Introduction

Ion trap mass spectrometers have become a powerful tool to study gas-phase chemistry, especially ion-molecule reactions [1, 2]. Intriguing mechanistic insights into ion-molecule reactions have been obtained on low-resolution mass spectrometers, analyzing reactions of ions with various neutral molecules such as CO2, O2, and ethene [3,4,5,6,7,8]. A systematic evaluation of mass spectra has often been hindered by the low resolution of common instruments and in some cases even prevalent fragments have not been completely rationalized. The introduction of high-resolution accurate mass mass spectrometry (HRAM-MS) enables the systematic investigation of such gas-phase ions. This has been demonstrated with Fourier transformation ion cyclotron resonance mass spectrometry (FT-ICR-MS) instruments that have been, for example, employed for the investigation of gas-phase H/D exchange of proteins, nucleic acids, and mononucleotides [9,10,11]. The recently introduced gas chromatography (GC) electron ionization (EI) Orbitrap mass spectrometers offer now the possibility to direct, store, and react high-energy ions using commercially available instrumentation [12]. The power of this GC coupled HRAM-MS has already been demonstrated in pioneering investigations of fragmentation pathways in the EI source [13, 14]. Here, we demonstrate that the Orbitrap mass spectrometer offers also broad and powerful possibilities for the systematic investigation of in-trap reactions.

Reactions within the ion trap have not only a substantial influence on the relative ion intensities in EI spectra. The fact that ions are stored in a relatively small volume for a certain amount of time and undergo secondary reactions can also be utilized to address the reactivity and the kinetic energy of the ions inside the trap. The number of ions and other possible reactants are key parameters influencing the spectra [15, 16]. The occurrence of chemical gas-phase reactions in ion traps like in the C-Trap of an Orbitrap instrument is known but not yet fully described [17,18,19,20,21,22]. With this study, we address a number of ions that undergo chemical transformations in a Q Exactive GC mass spectrometer and systematically address the resulting mass spectra.

Experimental

Materials and Reagents

All chemicals used as standards had a minimum purity of 93% and were purchased from Fluka (Munich, Germany), Alfa Aesar (Karlsruhe, Germany), Acros Organics (Darmstadt, Germany), and Roth (Karlsruhe, Germany). Perfluorotributylamine (PFTBA, Thermo Fisher Scientific, Bremen, Germany) was used as calibrant. The derivatization reagent N, O-bis-trimethylsilyl-trifluoroacetamide (BSTFA) was purchased from Supelco Analytical (Munich, Germany). Pyridine 99.9% used for derivatization and labeled water (D2O, 99 atom % D; D218O, 99 atom % D, 95 atom % 18O) were obtained from Sigma Aldrich (Munich, Germany). BrCN, obtained from Alfa Aesar (Karlsruhe, Germany), also contained significant quantities of ClCN and ICN that were used for HXCN+/XCN•+ (X=Cl, Br, I) ratio determination. Purified water was obtained from a TKA MicroPure water purification system (Niederelbert, Germany). Solid phase micro extraction (SPME) was performed with polydimethylsilane/carboxen co-polymer fibers (75 μm, 23 gauge for manual holders) obtained from Supelco (Bellefonte, PA, USA).

Sample Preparation and Data Analysis

Analytes for derivatization were dried under reduced pressure in a desiccator overnight. Dry pyridine was kept stored on baked-out molecular sieve (4 Å). A mixture of the dried analytes benzoic acid, geraniol, 3-hydroxy benzoic acid, 4-hydroxyphenylacetic acid, α-bisabolol, farnesol, 3,4-dihydroxybenzoic acid, 3-hydroxycinnamic acid, hexadecanoic acid, linoleic acid, stearic acid, and cholesterol in dry pyridine was prepared with a final concentration 10 μg/mL per compound. Additionally, a mixture of benzoic acid, 3-hydroxybenzoic acid, 3,4-dihydroxybenzoic acid, 4-fluorobenzoic acid, 3-nitrobenzoic acid, and 2-nitrobenzoic acid was prepared in dry pyridine. Concentrations were chosen to generate similar intensities of the individual molecular ions (Supplementary Table S1). One volume of BSTFA (20 μL) was combined with one volume of the compound or acid mix (20 μL), respectively, and incubated at 60 °C for 1 h (modified after [23]). Subsequent GC-HRMS analysis was performed whereby the derivatized acid mix was measured in technical triplicates.

For measurement of PFTBA under D218O and D218O-free conditions, the calibrant gas flow was turned on and mass spectra were acquired for 5 min. MS measurements of PFTBA in the presence of D218O were performed after the D218O valve was opened for 1 h.

For determination of HXCN+/XCN•+ ratios, 40 mL of a 1 mm BrCN solution in purified water was transferred to a 50-mL glass vessel. The vessel was sealed with parafilm and incubated for 24 h to create a gas/liquid phase equilibrium. An SPME holder was pierced through the parafilm and the fiber was exposed to the headspace of the solution for 1 s, 4 s, and 10 s, respectively. Experiments were performed in triplicates. Supplementary Tables S2 and S3 list peaks selected for integration. Statistical tests were performed in SigmaPlot 11.0.

Mass Spectrometer Settings

A Q Exactive GC mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) was used in all experiments. The resolution was set to 120,000 (FWHM) throughout analysis of liquid samples under varying conditions, and for SPME samples of XCN resolution was set to 60,000 (FWHM; instrument setting at 200 u). Mass range was 50–600 u for measurements of liquid samples and 50–450 u for XCN measurements. Automated gain control (AGC target) was set to 1 × 106, and maximum inject time was set to “auto.” Auxiliary temperatures were set to 280 °C for both transfer lines 1 and 2. MS transfer line temperature was set to 250 °C and the temperature of the electron ionization source was set to 300 °C. EI was performed at 70 eV energy, and only for liquid samples a filament delay was set to 5.4 min. Nitrogen for supply of the C-Trap and HCD cell of the GC Orbitrap had a minimum purity of 99.999% (Linde AG, Munich, Germany), and was further dried, using a moisture filter (the vendor specifies a gas quality of 6.0 after passage; Thermo, Bremen, Germany).

For the comparative unit resolution spectrum, a TSQ 8000 (Thermo Fisher Scientific) EI triple quadrupole GC mass spectrometer was used operating in single quadrupole mode. The mass spectrometer was set to monitor the mass range between 50 and 650 u in EI+ (70 eV) mode. The MS transfer line and the ion source temperature were set to 300 °C.

Gas Chromatography

Identical GC-and autosampler instrumentation was used for both mass spectrometers. Liquid samples were injected via a TriPlus™ RSH autosampler (Thermo Fisher Scientific, Bremen, Germany), while SPME fibers were manually injected for desorption into a programmed temperature vaporizer (PTV) injector on a TRACE™ 1310 gas chromatograph (Thermo Fisher Scientific, Bremen, Germany). For liquid samples, the PTV initial temperature was set to 100 °C, and the flow was 50 mL/min. During evaporation, the flow was increased to 80 mL/min. During transfer phase, the injector temperature was increased by 14.5 °C/s to 350 °C, and the split flow was stopped for 1 min. After transfer on column, the injector was cleaned at 400 °C with a flow of 50 mL/min for 5 min. For analysis of XCN, the PTV injector was used in split mode (1:10 split flow) at an injector temperature of 260 °C. Analysis of the trimethylsilane-derivatized acid mix was carried out using the SSL injector in split mode (1:10 split flow) at a temperature of 300 °C, and a gas flow of 1 mL/min. Helium 5.0 was used as carrier gas with an additional moisture and oxygen trap (the vendor specifies a gas quality of 6.0 after passage), using a flow rate of 1 mL/min on a ZB-SemiVolatiles column (5% phenyl, 95% dimethylpolysiloxane; 30 m length, 0.25 mm inner diameter, 0.25-μm-film thickness, 10 m pre-column, Phenomenex, Aschaffenburg, Germany).

For liquid samples, an initial oven temperature of 80 °C was held for 2 min, and then increased to 120 °C using a rate of 20 °C/min. After holding the temperature for 1 min, temperature was increased to 250 °C, using a rate of 5 °C/min, to then further increase the temperature to 320 °C at a rate of 10 °C. This temperature was kept for 2 min.

For SPME samples, the initial oven temperature was set to 40 °C and kept for 3 min. The temperature was increased to 260 °C at a rate of 100 °C, and kept for 2 min.

Supply of Labeled Water to the C-Trap

To dope the nitrogen flow supplied to the C-Trap/HCD cell with labeled water, a bypass with a glass reservoir was installed between the moisture filter and the nitrogen gas connection of the Q Exactive GC-MS (Figure 1). Tubing was made from copper, with an inner diameter of 1 mm, connected via two 3-way ball valves made from brass (Swagelok, Solon, OH, USA) to allow on/off switching of the labeled water supply. The glass reservoir had a volume of 3 mL and was filled with 200 μL of labeled water (D2O or D218O). The overall distance from glass reservoir to the gas connection of the mass spectrometer was about 45 cm. The nitrogen flow towards the gas connection was 0.5 mL/min. Labeled water was allowed to evaporate at room temperature to avoid saturation and contamination of the instrument. Measurements of PFTBA and the compound mix under D218O conditions were performed after the bypass valve was opened for 1 h.

Figure 1
figure 1

Mass spectra of cyanogen bromide and experimental setup for water enrichment of the C-Trap gas supply of a Q Exactive GC-MS. (a) EI mass spectrum of BrCN obtained from a GC TSQ™ 8000 triple quadrupole mass spectrometer in single quadrupole mode. (b) EI mass spectrum of BrCN obtained from Q Exactive GC-MS measurement. (c) EI mass spectrum of BrCN, using SIM mode m/z 105.7–110.1, gray window in (b). (d) Bypass installed after the moisture filter with a reservoir to dope N2 supply for the C-Trap with labeled water. (e) EI mass spectrum of BrCN after doping with D2O for 2 h. (f) EI mass spectrum of BrCN after doping with D2O for 2 h (m/z 107 region magnified)

Full size image

Results and Discussion

Cyanogen Halides Indicate Water Reactions in the C-Trap

Cyanogen bromide (BrCN) EI-MS spectra measured on a GC triple quadrupole mass spectrometer in single quadrupole mode (Figure 1a) show fewer peaks, compared to spectra measured on a Q Exactive GC-MS (Figure 1b). The accurate mass of the dominant additional ions allowed unambiguous identification of the underlying molecular formulae as H79/81BrCN+ (Figure 1b). Mechanistic considerations suggest that H79/81BrCN+ are not any more radical cations, which is also confirmed by the accurate mass. Their formation can be explained by a hydrogen atom transfer reaction (HAT) to the initially generated BrCN•+ with a hydrogen donor like water and a subsequent radical loss of OH•.

The setup of the Orbitrap instrument allows two possible locations for the reaction: (1) The ion pathway after ionization, including the quadrupole, and (2) the C-Trap [20]. To distinguish between these two possibilities, a selected ion mode (SIM) experiment was conducted. An m/z window 105.7–110.1 (see Figure 1b, gray area) was chosen to specifically exclude the isotopologue 79BrCN•+ (m/z 104.9209). The resulting mass spectrum lacked the HAT product H79BrCN+ (m/z 105.9287) and its isotopologues (Figure 1c). This indicates the absence of these ions in-between the ion source and the quadrupole and allows assigning the location of their formation to the C-Trap. H79BrCN+ is missing since the precursor 79BrCN•+ (104.9) was excluded by the quadrupole. To further confirm that the reactions take place in the C-Trap, the nitrogen flow (~ 0.5 mL/min) towards the C-Trap was doped with deuterium oxide. Therefore, a bypass with a small glass reservoir was installed after the moisture filter (see Figure 1d). After guiding the nitrogen flow through the bypass for 2 h, low amounts of DBrCN+ could be detected (see Figure 1e, f) and after 46 h, an increase in DBrCN+ to 6% of all formed atom transfer products was observed. D2O did not quantitatively replace water in the system. The remaining additional water could arise from the carrier or dampening gas or the sample. Installation of a moisture filter to the nitrogen flow that retained residual amounts of water (≤ 5 ppm) did not prevent the formation of adducts in the collision cell [17, 19]. After closing the valve, the relative amount of DBrCN+/HBrCN+ fell under 1% within 120 h (Figure S1). These results confirm that the C-Trap of the Q Exactive GC-MS is also a gas-phase reactor where the residual water undergoes reactions with ions, not only by forming adducts or addition/elimination reactions as previously reported [14, 20] but also by gas-phase reactions with radical ions.

The intensities of signals due to HAT reaction products exceeded those of the molecular ions of BrCN, implying high reactivity of the radical cation of BrCN with water. The ratio of HBrCN+ to BrCN•+ was concentration dependent (Figure 2a). The more BrCN•+ ions were present in the C-Trap, the more BrCN•+ ions remained unaffected, increasing the relative amount of BrCN•+ in relation to HBrCN+. Whereas the absolute intensity of BrCN•+ increased by a factor of 127, the ratio of BrCN•+ to HBrCN+ only increased from 0.4 to 2.1. This might be explained by a second-order reaction with water inside the trap.

Figure 2
figure 2

(a) Impact of the absolute measured ion intensity on the HBrCN+ (light gray): BrCN•+ (dark gray) ratio. Mean absolute intensities of triplicate measurements of the M•+and M-H+ ions are shown. Relative abundances of BrCN•+:HBrCN+: 0.4:1 (mean intensity 4.4 × 107), 1.4:1 (mean intensity 7.3 × 108), 2.1:1 (mean intensity 5.7 × 109). Mean absolute intensities differed significantly by factor 16 and 127 compared to 4.4 × 107 (P < 0.001; one-way ANOVA). (b) Impact of halide of XCN on HXCN+/XCN•+ intensity ratio (mean intensity ~ 4.1 × 107). Relative intensities of HXCN+:XCN•+: HClCN+:ClCN•+, 9.4:1; HBrCN+:BrCN•+, 2.4:1; HICN+:ICN•+, 1:377. BrCN data were copied from (a) for sake of comparison. No significant difference between mean absolute intensities of measured cyanogen halides was observed (P = 0.64, one-way ANOVA). Error bars show standard deviation of triplicate measurements

Full size image

To analyze the influence of the halide on the HXCN+ formation, three cyanogen halides (XCN, X=Cl, Br, I) were analyzed and the respective HAT product formation was monitored (Figure 2b). Sums of absolute intensities of analyzed ions were kept at the same level for all of the three cyanogen halides (mean intensities ~ 4.1 × 107) to ensure comparability in the analysis of product formation. All three tested cyanogen halides formed HAT products. A clear trend of relative proportions of HClCN+>HBrCN+>HICN+ in comparison to the non-protonated XCN•+ ions was observed. Whereas HClCN+ and HBrCN+ were the major observed species, hydrogen adducts of ICN could only be detected in traces. Measured as well as calculated proton affinities and gas-phase basicities of HXCN+ showed the reverse trend dependent on the nature of the halogen [22]. This different behavior substantiated the radical character of the herein described gas-phase reaction, since increasing electronegativity of the halide impairs the stability of the radical cation [24] and increases its reactivity (Figure 2b).

Figure 3
figure 3

(a) EI mass spectrum of calibration gas PFTBA (positive mode) without D218O doping. (b) Mass spectrum of calibration gas PFTBA (positive mode) after doping of N2 flow gas with D218O for 1 h (18O containing peaks highlighted in orange, 16O containing peaks highlighted in turquoise, bold numbers in parentheses indicate the number of the respective structure in Figure 4). Fragments F1 and F2 have been previously described [25], and in accordance, the 18O containing F3 could be identified (compare Supporting Figure S2)

Full size image

New Fragment Ions from the Calibration Gas PFTBA

To gain further insights in gas-phase reactions with water in the C-Trap, we addressed fragmentation of perfluorotribulylamine (PFTBA or FC-43), the common calibration reagent for EI-MS. PFTBA fragments are known to undergo gas-phase reactions with residual water by electrophilic addition of water and HF elimination [14, 25]. HRAM-MS in combination with selective labeling of water allows now to gain insight into the pathways within the C-Trap. Therefore, the N2-flow was doped with D218O, allowing tracing both incorporation of D and 18O. By comparison of EI mass spectra of PFTBA, acquired in the presence and absence of D218O in the C-Trap, previously described masses of fragments and products from reactions with water could be detected (Fig. 3, Figure S2, see Table S4 for accurate masses and deviations). Moreover, two hitherto unknown gas-phase reaction products of water and PFTBA fragment ions, namely C4F4NO+ (4) (found m/z 153.99092, calcd. 153.99105, dev. − 0.8 ppm) and C4F5O2+ (9) (found m/z 174.98112, calcd. 174.98130, dev. − 1.0 ppm), and their 18O-labeled isotopologues could be identified (see Figure 4). Putative mechanisms explaining the formation of 4/5 and 9–11 are shown in Figure 4. As reported earlier, one of the major fragments of the molecular ion of PFTBA 1 is 2 (C8F16N+, m/z 413.97663) [25]. The elimination of a perfluorobutane leads to 3 (m/z 175.99278) which reacts with H2O to 4 (m/z 153.99092) under loss of 2 HF, or product ion 5 (m/z 155.99514) in presence of D218O under loss of 2 DF. The formation of 9–11 can be explained by two addition-elimation reactions of the highly reactive perfluoro butyl cation 6 C4F9+ with H2O/D218O under losses of either HF and/or DF.

Figure 4
figure 4

(a) Suggested sequence for fragment formation from PFTBA 1 in the presence of H2O and D218O. Elimination of perfluorobutane from fragment 2 leads to 3 which reacts to 4 under loss of 2 HF or 5 in presence of D218O under loss of 2 DF. (b) Reactions of fragment 6 with water in the C-Trap under neutral loss of 2 HF to 7, and further to 9 under a second neutral loss of 2 HF. D218O doping leads to fragment isotopologues 8, 10, and 11

Full size image

Gas-Phase Reactions with Water Explain Unusual Fragments in Ion Trap Mass Spectra of Silyl Ethers and Esters

Analyzing Orbitrap EI-MS data of silyl ethers and -esters revealed prominent ions that can be explained with gas-phase reactions as well (Table S6 and mass spectra in the supporting information). We systematically addressed these reactions by investigation of compounds from diverse classes. These include terpene alcohols, amino acids, fatty acids, phenols, aryl carboxylic acids, and halogenated compounds that were derivatized, using the BSTFA/pyridine method [23]. The derivatized compounds were analyzed on the GC Orbitrap MS under normal and D218O-enriched conditions. Comparing the obtained EI mass spectra manually revealed that [M+3]+-peaks can be detected in several analytes (Figure 5a, see Table S5 for accurate masses and deviations). HRAM-MS and evaluation of the intensities revealed that these ions cannot be explained by the isotopic pattern of the analytes. Under D218O-enrichment, a series of further peaks [M+4]+, [M+5]+, [M+6]+, and [M+7]+ are detected (Figure 5b) suggesting their origin from the reaction with residual water in the C-Trap. A detailed analysis of the ions reveals that they are reaction products of the common fragment [M−CH3•]+ from trimethylsilylated compounds with water. The structures in Figure 5b indicate a suggested pathway that explains the observed signals. The proposed initial loss of the silyl ester-CH3 followed by decarboxylation to 14 is supported by the mass spectrum of trimethylsilyl benzoate 18 which shows the same losses and comparable intensities. Specifically, the [M+3] ions can be described as [M−CH3•+H2O]+ which is also supported by accurate mass measurements ([M+2.98712] error − 0.4 ppm, Figure 5a). Additionally, the D218O labeling confirmed this hypothesis due to the presence of the [M+7]+= [M−CH3•+D218O]+ peak ([M+7.0039], error 0.5 ppm, Figure 5b, arrow F). The presence of [M+4]+, [M+5]+, and [M+6]+ ions in these treatments makes it obvious that additional unlabeled water in the trap can be involved in exchange reactions. The reactivity of the [M−CH3•]+ fragments might be explained by their electron deficit and oxophilic character [26]. In our compiled list the TMS-derivatives of aryl carboxylic acids showed the respective ions. In all analyzed fatty acid silyl esters, the [M+3]+ peak was more abundant than the molecular ion itself; even in a not deliberately water spiked system. This has major implications for the MS-based structure elucidation since [M+3]+ signals are rarely considered.

Figure 5
figure 5

(a) EI mass spectrum of the bis-trimethylsilyl 3-hydroxybenzoic acid (12) in presence of residual H2O in the C-Trap. Formation of a [M−CH3•+H2O]+ adduct of 12 led to a characteristic mass shift of 2.9871. (b) EI mass spectrum of 12 after D218O doping. Formation of [M−CH3•+H2O/HDO/D2O/H218O/HD18O/D218O]+ adducts could be observed. (c) EI-MS of fragment cation 14 under control conditions. Formation of [M−CH4+H2O]+ adduct of 14 led to a characteristic mass shift of 1.97925. Box shows hypothetical formation of fragment 14 derived from 12 via 13 after loss of a CH3 radical and a subsequent neutral loss of CO2. A loss of methane leads to 15 which tended to form water adduct 16. Note that the structures 13–16 have not been verified by calculation but rather illustrate a mechanistic proposal. (d) EI mass spectrum of 14 after D218O doping. Formation of [M−CH4+H2O/HDO/D2O/H218O/HD18O/D218O]+ adducts could be observed. See Table S5 for accurate masses and deviations

Full size image

We further analyzed the mass spectra of for fragments of which the masses shifted upon D218O exposure. In the case of the bis-trimethylsilyl m-hydroxybenzoic acid, the ion with m/z 223.09674 (14, [C11H19OSi2]+, error − 0.7 ppm) is the most plausible precursor of a [223.09674 + 1.97925]+ species (Figure 5c). Its formation can be explained with a loss of methane followed by the addition of water or alternatively with the loss of a methyl group followed by the addition of a hydroxyl radical. Experiments with D218O clearly indicated the presence of the [M−CH4+D218O]+ peak, confirming the methane loss (Figure 5d). The molecular identity of the water adduct ([C10H17Si2O2]+, found m/z 225.07599, error − 0.8 ppm) makes the reaction sequence 12 ➔ 16 plausible.

We found such [X+1.97925]+ signals only in fragments of analytes with at least two trimethylsilyl groups, e.g., in the derivatives m-hydroxy cinnamic acid (Table S6), further supporting the suggested reaction sequence.

Electron Density Distribution Influences Adduct Formation

As observed in the XCN spectra, the electron density distribution in the analytes strongly influences the in-trap reactivity. In a systematic approach, the [M−CH3•+H2O]+ formation was tested in the benzoic acid derivatives 12, 17–21 (Figure 6). To allow a systematic comparison, absolute intensities of the studied molecular ions and their water products in sum were kept at a similar level of 2.1 to 3.8 × 108. In benzoic acid-TMS (18) that served as reference, the M•+ and [M−CH3•+H2O]+ were present in a nearly 1:1 ratio. Introduction of an electron withdrawing fluorine in para position (19) of the aromatic ring led to higher relative intensity of the respective [M−CH3•+H2O]+ ion (< 1:5 ratio). This effect is caused by the negative inductive effect of the fluorine substituent. The relative M•+ intensity in the mass spectrum of trimethylsilyl 3-nitrobenzoate (20) was lower compared to the fluorinated derivative 19. The more pronounced electron withdrawing effect in the ortho configured trimethylsilyl 2-nitrobenzoate (21) led to a complete lack of detection of the molecular ion. The [M−CH3•+H2O]+ of 21 instead was present in 100%. In contrast, introduction of an electron donating group like an O-trimethylsilylether (12 in Figure 6) that causes a positive I and M effect led to a relative increase of the M•+, most likely because of the higher electron density on the corresponding cation which became less reactive towards water. Two O-trimethylsilyl groups (17) almost prevented any [M−CH3•+H2O]+ formation in the C-Trap of the mass spectrometer.

Figure 6
figure 6

Relative ratio of M•+ (dark gray) and [M−CH3•+H2O]+ (light gray) of different benzoic acid trimethylsilyl ester derivatives after integration of the corresponding peaks in the chromatograms using a 3-ppm window (see Table S3 for selected ions and integration parameters). Total intensities of analyzed ions were in a range of 2.1 × 108 to 3.9 × 108. Error bars show standard deviation of triplicate measurements

Full size image

Conclusions

We provide evidence for a series of water gas-phase reactions that take place under GC Orbitrap EI mass spectrometric conditions. Our findings give insight into these processes and open up further perspectives for the mechanistic investigation of gas-phase reactions using Orbitrap instruments. We conclude that for some substance classes, additional peaks occur in mass spectra caused by reactions of radical cations with residual water in the system. In certain cases, the molecular ion might not be detectable anymore. The occurrence of [M−CH3•+H2O]+ could be misleading during structure elucidation if not recognized as water adduct (see, e.g., 21 in Figure 6). The observations have thus implications for the structure elucidation using GC Orbitrap mass spectrometry since the frequently occurring [M−CH3•+H2O]+ might easily be misinterpreted as the molecular ion. A routine inspection of the spectra monitoring ions in a distance of 2.98709 u to the M•+ might easily reveal the true molecular ion. Similar routines to inspect spectra for potential adducts (e.g., M+Na+, M+K+) are already established for to electrospray ionization MS and might be easily adapted. As demonstrated with TMS-protected carboxylic acids, the relative abundance of [M−CH3•+H2O]+ ions is dependent on the electron density distribution in the analytes.