A novel method to generate chiral quaternary carbon centers of high enantiomeric purity using a highly stereoselective addition of vinylalanes to a chiral aldehyde

Article abstract of DOI:10.1002/1521-3773(20000602)39:11<1930::AID-ANIE1930>3.0.CO;2-J

The anti addition of cyanocuprates to chiral pivalates 2 derived from aldehyde 1 generates aldehydes 3 with a chiral quaternary carbon center. The generation of the alcohol precursors was achieved by a highly stereoselective addition of vinylalanes on the aldehyde auxiliary. Piv = pivaloyl = 2,2- dimethylpropyl.

Full text of DOI:10.1002/1521-3773(20000602)39:11<1930::AID-ANIE1930>3.0.CO;2-J

COMMUNICATIONS
0.025 mm3, space group C2/m, a ˆ 22.286(5), b ˆ 4.1760(8), c ˆ
ABi4Se7 in a single-crystal to single-crystal fashion. We
anticipate that the zipperlike action in these materials will
also be feasible using electrochemical means.
19.794(4) Š,
b ˆ 116.89(3)8, Vˆ 1642.9(6) Š3,
Z ˆ 4,
1calcd
ˆ
5.960 gcmÀ3, 2qmax ˆ 50.448, Tˆ 293(2) K, 4127 reflections collected,
1651 independent, 1186 observed [I > 2s(I)], m ˆ 61.13 mmÀ1, Tmax/min
ˆ
1.00/0.51, 80 parameters, final R indices [I > 2s(I)]: R1 ˆ 0.082, wR2 ˆ
0.209, max/min residual electron density 5.64/ À 3.75 eŠÀ3. CsBi4Se7:
Crystal dimensions 0.42  0.02 Â0.02 mm3, space group C2/m, a ˆ
22.224(7), b ˆ 4.178(1), c ˆ 19.742(6) Š, b ˆ 116.384(1)8, Vˆ
1642.2(9) Š3, Z ˆ 4, 1calcd ˆ 6.154 gcmÀ3, 2qmax ˆ 57.168, Tˆ 173.1(1) K,
7393 reflections collected, 2154 independent, 1201 observed [I > 2s(I)],
m ˆ 60.39 mmÀ1, Tmax/min ˆ 1.00/0.46, 80 parameters, final R indices [I >
2s(I)]: R1 ˆ 0.053, wR2 ˆ 0.123, max/min residual electron density 6.11/
À 3.59 eŠÀ3. Further details on the crystal structure investigations may
be obtained from the Fachinformationszentrum Karlsruhe, 76344
Eggenstein-Leopoldshafen, Germany (fax: (‡49)7247-808-666;
e-mail: crysdata@fiz-karlsruhe.de), on quoting the depository numbers
CSD-411071, -411072, -411073, and -411074).
Experimental Section
Rb2Bi4Se7: A mixture of Rb2Se (0.060 g, 0.240 mmol) and Bi2Se3 (0.314 g,
0.480 mmol) was transferred to a carbon-coated silica tube which was
flame-sealed under vacuum. The tube was heated at 8108C for 6 d, then
cooled to 5108C at 78ChÀ1 and to 508C over 10 h. The product consisted of
silver-gray long laths. Semiquantitative energy-dispersive analysis (EDS)
using a scanning electron microscope (SEM) on several needles gave an
average composition of Rb2Bi4Se6.8. Cs2Bi4Se7: The procedure was similar
to the one above but using a mixture of Cs2Se (0.060 g, 0.174 mmol) and
Bi2Se3 (0.228g, 0.348mmol). A single-phase product was obtained. The
optical band gaps of these two compounds were measured spectroscopi-
cally and are 0.53 and 0.56 eV, respectively.
[8] a) SMART versions4 and 5 (1996 ± 1998), SAINT versions 4, 5, and 6
(1994 ± 1999), SADABS, SHELXTL V-5, Bruker Analytical Xray
Systems Inc. Madison, Wisconsin 53719 USA; b) G. M. Sheldrick,
University of Göttingen, Germany.
ABi4Se7 (A ˆ Rb, Cs): Rb2Bi4Se7 (0.1 g, 0.064 mmol) was ground into
powder and added to a 2.5mm solution of I2 in degassed, wet CH3CN
(100 mL) at 238C. The mixture was continuously stirred under an N2
atmosphere. The oxidation to ABi4Se7 was complete in less than 1h. The
optical band gaps of both compounds are ꢀ0.4 eV.
A Bruker SMART Platform CCD diffractometer using graphite mono-
chromatic MoKa radiation (l ˆ 0.71073 Š) was used for data collection in all
cases. Several different sets of frames, covering a random area of the
reciprocal space, were collected using 0.38 steps in w. The SMART[8a]
software was used for data acquisition and the SAINT[8a] program for data
extraction. The absorption correction was carried out with SADABS[8a] and
the structure solution (direct methods) and refinement (full-matrix least
squares on F 2) were done with the SHELXTL[8a] and/or the SHELX97[8b]
package of crystallographic programs.
A Novel Method To Generate Chiral
Quaternary Carbon Centers of High
Enantiomeric Purity Using a Highly
Stereoselective Addition of Vinylalanes
to a Chiral Aldehyde**
Claude Spino* and Christian Beaulieu
Received: December 27, 1999 [Z14475]
In memory of Larry Weiler
[1] a) K. Cheng and B. M. Foxman, J. Am. Chem. Soc. 1979, 99, 8102; b) K.
Honda, F. Nakanishi, N. Feeder, J. Am. Chem. Soc. 1999, 121, 8246;
c) M. Leibovitch, G. Olovsson, J. R. Scheffer, J. Trotter, J. Am. Chem.
Soc. 1998, 120, 12755, and references therein; d) M. Leibovitch, G.
Olovsson, J. R. Scheffer, J. Trotter, J. Am. Chem. Soc. 1997, 119, 1462;
e) T. Suzuki, Pure Appl. Chem. 1996, 68, 28 1.
[2] a) S. Rabe, U. Müller, Z. Anorg. Allg. Chem. 1999, 625(8), 1367; b) U.
Englert, B. Ganter, T. Wagner, W. Klaeui, Z. Anorg. Allg. Chem. 1998,
624(6), 970.
[3] a) C. J. Kepert, M. J. Rosseinsky, Chem. Commun. 1999, 375; b) H. Li,
M. Eddaoudi, M. OꢁKeefe, O. M. Yaghi, Nature 1999, 402, 276.
[4] W. Feitknecht, H. R. Oswald, U. Feitknect-Steinmann, Helv. Chim.
Acta 1960, 43, 1947.
We recently reported on a novel alternative to the
alkylation of chiral enolates using the SN2' addition of
cuprates to chiral carbonates 2 derived from menthone (1,
Scheme 1).[1] Oxidative cleavage of 3 furnished aldehydes,
carboxylic acids, or alcohols having a tertiary a-chiral center
(R2 ˆ H) with enantiomeric purities greater than 99%.
[5] a) T. J. McCarthy, S.-P. Ngeyi, J.-H. Liao, D. C. DeGroot, J. Schindler,
C. R. Kannewurf, M. G. Kanatzidis, Chem. Mater. 1993, 5, 331; b) D.-Y.
Chung, K.-S. Choi, L. Iordanidis, J. L. Schindler, P. W. Brazis, C. R.
Kannewurf, B. Chen, S. Hu, C. Uher, M. G. Kanatzidis, Chem. Mater.
1997, 9, 3060; c) D.-Y. Chung, L. Iordanidis, K.-S. Choi, M. G.
Kanatzidis, Bull. Korean Chem. Soc. 1998, 19, 1283.
[6] a) Rb2Bi4Se7: Crystal dimensions 0.65 Â 0.025 Â 0.025 mm3, space
group P21/m, a ˆ 13.062(2), b ˆ 4.2072(5), c ˆ 15.301(2) Š, b ˆ
93.137(2)8, Vˆ 839.61(17) Š3, Z ˆ 2, 1calcd ˆ 6.169 gcmÀ3
,
2qmax
ˆ
50.188, Tˆ 173(2) K, 3887 reflections collected, 1689 independent,
1201 observed [I > 2s(I)], m ˆ 62.69 mmÀ1, Tmax/min ˆ 0.064/0.029, 85
Scheme 1. Generation of tertiary chiral carbon centers from menthone (1).
parameters, final R indices [I > 2s(I)]: R1 ˆ 0.066, wR2 ˆ 0.16, max/min
À3
residual electron density 6.60/ À 3.15 e Š
. b) Cs2Bi4Se7: Crystal
dimensions 0.5  0.025  0.015 mm3, space group P21/m, a ˆ 13.227(1),
[*] Prof. C. Spino, C. Beaulieu
b ˆ 4.1857(4), c ˆ 15.556(2) Š, b ˆ 94.592(2)8, Vˆ 858.49(14) Š3, Z ˆ 2,
Â
Universite de Sherbrooke
1calcd ˆ 6.400 gcmÀ3
2qmax ˆ 56.928, Tˆ 173.1(1)K, 5978reflections
,
Â
Departement de Chimie
collected, 2224 independent, 1975 observed [I > 2s(I)], m ˆ
Sherbrooke, PQ, J1K 2R1 (Canada)
Fax : (‡1)819-821-8017
59.86 mmÀ1, Tmax/min ˆ 1.00/0.43, 86 parameters, final R indices [I >
2s(I)]: R1 ˆ 0.036, wR2 ˆ 0.094, max/min residual electron density
E-mail: cspino@courrier.usherb.ca
À3
3.79/ À 3.03 eŠ
.
[**] We acknowledge the Natural Sciences and Engineering Council of
Canada and Merck ± Frosst for financial support.
[7] RbBi4Se7: Data on this compound were collected on two different
crystals. One of them was the same crystal (now transformed) initially
used to determine the structure of Rb2Bi4Se7. The other was selected
from a converted batch of sample: Crystal dimensions 0.35 Â 0.025 Â
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
1930
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3911-1930 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 11

COMMUNICATIONS
Quaternary centers of high optical purity are notoriously
difficult to construct. Several ingenious strategies have helped
to surmount this synthetic obstacle, but many compromise on
the generality for the benefit of selectivity.[2±4] The prospect of
making quaternary carbon centers using the approach descri-
bed above (Scheme 1, R2 ˆ Me) was appealing because the
cuprate addition step should be selective regardless of the
substitution on the allylic moiety.[5] However, trisub-
stituted carbonates such as 2 (R2 ˆ Me) did not react with
cuprate reagents because of the high energy required to attain
one of the two reactive conformations A or C (Scheme 2, top).
vinyllithium reagents gave, each time, approximately 2:1
mixtures of two easily separable diastereomeric alcohols.
However, upon adding aldehyde 5 directly to a reaction
mixture containing the carboalumination[6] intermediate of
1-hexyne obtained from a zirconium-catalyzed reaction, we
obtained a surprising ratio of 13:1 of two easily separable
alcohols 6 and 7 in 65% nonoptimized yield. The major
alcohol was shown to have the configuration corresponding to
the Felkin ± Anh[7] addition product 6 f (Scheme 3, Table 1) by
Table 1. Addition of vinylalanes to aldehyde 5.
Entry
8
R1
Yield [%][a]
6:7[b]
OH
OH
OH
1
2
3
4
5
6
7
a
b
c
d
e
cyclohexyl
PhCH2
Ph
75
76
63
688
0
15:1
12:1
20:1
Me
Me
Me
iPr
iPr
iPr
R2
Me
R2
Me
R2
(CH2)3OT:B1 S
À
Me
À
CMe2CH2OTBS
±
A
B
C
À
f
f[c]
(CH2)3CH3
65
93
13:1
17:1[d]
R2 = H, ca. 3 kcal mol–1
R2 = Me, ca. 6 kcal mol–1
0 kcal mol–1
R2 = H, ca. 7 kcal mol–1
R2 = Me, ca. 14 kcal mol–1
À
(CH2)3CH3
[a] Yields of isolated products. [b] Ratios of the isolated products.
[c] Addition on PhMeCHCHO. [d] Ratio determined by NMR integration
of crude mixtures.
OH
OH
H
H
Me
Me
Me
Me
converting it, and its minor diastereomer 7 f, into the
corresponding Mosher esters. This ratio was later reproduced
with other vinylalanes prepared in a similar way. Also, the
addition of vinylalane 8 f to phenylpropanal gave a 93% yield
of a mixture of diastereomeric alcohols in a 17:1 ratio based
on NMR integration of the crude mixture. Preliminary results
therefore indicate that the preference of vinylalanes for the
Felkin ± Anh mode of addition on a-chiral aldehydes could be
general. Instances of increased Felkin ± Anh selectivity with
increasing size of the metal have been noted before.[7] If the
selectivity obtained here was a surprise, the chemical yields of
these additions were also noteworthy. We are aware of only a
few reports of vinylalanes being added directly to ketones or
aldehydes, and usually in only modest yields (30 ± 50%).[8±12]
Trialkylalanes have been added selectively to b-ketosulfox-
ides and chiral b-ketoamides.[13,14] Alkenylalanes prepared
from hydro- or carboalumination of alkynes are usually
treated with other electrophiles[6] or converted into the iodide
or bromide first or transmetalated directly with zinc, cop-
per,[15] lithium (ate complex), or other metals before addition
to carbonyl electrophiles.[16±18] The scope and use of this
protocol to make diastereomerically enriched allylic alcohols
is currently being investigated.
D
E
0 kcal mol–1
2.9 kcal mol–1
Scheme 2. Calculations on the reactive conformations for models of 2 (A,
B, C) and 11 (D, E).
We considered the chiral aldehyde 5 as a surrogate auxiliary
(Scheme 3). It was prepared in two easy steps from menthone
(Wittig olefination with ClPh3PCH2OMe followed by hydro-
lysis with HCl/chloroform). Calculations on a model allylic
Scheme 3. Generation of quaternary chiral carbon centers from 5. Piv ˆ
pivaloyl ˆ 2,2-dimethylpropyl.
On the basis of the work of Goering and co-workers we
expected that derivatization of alcohols 6 into the corre-
sponding pivalates 11 (pivalate ˆ 2,2-dimethylpropanoate)
and the use of cuprate reagents of the type RCuCNMgBr
should lead to highly regioselective reactions in favor of the
SN2' product.[19] This was indeed the case, while other types of
cuprate reagents (for example, R2CuLi) or other leaving
groups (for example, OAc, OCO2Me) gave unsatisfactory
mixtures of regioisomers. Also, the use of a catalytic amount
of CuCN led to the recovery of starting material only.[19]
Reactions using Goeringꢁs procedure with a range of alkyl
cuprates proceeded to give a good yield of the cross-coupling
products 9 and 10 with excellent stereoselectivities (Table 2).
system indicated a 3 kcalmolÀ1 energy difference between the
two reactive conformations D and E (Scheme 2, bottom),
which should provide enough bias to allow the anti-selective
cuprate addition to take place on conformer D only. However,
this change in chiral auxiliary (from menthone to 5) created
some uncertainties concerning issues crucial to the success of
the method. Would the addition of vinyl anions to aldehyde 5
proceed stereoselectively? How could we ensure an SN2'
addition of the cuprate reagent on the corresponding carbo-
nate or ester?
The answer to the first question was initially a resounding
no. Not surprisingly, di- or trisubstituted vinylmagnesium and
Angew. Chem. Int. Ed. 2000, 39, No. 11
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3911-1931 $ 17.50+.50/0
1931

COMMUNICATIONS
Table 2. SN2' displacements of cuprates on the pivalates of alcohols 6.
The present method should prove a valuable addition to the
arsenal of asymmetric reactions available for the synthesis of
natural or unnatural chiral compounds.
Entry
11
R1
R3
9
Yields [%]
d.e.[a] [%]
1
2
3
4
5
6
a
b
c
d
d
e
cyclohexyl
PhCH2
Ph
iPr
iPr
iPr
iPr
Et
nC7H15
a
b
c
d
e
f
95
92
90
95
89
90
> 98
> 98
91
> 98
> 98[b]
> 95[c]
Received: January 20, 2000 [Z14571]
À
(CH2)3OTBS
À
(CH2)3OTBS
[1] a) C. Spino, C. Beaulieu, J. Am. Chem. Soc. 1998, 120, 11832 ± 11833;
b) C. Beaulieu, C. Spino, Tetrahedron Lett. 1999, 40, 1637 ± 1640.
[2] K. Fuji, Chem. Rev. 1993, 93, 2037 ± 2066.
[3] E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110, 402 ± 415;
Angew. Chem. Int. Ed. 1998, 37, 388 ± 401.
[4] K. A. Lutomski, A. I. Meyers, Asymmetric Synthesis. Stereodifferen-
tiating Reactions, Part B, Vol. 3 (Ed.: J. D. Morrison), Academic Press,
New York, 1984, p. 213.
nBu
[a] Determined by comparison of the HPLC traces with authentic mixtures
of 9 and 10. [b] Inferred from the optical rotation in comparison with
literature value. [c] d.e. value based on differences in 13C NMR signals.
[5] T. Ibuka, N. Akimoto, M. Tanaka, S. Nishii, Y. Yamamoto, J. Org.
Chem. 1989, 54, 4055 ± 4061.
[6] For leading references, see E.-i. Negishi, D. Y. Kondakov, Chem. Soc.
Rev. 1996, 417 ± 426, and references therein.
[7] For a review on the stereoselectivity of addition to chiral carbonyl
groups, see A. Mengel, O. Reiser, Chem. Rev. 1999, 98, 1191 ± 1223,
and references therein.
[8] H. Newman, Tetrahedron Letters 1971, 47, 4571 ± 4572.
[9] P. Garner, J. M. Park, E. Malecki, J. Org. Chem. 1988, 53, 4395 ± 4398.
[10] R. S. Coleman, A. J. Carpenter, Tetrahedron Lett. 1992, 1697 ± 1700.
[11] T. Tsuda, T. Yoshida, T. Kawamoto, T. Saegusa, J. Org. Chem. 1987, 52,
1624 ± 1627.
[12] T. Tsuda, T. Yoshida, T. Saegusa, J. Org. Chem. 1988, 53, 1037 ± 1040.
[13] T. Fujisawa, A. Fujimura, Y. Ukaji, Chem. Lett. 1988, 1541 ± 1544.
[14] M. Taniguchi, H. Fujii, K. Oshima, K. Utimoto, Tetrahedron Lett.
1992, 33, 4353 ± 4356.
[15] P. Wipf, Synthesis 1993, 537 ± 557.
[16] E.-i. Negishi, D. Y. Kondakov, D. Choueiry, K. Kasai, T. Takahashi, J.
Am. Chem. Soc. 1996, 118, 9577 ± 9588.
tert-Butyl, phenyl, and benzyl cyanocuprates were not reac-
tive enough to take part in the reaction. The phenyl and
benzyl groups could nonetheless be introduced starting from
phenyl- or benzylacetylene. However, neopentyl alkynes
could not be carboaluminated efficiently. Also, the stereo-
selectivity in the cuprate addition was slightly lower in the
case of phenyl-substituted allyl pivalates 6c and 7c. This
decrease was also observed in our first generation system with
aryl-substituted allyl carbonates (Scheme 1, R1 ˆ Ar).[1b]
The disubstituted double bonds in adducts 9 were easily
cleaved with ozone or with KMnO4/NaIO4 leading to good
yields of aldehydes, carboxylic acids, or alcohols depending on
the work-up performed. For example, 9d was cleaved with O3
in dichloromethane to yield 65% of the aldehyde 12
(Scheme 4). The chiral auxiliary was recovered in 80% yield
[17] K. Maruoka, H. Yamamoto, Tetrahedron 1988, 44, 5001 ± 5032.
[18] For
a review on organoaluminium reagents, see J. R. Hauske,
Additions to C-X p-Bonds, Part 1, Vol. 1 (Ed.: S. L. Schreiber),
Pergamon, Exeter, 1991, pp. 49 ± 75.
[19] C. C. Tseng, S. D. Paisley, H. L. Goering, J. Org. Chem. 1986, 51,
2884 ± 2891.
TBSO
O
a) O3, CH2Cl2
iPr
–78 °C
(CH2)3OTBS
+
5
H
iPr
iPr
b) Ph3P
Me
Me
[20] A. S. C. Prakasa Rao, V. K. Bhalla, U. R. Nayak, S. Dev, Tetrahedron
1973, 29, 1127 ± 1130.
9d
12
Scheme 4. Oxidative cleavage of 9d. TBS ˆ tert-butyldimethylsilyl.
after chromatography and could be re-used. The conversion
of 9e into 14 (Scheme 5) was confirmed by comparison of its
optical rotation ([a]D) with that of its enantiomer reported in
the literature ([a]D ˆ À9.78, CHCl3, c ˆ 0.3; literature [a]D ˆ
‡9.88, c ˆ 0.3).[20] This observation indicated that the stereo-
chemistry of 9e results from an anti-selective cuprate addition
on the configuration and conformation corresponding to D
(Scheme 2).
59
Exceptional Deshielding of Co Caused by
Deuteration of the Hydrogen Bonds in
Cobaloximes**
Fioretta Asaro, Lorenza Liguori, and Giorgio Pellizer*
Transition metal shieldings are useful probes of structure
and reactivity for coordination compounds, since they allow
even tiny variations at the coordination site to be detected.[1]
The isotope 59Co exhibits the largest known shielding range.
Its NMR properties have been extensively studied for
MeO2C
a) TBAF / THF, 100%
TBSO
iPr
iPr
b) Jones’
c) CH2N2 / Et2O
85% for 2 steps
Et
Me
Et
Me
9 e
13
[*] Prof. G. Pellizer, Dr. F. Asaro, Dr. L. Liguori
Dipartimento di Scienze Chimiche
MeO2C
MeO2C
a) KMnO4 / NaIO4
acetone / H2O
Á
Universita di Trieste
via L. Giorgieri 1, 34127 Trieste (Italy)
Fax : (‡39)040-6763903
b) CH2N2, Et2O
40% for 2 steps
Et
Me
14
E-mail: pellizer@dsch.univ.trieste.it
Scheme 5. Proof of the absolute configuration of 9e. TBAF ˆ tetrabutyl-
[**] This work was supported by MURST under the Project Cofin
(MURST 97 CFSIB).
ammonium fluoride.
1932
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3911-1932 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 11