Determination of Methylphosphonofluoridic Acid in the Environment by Derivatization and LC/MS/MS Analysis

Abstract

Rapid and specific detection of toxic Novichok agents (A230, A232, A234) is crucial for forensic investigations and the prevention of chemical weapon misuse. While A232 and A234 are relatively stable, A230 is less stable and primarily undergoes hydrolysis via P–F bond cleavage. This product indicates the presence of the Novichok core but does not indicate the agent’s prior existence. In this study, a method with high sensitivity for determining the presence of the minor A230 hydrolysis product—namely methylphosphonofluoridic acid (MPFA), which is generated via P-N bond cleavage—in environmental matrices was established. 2-[(Dimethylamino)methyl]phenol (2-DMAMP) was found to be effective for the derivatization of MPFA in water. The derivatization protocol after optimization involved adding 2-DMAMP followed by agitating for 72 h at 50 °C before LC–MS/MS analysis. The derivatized MPFA, analyzed by ESI–MS/MS, showed two main fragment ions with m/z values of 185.0 and m/z 107.0. The approach was applied to tap water, aqueous soil extract, and saline samples. While intact MPFA exhibited reduced detectability due to strong matrix effects, derivatization enhanced its stability and minimized interferences, resulting in its significantly higher detection sensitivity. The detection of MPFA provides a clear indication that the toxic Novichok compound was present prior to hydrolysis.

1. Introduction

Novichok chemical warfare agents (CWAs), developed in the Soviet Union, are classified as fourth-generation organophosphorus toxins. Their use in several incidents in recent years has raised widespread international concern, resulting in their inclusion in the Chemical Weapons Convention’s Schedule 1 in June 2020 [1,2,3]. Consequently, research on Novichok agents has expanded significantly in recent years, and a number of analytical approaches have been implemented for their determination. Techniques such as gas chromatography–mass spectrometry (GC–MS) and liquid chromatography–tandem mass spectrometry (LC–MS/MS) are frequently employed for the detection of Novichok agents, their degradation products, and specific biomarkers in biological samples [4,5,6,7,8,9]. Novel derivatization approaches have also been introduced to enhance detection sensitivity and selectivity [4,9]. Additionally, mass spectrometric analysis of protein adducts, such as those formed with butyrylcholinesterase, has been reported as a reliable method for verifying exposure to Novichok agents in forensic and biomedical investigations [5,6,10]. In addition to the development of detection methods, numerous studies have investigated the decontamination and environmental fate of Novichok agents [7,11,12,13,14,15]. Recent studies have shown that Novichok agents A230, A232, and A234 undergo hydrolytic degradation influenced by environmental factors such as pH and decontaminants [7,16,17,18]. For these nerve agents, three primary and specific hydrolysis products have been identified, which are particularly valuable for trace analysis in environmental samples. However, minor hydrolysis products are also generated, such as methylphosphonofluoridic acid (from A230), methyl hydrogen phosphorofluoridate (from A232), and ethyl hydrogen phosphorofluoridate (from A234). Among the three agents, A230 is the least stable, degrading rapidly with a half-life of less than two minutes under various conditions, whereas A232 and A234 degrade more slowly [16,19]. The methylphosphonofluoridic acid (MPFA) is generated via P–N bond cleavage, resulting in protonation on the nitrogen atom and the release of an amidine group. Notably, the presence of MPFA alone is not conclusive evidence of A230 use, as the entire core structure of the agent is absent. Moreover, MPFA is not only a breakdown product of Novichok agents but also a degradation product of nerve agents, e.g., GB and GD [20].

The major hydrolysis product of Novichok agents is formed via P–F bond cleavage and provides information about the entire molecule. However, the presence of MPFA, in addition to the major hydrolysis product, can serve as complementary evidence supporting the conclusion that the active Novichok compound was present prior to hydrolysis.

LC–ESI–MS/MS using negative ion detection is the preferred technique for MPFA detection and identification. However, the minimum detectable levels (LODs) for this highly polar, low-molecular-mass compound are susceptible to matrix effects.

In recent years, we have established a straightforward LC–ESI–MS/MS strategy for detecting G-series nerve agents, which feature an electrophilic phosphonic compound and a suitable leaving group [21]. The method involves the incorporation of 2-[(dimethylamino)methyl]phenol (2-DMAMP) to water-based samples prior to analysis. This work focused on developing a chemical derivatization approach for the sensitive determination of the minor hydrolysis product of the A230 agent, bearing an electrophilic phosphonate group along with a good leaving group, e.g., F⁻, before its degradation to non-characteristic hydrolysis products such as methylphosphonic acid (MPA). Recently, Pardasani et al. proposed a GC-MS derivatization method for the detection of MPFA [22]. Our study is the first to report improved sensitivity for MPFA detection through derivatization in aqueous media by LC-MS.

2. Materials and Methods

2.1. Materials and Reagents

The hydrolysis product of A230, methylphosphonofluoridic acid (MPFA), was synthesized in-house and characterized by ¹H-NMR, ³¹P NMR, and ¹⁹F NMR (Figure S1). 2-[(Dimethylamino)methyl]phenol (2-DMAMP) and dimethylformamide (DMF) were purchased from Merck (St. Louis, MO, USA). LC–MS-grade water, methanol, and acetonitrile (ACN) were purchased from Biolab (Jerusalem, Israel). Tap water was taken from the municipal water system, arid soil samples were collected from Negev, Israel.

A solution of a high concentration and a series of standard solutions at the concentrations of 10 ng/mL, 100 ng/mL, and 1 µg/mL of MPFA were prepared using ACN as the solvent. Subsequently, 5 or 10 µL of each working solution was spiked to achieve a final volume of 1 mL of aqueous medium before derivatization. The final concentrations of the standards in the aqueous medium covered the range of 0.1 ng/mL to 10 ng/mL.

2.2. Instrumentation

2.2.1. LC-ESI-HRMS/MS (Orbitrap Mass Spectrometry)

The analysis was carried out using an HPLC–Q-Exactive Plus Orbitrap MS system. An Agilent 1290 HPLC system (Palo Alto, CA, USA) was employed for analyte separation. The LC chromatographic parameters are detailed in our previous study [21].

2.2.2. LC-ESI-LRMS/MS (QTRAP Mass Spectrometry)

Analyte separation was achieved utilizing an Agilent 1290 Infinity UHPLC system (Palo Alto, CA, USA). MRM analyses were performed using an SCIEX 5500 QTRAP mass spectrometer (Foster City, CA, USA) operated using Analyst software (version 1.6.2). MS parameters are described in our previous study [21]. Multiple reaction monitoring (MRM) transitions, collision energies, and other analyte-specific parameters were optimized individually for every individual compound as described in the section below. Samples were injected at a volume of 10 µL.

2.2.3. LC–MS/MS MPFA Analysis Before and After Chemical Derivatization

The detection and identification of MPFA pre- and post-derivatization were performed in both negative- and positive-ion MRM mode, respectively, using four product ions, as described in

Table 1. LC–MS/MS (MRM) settings for intact MPFA and the derivatized MFPA obtained using the LC–QTRAP system.

Analyte	Precursor Ion (m/z)	Fragment Ion	Declustering Potential (eV)	Collision Energy (eV)	Intensity Ratio	Retention Time (min)
MPFA	(−)97.0	(−)82.0	(−)50	−36	1.0	2.2
		(−)77.0		−22	15.0
		(−)63.0		−36	3.0
		(−)49.0		−60	1.0
Derivatized MPFA	(+)230.1	(+)185.0	(+)70	+27	3.5	3.4
		(+)107.0		+40	3.2
		(+)79.0		+55	1.0
		(+)77.0		+55	1.0

.

2.3. Sample Preparation

2.3.1. Preparation of Soil Extract

In total, 10 mL of water was added to 10 g of soil (1:1 w/v ratio). The mixture was then vortexed for 2 min to homogenize the soil suspension, followed by centrifugation.

2.3.2. Chemical Reaction of MPFA with 2-DMAMP-Derivatizing Agent in Environmental Samples

Five microliters of 2-DMAMP were added to a vial containing 995 µL of pure water, tap water, saline solution, or soil extract previously spiked with MPFA (0.1–10 ng). The solution was stirred for 72 h at 50 °C and analyzed directly without any additional sample preparation.

3. Results and Discussion

The determination of methylphosphonofluoridic acid (MPFA), which is one of the hydrolysis products of A230, presents significant analytical challenges due to its low volatility and high polarity. These properties make it poorly suited for direct analysis by GC–MS without preliminary derivatization. Even when LC–MS is utilized, effective separation and sensitive detection are challenging. In particular, when using the most common reversed-phase C18 columns, MPFA elutes only twice the dead time of the column [20], resulting in increased susceptibility to matrix effects and interferences. Hydrophilic interaction liquid chromatography (HILIC) has shown promise in detecting alkyl methylphosphonic acids (AMPAs), as it is expected to increase retention and improve LC-MS sensitivity [23]. However, we assume that the irreversible retention of MPFA on silica due to the fluorine atom remains a potential concern.

3.1. Evaluation of the Detection Sensitivity of Intact MPFA and Characterization of Its Chromatographic Behavior in Aqueous Media

First, underivatized MPFA was spiked into LC–MS-grade water with concentrations ranging from 0.1 to 10 ng/mL and subjected to analysis using MS/MS (MRM) to evaluate its potential detection limit (LOD). The LOD, defined as the analyte concentration giving a signal three times the noise (S/N = 3), was established at 0.1 ng/mL. MPFA fortified to 1 ng/mL in LC–MS-grade water is depicted in Figure 1a. As is clearly observed, a good peak shape was observed. Subsequently, the chromatographic peak and detection sensitivity of MPFA were investigated in three environmental matrices, representing real conditions: tap water, saline solution, and aqueous soil extract (with high salt content). Saline was used as a simplified model to approximate the ionic strength and osmolarity of plasma, with a consistent ionic strength of approximately 0.15 M. In contrast, soil solution ionic strength varies depending on the soil type. We expected that matrix complexity and high background interference, particularly in low-mass regions, would cause ion suppression and/or degraded peak shape. This is due to the polar nature of MPFA and the fact that it elutes just above the dead volume (Rt = 2.2 min). In addition, its relative instability in aqueous media further limits detectability. MPFA spiked at a concentration of 1 ng/mL in tap water, saline solution, and aqueous soil extract is shown in Figure 1b–d. Several chromatographic parameters were evaluated: elution time, peak height, and baseline width. A well-defined signal (12 s width at baseline) with a relatively intense signal was obtained for MPFA in LC–MS-grade water. However, in tap water, peak broadening was observed (18 s width at baseline), and peak intensity was decreased by a factor of two compared with that obtained using LC–MS-grade water. The broad peak obtained in tap water was attributed to matrix effects. In saline, the MPFA signal at 1 ng/mL was significantly affected by the high salt content, completely disappearing with no detectable signal observed. In soil samples, co-eluting interferences were detected at two specific MRM transitions at the same retention time, accompanied by a signal intensity that was four times lower than that observed in analyses using LC-MS-grade water. To address the issues of low sensitivity and poor chromatographic performance in environmental matrices, chemical derivatization was employed.

3.2. Derivatization Strategy

Two alternative chemical strategies can be employed to derivatize methylphosphonofluoridic acid (MPFA): O-alkylation using an alkylating agent, such as the well-known pentafluorobenzyl bromide or N-(2-(bromomethyl)benzyl)-N,N-diethylethanaminium bromide (CAX-B), which have recently been used for the derivatization of phenols and phosphonic acid derivatives [24]; and nucleophilic substitution, in which the fluoride is replaced by a nucleophile. One major limitation of the O-alkylation reaction for phosphonic acid derivatives is its reliance on polar aprotic organic solvents, such as acetone, under anhydrous conditions. Because MPFA is highly soluble in water and typically extracted using aqueous solutions, it is crucial to develop a derivatization method suitable for real applications in aqueous media—a significant challenge. Currently, no established analytical approach demonstrably enhances sensitivity, particularly one based on derivatization strategies.

3.3. Optimization of the Derivatization Reaction

MPFA contains an electrophilic phosphonate group and a leaving group, such as fluoride, similar to G-nerve agents, which facilitates its reaction with highly phenolic nucleophiles such as 2-DMAMP. Nevertheless, MPFA is expected to exhibit lower reactivity toward 2-DMAMP compared with G-type nerve agents due to the presence of a hydroxyl group bonded to the phosphorus atom. Under most pH conditions, this hydroxyl group becomes deprotonated, diminishing the electrophilicity of the phosphorus atom and consequently reducing its reactivity. Therefore, we examined the reaction kinetics of 2-DMAMP with MPFA. Multiple reaction conditions were studied and fine-tuned for the reaction with 2-DMAMP: reaction solvent (DMF, ACN, ACN:water 2:1, and water), reagent amount (2–20 µL of 2-DMAMP), reaction temperature (25 °C, 50 °C, and 70 °C), and reaction time (1–96 h). We observed that the reaction proceeded best in water, while in other solvent compositions, the signal intensity of the derivatized MPFA was relatively low (by factor five at least). Therefore, water was adopted as the reaction medium. The conversion of MPFA increased with reaction time, showing a threefold increase between 12 h and 72 h, after which no further enhancement in signal intensity was observed. The optimal reaction temperature was 50 °C. Lowering the temperature to 25 °C resulted in a twofold decrease in signal intensity, whereas raising the temperature to 70 °C had no effect on the signal. To assess the minimum quantity of the derivatization reagent required to achieve efficient conversion and high-intensity product ion signal, the modification was examined across a range of reagent concentrations. Increasing the reagent volume from 2 µL to 5 µL enhanced the signal intensity (by twofold), whereas further increases up to 20 µL produced no additional improvement. Finally, the optimized derivatization conditions were established as a 72 h reaction at 50 °C with the addition of 5 µL of 2-DMAMP reagent in 0.995 mL of water. The degradation products of A230 and the chemical modification of MPFA with 2-DMAMP are depicted in Figure 2a,b.

3.4. High-Resolution Orbitrap–MS/MS and Low-Resolution-MS/MS Analysis

Following derivatization with 2-DMAMP, MPFA was examined using Orbitrap-ESI –MS/MS. A plausible fragmentation pattern of the derivatization product is depicted in Figure 2c. Figure 3 presents the MS/MS spectra of MPFA post-derivatization. The precursor ion of derivatized MPFA was observed at m/z 230.0941. Two major product ions were detected: the m/z 185.0363 ion, due to the loss of diethylamine from the MH⁺, and the m/z 107.0491 ion (hydroxybenzyl cation). Two additional minor fragment ions were found at m/z 79.0541 and m/z 77.0386 (which was not observed at the presented collision energy of 30 eV).

LC, coupled to low-resolution MS/MS, was employed for both the intact MPFA and its derivatization product, with collision energies ranging from 10 to 60 eV. Four characteristic product ions were identified for each compound and selected as MRM transitions for quantitative determination (outlined in the experimental section).

Figure 2. The three degradation products of Novichok A230 (a). The reaction between MPFA and 2-[(dimethylamino)methyl]phenol (2-DMAMP) (b); and plausible fragmentation pattern of the derivatization product (c).

Figure 2. The three degradation products of Novichok A230 (a). The reaction between MPFA and 2-[(dimethylamino)methyl]phenol (2-DMAMP) (b); and plausible fragmentation pattern of the derivatization product (c).

Figure 3. Orbitrap-ESI-MS² spectra of the derivatized MPFA at a collision energy of 30 eV.

Figure 3. Orbitrap-ESI-MS² spectra of the derivatized MPFA at a collision energy of 30 eV.

3.5. Validation of Analytical Procedures

For the validation of the developed method’s suitability, MS-grade water samples were fortified with MPFA and derivatized using 2-DMAMP. The detection, quantification, and identification limit (LOD, LOQ, and LOI), as well as the linearity and reproducibility under the refined conditions were examined. The calibration curve of derivatized MPFA was linear between 0.1 and 10 ng/mL. Five concentrations (0.1, 0.5, 1, 5, and 10 ng/mL) were analyzed, yielding R² values exceeding 0.99 (Figure S2). Repeatability was evaluated using the derivatized FMPA spiked into water. The relative standard deviation (RSD) from triplicate analyses was under 15%. The derivatized MPFA maintained stability for 72 h at room temperature. Both main significant MRM transitions, MH⁺ > 185.0 (quantifier) and MH⁺ > 107.0 (qualifier), were used for detectability and confirmatory analysis, respectively. According to the European Commission criteria [25], the ratio of these transitions confirmed the analyte identity. The LOI, which required detection of MH⁺ > 185.0 and MH⁺ > 107.0 with a ~1.1:1.0 ratio (S/N ≥ 3), was 0.1 ng/mL.

3.6. Analytical Performance of MPFA in Environmental Matrices (Tap Water, Saline, and Soil Extract)

Determining the LODs and LOIs of MPFA in the different environments is complex, not only because of its instability in environmental and aqueous solutions but mostly due to the strong influence of matrix effects. As a result, the detection of intact MPFA may be short. Matrix effects, such as suppression or enhancement, may impact the reaction rate, ionization efficiency, and chemical stability of derivatized MPFA. These effects were examined via the comparison of blank matrix extracts fortified with MPFA to fortified water, which were both succeeded by chemical reaction and analysis. Aqueous extracts of tap water, soil, and saline solution were spiked with MPFA at five concentrations between 0.1 ng/mL and 10 ng/mL (0.1, 0.5, 1, 5, and 10 ng/mL) for tap water, and four concentrations (0.1, 0.5, 1 and 10 ng/mL) for soil and saline were analyzed, after which 2-DMAMP was added.

The MRM chromatograms showed an absence of background peaks around the retention time and were dominated by the derivatized MPFA (Figure 4). The matrix effect was within ±20%, with relative standard deviations (RSDs) from triplicate analyses below 15%. The signal shape and S/N ratio of the two main MRM signals (MH⁺ > 185.0 and MH⁺ > 107.0) after the chemical reaction were compared with those obtained from derivatization in pure water and exhibited a consistent signal intensity ratio of ~1.1:1.0. Recoveries > 80% (n = 3, across the entire dynamic range) were observed for tap water, soil, and saline solution. The derivatized MPFA calibration curve exhibited linearity between 0.1 and 10 ng/mL, with correlation coefficients (R²) above 0.999 (

Table 2. Validation parameters include the linearity range, correlation coefficient (R²), limit of quantification (LOQ), limit of identification (LOI), and measurement precision.

Matrix	Dynamic Range (ng/mL)	R2	LOQ A (ng/mL) (S/N ≥ 10)	LOI B (ng/mL) (S/N ≥ 3)	RSD (%)
Tap water	0.1–10	0.999	0.1	0.1	13
Saline	0.1–10	0.999	0.1	0.1	15
Soil extract	0.1–10	0.999	0.1	0.1	15

The statistical evaluation was performed using triplicate analyses. MPFA was quantified using the MRM transition 230.1 → 185.0, while the transition 230.1 → 107.0 served as a confirmatory marker (LOI). A: The most intense transition (230.1 → 185.0) was applied. B: Both transitions (230.1 → 185.0 and 230.1 → 107.0) were employed.

and Figure S2). These findings reveal that the reaction of MPFA with 2-DMAMP is successful across all aqueous media, despite variations in the sample matrix. It is notably observed that the modification yield of 2-DMAMP with MPFA in water-based soil extracts and saline solution, which have salt-rich contents, resembled the observed in water. In addition, the derivatized MPFA retained stability for 72 h or more at 25 °C in all extracts. Representative chromatograms demonstrate the LOI and LOQ level of MPFA after derivatization are depicted in Figure 5. Overall, we tracked four MRM transitions for the derivatized MPFA to further reduce the likelihood of false positives arising from similar chemical reactivity. A pair of MRM transitions with their respective signal intensity ratios, combined with a fixed retention time, satisfy the standard European identification requirements.

Overall, LC–MS/MS(+) following derivatization provides a highly effective strategy for detecting MPFA in environmental matrices. This approach stabilizes reactive MPFA, significantly enhancing its detectability and sensitivity in both tap water and aqueous extracts compared with the intact compound.

4. Conclusions

This work describes a protocol for the sensitive determination of MPFA, the minor hydrolysis product of A230, which was developed. Owing to its high polarity and low molecular mass, MPFA is particularly susceptible to matrix effects, which hinder its detection without appropriate sample pretreatment. The use of a water-soluble derivatizing agent (2-DMAMP), which reacts under mild conditions and without any additives, makes this reagent suitable and practical for determining MPFA in various environmental matrices for applications requiring detection of trace amounts. Although the method takes a considerable amount of time, it is simple and is the first to enable the improved detection of MPFA in an aqueous environment. Notably, MPFA does not occur naturally in the environment. It is primarily formed in the synthesis and hydrolysis of CWAs. The presence of MPFA, in addition to the major hydrolysis product, can serve as complementary evidence supporting the conclusion that the active Novichok was present prior to use.