Specificity enhancement by electrospray ionization multistage mass spectrometry--a valuable tool for differentiation and identification of 'V'-type chemical warfare agents.

Research article Received: 28 August 2013

Revised: 15 October 2013

Accepted: 15 October 2013

Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/jms.3298

Specificity enhancement by electrospray ionization multistage mass spectrometry – a valuable tool for differentiation and identification of ‘V’-type chemical warfare agents Avi Weissberg,* Nitzan Tzanani and Shai Dagan The use of chemical warfare agents has become an issue of emerging concern. One of the challenges in analytical monitoring of the extremely toxic ‘V’-type chemical weapons [O-alkyl S-(2-dialkylamino)ethyl alkylphosphonothiolates] is to distinguish and identify compounds of similar structure. MS analysis of these compounds reveals mostly fragment/product ions representing the amine-containing residue. Hence, isomers or derivatives with the same amine residue exhibit similar mass spectral patterns in both classical EI/MS and electrospray ionization-MS, leading to unavoidable ambiguity in the identification of the phosphonate moiety. A set of five ‘V’-type agents, including O-ethyl S-(2-diisopropylamino)ethyl methylphosphonothiolate (VX), O-isobutyl S(2-diethylamino)ethyl methylphosphonothiolate (RVX) and O-ethyl S-(2-diethylamino)ethyl methylphosphonothiolate (VM) were studied by liquid chromatography/electrospray ionization/MS, utilizing a QTRAP mass detector. MS/MS enhanced product ion scans and multistage MS3 experiments were carried out. Based on the results, possible fragmentation pathways were proposed, and a method for the differentiation and identification of structural isomers and derivatives of ‘V’-type chemical warfare agents was obtained. MS/MS enhanced product ion scans at various collision energies provided information-rich spectra, although many of the product ions obtained were at low abundance. Employing MS3 experiments enhanced the selectivity for those low abundance product ions and provided spectra indicative of the different phosphonate groups. Study of the fragmentation pathways, revealing some less expected structures, was carried out and allowed the formulation of mechanistic rules and the determination of sets of ions typical of specific groups, for example, methylphosphonothiolates versus ethylphosphonothiolates. The new group-specific ions elucidated in this work are also useful for screening unknown ‘V’-type agents and related compounds, utilizing precursor ion scan experiments. Copyright © 2013 John Wiley & Sons, Ltd. Additional supporting information may be found in the online version of this article at the publisher’s web site. Keywords: MS3; ‘V’ type isomers; Chemical warfare agents (CWAs); QTRAP; EPI

1340

The possible use of chemical warfare agents (CWAs) is a current concern, not only in view of the obvious military scenario but also in chemical terror events. Among the various CWAs, ‘V’-type agents [O-alkyl S-(2-dialkylamino)ethyl alkylphosphonothiolates] are extremely toxic and may be used for mass destruction. One of the challenges in monitoring these compounds is to distinguish and identify similar ‘V’-type structures. Difficulties in that respect are due to a large number of structures exhibiting similar mass spectral patterns and chromatographic behaviors. The general chemical structure of the O-alkyl S-(2-dialkylamino)ethyl alkylphosphonothiolate ‘V’-type agents is shown in Table 1. The GC-MS under both EI and CI conditions has traditionally been applied for the mass spectrometric characterization of O-ethyl S-(2-diisopropylamino)ethyl methylphosphonothiolate (VX) and related compounds.[1–4] Under EI conditions, the mass spectra acquired for many V-related compounds were remarkably similar.[5,6] EI data generally did not exhibit a molecular ion and were dominated by a base peak corresponding to [CH2N(Alkyl)2]+ and related secondary fragment ions. These ions represented the amine containing residue, while the phosphonate moiety remained unidentified. Molecular weight information was only obtained

J. Mass Spectrom. 2013, 48, 1340–1348

following CI/MS analysis, where fragmentation revealed the amine containing group mass, but still lacked information on the phosphonate group structure. Kireev and coworkers have previously carried out a thorough study of EI/MS behavior of 79 O-alkyl S-(2-dialkylamino)ethyl alkylphosphonothiolate derivatives.[6] Based on their fragmentation study, a set of EI-based fragment ions, typical of certain groups, were established. Liquid chromatography electrospray ionization MS (LC/ESI/MS) techniques are now routine in the analysis and identification of ‘V’-type CWAs and related compounds, mainly because of their superior sensitivity.[7–9] Acquisition of two multiple reaction monitoring (MRM) transitions in an MS–MS experiment, together with the chromatographic retention time (RT), is a widely accepted

* Correspondence to: Avi Weissberg, Analytical Chemistry Department, IIBR, P.O.B. 19, Ness-Ziona, Israel. E-mail: [email protected] Analytical Chemistry Department, Israel Institute for Biological Research (IIBR), P.O.B. 19, Ness-Ziona, Israel

Copyright © 2013 John Wiley & Sons, Ltd.

Distinction between ‘V’-type CWAs by ESI/MS3 Table 1. Structures of O-alkyl-S-(2-dialkylamino)ethyl alkyl phosphonothiolates studied in this work

[M + H] 1 2 3 4 5

VX V2 RVX VM V5

268 268 268 240 338

+

R1 Methyl Ethyl Methyl Methyl Ethyl

R2

R3

Ethyl Methyl Isobutyl Ethyl Tetrahydrofurfuryl

Isopropyl Isopropyl Ethyl Ethyl Isopropyl

J. Mass Spectrom. 2013, 48, 1340–1348

Experimental Chemicals and reagents Water (LC-MS grade), methanol (LC-MS grade) and formic acid (AR 99%) were purchased from Biolab company. Ammonium formate was obtained from Sigma-Aldrich (St. Louis, MO, USA). The VX and its derivatives were prepared according to existing methods.[16] Stock solutions were prepared in methanol (1 mg/mL) and diluted in water to trace concentrations in the range of 1–200 ng/mL.

Instrumentation An LC/QTRAP/MS setup was used

The analytes were separated using an Agilent 1290 highperformance LC system (Palo Alto, CA, USA), which consisted of a 1290 Infinity Binary Pump containing a Jet Weaver V35 Mixer, 1290 Infinity Autosampler and a 1290 Infinity Thermostatted Column Compartment. High-performance LC conditions: Gradient elution was performed on a reverse phase separation column (Gemini C18, 3.0 μM, 150 mm, 2.1 mm ID by Phenomenex, Switzerland) with a flow rate of 0.3 mL/min. The column was maintained at 40 °C in all the experiments. The mobile phase consisted of water containing 1 mM ammonium formate and 0.02% formic acid (solvent A) and MeOH also containing 1 mM ammonium formate (solvent B). The following elution gradient was used: 5%B/95%A to 100%B over 8 min. Mass spectra of the analytes were obtained using an Applied Biosystems 5500 QTRAP LIT quadrupole mass spectrometer (AB Sciex, Foster City, CA, USA) with Analyst software (version 1.6), equipped with a Turbo V ion source operated in the positive ESI mode. The ESI inlet conditions were gas 1, nitrogen (40 psi); gas 2, nitrogen (60 psi); ion spray voltage, 4500 V; ion source temperature, 600 °C; curtain gas, nitrogen (35 psi). Enhanced product ion MS2 experiments – the settings for EPI scans were as follows: the collision gas was set at ‘high’, the collision energy was set between 7 and 70 eV. The fixed LIT fill time was set to 100 ms. All EPI MS/MS spectra were recorded with a collision energy spread of ±10 eV. The MS3 experiments – the optimal settings for MS3 experiments were as follows: Q3 entry barrier 8 V, excitation time 25 ms.


wileyonlinelibrary.com/journal/jms

1341

criterion for a safe confirmation assay.[10] However, attention is scarcely paid to the selectivity of the MRM transitions. For ‘V’-type compounds, distinction between structural isomers may usually not be accomplished, unless further mass spectral information is pursued. In fact, ‘V’-type isomers with the same amine residue will share common ESI/MS/MS product ions. Together with a similar or slightly shifted LC RT, a false–positive might be hypothesized, and no differentiation can therefore be obtained, despite all the fulfilled identification requirements following current guidelines.[11,12] When an MS/MS enhanced product ion (EPI) scan experiment, which usually provides extensive mass spectral information, was performed, only low intensities of product ions indicating the phosphonate moiety were observed, and still two main product ions of the amine group dominated.[8] Thus, the MS/MS spectra of many isomers and derivatives remained practically identical. Bell and coworkers have studied the ESI ion-trap MS (ESI/ITMS) of O-isobutyl S-(2-diethylamino)ethyl methylphosphonothiolate (RVX) and VX.[13] The dissociation pathway that provides information on the phosphonate moiety in RVX was undetectable. Steinborner and coworkers studied the (ESI/ITMS) fragmentation of O,O-diethyl S-[2-(diethylamino)ethyl] phosphorothiolate (Amiton), which has a structure closely related to VX. Because of an intrinsic low mass cutoff, no product ions below m/z 100 were observed, and all the ions were very similar to those of RVX, representing the diethylamine moiety.[14] To the best of our knowledge, the dissociation processes of other ‘V’-type compounds were not reported. A quadrupole linear IT (QqLIT), platform where two quadropoles precede a linear IT mass analyzer, combines the attributes of triple quadropole and IT operation modes. The use of the third quadrupole (Q3) as a LIT with axial ion ejection and larger trapping volume enhances the full scan sensitivity. Because the isolation and fragmentation steps can be performed outside the trap space in the MS/MS mode, the fragmentation patterns are similar to QqQ instruments, without the typical low mass cutoff characteristic of three-dimensional traps.[15]Therefore, QqLIT instruments are highly suitable for sensitive and informative MS3 scan experiments. This paper reports, for the first time, a method incorporating MS/MS EPI scan experiments combined with MS3 experiments for unambiguous verification of structural isomers of ‘V’-type compounds. A fragmentation study of five representative ‘V’-type compounds (1–5 in Table 1) was performed. Those include the well-known VX, RVX,[13,14] O-ethyl S-(2-diethylamino)ethyl

methylphosphonothiolate (VM) and two additional derivatives (V2 and V5). These five compounds have been chosen as they contain various phosphonate groups and amine moieties while exhibiting real-world distinction challenges and thus require new approaches to achieve isomer confirmation. The broad energy range of EPI experiments provided information-rich MS2 spectra, and the MS3 experiments allowed the enhancement of the selectivity for low abundance phosphonate-containing product ions, revealing the phosphonate group structural information. As a consequence, this method enables differentiation between ‘V’-type isomers at sub nanogram/milliliter concentrations. In addition, elucidation of fragmentation pathways was carried out, leading to the selection of appropriate group-specific ions, which enable untargeted screening of ‘V’-type compounds through precursor ion scan experiments.

A. Weissberg, N. Tzanani and S. Dagan

Results and discussion With a view to differentiating structural isomers and similar derivatives, dissociation processes of five model compounds were investigated and are discussed in the succeeding texts. VX and V2 (O-methyl S-(2-diisopropylamino)ethyl ethylphosphonothiolate) These two isomers are composed of an identical amine residue (2-diisopropylamino)ethyl, with different phosphonate groups, where methyl and ethyl are switched between the R1 alkyl group and the OR2 alkoxy group. Distinction between these structural isomers utilizing the sensitive MRM targeted approach is not feasible. The two most intense ions are usually used for MRM transitions, one for quantification and the other one for confirmation of the identity. In this approach, structural specificity is often sacrificed, as is with VX and V2. These isomers share the same two dominant product ions serving for MRM (m/z 268–128, m/z 268–86) and elute at practically the same LC RT (3.17 min for VX and 3.19 min for the O-methyl S-(2-diisopropylamino)ethyl ethylphosphonothiolate). The more informative EPI MS2 spectra of the m/z 268 ion, [VX + H]+ (O-ethyl S-(2-diisopropylamino)ethyl methylphosphonothiolate), and [V2 + H]+ (O-methyl S-(2-diisopropylamino)ethyl ethylphosphonothiolate), are shown in Fig. 1.

1342

Figure 1. Enhanced product ion (EPI) MS/MS spectra of protonated V2 (O-methyl S-(2-diisopropylamino)ethyl ethylphosphonotiolate) (retention time = 3.19 min, collision energy 30 eV ± 10) (a) and VX (retention time = 3.17 min, collision energy 30 eV ± 10) (b). Almost identical spectra are obtained for these isomers.


3

Figure 2. Sequential MS spectra of several product ions of V2 (a) and VX 3 (b) presenting the differences in MS spectra. Both isomers possess a molecular weight of 267 Da.


J. Mass Spectrom. 2013, 48, 1340–1348

Distinction between ‘V’-type CWAs by ESI/MS3 The most abundant product ion in both EPI spectra is at m/z 128 and corresponds to the charge-favorable N,Ndiisopropylaziridinium ion. Further fragmentation of this primary product ion is straightforward with sequential elimination of propylene to yield the second intense product ion at m/z 86.[13] The third most intense product ion at m/z 79 is attributed to the phosphonate group but is common to both spectra. The intensities of other product ions in the spectra were very low, making the differentiation between the isomers based on these product ions impractical. In order to reveal spectral differences between these ‘V’-type isomers, sequential MS3 experiments were performed on each product ion and provided much more information. The different fragmentation pathways are illustrated in Fig. 2. A low intensity product ion at m/z 167, present in both EPI spectra, that corresponds to the phosphonate moiety (elimination of diisopropylamine), was sequentially fragmented by MS3 and resulted in the formation of two highly abundant secondary product ions in the case of VX; the major one at m/z 139 exhibiting loss of ethylene, (see discussion in the next paragraph), and a minor ion at m/z 79. On the other hand, sequential MS3 spectra of the same low abundance m/z 167 product ion from protonated V2 resulted in the formation of four abundant ions at m/z 125, 123, 107 and 79. The m/z 125 ion corresponds to [EtP(OH2)(O)OMe]+ and is presumably a result of water addition to the m/z 107 ion. This pathway of water addition to organophosphates in the gas phase was previously reported.[17] The m/z 123 ion corresponds to [EtP(S)OMe]+, m/z 107 corresponds to [EtP(O)OMe]+ and m/z 79 corresponds to [HP(O) OMe]+. Formation of the m/z 153 product ion that corresponds to a methyl loss from the phosphonate moiety is not observed. This is in agreement with our fragmentation study previously reported,[18] where O-alkyl cleavages were found to be favored only when the alkyl was larger than methyl. The m/z 123 ion corresponds to cleavage of the C2H4O group as illustrated in Scheme 1. MS3 fragmentation of the m/z 139 product ion in VX resulted in the formation of two ions at m/z 95 and m/z 61. The m/z 95 ion corresponds to the elimination of C2H4O, and m/z 61 corresponds to a

protonated thiirane. MS3 fragmentation of the m/z 123 product ion in V2 resulted in the formation of two ions, at m/z 95, which corresponds to loss of C2H4 (P–C bond cleavage) and m/z 63. MS3 fragmentation of the ion at m/z 95 resulted in the formation of the ion at m/z 63, which corresponds to loss of MeOH to produce [PS]+ (analogous to [PO]+ previously proposed).[19] Hence, these two isomers, VX and V2, were distinguished via their different MS3 spectral patterns. Other fragmentation routes similar to VX and V2 are the dissociation of the P–S bond to yield the m/z 107 product ion, which corresponds to [MeP(O)OEt]+ in VX and its isomer [EtP(O)OMe]+ in V2. The m/z 107 ion is fragmented to yield the m/z 79 ion, which corresponds to [MeP(O)OH]+ in VX and its isomer [HP(O) OMe]+ in V2. The entire proposed dissociation processes is illustrated in Scheme 1. In summary, the aforementioned differences between the isomers could be clearly observed only with MS3 experiments, revealing the different fragmentation pathways for VX and V2. Interestingly, for some minor peaks in the EPI spectra, it was hard to determine whether they originated from the ‘V’ analyte or coeluting contaminants. For example, the m/z 136 ion, observed in the EPI (MS/MS) spectrum of V2, was subsequently fragmented in MS3 and produced two ions at m/z 119 and m/z 94, both are not characteristic of any plausible structure of the ‘V’ compounds and originate from a possible trace contamination. Returning to the m/z 139 ion of VX, this ion is the result of an ethylene loss from the m/z 167 ion that could originate from either O–C bond cleavage (ethyl moiety) or S–C bond cleavage (ethylene group attached to the sulfur). This cleavage is reported to arise from the ethyl moiety (C–O bond cleavage).[13] In order to provide more convincing evidence for that hypothesis, MS/MS and MS3 experiments on the hydrolysis product of VX (S-2(diisopropylamino)ethyl methylphosphonothioic acid), in which the OEt group attached to the phosphorus is replaced by an OH group, were carried out. EPI (MS/MS) experiments of the m/z 240-(M + H)+ ion of the hydrolysis product reveal, in addition to the expected amine containing product ions, a product ion at

1343

Scheme 1. Proposed fragmentation pathways for protonated VX and V2.

J. Mass Spectrom. 2013, 48, 1340–1348



A. Weissberg, N. Tzanani and S. Dagan m/z 139 that corresponds to [MeP(O)(OH)(SCH2CH2)]+ (here, the remaining ethylene group has only one possible origin: S–Et bond). This ion was sequentially fragmented in MS3 experiments and resulted in the formation of two product ions at m/z 95 and m/z 61, as we previously observed with VX. Further evidence can be concluded from the fact that in other two ‘V’-type derivatives explored in this study (VM and RVX), the same m/z 139 ion appears, whereas m/z 153 appears for the ethyl-P analogue (V5), with the well-established dissociation pattern producing ions at m/z 95 for methylphosphonothiolates, m/z 109 for ethylphosphonothiolates and m/z 61 for both. In these three cases, O-alkyl modifiers on the phosphorus atom tend to cleave,

1344

Figure 3. Product ion spectra of protonated RVX at collision energies (CE) of 12, 30 and 60 eV with collision energy spread (CES) of ±10 eV.


while the C–S bond is still retained. In addition, S–Et bond remained intact also in V2 in the dissociation process of the m/z 167 ion, which corresponds to [(MeO)EtP(SCH2CH2)]+. In conclusion, no evidence for S–Et cleavage in any of the ‘V’-type derivatives was observed prior to O-alkyl cleavage. Consequently, the ethylene loss in VX, from m/z 167 to m/z 139 product ion, arises from loss of the ethyl moiety (C–O bond cleavage) rather than from the ethylene group attached to the sulfur atom. O-isobutyl S-(2-diethylamino)ethyl methylphosphonothiolate (RVX) Another structural isomer of VX is RVX: (O-isobutyl S-(2diethylamino)ethyl methylphosphonothiolate), [M + H]+ also at m/z 268. RVX is composed of a different amine residue (2-diethylamino)ethyl and a different phosphonate group (O-isobutyl methylphosphonothiolate). In case the amine residue is different, distinction is easier, and the most abundant product ions are at m/z 100, which corresponds to the amine residue N,Ndiethylaziridinium ion and m/z 72, which corresponds to the sequential loss of ethylene. However, in order to determine sets of specific ions for each class of the phosphonate derivative, we further investigated the dissociation pathways of RVX. First, we performed MS/MS EPI scans at different collision energies ranging from 7 to 70 eV. Using a low collision energy (12 eV), the [RVX + H]+ precursor ion undergoes fragmentation to exhibit two high mass product ions: a major ion at m/z 212 and a low abundance ion at m/z 195 as illustrated in Fig. 3. The m/z 212 ion is a result of a cleavage of the larger alkyl group C4H8 from the phosphonate moiety (unlike the more stable C2H4 in VX) as illustrated in Scheme 2. This fragmentation route is apparently different from that observed in VX where elimination of ethylene is not observed. This observation may be justified by the greater acidity of the proton on the tertiary carbon of the isobutyl moiety compared to that of the methyl protons on the ethyl moiety, if a four center mechanism[20] or McLafferty rearrangement[13] are assumed. This alkene cleavage provides vital information on the OR2 substituent of the phosphonate group, making the identification of the phosphonate structure easier, utilizing only EPI (MS2) experiments. MS3 fragmentation of the m/z 212 product ion produced two ions (m/z 134, 100). The ion at m/z 134 arises from the loss of CH3PO2 (P–S bond cleavage), and the ion at m/z 100 arises from C–S bond cleavage to yield the aziridinium

Scheme 2. Proposed fragmentation pathways for protonated RVX.


J. Mass Spectrom. 2013, 48, 1340–1348

Distinction between ‘V’-type CWAs by ESI/MS3 ion (Fig. 4). MS3 fragmentation of the m/z 134 product ion resulted in the formation of m/z 74 ion (loss of CH2CHSH), and MS3 fragmentation of the m/z 100 product ion resulted in the formation of m/z 72 ion (loss of C2H4). These two pathways support the structural elucidation of the amine residue but do not provide any additional information on the phosphonate moiety. A minor fragmentation pathway of the precursor ion, which provides essential information on the phosphonate group, involves the loss of diethylamine to produce a sulphonium ion at m/z 195 in an analogue manner to that seen with VX and compound V2 (m/z 167). Sequential MS3 of the low intensity m/z 195 product ion resulted in the formation of the m/z 139 ion, which manifested the loss of the isobutyl group and corresponds to [MeP(O)OH (SCH2CH2)+. MS3 of the m/z 139 product ion produced two major ions at m/z 95 (which corresponds to [MeP(S)OH]+) and m/z 61, as well as a minor ion at m/z 79 as observed with VX. The whole set of MS3 experiments allowed us to generate informative fourgeneration fragmentation pathway scheme for RVX (Scheme 2). This fragmentation study emphasizes an additional distinction aspect. While nitrogen-containing ions appear at an even mass number (m/z 212, 134, 100, 74, 72), phosphonate product ions (lacking the amine atom) appear at an odd mass number (m/z 195, 139, 95, 79, 61). All the product ions are even electron ions here, (not always the case in ESI fragmentation[21]) and this enables this distinction in ‘V’ agents. O-ethyl S-(2-diethylamino)ethyl methylphosphonothiolate (VM) Another well-known ‘V’-type agent is VM, which is composed of a (2-diethylamino)ethyl group as with RVX and a phosphonate group identical to that in VX. EPI of the [VM + H]+ precursor ion (m/z 240) undergoes fragmentation to exhibit two major product ions at m/z 100 and m/z 72. The m/z 100 product ion corresponds to N,N-diethylaziridinium ion, and an m/z 72 ion corresponds to the sequential loss of ethylene as with RVX. A low intensity product ion at m/z 167, present in EPI spectra, that corresponds to the phosphonate moiety (elimination of diethylamine) was sequentially fragmented by MS3 and resulted in the formation of the same ions we previously observed with VX, (as both contain the same phosphonate group), (see VM fragmentation data in Supporting Information, Figs. S1–S3). From the study of the three ‘V’-type agents (VX, RVX and VM), we can assume that the presence of ions at m/z 139, 95 and 79 may be indicative of methylphosphonothiolates in case the substituent at R2 is larger than methyl. V5-(O-tetrahydrofurfuryl S-(2-diisopropylamino)ethyl ethylphosphonothiolate)

J. Mass Spectrom. 2013, 48, 1340–1348

3

Figure 4. Sequential MS spectra for several product ions of protonated RVX.



1345

Another structure we studied is composed of the interesting tetrahydrofurfuryl group at the R2 phosphonate substituent. In this case, the simple alkyl group (as with VX, V2, RVX, VM) is replaced by a cyclic ether, whereas the amine residue is composed of a (2-diisopropylamino)ethyl group as with VX. We wondered whether the ether substituent would affect the dissociation process on the phosphonate moiety. The proton on the carbon attached to a cyclic oxygen ether is not acidic because of the inductive effect of the oxygen. Thus, if a four center mechanism[20] or McLafferty rearrangement[13] are assumed, a dominant fragmentation of the precursor ion to lose the tetrahydrofurfuryl group, as observed with isobutyl in RVX, is not favored. EPI spectra of the protonated V5 at different collision energies ranging from 7 to 70 eV, reveal only two major product ions at m/z 128

A. Weissberg, N. Tzanani and S. Dagan

Scheme 3. Proposed fragmentation pathways for protonated V5.

diisopropylamino ethanethiol moiety and corresponds to [EtP(O) O-tetrahydrofurfuryl]+. The third fragmentation pathway involved the formation of a product ion at m/z 153, which corresponds to [(OH)EtP(O)(SCH2CH2)]+. This is an ethyl analogue to the m/z 139 ion, which corresponds to [MeP(O)OHSCH2CH2] observed in VX, RVX and VM. Sequential MS3 of the m/z 153 product ion resulted in two major ions at m/z 109 [EtP(OH)S]+ (the ethyl analogue to m/z 95 ion that corresponds to [MeP(OH)S]+ observed in VX, RVX and VM) and m/z 61, which corresponds to a protonated thiirane (Fig. 6). Sequential MS3 of the m/z 177 product ion produced three major ions; m/z 93 indicating the presence of a [EtPO(OH)]+ ion that is assumed to react rapidly with water to yield the m/z 111 ion [EtPOH(H2O)]+ as suggested by Bell et al., [13,17] and m/z 67, probably related to the furfuryl group. Sequential MS3 of the m/z 111 product ion produced two ions at m/z 93 (water loss) and m/z 65, which corresponds to [HP(O)OH]+. The m/z 93 product ion was fragmented to produce the ion at m/z 65 (loss of ethylene). The proposed dissociation processes are illustrated in Scheme 3. We conclude that the presence of ions at m/z 153, 109 and 93 may be an indicative of ethylphosphonothiolates in case the substituent on the R2 is larger than methyl.

Summary and conclusions

Figure 5. Product ion spectra of protonated V5 (O-tetrahydrofurfuryl S(2-diisopropylamino)ethyl ethylphosphonothiolate) at collision energies (CE) of 10, 30 and 50 eV with collision energy spread (CES) of ±10 eV.

1346

and m/z 86 as observed with VX, which correspond to the amine moiety, whereas other product ions are at low intensities (Fig. 5). In addition to the major product ions, 30eV CID of protonated V5 induces fragmentation to very low abundance product ions at m/z 254, 177 and 153. The m/z 254 ion is a result of a loss of the large tetrahydrofurfuryl group (Scheme 3). This can explained by the four center mechanism or by MaCLafferty rearrangement.[13,20] A similar dissociation pattern was observed with RVX (loss of isobutyl) to yield a high abundance m/z 212 ion, in contrast to the low abundance m/z 254 ion observed in the dissociation process of V5. The m/z 177 product ion is a result of the loss of the


The information derived from routine MS/MS EPI scan experiments proved insufficient for isomer differentiation and structural elucidation of suspected ‘V’ CWA compounds because of poor spectral representation of the phosphonate group and possible co-elution with contaminant ions. The fragmentation study employing multistage tandem MS utilizing the QTRAP technology allowed better structure elucidation. MS3 produced informative spectra with specific structural feature ions characteristic of the phosphonate substructure, enabling clear differentiation and a higher degree of certainty in analyte identification. All the tested compounds followed the ‘nitrogen rule’, not only at the protonated molecule but also with all the ESI product ions, enabling to distinguish between ions of the amine group, (appear at an even mass number) and ions of the phosphonate group, which usually lack the nitrogen atom (appear at an odd mass number). Investigation of the fragmentation pathways enabled us to propose diagnostic ions for O-alkyl S-(2-dialkylamino)ethyl alkylphosphonothiolate derivatives as illustrated in Scheme 4(a) and Scheme 4(b).


J. Mass Spectrom. 2013, 48, 1340–1348

Distinction between ‘V’-type CWAs by ESI/MS3

Scheme 4. Unified specific fragmentation paths and ions proposed for methylphosphonothiolates (a), ethylphosphonothiolates (b) and generalized for alkylphosphonothiloates (c).

The group specific ions illustrated in Scheme 4 may be useful for screening unknown ‘V’-type and related compounds, implementing precursor scan screening strategies. For the general ‘V’ structure R1PO(OR2 > Me)SCH2CH2N(R3)2, one example is m/z 139, indicative for R1 = methyl, whereas m/z 153 is indicative for R1 = ethyl, both correspond to R1PO(OR2 > Me)SCH2CH2. Other group specific ions are at m/z 79 for R1 = methyl and m/z 93 for R1 = ethyl that correspond to R2PO(OH). Ions at m/z 95 for R1 = methyl and m/z 109 for R1 = ethyl correspond to R1P(OH)S. Although interpretation to all of the measured m/z values was proposed, some of the suggested structures may benefit from further confirmation by accurate mass determination. Although all the model compounds contain only up to C4 alkyl at the OR2 substituent and C2 alkyl at the R1 substituent, we feel confident to generalize our observation to higher alkyl groupcontaining compounds as illustrated in Scheme 4(c). This generalization is based on ESI/MS/MS fragmentation rules previously derived for organophosphorus esters and similar derivatives; when up to a C8 alkyl group was attached to an oxygen, an alkene cleavage was observed.[18] This observation is in accordance with other published results both in EI[22] and ESI.[23] In summary, the QTRAP technology offers improved sensitivity for MS3 experiments, allowing sub nanogram/milliliter limits for possible distinction between ‘V’-type derivatives leading to improved identification compared to MS2-based methods. The dissociation processes explored here can be extended to other derivatives of ‘V’ CWAs. Acknowledgement The authors are grateful to Dr. Yossi Zafrani for having fruitful discussions and for his useful suggestions.

References Figure 6. Sequential MS spectra of several product ions of protonated V5.

J. Mass Spectrom. 2013, 48, 1340–1348

[1] D. K. Rohrbaugh. Characterization of equimolar VX-water reaction products by gas chromatography–mass spectrometry. J. Chromatogr. A 1998, 809, 131.



1347

3

A. Weissberg, N. Tzanani and S. Dagan [2] G. L. Hook, G. Kimm, D. Koch, P. B. Savage, B. Ding, P. A. Smith. Detection of VX contamination in soil through solid-phase microextraction sampling and gas chromatography/mass spectrometry of the VX degradation product bis(diisopropylaminoethyl)disulfide. J. Chromatogr. A 2003, 992, 1. [3] S. Sass, T. L. Fisher. Chemical ionization and electron impact mass spectrometry of some organophosphonate compounds. Org. Mass. Spectrom. 1979, 14, 257. [4] D. K. Rohrbaugh, F. J. Berg, L. J. Szafraniec, D. I. Rossman, H. D. Durst, S. Munavalli. Synthesis and mass spectral characterization of diisopropylamino-ethanethiol, -sulfides and -disulfides and vinyl sulfides. Phosphorus, Sulfur Silicon Relat. Elem. 1999, 149, 95. [5] P. A. Smith, C. R. J. Lepage, P. B. Savage, C. R. Bowerbank, E. D. Lee, M. J. Lukacs. Use of a hand-portable gas chromatograph-toroidal ion trap mass spectrometer for self-chemical ionization identification of degradation products related to O-ethyl S-(2-diisopropylaminoethyl)methyl phosphonothiolate (VX). Anal. Chim. Acta 2011, 690, 215. [6] A. F. Kireev, I. V. Rybal’chenko, V. I. Savchuk, V. N. Suvorkin, I. A. Tipukhov, B. A. Khamidi. Qualitative chromatographic-mass spectrometric analysis of highly toxic alkyl phosphonate and alkyl thiophosphonate derivatives. J. Anal. Chem. 2002, 57, 842. [7] P. A. D’Agostino, J. R. Hancock, L. R. Provost. Analysis of O-ethyl S-[2(diisopropylamino)ethyl] methylphosphonothiolate (VX) and its degradation products by packed capillary liquid chromatographyelectrospray mass spectrometry. J. Chromatogr. A 1999, 837, 93. [8] S. J. Stubbs, R. W. Read. Liquid chromatography tandem mass spectrometry applied to quantitation of the organophosphorus nerve agent VX in microdialysates from blood probes. J. Chromatogr. B 2010, 878, 1253. [9] F. L. Ciner, C. E. McCord, R. W. Plunkett Jr, M. F. Martin, T. M. Croley. Isotope dilution LC/MS/MS for the detection of nerve agent exposure in urine. J. Chromatogr. B 2007, 846, 42. [10] B. Kmellar, L. Abranko, P. Fodor, S. J. Lehotay. Routine approach to qualitatively screening 300 pesticides and quantification of those frequently detected in fruit and vegetables using liquid chromatography tandem mass spectrometry (LC-MS/MS). Food Addit Contam. A 2010, 27, 1415. [11] C. Baumann, M. A. Cintora, M. Eichler, E. Lifante, M. Cooke, A. Przyborowska, J. M. Halket. A library of atmospheric pressure ionization daughter ion mass spectra based on wideband excitation in an ion trap mass spectrometer. Rapid Commun. Mass Spectrom. 2000, 14, 349. [12] A. Asperger, J. Efer, T. Koal, W. Engewald. On the signal response of various pesticides in electrospray and atmospheric pressure chemical ionization depending on the flow-rate of eluent applied in liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 2001, 65, 937.

[13] A. J. Bell, J. Murrell, C. M. Timperley, P. Watts. Fragmentation and reactions of two isomeric O-alkyl S-(2-dialkylamino)ethyl methylphosphonothiolates studied by electrospray ionization/ion trap mass spectrometry. J. Am. Soc. Mass Spectrom. 2001, 12, 902. [14] S. Ellis-Steinborner, A. Ramachandran, S. J. Blanksby. The fragmentation pathways of protonated Amiton in the gas phase: towards the structural characterisation of organophosphorus chemical warfare agents by electrospray ionisation tandem mass spectrometry. Rapid Commun. Mass Spectrom. 2006, 20, 1939. [15] H. V. Botitsi, S. D. Garbis, A. Economou, D. F. Tsipi. Current mass spectrometry strategies for the analysis of pesticides and their metabolites in food and water matrices. Mass Spectrom. Rev. 2011, 30, 907. [16] T. R. Fukuto, E. M. Stafford. The isomerization of O,O-diethyl O-2diethylaminoethyl phosphorothionate. J. Am. Chem. Soc. 1957, 79, 6083. [17] J. Bell, D. Despeyroux, J. Murrell, P. Watts. Fragmentation and reactions of organophosphate ions produced by electrospray ionization. Int. J. Mass Spectrom. Ion Proc. 1997, 533, 165. [18] A. Weissberg, S. Dagan. Interpretation of ESI(+)-MS/MS spectra – towards the identification of ‘unknowns’. Int. J. Mass. Spectrom. 2011, 299, 158. [19] P. A. D’Agostino, C. L. Chenier. Desorption electrospray ionization mass spectrometric analysis of organophosphorus chemical warfare agents using ion mobility and tandem mass spectrometry. Rappid Commun. Mass Spectrom. 2010, 24, 1617. [20] W. M. A. Niessen. Group-specific fragmentation of pesticides and related compounds in liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 2010, 1217, 4061. [21] E. M. Thurman, I. Ferrer, O. J. Pozo, J. V. Sancho, F. Hernandez. The even-electron rule in electrospray mass spectra of pesticides. Rapid Commun. Mass Spectrom. 2007, 21, 3855. [22] R. Karthikraj, L. Sridhar, S. Prabhakar, N. Prasada Raju, M. R. V. S. Murty, M. Vairamani. Mass spectral characterization of the CWC-related isomeric dialkyl alkylphosphonothiolates/alkylphosphonothionates under gas chromatography/mass spectrometry conditions. Rapid Commun. Mass Spectrom. 2013, 27, 1461. [23] A. Schwarzenberg, F. Ichou, R. B. Cole, X. MacHuron-Mandard, C. Junot, D. Lesage, J. C. Tabet. Identification tree based on fragmentation rules for structure elucidation of organophosphorus esters by electrospray mass spectrometry. J. Mass Spectrom. 2013, 48, 576.

Supporting information Additional supporting information may be found in the online version of this article at the publisher’s web site.

1348 wileyonlinelibrary.com/journal/jms


J. Mass Spectrom. 2013, 48, 1340–1348

Multistage Reactive Transmission-Mode Desorption Electrospray Ionization Mass Spectrometry.

Direct quantification of chemical warfare agents and related compounds at low ppt levels: comparing active capillary dielectric barrier discharge plasma ionization and secondary electrospray ionization mass spectrometry.

Signal and Charge Enhancement for Protein Analysis by Liquid Chromatography-Mass Spectrometry with Desorption Electrospray Ionization.

Electrospray ionization mass spectrometry of platinum anticancer agents.

Electrospray ionization mass spectrometry--a powerful new analytical tool.

Electrospray ionization mass spectrometry.

Chemical warfare agents: II. Nerve agents.

Correction to Sensitive and comprehensive detection of chemical warfare agents in air by atmospheric pressure chemical ionization ion trap tandem mass spectrometry with counterflow introduction.

Systematic optimization of ion-pairing agents and hexafluoroisopropanol for enhanced electrospray ionization mass spectrometry of oligonucleotides.

Rapid screening of N-oxides of chemical warfare agents degradation products by ESI-tandem mass spectrometry.

Separation and identification of oligomeric phenylethoxysiloxanols by liquid chromatography-electrospray ionization mass spectrometry.

Identification of phosphorylation sites in phosphopeptides by positive and negative mode electrospray ionization-tandem mass spectrometry.

Separation and identification of oligomeric ethyl silicates by liquid chromatography with electrospray ionization mass spectrometry.

Nanodiscs and electrospray ionization mass spectrometry: a tool for screening glycolipids against proteins.

Agents of chemical warfare: sulfur mustard.

Identification and quantification of sea lamprey gonadotropin-releasing hormones by electrospray ionization tandem mass spectrometry.

membrane-interfaced electron ionization quadrupole mass spectrometry instrument with Tri-Bed concentrator.

Strategies for differentiation of isobaric flavonoids using liquid chromatography coupled to electrospray ionization mass spectrometry.

Characterization of large, heterogeneous proteins by electrospray ionization-mass spectrometry.

Degradation pathways study of the natriuretic and β-adrenoceptor antagonist tienoxolol using liquid chromatography-electrospray ionization multistage mass spectrometry.

Studies on residue-free decontaminants for chemical warfare agents.

The identification of trimethylsilylated dipeptides with chemical ionization mass spectrometry.

electrospray ionization mass spectrometry for radical identification.

Active hydrogen by chemical ionization mass spectrometry.