Stereoselectivities of Nucleophilic Additions to Cycloheptanones. Experimental and Theoretical Studies and General Purpose Force Field for the Prediction of Nucleophilic Addition Stereoselectivities

Article abstract of DOI:10.1021/jo971627b

The selectivities of nucleophilic additions of ethyl, vinyl, and ethynyllithium and Grignard reagents to a 2-(3′-phenylpropyl)cycloheptanone were investigated experimentally. In all cases, cis-cycloheptanol was formed preferentially (40:1-6:1). Theoretical studies were performed on the stereoselectivities of nucleophilic additions of hydride and ethynyllithium reagents to cycloheptanones. An empirical force field for transition states of hydride and ethynyl reagents was used for transition-state geometries and conformational searches. Quantum mechanical calculations on HLi addition to cycloheptanones were compared to the corresponding cyclohexanone calculations. These results show that the formation of 2-alkyl-substituted cis-cycloheptanols is preferred for all nucleophiles studied as a result of torsional and steric effects in the transition states of these reactions.

Full text of DOI:10.1021/jo971627b

3
196
J . Org. Chem. 1998, 63, 3196-3203
Ster eoselectivities of Nu cleop h ilic Ad d ition s to Cycloh ep ta n on es.
Exp er im en ta l a n d Th eor etica l Stu d ies a n d Gen er a l P u r p ose F or ce
F ield for th e P r ed iction of Nu cleop h ilic Ad d ition
Ster eoselectivities
Kaori Ando,1a Kevin R. Condroski,1b and K. N. Houk*
Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569
Yun-Dong Wu*
The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
Sylvie Kim Ly and L. E. Overman*
Department of Chemistry, University of California, Irvine, California 92717
Received September 3, 1997
The selectivities of nucleophilic additions of ethyl, vinyl, and ethynyllithium and Grignard reagents
to a 2-(3′-phenylpropyl)cycloheptanone were investigated experimentally. In all cases, cis-
cycloheptanol was formed preferentially (40:1-6:1). Theoretical studies were performed on the
stereoselectivities of nucleophilic additions of hydride and ethynyllithium reagents to cyclohep-
tanones. An empirical force field for transition states of hydride and ethynyl reagents was used
for transition-state geometries and conformational searches. Quantum mechanical calculations
on HLi addition to cycloheptanones were compared to the corresponding cyclohexanone calculations.
These results show that the formation of 2-alkyl-substituted cis-cycloheptanols is preferred for all
nucleophiles studied as a result of torsional and steric effects in the transition states of these
reactions.
In tr od u ction
The stereoselectivities of nucleophilic additions to
developed from studies described in a number of previous
papers.1
0-13
This force field has been incorporated into
the MM2* force field in Macromodel and is now generally
available for the prediction of the stereoselectivities of
lithium aluminum hydride reductions and organometallic
additions to ketones.
cyclohexanones have been extensively studied both ex-
perimentally2 and computationally, and the origin of
,3
4,5
these selectivities has been the subject of recent intensive
debate.4
-9
By contrast, the stereoselectivities of nucleo-
philic additions to cycloheptanones and larger cyclic
ketones have not been studied systematically. We wish
to report both experiments and calculations on nucleo-
philic additions to 2-substituted cycloheptanones that
reveal a general pattern of stereoselectivity for such
molecules. We also describe a transition-state force field,
Ba ck gr ou n d
We previously studied the stereochemistries of nucleo-
philic additions to carbonyl compounds computationally
and reported quantitative support for the Felkin model
1
1-14
for both acyclic and cyclic carbonyl compounds.
The
stereoselectivity of nucleophilic additions to cyclohex-
anones substituted by relatively nonpolar substituents
is influenced by two factors: (1) the steric interaction of
the incoming group with 3,5-axial substituents and (2)
the torsional strain of the incoming group with respect
to all R-bonds.
(
1) (a) Permanent address: College of Education, University of the
Ryukyus, Nishihara-cho, Okinawa 903-01, J apan. (b) Present ad-
dress: Computational Molecular and Cellular Science, Abbott Labo-
ratories, Abbott Park, IL 60064.
(2) For a review: Ashby, E. C.; Laemmle, J . T. Chem. Rev. 1975,
7
5, 521-546.
(
(
3) Eliel, E. L.; Senda, Y. Tetrahedron 1970, 26, 2411-2428.
4) Wu, Y.-D.; Tucker, J . A.; Houk, K. N. J . Am. Chem. Soc. 1991,
1
13, 5018-5027. Wu, Y.-D.; Houk, K. N.; Paddon-Row, M. N. Angew.
(10) Ando, K.; Houk, K. N.; Busch, J .; Menass e´ , A.; S e´ quin, U. J .
Org. Chem. 1998, 63, 1761.
(11) Wu, Y.-D.; Houk, K. N. J . Am. Chem. Soc. 1987, 109, 908-
910.
Chem., Int. Ed. Engl. 1992, 31, 1019-1021. Wu, Y.-D.; Houk, K. N.
J . Am. Chem. Soc. 1993, 115, 10992-10993.
(5) Frenking G.; K o¨ hler, K. F.; Reetz, M. T. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1146-1149.
6) Cherest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968,
199-2204. Cherest, M.; Felkin, H. Ibid. 1968, 2205-2208.
7) Cieplak, A. S. J . Am. Chem. Soc. 1981, 103, 4540-4552. Cieplak,
(12) Wu, Y.-D.; Houk, K. N.; Florez, J .; Trost, B. M. J . Org. Chem.
1991, 56, 3656-3664.
(
2
(13) Houk, K. N.; Duh, H.-Y.; Wu, Y.-D.; Moses, S. R. J . Am. Chem.
Soc. 1986, 108, 2754-2755. Spellmeyer, D. C.; Houk, K. N. J . Org.
Chem. 1987, 52, 959-974. Broeker, J . L.; Houk, K. N. J . Org. Chem.
1991, 56, 3651-3655. Thomas, B. E.; Loncharich, R. J .; Houk, K. N.
J . Org. Chem. 1992, 57, 1354-1362. Brown, F. K.; Raimondi, L.; Wu,
Y.-D.; Houk, K. N. Tetrahedron Lett. 1992, 33, 4405-4408. Eurenius,
K. P.; Houk, K. N. J . Am. Chem. Soc. 1994, 116, 9943-9946. For a
general review of this approach, see: Eksterowicz, J . E.; Houk, K. N.
Chem. Rev. 1993, 93, 2439-2461.
(
A. S.; Tait, B. D.; J ohnson, C. R. J . Am. Chem. Soc. 1989, 111, 8447-
8
9
462. Cieplak, A. S.; Wiberg, K. B. J . Am. Chem. Soc. 1992, 114, 9226-
227. Senju, T.; Tomoda, S. Chem. Lett. 1997, 431. Cieplak, A. S. J .
Org. Chem. 1998, 63, 521.
8) Hahn, J . M.; le Noble, W. J . J . Am. Chem. Soc. 1992, 114, 1916-
917.
9) Reviews: Cieplak, A. S. Structure Correlation; B u¨ rgi, H.-B.,
(
1
(
Duniz, J . D., Eds.; Verlag Chemie: Weinheim, 1994; pp 205-302.
Gung, B. W. Tetrahedron 1996, 52, 5263-5301.
(14) Kaufmann E.; Schleyer, P. v. R.; Houk, K. N.; Wu, Y.-D. J . Am.
Chem. Soc. 1985, 107, 5560-5562.
S0022-3263(97)01627-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/30/1998

Nucleophilic Additions to Cycloheptanones
J . Org. Chem., Vol. 63, No. 10, 1998 3197
Sch em e 1
Ta ble 1. Ad d ition of Eth yl, Vin yl a n d Eth yn yl
Or ga n om eta llics to Cycloh ep ta n on e 1
organometallica
solvent
yield, %
b
cis:transc
entry
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
EtMgBrd
THF
Et2O
76
63
81
71
82
72
78
85
84
83
87
80
77
72
76
24:1
13:1
9:1
EtMgBrd
d
e
EtMgBr
MePh
EtLid
THFe
Et2O
24:1
23:1
14:1
39:1
41:1
31:1
27:1
19:1
15:1
8:1
d
EtLi
EtLid
MePhe
THF
f
CH2dCHMgBr
CH2dCHMgBrf
Et2Og
f
g
CH2dCHMgBr
MePh
1
1
1
1
1
1
CH2dCHLif
THFe
Et2O
MePhe
THF
CH2dCHLif
f
CH2dCHLi
h
HCtCLi
HCtCLih
HCtCLih
Et2O
MePh
7:1
6:1
a
Organometallic nucleophiles were added dropwise to a 0.20-
0
.23 M solution of 1 at 0 °C. b Isolated yield of the mixture of
stereoisomeric alcohols 2 and 3. Capillary GLC analysis of silyl
ether derivatives 4 and 5. FID detector response was assumed to
c
d
be identical for isomer pairs. 2 equiv of organometallic reagent
was used. e Solution contains 8-10% of Et2O. f 1 equiv of organo-
Fewer studies have been reported for cycloheptanones.
In an early application of 13C NMR chemical shifts to the
elucidation of stereochemistry, Christl and Roberts re-
ported that the sole product produced (81% yield) from
addition of MeLi to 2-methylcycloheptanone in refluxing
ether was cis-1,2-dimethylcycloheptanol.15a The addition
of vinylmagnesium bromide to 2-allylcycloheptanone has
been reported to give a 98:2 mixture of two isomers; the
major isomer was assumed to be the cis-alcohol.1
Addition of (E)-butadienyllithium to 2-butadienylcyclo-
heptanone afforded a single stereoisomeric product,
g
h
metallic reagent was used. Solution contains 8-10% of THF. 5
equiv of the ethylenediamine complex was employed.
lithium, vinylmagnesium bromide, and ethynyllithium)
to 2-(3′-phenylpropyl)cycloheptanone (1) at 0 °C was
studied in three solvents; THF, Et O, and toluene.
2
Product ratios were determined by capillary GLC analy-
sis after silylation of the alcohol products 2 and 3 with
trimethylsilyl chloride (Table 1). In all cases, cis-cyclo-
heptanol 2 was formed preferentially. The highest
stereoselectivities (40:1) were observed for vinylmagne-
sium bromide (entries 7-9) and the lowest for ethynyl-
lithium (entries 13-15). Stereoselection in general was
highest in THF and lowest in toluene.
The ethynyl silyl ether stereoisomers 4c and 5c were
separated by column chromatography and desilylated to
provide pure samples of the corresponding alcohols 2c
and 3c. These propargylic alcohols were chemically
correlated, respectively, with 2b and 3b by reduction with
5b
1
5c
which was also assumed to be the cis-alcohol.
tions of 2-methylcycloheptanone with LiAlH
7), BH (ds ) 74:26), disiamylborane (ds ) 64:26),
dicyclohexylborane (ds ) 97:3), di-3-pinanylborane (ds )
8:2), Na (ds ) 56:44), and Al(O-i-Pr) (ds ) 65:35) afford
Reduc-
4
(ds ) 73:
2
3
9
3
1
6
predominantly cis-2-methylcycloheptanol. Because of
the conformational flexibility of cycloheptanones, it is not
clear which factor controls these stereochemistries. In
a benzocycloheptanone system, Mukherjee et al. pre-
dicted that equatorial attack will be preferred, and
subsequent experimental studies confirmed this predic-
tion.17 The prediction was made with a force field
described in this paper, and the results were rationalized
by torsional effects.
4
LiAlH and with 2a and 3a by catalytic hydrogenation.
Stereochemical assignments were secured by single-
18
crystal X-ray analysis of hydroxy acid 6, which was
19
4
obtained by RuO oxidation of 2a .
Th eor etica l Stu d ies. Because seven-membered rings
enjoy more conformational flexibility than their six-
membered counterparts, cycloheptanone was optimized
It occurred to both of our groups that the stereoselec-
tive formation of cis-cycloheptanols might be a general
phenomenon. This paper reports experimental and
theoretical results that show that this is true.
20
by employing a Monte Carlo conformational search
21
available in Still’s Macromodel program. A total of 1000
conformations were generated and subsequently mini-
mized using the MM2* force field. The five unique
conformers located in this search (7-11) are illustrated
in Figure 1. The relative energies are given next to the
structures. Conformer 10, a chair with the carbonyl on
the two-carbon stern fragment, was found to be the global
Resu lts a n d Discu ssion
Exp er im en ta l Stu d ies. The addition of five nucleo-
philes (ethyllithium, ethylmagnesium bromide, vinyl-
(
15) (a) Christl, M.; Roberts, J . D. J . Org. Chem. 1972, 37, 3443-
452. (b) Avasthi, K.; Salomon, R. G. Ibid. 1986, 51, 2556-2562. (c)
Angoh, A. G.; Clive, D. L. J . J . Chem. Soc., Chem. Commun. 1984,
3
(18) The authors have deposited atomic coordinates for 4 with the
Cambridge Crystallographic Data Centre. The coordinates can be
obtained, on request, from the Director, Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge, CB2, 1EZ, U.K.
(19) Carlsen, P. H. J .; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J . Org. Chem. 1981, 46, 3936-3938.
(20) Chang, G.; Guida, W. C.; Still W. C. J . Am. Chem. Soc. 1989,
111, 4379-4386. Saunders: M.; Houk, K. N.; Wu, Y.-D.; Still, W. C.;
Lipton, M.; Chang, G.; Guida, W. C. J . Am. Chem. Soc. 1990, 112,
1419-1427.
(21) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp. R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J .
Comput. Chem. 1990, 11, 440-467.
5
34-536.
(
16) (a) Brown, H. C.; Varma, V. J . Am. Chem. Soc. 1966, 88, 2871-
2
1
6
6
871. (b) Brown, H. C.; Varma, V. J . Org. Chem. 1974, 39, 1631-
636. (c) Brown, H. C.; Krishnamurthy, S. Tetrahedron 1979, 35, 567-
07. (d) Huckel, W.; Honecker, O. J ustus Liebigs Ann. Chem. 1964,
78, 10-19. (e) Huckel, W.; Wachter, J . Ibid. 1964, 672, 62-77. (f)
Sehgal, R. K.; Koenigsberger, R. U.; Howard, T. J . J . Org. Chem. 1975,
0, 3073-3078; (g) Tetrahedron Lett. 1974, 47, 4173-4176. (h)
Chhatwal, V. K.; Gautam, R. K.; Saharia, G. S. J . Inst. Chem. (India)
983, 55, 55-57.
17) Mukherjee, D.; Wu, Y.-D.; Fronczek, F. R.; Houk, K. N. J . Am.
Chem. Soc. 1988, 110, 3328-3330.
4
1
(

3
198 J . Org. Chem., Vol. 63, No. 10, 1998
Ando et al.
special substructure within the MM2* force field present
in Macromodel. The parameters are listed in the Ap-
pendix. Extensive testing verified that the MM2* imple-
mentation of the force field afforded reliable energies that
closely paralleled those obtained using the original
implementation using Allinger’s MM2 program and force
field.1
1,12
The MM2* force field provides energies in close
agreement with experimental data from the literature
for a wide range of carbonyl compounds, including mono-
and polycyclic cyclohexanones and cyclohexenones, cy-
clopentanones, and heterocyclic ketones.12
A Monte Carlo conformational search was performed
as described above to locate the lower energy conformers
available for hydride addition to the flexible cyclohep-
tanone system. This search yielded nine unique con-
formers that lie within 4.2 kcal/mol of the global mini-
mum. Because the axial and equatorial nature of the
attack is not always apparent, we describe attack as “top”
and “bottom” in the discussion. The two lowest energy
conformers are virtually identical in energy; these are
bottom hydride addition to 7, and bottom (equivalent to
top addition in this case by C
2
symmetry) hydride
addition to 8. Bottom approach of hydride to 10 also lies
close in energy, only 0.3 kcal/mol above the global
minimum. Finally, bottom addition to 9 is 1.3 kcal/mol
higher in energy than the global minimum. Note that
attempts to model top addition to 10 resulted in a
conformational change to give conformer 8; this is likely
due to steric interactions with the methylene groups
which shield 10 from top attack.
A methyl substituent at the C2 position of cyclohep-
tanone significantly increases the number of conforma-
tions available to this substrate. In this case, Monte
Carlo conformational searches were employed separately
for hydride additions leading to trans- and cis-2-meth-
ylcycloheptanol products. The low-energy conformers are
essentially derived from placing a methyl group at the
carbonyl R positions on the conformers 7-11. The six
transition structures shown in Figure 2 represent the
four lowest energy structures leading to cis products (12,
1
5, 16, and 17), along with the two lowest energy
structures leading to trans products (13 and 14). The
enantiomeric pairs are depicted by a single enantiomer
for the purpose of clarity. In all cases, the methyl takes
the pseudoequatorial position. Structures 12 and 13 are
derived from attack of hydride on opposite faces of
conformer 8, and they result in two of the most favorable
transition structures leading to the cis and trans prod-
ucts, respectively. The energies of 12 and 16 are es-
sentially equal, since the main difference between them
is the position of the remote C4. The low-energy struc-
ture 12 has a Felkin-type arrangement3 with inside
methyl and anti methylene on the front side and outside
methylene on the rear. Structure 13 has a less favorable
inside methyl and outside methylene on the front side
and nearly eclipsing conformation on the rear side. The
source for this energy difference is due to both the
torsional strain with C7-H and the steric hindrance from
a pseudoaxial hydrogen introduced by the 2-methyl group
in 13.
F igu r e 1. MM2* relative energies of cycloheptanone conform-
ers and the MM2* relative energies of transition states for
axial and equatorial attack by hydride on each of these
conformers.
,11
minimum. Conformer 8, a twist conformer, is only 0.34
kcal/mol higher in energy. All of the other structures
were higher in energy than 10 by more than 1.3 kcal/
mol.
Transition structures for hydride addition to cyclohep-
tanone were calculated with an MM2* version of the
MM2 transition structure force field, which has been
successfully applied to stereoselectivities of reductions of
Structures 14 and 15 correspond to top and bottom
hydride attack, respectively, on conformer 9. Structure
15 is more stable than 14 by 1.2 kcal/mol. Structure 9
is similar to benzocycloheptenone; both possess a ring
that is more puckered than cyclohexanone. The calcu-
lated preference for pseudoequatorial (bottom of 9) attack
cyclohexanones,11 2-cyclohexenones, cyclopentanones,
12
11
and benzocycloheptenones,17 as well as Grignard and
organolithium additions to highly sterically hindered
cyclohexanones.10 In the present study, this force field
model was modified and adapted to be included as a

Nucleophilic Additions to Cycloheptanones
J . Org. Chem., Vol. 63, No. 10, 1998 3199
F igu r e 2. Favored transition structures for nucleophilic additions to 2-methylcycloheptanone. Structures 13 and 14 lead to
trans product; structures 12, 15, 16, and 17 lead to cis product.
of hydride is also similar to that of benzocycloheptenone;
the latter was confirmed by experiments.17 Structures
1). Recently, the MM2* transition-state force field
developed for lithium aluminum hydride reduction was
extended to the addition of acetylide anions to carbonyl
compounds.10 This force field was found to give rather
accurate predictions for the diastereoselectivity of acetyl-
ide anion addition to the substituted cyclohexanones.
Monte Carlo conformational searches were performed to
locate the lowest energy conformers of 1 for the transition
structure leading to 2c and 3c. The calculations predict
that acetylide anion will prefer to attack in a trans
fashion to afford the cis-cycloheptanol. The lowest energy
transition structures A and B leading to the cis and the
trans products are shown in Figure 5. A is favored over
B by 1.4 kcal/mol. This amounts to a predicted ratio of
93:7 (cis:trans) at 0 °C. The main source for this
difference can be seen in the dihedral angles with the
vicinal hydrogens (A, 48°; B, 18°). A and B correspond
to 16 and 13, respectively. The transition structure
corresponding to 12 is higher in energy than A by 0.13
kcal/mol. Although our calculation predicts the stereo-
selectivity of 1 with lithium acetylide correctly, the
molecule 1 is so flexible that there may be many
conformations accessible in the transition state. Conse-
quently, we also performed Monte Carlo conformational
searches for the transition structures of 2-methylcyclo-
heptanone, as a model compound, for the reaction of 1
1
6 and 17 both correspond to the bottom hydride attack
on conformers 7 and 10, respectively. These have a
Felkin-type arrangement (S (H) outside, M (Me) inside,
L (CH
2
) anti) and lead to the cis preference.11
To test the MM2* transition-state force field for its
applicability to a cycloheptanone system, we performed
ab initio calculations for LiH attack on 7. The energy
difference obtained is 3.8 kcal/mol by RHF/3-21G, 2.1
kcal/mol by RHF/6-31G*, and 2.3 kcal/mol by MP2/6-
3
3
1G*//RHF/3-21G, favoring bottom attack on 7 (Figure
). These results are comparable to the 3.6 kcal/mol
difference from the MM2* force field (Figure 1). Ab initio
calculations were also performed on the HLi addition to
2
-methylcycloheptanone. We chose the transition struc-
tures 12 and 13, which are the most favorable ones
leading to the cis and trans products, respectively. The
equatorial entry is favored by 1.1 kcal/mol by RHF/3-21G,
1
6
.0 kcal/mol by RHF/6-31G*//3-21G, 0.7 kcal/mol by MP2/
-31G*//RHF/3-21G. Both the energy differencies and
the transition structure conformations are similar to
those from the MM2* force field (Figure 4).
The nucleophilic addition of lithium acetylide to 1 gave
a mixture of the cis and trans alcohols in a ratio of 89:
1
1-86:14 at 0 °C experimentally (entries 13-15 in Table

3
200 J . Org. Chem., Vol. 63, No. 10, 1998
Ando et al.
F igu r e 3. Ab initio transition structures of the LiH addition reaction on 7.
F igu r e 4. Ab initio transition structures corresponding to 12 and 13, respectively.
F igu r e 5. MM2* transition state models for lithium acetylide addition to 1.
with acetylide anion. The lowest energy transition
structures, C and D, leading to the cis and trans alcohols,
transition structure corresponding to A is 0.19 kcal/mol
less stable than C. In this case, the energy difference
between C and D can be also due to the torsional effect
(see Figure 6).
1
8 and 19, respectively, are shown in Figure 6. The
equatorial transition structure C is more stable than D
by 1.1 kcal/mol, which corresponds to a product ratio 18:
1
9 ) 88:12 at 0 °C. We found 12 optimized transition
Con clu sion s
structures leading to 18 and eight optimized transition
structures leading to 19 within 3.0 kcal/mol of the global
minimum transition structure. On the basis of a Boltz-
mann distribution including all transition structures
within 3.0 kcal/mol, the product ratio is predicted to be
Nucleophilic additions of ethyl-, vinyl-, and ethynyl-
lithium and Grignard reagents to a 2-alkyl-substituted
cycloheptanone were performed experimentally. In all
cases, cis-cycloheptanol was formed preferentially. The
MM2* force field was applied to analyze nucleophilic
attack of hydride and ethynyllithium on 2-methylcyclo-
9
3:7. C and D correspond to 12 and 13, respectively. C
has a different cycloheptanone conformation from A. The

Nucleophilic Additions to Cycloheptanones
J . Org. Chem., Vol. 63, No. 10, 1998 3201
F igu r e 6. MM2* transition-structure models for lithium acetylide addition to 2-methylcycloheptanone.
heptanone and gave results comparable to those obtained
from experiments and ab initio calculations. This success
provides support for the proposal that the cis products
from the nucleophilic attack on 2-alkyl-substituted cy-
cloheptanones are favored due to torsional effects.
In this study, we observed higher selectivity when the
size of the nucleophile is increased. With cyclohexanones
this change results in a substantial decrease in the axial
preference, with an equatorial approach being typically
favored with alkyl and vinyl organometallics. With
cycloheptanones, the preference for the cis product is
increased as the size of the nucleophile increases. These
trends result from a combination of steric repulsions and
torsional effects in the transition state.
g, 99.0 mmol) at -78 °C as described to provide crude 1. Flash
column chromatography (silica gel, 5:1 hexanes-EtOAc) gave
1
1
0.0 g (40%) of pure 1: >99% pure by GLC analysis; H NMR
(
2
1
500 MHz, CDCl ) δ 7.25-7.28 (m, 2H), 7.16-7.19 (m, 3H),
3
.60 (app dd, J ) 12.9, 7.4 Hz, 2H), 2.38-2.52 (m, 3H), 1.82-
.86 (m, 4H), 1.72 (app dt, J ) 15.5, 6.7 Hz, 1H), 1.59 (app
13
pentuplet, J ) 7.7 Hz, 3H), 1.27-1.41 (m, 4H); C NMR (125
MHz, CDCl
5.9, 31.9, 31.2, 29.4, 29.0, 28.4, 24.4 ppm; IR (film) 1700 cm
MS (CI) m/z 231.1745 (231.1748 calcd for C16 23O, MH).
3
) δ 216.3, 142.2, 128.3, 128.2, 125.6, 52.1, 42.6,
-1
3
;
H
Gen er a l P r oced u r e for Rea ction of Or ga n om eta llic
R ea gen t s w it h Cycloh ep t a n on e 1. P r ep a r a t ion of
(1R*,2R*)-1-E t h yn yl-2-(3′-p h en ylp r op yl)-1-(t r im et h ylsi-
loxy)cycloh ep ta n e (4c) a n d (1R*,2S*)-1-Eth yn yl-2-(3′-
p h en ylp r op yl)-1-(tr im eth ylsiloxy)cycloh ep ta n e (5c). A
solution of 1 (0.262 g, 1.11 mmol) and THF (5 mL) was cooled
to 0 °C and added dropwise by cannula to a 0 °C solution of
lithium acetylene-ethylenediamine complex (0.515 g, 5.54
mmol) and THF (5 mL). The resulting solution was allowed
to warm to room temperature over 4 h , and after 12 h was
Exp er im en ta l Section 22
2
-(3′-P h en ylp r op yl)cycloh ep ta n on e (1). Following the
2
6
general procedures of Corey and Enders, a THF solution (40
mL) of cycloheptanone N,N-dimethylhydrazone (13.9 g, 90.0
mmol) was deprotonated with LDA (1 M in THF, 99 mmol,
4
quenched with aqueous NH Cl (20 mL). The aqueous phase
was extracted with EtOAc (4 × 20 mL), and the combined
organic layers were washed with brine (4 × 10 mL), dried
-
78 °C for 1 h and 0 °C for 1 h), a solution of 3-bromo-1-
4
(MgSO ), and concentrated. The residue was purified by flash
phenylpropane (19.7 g, 99.0 mmol) and THF (60 mL) was
added at -78 °C, and the resulting colorless solution was
allowed to warm to room temperature. Aqueous workup
provided the alkylated hydrazone, which was dissolved in
column chromatography (10:3, hexanes-ether) to give 0.31 g
1
(76%) of an inseparable mixture of 2c and 3c (6:1 by H NMR
analysis) as a colorless oil.
CH
2
Cl
2
and cleaved with m-chloroperbenzoic acid (81%, 21.1
A sample of a comparable mixture of 2c and 3c (0.388 g,
1
.51 mmol), TMSCl (0.411 g, 3.24 mmol), DMF (15 mL), and
(
22) Tetrahydrofuran (THF) and diethyl ether (Et
2
O) were distilled
imidazole (0.55 g, 8.1 mmol) was maintained at room temper-
ature for 12 h and then cooled in an ice-water bath and
2
quenched with H O (20 mL). The resulting mixture was
from sodium and benzophenone, while toluene was distilled from CaH
2
under reduced pressure. Capillary GC analyses were performed on a
Hewlett-Packard Model 5790A gas chromatograph equipped with a
flame ionization detector and a dimethylpolysiloxane column. Stere-
oisomers 4 and 5 were separated under isothermal conditions at 170
C. Lithium acetylide-ethylenediamine complex, ethylmagnesium
bromide (1 M in THF), and vinylmagnesium bromide (1 M in THF)
were purchased from Aldrich Chemical Co. and used as received.
extracted with ether (50 mL), and the extracts were washed
), and concentrated to
with brine (3 × 30 mL), dried (MgSO
4
°
give a yellow oil that GLC analysis showed was a 7:1 mixture
of 4c and 5c, respectively. Separation by flash column
chromatography (100:1 hexane-CH
pure samples of 4c (0.37 g, 74%) and 5c (0.11 g, 22%). 4c:
NMR (500 MHz, CDCl ) δ 7.26-7.29 (m, 2H), 7.16-7.71 (m,
3H), 2.66-2.71 (m, 1H), 2.57-2.63 (m, 1H), 2.38 (s, 1H, 2.08),
2 2
Cl ) gave isomerically
Ethyllithium (1.2 M in Et
2
O) was made from lithium metal and ethyl
1
bromide.23 Solid vinyllithium was prepared from tetravinyltin. The
24
H
molarity of solutions of these reagents in THF, toluene, or Et O were
2
3
determined by titration with sec-butyl alcohol in xylene and 1,10-
phenanthroline.2
5a
Vinyllithium was titrated using 1,3-diphenylac-
2
.13-2.08 (dd, J ) 14.3, 8.8 Hz, 1H), 1.26-1.83 (m, 14H), 0.19
2
5b
etone p-tosylhydrazone.
described in: Deng, W.; Overman, L. E. J . Am. Chem. Soc. 1994, 116,
1241-11250.
(
Other general experimental details are
13
(s, 9H); C NMR (125 MHz, CDCl
3
) 143.1, 128.6, 128.4, 125.7,
9
0.1, 75.0, 72.0, 51.5, 43.7, 36.6, 31.6, 30.4, 28.8, 27.7, 26.1,
1
23) Bryce-Smith, D.; Turner, E. E. J . Chem. Soc. 1953, 861-867.
24) Seyferth, D.; Weiner, M. A. J . Am. Chem. Soc. 1961, 83, 3583-
21.4, 2.1 ppm; MS (CI) m/z 329.2310 (329.2301 calcd for
33OSi, MH). 5c: 1H NMR (500 MHz, CDCl
) δ 7.26-7.30
m, 2H), 7.17-7.21 (m, 3H), 2.60-2.65 (m, 2H), 2.49 (s, 1H),
(
C
H
21
3
3
586.
(
(
25) (a) Watson, S. C.; Eastham, J . F. J . Organomet. Chem. 1967,
1
0
.93 (app ddd, J ) 3.7, 2.8, 1.8 Hz, 2H), 1.22-1.81 (m, 13H),
9
, 165-168. (b) Lipton, M. F.; Sorensen, C. M.; Sadler, A. C.; Shapiro,
.17 (s, 9H); 13C NMR (125 MHz, CDCl
) δ 143.0, 128.4, 128.2,
R. H. J . Organomet. Chem. 1980, 186, 155-158.
3
(
26) Corey, E. J .; Enders, D. Tetrahedron Lett. 1976, 3-6.
125.5, 86.7, 76.5, 74.0, 51.5, 43.8, 36.3, 31.5, 29.9, 27.6, 27.0,

3
202 J . Org. Chem., Vol. 63, No. 10, 1998
Ando et al.
Ch a r t 1

Nucleophilic Additions to Cycloheptanones
6.9, 20.9, 2.0 ppm; MS (CI) m/z 328.2210 (328.2222 calcd for
J . Org. Chem., Vol. 63, No. 10, 1998 3203
2
) 7.5 Hz, 3H); 13C NMR (125 MHz, CDCl
) δ 142.8, 128.4,
3
C
21
H
32OSi, M).
128.3, 125.6, 77.1, 49.9, 39.7, 36.4, 30.6, 30.3, 29.5, 29.3, 28.6,
(
1R*,2S*)-1-E t h en yl-2-(3′-p h en ylp r op yl)cycloh ep t a n -
27.6, 21.2, 7.4 ppm; MS (CI) m/z 260.2127 (260.2140 calcd for
1
-ol (3b). Following the general procedures of Corey and co-
C
18
H
28O, M).
(1R *,2R *)-1-E t h yl-2-(3′-p h e n ylp r op yl)cycloh e p t a n -
1-ol (2a ) was prepared from pure 2c in an identical fashion:
2
7
workers, a mixture of LiAlH
NaOCH (0.060 g, 1.1 mmol), and THF (2 mL) was stirred at
room temperature for 30 min, and a solution of isomerically
pure alcohol 3c (0.050 g, 0.27 mmol) and THF (2 mL) was then
added by cannula. The resulting mixture was heated at reflux
4
(1 M in THF, 0.550 mL), solid
3
1
3
H NMR (500 MHz, CDCl ) δ 7.26-7.29 (m, 2H,) 7.16-7.22
(m, 3H), 2.67-2.69 (m, 1H), 2.58-2.66 (m, 1H), 1.88 (t, J )
14.1, 8.0 Hz, 1H), 1.36-1.85 (m, 14H), 1.22-1.30 (q, J ) 8.8
Hz, 2H), 0.92-1.12 (q, J ) 9.8 Hz, 1H), 0.87-0.90 (t, J ) 7.5
(
oil bath, 80 °C) for 16 h and after being cooled to room
temperature was quenched with aqueous NH Cl (15 mL). The
aqueous layer was extracted with ether (3 × 25 mL), and the
combined organic layers were dried (MgSO ) and concentrated
to give 0.138 g of a 13:1 mixture ( H NMR analysis) of 3b and
c as a colorless oil. Since these compounds could not be
separated, this mixture was dissolved in toluene (0.5 mL) and
heated at 100 °C in the presence of Ag CO (0.021 g, 0.077
Hz, 3H); 13C NMR (125 MHz, CDCl
) δ 142.6, 128.2, 128.1,
4
3
125.5, 76.6, 47.4, 38.2, 36.3, 32.4, 30.4, 29.9, 29.5, 28.4, 27.1,
22.0, 7.8 ppm; MS (CI) m/z 260.2131 (260.2140 calcd for
C
4
1
18
H
28O, M).
3
3-[(1′R*,2′R*)-2-et h yl-2-h yd r oxycycloh ep t yl]bu t a n oic
1
9
Acid (6). Following the general procedure of Sharpless,
a
2
3
mixture of isomerically pure 2a (0.328 g, 1.26 mmol), NaIO
(4.9 g, 23 mmol), RuCl ‚H O (0.0057 g, 0.028 mmol), H O (20
mL), CCl (10 mL), and MeCN (10 mL) was stirred for 48 h at
room temperature and then quenched with CH Cl (20 mL)
and H O (30 mL). The organic and the aqueous layers were
separated, the aqueous layer was extracted with CH Cl
4
2
8
mmol) to fragment 3c back to cycloheptanone 1. After 16 h,
the reaction was allowed to cool to room temperature and was
diluted with ether (20 mL) and filtered through a short column
of Celite. The filtrate was concentrated, and the residue was
purified by flash chromatography (10:3 hexanes-ether) to give
0
3
2
2
4
2
2
2
2
2
(3 ×
1
.11 g (87%) of pure 3b as a nearly colorless oil: H NMR (500
20 mL) and EtOAc (3 × 20 mL), and the combined organic
MHz, CDCl
3
) δ 7.34-7.37 (m, 2H), 7.24-7.76 (m, 3H), 6.01
layers were washed with aqueous HCl (2 mL, 1 M), dried
(dd, J ) 17.2, 11.0 Hz, 1H), 5.30 (dd, J ) 17.2, 1.1 Hz, 1H),
4
(MgSO ), and filtered. After concentration, the resulting
5
1
1
1
2
.14 (d, J ) 11.0 Hz, 1H,), 2.67-2.73 (m, 1H), 2.61-2.64 (m,
H), 1.89-1.93 (m, 2H), 1.29-1.82 (m, 13H), 1.04-1.07 (m,
yellow oil was triturated with ether (50 mL), and the ether
solution was filtered through a short column of Celite. This
filtrate was concentrated, and the residue was purified by flash
chromatography (1:1 hexanes-EtOAc, containing 1% HOAc)
to give 0.0703 g (25%) of 6 as a colorless solid. Recrystalliza-
1
3
H); C NMR (125 MHz, CDCl ) δ 142.9, 142.8, 128.3, 128.2,
3
25.6, 111.6, 78.3, 49.1, 42.9, 36.3, 31.9, 30.4, 30.2, 29.9, 28.4,
1.7 ppm; MS (CI) m/z 259.2063 (259.2061 calcd for C18
H
27O,
MH).
1R*,2R*)-1-E t h en yl-2-(3′-p h en ylp r op yl)cycloh ep t a n -
-ol (2b) was prepared from pure 2c in an identical fashion:
tion from (hexanes-EtOAc) gave single crystals: mp 90-91
1
(
°C; H NMR (500 MHz, CDCl
3
) δ 5.25-6.25 (s, 1H), 2.31-2.41
1
(m, 2H), 1.78-1.90(m, 3H), 1.09-1.69 (m, 15H), 0.86-0.89 (t,
1
13
H NMR (500 MHz, CDCl
m, 3H), 5.93 (dd, J ) 17.2, 10.6 Hz, 1H), 5.18 (dd, J ) 17.2,
3
) δ 7.25-7.29 (m, 2H), 7.16-7.18
J ) 7.5 Hz, 3H); C NMR (500 MHz, CDCl
3
) δ 179.2, 77.1,
(
47.4, 38.2, 34.3, 32.4, 29.8, 29.5, 28.4, 27.1, 23.6, 22.1, 7.9 ppm;
IR (film) 3448, 2807, 1709, 1686 cm .
-
1
1
1
.47 Hz, 1H), 5.03 (dd, J ) 10.6, 1.2 Hz, 1H), 2.60-2.63 (m,
H), 2.51-2.55 (m 1H), 1.26-1.87 (m, 15H), 1.03-1.09 (m, 1H);
1
3
Ack n ow led gm en t. We are grateful to the National
Institute of General Medical Sciences, National Insti-
tutes of Health for financial support (K.N.H. and L.E.O.)
and to Fujisawa Foundation for a scholarship to K.A.
C NMR (125 MHz, CDCl ) δ 146.4, 142.7, 128.3, 128.2, 125.6,
3
1
10.6, 77.7, 47.6, 41.8, 36.2, 30.3, 30.3, 29.4, 29.0, 26.7, 21.5
27O, MH).
1R *,2S *)-1-e t h yl-2-(3′-p h e n ylp r op yl)cycloh e p t a n -
-ol (3a ). A mixture of isomerically pure 3c (0.038 g, 0.15
ppm; MS (CI) m/z 259.2053 (259.2061 calcd for C18
H
(
1
mmol), EtOAc (10 mL), and 10% palladium on carbon (0.023
g) was maintained at room temperature under an atmosphere
Ap p en d ix
Chart 1 lists the parameters used for the MM2* force
field in Macromodel.
of H
being purged with N
column of Celite and the filtrate was concentrated to give
2
until 3c was no longer detected by GLC analysis. After
2
, the mixture was filtered through a short
Su p p or tin g In for m a tion Ava ila ble: Copies of 13C NMR
spectra (125 MHz, CDCl ) for compounds 1, 2a , 2b, 3a , 3b, 4c
3
and 5c (7 pages). This material is contained in libraries on
microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
1
0
.0392 g (∼100%) of 3a as a yellow oil: H NMR (500 MHz,
CDCl ) δ 7.26-7.30 (m, 2H), 7.16-7.20 (m, 3H), 2.66-2.68 (m,
H), 2.57-2.61 (m, 1H), 1.12-1.80 (m, 18H), 0.87-0.92 (t, J
3
1
(
27) Corey, E. J .; Katzenellenbogen, J . A.; Posner, G. H. J . Am.
Chem. Soc. 1967, 89, 4245-4247.
28) Lenz, G. R. J . Chem. Soc., Chem. Commun. 1972, 468468.
(
J O971627B