Introduction

High performance liquid chromatography-mass spectrometry (HPLC-MS) is acknowledged to be an extremely versatile and indispensable instrumental technique for determination of pesticide residues as well as other toxic chemicals [1, 2]. In this instrumentation, the HPLC is attached to the MS via an ionization interface. The most commonly used interface for HPLC-MS is atmospheric pressure ionization (API), which mainly includes electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI). ESI and APCI are well-known as soft ionization, since there is very little fragmentation. The analytes are converted into quasi-molecular ions through protonation, deprotonation, or formation of other ionic adducts ([M+H]+, [M–H]–, [M+Na]+, [M+AcO]–, etc.) in the API ion source, which is advantageous in the sense that the molecular weight of the analyte can be easily obtained. However, ESI and APCI do not always produce these conventional precursor ions for all the compounds. The exceptions are often reported. One of our authors had reported the formation of unexpected [M–H]+ ions of 1,4-diphenyl-3-benzoyl-1,4-dihydropyridines in the ESI positive mode [3]. The mechanism was attributed to an electrochemical oxidation reaction. Owing to the high voltage in the ESI interface, the 1,4-dihydropyridine was oxidized to the corresponding aromatic pyridine. More examples about electrochemical redox reactions leading to unexpected ions in the ESI and APCI interfaces had also been observed by other mass spectrometrists [4,5,6,7]. Pan et al. had reported the formation of unexpected [M+15]+ ions during the analysis of aromatic aldehydes in ESI positive mode [8]. They explained that a gas-phase aldol reaction between the aromatic aldehydes and methanol (the solvent) occurred in the ion source to produce the [M+15]+ ions. A similar result was also presented by Jiang et al. in the analysis of heteroaromatic aldehydes using ESI-MS [9]. They observed the unusual [M+15]+, [M+33]+, and [M+47]+ ions in the ESI-MS spectra, which were the intermediates of in-source aldolization reactions between the heteroaromatic aldehydes and the solvent methanol. In addition, the formation of other kinds of unusual ions in the ESI or APCI interfaces, such as open-shell radical molecular ions [10,11,12,13], covalently bound dimers [14, 15], oxygen addition species [16], and so on, had been recognized and reported in literature. These side reactions in the ion source are mechanistically interesting that can help us to more deeply understand the ionization mechanism, and are potentially useful to design novel ion sources, such as active capillary dielectric barrier discharge ionization that is significantly softer than APCI [17,18,19]. However, theses anomalies during ionization processes must draw attention of analytical chemists. The side reactions may interfere with obtaining reliable molecular mass of the analyte, generating false information about the composition of the sample, and reducing the sensitivity and accuracy of quantitative analysis.

The neonicotinoid insecticides have been used widely for long-term control of sucking insects in vegetables, fruits, tea, and other crops. HPLC-MS is an important tool to study the residues, degradation, and metabolism of neonicotinoid insecticides in agrifoods [20,21,22,23,24]. In our recent study, when the standards of neonicotinoid insecticides (the structures are shown in Table 1) were analyzed by ESI and APCI mass spectrometry, we found that their degradation products could be detected, and under certain conditions, the abundance of degradation product was more intense than that of the parent. This observation is worthy for us to carry out an in-depth study.

Table 1 Chemical Structures of the Five Neonicotinoid Insecticides and their Degradation Products Studied
Full size table

Experimental

Materials

Thiamethoxam (98.5%), clothianidin (99.0%), imidacloprid (98.0%), acetamiprid (98.1%), and thiacloprid (98.0%) standards used in this study were purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany) and used as received. Water-18O (97% atom, 18O) was purchased from Beijing InnoChem Science and Technology Co., Ltd. (Beijing, China). Methanol (HPLC grade) was purchased from Tedia (Fairfield, OH, USA). Deionized water was purified with a Mill Q-Plus system (Millipore, Billerica, MA, USA).

The thiamethoxam urea was synthesized according to previous reported procedures [25]. The structure was confirmed by NMR spectroscopy and mass spectrometry. 1H NMR (400 MHz, CDCl3) δ 7.38 (s, 1H), 4.79 (s, 2H), 4.74 (s, 2H), 4.55 (d, J=0.6 Hz, 2H), 2.90 (s, 3H). 13C NMR (100 MHz, CDCl3) δ 153.94, 152.99, 139.62, 137.40, 80.02, 78.38, 41.03, 31.36.

The neonicotinoid insecticide standards were prepared in methanol at a final concentration of 1 μg mL–1. After preparation, the samples were analyzed by MS immediately.

Mass Spectrometry

The HPLC-MS analysis was performed on an UltiMate 3000 system (ThermoFisher Scientific, Bremen, Germany) coupled with a hybrid quadrupole-Orbitrap mass spectrometer, Q-Exactive (ThermoFisher Scientific, Bremen, Germany). A ThermoFisher Hypersil GOLD column (C18, 100 × 2.1 mm, 3 μm) was used for chromatographic separation. The column temperature was set at 30 °C and the autosampler temperature was set at 10 °C. The isocratic mobile phase consisted of methanol/water (30/70, v/v) at a flow rate of 0.2 mL min–1. The injection volume was 3 μL. Ionization in APCI positive mode was performed using the following settings: spray current 5 μA, sheath gas (N2) flow rate 35 arb, auxiliary gas (N2) flow rate 10 arb, sweep gas (N2) flow rate 3 arb, heated capillary temperature 320 °C, auxiliary gas heater temperature 320 °C, S-lens radio frequency (rf) level 50. Ionization in ESI positive mode was performed using the following settings: spray voltage 3.5 kV, sheath gas (N2) flow rate 35 arb, auxiliary gas (N2) flow rate 10 arb, heated capillary temperature 320 °C, S-lens rf level 50. Orbitrap resolution was set to 70,000 full-width at half-maximum (FWHM) at m/z=200. The MS/MS spectra were obtained with nitrogen (99.99%) as the collision gas after isolation of the desired precursor ion. The normalized collision energy (NCE) was varied in a range of 10%–50% to provide suitable energy for the fragmentation of the [M+H]+ ions.

The samples were also analyzed by using flow inject analysis (FIA) method without chromatographic separation. Unless otherwise indicated, the HPLC and MS conditions were the same as those in HPLC-MS analysis. The mobile phase was methanol/water (90/10, v/v) and methanol/water (50/50, v/v) in APCI and ESI, respectively. Different auxiliary gas heater temperatures (220 °C, 320 °C, and 420 °C) in APCI were investigated.

18O Labeling Experiment

Thiamethoxam and acetamiprid were dissolved in methanol/water-18O (50/50, v/v) to a final concentration of 1 μg mL–1. Solutions were infused to the mass spectrometer with a syringe pump at a flow rate of 20 μL min–1. The MS conditions were the same as those in FIA-MS analysis.

Results and Discussion

In the reported literature, the ESI interface has been commonly used in HPLC-MS analysis of neonicotinoid insecticides [20,21,22,23,24]. In view of APCI being often used for pesticide analysis [26,27,28,29], we tried to use HPLC-APCI-MS to determine the residual of neonicotinoid insecticides and their metabolites in tea. During method development, thiamethoxam was selected as a representative example. The HPLC-APCI-MS chromatogram and the full-scan mass spectrum of thiamethoxam standard are shown in Figure 1. Thiamethoxam is found at retention time (Rt) 2.7 min. Surprisingly, at Rt 2.7 min the most abundant ion in the mass spectrum is m/z 248 instead of protonated thiamethoxam (m/z 292). In the MS/MS spectrum of protonated thiamethoxam (Figure 2a), the main fragment ions are m/z 211 and 132, which are also observed in the full-scan mass spectrum (Figure 1b). Although m/z 248 is also a fragment ion of protonated thiamethoxam, its abundance is very low. Thus the in-source fragmentation of thiamethoxam cannot explain the unusually high intensity of the ion m/z 248 at Rt 2.7 min.

Figure 1
figure 1

The HPLC-APCI-MS chromatogram (a) and the full-scan mass spectrum (b) of thiamethoxam. The occurrence of significant M+2 peaks at m/z 250 and 294 (~30%) in the full-scan mass spectrum indicates that m/z 248 and 292 contain a chlorine atom

Full size image
Figure 2
figure 2

The MS/MS spectra of protonated thiamethoxam at m/z 292.0269 (a) and the ion at m/z 248.0260 (b)

Full size image

By analyzing the MS/MS spectrum and fragmentation pattern of m/z 248 (Figure 2b), the structure of this ion is assumed to be the protonated thiamethoxam urea, which is the metabolite or degradation product of thiamethoxam [30, 31]. Initially, we thought that the thiamethoxam had degraded during storage. However, the two compounds have the same retention time at Rt 2.7 min. To clarify this, we synthesized a thiamethoxam urea standard. The HPLC-APCI-MS chromatogram and the MS/MS spectrum of thiamethoxam urea standard are given in Figure 3. The retention time of thiamethoxam urea is 5.2 min, but its MS/MS spectrum is the same as that of the compound at Rt 2.7 min in Figure 1. Based on these experimental data, it can be concluded that the ion m/z 248 that occurred in the full-scan mass spectrum of thiamethoxam is the thiamethoxam urea, but it is not present in the standard solution. This thiamethoxamc urea should be generated from thiamethoxam after chromatographic separation, probably in the interface by in-source reaction.

Figure 3
figure 3

The HPLC-APCI-MS chromatogram of thiamethoxam urea (a) and the MS/MS spectrum of protonated thiamethoxam urea (m/z 248.0259)

Full size image

To investigate the universality of the above phenomenon, five kinds of neonicotinoid insecticides, including thiamethoxam, clothianidin, imidacloprid, acetamiprid, and thiacloprid, were studied by APCI and ESI mass spectrometry. The mass spectrometric results of these five neonicotinoid insecticides are given in Table 2. According to different pharmacophores (C=N-R), these five neonicotinoid insecticides can be divided into two categories, nitroguanidines (R=NO2) and cyanoamidines (R=CN). By using APCI, the degradation products of the five neonicotinoid insecticides all can be observed, whereas by using ESI, only the nitroguanidines can generate the degradation products, and their relative abundances are much lower. It is worth noting that every kind of the degradation product of the neonicotinoid insecticide has the same retention time as its parent. Especially to our surprise, the cyanoamidines (acetamiprid and thiacloprid) themselves do not have an oxygen atom, but their degradation products formed in mass spectrometry contain an extra oxygen atom. Therefore, the degradation mechanism of neonicotinoid insecticides in mass spectrometry seems to be interesting and requires carefully investigation.

Table 2 The mass spectrometric results of the five neonicotinoid insecticides studied using HPLC-MS method a
Full size table

Taking thiamethoxam as an example, we proposed three degradation mechanisms in the interface of the mass spectrometer (Figure 4). The first is in-source fragmentation. The fragmentation behavior of protonated thiamethoxam is similar to that of other protonated nitroguanidine compounds, the fragmentation mechanism of which has been investigated in extensive detail by O’Hair’s group [32,33,34]. Their major fragmentation pathways involve fragmentation of the nitroguanidine group in all cases. For the protonated thiamethoxam, a minor fragmentation route is observed in which the oxygen of the nitro group attacks the carbon of the guanidine group via intramolecular nucleophilic reaction leading to the loss of N2O and the formation of the protonated thiamethoxam urea. By comparing the full-scan and MS/MS mass spectra of thiamethoxam, the fragment ions (including m/z 248) of protonated thiamethoxam indeed can be observed in the APCI and ESI full-scan mass spectra (supplementary Figures S1 and S6), but the relative abundance of protonated thiamethoxam urea (m/z 248) in the MS/MS spectrum is very low. This mechanism is only applicable to the nitroguanidine neonicotinoid insecticides bearing a nitro group. However, it is not consistent with the APCI mass spectrometric results, in which the degradation products have relatively high abundance and can be formed both for the nitroguanidines and the cyanoamidines (Table 2). This mechanism can well explain the ESI mass spectrometric results of the five neonicotinoid insecticides as shown in Table 2. In the ESI full-scan mass spectra of other two nitroguanidines (clothianidin and imidacloprid), the [M+H]+ ions of their degradation products also can be observed with low relative abundance (Supplementary Figures S8 and S9). The degradation products of the cyanoamidines (acetamiprid and thiacloprid) are not observed in their ESI full-scan mass spectra (Supplementary Figures S7 and S10).

Figure 4
figure 4

The proposed degradation mechanisms of thiamethoxam in APCI

Full size image

The second is in-source pyrolysis. In APCI, the analyte in solution is introduced into a heated nebulizer. By the combination effects of heat and nebulizer N2 gas flow, the analyte and solvent are converted into a gas stream. The analyte may undergo pyrolysis in this process [35, 36]. For thiamethoxam, the oxygen of the nitro group attacks the carbon of the guanidine group via intramolecular nucleophilic reaction leading to loss of N2O. Then the degradation product is ionized to produce the protonated ion. This mechanism is also only applicable to neonicotinoid insecticides bearing a nitro group. The pyrolysis is sensitive to temperature. The relative abundances of the [M+H]+ ions of thiamethoxam and acetamiprid and their degradation products with different auxiliary gas heater temperatures are presented in Figure 5. With the increase of the auxiliary gas heater temperature, the intensities of the protonated thiamethoxam (R=NO2) and protonated thiamethoxam urea are distinctly reduced and increased, respectively, while the intensity changes of the protonated acetamiprid (R=CN) and protonated acetamiprid-amide are not significant. Other neonicotinoid insecticides studied herein (Supplementary Table S1) exhibit similar mass spectrometric behavior in APCI. This experiment indicates that in-source pyrolysis is an important reason for the degradation of nitroguanidine neonicotinoid insecticides in APCI. This pyrolysis mechanism can be ignored in ESI.

Figure 5
figure 5

The relative abundances of acetamiprid (a) and thiamethoxam (b) and their degradation products with different auxiliary gas heater temperatures (220 °C, 320 °C, and 420 °C) in APCI

Full size image

The third is in-source ion/molecule reaction. In APCI, the sample vapor produced in the heated nebulizer is ionized by ion/molecule reaction. Protonation is the predominant reaction in APCI positive mode, but side reactions may also occur [37, 38]. Inside the APCI source, water is one of the important reactant ions that can act as a source of protons [39]. Water is also a good source of oxygen atom, so it reacts with the protonated thiamethoxam through nucleophilic substitution reaction (loss of NH2NO2) giving rise to the protonated thiamethoxam urea. This mechanism is applicable to all the five neonicotinoid insecticides studied. To confirm that the carbonyl oxygen of the degradation product is derived from water, water-18O is added to the solvent. The full-scan APCI mass spectra of thiamethoxam and acetamiprid using CH3OH/H2 18O as the solvent are shown in Figure 6. Both thiamethoxam urea-16O and thiamethoxam urea-18O are observed in the full-scan mass spectrum of thiamethoxam, and only the acetamiprid-amide-18O is formed in the full-scan mass spectrum of acetamiprid. The MS/MS spectra of these 18O-labeled degradation products are consistent with that of the unlabeled degradation products (Supplementary Figures S11 and S12). These isotope labeling experiments indicate that in-source ion/molecule reaction with water is probably the primary reason for the degradation of neonicotinoid insecticides in APCI.

Figure 6
figure 6

The full-scan APCI mass spectra of thiamethoxam (a) and acetamiprid (b) using CH3OH/H2 18O (1:1, v/v) as the solvent. The occurrence of M+2 peaks at m/z 252 and 203 (~30%) in the full-scan mass spectra is consistent with the molecular formula of degradation products-18O that contain a chlorine atom

Full size image

Conclusions

Pesticide degradation or metabolism is an important research topic and the mass spectrometer has become an indispensable tool to detect the structure and concentration of the degradation products and metabolites. In the degradation or metabolism of neonicotinoid pesticides, one of the pathways is that their pharmacophore (C=N-R, R is a substituent) is converted into a carbonyl group (C=O). In this study, the degradation of neonicotinoid pesticides (thiamethoxam, clothianidin, imidacloprid, acetamiprid, and thiacloprid) inside the APCI and ESI ion sources was revealed and investigated based on various experimental methods. Their degradation in the APCI was clearly faster than that at normal conditions in the environment because of some in-source ion/molecule reactions with water molecules (based on 18O-labeling experiments). In addition, intramolecular fragmentation (the oxygen of the nitro group attacked the carbon of the imine group accompanying with loss of N2O) in the ion source also contributed to the degradation of nitroguanidine neonicotinoid pesticides (R=NO2) in APCI and ESI. The degradation of neonicotinoid pesticides inside the ion source may interfere with the qualitative and quantitative analysis of these compounds, including their degradation products and metabolites. This work permits us to highlight the importance of understanding unexpected reactions in mass spectrometry and improves our knowledge on the ionization process of atmospheric pressure ionization.