Theoretical Study of Adsorption of Sarin and ... - ACS Publications

Sep 22, 2006 - Department of CiVil and EnVironmental Engineering, FAMU-FSU College of Engineering, Florida State. UniVersity, Tallahassee, Florida...
10 downloads 0 Views 323KB Size
Download PDF
J. Phys. Chem. B 2006, 110, 21175-21183

21175

Theoretical Study of Adsorption of Sarin and Soman on Tetrahedral Edge Clay Mineral Fragments A. Michalkova,† J. Martinez,‡ O. A. Zhikol,§ L. Gorb,†,| O. V. Shishkin,§ D. Leszczynska,⊥ and J. Leszczynski*,†,| Computational Center of Molecular Structure and Interactions, Department of Chemistry, Jackson State UniVersity, 1400 J. R. Lynch Street, P.O. Box 17910, Jackson, Mississippi 39217, Departamento de Quimica Fisica, Facultad de Quimica, QTC, Pontificia UniVersidad Catolica de Chile, Santiago, Chile, STC “Institute for Single Crystals”, National Academy of Sciences of Ukraine, 60 Lenina AVenue, 61001 KharkiV, Ukraine, U.S. Army Engineer Research and DeVelopment Center (ERDC), Vicksburg, Mississippi 39180, and Department of CiVil and EnVironmental Engineering, FAMU-FSU College of Engineering, Florida State UniVersity, Tallahassee, Florida ReceiVed: April 13, 2006; In Final Form: June 6, 2006

This study provides details of the structure and interactions of Sarin and Soman with edge tetrahedral fragments of clay minerals. The adsorption mechanism of Sarin and Soman on these mineral fragments containing the Si4+ and Al3+ central cations was investigated. The calculations were performed using the B3LYP and MP2 levels of theory in conjunction with the 6-31G(d) basis set. The studied systems were fully optimized. Optimized geometries, adsorption energies, and Gibbs free energies of Sarin and Soman adsorption complexes were computed. The number and strength of formed intermolecular interactions have been analyzed using the AIM theory. The charge of the systems and a termination of the mineral fragment are the main contributing factors on the formation of intermolecular interactions in the studied systems. In the neutral complexes, Sarin and Soman is physisorbed on these mineral fragments due to the formation of C-H‚‚‚O, and O-H‚‚‚O hydrogen bonds. The chemical bond is formed between a phosphorus atom of Sarin and Soman and an oxygen atom of the -2 charged clusters containing an Al3+ central cation and -1 charged complex containing a Si4+ central cation (chemisorption). Sarin and Soman interact mostly in the same way with the same terminated edge mineral fragments containing different central cations. However, the interaction energies of the complexes with an Al3+ central cation are larger than these values for the Si4+ complexes. The interaction enthalpies of all studied systems corrected for the basis set superposition error were found to be negative. However, on the basis of the Gibbs free energy values, only strongly interacting complexes containing a charged edge mineral fragment with an Al3+ central cation are stable at room temperature. We can conclude that Sarin and Soman will be adsorbed preferably on this type of edge mineral surfaces. Moreover, on the basis of the character of these edge surfaces, a tetrahedral edge mineral fragment can provide effective centers for the dissociation.

Introduction For many years, several chemical compounds have been developed for using as chemical weapons in military conflicts. Chemical warfare agents are poisonous vapors, aerosols, liquids, or solids that have toxic effects on people, animals, or plants.1 They can have an immediate effect (few seconds to few minutes) or a delayed effect (several hours to several days). In protection against chemical warfare agents, decontamination is an important unavoidable part. The aim of decontamination is to rapidly and effectively render harmless or remove poisonous substances both on personnel and equipment. One group of particularly toxic chemical weapons is classified as nerve agents because of their extreme toxicity based on the disruption of nerve impulses in humans.2 Nerve agents are readily adsorbed by inhalation, ingestion, and dermal contact. Nerve agents belong to a class of phosphorus-containing organic * To whom correspondence should be addressed. E-mail: [email protected]. Telephone: 001-601-9797824. Fax: 001-601-9797823. † Jackson State University. ‡ Pontificia Universidad Catolica de Chile. § National Academy of Science of Ukraine. | ERDC. ⊥ FAMU-FSU.

chemicals that inhibit the acetycholinesterase enzyme (Ach), one of the most important neurotransmitter in many organisms including humans.3 Sarin (GB, isopropyl methylphosphonofluoridate (C4H10FO2P))4 and Soman (GD, 3,3-dimethyl-2-butyl methylphosphonoflouridate (C7H16FO2P)) are substances belonging to this group of chemical weapons. They have an extreme volatility (22 000 mg/m3 for Sarin and 3900 mg/m3 for Soman at 20 °C) at ambient temperature. That makes them extremely dangerous. Sarin and Soman exist in several stereoisomers5 and conformers.6 It was found that small areas of terrain, e.g., first aid stations or gun sites, may be decontaminated by removal of the topsoil.7 One part of the soil is clay minerals (layered aluminosilicates that consist of a continuous sheet of corner sharing tetrahedra bound to parallel sheets of edge sharing metal octahedra). Because of their high surface areas, clay minerals are often used as adsorbents and catalysts. Basal surfaces of clay minerals are characterized exclusively by charge-saturated and extremely stable siloxane bonds. Defect structures arise when the polar covalent bonds between the oxygen atoms and central cations of the tetrahedral sheets are

10.1021/jp062306j CCC: $33.50 © 2006 American Chemical Society Published on Web 09/22/2006

21176 J. Phys. Chem. B, Vol. 110, No. 42, 2006 broken. Such broken surface fragments consist predominantly of the oxygen anions [for a review, see e.g, ref 8]. Partially coordinated oxygen anions can strongly bind protons, resulting in neutralization of the negative surface charges and the formation of surface hydroxyl groups. Presence of the edge OH groups is partially responsible for electronegativity of the clay fragments and their capacity to absorb cations. The sorption mechanism of inorganic cations and anions on the edge clay surfaces9 and binding of organic matter on the edges10 has been studied. It was found that edge-sorption mechanisms are related to the acid-base properties of the clay edges. The edge surfaces are characterized by broken bonds and a well-known tendency to form inner-sphere complexes with protons and other cations.11-14 Because of the edge structure, this type of mineral surface is much more reactive than a regular basal mineral surface. For instance, it was experimentally demonstrated that the edge structures of biotite are approximately 250 times more reactive than the basal surface.15 Therefore, acid hydrolysis of 2:1 phyllosilicates essentially proceeds exclusively at the edge surfaces through most of the reactions (ref 16 and references therein). Our previous work was devoted to an adsorption of Sarin and Soman study on the surface of dickite (1:1 dioctahedral clay mineral of the kaolinite group).17 An analysis of the topological characteristics of electron density distribution in the inner part of the dickite-Sarin and the dickite-Soman systems at the B3LYP/6-31G(d,p) level of theory reveals that the adsorption of Sarin and Soman on the surface of dickite occurs through the formation of multiple hydrogen bonds. These hydrogen bonds were found between an oxygen atom and the methyl groups of Sarin and Soman and the hydroxyl groups (the adsorption on the octahedral surface) and between the methyl groups of Sarin and Soman and the basal oxygen atoms (the adsorption on the tetrahedral surface). Except this study, only an ab initio investigation of the adsorption of the phosphate groups on silica hydroxyls18 and a molecular modeling study of the tributhyl phosphate complex of europium nitrate in the clay hectorite (a trioctahedral clay mineral of the smectite group) have been carried out.19 It was found that both the polarizing and electron density abstracting abilities of silanols cause electron density redistribution, giving rise to strengthening of the phosphoester P-O bonding, weakening of the phosphinyl PdO and ester C-O bonding in adsorbed phosphate moieties. Force field calculations and MD studies serve to validate the use of these methods for predicting the behavior of lanthanide and actinide complexes in clays.19 These theoretical studies are supplemented by several experimental works devoted to the adsorption of phosphate molecules such as glyphosate, dimethyl methylphosphonate, and tricresyl phosphate on the surface of clay minerals were published, especially on montmorillonite.20-28 Except this clay mineral, the adsorption of tricresyl phosphate isomers onto kaolin and alumina was also investigated.29 But it was found that this phosphate molecule is adsorbed much better on montmorillonite than on these two other types of surfaces. Chromatographic methods were used to study the degradation of warfare nerve agents in soil.30-33 For example, D’Agostino and co-workers used this method to determine Sarin, Soman, and their hydrolysis products in soil.33 To extend our study on the adsorption of nerve agents on clay minerals,17 the investigation of the interactions of Sarin and Soman with tetrahedral edge clay mineral surfaces has been carried out. The purpose of this theoretical study is to provide more insight into the usefulness of clay minerals as adsorbents.

Michalkova et al. TABLE 1: Abbreviations of Calculated Adsorption Systems Used in Text and Chemical Formula of the Mineral Fragment of the Model investigated system one tetrahedra with an Al central cation and terminated O atom: Sarin and Soman one tetrahedra with an Al central cation and terminated OH group: Sarin and Soman one tetrahedra with a Si central cation and terminated O atom: Sarin and Soman one tetrahedra with a Si central cation and terminated OH group: Sarin and Soman

abbreviation

composition of the mineral parta

AlO(t)

[AlO(OH)3]2-

AlOH(t)

[Al(OH)4]-

SiO(t)

[SiO(OH)3]-

SiOH(t)

Si(OH)4

a With inclusion of terminal hydrogen atoms and with exclusion of an organic molecule.

It could lead to a finding of the way that these compounds can be used for the decomposition of nerve agents. Quantum chemical calculations are used in order to obtain detailed information on the thermodynamic stability of the various structural alternatives of nerve agent-mineral fragment complexes and their relative stabilities. Computational Details The adsorption of Sarin and Soman on the tetrahedral edge surfaces of clay minerals has been studied at the B3LYP/631G(d)34-40 and MP2/6-31G(d)41,42 levels of theory using the Gaussian98 program package.43 The complexes of Sarin and Soman with clay mineral fragment were fully optimized. Models of tetrahedral mineral fragments have been simulated by representative cluster models of the metal tetrahedra containing the Si4+ or Al3+ central cations. These two cases are the main situations that can occur on edge clay mineral surfaces. A different termination of the mineral fragments interacting with small organic molecules was tested in our previous work,44 devoted to the adsorption of methyl tert-butyl ether on broken clay mineral surfaces. It was found that a different termination of a mineral cluster has a small effect on the structure and interactions of the studied systems. A variety of positions of Sarin and Soman with respect to the mineral fragment were examined to find the most stable orientation of these target molecules. Abbreviations of the calculated adsorption systems used in the text and the chemical formula of the mineral fragment of a model are presented in Table 1. For practical reasons we will use the notation “R” for the -CH(CH3)2 fragment of Sarin and for the -CH(CH3)-C(CH3)3 fragment of Soman. The structure of adsorbed complexes and their interaction energies have been calculated. The interaction energies have been corrected by applying the basis set superposition error (BSSE) method.45 The values of interaction enthalpy (∆H), and interaction Gibbs free energy (∆G) have been calculated at room temperature (T ) 298.15 K) using the B3LYP/6-31G(d) level of theory. The entropy S(T) values have been calculated using rigid rotor-harmonic oscillator-ideal gas approximation based on the vibrational frequencies and optimized structures of the studied systems obtained at the same level of theory. The vibrational frequencies for thermal contributions to entropy (Svib(T) at 298.15 K) and to enthalpy (∆Hvib(T)) were scaled using the recommended frequency scaling factors in the work of Scott and Radom46 (used Svib(T) and ∆Hvib(T) scaling factors are 0.9989 and 1.0015 for the B3LYP/6-31G(d) method).

Adsorption of Sarin and Soman on Tetrahedral Edge Clay

J. Phys. Chem. B, Vol. 110, No. 42, 2006 21177

Figure 1. Optimized geometry of Sarin in the Sarin-[AlO(OH)3]2- (a), Sarin-[Al(OH)4]- (b), Sarin-[SiO(OH)3]- (c), and Sarin-Si(OH)4 (d) systems obtained at the B3LYP/6-31G(d) level of theory.

An analysis of topological characteristics of electron density was performed following Bader’s atoms in molecules theory (AIM).47 This analysis can be used to characterize hydrogen bonding solely from the charge density. There have been formulated several effects occurring in the charge density, like, for example, the location of the so-called (3, -1) bond critical points (BCPs, (i.e., points where F is minimum along the bond path and maximum in the other two directions,) on the surface of the total charge density and an analysis of the electron density (F) and analysis of the Laplacian of electron density (∇2F) at the BCPs, which are indicative of hydrogen bonding. These effects can be viewed as necessary criteria to conclude that a hydrogen bond is present.48,49 On the basis of these criteria, we assume that the C-H‚‚‚O bond is formed if the value of electron density at the BCP (F) amounts to between 0.002 and 0.035 e/au3 and the value of the Laplacian of electron density ∇2(F) is in the range of 0.024-0.139 e/au.549 Recently, these criteria were supplemented by analysis of ellipticity value in bond critical points of hydrogen bonds and distances between the BCP and nearest (3, +1) ring critical point (RCP) in order to distinguish weak hydrogen bonds and strong electrostatic interactions.50 It was demonstrated that ellipticity at the BCP of true H bonds should be smaller than 0.22. This rule also requires that distance between BCP of hydrogen bond and nearest RCP should be longer than 0.62 Å.50 Results and Discussion Topological and Geometrical Characteristics of Intermolecular Interactions of Sarin. The optimized structures of Sarin

interacting with tetrahedral edge fragments containing the Al3+ or Si4+ central cations (the [AlO(OH)3]2-, [Al(OH)4]-, Si(OH)4, and [SiO(OH)3]- fragments are drawn in Figure 1). Table 2 presents geometrical and topological characteristics of the hydrogen bonds formed between the edge tetrahedral mineral fragments and Sarin. On the basis of geometrical parameters and characteristics of the electron density distribution in bond critical points obtained at the B3LYP/6-31G(d) level of theory one may conclude that the target molecule is adsorbed due to the formation of two types of hydrogen bonds (typical O-H‚‚‚O hydrogen bond, in which the SiOH and AlOH groups of the mineral fragments act as proton donors to the oxygen of Sarin and Soman, and weak C-H‚‚‚O hydrogen bond, in which the oxygen atoms of the mineral fragments act as proton acceptors to the C-H groups of Sarin and Soman), strong electrostatic C-H‚‚‚O interactions, and a P-O chemical bond with the mineral fragment. We expect that the Sarin-[AlO(OH)3]2- system will be the most stable because of the formation of a P-O chemical bond between the phosphorus atom of Sarin and terminating oxygen atom of the mineral fragment. The total number of formed H bonds amounts to four in the Sarin-[Al(OH)4]- and Sarin-[AlO(OH)3]2- systems. Among them are three C-H‚‚‚O and one O-H‚‚‚O patterns. More detailed analysis of the electron density characteristics in BCP for the C-H‚‚‚O interactions reveals that some of them cannot be considered as true hydrogen bonds. In particular, ellipticity of HB2 interaction in Sarin-AlO(t) system is extremely high and the distance of BCP with nearest ring critical point (RCP) is only 0.267 Å. This indicates a significant instability of electron

21178 J. Phys. Chem. B, Vol. 110, No. 42, 2006

Michalkova et al.

TABLE 2: Calculated B3LYP/6-31G(d) H‚‚‚Y and X‚‚‚Y Distances (Å) (in parentheses), XsH‚‚‚Y Angles (deg) and Characteristics of (3, -1) Bond Critical Points (BCP) G (e/au3), ∇2G (e/au5), Ellipticity (E), and Distance between BCP and the Nearest (3, +1) Ring Critical Point (rBCP-RCP, Å) of Hydrogen Bonds in Studied Systems with Sarin and Somana H‚‚‚Y (X‚‚‚Y) X-H‚‚‚Y

F

∇2F

2.480 (3.064) 2.255 (2.635) 2.637 (3.628) 1.971 (2.905)

112.2 97.7 150.4 157.6

HB1 2.657 (3.566) HB2 1.756 (2.731) HB3

140.2 169.9

Sarin-AlO(t) 0.012083 0.043829 0.020854 0.087435 0.008491 0.027476 0.027429 0.076355 Sarin-SiOH(t) 0.006965 0.024936 0.038008 0.125790

HB1 2.657 (3.645) HB2 1.991 (2.920) HB3 2.495 (3.140) HB4 HB5 HB1 1.754 (2.732) HB2 2.542 (3.523)

HB1 HB2 HB3 HB4

a



rBCP-RCP H‚‚‚Y (X‚‚‚Y) X-H‚‚‚Y

0.792 1.574 0.340 0.008

0.323 0.267 0.463 0.838

2.409 (3.329) 2.024 (3.105) 2.166 (3.256) 2.235 (3.114)

140.8 166.0 172.6 150.3

0.078 0.015

1.359 1.255

2.530 (3.112) 2.339 (2.706) 1.713 (2.664)

112.2 97.3 157.2

149.9 157.0 116.5

Soman-AlO(t) 0.008118 0.026492 0.360 0.026373 0.073557 0.008 0.011349 0.040440 0.469

0.450 0.828 0.390

2.130 (3.213) 2.576 (3.458) 2.598 (3.606) 2.399 (3.435) 2.505 (3.589)

167.0 137.1 153.0 157.0 169.1

170.8 148.8

Soman-SiOH(t) 0.037958 0.126096 0.014 0.008699 0.029517 0.017

1.243 1.268

2.185 (2.690) 1.707 (2.657)

105.3 156.7

F

∇2F

Sarin-AlOH(t) 0.012075 0.037782 0.026825 0.069950 0.020530 0.054646 0.014535 0.048067 Sarin-SiO(t) 0.010597 0.039680 0.017631 0.078899 0.045969 0.138223 Soman-AlOH(t) 0.021728 0.056982 0.007595 0.030253 0.008404 0.029851 0.012303 0.037309 0.010219 0.031159 Soman-SiO(t) 0.021682 0.081913 0.046655 0.140116



rBCP-RCP

0.034 0.049 0.027 0.049

1.231 1.443 1.022 0.952

0.836 6.496 0.012

0.316 0.068 0.879

0.049 0.165 0.329 0.091 0.064

1.043 0.606 0.688 0.913 1.244

0.494 0.013

0.464 0.879

True hydrogen bonds are listed in bold.

density distribution in this area. Therefore, according to the results of previous studies,50,51 this interaction should be considered as strong electrostatic interactions rather than a true hydrogen bond. The same conclusion should be made for HB1 and HB3 interactions in the same system. A comparison of the C-H‚‚‚O interactions in Sarin-AlO(t) and Sarin-AlOH(t) systems demonstrates (Table 2) that a presence of the terminal hydroxyl group results in a transformation of strong electrostatic interactions into true C-H‚‚‚O hydrogen bonds. This means that the electron density distribution in the Sarin-AlOH(t) system is significantly more stable despite of absence of the P-O chemical bond. Probably, this reflects two different cases of adsorption of Sarin (namely chemically adsorbed Sarin in the case of the Sarin-AlO(t) system and physically adsorbed in the Sarin-AlOH(t) system). The results of the AIM analysis show that edge tetrahedral fragments containing a Si4+ cation instead of an Al3+ cation form the same type of the intermolecular interactions with Sarin. On the other side, a different termination of the cluster models (differently charged mineral fragments) leads to a different number of C-H‚‚‚O hydrogen bonds (see Figure 1) and a different strength of intermolecular hydrogen bonds (see Table 2). On average, Sarin forms less strong C-H‚‚‚O hydrogen bonds (one such H bond is formed in the Sarin-Si(OH)4 system and two strong electrostatic C-H‚‚‚O interactions in the Sarin[SiO(OH)3]- system) and stronger O-H‚‚‚O bonds (one O-H‚‚‚O hydrogen bond in both systems) with the Si4+ containing fragments than with the fragments containing an Al3+ central cation. The H‚‚‚O distances of C-H‚‚‚O H bonds and strong electrostatic interactions in the systems with an Al3+ central cation are on average about 0.1-0.2 Å shorter than the hydrogen bonds formed with the Si(OH)4 and [SiO(OH)3]fragments. The stronger interactions in the [AlO(OH)3]2- and [Al(OH)4]- systems correspond with larger values of the electron density and the Laplacian of the electron density than in the systems with a Si4+ central cation (for example, in the case of the Sarin-Si(OH)4 system, the F and ∇2F values are two and more times smaller than in the Sarin-[Al(OH)4]system). Comparing all considered Sarin-tetrahedral edge mineral fragment complexes, the strongest O-H‚‚‚O hydrogen bond is formed in the Sarin-[SiO(OH)3]- system because of the most preferable orientation of Sarin toward the fragment. This O-H‚‚‚O hydrogen bond corresponds with F value about

0.05 e/au3 and ∇2F value about 0.14 e/au.5 In comparison of the complexes with a Si4+ central cation, Sarin is the best stabilized by the interacting with the [SiO(OH)3]- fragment, where a chemical bond is formed between a phosphorus atom of the target molecule and an oxygen atom of the mineral fragment. To finish this section, we would like to mention that comparison of interatomic distances in adsorbed (see Figure 1 where the P-O1, P-F, and P-O2 bond lengths are illustrated) and isolated Sarin (see the data in ref 17) reveals that, for weakly interacting systems (the Sarin-Si(OH)4, and Sarin-[Al(OH)3]complexes), the adsorption results in small changes (less than 0.01 Å) in these bond lengths. On the other side, for the Sarin[SiO(OH)3]- and Sarin-[AlO(OH)3]2- systems where the P-O covalent bond is created, we have found a significant enlargement of these bond lengths. The largest change was found for the Sarin-[AlO(OH)3]2- systems in which, for example, the P-F bond is enlarged about 0.17 Å. In these two strongly interacting systems, the bond length changes of Sarin are larger for the systems with an Al3+ central cation than that for the systems with a Si4+ central cation. Topological and Geometrical Characteristics of Intermolecular Interactions of Soman. Figure 2 displays the optimized structure of Soman interacting with the [Al(OH)4]- and [AlO(OH)3]2- edge tetrahedral fragments. Table 2 contains characteristics of the electron density distribution at BCP for intermolecular hydrogen bonds formed in the Soman-[Al(OH)4]- and Soman-[AlO(OH)3]2- complexes. According to the AIM results, systems containing an Al3+ cation and Soman differ in the number of formed C-H‚‚‚O hydrogen bonds compared to similar species containing Sarin. Namely, two C-H‚‚‚O strong electrostatic interactions in the Soman-[AlO(OH)3]2- complex and five true C-H‚‚‚O hydrogen bonds in the Soman-[Al(OH)4]- system were found according to criteria mentioned above (see computational details).49-51 As one can see, an AIM analysis reveals a difference in the formation of hydrogen bonds in the systems with Sarin and Soman and the [Al(OH)4]- fragment. Because of its orientation to this fragment, Soman does not create the O-H‚‚‚O hydrogen bond as Sarin does (Sarin is oriented toward the mineral fragment by an O1 atom). Therefore, hydrogen bonds in complexes with Sarin and Soman are characterized by modified geometrical and topological parameters. For the C-H‚‚‚O hydrogen bonds, the values

Adsorption of Sarin and Soman on Tetrahedral Edge Clay

J. Phys. Chem. B, Vol. 110, No. 42, 2006 21179

Figure 2. Optimized geometry of Soman in the Soman-[Al(OH)4]- (a), Soman-[AlO(OH)3]2- (b), Soman-[SiO(OH)3]- (c), and Soman-Si(OH)4 (d) systems obtained at the B3LYP/6-31G(d) level of theory.

of the electron density range between 0.008 and 0.022 e/au3 and the values of the Laplacian of the electron density amount from 0.03 to 0.06 e/au5 and the C‚‚‚O distances are typical for this type of hydrogen bond (between 3.1 and 3.6 Å). The O-H‚‚‚O hydrogen bonds in the studied systems are characterized by similar distances, as was found for Sarin adsorbed on edge mineral fragments and in previously mentioned theoretical work studying the interactions between broken mineral surfaces and small organic species.52 Despite steric reasons, Soman is oriented toward the mineral fragment in the same way as Sarin (except a small change in a Soman orientation of the Soman[Al(OH)4]- complex). The strongest interaction that possesses a chemical bond character is formed in the Soman-[AlO(OH)3]2- system between a phosphorus atom of Soman and a terminal oxygen atom of the mineral fragment. Similarly to the complexes with Sarin, in this system, the O-H‚‚‚O hydrogen bond formed between an O1 atom of Soman and the hydroxyl group of the mineral fragment and two C-H‚‚‚O strong electrostatic interactions (formed between the C-H groups of the target molecule (C1-H, C3-H, and C-H groups of the R

part and the oxygen atoms of the mineral fragment) provide additional stabilization to this complex. Figure 2 presents also the optimized structure of Soman interacting with the electroneutral Si(OH)4 fragments and the [SiO(OH)3]- fragments. According to the geometrical and topological characteristics presented in Table 2, Soman interacts in a similar way with the [SiO(OH)3]- fragment, as was found for the Sarin-[SiO(OH)3]- and Soman-[AlO(OH)3]2- systems. This means the formation of a chemical bond in this system and additional stabilization by O-H‚‚‚O hydrogen bond and one C-H‚‚‚O strong electrostatic interaction. The only difference with the Soman-[AlO(OH)3]2- complex is in the formation of two strong electrostatic C-H‚‚‚O interactions. In the Soman-Si(OH)4 system, a Soman orientation is different than in the Soman-[Al(OH)4]- system but the same as in the SarinSi(OH)4 complex (the formation of one O-H‚‚‚O and one C-H‚‚‚O hydrogen bond). This indicates that, in these systems, the charge plays a crucial role in the stabilization, while the size of an organic molecule affects insignificantly the properties of these complexes.

21180 J. Phys. Chem. B, Vol. 110, No. 42, 2006

Michalkova et al.

TABLE 3: Calculated MP2/6-31G(d) H‚‚‚Y and X‚‚‚Y Distances (Å) (in parentheses), XsH‚‚‚Y Angles (deg) and Characteristics of (3, -1) Bond Critical Points (BCP) G (e/au3), ∇2G (e/au5), Ellipticity (E), and Distance between BCP and the Nearest (3, +1) Ring Critical Point (rBCP-RCP, Å) of Hydrogen Bonds in Studied Systems with Sarin and Somana H‚‚‚Y (X‚‚‚Y) X-H‚‚‚Y HB1 HB2 HB3 HB4 HB5

2.663 (3.654) 2.252 (2.606) 2.376 (2.986) 1.989 (2.921)

HB1 2.658 (3.236) HB2 1.795 (2.758) HB3 2.748 (3.521) HB1 2.675 (3.663) HB2 1.999 (2.930) HB3 2.427 (3.074) HB4 HB5 HB6 HB1 HB2 HB3 HB4 a

1.778 (2.750) 2.634 (3.673) 2.570 (3.500) 2.533 (3.518)

150.8 96.1 113.7 157.2

F

∇ 2F

Sarin-AlO(t) 0.008111 0.027582 0.022240 0.095427 0.014916 0.05292 0.026170 0.075107

Sarin-SiOH(t) 0.007527 0.030050 0.034162 0.116192 0.006192 0.024050 Soman-AlO(t) 150.5 0.007898 0.027041 157.2 0.025645 0.073724 116.6 0.013267 0.047114 112.4 164.6 127.6

168.0 158.1 142.7 149.3

Soman-SiOH(t) 0.035297 0.12078 0.007587 0.02663 0.008349 0.02974 0.009195 0.03160



F

rBCP-RCP H‚‚‚Y (X‚‚‚Y) X-H‚‚‚Y

0.374 2.398 0.571 0.012

0.425 0.194 0.403 0.826

2.189 (3.095) 2.664 (3.284) 2.089 (3.139) 2.216 (3.300 2.464 (3.259)

154.8 115.4 159.1 171.1 128.7

0.236 0.012 0.224

0.549 1.360 0.695

2.441 (3.047) 1.741 (2.683)

113.7 155.7

0.390 0.009 0.480

0.418 0.819 0.395

2.154 (3.229) 2.525 (3.338) 2.892 (3.449) 2.458 (3.483) 2.395 (3.411) 2.472 (3.533)

165.9 130.3 111.5 155.5 153.7 162.7

0.013 0.134 0.049 0.067

1.147 0.625 1.141 1.005

2.174 (3.928) 1.725 (2.665)

102.7 154.8

∇2F



rBCP-RCP

0.035 1.143 0.049 0.028 0.036

0.947 0.375 1.311 0.994 0.870

0.01286 0.04685 0.598 0.04261 0.13114 0.013 Soman-AlOH(t) 0.020908 0.056904 0.060 0.008862 0.035408 0.238 0.005439 0.023514 28.81 0.011792 0.038569 0.229 0.012717 0.039663 0.101 0.010806 0.035027 0.067 Soman-SiO(t) 0.02325 0.08969 0.663 0.04418 0.13610 0.015

0.393 0.866

Sarin-AlOH(t) 0.015718 0.052791 0.006797 0.029035 0.023836 0.06449 0.018794 0.052497 0.011296 0.038229 Sarin-SiO(t)

0.990 0.560 0.034 0.758 0.842 1.302 0.441 0.873

True hydrogen bonds are listed in bold.

The AIM analysis performed at the MP2 level of theory for the studied systems with Sarin (see Table 3) result in the formation of one additional C-H‚‚‚O strong electrostatic interaction (5 and 3) in the Sarin-AlOH(t) and Sarin-SiOH(t) systems than was found using the B3LYP level (4, and 2 H bonds). These interactions are less strong than the C-H‚‚‚O H bonds found at the B3LYP level of theory. On the other hand, in the Sarin-SiO(t) system, the analysis at the B3LYP level reveals the existence of one additional C-H‚‚‚O strong electrostatic interaction, which cannot be considered as a true hydrogen bond in comparison with the MP2 results. A similar difference in the number and strength of the C-H‚‚‚O hydrogen bonds was found for the complexes with Soman (see Table 3). In the Soman-AlOH(t), and Soman-SiOH(t) systems according to the F and ∇2F values obtained using the MP2 level of theory, one additional C-H‚‚‚O strong electrostatic interaction (Soman-AlOH(t)) and two additional C-H‚‚‚O hydrogen bonds (Soman-SiOH(t)) are formed, in comparison with the B3LYP results. Therefore, we concluded that, at both levels, the same orientation and character of stronger interactions were found. However, the characterization of the C-H‚‚‚O hydrogen bonds formation is slightly sensitive to the used method. We would like to emphasize that the structure of adsorbed Soman was modified by the adsorption in the same way as was discussed in a previous section for the Sarin-containing systems. This means that there are no significant changes in geometrical parameters of adsorbed Soman due to weak intermolecular interactions with the edge mineral fragments (see Figure 2 for bond lengths of adsorbed Soman and ref 17 for geometrical parameters of isolated Soman). On the other hand, the formation of the P-O covalent bond in the GD-[SiO(OH)3]- and GD[AlO(OH)3]2- complexes results in large bond length changes of adsorbed Soman. The orientation and intermolecular interactions of Sarin and Soman with the basal mineral surfaces have been already analyzed.17 We would like to highlight that Sarin and Soman interact differently with edge mineral surfaces than with the regular basal surface of clay minerals. Sarin and Soman do not form any chemical bonds with the regular basal surface of dickite. The H bond interactions formed between Sarin, Soman,

TABLE 4: Calculated B3LYP/6-31G(d) and MP2/6-31G(d) Values of ∆E, ∆H, and ∆G (kcal/mol) for Interactions between Sarin or Soman and the Mineral Fragment (T ) 298.15 K) ∆E

∆H

∆G

system

B3LYP

MP2

B3LYP

MP2

B3LYP

MP2

Sarin-AlOH(t) Sarin-AlO(t) Sarin-SiOH(t) Sarin-SiO(t) Soman-AlOH(t) Soman-AlO(t) Soman-SiOH(t) Soman-SiO(t)

-10.7 -53.0 -7.6 -8.2 -11.6 -55.1 -7.2 -9.4

-9.8 -48.5 -7.7 -6.0 -9.8 -52.4 -7.4 -6.7

-8.9 -50.7 -5.9 -6.9 -10.1 -52.9 -5.6 -8.3

-8.0 -46.2 -6.0 -4.7 -8.3 -50.2 -5.8 -5.6

3.3 -38.9 4.0 6.0 0.3 -41.1 4.4 4.4

4.2 -34.4 3.9 8.2 2.1 -38.4 4.2 7.1

and the regular tetrahedral surface of dickite (mineral fragments contain seven tetrahedral rings) are less strong than the interactions with edge mineral fragments as one can see from the comparison of X‚‚‚Y distances (see Table 4 in ref 17). Also, the values of the electron density and the Laplacian of the electron density are much larger for the Sarin- and Soman-edge mineral fragment systems than for the adsorption complexes of Sarin and Soman on the regular tetrahedral surface of dickite. It confirms that the edge mineral surfaces are more reactive than the regular basal mineral surfaces in the interactions with small organic species. Thermodynamics of Sarin and Soman Intermolecular Interactions. The interaction energies corrected by the basis set superposition error are presented in Table 4. As follows from these data for the Sarin- and Soman-[AlO(OH)3]2- complexes, the target molecule is adsorbed much more strongly on the mineral fragment where the formation of the new P-O covalent bond is possible. The interaction energy values of these two systems are much larger than these values for the rest of the systems (about 50 kcal/mol). Such values usually correspond to a destructive adsorption of Sarin and Soman (adsorption that is accompanied by the decomposition of these nerve agents as was shown in several theoretical and experimental studies of nerve agents or their simulants on magnesium oxide).53-55 This indicates also that, in the case of presence of such adsorption sites, the Sarin and Soman molecules adsorbed on the

Adsorption of Sarin and Soman on Tetrahedral Edge Clay

J. Phys. Chem. B, Vol. 110, No. 42, 2006 21181

Figure 3. Graph of dependence between the interaction energy and P-F bond length in studied systems obtained at the B3LYP/6-31G(d) level of theory.

[AlO(OH)3]2- mineral fragment will be most probably destroyed during such a type of process. This hypothesis can be partially confirmed by the largest changes in bond lengths of the target molecule, which were discussed in two previous sections. Figure 3 illustrates a graph of dependence between the interaction energy and P-F bond length of the studied systems. The P-F bond length was found to be the weakest in the Sarin- and Soman-[AlO(OH)3]2- complexes. This means that, in these two systems, the P-F bond can be easily broken and such adsorption can lead to a decomposition of a nerve agent. An analogical way of decomposition of dimethyl methylphosphonate (DMMP, this species is often used as a simulant of the organophosphorus compounds such as Sarin and Soman because it possess a similar structure) adsorbed on aluminum oxide54,56-58 and on silicon oxide59 was predicted experimentally, where the methoxy group was removed from DMMP. The scheme of the adsorption and reaction of DMMP on the alumina surface54 shows at first the formation of a P-O chemical bond between phosphorus and the oxygen of the surface hydroxyl. In the next step, the reaction involves the loss of the methoxy group and the formation of a phosphonate species. Sarin and Soman are more stable in the Sarin-[SiO(OH)3]and Soman-[SiO(OH)3]- systems than in the neutral Sarinand Soman-Si(OH)4 complexes because of the P-O bond formation. However, these systems containing a Si4+ central cation have much lower interaction energy than systems containing an Al3+ central cation. Partially, it can be caused by a different charge of the mineral part. A twice negatively charged mineral fragment attracts Sarin and Soman more strongly than a single negatively charged fragment. We have performed additional calculations of the PO(OH)HF-[SiO(OH)3]-, -[AlO(OH)3]2- systems to explain this large difference in the interaction energies (see Figure 4). The AIM analysis reveals that, in the PO(OH)HF-[AlO(OH)3]2- system, two, while in the PO(OH)HF-[SiO(OH)3]- complex, only one O-H‚‚‚O hydrogen bond is formed (this means existence of repulsion between OH groups and mineral and the organic species in the PO(OH)HF-[SiO(OH)3]- complex). Moreover, the P-O bond in the PO(OH)HF-[AlO(OH)3]2- system (1.71 Å, F value is 0.14347 e/au3, and ∇2F value 0.43017 e/au5) was found to be much stronger than in the PO(OH)HF-[SiO(OH)3]- complex (1.97 Å, F value is 0.08769 e/au3, and ∇2F value 0.02582 e/au5). Another explanation involves a basic character of [SiO(OH)3]and [AlO(OH)3]2- fragments that can play a significant role in the intermolecular interactions with Sarin and Soman (they have an acidic character). From this point of view, the target molecule interactions with the edge [AlO(OH)3]2- mineral fragment will be more favorable than those with the [SiO(OH)3]1- fragment.

Figure 4. Optimized geometry of the PO(OH)HF-[AlO(OH)3]2- (a), and PO(OH)HF-[SiO(OH)3]- (b) systems obtained at the B3LYP/631G(d) level of theory.

TABLE 5: Calculated B3LYP/6-31G(d) Values of ∆Ezpe, ∆Etherm, T∆Strans, T∆Srot, T∆Svib, and T∆Stot (kcal/mol) for Interactions between Sarin or Soman and the Mineral Fragment (T ) 298.15 K) system Sarin-AlOH(t) Sarin-AlO(t) Sarin-SiOH(t) Sarin-SiO(t) Soman-AlOH(t) Soman-AlO(t) Soman-SiOH(t) Soman-SiO(t)

∆Ezpe ∆Etherm T∆Strans T∆Srot T∆Svib

T∆Stot

-11.34 -11.33 -11.34 -11.34 -11.42 -11.42 -11.43 -11.42

-12.14 -11.77 -9.93 -12.93 -10.46 -11.88 -10.04 -12.75

1.75 1.73 1.13 1.34 1.18 1.66 1.05 1.20

1.85 2.35 1.73 1.29 1.50 2.16 1.57 1.07

-4.18 -4.27 -4.05 -4.18 -7.09 -7.17 -6.94 -7.05

6.95 7.41 9.03 6.16 8.05 6.71 8.34 5.72

This effect seems to be significant for strong intermolecular interactions. On the other hand, for weakly interacting systems, this effect is much less significant. Soman interacts more weakly with the tetrahedral mineral fragment terminated by the OH group in the complex with a Si4+ central cation than does Sarin. The Soman-Si(OH)4 system is the weakest interacting from all the studied complexes found at the B3LYP/6-31G(d) level of theory. In all other cases, Soman is better stabilized on studied tetrahedral mineral surfaces than Sarin. To make our predictions relevant to experimental conditions, we have calculated the change in enthalpies (∆H) and Gibbs free energies (∆G) at room temperature. These values are listed in Table 4. As is expected, the ∆H values are slightly decreased in comparison with ∆E values (see Table 5, thermal contributions to ∆H amount to about 1-2 kcal/mol, which provides about a 10-20% decrease in the ∆H value for weakly interacting complexes). However, the values of ∆G are decreased significantly due to the influence of a compensation effect.60 The most interesting result is that the majority of the ∆G values are positive. This indicates that, at room temperature, the destabilizing entropy contribution for the studied systems is larger than the stabilizing contribution arising from the change in enthalpy. Therefore, we may expect that, at room temperature, due to the destabilizing entropy effect, only strongly interacting complexes of Sarin and Soman adsorbed at the considered adsorption sites are stable. Among them are the Sarin-[AlO(OH)3]2- and Soman-[AlO(OH)3]2- systems.

21182 J. Phys. Chem. B, Vol. 110, No. 42, 2006 TABLE 6: BSSE Corrected Interaction Energies (kcal/mol) of the D-Sarin and D-Soman Systems Calculated Using the ONIOM(B3LYP/6-31G(d,p):LL) Method ∆E

system LL method

PM3

HF/3-21G

D(t)-Sarin D7(t)-Sarin D(t)-Soman D7(t)-Soman

-3.9 -7.4 -4.0 -9.1

-4.9 -5.7

As was mentioned before, similar investigation of the interactions of Sarin and Soman with the basal surface of clay minerals has been performed.17 Therefore, it would be interesting to investigate what conclusion concerning thermodynamics of this type of mineral surface could be made. To address this issue, an estimation of the ∆G values for the adsorption of Sarin and Soman on the regular basal surface of clay mineral models containing one or seven rings of dickite was performed. Table 6 contains the values of interactions energies that we have obtained previously.17 Most of the interaction energy values in Table 6 (characterizing adsorption on regular mineral surfaces) do not exceed the ∆E values in Table 4 (adsorption on the edge mineral surfaces). Moreover, we can assume that the T∆S values are in the same range as was found in this work. On the basis of these facts, one can predict that the ∆G values for these regular basal mineral surfaces will be rather positive. This means that, at room temperature, Sarin and Soman will be preferably adsorbed on charged edge mineral surfaces rather than on regular electroneutral mineral surfaces. To better understand such significant enthalpy-entropy compensation, we analyze the entropy terms more carefully. As is known, the entropy terms are divided into translational, rotational, and vibrational contributions. The corresponding components of T∆Stot are listed in Table 5 (the entropy terms presented in this Table are not scaled by frequency scaling factors, but scaling factor was applied in the calculations of ∆H and ∆G values presented in Table 4). The adsorption complexes in the gas phase are characterized by large negative T∆Stot values. The data collected in Table 5 show that the largest contribution to the T∆Stot values is provided by the T∆Strans terms, of which the values are all negative and approach the values of T∆Stot. Such significant negative contribution is not surprising because it reflects the loss of three translational degrees of freedom during the adsorption process. Another negative contribution is associated with the loss of some of Sarin and Soman rotational degrees of freedom. Finally, the positive contribution (smaller than translational but larger than rotational) is related to an increase of the number of vibrations when the complex is formed during the adsorption. This is a general phenomenon for all associative reactions of the A + B f C type.61 The largest T∆Stot values were found for the GB- and GD-[SiO(OH)3]- systems and the smallest values were predicted for the GB- and GD-[Si(OH)4] complexes. The T∆Strans values for individual studied systems with a different target molecule are very similar (for instance, these values for the GBand GD-[AlO(OH)]2- strongest interacting systems differ only about 0.1 kcal/mol). On the other hand, the rotational entropic contributions of the adsorption complexes with Sarin are smaller (about 3 kcal/mol) than the T∆Srot values for the adsorption complexes containing Soman. The vibrational contributions are mostly larger for the complexes containing Sarin than Soman, but this difference is less significant than for the T∆Srot values. In summary, previously discussed results obtained at both B3LYP and MP2 levels of theory suggest that Sarin and Soman will be preferably adsorbed on charged mineral edges. Therefore,

Michalkova et al. such edge mineral fragments can be considered as the strong centers for Sarin and Soman interaction with clay minerals. This work leads to a conclusion that the adsorption of nerve agents on clays is a complex phenomenon that depends not only on the type of the mineral surface but also on the type of the central cation, charge, and defect of the mineral fragment. According to the general information on chemical nerve agent spillage disposal, the puddles of liquid must be contained by covering with clay.62,63 Therefore, these theoretical predictions can potentially add to the overall understanding of surface chemical properties of clay minerals and their relation to the nerve agent contaminants. It can also help to better understand the decontamination process and to develop more efficient ways of spillage disposal. Conclusions The adsorption of Sarin (GB) and Soman (GD) on the edge tetrahedral fragments of clay minerals have been studied at the B3LYP/6-31G(d) and MP2/6-31G(d) levels of theory. Several different orientations of Sarin and Soman on tetrahedral edge mineral surfaces were found to depend on the charge and a mineral cluster termination. Sarin and Soman can be physisorbed or chemisorbed on small edge tetrahedral mineral fragments. The chemisorption occurs due to the formation of a P-O chemical bond created between a phosphorus atom of the target molecule and a terminal oxygen atom of the mineral cluster. In the case of the physisorbed target molecule, the formation of typical O-H‚‚‚O created between an oxygen atom of the target molecule and the OH groups of the mineral fragment is predicted. Moreover, the C-H‚‚‚O interactions, which can be classified as strong electrostatic interactions or true weak hydrogen bonds created between the C-H groups of Sarin and Soman and oxygen of the mineral fragment, provide the additional stabilization of the studied complexes. Topological characteristics of electron density distribution at BCP for all these interactions may be used for a further classification of other intermolecular interactions in adsorption complexes of Sarin and Soman on edge mineral fragments. The strongest adsorption of Sarin and Soman was found in the case of a (-2) charged [AlO(OH)3]2- mineral fragment (the interaction energy that amounts to 49 and 52 kcal/mol at the MP2/6-31G(d) level of theory), where the P-O covalent bond is created. This type of the adsorption is accompanied by a significant enlargement of distances of phosphorus-containing bonds in a nerve agent (so these bonds become much weaker). It is caused by a participation of atoms connecting with phosphorus in the interactions with the mineral fragments. On the basis of the results of this study and also according to other theoretical and experimental investigations,51-53 we conclude that such adsorption can possibly lead to a decomposition of Sarin and Soman. The values of the interaction energies of Sarin and Soman physisorbed systems on edge tetrahedral fragments are relatively low. ∆G values for weakly interacting systems (with weakly negative formation enthalpies) were found to be positive. On the basis of these results, we concluded that only strongly bound complexes of Sarin and Soman interacting with charged edge tetrahedral mineral fragments (the Sarin- and Soman-[AlO(OH)3]2- complexes) are stable at room temperature. It was found that, in the case of Sarin and Soman, adsorption on charged edge tetrahedral mineral surfaces is more favorable than an adsorption on regular electroneutral mineral surfaces. Acknowledgment. This work was facilitated by support from the Office of Naval Research grant no. N00034-03-1-0116,

Adsorption of Sarin and Soman on Tetrahedral Edge Clay the Army High Performance Computing Research Center under the auspices of the Department of the Army grant no. DAAD1901-2-0014, and the Army Research Laboratory Cooperative agreement no. DAAH04-95-2-0003/contract no. DAAH04-95C-0008, the content of which does not necessarily reflect the position or policy of the government, and no official endorsement should be inferred. This work was also facilitated by use of the Network Visualization System for Computational Chemistry (NVSCC) (http://www.ccmsi.us/nvscc). References and Notes (1) http://sis.nlm.nih.gov/Tox/ChemWar.html. (2) http://faculty.washington.edu/chudler/weap.html. (3) http://www.rand.org/publications/MR/MR1018.5/ MR1018.5.chap5.html. (4) http://cfrterrorism.org/weapons/sarin.html. (5) Walker, A. R. H.; Suenram, R. D.; Samuels, A.; Jensen, J.; Ellzy, M. W.; Lonchner, J. M.; Zeroka, D. J. Mol. Spectrosc. 2001, 297, 77. (6) Kaczmarek, A.; Gorb, L.; Sadlej, A. J.; Leszczynski, J. Struct. Chem. 2003, 15, 517. (7) http://www.ohio.vermeer.com/vcom/TerrainLeveler/Index.jsp?NewsID)12980. (8) Ward, D. B.; Brady, P. B. Clays Clay Miner. 1998, 46, 453. (9) Sposito, G. Surface Chemistry of Soils; Oxford University Press: New York, 1984; p 223. (10) Claret, F.; Bauer, A.; Scha¨fer, T.; Griffault, L. and Lanson, B. Clays Clay Miner. 2002, 50, 632. (11) White G. N.; Zelazny, L. W. Clays Clay Miner. 1989, 36, 141. (12) Zachara, J. M.; McKinley, J. P. Aquat. Sci. 1993, 55, 250. (13) Schlegel, M. L.; Maceau, A.; Chateigner, D.; Charlet, L. J. Colloid Interface Sci. 1999, 215, 140. (14) Charlet, L.; Schindler, P. W.; Spadini, L.; Furrer, G.; Zysset, M. Aquat. Sci. 1993, 55, 291. (15) Turpault, M. P.; Trotignon, L. Geochim. Cosmochim. Acta 1994, 58, 2761. (16) Nagy, K. L. Dissolution and Precipitation Kinetics of Sheet Silicates. In Chemical Weathering Rates of Silicate Minerals; White, A. F., Brantley, S. L., Eds.; Reviews in Mineralogy 31; Mineralogical Society of America: Washington, D. C., 1995; pp 173-233. (17) Michalkova, A.; Gorb, L.; Ilchenko, M.; Zhikol, O. A.; Shishkin, O. V.; Leszczynski, J. J. Phys. Chem. B 2004, 108, 1918. (18) Murashov, V. V.; Leszczynski, J. J. Phys. Chem. A 1999, 103, 1228. (19) Hartzell, C. J.; Cygan, R. T.; Nagy, K. L. J. Phys. Chem. A 1998, 102, 6722. (20) Bowen, J. M.; Powers, C. R.; Ratcliffe, A. E.; Rockley, M. G.; Hounslow, A. W. EnViron. Sci. Technol. 1988, 22, 1178. (21) Bowen, J. M.; Compton, S. V.; Blanche, M. S. Anal. Chem. 1989, 61, 2047. (22) Levesque, M.; Schnitzer, M. Soil Sci. 1967, 103, 183. (23) Hance, R. J. Pest. Sci. 1976, 23, 363. (24) Shoval, S.; Yariv, S. Clays Clay Miner. 1979, 27, 19. (25) Sprankle, P.; Meggitt, W. F.; Penner, D. Weed Sci. 1975, 23, 229. (26) Morillo, E.; Undabeytia, T.; Maqueda, C. EnViron. Sci. Technol. 1997, 31, 3588. (27) McConnell, J. S.; Hossner, L. R. J. Agric. Food Chem. 1985, 33, 1075. (28) Glass, R. L. J. Agric. Food Chem. 1987, 35, 497. (29) Takimoto, K.; Ito, K.; Mukai, T.; Okada, M. EnViron. Sci. Technol. 1998, 32, 3907. (30) Kataoka, M.; Tsunoda, N.; Ohta, H.; Tsuge, K.; Takesako, H.; Seto, Y. J. Chromatogr., A 1998, 824, 211.

J. Phys. Chem. B, Vol. 110, No. 42, 2006 21183 (31) Vermilion, W. D.; Crenshaw, M. D. J. Chromatogr., A 1997, 770, 253. (32) Kataoka, M.; Tsuge, K.; Takesako, H.; Hamazaki, T.; Seto, Y. EnViron. Sci. Technol. 2001, 35, 1823. (33) D’Agostino, P.; Hancock, J. R.; Provost L. R. J. Chromatogr., A 2001, 912, 291. (34) Becke, D. J. Chem. Phys. 1993, 98, 5648. (35) Yang, Lee W.; Parr, R. G. Phys. ReV. B 1998, 37, 785. (36) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724. (37) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257. (38) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209. (39) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163. (40) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213. (41) Head-Gordon, M.; Pople, J. A.; Frisch, M. A. Chem. Phys. Lett. 1988, 153, 503. (42) Frisch, M. J.; Head-Gordon, M.; Pople, J. A. Chem. Phys. Lett. 1990, 166, 275. (43) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998. (44) Michalkova, A.; Johnson, L. D.; Gorb, L.; Zhikol, O. A.; Shishkin, O. V.; Leszczynski, J. Int. J. Quantum Chem. 2005, 105, 325. (45) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553. (46) Scott, A. P.; Radom, S. L. J. Phys. Chem. 1996, 100, 16502. (47) Bader, R. W. F. Atoms in Molecules: A Quantum Theory; Oxford University Press: Oxford, 1990. (48) Koch, U.; Popelier, P. L. A. J. Phys. Chem. 1995, 99, 9747. (49) Popelier, P. A. L. J. Phys. Chem. A 1998, 102, 1873. (50) Shishkin, O. V.; Palamarchuk, G. V.; Gorb, L.; Leszczynski, J. J. Phys. Chem. B 2006, 110, 4413. (51) Hocquet, A. Phys. Chem. Chem. Phys. 2001, 3, 3192 (52) Aquino, A. J. A.; Tunega, D.; Haberhauer, G.; Gerzabek, M. H.; Lischka, H. J. Comput. Chem. 2003, 24, 1853. (53) Michalkova, A.; Gorb, L.; Ilchenko, M.; Zhikol, O. A.; Shishkin, O. V.; Leszczynski, J. J. Phys. Chem. B 2004, 108, 5294. (54) Mitchell, M. B.; Seinker, V. N.; Mintz, E. A. J. Phys. Chem. B 1997, 101, 11192. (55) Wagner, G. W.; Bratram, P. W.; Koper, O.; Klabunde, K. J. J. Phys. Chem. B 1999, 103, 3225. (56) Templeton, M. K.; Weinberg, W. H. J. Am. Chem. Soc. 1985, 107, 774. (57) Templeton, M. K.; Weinberg, W. H. J. Am. Chem. Soc. 1985, 107, 97. (58) Wagner, G. W.; Procell, L. R.; O’Connor, R. J.; Munavalli, S.; Carnes, C. L.; Kapoor, P. N.; Klabunde, K. J. J. Am. Chem. Soc. 2001, 123, 1636. (59) Henderson, M. A.; Jin, T.; White, J. M. J. Phys. Chem. 1986, 90, 4607. (60) Dunitz, J. D. Chem. Biol. 1995, 2, 709. (61) Searle, M. S.; Williams, D. H. J. Am. Chem. Soc. 1992, 114, 10690. (62) monitoring.http://www.bt.cdc.gov/agent/tabun/ctc0002.asp. (63) http://www.asod.org/chemical.htm.