Computational Mechanistic Study of the Julia–Kocieński Reaction

Feb 16, 2015 - except sulfur for which the 6-311+G(2df,p) basis set was used. .... Suggested Mechanistic Pathways A and B of the Julia−Kocieński Re...
0 downloads 0 Views 4MB Size
Subscriber access provided by GEORGIAN COURT UNIVERSITY

Article

Computational mechanistic study of the Julia-Kocie#ski reaction Laura Legnani, Alessio Porta, Pierluigi Caramella, Lucio Toma, Giuseppe Zanoni, and Giovanni Vidari J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.5b00008 • Publication Date (Web): 16 Feb 2015 Downloaded from http://pubs.acs.org on February 18, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Organic Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Computational mechanistic study of the Julia-Kocieński reaction Laura Legnani,* Alessio Porta, Pierluigi Caramella, Lucio Toma, Giuseppe Zanoni, Giovanni Vidari* Università di Pavia - Dipartimento di Chimica – Sezione di Chimica Organica, Via Taramelli 12, 27100 Pavia, Italy Tel: (+39) 0382987311. Fax: (+39) 0382987323. E-mail: [email protected] (L.L.); Tel: (+39) 0382987322. Fax: (+39) 0382987323. E-mail: [email protected] (G.V)

R

R'

R CHO + R' CH2SO2Ar

Abstract. This paper describes the first detailed computational mechanistic study of the JuliaKocienski olefination between acetaldehyde (1) and ethyl(1-phenyl-1H-tetrazol-5-yl)sulfone (2), considered a paradigmatic example of the reaction between unsubstituted alkyl PT sulfones and linear aliphatic aldehydes. The theoretical study was performed within the density functional approach through calculations at the B3LYP/6-311+G(d,p) level for all atoms except sulfur for which the 6-311+G(2df,p) basis set was used. All the different intermediates and transition states encountered along the reaction pathways leading to final E and Z olefins have been located and the relative energies calculated, both for the reactions with potassium and lithium-metallated sulfones, in THF and toluene, respectively. We have essentially confirmed the complex multistep mechanistic manifold proposed by others; however, the formation of a spirocyclic intermediate in the Smiles rearrangement was excluded. Instead, we found that this step involves a concerted, though asynchronous, mechanism. Moreover, our calculations nicely fit with the diastereoselectivities observed experimentally for potassium and lithium-metallated sulfones, in THF and toluene, respectively.

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction

The total synthesis of complex natural product molecules often requires the use of olefination reactions capable of joining together highly functionalized fragments to form alkene double bonds in highly regio- and stereoselective fashion. In addition, starting products must be easily obtainable and, ideally, the reaction should be atom economical and produce nontoxic products. No single method developed so far provides a universal solution to these problems.1 Direct olefinations of carbonyl compounds probably still remain the most generally applicable methods.1 They include, among others, the very well-known Wittig reaction, the Horner-Wittig and Horner-Wadsworth-Emmons (HWE) variants, and the Peterson, Johnson, and Julia olefinations.2 The last one has gained a great popularity in the last decades, thanks to different generally applicable methods for incorporating sulfone moieties into synthetic fragments which then smoothly react with counterpart carbonyl compounds. The absence of toxicity of the sulfur derivatives, and the usually high and predictable regio- and stereoselectivity of formed olefins are additional important features of this methodology. For these characteristics and the operationally simple procedures, the classical Julia reaction and the modified variants have successfully been used in the synthesis of a great number of different biologically active natural product molecules.3 The original classical Julia olefination, discovered by Marc Julia, also known as the Julia-Lythgoe olefination,4 consists of a multi-step sequence comprising the nucleophilic attack of an α-metallated aromatic sulfone to an aldehyde affording a β-hydroxy sulfone, the conversion of the hydroxyl group to a better leaving group, typically an acyloxy or a sulfonate group, and the final reductive elimination as an olefin forming step. This original procedure was later significantly improved by the so-called one-pot olefination protocol, at first developed by Sylvestre Julia and further by Kocieński, which is commonly known as the Julia-Kocieński (J-K) reaction.3f,5 Actually, these variants encompass the need of the functionalization step; instead, olefins are directly prepared in

ACS Paragon Plus Environment

Page 2 of 22

Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

one-pot, under mild reaction conditions, from carbonyl compounds and benzothiazol-2-yl sulfones (BT sulfones) or 1-phenyl-1H-tetrazol-5-yl sulfones (PT sulfones) or other heroarylalkylsulfones, upon the presumed Smiles rearrangement of an intermediate alkoxide.3c,3f Moreover, there are fewer problems with scale-up than with the classical variant, and the E/Z-selectivity can be controlled to some extent by varying the sulfonyl group, the carbonyl component, the solvent polarity, and the counterion of the base. The commonly accepted mechanistic pathways of the J-K leading to the E-olefin (A) or the Z-olefin (B) are depicted in Scheme 13c,3e,3f,6 for the reaction between acetaldehyde and ethyl PTsulfone. They comprise four main steps; i) the addition of a metallated PT-sulfone to an aldehyde to give alkoxides 3E (anti) or 3Z (syn); ii) the Smiles rearrangement of the folded conformations 3Eg and 3Zg, to give the cyclic Smiles intermediates 4E and 4Z, respectively, which undergo cleavage to the sulphinates 5Eg and 5Zg, respectively; iii) the conformational change of the conformers 5Eg and 5Zg, resulting from the Smiles rearrangement, into geometries 5Ea and 5Za, respectively, having antiperiplanar oriented sulfinate and OPT groups; iv) the antiperiplanar β-elimination of sulfur dioxide and the heterocyclic moiety through the transition states TS6E and TS6Z to deliver olefin A: E-pathway SO2PT

O

3'

TS1E Me

Me

Me

and

6E

SO2PT Me O

1

2

3E anti

N N N Ph N

4'

N N 5' 2' N N Ph 1' O O S O Me 2 3 H 1 H Me4 3Eg

SO2PT

O Me

Me

TS1Z Me

SO2PT Me O

1

2

3Z

syn

Me H H 3Zg

O SO2

TS3E Me Me

O TS4E

H H

O2S

TS3Z O Me

Me H 4Z

Me H

O Me

5Eg

S

OPT TS5Z Me

5Zg

6E

O Me

Me

Me

Me

Me

5Ea

O O2S

TS6E

OPT - SO2 - OPT

Me

N N N Ph N O SO2 TS4Z

S

OPT TS5E Me

Me

4E

B: Z-pathway N N N N Ph O O S

respectively.

6Z,

OPT 5Za

TS6Z - SO2 - OPT

Me Me 6Z

Scheme 1. Suggested mechanistic pathways A and B of the Julia-Kocienski reaction, leading to (E)but-2-ene and (Z)-but-2-ene, respectively.3c The syn and anti notations indicate the relative stereochemistry of the sulfonyl and alkoxy substituents in the adducts, while the a and g symbols

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

denote their antiperiplanar and gauche orientations, respectively. The counterion of anionic species is omitted in the scheme for the sake of clarity. In the case of aliphatic PT sulfones the first step has been demonstrated experimentally to be irreversible;3f and the final olefin ratio is then considered to be determined by the initial 3E/3Z ratio. Pairing of polar and strongly coordinating solvent with a large counterion of metallated sulfones, such as a potassium cation, enhances the stereoselectivity in favor of the E-olefin. In contrast, in less polar and coordinating solvents, such as in toluene, the amount of the Z-olefin increases significantly, thus dramatically decreasing the reaction stereoselectivity.3c,7 This effect is further enhanced when lithium is the counterion of metallated sulfones. To the best of our knowledge, this complex multistep mechanistic manifold has never been the object of in depth computational studies aimed at confirming the proposed different steps by locating the different intermediates and transition states (TSs) encountered along the reaction pathways leading to final olefins.8 Besides confirming the hypothesized mechanism, we intended to shed some light upon the dependence of the reaction stereoselectivity on the type of metal counterion and the solvent polarity. To this aim the alternative reaction pathways leading to E and Z-olefins were considered, starting from the addition to aldehyde of either potassium metallated sulfone in THF or lithium metallated sulfone in toluene. Here we report a detailed computational mechanistic study of the J-K reaction between acetaldehyde (1) and ethyl(1-phenyl-1H-tetrazol-5-yl)sulfone (2), which we considered a paradigmatic example of the reaction between unsubstituted alkyl PT sulfones and linear aliphatic aldehydes.

ACS Paragon Plus Environment

Page 4 of 22

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Results and discussion.

Modeling of the Julia-Kocieński E-pathway with potassium-metallated sulfone 2 in THF

Our computational study started by modeling the J-K reaction under the conditions highly favoring the formation of E-olefins. Experimentally, among the solvents and bases tested for the reaction between aliphatic sulfones and aliphatic aldehydes, pairing of a base having potassium as the counterion, usually KHMDS, and 1,2-dimethoxyethane (DME) as the solvent, afforded Eolefins with remarkable selectivity in all cases studied. It has been suggested that, under this condition, the big cation K+ is largely coordinated by the polar solvent, favoring a dissociated carbanion, which, in the first step of the reaction, would afford anti adduct through an open transition state, that would eventually lead to the E-olefin.3c In our studies, among the solvents implemented in Gaussian, tetrahydrofuran (THF) was selected for the modeling, since it was considered the most similar to DME, as a polar coordinating ether solvent. Moreover, K+ was selected as the base counterion and the anionic species were thus considered, on first approximation, completely dissociated from the cation.

2.56 2.40

TS1E_A 6.18 (8.43)

TS1E_B 5.70 (8.11)

TS1E_C 8.17 (10.73) 1.82 2.55

1.81

2.18

2.24

TS1E_D 4.08 (6.88) 1.81 2.56

1.82

2.78 2.49

3Eg_A -5.18 (-5.04)

3Eg_B -2.19 (-2.99)

3Eg_C -4.10 (-4.11)

ACS Paragon Plus Environment 3Ea_A 4.00 (0.88)

3Eg_D -3.13 (-3.27)

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. 3D plots of the possible conformers of TS1E and intermediate 3E in the E pathway of the Julia-Kocieński reaction of acetaldehyde with potassium metallated ethyl(1-phenyl-1H-tetrazol-5yl)sulfone in THF. The corresponding relative electronic energies (kcal/mol) in vacuo and in solvent (in parenthesis), with respect to the isolated reacting species, are reported. Bond distances (Å) are indicated on the 3D-plots.

The geometry of isolated reacting species, i.e. acetaldehyde 1 and the α-anion of the PTsulfone 2, were at first optimized; subsequently, the modeling study examined the first step, namely the addition of 2 to 1 to give the anti adduct (2R*,3S*)-3E. All the possible minimum energy conformers of the adduct 3E were located, but only the four geometries 3Eg_A-D (Figure 1), corresponding to the gauche conformer 3Eg (Scheme 1), having the anionic oxygen atom spatially close to the tetrazole ring, resulted significantly populated. The other conformers were less stable by about 10 kcal/mol in vacuo, and 6-8 kcal/mol in THF. In particular, a geometry of type 3Ea_A (Figure 1), in which the dipoles associated with the carbonyl and the sulfonyl groups are anti-oriented, was calculated to be 9.2 kcal/mol in vacuo and 5.9 kcal/mol in THF higher in energy than the most stable conformer 3Eg_A. In each of the four transition states TS1E_A-D (Figure 1), collapsing to 3Eg_A-D, respectively, the two dipoles were gauche oriented, and the least sterically crowded transition state TS1E_D was found to be preferred by about 1-4 kcal/mol over the other three pathways. However, it is worthy pointing out that the most stable conformer of 3E was calculated to be 3Eg_A, deriving from a TS about 2 kcal/mol less stable than TS1E_D. Considering the short distance between the anionic oxygen atom and the tetrazolyl carbon atom (2.49-2.78 Å) in conformers 3Eg_A-D, a quasi five-membered ring could be envisioned, showing different puckering modes and orientations of the substituents. The phenyl group of the PT moiety was trans oriented with respect to the adjacent methyl groups in 3Eg_A and 3Eg_B, while was positioned in cis fashion in 3Eg_C and 3Eg_D. ACS Paragon Plus Environment

Page 6 of 22

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Each conformation 3Eg_A-D of intermediate 3E was then the starting point of four reaction pathways converging to (E)-but-2-ene, whose energy profiles were all located (Figure 2). As a typical example, the 3D plots of the calculated TSs and intermediates for the different steps of the A profile are reported in Figure 3. The graphic clearly shows that after the initial formation of adduct 3E, the reaction readily evolves through the following steps to the final product. Noteworthy, the calculated activation energy barrier of 2-3 kcal/mol for the interconversion between conformations 3Eg_A and 3Eg_B, and between 3Eg_C and 3Eg_D, through rotation around the single bond C(2)C(3), was comparable with the activation energies of the different reaction steps. Therefore, due to the rapid conformer interconversion, the A pathway was indistinguishable from B, and the C one was indistinguishable from D. On the contrary, the conversion of 3Eg_A to 3Eg_D, and 3Eg_B to 3Eg_C, via rotation around the single bond C(1’)-S, was much more sluggish, due to an energy barrier of about 17 and 13 kcal/mol, respectively. In the suggested mechanism of the J-K reaction,3c,3f,6 intermediate 3Eg is converted to sulfinate 5Eg via an intramolecular nucleophilic aromatic substitution, the so-called Smiles reaction.9 It has been proposed that such rearrangement would occur through the so-called Smiles intermediate 4E (Scheme 1), even if evidences for both a stepwise and a concerted mechanism exist in literature for the Smiles reaction.9 Both the mechanisms were considered in our study; however, the search of an energy minimum along the reaction coordinate corresponding to the spirocyclic intermediate 4E met with no success. In fact, every starting geometry corresponding to the predictable features of the supposed Smiles intermediate collapsed either to the alkoxide 3Eg or to the sulphinate 5Eg, while the cyclic structure 4E was not located as an energy minimum. In striking contrast, the geometry TS3E was located as a low energy transition state.

ACS Paragon Plus Environment

The Journal of Organic Chemistry

10.73

10 8 6

TS1E_C

8.43 TS1E_B TS1E_D

TS1E_A 8.11 6.88

4 2 0

al/mo l)

-2

∆E(kc

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

-4 -6

TS3E_C

0.00 R 3Eg_B 3Eg_D 3Eg_C 3Eg_A

-0.25 -2.23

-2.99 -3.27 -4.11 -5.04

-3.16 -3.20

TS3E_B TS3E_D TS3E_A

TS6E_A

-32 -33.34

5Eg_A -35.54

-34 -36

TS5E_A

5Ea_A

-31.94 -34.80

-35.36

-34.80

TS5E_C -37.23

-38 -40 5Eg_C

-42

-3760

5Eg_B -40.23 -40.47

-35.02

TS5E_B

-39.57 TS5E_D

-42.65 5Eg_D

TS6E_B TS6E_D TS6E_C

5Ea_B -40.51 -40.51 -42.05

5Ea_D

5Ea_C 6E

-48

-48.96

Reaction Coordinate

Figure 2. Energy profiles A (black), B (red), C (blue), and D (green) for the E pathway of the JuliaKocieński reaction of acetaldehyde with potassium metallated ethyl(1-phenyl-1H-tetrazol-5yl)sulfone in THF.

1.87

3Eg_A

TS3E_A

5Eg_A

1.82

TS5E_A

2.13

1 2

5Ea_A

TS6E_A

6E

Figure 3. Julia-Kocieński reaction of acetaldehyde with potassium metallated ethyl(1-phenyl-1Htetrazol-5-yl)sulfone in THF: 3D plots of calculated intermediates and TSs in the A profile of the E pathway. Bond distances (Å) are reported on the 3D-plots. ACS Paragon Plus Environment

Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

The IRC analysis (Figure 4), starting from TS3E_A, was performed in both the forward and the backward directions, and confirmed this geometry for the transition state of the conversion of adduct 3Eg_A to intermediate 5Eg_A. Moreover, the accurate examination of the bond forming and breaking timing indicated that they occurred in a concerted, though asynchronous fashion. Actually, the length of 1.86 Å for the partial C-S bond in TS3E_A almost corresponded to the intact single bond (1.81 Å) in 3Eg_A, whereas the length of 1.87 Å for the partial C-O bond in TS3E_A indicated that the formation of this new bond had already progressed significantly. In conclusion, the conversion of 3Eg to 5Eg did not involve the formation of an intermediate, like the Smiles spirocyclic derivative 4E; on the contrary, the process corresponded to a fast intramolecular nucleophilic substitution occurring in a concerted, though asynchronous manner.

d(Å)

Reaction coordinate

Figure 4. Length (Å) of the forming C-O and breaking C-S bonds [d(C-O) and d(C-S), respectively] at the forward (positive) and backward (negative) points in the IRC analysis, starting from TS3E_A (point 0), of the conversion of intermediate 3Eg_A to 5Eg_A.

-0.956 -0.948 -0.936 2.119 -0.241 -0.073

0.161

-0.937 -0.945

-1.075

2.094 -0.089

-0.332 0.266

-0.753

-0.135

-0.077

-0.278 0.687

-0.436 -0.316

-0.070

-0.351 -0.578

-0.087

3Eg_A

TS3E_A ACS Paragon Plus Environment

5Eg_A

1.454 -1.034

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. NBO charges distribution in 3Eg_A, TS3E_A, and 5Eg_A.

In this context, we examined the charges distribution in 3Eg_A, TS3E_A, and 5Eg_A (Figure 5), in particular, the distribution of the negative charge. In fact, the examination of the stepwise mechanism proposed for the Smiles rearrangement (Scheme 1) clearly shows that a net transfer of negative charge should occur from the alkoxide oxygen atom in 3Eg_A to the N(5’) atom of the tetrazole ring in the Smiles intermediate 4E. In striking contrast, instead, our data showed that in the transition state TS3E_A most of the negative charge is still retained by the alkoxide oxygen atom attacking the positively charged carbon of the tetrazole ring, while it is transferred to N(5’) only marginally. Moreover, while the TS collapses to the final product 5Eg_A, the negative charge on the oxygen is mostly transferred onto the positively charged sulfur atom of the sulfonyl group, which thus becomes a good leaving group and finally detaches as a sulphinate. Thus the ability of the leaving group in stabilizing the negative charge appears to be a crucial factor to force the Smiles reaction of intermediate 3Eg to proceed through a concerted pathway. Similar considerations have been applied to other concerted nucleophilic displacement reactions at sp2 trigonal carbons, including those in (aryloxy)triazines10a and some nucleophilic acyl substitution reactions involving excellent leaving groups.10b-d We then examined the last two steps of the E-olefin formation, namely: i) the rotation of sulphinate 5Eg around the C(2)-C(3) single bond to the antiperiplanar conformation 5Ea, having the stereoelectronically required geometry for triggering the following β-elimination step; ii) the extrusion of sulfur dioxide with concerted elimination of the PT heterocyclic moiety. The rotational energy barriers TS5E between 5Eg and 5Ea was found to be low (2-3 kcal/mol) while the antiperiplanar β-elimination to the E-olefin 6E was estimated to take place through a slightly higher energy barrier, involving the transition state TS6E in which the rupture of the C-S bond was more advanced than that of the C-OPT bond.

ACS Paragon Plus Environment

Page 10 of 22

Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Two alternative mechanisms for the elimination of the sulfinate and the OPT moieties from the intermediate 5Eg to give the final olefin were also examined. In the first, the direct synperiplanar elimination of the two groups were considered, which would lead to the cis-olefin 6Z. Although stereoelectronically disfavored compared to the antiperiplanar elimination, this mechanism would not require the preliminary rotation of 5Eg around the C(2)-C(3) single bond to give 5Ea. The calculated energy barrier of the synperiplanar elimination was, however, calculated to be 8.7 kcal/mol in vacuo and 10.4 kcal/mol in THF higher than the barrier for the antiperiplanar elimination, clearly indicating that the former mechanism was an unlikely process. The other mechanism considered for the formation of the E-olefin from 5Eg was a kind of two steps E1cb-like process. Actually, we envisioned the possibility that the spontaneous departure of sulfur dioxide from 5Eg could generate a carbanion on C(2), which would then expel the OPT group to afford the final double bond. In the event, the search of an energy barrier for such process was, however, unfruitful, and no TS was located, thus excluding also this stepwise mechanism.

Modeling of the Julia-Kocieński Z-pathway with potassium-metallated sulfone 2 in THF

Our calculations were then extended to the alternative pathway of the reaction between acetaldehyde 1 and potassium-metallated PT-sulfone 2 in THF, leading to the less favored Z-olefin 6Z. At first, the addition step giving the intermediate 3Z was modeled. Only the specific arrangement represented as 3Zg, with the anionic oxygen facing the tetrazole ring, resulted significantly populated, while the other geometries were less stable of 7-15 kcal/mol in vacuo and 6-9 kcal/mol in THF. Similarly to intermediate 3Eg in the E pathway, alkoxide 3Zg existed in four different conformations (3Zg_A-D). Due to the unfavourable gauche interactions between one methyl group and the tetrazole ring, conformations 3Zg_A and 3Zg_D, though having the two methyls in an antiperiplanar orientation, were less stable than the conformations B and C having

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

gauche methyl groups. Moreover, 3Zg_A quickly converted to 3Zg_B through a low energy barrier of about 0.60 kcal/mol, both in vacuo and in THF, while the conversion of 3Zg_D to 3Zg_C was even faster. Noteworthy, adducts B and C and the corresponding TSs showed an inverted stability. In fact, although 3Zg_C was the most stable conformer of 3Zg, the corresponding transition state TS1Z_C was 1.80 kcal/mol in vacuo and 1.75 kcal/mol in THF higher in energy than the lowest energy transition state TS1Z_B, collapsing to 3Zg_B. Attempts to locate TS1Z_A failed, while a relaxed potential energy surface scan of the cleavage of the C-C bond of 3Zg_A afforded TS1Z_B.

2.38

2.23

2.34

TS1Z_C 6.42 (9.24)

TS1Z_B 4.62 (7.49)

TS1Z_D 5.67 (8.17) 1.81 1.82

1.82 2.59

3Zg_A -3.74 (-3.14)

2.46

1.82 2.34

3Zg_B -4.76 (-4.83)

2.53

3Zg_C -6.61 (-6.59)

3Zg_D -1.47 (-1.43)

Figure 6. 3D plots of the possible conformers of TS1Z and intermediate 3Zg in the Z pathway from potassium-metallated sulfone 2. The corresponding relative electronic energies (kcal/mol) in vacuo and in THF (in parenthesis), with respect to the isolated reacting species, are reported. Bond distances (Å) are reported on the 3D-plots.

The reaction then proceeded from 3Zg_B and 3Zg_C to the final product (Z)-but-2-ene through intermediates and TS analogous to the E-pathway affording (E)-but-2-ene (Figure 7). In

ACS Paragon Plus Environment

Page 13 of 22

Figure 8 the 3D plots of the transition states and intermediates in the reaction pathway proceeding from 3Zg_C are reported, as an indicative example.

8

TS1ZC

al/mo

l)

0

9.24 TS1ZB 7.49

4

∆E(kc

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

0.00 R

-2 -4

3Zg_B

-6

3Zg_C

-4.83

TS3Z_B -4.71 -4.80 TS3Z_C

-6.59

-32

TS6Z_C -32.03

-34

-33.40

TS6Z_B

-36

TS5Z_C -37.30

-38

5Zg_C -39.43

TS5Z_B -38.36

-37.59 5Za_C

-40 -41.63 5Zg_B

-42

-42.19 5Za_B

-44 -46

-46.60 6Z

Reaction Coordinate

Figure 7. Energy profiles B (red), and C (blue) for the Z pathway of the Julia-Kocieński reaction of acetaldehyde with potassium metallated ethyl(1-phenyl-1H-tetrazol-5-yl)sulfone in THF.

1.81

2.53

1.87

3Zg_C

TS3Z_C

5Zg_C

TS5Z_C

1.77

1.55

2.13 1.98

5Za_C

TS6Z_C

6Z

Figure 8. 3D plots of the intermediates and the transition states in the Z pathway of the JuliaKocieński reaction proceeding from the conformer 3Zg_C of adduct 3Zg. Bond distances (Å) are reported on the 3D-plots.

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

The 3Zg_C adduct was then rapidly converted to sulphinate 5Zg_C via the transition state TS3Z in which the formation of the new C-O bond (1.87 Å) proceeded in a concerted, however, asynchronous fashion with the breaking of the C-S bond (1.84 Å). Noteworthy, as observed for the E pathway, the existence of the presumed spirocyclic Smiles structure 4Z was not detected, likely due to the same electronic factors discussed before. Subsequently, a rapid rotation around the C(2)C(3) single bond allowed the sulfinate 5Zg_C to assume the conformation 5Za_C, in which the two methyl groups were gauche- oriented. In the final step of the pathway, the concomitant antiperiplanar β-elimination of the PTO moiety and sulfur dioxide occurred via the transition state TS6Z, affording the Z-olefin 6Z. According to the mechanistic manifold for irreversible J-K reactions between aldehydes and the α-anion of PT sulfones, such as 1 and 2, the final olefins ratio is determined by the relative rate of formation of adducts 3E and 3Z in the first step.3c Our calculations predicted an energy difference of 0.54 Kcal/mol in vacuo and 0.61 kcal/mol in THF for the transition states TS1E_D and TS1Z_B in the E and Z pathways, respectively (Figures 2 and 7), corresponding to an E/Zselectivity = 75:25 in vacuo, and 78:22 in THF at -78 °C. These values were nicely comparable with the E/Z ratio = 86:14 determined for the reaction between hexanal and potassium metallated pentylphenyltetrazolyl sulfone at -78 °C, considered as the reference reaction.7

Modeling of the Julia-Kocieński E- and Z-pathways with lithium-metallated sulfone 2 in toluene

2.05

2.06 1.97

1.91

1.93

1.77 1.87

1.85

1.96 1.76

2.45 2.25 2.60

TS1E’

2.19

TS1E”

1.99

2.14 2.09

1.66

TS1Z”

TS1Z’

2.00

2.14 2.08

1.66

1.73

1.74

ACS Paragon Plus Environment 3Eg’

3Eg”

3Zg’

3Zg”

Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Figure 9. 3D plots of the intermediates and the transition states having a di- and a tricoordinated lithium ion in the Z and E pathways of the Julia-Kocieński reaction with lithium-metallated sulfone 2 in toluene. Bond distances (Å) are reported on the 3D-plots. . A decrease in the E/Z olefins ratio has been observed experimentally for the Julia-Kocienski reactions as the counterion of metallated sulfones is changed from K+ to Li+ in an apolar, poorly coordinating solvent, for example toluene.3c It has been suggested that this significant drop in the diastereoselectivity would be due to an increase of populated chelated transition states in the first step of the J-K reaction,3c that would be largely favored by small cations in non-coordinating apolar solvents. The chelation effect would lead, preferentially, to 3Z-like syn adducts instead of 3E-like anti ones, resulting, at the end of the reaction pathway, in an increasing proportion of the Z-olefin. To test this hypothesis, at first we modeled the transition states leading to lithium chelated syn and anti adducts for the reaction between aldehyde 1 and lithium metallated sulfone 2 in toluene at -78 °C, by introducing the lithium ion in the calculations.11 Two types of transition states were found for both the E- and Z-pathways, in which lithium resulted to be tricoordinated and dicoordinated, respectively. Figure 9 shows two representative TSs for the syn addition, TS1Z’ and TS1Z”, and two for the anti addition, TS1E’ and TS1E”. In TS1Z’ and TS1E’ the cation was chelated to the N(3’) of the tetrazole ring, the alkoxy anion and one oxygen of the sulfone moiety, while in TS1Z” and TS1E” the lithium was only chelated to N(3’) and the alkoxy anion (Figure 9). The energy values (see SI) allowed to calculate energy barriers corresponding to a final ratio of (E) and (Z)-but-2-ene of 10:90 for the tricoordinated lithium pathway and 51:49 for the less favored dicoordinated one. Thus, the modeling clearly indicated the major role played by the lithium coordination in apolar solvents in favoring the formation of Z-olefins in the J-K reaction, though the calculated E/Z ratio did not exactly reproduce the stereoselectivity found experimentally,7 due to an overextimated proportion of the Z-olefin. Besides to an incomplete modeling accuracy, this

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

discrepancy may be simply due to the presence of competing transition states, in which anionic species are loosely coordinated to lithium. Actually, these transition states would resemble those modeled with potassium as the counterion (see above), which would preferentially afford (E)-but-2ene. 1.97 1.92

1.92 2.16

1.73

2.16

1.97

1.97 1.99

1.73

TS3E’

TS3Z’

5Ea’

5Zg’

Figure 10. 3D plots of TS3E’, TS3Z’ and their sulfinates 5 coordinated to lithium ion in the E and Z pathways of the Julia-Kocieński reaction. Distances between lithium and coordinated atoms (Å) are reported on the 3D-plots. The four transition states TS1Z’, TS1E’, TS1Z”, and TS1E” collapsed to the two lithiumtricoordinated adducts 3Zg’ and 3Eg’, and the two metal-dicoordinated adducts 3Zg” and 3Eg”, respectively. As expected, the tricoordinated species were more stable than the corresponding dicoordinated ones. It was then interesting to explore whether, according to the general mechanistic manifold suggested for the J-K reaction (Scheme 1),3c,3f,6 lithium chelated intermediates would convert to the corresponding sulfinates through the corresponding Smiles intermediates. As found for the reaction pathway with potassium-metallated sulfones, modeled above, the search of an energy minimum along the reaction coordinate corresponding to spirocyclic intermediates-like 4E and 4Z met with no success. In fact, these cyclic structures were not located as energy minima. In striking contrast, and quite remarkably, both 3Zg’ and 3Zg” converged to the same sulfinate 5Zg’ via a concerted asynchronous mechanism involving the same transition state TS3Z’ (Figure 10). Analogously, in the E pathway, the corresponding di- and tricoordinated 3E adducts converged to the same 5Ea’

ACS Paragon Plus Environment

Page 17 of 22

sulfinate through an identical transition state (Figure 10). For the two TSs, in which the new C-O bond was only partially formed, similarly chelated tricoordinated structures were observed (Figure 10). Conversely, sulfinates 5Ea’ and 5Zg’, in which the C-O bond was completely formed, existed as dicoordinated structures, having the lithium ion linked only to N(3’) and one sulfone oxygen (Figure 10).

Pathway A 18 TS1E"

17.81

16 14 12

TS1E' 11.12

10 8 4 0 -2 -4

cal/m ol)

-6

∆E ( k

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

TS3E' R"

8.39

6.23 R' 0.00 3Eg" -4.06 -4.27 3Eg'

-8

-24 TS6E'

-26

-26.60

-28 -32

-33.49

-34

5Ea'

-40 -44 -46 -48.23 6E'

Reaction Coordinate

ACS Paragon Plus Environment

The Journal of Organic Chemistry

Pathway Ba 18 TS1Z"

17.58

16 14 12 10 8 4 0 -2 -4

cal/m o

l)

-6

∆E ( k

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

TS1Z' 9.12 TS3Z' R"

6.91

5.85 R' 0.00 3Zg" -3.62 -4.57 3Zg'

-8 TS6Z' -24.06

-24 -26 -28

-28.34 5Zg'

-34

a -36.19

-36

5Za'

-38

-36.82 6Z'

Reaction Coordinate

Figure 11. Energy profiles for the Julia-Kocieński reaction of acetaldehyde with lithium-metallated sulfone 2 in toluene, leading to (E)-but-2-ene (pathway A) and (Z)-but-2-ene (pathway B), respectively. aAll attempts to locate TS5Z’ in pathway B failed, likely due to the rapid conversion of 5Zg’ to 5Za’.

The reaction then proceeded uneventfully from intermediates 5Ea’ and 5Zg’ to the final olefins 6E and 6Z, respectively, as highlighted by the energy pathways A and B depicted in Figure 11. Interestingly, in 5Ea’ the PTO and the sulfinate moieties were already positioned in the antiperiplanar orientation required for triggering the subsequent β-elimination step, while 5Zg’ attained such geometry through fast rotation around the single bond C(2)-C(3).

ACS Paragon Plus Environment

Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

Conclusions The Julia-Kocieński reaction between heterocyclic sulfones and aldehydes is one of the most popular procedure to obtain substituted olefins with usual excellent E-selectivity.2,3 In this paper we have reported the first detailed mechanistic study of the J-K reaction between acetaldehyde (1) and metallated ethyl(1-phenyl-1H-tetrazol-5-yl)sulfone (2), which we considered a paradigmatic example of the reaction between unsubstituted alkyl PT sulfones and linear aliphatic aldehydes. The theoretical study was performed within the density functional approach through calculations at the B3LYP/6-311+G(d,p) level for all atoms except sulfur for which the 6-311+G(2df,p) basis set was used.12 Moreover, single-point calculations by using a solvent polarizable continuum model13 were performed to take solvent effects into account. All the different intermediates and transition states encountered along the reaction pathways leading to final (E)-but-2-ene and (Z)-but-2-ene have been located and the relative energies have been calculated, both for the reaction with potassiummetallated sulfone in THF and with lithium-metallated sulfone in toluene. We have essentially confirmed the complex multistep mechanistic manifold proposed by others,3c,3f,6 with the remarkable exception of the formation of the intermediate Smiles spirocyclic derivatives 4E and 4Z. Actually, these species have been proposed as energy minimum intermediates in the conversion of the initial adducts 3E and 3Z to sulfinates 5E and 5Z, respectively.3c,3f,6 On the contrary, we found that the formation of the new C-O bond proceeded in a concerted, however, asynchronous fashion with the breaking of the C-S bond, so that the process corresponded to a fast intramolecular nucleophilic aromatic reaction proceeding in a concerted, though asynchronous fashion. Moreover, our calculations nicely fit with the experimental data7 showing that metal-free sulfone anions, such as those formed by metallation with a potassium base in polar and coordinating solvent, such as THF, enhances the reaction E-diastereoselectivity. In contrast, with lithium as sulfone anion counterion in a less polar and coordinating solvent, such as toluene, the amount of the Z-olefin was found to increase significantly, resulting in a dramatic decrease of the reaction stereoselectivity.

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Our modeling study will now be extended to more complex substrates, including the JuliaKocieński reactions between substituted sulfones and aldehydes. Our findings will be reported in due time.

Experimental Section

Computational Methods All the calculations were carried out using the GAUSSIAN09 program package.14 All the structures of reactants, transition states, intermediates, and products were optimized in the gasphase at the B3LYP/6-311+G(2df,p) level for the S atom and 6-311+G(d,p) level12 for the other atoms to correctly describe geometries and electronic properties of compounds containing a sulfur atom, belonging to the third period. The reaction pathways were confirmed by IRC analyses performed at the same level as above. Vibrational frequencies were computed at the same level of theory to define the optimized structures as minima or transition states, which present an imaginary frequency corresponding to the forming bonds. Thermodynamics at 298.15 K allowed the enthalpies and the Gibbs free energies to be calculated. The solvent effects were considered by single-point calculations, at the same level as above, on the gas-phase optimized geometries, using a self-consistent reaction field (SCRF) method, based on the polarizable continuum model (PCM).13 NBO analysis was performed at the same level of calculations.15

Acknowledgments Prodest scarl is warmly thanked for a postdoctoral fellowship to L.L. The Italian MIUR is acknowledged for financial support (funds PRIN 2010-2011).

Supporting Information Available: B3LYP/6-311+G(d,p) (for C, H, O, N, Li atoms) and 6311+G(2df,p) level (for S atom) Energies and Cartesian Coordinates of all intermediates and

ACS Paragon Plus Environment

Page 20 of 22

Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Organic Chemistry

transition states along the reaction pathways. This material is available free of charge via the Internet at http://pubs.acs.org

References and Footnotes 1. (a) Preparation of Alkenes: A Practical Approach; Williams, J. M. J., Ed.; Oxford University Press: Oxford, 1996. (b) Kelly, S. E. Alkene Synthesis. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991, Vol. 1, pp. 729–817. 2. (a) Kürti, L.; Czakó, B. Strategic Applications of Named Reactions in Organic Synthesis; Elsevier Academic Press: Burlington, USA, 2005. (b) Johnson, C. R.; Shanklin, J. R.; Kirchhoff, R. A. J. Am. Chem. Soc. 1973, 95, 6462–6463. 3. (a) Chatterjee, B.; Bera, S.; Mondal, D. Tetrahedron: Asymmetry 2014, 25, 1−55. (b) Lorente, A.; Albericio, F.; Álvarez, M. J. Org. Chem., 2014, 79 , 10648–10654. (c) Aïssa, C. Eur. J. Org. Chem. 2009, 1831–1844. (d) Zanoni, G.; Brunoldi E.M.; Porta, A.; Vidari, G. J. Org. Chem. 2007, 72, 9698–9703. (e) Plesniak, K.; Zarecki, A.; Wicha, J. Top. Curr. Chem., 2007, 275, 163–250. (f) Blakemore, P. R. J. Chem. Soc., Perkin Trans I, 2002, 2563–2585. (g) Nájera, C.; Yus, M. Tetrahedron 1999, 55, 10547-10658. 4. Julia, M.; Paris, J.-M. Tetrahedron Lett. 1973, 14, 4833–4836. 5. Baudin, J. B.; Hareau, G.; Julia, S. A.; Ruel, O. Tetrahedron Lett. 1991, 32, 1175−1178. 6. Baudin, J. B.; Hareau, G.; Julia, S. A.; Lorne, R.; Ruel, O. Bull. Soc. Chim. Fr. 1993, 130, 856−878. 7. Blakemore, P. R.; Cole, W. J.; Kocieński, P. J.; Morley, A. Synlett 1998, 26−28. 8. For computational mechanistic studies for the fluoro-Julia-Kocienski olefination of aldehydes with BTFP sulfones, see: Alonso, D.A.; Fuensanta, M.; Gómez-Bengoa, E.; Nájera, C. Adv. Synth. Catal. 2008, 350, 1823−1829. 9. Truce, W. E.: Kreider, E. M.; Brand, W. W. Org. Reactions 1970, 18, 99–215.

ACS Paragon Plus Environment

The Journal of Organic Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

10. (a) Hunter, A.; Renfrew, M.; Rettura, D.; Taylor, J. A.; Whitmore, J. M. J.; Williams, A. J. Am. Chem. Soc. 1995, 117, 5484−5491. (b) Concerted organic and bio-organic mechanisms; Williams, A., CRC Press: Boca Baton, FL, 2000. (c) Williams, A. Chem. Soc. Rev., 1994, 23, 93−100. (d) Williams, A. Acc. Chem. Res., 1989, 22, 387−392. 11. The solvation sphere of the lithium ion was not considered to simplify the computational studies. 12. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789. 13. (a) Cancés, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3032−3042. (b) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem. Phys. Lett. 1996, 255, 327−335. (c) Barone, V.; Cossi, M.; Tomasi, J. J. Comput. Chem. 1998, 19, 404−417. 14. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision B.01; Gaussian, Inc., Wallingford, CT, 2010. 15. (a) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211-7218. (b) Reed, A. E., Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735-746. ACS Paragon Plus Environment

Page 22 of 22