoring the most significant ions of each compound (Table 2). The identity of the detected species was confirmed by their chromatographic retention time, the presence of the target and the qualifier
6
ions, as well as the qualifier-to-target ion ratios (±20% relative deviation) for each compound. An ultrasonic processor UP 200 H (Dr. Hielscher, Teltow, Germany), with an effective output of 200 W in liquid media equipped with a titanium sonotrode
Tuttlingen, Germany) was used to disrupt the sample emulsions.
SC R
2.3. Samples and analytical procedure
IP T
(7 mm i.d.), was used for emulsion generation. An EBA 20 centrifuge (Hettich,
Sixteen different samples were obtained from different sites, including
U
rivers, irrigation channels and marshes, all located in south-east Spain. The
N
samples were directly collected from the central area of the surface water
A
source into the containers, amber glass lab vessels, without disturbing the
M
bottom sediment. They were immediately placed in a portable cooler, and latter
ED
stored in a freeze at 4 ºC until analysis, which was carried out 72 hours after collection. No other pretreatment was carried out since they had pH values in
PT
the 6-8 range.
For in situ IL extraction, a 10 mL-aliquot of the sample was placed in a 15
CC E
mL conical bottomed glass tube. The ionic strength of the sample was adjusted by addition of 0.5 g of sodium chloride. Samples were spiked with the IS (TBP)
A
at a concentration of 1 ng L-1. Next, 240 μL of a 1 M Li[NTf2] solution and 200 μL of a 1 M [C4MIm]Cl solution were injected into the sample, leading to the formation of a cloudy dispersion. This emulsion consists of very fine droplets of IL dispersed through the sample solution, from which the OPPs are extracted. The emulsion was broken by centrifuging for 2 min at 3000 rpm. The IL phase
7
settled as a drop in the bottom of the conical bottomed glass tube, and it was recovered using a microsyringe. A 20 μL volume of the IL extract was placed in a 150 µL glass microvial and introduced into a TDU glass desorption tube. The desorption tube was submitted to a heating programme in the TDU. For recovery studies, three water samples, in which the analytes were
IP T
confirmed to be absent, were spiked at two different concentration levels (0.2 and 1 μg L-1) for validation purposes. Three replicates were analyzed in each
U
SC R
case.
A
N
3. Results and discussion
M
3.1. Ionic liquid microextraction procedure
ED
The extraction with ILs usually requires relatively long extraction times, enough to ensure that the equilibrium partition between the aqueous sample
PT
and the extracting phase is reached. This extraction process can be speeded
CC E
up by increasing the contact surface area between both phases, which can be achieved through the formation of emulsions. The in situ generation of the IL, via a metathesis reaction between its anionic and cationic components, leads to
A
the formation of water-insoluble IL nanodroplets dispersed through the sample medium. OPP pesticides are extracted into this IL dispersed phase, and the equilibrium between phases is reached after few seconds. The IL containing the analytes is recovered after a centrifugation step.
8
As a previous stage, different ILs, including [C4MIm][NTf2], [C6MIm][NTf2], [C8MIm][NTf2], [C10MIm][NTf2] and [C12MIm][NTf2], were first assayed as the extraction solvents. For this, 100 μL of the IL were added to the sample, and dispersed with the aid of ultrasounds, until a cloudy solution was obtained. The resulting emulsion was broken by centrifugation, and 20 μL of the recovered IL
IP T
drop was submitted to TD. The results obtained are shown in Fig. 1A. Due to
the relative polar nature of the studied analytes, extraction efficiency was
SC R
significantly lower with long alkyl chain ILs. The most suitable was found to be [C4MIm][NTf2].
After selecting the IL to be used as extracting phase, its in situ generation
U
was studied. A cloudy solution was achieved by adding, to the aqueous sample,
N
a solution containing the cationic part of the IL ([C4MIm]Cl), followed by the
A
addition of the same volume of a solution containing the anionic part (Li[NTf2]).
M
Different volumes of these solutions, ranging from 150 to 250 μL were
ED
considered for this purpose. As shown in Fig. 1B, best responses were obtained when 200 μL of both solutions were used.
PT
The ratio between the added volume of the anionic and cationic parts of the IL may modify the obtained response. The influence of this ratio was
CC E
evaluated in the 0.8-1.2 range, covering both an excess of the anionic and cationic parts. For this purpose, the added volume of the cationic part solution
A
was set to 200 μL, while the added volume of the anionic part solution was varied from 160 to 240 μL. Taking into account the average response of the studied compounds (Fig. 1C), the ratio 1.0:1.2 was finally selected. The possible influence of the order of addition to the sample of each of the IL part was also evaluated and no significant differences were observed.
9
An increase in the ionic strength of the medium may reduce the solubility of the OPPs, and shift their extraction equilibrium from the aqueous to the organic phase. On the other hand, an increase in the salt content of the medium may reduce the solubility of the IL extracting phase, leading to higher IL recovered volumes that, due to a dilution effect, may decrease sensitivity. The
IP T
influence of the ionic strength of the medium was evaluated by the addition of
different amounts of sodium chloride, ranging from 0 to 10% (w/v), to the
SC R
sample. The obtained responses are shown in Fig. 2. In general, an increase in the ionic strength increased the analyte responses, although a high salt concentration led to a decrease in the analyte areas. Therefore, and taking into
U
account the obtained average response, a compromise value corresponding to
ED
3.2 Thermal desorption
M
A
N
5.0% (w/v) NaCl was selected.
In order to introduce the previously extracted OPPs into the GC system,
PT
20 µL of the settled IL phase were placed into a glass microvial using a
CC E
microsyringe, and submitted to thermal desorption. This step may be influenced by desorption temperature, desorption time and the gas flow rate. In general, high desorption temperatures facilitate the vaporization of the
A
analytes, increasing the obtained responses. However, despite the thermal stability of ILs, high temperatures (above 290 ºC) lead to their thermal decomposition, which causes a significant increase in the chromatographic background and may damage analytical instrumentation. Thermal desorption step usually takes some minutes due to the relatively large volume of IL placed 10
in the glass microvial (20 µL) and the small surface of the drop. The desorption temperature should be maintained for a short period of time, which should be long enough to ensure complete vaporization. While the analytes are being thermally desorbed, a helium carrier gas flow impels the vaporized compounds from the heated TDU to the cooled PTV.
IP T
These three parameters were carefully optimized using a central
composite design (α = 1.5, 8 cube points, 6 axial points and 2 central points, in
SC R
duplicate), in the following ranges: 230-290 ºC for the desorption temperature,
1-7 min for the desorption time and 40-100 mL min-1 for the gas flow rate. The effects of these factors on the response obtained for each analyte are shown in
U
Fig. 3. Taking into account the composite response, which summarizes the
N
behaviour of all OPPs, the sensitivity of the method was highest using 242 ºC
A
as desorption temperature, 3.8 min as desorption time and 40 mL min -1 as
M
carrier gas flow rate.
ED
Before they entered the chromatographic column, the vaporized compounds were focused and retained in the PTV. The effect of different filling
PT
materials in the liner of PTV (fiberglass, Tenax, Carbotrap and PDMS foam) as well as a deactivated baffled liner were evaluated for this purpose. The best
CC E
results for all compounds were obtained using the baffled liner (Fig. 4A). Lower temperatures in the PTV increase the retention efficiency and
A
minimize losses of the more volatile compounds. A Peltier unit was used for cooling purposes, and temperatures of 10, 15 and 20 ºC were tested. As expected, analytes responses were higher with lower cooling temperatures (Fig. 4B), especially in the case of the more volatile compounds, such as TE.
11
After the vaporization step, the PTV was heated in order to transfer the compounds trapped in the liner to the chromatographic column. Different heating temperatures, from 270 to 290 ºC, were tested, and no significant differences in analytical performance for OPPs were observed. An intermediate value of 280 ºC was selected as heating temperature. Therefore, a program
IP T
temperature increasing from 10 to 280 ºC at 540 ºC min -1, with a hold time of 2
SC R
min, was applied in the PTV.
U
3.3. Method performance
N
Tributylphosphate (TBP), which shows a relatively similar chemical and
A
chromatographic behaviour to the studied compounds, and was not present in
M
any of the analyzed samples, was evaluated for use as internal standard (IS) to minimize any matrix effect. This led to an enhancement of the repeatability of
ED
the proposed method, as was evaluated in water samples (n=10) spiked with the studied compounds (1 μg L-1). The RSD values decreased from 5.9-16.2%
PT
when no IS was added, to 4.1-9.7% depending on the compound, when IS was
CC E
added (Table 3).
Calibration graphs were obtained for analyte concentrations ranging from
0.1 to 2 μg L-1. Linear regression of the obtained areas for each compound
A
divided by the IS area provided good correlation coefficient values (r > 0.99) in the evaluated range. The slopes of different aqueous calibrations were compared using an ANOVA test. No statistically significant differences were observed among calibrations carried out on three different days, with “p” values in the 0.06-0.73 12
range, or with waters obtained from three different origins, with “p” values in the 0.09-0.83 range. These results confirm the good reproducibility for the proposed method. LODs and quantification limits (LOQs) were calculated taking into account two different criteria: the standard deviation of the y-intercepts of the
IP T
calibration graphs and signal-to-noise ratios. Values within the 10-19 ng L-1 and 33-64 ng L-1 ranges were obtained for LOD and LOQ, respectively, using
SC R
standard deviation of the y-intercept, while values within the 5-16 ng L-1 and 1754 ng L-1 ranges were obtained for LOD and LOQ, respectively, using signal-tonoise ratios (Table 3).
U
The sensitivity attained is enough to use the procedure for the control of
N
OPPs level in water, a quality standard of 0.1 μg L-1 maximum level being
A
established for pesticides [29]. Moreover, the described procedure could be
M
adapted without difficulty to meet a reliable control of the lowest established EU
ED
maximum pesticide residues levels in foods for these organophosphorus compounds (0.01-0.05 mg kg-1) [30]. As shown in Table 4, the sensitivity of the
PT
method is in line with the reported sensitivity of other methods using similar preconcentration methodologies based on microextraction using conventional
CC E
solvents and specific phosphorus detectors [31–35].
A
3.4. Analysis of samples and recovery studies
The optimized procedure was applied to the analysis of sixteen different samples obtained from different watercourses in the south-east of Spain. Only one of the studied compounds, MPT, was present in two of the analyzed
13
samples, at concentrations corresponding to 55±5 ng L-1 and 65±16 ng L-1, respectively (n=3). These two samples were obtained from irrigation drain channels of an area in which the presence of this OPP has been previously reported [36]. Fig. 5 shows a typical chromatogram obtained by in situ IL-DLLME direct
IP T
microvial thermal desorption in combination with GC-MS for a spiked water sample. The analysis of the blank samples confirmed the absence of interfering
SC R
peaks at the analyte retention times, the OPPs being identified by comparing their retention time and mass spectra in samples and standard solutions. Recovery assays, fortifying three different water samples at two
U
concentration levels, corresponding to the pesticides MRL in waters [29], 0.1 μg
N
L-1, and ten times this level, 1 μg L-1, provided recovery values in the 85-118%
PT
4. Conclusion
ED
M
A
range for the lowest level, and in the 97-113% for the highest one.
ILs were seen to be suitable solvents for the extraction and
CC E
preconcentration of OPPs by in situ DLLME, avoiding the use of toxic organic solvents. Direct microvial insert thermal desorption allowed the introduction of
A
the IL extracts directly into the GC system, facilitating the operation, while increasing sensitivity. The analytical method, based on the combination of these two approaches, is useful for the quantification of OPPs in water samples, due to its good sensitivity, precision, accuracy and selectivity.
14
Acknowledgements The authors acknowledge the financial support of the Comunidad Autónoma de la Región de Murcia (CARM, Fundación Séneca, Project 19888/GERM/15), the Spanish MINECO (Project CTQ2015-68049-R) and the
IP T
European Commission (FEDER/ERDF). J.I. Cacho also acknowledges a
A
CC E
PT
ED
M
A
N
U
SC R
fellowship from the University of Murcia.
15
References [1]
R. Liu, J. Liu, Y. Yin, X. Hu, G. Jiang, Ionic liquids in sample preparation, Anal. Bioanal. Chem. 393 (2009) 871–883.
[2]
E. Aguilera-Herrador, R. Lucena, S. Cárdenas, M. Valcárcel, The roles of
IP T
ionic liquids in sorptive microextraction techniques, TrAC Trends Anal. Chem. 29 (2010) 602–616.
V. Vičkačkaitė, A. Padarauskas, Ionic liquids in microextraction
SC R
[3]
techniques, Cent. Eur. J. Chem. 10 (2012) 652–674. [4]
A. Spietelun, Ł. Marcinkowski, M. de la Guardia, J. Namieśnik, Green
M.J. Trujillo-Rodríguez, P. Rocío-Bautista, V. Pino, A.M. Afonso, Ionic
M
[5]
A
N
techniques, Talanta 119 (2014) 34–45.
U
aspects, developments and perspectives of liquid phase microextraction
liquids in dispersive liquid-liquid microextraction, TrAC Trends Anal.
[6]
ED
Chem. 51 (2013) 87–106.
J. López-Darias, V. Pino, J.H. Ayala, A.M. Afonso, In-situ ionic liquid-
PT
dispersive liquid-liquid microextraction method to determine endocrine
CC E
disrupting phenols in seawaters and industrial effluents, Microchim. Acta 174 (2011) 213–222.
A
[7]
[8]
E. Aguilera-Herrador, R. Lucena, S. Cardenas, M. Valcárcel, Direct coupling of ionic liquid based single-drop microextraction and GC/MS, Anal. Chem. 80 (2008) 793–800. A. Chisvert, I.P. Román, L. Vidal, A. Canals, Simple and commercial readily-available approach for the direct use of ionic liquid-based singledrop microextraction prior to gas chromatography, J. Chromatogr. A 1216
16
(2009) 1290–1295. [9]
X. Du, M. Qian, Quantification of 2,5-dimethyl-4-hydroxy-3(2H)-furanone using solid-phase extraction and direct microvial insert thermal desorption gas chromatography-mass spectrometry, J. Chromatogr. A 1208 (2008)
IP T
197–201. [10] J.I. Cacho, N. Campillo, P. Viñas, M. Hernández-Córdoba, Improved sensitivity gas chromatography–mass spectrometry determination of
SC R
parabens in waters using ionic liquids, Talanta 146 (2016) 568–574.
[11] J.I. Cacho, N. Campillo, P. Viñas, M. Hernández-Córdoba, Cloud point
U
extraction and gas chromatography with direct microvial insert thermal
N
desorption for the determination of haloanisoles in alcoholic beverages,
A
Talanta 160 (2016) 282–288.
M
[12] S.O. Pehkonen, Q. Zhang, The degradation of organophosphorus
ED
pesticides in natural waters: a critical review, Crit. Rev. Environ. Sci. Technol. 32 (2002) 17–72.
PT
[13] S. Magdic, A. Boyd-Boland, K. Jinno, J.B. Pawliszyn, Analysis of organophosphorus insecticides from environmental samples using solid-
CC E
phase microextraction, J. Chromatogr. A 736 (1996) 219–228.
[14] K.V.
Ragnarsdottir,
Environmental
fate
and
toxicology
of
A
organophosphate pesticides, J. Geol. Soc. London 157 (2000) 859–876.
[15] C.N. Pope, Organophosphorus pesticides: Do they all have the same mechanism of toxicity?, J. Toxicol. Environ. Heal. Part B 2 (1999) 161– 181. [16] L. 17
Karalliedde,
Organophosphorus
poisoning
and
anaesthesia,
Anaesthesia 54 (1999) 1073–1088. [17] B. Eskenazi, A. Bradman, R. Castorina, Exposures of children to organophosphate pesticides and their potential adverse health effects, Environ. Health Perspect. 107 Suppl (1999) 409–419.
IP T
[18] EU Regulation Nº 1007/2009, Off. J. Eur. Union. L309 (2009) 1–50. [19] M. Tankiewicz, J. Fenik, M. Biziuk, Determination of organophosphorus
SC R
and organonitrogen pesticides in water samples, TrAC - Trends Anal. Chem. 29 (2010) 1050–1063.
[20] M. Tankiewicz, J. Fenik, M. Biziuk, Solventless and solvent-minimized
U
sample preparation techniques for determining currently used pesticides
N
in water samples: a review, Talanta 86 (2011) 8–22.
A
[21] J. Beltran, F.J. López, F. Hernández, Solid-phase microextraction in
M
pesticide residue analysis, J. Chromatogr. A 885 (2000) 389–404.
ED
[22] M.I. Pinto, G. Sontag, R.J. Bernardino, J.P. Noronha, Pesticides in water and the performance
of
the
liquid-phase
microextraction based
PT
techniques. A review, Microchem. J. 96 (2010) 225–237.
CC E
[23] L.M. Ravelo-Pérez, J. Hernández-Borges, M. Asensio-Ramos, M.Á. Rodríguez-Delgado,
Ionic
liquid
based
dispersive
liquid-liquid
microextraction for the extraction of pesticides from bananas, J.
A
Chromatogr. A 1216 (2009) 7336–7345.
[24] L.M. Ravelo-Pérez, J. Hernández-Borges, A.V. Herrera-Herrera, M.Á. Rodriguez-Delgado, Pesticide extraction from table grapes and plums using ionic liquid based dispersive liquid-liquid microextraction, Anal. Bioanal. Chem. 395 (2009) 2387–2395. 18
[25] T. Trtić-Petrović, A. Dimitrijević, Vortex-assisted ionic liquid based liquidliquid microextraction of selected pesticides from a manufacturing wastewater sample, Open Chem. 12 (2014) 98–106. [26] B. Tadesse, E. Teju, A. Gure, N. Megersa, Ionic-liquid-based dispersive liquid-liquid microextraction combined with high-performance liquid
IP T
chromatography for the determination of multiclass pesticide residues in water samples, J. Sep. Sci. 38 (2015) 829–835.
SC R
[27] H.M. Albishri, N.A.M. Aldawsari, D.A. El-Hady, Ultrasound-assisted temperature-controlled ionic liquid dispersive liquid-phase microextraction
U
combined with reversed-phase liquid chromatography for determination of
N
organophosphorus pesticides in water samples, Electrophoresis 37
A
(2016) 2462–2469.
M
[28] C. Yu, S. Zhang, J. Zhang, S. Li, W. Zhou, H. Gao, E. Lu, An in situ ionic liquid dispersive liquid-liquid microextraction method for the detection of
ED
pyrethroids by LC-UV in environmental water samples, J. Brazil. Chem.
PT
Soc. 24 (2013) 1034–1040.
[29] Directive 2006/118/EC, Off. J. Eur. Union. L372 (2006) 19–31.
CC E
[30] EU Regulation Nº 396/2005, Off. J. Eur. Union. L70 (2005) 1–16. [31] Y. Abdollahzadeh, Y. Yamini, A. Jabbari, A. Esrafili, M. Rezaee,
A
Application of ultrasound-assisted emulsification microextraction followed by gas chromatography for determination of organophosphorus pesticides in water and soil samples, Anal. Methods 4 (2012) 830.
[32] Q. Xiao, B. Hu, C. Yu, L. Xia, Z. Jiang, Optimization of a single-drop microextraction procedure for the determination of organophosphorus 19
pesticides in water and fruit juice with gas chromatography-flame photometric detection, Talanta 69 (2006) 848–855. [33] M.R. Khalili-Zanjani, Y. Yamini, N. Yazdanfar, S. Shariati, Extraction and determination of organophosphorus pesticides in water samples by a new
detection, Anal. Chim. Acta 606 (2008) 202–208.
IP T
liquid phase microextraction-gas chromatography-flame photometric
[34] F. Ahmadi, Y. Assadi, S.M.R.M. Hosseini, M. Rezaee, Determination of pesticides
in
water
samples
by
single
SC R
organophosphorus
drop
microextraction and gas chromatography-flame photometric detector, J.
U
Chromatogr. A 1101 (2006) 307–312.
liquid–liquid
microextraction
combined
with
gas
A
Dispersive
N
[35] S. Berijani, Y. Assadi, M. Anbia, M.-R. Milani Hosseini, E. Aghaee,
M
chromatography-flame photometric detection, J. Chromatogr. A 1123
ED
(2006) 1–9.
[36] R. Moreno-González, J.A. Campillo, V.M. León, Influence of an intensive
PT
agricultural drainage basin on the seasonal distribution of organic pollutants in seawater from a Mediterranean coastal lagoon (Mar Menor,
A
CC E
SE Spain), Mar. Pollut. Bull. 77 (2013) 400–411.
20
Figure captions
Fig. 1. Influence of IL nature (A), IL volume (B), and ratio of cationic to anionic
A
CC E
PT
ED
M
A
N
U
SC R
IP T
IL components (C) on OPPs extraction efficiency.
21
Fig. 2. Effect of sodium chloride concentration in the aqueous medium on the
N
U
SC R
IP T
analyte responses.
A
Fig. 3. Effects of thermal desorption parameters (temperature, time and gas
A
CC E
PT
ED
M
flow rate) on response plots.
22
IP T SC R U N A M
ED
Fig. 4. Influence of PTV liner filling material (A) and PTV focusing temperature
A
CC E
PT
(B) on OPPs responses.
23
IP T SC R U N A M
ED
Fig. 5. GC-MS elution profile corresponding to a water sample spiked with the
A
CC E
PT
studied OPPs at a concentration of 0.2 μg L-1.
24
Table 1 Experimental conditions of the TD-GC-MS procedure Thermal Desorption Unit Mode
Splitless
Temperature program
75 – 242 ºC at 360 ºC min-1, held 3.8 min
Desorption flow
40 mL min-1
Solvent Venting
Liner
Empty baffled, deactivated, 2 mm i.d.
Temperature programme
10 – 280 ºC (2 min) at 540 ºC min-1
SC R
Mode
IP T
Cooled Injector System
GC-MS
HP-5MS, 5% diphenyl-95% dimethylpolysiloxane
U
Capillary column
(30 m x 0.25 mm, 0.25 μm) Helium (1 mL min-1)
Oven programme
75 ºC held 0.5 min
A
N
Carrier gas
M
75 – 215 ºC at 20 ºC min-1, held 1.25 min 215 – 240 ºC at 25 ºC min-1,
Transfer line temperature
PT
Quadrupole temperature
ED
240 – 280 ºC at 40 ºC min-1, held 2.25 min
CC E
Ion source temperature
A
Ionization mode
25
300 ºC 150 ºC 230 ºC Electron-impact mode (70 eV)
RT, min
Monitorized ions, m/z
TE
4.4
65, 97 (79), 121 (75), 198 (69)
TN
7.4
97, 96 (92), 107 (72), 143 (46)
IS
7.6
99, 155 (11), 211 (6)
SF
7.8
322, 202 (48), 174 (33), 238 (30)
PR
8.0
75, 121 (25), 93 (12), 260 (7)
DM
8.5
87, 93 (67), 125 (45), 229 (4)
DS
8.8
88, 153 (10), 186 (8), 274 (4),
MPT
9.5
109, 125 (74), 263 (46), 79 (37)
PT
10.1
97, 109 (89), 291 (52), 139 (42)
FM
11.6
218, 125 (41), 93 (40), 109 (19)
N
U
SC R
Compound
A
Underlined numbers correspond to m/z of the target ion.
A
CC E
PT
ED
M
Values into brackets correspond to the qualifier-to-target ion ratios (%)
Table 3 Method characteristics
26
IP T
Table 2 Method characteristics
RSDa, LOQb, LOQc, -1 % ng L ng L-1 TE 7.8 26 43 TN 9.7 17 37 SF 9.5 24 64 PR 5.2 28 33 DM 7.5 39 57 DS 6.2 32 41 MPT 6.7 54 51 PT 4.1 36 54 FM 5.9 36 55 a b b n=10. Calculated for S/N=10. Calculated for Sy-intercept.
SC R
IP T
Compound
U
Table 4 Comparison with other methods based on microextraction for the determination of OPPs
A
CC E
PT
ED
M
A
N
Microextraction Detection LOD, Extraction RSD, % Reference technique system ng L-1 time, min USAEME GC-FID 10 - 100 6.9 - 8.7 0.5 [31] SDME GC-FPD 200 - 600 1.7 - 10.0 20 [32] SDME GC-FPD 10 - 40 3.5 - 8.9 20 [33] SDME GC-FPD 1-5 1.1 - 8.6 40 [34] DLLME GC-FPD 3 - 20 1.2 - 5.6 [35] IL-DLLME GC-MS 5 - 16 4.1 - 9.7 This work USAEME: Ultrasound assisted emulsification microextraction, SDME: Single drop microextraction, FID: Flame ionization detector, FPD: Flame photometric detector
27