Application of 1,2:5,6-di-O-cyclohexylidene-D-mannitol as the chiral director in Matteson's asymmetric homologation

Article abstract of DOI:10.1016/S0022-328X(99)00066-2

1,2:5,6-Di-O-cyclohexylidene-D-mannitol is the good chiral director in Matteson's asymmetric homologation as indicated by the high enantiomeric excesses (ee's) of the secondary alcohols produced by treatment of the homologation products with alkyllithium

Full text of DOI:10.1016/S0022-328X(99)00066-2

Journal of Organometallic Chemistry 581 (1999) 66–69
Application of 1,2:5,6-di-O-cyclohexylidene- -mannitol as the
D
chiral director in Matteson’s asymmetric homologation
Guisheng Li, George W. Kabalka *
Departments of Chemistry and Radiology, The Uni6ersity of Tennessee, Knox6ille, TN 37996-1600, USA
Received 2 July 1998; received in revised form 3 September 1998
Abstract
,2:5,6-Di-O-cyclohexylidene-D-mannitol is the good chiral director in Matteson’s asymmetric homologation as indicated by the
1
high enantiomeric excesses (ee’s) of the secondary alcohols produced by treatment of the homologation products with alkyllithium
or Grignard reagents followed by oxidation. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Boronates; 1,2:5,6-Di-O-cyclohexylidene-D-mannitol; Homologation
1. Introduction
Asymmetric homologation via a-chloroboronic esters
provides an extremely efficient method for constructing
chiral centers [1]. The reaction has been extensively
employed in syntheses of biologically important
molecules including pheromones [2], homoallylic alco-
hols [3], sugars [4], amino acids [5] and aminoboronic
acids [6]. Pinanediol is often used as the chiral director
but the hydrolytic stability of pinanediol boronates is
In light of the fact that the enantioselectivity in-
creases as the size of the substituent increases in 2a–c,
we felt that the C symmetric 1,2:5,6-di-O-cyclohexyli-
2
problematic in some syntheses. The lack of C symme-
dene-D-mannitol, 3, which is sterically similar to 2b and
2
try is another disadvantage associated with pinanediol
and it accounts for the poor diastereoselectivities ob-
served in the reactions of pinanediol dichloromethyl-
boronate with organolithium or Grignard reagents. To
is commercially available at a modest price, might serve
as an effective chiral director. We wish to report the
results of an investigation in which 3 was used as the
chiral director in asymmetric homologation reactions.
circumvent these difficulties, C symmetric 1,2-diols are
2
being developed as chiral ligands. Diacetone mannitol
and tartrates [7] are ineffective but (R,R)-2,3-butanediol
gives satisfactory results when used as the chiral direc-
tor [8]. 1,2-Di-isopropylethanediol [9] and 1,2-dicyclo-
hexylethanediol [1c,d, 3c] are also effective chiral
directors, providing diastereomeric purities in excess of
2. Results and discussion
The starting 1,2:5,6-di-O-cyclohexylidene-D-mannitol
boronic esters, easily prepared via the quantitative reac-
tions of the corresponding boronic acids and 3, are
stable in air and can be purified by column chromatog-
raphy on silica gel to give crystalline solid products in
98%. However, these chiral diols (1 and 2) are either
expensive or are not yet commercially available.
9
0–95% isolated yields. In order to evaluate 3 as the
chiral director in asymmetric homologations, a series of
reactions (scheme) were carried out under conditions
similar to those described by Matteson [2b]. Preformed
*
Corresponding author. Tel.: +1-423-9743260; fax: +1-423-
9
0
742997.
022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00066-2

G. Li, G. W. Kabalka / Journal of Organometallic Chemistry 581 (1999) 66–69
67
dichloromethyllithium, generated via the slow addition
yield and in 91% ee (entry 1, the absolute configuration
was confirmed by optical rotation). This was quite
promising since pinanediol phenylboronate often pro-
vides lower ee’s than other boronates due to the well-
documented epimerization of the resulting benzylic
a-chloroboronates in the presence of LiCl [10]. For
example, an 88% ee was reported for (S)-1-phenyl-1-
ethanol when pinanediol was used as the chiral director
under otherwise identical conditions [11]. Our results
indicated that the % ee of the final alcohol 7 is depen-
of BuLi to CH Cl in THF at ca. −100°C, readily
2
2
reacted with the boronates to form borate complexes.
After the addition of anhydrous ZnCl , the reaction
2
mixture was warmed to room temperature and stirred
1
for 24 h except in the case of R =Ph (0°C for 1 h).
1
dent upon R . For example, n-butylboronate gave (R)-
1
-phenyl-1-pentanol in 90% ee (entry 2) while the
isopropylboronate produced the corresponding product
in 96% ee (entry 4). t-Butylboronate generated the
desired alcohol in 87% ee but the reaction required
additional time (entry 6). In contrast to reports that the
highest % ee is obtained when an extra equivalent of
ZnCl is added for each oxygenated substituent in the
2
1
substrates [4] , the addition of five equivalents of ZnCl
2
(
4 additional equivalents for the four oxygen atoms in
the mannitol derivative) resulted in a dramatic drop in
ee (entry 3). Apparently, excess ZnCl significantly
The mixture was cooled to −78°C prior to the addi-
tion of R M. Warming to room temperature for 24 h
2
%
converted the a-chloroboronates 5 to boronates 6. The
routine alkaline oxidation of 6 furnished the desired
alcohol 7, along with recovered chiral diol 3 which was
readily separated by column chromatography. The per-
centage enantiomeric excess (% ee) of product alcohols
2
accelerated the epimerization of the intermediate 5 [10].
The results of reactions in which R is cyclohexyl
paralled those obtained with the isopropyl boronate,
yielding the corresponding products in 95% ee (entry 7
and 8). It is noteworthy that the highest ee (99%) was
obtained when the a-chloroboronate (R =c-C H , en-
try 9) was reacted with vinylmagnesium bromide. In
these reactions, a byproduct identified as a mixture of
1
7
were determined by GC analysis of the (+)-menthyl
1
chloroformate derivatives and the results are summa-
rized in Table 1.
The reaction of phenylboronate was examined ini-
tially and (S)-1-phenyl-1-pentanol was obtained in 50%
6
11
(
E)- and (Z)-3-cyclohexyl-2-propen-1-ol (E/Z=85/15)
was isolated in 25% yield. The byproduct could have
resulted from the slow 1,3-allylic rearrangement of the
intermediate boronate
Table 1
1
2
6
(R =c-C H , R =
6 11
Homologation of 1,2:5,6-di-O-cyclohexylidene-
D-mannitol boronates
CH ꢀCH). Such a rearrangement could explain the
using dichloromethyllithium
2
enhanced ee (99%) compared with 95% ee in other
cases, assuming the minor diastereomer of 6 undergoes
a faster allylic rearrangement leaving 6 diastereomeri-
cally enriched. Furthermore, the 1,3-rearrangement of
Entry R1
R M
2
Yield of 7 ee of 7 (%)
b
a
(
%)
1
2
3
4
5
6
7
8
9
1
Ph
n-BuLi
PhLi
PhLi
PhLi
PhLi
PhLi
CH ꢀCHCH MgBr 77
MeMgBr
CH ꢀCHMgBr
n-C H CꢁCLi
50
71
72
65
54
68
91
90
36
96
70
87
95
95
99
–
2
propargylboronate (entry 10, R =n-C H CꢁC) was so
n-Bu
n-Bu
i-Pr
t-Bu
t-Bu
c-C H
c-C H
c-C H
c-C H
3
7
c
rapid that the only isolated product was the corre-
sponding a,b-unsaturated ketone, 1-cyclohexyl-1-hex-
ene-3-one (in 54% yield), which was produced by
oxidation of the allenylic boronate. It is important to
d
6
11
11
11
11
2
2
note that the 1,2:5,6-di-O-cyclohexylidene-D-mannitol
recovered from the reaction in about 85% yield can be
71
41
6
e
6
2
f
0
–
6
3
7
a
1
Isolated yields based on 4.
Determined by capillary GC analysis of the corresponding (+)-
In the reactions of 1,2:5,6-di-O-cyclohexylidene-D-mannitol
b
boronates with dichloromethyllithium, little variation in the % ee was
menthyl chloroformate (99% ee) derivatives of 7.
noted within the range of 0.6 to 1.2 equivalents of ZnCl . 1,2:5,6-Di-
O-cyclohexylidene-D-mannitol is very different from its analog diace-
tone mannitol. When diacetone mannitol was used as the chiral
2
c
Five equivalents of ZnCl2 were used.
d
The reaction mixture was stirred at r.t. for 90 h after addition of
ZnCl .
director, 0.6 equivalent of ZnCl did not improve the diastereoselec-
2
2
e
A mixture of (E)- and (Z)-3-cyclohexyl-2-propen-1-ol (E/Z=85/
tivity while 2.7 equivalents raised the ee from 36% to 82–87% [4]. The
authors thank Dr Matteson for disclosing his unpublished findings
that diacetone mannitol undergoes ketal cleavage when large amounts
of ZnCl2 are used.
1
1
5) was isolated in 25% yield.
No propargylic alcohol was obtained; the major product was
-cyclohexyl-1-hexene-3-one (54%).
f

68
G. Li, G. W. Kabalka / Journal of Organometallic Chemistry 581 (1999) 66–69
repeatedly used without any noticeable deterioration in
asymmetric induction.
temperature (r.t.) for 24 h. The filtrate was concen-
trated and the residue purified by column chromatogra-
phy on silica gel (hexane/ethyl acetate=10/1) to afford
1
a crystalline solid (1.72 g, 90%). H-NMR (CDCl ) l
3
3. Conclusions
7.85–7.75 (m, 2H), 7.50–7.35 (m, 3H), 4.50–4.35 (m,
13
2
H), 4.20–3.90 (m, 6H), 1.80–1.25 (m, 20H). C-NMR
Preliminary results demonstrate that 1,2:5,6-di-O-cy-
(CDCl ) l 134.98, 131.62, 127.79, 110.48, 79.91, 76.36,
3
clohexylidene-
D
-mannitol is a useful chiral director in
65.96, 36.24, 34.51, 25.15, 24.02, 23.75.
Matteson’s asymmetric homologation due to its ready
availability and the high enantiomeric purity of the
homologation products.
4.3. 1,2:5,6-Di-O-cyclohexylidene-D-mannitol-n-
butylboronate
1
H-NMR (CDCl ) l 4.22–4.15 (m, 2H), 4.12–4.00
3
4. Experimental section
(m, 4H), 3.95–3.87 (m, 2H), 1.75–1.35 (m, 24H), 0.95–
13
0
.78 (m, 5H). C-NMR (CDCl ) l 110.35, 79.16, 76.31,
3
4.1. General methods
65.85, 36.17, 34.43, 26.12, 25.17, 23.99, 23.74, 13.84.
All air- and moisture-sensitive reactions were per-
4.4. 1,2:5,6-Di-O-cyclohexylidene-D-mannitol-iso-
formed in flame-dried glassware under a positive pres-
sure of nitrogen. Dry solvents were distilled under
nitrogen prior to use from an appropriate drying agent:
tetrahydrofuran and diethyl ether from the sodium
benzophenone ketyl, dichloromethane from phospho-
rous pentoxide. Boronic acids were prepared according
to literature methods (n-butyl-, isopropyl- and t-butyl-
boronic acids were prepared from triisopropyl borate
and the corresponding alkyllithium or Grignard
reagents followed by acidic hydrolysis [12] while cyclo-
hexylboronic acid was prepared by hydroboration of
cyclohexene with dibromoborane methyl sulfide com-
plex followed by hydrolysis [13]) and used immediately
propylboronate
1
H-NMR (CDCl ) l 4.25–4.15 (m, 2H), 4.12–4.00
3
(m, 4H), 3.95–3.85 (m, 2H), 1.70–1.30 (m,20H), 1.15
(m, 1H), 0.98 (d, 6H, J=6.40). C-NMR (CDCl ) l
110.32, 79.16, 76.31, 65.80, 36.17, 34.46, 25.15, 24.00,
23.75, 18.00.
13
3
4.5. 1,2:5,6-Di-O-cyclohexylidene-D-mannitol t-
butylboronate
1
H-NMR (CDCl ) l 4.25–4.15 (m, 2H), 4.12–4.00
3
(m, 4H), 3.96–3.86 (m, 2H), 1.70–1.32 (m, 20H), 0.95
13
in the preparation of 1,2:5,6-di-O-cyclohexylidene-
D
-
(s, 3H). C-NMR (CDCl ) l 110.27, 79.19, 76.33,
3
mannitol boronates. Anhydrous zinc chloride was pre-
pared by melting zinc chloride (reagent grade), and the
molten zinc chloride was cooled and crushed in a dry
box under nitrogen. All other reagents were purchased
65.75, 36.18, 34.49, 27.00, 25.17, 24.01, 23.77.
4.6. 1,2:5,6-Di-O-cyclohexylidene-D-mannitol cyclo-
hexylboronate
1
from Aldrich and used without further purification. H-
1
3
1
and C-NMR spectra were recorded on a Bruker AC
50 (250.13 and 62.89 MHz respectively) spectrometer.
H-NMR (CDCl ) l 4.23–4.15 (m, 2H), 4.12–3.97
3
1
3
2
(m, 4H), 3.95–3.85 (m, 2H), 1.72–1.28 (m, 31H). C-
1
13
Chemical shifts for H- and C-NMR are reported in
NMR (CDCl ) l 110.32, 79.09, 76.33, 65.81, 36.18,
3
ppm on the l scale with internal reference to SiMe (l
34.46, 27.95, 26.94, 26.70, 25.16, 23.99, 23.75.
4
0
.00) and CDCl (l 77.00), respectively. Coupling con-
3
stants (J) are given in hertz (Hz). GC analysis was
performed on a HP 5890 instrument equipped with a
capillary column (Carbowax, 30 m×0.25 mm ID×
4.7. (S)1-cyclohexyl-1-ethanol [14]
The synthesis is representative: n-BuLi (1.57 ml, 3.94
0.25 mm). Rotations were determined using a Perkin–
mmol) was slowly added to a solution of CH Cl (0.26
2
2
Elmer 241 Polarimeter at room temperature.
ml, 3.94 mmol) in THF (10 ml) cooled to −100°C
ethanol/liquid N ). After addition, the reaction mix-
(
2
4
.2. 1,2:5,6-Di-O-cyclohexylidene-
D
-mannitol-phenyl-
ture was stirred at −100°C for 20 min. To the resulting
white suspension was added a solution of 1,2:5,6-di-O-
boronate
cyclohexylidene-D-mannitol cyclohexylboronate (1.42 g,
The synthesis of 1,2:5,6-di-O-cyclohexylidene-
D
-man-
3.28 mmol) in Et O (5 ml) at −78°C. The clear solu-
2
nitol phenylboronate is representative. A mixture of
tion was stirred at −100°C for another 20 min before
phenylboronic acid (0.558 g, 4.58 mmol), 1,2:5,6-di-O-
anhydrous ZnCl (0.360 g, 2.62 mmol) was added. The
2
cyclohexylidene-
drous MgSO4 (2 g) in Et O was stirred at room
D
-mannitol (1.57 g, 4.58 mmol), anhy-
reaction mixture was gradually warmed to r.t. and
stirred at r.t. for 24 h. The mixture was cooled to
2

G. Li, G. W. Kabalka / Journal of Organometallic Chemistry 581 (1999) 66–69
69
13
−
78°C and CH MgBr (2.84 ml, 8.52 mmol) was added.
C-NMR (CDClS) l 139.81, 115.45, 77.76, 43.49,
3
After stirring at r.t. for 24 h, the mixture was treated
with 3 M NaOH (1.5 ml) and 30% H O (0.8 ml) at 0°C
28.75, 28.33, 26.51, 26.13.
2
2
and then stirred at r.t. overnight. After the usual work-
Acknowledgements
up, chromatographic separation provided (S)-1-cyclo-
3
hexyl-1-ethanol (0.299 g, 71%): [h] =3.51° (c 3.1,
1
We wish to thank the Department of Energy and the
Robert H. Cole Foundation for their support of this
research.
CHCl ); H-NMR (CDCl ) l 3.55 (quint, 1H, J=6.2),
3
3
1
3
1
.90–0.90 (m, 15H). C-NMR (CDCl ) l 71.63, 44.54,
3
28.12, 27.75, 25.92, 25.63, 25.54, 19.83. An ee of 95%
was determined by GC analysis of its (+)-menthyl
chloroformate derivative, compared with the racemic
References
samples.
1,2:5,6-Di-O-cyclohexylidene-D-mannitol
(
0.936 g, 83%) was also recovered.
[
1] (a) D.S. Matteson, Chem Rev. 89 (1989)1535. (b) D.S. Matteson,
Stereodirected Synthesis with Organoboranes, Springer Verlag,
Berlin, 1995. (c) D.S. Matteson, R.P. Singh, C.H. Sutton, J.D.
Verheyden, J. Lu, Heteroatom Chem (1997) 487. (d) D.S. Mat-
teson, H.-W. Man and O.C. Ho, J. Am. Chem. Soc. 118 (1996)
4
.8. (S)-1-Phenyl-1-pentanol [15]
3
1
[
h] = −36.81° (C 3.97, benzene). H-NMR (CDCl )
3
4
560.
l 7.32 (m, 5H), 4.63 (t, 1H, J=6.34), 2.00 (s, 1H), 1.75
(
(
[
[
2] (a) P.B. Tripathy, D.S. Matteson, Synthesis (1990) 200. (b) D.S.
Matteson, K.M. Sadhu, M.L. Peterson, J. Am. Chem. Soc. 108
(1986) 810. (c) D.S. Matteson, K.M. Sadhu, J. Am. Chem. Soc.
13
m, 2H), 1.31 (m, 4H), 0.88 (t, 3H, J=7.11). C-NMR
CDCl ) l 144.90, 128.37, 127.41, 125.86, 74.64, 38.76,
3
1
05 (1983) 2077.
3] (a) R.W. Hoffmann, B. Landmann, Chem. Ber. 119 (1986) 1039.
b) R.W. Hoffman, B. Landmann, Angew Chem. Int. Ed. Engl.
2
7.93, 22.56, 13.95.
(
4
.9. (R)-2-Methyl-1-phenyl-1-propanol [16]
25 (1986) 189. (c) R.W. Hoffmann, K. Ditrich, G. K o¨ ster, R.
St u¨ rmer, Chem. Ber. 122 (1989) 1783.
1
[4] D.S. Matteson, M.L. Peterson, J. Org. Chem. 52 (1987) 5116.
H-NMR (CDCl ) l 7.29 (m, 5H), 4.34 (d, 1H,
3
[
[
5] D.S. Matteson, E.C. Beedle, Tetrahedron Lett. 28 (1987) 4499.
6] (a) D.S. Matteson, K.M. Sadhu, G.E. Lienhard, J. Am. Chem.
Soc. 103 (1981) 5241. (b) D.S. Matteson, K.M. Sadhu,
Organometallics 3 (1984) 614.
J=6.84), 1.93 (m, 2H), 0.99 (d, 3H, J=6.68), 0.79 (d,
3
H, J=6.79).
[
[
7] D.S. Matteson, R. Ray, R.R. Rocks, D.J. Tsai, Organometallics
4
.10. (R)-2,2-Dimethyl-1-phenyl-1-propanol [16]
3
(1983) 614.
8] (a) K.M. Sadhu, D.S. Matteson, G.D. Hurst, J.M. Kurosky,
Organometallics 3 (1984) 804. (b) R.W. Hoffmann, B. Land-
mann, Chem. Ber. 119 (1986) 2013.
3
1
[
h] =31.22° (C 4.4, Et O). H-NMR (CDCl ) l 7.30
2 3
(
m, 5H), 4.40 (s, 1H), 1.35 (s, 1H), 0.92 (s, 9H).
[
9] (a) D.S. Matteson, E.C. Beedle, A.A. Kandil, J. Org. Chem. 52
(
1987) 5034. (b) D.S. Matteson, P.B. Tripathy, A. Sarkar, K.M.
Sadhu, J. Am. Chem. Soc. 111 (1989) 4399.
10] D.S. Matteson, E. Erdik, Organometallics 2 (1983) 1083.
[11] M.V. Rangaishenvi, B. Singaram, H.C. Brown, J. Org. Chem. 56
(1991) 3286.
4.11. (S)-1-Cyclohexyl-3-buten-1-ol [17]
[
3
1
[
h] = −10.15 (C 4.0, EtOH). H-NMR (CDCl ) l
3
5
2
1
2
.82 (m, 1H), 5.15 (m, 2H), 3.40 (m, 1H), 2.45 (m, 1H),
.15 (m, 1H) 1.90–1.00 (m, 12H). C-NMR (CDCl ) l
35.47, 117.91, 74.75, 43.10, 38.82, 29.11, 28.12, 26.53,
6.29, 26.15.
1
3
[
[
12] H.C. Brown, T.E. Cole, Organometallics 2 (1983) 1316.
13] H.C. Brown, N.G. Bhat, V. Somayaji, Organometallics 2 (1983)
3
1
311.
[
[
[
14] D. Seebach, H. Daum, Chem. Ber. 107 (1974) 1748.
15] J.P. Mazaleyrat, D.J. Cram, J. Am. Chem. Soc. 103 (1981) 4585.
16] R. McLeod, F.J. Welch, H.S. Mosher, J. Am. Chem. Soc. 82
4
.12. (S)-1-Cyclohexyl-2-propen-1-ol [18]
(
1960) 876.
17] P.A. Bartlett, W.S. Johnson, J.D. Elliott, J. Am. Chem. Soc. 105
1983) 2088.
[
[
3
o
1
[
h] = −14.32 (C 1.7, CHCl ). H-NMR (CDCl ) l
3 3
(
5
.85 (m, 1H), 5.18 (d, 1H, J=17.50), 5.3 (d, 1H,
J=12.50) 3.85 (t, 1H, J=7.30). 1.90–0.90 (m, 12H).
18] S.K. Aggarwal, J.S. Bradshaw, M. Eguchi, S. Parry, Tetrahe-
dron 43 (1987) 451.
.