Journal of Molecular Structure: THEOCHEM 950 (2010) 41–45

Contents lists available at ScienceDirect

Journal of Molecular Structure: THEOCHEM

journal homepage: www.elsevier .com/locate / theochem

Ramberg–Bäckland reaction of a-chloromethyl methyl sulfone: A DFT study

Hui Zhang, Lan-Xiang Zhang, Fei Yang, Feng-Ling Liu *

College of Chemistry, Chemical Engineering and Materials Science, Shandong Normal University, Jinan 250014, People’s Republic of China

a r t i c l e i n f o

Article history:Received 9 December 2009Received in revised form 18 March 2010Accepted 19 March 2010Available online 28 March 2010

Keywords:B3LYP/6-311+G(d,p)a-Chloromethyl methyl sulfoneRamberg–Bäckland reactionTransition stateEnergy barrier

0166-1280/$ - see front matter � 2010 Elsevier B.V. Adoi:10.1016/j.theochem.2010.03.022

* Corresponding author. Tel.: +86 531 86189772; faE-mail addresses: zhanghui0805@126.com (H. Zh

Liu).

a b s t r a c t

The mechanisms of decomposition of a-chloromethyl methyl sulfone (CH2ClSO2CH3) uncatalyzed andcatalyzed by hydroxide ion have been studied by using the density functional theory (DFT) at theB3LYP/6-311+G(d,p) level. The computational results indicate that the decomposition of a-chloromethylmethyl sulfone uncatalyzed is with two steps, and catalyzed by hydroxide ion is via three steps. The com-putational results show that the decomposition of a-chloromethyl methyl sulfone catalyzed by hydrox-ide ion namely Ramberg–Bäckland reaction is much easier.

� 2010 Elsevier B.V. All rights reserved.

1. Introduction

The formation of a carbon–carbon double bond plays an impor-tant role in organic chemistry, especially in the synthesis of com-plex natural products. Ramberg and Bäckland described areaction in which a-bromoethyl ethyl sulfone was converted tocis-2-butene [1] namely the Ramberg–Bäckland reaction, whichprovides a way to form a carbon–carbon double bond.

SO2

Although there are many ways to form alkenes, the Ramberg–Bäckland reaction shows the advantages of the good regioselec-tivity and stereoselectivity [2]. Furthermore, because of its abilityto convert sulfones into alkenes in one operation without re-course to the prior preparation of the a-halo sulfones by a sepa-rate step, the Mayers’ modification of the Ramberg–Bäcklandreaction has been used with considerable frequency in organicsynthesis [3].

The mechanism of the Ramberg–Bäckland reaction was pre-sumed by Bordwell and Paquette as follows [4–7].

ll rights reserved.

x: +86 531 82615258.ang), sdliufl@sina.com (F.-L.

S

H2C CHR1

R2

X +

S

HC CH

O O

R1

R2

X + H2OOH

OO

ð1Þ

S

HC CH

O O

R1

R2

X

S

HC CH

O O

R1

R2

+ X ð2Þ

S

HC CH

O O

R1

R2

HR1C CHR2 ++ 2OH SO32- + H2O ð3Þ

Treating the a-halo sulfone with base, it releases hydrogen ha-lide and converts to ring sulfone by 1,3-elimination, and then re-leases sulfur dioxide and forms alkene.

The Ramberg–Bäckland reaction has been applied widely in theorganic synthesis [2], and many investigations are to optimize thereaction conditions of composing alkenes. To the best of ourknowledge, the mechanism of the Ramberg–Bäckland reactionhas not been studied by using quantum-chemical methods. Thusthe decomposition mechanism of the a-chloromethyl methyl sul-fone has been studied by using DFT method at the B3LYP/6-311+G(d,p) and the MP2/6-311+G(d,p) level of theory in this work.

In this paper, we present the results of DFT calculations at theB3LYP/6-311+G(d,p) and the MP2/6-311+G(d,p) level of theory de-

Fig. 1. Potential energy surfaces obtained at the B3LYP/6-311G(d,p) level fora-chloromethyl methyl sulfone.

42 H. Zhang et al. / Journal of Molecular Structure: THEOCHEM 950 (2010) 41–45

signed to understand the mechanism of the decomposition reac-tion of a-chloromethyl methyl sulfone. Detailed information onthe reaction processes is provided in this paper.

2. Computational methods

All theoretical calculations are carried out with the GAUSSIAN-03 program package [8]. First, potential energy surfaces are ob-tained at B3LYP/6-311G(d,p) level in order to find the most stableconformation. The geometries of all reactants, transition states andproducts for the reactions are optimized by using density func-tional theory (DFT) at the B3LYP/6-311+G(d,p) level of theory. Allequilibrium states and transition states have been verified accord-ing to the number of imaginary frequency through the vibrationalanalysis. The reactants, products and intermediates after optimizedare the energy minimal points on the potential energy surfaces,and their frequencies are all positive. All transition states havebeen determined upon the vibrational analysis from the uniqueimaginary vibration mode and further confirmed with intrinsicreaction coordinate (IRC) theory.

The energy barrier Ea (theoretical activation energy) of each ele-mentary reaction is calculated according to the following equation:

Ea ¼ ðEmÞTS � ðEmÞRwhere (Em)TS and (Em)R are the sum of electronic and zero-pointenergies of the transition state and the reactant, respectively.

3. Results and discussion

MacLean and Chandler [9], and Krishnan et al. [10] suggestedthat 6-311+G(d,p) basis set is more suitable for the molecules con-taining S and Cl elements than 6-31G(d,p), so except the calcula-tions of potential energy surfaces, all the other calculations ofthis work are carried out at the B3LYP/6-311+G(d,p) level of theory.

3.1. The decomposition of a-chloromethyl methyl sulfone

The stable rotamers of a-chloromethyl methyl sulfone wereidentified in the potential energy surface (PES) through rotatingthe Cl–C–S–CH3 torsional angle obtained at the B3LYP/6-311++g(3df,3pd) level by Rittner et al. [11]. We consulted the methodused by Rittner et al. to study the conformation of a-chloromethylmethyl sulfone and obtained the stable rotamer. The potential en-ergy surfaces are built through rotating the Cl(10)–C(5)–C(1)–S(9)(the numbering system of atoms in Fig. 2) torsional angle, and theoptimized structures from �128.1� to 231.9� in step of 10� at theB3LYP/6-311G(d,p) level, see Fig. 1. There are two minima in ourPES, this is agreement with Rittner et al. result. From Fig. 1, itcan be seen that the most stable conformation is the torsionalangle at �128.1� or 231.9�. In the following calculations, all thecalculations about a-chloromethyl methyl sulfone are all on thisconformation.

Subsequent calculations of the reaction of a-chloromethylmethyl sulfone are calculated at the B3LYP/6-311+G(d,p) level.The results indicate that a-chloromethyl methyl sulfone is decom-posed to ethene, sulfur dioxide and hydrogen chloride via twosteps. Energies corrected with zero-point energies (ZPE) of reac-tants, transition states, intermediates and products, and energybarriers of the reaction are collected in Table 1. Fig. 2 shows thegeometries of a-chloromethyl methyl sulfone, the reaction transi-tion states and the intermediate, and some parameters are pre-sented in it. Fig. 3 presents the potential energy profile and thekey geometries for the reaction.

In the first step, the a-chloromethyl methyl sulfone produces acyclo-structure and a HCl molecule via a transition state of TSa1,

and the calculated energy barrier is 86.2 kcal/mol. A significant fea-ture of the transition structure TSa1 is the presence of an essentiallyplanar four-membered ring formed because of the shortening dis-tance of Cl(10) and H(2). Then the bonds of C(5)–Cl(10) and C(1)–H(2) are lengthened gradually till broken at the end, and the bondlength of Cl(10)–H(2) is shorten from 3.045 to 1.674 Å, while allbond lengths of C(1)–C(5), C(1)–S(9) and C(5)–S(9) are shorteneda little. In the progress of forming the intermediate, the bondlength of C(1)–C(5) is shortened from 2.368 to 1.579 Å, leadingthe C(1), C(5) and S(9) atoms to form a three-membered ring. Atthe same time, Cl(10) and H(2) atoms have been departed fromC(5) and C(1), respectively, leading to form a HCl molecule.

In the second step, the three-membered ring is decomposed viaa transition state TSa2 and the calculated energy barrier is15.2 kcal/mol. The bond lengths of S(9)–C(1) and S(9)–C(5) arelengthened, and the bond length of C(1)–C(5) is shortened gradu-ally till produced SO2 and ethene. The transition state has beenconfirmed with intrinsic reaction coordinate (IRC) theory whichis presented in Fig. 3.

For the reaction channel composed of several elementaryreactions, the reaction with highest activation energy is the speed-controlled step. So the first step of the decomposition of a-chloro-methyl methyl sulfone uncatalyzed is the speed-controlled, i.e. thekey step.

3.2. The a-chloromethyl methyl sulfone decomposition catalyzed byhydroxide ion

Since the energy barrier of the speed-controlled step is veryhigher, the calculated results show that the a-chloromethyl methylsulfone decomposition uncatalyzed to form ethene, sulfur dioxideand hydrogen chloride is strongly disfavored. However, the Ram-berg–Bäckland reaction can occur easily. So we study the reactionof a-chloromethyl methyl sulfone decomposition catalyzed byhydroxide ion at the B3LYP/6-311+G(d,p) level. The calculationalresults indicate that the reaction is carried out in 3 steps. All thetransition states with solid arrows indicating displacement vectorsare showed in Fig. 4. Fig. 5 shows the potential energy surface forthe a-chloromethyl methyl sulfone decomposition catalyzed byhydroxide ion.

The first step is a progress of H(11) atom combining withhydroxide ion and moving away from C(7) atom. In the vibrationalmode corresponding to the imaginary frequency (m = 755i cm�1),the dominant motion mi is H(11) atom being transferred betweenC(7) and O(12) atoms. In the progress of moving away from C(7),H(11) atom is close to O(12) atom gradually till forming a singlebond. The products contain a H2O molecule and a metastable carb-anion. Because of the strong attraction from the hydroxide ion, the

Fig. 2. A depiction of the optimized a-chloromethyl methyl sulfone (a), TSa1 (b), intermediate (c), and the TSa2 (d) structures.

Table 1Energies of all species and activation energies of the decomposition reaction of a-chloromethyl methyl sulfone uncatalyzed, catalyzed by hydroxide ion, and the reaction inmethanol.

Species a-Chloromethyl methyl sulfonedecomposition uncatalyzeda

a-Chloromethyl methyl sulfonedecomposition catalyzedby hydroxide iona

a-Chloromethyl methyl sulfonedecomposition catalyzedby hydroxide ionb

The reaction in methanola

Reactant �1088.03862 �1163.92064 — —TS1 �1087.90123 �1163.91907 — —Ea1 86.2 1.0 — —Intermediate1 �1088.00302 �1163.92683 �1162.00846 �1163.92842TS2 �1087.96822 �1163.90458 �1161.97640 �1163.91191Ea2 21.8 14.0 20.1 10.4Products/intermediate2 �1088.05561 �1163.95557 — —TS3 — �1163.94307 — —Ea3 — 7.8 — —Product — �1164.02282 — —

a Energies (a.u.) are obtained at the B3LYP/6-311+G(d,p) level of theory.b Energies (a.u.) are obtained at the MP2/6-311+G(d,p) level of theory. All corrected by zero-point energies. Ea in kcal/mol.

H. Zhang et al. / Journal of Molecular Structure: THEOCHEM 950 (2010) 41–45 43

H(11) atom is lose from C(7) atom easily, and the energy barrier(Eb1) is very low, being only 1.0 kcal/mol.

The second step is a progress that the carbanion loses chlorideion via TSb2. The dominant motion mi of the vibrational mode cor-responding to the imaginary frequency m = 391i cm�1) containsCl(10) atom deviating from C(4) atom and C(4), C(7) atomsclosing each other. The products are a metastable intermediatewith three-membered ring and a chloride ion. The energy barrier(Eb2) of this step is 14.0 kcal/mol, higher than that in the firststep.

In the third step, the three-membered ring is decomposed toSO2 and ethene. According to the results, the energy barrier is7.8 kcal/mol which is lower than Ea2, so the second step is the

key step to the reaction. This result is consistent with the experi-ments of Bordwell and Cooper [4].

In quest of the more accurate results of the key step, we recal-culated the IMb1 and TSb2 at the MP2/6-311+G(d,p) level. The ener-gies at the MP2/6-311+G(d,p) level also given in the Table 1, andthe energy barrier of the key step is 20.1 kcal/mol which is in rea-sonable agreement with experiment [12].

Since the solvent plays an important role in the Ramberg–Bäck-land reaction, and many reactions are studied in methanol [12], sowe calculate the speed-controlled step in methanol at the B3LYP/6-311+G(d,p) level to contrast the results. The results indicate thatthe energy barrier (Eb2) reduces to 10.4 kcal/mol from 14.0 kcal/mol most because of the reducing of the energy of the transition

Fig. 3. Minimum-energy pathway for the two steps reaction of a-chloromethyl methyl sulfone.

44 H. Zhang et al. / Journal of Molecular Structure: THEOCHEM 950 (2010) 41–45

state TSb2. This means that the Ramberg–Bäckland reaction is eas-ier in methanol. The energies of the intermediate and transitionstate of the step are collected in Table 1.

According to the results of this work, comparing with thedecomposition of a-chloromethyl methyl sulfone uncatalyzed,the reaction catalyzed by hydroxide ion is much easier. As we

know most organic reactions have activation energies in the rangeof 40–125 kJ mol�1 (10–30 kcal mol�1). Reaction with activationenergies less than 80 kJ mol�1 take place spontaneously at or be-low room temperature, whereas reactions with higher activationenergies normally require heating [13]. According to the result ofthis work, the key step of the energy barrier of the a-chloromethyl

Fig. 4. TSs with solid arrows indicating displacement vectors for a-chloromethylmethyl sulfone decomposition catalyzed by hydroxide ion.

Fig. 5. The potential energy surface for the a-chloromethyl methyl sulfonedecomposition catalyzed by hydroxide ion.

H. Zhang et al. / Journal of Molecular Structure: THEOCHEM 950 (2010) 41–45 45

methyl sulfone decomposition catalyzed by hydroxide ion is only20.1 kcal/mol, coincides with experiment conditions of theRamberg–Bäckland reaction [1].

4. Conclusions

DFT calculations at the B3LYP/6-311+G(d,p) level have beenused to study the a-chloromethyl methyl sulfone decomposition

uncatalyzed and catalyzed by hydroxide ion. The main conclusionsof our work at this level of theory can be collected as follows:

The decomposition reaction of a-chloromethyl methyl sulfoneuncatalyzed contains two steps separated by a metastable inter-mediate, and the first one is the key step. The higher energy barrierof the key step indicates that a-halo sulfone is difficult to decom-pose. The mechanism can be described by the following reactionsequences:

S

XH2C CH3

O O

S

H2C CH2

O O

+ HX ð1Þ

S

H2C CH2

O O

CH2H2C + SO2 ð2Þ

The a-chloromethyl methyl sulfone in the presence of base,such as hydroxide ion, is much easier to decompose forming eth-ene, sulfur dioxide and hydrogen chloride. The mechanism canbe described by the following reaction sequences:

S

H3CH2C X +

S

H2CH2C

O O

X + H2OOH

OO

ð1Þ

S

H2CH2C

O O

X

S

H2C CH2

O O

+ X ð2Þ

S

H2C CH2

O O

H2C CH2 + SO2 ð3Þ

The DFT study of this work shows that the decomposition of a-chloromethyl methyl sulfone catalyzed by hydroxide ion is withthree steps and much easier than the decomposition of a-chloro-methyl methyl sulfone uncatalyzed, which is accordance with themechanism presumed by of Bordwell and Paquette [4–7]. The reac-tion is much easier if methanol and hydroxide are used as solventand catalyst.

References

[1] L. Ramberg, B. Bäcklund, Arkiv. Kemi, Mineral. Geol. 13A (1940) 50.[2] X.-L. Wang, X.-P. Cao, Z.-L. Zhou, Youji Huaxue 23 (2003) 120.[3] T.-L. Chan, S. Fong, Y. Li, T.-O. Man, C.-D. Poon, J. Chem, Soc. Chem. Commun.

(1994) 1771.[4] F.G. Bordwell, G.D. Cooper, J. Am. Chem. Soc. 73 (1951) 5187.[5] F.G. Bordwell, Acc. Chem. Res. 3 (1970) 281.[6] L.A. Paquette, J.C. Philips, Tetrahedron Lett. 8 (1967) 4645.[7] L.A. Paquette Jr., R.E. Wingard, G.M. Photis, J. Am. Chem. Soc. 96 (1974) 5801.[8] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheesman,

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, G.A. Petersson, H.Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishoda, T.Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian,J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.Austin, R. Cammi, C. Pomelli, J. Otcherski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.Salvador, J.J. Danneberg, 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. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P.Piskorz, I. Kamaromi, L.R. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.Nanaykkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, G.Gonzalez, J.A. Pople, Gaussian 03. Revision B.03, Gaussian Inc., Pittsburgh, PA,2003.

[9] A.D. MacLean, G.S. Chandler, J. Chem. Phys. 72 (1980) 5639.[10] R. Krishnan, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 72 (1980) 650.[11] Lucas C. Ducati, Matheus P. Freitas, Claudio F. Tormena, Roberto Rittner, J. Mol.

Struct. 800 (2006) 45.[12] F.G. Bordwell, Joel M. Williams Jr., JACS 90 (2) (1968) 435.[13] J. McMurry, Fundamental of Organic Chemistry, fifth ed., Thomson Learning,

Asia Pte Ltd., 2003, p. 94.