Sulfonyl-stabilized allylic norbornenyl and norbornyl carbanions: Structure and stereoselectivity of reaction with electrophiles

FULL PAPER

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions:

Structure and Stereoselectivity of Reaction with Electrophiles

Hans-Joachim Gais,*^[a]Markus van Gumpel,^[a]Gerhard Raabe,^[a]Jürgen Müller,^[b]

Siegmar Braun,^[c]Hans Jörg Lindner,^[c]Susanne Rohs,^[a]and Jan Runsink^[a]

Keywords: Norbornenyl carbanions / Norbornyl carbanions / Lithioallyl sulfones / X-ray crystal structures / NMR

spectroscopy / Ab initio calculations / Stereoselective alkylation / Aggregation

The structures of the norbornenyl and norbornyl sulfones

exo-5, endo-5 and endo-6 have been determined

experimentally, by X-ray analysis, and theoretically by ab

initio calculations (HF/6–31+G*). X-ray crystal structure

more stable than the exo anions. There is good agreement

between the optimized structures of the free anions and the

experimentally determined structures of the anions of the

contact ion pairs in the crystal. Reactions of 3 and 4 with DX,

analyses of the lithiated allylic norbornenyl and norbornyl MeI, EtI, nPrI and nHeI occurred at the C2 atom under the

sulfones endo-3/ent-endo-3·2diglyme and endo-4/ent-endo-

4·2diglyme revealed dimeric O–Li contact ion pairs devoid of

C–Li bonds. The anions of endo-3/ent-endo-3·2diglyme and

selective formation of the corresponding endosulfones endo-

8a–e and endo-9a–e, respectively, in all cases. Thus, an

earlier report on the selective formation of the exosulfone

exo-9b in the reaction of 4 with MeI has to be revised.

Product ratios were independent of the configuration of the

starting sulfones and varied with the nature of the

electrophile. Selectivities were highest in the case of the

norbornyl species 4. Reaction of 3 with PhCHO occurred at

the α position (C2) to afford the alcohols endo-8f and exo-8f

endo-4/ent-endo-4·2diglyme

adopt

both

the

endo

conformation (C2–S) and are characterized by in the exo

direction pyramidalized anionic C atoms. The degree of

pyramidalization of the C2 atom of 3 is higher than that of 4.

Ab initio optimizations (HF/6–31+G*) of the structures of the

anions of methylenenorbornene I and methylenenorbornane

II resulted in local minima featuring non-planar C2 atoms (88:12) as single diastereomers and at higher temperatures

which are pyramidalized in the exo direction in both cases,

but to different degrees. In both cases cryoscopy of 3 and 4

in THF at –108.5 °C revealed approximately 1:1 mixtures of

monomers and dimers. The sulfones exo-5, endo-5, exo-6

and endo-6 as well as the lithiosulfones 3 and 4 were studied

by NMR spectroscopy. ¹H-NMR (400 MHz), ¹³C-NMR (100

MHz) and ⁶Li-NMR (44 MHz) spectroscopy of 3 and 4 at –100

°C in [D]₈THF revealed in each case only one set of signals,

independent of the configuration of the starting sulfones.

This indicates in both cases that attainment of both the

monomer-dimer and the endo/exo equilibria of 3 and 4 is fast

on the NMR time scale. According to ⁶Li{¹H}- and ¹H{¹H}-

NOE experiments of 3 and 4 the monomeric and dimeric

species endo-3 and endo-4, having endo anions, seem to be

preferred in THF solution. Ab initio calculations of the anions

at the γ position (C8), whereas reaction of 4 with PhCHO

took place at the γ position even at low temperatures.

Methylation of endo-5 and exo-5 at –105 °C by both the

stepwise method and by the in situ method gave different

ratios of exo- and endo-methylation products. The

selectivities of reaction of 3 and 4 with electrophiles have

been rationalized by the Curtin-Hammett/Winstein-Holness

concepts. It is proposed that endo-3 (endo-4) and exo-3 (exo-

4) are conformationally labile on the time scale defined by

the rate of their reactions with electrophiles, and are attacked

by electrophiles with high selectivities from the exo and the

endo face, respectively, because of the shielding by the

phenyl group and the direction of the pyramidalization of the

anionic C atom. Preferential formation of the endo sulfones

is thus ascribed to exo attack of electrophiles on endo-3

of 3 and 4 resulted in structures endo-3(–Li⁺), exo-3(–Li⁺), (endo-4) being faster than endo attack on exo-3 (exo-4)

endo-4(–Li⁺) and exo-4(–Li⁺) (HF/6–31+G*), whose atomic

point charges were calculated by the method of Kollman et

al. The C2 atoms of endo-3(–Li⁺) and endo-4(–Li⁺) are pyramidalized, to be more reactive than 4. The base-

because of Houk’s staggering effect. Methylation of 3 and 4

by the in situ method showed 3, whose C2 atom is stronger

pyramidalized in the exo direction whereas the C2 atoms of

exo-3(–Li⁺) and exo-4(–Li⁺) are pyramidalized in the endo

direction. According to the calculations, the endo anions are

catalyzed H/D exchange of sulfones endo-5, exo-5, endo-6

and exo-6 with NaOCD₃in CD₃OD proceeded in all cases

with a high degree of retention of configuration.

[a]

Institut für Organische Chemie der Rheinisch-Westfälischen

Introduction

Technischen Hochschule Aachen,

Professor-Pirlet-Straße 1, D-52056 Aachen, Germany

Fax: (internat.) ϩ 49(0)241/8888665

E-mail: Gais@RWTH-Aachen.de (H.-J. G.)

gk016ra@cluster.rz.rwth-aachen.de (G. R.)

Norbornenyl and norbornyl carbanions of type 1 (Scheme

1), carrying a stabilizing functional group in the 2 position,

have attracted much interest for both synthetic and stereo-

chemical reasons.^[1Ϫ8]Reactions of 1 with electrophiles gen-

erally proceed in a highly exo-selective fashion (product has

the functional group in the endo position) irrespective of

the configuration of the corresponding carbon acid and the

nature of the functional group. The origin of the exo selec-

tivity and its configurational invariance are, however, only

[b]

Institut für Anorganische Chemie der Universität Basel,

Spitalstraße 51, CH-4056 Basel, Switzerland

Fax: (internat.) ϩ 41(0)61/2671018

E-mail: Muellerju@ubaclu.unibas.ch (J. M.)

[c]

Institut für Organische Chemie der Technischen Universität

Darmstadt,

Petersenstraße 22, D-64287 Darmstadt, Germany

Fax: (internat.) ϩ 49(0)6151/165591

E-mail: Lindner@oc1.oc.chemie.tu-darmstadt.de (H. J. L.)

d11k@hrzpub.tu-darmstadt.de (S. B.)

Eur. J. Org. Chem. 1999, 1627Ϫ1651

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

1434Ϫ193X/99/0707Ϫ1627 $ 17.50ϩ.50/0

1627

H.-J. Gais et al.

FULL PAPER

tially yielded the corresponding exo sulfone. This was ra-

tionalized by proposing that 3 and 4 adopt opposing con-

formations of type endo-2 and exo-2, respectively, featuring

CϪLi and OϪLi bonds. It remained unclear, however, as

to why 3 and 4 should exhibit such a structural divergence.

Second, lithioallyl sulfones have gained considerable im-

portance in organic synthesis.^[13][14]However, only little is

known about their structure.^[15Ϫ18]It is particularly desir-

able to have knowledge of the mode of coordination be-

tween the Li atom and the allylic α-sulfonyl carbanion in

the contact ion pair in order to gain a better understanding

of the high regioselectivity of their reaction with electro-

philes. X-ray crystal structure analysis of an acyclic lithio-

allyl sulfone had revealed a coordination of the Li atom

only by the O atom but not by the C atoms of the allylic

moiety.^[16]However, because of the syn position of the sul-

fonyl group and the exocyclic double bond in 3 and 4, a

coordination of the Li atom not only by the O atom but

also by the allylic moieties might be enforced, and thus be

amenable to scrutiny. Thus, it was hoped that an investi-

gation of 3 and 4 would not only help to answer the afore-

mentioned stereochemical questions concerning 2, but also

yield information as to the structure of lithioallyl sulfones

in general.

In this paper efforts to elucidate the structure and reacti-

vity of 3 and 4 by using sulfones exo-5, endo-5, endo-6 and

exo-6^[19]as the starting carbon acid^[20]are described.

Results and Discussion

Scheme 1. Sulfonyl-stabilized 2-norbornenyl and 2-norbornyl

anions

1. Structure of the Sulfones

We began the structural investigation of 3 and 4 by the

determination of the crystal structures of sulfones exo-5,

endo-5 and endo-6 by X-ray analysis (see Figures

1Ϫ3).^[21][22]The bonding parameters of exo-5, endo-5 and

endo-6 are in the normal range and show no peculiarities^[23]

(Table 1). Presumably due to steric reasons, the phenylsul-

fonyl group adopts a C2ϪS conformation in which the

phenyl group is anti to the exocyclic double bond in all

three sulfones, as revealed by the values of the torsion angle

C9ϪSϪC2ϪC3.^[23]Intuitively one assumes that due to less

steric crowding the exo isomers of 5 and 6 are somewhat

more stable than the corresponding endo isomers. To esti-

mate the relative stability of the exo and endo isomers of 5

poorly understood. This seems to be at least in part due to

the scarcity of information concerning the structure of 1,^[9]

which is in sharp contrast to the wealth of knowledge on

the structure of norbornenyl and norbornyl cations.^[10]

A

particularly stereochemically interesting subclass of 1 en-

compasses the α-sulfonyl carbanions 2^[6]which, in princi-

pal, can adopt the two conformations: endo-2 and exo-

2.^[11,12aϪb]The possible existence of diastereomers endo-2

and exo-2 raises several pertinent stereochemical questions.

(1) Which height do the barriers for the endo/exo equili-

bration have, and what is the position of the endo/exo equi-

librium? (2) How do the sulfonyl group and the bicyclic ring

skeleton exert control over the reactivity of the dia-

stereomers towards electrophiles and the interplay between

that and the stereochemical control ? (3) Are the anionic C

atoms of the diastereomers planar? (4) If not, in which di-

rection are the anionic C atoms pyramidalized? In order to

obtain more specific information on these questions the al-

lylic carbanions 3 and 4 have been studied, which were

selected for several reasons. Trost et al. had reported on a

stereochemical dichotomy in the methylation of 3 and 4.^[6e]

While reaction of the norbornenyl species 3 with methyl

iodide gave the corresponding endo sulfone preferentially,

that of the norbornyl species 4 with methyl iodide preferen- Figure 1. Crystal structure of exo-5

1628 Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

together with the corresponding experimental results for the

molecules in the crystal. At this level of theory the endo

isomers of 5 and 6 are 1.7 kcal/mol and 1.9 kcal/mol higher

in energy than the corresponding exo isomers. When corre-

lation energy calculated by means of Møller-Plesset pertur-

bation theory to the second order (MP2) is added in single

point calculations (MP2/6Ϫ31ϩG*//HF/6Ϫ31ϩG*) the en-

ergy difference is slightly reduced to 1.5 kcal/mol for 5 and

1.7 kcal/mol for 6. The calculated values for the energy dif-

ference between sulfones endo-5 and exo-5 as well as sul-

fones endo-6 and exo-6 in the gas-phase compare favorably

with the results of a base-catalyzed equilibration of the sul-

fones, which in the case of the norbornenyl sulfones gave a

mixture of exo-5 and endo-5 in a ratio of 76:24, and in that

of the norbornyl sulfones a mixture of exo-6 and endo-6 in

a ratio of 86:14 (vide infra).

Figure 2. Crystal structure of endo-5

On the basis of the unequivocal determination of the

configurations of exo-5, endo-5 and endo-6 by X-ray struc-

ture analysis, a complete assignment of the signals in the

NMR spectra of exo-5, endo-5, endo-6 and exo-6 was ac-

complished by NOE experiments and two-dimensional

methods (H,H-COSY, C,H-COSY) (Tables 2Ϫ4). The signs

of the ∆δ (¹H) values on going from the endo to the exo

isomers are the same for the norbornenyl system 5 and the

norbornyl system 6. Because of the anisotropy effect of the

SO₂group, the signal for 7-H_synof exo-6 is significantly

shifted downfield as compared to that of endo-6 whereas it

is the signal of 6-H_endoof endo-6 which is shifted downfield

Figure 3. Crystal structure of endo-6

as compared to that of exo-6.^[6f,6h,24]Similar differences ex-

1

and 6, and because of structural reasons (vide infra), com- ist in the H-NMR spectra of exo-5 and endo-5 in regard

plete geometry optimizations for the four sulfones at the to the signals of 7-H_syn. Thus, in solution the phenylsulfonyl

one-determinant (Hartree-Fock, HF) level of ab initio the- group in exo-5, endo-5, exo-6 and endo-6 most likely prefer-

ory with the 6Ϫ31G* basis set were performed. Some opti- entially adopts a similar C2ϪS conformation as in the crys-

mized structural parameters of interest are listed in Table 1 tal.

˚

Table 1. Selected bond lengths [A], bond angles [°] and absolute values of dihedral angles [°] of sulfones exo-5, endo-5 and endo-6 as well

as of lithiosulfones endo-3/ent-endo-3·2diglyme and endo-4/ent-endo-4·2diglyme; the numbers in square brackets are structural parameters

calculated at the HF/6Ϫ31G* level

exo-5

endo-5

endo-3/

ent-endo-

3·2diglyme

endo-6

exo-6^[a]

endo-4/

ent-endo-

4·2diglyme

SϪC2

1.794(4) [1.802]

1.769(4) [1.776]

1.444(3) [1.439]

1.438(3) [1.440]

1.532(6) [1.535]

1.302(7) [1.316]

1.324(7) [1.322]

Ϫ

1.790(6) [1.798]

1.782(6) [1.776]

1.451(5) [1.440]

1.436(4) [1.436]

1.521(9) [1.534]

1.33(1) [1.316]

1.31(1) [1.321]

Ϫ

1.649(7)

1.783(4)

1.444(3)

1.453(3)

1.440(9)

1.320(8)

1.28(1)

1.776(2) [1.799]

1.761(3) [1.775]

1.436(2) [1.440]

1.442(2) [1.438]

1.530(4) [1.534]

1.319(4) [1.317]

1.550(5) [1.555]

Ϫ

[1.803]

[1.776]

[1.439]

[1.440]

[1.534]

[1.317]

[1.557]

Ϫ

1.633(3)

1.778(3)

1.443(2)

1.452(2)

1.434(4)

1.333(6)

1.483(6)

1.916(5)

1.930(5)

107.0(3)

122.7(2)

125.5(3)

102.2(3)

132.9(4)

116.8(1)

73.0(3)

SϪC9

SϪO1

SϪO2

C2ϪC3

C3ϪC8

C5ϪC6

O1ϪLi1

1.913(7)

1.935(7)

107.5(5)

119.1(5)

123.2(3)

102.1(5)

132.4(7)

116.7(2)

67.2(4)

O2ϪLi1

Ϫ

C1ϪC2ϪC3

C1ϪC2ϪS

C3ϪC2ϪS

C2ϪC3ϪC4

C2ϪC3ϪC8

OϪSϪO

101.7(3) [101.6]

115.0(3) [115.3]

112.0(3) [112.9]

104.7(4) [104.8]

128.0(4) [127.7]

118.2(2) [120.1]

59.0(3) [64.8]

174.4(3) [179.0]

55.1(6) [54.1]

102.2(5) [102.0]

116.9(4) [117.3]

112.7(4) [115.1]

105.3(5) [104.5]

127.7(6) [128.0]

119.3(3) [120.1]

59.0(5) [62.7]

177.0(4) [177.3]

51.3(8) [49.4]

102.5(2) [102.4]

118.6(2) [119.1]

113.9(2) [114.0]

104.5(2) [104.9]

127.4(3) [127.6]

118.3(1) [119.7]

46.9(2) [56.3]

167.6(2) [177.4]

43.5(3) [48.1]

[101.9]

[115.1]

[112.7]

[105.3]

[127.2]

[120.1]

[64.9]

[178.8]

[55.6]

C9ϪSϪC2ϪC1

C9ϪSϪC2ϪC3

SϪC2ϪC3ϪC8

73.8(4)

31.6(9)

79.5(3)

17.9(6)

[a]

No experimental data available: See text.

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1629

H.-J. Gais et al.

FULL PAPER

1

Table 2. H-NMR data (δ_H, ppm) of sulfones exo-5, endo-5, exo-6 and endo-6 in CDCl₃as well as of lithiosulfones 3 and 4 in [D]₈THF

at 25°C

[a]

exo-5

endo-5

3

∆δ₁/∆δ₂

exo-6

endo-6

4

∆δ₁/∆δ₂

H-1

3.16

3.50

3.28

6.28

Ϫ

3.03

4.20

3.29

6.13

Ϫ

3.40

Ϫ

2.97

5.66

Ϫ

ϩ0.24/ϩ0.37

Ϫ

Ϫ0.31/Ϫ0.32

Ϫ0.62/Ϫ0.47

Ϫ

Ϫ0.29/Ϫ0.26

Ϫ

Ϫ0.59/0

Ϫ0.21/Ϫ0.34

Ϫ0.87/Ϫ1.08

Ϫ1.13/Ϫ1.11

Ϫ0.33/Ϫ0.29

Ϫ0.45/Ϫ0.42

Ϫ0.54/Ϫ0.52

Ϫ

2.64

3.59

2.79

1.65

1.32

1.65

1.22

1.85

1.15

5.14

5.19

7.93

7.56

7.66

Ϫ

2.37

4.03

2.84

1.75

1.57

1.40

2.37

1.39

5.17

5.14

7.93

7.57

7.64

Ϫ

2.92

Ϫ

ϩ0.18/ϩ0.55

Ϫ

H-2

H-4

2.51

1.42

1.00

1.42

1.27

1.00

4.00

3.70

7.75

7.15

3.28

3.45/3.54

Ϫ0.28/Ϫ0.33

Ϫ0.23/Ϫ0.33

Ϫ0.32/Ϫ0.57

Ϫ0.13/ϩ0.02

ϩ0.05/Ϫ1.10

Ϫ0.58/Ϫ0.12

Ϫ0.15/Ϫ0.39

Ϫ1.14/Ϫ1.17

Ϫ1.49/Ϫ1.44

Ϫ0.18/Ϫ0.18

Ϫ0.41/Ϫ0.42

Ϫ0.51/Ϫ0.59

Ϫ

H_exo-5

H_endo-5

H_exo-6

H_endo-6

H_syn-7

H_anti-7

H_syn-8

H_anti-8

H-10/H-14

H-11/H-13

H-12

6.09

Ϫ

6.06

Ϫ

5.80

Ϫ

2.03

1.50

5.12

5.27

7.93

7.58

7.67

Ϫ

1.44

1.63

5.33

5.25

7.89

7.55

7.65

Ϫ

1.44

1.29

4.25

4.14

7.60

7.13

3.28

3.45/3.54

OCH₃

OCH₂

Ϫ

[a]

∆δ₁ϭ δ_{lithiosulfone}Ϫ δ_{exo sulfone}; ∆δ₂ϭ δ_{lithiosulfone}Ϫ δ_{endo sulfone}

.

Table 3. ¹H-NMR data (J [Hz]) of sulfones exo-5 and endo-5 in 2. Structure of the Lithiosulfones

CDCl₃as well as of lithiosulfone 3 in [D]₈THF at 25°C

a. Structure in the Crystal

exo-5

endo-5

3

Deprotonation of exo-5, endo-5, endo-6 and exo-6 with

nPrLi in ether and recrystallization of the solids formed

from diglyme gave the respective lithiosulfones 3 and 4

(each one containing a diglyme molecule per formula unit)

in yields of 71% and 81%, respectively. X-ray structure

analysis^[21]revealed the heterochiral dimers endo-3/ent-

endo-3·2diglyme and endo-4/ent-endo-4·2diglyme (Figures 4

and 5) whose Li atoms are bound to the sulfonyl O atoms

but not to the allylic moieties, despite a favorable s-cis ar-

rangement of the C2ϪS and the C3ϪC8 bonds.

The anions of endo-3/ent-endo-3·2diglyme and endo-4/

ent-endo-4·2diglyme both adopt a C2ϪS conformation in

which the anionic lone pair orbital is periplanar to the

SϪPh bond. X-ray crystal structure analysis of a number of

lithiosulfones^{[11,16,25,26]}had revealed similar dimeric OϪLi

contact ion pairs whose anions feature such a confor-

mation, which is mainly due to a stabilizing hyperconjug-

H-1,H-2

0

2.2

1.5

0.8

2.7

1.8

0.8

0

1.9

2.2

3.2

0.5

1.7

5.5

0.7

8.7

0

Ϫ

H-1,H-4

1.5

0.5

3.1

1.6

0.7

1.6

1.5

1.6

3.1

0.5

1.6

5.5

0.5

9.0

0

1.6

0.8

2.4

1.6

1.9

0

Ϫ

3.1

0.5

1.7

5.1

0.4

6.2

2.7

H-1,H-5

H-1,H-6

H-1,H_syn-7

H-1,H_anti-7

H-1,H_anti-8

H-2,H_anti-7

H-2,H_syn-8

H-2,H_anti-8

H-4,H-5

H-4,H-6

H-4,H_syn-7

H-4,H_anti-7

H-5,H-6

H-5,H_syn-7

H-6,H_syn-7

H_syn-7,H_anti-7

H_syn-8,H_anti-8

Table 4. ¹³C-NMR data (δ_C) of sulfones exo-5, endo-5, exo-6 and endo-6 in CDCl₃as well as of lithiosulfones 3 and in [D]₈THF at 25°C

[a]

exo-5

endo-5

3

∆δ₁/∆δ₂

exo-6

endo-6

4

∆δ₁/∆δ₂

C-1

45.7

67.4

142.8

50.6

139.9

135.6

46.3

111.1

139.0

128.8

129.2

133.7

Ϫ

45.4

68.9

143.1

52.3

135.0

133.0

49.8

109.9

139.4

128.6

129.2

133.6

Ϫ

49.1

ϩ3.4/ϩ3.7

ϩ3.3/ϩ1.8

ϩ11.2/ϩ10.9

ϩ5.6/ϩ4.9

Ϫ9.0/Ϫ4.1

ϩ2.1/ϩ4.7

ϩ8.0/ϩ5.1

Ϫ27.3/Ϫ26.1

ϩ10.4/ϩ10.0

Ϫ3.7/Ϫ3.5

Ϫ1.3/Ϫ1.3

Ϫ5.1/Ϫ5.0

Ϫ

40.2

70.7

146.6

45.1

29.6

28.3

35.8

110.0

138.8

128.9

133.4

Ϫ

40.5

69.4

146.0

47.0

28.1

22.8

39.0

108.4

140.2

128.2

129.2

133.5

Ϫ

43.9

ϩ3.7/ϩ3.4

ϩ1.8/ϩ3.1

ϩ10.2/ϩ10.8

ϩ6.5/ϩ5.6

ϩ0.9/ϩ2.4

ϩ3.4/ϩ8.9

ϩ6.9/ϩ3.7

Ϫ32.7/Ϫ31.1

ϩ12.1/ϩ10.7

Ϫ4.2/Ϫ3.5

Ϫ0.9/Ϫ1.1

Ϫ4.8/Ϫ4.9

Ϫ

C-2

70.7

72.5

C-3

154.0

57.2

156.8

51.6

C-4

C-5

130.9

137.7

54.9

30.5

C-6

31.7

C-7

42.7

C-8

83.8

77.3

C-9

149.4

125.1

127.9

128.6

59.0

150.9

124.7

128.1

128.6

59.0

C-10/C-14

C-11/C-13

C-12

OCH₃

OCH₂

Ϫ

71.3/72.9

Ϫ

71.1/72.9

Ϫ

[a]

∆δ₁ϭ δ_{lithiosulfone}Ϫ δ_{exo sulfone}; ∆δ₂ϭ δ_{lithiosulfone}Ϫ δ_{endo sulfone}

.

1630

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

action between the negative charge at C2 and the positively

charged S atom, as well as by a hyperconjugative n_C-σ*_SPh

interaction. According to experiment and theoretical calcu-

lations Coulomb interaction, hyperconjugation and polari-

zation are the mechanisms (in this order) by which a sul-

fonyl group stabilizes a negative charge.^[11,25Ϫ29]The short-

ening of the C2ϪC3 bond and the lengthening of the

C3ϪC8 bond would be compatible with both an allylic

delocalization of the negative charge and a Coulomb inter-

action stemming from a polarization of the C3ϪC8 double

bond by the negative charge at C2 (vide infra).

The S atoms of endo-3/ent-endo-3·2diglyme and endo-4/

ent-endo-4·2diglyme are bent out of the C1ϪC2ϪC3 plane

in the endo direction. Thus, the anionic C atoms (C2) of the

two lithioallyl sulfones are not planar but pyramidalized in

the exo direction, i. e., the apex of the pyramid at C2 points

to the exo side, as indicated by several criteria of planarity

(Table 5). The pyramidalization angle χ₂^[30a]of the C2 atom

of endo-3/ent-endo-3·2diglyme is 35.3(6)° and that of endo-

4/ent-endo-4·2diglyme is 24.0(4)°. Such values correspond

to a hybridization approximately half-way between sp²and

sp³.^[31]The pyramidalization of C2 of endo-3/ent-endo-

3·2diglyme and endo-4/ent-endo-4·2diglyme is of the same

magnitude as that found for the Cα atoms of

Figure 4. Crystal structure of endo-3/ent-endo-3·2diglyme

[(Me₂CϪSO₂Ph)Li·diglyme]₂(χ_α

ϭ

40.8°)^[25b]and

(Me₂CϪSO₂Ph)Li·[2.1.1]-cryptand (χ_αϭ 32.5°),^[25e]whose

anionic C atoms carry two methyl groups. In all of the

aforementioned lithiosulfones the direction of pyramidali-

zation of the anionic C atoms is the same, i. e., the apex of

the pyramid at the anionic C atom is syn to the sulfonyl O

atoms. According to experiment and theoretical calculation,

the Cα atoms of non-planar α-sulfonyl carbanions feature

such a direction of pyramidalization presumably because of

the stabilizing hyperconjugative n_C-σ*_SPhinteraction and

the minimization of torsional strain around the CαϪS

bond.^[11]One reason for the pyramidalization of the C2

atoms of endo-3/ent-endo-3·2diglyme and endo-4/ent-endo-

4·2diglyme is the minimization of strain energy in the bi-

cyclic ring systems by the reduction of the C1ϪC2ϪC3

bond angle. The relief of torsional strain around the C2ϪS

bond^{[25b,25e,25g]}and of the allylic strain around the C2ϪC3

bond could also be important contributing factors.

Table 5. Pyramidalization of the C2 atom in endo-3/ent-endo-

3·2diglyme and endo-4/ent-endo-4·2diglyme

Figure 5. Crystal structure of endo-4/ent-endo-4·2diglyme

ative n_C-σ*_SRinteraction.^[27]The phenyl groups of the

anions of endo-3/ent-endo-3·2diglyme and endo-4/ent-endo-

4·2diglyme are both in the endo position.

A comparison of the bonding parameters of endo-3/ent-

endo-3·2diglyme and endo-4/ent-endo-4·2diglyme with those

of the corresponding sulfones reveals a significant shorten-

ing of the C2ϪS and C2ϪC3 bonds and a less significant

lengthening of the C3ϪC8 bond (cf. Table 1). The shorten-

endo-3

endo-34

˚

∆₂[A]

0.29

0.19

Σ_{C2 [°]}

α [°]

349.8

355.2

67.2^[a]

73.8^[a]

35.3(6)^[b]

73.0^[a]

79.5^[a]

24.0(4)^[c]

β [°]

χ₂[°]

[a]

[b]

[c]

Absolute values. Ϫ

χ₂ϭ 180.0Ϫ144.7(6). Ϫ

χ₂ϭ

ing of the C2ϪS bond is mainly caused by a Coulomb inter- 180.0Ϫ156.0(4).

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1631

H.-J. Gais et al.

FULL PAPER

Interestingly, the degree of pyramidalization of the C2 similar in both anions. This correlates well with measure-

atom of the norbornyl species endo-4/ent-endo-4·2diglyme ments of the gas-phase acidities for the carbon acids I and

is less than that of the norbornenyl species endo-3/ent-endo- III using the flowing afterglow technique, which revealed

3·2diglyme. At first glance one is tempted to ascribe the them to be the same.^[33]In summary, the difference in the

greater planarization of the allylic moiety of the norbornyl degree of Cα pyramidalization in endo-3/ent-endo-

anion endo-4/ent-endo-4·2diglyme to a steric interaction be- 3·2diglyme and endo-4/ent-endo-4·2diglyme as well as II and

tween the phenyl group and the H atoms 5-H_endoand 6- IV is most likely due to the different ring strain of the nor-

H

_endo. However, the same direction and similar degrees of bornenyl and norbornyl ring systems.

pyramidalization as for endo-3/ent-endo-3·2diglyme and

endo-4/ent-endo-4·2diglyme were obtained when the struc-

tures of the free methylene norbornenyl anion II (χ₂

ϭ

37.7°) and methylene norbornyl anion IV (χ₂ϭ 23.1°)

(Scheme 2), formally derived from 3 and 4 by replacing the

PhO₂SϩLi group by a H atom, were optimized at the HF/

6Ϫ31ϩG* level. When correlation energy (MP2) is included

in additional single point calculations (MP2/6Ϫ31ϩG*//

HF/6Ϫ31ϩG*) planarization of the allylic moiety in II re-

quires 1.8 kcal/mol, indicating that the gain of allylic stabi-

lization energy upon planarization is obviously insufficient

to compensate for the presumed increase of strain energy.

While the lengths of the C2ϪC3 and C3ϪC8 bonds are

˚

1.413 A and 1.370 A in the pyramidalized species, their

˚

values change to 1.390 A and 1.382 A when the allylic moi- Scheme 2. Isodesmic reactions of methylenenorbornene and methy-

ety is planarized. These values are quite close to the length

lenenorbornane with the allyl anion

of the CϪC bonds in the unsubstituted planar allyl anion

˚

which is 1.388 A at the same level of theory. Intuitively, one

It is of interest to note in context with the exo pyrami-

expects that transfer of a negative unit charge from a planar dalization of the C2 atoms in the norbornenyl anion II and

allyl anion to an allylic moiety with a pyramidalized ter- in the norbornyl anion IV that, according to experimental

minal C atom concurs with an increase of energy. The studies^[30]and theoretical calculations,^[34]in norbornene

change of energy associated with the isodesmic reaction be- and its derivatives the C atoms of the double bond are also

tween the methylenenorbornene I and the unsubstituted al- pyramidalized in the exo direction. This has been ascribed

lyl anion, however, reveals that this is not necessarily the either to a minimization of the torsional strain around the

case (cf. Scheme 2). While this energy of reaction is indeed two C(sp²)ϪC(sp³) single bonds^[30a,34bϪd]or, alternatively,

positive at the HF/6Ϫ31ϩG*//HF/6Ϫ31ϩG* level (ϩ5.1 to the σ_C1-C6-π* interaction being more stabilizing than the

kcal/mol), it is Ϫ1.9 kcal/mol when correlation energy is σ_C1-C7-π* interaction.^[30b,34eϪk]Similarly, pyramidalization

included in single point calculations. At Ϫ3.3 kcal/mol this of the C2 atoms of II and IV in the exo rather than the

change of energy is even more negative for the hypothetical endo direction (cf. Scheme 2) could be ascribed to a mini-

reaction between the methylenenorbornane III and the un- mization of torsional strain around the C1ϪC2 bonds, or,

substituted planar allyl anion (cf. Scheme 2). This might at alternatively to the hyperconjugative n_C-σ*_C1ϪC6interac-

least in part be due to the fact that the degree of pyrami- tion being more stabilizing than the hyperconjugative n_C-

dalization of the anionic center (C2) is lower in anion IV σ*_C1ϪC7interaction. An NBO analysis of the molecular

than in anion II and, therefore, allylic stabilization is more wavefunctions of II and IV revealed that the n_C-σ*_C1ϪC6

effective in this case. This is also reflected by the C2ϪC3 interaction plays a much more important role in these

˚

and C3ϪC8 bond lengths of IV, which are 1.397 A and anions than the n_C-σ*_C1ϪC7interaction. However, working

˚

1.383 A and therefore much more similar to each other than at the HF level, it has so far not been been possible to

the corresponding bond lengths in II. The relative stabili- locate stationary points corresponding to those epimers of

zation of the negative charge in the anions of endo-3/ent- II and IV whose C2 atoms are pyramidalized in the endo

endo-3·2diglyme and endo-4/ent-endo-4·2diglyme can be es- direction. Changing the sign of the HϪC2ϪC1ϪC3 di-

timated approximately from the change of energy associated hedral angle and restarting the geometry optimization re-

with the isodesmic reaction between II and III (cf. Scheme sulted in II and IV, respectively. Therefore single point cal-

2). This energy of reaction can be calculated from those culations were performed for the starting geometries and

associated with the other hypothetical reactions in Scheme the resulting wave-functions subjected to an NBO analysis.

2, and amounts to Ϫ2.5 kcal/mol at the HF and Ϫ1.4 kcal/ The results of these calculations show indeed that in the

mol at the MP2/6Ϫ31ϩG*//HF/6Ϫ31ϩG* level. The sig- endo pyramidalized anions the n_C-σ*_C1ϪC7interaction is

nificant reduction of this energy of reaction upon inclusion more important than in the exo isomers while just the op-

of MP2 corrections might indicate that a more complete posite is true for the interaction of the anionic lone pair

treatment of the correlation energy will further reduce this with the σ*_C1ϪC6orbital. However, since no stationary

value, and that stabilization of the negative charge is quite points for the endo anions could be located so far, it remains

1632

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

1

uncertain whether the gain of stabilization energy due to per formula unit according to H-NMR spectroscopy. Ag-

increased n_C-σ*_C1ϪC7mixing is sufficient to compensate the gregation states of

n

ϭ

1.58 ±0.13 (c_nom

ϭ

simultaneous loss of n_C-σ*_C1ϪC6stabilization energy. 49.7Ϫ50.3 mmol/kg) and of n ϭ 1.62±0.06 (c_nom

ϭ

Finally, it seems to be of interest to compare the struc- 50.0Ϫ50.3 mmol/kg) were determined for the norbornenyl

tures of endo-3/ent-endo-3·2diglyme and endo-4/ent-endo- species 3 and the norbornyl species 4, respectively, at the

4·2diglyme with the crystal structure of the acyclic lithioal- freezing point of THF (Ϫ108.5°C). Hence, at low tempera-

lyl sulfone 7/ent-7·2diglyme (Figure 6).^[16]The structures of tures in THF lithiosulfones 3 and 4 exist in this concen-

all three species are quite similar with respect to the coordi- tration range as approximately 1:1 mixtures of monomers

nation of the Li atom, the conformation of the anions, the and dimers (cf. Experimental Section). For the monomeric

degree of alternation of the bond lengths of the allylic moi- species the OϪLi contact ion pair structures endo-3a؊c

eties, the length of the bond between the S atom and the (endo-4a؊c) and exo-3a؊c (exo-4a؊c) can be envisioned,

anionic C atom, and the direction of pyramidalization of in which the Li atom is coordinated either by one or two O

the anionic C atoms. The degree of pyramidalization of the atom(s) and where the phenyl group is either endo or exo

anionic C atom of the acyclic lithiosulfone, however, seems (Schemes 3 and 4).^[12c]A contact ion pair structure of type

to be less than in the bicyclic lithiosulfones as expressed by endo-3a,b (endo-4a,b) and exo-3a,b (exo-4a,b) was found for

χ_αϭ 22.4°. Interestingly, lithioallyl sulfoximines, the aza 12-crown-4 complexed (α-benzyl)phenyl lithiobenzyl sul-

analogues of the sulfones, adopt a similar dimeric contact fone in the crystal,^[40]while a contact ion pair structure of

ion pair structure as endo-3/ent-endo-3·2diglyme, endo-4/ type endo-3c (endo-4c) and exo-3c (exo-4c) was discovered

ent-endo-4·2diglyme and 7/ent-7·2diglyme in the crystal, be- for 18-crown-6 complexed phenyl potassiobis(trimethylsilyl)-

ing devoid of CϪLi bonds.^[35]

methyl sulfone and potassiomethyl tolyl sulfone in the crys-

tal,^[41][42]and located computationally for lithiomethyl

methyl sulfone.^[27b,27e]Likely structures for the dimeric

OϪLi contact ion pairs of 3 and 4 are endo-3/ent-endo-3

(endo-4/ent-endo-4) and exo-3/ent-exo-3 (exo-4/ent-exo-4),

which are similar to those found in the crystal except for

the replacement of the two diglyme molecules by four THF

molecules. It was previously observed that the OϪLi con-

tact ion pair structure of a dimeric lithiosulfone in the crys-

tal is basically not altered upon replacement of chelating

ligands by THF molecules.^[25c,25d]Based on this previous

observation that aggregation has only a minor influence

upon

the

anion

structure

of

lithiosul-

fones,^{[25c,25d,27b,27e,40Ϫ42]}the respective anions of monomers

endo-3a-c (endo-4a-c) and exo-3a-c (exo-4a-c), as well as of

dimers endo-3/ent-endo-3 (endo-4/ent-endo-4) and exo-3/ent-

exo-3 (exo-4/ent-exo-4) are expected to exhibit basically the

same structure.

In addition to the heterochiral dimers the corresponding

homochiral dimers also have to be considered. In the crystal

only dimers having endo anions were found. This may indi-

cate that in solution dimers with endo anions are also more

stable than those with exo anions (vide infra).

In order to gain further information about the monomer/

dimer and, especially, endo/exo equilibria and as to the elec-

tronic structure of the allylic moieties of 3 and 4, the lithio-

sulfones were studied by ¹H-, ¹³C- and ⁶Li-NMR spec-

troscopy. Lithiosulfones 3 and 4 were prepared for the spec-

Figure 6. Crystal structure of 7/ent-7·2diglyme; the view is con-

structed from the published coordinates;^[16]selected bond lengths

˚

[A]: SϪC1 1.668(5), C1ϪC2 1.445(5), C2ϪC3 1.352(9)

b. Structure in Solution

Acyclic lithiosulfones generally exist in THF solution as troscopic investigations by the separate treatment of the re-

monomeric OϪLi contact ion pairs being in equilibrium spective sulfones exo-5, endo-5, exo-6 and endo-6 with

6

with small amounts of the corresponding dimeric OϪLi nBuLi at Ϫ78°C in [D₈]THF, and Li-labeled 3 and 4 were

contact ion pairs.^{[25c,25d,36Ϫ38]}Because of the unknown ag- obtained by deprotonation of the corresponding sulfones

1

gregation of lithiosulfones of type 2, that of 3 and 4 in THF with nPr⁶Li. Interestingly, the H-NMR (400 MHz), ¹³C-

by cryoscopy were determined.^[39]Lithiosulfones 3 and 4 NMR (100 MHz) and ⁶Li-NMR (44 MHz) spectra of 3 and

were prepared for the cryoscopic experiments by treatment 4 at Ϫ100°C in THF showed only one set of signals, to-

of the corresponding sulfones endo-5, exo-5, endo-6 and gether with only minor shift differences and line broaden-

exo-6 with nBuLi in THF/n-hexane followed by removal of ing. No differences in the NMR spectra of 3 and 4 were

the solvents in vacuo. The microcrystalline lithiosulfones 3 observed starting either from the corresponding exo or endo

and 4 thus prepared each contained two molecules of THF sulfone. Provided the respective monomeric and dimeric

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1633

H.-J. Gais et al.

FULL PAPER

Scheme 3. Monomer-dimer equilibria of the lithioallyl sulfones endo-3 and endo-4

Scheme 4. Monomer-dimer equilibria of the lithioallyl sulfones exo-3 and exo-4

contact ion pairs are NMR-spectroscopically distinguish- those in the ortho position of the phenyl group for the

able, these results indicate that attainment of the monomer/ monomers and dimers, having endo and exo anions. No

dimer and the endo/exo equilibria (vide infra) of 3 and 4 is NOE effect was observed, however, between the Li atom

fast on the NMR time scale at low temperatures (cf. and the H atoms at C7(H_syn) and C6. The assumption of a

Schemes 3 and 4). The endo/exo equilibration of 3 and 4 close neighboring position of the Li atom and H-1 is cor-

requires, besides other processes, C2ϪS bond rotation and roborated by the downfield shift of the signal for H-1^[43]of

C2 inversion. D-NMR spectroscopy of the lithium salts of 3 and 4 as compared to sulfones endo-5, exo-5, endo-6 and

6

chiral acyclic secondary α-phenylsulfonyl carbanions in exo-6, respectively. The results of the Li{¹H}-NOE experi-

[D]₈THF revealed ∆G^϶values for their enantiomerization ments would be compatible with both the endo species endo-

of 9Ϫ10 kcal/mol.^[25c]Thus, in the case of 3 and 4 C2ϪS 3a؊c (endo-4a؊c) and endo-3/ent-endo-3 (endo-4/ent-endo-

bond rotation and C2 inversion could have indeed such low 4) as well as with the exo species exo-3a؊c (exo-4a؊c) and

barriers, allowing for a rapid equilibration of the endo/exo exo-3/ent-exo-3 (exo-4/ent-exo-4) (cf. Figures 5 and 6).

diastereomers at even Ϫ100°C.

Thus, no statement as to the position of the endo/exo equili-

The signals in the H- and ¹³C-NMR spectra of 3 and 4 bria is possible. Somewhat more informative, however, were

at 25°C in [D₈]THF were completely assigned by a combi- ¹H{¹H}-NOE experiments of 3 at 25°C in [D₈]THF, which

nation of NOE experiments and two-dimensional methods revealed NOE effects only between the ortho-H atoms of

(H,H-COSY and C,H-COSY) (cf. Tables 2Ϫ4). An inspec- the phenyl group and the H atoms at C6 and C8(H_syn), but

tion of Tables 2 and 4 reveals that the changes occurring in not between the ortho-H atoms and the H atom at C7(H_syn).

1

the NMR spectra on going from the acid to the anion are, Unfortunately, similar H{¹H}-NOE experiments were not

1

in both cases, similar in sign and magnitude.

possible in the case of 4 because of the almost identical

⁶Li,¹H-HOESY experiments^[43]of 3 and 4 in [D₈]THF at chemical shifts for the H atoms at C6(H_endo) and C7(H_syn).

Ϫ78°C revealed, on the average, close spatial proximities These results might perhaps be taken as an indication for

between the Li atom and the H atoms at C1, C8(H_syn) and the endo/exo equilibria of 3 in THF, and most likely that of

1634

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

4 also, being shifted to the side of the endo isomers (cf. negative charge at C2. The concept of a stabilization of al-

Scheme 3).

lylic anions by resonance has been challenged recently on

Of particular interest is the mode of the additional stabi- the basis of ab initio calculations.^[32b]It was proposed that

lization of the negative charge of 3 and 4 provided by the internal Coulomb interaction is the dominating mechanism

neighboring double bond.^[18]Chemical shift variations of of stabilization and that the contribution by resonance is

the ¹³C- and H-NMR signals on going from the carbon only minimal. Stabilization of the anions of 3 and 4 by

1

acid to anion provide a good tool for an estimation of the Coulomb interactions could indeed be important because

structural changes occurring upon deprotonation and for of the existence of the extended charge alternating structure

the charge distribution in the anion.^[28b]In case of 3 and 4 O^Ϫ(O^Ϫ)ϪS^ϩϪC2^ϪϪC3^ϩϪC8^Ϫ.^[46]Quite interesting in re-

the situation is rather complex because of the existence of gard to the question of the mode of the additional stabili-

the monomeric and dimeric OϪLi contact ion pairs having zation of the negative charge in 3 and 4 by the adjacent

exo and endo anions. Because of the fast attainment of the double bond is the observation that the ∆δ_8-Hand ∆δ_C8

various equilibria on the time scale of NMR spectroscopy, values for the norbornyl species are significantly larger than

only averaged signals are observed. Thus, only a qualitative those for the norbornenyl species (cf. Tables 2 and 4). This

interpretation of the NMR data of 3 and 4 in regard to the difference is most likely caused by a higher negative charge

above question is possible. An inspection of Tables 2 and 4 density at the C8 atom of 4 as compared to 3. One of the

reveals the following shift variations (∆δ) on going from main structural difference between endo-3/ent-endo-

the respective sulfones endo-5, exo-5, endo-6 and exo-6 to 3·2diglyme and endo-4/ent-endo-4·2diglyme in the crystal is

lithiosulfones 3 and 4: (i) a slight downfield shift of the ¹³C the higher degree of planarization of the allylic moiety of

1

signal for C2, (ii) a strong upfield shift of the ¹³C and H the norbornyl species. Since the accumulation of negative

signals for C8 and H-8, respectively, and (iii) a less strong charge at C8 of 3 and 4 by allylic delocalization but not by

downfield shift of the ¹³C signal for C3. These shift varia- polarization depends on the degree of planarization of the

tions correspond very well to those observed for other cyc- allylic moieties, the NMR data point to a significant allylic

lic and acyclic lithiated allylic sulfones.^{[15Ϫ17,25g,44]}It is gen- delocalization of the negative charge in 3 and 4.

erally accepted that two main effects, which oppose each

1

other, contribute to the chemical shift variation of the H-

3. Structure of the Free Anions

and ¹³C-NMR signals on going from the carbon acid to the

anion: an upfield shift due to the negative charge and a

The theoretical calculations of the free anions of 3 and 4

downfield shift in case of a change of the coordination ge- were carried out in order to obtain information as to the

ometry from tetrahedral to planar.^[28]The shift variations relative stabilities of the endo and exo conformers as well

recorded for the OϪLi contact ion pairs 3 and 4 would thus as to the distribution of their negative charge. In addition,

be qualitatively compatible with allylic moieties which are structural information about the exo anions were sought

characterized by (i) a coordination geometry of C2 between because of their experimental inaccessibility. The structures

tetrahedral and planar, (ii) an accumulation of negative of the endo and exo conformers of 3(ϪLi^ϩ) and 4(ϪLi^ϩ)

charge at C2 and C8 and (iii) an accumulation of positive have been optimized at the HF/6Ϫ31ϩG* level resulting in

charge at C3. In accordance with a degree of pyramidali- the local minima endo-3(ϪLi^ϩ), exo-3(ϪLi^ϩ), endo-

zation of the magnitude found in the crystal would also be 4(ϪLi^ϩ) and exo-4(ϪLi^ϩ) (Figures 7 and 8, Table 7). The

the magnitude of the one-bond ¹³C,¹³C coupling between C2 atoms of endo-3(ϪLi^ϩ) and endo-4(ϪLi^ϩ) are pyrami-

C2 and C3 and of the two-bond ¹³C,¹³C coupling between dalized into the exo direction as shown by the out-of-plane

C2 and C9 through the SO₂group^[45](Table 6).

bending of the S atoms in the endo direction. The degree of

pyramidalization is somewhat diminished as compared to

that found in endo-4/ent-endo-4·2diglyme and endo-4/ent-

endo-4·2diglyme. The C2 atoms of the exo anions exo-

3(ϪLi^ϩ) and exo-4(ϪLi^ϩ) are not planar but pyramidalized

into the endo direction, and the degree of pyramidalization

is similar to that of the endo anions endo-3(ϪLi^ϩ) and endo-

4(ϪLi^ϩ). The agreement between the structural parameters

of the free gas phase anions endo-3(ϪLi^ϩ) and endo-

4(ϪLi^ϩ) and the anions of the dimeric contact ion pairs

endo-3/ent-endo-3·2diglyme and endo-4/ent-endo-4·2diglyme

in the crystal is reasonably good. According to the calcu-

lations the endo anions endo-3(ϪLi^ϩ) and endo-4(ϪLi^ϩ) are

more stable than the exo anions exo-3(ϪLi^ϩ) and exo-

4(ϪLi^ϩ) as expressed by their relative energies (MP2/

6Ϫ31ϩG*//HF/6Ϫ31ϩG*) of ϩ2.40 kcal/mol and ϩ0.99

Table 6.¹³C-NMR data (J [Hz]) of lithiosulfones 3 and 4 in

[D]₈THF at 25°C

3

4

C-1,C-2

C-1,C-6

C-1,C-7

C-2,C-3

C-2,C-9

C-3,C-4

C-3,C-8

C-4,C-5

C-4,C-7

C-5,C-6

C-9,C-10

35

36

31

62

Ϫ

34

76

36

31

70

Ϫ

40

33

30

55

9

37

75

31

30

31

57

The two mechanisms leading to a charge distribution of kcal/mol, respectively.

type C2^ϪϪC3^ϩϪC8^Ϫin 3 and 4 are allylic delocalization

The distribution of the atomic point charges, which were

and/or polarization of the C3ϪC8 double bond by the calculated by the method of Kollman et al.,^[47]in endo-

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1635

H.-J. Gais et al.

FULL PAPER

Figure 7. Calculated (HF/6Ϫ31ϩG*) structures of endo-3(ϪLi^ϩ)

(a) and endo-4(ϪLi^ϩ) (b)

Figure 8. Calculated (HF/6Ϫ31ϩG*) structures of exo-3(ϪLi^ϩ) (a)

and exo-4(ϪLi^ϩ) (b)

3(ϪLi^ϩ), exo-3(ϪLi^ϩ), endo-4(ϪLi^ϩ) and exo-4(ϪLi^ϩ)

shows the pattern typical for the unsubstituted allyl

anion,^[32]i. e., negatively charged termini (C2, C8) and a

positive charge at the central carbon atom (C3) whereby the

negative charge is not equally distributed among C2 and

C8. However, significant amounts of the negative charge

are transferred onto the PhO₂S substituent, bonded to the

C2ϪC3ϪC8(H)₂segment.^[48]While in the unsubstituted al-

lyl anion the sum of the atomic charges of this segment

amounts to Ϫ1.60 e, it is Ϫ0.72 e in endo-3(ϪLi^ϩ) and

Ϫ0.70 e in endo-4(ϪLi^ϩ). The remaining negative charge in

endo-3(ϪLi^ϩ) and in endo-4(ϪLi^ϩ) is transferred com-

pletely to the phenylsulfonyl group while those atoms of the

norbornyl substituents which are not part of the allyl moi-

ety result in very small positive charges of 0.03 e [endo-

4(ϪLi^ϩ)] and 0.04 e [endo-3(ϪLi^ϩ)], respectively. No princi-

pal differences in the charge distribution of the endo and

exo anions were found. Charge transfer from C2 to C8 in

endo-4(ϪLi^ϩ) and exo-4(ϪLi^ϩ), whose allylic moieties are

Table 7. Structural parameters and atomic charges of endo-

3(ϪLi^ϩ), endo-4(ϪLi^ϩ), exo-3(ϪLi^ϩ) and exo-4(ϪLi^ϩ) at the HF/

6Ϫ31ϩG* level

endo-

exo-

3(ϪLi^ϩ)

4(ϪLi^ϩ)

3(ϪLi^ϩ)

4(ϪLi^ϩ)

˚

S1ϪC2 [A]

1.667

1.808

1.455

1.451

1.343

0.23

1.666

1.808

1.455

1.451

1.445

1.346

0.14

1.657

1.809

1.451

1.455

1.451

1.342

0.21

1.661

1.807

1.451

1.455

1.445

1.346

0.16

˚

S1ϪC9 [A]

˚

S1ϪO1 [A]

˚

S1ϪO2 [A]

˚

C2ϪC3 [A]

˚

C3ϪC8 [A]

˚

∆₂[A]

χ

[°]

26.6

17.0

24.2

18.5

q²(C2) [e]

q(C3) [e]

q(C8) [e]

q(S) [e]

Ϫ0.70

ϩ0.47

Ϫ1.02

ϩ1.10

Ϫ0.69

Ϫ0.65

ϩ0.48

Ϫ1.08

ϩ1.09

Ϫ0.67

Ϫ0.68

Ϫ0.50

ϩ0.34

Ϫ0.98

ϩ0.98

Ϫ0.65

Ϫ0.68

Ϫ0.41

ϩ0.30

Ϫ1.01

ϩ0.99

Ϫ0.66

Ϫ0.68

q(O1) [e]

q(O2) [e]

more planarized, is somewhat stronger than in endo- anion (132.2°). The calculated charge distributions of the

3(ϪLi^ϩ) and exo-3(ϪLi^ϩ). The C2ϪC3ϪC8 bond angles in free anions are qualitatively in good agreement with those

all four anions cover the range between 134.1Ϫ134.6° and, derived for the corresponding anions of the OϪLi contact

therefore, are quite similar to the value for the free allyl ion pairs by NMR spectroscopy (cf. Tables 2 and 4).

1636

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

4. Reaction of the Lithiosulfones with Electrophiles

While alkylation of the norbornenyl anion 3 with MeI,

allylic bromides and propargyl bromide had revealed the

selective formation of the corresponding endo sulfones

(87:13 to > 99:1 d.r.),^[6e,6f]methylation of the norbornyl

anion 4 with MeI was reported to give preferentially the

corresponding exo sulfone (81:19 d.r.).^[6e]No further reac-

tions of 4 with electrophiles have been described, however.

In the light of the results of the structural investigations of

3 and 4 outlined above and because of the questions elabor-

ated upon in the introduction, it was decided to study their

reaction with electrophiles on a broader basis. Especially

reaction of 3 and 4 at low temperatures were sought starting

separately from sulfones endo-5, exo-5, endo-6 and exo-6 in

the absence and in the presence of the electrophile (stepwise

and in situ method, respectively).

a. Norbornenyl Anion

Deuteration of 3, which was prepared by the separate

treatment of endo-5 and exo-5 at Ϫ78°C with nBuLi in

THF, with CF₃COOD at Ϫ78°C gave in both cases a mix-

ture of endo-8a and exo-8a in a ratio of 87:13 (Scheme 5)

(Table 8, entry 1). A similar result was obtained by using

DCl instead of CF₃COOD under otherwise identical con-

ditions (Table 8, entry 2). Reaction of 3 with MeI, EtI, nPrI

and nHeI under the same conditions (stepwise method) af-

forded mixtures of endo-8b؊e and exo-8b؊e, in which the

endo isomers dominated, in high yields (Table 8, entries

3Ϫ8). Selectivity of alkylation varied and was the highest

in the case of the sterically more demanding alkylating re-

agents. No difference in the ratio of the methylation prod-

ucts endo-8b and exo-8b was observed by starting either

from endo-5 or exo-5.^[6f,6g]The configurations of exo-8b؊e

Scheme 5. Reaction of the lithioallyl sulfone 3 with electrophiles

and endo-8b؊e were assigned unequivocally by NMR spec- ment of 3 with PhCHO at Ϫ78°C followed by addition of

troscopy in combination with NOE experiments. Thus, the AcOH at Ϫ78°C shortly afterwards gave alcohols endo-8f

norbornenyl species 3 reacts with alkylating reagents, in ac- and exo-8f in a ratio of 88:12 and none of alcohol γ-8f.

cordance with previous observations,^[6e,6f]under the prefer- Surprisingly, endo-8f and exo-8f were formed as single dia-

ential formation of the endo sulfones independent of the stereomers. Normally, hydroxyalkylation of lithiosulfones

configuration of the starting sulfones. Hydroxyalkylation of proceeds with low diastereoselectivity.^[13][14]While the con-

3 with PhCHO yielded different products depending on the figuration of endo-8f was secured by NOE experiments that

reaction temperature (Table 8, entries 9 and 10). Thus, treat- of exo-8f has not been determined. Warming the reaction

Table 8. Reaction of lithiosulfones 3 and 4 with electrophiles at Ϫ78°C in THF

Electrophile Entry

Product

R

endo/exo/γ Yield (%) Entry

Product

R

endo/exo/γ

Yield (%)

CF₃COOD

DCl

1

8a

D

87:13:0

88:12:0

77:23:0

76:24:0

83:17:0

89:11:0

88:12:0

95

91

85

96

95

84

87

80

12

9a

D

59:41:0

Ϫ

88

Ϫ

2

8a

D

Ϫ

MeI

3

8b^[a]

8b^[c]

8b^[e]

8c

Me

Et

13

14

15

16

17

18

19

20

21

9b^[b]

9b^[d]

9b^[f]

9c

Me

Et

nPr

nHe

81:19:0

80:20:0

95:5:0

97:3:0

90

87

91

95

81

79

66

80

MeI

4

MeI

5

EtI

6

nPrI

7

8d

nPr

nHex

9d

nHeI

8

8e

9e

PhCHO

Me₃SiCl

9

8f

PhC(H)OH 88:12:0^[g]88

9f

PhC(H)OH 3:0:97^[h]

PhC(H)OH 0:0:100^[k]

10

11

8f^[i]

8g

PhC(H)OH 0:0:100^[j]

Me₃Si 0:0:100

72

86

9f^[i]

9g

Me₃Si

0:0:100

[a]

[c]

[e]

From a mixture of endo-5 and exo-5. ؊ ^[b]From a mixture of endo-6 and exo-6. Ϫ From endo-5. Ϫ ^[d]From endo-6. Ϫ From exo-

[f]

[g]

[h]

[i]

[j]

[k]

5. Ϫ From exo-6. Ϫ Single diastereomers. Ϫ 87:10 d.r. Ϫ At 25°C. Ϫ 50:50 d.r. Ϫ 69:31 d.r.

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1637

H.-J. Gais et al.

FULL PAPER

mixture derived from 3 and PhCHO to room temperature nation with NOE experiments. A final proof for the prefer-

before quenching with aqueous NaHCO₃led to isolation ential formation of the endo sulfone endo-9b and not of the

only of alcohol γ-8f. This temperature dependent reactivity exo sulfone exo-9b in reaction of 4 with MeI was ac-

of 3 towards PhCHO is typical for lithiated allylic sulfones complished by chemical correlation (Scheme 7). Selective

which react with aldehydes under kinetic control in α posi- and partial hydrogenation of a 76:24 mixture of endo-8b

tion and under thermodynamic control in γ position.^[13][14]and exo-8b^[19]led to isolation of a 76:24 mixture of endo-

Presumably because of steric reasons, reaction of 3 with 9b and exo-9b in a yield of 86%.^[50]Thus, the major isomer

Me₃SiCl occurred almost exclusively in γ position and gave obtained by hydrogenation of the mixture of endo-8b and

γ-8g in 86% yield (Table 8, entry 11).

exo-8b and the major isomer formed in methylation of 4

are identical. Interestingly, reaction of 4 with PhCHO pro-

ceeded already at Ϫ78°C almost exclusively in γ position

and gave alcohol γ-9f (Table 8, entries 19 and 20). The con-

figurations of the other sulfones derived from 4 were also

assigned by NMR spectroscopy in combination with NOE

experiments. Presumably because of steric reasons, reaction

of 4 with Me₃SiCl occurred almost exclusively in γ position

and gave γ-9g in a yield of 80% (Table 7, entry 21).

b. Norbornyl Anion

Deuteration of 4, which was prepared by the separate

treatment of endo-6 and exo-6 at Ϫ78°C with nBuLi in

THF, with CF₃COOD at Ϫ78°C gave in both cases a mix-

ture of endo-9a and exo-9a in a ratio of 59:41 (Scheme 6)

(Table 8, entry 12). Reaction of 4, which was prepared in

the same manner as above, with MeI, EtI, nPrI and nHeI

(stepwise method) proceeded in the same manner as that of

3 and gave preferentially the endo sulfones endo-9b؊e

(Table 8, entries 13Ϫ18). Here, too, selectivity of alkylation

varied and was highest with the sterically more demanding

alkylating reagents. We note that selectivities of alkylation

of 4 were higher than those of 3. Treatment of 4 with MeI

in THF at Ϫ78°C afforded a mixture of endo-9b and exo-

9b in a ratio of 80:20 (Table 8, entries 13Ϫ15) and not in

an opposite ratio of 19:81 as reported previously.^[6e,49]No

difference in the ratio of the methylation products endo-9b

and exo-9b was observed by starting either from endo-6 or

exo-6. The configurations of endo-9b and exo-9b were un-

equivocally determined by NMR spectroscopy in combi-

Scheme 7. Stereochemical correlation of the unsaturated sulfones

endo-8b/exo-8b with the saturated sulfones endo-9b/exo-9b

c. In situ Methylation

Because of the fast isomerization of the endo and exo

diastereomers of 3 and 4 even at Ϫ105°C, equilibria are

fully established before addition of the electrophile occurs

(stepwise method). Therefore the separate deprotonation of

endo-5, exo-5, endo-6 and exo-6 with nBuLi at low tempera-

tures in the presence of MeI (in situ method) was studied.

The in situ method has been successfully pioneered by

Hoffmann et al. for the determination of the configura-

tional stability of heteroatom-substituted chiral organo-

lithium compounds.^[51]Deprotonation of endo-5 with

nBuLi at Ϫ78°C in the presence of MeI (11 equiv.) led to

isolation of a mixture of endo-8b and exo-8b in a ratio of

76:24, which is not different from that recorded by the step-

wise method (Table 9, entry 1). However, deprotonation of

endo-5 with nBuLi at Ϫ105°C in the presence of MeI (11

equiv., c ϭ 0.044 mol/L) led to isolation of a mixture of

sulfones endo-8b and exo-8b in a ratio of 68:32 (Table 9,

entry 2). The ratio of endo-8b and exo-8b changed even to

61:39 (Table 9, entry 3) by raising the amount of MeI in

the in situ methylation to 100 equiv. (c ϭ 4.0 mol/L) under

otherwise identical conditions. In a control experiment

endo-5 was treated with nBuLi at Ϫ78°C followed by ad-

dition of MeI at Ϫ105°C to give a mixture of endo-8b and

Scheme 6. Reaction of the lithioallyl sulfone 4 with electrophiles

1638

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

Table 9. In situ methylation of sulfones endo-5, exo-5, endo-6 and exo-6

Entry

Sulfone

Lithiosulfone

T [°C]

Equiv. MeI

Product

endo/exo

Yield (%)

1

2

endo-5

exo-5

3

4

Ϫ78

11

8b

9b

76:24

68:32

61:39

76:24

65:35

57:43

45:55

80:20

81:19

80:20

96

98

96

98

99

Ϫ105

Ϫ105^[a]

Ϫ78

3

100

11

4

5

11

6

exo-5

Ϫ105

Ϫ78

Ϫ105

7

exo-5

100

11

8

exo-6

9

exo-6

11

10

endo-6

[a]

Deprotonation at Ϫ78°C and addition of MeI at Ϫ105°C after 20 min.

exo-8b in the same ratio (Table 9, entry 4) as recorded by ness concepts^[52]should apply to the endo/exo selectivity of

the stepwise method at Ϫ78°C. Thus, the differences ob- reaction of 3 and 4 with electrophiles. Before an attempt is

served in the stepwise and the in situ methylation of the made to rationalize the stereoselectivity three questions

norbornenyl sulfone endo-5 can not be due to a temperature have to be addressed. (1) What are the reactive species? (2)

dependency of the diastereoselectivity of the methylation or What is the stereoselectivity of their reaction with electro-

of the equilibria between the various contact ion pairs. philes? (3) Although of less importance in the present case,

Rather, the lower diastereomer ratios of sulfones endo-8b what is the stereoselectivity of the deprotonation of the endo

and exo-8b obtained by the in situ method, especially at a and exo isomers of 5 and 6? In THF solution the respective

higher concentration of MeI, indicate that the trapping of monomeric contact ion pairs endo-3a؊c (endo-4a؊c) and

the endo and exo anionic species now competes more with exo-3a؊c (exo-4a؊c) as well as the corresponding homo-

their isomerization. The in situ methylation of the exo nor- and heterochiral dimers have to be considered primarily as

bornenyl sulfone exo-5 took a similar course but resulted in reactive species (cf. Schemes 3 and 4). An inspection of

an inversion of the ratio of endo-8b and exo-8b (Table 9, space-filling models of monomers endo-3c (endo-4c) and

entries 5Ϫ7). Whereas the stepwise methylation of exo-5 at exo-3c (exo-4c) as well as of dimers endo-3/ent-endo-3

Ϫ78°C gave endo-8b and exo-8b in a ratio of 76:24, its in (endo-4/ent-endo-4) and exo-3/ent-exo-3 (exo-4/ent-exo-4)

situ methylation at Ϫ105°C yielded the two sulfones in a shows, however, that their anionic C atoms ought to be sev-

ratio of 45:55. These results strongly support the notion erely shielded by the THF molecule(s) in the syn position.

that 3 forms an equilibrium mixture of endo-3 and exo-3 in Thus, more likely candidates should be the monomers endo-

THF solution (vide infra). In contrast to the in situ meth- 3a,b (endo-4a,b) and exo-3a,b (exo-4a,b) in which the C2

ylation of the norbornenyl sulfones that of the norbornyl atom is less shielded by the THF solvated Li atom. It

sulfones exo-6 and endo-6 revealed no differences as com- should be noted, however, that endo-3c (endo-4c) and exo-

pared to the stepwise method at Ϫ78°C (Table 9, entries 3c (exo-4c) might be in equilibrium with the corresponding

8Ϫ10). At a first glance it seems odd that only the in situ mono-solvated species whose C2 atom is less shielded

methylation of endo-5 and exo-5 gives different product ra- towards electrophiles. Although it has been concluded from

tios upon lowering the temperature from Ϫ78°C to Ϫ105°C determination of ion pair acidities and conductivities that

or by raising the concentration of MeI. We have observed, lithiosulfones are OϪLi contact ion pairs (CIP) in THF,^[38]

however, that during the dropwise addition of nBuLi to a the existence of minute amounts of solvent separated ion

mixture of exo-6 or endo-6 and MeI (11 equiv.) at Ϫ105°C pairs (SSIP), which might have a higher reactivity than

in THF a transient yellow color developed whereas in the CIPs, cannot be excluded. However, based on X-ray crystal-

case of the similar in situ methylation of exo-5 or endo-5 no lography of lithium and tetrabutyl ammonium salts of α-

such phenomenon was observed. Since formation of 3 and sulfonyl carbanions^[25c,25d]it can be safely assumed that the

4 is associated with the development of an intensive yellow anions of the corresponding CIPs and SSIPs will have simi-

color of their solutions, this indicates that the norbornyl lar structures.^[11,25Ϫ27]

species 4 has a lower reactivity towards MeI than the nor-

bornenyl species 3 and, as a consequence, the endo/exo equi-

librium of 4 is presumably fully attained before methylation

occurs even in the presence of the electrophile.

A study of the reactivity of conformationally stable

acyclic chiral S-tert-butyl- and S-trifluoromethylsulfonyl

carbanions had revealed a very high propensity for electro-

philes to attack the anionic C atom, irrespective of its coor-

dination geometry, syn to the O atoms and, thus, anti to the

d. Model for the endo/exo Selectivity of Anion Reaction

tert-butyl and trifluoromethyl groups.^[25f,53]This has been

According to the aforementioned results, the endo and attributed to a shielding of the face of the anionic C atom

exo diastereomers of 3 and 4 are conformationally labile anti to the O atoms by the substituent at the S atom. In

on the time scale defined by the rate of their reaction with addition a possible energy lowering interaction between the

electrophiles. Hence, the Curtin-Hammett/Winstein-Hol- electrophile and the Li atom, which is coordinated to the O

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1639

H.-J. Gais et al.

FULL PAPER

atom of the sulfonyl group, may contribute to the high syn C1ϪC6 bond are nearly ideal staggered whereas in the tran-

selectivity, too. Thus, because of similar parameters for the sition state of the endo attack these bonds are more eclips-

freely rotating phenyl group (E_sϭ 1.01_depth, E_sϭ 3.82_width), ing.^[34c,34d]Thus, the exo attack of electrophiles on endo-

the tBu group (E_sϭ 2.78) and the CF₃group (E_sϭ 2.4),^[54]3a,b (endo-4a,b) should be faster than the endo attack on

it is believed that endo attack on endo-3a,b (endo-4a,b) and exo-3a,b (exo-4a,b) because in the transition state of the exo

exo attack on exo-3a,b (exo-4a,b) are sterically hindered attack on endo-3a,b (endo-4a,b) the developing C2ϪR bond

and that electrophiles attack endo-3a,b (endo-4a,b) with a and the C1ϪC6 bond are nearly staggered whereas in the

high preference at the exo face and exo-3a,b (exo-4a,b) at transition state of endo attack on exo-3a (exo-4a) those

the endo face (cf. Schemes 3 and 4). The exo attack on endo- bonds are eclipsed (Figure 9). In an alternative but closely

3a,b and endo-4a,b should be likewise preferred because of related view the faster exo attack on endo-3a,b (endo-4a,b)

the pyramidalization of the anionic C atoms^[55][56]in the can be ascribed to the developing C2ϪR bond and the

exo direction. Unfortunately, experimental information as C1ϪC7 bond in the transition state of the exo attack on

to the coordination geometry of the C2 atom of the exo endo-3a,b (endo-4a,b) being less eclipsing than the de-

anions exo-3a-c, exo-4a-c, exo-3/ent-exo-3 and exo-4/ent- veloping C2ϪR bond and the C1ϪC6 bond in the tran-

exo-4 is not available. However, the ab initio calculation of sition state of endo attack on exo-3a (exo-4a). It should be

the free exo anions exo-3(ϪLiϩ) and exo-4(ϪLi^ϩ) have re-

vealed a pyramidalization of C2 in the endo direction which

is of the same magnitude as found in endo-3(ϪLi^ϩ) and

endo-4(ϪLi^ϩ). Thus, it seems safe to assume that the C2

atoms of exo-3a؊c, exo-4a؊c, exo-3/ent-exo-3 and exo-4/

ent-exo-4 are pyramidalized in the endo direction. This

should provide an additional bias for a preferential attack

of electrophiles on exo-3a؊c, exo-4a؊c, exo-3/ent-exo-3

and exo-4/ent-exo-4 at the endo face.

On the basis of the above assumptions the preferential

formation of endo-8a؊f (endo-9a؊e) in reaction of 3 (4) is

ascribed to the exo attack of the electrophiles on endo-3a,b

(endo-4a,b) being faster than endo attack on exo-3a,b (exo-

4a,b) in combination with the endo/exo equilibration of the

anionic species being faster than their reaction with the

electrophiles (cf. Schemes 5 and 6). The underlying kinetic

scheme is depicted in Scheme 8.^[57]It should be emphas-

ized, however, that this scheme represents an oversimpli-

fication since other possibly reacting species have not been

considered and because of the limiting assumptions which

have been made above.

Scheme 8. Simplified kinetic scheme for the reaction of the endo

and exo conformers of carbanions 3 and 4 with electrophiles

Now the question has to be addressed as to why endo-

3a,b (endo-4a,b) should be more reactive towards electro-

philes than exo-3a,b (exo-4a,b). It has long been recognized

that the exo attack of electrophiles on norbornenes^[58]is

preferred over the endo attack. Several explanations for the

exo preference such as the torsional effect, the exo pyrami-

dalization, the endo deformability and the staggering effect

have been advanced.^{[30a,34cϪ34e,34h,34i,59]}According to

Figure 9. Modell for the trajectory of attack of electrophiles on the

exo face of endo-3(ϪLi^ϩ) (a) and on the endo face of exo-3(ϪLi^ϩ)

HoukЈs staggering effect, the exo attack on norbornene is

preferred over the endo attack because in the transition state

(b); the views are constructed from the calculated coordinates of

endo-3(ϪLi^ϩ) and exo-3(ϪLi^ϩ) by picturing the trajectory ma-

of the exo attack the developing C2ϪR bond and the nually

1640

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

emphasized, however, that although the staggering effect of- of 64% endo-5 and 36% endo-8a containing none of the exo

fers a rationalization for the selectivity of reaction of 3 and isomer exo-8a (Table 10). In a similar experiment exo-5 was

4 with electrophiles, it may well be that other not yet re- treated with 2.5 mol-% NaOCD₃in CD₃OD to give, after

vealed factors are contributing, too.

a reaction time of 21 h, a mixture of 42% exo-5 and 58%

The higher selectivities of alkylation of 4 as compared to exo-8a uncontaminated by the endo isomer endo-8a (Table

3 could be ascribed within this scheme to the endo attack 11). In both cases a mixture of exo-8a and endo-8a in a

on exo-4a being even further disfavored as compared to the ratio of 76:24 was obtained after a prolonged reaction time

exo attack on endo-4a because of a shielding of the endo of 56.5 h. Similar observations were made in the case of the

face of the C2 atom in exo-4a by 6-H_endo(cf. Scheme 6).^[1c]H/D exchange of the norbornyl sulfones endo-6 and exo-6.

Finally, it is tempting to ascribe the higher reactivity of the While the endo sulfone endo-6 gave after a reaction time of

norbornenyl species 3 as compared to the norbornyl species 8 h a mixture of 31% endo-6 and 69% endo-9a and none of

4 in alkylation to the higher degree of the pyramidalization exo-9a (Table 12), the exo sulfone exo-6 yielded, after a re-

of its C2 atom (vide supra).

action time of 29 h, a mixture of 49% exo-6 and 51% exo-

9a uncontaminated by endo-9a (Table 13). In both cases the

equilibrium mixture contained exo-9a and endo-9a in a ra-

tio of 86:14. These results show, that the base-catalyzed H/

e. H/D Exchange of the Sulfones

Although the endo/exo selectivities of deprotonation of D exchange of endo-5, exo-5, endo-6 and exo-6, which is

sulfones exo-5, endo-5, endo-6 and exo-6 with nBuLi should faster than diastereomerization, takes place under a high

have only little bearing upon the selectivity of reaction of 3 degree of retention of configuration. Thus, at least under

and 4 with electrophiles, their knowledge would, neverthe-

less, be of interest. It had been observed previously that

deprotonation of chiral acyclic tert-butyl and trifluoro-

methyl sulfones with organolithium compounds generally

occurs with high selectivity apparently in a CαϪS confor-

mation in which the α-H atom is gauche to both O

atoms.^[25f,53]This was rationalized on the basis of an intra-

molecular deprotonation following complexation of RLi by

the sulfonyl O atom^[60]in combination with a preferred

CαϪS conformation resembling that of the α-sulfonyl car-

banion formed. Thus in the case of sulfones exo-5 (exo-6)

and endo-5 (endo-6) transition states of type exo-10 (exo-

11) and endo-10 (endo-11) (Scheme 9), respectively, would

have to be considered. However, judged by molecular mod-

els these transition states appear to be less likely because

of a steric interaction between the phenyl group and the

methylene (C7) and ethano/etheno (C5, C6) bridges, respec-

tively. This seemed to be an especially serious argument if

one considers that in the transition state of sulfone depro-

Scheme 9. Transition state models for the base-catalyzed H/D ex-

tonation only

a minor structural reorganization oc-

change of sulfones exo-5, endo-5, and exo-6

Table 10. H/D-Exchange of sulfone endo-5

curs.^[64][65]Cram et al.,^[61]Corey et al.^[62]and Goering et

al.^[63]had studied the H/D exchange of optically active

acyclic sulfones with alkoxides in deuterated alcohols and

observed a high degree of retention of configuration. This

observation was rationalized by proposing a retention

mechanism which features (i) an intramolecular depro-

tonation of the sulfone in a conformation in which the H

atom is gauche to both O atoms following a coordination

of the base and the deuterated solvent molecules to the O

atoms, and (ii) an intramolecular transfer of a D atom syn

to the O atoms before CαϪS rotation and, thus, racemi-

zation can take place. It was thus of interest to determine

the stereochemistry of the H/D exchange of sulfones exo-5,

endo-5, endo-6 and exo-6 in order to obtain information

as to the likelihood of transition states of deprotonation,

featuring a C2ϪS conformation in which the 2-H atom bi-

sects the OϪSϪO angle. Treatment of the norbornenyl sul-

fone endo-5 with 2.5 mol-% NaOCD₃in CD₃OD at 25°C

led, after a reaction time of 4.5 h, to isolation of a mixture

t [h]

endo-5 (%)

endo-8a (%)

exo-8a (%)

0

100

93

64

13

Ϫ

7

Ϫ

9

18

38

61

69

71

76

0.67

2.5

4.5

12

36

78

82

62

39

31

29

24

21

27.5

36

43

46.5

56.5

Ϫ

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1641

H.-J. Gais et al.

FULL PAPER

Table 11. H/D-Exchange of sulfone exo-5

the conditions employed, the attainment of transition states

of deprotonation of type exo-12 (exo-13) and endo-12

(endo-13), in which the H-2 atom bisects the OϪSϪO angle,

seems feasible.

Conclusion

t [h]

exo-5 (%)

exo-8a (%)

endo-8a (%)

The lithioallyl sulfones 3, 4, and 7 crystallize preferen-

tially as dimeric contact ion pairs whose Li atoms are coor-

dinated to the O atoms and not to the allylic moieties, de-

spite a favorable arrangement of the allylic moiety and the

sulfonyl group in the first two cases. This is in sharp con-

trast to the known crystal structures of other heteroatom-

substituted allyllithium compounds, where the Li atoms

make contact, not only to the heteroatoms, but also to the

allylic moieties.^[66]The anions of the three lithioallyl sul-

fones adopt the typical CαϪS conformation of α-sulfonyl

carbanions and feature reduced CαϪS bond lengths. Their

allylic moieties are characterized by non-planar anionic C

atoms and by bond alternation. The degree of pyramidali-

zation corresponds to a hybridization approximately half-

way between sp²and sp³and the direction of pyramidali-

zation is such that the apex of the pyramid at Cα is syn to

the O atoms. In the crystal both anions of 3 and 4 prefer

the endo conformation and possess in the exo direction py-

ramidalized anionic C atoms. The C2 atoms of endo-3/ent-

endo-3·2diglyme and endo-4/ent-endo-4·2diglyme show a

different degree and the same direction of pyramidalization

which points to a minimization of strain energy as the com-

mon cause for this difference. This is supported by the ab

initio calculations of the parent anions II and IV. The crys-

tallographic results are nicely corroborated by the ab initio

calculation of the free anions endo-3(ϪLi^ϩ), exo-3(ϪLi^ϩ),

endo-4(ϪLi^ϩ) and exo-4(ϪLi^ϩ), which revealed similar de-

grees but opposite directions of the pyramidalization of the

C2 atoms.

0

100

98

91

89

66

42

4

Ϫ

2

Ϫ

13

21

22

24

1

3

9

4.5

12.5

21

11

34

58

83

79

78

76

27.5

36

Ϫ

43

Ϫ

56.5

Table 12. H/D-Exchange of sulfone endo-6

t [h]

endo-6 (%)

endo-9a (%)

exo-9a (%)

0

100

98

81

59

45

31

10

Ϫ

2

Ϫ

10

17

23

36

67

72

86

0.33

1.75

3

19

41

55

69

80

83

77

64

33

28

14

5

8

14.5

24

28.5

39.5

65

72

91.5

Ϫ

In THF solution the structural picture of 3 and 4 is much

more complex. Monomeric and dimeric OϪLi contact ion

pairs, having endo and exo anions, are in rapid equilibrium,

even at low temperatures, whereby the endo species seem

to be the preferred one. This is supported by the ab initio

calculations of the free anions which revealed small energy

differences in favor of the endo isomers. Based on the NMR

data the extra stabilization of the negative charge of 3 and

4 by the double bond seems to be due to allylic delocali-

zation. However, because of the charge-alternating struc-

ture O^ϪϪ(O^Ϫ)S^ϩϪC2^ϪϪC3^ϩϪC8^Ϫ, internal Coulomb in-

teraction may provide an additional and important stabiliz-

ation. Such charge-alternating structures were also located

by the ab initio calculations of endo-3(ϪLi^ϩ) and endo-

4(ϪLi^ϩ), exo-3(ϪLi^ϩ) and exo-4(ϪLi^ϩ).

Table 13. H/D-Exchange of sulfone exo-6

t [h]

exo-6 (%)

exo-9a (%)

endo-9a (%)

0

100

98

95

93

80

59

49

28

4

Ϫ

2

Ϫ

7

0.67

2

5

8

15

24.5

29

39.5

65.5

72

7

Both lithiosulfones 3 and 4 react with electrophiles with

high selectivities under formation of the corresponding

endo-sulfones. Thus, the previous report on a selective endo-

methylation of 4^[6e]has to be revised. The similar endo/exo

selectivities of reaction of 3 and 4 with electrophiles can be

rationalized by assuming that (i) endo and exo species are

20

41

51

65

84

86

12

14

Ϫ

1642

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

eters used in the X-ray measurements and in the crystal structure

analyses are reported in Tables 14 and 15. The crystal structures

of endo-5, exo-5 and endo-6 were solved using direct methods as

implemented in the XTAL3.2 package of crystallographic rou-

tines,^[75]employing GENSIN^[76]to generate structure-invariant re-

lationships and GENTAN^[77]for the general tangent phasing pro-

cedure. The crystal structures of endo-3/ent-endo-3·2diglyme and

endo-4/ent-endo-4·2diglyme were solved by direct methods and re-

fined using SHELX-97.^[78]Experimental and theoretical molecular

structures were visualized with the program SCHAKAL 92.^[79]

configurationally labile on the time scale defined by the rate

of reaction with electrophiles, (ii) endo anions react with

electrophiles highly selective from the exo side and exo

anions from the endo side, and (iii) endo anions are more

reactive than exo anions. It will be interesting to see

whether the replacement of the phenyl group in 3 and 4 by

a tert-butyl group will lead, as in other cases, to anions

which are conformationally stable on the time scale defined

by the rate of their reaction with electrophiles.

Cryoscopy: The cryoscopic experiments were carried out according

to Bauer and Seebach.^[39]Temperature measurements were made

by using a Pt 100 platin thermo-couple of type 42943 (Burster

Präzisionsmeßtechnik, Gensbach) in connection with a high-pre-

cision thermometer Kelvimat Typ 4321 (Burster Präzisionsmeß-

technik, Gensbach) (± 0.002 K). Every 3 s the temperature was

registrated by a personal computer. For the measurements, the

solution of the lithiosulfone was filled with a double-ended needle

into the cryoscopic vessel up to a gauge mark (approximately 60

mL) and kept under argon at atmospheric pressure. The cryoscopic

vessel was placed in a conical glass jacket, which was evacuated to

10^Ϫ1mbar and cooled by dipping into liquid nitrogen. A cooling

rate of initially 5 K/min resulted, which decreased to 1 K/min at

the freezing point. Several independent measurements of the melt-

Experimental Section

Computational Methods: The geometries of the anions were optim-

ized at the HF level of ab initio theory employing the 6Ϫ31ϩG*

basis set,^[67][68]while the 6Ϫ31G* set of gaussians was used for the

sulfones.^[69Ϫ72]In order to include correlation energy single point

calculations at the HF/6Ϫ31ϩG*- and HF/G-31G*-optimized

structures using Møller-Plesset perturbation theory^[73]to the se-

cond order (MP2/basis set//HF/basis set) were performed. All ab

initio calculations were carried out employing the GAUSSIAN 94

set of quantum chemical routines^[74]running on an SNI-s600/20

computer of the Rechenzentrum at the RWTH Aachen.

X-ray Analyses: Suitable single crystals of endo-3/ent-endo-

3·2diglyme and endo-4/ent-endo-4·2diglyme were sealed in capillary ing point of THF gave a value of 164.584 ± 0.0016 K. The cryo-

tubes. The crystal data and the most salient experimental param- scopic constant E_Kϭ 1.856 ± 0.016 Kиkg/mol was determined by

Table 14. Crystal data and parameter of data collection for the sulfones endo-5, exo-5 and endo-6

Compound

endo-5

exo-5

endo-6

(a) Crystal data

Formula

M_r

C₁₄H₁₄O₂S

246.33

C₁₄H₁₄O₂S

246.33

C₁₄H₁₆O₂S

248.35

Color and habit

Crystal size [mm]

Crystal system

Space group

colorless, irregular

0.3 ϫ 0.3 ϫ 0.5

orthorhombic

P2₁2₁2₁(No. 19)

9.5497(9)

11.2103(9)

11.5464(7)

90

colorless, irregular

0.3 ϫ 0.3 ϫ 0.5

orthorhombic

P2₁2₁2₁(No. 19)

8.0072(5)

11.393(1)

13.622(1)

90

colorless, irregular

0.3 ϫ 0.3 ϫ 0.3

monoclinic

P2₁/a (No. 14)

12.185(3)

8.968 (1)

12.305(3)

90

˚

a, [A]

˚

b, [A]

˚

c, [A]

α [°]

β [°]

γ [°]

90

111.497(9)

90

3]

˚

V [A

1236.1

1242.68

1251.07

Z

4

D_calcd.[g cm^Ϫ3

]

1.323

1.317

1.318

µ [cm^Ϫ1

]

21.68

21.57

21.43

(b) Data collection

diffractometer

T [°C]

CDA4 Enraf-Nonius

25

CDA4 Enraf-Nonius

25

CDA4 Enraf-Nonius

25

Radiation

Cu-K_α

1.54179

graphite

Ω/2Θ

Cu-K_α

1.54179

graphite

Ω/2Θ

Cu-K_α

1.54179

graphite

Ω/2Θ

˚

λ [A]

Monochromator

Scan method

Θ_max[°]

75.2

No. of data collected

No. of unique data

Observation criterion

(c) Refinement

3297

5785

5898

1492

1506

2750

I > 2 σ(I)

No. of parameters refined

155

154

No. of data observed in refinement

1074

1288

2079

[a]

R, R_w

0.056, 0.047

not refined

Ϫ0.4/ϩ0.3

2.522

0.046, 0.040

not refined

Ϫ0.4/ϩ0.3

2.401

0.054, 0.049

not refined

Ϫ0.4/ϩ0.3

2.058

Extinction r*

Ϫ3

˚

∆(ρ) [e A

]

GOF

^[a]R ϭ (ΣԽԽF_oԽ Ϫ ԽF_cԽԽ)/ΣԽF_oԽ; R_wϭ (Σw(ԽF_oԽ Ϫ ԽF_cԽ)²/ΣwԽF_oԽ²)^0.5; w ϭ 1/σ²(F_o) where F_oand F_care observed and calculated structure factors.

Eur. J. Org. Chem. 1999, 1627Ϫ1651 1643

H.-J. Gais et al.

FULL PAPER

0.063Ϫ0.200 mm. Ϫ TLC: Merck silica gel 60 F₂₅₄plates. Ϫ El-

ementary analyses: Microanalytical laboratories of the Institut für

Organische Chemie at the RWTH Aachen and the Institut für Or-

ganische Chemie und Biochemie at the University of Freiburg.

Table 15. Crystal data and parameter of data collection for the

lithiosulfones 3 and 4

Compound

3

4

(±)-endo- and (±)-exo-3-Methylene-2-(phenylsulfonyl)bicyclo[2.2.1]-

hept-5-ene (endo-5 and exo-5): 1,2-Phenyl propadienyl sulfone

(12.6 g, 0.07 mol) and cyclopentadiene (13.2 g, 0.20 mol) in toluene

(150 mL) were heated to reflux for 2 h. Evaporation of the solvent

gave a mixture of endo-5 and exo-5 in a ratio of 2:1 as an orange

oil. Chromatography (EtOAc/n-hexane, 1:2) and crystallization

from EtOAc/n-hexane (1:5) afforded endo-5 (7.6 g, 44%) and exo-5

(3.6 g, 21%) as colorless crystals.

endo-5: M.p. 64°C. Ϫ MS; m/z (%): 246 [M^ϩ] (12), 121 (39), 106

(10), 105 (100), 103 (18), 79 (28), 78 (11), 77 (42), 66 (14), 51 (16),

44 (11), 40 (23), 39 (14). Ϫ C₁₄H₁₄O₂S (246.3): calcd. C 68.27, H

5.73; found C 68.49, H 5.80.

(a) Crystal data

Formula

(C₂₀H₂₇LiO₅S)₂

772.94

(C₂₀H₂₉LiO₅S)₂

776.98

M_r

Color and habit

Crystal size [mm]

Crystal system

Space group

yellow, flat prism

1.30 ϫ 0.62 ϫ 0.25 1.62 ϫ 0.65 ϫ 0.20

monoclinic

P21/c

triclinic

PϪ1 (No. 2)

11.118(8)

10.840(8)

9.388(8)

80.62(3)

76.45(3)

76.50(3)

1062.5(14)

2

˚

a [A]

9.161(9)

21.240(3)

12.690(2)

90.0

˚

b [A]

˚

c [A]

α [°]

β [°]

γ [°]

122.18(5)

90.0

3

˚

V [A ]

2089.9(21)

4

Z

D_calcd.[g cm^Ϫ3

]

1.228

0.171

1.214

0.169

exo-5: M.p. 53°C. Ϫ MS; m/z (%): 246 [M^ϩ] (3), 121 (39), 106 (11),

105 (100), 103 (13), 79 (36), 77 (37), 65 (6), 51 (9), 39 (5). Ϫ

C₁₄H₁₄O₂S (246.3): calcd. C 68.27, H 5.73; found C 68.00, H 5.66.

µ [cm^Ϫ1

]

(b) Data collection

Diffractometer

T [°C]

Stoe-STADI4

23

Stoe-STADI4

25

(±)-endo-3-Methylene-2-(phenylsulfonyl)bicyclo[2.2.1]heptane

(endo-6): A mixture of Pd/C (10%) (1 g) and endo-5 (0.500 g,

2.0 mmol) in MeOH (70 mL) was vigorously stirred under H₂at

room temp. After 17 min, the theoretical amount of H₂had been

taken up and hydrogenation was terminated by removal of the cata-

lyst by filtration through Celite. Concentration of the solution in

vacuo gave endo-6 (360 mg, 72%) as colorless crystals. Ϫ M.p.

100°C (EtOAc/n-hexane, 1:3). Ϫ MS; m/z (%): 248 [M^ϩ] (2), 220

(5), 108 (10), 107 (87), 105 (11), 91 (41), 80 (13), 79 (100), 78 (22),

77 (78), 65 (16), 53 (15), 51 (9). Ϫ C₁₄H₁₆O₂S (248.3): calcd. C

67.71, H 6.49; found C 67.33, H 6.50.

Radiation

Mo-K_α

0.71069

graphite

Ω/2Θ

Mo-K_α

0.71069

graphite

Ω

˚

λ [A]

Monochromator

Scan method

Θ_max[°]

20.03

20.09

No. of data collected 2606

No. of unique data

2449

1957

2007

Observation criterion I > 2 σ(I)

No. of data observed 1571

(c) Refinement

I > 2 σ(I)

1743

No. of parameters

refined

325

331

[a]

R, R_w

0.0527, 0.1382

not refined

Ϫ0.3/ϩ0.2

1.061

0.0383, 0.1067

not refined

Ϫ0.2/ϩ0.2

1.031

Extinction r*

∆(ρ) [e A

Ϫ3

˚

(±)-exo-3-Methylene-2-(phenylsulfonyl)bicyclo[2.2.1]heptane (exo-

6): A mixture of Pd/C (10%) (1 g) and exo-5 (0.500 g, 2.0 mmol) in

MeOH (70 mL) was vigorously stirred under H₂at room temp.

After 15 min, the theoretical amount of H₂had been taken up and

hydrogenation was terminated by removal of the catalyst by fil-

tration through Celite. Concentration of the solution in vacuo gave

exo-6 (420 mg, 84%) as colorless crystals. Ϫ M.p. 71°C (EtOAc/n-

hexane, 1:3). Ϫ MS; m/z (%): 248 [M^ϩ] (4), 220 (3), 109 (12), 108

(8), 107 (100), 91 (12), 79 (57), 78 (6), 77 (16), 65 (3), 51 (5). Ϫ

C₁₄H₁₆O₂S (248.3): calcd. C 67.71, H 6.49; found C 67.49, H 6.54.

]

GOF

[a]

R ϭ (ΣԽԽF_oԽ Ϫ ԽFcԽԽ)/ΣԽF_oԽ; R_wϭ (Σw(ԽF_oԽ Ϫ ԽF_cԽ)²/ΣwԽF_oԽ²)^0.5

;

w ϭ 1/σ²(F_o) where F_oand F_care observed and calculated struc-

ture factors.

measurement of solutions of naphthalene and trans-stilbene in

THF (c ϭ 0.044Ϫ0.045 mol/kg).

General: All reactions were carried out in absolute solvents under

argon with syringe and Schlenk techniques in oven-dried glassware.

Glass tubes containing solutions of lithiosulfones for NMR spec-

troscopy were sealed under argon. THF and ether were distilled

under argon from potassium/benzophenone and sodium/benzo-

phenone, respectively. Toluene was distilled from sodium and meth-

anol was dried with magnesium and distilled. Ϫ ¹H, ¹³C and ⁶Li

NMR: Varian VXR 300, Varian Gemini 300, Varian Unity 500,

Bruker AC-300 spectrometers, chemical shifts reported relative to

TMS (¹H, ¹³C) and LiBr (⁶Li), NMR chemical shifts of minor iso-

mers obtained as mixture with major isomers given only when an

unequivocal assignment was possible. Ϫ GC: Chrompack CP-9000

(DB-5: 30 m, 0.32 mm; 50 kPa H₂; S1: 100°C, 5 min, 20°C/min,

250°C, 5 min, 30°C/min, 300°C, 15 min; S2: 50°C, 5 min, 30°C/

Lithium·2digylme (±)-3-Methylene-2-(phenylsulfonyl)bicyclo[2.2.1]-

hept-5-ene-2-ide (endo-3/ent-endo-3·2diglyme): A solution of exo-5

(100 mg, 0.40 mmol) in ether (2 mL) was added at Ϫ78°C to a

solution of nPrLi (0.5 mL, 0.84  in n-hexane, 0.40 mmol) in ether

(2 mL), whereby the solution turned orange. After addition of di-

glyme (4 mL, 28 mmol) to the solution, it was warmed to room

temp. and concentrated in vacuo until most of n-hexane and ether

had been removed. The resulting suspension was warmed to 50°C,

until all of the solid was dissolved. The solution was allowed to

cool slowly to room temp. and kept for 2 d at this temp. The solid

was removed by filtration, washed with n-hexane and dried in va-

cuo to give endo-3/ent-endo-3·2diglyme (110 mg, 71%) as yellow

6

crystals. Ϫ Li NMR (44.17 MHz, [D₈]THF): δ ϭ Ϫ0.74 (s).

min, 150°C, 2 min, 20°C/min, 250°C, 2 min, 10°C/min, 300°C, Lithiumи2diglyme (±)-3-Methylene-2-(phenylsulfonyl)bicyclo[2.2.1]-

15 min) and Carlo Erba Mega Series 5300 [permethyl-β-cyclodex- heptane-2-ide (endo-4/ent-endo-4·2diglyme): A solution of exo-6

trin/polysiloxane (β-CD): 25 m, 0.25 mm; 100 kPa H₂; 100°C, (100 mg, 0.40 mmol) in ether (2 mL) was added at Ϫ78°C to a

2 min, 150°C, 5 min, 170°C, 5 min, 200°C, 30 min] instruments. Ϫ solution of nPrLi (0.5 mL, 0.84  in n-hexane, 0.40 mmol) in ether

GC MS: Magnum Finnigan (HT-5: 25 m, 0.25 mm; 50 kPa He, CI,

40 eV, MeOH). Ϫ MS: Varian MAT 212S (EI, 70 eV), decisive sig-

nals of the MS spectra and those with an intensity higher than

(2 mL) whereby the solution turned red. After addition of diglyme

(4 mL, 28 mmol) to the solution, it was warmed to room temp. and

concentrated in vacuo until most of n-hexane and ether had been

10% are listed. Ϫ Column chromatography: Merck silica gel 60, removed. The resulting suspension was warmed to 50°C, until all

1644 Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

of the solid was dissolved. The solution was allowed to cool slowly

to room temp. and kept at this temp. for 2 d. The solid was removed

by filtration, washed with n-hexane and dried in vacuo to give endo-

4/ent-endo-4·2diglyme (134 mg, 81%) as yellow crystals. Ϫ ⁶Li

NMR (44.17 MHz, [D₈]THF): δ ϭ Ϫ0.76 (s).

The combined organic phases were washed with satd. aqueous

NaHCO₃(40 mL), dried (MgSO₄) and concentrated in vacuo. Puri-

fication of the residue by chromatography (EtOAc/n-hexane, 1:2)

afforded a mixture of the corresponding endo and exo sulfones as

colorless solids or oils.

Determination of Aggregation of Lithiosulfones 3 and 4: The sulfone

(2.70 mmol) was placed in an argon-filled weighed Schlenk flask

and THF (10 mL) was added followed by the dropwise addition of

nBuLi (1.69 mL, 1.6  in n-hexane) at Ϫ78°C. After warming the

solution to room temp., the volatiles were removed in vacuo and

the residue dried for 10 min at 10^Ϫ3mbar. The amount of lithiosul-

fones was determined by differential weighing. The amount of co-

ordinated THF was determined by ¹H-NMR spectroscopy in an

independent experiment. The lithiosulfone was dissolved by ad-

dition of THF (63 mL) and the flask was weighed again. Thus,

solutions with c_nomϭ 0.0497Ϫ0.0503 mol/kg were obtained. In

each cryoscopic experiment the solution was frozen and thawed

and the cooling curves were measured at least seven times, whereby

attention was paid that at the cooling branch no precipitation of

the lithiosulfone occurred. After the registration of the cooling

curves, the solution was warmed to Ϫ80°C and treated with

CF₃COOD (5 mL, 10 mmol). The solution was washed with satd.

aqueous NaHCO₃and brine, dried (MgSO₄) and concentrated in

vacuo to give the starting sulfone, whose degree of deuteration (C2)

(±)-endo- and (±)-exo-2-Deuterio-3-methylene-2-(phenylsulfonyl)-

bicyclo[2.2.1]hept-5-ene (endo-8a and exo-8a): CF₃COOD (0.09 mL,

1.2 mmol) or DCl/D₂O (0.2 mL, 10 M, 2 mmol) were added to the

solution of 3, which became immediately colorless. Workup gave a

mixture of endo-8a and exo-8a (47 mg, 95%) in a ratio of 87:13

(GC) or a mixture of endo-8a and exo-8a (45 mg, 91%) in a ratio

of 88:12 (GC) as colorless oils.

1

endo-8a: GC: t_Rϭ 13.01 min (DB-5, S2). Ϫ H NMR(300 MHz,

CDCl₃): δ ϭ 1.45 (dm, J_7s,7aϭ 8.7 Hz, 1 H, H-7_s), 1.64 (dm,

J_7a,7sϭ 8.7 Hz, 1 H, H-7_a), 3.03 (s, 1 H, H-1), 3.30 (s, 1 H, H-4),

5.26 (s, 1 H, H-8_a), 5.33 (s, 1 H, H-8_s), 6.06 (dd, J_6,5ϭ 5.7, J_6,1

ϭ

2.6 Hz, 1 H, H-6), 6.14 (dd, J_5,6ϭ 5.7, J_5,4ϭ 3.3 Hz, 1 H, H-5),

7.55 (tm, 2 H, H-11,13), 7.65 (tm, 1 H, H-12), 7.89 (dm, 2 H, H-

10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 45.34 (C-1), 49.77 (C-

7), 52.26 (C-4), 110.01 (C-8), 128.67 (C-10,14), 128.99 (C-11,13),

133.06 (C-6), 133.59 (C-12), 135.09 (C-5), 139.43 (C-9), 143.11 (C-

3). Ϫ GC MS; m/z (%): 248 [M^ϩϩ 1] (5), 184 (8), 122 (3), 106 (8),

89 (100), 61 (46).

exo-8a: GC: t_Rϭ 12.90 min (DB-5, S2). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.51 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 2.03 (dm,

J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 3.17 (s, 1 H, H-1), 3.29 (s, 1 H, H-4),

1

was determined by H-NMR spectroscopy. The melting point de-

pression ∆T of each solution was determined by linear regressions

of the successive cooling curves and by averaging the results. The

equation n ϭ c_nomиE_K/∆T was used for determination of the degree

of aggregation (Table 16).

5.13 (d, 1 H, H-8_s), 5.29 (s, 1 H, H-8_a), 6.11 (dd, J_6,5ϭ 5.3, J_6,1

ϭ

3.0 Hz, 1 H, H-6), 6.28 (dd, J_5,6ϭ 5.3, J_5,4ϭ 3.0 Hz, 1 H, H-5),

7.58 (tm, 2 H, H-11,13), 7.67 (tm, 1 H, H-12), 7.93 (dm, 2 H, H-

10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 45.65 (C-1), 46.29 (C-

7), 50.67 (C-4), 111.15 (C-8), 128.88 (C-10,14), 129.16 (C-11,13),

133.68 (C-12), 135.55 (C-6), 139.04 (C-9), 139.94 (C-5), 142.83 (C-

3). Ϫ GC MS; m/z (%): 248 [M^ϩϩ 1] (16), 122 (15), 106 (15), 89

(100), 61(44).

The D content of the recovered sulfone was only in the range of

94.5Ϫ98%. Several factors including the use of an insufficient

amount of nBuLi, an incomplete deuteration and a hydrolysis of

the lithiosulfone during preparation and/or measurement could be

responsible for the partial recovery of the H-sulfone. We think that

the major factor is hydrolysis of the lithiosulfone. We have thus

corrected the aggregation number under the assumption that for-

mation of the H-sulfone is solely caused by hydrolysis of the lithio-

sulfone under formation of H-sulfone and LiOH and that both are

monomeric in THF.

endo-8a/exo-8a: C₁₄DH₁₃O₂S (247.3): calcd. C 67.99, H 5.30; found

C 68.14, H 5.41.

(±)-endo- and (±)-exo-2-Deuterio-3-methylene-2-(phenylsulfonyl)-

bicyclo[2.2.1]-heptane (endo-9a and exo-9a): CF₃COOD (0.09 mL,

1.2 mmol) was added to the solution of 4, which became immedi-

ately colorless. Workup gave a mixture of endo-9a and exo-9a

(44 mg, 88%) in a ratio of 59:41 (GC) as a colorless oil.

General Procedure for Reaction of Lithiosulfones 3 and 4 with Elec-

trophiles: nBuLi (0.14 mL, 1.6  in n-hexane, 0.22 mmol) was added

dropwise at Ϫ78°C to a solution of endo-5 or exo-5 and endo-6 or

exo-6 (50 mg, 0.2 mmol) in THF (5 mL). After addition of the first

drop of the base, the solution turned immediately yellow and yel-

low/orange, respectively. Stirring was continued for 40 min to give

deep yellow and yellow/orange solutions of 3 and 4, respectively.

endo-9a: GC: t_Rϭ 23.48 min (β-CD). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.39 (sm, 2 H, H-7_s, H-7_a), 1.40 (m, 1 H, H-6), 1.58

(m, 1 H, H-5), 1.73 (m, 1 H, H-5), 2.37 (m, 2 H, H-1,H-6), 2.84 (s,

The electrophile was added to the solution of 3 and 4 at Ϫ78°C, 1 H, H-4), 5.14 (s, 1 H, H-8_a), 5.17 (s, 1 H, H-8_s), 7.57 (tm, 2 H,

whereby decolorization occurred except otherwise stated. After stir-

H-11,13), 7.65 (tm, 1 H, H-12), 7.94 (dm, 2 H, H-10,14). Ϫ ¹³C

ring the solution for 10Ϫ15 min at this temp., it was warmed to NMR (75 MHz, CDCl₃): δ ϭ 22.90 (C-6), 28.18 (C-5), 39.17 (C-

room temp. during 1.5 h. Subsequently, satd. aqueous NaHCO₃(20 7), 40.55 (C-1), 47.06 (C-4), 108.57 (C-8), 128.30 (C-10,14), 129.27

mL) was added and the mixture was extracted with ether (20 mL).

(C-11,13), 133.55 (C-12), 140.35 (C-9), 146.04 (C-3). Ϫ GC MS;

Table 16. Cryoscopy of lithiosulfones 3 and 4 in THF

[a]

Lithiosulfone

c_nom[mol/kg]

∆T [K]

c_exp[mol/kg]

n

D content of sulfone [%] n_cor

3

4

0.0497

0.0498

0.0503

0.0500

0.0503

0.058

0.071

0.064

0.062

0.058

0.0313

0.0383

0.0345

0.0334

0.0313

94.5

97

98

1.45±0.08^[b]

1.56±0.06^[b]

1.58±0.13^[b]

1.62±0.06^[b]

[a]

[b]

n_corϭ c_nomϪ c_H-sulfone/c_expϪ 2 c_H-sulfone. Ϫ Mean value.

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1645

H.-J. Gais et al.

FULL PAPER

m/z (%): 250 [M^ϩϩ 1] (77), 215 (31), 184 (2), 124 (5), 108 (100),

80 (8), 61(2).

exo-9b: GC: t_Rϭ 13.54 min (DB-5, S2). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.21 (dm, J_7a,7sϭ 10.0 Hz, 1 H, H-7_a), 1.29 (m, 1 H,

H-5), 1.52 (s, 3 H, Me), 1.53 (m, 2 H, H-6), 1.72 (m, 1 H, H-5),

2.32 (dm, J_7s,7aϭ 10.0 Hz, 1 H, H-7_s), 2.55 (s, 1 H, H-1), 2.80 (s,

1 H, H-4), 5.17 (s, 2 H, H-8_a, H-8_s), 7.55 (tm, 2 H, H-11,13), 7.64

(tm, 1 H, H-12), 7.91 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz,

CDCl₃): δ ϭ 21.54 (Me), 24.98 (C-6), 29.42 (C-5), 37.50 (C-7),

44.76 (C-1), 46.23 (C-4), 72.03 (C-2), 108.57 (C-8), 128.76 (C-

11,13), 130.53 (C-10,14), 133.54 (C-12), 137.67 (C-9), 153.53 (C-3).

Ϫ GC MS; m/z (%): 263 [M^ϩϩ 1] (9), 241 (8), 121 (100), 93 (14),

89 (33), 79 (10), 61 (21).

exo-9a: GC: t_Rϭ 22.66 min (β-CD). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.15 (dm, 1 H, H-7_s), 1.23 (m, 1 H, H-6), 1.33 (tm, 1

H, H-5), 1.66 (dm, 2 H, H-5, H-6), 1.86 (dm, 1 H, H-7_a), 2.64 (s,

1 H, H-1), 2.79 (s, 1 H, H-4), 5.14 (s, 1 H, H-8_s), 5.19 (s, 1 H, H-

8_a), 7.56 (tm, 2 H, H-11,13), 7.66 (tm, 1 H, H-12), 7.91 (dm, 2 H,

H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 28.43 (C-6), 29.84

(C-5), 36.05 (C-7), 40.27 (C-1), 45.32 (C-4), 110.19 (C-8), 129.07

(C-10,14), 129.15 (C-11,13), 133.68 (C-12), 139.00 (C-9), 146.81 (C-

3). Ϫ GC MS; m/z (%): 250 [M^ϩϩ 1] (84), 215 (25), 184 (2), 171

(3), 124 (3), 108 (100), 80 (13), 61(7).

endo-9b/exo-9b: C₁₅H₁₈O₂S (262.4): calcd. C 68.67, H 6.91; found

C 68.67, H 6.91.

endo-9a/exo-9a: C₁₄DH₁₅O₂S (249.3): calcd. C 67.44, H 6.06; found

C 67.20, H 6.03.

(±)-endo- and (±)-exo-2-Ethyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]hept-5-ene (endo-8c and exo-8c): Neat EtI (0.17 mL,

2.0 mmol) was added to the solution of 3, which became light yel-

low. The solution was stirred for 30 min at Ϫ78°C and warmed

within 15 min to Ϫ45°C whereby it turned colorless. Workup gave

a mixture of endo-8c and exo-8c (46 mg, 84%) in a ratio of 83:17

(GC) as a colorless solid.

(±)-endo- and (±)-exo-2-Methyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]hept-5-ene (endo-8b and exo-8b): Neat MeI (0.14 mL,

2.2 mmol) was added to the solution of 3, which turned immedi-

ately pale yellow. Workup gave a mixture of endo-8b and exo-8b

(44 mg, 85%) in a ratio of 77:23 (GC) as a colorless solid.

endo-8b: GC: t_Rϭ 23.63 min (β-CD). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.67 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 1.73 (dm,

J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 1.73 (s, 3 H, Me), 2.83 (s, 1 H, H-1),

3.28 (s, 1 H, H-4), 5.29 (s, 1 H, H-8_a), 5.42 (s, 1 H, H-8_s), 5.98 (dd,

1

endo-8c: GC: t_Rϭ 11.32 min (DB-5, S2). Ϫ H NMR (300 MHz,

CDCl₃): δ ϭ 1.20 (t, J ϭ 7.39 Hz; 3 H, Me), 1.59 (dm, J_7a,7s

ϭ

9.0 Hz, 1 H, H-7_a), 1.69 (dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 1.95 (m,

1 H, CH₂), 2.57 (m, 1 H, CH₂), 3.21 (s, 1 H, H-4), 3.28 (s, 1 H, H-

1), 5.25 (s, 1 H, H-8_a), 5.36 (s, 1 H, H-8_s), 5.67 (dd, J_5,6ϭ 5.3,

J_5,4ϭ 3.0 Hz, 1 H, H-5), 6.12 (dd, J_6,5ϭ 5.3, J_6,1ϭ 3.3 Hz, 1 H,

H-6), 7.42 (tm, 2 H, H-11,13), 7.54 (tm, 1 H, H-12), 7.83 (dm, 2

H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 10.24 (Me), 30.57

(CH₂), 48.22 (C-1), 48.67 (C-7), 52.07 (C-4), 77.85 (C-2), 110.35

(C-8), 127.91 (C-11,13), 130.31 (C-10,14), 132.97 (C-12), 135.67 (C-

5), 136.80 (C-6), 140.70 (C-9), 149.59 (C-3). Ϫ GC MS; m/z (%):

275 [M^ϩϩ 1] (1), 237 (1), 133 (100), 105 (4), 89 (4), 61 (2).

exo-8c: GC: t_Rϭ 11.33 min (DB-5, S2). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.06 (t, J ϭ 7.39 Hz; 3 H, Me), 1.40 (m, 1 H, CH₂),

1.51 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 2.06 (m, 1 H, CH₂), 2.26

(dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 3.26 (s, 1 H, H-4), 3.55 (s, 1 H,

H-1), 4.95 (s, 1 H, H-8_s), 5.22 (s, 1 H, H-8_a), 6.22 (dd, J_6,5ϭ 5.3,

J_6,1ϭ 2.6 Hz, 1 H, H-6), 6.27 (dd, J_5,6ϭ 5.3, J_5,4ϭ 3.0 Hz, 1 H,

H-5), 7.50 (tm, 2 H, H-11,13), 7.63 (tm, 1 H, H-12), 7.92 (dm, 2

H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 10.51 (Me), 31.04

(CH₂), 47.19 (C-7), 48.67 (C-1), 51.56 (C-4), 110.23 (C-8), 128.58

(C-11,13), 130.45 (C-10,14), 133.43 (C-12), 135.36 (C-5), 139.22 (C-

9), 139.98 (C-6), 150.08 (C-3). Ϫ GC MS; m/z (%): 275 [M^ϩϩ 1]

(5), 133 (2), 89 (100), 61 (44).

J_6,5ϭ 5.7 Hz, J_6,1ϭ 3.0 Hz, 1 H, H-6), 6.21 (dd, J_5,6ϭ 5.7, J_5,4

ϭ

3.0 Hz, 1 H, H-5), 7.50 (tm, 2 H, H-11,13), 7.58 (tm, 1 H, H-12),

7.87 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 24.67

(Me), 49.33 (C-7), 51.55 (C-1), 52.42 (C-4), 73.01 (C-2), 109.97 (C-

8), 128.49 (C-11,13), 130.14 (C-10,14), 133.38 (C-5), 134.94 (C-12),

136.40 (C-6), 138.97 (C-9), 149.46 (C-3). Ϫ GC MS; m/z (%): 261

(91) [M^ϩϩ 1], 237 (5), 209 (3), 197 (13), 181 (3), 135 (6), 119 (100),

91 (6).

exo-8b: GC: t_Rϭ 24.07 min (β-CD). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.40 (s, 3 H, Me), 1.55 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-

7_a), 2.57 (dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 3.10 (s, 1 H, H-1), 3.34

(s, 1 H, H-4), 5.19 (s, 1 H, H-8_s), 5.32 (s, 1 H, H-8_a), 6.09 (dd,

J_6,5ϭ 5.3, J_6,1ϭ 3.0 Hz, 1 H, H-6), 6.31 (dd, J_5,6ϭ 5.3, J_5,4

ϭ

3.0 Hz, 1 H, H-5), 7.56 (tm, 2 H, H-11,13), 7.66 (tm, 1 H, H-12),

7.93 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 24.23

(Me), 48.01 (C-7), 50.41 (C-1), 51.81 (C-4), 72.46 (C-2), 110.86 (C-

8), 128.81 (C-11,13), 130.47 (C-10,14), 133.62 (C-12), 135.90 (C-5),

137.97 (C-9), 139.83 (C-6), 148.76 (C-3). Ϫ GC MS; m/z (%): 261

[M^ϩϩ 1] (34), 233 (2), 197 (4), 183 (3), 135 (3), 119 (100), 91 (11),

89 (14), 61 (9).

endo-8b/exo-8b: C₁₅H₁₆O₂S (260.4): calcd. C 69.20, H 6.19; found

C 69.13, H 6.21.

endo-8c/exo-8c: C₁₆H₁₈O₂S (274.4): calcd. C 70.04, H 6.61; found

C 69.99, H 6.64.

(±)-endo- and (±)-exo-2-Methyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]heptane (endo-9b and exo-9b): Neat MeI (0.14 mL,

2.2 mmol) was added to the solution of 4, which immediately

turned pale yellow. Workup gave a mixture of endo-9b and exo-9b

(47 mg, 90%) in a ratio of 81:19 (GC) as a colorless solid.

(±)-endo- and (±)-exo-2-Ethyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]heptane (endo-9c and exo-9c): Neat EtI (0.17 mL,

2.0 mmol) was added to the solution of 4 which turned yellow. The

solution was stirred for 30 min at Ϫ78°C and warmed within

15 min to Ϫ45°C whereby it became pale yellow. Workup gave a

mixture of endo-9c and exo-9c (50 mg, 91%) in a ratio of 95:5 (GC)

as a colorless solid.

1

endo-9b: GC: t_Rϭ 13.47 min (DB-5, S2). Ϫ H NMR (300 MHz,

CDCl₃): δ ϭ 1.36 (dm, J_7a,7sϭ 10.0 Hz, 1 H, H-7_a), 1.44 (s, 3 H,

Me), 1.49 (m, 1 H, H-6), 1.63 (dm, J_7s,7aϭ 10.0 Hz, 1 H, H-7_s),

1

1.70 (m, 2 H, H-5), 2.23 (m, 1 H, H-1), 2.65 (m, 1 H, H-6), 2.85 endo-9c: GC: t_Rϭ 11.61 min (DB-5, S2). Ϫ H NMR (300 MHz,

(s, 1 H, H-4), 5.12 (s, 1 H, H-8_a,s), 5.13 (s, 1 H, H-8_a,s), 7.55 (tm, CDCl₃): δ ϭ 1.01 (t, J ϭ 7.39 Hz; 3 H, Me), 1.33 (dm, J_7s,7a

ϭ

2 H, H-11,13), 7.64 (tm, 1 H, H-12), 7.96 (dm, 2 H, H-10,14). Ϫ 10.0 Hz, 1 H, H-7_s), 1.50Ϫ1.90 (m, 6 H, H-5, H-5, H-6, H-7_a,

¹³C NMR (75 MHz, CDCl₃): δ ϭ 25.41 (Me), 25.58 (C-6), 26.74

CH₂), 1.60 (dm, J_7a,7sϭ 10.0 Hz, 1 H, H-7_a), 1.75 (m, 1 H, CH₂),

(C-5), 38.46 (C-7), 47.05 (C-1), 47.26 (C-4), 72.73 (C-2), 107.79 (C-

1.84 (m, 1 H, CH₂), 2.58 (m, 1 H, H-6), 2.65 (s, 1 H, H-4), 2.82 (s,

8), 128.47 (C-11,13), 130.15 (C-10,14), 133.43 (C-12), 138.60 (C-9), 1 H, H-1), 4.97 (s, 1 H, H-8_a,s), 5.13 (s, 1 H, H-8_a,s), 7.52 (tm, 2 H,

153.04 (C-3). Ϫ GC MS; m/z (%): 263 [M^ϩϩ 1] (7), 241 (7), 121

(100), 93 (10), 89 (5), 79 (4), 61 (4).

H-11,13), 7.63 (tm, 1 H, H-12), 7.97 (dm, 2 H, H-10,14). Ϫ ¹³C

NMR (75 MHz, CDCl₃): δ ϭ 9.81 (Me), 25.58 (CH₂), 27.28 (C-6),

1646

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

30.37 (C-5), 37.96 (C-7), 44.23 (C-1), 47.24 (C-4), 76.88 (C-2),

FULL PAPER

1), 47.23 (C-4), 76.93 (C-2), 107.75 (C-8), 128.56 (C-11,13), 130.10

108.06 (C-8), 128.64 (C-11,13), 130.24 (C-10,14), 133.37 (C-12), (C-10,14), 133.32 (C-12), 139.79 (C-9), 153.36 (C-3). Ϫ GC MS;

139.84 (C-9), 153.03 (C-3). Ϫ GC MS; m/z (%): 277 [M^ϩϩ 1] (33),

m/z (%): 291 [M^ϩϩ 1] (40), 149 (7), 125 (3), 89 (100), 61 (46).

135 (12), 125 (3), 89 (100), 61 (53).

exo-9d: GC: t_Rϭ 11.85 min (DB-5, S1). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 4.98 (s, 1 H, H-8_a,s), 5.13 (s, 1 H, H-8_a,s). Ϫ GC MS;

exo-9c: GC: t_Rϭ 11.88 min (DB-5, S2). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 1.14 (t, J ϭ 7.05 Hz; 3 H, Me) 4.93 (s, 1 H, H-8_a,s), m/z (%): 291 [M^ϩϩ 1] (1), 149 (100), 121 (2), 107 (4), 89 (17), 79

5.11 (s, 1 H, H-8_a,s), 7.87 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (4), 61 (10).

(75 MHz, CDCl₃): δ ϭ 128.90 (C-11,13), 129.49 (C-10,14). Ϫ GC

endo-9d/exo-9d: C₁₇H₂₂O₂S (290.4): calcd. C 70.31, H 7.63; found

MS; m/z (%): 277 [M^ϩϩ 1] (1), 137 (1), 125 (2), 89 (100), 61 (53).

C 70.03, H 7.69.

endo-9c/exo-9c: C₁₆H₂₀O₂S (276.4): calcd. C 69.53, H 7.29; found

(±)-endo- and (±)-exo-2-Hexyl-3-methylene-2-(phenylsulfonyl)bicy-

C 69.41, H 7.29.

clo[2.2.1]hept-5-ene (endo-8e and exo-8e): Neat nHeI (0.3 mL,

(±)-endo- and (±)-exo-3-Methylene-2-(phenylsulfonyl)-2-propylbi-

cyclo[2.2.1]hept-5-ene (endo-8d and exo-8d): Neat nPrI (0.2 mL, change. The solution was warmed within 35 min to Ϫ60°C whereby

2.0 mmol) was added to the solution of 3, which became light yel- it became colorless. Workup gave a mixture of endo-8e and exo-8e

2.0 mmol) was added to the solution of 3, whose color did not

low. The solution was warmed within 35 min to Ϫ60°C whereby it (53 mg, 80%) in a ratio of 88:12 (GC) as a colorless oil.

became colorless. Workup gave a mixture of endo-8d and exo-8d

(50 mg, 87%) in a ratio of 89:11 (GC) as a colorless solid.

endo-8e: GC: t_Rϭ 9.51 min (DBϪ5, 150°C, isotherm, 100 kPa H₂).

Ϫ H NMR (300 MHz, CDCl₃): δ ϭ 0.88 (m, 3 H, Me), 1.2Ϫ1.8

1

endo-8d: GC: t_Rϭ 9.50 min (DBϪ5, 150°C, isotherm, 100 kPa

H₂). Ϫ ¹H NMR (300 MHz, CDCl₃): δ ϭ 0.91 (t, J ϭ 7.3 Hz, 3 1.71 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 1.85 (m, 1 H, CH₂), 2.47 (m,

H, Me), 1.59 (dm, J_7s,7aϭ 8.7 Hz, 1 H, H-7_s), 1.71 (dm, J_7a,7s1 H, CH₂), 3.19 (s, 1 H, H-4), 3.27 (s, 1 H, H-1), 5.25 (s, 1 H, H-

8.7 Hz, 1 H, H-7_a), 1.75 (m, 2 H, CH₃ϪCH₂ϪCH₂), 1.84 (dt, J ϭ 8_a,s), 5.35 (s, 1 H, H-8_a,s), 5.68 (dd, J_5,6ϭ 5.3, J_5,4ϭ 3.3 Hz, 1 H,

(m, 10 H, H-7, H-7, CH₂), 1.59 (dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s),

ϭ

12.7, J ϭ 4.7 Hz, 1 H, CH₂), 2.44 (dt, 1 H, J ϭ 12.7, J ϭ 4.7 Hz,

CH₂), 3.19 (s, 1 H, H-4), 3.27 (s, 1 H, H-1), 5.25 (s, 1 H, H-8_a,s),

H-5), 6.13 (dd, J_6,5ϭ 5.3, J_6,1ϭ 3.0 Hz, 1 H, H-6), 7.42 (tm, 2 H,

H-11,13), 7.54 (tm, 1 H, H-12), 7.84 (dm, 2 H, H-10,14). Ϫ ¹³C

5.35 (s, 1 H, H-8_a,s), 5.67 (dd, J_5,6ϭ 5.3, J_5,4ϭ 3.3 Hz, 1 H, H-5), NMR (75 MHz, CDCl₃): δ ϭ 14.05 (Me), 22.61 (CH₂), 25.19

6.12 (dd, J_6,5ϭ 5.3, J_6,1ϭ 3.0 Hz, 1 H, H-6), 7.42 (tm, 2 H, H- (CH₂), 29.87 (CH₂), 31.52 (CH₂), 37.73 (CH₂), 48.65 (C-1, C-7),

11,13), 7.55 (tm, 1 H, H-12), 7.82 (dm, 2 H, H-10,14). Ϫ ¹³C NMR 52.06 (C-4), 77.71 (C-2), 110.18 (C-8), 127.91 (C-11,13), 130.27 (C-

(75 MHz, CDCl₃): δ ϭ 14.73 (Me), 18.77 (CH₂), 39.96 (CH₂), 48.78

10,14), 132.95 (C-12), 135.62 (C-5), 136.82 (C-6), 140.66 (C-9),

(C-1, C-7), 52.18 (C-4), 77.91 (C-2), 110.32 (C-8), 128.02 (C-11,13), 149.85 (C-3). Ϫ GC MS; m/z (%): 331 [M^ϩϩ 1] (2), 205 (1), 189

130.38 (C-10,14), 133.08 (C-12), 135.71 (C-5), 136.94 (C-6), 140.76 (100), 147 (2), 133 (4), 119 (4), 105 (3), 89 (17), 61 (10).

(C-9), 149.89 (C-3). Ϫ GC MS; m/z (%): 289 [M^ϩϩ 1] (2), 147

(100), 119 (1), 89 (1), 69 (1).

exo-8e: GC: t_Rϭ 9.43 min (DBϪ5, 150°C, isotherm, 100 kPa H₂).

Ϫ H NMR (300 MHz, CDCl₃): δ ϭ 0.88 (m, 3 H, Me), 1.2Ϫ1.8

1

exo-8d: GC: t_Rϭ 9.43 min (DBϪ5, 150°C, isotherm, 100 kPa H₂).

Ϫ ¹H NMR (300 MHz, CDCl₃): δ ϭ 0.78 (t, J ϭ 7.3 Hz; 3 H, Me), 1 H, H-4), 3.54 (s, 1 H, H-1), 4.97 (s, 1 H, H-8_s), 5.22 (s, 1 H, H-

1.50 (dm, J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 1.70 (m, 2 H, MeϪCH₂), 8_a), 6.19 (dd, J_6,5ϭ 5.3, J_6,1ϭ 3.0 Hz, 1 H, H-6), 6.27 (dd, J_5,6

1.84 (m, 1 H, CH₂), 2.23 (dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 2.44 (m, 5.3, J_5,4ϭ 2.3 Hz, 1 H, H-5), 7.42 (tm, 2 H, H-11,13), 7.55 (tm, 1

1 H, CH₂), 3.27 (s, 1 H, H-4), 3.55 (s, 1 H, H-1), 4.97 (s, 1 H, H-

H, H-12), 7.91 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃):

8_s), 5.22 (s, 1 H, H-8_a), 6.20 (dd, J_6,5ϭ 5.3, J_6,1ϭ 3.0 Hz, 1 H, H- δ ϭ 14.00 (Me), 22.60 (CH₂), 25.50 (CH₂), 29.77 (CH₂), 31.43

(m, 1 H, CH₂, H-7_a), 2.26 (dm, J_7s,7aϭ 9.7 Hz, 1 H, H-7_s), 3.27 (s,

ϭ

6), 6.27 (dd, J_5,6ϭ 5.3, J_5,4ϭ 3.3 Hz, 1 H, H-5), 7.51 (tm, 2 H, H-

(CH₂), 38.28 (CH₂), 47.22 (C-7), 48.51 (C-1), 51.46 (C-4), 110.18

11,13), 7.63 (tm, 1 H, H-12), 7.91 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (C-8), 128.55 (C-11,13), 130.40 (C-10,14), 133.43 (C-12), 135.36 (C-

(75 MHz, CDCl₃): δ ϭ 14.72 (Me), 19.09 (CH₂), 40.49 (CH₂), 47.35 5), 139.97 (C-6). Ϫ GC MS; m/z (%): 331 [M^ϩϩ 1] (1), 205 (1),

(C-7), 48.65 (C-1), 51.56 (C-4), 76.73 (C-2), 110.24 (C-8), 128.69

(C-11,13), 130.49 (C-10,14), 133.55 (C-12), 135.49 (C-5), 140.10 (C-

6). Ϫ GC MS; m/z (%): 289 [M^ϩϩ 1] (1), 147 (100), 119 (4), 105

(8), 89 (26), 69 (4), 61 (12).

189 (100), 147 (2), 133 (5), 119 (6), 105 (4), 89 (9), 61 (5).

endo-8e/exo-8e: C₂₀H₂₆O₂S (330.5): calcd. C 72.69, H 7.93; found

C 72.69, H 8.00.

(±)-endo- and (±)-exo-2-Hexyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]heptane (endo-9e and exo-9e): Neat nHeI (0.3 mL,

2.0 mmol) was added to the solution of 4 whose color did not

change. The solution was warmed slowly to Ϫ40°C whereby it be-

endo-8d/exo-8d: C₁₇H₂₀O₂S (288.4): calcd. C 70.80, H 6.99; found

C 70.63, H 6.99.

(±)-endo- and (±)-exo-3-Methylene-2-(phenylsulfonyl)-2-propylbi-

cyclo[2.2.1]heptane (endo-9d and exo-9d): Neat nPrI (0.2 mL, came colorless. Workup gave a mixture of endo-9e and exo-9e

2.0 mmol) was added to the solution of 4, whose color did not

change. The solution was warmed slowly to Ϫ40°C whereby it be-

came colorless. Workup gave a mixture of endo-9d and exo-9d

(55 mg, 95%) in a ratio of 97:3 (GC) as a colorless solid.

(54 mg, 81%) in a ratio of 97:3 (GC) as a colorless oil.

endo-9e: GC: t_Rϭ 8.97 min (DBϪ5, 150°C, isotherm, 100 kPa H₂).

Ϫ

¹H NMR (300 MHz, CDCl₃): δ ϭ 0.86 (t, 3 H, J ϭ 7.05 Hz,

Me), 1.00Ϫ1.90 (m, 15 H, CH₂, H-7, H-7, H-6, H-5, H-5), 1.34

1

endo-9d: GC: t_Rϭ 11.93 min (DB-5, S1). Ϫ H NMR (300 MHz, (dm, J_7s,7aϭ 10.0 Hz, 1 H, H-7_s), 1.67 (dm, J_7a,7sϭ 10.0 Hz, 1 H,

CDCl₃): δ ϭ 0.79 (t, J ϭ 7.0 Hz; 3 H, Me), 1.30Ϫ1.80 (m, 9 H, H-

5, H-5, H-6, H-7, H-7, CH₂), 1.34 (dm, J_7s,7aϭ 10.0 Hz, 1 H, H-

H-7_a), 2.59 (m, 1 H, H-6), 2.65 (s, 1 H, H-4), 2.81 (s, 1 H, H-1),

4.97 (s, 1 H, H-8_a,s), 5.11 (s, 1 H, H-8_a,s), 7.53 (tm, 2 H, H-11,13),

7_s), 1.67 (dm, J_7a,7sϭ 10.0 Hz, 1 H, H-7_a), 2.59 (m, 1 H, H-6), 2.65 7.63 (tm, 1 H, H-12), 7.96 (dm, 2 H, H-10,14). Ϫ ¹³C NMR

(s, 1 H, H-4), 2.81 (s, 1 H, H-1), 4.96 (s, 1 H, H-8_a,s), 5.11 (s, 1 H,

(75 MHz, CDCl₃): δ ϭ 14.04 (Me), 22.57 (CH₂), 24.69 (CH₂), 25.52

H-8_a,s), 7.52 (tm, 2 H, H-11,13), 7.63 (tm, 1 H, H-12), 7.96 (dm, 2 (C-6), 27.21 (C-5), 29.73 (CH₂), 31.52 (CH₂), 37.63 (C-7), 37.86

H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 14.43 (Me), 18.13

(CH₂), 25.54 (C-6), 27.20 (C-5), 37.86 (C-7), 39.75 (CH₂), 44.53 (C-

(CH₂), 44.51 (C-1), 47.23 (C-4), 76.84 (C-2), 107.77 (C-8), 128.55

(C-11,13), 130.10 (C-10,14), 133.32 (C-12), 139.80 (C-9), 153.37 (C-

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1647

H.-J. Gais et al.

FULL PAPER

2

3

2

3). Ϫ GC MS; m/z (%): 333 [M^ϩϩ 1] (19), 191 (100), 149 (6), 135 3.06 (dd, J ϭ 13.4, J ϭ 4.7 Hz, 1 H, CH₂), 3.31 (dd, J ϭ 13.4,

(18), 121 (14), 109 (8), 95 (11), 89 (73), 81 (8), 61 (43).

³J ϭ 8.3 Hz, 1 H, CH₂), 3.56 (sm, 1 H, H-4), 3.64 (sm, 1 H, H-1),

4.95 (m, 1 H, CH), 6.48 (m, 2 H, H-5, H-6), 7.30 (m, 1 H, H-12,

CPh), 7.36 (tm, 2 H, H-11,13, CPh), 7.43 (dm, 2 H, H-10,14, CPh),

7.50 (tm, 2 H, H-11,13), 7.58 (tm, 1 H, H-12), 7.75 (dm, 2 H, H-

10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 39.04 (C-7), 52.37 (C-

4), 57.53 (C-1), 71.48 (CH₂), 73.14 (CHOH), 125.79 (C-10,14),

127.28 (C-11,13), 127.76 (C-12), 128.52 (C-10,14, CPh), 129.11 (C-

11,13, CPh), 133.08 (C-12, CPh), 139.82 (C-2), 140.27 (C-5), 142.48

(C-6), 143.88 (C-9), 147.89 (C-9, CPh), 167.01 (C-3).

exo-9e: GC: t_Rϭ 8.88 min (DBϪ5, 150°C, isotherm, 100 kPa H₂).

1

Ϫ H NMR (300 MHz, CDCl₃): δ ϭ 0.81 (m, 3 H, CH₃), 4.96 (s,

1 H, H-8_a,s). Ϫ GC MS; m/z (%): 191 [M^ϩϪ SO₂Ph] (100), 149

(7), 135 (24), 121 (15), 109 (11), 95 (12), 89 (69), 81 (8), 69 (4),

61 (40).

endo-9e/exo-9e: C₂₀H₂₈O₂S (332.5): calcd. C 72.25, H 8.49; found

C 71.92, H 8.32.

(±)-(1R*,9R*)-endo-2-[(1-Hydroxy-1-phenyl)methyl]-3-meth-

ylene-2-(phenylsulfonyl)bicyclo[2.2.1]hept-5-ene and (±)-exo-2-[(1-

Hydroxy-1-phenyl)methyl]-3-methylene-2-(phenylsulfonyl)bicyclo-

[2.2.1]hept-5-ene (endo-8f and exo-8f): Freshly distilled PhCHO (0.2

mL, 2.0 mmol) was added to the solution of 3, which became im-

mediately colorless. After 5 min, a 1:1 mixture of AcOH and THF

(4 mL) was added at Ϫ 78°C. Workup gave a mixture of endo-8f

(single diastereomer) and exo-8f (single diastereomer) (62 mg, 88%)

in a ratio of 88:12 (GC) as a colorless solid.

γ-8f (Mixture of Diastereomers): C₂₁H₂₀O₃S (352.4): calcd. C 71.57,

H 5.72; found C 71.66, H 5.96.

(±)-3-[(2-Hydroxy-2-phenyl)ethyl]-2-(phenylsulfonyl)bicyclo-

[2.2.1]hept-2-ene (γ-9f) and (±)-endo-2-[(1-Hydroxy-1-phenyl)-

methyl]-3-methylene-2-(phenylsulfonyl)bicyclo-[2.2.1]-heptane (endo-

9f): Freshly distilled PhCHO (0.2 mL, 2.0 mmol) was added to the

solution of 4, which became immediately colorless. After 5 min, a

1:1 mixture of AcOH and THF (4 mL) was added to the solution

at Ϫ78°C. Workup gave a mixture of γ-9f (diastereomer 1), γ-9f

(diastereomer 2) and endo-9f (only one diastereomer) (56 mg, 79%)

in a ratio of 87:10:3 (GC) as a colorless solid.

1

endo-8f: GC: t_Rϭ 12.93 min (DB-5, S2). Ϫ H NMR (300 MHz,

CDCl₃): δ ϭ 0.73 (dm, J_7s,7aϭ 9.0 Hz, 1 H, H-7_s), 1.18 (dm,

J_7a,7sϭ 9.0 Hz, 1 H, H-7_a), 2.91 (s, 1 H, H-4), 3.63 (s, 1 H, H-1),

4.26 (d, J_OH,Hϭ 5.0 Hz, 1 H, OH), 5.47 (s, 1 H, H-8_a,s), 5.49 (m,

γ-9f (Diastereomer 1): GC: t_Rϭ 15.31 min (DB-5, S1).

1 H, H-5), 5.67 (s, 1 H, H-8_a,s), 5.83 (d, J_H,OHϭ 5.0 Hz, 1 H, CH), γ-9f (Diastereomer 2): GC: t_Rϭ 15.38 min (DB-5, S1).

6.19 (m, 1 H, H-6), 7.29 (m, 3 H, H-12, H-11,13, CPh), 7.41 (tm,

1

endo-9f: GC: t_Rϭ 15.55 min (DB-5, S1). Ϫ H NMR (300 MHz,

2 H, H-11,13), 7.49 (dm, 2 H, H-10,14, CPh), 7.54 (tm, 1 H, H-

CDCl₃): δ ϭ 0.90 (dm, 1 H, H-7_s), 2.49 (m, 2 H, H-1, H-6), 3.20

12), 7.86 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ

(sm, 1 H, H-4), 4.40 (d, J_OH,Hϭ 2.2 Hz, 1 H, OH), 4.97 (s, 1 H,

47.54 (C-1), 48.75 (C-7), 51.77 (C-4), 76.52 (CHOH), 85.01 (C-

H-8_a), 5.30 (s, 1 H, H-8_s), 8.11 (dm, 2 H, H-10,14).

2), 114.06 (C-8), 127.76 (C-11,13), 127.79 (C-10,14), 128.11 (C-12),

γ-9f (Mixture of Diastereomers)/endo-9f: C₂₁H₂₂O₃S (354.5): calcd.

128.62 (C-11,13, CPh), 131.04 (C-10,14, CPh), 133.25 (C-5), 138.35

(C-6), 138.47 (C-12, CPh), 139.81 (C-9), 141.07 (C-9, CPh), 145.18

(C-3).

C 71.16, H 6.26; found C 70.89, H 6.24.

(±)-3-[(2-Hydroxy-2-phenyl)ethyl]-2-(phenylsulfonyl)bicyclo-

[2.2.1]hept-2-ene (γ-9f): Freshly distilled PhCHO (0.2 mL,

2.0 mmol) was added to the solution of 4, which became immedi-

ately colorless. After warming the solution to room temp., it turned

yellow again. Workup gave γ-9f (47 mg, 66%) as a mixture of dia-

stereomers in a ratio of 69:31 (GC) as a colorless oil.

exo-8f: GC: t_Rϭ 12.82 min (DB-5, S2). Ϫ ¹H NMR (300 MHz,

CDCl₃): δ ϭ 0.48 (dm, J_7,7ϭ 9.0 Hz, 1 H, H-7_a,s), 0.95 (dm, J_7,7ϭ

9.0 Hz, 1 H, H-7_a,s), 2.99 (s, 1 H, H-4), 3.27 (sm, 1 H, H-1), 5.00

(d, J_OH,Hϭ 2.3 Hz, 1 H, OH), 5.57 (s, 1 H, H-8_a,s), 5.72 (s, 1 H,

H-8_a,s), 5.90 (d, J_H,OHϭ 2.3 Hz, 1 H, CH), 6.06 (m, 2 H, H-5, H-6).

γ-9f (Diastereomer 1): GC: t_Rϭ 15.28 min (DB-5, S1). Ϫ ¹H NMR

(300 MHz, CDCl₃): δ ϭ 1.03 (dm, J_7,7ϭ 8.7 Hz, 1 H, H-7_a,s), 1.04

(m, 1 H, H-5), 1.12 (m, 1 H, H-6), 1.36 (dm, J_7,7ϭ 8.7 Hz, 1 H,

endo-8f/exo-8f: C₂₁H₂₀O₃S (352.4): calcd. C 71.57, H 5.72; found

C 71.41, H 5.76.

(±)-3-[(2-Hydroxy-2-phenyl)ethyl]-2-(phenylsulfonyl)bicyclo-

[2.2.1]hepta-2,5-diene (γ-8f): Freshly distilled PhCHO (0.2 mL,

2.0 mmol) was added to solution of 3, which became immediately

colorless. After warming the solution to room temp., it turned yel-

low again. Workup gave γ-8f (51 mg, 72%) as a mixture of dia-

stereomers in a ratio of 50:50 (GC) as a colorless oil.

2

H-7_a,s), 1.63 (m, 2 H, H-6, H-5), 2.67 (s, 1 H, H-4), 2.93 (dd, J ϭ

3

13.4, J ϭ 7.3 Hz, 1 H, CH₂), 3.01 (sm, 1 H, H-1), 3.05 (sm, 1 H,

OH), 3.38 (dd, ²J ϭ 13.4, ³J ϭ 4.7 Hz, 1 H, CH₂), 5.06 (ddm, ³J ϭ

3

7.3, J ϭ 4.3 Hz, 1 H, CH), 7.30 (tm, 1 H, H-12, CPh), 7.36 (tm,

2 H, H-11,13, CPh), 7.44 (dm, 2 H, H-10,14, CPh), 7.51 (tm, 2 H,

H-11,13), 7.59 (tm, 1 H, H-12), 7.82 (dm, 2 H, H-10,14). Ϫ ¹³C

NMR (75 MHz, CDCl₃): δ ϭ 24.77 (C-5), 26.19 (C-6), 37.72 (C-

γ-8f (Diastereomer 1): GC: t_Rϭ 24.46 min (βϪCD, 140°C, iso-

therm, 60 kPa H₂). Ϫ ¹H NMR (300 MHz, CDCl₃): δ ϭ 1.85 (dm, 7), 44.97 (C-4), 47.42 (CH₂), 49.77 (C-1), 72.77 (CHOH), 125.77

J_7a,7sϭ 6.7 Hz, 1 H, H-7_a), 2.04 (m, 1 H, H-7_s), 2.67 (s, 1 H, OH), (C-10,14), 127.38 (C-11,13), 127.69 (C-12), 128.50 (C-10,14, CPh),

2

3

2

3.16 (dd, J ϭ 13.4, J ϭ 7.7 Hz, 1 H, CH₂), 3.26 (dd, J ϭ 13.4,

129.11 (C-11,13, CPh), 133.20 (C-12, CPh), 140.45 (C-9), 141.33

³J ϭ 4.7 Hz, 1 H, CH₂), 3.42 (sm, 1 H, H-4), 3.64 (sm, 1 H, H-1), (C-9, CPh), 143.94 (C-2), 159.17 (C-3).

5.02 (m, 1 H, CH), 6.48 (m, 2 H, H-5, H-6), 7.30 (m, 1 H, H-12,

γ-9f (Diastereomer 2): GC: t_Rϭ 15.36 min (DB-5, S1). Ϫ ¹H NMR

CPh), 7.36 (tm, 2 H, H-11,13, CPh), 7.43 (dm, 2 H, H-10,14, CPh),

7.50 (tm, 2 H, H-11,13), 7.58 (tm, 1 H, H-12), 7.78 (dm, 2 H, H-

10,14). Ϫ ¹³C NMR (75 MHz, CDCl₃): δ ϭ 39.11 (C-7), 52.37 (C-

4), 57.60 (C-1), 71.02 (CH₂), 72.52 (CHOH), 125.79 (C-10,14),

127.42 (C-11,13), 127.76 (C-12), 128.50 (C-10,14, CPh), 129.11 (C-

11,13, CPh), 133.12 (C-12, CPh), 139.57 (C-2), 139.93 (C-5), 142.52

(C-6), 143.59 (C-9), 147.89 (C-9, CPh), 166.68 (C-3).

(300 MHz, CDCl₃): δ ϭ 1.10 (dm, J_7,7ϭ 8.7 Hz, 1 H, H-7_a,s), 1.15

(m, 1 H, H-5), 1.19 (m, 1 H, H-6), 1.46 (dm, J_7,7ϭ 8.7 Hz, 1 H,

2

3

H-7_a,s), 1.71 (m, 2 H, H-5, H-6), 2.69 (dd, J ϭ 13.4, J ϭ 4.3 Hz,

1 H, CH₂), 2.86 (dm, 1 H, OH), 2.91 (m, 1 H, H-4), 3.07 (sm, 1

2

3

H, H-1), 3.49 (dd, J ϭ 13.4, J ϭ 9.4 Hz, 1 H, CH₂), 4.95 (ddm,

³J ϭ 9.4, J ϭ 4.3 Hz, 1 H, CH), 7.30 (tm, 1 H, H-12, CPh), 7.36

3

(tm, 2 H, H-11,13, CPh), 7.44 (dm, 2 H, H-10,14, CPh), 7.51 (tm,

γ-8f (Diastereomer 2): GC: t_Rϭ 24.59 min (βϪCD, 140°C, iso-

2 H, H-11,13), 7.59 (tm, 1 H, H-12), 7.87 (dm, 2 H, H-10,14). Ϫ

therm, 60 kPa H₂). Ϫ ¹H NMR (300 MHz, CDCl₃): δ ϭ 1.88 (dm, ¹³C NMR (75 MHz, CDCl₃): δ ϭ 24.66 (C-5), 26.10 (C-6), 37.72

J_7a,7sϭ 6.7 Hz, 1 H, H-7_a), 2.07 (m, 1 H, H-7_s), 2.67 (s, 1 H, OH), (C-7), 44.97 (C-4), 47.64 (CH₂), 49.14 (C-1), 73.09 (CHOH), 125.83

1648

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

(C-10,14), 127.38 (C-11,13), 127.78 (C-12), 128.59 (C-10,14, CPh), Partial Hydrogenation of (±)-endo- and (±)-exo-2-Methyl-3-methyl-

FULL PAPER

129.24 (C-11,13, CPh), 133.20 (C-12, CPh), 140.39 (C-9), 141.33

(C-9, CPh), 144.32 (C-2), 159.51 (C-3).

ene-2-(phenylsulfonyl)bicyclo[2.2.1]hept-5-ene: A mixture of endo-8b

and exo-8b (88 mg, 0.34 mmol) in a ratio of 76:24 and 10% Pd/C

(200 mg) in MeOH (30 mL) was vigorously stirred under H₂at

room temp. After 5 min, the theoretical amount of H₂had been

consumed and hydrogenation was terminated by the removal of the

catalyst by filtration through Celite. Concentration of the solution

in vacuo gave a mixture of endo-9b and exo-9b (76 mg, 86%) in a

ratio of 76:24 (GC) as a colorless solid.

γ-9f (Mixture of Diastereomers): C₂₁H₂₂O₃S (354.5): calcd. C 71.16,

H 6.26; found C 70.89, H 6.28.

(±)-2-(Phenylsulfonyl)-3-[(trimethylsilyl)methyl]bicyclo-

[2.2.1]hepta-2,5-diene (γ-8g): Neat Me₃SiCl (0.13 mL, 1.0 mmol)

was added to the solution of 3, whose color did not change. The

solution was stirred for 30 min at Ϫ78°C and then warmed slowly

to Ϫ35°C, whereby it became colorless. Workup gave γ-8g contami-

nated only by traces of endo-8g and exo-8g (55 mg, 86%) as a color-

less oil. Ϫ GC: > 95% (t_Rϭ 13.36 min, DB-5, S2). Ϫ ¹H NMR

H/D Exchange of Sulfones exo-5, endo-5, exo-6 and endo-6: A solu-

tion of CD₃ONa (0.005 mmol) in CD₃OD (0.1 mL) was added at

25°C to a solution of sulfone (0.2 mmol) in CD₃OD (0.7 mL), con-

1

tained in a NMR tube, and H-NMR spectra were recorded after

(300 MHz, CDCl₃): δ ϭ 0.15 (s, 9 H, SiMe₃), 1.87 (dm, J_7a,7s

ϭ

the time given in Tables 10Ϫ13.

6.7 Hz, 1 H, H-7_a), 2.05 (dm, J_7s,7aϭ 6.7 Hz, 1 H, H-7_s), 2.20 (d,

2

²J ϭ 11.4 Hz, 1 H, CH₂), 2.60 (d, J ϭ 11.4 Hz, 1 H, CH₂), 3.44

(sm, 1 H, H-1), 3.63 (sm, 1 H, H-4), 6.50 (sm, 2 H, H-5, H-6), 7.49

(tm, 2 H, H-11,13), 7.55 (tm, 1 H, H-12), 7.79 (dm, 2 H, H-10,14).

Ϫ

Acknowledgments

¹³C NMR (75 MHz, CDCl₃): δ ϭ Ϫ0.69 (SiMe₃), 23.63 (CH₂),

Financial support of this work by the Deutsche Forschungsgemein-

schaft [Sonderforschungsbereich “Asymmetric Synthesis with

Chemical and Biological Methods” (SFB 380)], the VW Stiftung

and the Fonds der Chemischen Industrie is gratefully acknowl-

edged. The authors thank Professors Dr. Barry M. Trost, Dr. Ken

Kanematsu and Dr. Hans-Josef Altenbach for providing samples,

NMR spectra and synthetic procedures of the sulfones. Dr. Jürgen

Vollhardt is thanked for his help in the preparation of single crys-

tals of the lithiosulfones. The authors are grateful to Priv.-Doz. Dr.

Walter Bauer and Dr. Alexander Giesen for their advice concerning

the cryoscopic experiments. We thank Dipl.-Phys. Thomas Eifert

(Rechenzentrum der RWTH Aachen) for generous technical sup-

port.

51.89 (C-1), 58.92 (C-4), 69.69 (C-7), 127.21 (C-11,13), 128.93 (C-

10,14), 132.64 (C-12), 138.77 (C-6), 139.34 (C-9), 140.77 (C-3),

143.44 (C-5), 170.55 (C-2). Ϫ GC MS; m/z (%): 319 [M^ϩϩ 1] (3),

303 (1), 177 (2), 105 (8), 89 (100), 75 (3), 61 (48). Ϫ C₁₇H₂₂O₂SSi

(318.5): calcd. C 64.11, H 6.96; found C 64.30, H 6.87.

(±)-2-(Phenylsulfonyl)-3-[(trimethylsilyl)methyl]bicyclo[2.2.1]hept-2-

ene (γ-9g): Neat Me₃SiCl (0.13 mL, 1.0 mmol) was added to the

solution of 4, which became light yellow. After stirring the solution

for 30 min at Ϫ78°C, it became colorless. Workup gave γ-9g

(51 mg, 80%) as a colorless oil. Ϫ GC: 100% (t_Rϭ 14.26 min, DB-

1

5, S2). Ϫ H NMR (300 MHz, CDCl₃): δ ϭ 0.15 (s, 9 H, SiMe₃),

0.87 (m, 1 H, H-5 or H-6), 1.08 (m, 1 H, H-5 or H-6), 1.11 (dm,

J_7a,7sϭ 8.7 Hz, 1 H, H-7_a), 1.47 (dm, J_7s,7aϭ 8.7 Hz, 1 H, H-7_s),

1.55 (m, 1 H, H-5 or H-6), 1.69 (m, 1 H, H-5 or H-6), 1.90 (d,

[1]

[1a]

EWG ϭ COOR:

Soc. 1975, 97, 6908Ϫ6909. Ϫ

Am. Chem. Soc. 1977, 99, 3101Ϫ3113. Ϫ

A. Dundulis, J. Org. Chem. 1980, 45, 3236Ϫ3245. Ϫ

Paquette, C. W. Doecke, F. R. Kearney, A. F. Drake, S. F. Ma-

E. J. Corey, H. E. Ensley, J. Am. Chem.

2

²J ϭ 11.4 Hz, 1 H, CH₂), 2.70 (d, J ϭ 11.4 Hz, 1 H, CH₂), 2.76

[1b]

B. M. Trost, Y. Tamaru, J.

[1c]

(sm, 1 H, H-1), 3.06 (sm, 1 H, H-4), 7.50 (tm, 2 H, H-11,13), 7.57

(tm, 1 H, H-12), 7.88 (dm, 2 H, H-10,14). Ϫ ¹³C NMR (75 MHz,

CDCl₃): δ ϭ Ϫ0.66 (SiMe₃), 21.03 (CH₂), 24.85 (C-6), 26.68 (C-5),

44.53 (C-1), 46.81 (C-7), 51.03 (C-4), 127.20 (C-11,13), 129.00 (C-

10,14), 132.69 (C-12), 132.90 (C-9), 142.61 (C-3), 163.52 (C-2). Ϫ

GC MS; m/z (%): 321 [M^ϩϩ 1] (3), 305 (31), 277 (2), 199 (2), 135

(1), 89 (100), 75 (3), 61 (39). Ϫ C₁₇H₂₄O₂SSi (320.5): calcd. C 63.70,

H 7.55; found C 63.79, H 7.48.

A. P. Krapcho, E.

[1d]

L. A.

[1e]

son, J. Am. Chem. Soc. 1980, 102, 7228Ϫ7233. Ϫ

W. Op-

polzer, E. Francotte, K. Bättig, Helv. Chim. Acta 1981, 64,

478Ϫ481. Ϫ ^[1f]A. G. M. Barrett, R. A. E. Carr, S. V. Attwood,

G. Richardson, N. D. A. Walshe, J. Org. Chem. 1986, 51,

4840Ϫ4856. Ϫ ^[1g]T. van der Does, G. W. Klumpp, M. Schakel,

[1h]

Tetrahedron Lett. 1986, 27, 519Ϫ520. Ϫ

G. Helmchen, K.

[1i]

Ihrig, H. Schindler, Tetrahedron Lett. 1987, 28, 183Ϫ186. Ϫ

M. Bertrand, J. P. Zahra, Tetrahedron Lett. 1989, 30,

In Situ Methylation of Lithiosulfones 3 and 4 ؊ (±)-endo- and (±)-

exo-2-Methyl-3-methylene-2-(phenylsulfonyl)bicyclo[2.2.1]hept-5-ene

(endo-8b and exo-8b): Neat MeI (0.14 mL, 2.2 mmol) was added at

Ϫ78°C (Ϫ105°C) to a solution of endo-5 or exo-5 (50 mg,

0.2 mmol) in THF (5 mL). Subsequently, the solution was treated

dropwise with nBuLi (0.14 mL, 1.6  in n-hexane, 0.22 mmol) at

this temp., whereby it remained colorless. Stirring of the mixture

was continued at Ϫ78°C (Ϫ105°C) for 10 min. Workup gave a mix-

ture of endo-8b and exo-8b (50 mg, 96%) in a ratio of 76:24 (68:32

or 57:43) (GC) as a colorless solid.

[1j]

4117Ϫ4120. Ϫ

K. Miyaji, Y. Ohara, Y. Takahashi, T. Tsu-

ruda, K. Arai, Tetrahedron Lett. 1991, 32, 4557Ϫ4560. Ϫ ^[1k]E.

J. Corey, N. Imai, S. Pikul, Tetrahedron Lett. 1991, 32,

[1l]

7517Ϫ7520. Ϫ

J. Ibarzo, R. M. Ortuno, Tetrahedron 1994,

[1m]

40, 9825Ϫ9830. Ϫ

C. A. Dvorak, V. H. Rawal, J. Chem.

Soc., Chem. Commun. 1997, 2381Ϫ2382.

EWG ϭ CHO: P. Groenewegen, H. Kallenberg, A. van der

Gen, Tetrahedron Lett. 1979, 2817Ϫ2820.

[2]

[3]

[4]

[5]

EWG ϭ C(NR)OR: Y. Langlois, A. Pouilhes, Tetrahedron:

Asymmetry 1991, 2, 1223Ϫ1230.

EWG ϭ NO₂: F. G. Bordwell, K. C. Yee, J. Am. Chem. Soc.

1970, 92, 5939Ϫ5944.

EWG ϭ CN: W. H. Boehme, E. Schipper, W. G. Scharpf, J.

Nichols, J. Am. Chem. Soc. 1958, 80, 5488Ϫ5495; W. H.

Boehme, E. A. Siegmund, W. G. Scharpf, E. Schipper, J. Med.

Pharm. Chem. 1962, 5, 451Ϫ464; J.-C. Muller, J.-P. Fleury, Bull.

Soc. Chim. Fr. 1970, 738Ϫ742; E. Vedejs, J. E. Telschow, J. Org.

Chem. 1976, 41, 740Ϫ741; R. W. Freerksen, S. J. Selikson, R.

R. Wroble, K. S. Kyler, D. S. Watt, J. Org. Chem. 1983, 48,

4087Ϫ4096; Y. Kawakami, H. Hisada, Y. Yamashita, Tetra-

(±)-endo- and (±)-exo-2-Methyl-3-methylene-2-(phenylsulfonyl)bicy-

clo[2.2.1]heptane (endo-9b and exo-9b): Neat MeI (0.14 mL,

2.2 mmol) was added at Ϫ78°C (Ϫ105°C) to a solution of endo-6

or exo-6 (50 mg, 0.2 mmol) in THF (5 mL). Subsequently, the mix-

ture was treated dropwise with nBuLi (0.14 mL, 1.6  in n-hexane,

0.22 mmol) at this temp., whereby it remained colorless at Ϫ78°C

but became transient yellow at Ϫ105°C. Stirring was continued for

10 min at Ϫ78°C (Ϫ105°C). Workup gave a mixture of endo-9b

and exo-9b (50 mg, 95%) in a ratio of 80:20 (80:20 or 81:19) (GC)

as a colorless solid.

hedron Lett. 1985, 26, 5835Ϫ5836.

[6]

[6a]

EWG ϭ SO₂R:

H. Hogeveen, F. Montanari, F. Taddei, J.

[6b]

Chem. Soc. 1964, 965Ϫ974. Ϫ

G. Maccagnani, F. Montan-

ari, F. Taddei, J. Chem. Soc. B 1968, 453Ϫ458. Ϫ ^[6c]D. I. Dav-

[6d]

ies, P. J. Rowley, J. Chem. Soc. C 1968, 1832Ϫ1836. Ϫ

L.

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1649

H.-J. Gais et al.

FULL PAPER

Veniard, J. Bena¨ım, G. Pourcelot, C. R. Acad. Sci., Ser. C 1971,

12 Union Road, Cambridge CB2 1EZ, UK [Fax: (internat.) ϩ

44-1223/336-033, E-mail: deposit@ccdc.cam.ac.uk₎.

Because of severe disorder problems the crystal structure of

exo-6 could not be solved.

For the crystal structures of norbornenyl sulfones, sulfoxides

and sulfoximines, see: G. D. Andreetti, G. Bocelli, P. Sgara-

botto, Cryst. Struct. Commun. 1979, 8, 173Ϫ177; X. Ji, D. van

der Helm, R. V. Williams, W. J. Ebey, Acta Crystallogr. 1989,

B45, 93Ϫ99; ref.^[7]

482Ϫ485. Ϫ ^[6e]B. M. Trost, N. R. Schmuff, J. Am. Chem. Soc.

[6f]

[22]

[23]

1985, 107, 396Ϫ405. Ϫ

K. Hayakawa, H. Nishiyama, K.

[6g]

Kanematsu, J. Org. Chem. 1985, 50, 512Ϫ517. Ϫ

N. P.

Singh, B. Metz, J. F. Biellmann, Bull. Soc. Chim. Fr. 1990, 127,

[6h]

98Ϫ107. Ϫ

R. V. Williams, G. W. Kelley, J. Loebel, D. van

der Helm, P. C. Bulman Page, J. Org. Chem. 1990, 55,

3840Ϫ3846.

[7]

[8]

[9]

EWG ϭ S(O)(NR)Tol: R. S. Glass, K. Reineke, M. Shanklin,

J. Org. Chem. 1984, 49, 1527Ϫ1533.

EWG ϭ S(O)R: R. V. Williams, X. Lin, J. Chem. Soc., Chem.

Commun. 1989, 1872Ϫ1873.

For example, a search in the Cambridge Crystallographic Data

Base in October 1998 revealed no entry for a carbanion of

type 1.

Stable Carbocation Chemistry (Eds.: G. K. S. Prakash, P. v. R.

Schleyer), Wiley, New York, 1997.

[24]

G. Barbarella, M. Cinquini, S. Colonna, J. Chem. Soc., Perkin

Trans. 1 1980, 1646Ϫ1649.

[25] [25a]

H.-J. Gais, H. J. Lindner, J. Vollhardt, Angew. Chem. 1985,

97, 895Ϫ860; Angew. Chem. Int. Ed. Engl. 1985, 24, 865. Ϫ ^[25b]

H.-J. Gais, J. Vollhardt, G. Hellmann, H. Paulus, H. J. Lindner,

[25c]

Tetrahedron Lett. 1988, 29, 1259Ϫ1262. Ϫ

H.-J. Gais, G.

[10]

[11]

Hellmann. H. Günther, F. Lopez, H. J. Lindner, S. Braun, An-

gew. Chem. 1989, 101, 1061Ϫ1063; Angew. Chem., Int. Ed. Engl.

[25d]

1989, 28, 1025Ϫ1028. Ϫ

H.-J. Gais, G. Hellmann, H. J.

For a review on the structure of α-sulfonyl carbanions, see: G.

Boche, Angew. Chem. 1989, 101, 286Ϫ306; Angew. Chem., Int.

Ed. Engl. 1989, 28, 277Ϫ297.

Lindner, Angew. Chem. 1990, 102, 96Ϫ103; Angew. Chem. Int.

[25e]

Ed. Engl. 1990, 29, 100Ϫ103. Ϫ

H.-J. Gais, J. Müller, J.

Vollhardt, H. J. Lindner, J. Am. Chem. Soc. 1991, 113,

[12] [12a]

The stereodescriptors endo and exo designate the position

[25f]

4002Ϫ4003. Ϫ

1997. Ϫ

R. Scholz, Ph. D. Thesis, RWTH Aachen

[12b]

of the phenyl group. Ϫ

In the three-dimensional graphical

[25g]

T. Hess, Ph. D. Thesis, RWTH Aachen 1998.

representations of the SO₂group the SϪO double bond is rep-

[26]

G. Boche, M. Marsch, K. Harms, G. M. Sheldrick, Angew.

Chem. 1985, 97, 577Ϫ578; Angew. Chem. Int. Ed. Engl. 1985,

24, 573Ϫ574; W. Hollstein, K. Harms, M. Marsch, G. Boche,

Angew. Chem. 1987, 99, 1279Ϫ1280; Angew. Chem. Int. Ed.

Engl. 1987, 26, 1287Ϫ1288; W. Hollstein, K. Harms, M.

Marsch, G. Boche, Angew. Chem. 1988, 100, 868Ϫ869; Angew

Chem. Int. Ed. Engl. 1988, 27, 846Ϫ847;

[12c]

resented only by one line because of clarity. Ϫ

The same

numbering scheme has been used for all compounds with a

methylenenorbornenyl and methylenenorbornyl skeleton be-

cause of comparison.

[13]

For reviews, see: J.-F. Biellman, J.-B. Ducep, Org. React. 1982,

27, 1Ϫ344; S. Oae, Y. Uchida in The Chemistry of Sulfones and

Sulfoxides (Eds.: S. Patai, Z. Rappoport, C. J. M. Stirling, ),

Wiley, New York, 1988, pp. 583Ϫ664; K. Ogura in Comprehen-

sive Organic Synthesis (Ed.: S. L. Schreiber), Pergamon Press,

Oxford, 1991, vol. 1, pp. 505Ϫ539; A. Krief in Comprehensive

Organic Synthesis (Ed.: G. Pattenden), Pergamon Press, Oxford,

1991, vol. 3, pp. 85Ϫ191; Y. Yamamoto in Comprehensive Or-

ganic Synthesis (Ed.: C. H. Heathcock), Pergamon Press, Ox-

ford, 1991, vol. 3, pp. 55Ϫ79; N. S. Simpkins, Sulphones in Or-

ganic Synthesis, Pergamon Press: Oxford, 1993; M. Braun in

Stereoselective Synthesis (Houben-Weyl) (Eds.: G. Helmchen,

R. W. Hoffmann, J. Mulzer, E. Schaumann), Thieme, Stuttgart

1995, vol. E21b, pp. 1713Ϫ1735; S. Pyne in Stereoselective Syn-

thesis (Houben-Weyl) (Eds.: G. Helmchen, R. W. Hoffmann,

J. Mulzer, E. Schaumann), Thieme, Stuttgart 1995, vol. E21b,

pp. 2068Ϫ2086.

For selected illustrative examples, see: M. Julia, A. Guy-

Rouault, Bull. Soc. Chim. Fr. 1967, 1411Ϫ1411; M. Julia, D.

Arnold, Bull. Soc. Chim. Fr. 1973, 743Ϫ746; B. Lythgoe, I.

Waterhouse, J. Chem. Soc., Perkin Trans. 1 1979, 2429Ϫ2436;

B. M. Trost, N. R. Schmuff, M. J. Miller, J. Am. Chem. Soc.

1980, 102, 5979Ϫ5981; S. E. Denmark, M. A. Harmata, J. Am.

Chem. Soc. 1982, 104, 4972Ϫ4974; T. Mandai, T. Yanagi, K.

Araki, Y. Morisaki, M. Kawada, J. Otera, J. Am. Chem. Soc.

1984, 106, 3670Ϫ3672; B. M. Trost, M. Reza Ghadiri, J. Am.

Chem. Soc. 1984, 106, 7260Ϫ7261; J. A. Marshall, R. C. And-

rews, J. Org. Chem. 1985, 50, 1602Ϫ1606; S. D. A. Street, C.

Yeates, P. Kocienski, S. F. Campbell, J. Chem. Soc., Chem. Com-

mun. 1985, 1386Ϫ1388; T. Cuvigny, C. Herve du Penhoat, M.

Julia, Tetrahedron 1986, 42, 5329Ϫ5336; B. M. Trost, C. A.

Merlic, J. Am. Chem. Soc. 1988, 110, 5216Ϫ5218; H.-J. Gais,

W. A. Ball, J. Bund, Tetrahedron Lett. 1988, 29, 781Ϫ784; A.

Padwa, S. H. Watterson, Z. Ni, J. Org. Chem. 1994, 59,

3256Ϫ3258; Z. Jin, P. L. Fuchs, J. Am. Chem. Soc. 1994, 117,

3022Ϫ3028.

[27] [27a]

S. Wolfe, Stud. Org. Chem. (Amsterdam) 1985, 19,

133Ϫ190. Ϫ ^[27b]D. A. Bors, A. Streitwieser, J. Am. Chem. Soc.

[27c]

1986, 108, 1397Ϫ1404. Ϫ

G. Boche, J. C. W. Lohrenz, J.

Cioslowski, W. Koch in The chemistry of functional groups, Sup-

plement S: The chemistry of sulphur-containing functional groups

(Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1993, pp.

[27d]

339Ϫ362. Ϫ

T. Hoz, H. Basch in The chemistry of func-

tional groups, Supplement S: The chemistry of sulphur-containing

functional groups (Eds.: S. Patai, Z. Rappoport), Wiley, New

[27e]

York, 1993, pp. 1Ϫ100. Ϫ

R. Koch, E. Anders, J. Org.

[27f]

Chem. 1994, 59, 4529Ϫ4534. Ϫ

G. Raabe, H.-J. Gais, J.

Fleischhauer, J. Am. Chem. Soc. 1996, 118, 4622Ϫ4630.

[28] [28a]

S. Bradamante, G. A. Pagani, J. Chem. Soc., Perkin Trans.

[28b]

2 1986, 1035Ϫ1046. Ϫ

Carbanion Chem. 1996, 2, 189Ϫ263.

S. Bradamante, G. A. Pagani, Adv.

[14]

[29]

F. G. Bordwell, J. C. Branca, C. R. Johnson, N. R. Vanier, J.

Org. Chem. 1980, 45, 3884Ϫ3889.

[30] [30a]

O. Ermer, P. Bell, S. A Mason, Angew. Chem. 1989, 101,

1298Ϫ1301; Angew. Chem. Int. Ed. Engl. 1989, 28, 1239Ϫ1242

and references cited therein. Ϫ

[30b]

J. Benet-Buchholz, T. Hau-

mann, R. Boese, Chem. Commun. 1998, 2003Ϫ2004.

χ₂ϭ 180 Ϫ ԽS1ϪC2ϪC1ϪC3Խ. Note that for the planar C2

atom χ₂ϭ 0° and for the tetrahedral C2 atom χ₂ϭ 60°.

[31]

^{[32] [32a]}P. v. R. Schleyer, J. Am. Chem. Soc. 1985, 107, 4793Ϫ4794.

[32b]

Ϫ

K. B. Wiberg, C. M. Breneman, T. J. LePage, J. Am.

Chem. Soc. 1990, 112, 61Ϫ72.

[33]

R. E. Lee, R. R. Squires, J. Am. Chem. Soc. 1986, 108,

5078Ϫ5086.

[34] [34a]

G. Wipff, K. Morokuma, Tetrahedron Lett. 1980, 21,

[34b]

4445Ϫ4448. Ϫ

U. Burkert, Angew. Chem. 1981, 93,

602Ϫ603; Angew. Chem. Int. Ed. Engl. 1981, 20, 572Ϫ573. Ϫ

[34c]

N. G. Rondan, M. N. Paddon-Row, P. Caramella, J. Ma-

reda, P. H. Mueller, K. N. Houk, J. Am. Chem. Soc. 1982, 104,

[34d]

[15]

[16]

4974Ϫ4976. Ϫ

K. N. Houk, N. G. Rondan, F. K. Brown,

S. Bradamante, A. Mangia, G. Pagani, J. Chem. Soc. B 1971,

545Ϫ548.

H.-J. Gais, J. Vollhardt, H. J. Lindner, Angew. Chem. 1986, 98,

916Ϫ917; Angew. Chem., Int. Ed. Engl. 1986, 25, 939Ϫ941.

M. Christl, C. Herzog, Tetrahedron Lett. 1987, 28, 187Ϫ190.

F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456Ϫ463.

W. L. Jorgensen, J. D. Madura, D. C. Spellmeyer, J. Am. Chem.

[34e]

Soc. 1983, 105, 5980Ϫ5988. Ϫ

iter, Tetrahedron 1983, 39, 3345Ϫ3350. Ϫ

Vogel, J. Mol. Struct. (Theochem.) 1985, 124, 9Ϫ23. Ϫ

J. Spanget-Larsen, R. Gle-

[34f]

P.-A. Carrupt, P.

[34g]

C.

[17]

[18]

[19]

R. Castro, R. Dutler, A. Rauk, H. Wieser, J. Mol. Struct.

[34h]

(Theochem.) 1987, 152, 241Ϫ253. Ϫ

Y. Kurita, C. Takay-

[34i]

L. Veniard, J. Bena¨ım, G. Pourcelot, C. R. Acad. Sci., Ser. C

1968, 266, 1092Ϫ1094.

ama, Tetrahedron 1990, 46, 3789Ϫ3802. Ϫ

N. Koga, T. Oz-

awa, K. Morokuma, J. Phys. Org. Chem. 1990, 3, 519Ϫ533. Ϫ

[20]

[21]

[34j]

In all experiments with sulfones 5 and 6 described in this work

the racemates were used.

Crystallographic data (excluding structure factors) for the struc-

tures reported in this paper have been deposited with the Cam-

bridge Crystallographic Data Centre as supplementary publi-

cation no. CCDC-102801 and CCDC-102802. Copies of the

data can be obtained free of charge on application to CCDC,

A. Rastelli, M. Cocchi, E. Schiatti, R. Gandolfi, M. Burd-

[34k]

isso, J. Chem. Soc., Faraday Trans. 1990, 86, 783Ϫ787. Ϫ

M. C. Holthausen, W. Koch, J. Phys. Chem. 1993, 97,

10021Ϫ10027.

[35]

H.-J. Gais, D. Lenz, G. Raabe, Tetrahedron Lett. 1995, 36,

7437Ϫ7440; J. F. K. Müller, R. Batra, B. Spingler, M. Zehnder,

Helv. Chim. Acta 1996, 79, 820Ϫ826.

1650

Eur. J. Org. Chem. 1999, 1627Ϫ1651

Sulfonyl-Stabilized Allylic Norbornenyl and Norbornyl Carbanions

FULL PAPER

nisms and Novel Reactions, American Chemical Society, Wash-

ington, DC, 1994, pp. 120Ϫ123.

W. E. Truce, M. F. Amos, J. Am. Chem. Soc. 1951, 73,

3013Ϫ3017.

J. Cram, W. D. Nielsen, B. Rickborn, J. Am. Chem. Soc. 1960,

82, 6415Ϫ6416; D. J. Cram, R. D. Trepka, P. S. Janiak, J. Am.

Chem. Soc. 1966, 88, 2749Ϫ2759; J. N. Roitman, D. J. Cram, J.

Am. Chem. Soc. 1971, 93, 2225Ϫ2231.

E. J. Corey, E. T. Kaiser, J. Am. Chem. Soc. 1961, 83, 490Ϫ491;

E. J. Corey, T. H. Lowry, Tetrahedron Lett. 1965, 793Ϫ801.

H. L. Goering, D. L. Towns, B. Dittmar, J. Org. Chem. 1962,

27, 736Ϫ739.

[36]

G. Boche, personal communication, February 26, 1997; F.

Haller, Diploma Thesis, Universität Marburg, 1990.

[60]

[61]

[37]

A. Giesen, Ph. D. Thesis, RWTH Aachen, 1998; S. Rohs, H.-J.

Gais, unpublished results.

[38]

M. J. Kaufman, S. Gronert, D. A. Bors, A. Streitwieser, J. Am.

Chem. Soc. 1987, 109, 602Ϫ603; A. Streitwieser, G. P. Wang,

D. A. Bors, Tetrahedron 1997, 53, 10103Ϫ10112.

[39]

W. Bauer, D. Seebach, Helv. Chim. Acta 1984, 67, 1972Ϫ1988.

[62]

[63]

[64]

[65]

[66]

[40]

O. Luche, Ph. D. Thesis, RWTH Aachen, 1994.

[41]

H.-J. Gais, J. Vollhardt, C. Krüger, Angew. Chem. 1988, 100,

1108Ϫ1110; Angew. Chem. Int. Ed. Engl. 1988, 27, 1092Ϫ1094.

[42]

P. G. Cozzi, C. Floriani, Organometallics 1995, 14, 1756Ϫ1760.

[43]

S. Wodzinski, J. W. Bunting, J. Am. Chem. Soc. 1994, 116,

6910Ϫ6915 and references cited therein.

F. Terrier, E. Kizilian, R. Goumont, N. Faucher, C. Wakselman,

J. Am. Chem. Soc. 1998, 120, 9496Ϫ9503.

For a review, see: W. Bauer, P. v. R. Schleyer, Adv. Carbanion

Chem. 1992, 1, 89Ϫ175.

[44]

W. A. Ball, Ph. D. Thesis, TH Darmstadt, 1987; J. Vollhardt,

Ph. D. Thesis, TH Darmstadt, 1990; H.-J. Gais, F. Gerhards,

D. Seebach, T. Maetzke, R. K. Haynes, M. N. Paddon-Row, S.

S. Wong, Helv. Chim. Acta 1988, 71, 299Ϫ311; H. Ahlbrecht,

G. Boche, K. Harms, M. Marsch, H. Sommer, Chem. Ber. 1990,

123, 1853Ϫ1858; M. Marsch, K. Harms, O. Zschage, D. Hoppe,

G. Boche, Angew. Chem. 1991, 103, 338Ϫ339; Angew. Chem.,

Int. Ed. Engl. 1991, 30, 321Ϫ322; D. J. Pippel, G. A. Weisen-

burger, S. R. Wilson, P. Beak, Angew. Chem. 1998, 110,

2600Ϫ2602; Angew. Chem. Int. Ed. Engl. 1998, 37, 2522Ϫ2524.

T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. v. R. Schleyer.

J. Comput. Chem. 1983, 4, 294Ϫ301.

N. Spalthoff, unpublished results.

[45]

C. Rücker, H. Fritz, Magn. Res. Chem. 1988, 26, 1103Ϫ1108.

[46]

For the NMR-spectroscopic detection of a similar charge alter-

nation in lithioallyl phosphonic diamides, see: E. Denmark, C.

J. Cramer, J. Org. Chem. 1990, 55, 1806Ϫ1813.

[47]

U. C. Singh, P. A. Kollman, J. Comput. Chem. 1984, 5,

129Ϫ145; B. H. Besler, K. M. Merz, Jr., P. A. Kollman, J.

Comp. Chem. 1990, 11, 431Ϫ439.

[67]

[68]

[69]

[70]

[48]

For calculations of the charge distribution in non-allylic α-sul-

fonyl carbanions, see: ref.^{[27c,27d,27f]}; P. Speers, K. E. Laidig, A.

Streitwieser, J. Am. Chem. Soc. 1994, 116, 9257Ϫ9261; K. B.

Wiberg, H. Castejon, J. Am. Chem. Soc. 1994, 116,

10489Ϫ10497.

M. J. Frisch, J. A. Pople, J. S. Binkley, J. Chem. Phys. 1984,

80, 3265Ϫ3269.

R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971,

54, 724Ϫ728.

[49]

An interchange of the NMR data of endo-9a and exo-9a in

W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972,

56, 2257Ϫ2261.

ref.^[19]seems to be responsible for this erroneous report in

ref.^[6e]

[71]

[72]

P. C. Hariharan, J. A. Pople, Mol. Phys. 1974, 27, 209Ϫ214.

P. C. Hariharan, J. A. Pople, Theoret. Chim. Acta 1973, 28,

213Ϫ222.

[50]

In addition, sulfones endo-8b and exo-8b were chemically corre-

lated with endo-9b and exo-9b by a partial hydrogenation of

endo-8b and exo-8b (85:15), which afforded endo-9b and exo-

9b (85:15).

[73]

[74]

C. Møller, M. S. Plesset, Phys. Rev. 1934, 46, 618Ϫ622.

M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. W. Gill, B. G.

Johnson, M. A. Robb, J. R. Cheeseman, T. A. Keith, G. A.

Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-Lan-

tham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslow-

ski, B. B. Stefanov, A. Nanayakara, M. Challacombe, C. Y.

Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S.

Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley,

D. J. Defrees, J. Baker, J. J. P. Stewart, M. Head-Gordon, C.

Gonzalez, J. A. Pople, GAUSSIAN 94 (Revision D.4), Gaus-

sian, Inc., Pittsburgh PA, 1995.

S. R. Hall, H. D. Flack, J. M. Stewart (Eds.), XTAL3.2 Refer-

ence Manual, Universities of Western Australia, Geneva, and

Maryland, Lamb, Perth, 1992.

S. R. Hall, V. Subramanian, GENSIN, XTAL3.2 Reference

Manual (Eds.: S. R. Hall, H. D. Flack, J. M. Stewart), Univer-

sities of Western Australia, Geneva, and Maryland, Lamb,

Perth, 1992, pp. 131Ϫ139.

[51]

R. Hirsch, R. W. Hoffmann, Chem. Ber. 1992, 125, 975Ϫ982.

[52]

For a review, see: J. I. Seeman, Chem. Rev. 1983, 83, 83Ϫ134.

[53]

G. Hellmann, Ph. D. Thesis, Universität Freiburg, 1991; H.-J.

Gais, G. Hellmann, J. Am. Chem. Soc. 1992, 114, 4439Ϫ4440;

D. Özdemir, Ph. D. Thesis, RWTH Aachen, 1995; E. Thiemer-

mann, Ph. D. Thesis, RWTH Aachen, 1997; A. Hack, Ph. D.

Thesis, RWTH Aachen, 1998.

[54]

For a review, see: S. H. Unger, C. Hansch, Progr. Phys. Org.

Chem. 1976, 12, 91Ϫ118.

[55]

[75]

[76]

B. Rosenstock, H.-J. Gais, E. Herrmann, G. Raabe, P. Binger,

A. Freund, P. Wedemann, C. Krüger, H. J. Lindner, Eur. J. Org.

Chem. 1998, 257Ϫ273 and references cited therein.

[56]

A. S. Cieplak in Structure Correlation (Eds.: H.-B. Bürgi, J. D.

Dunitz), VCH, Weinheim, 1994, vol. 1, pp. 205Ϫ302.

For examples and kinetic analyses of such systems, see ref.^[52]

[57]

and literature cited therein.

[58]

[77]

T. G. Traylor, Acc. Chem. Res. 1969, 2, 152Ϫ160; J. March,

S. R. Hall, GENTAN, XTAL3.2 Reference Manual (Eds.: S. R.

Hall, H. D. Flack, J. M. Stewart), Universities of Western Aus-

tralia, Geneva, and Maryland, Lamb, Perth, 1992, pp.

139Ϫ150.

Advanced Organic Chemistry, Wiley, New York, 1992, pp.

753Ϫ754; “Stereoselective Synthesis” in Methoden Org. Chem.

(Houben-Weyl), 1995, vol. E21bϪe.

[59]

[78]

[79]

P. v. R. Schleyer, J. Am. Chem. Soc. 1967, 89, 701Ϫ703; R.

G. M. Sheldrick, Acta Crystallogr. A 1990, 46, 467Ϫ473; G. M.

Sheldrick, SHELXL-97, Universität Göttingen, 1997.

E. Keller, SCHAKAL 92, Universität Freiburg, Germany, 1992.

Received January 15,1999

Huisgen, P. H. J. Ooms, M. Mingin, N. L. Allinger, J. Am.

Chem. Soc. 1980, 102, 3951Ϫ3953; W. H. Watson, J. Galloy,

P. D. Bartlett, A. A. M. Roof, J. Am. Chem. Soc. 1981, 103,

2022Ϫ2031; R. Huisgen, The Adventure Playground of Mecha-

[O99017]

Eur. J. Org. Chem. 1999, 1627Ϫ1651

1651

Article Doi

Sulfonyl-stabilized allylic norbornenyl and norbornyl carbanions: Structure and stereoselectivity of reaction with electrophiles

DOI: 10.1002/(SICI)1099-0690(199907)1999:7<1627::AID-EJOC1627>3.0.CO;2-9

Source and publish data:

Authors:

Article abstract of DOI:10.1002/(SICI)1099-0690(199907)1999:7<1627::AID-EJOC1627>3.0.CO;2-9

Full text of DOI:10.1002/(SICI)1099-0690(199907)1999:7<1627::AID-EJOC1627>3.0.CO;2-9

Products guided by the article

R&D Labs maybe for 19720-03-7

Relevant to this article

Hot Product