Chemical Warfare Agent Simulants in Gamble ’ s Fluid : Is the Fluid Toxic ? Can It BeMade Safer by Inclusion of Solid Nanocrystalline Metal Oxides ?

e reactions of chemical warfare agent simulants, 2-chloroethyl ethyl sul�de (2-CEES) and di-i-propyl �uoro phosphate (DFP), in �uids have been investigated. Data analyses con�rm the major degradation pathway to be hydrolysis of 2-CEES to 2-hydroxyethyl ethyl sul�de, along with minor self-condensation products. Among the three �uids examined, 2-CEES degradation was the fastest in �amble�s �uid during a 96 h period. Upon addition of Exceptional HazardAttenuationMaterials (EHAMs) to 2-CEES containing �amble�s �uid, degradation was generally improved during the �rst 24 h period. e 96 h outcome was similar for �uid samples with or without EHAM 2 and EHAM 4. EHAM 1-added �uid contained only one degradation product, 2-nitroethyl ethyl sul�de. DFP degradation was the slowest in �amble�s �uid, but was enhanced by the addition of EHAMs. F�IR and solid state P NMR con�rm the destructive adsorption of 2-CEES and DFP by the EHAMs. e results collectively demonstrate that 2-CEES and DFP decompose to various extents in �amble�s �uid over a 96 h period but the �uid still contains a considerable amount of intact simulant. EHAM 1 appears to be promising for 2-CEES and DFP mitigation while EHAM 2 and EHAM 4 work well for early on concentration reduction of 2-CEES and DFP.


Introduction
Use of chemical and biological warfare agents (CWAs and BWAs) in conventional or terrorism related incidents is not new [1][2][3].Despite the longtime use of CWAs, very little is understood about the possibilities of secondary contamination of persons who handle fatalities from these incidents.is lack of understanding and dearth of reliable data makes designing protective systems and processes for dealing with contaminated remains extremely challenging.In fact, fatality management has always been a highly sensitive issue [4].e various operational, political, and religious constraints associated with the proper search, recovery, identi�cation, and disposition of remains pose a complex problem.A catastrophic event will produce appreciable contaminated casualties that will require very speci�c approaches to contaminated remains processing and sound decisions have to be quickly made concerning proper handling of the remains.
e handling and transport of contaminated human remains is an important consideration of both combative and humanitarian military actions.Current protocol has remained largely unchanged for decades, although significant advances in decontamination and incident response have been made.Handlers of contaminated and potentially contaminated human remains face signi�cant biological and chemical threats [5,6].Contamination may arise from a deliberate enemy action, such as an Improvised Explosive Device (IED) which also unleashes a toxic gas or hazardous chemical or direct exposure to CWAs, or it may be incidental due to poor sanitation and looser environmental standards found in some foreign nations.While the causes of contamination may differ, the end result is the same.ere is a need for a system to provide for the safe transportation of contaminated remains from the point of fatality to a Mortuary Affairs Decontamination Collection Point (MADCP) and, following processing and release, on to the Continental United States (CONUS).is system must be sufficiently robust to permit the transportation of remains via multiple means (i.e., helicopter, airplane, boat, or wheeled transport) while being sufficiently versatile to contain and mitigate a variety of contaminant types.
In recognition of this need, NanoScale Corporation (NanoScale), in collaboration with Kappler, Inc., has recently produced a human remains pouch which was speci�cally designed to allow for safe handling and transport of contaminated human remains (http://www.nanoscalecorp.com/,http://www.kappler.com/home/).is contaminated human remains pouch (CHRP) has been shown to dramatically reduce or eliminate the permeation of multiple toxic industrial chemicals (TICs), as well as CWAs [7].However, the CHRP, as well as other protective devices, could be made even more effective if more were known about the fate of CWAs on human remains.NanoScale has produced many extremely effective formulations for decontamination, including their �agship product, FAST-ACT R ⃝ , which is capable of mitigating hazards arising from multiple toxic chemicals in different chemical classes.Biologically active formulations able to kill viruses, bacteria, and spores have also been produced.e main focus of NanoScale's business has been decontamination; therefore, NanoScale is keenly interested in exploring the question "What is the outcome of postmortem interaction of CWAs with physiologically active body �uids�� e objective of this study was thus twofold: (1) to develop highly efficient, dually active (against CWAs and BWAs) solid decontaminants based on nanocrystalline metal oxides, and (2) to develop reliable analytical methods and generate data to quantify the residual CWA simulants and degradation products that may be present in contaminated human remains.Speci�cally, we have investigated the degradation of CWA simulants in Gamble's �uid, a synthetic equivalent of physiological �uid, with and without the inclusion of novel dually active solid decontaminants.Over the past several years, NanoScale has developed a series of highly reactive nanocrystalline metal oxides and their derivatives, which have been shown to be effective in destructive adsorption of a number of toxic compounds, including the CWAs, soman (GD), VX, and mustard (HD) [8][9][10][11].Destructive adsorption by nanocrystalline materials leads to chemical conversion of the CWAs into far less toxic compounds, thereby permanently eliminating the threat.e unique properties of these mesoporous adsorbents are due to their high surface areas, unusual morphologies and high surface concentrations of reactive defect sites.In oxides such as MgO, Al 2 O 3 , and TiO 2 , nerve agents are adsorbed by means of hydrogen-bonds between the surface hydroxyl groups and the P = O moieties, followed by sequential removal of the functional groups by hydrolysis.is leaves behind a stable phosphate attached to the surface (Scheme 1).Similarly, HD undergoes decomposition by elimination/hydrolysis induced by the metal oxide materials.Despite the intense interest in exploiting the novel properties of these metal oxides as dry materials, studies directed towards their performance in a �uid environment are scarce.
Herein, efforts to evaluate the fate of CWA simulants di-i-propyl �uoro phosphonate (DFP; simulant for nerve agent) and 2-chloroethyl ethyl sul�de (2-CEES; simulant for mustard) in a variety of �uids with and without the inclusion of solid decontaminants are reported.Analytical techniques involving gas chromatography (GC), Fourier transform infrared (FTIR) spectroscopy, and nuclear magnetic resonance (NMR) spectroscopy were used.Results presented herein should aid in developing a preliminary assessment of potential human health and safety risks that may arise from exposure to residual levels of agents present in the contaminated remains.Saline solutions were prepared at a concentration of 0.9% (w/w).Gamble's �uid was prepared according to the published literature [12].e composition of Gamble's �uid is sodium chloride (6.78 g/L); ammonium chloride (0.53 g/L); sodium bicarbonate (2.27 g/L); glycine (0.45 g/L); L-Cysteine (0.12 g/L); sodium citrate tribasic (0.06 g/L); calcium chloride (0.02 g/L); sodium phosphate dibasic (0.17 g/L).For non-NMR sample preparations, Elix water was used.For saline and Gamble�s �uid used in NMR experiments, D 2 O was used in place of water.

pH
Analysis.e fates of the individual simulants, 2-CEES and DFP, in �uids were investigated by measuring pH of the solutions as a function of time, as likely reaction products were expected to increase the acidity of a given simulant solution.us, pH measurements were made on aqueous, 0.9% saline, and Gamble�s �uid solutions of 2-CEES and DFP with a Fisher Scienti�c, accumet AB15 pH meter.In a representative procedure, 20 L of simulant was added to 2 mL of �uid (water, saline, or Gamble�s �uid), stirred, and the pH was measured aer 10 min, 1 h, 96 h, and 8 days of the preparation.e pH of the �uid was measured prior to the addition of simulant for comparison.

GC Analysis. GC analysis was conducted to determine
the fate of the simulant in �uids with and without any EHAM present.A Hewlett Packard 5890 Series II GC equipped with a Hewlett Packard 5972 Mass Selective Detector and a HP-5MS capillary column was used for sample analysis.Since 2-CEES and HEES coeluted on this column, quantitation of 2-CEES and HEES was performed using a Hewlett Packard 5890 Series II GC equipped with a Flame Photometric Detector utilizing a 5% SE-30 ON CW-HP 80/100 packed column.In a typical experiment, 5 L simulant (2-CEES or DFP) was added to 0.5 mL �uid (water, saline, or Gamble�s �uid), vortexed, and le for 1, 24, or 96 h.e solution was extracted with 1 mL ethyl ether and analyzed by GC-MS and GC-FPD.Control samples containing 5 L simulant in 1 mL ether were used to calculate the amount of simulant remaining.A second set of GC experiments was carried out to determine the fate of the simulants in the presence of EHAMs.For this study, 20 L simulant was added to 2 mL of Gamble�s �uid, along with 600 mg of EHAM.Samples were vortexed brie�y and allowed to sit at room temperature for a contact time of 1, 24, or 96 h.Subsequently, a 0.5 mL sample was removed using a syringe �lter.is �uid was extracted with 1 mL ethyl ether and analyzed by GC-MS and GC-FPD.Control samples containing 20 L simulant (2-CEES or DFP) in 2 mL ether were used to calculate the relative amounts of the remaining simulant and any by-products produced.

FTIR Analysis.
FTIR spectra of the powder samples were recorded in the absorbance mode using a ermo Scienti�c Nicolet 6700 spectrophotometer.e FTIR unit was equipped with a HeNe laser source, and an MCT detector cooled with liquid nitrogen.A Praying Mantis Di�use Re�ectance Infrared Fourier Transform spectroscopy (DRIFTS) unit with purging nitrogen was used for the sample analysis at ambient temperature.All spectra were collected by coadding 1032 scans at a resolution of 4 cm −1 .e spectra were recorded from 4000 to 500 cm −1 .
Due to the instrumentation detection limit, a higher agent concentration was needed to perform the DRIFTS experiments.As a result, all of the FTIR related samples were prepared in duplicate with a 2-CEES to sorbent (EHAM 1, EHAM 2, or EHAM 4) ratio of 1 : 10 (wt : wt).In a typical reaction, 9.3 L 2-CEES was placed in a GC vial, followed by the addition of 100 mg of the test sorbent.e GC vial was capped, the mixture was allowed to react under ambient conditions for 1, 24, and 96 h, and the IR spectra of the samples were acquired.Background subtraction of the acquired spectra was performed using the appropriate sorbent spectrum to obtain the �nal FTIR spectrum of the reacted sorbent.

NMR Analysis.
Solution NMR experiments were performed at Kansas State University, Manhattan, KS, USA, using a11.75 T (500 MHz for 1 H) Varian UNITYplus spectrometer (Varian, Palo Alto, CA, USA).NMR samples were prepared by adding 20 L of simulant (2-CEES or DFP) to 2 mL solvent (CDCl 3 , D 2 O, saline, or Gamble's �uid) and then vortexing brie�y to mix.For samples involving EHAM sorbents, 20 L of agent was added to 2 mL solvent, quickly mixed by vortexing, followed by the addition of 600 mg of sorbent.e mixture was stirred for the desired contact time (1, 24, or 96 h) before being �ltered for NMR analysis.
1 H NMR spectra were acquired at room temperature, using a spectral width of 6010 Hz over 8 K data points.A relaxation delay of 1 s was employed for a 45 degree �ip angle.Four scans were coadded for each sample and Fourier-transformed with a line-broadening factor of 0.5 Hz to improve the S/N ratio.Chemical shis were referenced against the solvent peak (7.24 ppm for CDCl 3 and 4.8 ppm for HOD).Relative amounts of various products formed were estimated from corresponding methyl peak intensities expressed as percentages of the total intensity of all methyl peaks in a spectrum.is permitted comparison of NMR data as a function of time not only for a given sample, but also between samples to obtain semiquantitative estimates of reaction rates.
Solid state proton-decoupled 31 P NMR spectra were obtained at Iowa State University, Ames, IA, USA, with a Bruker AVANCE II 14.1 T (242.9MHz for 31 P) equipped with a Bruker PH MASDVT600WB H/X/Y probe under the condition of magic angle spinning (MAS) at a rate of 5-10 KHz, using a spectral width of 73529 Hz over 4 k time domain points; a relaxation delay of 10 s was used for a 90 degree pulse; 32 scans were collected for each sample, a linebroadening factor of 20 Hz was employed to improve the S/N ratio.Chemical shis are reported relative to the external standard, phosphoric acid, which is assigned a value of 0 ppm.Typically, 20 L of agent was added to 600 mg of sorbent and vortexed brie�y to mix.Aer the desired contact time, the sample was homogenized thoroughly with a spatula and packed into a Bruker 4 mm zirconia MAS rotor for data collection.

pH Analysis. pH values of individual solutions of 2-
CEES and DFP in water, 0.9% saline solution, and Gamble's �uid were measured aer 10 min, 1 h, 96 h, and 8 days of preparation.e results (Figure 1) indicate hydrolysis and degradation of these compounds into products that increase acidity.

GC Analysis.
Agent-contacted �uid and sorbent were extracted with ethyl ether aer 1, 24, and 96 h of preparation and analyzed by gas chromatography.Results were analyzed in terms of all the possible 2-CEES and DFP degradation products, as summarized in Schemes 2 and 3, respectively.e assignments were con�rmed by GC-MS fragmentation pattern as well as NMR data and by comparison with data from authentic samples.e �uids and sorbent reacted with 2-CEES and generated several by-products within 1 h and the reaction(s) continued up to 96 h.e relative amounts (normalized against an external control) of remaining 2-CEES and by-products are reported in Tables 1 and 2.
Similarly, the �uids and sorbent reacted with DFP within the �rst hour of contact and the amount of DFP present continued to decrease during the 96 h period (Figures 2(a) and 2(b)).Because of nonvolatility of phosphonic acid-based reaction products, only the relative amounts of the remaining DFP at various time points are reported.Among the examined �uids (Figure 2(a)), degradation was the slowest in Gamble's.e results presented in Figure 2(b) show that EHAMs 2 and 4 are especially efficient for early on reduction of DFP.

FTIR Studies
. e FTIR spectrum of neat 2-CEES is shown in Figure 3, and the bands characteristic of C-Cl functionality are marked (1215 cm −1 Cl-C-H 2 scissor and 695 cm −1 C-Cl stretch).e FTIR spectrum for the reaction of dry EHAM 1 and 2-CEES (Figure 4) remained identical for the 1, 24, and 96 h time periods.Even though several bands were observed in all of the spectra, the bands due to C-Cl functionality were not present.Two new bands (not observed in neat 2-CEES) at 1640 cm −1 and 1280 cm −1 were present in all spectra.
In contrast to EHAM 1, EHAM 2-treated samples (Figure 5) exhibited time sensitive spectral variations.e variations were speci�cally obvious in the 3700-3500 and 1550-1400 cm −1 regions.e IR bands due to Cl-C-CH 2 scissor and C-Cl stretch were absent.Overall, the spectral bands for EHAM 4-contacted samples remained relatively identical for the three time periods (Figure 6).As mentioned earlier for EHAM 2, the bands assigned to the C-Cl functionality were not present.

Solution 1 H NMR Studies.
An assigned 1 H NMR spectrum of 2-CEES in CDCl 3 is shown in Figure 7.When 2-CEES was examined in an aqueous environment, several new peaks were observed.e 1 H NMR spectra of 2-CEES in D 2 O, saline, and Gamble's �uid were very similar.e spectra of 2-CEES in Gamble's �uid, examined aer 1, 24, and 96 h of preparation, are shown in Figure 8. e spectra become complex because of formation of at least three degradation products (Figure 8), as deduced from the appearance of four groups of methyl peaks.Changes in relative intensities of these methyl peaks allowed us to monitor the time course of disappearance of 2-CEES and formation of new reaction products (Table 3).minor product decreasing over the 96 h period, suggesting further degradation of this compound.e peak patterns suggest that both products have the basic structure similar to HEES and CEES.e identity of the major product is inferred to be nitroethyl ethyl sul�de (NEES) on the basis of GC-MS fragmentation pattern.In contrast, both EHAM 2 and EHAM 4 yielded multiple products, when reacted with 2-CEES in Gamble's �uid (Figures 9(b) and 9(c), resp.), as seen in the up�eld methyl region (1.1-1.4 ppm).Table 4 summarizes the relative intensity data for all the products present in the �uid.deduced that EHAMs 2 and 4 were very efficient in degrading DFP.Table 6 displays the related methyl peak intensity data for this set of samples.degradation appears to be complete within 24 h, as indicated by the appearance of a broad peak at −3.12 ppm.Similar results are obtained with EHAM 4 (Figure 12(c)).Aer 1 h of reaction, two broad peaks were seen at −5.67 and −12.9 ppm, and within 24 h only a broad peak at −4.54 ppm remained.e increased broadness of this peak aer 96 h suggests the presence of multiple phosphorus species in this sample.

pH Analysis.
Changes in pH of the simulant-contacted �uids (water, saline, and Gamble�s �uid) indicate the occurrence of a reaction or reactions that produce acidic compounds.ese changes were observed for both 2-CEES and DFP.Gamble�s �uid slows down the rate of drop in pH, likely due to the buffering effect of salts and amino acids present in the solution.

GC Analysis.
As seen from Table 1, the amount of extracted 2-CEES was signi�cantly lower than the control (2-CEES in ethyl ether); however, extracted HEES increased signi�cantly from 1 h to 96 h for all the �uids.�nly trace levels of condensed (dimer or trimer derived) products were extracted.Similarly, trace levels of DEDS were observed aer 1 h and 24 h time points for all the �uids, but were not seen by 96 h.is could be due to evaporative loss of DEDS.e origin of DEDS in these samples is not known.A similar trend was observed for EVS as well.
When experiments were performed using EHAMs in Gamble�s �uid (Table 2), the amount of recovered 2-CEES was signi�cantly lower when compared to �uids alone (Table 1).e only exception to this was EHAM 1 sample with 1 h contact time.It is likely that hydrolysis of 2-CEES in the presence of EHAM 1 is much more controlled due to its a�nity for water.Interestingly, �uids from EHAMs 2 and 4 contained as much HEES as did Gamble�s �uid.In contrast, EHAM 1 �uid had only minor amounts of HEES as the reaction product, which could be attributed to the high adsorptive power of EHAM 1.A minor amount of NEES was observed only with EHAM 1.
As attested by GC analysis (Figure 2), degradation of DFP was signi�cantly faster and essentially complete in water and saline.However, in Gamble� �uid, DFP degradation was only partial (Figure 2(a)).is sharp difference in hydrolytic power of the �uids was also further con�rmed by NMR results (Figure 10 and Table 5).In contrast, samples involving �F� exposed to EHAMs in Gamble�s �uid had signi�cantly lower amounts of extracted �F� than Gamble�s �uid alone (compare Figures 2(a 4.3.FTIR Analysis.e negative absorption feature at ∼3700 cm −1 from reaction of 2-CEES and EHAM 1 (Figure 4) indicates that hydrogen bonding was involved in the adsorption of 2-CEES by EHAM 1.It was previously reported that the surface MO-H groups can readily hydrogen bond to the sulfur and chlorine atoms in 2-CEES [13].e absence of the C-Cl functionality suggested that 2-CEES was consumed; the newly formed bands (1640 cm −1 , ONO 2 asymmetric stretch; 1280 cm −1 , ONO 2 symmetric stretch) were attributable to the NEES product [14].In contrast, EHAMs 2 and 4 display a positive hydroxyl band attesting to the formation of signi�cant amounts of hydroxyl product.
All the dry EHAM formulations show the presence of NEES as one of the degradation products of 2-CEES.

NMR Analysis
4.4.1.Solution 1 H NMR Studies.2-CEES undergoes hydrolysis to yield at least three products aer 96 h.One of the products is identi�ed as HEES, while the other two products are likely to be the two sulfonium condensation products (dimer and/or trimer in Scheme 2) [15].ese conclusions are supported by GC-MS data.It is difficult to differentiate between the various condensed products (R-Cl and R-OH or dimer versus trimer) by NMR because their structures are highly similar with chemically equivalent hydrogens.
Very similar changes are observed for 2-CEES in all the three �uids (Table 3).is reinforces the inference drawn that the slower pH drop in Gamble�s �uid is due to its buffering capacity, rather than a reduced rate of hydrolysis and degradation of the simulant in that solution.In D 2 O and saline, aer 96 h, roughly equal amounts of 2-CEES, HEES, and the condensed products result, whereas in Gamble's �uid, HEES is formed as a major product.
Differences in the amounts of reactions products formed, as monitored by the two methods, may arise out of degradation of the condensed products, back into 2-CEES and HEES on the GC column [15].
Interestingly, EHAM 1-contacted Gamble's �uid contained only one product, namely, NEES.Additional products might have formed and been trapped on the sorbent.EHAM 2 and EHAM 4 showed similar degradation results with the amount of HEES formed being around 50% aer 96 h.However, in both cases, higher levels of the condensed products were observed aer 96 h, as compared to 2-CEES in Gamble's �uid or EHAM 1. e NMR results are consistent with the GC-MS data, especially those related to extractable amounts of HEES from all the EHAMs (Tables 2 and 4).Overall, it is concluded that EHAMs accelerate the degradation of 2-CEES with EHAM 1 being the most e�cient in keeping the �uid less toxic.
In the case of DFP, analysis of its degradation products in D 2 O, saline, and Gamble's �uid, as deduced from relative intensities of various peaks (Table 5), suggests that DFP undergoes loss of �uorine to yield DIPP, which is subsequently degraded to form IPA and IPP as minor products [16].e hydrolysis rate is the slowest in Gamble's �uid with DFP as a major component of the products observed even aer 96 h.e solution 1 H NMR results of DFP in �uids are consistent with the GC-MS data (Figure 2).e relative intensities of various methyl peaks in the spectra of DFP and EHAMs are shown in Table 6.e most notable result is that complete degradation of DFP is readily achieved by EHAM inclusion.As with DFP in �uids, the amount of DFP is decreased and the major product is DIPP (Peak i).EHAM 1 reacts with DFP slowly since 92% of the methyl peaks are from DFP aer 1 h, while with EHAMs 2 and 4, less than 1% is seen for the same time period.However, aer 96 h, DFP has been degraded nearly completely by all of the EHAMs.EHAMs 2 and 4 are very quick to destroy DFP with almost the entire product formed being the monohydrolyzed product (DIPP).It appears that EHAM 1 further hydrolyzes DFP to IPP, as seen by the higher levels of IPA (Peak g) and IPP (Peak j).e results in the presence of EHAMs also match very closely for both GC and NMR analysis.In both experiments, the level of DFP decreases to 0% aer 96 h for EHAM 1, but for EHAMs 2 and 4, the level drops to 0% within 1 h.4.4.2.Solid State NMR.Solid state 31 P NMR spectra of DFP treated with EHAM 1 are consistent with the formation of hydrolysis products as observed in solution 1 H NMR spectra.e initial spectrum of the DFP control shows a very sharp doublet and does not appear to be bound to the low surface area TiO 2 support.Broadening of the two peaks aer 1 h suggests binding of the compound to the EHAM surface.Aer 16 h, the peaks broaden further and begin to merge into a broad band at −10.8 ppm, thus indicating the cleavage of �uorine atom to yield DIPP [17].In addition, a new broad peak appears at −3.15 ppm due to further hydrolysis of DIPP to IPP.IPP has a similar structure to phosphoric acid.is is consistent with the solution 1 H NMR data when DFP is treated with EHAM 1 in Gamble's �uid (Table 6).By 96 h, the hydrolysis reaction is complete to form DIPP (−10.8 ppm), which is further hydrolyzed to IPP (−3.59 ppm).Between 96 h and 11 d, the hydrolysis reaction is complete, as no further changes are noted up to 21 d.
Reactions of DFP with EHAMs 2 and 4 yielded similar results to EHAM 1, except that the reactions occurred much faster, as noted from solution 1 H NMR studies.Slight variations in chemical shis are expected for the reaction products based on different metal oxide surfaces [18].

Conclusions
pH measurement of �uids is a good indicator of CWA simulant degradation, since acid by-products result from hydrolysis of these compounds.Similarly, solvent extraction of CWA simulant-contacted �uids and subsequent GC analysis provide an overview of various degradation products.For the safe handling of contaminated human remains, it is necessary to use pouches enhanced with EHAMs because contaminated body �uids are expected to contain appreciable amounts of intact agent (as seen by 2-CEES and DFP analysis) or their derivatives (2-CEES analysis).NMR analysis of simulantcontacted �uids revealed that the rate of degradation is highly �uid dependent for DFP (Gamble's < D 2 O ∼ saline) and 2-CEES (Gamble's > D 2 O ∼ saline).Inclusion of EHAMs facilitates degradation of both DFP and 2-CEES in these �uids, especially during the early hours of contamination.e toxicity of the contaminated Gamble's �uid can be minimized by including an appropriately formulated EHAM sorbent.e ongoing threat of homicidal use of CWAs during military con�icts and by terrorists underlines the necessity for development of clear guidelines that will guarantee foolproof protection of personnel dealing with victims of these serious incidents.Early recognition and protective measures are essential when dealing with CWA-related incidents, otherwise, responding human remains handlers will only add to the list of victims.

S 1 :
Decomposition of nerve/blister agents on metal oxide surfaces.

1 HF 1 :S 2 :
NMR spectra of 2-CEES in �uids in the presence of EHAMs are shown in Figures 9(a)-9(c).e spectra for 2-CEES in Gamble's �uid treated with EHAM 1 indicate the formation of two products initially, with the amount of the pH of simulant-containing �uids at various time periods.Degradation products of 2-CEES.

1 HS 3 :
NMR spectra of DFP in CDCl 3 , D 2 O, saline, and Gamble's �uid are shown in Figures 10(a)-10(d), respectively.At least, four different products containing isopropyl group were seen in addition to intact DFP aer 96 h in Gamble's Degradation products of DFP.

F 3 :
FTIR spectrum of 2-CEES.�uid (Scheme 3).In sharp contrast, complete/near complete degradation was observed in D 2 O and saline, respectively.All compounds containing an isopropyl group yielded a doublet around 1-1.3 ppm and a multiplet around 3-5 ppm.One of these compounds was identi�ed as IPA on the basis of a reference spectrum obtained of that compound in D 2 O.

F 4 :
Subtracted FTIR spectra of EHAM 1 and 2-CEES.Table 5 summarizes the relative intensity data for all the products present in the �uid.Time-dependent 1 H NMR spectra of DFP in Gamble's �uid containing EHAMs 1, 2, and 4 are displayed in Figures 11(a)-11(c), respectively.On the basis of peak patterns observed in the 1.0-1.3ppm region, it is readily

F 6 : 3 F 7 :
31 P NMR Studies.A reference protondecoupled solid state 31 P NMR of DFP was collected by adding DFP to low surface area conventional TiO 2 (FS-TiO 2 , SSA = 11 m 2 /g, shown as DFP control in Figure12).e doublet arises from the scalar coupling between �uorine and phosphorous atoms ( 1  PF = 969 Hz).e nonporous, low surface area TiO 2 , even though chemically similar to the TiO 2 present in the EHAMs, did not interact signi�cantly with DFP.When DFP was allowed to react with EHAM 1, the doublet broadened and began to coalesce (Figure12(a)), indicating hydrolysis and binding to the surface.With EHAM 2 (Figure 12(b)), the reaction occurred much faster, as was observed by solution 1 H NMR studies (Figure 11(b)).e Subtracted FTIR spectra of EHAM 4 and 2-CEES. 1 H NMR spectrum of 2-CEES in CDCl3.
T 5: Comparison of methyl-peak intensities of DFP reaction products.