Pesticides and Their Degradation Products Including Metabolites: Chromatography-Mass Spectrometry Methods

This chapter reviews the selection of chromatography-mass spectrometry methods for the analysis of organophosphorus pesticides, pyrethroid insecticides, carbamates, and phenylureas. Options with different GC-MS, GC-MS/MS, and LC-MS/MS methods will be discussed for inclusion of the targeted pesticides. In addition, methods for the analysis of metabolites of these chemical classes of pesticides are investigated, including the feasibility of simultaneous analysis with parent pesticides. In some cases, a targeted approach is required for the analyses of metabolites. These methods apply to a wide variety of sample matrices including environmental (air, water, and soil), food (fruits, vegetation, or food products), and biological samples (urine and blood). The focus of the chapter is on MS detection approaches with consideration of the chromatographic separation conditions as required. A short discussion of multiresidue analysis methods and/or where feasible, other chemical classes or selected pesticides from these chemical classes can be analyzed in existing methods is included.


Introduction
Organophosphorus pesticides, pyrethroids, carbamates, and phenylureas remain important chemical classes of pesticides that require chemical analysis by gas chromatography-mass spectrometry (GC-MS), gas chromatography-tandem mass spectrometry (GC-MS/MS), or liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods. The most diverse range of chromatography-mass spectrometry methods is available for these chemical classes of pesticides with method selection often based upon sensitivity and selectivity needs (see Figure 1). The chapter will discuss selection of methods for chemical analysis for each of these chemical classes of pesticides along with the feasibility of separate or simultaneous analysis of metabolites and degradation products of these parent pesticides. The focus of this chapter is on the chromatography-mass spectrometry aspects of the methods. Extraction and clean-up or pre-concentration procedures for the target analytes from sample matrices will also influence the magnitude of matrix enhancement or suppression in the MS detection and column choice (or separation conditions used) to minimize the influence of matrix peaks. Further discussion on sample preparation procedures has been recently reviewed [1,2].
Selection ion monitoring (SIM) with EI does not always meet sensitivity or selectivity needs or provide information on the molecular weight for some OPs due to the high amount of fragmentation in the EI source. OPs are prone to fragmentation in the EI source such that the molecular ion is often too low in abundance to monitor such that fragment ions are used for quantitation and confirmation analysis [3-7, 9, 10]. Positive or negative chemical ionization may be selected to obtain molecular weight confirmation, however, even with negative chemical ionization (NCI) significant amount of fragmentation of OPs may occur in the ion source although typically few fragment ions are observed in NCI as compared to EI [7,10]. Electron capture in NCI can occur by dissociate electron capture and the structure of the OP may lead to more stable negatively charged fragment ions than the molecular ion. PCI is generally not selected for quantitative analysis as it does not provide significant improvements in selectivity over EI, while NCI is used for OPs, organochlorines (OCs), and pyrethroids when additional sensitivity or selectivity is required [7,10]. OPs, organochlorines, and pyrethroids that contain halogen atoms or nitro groups often have lower detection limits with NCI than EI. For example, diazinon and malathion (see structures in Figure 2) have better sensitivity with GC-EI-MS than GC-NCI-MS, while chlorpyrifos-ethyl (chlorinated) and parathion-ethyl (contains a nitro group) have good sensitivity with GC-NCI-MS [7]. The 37 Cl or 81 Br isotopes of the molecular ion or fragment ions can be used for confirmation analysis with GC-EI-MS such as for chlorpyrifos methyl (m/z = 288); however, as there is a high degree of fragmentation of OPs with EI, generally more than two fragment ions of higher abundance than the isotope ions can be selected for quantitation and confirmation [3, 5-7, 9, 10].
Most halogenated OPs observed better sensitivity with GC-NCI-MS than GC-EI-MS or GC-EI-MS/MS [7]. To provide additional selectivity, GC-EI-MS/MS has been used; however, when the molecular ion is selected as the precursor ion for collision-induced dissociation (CID), the sensitivity is lower than when NCI in SIM mode is used [7]. If the OR 1 group is an ethoxy group, CID of the molecular ion may lead to loss of ethene (C 2 H 4 ) from the ethoxy group and if the OP is halogenated, the loss of halogen radical (e.g., Cl radical) is also frequently observed [6]. For example, the SRM 349→286 of chlorpyrifos corresponds to CID of the molecular ion (M +• ) to form fragment ion F + (Cl 2 NC 4 HOPS(OC 2 H 5 )(OH) + ) as a result of loss of C 2 H 2 from an ethoxy group and Cl radical from the aromatic R group. Phorate observes loss of ethyl from the aliphatic R group (SRM: 260→231) to form ( + SCH 2 SPS(OC 2 H 5 ) 2 ) [5,7]. As phorate has an aliphatic R group, fragmentation within the R group can result in a stable fragment ion CH 3 CH 2 SCH 2 + at m/z = 75 (SRM: 260→75). The fragment ion at m/z = 231 can undergo further fragmentation through loss of two molecules of ethene from the two ethoxy groups and neutral loss of SCH 2 to form SPS(OH) 2 + corresponding to ion at m/z=129 (SRM: 231→129 observed). For (RO)PS(OR 1 ) 2 where OR 1 is methoxy, CID of the molecular ion will either form [PS(OR 1 ) 2 ] + with loss of OR radical or a thiono-thiolo rearrangement may occur such that [PO(OR 1 ) 2 ] + is formed with loss of SR radical as observed for fenthion 278→125 and 278→109, respectively [3,5]. Thiono-thiolo rearrangements have been proposed for fragmentation of diazinon in LC-MS/MS [15].
To improve the sensitivity of GC-EI-MS/MS, the precursor ion can be selected as an abundant fragment ion rather than the molecular ion (see Table 1). For bromophos-methyl ( monoisotopic mass 364) and bromophos-ethyl (monoisotopic mass 392), the fragment ions at m/z = 331 and m/z = 359, respectively are selected for precursor ions (SRM 331→286 and 359→303, respectively; see Table 1) and correspond to the either the 37 Cl or 81 Br isotope of [M-Cl] + [3,6]. The R groups of OPs vary substantially and can play a significant role in the fragmentation pathway that dominates. For some OPs, the most abundant fragment ion available for CID is R+. For example, azinphos-methyl and azinphos-ethyl fragmentation at S-R bond of RS(OR 1 ) 2 PS to produce R + and ion at m/z = 160 is the dominant fragment ion formed by loss of the S(OR 1 ) 2 PS radical in the EI ion source. Both azinphos-ethyl and azinphos-methyl monitor the SRM transitions at m/z of 160→105, and 160→132 for quantitation and confirmation analysis [3,6]. The m/z = 160 fragment ion undergoes collision-induced dissociation through loss of N 3 CH or C 2 H 2 to give fragment ions at 105 and 132, respectively.
An additional reason why LC-ESI + -MS/MS is chosen over GC-MS methods for OPs is the ability to analyze OPs and OP sulfones, sulfoxides, and oxons simultaneously with often comparable sensitivities to their parent OPs [6,11,26,28,29,32,35]. Molecular weight confirmation is available as the protonated molecular ion is high in abundance and generally selected for the precursor ion *LC-ESI + -QTOF-MS. for LC-MS/MS (Table 3). Similar to the OPs, mobile phase containing methanol (and gradient elution) is often preferred for optimal sensitivity of OP degradation products. However, when OPs (or their degradation products) are included in multiclass methods, acetonitrile may be selected due to the sensitivity needs of other target chemical classes of pesticides and to reduce run times.
Alkylphosphates and alkylthiophosphates can also be analyzed by LC-MS/MS but to achieve the required sensitivity LC-ESI --MS/MS is selected such that they are typically analyzed in a separate method from OPs (see Table 4) [11,24,27,31,33,34,39]. To provide the best sensitivity, acetonitrile rather than methanol is selected as the organic modifier in the mobile phase with either acetic or formic acid as a mobile phase additive. Chlorpyrifos degradation product 3,5,6-trichloro-2-pyridinol has been widely studied and can be included in LC-ESI + -MS/MS methods with approximately a 50 times higher detection limit than OPoxons [11]. LC-ESI --MS/MS has also been widely used, however, collision-induced dissociation only produces the Clfragment ion such that it is more common to monitor the 35
To improve sensitivity and extend the range of carbamates amenable to GC-EI-MS methods derivatized prior to analysis with 9-xanthydol, trimethylphenylammonium hydroxide and trimethylsulfonium hydroxide or sodium hydride has been used [43][44][45]. Metabolites of carbofuran and carbaryl have been analyzed after derivatization using trifluoroacetic acid with trimethylamine to produce volatile derivatives that can be analyzed by GC-EI-MS [46]. Photodegradation products (phenols and para-hydroxybenzamides) of carbamates were analyzed directly by GC-EI-MS/MS method [47].
For LC-ESI + -MS/MS the precursor ion is generally selected as the protonated molecular ion [M+H] + (see Table 6). Both methanol and acetonitrile have been used as the organic modifier in the mobile phase for the separation of carbamates and when both chemical classes are analyzed together; however, acetonitrile provides the best overall sensitivity. Sodium adducts of carbamates can also be observed with ESI + and have been attributed to impurities in methanolic mobile phases or sodium from metal tubing [51]. Both 0.1% formic acid and 5 mM ammonium acetate should be added to the mobile phase to improve sensitivity and to provide for ammonium adduct [M+NH 4 ] + formation for aldicarb, methiocarb sulfone, and oxamyl (see Table 6) [51,53]. Ammonium acetate can also improve the peak shapes observed in the separation. Aldicarb sulfone and methiocarb sulfone observed both the protonated molecular ion and ammonium adduct under these conditions [53]. The addition of ammonium acetate to the mobile phase also minimizes sodium adduct formation which was observed in this work and others for aldicarb, aldicarb sulfone, aldicarb sulfoxide, 3-hydroxycarbofuran, siduron, and diuron [51]. The common, group-specific fragmentation pathway for N-methylcarbamates is the neutral loss of methyl isocyanate (CH 3 -N=C=O), while for phenylureas, loss of the substituted aniline ring is common. For methomyl-oxime only one significant fragment ion was formed. The RSD of the ratio of areas SRM1/SRM2 was less than 20% for the majority of the compounds (see Table 6 Atmospheric pressure chemical ionization in positive and negative modes (APCI + or APCI -) can give similar range of sensitivity and structural information as ESI + and can provide added selectivity for the LC-MS/MS analysis of carbamates [51]. Sodium adducts of the molecular ion do not form with APCI + and sensitivity is better in positive ion mode than in negative ion mode, partially due to greater fragmentation with to [M-CONHCH 3 ] − in the APCI − ion source [52,59]. LC-APCI + -MS has also been found to be more sensitive for some phenylureas [60].
In general, only a few pyrethroids have been included in LC-ESI + -MS/MS multiclass methods.

Other considerations
Generally, there is a larger diversity of azole fungicides and strobilurin fungicides that can be analyzed with LC-ESI + -MS/MS methods as compared to those amenable to GC-MS methods [76,79,80,85,86]. For pesticides that are halogenated, GC-NCI-MS should be considered as an option to improve the sensitivity or selectivity of the analysis. Dissociative electron capture is often observed in negative chemical ionization for OPs, OCs, pyrethroids, azole fungicides, and strobilurin fungicides. GC-EI-MS/MS methods may also provide added selectivity; however, as many pesticides from these chemical classes fragment easily in an EI ion source, the precursor ion may need to be selected as a fragment ion which is capable of undergoing further collision-induced dissociation to achieve the required sensitivity. OP metabolites (OP oxons, sulfones, sulfoxides, and selected others) can be analyzed by LC-ESI + -MS/MS, while alkylphosphates or alkylthiophosphates should be analyzed by LC-ESI --MS/MS or following derivatization by GC-MS. Pyrethroid metabolites are still commonly analyzed following derivatization with GC-EI-MS methods with a small selection of common pyrethroid metabolites also frequently analyzed by LC-MS/MS.

Conclusions
A larger number of OPs including organophosphates and organothiophosphates have been analyzed by GC-MS or GC-MS/MS methods as compared to LC-ESI + -MS/MS. GC-EI-MS or GC-EI-MS/MS is most commonly selected for analysis of OPs, and GC-EI-MS provides excellent confirmation of identity of the OP through spectral library matches. When added selectivity is required, such as when matrix remains after sample clean-up, analysis of OPs by GC-NCI-MS or GC-EI-MS/MS should be selected. GC-NCI-MS analysis of halogenated (or nitro substituted) OPs generally provides better sensitivity than GC-EI-MS/MS, particularly when the precursor ion selected for CID is the molecular ion. Although NCI is a softer ionization process than EI, fragment ions are still often observed as a result of dissociative electron capture. Sensitivity of GC-EI-MS/MS can be improved by selection of an abundant fragment ion for the precursor ion rather than the molecular ion which may be too low in abundance. The number of applications using LC-ESI + -MS/MS for the analysis of OPs has increased in the past ten years and for those OPs that can be ionized efficiently by ESI, the sensitivity may be better than with GC-MS methods (particularly for OPs that elute later in the GC separations). Another advantage of LC-ESI + -MS/MS is that it is feasible to analyze OP degradation products (OP oxons, OP sulfones, or OP sulfoxides) simultaneously with parent OPs. Derivatization of alkylphosphates and alkylthiophosphates metabolites of OPs is required to achieve the desired sensitivity when analyzed by GC-MS or GC-MS/MS methods. Alkylphosphate metabolites can also be analyzed by LC-ESI --MS/MS. Carbamates and phenylureas are commonly analyzed by LC-ESI + -MS/MS. Selected carbamates can be analyzed by GC-MS methods, but a derivatization step is required prior to analysis. The main degradation products of carbamates including carbamate sulfone or sulfoxides can be analyzed by LC-ESI + -MS/MS simultaneously with carbamates and phenylureas. APCI and APPI in positive ion mode have also been used to ionize metabolites of carbamates to achieve better sensitivity than ESI. APCI + is also not prone to sodium adduct formation. Mobile phase additives used for the LC-ESI + -MS/MS separation of both OPs, carbamates and phenylureas include 0.1% formic acid and 5 mM ammonium acetate. Better sensitivity for OPs is obtained when methanol is used as the organic modifier for gradient elution, while acetonitrile is more commonly used for the separation of carbamates to obtain optimal sensitivity. Carbamates are prone to adduct formation (reduce sensitivity) in mobile phases containing methanol, and ammonium formate or ammonium acetate is generally used to reduce sodium adduct formation. Other pesticides that can be analyzed by LC-ESI + -MS/MS include azole fungicides, neonicotinoid insecticides, and strobilurin fungicides. Pending the target list of pesticides, it is feasible to obtain simultaneous analysis of all these chemical classes; however, if optimal sensitivity is required then class-specific methods will achieve better results.