Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67 www.elsevier.com/locate/theochem
Theoretical investigation of sulfur and halogen-substituted carbocations Paul Kiprof *, Stephen R. Miller, Melissa A. Frank Department of Chemistry and Biochemistry, University of Minnesota Duluth, 1039 University Drive, Duluth, MN 55812, USA Received 13 December 2005; accepted 17 February 2006 Available online 5 April 2006
Abstract The stabilization of sulfur, fluorine and chlorine substituted carbocations was studied using hydride exchange reactions at the G2 level of theory; evaluation of the p bonding in these was performed using variation of the dihedral angle in the case of sulfur compounds and NRT and other bond order calculations for all compounds. Resonance contributors were estimated using the NRT method. Sulfur substitution stabilizes carbocations comparable to oxygen substitution covered previously. The stability of trisubstituted carbocations is enhanced over the oxygen analogues. Halogen substitution leads to destabilization, which is more extreme for the fluorine-substituted carbocations. q 2006 Elsevier B.V. All rights reserved. Keywords: Carbocation; Sulfur; Fluorine; Chlorine; G2; Hydride exchange; Resonance; NBO; NRT; Bond order
1. Introduction Carbocations are important intermediates in organic chemistry [1]. Heteroatom substitution of intermediate carbocations can lead to their stabilization and enables organic reactions, that normally would not occur were the carbocation not stabilized [2]. In an earlier study, we addressed the effect of oxygen substitution on the stability of carbocations [3]. This work comprises the study of sulfur and halogenstabilized carbocations. Sulfur has a much lower electronegativity than oxygen, closer to the electronegativity of carbon, whereas the electronegativity of the halogens covers a wide range, from the electronegativity of fluorine being higher than the electronegativity of oxygen and the electronegativity of iodine close that of carbon [4]. The ability to form p-bonds is the highest for the second row elements oxygen (less so for fluorine) and decreases going to the third and higher periods [4]. We studied the stabilization of heteroatom substituted carbocations and evaluated the effect of each heteroatom substitution energetically and through calculation of charges using various models and analysis of the bonding situation in these carbocations, including a study of the natural * Corresponding author. Tel.: C1 218 726 8021; fax: C1 218 726 7394. E-mail address:
[email protected] (P. Kiprof).
0166-1280/$ - see front matter q 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.theochem.2006.02.009
resonance structures using NRT theory as implemented in NBO. 2. Methods All calculations were done using the GAUSSIAN 03 program system [5] and the NBO 5 program [6]. Mayer bond orders were calculated using the program BORDER [7]. High-level energy calculations were done using the standard G2 method as implemented in GAUSSIAN 03. The energies of carbocations and analogous alkanes were calculated. These energies were used to construct isodesmic reactions, essentially hydride transfer reactions, that compare the stability of the respective carbocation to the tert-butyl cation. Furthermore, optimized geometries at the MP2(full)/6-311G** level of theory were calculated to study charges and bonding of the carbocations. For all charge, bonding and resonance calculations the SCF wave function was used. The variation of the dihedral angles C–S–C–C for the sulfur compounds was performed at the B3LYP/6-31G* and B3LYP/6-311G** levels of theory. In this calculation, the dihedral angle was varied from 0 to 3608 in increments of 108 and kept constant for each calculation while all other structural parameters were allowed to optimize. Charges were calculated as Mulliken, CHELPG charges as implemented in GAUSSIAN 03 and natural charges. Bonding parameters were calculated as Wiberg bond indices and bond orders from NRT calculations. Additionally, natural resonance theory (NRT) analyses were performed on all carbocations
62
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67
Scheme 1.
to calculate resonance structures and their contributions. Different solutions from the NRT routine were evaluated concerning their feasibility and figure of merits. Single resonance structures were also analyzed of how well they describe the electron density as a Lewis structure. In most cases, it was found that resonance hybrids were necessary for the correct description. The final result was verified by input of all reasonable resonance structures as a starting point to calculate the resonance hybrid.
3. Results and discussion 3.1. Isodesmic hydride exchange reactions The results of the hydride exchange isodesmic reactions are shown in Schemes 1–3. Sulfur substitution results in considerably enhanced cation stabilization compared with the tert-butyl cation. The first substitution of a methyl group with a methylthio group makes
Scheme 2.
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67
63
Scheme 3.
the carbosulfonium ion more stable by 23.12 kcal/mol. This is quite comparable to the stabilization by a methoxy group of 25.73 kcal/mol [3]. This is consistent with studies by Apeloig and Karni on the hydride exchange between CH2OHC and CH3SH [8]. The second substitution adds another 11.76 kcal/ mol to raise the total stabilization to 34.88 kcal/mol. The third substitution adds another 7.07 kcal/mol to the stability of the carbocation resulting in a total energy of stabilization of 41.95 kcal/mol. Unlike the oxygen-stabilized carbocations, the second incremental stabilization in the sulfur system is smaller, while the final substitution gives a higher incremental stabilization energy [3]. This corresponds to the trend observed ¨ sapay in experimental gas phase basicity determinations by O et al. [9]. The fluoro and chloro substituted carbocations are all destabilized compared with the tert-butyl cation, with the first substitution leading to a destabilization of 8.77 kcal/mol for the fluoro and 8.66 kcal/mol for the chloro case. The destabilization values for the second and third substitution are much higher for the fluorine case (25.89 and 60.26 kcal/mol) than for the chloro substitution (16.32 kcal/mol for the second and 22.79 kcal/mol for the third substitution).
to stabilization of the system. Maxima in energy are reached at dihedral angles of 90 and 2708, where no constructive p overlap between carbon and sulfur exists (Fig. 1). The curves for the B3LYP/6-31G* and B3LYP/6-311G** levels of theory are practically identical. In contrast to the analogous oxygen case, the range of stabilization/destabilization is much larger (30.11 kcal/mol vs. 16.6 kcal/mol). For the carbosulfonium ion, absence of p bonding leads to net destabilization vs. the tert-butyl cation,
3.2. Evaluation of p bonding contributions to the stabilization of carbosulfonium ions The influence of p bonding on the stabilization energy of the methylthio dimethyl cation was studied by variation of the dihedral angle about the C–S bond. At dihedral angles 0, 180, and 3608, the relative energy is at a minimum since there is maximal p-overlap between carbon and sulfur leading
Fig. 1. C–S bond rotation barrier in the methylthiodimethyl carbocation at the B3LYP/6-311G** level of theory.
64
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67 Table 1 Calculated charges of the cations Molecule
Fig. 2. C–S bond rotation barrier in the bis(methylmercapto)methyl carbocation at the B3LYP/6-311G** level of theory.
making p bonding an important part in the stabilization of carbosulfonium ions. In the case of two methylthio substituents, the overall range of energies is less when the p bonding for one C–S bond is lost, since stabilization of the carbocation is partially maintained by the second methylthio group (Fig. 2) and therefore, the destabilization caused by lack of p bonding of one sulfur group is less than 19 kcal/mol.
Atom
Mulliken a
NBO
cHelpg a
0.68a
C
0.05
0.67
C S S
K0.29 0.40 0.39
K0.14 0.51 0.46
0.27 0.04 0.08
C F
0.50 K0.19
1.07 K0.30
0.87 K0.19
C F
0.96 0.01
1.65 K0.22
1.21 K0.07
C Cl Cl
K0.03 0.00 K0.01
0.28 0.28 0.28
0.27 0.22 0.22
C S
K0.16 0.45
0.18 0.57
0.39 0.08
C S
K0.40 0.38
K0.42 0.45
0.15 0.06
C F
0.72 K0.05
1.38 K0.26
1.02 K0.10
C Cl
K0.01 0.24
0.48 0.25
0.44 0.16
C Cl
K0.04 0.35
0.06 0.31
0.15 0.28
3.3. Charge determination The charges of the cations are listed in Table 1. The Mulliken and NBO charges generally give a more negative (i.e. less positive) value for the carbon in the sulfur containing cations. As a general trend, the charge on the carbon moves to less positive values with increasing sulfur substitution. In the case of the fluorine-containing ions, the NBO method gives the most positive values for the charges. The general trend here is an increase of the positive charge on the cationic carbon center. For the chlorine-substituted cations, the NBO and CHELPG methods give very similar values and both show a decrease in the charge of the cationic carbon with increasing substitution. The charge of the cationic carbon for sulfur and chlorine substitution is less positive than the charge of the tert-butyl central carbon, and all fluorine-substituted structures have more positive cationic carbon charges than the tert-butyl ion. Upon fluorine substitution, the carbocation is destabilized and the charge of the cationic carbon increases. The opposite is true for the sulfur substitution. The added methylthio substituents increase the stability and lower the charge of the cationic carbon. Since the chlorine case does not follow this trend, one has to assume that lowering of the charge is only one component of the stabilization of the carbocation.
a
Ref. [3].
3.4. Resonance structures and bonding All cationic structures were analyzed by an NBO calculation. For the carbosulfonium ions (Table 2), the bond orders of all methods are in good agreement. For the methylthio cation the NRT bond order is higher then the other two methods.
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67
65
Table 2 Natural resonance theory bond orders, Wiberg bond indices, and Mayer bond orders of the carbosulfonium ions Molecule
NRT bond order Wiberg bond index Mayer bond order
1.90 1.65 1.60
1.49, 1.46 1.43, 1.39 1.36, 1.31
In this latter case, there is a strong ionic character of the C–S bond, which manifests itself mostly in the p bonding interaction between carbon and sulfur, which is slightly tilted to a higher contribution of the sulfur atom, while the s bond has an equitable contributions from both atoms. The same applies to the bis(methylmercapto)cation, where the overall bond order is lower. The agreement in this case for all three methods is better. For the tri-substituted cation, the Mayer bond order is significantly lower than the NRT or Wiberg bond index. In the fluorine-substituted systems, the NRT analysis finds a significantly higher bond order for the C–F bond than the other two methods. The C–F bond is found to be highly polar with a significant ionic character. A strong p interaction exists within each C–F bond, which has contributions mostly from fluorine orbitals. The NRT analysis therefore, finds the bond order higher (Figs. 3–5). In the case of the trifluorocation, no significant contribution of the resonance structure with all single bounds could be detected in the NRT analysis. In general, for the fluorine systems, the C–F bonding is polar, because of the difference in electronegativity. This is true for p or s bonding. Both bonding interactions have high
1.33 1.29 1.19
1.97 1.88 1.94
0.99 0.99 1.00
ionic character. The presence of a p bonding interaction does not automatically lead to a charge on the fluorine atom, as would be suggested by the formal charges. Instead, both, p and s interactions, are mostly ionic in nature. One description is to assume a dative p bond between carbon and fluorine. The natural bonding orbitals have mostly contributions from fluorine rather than from carbon. The opposite is true for the natural anti-bonding orbitals. The ability to describe the bonding in terms of Lewis structures and resonance hybrids becomes limited.
Fig. 4. Resonance structures of the fluorine compounds.
Fig. 3. Resonance structures of the sulfur compounds.
Fig. 5. Resonance structures of the chlorine compounds.
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67
1.33 1.35 1.30
1.01 0.97 1.00
66
In the case of the chlorine-substituted carbocations, the NRT bond order is also higher than for the other two methods, however, the difference is not as large as for the fluoro carbocations. This is manifested also in the less ionic character of the C–Cl bonds. The covalent character, in turn, is higher than that of the C–F bonds. In the natural bond orbital analysis, s and p bonding interactions are found for the most prevalent Lewis structure, and the contributions of carbon and chlorine to the s bonding interaction are close to equal. The p bonding interaction has, like in the case of the C–F bonds, a significantly higher contribution of the halogen (Table 3).
1.86 1.43 1.49 1.00 0.78 0.80
Sulfur substitution leads to the greatest stabilization of carbocations, even compared with carboxonium ions. Chlorine/fluorine substitution leads to destabilization of the carbocation. The destabilizing effect is greater for fluorine compounds than for chlorine compounds. In all cases substantial p interaction of the cationic carbon and the heteroatom was seen. However, neither p interaction, nor the charge of the cationic carbon can be related to the stabilization energy of the compound. In the cases of the halogen substitution, bonds have a high ionic component and are very polar, such that p interactions can exist, however, the positive charge remains mostly on the cationic center. In the case of such highly polar bonds, there is a discrepancy in the methods determining bond order. Acknowledgements
NRT bond order Wiberg bond index Mayer bond order
1.85 1.04 1.19
1.48 1.08 1.27
1.33 1.12 1.34
We would like to thank the Supercomputing Institute for Digital Simulation and Advanced Computation of the University of Minnesota and the Visualization and Digital Imaging Laboratory at UMD for valuable computer resources.
Molecule
Table 3 Natural resonance theory bond orders, Wiberg bond indices, and Mayer bond order of the halogen-substituted carbocations
1.46 1.39 1.38
4. Conclusion
References [1] (a) P.A. Krasutsky, I.V. Kolomitsyn, P. Kiprof, R.M. Carlson, A.A. Fokin, J. Org. Chem. 65 (2000) 3926; (b) P.A. Krasutsky, I.V. Kolomitsyn, P. Kiprof, R.M. Carlson, N.A. Syderenko, A.A. Fokin, J. Org. Chem. 66 (2001) 1701; (c) G.A. Olah, K.K. Laali, Q. Wang, G.K.S. Prakash, Onium Ions, Wiley, New York, 1998; (d) G.A. Olah, J. Org. Chem. 66 (2001) 5943 (and references therein); (e) G.A. Olah, D.G. Parker, N. Yoneda, Angew. Chem., Int. Ed. Engl. 17 (1978) 909. [2] (a) A. Bayer, V. Villiger, Ber. 32 (1889) 3625; (b) A. Bayer, V. Villiger, Ber. 33 (1900) 858; (c) R. Criegee, R. Chem. Ber. 77 (1944) 722. [3] S.R. Miller, S. Krasutsky, P. Kiprof, J. Mol. Struct. (Theochem) 674 (2004) 43–47. [4] G. Wulfsberg, Inorganic Chemistry, 2000, University Science Books, Sausalito, CA, 2000. [5] 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,
P. Kiprof et al. / Journal of Molecular Structure: THEOCHEM 764 (2006) 61–67 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 C.01, Gaussian, Inc., Wallingford, CT, 2004.
67
[6] NBO 5.0. E.D. Glendening, J.K. Badenhoop, A.E. Reed, J.E. Carpenter, J.A. Bohmann, C.M. Morales, F. Weinhold, Theoretical Chemistry Institute, University of Wisconsin, MI, 2001. [7] I. Mayer, Program ‘BORDER’, Version 1.0 (Chemical Research Center, Hungarian Academy of Sciences), Budapest, 2005. [8] Y. Apeloig, J. Karni, J. Chem. Soc., Perkin Trans. 2 (1988) 625. ¨ sapay, J. Delhalle, K.M. Nsunda, E. Rolli, R. Houriet, L. Heves, J. Am. [9] K. O Chem. Soc. 11 (1989) 5028.