GB2331298A - Decontamination material - Google Patents

Abstract

A material comprises an acidic resin which reacts with at least one toxic chemical to convert it to a non-hazardous residue. The acid resin may be selected from AMBERLYST (RTM), AMBERLlTE (RTM), or DOWEX (RTM) and is preferably an adsorbent support material. A composition comprising the material is also disclosed, wherein the material supports an active reagent which may react with a toxic chemical to produce a non-hazardous product. The active reagent is typically an oxidising agent and may be MMPP (magnesium monoperoxyphthalate), oxone (potassium peroxymonosulphate), MCPBA (m-chloroperoxybenzoic acid), IBDA (iodobenzene diacetate), calcium hypochlorite, or fichlor (sodium dichloroisocyanurate). A further active agent may be provided which is active against a different toxic chemical. A method for removing a toxic chemical from a contaminated area using the composition is also described.

Description

L 1 Decontamination Compositions 2331298 The present invention relates to

compositions useful in the removal of toxic materials in particular chemical toxins such as chemical warfare (CW) agents and toxic industrial chemicals, to methods of removing toxic materials using the compositions and to methods for their preparation.

Spillage of toxic chemicals, either accidentally in the case of industrial applications, or deliberately for example in battleground situations presents a significant hazard to personnel coming into contact with these chemicals.

At present fullers' earth (FE) powder is generally used for the decontamination of personnel and personal equipment who have been exposed to CW agents, because powder decontaminants offer a rapid and simple method for the removal of liquid CW agent. However, FE adsorbs, rather than deactivates, liquid chemical agent with the result that the contaminated powder may present a desorption or contact hazard. In addition, agent may be leeched from the powder (for example, by the action of water).

The applicants have found identified reactive powders which can not only absorb but also detoxify toxic liquid chemicals such as CW agents to produce a nonhazardous residue. The effectiveness and speed of the decontamination process will be improved, leading to enhanced operational performance and a reduction in the logistic burden. Such systems are also likely to offer higher agent capacities, meaning that less decontaminant should be required.

According to the present invention there is provided a composition comprising an absorbent support material and an active reagent which reacts with at least one toxic chemical and converts it to an non-hazardous residue.

2 Suitably the absorbent support material is in the form of a flowable powder.

Examples of absorbent support materials include zeolites, clays, resins, silicas, aluminas or carbons. Factors influencing choiceinclude the nature of the porosity (which affects the adsorption rate and capacity of the solid), catalytic activity (if any), and the compatibility of the powder with the oxidants. Lamellar clays and large-pore 10 zeolites might possess appropriate properties, whereas unimpregnated activated carbons are relatively inert and more effective for adsorption of agent without chemical reaction.

Particular examples of powder supports include Fullers' Earth (FE), acid leached montmorillonite (K1GY,,. silicalite,-KSF (acid impregnated montmorillonite), XE555 (resin/carbon mixture), neutral alumina, acidic alumina and fine alumina.

Some of these powders may promote reaction between the active reagent and the toxic chemical.

1 35 Preferably however, for reasons of availability and cost, the support material is a clay, in particular Fuller's Earth or acid leached montmorillonite such as K10.

Reagents which might usefully be added to suitable powder supports include bases and nucleophiles, oxidising agents, quaternising agents or reducing agents. The selection will depend upon the nature of the particular toxic chemical which is likely to be encountered. For industrial purposes, the hazard is likely to be known in advance. For military purposes however, it would be preferable to use a reagent which reacts with more than one CW agent. Alternatively or additionally, a range of potential oxidant impregnated powders may be prepared so as tackle a wider range of CW agents likely to be encountered.

Agents which may be removed using the compositions of the invention include organophosphorus and organosulphur compounds 3 including the CW agents H, G and V agents. For example, the well known H agent is sulphur mustard or HD. Examples of G agents include GA (Tabun), GB (Sarin) and GD (Soman) whilst an example of a V agent which may be encountered is Oethyl S-2(diisopropylamino)ethylmethylphosphonothiolate M).

It has been found that particularly suitable compositions for the removal of HD include oxidising agents as illustrated hereinafter. Particularly preferred oxidising agents are selected from MMPP (magnesium monoperoxyphthalate), Oxone (potassium peroxymonosulphate), MCPBA (m-chloroperoxybenzoic acid), IBDA (iodobenzene diacetate), calcium hypochlorite or Fichlor (sodium dichloroisocyanurate). Preferably these agents are supported on K10 material. All six of these oxidant and powder support combinations were found torapidly decontaminate HD under ambient conditions.

Of the systems studied, FE or K10 / Oxone, FE or K10 / IBDA and FE or K10 1 Fichlor are particularly preferred since they selectively and rapidly oxidise HD oxidised to the sulphoxide.

Work with sulphur mustard detoxification showed that the FE or K10 / IBDA and FE or K10 / Fichlor systems rapidly detoxified the agent with the highest selectivity for the non-toxic product. The reaction with Oxone was slower, and resulted in the formation of a toxic reaction product. It was, however, acceptable for VX detoxification. The two organic oxidants, Fichlor and IBDA, in the presence of clay are likely to be the materials of choice for decontamination of both VX and HD.

Until measurements of the stability of the preferred systems have been carried out, Fichlor, IBDA and Oxone remain candidates.

When VX is the CW agent contaminant, suitable active reagents include bases and other nucleophiles (e.g. alkoxides, amines and alkanethiolates), oxidising agents (e.g. hypochlorites, hydroperoxides and peracids), quaternising agents (e.g. acids and alkyl halides) or reducing agents (e. g. bisulphite, 1\i- 4 dithionite and borohydride). Preferably however, the active reagent is one which will also be suitable for HD detoxification such as Fichlor, IBDA as listed above.

Preferably in this instance, the reagents are supported on Fullers earth.

It has been found that the performance of the FE / oxidant systems is, in descending order of preference for VX decontamination, Fichlor / clay > Oxone / clay =_ IBDA / clay 10 MMPP / clay >> calcium hypochlorite / clay.

In the present case, of the oxidant and powder systems studied, only the FE / Fichlor, FE / Oxone and K10 / Fichlor systems were found to rapidly detoxify VXto yield non toxic products (in 30 minutes or less).

A range of agents active against GD have been found as illustrated hereinafter. In the main, suitable agents are bases. one or more of these agents are suitably combined with agents active against HD and/or VX to provide a composition which is active against a range of threat agents.

Where the compositions of the invention more than one active reagent, it is possible that they may be supplied separately, for example in a twopack formulation for admixture with the support material in situ. For example, where one active reagent is acidic and/or the support is an acidic material such as a clay, and a reactive agent is basic, it may be preferable for shelf-life and storage properties, that the basic reagent is kept separate from the other components of the composition until required for use. Two-pack formulations, for example comprising support material and optionally one active reagent in one compartment and an active reagent or a different active reagent in another may be provided in a manner such that mixing is facilitated.

Thus a further aspect of the invention comprises a container having a plurality of compartments, each compartment comprising at least one component of a composition as described above.

The amount of active reagent which will be present in the composition of the invention depends to a large extent on the particular nature of the support material and reagent. In general however, compositions of the invention suitably comprise from 5-20wt% of active reagent. They may also include water. For example, powder support materials suitably contain sufficient water to ensure that they are free flowing. In general, this means that the compositions will contain from 20 to 25wt% water.

They are prepared by mixing reagent with the support powder. This can be done using a simple powder M-1ker, for example a rolling mill.

The invention further provides a method for removing a toxic chemical from a particular area, said method comprising applying to the toxic chemical a compostion as described above, and then removing the resultant solid.

The composition is suitably applied in sufficient amounts to ensure that there is enough active reagent applied to decompose all the toxic chemical present. This means that, in general, sufficient powder is applied to ensure that there is an excess of reagent as compared to toxic chemical, in general at least a 3 fold molar excess, and preferably at least a 5 fold molar excess, is applied.

It has been found that VX can be detoxified using a number of solid oxidants, which have also been shown to detoxify sulphur mustard. VX could not be extracted from K10 powder in the absence of an oxidant, but was recovered (up to 12%) from FE, in addition to the degradation product VPA. That a W reaction product was extracted from the FE indicates that this powder promotes (partial) degradation of W. The precise fate of the agent on the K10 powder is unknown. None of the other solids 6 investigated (Silicalite, XE-555, neutral, fine and acidic alumina) offered a measurable advantage over the FE and K10 powders in the absence of oxidant. No VX and / or reaction products were recovered from XE-555, due to sorption into the porosity of the resin mixture. The XE-555 is unsuitable because the requirement is to identify a reactive decontaminating powder. The KSF powder behaved in a similar manner to the K10 powder. Some combinations of the powders and oxidants provided an effective means for VX detoxification. No VX could be recovered from the FE / Fichlor, FE / Oxone, or FE / MMPP systems.

k is The invention will now be particularly described by way of example with reference to the accompanying drawings in which:

Figure 1 shows a series of 'H NMR results for various components: specifically A: HD (Bis(2-chloroethyl) sulphide), B: HDO (Bis(2chloroethyl) sulphoxide), C: HD02(Bis(2chloroethyl)sulphone, D: an extract from treated FE/fichlor, 20 and E: an extract from treated FE/IBDA; Figure 2 shows the NMR results for various systems including A: 'H NMR for VS ( (S-(diisopropylamino) ethyl thiol), B: 'H NMR for VSO (2diisopropylamino)-1-ethanesulphenic acid), C: 'H NMR 25 for VX (O-ethyl S2(diisopropylamino)ethylmethylphosphonothiolate), D: 'H NMR for VXO (Oethyl S-2-(diisopropyl)amino) ethyl methylphosphonosulphoxide), E: 'H NMR for VXO (expansion of D) and F: 31p NMR for extract from Fullers earth/fichlor; Figure 3 is an ion chromatograph for a K10/Oxone mixture; and Figure 4 shows the chemical structures of VX, VS and the reaction products mentioned in the application.

In the examples, the following materials were used. Oxidants were supplied by Aldrich and were used as supplied without further purification. The purities of the calcium hypochlorite 7 (35% w/w available as chlorine), m-chloroperoxybenzoic acid (MCPBA) (50 - 60% purity) and hydrogen peroxide (30% v/v) were known, the remaining oxidants being assumed to be 100% pure. DBS (Aldrich, 96%) was distilled under nitrogen, the fraction collected in the temperature range 186-188'C being used.

CEES (Aldrich, 98%) was used after purification by distillation under reduced pressure [3]. HD (98% by NMR) and hexamethyldisiloxane (HMDSO, 99. 5+%) were supplied by CBD Porton Down and Aldrich, respectively. Dichloromethane (DCM, Fisons AR Grade, >99.8%) was dried before use over 3A molecular sieve. Helium (BOC Ltd.) for use in GC analyses was dried and deoxygenated by passage through a Puritube (Phase Separations Ltd.) and an oxygen trap (Phase Separations Ltd.).

K10 powder was used as received (ca. 6 w/w% water) and after adjustment of the water content (addition of 2 g distilled water to 10 g of K10, the water being added in eight aliquots to produce a free-flowing powder of water content ca. 26 wlw%). Adjustment of the water content was made after outgassing the K10 (16 h, 150'C). FE powder was used as received (ca. 16.5 w/w% water). The moisture content of the powders was determined using a moisture balance (Denver Instruments, model IR-200). Zeolite molecular sieves were used in both the Na+ form and the H+ form. Ion-exchange was carried out to produce the Na and H+ forms via heating under reflux in either an aqueous solution of sodium acetate (1.0M, 200 ml) or ammonium acetate (1.0M, 200 ml) [41. The solid was removed by filtration and washed with distilled water, the Na+ form being prepared by calcination of the product derived from reflux with sodium acetate in air at 4000C. The H+ form was prepared by calcination of the product derived from reflux with ammonium acetate in air at 5SO'C. Zeolite was wetted by addition of four 0.1 g aliquots of distilled water to 2 g of zeolite (16.7 w/w% water).

8 Example 1

Removal of DBS (HD simulant) is In order to allow a large number of compositions to be tested, a number of HD simulants were used in preliminary experiments.

The first of these was dibutyl sulphide (DBS). This compound was selected because the oxidation reaction is less complex than that of chloroalkyl sulphides (e.g. HD), from which hydrolysis products may be formed in the presence of moisture (Y-C Yang et al., Chem Rev. (1992) 92, 1729). Some of those reagents identified as suitable for DBS oxidation should also be active toward a'second simulant, 2-chloroethyl ethyl sulphide (CEES). Secondary screening (s.ee..Example 2 below) with CEES was expected to yield a limited number of oxidants which would be active toward HD.

DBS oxidation was studied using a similar procedure to that described elsewhere (Hirano et al., Bull Chem Soc. Jpn, (1991), 64, 3752), Hirano et al., Chem Lett., (1991) 523). A 0.20M solution of DBS in DCM (dichloromethane)(10.0 ml) was added to a mixture of K10 (2.4 g) containing 26.5wt% water and the oxidant (5 mmol). The mixture was thoroughly stirred at room temperature for at least three hours, aliquots being removed at various time intervals in order to follow the reaction by GC. A stock solution of DBS in DCM containing 0.0040 M nonadecane was prepared, the nonadecane being used as an internal standard for the GC analysis. All the experiments were carried out under air, no specific precautions being taken to exclude moisture.

Samples of the reaction mixtures (ca. 0.1 ml) were removed at various times and diluted with sufficient DCM to allow filtration through glass wool to remove solid particulates.

The filtrates were then frozen using liquid nitrogen until analysis was carried out. Samples were thawed immediately t:l_ 9 prior to analysis and then refrozen after filling the microsyringe prior to injection into the chromatograph.

An internal standard (0.0040 M nonadecane) was added prior to 5 analysis by gas chromatography.

The chromatograph (fitted with a flame ionisation detector (FID)) was calibrated using DBS, dibutyl sulphoxide and dibutyl sulphone using a method described elsewhere (J.B.

Pattisonet al., "" A programmed Introduction to Gas-Liquid Chromatography" Heydon and Son Ltd., London 2nd Edition, 1973). Separation of the components was carried out in helium using a 30 metre Carbowax capillary column (Alltech EconoCap, column ID: 0.25 mm) fitted to an HP 5890 Series II chromatograph. The chromatograph was operating in split mode with temperature programming and a column head pressure of 26 psi. The conditions of temperature programming were: initial temperature 140'C for two minutes, followed by a thermal ramp (300C min-1) to 220'C, and then held at this temperature for 2 minutes. The injection block temperature was 28TC and the detector temperature was 300'C.

The types of product obtained using each system were measured using 1H NMR. A comparison of the 'H NMR of two of the systems (FE/Fichlor and FE/IBDA) with the 'H MNR of HD and its oxidation products is illustrated in Figure 1. The products detected using each of the test systems is shown in Table 1.

Table 1

System 1 Products Recovered DBS Oxidation K10 Oxone Sulphone K10 KMn04 Sulphone K10 Pb (0Ac) 4. DBS and Sulphoxide K10 Tl (N03) 3. Sulphoxide K 10 Fe (N03) 3 Sulphoxide K10 MMPP Sulphoxide and Sulphone K10 MCPBA Sulphoxide and Sulphone K10 / IBDA Sulphoxide (major) and Sulphone K10 / Fichlor (minor) K10 / calcium hypochlorite Sulphone (minor) K10 / Na104 Sulphone K10 / NaB03 DBS and Sulphoxide K10 / Na2C03.1.5H202 DBS (major) and Sulphoxide DBS (major) and Sulphoxide Lead and Thallium are environmental toxins. Extractions were after 24 hours. The Sulphoxide is the desired reaction product.

oxidation of DBS in the presence of K10 was achieved using a range of organic and inorganic compounds.

Reaction with Oxone was rapid, yielding the sulphoxide, and then, by further oxidation, the sulphone, which was the only reaction product detected in solution after 3 hours. The precise results are shown in Table II. Results are expressed as percentage yield/recovery of each component relative to the initial amount of DBS (2mmol).

lo- 11 Table II a Determined by GC. The results are the mean of three analyses.

Weight of K10 = 2.407 g; weight of Oxone = 3.081 g.

Results were obtained in a similar manner for other systems tested. oxidation was more rapid in the presence potassium permanganate, although as before, the sulphoxide reaction product was rapidly oxidised to the sulphone. MMPP oxidation to the sulphoxide was also rapid, further conversion to the sulphone again taking place. Whilst oxidation using metachloroperoxybenzoic acid (MCPBA) was rapid, the reaction was unselective, yielding both the sulphoxide and sulphone in approximately equal proportions. DBS was efficiently and selectively oxidised to the sulphoxide by lead acetate, thallium(III) nitrate or iron(III) nitrate (lower yield of the sulphoxide). Oxidation by IBDA was both very rapid and selective, yielding a high proportion of the sulphoxide reaction product (ca. 90+ %). In the presence of Fichlor, DBS was rapidly removed from free solution, although only small quantities of the corresponding sulphoxide and sulphone were Time % recovery % yield % yield (min) DBS' dibutyl sulphoxide a dibutyl sulphone' 0 106 0 0 81 22 trace 52 50 trace 16 60 3 2 77 9 trace 55 26 0 41 48 0 15 86 0 7 82 18 d- 0 0 100 12 detected. The low mass balance is possibly due to adsorption of the reactant and / or its oxidation products on the oxidant. A low mass balance was also obtained using calcium hypochlorite. However, the sulphone was the predominant reaction product, there being no significant quantities of the sulphide present five minutes after addition of the DBS. Other oxidants screened using DBS were inferior to those described above. Sodium periodate selectively yielded the sulphoxide, although the reaction was slow (ca. six hours). Similar results were obtained using sodium borate. DBS oxidation by sodium percarbonate was more rapid but less selective, a mixture of the oxidation products being formed. No detectable quantities of either dibutyl sulphoxide or dibutyl sulphone were detected when manganese oxide or ammonium nitrate was used, even after prolonged contact (18 and 6 hours respectively). Both sodium iodate and potassium dichromate were inactive. only trace quantities of the sulphoxide and sulphone were detected using either copper nitrate or sodium chlorate. DBS was selectively oxidised to the sulphoxide by hydrogen peroxide or tert-butyl hydroperoxide, albeit very slowly.

Other systems tested were generally less preferred than those listed above. For example, oxidation of DBS by K10 / Na104 was highly selective, the sulphoxide being the predominant product. However, the reaction was very slow (ca. six hours). The rate of DBS oxidation was much reduced compared to Oxone or KMn04. Similar observations were made using NaB03 oxidant in the presence of K10. A slow oxidation of DBS to the sulphoxide took place over ca. 180 minutes, with only trace quantities of dibutyl sulphone being formed. DBS oxidation by Na2C03.1.5H202 was somewhat more rapid but was less selective, yielding a mixture of the oxidation products.

13 Example 2

Interaction of DBS with oxidants in the Absence of K10 To determine the role of the powder support in the oxidation of DBS, selected oxidants were studied in the absence of the support. The methodology used was similar to that described above in Example 1. The results obtained using KMn04 demonstrated that only trace quantities of dibutyl sulphone were formed over three hours, with no dibutyl sulphoxide being detected. This contrasts with the results obtained in the presence of K10, where oxidation of DBS took place to exclusively yield dibutyl sulphone after a two hour reaction period. This result therefore demonstrates the catalytic role of the support in the oxidation of DBS in the presence of KMn04.

The catalytic role of K10 was further investigated using Na104. As before, only trace quantities of dibutyl sulphoxide and dibutyl sulphone were detected during a prolonged reaction period (20 hours). This contrasts with the results obtained in the presence of K10, where oxidation of DBS took place to exclusively yield dibutyl sulphoxide. Whilst this reaction was slow, it was more rapid and complete than the corresponding experiments carried out in the absence of the K10. The support was therefore catalytic in the presence of this oxidant.

It appears therefore that in many cases, the K10 support has a catalytic role in the compositions of the invention.

Example 3 Interaction of CEES with K10 and Oxidant Mixtures Secondary screening of the compositions was carried out using 2-chloroethyl ethyl sulphide (CEES) as a second HD simulant.

On the basis of the results obtained using DBS, eight systems were selected for study using CEES. These were Oxone, MMPP, IBDA, Fichlor, calcium hypochlorite, iron(III)nitrate, 14 potassium permanganate and MCPBA. Lead acetate and thallium(III)nitrate were not investigated further due to the environmental hazards associated with their use and disposal.

The reactions of CEES with K10/oxidant mixtures were performed in a similar manner to the reactions of DBS described in Example 1, but using a 0.20M solution of CEES in DCM. Analysis was by GC. Separation was carried out using a 30 metre capillary column (Restek Rtx0-1, 0.32 mm ID, 0.5 gm film) installed in a HP 5890 Series II chromatograph. The carrier gas was helium at a column head pressure of 7.5 psig. The chromatograph was operating in splitless mode with temperature programming. The conditions of temperature programming were WC followed by a thermal ramp (40C min-. ') to 140'C. The injection block temperature was 250'C and the detector temperature was 3000C. The results of the CEES experiments were determined as the percentage yield / recovery of each component of the reaction mixture detected relative to the initial amount of substrate added (2 mmol).

1 1 In addition to the sulphoxide and sulphone, it was likely that CEES oxidation would yield other products. The products detected are summarised in Table III.

is Table III

CEES Oxidation K10 KMn04 CEES and Sulphone K10 Fe(N03h CEES (minor) and Sulphoxide K10 Oxone (major) K10 MMPP Sulphone K10 MCPBA Sulphoxide and Sulphone K10 IBDA Sulphoxide and Sulphone (major) K10 / Fichlor Sulphoxide K10 / calcium hypochlorite Sulphone (low yield) Sulphone (major) In the case of Oxone and potassium permanganate, ethylvinyl sulphone, 0-chloroethyl ethylsulphoxide and P-chloroethyl ethylsulphone were detected.

FIGURE 1. CEES REACTION PRODUCTS (K10 / OXONE or K10 / KMn04) 0 0 S S 11 c 1 S c 0 0 Ethylvinylsulphone chloroethylethylsulphoxide chloroethylethylsulphone Oxidation of CEES by potassium permanganate was slow, the sulphone being the only reaction product detected. The rate of oxidation was therefore slower compared to DBS. In both cases, the sulphone was the principal reaction product, indicating that chloroalkyl sulphides are likely to oxidise in the same manner as alkyl sulphides. Oxidation by iron (III) nitrate was also slow, only small quantities of reaction products being detected (sulphoxide and sulphone). In contrast, DBS oxidation by this system was very rapid, yielding almost exclusively the sulphoxide. CEES oxidation by Oxone was rapid, the sulphone being the predominant reaction product. The rate of oxidation of CEES to the sulphoxide was again reduced compared to DBS 16 oxidation. The oxidation characteristics of CEES and DBS by MMPP or MCPBA were very similar, there being a rapid decrease in the amount of sulphide in conjunction with a rapid increase in the quantity of sulphoxide (the oxidation of CEES by MPBA was not influenced by the K10 support). In both cases, an accumulation of sulphone was observed. CEES oxidation by IBDA was very rapid, being essentially complete within five minutes, and only the sulphoxide being formed in appreciable yield (in the absence of K10, the rate of reaction of CEES with IBDA was significantly reduced, confirming the catalytic activity of the support: little or no reaction was observed in the presence of the powder alone). This contrasts with the results obtained using DBS, where the reaction was less selective, yielding a mixture of sulphoxide and sulphone. CEES oxidation by calcium hypochlorite or Fichlor was also rapid, the major product being the sulphone. The material balance was low, possibly due to further reaction of the products (for Fichlor, the chromatograms contained extraneous peaks). All the CEES and DBS experiments were carried out using wetted K10 powder. To establish the effect, if any, of the water content of the powder on the activity of the powder and oxidant systems toward CEES, oxidations were carried out using K10 with ca. 6% w/w water (as received) and each of the six oxidants (Oxone, Fichlor, IBDA, MMPP, MCPBA and calcium hypochlorite). With the exception of MPBA and IBDA, the rate and selectivity of the oxidation was reduced, clearly demonstrating the importance of the water content.

Example 4 Interaction of CEES with Zeolite and Oxidant Mixtures In addition to the studies with K10 powder, the performance of some zeolite supports was investigated, being selected on the basis of their reactivity and porosity (zeolites being significantly more chemically reactive than porous activated carbons). The zeolite systems studied were the H' and Na' forms of dry and wetted (water content ca. 16. 7 w/w%) zeolite A, w- 17 ZSM5, zeolite P, mordenite, zeolite Y and organophilic zeolite in combination with either MMPP or IBDA.

The oxidation of.CEES by zeolite/oxidant systems was carried out in a similar manner to that described in Example 3(10 ml of a 0.20M solution of CEES in DCM was added to a mixture of either dry zeolite (2.0 g) or wetted zeolite (16.7 % wlw water, 2.4 g) and the oxidant (5 mmol) under study). In each case, the mixture was stirred for 1.5 hours and an aliquot of the solution analysed by GC. The oxidants used were IBDA and MMPP. For those experiments using IBDA, the samples were stored in liquid nitrogen (IBDA being soluble in DCM).

It was found that a significant loss of the reaction components from free solution did not occur until the port diameter of the zeolite was greater than ca. 6A, demonstrating the geometric constraints imposed on the reaction components diffusing into the porosity of the zeolites. Time course experiments were carried out for some of the zeolite / oxidant systems, some of which promoted rapid and selective CEES oxidation (eg. H'P (wetted) / IBDA selectively oxidised CEES to the sulphoxide within 15 minutes). The Na+ZSM-5 (wetted) / MMPP and the Hp (wetted) / MMPP combinations resulted in the most efficient conversion of CEES to its oxidation products, but in both cases the sulphone was the major product. CEES oxidation by H+P (dry) / IBDA was again rapid. there being little CEES remaining in solution after 15 minutes. Initially, the major oxidation product was the sulphoxide, thereafter conversion to the sulphone took place. In this case, the zeolite water content did not influence the nature of the final reaction products. Na+A (wetted) / IBDAand H'ZSM-5 (dry) / IBDA were also found to selectively oxidise CEES to the sulphoxide. The results obtained indicated that the zeolites offered no advantages over K10.

Is Example 5 Interaction of HD with Powder and Oxidant mixtures. On the basis of the results obtained with CEES, a selection of powder and oxidant combinations were prepared and challenged with liquid HD. The powders used are listed in Table IV.

HD may be oxidised to the corresponding sulphoxide and/or sulphone, the latter product being a vesicant (albeit significantly less toxic than HD). The intention was therefore to identify powder and oxidant systems which would rapidly detoxify HD to yield (predominantly) the sulphoxide.

HD oxidation was studied by addition of 50 gl (0. 4 mmol) of agent to samples of the powder and oxid.ant (ca. 0.48 g. powder to 1 mmol of oxidant) in centrifuge tubes (without addition of solvent), the tubes being agitated using a vortex mixer prior to transfer to a motorised tube roller for up to 24 hours.

Analysis of HD reaction was by 'H NMR spectroscopy (Jeol GSX400 and Jeol Lambda 300 and 500 MHz instruments), extracts from the powder / oxidant being obtained by solvent extraction (1.5 ml d- chloroform). To allow quantification, the internal standards hexamethyldisilane (HMDS, 10 [il (4.7 x 10-5 moles)) or hexamethyldisiloxane (HMDSO, between 6 x 10-6 moles and 7 X 10-6 moles) were added. The 'H NMR spectra were sufficiently resolved to enable accurate integration of the signals at 33.0 ppm (dt, HD), 83.2 ppm (m, HD sulphoxide), 83.6 ppm (dt, HD sulphone) and 80 ppm (s, HMDS). The mass balance of each component was determined, the results being obtained as percentage yield / recovery relative to the initial quantity of HD. Some time course experiments were carried out, extracts being removed at various time intervals for analysis.

The products detected are summarised in Table IV.

19 Table IV

HD Oxidation FE / Fich.lor Sulphoxide and Sulphone (trace) FE calcium hypochlorite Sulphoxide and Sulphone (low FE Oxone yield) FE IBDA Sulphoxide and Sulphone FE MMPP Sulphoxide FE MWBA Sulphoxide (major) and Sulphone (low yield) Sulphoxide (low yield) and Sulphone (low yield) KIO Fichlor K10 calcium hypochlorite HD, Sulphoxide (low yield) and K10 Oxone Sulphone (low yield) HD (low) and Sulphoxide K10 IBDA HD, Sulphoxide and Sulphone (low KIO MCPBA yield) K10 MMPP Sulphoxide and Sulphone (trace) HD and Sulphone (trace) Sulphoxide Furthermore, with the exception of FE/Oxone, no HD or significant quantities of the toxic sulphone reaction product were recovered from the FE/oxidant systems after 24 hours. The mass balance calculations were generally low, indicating that significant quantities of HD (and/or reaction products) were retained on the powder. The exception was the FE / IBDA system, where about 75% of the HD was recovered as the nontoxic sulphoxide.

is Although the mass balance calculations were generally low (with the exception of IBDA), experiments where the powder support of an oxidant/powder system was destroyed (HF t-, digestion) after adsorption of CEES had demonstrated that the strongly adsorbed materials were oxidised almost completely, indicating that where the oxidant is shown to be active, then non-recovered material is likely to be present as strongly adsorbed reaction products. The use of an excess of oxidant had no significant impact on the nature (le. the reaction products) or efficiency of the reaction. Time course extractions using Fichlor, IBDA and Oxone demonstrated that the IBDA system resulted in the most rapid oxidation of HD to the sulphoxide. The Fichlor containing system was less effective, but nevertheless provided acceptable levels of agen t oxidation (to the sulphoxide) combined with very low HD recovery (after 2 hours). The Oxone containing system was also preferred, although there was evidence that further oxidation of the sulphoxide to the sulphone was taking place over a relatively short timescale (2 - 3 hours).

In the absence of additional moisture (le. greater than the as received water content) the K10 powder was less effective than the corresponding fullers' earth systems, with the exception of IBDA and MMPP, from which no HD was recovered. Systems prepared using K10 with an adjusted water content were generally as least as efficient as the corresponding fullers' earth systems after 24 hours contact. However, the differences found using Fichlor or IBDA (which were the most effective) were not very great.

Time course extractions indicated that HD oxidation was generally more rapid for K10/oxidant systems compared to the corresponding FE systems. Whilst the rate of HD oxidation using K10 / Fichlor was similar to the rate observed using FE / Fichlor, significant quantities of the toxic sulphone product were recovered from the K10- based system. The rate of HD oxidation by IBDA in the presence of fullers' earth or K10 was similar and rapid. Whilst differences were also observed using Oxone, these are less important because, unlike Fichlor and IBDA, significant quantities of sulphone were produced using both K10 and fullers' earth based systems. For those W_ 21 systems shown to be effective for HD detoxification (Fichlor and IBDA, with either FE or K10), reaction was complete within 30 minutes.

Example 6 Interaction of VX with oxidants, powders and mixtures as measured by NMR and GC The interaction of W with oxidants, powders and mixtures thereof, was investigated using nuclear magnetic resonance (NMR 'H and 13p) spectroscopic analysis of solvent extracts.

All of the solvent extraction experiments were carried out using powder, oxidant or powder/oxidant mixtures to which 20 1 of W was added. Powder experiments were carried out using 200 mg samples, correction being made for the water content such that the quantity of powder used was the same in each case. oxidant experiments were carried out using an excess of oxidant. Evaluation of powder / oxidant systems was carried out using 200 mg samples of powder to which an appropriate quantity of oxidant was added (except see 2.1.5). The quantity of oxidant was calculated such that there were 3 molar equivalents of active oxidant per mole of VX. This is based on previous work which identified that complete oxidation of W required 3 moles of active oxidant species. The oxidants used in this study were Oxone (potassium peroxymonosulphate), Fichlor (sodium dichoroisocyanurate), calcium hypochlorite, iodobenzene diacetate (IBDA), magnesium monoperoxyphthalate (MMPP) and meta-chloroperoxybenzoic acid (MCPBA). The quantities of oxidant required to yield a 3 molar equivalent for 20 gl of VX are shown below in Table VI:

22 Table VI oxidant Weight (mg) Oxone 207.5 Fichlor 74.22 calcium hypochlorite 48.3 iodobenzene diacetate 217.42 magnesium monoperoxyphthalate 333.9 meta-chloroperoxybenzoic acid 194.2 Samples were weighed into centrifuge tubes fitted with screw cap lids. After addition of VX, the samples were thoroughly mixed using a vortex mixer. Thetubes-We-re then transferred to a multimix (motorised tube roller) and left for approximately 2-4 hours, after which the mixture was agitated to ensure that no free liquid agent remained unadsorbed. After a total of 24 hours, solvent (either 2m1 of dchloroform, d4-methanol or d3nitromethane) was added and the samples replaced on the multimix for a further 30 minutes. The tubes were then transferred to a centrifuge (10 minutes @ 200Orpm). The supernatant liquid (ca. lml) was transferred to an NMR tube via a fibre filter to remove any entrained powder particles. A pre-weighed quantity of the NMR reference hexamethyldisiloxane (HMDSO) was added to each tube prior to measurement of the spectra. NMR spectra were recorded using Jeol Lambda 300 Mhz ('H) and 500 Mhz (31p) instruments. At each stage of the process, all relevant weights were recorded to facilitate subsequent quantification of any reagent of products in the extract. Quantification was achieved by comparing the 'H integrals for specific signals due to VX and/or its reaction products with the integral obtained for the NMR reference compound: these values were then correlated with the original weights recorded to determine the quantity of VX and/or reaction products present in the extract. In addition to extractions carried out after addition of VX, some powder (or oxidant, or powder plus oxidant) samples were doped with ethyl methylphosphonic acid (VPA) or S23 diisopropylamino)ethyl thiol (VS). These chemicals are potential reaction products of VX and the purpose of testing these was to determine the extraction efficiencies of any reaction products formed and to aid in interpretation of the NMR spectra. The percentage recoveries recorded were relative to the initial quantity of VX (or VS/VPA) added to the oxidant, powder or oxidant/powder mixture.

GC-MS analysis was of the solvent extracts after completion of 10 the NMR analysis. Analysis was carried out using a Hewlett Packard GC-MS fitted with a 25m J&W Scientific M column (internal diameter 0.2mm, film 0.33pm). The thermal profile used for the analysis was 500C for 3 minutes followed by a thermal ramp of 200C min-1 to 32CC (held -for 3 minutes) When carried out, derivatisation was by methylation (diazomethane, to enable acid products to be identified). The MS was operated either in EI (electron impact) or CI (chemical ionisation) mode.

is The NMR results indicated that, when the spectra were sufficiently resolved, the reaction products were the same, no VX being recovered. The reaction products VPA (ethyl methylphosphonic acid) and VXO (0-ethyl S2(diisopropylamino)ethyl methylphosphonosulphoxide) were found (VXO was identified on the basis of the results of oxidation of VS (S(diisopropylamino) ethyl thiol). 'H NMR assignments shown in Table VII. NMR spectra of VS, VSO, VX and VXO as well as those obtained for are shown in figure 2.

24 Table VII

H NMR ASSIGNMENTS (d-chloroform) Compound Signal; Chemical Shift (ppm); Multiplicity (see structures) vx a; 1. 3; b;4. 1; c; 1.75; d;2.55; e;2.75; f;2.9; g; 1. 0; triplet multiplet doublet triplet multiplet septet doublet vX0 a; 1.4; b;4. I; cj.75; d;3.2; e;3.41; f;3.65; g;1.25; multiplet multiplet doublet multiplet multiplet multiplet multiplet vs d;2.45; e;2.55 f;2.95; g;0,95; triplet multiplet septet doublet VS0 - - d;12; e;3.45; f;3.65; g;1.35 triplet triplet septet double doublet VPA a; 1.3; b;4.0; c; 1.4; triplet multiplet doublet The results are summarised in Table VIII Table VIII

System T Products Recovered Oxidants VS Oxone VS0 VS Fichlor VS0 VX Oxone W0, VPA VX 1BDA VXO, VPA VX calcium hypochlorite W0, VPA (poorly resolved NMR) VX Fichlor W0, VPA (poorly resolved NMR) W MCPBA Unknown Powders FE VX, VPA K10 VPA silicalite VX, VPA, pyrophosphonate compound KSF VX, W0, VPA XE-555 None neutral alumina VX, W0, VPA, fine alumina diisopropylethylamine acidic alumina VX, VPA, pyrophosphonate compound VX, VPA, pyrophosphonate compound Powder Oxidant Combinations FE Oxone VPA FE Fichlor VPA, diisopropylethylamine FE calcium hypochlorite VX, VPA, pyrophosphonate compound FE IBDA VPA, VSO, W (trace) FE MMPP VPA, pyrophosphonate compound, diisopropylethylamine K10 Oxone VX, VPA, VSO, K10 Fichlor di(diisopropylamino)ethyl K10 calcium hypochlorite disulphide K10 IBDA VPA, VSO, diisopropylethylamine K10 MMPP VPA, di(diisopropylamino)ethyl disulphide, VX VX, VPA, VSO VPA, VSO, pyrophosphonate compound 1 is 1 26 Recovery of VX and / or reactions products from the powder supports using d-chloroform was generally very low. Irrespective of the water content (which significantly influenced HD detoxification) no VX could be recovered from the K10 powder (presumably due to salt formation via protonation of nitrogen on the acidic support). Only unreacted VX was recovered f rom FE (ca. 12% recovered). d4-Methanol was used to determine whether other VX degradation products were present which were not extracted using d-chloroform. VPA, the corresponding methyl ester, and VX were extracted from K10 or FE using d4- methanol to which W was added (extraction being significantly more effective from FE). The presence of free acid indicated that W decomposition occurred on the powder surface. The predominant product found was VPA and the VPA methyl ester (about 50%). Approximately 7.5% of the W was recovered. There were no clear signals which could be assigned to the sulphur containing fragment of the molecule. The results indicate that reaction of W takes place on the FE powder, but that extractable W remains for at least 24 hours.

The fate of W on the K10 powder was less clear, however, the absence of significant quantities of the VPA (or the methyl ester) suggests that W is largely unchanged (possibly due to salt formation between W and the powder surface).

It was not possible to further enhance recovery of VX or reaction products using d3-nitromethane, or acidified d4methanol (5% hydrochloric acid) to displace reactant or products from the surface of the powders. Much of the work concentrated on FE and K10. Other powder supports, identified during earlier work with sulphur mustard, were evaluated. Silicalite powder offered no advantages, small (ca. 0.2%) quantities of VX being recovered using dchloroform, and VX and VPA being recovered using d4methanol (24% and 8% respectively). VPA and VXO were recovered from KSF powder (30% and 27% respectively). That significant quantities of reaction products, but no VX, was recovered highlights the reactivity of KSF powder. No W or reaction products were recovered from the XE-555 powder, due to the presence of porous carbon in the 27 formulation which resulted in adsorption of VX into the pores. If developing a non-reactive adsorbent powder, then a carbon based system would be the obvious choice. In the present case, carbon based systems are unsuitable because the agent must be accessible to the oxidant. Prior work showed that a VX simulant adsorbed onto such a powder was not decomposed [Wagner, G.W. et al., J. Molec. Catal. A, 99, 1995, 175], meaning that the powder may pose a potential desorption hazard. The powder also possesses a lower density (compared to eg. fullers earth), with the consequence that it is likely to be more prone to uncontrolled dissipation. Small quantities of VX were found in the extract from neutral alumina, as were signals consistent with diisopropylamine, VPA and VXO. Significant (ca. 75% and 67%) quantities of VX were recovered from the fine alumina and acidic alumina supports in addition to small amounts of VPA.

On the basis of the results, FE and K10 powders and various oxidants were selected for evaluation of VX destruction efficiency. Whilst the water content of the powder for FE Oxone had no marked effect on VX reaction, slightly more VPA was recovered from the powder with the higher water content (maximum VPA recovery was 16%). No VX was recovered, and no signals were observed in the 31 P spectrum. GC-MS analysis indicated that diisopropylethylamine and VPA were present, neither of which are known to be toxic. VX was not found in the 1 H spectrum of the d4-methanol extract from K10 / Oxone, although signals in the 1 H spectrum indicative of VXO / VX and VPA were found. VPA recovery was ca. 20%. A small signal at ca. 56 ppm, indicative of VX, was observed in the 31 P spectrum, which was confirmed by GCMS (figure 3).

NO VX was recovered from FE / Fichlor, although VPA was present (about 25%). VS, or the analogous disulphide, was observed, although definitive assignment was not possible due to insufficient spectral peaks. VPA and the VPA methyl ester C- 28 were detected in the 31 P spectrum, no VX being present (figure 6). No VX was detected in the 1 H and 31 P spectra of the extract from K10 / Fichlor. This was confirmed by GC-MS. VPA, present as the free acid, or as the methyl ester, was present in the 31 P spectrum. Recovery of the VPA ('H spectrum, which also indicated the presence of VSO) was ca. 30%. Diisopropylethylamine was also detected (GC-MS).

According to 'H NMR no VX could be extracted from FE / IBDA, although VPA and VSO were present (VPA recovery was 24%). Two weak signals were observed in the 31 P spectrum, which were due to VX (56 ppm) and the pyrophosphonate compound (ca. 20 ppm). VX was detected b y 31 P NMR in only a sma. 11---quantity. VPA and (signals consistent with) VSO were found in the 1 H spectrum of the extract from the K10 / IBDA system. VPA recovery was approximately 60%. The 31 P spectrum indicated that (trace) amounts of VX were present.

Most of the extract from FE / calcium hypochlorite was VX (1H and 31 P spectra), about 12% being recovered. VPA and the pyrophosphonate were also detected. The pyrophosphonate product is toxic (about ten times less toxic than VX (Yang, Y C et al., J. Am Chem. Soc. (1990) 112, 6621). Trace quantities of VX were detected by GC-MS, as was the disulphide compound.

Analogous behaviour was found for K10 / calcium hypochlorite (ca. 9% VX recovered). VPA (recovery about 15%), and the disulphide (recovery about 60%) were also detected ( 1 H). The pyrophosphonate compound was detected in the 31 P spectrum, and GC-MS indicated that VX (trace), VS and diisopropylethylamine were present in the extract.

The spectra obtained for FE / MMPP were poorly resolved, although the VPA methyl ester and pyrophosphonate compound were present ( 31 P). Diisopropylethylamine was detected (GC- 1-, 29 MS), but no VX. The 1 H spectra obtained for K10 /MMPP were also poorly resolved: hence calculation of recovery was not possible. However, some signals consistent with VPA and VSO were found. The pyrophosphonate compound was detected in the 31 P spectrum.

Example 7 VX Detoxication rate studies In order to obtain a measure of the rate of VX detoxification, selected samples were prepared and analysed as described in Example 6 above except that analysis was carried out after extracting the samples at varying time intervals (typically 0.5, 1, 2 and 5 hours after addition of,VX). Agitation to ensure that no free liquid agent remained unadsorbed was carried out immediately after addition of the VX.

Measurement of the rate of VX detoxification by the FE / Oxone, FE / Fichlor or K10 / Fichlor systems (irrespective of the ratio of powder and oxidant to the (fixed) quantity of VX) demonstrated that no VX could be recovered after 30 minutes (or thereafter). Where products were found, only VPA and / or the VPA methyl ester were detected. None of these extracts showed anticholinesterase activity. The results are illustrative of the effectiveness of these particular systems. All of the 'H spectra of the extracts from the K10 / Oxone and K10 / IBDA samples were consistent with the presence of VXO. The use of greater quantities of powder / oxidant or longer contact times did not diminish the signal due to this product.

W1 I Example 8 Vapour desorption Vapour desorption measurements will identify the potential vapour desorption hazard from the mixtures, but will not indicate the potential for agent leeching from the powders by the action of eg. water.

Vapour desorption measurements were carried out using a sealed desorption cell apparatus. Samples were subjected to an air purge (200 ml min -1), the effluent air being passed through a bubbler trap containing 2 ml (4 ml for overnight collection) of diethyl succinate. The relative humidity of the purge air supply was ca. 60% at 25 0 C and ca. 20% at 500C (sparger containing a saturated solution of potassium bromide, temperature control being via an environmental cabinet). Collection of agent vapour was made over at least 24 hours. 400 mg samples of powder, to which an appropriate quantity of oxidant was added, were used. Analysis was by GC (Perkin Elmer, fitted with an M operating in phosphorus mode; a 15 m x 0.53 mm I. D fused-silica capillary column (DB1701) with a 50 cm deactivated silica retention gap was used; the initial temperature was 170 0 C (1 minute), thereafter rising at 30 0 C min- 1 to 260 0 C for 2 minutes. The injector port and detector temperatures were 2000C and 250 0 C respectively. The carrier gas was helium at 10 ml min- 1).

No VX or volatile reaction products were detected 25 0 C f rom any of the powders and powders and oxidants. VX was detected in some of the air samples collected from the FE (wetted sample only, 26% w/w water: no desorption was observed from the as received sample - 16.5% w/w water) (maximum 0.46 ig over 30 minutes - equivalent to 6 litres of air). VX was also detected in two of the air samples collected from the XE-555 powder at 49 0 C 10C (0.44 n over 15 hours). That VX was not 31 found in air samples from the powder and oxidant systems indicates the value of incorporating reactive species into the powder, particularly in hot environments where vapour desorption is more likely.

The acetylcholinesterase inhibition activity of some of the extracts was also measured to ensure that they did not contain any toxic products (such as trace amounts of VX).

Measurements of acetylcholinesterase inhibition activity demonstrated that systems such as FE / Oxone and FE / Fichlor would not pose a hazard, the agent being destroyed to yield inactive products. Those systems from which the pyrophosphonate was recovered did inhibit acetylcholinesterase.

Example 9 GD Decontamination A range of solid materials was examined for the ability to decontaminate GD. For these tests, liquid GD (20gl) was added to a nine molar excess of solid (in the case of resins, an arbitrary weight of 200mg was used). After thorough mixing, each sample was agitated using a rotary mixer for 24 hours. Each sample was then extracted using CDC13 (2m1) and the liquid extract examined by nmr spectroscopic analysis. The results are summarised in Table IX.

32 Table IX f System GD present imidazole No 2-(dimethylamino)-methylhydroxypyridine yes 2-pyridinaldoxime yes 2-mercaptoethylamine yes 1,12-diaminododecane yes 19H,31H-phthalocyanine yes sodium metatungstate yes sodium vanadate no sodium molybdate yes cerium (III) chloride yes sodium thiosulphate yes copper (II) hydroxide no urea yes thiourea yes urea/hydrogen peroxide yes histidine yes pyrogallol yes 2,21-dipyridylamine yes calcium hydroxide no imidazole sodium salt no catechol yes 2-mercaptobenzimidazole yes 2-aminophenol yes sodium phenoxide no Fichlor yes IBDA yes Oxone yes Amberlyst 15 (wet) no Amberlite IR-118 no Dowex HCR-W2 no sodium percarbonate no sodium perborate yes Fullers earth (FE)(as received) yes K10 (wet) yes K10 (as received) yes Fullers Earth (wet) yes 33 Degradation products were obtained. From the data presented, it can be seen that ten powders examined destroy GD over 24 hours, namely imidazole, sodium vanadate, copper(I1)hydroxide, imidazole (sodium salt), sodium phenoxide, Amberlyst 15 (wet), Amberlite IR-118, Dowex HCR-W2 and sodium percarbonate. At present, time course experiments are being carried out to determine which of these ten powders is the most rapid at destroying GD.

L- Materials active for detoxification of GD, such as the ten materials described above, when combined with clay/oxidant systems, would allow the production of a general purpose decontaminant for H, G and V agents. Preferred compositions will contain more than one active agent-to ensure activity 15 against a range of threat materials.

d 34

Claims (10)

Claims

1. A material which reacts with at least one toxic chemical and converts it to a non- hazardous residue comprising an acidic resin.

2. A material as claimed in claim 1 wherein said acid resin is selected from Amberlyst (wet), Amberlite IRA 18, Dowex HCR-WI

3. A material as claimed in any preceding claim wherein said acid resin is also an absorbent support material.

4. A composition comprising a material as claimed in claim 3 supporting an active reagept which reacts with at least one toxic chemical and converts it to a nonhazardous residue.

5. A composition as claimed in claim 4 wherein said active reagent is a base, a nucleophile an oxidising agent a quarternising agent or a reducing agent.

6. A composition as claimed in claim 5 wherein the oxidising agent is N (magnesium monoperoxyphthalate), Oxone (potassium peroxymonosulphate), MCPBA (m-chloroperoxybenzoic acid), 1BDA (lodobenzene diacetate) calcium hypochlorite or Fichlor (sodium dichloroisocyanunarate)

7.

A composition as claimed in claims 4 to 6 which comprises 5 to 20wt% of active reagent.

A composition as claimed in claim 3 which support at least a further active reagent which is active against a different toxic chemical.

9. A composition according to claim 8 wherein said first reactive agent is active against HD and VX and a second active reagent is active against GD.

10. -A method for removing at least one toxic chemical from an area, said method comprising applying a composition as claimed in any of claims 1 to 9.