Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 May 17;9(21):23013–23020. doi: 10.1021/acsomega.4c02629

Rapid and Convenient Potentiometric Method for Determining Fluorosulfate, a Byproduct of the Fumigant and Greenhouse Gas Sulfuryl Fluoride

Zhihao Chen †,‡, Chengjin Wang †,§, Joseph J Pignatello †,*
PMCID: PMC11137687  PMID: 38826532

Abstract

graphic file with name ao4c02629_0008.jpg

A fluorosulfate ion (FSO3–) is a hydrolysis product of sulfuryl fluoride (SO2F2), which is widely used to fumigate buildings, soil, construction materials, and postharvest commodities, and is a potent greenhouse gas. It is a potential marker for biological exposure to SO2F2 and for monitoring the progress of reactions used to scrub SO2F2 from fumigation vent gases. Here, we report a simple and inexpensive potentiometric method for determining FSO3– using a commercial nitrate-selective electrode and discuss its application. The method is suitable for solutions between 0.0025 mM and 660 mM FSO3– at initial pH between 5 and 9. Halide interference depends on its molar ratio to FSO3– and follows the sequence, F– < Cl– < Br– ≪ I–. Halide interference can be eliminated by adding silver sulfate. Interference by bicarbonate can be eliminated by H2SO4 pretreatment, and interference by phosphate or pyrophosphate by MgSO4 addition. Sulfate does not interfere, as it does in ion chromatography. Satisfactory method detection limits for FSO3– in spiked aqueous extracts of 11 fruits were obtained. The method accurately quantified the yield of FSO3– relative to that of F– in base hydrolysis of SO2F2. This study demonstrates that the developed method is highly selective, convenient, and sensitive and thus can be of great value in practice.

1. Introduction

Fluorosulfate (FSO3–), the conjugate base of fluorosulfuric acid (pKa, −10), is a chemical intermediate in inorganic1−4 and organic synthesis,5−15 and is produced in certain reactions of interest in agricultural and environmental contexts.16,17 A major source of FSO3– is the decomposition of sulfuryl fluoride (SO2F2), which is widely used to rid pests in buildings and structures, agricultural soils, and postharvest commodities in quarantine and preshipment operations.18−26 Sulfuryl fluoride is increasingly used as a replacement fumigant for the ozone-depleting gas, methyl bromide. Foods treated with SO2F2 include wheat,20 flours,23 animal-based stored products,27 cereals and cereal products, pulses and pulse products, dried fruits, tree nuts,28 dried beef, gluten, and many others.29 Spent fumigation gases of SO2F2 are generally vented to the atmosphere. Of concern is the greenhouse gas potential of SO2F2.30,31 Atmospheric SO2F2 has a global warming potential 4780 times greater than carbon dioxide32,33 and a lifetime of 36 ± 11 years.24,25 The mole fraction of SO2F2 in the atmosphere grew by 5 ± 1% per year from 1978 to 2007, with estimated yearly emissions increasing from 0.6 Gg in 1978 to 1.9 Gg in 2007;24,25 by 2019, global emissions had increased to 2.9 ± 0.4 Gg/yr.24

The potency of SO2F2 as a greenhouse gas has prompted investigative efforts to remove it from fumigation vent streams, such as solvent adsorption and dielectric barrier discharge plasma.19,33 A potentially more convenient method is scrubbing with aqueous hydroxide solution, in which OH– acts as a nucleophile to attack the central sulfur and displace the fluorine with generation of equimolar amounts of FSO3– and F– (eq 1).

1. 1

Sulfuryl fluoride is readily absorbed by tissues of fumigated products or animals,23 where it can undergo nucleophilic and hydrolytic reactions leading to sulfate, fluoride, and fluorosulfate. Whereas SO2F2 itself is acutely toxic to fish, as well as to humans at high inhalation exposure,34 the chronic neurotoxicity and other chronic toxicities of SO2F2 have been attributed to its breakdown products.34,35 Fluoride and sulfate were found in a variety of fumigated foodstuffs including beef.21 Fluorosulfate was identified in the urine and blood of rats exposed to SO2F2.34 Sulfuryl fluoride has also been used in click chemistry to convert alcohols and amines to fluorosulfonate derivatives;8−11 the fluorosulfonate group is then removed by base hydrolysis, producing FSO3–.12−15 Metal fluorosulfates (such as Li+) have been proposed for use as electrolytes in batteries.36 Lastly, FSO3– is found naturally in the mineral Reederite-(Y) from Mont-Saint-Hilaire.17 Fluorosulfate can be regarded at this time as persistent and mobile in the environment since it is stable to hydrolysis in water at ordinary pH37,38 and only weakly coordinates to metal cations38−40 and mineral surfaces.17 While its toxicity is unknown, one study noted that FSO3– inhibits crucial metabolic enzymes including acetylcholinesterase, butyrylcholinesterase, sulfatase, and glutathione S-transferase.41 The potential toxicity of FSO3– stresses the need for analytical methods for its determination.

A rapid and convenient method for FSO3– would assist investigators in monitoring the decomposition of SO2F2 in fumigation applications, its presence in fumigated foodstuffs, and as an aid in chemical investigations where FSO3– is a byproduct. However, rapid and convenient analytical methods for FSO3– are lacking. So far, the only methods reported are based on ion chromatographic (IC) separation.19,42 However, we find that the peaks for FSO3– and SO42– in the ion chromatogram are difficult to separate (i.e., they have nearly the same retention time; vide infra). Sulfate is a natural component of water and tissues and is often coproduced in reactions involving SO2F2 (eqs 1 and 2).21,34 Therefore, due to the interference of sulfate, IC is not a good analytical tool for FSO3–. Moreover, IC equipment is expensive and unsuitable for on-site use.

1. 2

Potentiometric analysis is rapid and convenient and requires only inexpensive equipment.43 We reasoned that a particular commercial nitrate-selective electrode might be workable for FSO3– analysis, since the perchlorate ion (ClO4–), which is isoelectronic and structurally similar to FSO3–, was claimed by the supplier to strongly interfere with nitrate analysis. Our prediction turned out to be correct; in fact, the electrode is more sensitive to fluorosulfate than nitrate for which it was designed. The main objectives of this study were to 1) test electrode selectivity in the presence of potentially interfering ions, 2) validate the method in a known reaction, namely, alkaline hydrolysis of SO2F2 (eq 1) and 3) test the method for determination of FSO3– in tissues of fumigated food products.

2. Experimental Section

2.1. Chemicals

All chemicals were of analytical grade or higher purity and used as received unless stated. Potassium fluorosulfate (KFSO3, 99.5%), potassium chloride (KCl, 99.0%), lithium chloride (LiCl, 99.7%), magnesium chloride hexahydrate (MgCl2·6H2O, 99%), sodium bromide (NaBr, 99.5%), potassium iodide (KI, 99+%), sulfuric acid (H2SO4, 95.0–98.0 w/w % Plus grade), and potassium hydroxide (KOH, 99.98%) were purchased from Thermo Fisher Scientific. Potassium fluorosulfate was stored in a desiccator. Hydrogen peroxide solution (≥30%), sodium pyrophosphate decahydrate (Na4P2O7·8H2O, ≥99%), calcium chloride dihydrate (CaCl2·2H2O, 99%), sodium perchlorate (NaClO4, 98%), potassium bicarbonate (KHCO3, 99.7%), phosphorus pentoxide (P2O5, ≥ 99%) and magnesium sulfate (MgSO4) were obtained from Sigma-Aldrich, US. A cylinder of SO2F2 was obtained gratis from Douglas Products. Potassium fluoride standard solution of 1000 mg-F/L was purchased from Cole-Parmer and stored at 4 °C. Diluted KOH and H2SO4 solutions were used for pH adjustment; HCl, HNO3, and HClO4 were avoided because their anions interfere (see below).

To determine whether the purchased KFSO3 contained absorbed water, samples were either oven-dried at 105 °C or placed in a desiccator containing a bed of anhydrous CaCO3 (with <2% CoCl2 as a colorimetric indicator) or phosphorus pentoxide (P2O5) desiccant. Oven drying resulted in continuous mass loss and evidence of sublimation and/or decomposition (Figure S1 and Text S1). Desiccation resulted in the leveling-off of mass loss corresponding to 0.54% (CaCO3) or 0.18% (P2O5) moisture content of the original commercial product (Text S1, Table S1). We made no correction for water content of the purchased KFSO3 when making standard solutions.

2.2. Measurement of FSO3– and F– by Potentiometric and IC Methods

A commercial glass nitrate-selective combination electrode (Cole-Parmer, # EW-27502–31, US) connected to a pH/ion analyzer (Corning 350, US) was used to quantify fluorosulfate. The reference electrode (Ag/AgCl (3 M KCl)) was a double junction electrode, of which the inner compartment is filled with KCl solution and the outer compartment with 0.1 M (NH4)2SO4 solution. The nitrate ion exchanger is contained within a gelled, organophilic electrode membrane. When the electrode is inserted into a solution, ion exchange will take place between the membrane and specific ions, resulting in a redox potential difference obeying the Nikolsky-Eisenman equation (eq 3), a generalization of the Nernst equation,44−46

2.2. 3

where E is the measured electrode potential; E0 is constant potential determined by the electrode and background electrolyte; R is the gas constant (8.314 J/K/mol); T is the absolute temperature (K); zi is the charge number of the primary ion (FSO3–); F is the Faraday constant (96 487 C/mol); ai is the activity of freely dissolved FSO3–; aj is the activity of interfering ion j; Kij is the potentiometric selectivity coefficient (SC) for ion j with respect to FSO3–; and zj is the charge number of ion j. Here, activity was assumed equal to the ion concentration. In a background of high and constant ionic strength (IS), eq 3 is influenced only by ions that the electrode is sensitive to. It is the ion activity change that determines the obtained potential values and no obvious electrolysis takes place at the electrode membrane.47 Theoretically, the electrode response is proportional to the log of the FSO3– activity with a slope of 59 mV/decade at room temperature. The background ionic strength (IS) of the samples was controlled at 0.12 M according to the manufacturer’s protocol for the electrode by adding an ionic strength adjuster (ISA, 2.0 M (NH4)2SO4, pH 5.5) to sample at the volume ratio of 1:50. With slow and continuous stirring, the electrode response became stable within 2–5 min near the detection limit for FSO3–, but within 1 min at higher concentrations. The solution pH mentioned below for FSO3– analysis corresponds to the value prior to ISA addition. Based on our experience, with frequent use, the electrode should last for one year if used and stored according to directions, indicating the low-cost of the analysis.

Fluoride was measured using a different, fluoride-selective glass combination electrode (Cole-Parmer, # EW 27502–19, US). Sample solutions were prepared and adjusted to pH 5.3 with acetate buffer according to the supplier’s protocol. Fluoride calibration curves were constructed based on the purchased standard solution (1000 mg-F/L).

The electrode method was compared against an ion chromatographic method (Metrohm 930 Compact IC Flex, Switzerland) employing a Metrosep A Supp 5–150/4.0 (poly(vinyl alcohol) with quaternary ammonium groups), which is widely used in IC. The mobile phase consisted of 3.2 mM Na2CO3 and 1.0 mM NaHCO3 at a flow rate of 0.8 mL/min.

2.3. Effects of pH and Interfering Ions

The measured electrode potential can be sensitive to pH,43,48,49 as well as certain ions besides FSO3–. Calibration curves were constructed at different pH values of 3.0, 5.0, 7.0, 9.0, and 11.0 of the samples adjusted prior to combining with the ISA. Samples with different concentrations of potentially interfering ions were adjusted to the desired pH prior to combining with the ISA.

2.4. Food Tissue Extraction and Analysis

The fluorosulfate ion may be generated in foods that have been fumigated with SO2F2.23,28 To determine whether the proposed potentiometric method is suitable for determining FSO3– residuals in food tissues, 12 foods were selected for analysis including broccoli, cucumber, sweet onion, sweet potato, strawberries, apple, carrot, yellow squash, blueberries, cauliflower, kiwi, and dried dog food. The raw foods were first chopped and blended. Then, except for the dog food, the food cells were lysed by subjecting the blended material to three freeze–thaw cycles (about 5 min each cycle) in liquid nitrogen to rupture cells.50,51 The lysed tissues were further ground in a mortar and 2.5 g of each tissue were transferred to a small sealable flask with a desired volume of water. The flasks were heated in an oven at 65 °C for 24 h and then sonicated for 1 h to ensure that cellular components were completely released. After centrifugation at 6000 rpm for 15 min, the collected aqueous supernatant was filtered through a 0.45 μm membrane prior to addition of a known amount of FSO3–.

It was assumed that the extraction protocol efficiently extracted FSO3–. To verify this, raw food samples were directly spiked with FSO3– and then processed by the extraction protocol. The recovery of FSO3– was 91–103% (Text S2 and Table S2). As shown in Figure S2, all extracts of 10 (g tissue)/(L water) showed absorbance from 200 to 400 nm, indicating the presence of soluble organic or inorganic substances lysed from cells.

2.5. Determination of FSO3– and F– Produced by Base Hydrolysis of SO2F2

To validate the proposed method, we measured FSO3– and F– formed simultaneously during base (OH–) hydrolysis of SO2F2. This reaction is known to produce FSO3– and F– in an equimolar yield (eq 1).52 The reactions were performed in a brown glass bottle (Microsolv Technology Corporation, US) sealed by a cap with a PTFE-lined silicone rubber septum. The liquid phase (100 mL) consisted of NaOH solution at pH 11.6 or 12.5 and the headspace volume was 25 mL. A stir rate of 700 rpm was used, which caused a vortex reaching to the 1/3 level of the flask. Sulfuryl fluoride (4 mL, room temperature) was transferred from its gas cylinder to the bottle via a Tedlar Sample Bag (SKC Co., Ltd.), which was flushed 3–4 times, and a gastight syringe. A second syringe with a long needle was used to withdraw liquid samples (2.5 mL) at fixed intervals. The samples were supplemented with 50 μL of acetic acid, which reduced the pH to 5.3 and stopped hydrolysis.52 Then, aliquots were diluted 5-fold and analyzed for F– and FSO3– separately.

2.6. Data Analysis and Limit of Detection

Deviation analysis was used to evaluate the effect of different interfering substances (ions, food extracts) on the measurement. The deviation percentage error (α) is given by eq 4:

2.6. 4

where c0 and ci represent the concentration obtained by the proposed potentiometric method without and with interfering substance, respectively. According to the Nikolsky-Eisenman principle,44,46c0 corresponds to ai while ci is the sum of (ai + Kijaj) in eq 3. The α is related to the selectivity coefficient by Kij = αc0/aj (eq 3). We choose α here to clearly show the interference. To minimize the artifact of electrode response drift, measurements on samples with and without interfering ions taken at each reaction time were made immediately following one another. We considered α values of ≤10% to represent no significant effect of interfering matter.53

The limit of detection (LOD) is defined as the lowest detectable concentration of an analyte and is given by, LOD = 3.3 × SD/S, where SD is the standard deviation of the y-intercept and S is the slope of the regression line.

3. Results and Discussion

3.1. Construction of Calibration Curves and the Effect of pH

Figure 1 shows a calibration curve constructed from serial dilution of a stock solution of KFSO3 at 660 mM (65.4 g/L FSO3–, close to the limit of solubility) adjusted to pH 6.5 with KOH. It was divided into two linear regions, a common practice with selective electrodes.49,54,55 Between 0.0075 mM and 660 mM, the slope is −60.85 mV/decade, which is close to the theoretical slope of the Nikolsky-Eisenman equation (eq 3), and the r2 value is 0.997. Between 0.0025 mM and 0.0075 mM, the slope (−53.07 mV/decade) deviates from theoretical, and the fit is poorer (r2 = 0.935) (Figure 1, inset). We calculate the LOD to be 0.0007 mM and the limit of quantification to be 3.3 × LOD, or approximately 0.0023 mM (230 μg/L).

Figure 1.

Figure 1

Open in a new tab

Calibration curve for FSO3– solutions with the ion-selective electrode. The solution pH was ∼6.5 prior to mixing with the ISA. Linear fitting curves are shown between 0.0075 mM and 500 mM and between 0.0025 mM and 0.0075 mM (inset).

The FSO3– and NO3– calibration curves are compared in Figure S3. The FSO3– curve is slightly steeper than the NO3– curve (slope, −61.72 vs −57.22), but both are close to the theoretical slope. More importantly, NO3– gives an LOD (0.009 mM) about 10 times greater than the LOD for FSO3–. This means the nitrate-selective electrode is more selective for FSO3– than it is for NO3–.

The potentiometric response may depend on the pH of the sample before it is combined with the ISA due to the high mobility of hydrogen and hydroxide ions which will interfere the liquid-junction potential of the electrode.43,45,48,56 Confirming this (Figure 2), the slope of the fitting line ranges from −74.99 mV/decade at pH 3.0 to −40.86 mV/decade at pH 11.0. At the extreme pH values of 3.0 and 11.0, the fits are noticeably poorer, and the slopes deviate substantially from the theoretical value of −59 mV/decade compared to the corresponding fits and values at pH 5, 7, and 9. We conclude that the acceptable pH range of the sample is 5.0–9.0, and for the best results we recommend that the calibration pH should be close to the expected sample pH, or else the samples should be neutralized with KOH or H2SO4 before adding the ISA.

Figure 2.

Figure 2

Open in a new tab

Calibration curves of FSO3– at different solution pH values (3.0–11.0) prior to mixing with the ISA. The pH and corresponding r2 values are 3.0 (0.95), 5.0 (0.99), 7.0 (1.00), 9.0 (1.00) and 11.0 (0.98).

3.2. Potential Interferences

We tested the effects on electrode response of some ions that may coexist with FSO3– in applications, including halides, selected oxoanions, and selected metal ions. The results are shown in Figures 3, 4, S4, and S5. We considered α within ±10% to indicate noninterference (eq 4).

Figure 3.

Figure 3

Open in a new tab

Effect of different concentrations of halide ions on the potentiometric response of 0.05 mM, 0.50 mM, and 5.00 mM FSO3–. The solid symbols refer to the left axis while the open symbols refer to the right axis. The dotted lines represent 10% error (eq 4). Break on y axis in (a–d) ranges from10.5 to 10.501.

Figure 4.

Figure 4

Open in a new tab

Effects of some common oxoanions on the potentiometric response of 0.05 mM, 0.50 mM, and 5.00 mM FSO3–: (a) perchlorate at pH 6.8 prior to mixing with ISA; (b) bicarbonate at pH 8.2; (c) phosphate at pH 7.4; (d) pyrophosphate at pH 6.5. For (a,b), the solid symbols refer to the left axis while the open symbols refer to the right axis. Dotted lines represent 10% error (eq 4). For (c,d), open and solid symbols represent with or without addition of 10 mM MgSO4, respectively, and the α values are shown in Table S3.

3.2.1. Effects of Halide Ions

Fluoride is a potential interferant, as hydrolysis of both SO2F2 and FSO3– gives F– as a product. Figure 3a shows that F– does not interfere with FSO3– determination except at very low FSO3– and very high F– such that [F–] is at least 1000 times greater than [FSO3–]. Chloride is ubiquitous in environmental waters and tissues.57 Chloride added as KCl does not interfere except at very low FSO3– and very high Cl– such that [Cl–] is at least 100 times [FSO3–] (Figure 3b). The chloride counterion, Ca2+, Mg2+, or Li+, has no effect (Figure S4).

Bromide added as NaBr does not interfere until [Br–] exceeds 20–100 times [FSO3–] (Figure 3c). Iodide as KI strongly interferes if [I–] is more than twice the [FSO3–] (Figure 3d). In summary, interference by halides follows the order, F– < Cl – < Br– ≪ I–. According to the supplier of the electrode, interference by Cl –, Br–, and I– of nitrate analysis can be mitigated by adding silver sulfate to the solution. This was verified in the case of fluorosulfate (0.05 mM) for chloride (100 and 500 mM).

3.2.2. Effects of Oxoanions

Oxoanions of potential interest as potential interferents include nitrate, perchlorate, bicarbonate, sulfate, phosphate, and pyrophosphate. These ions may be present naturally in water samples or tissues, and their salts are commonly used as buffers or background electrolytes in chemical studies. Sulfate also is a recognized product of FSO3– and SO2F2 hydrolysis.37,38 The ISA introduces 0.04 M sulfate to all samples after mixing.

Despite being designed specifically for NO3–, the electrode is less sensitive to NO3– than FSO3– analysis, and NO3– added in equivalent concentration to FSO3– has no interference based on their separate potential response as shown in Figure S3. For example, at a potential of 196 mV, the concentration of NO3– (2.6 mM) is 52 times that of FSO3– (0.05 mM), while at 104 mV, the concentration of NO3– (89.1 mM) is 63 times that of FSO3– (1.41 mM). This means that NO3– would have to be much higher in concentration than FSO3– for them to reach the same response intensity.

Perchlorate is known to interfere with NO3– determination by this electrode, so it was not surprising to find that ClO4– strongly interferes with FSO3– determination, starting at concentrations as low as 20% that of FSO3– (Figure 4a). (Bi)carbonate at pH 8.2 begins to interfere when its concentration reaches only double that of FSO3– (Figure 4b). However, bicarbonate interference can be removed by prior acidification of the sample with sulfuric acid followed by gas sparging. Sulfate does not interfere at concentrations as high as 40 mM which is obtained anyway by adding the ISA. Phosphate at pH 7.4 strongly interferes even at 0.01 mM, but phosphate interference can be mitigated by addition of MgSO4 which induces its precipitation after mixing with the ISA, presumably as NH4MgPO4 (Figure 4c).58,59 The electrode response to FSO3– is less sensitive to pyrophosphate interference at pH 6.5 than to phosphate interference at pH 7.4. Similar to phosphate, pyrophosphate interference can be eliminated by addition of MgSO4 (Figure 4d). The electrode potentials in response to FSO3– were higher with than without MgSO4, which means that a separate calibration curve is needed if MgSO4 is added to mitigate interference by phosphate or pyrophosphate (Table S3).

A strategy we are investigating to remove SO2F2 from fumigation vent streams is hydrogen peroxide (H2O2)-assisted alkaline defluorination.60 In this reaction, H2O2 (as HO2–) displaces both F atoms as F– in successive steps and generates the oxo-anions, peroxymonosulfate (PMS; HOOSO3–), and peroxydisulfate (PDS; –O3SOOSO3–) as minor byproducts. We therefore tested for potential interference by PMS and PDS (Figure S5). The monoanion PMS began to interfere when its concentration exceeded 20-times that of FSO3–. The dianion PDS began to interfere when its concentration exceeded only twice that of FSO3–. As expected, H2O2 did not interfere (data not shown).

Given the results in Figures 3, 4, S4, and S5, the reader is alerted that oxoanions besides those tested here may interfere.

3.3. Validation and Practical Application

3.3.1. Comparison with Ion Chromatography

We attempted to compare the electrode method with an ion chromatographic (IC) method for FSO3– analysis. Nie et al.19 used a strong base quaternary ammonium anion exchange resin column to detect FSO3– as a base hydrolysis product of SO2F2. They assigned the peak eluting at about 11.4 min to SO42– and the peak at about 13.7 min to FSO3–, an elution order that seems counterintuitive due to the difference in charge. Few details were given, and the authors did not attempt to quantify yields by IC. However, using a similar column from the same supplier and the same eluent we found it was not possible to separate the peaks for SO42– and FSO3–, in each case appearing at about 13.3 min. Details are provided in Text S3 and Figure 5. The source of the discrepancy between our results and those of Nie et al.19 is unclear. In any case, it appears that SO42– has the potential to significantly interfere with FSO3– determination by IC.

Figure 5.

Figure 5

Open in a new tab

Ion chromatogram results for mixtures of F–/SO42–, F–/FSO3– and F–/FSO3–/SO42–. Concentrations respective FSO3– and SO42– are 0.10 mM in (a) and 0.5 mM in (b). For (a), areas of F– are 9.106, 9.319, and 9.159, and the corresponding peak areas at 13.4 min are 9.989, 5.545, and 15.278. For (b), areas of F– are 9.505, 9.370, and 9.369, and the corresponding peak areas at 13.3 min are 1.960, 1.008, and 3.103. F– of 0.10 mM was used as an internal standard for all samples.

3.3.2. Validation against a Second Product in Known Yield: The Case of Base Hydrolysis of SO2F2

The purpose of this set of experiments was to validate the electrode method by employing it in a reaction that generates FSO3– in a known ratio to a second product that can be determined independently. The reaction is strong base hydrolysis of SO2F2, which is reported to generate F– and FSO3– in a ratio slightly greater than 1 (eq 1). The ratio is not equal to 1 because FSO3– undergoes further slow hydrolysis to sulfate according to eq 2.52 Simultaneous measurement of FSO3– using the proposed nitrate-selective electrode and F– using the fluoride-selective electrode was performed during base hydrolysis of SO2F2 at pH 11.6 or 12.5 (Figure 6). The molar ratio of F– to FSO3– was found to be ∼1.1, slightly greater than the 1:1 stoichiometry of eq 1, and as anticipated. Thus, the developed method for FSO3– is validated against a second product in known yield. Furthermore, this reaction constitutes a realistic application of the method.

Figure 6.

Figure 6

Open in a new tab

Evolution of F– and FSO3– (left) and their ratio (right, blue curves) during base hydrolysis of SO2F2 at initial pH values of 11.6 and 12.5 in a sealed vial. Final pH values were 11.1 and 12.3, respectively. Injected SO2F2 volume is 4.0 mL, which corresponds theoretically to 1.6 mM SO2F2 in the liquid phase. The rate is faster at pH 12.5 because it is first order in hydroxide concentration.

3.3.3. Determination of Fluorosulfate in Food Tissues

The need for analytical tools to determine FSO3– residuals in food tissues will likely increase as SO2F2 use in the fumigation of foods expands.21,28 SO2F2 is absorbed by protein-rich stored products.61 To validate the use of the developed method under realistic conditions, FSO3– was spiked into the aqueous food extracts of 12 different commercially available foods including 11 vegetables and a dog food product. Recovery of FSO3– directly injected to vegetable tissue just prior to workup prior to workup was quantitative (Section 2).

Measured electrode potentials in the presence of different concentrations of tissue extracts are listed in Table S3, and the corresponding amounts of FSO3– (g) per gram tissue are listed in Table S4. The corresponding deviations (α) are shown in Figure 7. Generally, α decreases with increasing concentration of FSO3– in tissue extract (Figure 7).

Figure 7.

Figure 7

Open in a new tab

Electrode response deviation (α, 100%) from the control with an increasing amount of FSO3– in various fresh food tissues of (a) broccoli, cucumber, sweet onion, strawberries, carrots, yellow squash, and blueberries; (b) sweet potato, apple, kiwi, and cauliflower. The short dot line is where the deviation value is 10% (eq 4), above which this potentiometric method is considered as inappropriate.

The substances in the specific food extract have a significant effect on the FSO3– method detection limit. For broccoli, cucumber, sweet onion, strawberries, carrots, yellow squash, and blueberries, α remained below 10% above 4.95 × 10–5 (g FSO3–)/(g tissue) (Figure 7a). The method detection limit for these products thus lies in the range of 1.98–4.95 × 10–5 (g FSO3–)/(g tissue). For sweet potato, apple, and kiwi, the method detection limit lies in the range of 0.99–4.95 × 10–5 (g FSO3–)/(g tissue) (Figure 7b). Cauliflower extract had a higher method detection limit of 0.99–4.95 × 10–4 (g FSO3–)/(g tissue) (Figure 7a). Therefore, depending on the food, the detection limit for FSO3– may vary from 0.99 × 10–5 (g FSO3–)/(g tissue) to 4.95 × 10–4 (g FSO3–)/(g tissue). The method detection limit for FSO3– in dog food is 0.50–4.95 × 10–1 (g FSO3–)/(g tissue), which is much higher compared to the other foods (Figure 7 and Tables S3 and S4). The higher detection limit for dog food is probably because dog food has more soluble organic material and contains added ingredients.62 This is consistent with dog food extract having the strongest absorbance in the UV–visible region among the tissue extracts (Figure S2). It is best to prepare calibration solutions at the same pH as that of the food tissue extract. Overall, the developed method can be used for FSO3– determination in many food tissues.

4. Conclusions

We report a potentiometric method for quantification of FSO3–, a residual found in SO2F2-fumigated commodities and a byproduct of SO2F2 removal technologies, adopting a commercially available nitrate-selective electrode. The electrode is more sensitive to FSO3– than to NO3– in practice. It has a quantifiable range of 0.0025–660 mM FSO3–. Unlike ion chromatography, sulfate does not interfere. Selectivity for FSO3– is satisfactory in the presence of most halides, many oxyanions, and food tissue extracts. Although some buffer ions (bicarbonate, phosphate, and pyrophosphate) interfere, their interference can be mitigated by addition of other reagents (sulfuric acid for bicarbonate; magnesium sulfate for phosphate and pyrophosphate). The method appears to be suitable for analyzing FSO3– in many produce commodities but may be less suitable or unsuitable for complex processed foods like dog food.62 The method has been validated against the known ratio of FSO3– to F– byproducts formed by base hydrolysis of sulfuryl fluoride. Overall, the potentiometric method is rapid, convenient, and economical. The only equipment needed is a pH meter, which is available in portable models. We are currently using the method to monitor FSO3– in experimental systems designed to remove SO2F2 from spent fumigation vapors and convert it to harmless products.60

Acknowledgments

This project was funded by a grant (PN 21-02) from the U.S. Department of Agriculture FAS/TASK (Technical Assistance for Specialty Crops) Program managed by the California Prune Board (CPB).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c02629.

  • Text and tables on purity, recovery in plant tissues, and attempts to determine KFSO3 by ion chromatography. Figures on spectra of food extracts, calibration curves of FSO3– vs NO3–, and electrode response effects with respect to cations and persulfate ions (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao4c02629_si_001.pdf (442.6KB, pdf)

References

  1. Gillespie R. J. Fluorosulfuric acid and related superacid media. Acc. Chem. Res. 1968, 1, 202–209. 10.1021/ar50007a002. [DOI] [Google Scholar]
  2. Jache A. W. Fluorosulfuric Acid, Its Salts, and Derivatives. Adv. Inorg. Chem. Radiochem. 1974, 16, 177–200. 10.1016/S0065-2792(08)60292-7. [DOI] [Google Scholar]
  3. Bartmann K.; Mootz D. Structures of two strong broensted acids: (I) fluorosulfuric acid and (II) trifluoromethanesulfonic acid. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1990, 46, 319–320. 10.1107/S0108270189010395. [DOI] [Google Scholar]
  4. Engelbrecht A. Fluorosulfuric acid as a reaction medium and fluorinating agent in inorganic chemistry. Angew. Chem., Int. Ed. 1965, 4, 641–645. 10.1002/anie.196506411. [DOI] [Google Scholar]
  5. Revathi L.; Ravindar L.; Leng J.; Rakesh K. P.; Qin H. L. Synthesis and chemical transformations of fluorosulfates. Asian J. Org. Chem. 2018, 7, 662–682. 10.1002/ajoc.201700591. [DOI] [Google Scholar]
  6. Krespan C. G.; Dixon D. A. Fluorosulfonation. Insertion of sulfur trioxide into allylic carbon-fluorine bonds. J. Org. Chem. 1986, 51, 4460–4466. 10.1021/jo00373a023. [DOI] [Google Scholar]
  7. Smart B. E. A novel reaction of sulfur trioxide with fluoro olefins. J. Org. Chem. 1976, 41, 2353–2354. 10.1021/jo00875a032. [DOI] [Google Scholar]
  8. Wei M. J.; Liang D. C.; Cao X. H.; Luo W. J.; Ma G. J.; Liu Z. Y.; Li L. A broad-spectrum catalytic amidation of sulfonyl fluorides and fluorosulfates. Angew. Chem., Int. Ed. 2021, 133, 7473–7480. 10.1002/ange.202013976. [DOI] [PubMed] [Google Scholar]
  9. Luy J. N.; Tonner R. Complementary base lowers the barrier in SuFEx click chemistry for primary amine nucleophiles. ACS Omega 2020, 5, 31432–31439. 10.1021/acsomega.0c05049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dong J.; Krasnova L.; Finn M. G.; Sharpless K. B. Sulfur (VI) fluoride exchange (SuFEx): Another good reaction for click chemistry. Angew. Chem., Int. Ed. 2014, 53, 9430–9448. 10.1002/anie.201309399. [DOI] [PubMed] [Google Scholar]
  11. Qin H. L.; Zheng Q.; Bare G. A.; Wu P.; Sharpless K. B. A Heck–Matsuda process for the synthesis of β-arylethenesulfonyl fluorides: Selectively addressable bis-electrophiles for SuFEx click chemistry. Angew. Chem., Int. Ed. 2016, 55, 14155–14158. 10.1002/anie.201608807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lee C.; Cook A. J.; Elisabeth J. E.; Friede N. C.; Sammis G. M.; Ball N. D. The emerging applications of sulfur (VI) fluorides in catalysis. ACS Catal. 2021, 11, 6578–6589. 10.1021/acscatal.1c01201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ball N. D.; In Emerging Fluorinated Motifs: synthesis, Properties, and Applications, Vol. 2.; Properties and Applications of Sulfur (VI) Fluorides; Wiley-VCH Verlag GmbH & Co. KGaA: 2020; pp. 621–671.. [Google Scholar]
  14. Deng X. Y.; Zhu X. H. ″Recent Advances of S–18F Radiochemistry for Positron Emission Tomography″. ACS Omega 2023, 8, 37720–37730. 10.1021/acsomega.3c05594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Wang N. X.; Yang B.; Fu C. Y.; Zhu H.; Zheng F.; Kobayashi T.; Liu J.; Li S. S.; Ma C.; Wang P. G.; et al. Genetically encoding fluorosulfate-L-tyrosine to react with lysine, histidine, and tyrosine via SuFEx in proteins in vivo. J. Am. Chem. Soc. 2018, 140, 4995–4999. 10.1021/jacs.8b01087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cao A.; Guo M. X.; Yan D. D.; Mao L. G.; Wang Q. X.; Li Y.; Duan X. Y.; Wang P. S. Evaluation of sulfuryl fluoride as a soil fumigant in China. Pest Manag. Sci. 2014, 70, 219–227. 10.1002/ps.3535. [DOI] [PubMed] [Google Scholar]
  17. Grice J. D.; Gault R. A.; Chao G. Y. Reederite-(Y), a new sodium rare-earth carbonate mineral with a unique fluorosulfate anion. Am. Mineral. 1995, 80, 1059–1064. 10.2138/am-1995-9-1023. [DOI] [Google Scholar]
  18. U.S. EPA; RED (Reregistration Eligibility Document) Facts Sulfuryl Fluoride.; Office of Prevention. In Pesticides And Toxic Substances, EPA; 1993, Vol. 9. (H-7508W). EPA-738-F-93–012 [Google Scholar]
  19. Nie Y.; Liang X. J.; Ji J. B.; Lu M. Z.; Yu F. W.; Gu D. Y.; Xie Q. L.; Min M. Harmless treatment of sulfuryl fluoride by chemical absorption. Environ. Eng. Sci. 2015, 32, 789–795. 10.1089/ees.2015.0021. [DOI] [Google Scholar]
  20. Meikle R. W. Fumigant decomposition, fate of sulfuryl fluoride in wheat flour. J. Agr. Food Chem. 1964, 12, 464–467. 10.1021/jf60135a022. [DOI] [Google Scholar]
  21. Scheffrahn R. H.; Hsu R. C.; Osbrink W. L. A.; Su N. Y. Fluoride and sulfate residues in foods fumigated with sulfuryl fluoride. J. Agr. Food Chem. 1989, 37, 203–206. 10.1021/jf00085a046. [DOI] [Google Scholar]
  22. Derrick M. R.; Burgess H. D.; Baker M. T.; Binnie N. E. Sulfuryl fluoride (Vikane): A review of its use as a fumigant. J. Am. Inst. Conserv. 1990, 29, 77–90. 10.1179/019713690806046082. [DOI] [Google Scholar]
  23. Hwaidi M.; Collins P. J.; Sissons M.; Pavic H.; Nayak M. K. Sorption and desorption of sulfuryl fluoride by wheat, flour and semolina. J. Stored Prod. Res. 2015, 62, 65–73. 10.1016/j.jspr.2015.04.005. [DOI] [Google Scholar]
  24. Gressent A.; Rigby M.; Ganesan A. L.; Prinn R. G.; Manning A. J.; Mühle J.; Salameh P. K.; Krummel P. B.; Fraser P. J.; Steele L. P.; et al. Growing atmospheric emissions of sulfuryl fluoride. J. Geophys. Res.: Atmos. 2021, 126, e2020JD034327 10.1029/2020JD034327. [DOI] [Google Scholar]
  25. Mühle J.; Huang J.; Weiss R. F.; Prinn R. G.; Miller B. R.; Salameh P. K.; Harth C. M.; Fraser P. J.; Porter L. W.; Greally B. R.. et al. Sulfuryl fluoride in the global atmosphere. J. Geophys. Res.: Atmos. 2009, 114. 10.1029/2008JD011162. [DOI] [Google Scholar]
  26. Friedemann A. E. R.; Andernach L.; Jungnickel H.; Borchmann D. W.; Baltaci D.; Laux P.; Schulz H.; Luch A. Phosphine fumigation - Time dependent changes in the volatile profile of table grapes. J. Hazard. Mater. 2020, 393, 122480. 10.1016/j.jhazmat.2020.122480. [DOI] [PubMed] [Google Scholar]
  27. Hasan M. M.; Aikins M. J.; Schilling M. W.; Phillips T. W. Sulfuryl fluoride as a methyl bromide alternative for fumigation of Necrobia rufipes (Coleoptera: Cleridae) and Tyrophagus putrescentiae (Sarcoptiformes: Acaridae), major pests of animal-based stored products. J. Stored. Prod. Res. 2021, 91, 101769. 10.1016/j.jspr.2021.101769. [DOI] [Google Scholar]
  28. Sriranjini V. R.; Rajendran S. Sorption of sulfuryl fluoride by food commodities. Pest Manag. Sci. 2008, 64, 873–879. 10.1002/ps.1577. [DOI] [PubMed] [Google Scholar]
  29. Meikle R. W.; Stewart D. Structural fumigants, the residue potential of sulfuryl fluoride, methyl bromide, and methanesulfonyl fluoride in structural fumigations. J. Agric. Food Chem. 1962, 10, 393–397. 10.1021/jf60123a011. [DOI] [Google Scholar]
  30. Erickson B. E. Fumigant under scrutiny for greenhouse gas emissions. C&EN 2022, 100, 17. 10.47287/cen-10039-polcon1. [DOI] [Google Scholar]
  31. Evans J.; Getz C.; Sellen J.; 2022/10/27. Petition to Regulate Sulfuryl Fluoride to Reduce the Use of the High Global Warning Potential Pesticide-A Legal Petition. [Google Scholar]
  32. Papadimitriou V. C.; Portmann R. W.; Fahey D. W.; Mühle J.; Weiss R. F.; Burkholder J. B. Experimental and theoretical study of the atmospheric chemistry and global warming potential of SO2F2. J. Phys. Chem. A. 2008, 112, 12657–12666. 10.1021/jp806368u. [DOI] [PubMed] [Google Scholar]
  33. Nie Y.; Zheng Q. F.; Liang X. J.; Gu D. Y.; Lu M. Z.; Min M.; Ji J. B. Decomposition treatment of SO2F2 using packed bed DBD plasma followed by chemical absorption. Environ. Sci. Technol. 2013, 47, 7934–7939. 10.1021/es400786p. [DOI] [PubMed] [Google Scholar]
  34. Tsai W. T. Environmental and health risks of sulfuryl fluoride, a fumigant replacement for methyl bromide. J. Environ. Sci. Health, Part C 2010, 28, 125–145. 10.1080/10590501.2010.481806. [DOI] [PubMed] [Google Scholar]
  35. Dong Q. X.; Lewis C. M.; Koshlukova S.; California Department of Pesticide Regulation Memorandom. In Establishing Sulfuryl Fluoride Uncertainty Factors for Acute and Short-term Exposures; Department of Pesticide Regulation, 2017. [Google Scholar]
  36. Zou H. L.; Chen P. P.; Peng C.; Liang C. D.; Lithium ion battery, battery module, battery pack and power consuming device, Contemporary Amperex Technology CO, 2023. [Google Scholar]
  37. Jones M. M.; Lockhart W. L. Kinetics of decomposition of the fluorosulphate ion in aqueous solution. J. Inorg. Nucl. Chem. 1968, 30, 1237–1243. 10.1016/0022-1902(68)80550-0. [DOI] [Google Scholar]
  38. Lawrance G. A. Coordinated trifluoromethanesulfonate and fluorosulfate. Chem. Rev. 1986, 86, 17–33. 10.1021/cr00071a002. [DOI] [Google Scholar]
  39. Michałowski T.; Malinowski P. J.; Grochala W. Synthesis, crystal structures, and selected properties of metal fluorosulfates(VI). J. Fluor. Chem. 2016, 189, 102–118. 10.1016/j.jfluchem.2016.07.024. [DOI] [Google Scholar]
  40. Alleyne C. S.; Thompson R. C. Coordinating Properties of the fluorosulfate ion. Tetrakis (pyridine) complexes of zinc (II), copper (II), and nickel (II) fluorosulfates. Can. J. Chem. 1974, 52, 3218–3228. 10.1139/v74-474. [DOI] [Google Scholar]
  41. Kodani S. D.; Biochemical toxicology of fluorosulfate, an intermediary metabolite of the fumigant sulfuryl fluoride. In Disseration; Spring: University of California, Berkeley, 2012. [Google Scholar]
  42. U.S. EPA Reviewer; Data Evaluation Record of Sulfuryl Fluoride/078003: Neurotoxicity and Toxicokinetics – Rat; EPA, 2015. [Google Scholar]
  43. Arnold M. A.; Meyerhoff M. E. Ion-selective electrodes. Anal. Chem. 1984, 56, 20–48. 10.1021/ac00269a003. [DOI] [Google Scholar]
  44. Fakih I.; Durnan O.; Mahvash F.; Napal I.; Centeno A.; Zurutuza A.; Yargeau V.; Szkopek T. Selective ion sensing with high resolution large area graphene field effect transistor arrays. Nat. Commun. 2020, 11 (1), 3226. 10.1038/s41467-020-16979-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Hasan N.; Kansakar U.; Sherer E.; DeCoster M. A.; Radadia A. D. Ion-selective membrane-coated graphene–hexagonal boron nitride heterostructures for field-effect ion sensing. ACS Omega 2021, 6, 30281–30291. 10.1021/acsomega.1c02222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Buck R. P.; Lindner E. Recommendations for nomenclature of ionselective electrodes (IUPAC Recommendations 1994). Pure Appl. Chem. 1994, 66, 2527–2536. 10.1351/pac199466122527. [DOI] [Google Scholar]
  47. Wang A. L.; Jiang Y.; Yan Y. Q.; Bu L. J.; Wei Z. S.; Spinney R.; Dionysiou D. D.; Xiao R. Y. Mechanistic and quantitative profiling of electro-Fenton process for wastewater treatment. Water Res. 2023, 235, 119838. 10.1016/j.watres.2023.119838. [DOI] [PubMed] [Google Scholar]
  48. Bailey P. L., 2nd. ed.Analysis with Ion Selective Electrodes. Heyden & Son Ltd,: London, 1976. [Google Scholar]
  49. Chen X. V.; Mousavi M. P. S.; Bühlmann P. Fluorous-phase ion-selective pH electrodes: Electrode body and ionophore optimization for measurements in the physiological pH range. ACS Omega 2020, 5, 13621–13629. 10.1021/acsomega.0c00582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Dura B.; Choi J. Y.; Zhang K.; Damsky W.; Thakral D.; Bosenberg M.; Craft J.; Fan R. scFTD-seq: Freeze-thaw lysis based, portable approach toward highly distributed single-cell 3′ mRNA profiling. Nucleic Acids Res. 2019, 47, e16 10.1093/nar/gky1173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Chen Z. H.; Li J. Y.; Chen M. Q.; Koh K. Y.; Du Z. R.; Gin K. Y.; He Y. L.; Ong C. N.; Chen J. P. Microcystis aeruginosa removal by peroxides of hydrogen peroxide, peroxy monosulfate and peroxydisulfate without additional activators. Water Res. 2021, 201, 117263. 10.1016/j.watres.2021.117263. [DOI] [PubMed] [Google Scholar]
  52. Cady G. H.; Misra S. Hydrolysis of sulfuryl fluoride. Inorg. Chem. 1974, 13, 837–841. 10.1021/ic50134a016. [DOI] [Google Scholar]
  53. Causey A. G.; Hill H. M.; Phillips L. J. Evaluation of criteria for the acceptance of bioanalytical data. J. Pharm. Biomed. Anal. 1990, 8 (8–12), 625–628. 10.1016/0731-7085(90)80092-4. [DOI] [PubMed] [Google Scholar]
  54. Mohamed G. G.; El-Shahat M. F.; Al-Sabagh A. M.; Migahed M. A.; Ali T. A. Septonex–tetraphenylborate screen-printed ion selective electrode for the potentiometric determination of Septonex in pharmaceutical preparations. Anal 2011, 136, 1488–1495. 10.1039/c0an00662a. [DOI] [PubMed] [Google Scholar]
  55. Paul A.; Nair R. R.; Chatterjee P. B.; Srivastava D. N. Fabrication of a Cu (II)-Selective electrode in the polyvinyl chloride matrix utilizing mechanochemically synthesized rhodamine 6g as an ionophore. ACS Omega 2018, 3, 16230–16237. 10.1021/acsomega.8b02870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Xu K. B.; Li Y.; Li M. Potentiometric phosphate ion sensor based on electrochemical modified tungsten electrode. ACS Omega 2021, 6, 13795–13801. 10.1021/acsomega.1c00195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Tratnyek P. G.; Grundl T. J.; Haderlein S. B.; Aquatic redox chemistry, American Chemical Society, 2011. [Google Scholar]
  58. Sugiyama S.; Yokoyama M.; Ishizuka H.; Sotowa K. I.; Tomida T.; Shigemoto N. Removal of aqueous ammonium with magnesium phosphates obtained from the ammonium-elimination of magnesium ammonium phosphate. J. Colloid Interface Sci. 2005, 292, 133–138. 10.1016/j.jcis.2005.05.073. [DOI] [PubMed] [Google Scholar]
  59. Lee S. I.; Weon S. Y.; Lee C. W.; Koopman B. Removal of nitrogen and phosphate from wastewater by addition of bittern. Chemosphere 2003, 51, 265–271. 10.1016/S0045-6535(02)00807-X. [DOI] [PubMed] [Google Scholar]
  60. Pignatello J. J.; Wang C. J.; Chen Z. H.. Method for Scrubbing Sulfuryl Fluoride from A Fluid 18/703097US, 2024 [Google Scholar]
  61. Austel N.; Schubert J.; Gadau S.; Jungnickel H.; Budnik L. T.; Luch A. Influence of fumigants on sunflower seeds: Characteristics of fumigant desorption and changes in volatile profiles. J. Hazard. Mater. 2017, 337, 138–147. 10.1016/j.jhazmat.2017.04.070. [DOI] [PubMed] [Google Scholar]
  62. Raditic D. M.; Remillard R. L.; Tater K. C. ELISA testing for common food antigens in four dry dog foods used in dietary elimination trials. J. Anim. Physiol. Anim. Nutr. 2011, 95, 90–97. 10.1111/j.1439-0396.2010.01016.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao4c02629_si_001.pdf (442.6KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES