skip to: page content | site navigation | section menu
Journal of Medical Chemical, Biological and Radiological Defense
J Med CBR Def  |  Volume 3, 2005
Received 03 Aug 2004 |  Accepted 20 Aug 2005  |  Published 20 Sept 2005

Application of NMR and EPR Spectroscopy to the Analysis of the Reaction of Phosphovanadomolybdate Polyoxometalate (H5PV2Mo10O40) with Chloroethyl Sulfides (Half-Sulfur Mustard and Sulfur Mustard)

Carmen M. Arroyo 1,*, James M. Sankovich2, Damon L. Burman3, David W. Kahler1, Sunil-Datta Soni1, Brennie E. Hackley Jr.1,4

1 U.S. Army Medical Research Institute of Chemical Defense Drug Assessment Division, 3100 Ricketts Point Road, Edgewood Area, Aberdeen Proving Ground, MD 21010-5400

2 USAMRD, MCMR-UWB-L, Brooks AFB, TX 78235-5138

3 ORISE, Oak Ridge Institute for Science and Education, Oak Ridge, TN

4  Scientific Advisor

* Corresponding author:

Commander,

U.S. Army Medical Research Institute of Chemical Defense

ATTN: MCMR-UV-DA, Dr. C. M. Arroyo

Aberdeen Proving Ground, Maryland, 21010-5400

Tel.: 410.436.4454 |  Fax: 410.436.4147 |  E-mail: Carmen.Arroyo@amedd.army.mil

abstract

Magnetic resonance spectroscopy, nuclear magnetic resonance (NMR) and electron paramagnetic resonance (EPR) techniques support the oxidation of sulfur mustard (bis- (2-chloroethyl) sulfide; military designation, HD) and its analogous compound (2-chloroethyl ethyl sulfide; H-MG) by the vanadium-substituted heteropoly acid, H5PV2Mo10O40, readily accessible and significantly selective. All detected reaction products were characterized by 13C-NMR spectroscopy. The detected products are the reductively dehalogenated 2-ethylthioethanol, an organosulfur oxyacid, a cyclic sulfoxide, and the sulfone. 31P-NMR spectra exhibited significant up-field chemical shift changes in the presence of the sulfur mustard and its analogous compound. The distinction in the up-field chemical shift was modest in the 31P-NMR spectra; however, the up-field chemical shift in the 51V-NMR spectra was exceptionally conspicuous. V2P resonance at δ 523.12 ppm was shifted up-field by more than twice the amount of any other PVV2Mo10O40-5 resonances. The fact that an up- field shift was observed in the presence of the sulfur mustard or its analogous compound supports the presence of the V+4 paramagnetic cation. The presence of the paramagnetic V+4 ion was verified by EPR since the reduced form of the polyoxometalate, PVIV2Mo10O4040-7 was detected. It was noted that the most characteristic and distinctive feature of H5PV2Mo10O40 consists of its propensity for easy O-atom transfer to the donor substrate (S) yielding oxidation product SO. Decamolybdodivanadophosphate (H5PVV2Mo10O40) directly oxidizes the substrate (HD or H-MG) and a terminal oxidant reoxidizes the reduced form of the polyoxometalate (PVIV2Mo10O40 -7):

INTRODUCTION

Polyoxometalates (POMs) have been used to catalyze the oxidative breakdown of halogenated organic compounds and are good candidates to inactivate chemical warfare agents (CWAs) (Khenkin, 2001). Many of the POMs undergo rapid, reversible redox changes. Hill (Johnson, 1999) introduced the concept that POMs could be used as a reactive component in the existing topical skin protectant (TSP) (McCreery, 1997), or barrier cream, against chemical warfare agents (CWAs).

Sulfur mustard (HD) received its name from its smell (onion, garlic, mustard) (Medema, 1986). When mustard was first used by the Germans, the Allies called it Hun Stoffe (German stuff), abbreviated HS; later it became known as H (Medema, 1986). Distilled, or near pure, sulfur mustard is known as HD. The aim of the present was to apply magnetic resonance techniques, namely nuclear magnetic resonance (NMR) and electron paramagnetic resonance (EPR), to obtain a more complete picture on the mechanism of the chemical reaction of a selected POM with sulfur mustard and its analogous compound.

Here, we present evidence that vanadium containing polyoxometalates of Keggin (Haimov, 2002; Neumann, 1997; Neumann, 1998; Khenkin, 2000; Pope, 1983) structure H5PVV2Mo10O40 (Figure 1) can activate sulfides and oxidizes such substrates yielding the product and reduced catalyst. Reoxidation of the catalyst by dioxygen and formation of water then follows. The regeneration of the original oxidized form of the catalyst by reaction with molecule of oxygen is very fast suggesting that H5PVV2Mo10O40 could be used as catalyst for the decontamination of sulfur mustard and its analogous compound.

Experimental Section
Materials and Instruments:

NMR solvents, available from commercial sources (Sigma Chemical Company and Aldrich Chemical Company, St. Louis, MO, USA), were of the highest purity available, and were used without additional purification. The metal-free water was obtained from a Millipore water purification system (Quantum EX Ultrapure Organex cartridge, Bedford, MA, USA). Half-sulfur mustard (2-chloroethyl ethyl sulfide; H-MG) was obtained from Sigma Chemical Co. ( St. Louis, MO, USA). Sulfur mustard (bis- (2-chloroethyl) sulfide; HD) was acquired from the United States Army Soldier and Biological Chemical Command (Aberdeen Proving Ground, MD, USA). Professor Craig L. Hill, Department of Chemistry, Emory University, Atlanta, GA, USA provided the H5PV2Mo10O40 polyoxometalate. H5PV2Mo10O40 polyoxometalate was used as received. No further purification was made. Hydrated PVV2Mo10O40-5 was prepared by drying the POM at 120° C for 24 h. The quaternary ammonium salt [(C4H9)4N]5PV2Mo10O40 was prepared by mixing 10 equiv of tetrabutylammonium bromide [(C4H9)4NBr; Sigma Chemical Co.; T 2390] dissolved in water and an aqueous solution of H5PVV2Mo10O40. The precipitate was removed by filtration and dried overnight in a vacuum oven at 85° C. The thermogravimetric analysis showed the absence of water (data not shown).

Nuclear Magnetic Resonance (NMR):

13C-NMR (150.81 MHz, internal standard) and 31P-NMR (242.78 MHz, 85% H3PO4 external standard) measurements were taken on a VARIAN 600 MHz INOVA high-resolution spectrometer (VARIAN Nuclear Magnetic Resonance Instruments, Palo Alto, CA, USA), equipped with an Oxford Instruments LTD magnet. The 13C and 31P NMR measurements were obtained on a 600 MHz 5mm PENTA PFG probe. 51V- NMR [157.66 MHz, vanadium (V) oxytrichloride (VOCl3) used as external standard] measurements were taken using a 600 MHz SWITCHABLE probe. All one-dimensional experiments were performed at temperatures ranging from 25° - 37°C. Data acquisition, processing, display, and analysis were performed on SUN ULTRA 10 using VNMR 6.1C software.

Electron Paramagnetic Resonance (EPR):

Fifty microliters ( µL) of a stock solution of H5PVV2Mo10O40 (4 mM) in 950 mL of metal-free water were transferred under N2 gas at 37° C into an EPR quartz capillary tube (50 µL; dimensions: o.d. 3 mm; i.d. 1.99 mm; wall 0.48 mm; Wilmad Glass, Buena, NJ, USA). The EPR spectrum was immediately recorded at room temperature. The EPR powder spectra were obtained from 35 mg of hydrated PV2Mo10O40-5, transferred into an EPR quartz tube at 25° C. All EPR spectra were recorded on a Bruker ESP 300E X-band EPR spectrometer at 100-KHz magnetic field modulation and a microwave frequency 9.78 GHz. The magnetic field was set at 348.0 mT, the microwave power at 5.02 mW , the modulation amplitude at 0.1 mT (unless otherwise indicated), the time constant at 655 ms , the sweep time at 335 s , the receiver gain at 5.0 x 104, and the sweep width at 300.0 mT. The Bruker EPR software tools were used for the computer simulation.

Determination of Rate Law for the Reaction of the HD with H5PVV2Mo10O40:

The initial–rate method was employed to determine the reaction order. The initial rate was obtained by following spectrophotometrically the formation of the reduced form of H5PVV2Mo10O40 (λ max = 780 nm / e = 1430 M-1 cm-1 in CD3CN). For the determination of the order with respect to HD, six vials each with 0.0280 g of H5PVV2Mo10O40 , 40 μL of D2O and 4 mL of CD3CN, were allowed to stand overnight to ensure complete equilibration. Six vials each with 1 mL of CD3CN and varying amounts of HD from 0.25 - 4.5 μL were prepared. The reactions were conducted in a 1-cm pathlength quartz cuvette with an affixed glass stopcock. Immediately before the experiment, both the H5PVV2Mo10O40 and HD solutions were degassed by a repeated vacuum/nitrogen gas cycle three times. Next, 1.0 mL of the solution was injected through a rubber septum stopper into a nitrogen-filled cuvette, and then the cuvette was placed into a UV-Vis (Varian) spectrometer. The reaction was initiated by adding 0.25 to 4.50 μL of HD solution into the cuvette, in which the contents were magnetically stirred. The timer was started on the addition of the HD. The absorbance was measured every 30 s for 8 min and then less frequently until it leveled off. The final concentrations of the reactants in the cuvettes were as follows: H5PVV2Mo10O40, 1.60 x 10-4 M; D2O, 2.20 x 10-1 M; 0.40% by volume; HD from 4.40 x 10-1 to 1.70 x 10-1 M.

Results
NMR Spectroscopy:

The oxidation of half sulfur mustard (I) catalyzed by the heteropolyacid, H5PVV2Mo10O40 , was monitored by NMR. This reaction was performed in deuterated water at 37° C generating several products given in Scheme 1. Figure 2 represents a continuous 13C-NMR 1-D experiment recorded for 15-hrs and shows a time averaged product distribution.

 

Scheme 1.

In addition to the excess 2-chloroethy ethyl sulfide (I), with the characteristic 13C-NMR resonances δ C4 42.31, C3 34.89, C2 25.40 and C1 14.24 ppm (Figure 2), the 2-ethylthioethanol (II) was observed at δ C4 60.52, C3 33.04, C2 25.27 and C1 13.98 ppm. An organosulfur oxyacid, a sulfenic acid, was detected displaying chemical shifts of 8.34 and 40.08 ppm assigned to be III . As the reaction proceeded for 15-hrs the 13C-NMR spectrum exhibited resonance peaks at 25.44 and 56.34 ppm corresponding to a cyclic sulfoxide product IV. The yield of the reaction was 99.9% with the final relative product distribution of 10% for compound I, 20% for compound II, 20% for compound III and 50% for compound IV.

 

When the reaction with the sulfur mustard was carried out in deuterated water at 37°C a sulfone V was the only product observed after18 hrs (Scheme 2).

Scheme 2.

Figure 3 shows the 13C-NMR spectrum of the sulfone V with resonances at δ 6.99 and δ 33.97 ppm. Since each H5PVV2Mo10O40 molecule can be formally considered as a two-electron oxidant, the conversion into sulfone was 99.9 %.

Multiple 31P and 51V NMR resonances for the decamolybdodivanadophosphate in deuterated aqueous solutions evidence the coexistence of multiple positional isomers. These positional isomers exhibit, among them, five resolvable 31P NMR resonances, and six resolvable 51V NMR resonances, Figure 4 and Figure 6, respectively. Equilibrated solutions containing decamolybdodivanadophosphate exhibited identical 31P and 51V NMR spectra, when at the same concentration and pH³ 2.

The five- 31P NMR chemical shifts varied over HD concentrations that ranged from 0.0 to 4.4 x 10-4 M; no additional resonance s corresponding to H5PVV2Mo10O40 species were revealed. 31P-NMR resonances at δ–3.64 ppm and δ –3.47 ppm exhibit significant chemical-shift changes in the presence of HD (Figure 4). The up-field shift of δ –3.64 ppm fits the normal expectation that a decreased electron density on phosphorus increases shielding, decreases p-orbital unbalancing, and decreases deshielding (downfield shift) of the phosphorus nucleus (Gorenstein, 1984). These shift variations may also relate to O-P-O torsional-angle changes with increasing HD concentrations (Gorenstein, 1984). The detected up-field shift in the presence of H-MG or HD supports the presence of V+4 cations.

A typical 51V-NMR spectrum of H5PVV2Mo10O40 in D2O is shown in Figure 5. In the 31P-NMR spectra of the reaction of H5PVV2Mo10O40 with HD, the up-field chemical-shift change is moderate (Figure 4), but in the 51V-NMR spectra (Figure 6) of the reaction of H5PVV2Mo10O40 with HD, it is exceptionally evident. The 51V2P resonance at δ–523.12 ppm is shifted up-field by more than twice the amount of any other H5PVV2Mo10O40 resonance, as shown in Figure 6. The 51V chemical shifts are not measurably dependent on the concentration of the decamolybdodivanadophosphate, but the 31P chemical shifts are. Presumably, this is a bulk susceptibility effect. The 31P chemical shifts are also dependent on the ionic medium concentration, which is held constant in the present study.

EPR spectroscopy:

The V+4 cation has a single unpaired electron in its lowest lying dxy orbital and its interaction with the 51V nucleus (99.7% abundance, I = 7/2) yields a sharp eight-line EPR isotropic spectrum in solution at room temperature (Bolton, 1972). The fifteen-line EPR spectrum illustrated in Figure 7 shows the anisotropic hyperfines of the hydrated powder of PVV2Mo10O40-5. In the aprotic dimethyl sulfoxide (DMSO) solvent, the EPR spectrum of the decamolybdodivanadophosphate shows the lowest intensity (Figure 8A), which indicates the presence of a small amount of V+4 cations . The reaction of the H5PVV2Mo10O40 polyoxometalate with the sulfur mustard increases the quantity of V+4 cations , since the EPR signal intensity is raised by 1.6 fold (Figure 8B) . Evidence for the appearance of the reduced species PVIV2Mo10O40.

We attempted to detect the ethylenesulfonium radical cation in situ by applying EPR-spin trapping techniques (Arroyo,1999); however, we were unable to observe the ethylene sulfonium radical cation under these experimental conditions used. Also previously, we showed that the radical cation of the half-sulfur mustard (I) or sulfur mustard could not be observed by EPR-spin trapping techniques (Arroyo, 1999).

The Reaction Rate of the Sulfur Mustard with H5PVV2Mo10O40 :

 

The course of the reaction

(ClCH2CH2) 2S + 2H2O + H5PVV2Mo10O40 -----> (CH3CH2) 2S = O + 2HCl + H7PVIV2Mo10O40 + ½ O2 (Reaction 1)

Scheme 3.

was evaluated by using the integrated first-order rate law (Eq. 1) under pseudo-first-order conditions.

 

At = A + (Aο– A) exp (-kobst) (Eq. 1)

 

The concentrations of HD, H2O, and H5PVV2Mo10O40 were 4.4 x 10-1 M, 2.0 x 10-1 M and 4.61 x 10-4 M, respectively. The following values were determined experimentally: A = 3.366, Aο = 1.110 and slope = 0.016; the calculated rate constant from Eq. 1 was 3.40 ± 0.05 s-1. The reaction in Eq. 1 take place by initial thioether reduction of polyoxometalate and subsequent reoxidation of the reduced polyoxometalate by the terminal oxidant. This reaction has been previously reported by Hill (Johnson, 1999), which indicates that the polyoxometalate directly oxidizes the substrate and a terminal oxidant reoxidizes the reduced form of the polyoxymetalate.

Discussion

Sulfide oxidation by H5PVV2Mo10O40 was monitored via 13C-NMR to determine the identity of the products. The oxidation was observed by the change in the 13C-NMR spectra of the sulfide and the changes the complexes 51V-NMR spectra. It was noted that the most characteristic and distinctive feature of H5PVV2Mo10O40 consists of its propensity for easy O-atom transfer to the donor substrate (S:) yielding oxidation product SO. Decamolybdodivanadophosphate (H5PVV2Mo10O40 ) facilitates the oxidation of sulfur mustard by directly oxidizing the substrate (HD or H-MG) and an oxidant reoxidizes the reduced form of the polyoxometalate (PVIV2Mo10O40-7):

 

HD or H-MG + H5PVV2Mo10O40 ------> sulfone + H7PVIV2Mo10O40 (Eq. 2a)

H7PVIV2Mo10O40 + ½ O2 ------> H5PVV2Mo10O40 + H2O (Eq. 2b)

 

This was verified by the V+5 <-> V+4 transformation, this transformation was observed by EPR spectroscopy. The V+5<-> V+4 is actually responsible for the redox activity of the heteropolyacid, H5PVV2Mo10O40 (Khenkin, 2000). Furthermore, the vanadium substituted heteropolyanions have a fairly high value of oxidation potential (0.7 relative to normal hydrogen electrode (NHE)) and are capable of oxidizing substrates ranging from organic to inorganic compounds. They are reversibly acting oxidants i.e., their reduced forms can be reoxidized to the original form by oxygen in mild conditions.

Our magnetic resonance data indicate that the reaction under study takes place by initial reduction of polyoxometalate with chloroethyl sulfide, to afford the reduced form PVIV2Mo10O40-7 , as was observed by EPR. In accord with reported work on the reduction of polyvanadomolybdic acids (Sharer, 1991; Kozhevnikov, 1998; Muller, 1998; Petterson, 1994; Weinstock, 1998) and the known redox potentials, only the VV and not the MoVI ions are reduced (Khenkin, 2000). The observed chemical up-field shift for the 31P or 51V metal centers most likely results from the paramagnetic V4+ cations of the reduced form of the polyoxometalate, PVIV2Mo10O40-7 when HD or H-MG are present. This observation was confirmed by the appearance of a reduced blue species, PVIV2Mo10O40-7 , at 750 nm in the visible spectrum (data not shown).

Aerobic heterogeneous catalytic oxygenation of sulfides with transition metal oxides involves activation of S-R’2 bond and transfer of lattice oxygen from the polyoxometalate to yield the product and the reduced catalyst. Re-oxidation of the catalyst by dioxygen and formation of water then follows.

 

Substrates (HD or H-MG) + Catalystox ------> Product + Catalystred

Catalystred + 2 H+2 + ½ O2 ------> Catalystox + H2O

 

The experiments described above indicate that oxygenation occurred by oxygen transfer from a lattice oxygenation of the H5PVV2Mo10O40 to the sulfide substrate. Further aspects of the oxidation mechanism and catalytic cycle will be studied in the near future.

tables

Table 1. 13C-NMR Chemical Shifts of the Half Sulfur Mustard and the Identified Reaction Products [go to text reference]

Chemical Structure
Identified
13
C atoms
Observed
δTMS Scale (ppm) *
Predicted
δ (13C) Values
Configuration Limits
(calculated error)

I
C3
33.04
35.60
1.10
C4
42.31
42.76
1.10
C2
25.40
26.38
1.10
C1
14.24
15.11
1.10
       

II
C3
34.89
34.30
0.50
C4
60.52
61.00
0.50
C2
25.27
26.00
0.50
C1
13.98
14.68
0.50
       

III
C2
25.44
24.23
1.02
C1
34.89
34.25
1.00

IV
C2
25.44
1.12
1.27
C1
56.34
1.43
1.10

V
C2
42.25
44.30
2.10
C4
8.34
7.03
1.30

* δTMS scale (ppm) with tetramethylsilane (TMS) as internal reference.

figures

Figure 1.  Purification of BChE by ion exchange at pH 4.0

Figure 1. [go to text reference]

Ball-and-stick representation of the H5PVV2Mo10O40 polyoxometalate; center pink sphere represents the phosphorous (P) atom, the two black spheres the vanadium (V) atoms, the ten brown spheres the molybdenum (Mo) atoms, and the forty white spheres the oxygen (O) atoms.(For simplicity, hydrogens atoms are not shown)

Figure 2. SDS gel stained with Coomassie blue

Figure 2. [go to text reference]

13 C-NMR spectrum of the reaction betweenH5PVV2Mo10O40 and 2-chloroethyl ethyl sulfide (half-sulfur mustard, H-MG); reaction conditions: 3.0 μmol H5PVV2Mo10O40 and 30.0 μmol of the H-MG in 1.0 mL of deuterated water (D2O) at 37° C under aerobic conditions (O2 gas). The NMR acquisition parameters were: number of transients 74000 scans, acquisition time 1.29 s, first delay (d1) 1.00 s, pulse width (pw) 4.75 μsec and 90° pulse width (pw90) 9.50 μsec; the recording time was 15 h.

Figure 3. Affinity column purification of BChE

Figure 3. [go to text reference]

13 C-NMR spectrum of the reaction of H5PVV2Mo10O40 with bis-2-chloroethyl sulfide (sulfur mustard, HD); reaction conditions: 30.0 μmol of HD in deuterated water with H5PVV2Mo10O40 (120 μmol/ mL) at 37° C under aerobic conditions. The NMR acquisition parameters were: number of transients 8075 scans, acquisition time 3.8 s, first delay (d1) 2.0 s, pulse width (pw) 7.3 μsec and 90° pulse width (pw90) 14.60 μsec; recording time was 13:60:46 h.

Figure 4. HPLC purification of BChE

Figure 4. [go to text reference]

A) 31P-NMR spectrum of a solution of 3.0 μmol H5PVV2Mo10O40 and 30.0 μmol of HD in 1.0 mL of D2O at 37° C under aerobic conditions with resolution enhancement. The NMR acquisition parameters were: number of transients 512 scans, acquisition time 1.6 s, first delay (d1) 1.0 s, pulse width (pw) 10.65 μsec and 90° pulse width (pw90) 27.40 μsec; the recording time was 00:30:48 h. Inserts are 31P-NMR spectra windows at various concentrations of H5PVV2Mo10O40 after reaction with HD (4.4 x 10 -4 M): a) 5.0 mg/600 μL in D2O; b) 10.0 mg/600 μL in D2O, and c) 15.0 mg/600 μL in D2O at 37° C under aerobic conditions.

Figure 5. HPLC fractions on SDS gel stained with Coomassie blue

Figure 5. [go to text reference]

51 V-NMR spectrum of phosphovanadomolybdate polyoxometalate, H5PVV2Mo10O40 (35.0 mg/ 600 μL D2O); the 51V-NMR spectrum was referenced to pure vanadium (V) oxytrichloride (VOCl3, 99.99%). The NMR acquisition parameters were: number of transients 4096 scans, acquisition time 0.10 s, first delay (d1) 1.0 s and pulse width (pw) 9.2 μsec; the recording time was 00:37:51 h. The large linewidth of the signal at δ–535 ppm is indicative of an equilibrium mixture of α- and β- Keggin structural isomers.

Figure 6. Nondenaturing 4-30% gradient gel stained for BChE activity

Figure 6. [go to text reference]

Progressive 51V-NMR spectra of a solution of H5PVV2Mo10O40 (25.0 mg/600 μL D2O) with HD (4.40 x 10 -4 M) at 37°C as a function of time; the NMR spectra were recorded immediately after the reaction started and then every hour for 12 h. The NMR acquisition parameters were: number of transients 4096 scans, acquisition time 0.1 s, first delay (d1) 1.0 s and pulse width (pw) 9.2 μsec . An external reference VOCl3 was used.


Figure 7. [go to text reference]

Figure 7a is the first-derivative powder EPR spectrum of phosphovanadomolybdate, PVV2Mo10O40-5 (35.0 mg ), at 25o C. The graph in Figure 7b shows the decay of the anisotropic hyperfine splitting of the vanadyl transition metal in PVV2Mo10O40-5 (35.0 mg) for the reaction with 0.0 mg/mL, 1.7 mg/mL, 3.4 mg/mL, and 5.1 mg/mL of H-MG under aerobic condition for 20 min at 37o C; the EPR parameters are given in the Experimental Section.

Figure 8. Residence time of purified human BChE in mice

Figure 8. [go to text reference]

A) EPR spectrum of a solution polyoxometalate (35.0 mg) in 1.0 mL of dimethyl sulfoxide (DMSO) under aerobic conditions at 37° C; B) EPR spectrum of the reaction of PVV2Mo10O40-5 polyoxometalate with HD (4.5 μL of 9.35 mg/mL) in DMSO under aerobic conditions at 37° C . The EPR spectrum (A) of the polyoxometalate in DMSO shows a small contribution of the 51V+4 paramagnetic cation. The presence of HD (B) increases the quantity of 51V+4 cation by 1.6 fold. The eight hyperfine lines are due to specific transitions of 51V+4 cation (I = 7/2). The EPR parameters were: receiver gain 5 x 104, time constant 655.36 ms, sweep time 335.54 s, sweep width 300.0 mT and microwave power 50.2 mW.

references

Arroyo C.M. (1999) Encyclopedia of Spectroscopy and Spectrometry, Academic Press Ltd., London, UK, pp. 2189-2198.

Bolton J.R. (1972) Biological Applications of Electron Spin Resonance, Wiley Interscience, John Wiley & Sons, Inc., first ed., 52, New York, USA.

Gorenstein D.G. (1984) Phophorus-31 NMR, principles, and applications. Academic Press, Inc. Orlando, Florida, USA, pp. 14.

Haimov A., Neumann R. (2002) Polyethylene glycol as a non-ionic liquid solvent for polyoxometalate catalyzed aerobic oxidation. Chem. Commun. (Camb).8, 876-877.

Johnson RP., Hill CL. (1999) Polyoxometalate oxidation of chemical warfare agent simulants in fluorinated media. J. Appl. Toxicol. 19, S71-75.

Khenkin AM., Neumann R. (2000) Low-Temperature Activation of Dioxygen and Hydrocarbon Oxidation Catalyzed by a Phosphovanadomolybdate: Evidence for a Mars-van Krevelen Type Mechanism in a Homogeneous Liquid Phase This research was supported by the Basic Research Foundation administered by the Israeli Academy of Science and Humanities. Angew. Chem. Int. Ed. Engl. 39, 4088-4090.

Khenkin AM., Weiner L., Wang Y., Neumann R. (2001) Electron and oxygen transfer in polyoxometalate, H(5)PV(2)Mo(10)O(40), catalyzed oxidation of aromatic and alkyl aromatic compounds: evidence for aerobic Mars-van Krevelen-type reactions in the liquid homogeneous phase. J. Am. Chem.Soc.123, 8531-8542.

Kozhevnikov I.V. (1998) Catalysis by Heteropoly Acids and Multicomponent Polyoxometalates in Liquid-Phase Reactions. Chem. Rev. 98, 171-198.

McCreery M.J. (1997) Topical skin protectants. US Patent No. 5,607,979.

Medema J. (1986) Mustard gas: The science of H. Nuclear Biology Chemical Defense Technol. Int. 1, 66-77.

Muller A., Peters F., Pope MT. (1998) Gatteschi D. Polyoxometalates: Very Large Clusters-Nanoscale Magnets. Chem. Rev. 98, 239-272.

Neumann R, Dahan M. (1997) Molecular oxygen activation: a ruthenium substituted polyoxometalate as an inorganic dioxygenase catalyst. Nature388, 353-355.

Neumann R, Dahan M. (1998) Molecular oxygen activation by a ruthenium substituted "sandwich" type polyoxometalate. J. Am. Chem. Soc.120, 11969-11976.

Petterson L, Anderson J, Selling, Grate J.H. (1994) Multicomponent Polyanions.46. Characterization of the isomeric Keggin decamolybdodivanadophosphate ions in aqueous solution by 31P and 51V NMR. Inor. Chem. 33, 982-993.

Pope M.T. (1983) Heteropoly and Isopoly Oxometalates. Springer Verlag, USA.

Sharer J.E., Duescher R.J., Elfarra A.A. (1991) Formation, stability, and rearrangements of the glutathione conjugates of butadiene monoxide: evidence for the formation of stable sulfurane intermediates. Chem. Res. Toxicol.4, 430-436.

Weinstock I.A. ( 1998) Homogeneous-Phase Electron-Transfer Reactions of Polyoxometalates. Chem. Rev.98,113-170.