Efficient Decontamination of HD by an Electrophilic Iodine/Carboxylate Composite as an Active Sorbent.

The development of new and efficient decontamination methods has become more relevant in recent years, especially with regard to solid-based decontamination and detoxification systems. The majority of powders used today are dealing with the physical adsorption of chemical warfare agents (CWAs) and their removal from sites without actively destroying them. In this work, we have designed and developed an active solid composite matrix combining organic carboxylate salts and N-iodosuccinimide (NIS) for HD decontamination via oxidation. All the reactions and mechanistic studies for the sorption and degradation of CWAs were conducted using direct polarization and cross polarization solid-state magic-angle spinning nuclear magnetic resonance techniques. Performance toward the sorption and detoxification of HD was tested, exhibiting oxidation within minutes in a mild and selective manner to the nontoxic sulfoxide derivative followed by visible formation of iodine. The results indicate that carboxylate moieties in the matrix are important for stabilizing the positively charged sulfonium ion intermediate and for supplying oxygen for hydrolysis in a water-deficient environment. The NaOBz/NIS composite was shown to be the most efficient in sorbing and converting the water-insoluble agent HD to its nontoxic, water-soluble sulfoxide, which could then be removed from the site with mere water, resulting in less environmental damage and quick remediation.

Introduction

Sulfur mustard (HD, 1, Figure ) was first employed by the Germans during World War I in 1917 as an offensive weapon against the British at Ypres, Belgium. Over the course of the 20th century, several attacks using HD against soldiers and civilians were reported. Infamous, the Iraqi attack on the Kurdish population of Halbja in 1988 resulted in many civilian casualties.[1] Past events have demonstrated the toxicity of sulfur mustard on humans, which causes local effects on the eyes and respiratory tissues and broad lesions on the skin along with systemic effects on the nervous, cardiovascular, and digestive systems. Known for its vesicant properties, HD is considered to be a blister agent. Due to its high toxicity, chemical stability, and environmental persistence (nonvolatile and water-insoluble), HD is broadly stockpiled.[2−5] Numerous detoxification approaches have been developed over the past decades to deal with this threat, most of which are based on hydrolysis/elimination processes that commonly demand strong bases and harsh conditions, while other approaches based on oxidation typically use corrosive oxidants, such as hypochlorite and STB (Super Tropical Bleach).[6,7] The lack of selectivity of some of these decontaminants can lead to the formation of toxic side products. HD overoxidation, for example, can result in the formation of the sulfone vesicant HDO2, 3, rather than the desired nonvesicant sulfoxide HDO, 2 (Figure ). Furthermore, since most of the aforementioned decontaminants are water-soluble or aqueous formulations, devising decontamination procedures for hydrophobic substances, such as HD, requires special attention.

Figure 1

HD and common oxidation products.

Another method of CWA decontamination is the use of powders. In recent years, an increasing number of studies have been conducted on reactive solids and their ability to destroy CWAs. Systems based on various materials like activated carbon,[8] zeolites,[9] POMs,[10−12] MOFs,[13,14,23,15−22] inorganic oxides such as alumina (Al2O3),[24−30] silica (SiO2),[31] and titania (TiO2),[32−34] and zirconium compounds[35−38] were introduced to the field of CWA decontamination. Due to their high surface area and unique physical and chemical properties that include a large number of basic and acidic sites, these systems can significantly accelerate hydrolysis reactions. Conversely, only a handful of examples have been shown to act via oxidation, most of which are catalyzed by photoinduced singlet oxygen formation.[18,19,39,40] Therefore, we sought to develop an active powder that would contain and degrade lipophilic CWAs via a selective oxidation pathway. In our previous study, we demonstrated that N-iodosuccinimide (NIS) can efficiently detoxify sulfur-containing CWAs (HD and VX) by oxidizing the sulfur to the corresponding iodo-sulfonium ion, which further reacts with water to give solely nontoxic products.[41] We postulated that this type of transformation could be carried out on solid supports containing carboxylate moieties, which are able to catalyze the hydrolysis of sulfonium ions to the corresponding sulfoxides as demonstrated by Higuchi et al.[42] and Nagy and coworkers.[43,44] In this work, we present the design and preparation of solid composite matrices composed of organic salts with carboxylate functional groups and NIS (RCOO–Na+/NIS) and their performance toward the sorption and detoxification of HD. Solid-state MAS NMR direct polarization (DP) and cross polarization (CP) techniques were applied in order to investigate the interactions between the liquid (CWA) and solid (sorbent) phases and to shed light on the mechanisms of detoxification.

Experimental Section

Caution! These experiments should only be performed by trained personnel using applicable safety procedures.

Chemicals

HD* (13C labeled 1) was obtained locally at IIBR (>99% purity). This compound contains only one labeled carbon atom per chloroethyl arm, and thus, undesired 13C–13C couplings are avoided (SI, Figures S1 and S2). N-Iodosuccinimide (NIS), dibutylsulfide, sodium polyacrylate (NaPA, MW of ∼5000), sodium acetate (NaOAc), sodium trifluoroacetate (NaTFA), sodium benzoate (NaOBz), and all solvents (HPLC grade) were purchased from commercial suppliers and used without further purification. Deionized water was obtained from a laboratory water purification system.

NMR Spectroscopy

Solution 1H and 13C{1H} NMR spectra were obtained at room temperature at 500 and 125 MHz, respectively, on an 11.7 T (500 MHz) Bruker spectrometer (Avance III HD). Chemical shifts were calibrated to TMS as 0 ppm. The spectra were recorded using the standard parameters of TopSpin NMR software (version 3.5). 13C experiments were carried out with a zgpg30 pulse program using 64 scans, with a spectral width of 240 ppm, a relaxation delay of 2 s, and an acquisition time of 1.1 s. 1H experiments were carried out with a zg30 pulse program using 16 scans, with a spectral width of 20 ppm, relaxation delay of 2 s and acquisition time of 3.27 s.

Solid-State NMR Spectroscopy

13C MAS NMR measurements were carried out on an 11.7 T (500 MHz) Bruker spectrometer (Avance III HD) equipped with a 4 mm standard cross polarization (CP) magic-angle spinning (MAS) probe with zirconia rotors. Samples were spun at 5 kHz. CP MAS experiments via the Hartmann–Hahn matching condition were carried out with RF levels of the X channel (13C) set to 65 kHz, while the RF level of the 1H channel was ramped between 50 and 100 kHz, with a contact time of 2000 μs. Experiments were repeated using a relaxation delay of 5 s, 128 scans, an acquisition time of 0.05 s, and a spectral width of 300 ppm. Direct polarization (DP) experiments were carried out using a zg pulse. For each spectrum, 110 scans were collected. The acquisition time was set to 0.05 s. Experiments were repeated with relaxation delays of 5 s and a spectral width of 300 ppm.

SEM/EDS Analysis

Surface morphologies were obtained using a Phenom Pro scanning electron microscope (SEM) (Phenom Company, The Netherlands). EDS elemental compositions were determined using a Quanta 200 FEG SEM equipped with an EDS detector (EDAX, Ametek, The Netherlands).

Preparation of NIS Composite Systems

General Procedure (NIS, 17% w/w)

A solution of NIS (100 mg) in ACN (4 mL) was added to the relevant sodium salt (500 mg). In the case of NaOBz, the solid was ground using a mortar and pestle prior to the addition of the solution. The suspension was stirred well, and the solvent was evaporated under vacuum using a rotary evaporator. The solid composite obtained was ground to a fine powder.

NaI Test

In order to test the composite obtained, a small quantity was dissolved in water, and the NaI salt was added. In all cases, iodine was formed, indicating the presence of reactive NIS in the composite.

Application of HD: MAS NMR Measurements

In a typical experiment, the powder was charged into a 4 mm MAS zirconia rotor. One microliter (1 μL) of HD* was applied on top of the powder, and the rotor was capped with a Kel-F plug. The reaction was monitored at room temperature by 13C MAS NMR over several periods of time.

Dibutylsulfide Simulant Experiments

NaOBz (197 mg, 1.37 mmol), NIS (307.5 mg, 1.37 mmol), and dry acetonitrile-d3 (ampoule, 0.75 mL × 2) were placed in a 4 mL vial. To the mixture was added dibutylsulfide (23.9 μL, 0.137 mmol). The mixture was filtered through a cotton plug into an NMR tube, and the reaction was monitored by 13C NMR. A control experiment was conducted in the same manner sans NaOBz and filtration. The supernatant was transferred into an NMR tube, and the reaction was monitored by 13C NMR. After identification of the sulfonium species, 100 μL of water was added to the NMR tube, and the reaction was monitored by 13C NMR.

Water Removal Experiments

A 4 mL vial was charged with NaOBz/NIS (82 mg). HD* (1 μL) was applied to the vial bottom and then covered with composite powder. A control experiment was conducted under the same conditions with NaOBz powder (83 mg). The vials were capped and left to rest overnight. Water (1 mL) was added, and the vial was stirred well until all components were dissolved. The solution was transferred into an NMR tube and was monitored by 13C NMR.

Results and Discussion

As part of this research, we sought to determine which properties govern the processes of HD sorption and degradation on various solid carboxylic supports. We chose the following carboxylate salts to be used as a support in the solid matrices: sodium acetate (NaOAc), sodium trifluoroacetate (NaTFA), sodium polyacrylate (NaPA) as a polymeric form of acetate, and sodium benzoate (NaOBz), which introduces an aromatic moiety. Solid-based NIS composites were prepared, characterized, and tested. SEM analysis of the composites showed a clear change in the morphology compared to the salt support (Figure A–C; NaTFA/NIS was not measured due to hygroscopicity). EDS measurements showed a strong iodine signature indicating an integrated NIS within the solid matrix (SI, Figures S3–S8). Using 13C coupled experiments, we were able to track the agent fate and to study the mechanisms of degradation on the different composites. 13C solid-state NMR spectra were obtained using direct polarization (DP) and cross polarization (CP) techniques. DP is based on C-nuclide spin lattice relaxation, therefore favoring more mobile systems, while CP is based on 1H-X heteronuclear dipolar interactions, therefore more sensitive to internuclear distances and favoring a more rigid system. The DP method was used to evaluate the sorption magnitude of 1 on the solid matrices, and the half-height width of the signal was determined in Hz. In the case of sorption, line broadening is correlated with molecular-level motion restriction due to anisotropic interactions.[45−47]

Figure 2

SEM images of the NIS-free support vs the NaSalt/NIS composites (scale bar of 100 μm). (A1) Neat NaPA salt; (A2) NaPA/NIS composite; (B1) neat NaOAc salt; (B2) NaOAc/NIS composite; (C1) neat NaOBz salt; (C2) NaOBz/NIS composite.

Interaction of HD, 1, with Sodium Carboxylate Salts (NIS-Free)

The sorption measurements of 1 were performed on the salt supports free of the NIS oxidant. The resonance signals of 13C-enriched 1 appear as two sets of triplets in DP MAS NMR spectra (Figure A1–D1 controls). The signal at 37 ppm, which corresponds to the methylene adjacent to the sulfur in HD, was deconvoluted in order to better measure the half-height width [Hz]. Using this parameter, we could estimate the degree of sorption on the surface. The broader the peak, the higher the sorption. The largest line broadening was seen in NaOBz (Figure C1 control, 136.6 Hz), while NaPA (Figure A1 control, 40 Hz) and NaOAc (Figure B1 control, 39 Hz) showed the smallest line broadening. Due to its more lipophilic nature, NaTFA was a slightly better sorbent of 1 (Figure D1 control, 49 Hz) than its nonfluorinated analogues NaOAc and NaPA. This indicates that NaOBz is a better sorbent of HD than the tested acetate-based sodium salts.

Figure 3

13C MAS NMR spectra (DP/CP) of HD on different salt/NIS composites over time. (A1) DP NMR of the NaPA composite; (A2) CP NMR of the NaPA composite; (B1) DP NMR of the NaOAc composite; (B2) CP NMR of the NaOAc composite; (C1) DP NMR of the NaOBz composite; (C2) CP NMR of the NaOBz composite; (D1) DP NMR of the NaTFA composite; (D2) CP NMR of the NaTFA composite.

Oxidative Degradation of HD, 1, with NaSalt/NIS Composites

In our previous study, we found that oxidation of 1 by NIS occurs via the postulated sulfonium intermediate followed by iodine release and color change (Scheme , species 4).[41] The oxidation step is fast, and the rate-determining step is the acyl transfer of this species to the corresponding sulfoxide 2.

Scheme 1

Postulated Oxidation Pathway of HD with NIS on a Carboxylate Solid Support

Monitoring the reactions by MAS NMR, the four NIS composite systems exhibited different reactivities toward the oxidation of 1. Applying 1 on the NaPA/NIS composite, 13C DP MAS NMR measurements showed a reduced intensity and broadening of the triplet signal at 37 ppm (carbons adjacent to the sulfur atom) relative to the triplet signal at 47 ppm (carbons at the β position to the sulfur) (Figure A1). Furthermore, the corresponding signals also appeared in the CP spectra (Figure A2), indicating a more limited atomic-level motion as a result of sorption. These results can be explained by the oxidation of 1 and the formation of the postulated iodo-sulfonium intermediate 4 (Scheme ). Species 4, which has a positive charge, is stabilized by the negatively charged carboxylate groups, leading to iodide displacement and chemisorption of species 5 to further confine the motion of the α-carbon of the sulfonium. Similar sorption dynamics was observed with NaOBz/NIS and NaTFA/NIS composites (Figure C,D, respectively). However, with the NaOAc/NIS composite, no apparent changes in both DP and CP MAS NMR spectra were observed, indicating that no significant reaction took place. The dynamics of these species (1, 5, and 2) appear to vary on the different solid supports. This was more prominent with the NaTFA/NIS composite (Figure D1,D2). In this system, 1 itself does not show significant sorption on the surface of the NIS-free solid support (Figure D1,D2, control), while fast sorption of oxidized species 5 on the NIS composite was observed in the CP spectrum (Figure D2, 88 min), correlated with the broadening of the triplet at 37 ppm in the DP spectrum (Figure D1, 84 min). By the end of the measurement, species 5 was not observed in DP, suggesting that the reaction was complete (Figure D1, 1383 min). However, species 5 was visible as the sorbed form in CP (Figure D2, 1335 min). This example demonstrates strongly how the CP method is a valuable complement to the DP method, enabling us to acquire important information on the kinetics and dynamics of species involved in reactions on solid matrices.

Comparatively to the other tested composites, NaOBz/NIS afforded the best results for the sorption and oxidation of 1, the sorption of species 5, and the hydrolysis to the corresponding sulfoxide product 2 (Figure C1).

It is important to emphasize that the oxidation of HD in all the reactive composites (with the exception of NaOAc/NIS) occurs in minutes (complete before the NMR measurement starts), and therefore, there is no HD present. The transition from oxidized species 4 to 5 is slow and evolves over time, as can be seen visually by the iodine formation (Figure ) and the increasing intensity of the signals of species 5 in CP experiments (Figure A2,D2) due to sorption.

Figure 4

Left to right: NaOBz/NIS (no HD); NaOBz/NIS+HD (15 min); NaOBz/NIS+HD (45 min).

The results demonstrated that the ability to induce oxidative degradation of 1 is dependent on the lipophilicity of the composite, which can be represented by the partition coefficient of the corresponding conjugated acids of the sodium salt support (Table , log P). The NaOAc/NIS composite has the lowest partition coefficient of all sodium salts tested, and no reaction has been observed (Table , entry 1). The NaOAc/NIS matrix seems to be too polar to induce sorption interactions that could lead to significant reactivity. NaPA/NIS, in contrast to NaOAc/NIS, is sufficiently lipophilic to allow the sorption of 1 and to induce a reaction (Table , entry 2). In this case, the NMR spectra were very indicative compared to the control spectrum (Figure A1). In the NaTFA/NIS system, the sorption and reaction trends of 1 were consistent with the lipophilicity of the system, which facilitates the fast reaction and the binding of the oxidized species 5 (Table , entry 3). Among the composites tested, the NaOBz/NIS composite, with the highest lipophilicity, provided the best results for the sorption and oxidation of 1 (Table , entry 4). Based on these results, the sorption of 1 on the NIS-free salts and the reactivity of the NaSalt/NIS composites toward the oxidation of 1 are in good agreement with the lipophilic nature of the sodium salt support.

Table 1

Sorption of 1 and 5 on a Carboxylate Salt Support (DP MAS NMR Measurements)

entry	salt	log P (corresponding acid)a	salt (no NIS) 1 (control)b [Hz]	NaSalt/NIS species 5b [Hz]
1	NaOAc	–0.28	39	no reaction
2	NaPA	0.26 (acid polymer)	40	69
3	NaTFA	0.5	49	84
4	NaOBz	1.87	137	166c

From databases.

Triplet deconvolution in the 37 ppm region. Half-height width in Hz. Up to 1.5 h from addition of 1.

Due to product overlap and difficulty in deconvolution, the triplet at 47 ppm was measured.

Carboxylate-Assisted Hydrolysis of HD

The carboxylate moieties in the matrix have a dual mode of action: first, as mentioned above, they stabilize the positively charged sulfonium ion in species 4, and second, in a water-deficient environment such as our composite systems, the carboxylate itself can provide oxygen for the formation of sulfoxide via a postulated intermediate 5 and the formation of anhydride 6 (Scheme ).

A similar mechanism was previously reported by Higuchi et al.[42] and Young and Hsieh[48] concerning the ability of carboxylates to catalyze the iodine-generated iodo-sulfonium species and its hydrolysis to the corresponding sulfoxide in aqueous solutions (Scheme A,B). To better understand the species generated in the mechanism of sulfide oxidation and to determine the carboxylate ability to act as an oxygen source, we conducted a series of experiments under dry conditions. In order to avoid safety procedures, we used dibutylsulfide as a simulant for HD, which was reacted with NIS in the presence of NaOBz in dry acetonitrile. The color changed immediately due to iodine emission, indicating that oxidation had occurred (Scheme C). Monitoring the reaction by 13C NMR revealed that dibutylsulfide (Figure A) had been fully converted into the corresponding sulfoxide (Figure D, Bu2S=O). In the control experiment without NaOBz, a new species was identified with distinct four new peaks (Figure B, “Bu2S+”) attributed to the sulfonium succinimide formation in this system (Scheme C).[49] Some amount of the sulfoxide product was present as a result of a small amount of water in the system; however, no further changes in the reaction were observed over time (Figure B, Bu2S=O). Addition of water led to full hydrolysis and the formation of the corresponding dibutylsulfoxide (Figure C). These results clearly demonstrate that the carboxylate groups can serve as oxygen donors under water-deficient conditions and assist in the acyl transfer of sulfonium ions to the corresponding sulfoxide.

Scheme 2

(A–C) The Role of Carboxylates in the Oxidation Mechanism of Sulfoxides with Electrophilic Iodine

(A) General reaction of sulfonium with carboxylates in aqueous solutions. (B) General reaction of sulfonium with dicarboxylate in aqueous solutions. (C) Suggested sulfonium reaction mechanism on a carboxylate solid support (this work).

Adapted with permission from ref (48). Copyright 1982 American Chemical Society.Adapted with permission from ref (42). Copyright 1968 American Chemical Society.

Figure 5

13C NMR spectra of dibutylsulfide (Bu2S) oxidation with NIS under dry and hydrolytic conditions (in ACN-d3). (A) Neat Bu2S; (B) control: Bu2S with NIS under dry conditions; (C) control experiment (B) with added water; (D) Bu2S with NaOBz and NIS under dry conditions.

Water Removal of the NaOBz/NIS Matrix Containing Detoxified HD

Following the successful sorption and detoxification of 1 by NaOBz/NIS, we continued to investigate the possibility of using our composite systems in a more realistic scenario. We reacted NaOBz/NIS with 1 for 24 h, following full dissolution of the contaminated solids in water, simulating onsite removal by water. The solution was taken immediately to NMR analysis.for analysis. Full and highly selective detoxification was observed yielding the nonvesicant oxidation product 2 as the sole product, without the overoxidation product 3 or halogenation of the ethylene groups (SI, Figures S10 and S11). Based on these results, the NaOBz/NIS composite is not only sufficiently lipophilic to sorb and decontaminate HD but also hydrophilic enough to dissolve in water, allowing rapid and easy environmental remediation.

As an added benefit, the proposed composite contains a mild oxidant (NIS) and a biodegradable salt (NaOBz), which is nonvolatile, easy to store, environmentally benign, and biodegradable in comparison to most detoxification procedures currently used, which are mostly composed of organic solvents, corrosive reagents, and environmentally toxic and hazardous materials.

Conclusions

Four powders composed of organic carboxylate salts with NIS were prepared and evaluated for their ability to sorb and detoxify HD. With the exception of NaOAc/NIS, all of these composites reacted within minutes with HD. Accordingly, the lipophilicity of the salt support has a dominant impact on the reaction and thus affects the NIS composites’ ability to decontaminate HD, namely, the more lipophilic the salt support, the better the reactivity of the composite: NaOBz/NIS ≪ NaTFA/NIS > NaPA/NIS. Using both DP and CP NMR techniques enabled us to follow the sorption of HD on the matrix and determine the fate of the agent. This study also highlights the role of the carboxylate moiety in the matrix, which stabilizes the sulfonium ion intermediate and provides oxygen for hydrolysis in a water-deficient environment. In this work, we have developed a system that can sorb and efficiently react with the water-insoluble agent HD, converting it to its nontoxic, water-soluble sulfoxide, which can then be removed from the site simply by using water, resulting in less environmental damage and quick remediation.

Currently, we are evaluating the efficiency of our NIS composite systems for decontamination of VX and other hazardous chemicals, and the results will be reported in the near future.