Polymeric Sorbent with Controlled Surface Polarity: An Alternate for

ABSTRACT: Extraction and identification of lethal nerve agents and their ... the forensic & verification view point of the Chemical Weapons Convention...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/ac

Cite This: Anal. Chem. XXXX, XXX, XXX−XXX

Polymeric Sorbent with Controlled Surface Polarity: An Alternate for Solid-Phase Extraction of Nerve Agents and Their Markers from Organic Matrix Kanchan Sinha Roy, Ajay Kumar Purohit, Buddhadeb Chandra, D. Raghavender Goud, Deepak Pardasani, and Devendra Kumar Dubey* VERTOX Laboratory, Defence Research and Development Establishment, Jhansi Road, Gwalior, 474002 Madhya Pradesh, India S Supporting Information *

ABSTRACT: Extraction and identification of lethal nerve agents and their markers in complex organic background have a prime importance from the forensic and verification viewpoint of the Chemical Weapons Convention (CWC). Liquid−liquid extraction with acetonitrile and commercially available solid phase silica cartridges are extensively used for this purpose. Silica cartridges exhibit limited applicability for relatively polar analytes, and acetonitrile extraction shows limited efficacy toward relatively nonpolar analytes. The present study describes the synthesis of polymeric sorbents with tunable surface polarity, their application as a solid-phase extraction (SPE) material against nerve agents and their polar as well as nonpolar markers from nonpolar organic matrices. In comparison with the acetonitrile extraction and commercial silica cartridges, the new sorbent showed better extraction efficiency toward analytes of varying polarity. The extraction parameters were optimized for the proposed method, which included ethyl acetate as an extraction solvent and n-hexane as a washing solvent. Under optimized conditions, method linearity ranged from 0.10 to 10 μg mL−1 (r2 = 0.9327−0.9988) for organophosphorus esters and 0.05−20 μg mL−1 (r2 = 0.9976−0.9991) for nerve agents. Limits of detection (S:N = 3:1) in the SIM mode were found in the range of 0.03−0.075 μg mL−1 for organophosphorus esters and 0.015−0.025 μg mL−1 for nerve agents. Limits of quantification (S:N = 10:1) were found in the range of 0.100−0.25 μg mL−1 for organophosphorus esters and 0.05−0.100 μg mL−1 for nerve agents in the SIM mode. The recoveries of the nerve agents and their markers ranged from 90.0 to 98.0% and 75.0 to 95.0% respectively. The repeatability and reproducibility (with relative standard deviations (RSDs) %) for organophosphorus esters were found in the range of 1.35−8.61% and 2.30−9.25% respectively. For nerve agents, the repeatability range from 1.00 to 7.75% and reproducibility were found in the range of 2.17−6.90%.

N

erve agents such as sarin, cyclosarin, soman, tabun and Vx are considered to be the most powerful and deadliest class of the chemical warfare agents (CWAs).1−6 These CWAs are the most powerful inhibitors of the enzyme acetylcholinesterase, resulting in impaired nerve impulses due to accumulation of neurotransmitter acetylcholine.7−9 These synthetic organophosphate neurotoxins are highly potent, volatile and colorless liquids and have deliberately been used in many incidents related to military and terrorist activities.10−13 Syria and Malaysia are examples of recent events.14,15 To prohibit the development, production, stockpiling, transfer and usage of chemical warfare agents, an international treaty, known as the Chemical Weapon Convention16,17 (CWC), has become operational since 1997. The Organization for Prohibition of Chemical Weapons (OPCW) located at The Hague, The Netherlands, administers the CWC through its strict verification mechanism.18 The verification regime involves unambiguous identification of CWAs and their markers (including precursors, starting compounds and degradation products) in environmental as well as synthetic samples.18,19 © XXXX American Chemical Society

These samples are collected from the production, storage and suspected sites of CW activity20 and they contain complex matrices, such as soil, water and organic liquids. Identification of the aforementioned analytes is highly desired in these matrices.18−20 Organic matrix has high importance from the CWC verification viewpoint because it can be used as a vehicle for the spread of CWAs, or it can be used as a solvent during synthesis. Analysis of the organic sample is also required from the suspected area or production plant. Analysis of organic samples from different sources like engine oil, machine fuel, silicon oil, vacuum pump oil or incineration waste, also gives information about the use or production of convention related compounds (CRCs). But detection and identification of CRCs from such organic samples are very challenging tasks because of Received: March 29, 2018 Accepted: April 26, 2018 Published: April 26, 2018 A

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry the similar chemical nature of analytes and matrix.19,20 In addition, most of the organic samples generally contain high hydrocarbon background. Gas chromatography coupled to mass spectrometry (GC-MS) is the most sought after analytical technique for the retrospective detection and identification of CWAs.21−24 Straight analysis of such organic samples leads to false negative results because the analytes peak gets masked by the co-eluting background peaks.25 Therefore, extraction and sample cleanup are of utmost importance for the analysis of a complex sample. Sample cleanup through an extraction process is the most reliable tool for chromatographic separations and analysis. However, only a few sample preparation methods are reported in the form of the recommended operating procedures26 (ROPs) for the analysis of CRCs in organic samples in the presence of high matrix background.20,25−29 These include concentration, dilution, solvent exchange, liquid−liquid extraction (LLE) and solid phase extraction (SPE). Each type of sample clean up procedure has its own limitations. The applicability of these extraction modes depends on the nature of the matrix and background as well as the type of analytes. Liquid−liquid extraction (LLE) with acetonitrile (ACN) and solid phase extraction (SPE) using a silica based sorbent are the most common methods for the extraction of CWAs from hydrocarbon containing organic matrix.25,28 The ACN extraction shows limited efficacy for relatively non polar analytes and further back extraction or solvent exchange is also required.26 SPE delivers a straightforward and robust means for sample preparation that should be chromatographically more superior than LLE. SPE plays an important role in eliminating background by trapping the analyte on the sorbent used in the SPE column. The elimination of background chemicals is achieved by partitioning the analytes from the background matrix to the sorbent surface. Silica based SPE sorbent (silanol) is generally used for the analysis of CWAs in the presence of a hydrocarbon and nonpolar background.25,26 From silica cartridges, the extraction of the nitrogen-containing basic analytes is difficult due to their strong adsorption on the acidic silica surface. Even in ROP, it is clearly mentioned that polar analytes are not properly extracted from silica cartridges.26 The presence of polar moieties, such as P O, P−CN, P−S and P−F makes nerve agents reactive toward hydroxyl functionality present on the sorbent and extractant, thereby causing reduction in recoveries. Methanol, being a protic solvent, is expected to react with labile bonds, such as P− F, P−CN and P−S. Therefore, the use of methanol in a silica cartridge as an eluting or extracting solvent may result in the hydrolysis of the real agents. Thus, there is a need for the development of new SPE materials that can be used for analytes of varying polarities. Many sorbents are commercially available and also prepared in-house for SPE. They are usually categorized into silica-based, carbon-based and polymeric sorbents.30−34 Among these, polymeric sorbents have received great attention towards the SPE due to their stability and tunability for broad range analytes.30,32 For SPE, polymers with various functional characteristics had been developed.35−37 The partition coefficients of these sorbents either do not favor the transfer of the target analytes to the sorbent bed or there is irreversible sorption.30 This suggests that the use of compoundspecific and class-specific sorbents requires a complex sample containing huge interfering species. The tailored materials were successfully applied for the extraction of various compounds.30−36 In this study, we have

sequentially synthesized four different kinds of polymers and evaluated them as extractant against polar as well as nonpolar CRCs/CWAs present in an organic sample. To circumvent the problems associated with the conventional silica SPE, we selected the polymeric sorbent with a higher surface polarity. The extraction efficiency of the selected polymeric sorbent was compared with the existing silica based SPE. The polymeric sorbent was finally applied for the solid phase extraction of nerve agents and their markers from organic matrix containing high nonpolar background.

■

EXPERIMENTAL SECTION Chemicals. 2,2-Azobis(2-methyl-propionitrile) (AIBN), 2hydroxyethyl methacrylate (HEMA), methacrylic acid (MAA), divinylbenzene (DVB), ethylene glycol dimethacrylate (EGDMA), alumina column and tri-n-butyl phosphate (TBP) (chromatographic standard) were purchased from SigmaAldrich (New Delhi, India). AIBN was recrystallized twice from ethanol and dried in a vacuum oven for 24 h. HEMA, MAA, DVB, and EGDMA were passed through an alumina column and then vacuum distilled before use. HPLC grade solvents n-hexane, dichloromethane, chloroform, ethyl acetate, acetone, acetonitrile, methanol and silicon oil were purchased from E. Merck (Mumbai, India) and used as received. The normal phase SPE silica cartridges (200 mg, 3 mL) were obtained from Agilent Technologies (USA). The empty polypropylene SPE tube (3 mL) with frits was also procured from Sigma-Aldrich (New Delhi, India). The diesel was purchased from a local petrol filling station (Bharat Petroleum, Gwalior, India). CWAs simulants, such as O,O-dimethyl methylphosphonate (DMMP), O,O-diethyl methylphosphonate (DEMP), O,O-diisopropyl methylphosphonate (DIMP), O,O-diethyl ethylphosphonate (DEEP), O,O-diethyl propylphosphonate (DEPP), O,O-dipropyl methylphosphonate (DPMP), O,O-dibutyl methylphosphonate (DBMP), O,Odiethyl-N,N-diethylphosphoramidate (DEDEPA), O,O-dimethyl-N,N-dipropylphosphoramidate (DEDEPA), O,O-dioctyl methylphosphonate (DOMP) and O,O-dicyclohexyl methylphosphonate (DHMP) were synthesized in our laboratory in microgram quantities and purified as per the reported procedures.38,39 The real agents O-isopropyl methylphosphonofluoridate (sarin), O-pinacolyl methylphosphonofluoridate (soman), diisopropyl fluorophosphates (DFP), O-cyclohexyl methylphosphonofluoridate (cyclosarin) and O-ethyl-N,Ndimethylphosphoramidocyanidate (tabun) were used in a declared schedule-1 facility of our institute. Caution. Chemical warfare agents are very toxic in nature. These chemicals should be handled in the presence of trained professionals in an efficient fume hood equipped with an alkali scrubber, and they must wear appropriate protective gear. Decontamination solution (approximately 15% w/v solution of bleach and alkali) must be kept at the workplace, and the solvent waste must be decontaminated in a proper way. Instruments. All the GC-MS analyses were performed in the electron ionization (EI) mode with an Agilent 6890N gas chromatograph equipped with a 5973N mass-selective detector (Agilent Technologies, USA). A DB-5MS (Agilent Technologies) capillary column (30 m × 0.25 mm I.D., 0.25 μm film thickness) was used as the stationary phase. The temperature for the GC oven was programmed from 50 °C (2 min) to 300 °C @ 20 °C min−1 (2 min). Helium was used as the carrier gas at a constant flow rate of 1.0 mL min−1. The samples were analyzed in the splitless mode at an injection temperature of B

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry 250 °C. A 10 μL micro syringe was used to inject the samples, and the injection volume was kept at 1 μL. The EI source was kept at 230 °C; ionization energy was 70 eV, and quadruple temperature was 150 °C. All of the qualitative studies were performed in the full scan mode with a scan range of m/z 35− 450 (3.47 scans per second) and in the selected ion monitoring (SIM) mode, the dwell time 100 ms was used for quantitation studies. Infrared spectroscopy data were collected on a Thermo FTIR Nicolet 6700 (Madison, WI, USA) spectroscope equipped with a deuterated tri glycine sulfate (DTGS) detector between 4000 and 450 cm−1 as KBr pellets. The FT-IR spectrum (Figure S1 in Supporting Information) showed the presence of characteristic functional groups in the polymer. Thermal stability of the polymer was studied by thermo gravimetric analysis (TGA) using Pyris-1 TGA, (make PerkinElmer, USA). P1 and P2 sorbents were separately treated with the heating rate of 20 °C min−1 up to 800 °C under a N2 atmosphere. The thermograms (Figure S2) showed decomposition of P1 and P2 at 230 °C and 260 °C, respectively. Surface characteristics of polymer particles were characterized through BET-surface area, pore volume, and pore sizes. These are summarized in Table 2. The shape and size of the polymer resins were visualized by environmental scanning electron microscopy (ESEM) (FEIQuanta 400 ESEM FEG, FEI, Hillsboro, OR) operated at 10 kV. Images were recorded at 40000× magnification (Figure S3). Synthesis of Materials. In a typical reaction, MAA (20.0 mmol) and EGDMA (40.0 mmol) were dissolved in 25 mL of chloroform for “P1” and 25 mL of acetonitrile for “P2”. The mixture was stirred at room temperature and degassed with nitrogen for 15 min to remove dissolved O2. Polymerization was induced by the addition of AIBN (0.40 mmol) and heating the mixture at 65 °C in a thermostat for 18 h after sealing the reaction vessel. The obtained bulk polymers were grounded manually with mortar. Unreacted monomers, homomers, and oligomers were removed by Soxhlet extraction with methanol over period of 48 h. Finally, polymer powder was dried at 60 °C in a vacuum overnight. Standard and Spiking Solution. Stock solutions of individual analytes (10000 μg mL−1) were prepared in nhexane. A combined stock solution was prepared in n-hexane containing DMMP, DEMP, DIMP, DEEP, DEPP, DPMP, DEDEPA, DMDPPA, DBMP, DHMP and DOMP each at 500 μg mL−1 and was used for the method optimization. Also a combined stock solution for real agents containing sarin, soman, DFP, tabun and cyclosarin was prepared at 500 μg mL−1 in n-hexane. A series of working standards were prepared by properly diluting the combined stock solution with n-hexane in a 10 mL volumetric flask. All of the standard solutions were stored at −4 °C in the dark. A stock of diesel (10000 μg mL−1) was also prepared in n-hexane for the hydrocarbon background. Method optimization was performed in the n-hexane matrix, and the matrix was spiked with a concentration of 3 μg mL−1 for each analyte (nerve agents as well as their markers) using the intermediate stock (500 μg mL−1) of analytes. To mimic the real scenario, the matrix was spiked with 5000 μg mL−1 of hydrocarbon background using the intermediate diesel stock (10000 μg mL−1). Solid Phase Extraction Procedure. Two different types of SPE cartridges were homemade by separately packing a polypropylene SPE tube with 50 mg of polymeric sorbent P1 and P2 particles. The preconditioning of P1, P2 and normal

phase silica SPE cartridges were done with 1 mL of acetone followed by equilibration with 2 mL of n-hexane. Then 1 mL of organic sample spiked with the selected analytes was loaded on all of the preconditioned cartridges separately and passed through under a mild nitrogen flow. All of the cartridges were then washed with n-hexane. Finally the elution was done with suitable solvents. Elutate was concentrated up to 500 μL under a gentle nitrogen stream. The tri-n-butyl phosphate was used as a chromatographic standard and was added (25 μL of stock 100 μg mL−1 solution) to the eluate and 1 μL of that sample was analyzed with GC-MS in the EI mode. Recoveries of all the analytes were calculated by comparing the peak areas of analytes in control and treated samples. Control samples were prepared by adding analytes at concentration of 6 μg mL−1 (maximum possible extraction of analyte) to the blank extracting solvent just before the GC-MS analysis. Thus, matrix matched control samples were prepared similar to treated samples (i.e., preconditioned, loaded, washed, and eluted with extracting solvent) with blank n-hexane containing diesel background that did not contain the analytes. All of the experiments were performed in triplicate runs, and the values shown in the figures are the averages of the triplicate runs.

■

RESULTS AND DISCUSSION Synthesis and Selection of Sorbents. The study was aimed to develop a polymeric sorbent that can overcome

Table 1. Synthesis of Different Polymers with Different Monomers Composition monomers polymer A B C D

a

MAA 1 1 0.5 1

HEMA

EGDMA

0.5

1 2 2

DVB 2 1

a

All of the polymer was synthesized in chloroform using bulk polymerization techniques.

Figure 1. Structures of nerve agents and their environmental markers as well as the CWC-related organophosphorus esters.

limitations of the currently used sorbents. Generally, the hydrophobicity of the polymeric backbone favors the extraction C

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 2. Extraction performances (%) of different polymeric sorbents in chloroform for selected organophosphorus esters.

Figure 4. Extraction performance of (a) commercial silica cartridges and (b) polymeric sorbent P2 for nerve agents.

Figure 5. Recovery (%) of the selected organophosphorus esters with different volume amounts of EtOAc.

combination of methacrylic acid (MAA) and divinylbenzene (DVB) at a molar ratio of 1:2 in CHCl3 (Table 1). The performance of polymer “A” was tested against organophosphorus esters, as shown in Figure 1 (spiked concentration of each analytes was 3 μg mL−1). The results summarized in Figure 2 indicate that the recovery of the polar analytes is less than that of the nonpolar analytes, which could be due to the nonpolar surface of the sorbent “A”. Therefore, to impart surface polarity, different polymeric sorbents “B”, “C” and “D” were synthesized in CHCl3 by varying the functional monomers and cross-linkers (Table 1). The best extraction efficiency for polar analytes was found with the sorbent “D” (Figure 2). This is because of the higher surface polarity of sorbent “D” compared to that of the other sorbents. This surface polarity was imparted by the presence of carboxyl and carbonyl functional groups of MAA and EGDMA. Therefore, sorbent “D” was chosen for further study. In addition to the functional groups of the monomer and cross-linker, the polarity of the solvent used for the synthesis of the polymer, also plays an important role in deciding the surface polarity.40,41 Despite the pore generating ability of the solvent, it also governs the pre-arrangement of the monomers and cross-linkers according to its polarity.42,43 The nonpolar solvents make the polymer surface hydrophobic.42,43 This is because during the synthesis, the polar parts of the monomer and cross-linker are less solvated by the nonpolar solvent. As a result, the relatively polar parts of the monomer and cross-linker remain situated inside

Figure 3. Recovery % of selected organophosphorus esters in different extracting solvents using (a) commercial silica cartridges, (b) polymeric sorbent P1 and (c) polymeric sorbent P2.

Table 2. Surface Areas, Total Pore Volumes, and Average Pore Sizesa characteristics

P1

P2

BET surface area (m2g−1) pore volume (cm3g−1) pore size (nm)

106.55 0.26 9.677

53.49 0.17 12.451

a

Surface properties of the polymeric sorbent P1 and P2 were obtained from the BET studies.

of nonpolar analytes. It becomes more challenging when the background matrix is also hydrophobic. To tame the hydrophobicity of the polymeric sorbent, its surface should be endowed with adequate number of polar groups to make it interactable with both polar as well as nonpolar analytes. The appropriate selection of the polymeric sorbent with balanced surface polarity has been achieved by the proper selection of the monomer, cross-linker and solvent in the synthesis step. For this, we have first synthesized polymeric sorbent “A”, by the D

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

and P2 was compared with those of the commercial solid phase silica cartridges (200 mg) (Figure 3). It is evident from Figure 3 that P1 and P2 sorbents were more efficient than the silica sorbent obtained commercially. Only methanol and to some extent, acetone were able to extract the analytes from silica. Whereas both P1 and P2 performed better with all the types of extracting solvents. A closes look at Figure 3b and Figure 3c reveal that sorbent P2 gave better recoveries of the polar analytes with relatively more polar extractants than those obtained from sorbent P1. It could be attributed to the high surface polarity of P2 over P1, despite having its lower surface area than P1 (Table 2). Therefore, sorbent P2 was selected for further method optimization. Characterization of Poly(MAA-co-EGDMA) Sorbent. Synthesized P1 and P2 sorbents were characterized by the FTIR analysis. FT-IR spectra are given in Figure S1. The similarity of the IR spectra indicates a similar chemical contribution of the polymer matrix of P1 and P2. The stretching vibration of (C O) the carboxylic acid, aliphatic C−H groups and (−C−O− C−) moiety shows the strongest peaks at 1725 cm−1, 2926 cm−1 and 1160 cm−1, respectively. The broad peak at 3445 cm−1 is due to the stretching vibration of the free −OH group of the carboxylic acid. It confirmed the formation of the MAAco-EGDMA backbone for both P1 and P2. The thermal stability of the polymeric sorbents was checked by thermo gravimetric analysis (TGA). The thermograms (Figure S2) showed decomposition temperatures of 230 °C and 260 °C, respectively, for P1 and P2. The SEM images shown in Figure S3 provided the information regarding the particle size and shape of P1 and P2. From the SEM images, it is evident that the sorbents P1 and P2 have homogeneous regular morphology with an average particle diameter of 150−250 nm. The results of the BET surface area analysis are shown in Table 2, indicating a higher surface area of P1 than that of P2. Sorbent P2 exhibits slightly higher porosity than that of P1, with respective pore sizes of 9.667 nm and 12.451 nm. The total pore volume of P1 and P2 was found to be 0.26 cm3g−1 and 0.17 cm3g−1, respectively. Optimizations of Extraction Conditions. To obtain the best enrichment factors from sorbent P2, the governing parameters including the elution solvent, elution solvent volume, washing solvent and its amount were optimized. Considering the extreme toxicity of CWAs, the initial optimizations were done with the nontoxic organophosphorus esters. Optimization of Elution Solvent and its Volume. The elution of retentates is crucial to determine the overall

Figure 6. Total ion chromatogram obtained from GC-EI-MS analyses of the (a) selected analytes spiked in the organic matrix with a diesel background, (b) after washing the diesel background for the organophosphorus esters and (c) after washing the diesel background for the nerve agents.

the bulk of the polymer matrix and make the surface of the polymer hydrophobic. Whereas, the polar solvent strongly solvates the polar functional groups of the monomer and crosslinker. It results in the exposure of the polar portions towards the outer periphery of the polymer.42,43 Therefore, for the fine tuning of the surface polarity of the polymeric sorbent “D”, its synthesis was performed in a more polar solvent (acetonitrile) than CHCl3. The extraction efficiency of both the polymers, which were synthesized in CHCl3 (P1) and in ACN (P2), was checked. The extraction ability of polymers P1

Table 3. Analytical Figures of Merit for Organophosphorus Esters under the Optimized SPE Conditions analytes

linearity range (μg mL−1)

coefficient of determination (r2)

intraday repeatability (RSD %)

interday reproducibility (RSD %)

LOD in SIM (μg mL−1)

LOQ in SIM (μg mL−1)

DMMP DEMP DIMP DEEP DEPP DPMP DEDEPA DMDPPA DBMP DHMP DOMP

0.25−10 0.25−10 0.20−5 0.20−5 0.18−5 0.18−5 0.15−5 0.15−5 0.10−5 0.10−5 0.10−5

0.9959 0.9988 0.9927 0.9327 0.9833 0.9924 0.9632 0.9741 0.9894 0.9889 0.9349

2.05 1.82 2.45 2.02 2.00 2.43 2.16 2.30 2.57 2.98 3.89

2.98 3.83 4.60 3.22 3.14 2.94 2.28 2.03 3.14 4.68 5.63

0.07 0.07 0.06 0.06 0.05 0.05 0.04 0.04 0.03 0.03 0.03

0.250 0.250 0.200 0.200 0.180 0.180 0.150 0.150 0.100 0.100 0.100

E

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Table 4. Analytical Figures of Merits for Nerve Agents under the Optimized SPE Conditions analytes

linearity range (μg mL−1)

coefficient of determination (r2)

intraday repeatability (RSD %)

interday reproducibility (RSD %)

LOD in SIM (μg mL−1)

LOQ in SIM (μg mL−1)

sarin DFP soman tabun cy. sarin

0.05−20 0.100−20 0.100−20 0.125−20 0.100−20

0.9976 0.9991 0.9996 0.9936 0.9981

2.97 1.00 1.50 3.13 2.46

2.72 4.91 2.75 6.28 3.87

0.015 0.025 0.030 0.035 0.025

0.050 0.100 0.100 0.150 0.100

Further experiments were done to optimize the volume of eluting solvent. Different amounts of ethyl acetate (1, 0.5 + 0.5, 1 + 0.5, and 2 mL) were used to study the effect of the eluting solvent volume. Results depicted in Figure 5 show that the highest recoveries (75−98%) were achieved with 1 mL of EtOAc when used in two portions of 0.5 mL of each. No further increase was observed with the higher amount of EtOAc. Therefore, 1 mL (0.5 + 0.5 mL) of EtOAc was fixed as the eluting solvent. Background Elimination. The presence of a complex background (such as diesel fuel) can mask the analytes in the total ion current (TIC) chromatogram of GC-MS analysis, rendering difficulty in their identification (Figure 6a). Therefore, elimination of background chemicals is essential and it is affected by the washing. The nature of the washing solvent and its volume must be such that it causes a minimum or no loss of the desired analytes. A highly complex hydrocarbon background such as diesel fuel, also remains adsorbed on the surface of polymer sorbent due to hydrophobic interactions. To eliminate the hydrocarbon background, n-hexane was considered as the washing solvent due to its affinity with hydrocarbon. Results of washings with 1 mL of n-hexane are depicted in Figure 6. After the washings, sufficient removal of the hydrocarbon background was observed for the organophosphorus esters (Figure 6b) and nerve agents (Figure 6c). Further increase of the washing solvent to 2 mL and 3 mL, did not result in further appreciable removal of the background, hence 1 mL of n-hexane was fixed as the washing solvent. Linearity, Sensitivity, and Reproducibility (LOD, LOQ, and RSD). The linearity range, limits of detection (LODs) and limits of quantification (LOQs) were determined under the optimized conditions. The calibration curves were established with different concentration levels of analytes. The linearity ranges of the method were found in a concentration range of 0.25−10.0 μg mL−1 for organophosphorus esters and 0.05− 20.0 μg mL−1 for nerve agents with good correlation coefficients (r2) between 0.9327 and 0.9991 for all the analytes. The recoveries of the nerve agents and their markers ranged from 90.0 to 98.0% and 75.0 to 95.0% respectively. The LODs and LOQs for the selected analytes were determined based on the signal-to-noise (S:N) ratio of 3:1 and 10:1 respectively. Tables 3 and 4 show the LODs and LOQs of all the analytes. The observed LODs of each analyte were re-evaluated by analysis of the identical spiked concentration in n-hexane. It was

Figure 7. Total ion chromatogram obtained from GC-EI-MS of the 26th OPCW OPT sample 841 after solid phase extraction using the P2 polymeric sorbent under optimized conditions.

recoveries of the analytes. To optimize this parameter, various eluting solvents (dichloromethane, acetone, ethyl acetate, chloroform, acetonitrile, and methanol) were tested to find the best solvent. Results depicted in Figure 3c indicate that ethyl acetate, acetone, acetonitrile, and methanol showed good recoveries of all of the analytes. However, owing to better chromatographic response in GC-MS analysis, ethyl acetate was selected as an eluting solvent. Methanol, being a protic solvent, is likely to react with the labile bonds (P−F, P−CN, and P−S) of real agents, hence it was not preferred as an eluting solvent. When real agents were attempted for elution from commercial silica cartridges, their recoveries were almost nil with DCM and CHCl3 (Figure. 4a). Only acetone and ethyl acetate could recover them. These observations are ascribed to the hydrogen bonding capability of fluorine and nitrogen containing agents with ‘−OH’ group of silica. The DCM and CHCl3 were unable to break the H-bond, whereas ethyl acetate (EtOAc) and acetone, owing to their hydrogen bond acceptor property, could recover nerve agents by disrupting the H-bonds. Results enumerated in Figure 4a further illustrate the fact because the recoveries of nerve agents from silica were relatively inferior with acetone than those obtained with EtOAc (acetone has less H-bond accepting properties than EtOAc). It is important to note that all of the extracting solvents were able to recover the nerve agents efficiently from sorbent P2. Here too, EtOAc showed the best performance among them all, hence it was used for subsequent extractions (Figure 4b).

Table 5. Comparison of the Current Method with Other Reported Methods method a

SE-GC-MS LLE-GC-MSa SPE-GC-MSa present study a

sample

analytes

recovery (%)

n-hexane n-hexane n-hexane n-hexane

CWAs CWAs phosphoramidates nerve agents and organophosphorus esters

80−99 69−85 70−85 75−98

LODsb (μg mL−1)

RSDs (%)

ref

1.0 0.200

2.30−3.31 3.23−4.01 8.2−10.3 1.00−9.28

28 28 25

LLE, Liquid−liquid extraction; SE, solvent exchange; MS, mass spectrometry; GC, gas chromatography; and SPE, solid phase extraction. bFull scan. F

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

analytes of varying polarity indicate applicability of the method in the real world samples.

also observed that the S:N ratio in LOD of each of the analytes was more than 3. The intraday/interday precision and accuracy were determined by relative standard deviation for the selected analytes at three quantification levels in triplicate for 3 consecutive days for 3 weeks. The intraday precision for the organophosphorus esters and for the nerve agents were found in the range of 1.31−8.61% and 1.10−7.75% respectively. The corresponding intraday mean precisions for the organophosphorus esters and nerve agents were in the range of 2.03−9.47% and 2.04−6.90% respectively, which indicates the decent applicability of the methods toward the analysis of the environmental samples. Application of Developed Method. The developed SPE method was used for the extraction of nerve agents and their nontoxic organophosphorus esters from different organic liquids like silicon oil and vacuum pump oil. The samples were spiked at three different concentrations of the selected analytes. Polymeric sorbent P2 was used to extract the analytes under the optimized conditions. The organophosphorus esters were extracted with good recoveries from silicon oil (68.62− 93.45%) with RSD ≤ 8.69% as well as from vacuum pump oil (65.26−90.61%) with RSD ≤ 8.62%. The nerve agents were also extracted with excellent recoveries from silicon oil (90.82− 98.96%) with RSD ≤ 7.20% as well as from vacuum pump oil (87.85−97.63%) with RSD ≤ 8.66% (Tables S1 and S2). The performance of the developed method also was tested with the sample provided by the OPCW in the 26th official proficiency test. This sample was successfully extracted using the polymer sorbent P2 under the optimized extraction conditions. Results of the analysis are presented in Figure 7. The spiked compound VX was clearly detected in this sample. A comparison of the developed method with the reported methods is also presented in Table 5. These data demonstrate that the present method has comparable recoveries and RSDs with other reported methods, with better LODs in the full scan mode.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.8b01428. Experimental section and additional contents as noted in text (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: dkdubey@rediffmail.com; Fax: (+) 0751-2341148. ORCID

Kanchan Sinha Roy: 0000-0001-7726-6788 Devendra Kumar Dubey: 0000-0003-3820-2233 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Kanchan Sinha Roy is thankful to DRDE, Gwalior and DRDO, New Delhi for funding and fellowship. We are thankful to Dr. Meehir Palit for scientific inputs, Dr. Pratibha Pandey, Dr. G.K Prasad and Dr. P.K Gutch for their support for characterization of materials. This article is assigned the DRDE accession no. DRDE/VTX/005/2018.

■

REFERENCES

(1) Brown, M. A.; Brix, K. A. J. Appl. Toxicol. 1998, 18, 393−408. (2) Chauhan, S.; D’cruz, R.; Faruqi, S.; Singh, K. K.; Varma, S.; Singh, M.; Karthik, V. Environ. Toxicol. Pharmacol. 2008, 26, 113−122. (3) Kim, K.; Tsay, O. G.; Atwood, D. A.; Churchill, D. G. Chem. Rev. 2011, 111, 5345−5403. (4) Romano, J. A., Jr; Salem, H.; Lukey, B. J. Chemical warfare agents: chemistry, pharmacology, toxicology, and therapeutics; CRC Press, 2007. (5) Sidell, F. R.; Borak, J. Annals of emergency medicine 1992, 21, 865−871. (6) Sidell, F. R.; Newmark, J.; McDonough, J. H. Textbooks of military medicine, medical aspects of chemical warfare; Department of the Army: United States, 2008; pp 155−219. (7) Ivarsson, U.; Nilsson, H.; Santesson, J. A FOA briefing book on chemical weapons: threat, effects, and protection; Forsvarets Forskningsanstalt: Sundbyberg, Sweden, 1992. (8) Marrs, T. T.; Maynard, R. L.; Sidell, F. Chemical warfare agents: toxicology and treatment; John Wiley & Sons, 2007. (9) Salem, H.; Katz, S. A. Inhalation toxicology; CRC Press, 2014. (10) United Nations Security Council Resolution 2118 series 2013; UN Secuity Council: New York, USA, 2013. (11) Enserink, M.; Science, 2013.341105010.1126/science.341.6150.1050 (12) Munro, N. B.; Watson, A. P.; Ambrose, K. R.; Griffin, G. D. Environ. Health Perspect. 1990, 89, 205−215. (13) Pita, R.; Domingo, J. Toxics 2014, 2, 391−402. (14) BBC NEWS Asia. Kim Jong-nam killing: “VX nerve agent” found on his face, http://www.bbc.com/news/world-asia-39073389, 2017. (15) BBC NEWS Middle East. Syria chemical “attack”: What we know, http://www.bbc.com/news/world-middle-east-39500947, 2017. (16) Convention on the Prohibition of the Development, Stockpiling and Use of chemical Weapons and on their Destruction; Technical Secretariat of The Organization for the Prohibition of Chemical Weapons: The Hague, The Netherlands, 2005. (17) Krutzsch, W.; Trapp, R. A commentary on the chemical weapons convention; Springer, 1994.

■

CONCLUSION This study has resulted in the development of a polymeric sorbent with optimized extraction parameters to extract and enrich the nerve agents and their markers from an organic sample. By varying the proportions of the monomer and crosslinker from 0.5:2 to 1:2, the surface polarity was tuned. The role of porogen was also investigated as a function of the surface polarity and pore size. The optimum composition of the sorbent was found to be MAA as the monomer and EGDMA as the cross-linker in the ratio of 1:2 respectively, using acetonitrile as the porogen. The sorbent was evaluated for the extraction of nerve agents and their polar as well as nonpolar markers from organic samples in the presence of hydrocarbon background. A comparison with commercially available silica was also made. The results indicated a better performance of polymeric sorbent than that of silica in terms of overall recoveries. The recoveries of the nerve agent’s markers ranged from 75.0 to 95.0% by the developed sorbent and 9.0 to 18.0% by silica. Various extraction parameters like ethyl acetate as an eluting solvent, n-hexane as a washing solvent and their amounts were optimized for the best enrichment of the analytes. The LODs were obtained in the range of 0.03−0.075 μg mL−1 for organophosphorus esters and 0.015−0.025 μg mL−1 for nerve agents. These values meet the requirement of analysis of CRCs for verification of CWC. The cleanup and enrichment ability of developed polymeric sorbent toward the G

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (18) Hooijschuur, E. W. J.; Kientz, C. E.; Brinkman, U. A. Journal of Chromatography A 2002, 982, 177−200. (19) Blum, M.-M.; Mamidanna, R. V. S. M. Anal. Bioanal. Chem. 2014, 406, 5067. (20) Mesilaakso, M. Chemical Weapons Convention chemicals analysis: sample collection, preparation and analytical methods; John Wiley & Sons, 2005. (21) D’Agostino, P. A.; Hancock, J. R.; Chenier, C. L. Eur. J. Mass Spectrom. 2003, 9, 609−618. (22) Kientz, C. E. Journal of Chromatography A 1998, 814, 1−23. (23) Smith, P. A.; Koch, D.; Hook, G. L.; Erickson, R. P.; Lepage, C. R. J.; Wyatt, H. D. M.; Betsinger, G.; Eckenrode, B. A. TrAC, Trends Anal. Chem. 2004, 23, 296−306. (24) Terzic, O.; Swahn, I.; Cretu, G.; Palit, M.; Mallard, G. Journal of Chromatography A 2012, 1225, 182−192. (25) Reddy, T. J.; Saradhi, U. V. R. V.; Prabhakar, S.; Vairamani, M. Journal of Chromatography A 2004, 1038, 225−230. (26) Vanninen, P. Recommended Operating Procedures For Analysis In The Verification of Chemical Disarmament; University of Helsinki: Helsinki, 2011. (27) Kuitunen, M. L. Encyclopedia of Analytical Chemistry 2006.10.1002/9780470027318.a0410 (28) Pardasani, D.; Palit, M.; Gupta, A. K.; Shakya, P.; Sekhar, K.; Dubey, D. K. Anal. Chem. 2005, 77, 1172−1176. (29) Pragney, D.; Saradhi, U. V. R. V. TrAC, Trends Anal. Chem. 2012, 37, 73−82. (30) Augusto, F.; Hantao, L. W.; Mogollón, N. G. S.; Braga, S. C. G. N. TrAC, Trends Anal. Chem. 2013, 43, 14−23. (31) Fontanals, N.; Marcé, R. M.; Borrull, F. Journal of Chromatography A 2007, 1152, 14−31. (32) Huck, C. W.; Bonn, G. K. Journal of Chromatography A 2000, 885, 51−72. (33) Lemos, V. A.; Teixeira, L. S. G.; Bezerra, M. d. A.; Costa, A. C. S.; Castro, J. T.; Cardoso, L. A. M.; de Jesus, D. S.; Santos, E. S.; Baliza, P. X.; Santos, L. N. Appl. Spectrosc. Rev. 2008, 43, 303−334. (34) Ravelo-Pérez, L. M.; Herrera-Herrera, A. V.; Hernández-Borges, J.; Rodríguez-Delgado, M. Á . Journal of Chromatography A 2010, 1217, 2618−2641. (35) Fontanals, N.; Marcé, R. M.; Borrull, F. TrAC, Trends Anal. Chem. 2005, 24, 394−406. (36) Płotka-Wasylka, J.; Szczepańska, N.; de la Guardia, M.; Namieśnik, J. TrAC, Trends Anal. Chem. 2016, 77, 23−43. (37) Vidal, L.; Riekkola, M.-L.; Canals, A. Anal. Chim. Acta 2012, 715, 19−41. (38) Barnaby, F.; Humanities Press, New York, 1975. (39) Ledgard, J. A laboratory history of chemical warfare agents; Jared Ledgard, 2006. (40) Chen, J.; Zhuang, H.; Zhao, J.; Gardella, J. A. Surf. Interface Anal. 2001, 31, 713−720. (41) Santora, B. P.; Gagné, M. R.; Moloy, K. G.; Radu, N. S. Macromolecules 2001, 34, 658−661. (42) Horemans, F.; Weustenraed, A.; Spivak, D.; Cleij, T. J. J. Mol. Recognit. 2012, 25, 344−351. (43) Zhang, Y.; Song, D.; Lanni, L. M.; Shimizu, K. D. Macromolecules 2010, 43, 6284−6294.

H

DOI: 10.1021/acs.analchem.8b01428 Anal. Chem. XXXX, XXX, XXX−XXX