Quantum chemical study of thiosulfinic acids and their anions

Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

Contents lists available at ScienceDirect

Journal of Molecular Structure: THEOCHEM journal homepage: www.elsevier.com/locate/theochem

Quantum chemical study of thiosulfinic acids and their anions Marian Mikołajczyk *, Marek Cypryk, Grzegorz Krasin´ski Center of Molecular and Macromolecular Studies, Polish Academy of Sciences, 90-363 Łódz´, Sienkiewicza Str. 112, Poland

a r t i c l e

i n f o

a b s t r a c t

Article history: Received 24 April 2008 Received in revised form 21 May 2008 Accepted 24 May 2008 Available online 9 July 2008 Keywords: Thiosulfinic acids Thiosulfinate anions Tautomerism Mesomerism DFT

The electronic and geometrical structures of thiosulfinic acids, RSSOH, 1, (1a, R = H; 1b, R = CH3; 1c, R = tC4H9; 1d, R = C6H5; 1e, R = F) and their anions have been investigated by the ab initio and density functional methods. The calculations show that the stability of the thiolo-tautomers of 1, RS(O)SH, and the thiono-tautomers, RS(S)OH, is almost the same in the gas phase, the energy of the thiolo-tautomers being slightly lower. The tautomers of 1 having the thiosulfone structure, RS(O)(S)H, were found to be the least stable. The thiosulfinate anions show ambident character with the negative charge dispersed over terminal O and S atoms and are stabilized by electronegative substituents. Ó 2008 Elsevier B.V. All rights reserved.

Thiosulfinic acids, RSSOH (1) represent a practically unexplored class of organic sulfur compounds which are interesting from both stereochemical and theoretical points of view [1]. They may be formally derived from the well-known sulfinic acids, RSO2H, by replacement of one of the two oxygen atoms by sulfur. However, in contrast to sulfinic acids, which are effectively achiral due to a fast proton exchange between two enantiomeric forms via the achiral sulfinic acid anion (Eq. (1)) [2], thiosulfinic acids 1 as well as their anions are chiral.

OH

O

S

S

R (R)

O

OH

O

1. Introduction

R

O

S R

SH

S R

H

R

O H

S R

O

S

S R

S

H S S R

-H

O S

O

O S

S R

S ð2Þ

OH

(S) ð1Þ

Theoretically, thiosulfinic acids 1 can exist in three tautomeric structures: thiolo-form 10 , thiono-form 100 and thiosulfone-form 1000 (Scheme 1). Accordingly, the thiosulfinate anion can be described by three mesomeric forms in which the negative charge is located on the terminal sulfur atom, oxygen atom and central sulfur atom, respectively (Eq. (2)). Investigation of the tautomerism of thiosulfinic acids and reactivity of the ‘tridentate’ anion derived from them would be of great interest.

* Corresponding author. Tel.: +48 42 681 5832; fax: +48 42 680 3260. E-mail address: [email protected] (M. Mikołajczyk). 0166-1280/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.theochem.2008.05.024

Esters of thiosulfinic acids are known to be relatively stable [1]. Isolation of the stable metal complexes with thiosulfinate ligands has also been reported, see for review [3]. Some time ago we prepared relatively stable salts of organic thiosulfinic acids RSSOH, R = t-Bu, Adamantyl, and Triptycenyl [4]. The single-crystal X-ray analysis of S-benzylthiuronium salts of adamantanethiosulfinic acid 1f allowed us to determine structural parameters of the corresponding thiosulfinic acid anions [4]. The stability of these salts may be attributed to steric protection by bulky groups R bonded to the central sulfur atom. However, we found that both esters and the free acids generated from their salts are very unstable. They are very sensitive to nucleophilic attack and cannot survive in the presence of anionic species [5]. Methylation of t-butylthiosulfinate salt gives the mixture of products indicating that a complex system of consecutive reactions takes place in solution [5]. Similarly, the acids undergo fast elimination of elemental sulfur and subsequent conversion of the transiently formed sulfenic acids

106

M. Mikołajczyk et al. / Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

S

O R

S

R

S H

S

1'

possible isomers (Scheme 2) have been calculated to compare their relative stabilities, which may be important for the prediction of possible rearrangement pathways.

S O H

R

S

H

O 1'''

1''

2. Theoretical methods

Scheme 1.

into the corresponding thiosulfinates, RS(O)SR [4]. This fact precluded experimental studies of the tautomerism in this class of thioacids. Therefore, we directed our attention to quantum chemical studies of thiosulfinic acids RSSOH 1a–e, where R = H (a), Me (b), t–Bu, (c), Ph (d) and F (e). First theoretical ab initio calculations using MP2/6-31G* treatment have been performed by Basch [6] for only two hypothetical tautomeric forms of the parent thiosulfinic acid (R = H), i.e. HS(O)SH, 1a0 and HS(S)OH, 1a00 (Eq. (3)).

O

S

HSSH

HSOH

1a'

1a"

ð3Þ

More detailed theoretical studies of five the most stable structural isomers (1a–3a) out of nine possible for the parent hydrogen-substituted species (Scheme 2) were performed at the MP2// 6-311G(d,p) level by Steudel et al. [7]. Methyl-substituted analogues were analyzed theoretically by the same method as model species in the study of the mechanism of flash vacuum pyrolysis of t-butylthiosulfinic acid S-t-butyl ester, t-BuS(O)SBu-t [8]. Thiosulfinate anions have not been studied by quantum chemical methods except the simplest one, HS(S)O, which was observed as the transient species formed during mass spectrometric neutralization–reionization experiments. Its energetics, structural features and fragmentation pathways were computationally investigated by advanced ab initio calculations at the B3LYP/augcc-pVTZ + d, CCSD(T)/aug-cc-pVTZ + d, and CBS-Q levels of theory [9]. In the present work, the electronic structures of thiosulfinic acids 1a–e and of the corresponding anions have been investigated by the density functional methods. The main objective of this study was to compare the steric and electronic effects of the substituent R on the stability of thiosulfinic acids. The stability of these acids in aqueous solution was also examined by SCRF calculations. In the case of parent hydrogen-substituted species the energies of all nine

O H

S

S

S

H

H

S

S O H

H

S

H

O 1a' H O

S

1a'' S

H

H

1a'''

S

O

2a

H

S

H

S

S

S

O

S

S

H O H

H O H

H

H

S

H

S 4a

3. Results and discussion 3.1. Parent thiosulfinic acid, H2S2O and its isomers (1a–7a) Geometries of the isomers 1a–3a including rotamers were studied in detail previously [7,8]. Our calculations at the B3LYP/6311 + G(2d,p)//B3LYP/6-31G(d) level (referred to as DFT further in text) confirm these predictions. Thus, we will not discuss the structural features of these species in detail. The most stable conformers are ()syn-clinal for 1a’, anti-periplanar for 1a’’, and transin the case of 2a and 3a, in accord with previous findings [8]. The B3LYP/6-31G(d) bond distances are slightly longer (by ca. 0.02 Å) than those calculated by the MP2/6-311G(d,p) method, partly due to the shortening of the 6-31G(d) basis set. Thus, we found the SAS bond length of 2.200 Å in 1a0 while the S@S bond lengths are 1.976 and 1.942 Å in 1a00 and 1a000 , respectively. The S@O bond lengths are 1.493 and 1.475 Å in 1a0 and 1a000 , respectively, and the SAO bond length in 1a00 is 1.689 Å. The geometries of the other isomers (4a–7a) require a brief discussion, since these structures have not been studied before. The equilibrium structures of the isomers 4a–7a together with the key bond lengths and Wiberg bond orders (in parentheses) are shown in Scheme 3. Optimization of 4a leads to essential loosening of the SAS bond and results in the complex of H2S with S@O. The SAS bond order is only 0.28 while the SAO

3a

O S

Geometry optimizations of intermediates, transition states, and products were carried out at the B3LYP/6-31G* and B3LYP/631 + G* (for anions and species in solution) level of theory [10– 12] using the Gaussian 03 suite of quantum chemical programs [13]. All stationary points on the respective potential energy surfaces were characterized at the same level of theory by evaluating corresponding Hessian indices. Careful verification of the unique imaginary frequencies for transition states has been carried out to check whether the frequency indeed pertains to the desired reaction coordinate. Further, intrinsic reaction coordinate (IRC) calculations were carried out to authenticate all transition states [14,15]. Single point energies were then calculated using a more flexible triple-f quality basis set, 6-311 + G(2d,p). Energies of the simplest model compounds were also calculated using the G3/ B3LYP (denoted as G3B3) theory [16]. Atomic charges and Wiberg bond orders were computed according to Natural Bond Orbital theory using the Gaussian built-in option [17] and the standalone GenNBO program [18] for the DFT density. Solvation energies in water were calculated using a continuum solvation model and the SCRF-PCM method [19–23] as implemented in Gaussian 03. UAHF atomic radii have been used for cavity definition. This energy in solution (Gsolv) comprises the electronic energy of the polarized solute, the electrostatic solute–solvent interaction energy, and the non-electrostatic terms corresponding to cavitation, dispersion, and short-range repulsions [19–23].

5a

6a Scheme 2.

7a

O 2.726

H2S (0.28) S

1.524 (1.66)

2.032

1.590

H2S (1.00)O 1.866 (070)

H2O

2.522 S 1.942 (0.13) (1.91)

S 4a

S

S

5a

6a

S

H2 O 7a

0

Scheme 3. B3LYP/6-31G(d) bond lengths (Å A) and Wiberg bond orders (in parentheses) for the singlet state species 4a–7a.

107

M. Mikołajczyk et al. / Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

bond order equals 1.66, which suggests a significant double bond character. Similarly, the equilibrium geometry of 6a should be interpreted as the H2OAS2 complex. The 7a structure is the complex of two sulfur atoms with H2O having biradical electron configuration (Scheme 3). It is the highest in energy among all species studied here. In general, the singlet states of the isomers 4a, 5a and 7a are higher in energy than their triplet states. This suggests that they are very unstable and may easily undergo fragmentation via crossing the triplet surface with formation of radical species. For energy calculations, we used also the G3/B3LYP method as a reference because of its demonstrated thermochemical accuracy [16]. The agreement between DFT and G3/B3LYP results is good, which allows us to conclude that the relatively inexpensive B3LYP/6-311 + G(2d,p)//B3LYP/6-31G(d) (DFT) method is sufficiently accurate and can be used for the analysis of larger structures. Table 1 summarizes the calculated enthalpies for nine H2S2O isomers examined, 1a–7a, both in the gas phase and in aqueous solution. Only the most stable conformers are considered, i.e., ()syn-clinal for 1a0 , anti-periplanar for 1a00 , and trans- for 2a and 3a, but the energy differences between the rotational isomers are as small as 1.5 kcal/mol or less. The energy differences obtained at the MP2/6-311G(d,p) level are systematically higher by 4– 7 kcal/mol, compared to those calculated by the DFT method [7,8].

Table 2 Relative enthalpies at 298 K for thiosulfinic acid isomers RS2OH, 1a–e, in the gas phase, in kcal/mol R = Ha

Isomer 0

1 10 0 10 0 0 3 0d 1 3 00d 1 2 3 a b c d

0 1.9 25.9 32.2 34.8 16.2 14.0

R = tBua

R = Me a

0 1.1 22.1 34.9 36.5 9.8 17.8

(1.3) (20.2)

(14.9) (14.1)

c

0 0.4

0 0.2 19.7

0 3.2 23.9

0 2.2 23.0

14.7

10.6 18.0

1.8 30.9

10.6 16.3

-0.874

-0.483

O

S

1.496 (1.33) 1.840 (0.92) -0.113

Me

1.980(1. 35) 1.831 (0.93)

1.131

S

-0.297

2.214 (0.81)

0.153

S 1.351 H

-0.080

Me

(0.96)

1b'

Isomer

Gas phase

Water

DH298 DFT 1a0 (anti) 1a0 0 (syn) 1a0 0 0 2a 3a 4a 5a 6a 7a a b c d

b

0 1.9 25.9 16.2 14.0 23.7 56.3 1.9 103.9

DE0 + ZPEa G3B3

MP2

0 1.3 20.2 14.9 14.1 24.9 55.5 1.6 104.7

0 0.4 27.6 22.8

l (D)

DH298

DGsolvd

c

DFT1 2.542 2.327 3.318 0.814 1.187 3.813 5.399 2.530 3.517

MP2/6-311G(d,p) [6,7]. DFT = B3LYP/6-311 + G(2d,p)//B3LYP/6-31G(d). DFT1 = B3LYP/6-311 + G(2d,p)//B3LYP/6-31 + G(d). DGsolv calculated by SCRF/DFT1.

0 1.0 23.6 19.1 14.8 19.3 51.2 3.3 85.9

1.1 2.4 6.5 7.1 2.7 5.6 11.0 7.9 21.5

0.949 -0.874 0.488 1.691 0.980 (0.76) (0.73)

S

O

-0.378

-0.839

S 1.926 (1.58)

1.467(1.45) 1.649 (0.65) -0.493

F

1.659 (0.64)

1.421

S

-0.250

2.178 (0.89)

H

1b''

O

Table 1 Relative enthalpies at 298 K for H2S2O isomers 1a–7a in the gas phase and aqueous solution (compared to the lowest energy isomer 2a) in kcal/mol and gas phase dipole moments in D

R = Phb

B3LYP/6-311 + G(2d,p)//B3LYP/6-31G(d); in parentheses DH298(G3B3) values. B3LYP/6-311 + G(2d,p)//B3LYP/6-31 + G(d,p). MP2/6-311G(d,p) [7]. Triplet states.

3.2. Comparison of the stabilities of thiosulfinic acids RS2OH 1a–e The relative enthalpies of isomers 1a–e, 2a–e and 3a–e in the gas phase are compared in Table 2. The stability order and the enthalpy differences are very similar for all organothiosulfinic acids. The only exception is fluoro-substituted analogue, in which the isomer 1e0 becomes more stable than the linear structure, 2e. Thus, only an extremely electronegative substituent can stabilize the thiosulfinic acid (‘‘branched”) structure. SCRF calculations show that this order does not change in aqueous solution. The bond lengths, charge distribution and Wiberg bond orders of methyl- and fluorothiosulfinic acids are compared in Scheme 4. NBO charge and bond order analysis suggests that the most significant changes induced by the substituent R occur in the S1 @ Y region (Y = O, S). Upon substitution of methyl by fluorine, the energy of electron delocalization from lone pairs on Y to antibonding r* orbitals of S1 with adjacent atoms increases. As a result, the S@Y bond shortens, bond order increases and the negative charge on Y decreases. Clearly, an electronegative substituent at S1 causes strengthening of the S1@Y double bond. The electron transfer from Y to F occurs mainly via negative hyperconjugation nY ! rSF . Consequently, the SAF bond

R = Fa

0.151

S 1.352 H

-0.487

F

1.223

S

1.650 (0.83)

(0.96)

-0.857

0.498

O 0.982 H (0.71)

1e'

1e'' 0

Scheme 4. B3LYP/6-31G* bond lengths (Å A), Wiberg bond orders (in parentheses) and NPA atomic charges (in italic) for thiolo- and thiono-forms of acids 1b and 1e.

order is only 0.65 compared to 0.92 for the SAC bond. The structure and electronic charge of the OH group in 10 0 and S2H in 10 remain practically unaffected by the change of the substituent R. Interestingly, the triplet states of 1a0 and 1a00 are by 32.2 and 34.8 kcal/mol, respectively, higher in energy than the singlet states. Substitution of hydrogen for methyl does not affect these relative energies significantly (Table 2). The energy of 10 is slightly lower than that of 100 due to the presence of a weak SAS bond. Indeed, 0 the SAS distance in triplet geometry (3.566 Å A) indicates well advanced dissociation to HASAO and SH fragments. 3.3. Reactivity of thiosulfinic acids Thiosulfinic acid 1a is thermodynamically unstable. The enthalpy of its hypothetical decomposition to water and elemental sulfur is 37.4 kcal/mol (Eq. (4)). In contrast to the MP2/6-311G(d,p) results [7], our calculations predict the thiolo-tautomer 1a0 to be slightly more stable than the thiono-form 1a00 , by 1.9 and 1.3 kcal/mol at the DFT and G3/B3LYP level, respectively. These energy differences are too small for a decisive conclusion about the relative stability of these tautomers. Moreover, our calculations predict that the dipole moments of the thiolo-tautomers are larger than those of the thiono-forms, but the difference is not significant (Table 1).

O HSSH

H2O + 1/4 S8

ð4Þ

To estimate the stabilization of thiosulfinic acids by polar solvents, we carried out the SCRF calculations of isomers 1a–7a in water. The stability order for isomers 1a–7a in water is very

108

M. Mikołajczyk et al. / Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

similar to that in the gas phase, with exception of isomer 7a, which is more effectively solvated, however, it is still the least stable species (Table 1). The free energy of solvation calculated by SCRF/B3LYP/6-311 + G(2d,p)//B3LYP/6-31 + G(d) method (referred to as DFT1) is not proportional to the dipole moment in the gas phase (Table 1). In particular, DGsolv for thiosulfinic acid 1a0 and 1a00 is exceptionally small. Thus, thiosulfinic acid 1a is less stable in solution than in the gas phase compared to linear HSSOH, 2a. In general, the energy differences between isomers 1a–7a in solution are reduced. The stability of both tautomeric forms 1a0 and 1a00 is approximately the same and they may be easily converted into each other depending on the properties of the reactive medium. Conversion between these tautomers may be accomplished by two routes: via a proton transfer mediated (i) by a water molecule and (ii) by an acid dimer (Eqs. (5) and (6)). The enthalpy barriers DHà for these reactions in the gas phase are 5.2 and 4.5 kcal/mol, respectively. Analogous reactions involving thiosulfinate anions should proceed with even lower energy barriers, due to their much higher nucleophilicities.

O R S

S

H

H

S

H2O

S

R

O

H

S

-H2O

R

H

O

S

1''

1' H

O 2R S

O H ð5Þ

S

S

O

S

S

O

H R

S

R

S

O H ð6Þ

H

1'

R in RS2OH à

DH racemization of 1’ DHà racemization of 1’’ DHà (Eq. (5)) DHà (Eq. (6)) DHà (Eq. (7)) DHà (Eq. (8)) RS(O)SH ? RS = O + SH RS(S)OH ? RS = S + OH

H (H2O)

Me

t-Bu

43.1 38.8 5.2 4.5 36.7 45.8

(40.9) (35.7)

45.5 42.7

42.8 39.5

(36.3) (45.2)

39.6 44.8 35.7 42.2

39.7 44.1

Ph

F 49.0 31.2

37.2 39.6

46.9 48.4 35.7 48.3

was obtained from calculations of the tautomeric structures of all thiosulfinic acids 1a–e, it is reasonable to assume that this is a general rule for thiosulfinic acids 1. Since one of the important feature of hypothetical thiosulfinic acids is its chirality, we have calculated the racemization barrier which determines their optical stability. The barriers of 30– 45 kcal/mol are comparable with those for the energies of the SAS and SAO bond dissociation. Thus, the fragmentation of thiosulfinic acids 1 can effectively compete with racemization. The barriers for the discussed reactions of thiosulfinic acids are listed in Table 3. 3.4. Thiosulfinate anions RS2O(), 1a–e()

S 2R

Table 3 Enthalpy barriers for reactions of thiosulfinic acids 1a–e (kcal/mol) at 298 K

1''

Removal of a proton from the acidic group in the structures 10 , 100 , or 1000 leads to the corresponding anions RS(O)(S)() 1() (Eq. (2)). The B3LYP/6-31 + G(d,p) geometrical parameters for the organothiosulfinic acid anions 1() are in good agreement with

0

Isomer 1 may also intramolecularly isomerize to the more stable linear form 2 (Eq. (7)). The barrier for this isomerization reaction in the gas phase is much higher, DHà = 36.7 kcal/mol. Isomer 100 may isomerize in an analogous way (Eq. (8)) to the higher in energy linear form 3 with an enthalpy barrier of 45.8 kcal/mol. SCRF calculations show that the solvation effect on the barriers of both reactions is negligible. Such high barriers, which are comparable with those for homolytic decomposition (the enthalpy for the SAS bond breaking in MeS(O)SH is 35.7 kcal/mol and the enthalpy for the SAO bond breaking in MeS(S)OH is 42.2 kcal/mol), indicate that these isomerizations should proceed on the fragmentation–recombination route rather than via an intramolecular rearrangement. An even smaller barrier of 18 kcal/mol was calculated for the enthalpy for the SAS bond breaking in HS(O)SH, although this value is probably underestimated [24].

S

S R

S

O H

S

O H

R S

S

1''-TS O S H S

H

1'-TS

Bond length (Å)

H, 1a()a Me, 1b()a t-Bu, 1c()a Ph, 1d()a F, 1e()a Adamantyl, 1f()b Triptycenyl, 1g()b a

S OH

ð7Þ

R

S

O

SH

ð8Þ

H

S

S1Ra

S1S2

RS1S2

OS1S2

1.541 1.540 1.541 1.538 1.502 1.536 1.526

1.392 1.856 1.938 1.851 1.805 1.845 1.843

2.084 2.077 2.073 2.072 2.007 2.059 2.004

96.7 99.7 103.3 101.9 102.1 104.7 105.8

115.6 114.3 113.2 114.8 112.9 109.7 109.2

S H

S

H O

79.8%

7.1%

6.0%

S

S

Me

S

Me

O

O

O

78.4%

7.1%

5.7%

S O 28.4%

S

F O

25.6%

S S O 5.2%

S

S F

S S O

5.7%

S

S Me

S F

S

O

S Me

S H

O

3

Among three possible tautomeric forms of the investigated thiosulfinic acids, the tautomers 1000 having the thiosulfone structure are the least stable. The calculated energy difference between 10 00 and the most stable isomers (10 or 2) is quite substantial ranging from 24 (R = F) to 40 kcal/mol (R = H). Such high energy values for 100 0 isomers are comparable to those of triplet excited states of thiosulfinic acids. Therefore, 1000 isomers should be very reactive and are expected to undergo facile decomposition. Since the same order of the relative stabilities

Valence angle (°)

S1O

B3LYP/631 + G(d,p). X-ray data.

b

2

R 1'

R

S

R S

R 1'' O

Table 4 Selected structural parameters of anions RS1(S2)(O)() (1())

S

S

F

S O

O 23.9%

7.3% ()

Scheme 5. Main resonance structures (NRT) for anions 1a

()

, 1b

and 1e().

109

M. Mikołajczyk et al. / Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

O -1.027 1.541(1.14) -0.025 1.392 H (0.86) S 0.764 2.085(1.07)

O -1.031

1.540(1.16) -0.185

S -0.763

Me

1.856 (0.90)

-0.605

S

F

0.971

2.077(1.09)

O

1.502(1.32) 1.805 (0.49)

S

-0.951

1.192

2.007(1.34)

S -0.756

S

-0.636

0

Scheme 6. B3LYP/6-31 + G(d,p) bond lengths (Å A), Wiberg bond orders (in parentheses) and NPA atomic charges (in italic) for anions 1a(), 1b() and 1e().

the results of the X-ray structural analysis for the S-benzylthiuronium salts of adamantanethiosulfinic acid 1f [4] and triptycenethiosulfinic acid 1g [25] (Table 4). The geometries of the anions are pyramidal, similar to the parent acids. The pyramidality of the central sulfur atom (defined as a sum of appropriate bond angles around it) in thiosulfinic acids and in the related anions is about 315–325°. The SAO and SAS bond lengths are intermediate between the corresponding bond lengths in precursor tautomers 10 and 100 . Analysis of charge distribution shows that the negative charge is almost equally distributed over O and S terminal atoms, whereas the central sulfur atom bears a significant positive charge. Thus, the attack of electrophilic reagents at both nucleophilic centers is probable. The resonance structures estimated by Natural Resonance Theory (NRT) analysis for 1a(), 1b() and 1e() are shown in Scheme 5. The distribution of the resonance structures for 1a() and 1b() is almost identical, while the pattern for 1e() is essentially different, showing extensive participation of the fluorine atom in the negative charge distribution. Electron-withdrawing effect of fluorine results in an increase of the double bond character of the SAO and SAS bonds and in reduction of the negative charge located on terminal atoms (Scheme 6). The gas phase acidities of acids 10 , 100 , 2 and 3 was calculated relative to water according to Eq. (9) (Table 5). The acidity increases in order 3  2 < 1. Substitution of hydrogen by an organyl group does not affect the acidities significantly. Small increase in acidity is observed when electron-withdrawing phenyl group is attached to sulfur. Fluorothiosulfinic acid isomers show enhanced acidities, as expected, due to stabilization of the conjugated anions by the strongly electronegative fluorine atom.

HA þ OHðÞ ! AðÞ þ H2 O

ð9Þ

Table 5 Relative acidities of acids 1–3 (free energies of proton exchange with water, Eq. (9), at 298 K, in kcal/mol) Acid 1

2

RS (O)S H, 1’ RS1(S2)OH, 1’’ RS1S2OH, 2 RS1OS2H, 3

H

Me

t-Bu

Ph

F

61.6 63.8 46.4 49.6

61.6 62.5 43.9 47.5

61.5 61.3 45.2 48.3

62.9 65.1 50.4 52.1

76.7 79.9 65.1 59.5

Proton abstraction from the central sulfur atom (S1) in the tautomeric structures 1a0 and 1a00 , leads to the anions 2a0 () and 2a00 () which are more stable than 1a00 (Scheme 7). Furthermore, anions 1a() may undergo isomerization which is significantly facilitated by the assistance of water molecule (Eq. (10)). The enthalpy barrier for isomerization 10 without intervention of water is 37.5 kcal/mol while that for water-assisted process is only 2.9 kcal/mol. Obviously, the replacement of a hydrogen at S1 by another group R makes this process impossible. The alternative pathway of isomerization of anions 100 () according to Eq. (11) has an extremely high energy barrier (DHà > 85 kcal/mol) and is not competitive with other reactions such as fragmentation. We could not locate the transition state for the reaction of 1() involving migration of sulfur, analogous to 11. Such reaction probably results in decomposition of the anion. Relative stabilities and reactivities of the anions are listed in Table 6.

H

O H S

S

H2 O

H O

O H

1a(-) S

R

S

S

S

S O

S

-H2O

HO S

O

ð10Þ

S

R S

S

O

R 1''

ð11Þ

2

1''-TS

4. Conclusions In summary, our calculations showed that (a) among the possible tautomeric forms of the parent thiosulfinic acids 1 the thiolo-forms are the most stable ones in the gas phase, (b) the thiosulfinate anions exist in the forms related to these tautomers, (c) the sulfur–sulfur bond is highly polar in these anions, and (d) the pathway for protonation of the thiosulfinate anion is determined by two matching factors – the charge distribution in the anion and the stability of the respective acid. Triplet states of thiosulfinic acids are relatively low-lying, and their energies are comparable to the energies of SAS and SAO bond breaking. The energies the SAS and SAO bond dissociation are in the range of 32–42 kcal/mol, much smaller than the energies of regular single bonds. Therefore, the main decomposition reaction of thiosulfinic acids is predicted

O HS1SH

-H+

1a'

OS1SH 2a'(-)

S HS1OH

-H+

1a'' Scheme 7.

SS1OH 2a''(-)

Table 6 Relative stabilities of anions 1()–3() and enthalpies of reactions (kcal/mol) at 298 K R in RS2O()

H (H2O)

Me

tBu

Ph

RS1(O)(S2)(), 1() RS1S2O(), 2’() ROS1S2(), 2’’() RS1OS2(), 3() DHà racemization DHà isomerization (Eq. (9)) DHà isomerization (Eq. (10))

0 0.2 7.3 30.1 40.2 37.5 37.5

0 7.4 6.4 30.7 45.7 (44.8)

0 6.7 4.9 31.5 45.7

0 1.8 4.0 26.0 30.5

0 13.6 a 47.6 27.8

90.6

90.5

85.7

108.2

a

(8.1) (18.1) (19.9) (38.3) (2.9) (36.3)

Isomerizes upon optimization to 1

()

.

F

110

M. Mikołajczyk et al. / Journal of Molecular Structure: THEOCHEM 863 (2008) 105–110

to be the homolytic fragmentation. Steric effect (in the range of substituents examined) and solvation seem to play a minor role in the stabilization of thiosulfinic acids. More significant changes are observed due to the inductive effect. Thus, strongly electronwithdrawing substituents such as fluorine stabilize both acid and anion. The observed higher stability of anions compared to acids may be explained by the resistance of the negatively charged anions to nucleophilic attack.

[9] [10] [11] [12] [13]

Acknowledgement _ ´ ski for his assistance The authors thank Dr. Remigiusz Zurawin in the initial calculations. References [1] T. Takata, T. Endo, in: S. Patai (Ed.), The Chemistry of Sulphinic Acids, Esters and Their Derivatives, Wiley, New York, 1990 (Chapter 18). [2] F. Wudl, D.A. Lightner, D.J. Cram, J. Am. Chem. Soc. 89 (1967) 4099. [3] W. Weigand, R. Wünsch, Chem. Ber. 129 (1996) 1409. _ [4] M. Mikołajczyk, P. Łyzwa, M. Wieczorek, G. Bujacz, Angew. Chem. Int. Ed. Engl. 28 (1989) 97. _ [5] J. Drabowicz, P. Łyzwa, B. Bujnicki, M. Mikołajczyk, Phosphorus Sulfur and Silicon 95–96 (1994) 293. [6] H. Basch, in: S. Patai (Ed.), The Chemistry of Sulphinic Acids, Esters and Their Derivatives, Wiley, New York, 1990 (Chapter 2). [7] R. Steudel, Y. Drozdova, R.H. Hertwig, W. Koch, J. Phys. Chem. 99 (1995) 5319. [8] A. Königshofen, M. Behnke, M. Hoverath, J. Hahn, Z. Anorg, Allg. Chem. 625 (1999) 1779.

[14] [15] [16] [17] [18]

[19] [20] [21] [22] [23] [24] [25]

G. de Petris, M. Rosi, A. Troiani, J. Phys. Chem. A 111 (2007) 6526. A.D. Becke, J. Chem. Phys. 98 (1993) 5648. A.D. Becke, Phys. Rev. A 38 (1998) 3098. C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1998) 785. M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision D. 01, Gaussian, Inc., Wallingford, CT, 2004. C. Gonzalez, H.B. Schlegel, J. Chem. Phys. 90 (1989) 2154. C. Gonzalez, H.B. Schlegel, J. Phys. Chem. 94 (1990) 5523. A.G. Baboul, L.A. Curtiss, P.C. Redfern, K. Raghavachari, J. Chem. Phys. 110 (1999) 7650. E.D. Glendening, A.E. Reed, F. Weinhold, NBO ver. 3.1 manual. E.D. Glendening, J.K. Badenhoop, A.E. Reed, J.E. Carpenter, J.A. Bohmann, C.M. Morales, F. Weinhold, NBO 5.0., Theoretical Chemistry Institute, University of Wisconsin, Madison, 2001. J. Tomasi, M. Persico, Chem. Rev. 94 (1994) 2027. J. Tomasi, R. Cammi, J. Comput. Chem. 16 (1995) 1449. M. Cossi, V. Barone, R. Cammi, J. Tomasi, Chem. Phys. Lett. 255 (1996) 327. V. Barone, M. Cossi, J. Tomasi, J. Chem. Phys. 107 (1997) 3210. E. Cancès, B. Mennucci, J. Tomasi, J. Chem. Phys. 107 (1997) 3032. R. Benassi, G.L. Fiandri, F. Taddei, Tetrahedron 50 (1994) 12469. M. Wieczorek, Personal Communication (2006).