Decontamination and Remediation of the Sulfur Mustard Simulant CEES with “Off‐the‐Shelf” Reagents in Solution and Gel States: A Proof‐of‐Concept Study

Abstract The decontamination and remediation of sulfur mustard chemical warfare agents remains an ongoing challenge. Herein, we report the use of “off‐the‐shelf” metal salts alongside commercially available peroxides to catalyze the degradation of the simulant 2‐chloroethyl ethyl sulfide (CEES) in solution and encapsulated within a supramolecular gel.


Experimental
General remarks: All reactions were performed under slight positive pressure of nitrogen using oven dried glassware. NMR spectra were determined on a Jeol ECS-400 spectrometer with the chemical shifts reported in parts per million (ppm), calibrated to the centre of the solvent peak set. All solvents and starting materials were purchased from chemical stores where available. Melting points were recorded in open capillaries on a Stuart SMP10 melting point apparatus and are uncorrected. Infrared (IR) spectra were recorded on a Shimadzu IR-Affinity 1, and reported in wavenumbers (cm -1 ). EPR Studies were performed using an ADANI CMS 8400 X-Band spectrometer at room temperature, on 10μl samples within quartz tubes. Solution state spectra were simulated using the EasySpin 1 toolbox for MatLab using the garlic core modelling function. Linewidths were allowed to freely refine to a mixed Gaussian/Lorenzian lineshape. All spectra were recorded with Modulation Amplitude of 200 mT and a scan rate of 1 mTs -1 . For primary solution state studies metal salts were purchased from commercial sources, for secondary solution state studies the metal salts were synthesised in house to test for ease of synthesis.
Synthesis of Cu(acac) 2 : As described published by Maverick and co-workers. 2 Synthesis of Cu(hfac) 2 .H 2 O: As described published by Glidewell. 3 Synthesis of 1: Dodecylamine (0.25 g, 1.35 mmol) was added to a stirring solution of carbonyl diimidazole (CDI) (0.26 g, 1.62 mmol) for two hours. After this time (1S,2S)-cyclohexane-1,2-diamine ( 0.07 g, 0.68 mmol) was added and the solution heated to 60 °C overnight. The reaction mixture was then washed with water (50 mL) and the resultant solid removed by filtration, titrated in ether (20 mL), followed by methanol (40 mL) and finally water (40 mL). The remaining white solid was removed by filtration and dried under reduced pressure. Yield 53 % (0.18 g, 0.36 mmol). NMR spectra were found to match those previously reported by Feringa and co-workers. 4 Synthesis of 2: Butylisocyanate (0.19 g, 1.96 mmol) was added to a stirring solution of cyclohexylamine (0.19 g, 1.96 mmol) in chloroform (10 mL) for three hours. The mixture was taken to dryness and the resulting oil dissolved in hexane (5 mL). The product precipitated as a white solid and was isolated by filtration. Yield 24 % (0.096 g, 0.46 mmol). NMR spectra were found to match those previously reported by King and co-workers. 5 Figure S1 1 H NMR of compound 1 in CDCl 3 at approximately 60 °C. Position of NH resonances differs slightly from those previously reported. We believe this is due to the concentration of the sample and the corresponding self-associative properties of the molecule combined with peak broadening effects.

EPR studies
Experimental Method: EPR samples were made up with laboratory analytical grade chloroform and no special measures were used to exclude air or moisture to maintain conditions comparable to those of the reaction. Cu(hfac) 2 .H 2 O (50 mg, 0.1 mmol) was stirred with gently warmed chloroform (1 cm 3 ) to give a saturated stock solution (Solution A) at 0.1M concentration. Likewise, CEES (12 μl, 0.1 mmol) was dissolved in chloroform (1 cm 3 ) to form a stock solution (Solution B) at 0.1M concentration. These stock solutions were then used to make up solutions of Cu(hfac) 2 with varying concentrations ratio of copper to CEES to attempt to drive the equilibrium towards domination by Cu(hfac) 2 .CEES. For Samples 1 to 3, the Cu(hfac) 2 concentration was kept approximately constant for ease of comparison.  6,7 Upon addition of increasing quantities of CEES, there is a gradual increase in g iso value and concomitant decrease in the A iso coupling to the copper nucleus as the concentration of CEES increases. This is accompanied by an increase in asymmetric line broadening, indicating a movement into the fastmotional regime.
In the absence of good solid state data for the complex to properly model rotational anisotropy, the values for Sample 4 should be regarded as indicative only. The EPR spectra were largely similar in appearance, and the spectra and simulations for Samples 1 and 4 are shown as exemplars. It is also plausible that the change in lineshapes relates to the formation of an higher order adduct e.g. Cu(hfac) 2 .(CEES) 2 , but given the significant linewidths involved, there are insufficient data to attempt to model that.

Oxidative breakdown of CEES Primary Solution state studies
Experimental Method: Samples were prepared as detailed in Table S3 in a single NMR tube per experiment, with the hydrogen peroxide layered on top of the CDCl 3 mixture containing catalyst and/or simulant as appropriate. The addition of the peroxide was taken as time = 0, at this point it was assumed that no oxidation products were present as 1 H NMR of the stock CEES in CDCl 3 showed no breakdown products to be present. The samples were then sealed and disturbed as little as possible for the course of the experiment with the temperature of the samples maintained at 18 ± 2 °C. All experiments were monitored by 1 H NMR. The percentage conversion of simulant to oxidised species was calculated through comparative integration of simulant and product peaks, the results of which are given in Table S3. The initial and final 1 H NMR spectra for each experiment are shown in Figures S7-S30. The identification of the corresponding sulfoxide 8 and sulfone 9 was achieved through comparison with previously published literature values. The sparsity of data points for the Mn, Fe and Ni catalysis tests are due to the paramagnetic broadening which precluded meaningful integration.

Secondary Solution state studies
Experimental Method: Samples were prepared, as detailed in Table S3, in a single NMR tube per experiment, with the hydrogen peroxide layered on top of the CDCl 3 mixture containing catalyst and/or simulant as appropriate. The addition of the peroxide was taken as time = 0, at this point it was assumed that no oxidation products were present as 1 H NMR of the stock CEES in CDCl 3 showed no breakdown products to be present. The samples were then sealed and briefly shaken by hand for approximately two seconds before being allowed to separate. The samples were disturbed as little as possible for the course of the experiment with the temperature of the samples maintained at approximately 18 ± 2 °C. All experiments were monitored by 1 H NMR. The percentage conversion of simulant to oxidised species was calculated through comparative integration of simulant and product peaks, the results of which are given in Table S4. The initial and final 1 H NMR spectra for each experiment are shown in Figures S31-34.

Solid state studies
Experimental Method: Samples were prepared, as detailed in Table S5, in a single NMR tube per experiment. Those sample containing compound 1 to gelate the sample were briefly heated for 15 seconds to initially dissolve the gelator so that the gel could from upon cooling. The addition of the final component of the mixture was taken as time = 0; at this point it was assumed that no oxidation products were present as 1 H NMR of the stock CEES in CDCl 3 showed no breakdown products to be present. The samples were then sealed and briefly shaken by hand for approximately two seconds to ensure the even distribution of components within the solution. The samples were disturbed as little as possible for the course of the experiment with the temperature of the samples maintained at 18 ± 2 °C. All experiments were monitored by 1 H NMR. The percentage conversion of simulant to oxidised species was calculated through comparative integration of simulant and product peaks, the results of which are given in Table S6. The initial and final 1 H NMR spectra for each experiment are shown in Figures S35-S37.

Sol and Gelator simulant studies
Experimental Method: Samples were prepared, as detailed in Table S7, in a single NMR tube per experiment. Those samples containing compound 1 were briefly heated for 15 seconds to initially dissolve the gelator. The addition of the final component of the mixture was taken as time = 0. At this point it was assumed that no oxidation products were present as 1 H NMR of the stock CEES in CDCl 3 showed no breakdown products to be present. The samples were then sealed and briefly shaken by hand for approximately two seconds to ensure the even distribution of components within the solution. The samples were disturbed as little as possible for the course of the experiment with the temperature of the samples maintained at approximately 18 ± 2 °C. All experiments were monitored by 1 H NMR. The percentage conversion of simulant to oxidised species was calculated through comparative integration of simulant and product peaks, the results of which are given in Table S8. The initial and final 1 H NMR spectra for each experiment are shown in Figures S39-S41.
Although a gel was found to form under these conditions in the presence of 1 (sample 27), addition of Cu(hfac) 2 .H 2 O (sample 28) prevented gel formation meaning this sample remained a sol. We believe that this is due to the interactions of the gelator with the Cu(II) complex altering material formation processes. These spectra suggest that under these reaction conditions the CEES undergoes primary oxidation to the sulfoxide. Secondary oxidation of sulfoxide to the sulfone was not observed by these analysis methods.