Titanium(IV)-Catalyzed Dynamic Kinetic Asymmetric Transformation of Alcohols, Silyl Ethers, and Acetals under Carbon Allylation

Article abstract of DOI:10.1002/anie.200352128

High yields and enantioselectivities are obtained in a dynamic kinetic asymmetric transformation in which the C-O bond of alcohols, silyl ethers, and acetals are substituted by a C-C bond. For example, the racemic indanyl silyl ether 1 is converted into (S)-2 in the presence of the chiral Lewis acidic Ti^IV complex 3.

Full text of DOI:10.1002/anie.200352128

Communications
Carbon Allylation
Titanium(iv)-Catalyzed Dynamic Kinetic
Asymmetric Transformation of Alcohols, Silyl
Ethers, and Acetals under Carbon Allylation**
Manfred Braun* and Wolfgang Kotter
Dedicated to Professor Reinhard W. Hoffmann
on the occasion of his 70th birthday
Reactions that allow the complete conversion of an achiral
starting material into an enantiomerically pure product are in
accord with the principle of “chiral economy”.[1] Comparable
efficiency and utility is provided by reactions in which both
enantiomers of a racemic substrate are converted into a
single, enantiomerically pure product. Such processes, which
are particularly advantageous when they are combined with
the formation of a carbon–carbon bond, rely on either
kinetically[2] or thermodynamically[3] controlled dynamic
resolutions. The large majority of the hitherto known dynamic
resolutions are mediated by enzymes,[2c,e] and their scope was
enhanced in the 1990s by combining enzymatic conversions
with transition-metal catalysis.[2e,4] The relatively rare non-
enzymatic methods rely partly on stoichiometric amounts of
chiral auxiliaries[5] or co-solvents[3,6] and partly on transition-
metal catalysts.[7]
Only very few dynamic resolutions have been disclosed
that lead to concomitant carbon–carbon bond formation at a
stereogenic center and do not require an auxiliary reagent in
stoichiometric amounts.[7e] A prerequisite for both enantiom-
ers of a racemic starting material to be converted into an
enantiomerically pure product is an in situ racemization. The
term “dynamic kinetic asymmetric transformation”
(DYKAT)[2d,7g] has been proposed for such cases in which
the racemization occurs as a result of the configurational
lability of organometallic intermediates, which subsequently
react to give the final products. Herein we describe a novel,
titanium(iv)-catalyzed substitution of a carbon–oxygen bond
by a carbon–carbon bond, a method that relies on a dynamic
kinetic conversion and leads to products that result from a
carbon allylation in a highly enantioselective manner.
The carbon allylation of tertiary alcohols, a transforma-
tion that opens a simple possibility for the formation of
quaternary carbon centers,[8] was observed in the course of the
geminal dialkylation of ketones;[9] it takes place upon treat-
ment with allylic silanes and is mediated by stoichiometric
amounts of strong Lewis acids.[10] Accordingly, treatment of
indanol 1a[11] with allyltrimethylsilane in the presence of
[*] Prof. Dr. M. Braun, Dipl.-Chem. W. Kotter
Institut für Organische und Makromolekulare Chemie
Heinrich-Heine-Universität Düsseldorf
Universitätsstrasse 1, 40225 Düsseldorf (Germany)
Fax: (+49)211-81-15079
E-mail: braunm@uni-duesseldorf.de
[**] This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie.
514
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200352128
Angew. Chem. Int. Ed. 2004, 43, 514 –517

Angewandte
Chemie
titanium tetrafluoride results in the carbon-allylation product
rac-2, which features a quaternary carbon center (Scheme 1).
As a stereoselective variant of the carbon allylation of tertiary
alcohols was unknown, we first studied whether racemic
alcohols such as 1a can be converted stereoselectively into
nonracemic allylation products when the reaction is mediated
by a chiral Lewis acid. The difluorotitanium complex 3,[12]
accessible from (R)-2-amino-1,1,2-triphenylethanol, seemed
to be a suitable chiral reagent with sufficient Lewis acidity
(Scheme 1).
found in the Me3SiOSiMe3 by-product, whose formation was
proven by gas chromatography by comparison with an
authentic sample.
The essentially complete conversion of the racemic
substrate rac-1b into the allylation product (S)-2 with high
enantiomeric excess is rationalized as a dynamic kinetic
asymmetric transformation (Scheme 2): It is assumed that,
First, the racemic alcohol rac-1a was allowed to react with
allyltrimethylsilane in the presence of a stoichiometric
amount of the titanium complex 3 (Scheme 1). The alkene
Scheme 2. Dynamic kinetic asymmetric carbon allylation of racemic 1b
into (S)-2, mediated by the titanium(iv) complex 3. 13C NMR chemical
shift values for the carbinyl carbon atoms of 1a and 1b as well as of
1a und 1b in the presence of equimolar amounts of 3 and TiF4 (sol-
vent: CD3CN).
Scheme 1. Conversion of racemic 1a or 1b into (S)-alkene 2.
first, the chiral Lewis acid 3 and both enantiomers of the silyl
ether 1b form two diastereomeric contact ion pairs, (R)-1b·3
and (S)-1b·3. They rapidly equilibrate via the planar, achiral
carbenium ion 4. In the subsequent reaction with allyltrime-
thylsilane, one of the ion pairs, (S)-1b·3, is postulated to react
distinctly faster than its diastereomer (R)-1b·3. Presumably,
the allyl residue attacks the indanyl cation from the face that
is not occupied by the titanium.[14]
(S)-2 was thus obtained with 81% ee. The fact that its
chemical yield amounted to 94% clearly shows that both
enantiomers of the starting material had been converted.
Therefore, the reaction cannot be a simple kinetic resolution.
Presumably, owing to the protic substrate 1a, the Lewis acid 3
is consumed in the course of the reaction. Thus, stoichiometric
amounts of the titanium complex are necessary—clearly an
unfavorable situation with respect to the “chiral economy” of
the reaction. In the search for a solution to this problem, the
hydroxy group in the substrate rac-1a was replaced by a
different leaving group.
When the silyl ether rac-1b, readily available by protec-
tion of rac-1a according to standard procedures,[13] was
submitted to a reaction with allyltrimethylsilane, it turned
out that just 0.1 equivalents of the titanium compound 3 was
sufficient to bring about a 96% yield of (S)-2. Furthermore,
the enantiomeric excess was enhanced to 98.9% ee, a clear
improvement over the reaction with 1a. In this catalytic
variant of the reaction, the titanium complex 3 is evidently not
blocked by the silyloxy leaving group. Instead, the latter is
The 13C NMR spectra of 1:1 mixtures formed from the
substrates 1a or 1b and the titanium complex 3 in CD3CN
give an indication of a significant partial positive charge at the
carbinyl carbon atom—clearly as a result of a strongly
polarized carbon–oxygen bond. Both the alcohol 1a and the
silyl ether 1b display substantial downfield shifts of the signals
for the carbinyl carbon atom. As shown in Scheme 2,
comparable downfield shifts also result when the substrates
1a or 1b are measured in the presence of equimolar amounts
of titanium tetrafluoride.[15]
To check the scope of this enantioselective carbon
allylation procedure, a series of further substrates 5–8 were
allowed to react with allyltrimethylsilane in the presence of
the chiral Lewis acid 3 (Table 1). In the case of the
tetrahydronaphthyl ether 5, a slightly lower enantiomeric
Angew. Chem. Int. Ed. 2004, 43, 514 –517
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
515

Communications
Table 1: Carbon allylation of ethers 5 and 6 as well as of the N,O- and
O,O-acetals 7 and 8 into alkenes 9–12.
with 9-BBN). The absolute configurations of the alkene 10
and the carbamate 11 were assigned by comparison of their
Substrate Alkene
Yield [%]
ee [%]
Conditions
96
93
[a]
62
82
90
56
[a]
[a]
optical rotations with the corresponding data described in the
literature.[18] The indanyl derivative 2 was converted into the
ester 15, whose configuration is known:[19] A rhodium(iii)
chloride catalyzed shift of the carbon–carbon double bond[20]
delivered the alkene 14, whose ozonolysis and subsequent
oxidative workup gave the corresponding carboxylic acid,
which was finally esterified by treatment with diazomethane
to yield (R)-15. Its configuration was determined by the
comparison of the optical rotations.[19] Bearing in mind the
formal inversion of the configuration owing to a change of
priorities in the course of the transformation, the S configu-
ration was assigned to 2. By analogy, the tetrahydronaphtha-
lene derivative 9 is assumed to have the same configuration.
By means of the chiral titanium(iv) complex 3, the
substitution of a hydroxy, silyloxy, or alkoxy group by an
allylic residue is possible, for the first time, in a dynamic
kinetic asymmetric transformation. Substrates that carry
leaving groups in strongly activated positions are suitable
starting materials, and the method allows the generation of
quaternary carbon centers in a highly enantioselective
manner. The procedure is not restricted to substrates whose
stereogenic carbon center is incorporated in a cyclic skeleton.
75
82
54
79
[b]
[c]
[a] 3: 10 mol%. [b] 3: 10 mol%, TiF4: 90 mol%. [c] 3: 200 mol%.
excess was observed in the alkene 9 relative to the corre-
sponding indanyl derivative. Furthermore, the silyl ether 6
derived from 1-phenylethanol was used as a substrate to find
out whether the method is limited to substrates whose
stereogenic center is incorporated in a carbocycle or hetero-
cycle. Evidently, this limitation does not apply: The allylation
product (S)-10 was also obtained with an enantiomeric excess
of 90%. This result also shows that the method is not
restricted to the silyl ethers of tertiary alcohols. Finally,
heterocyclic substrates were also used. The benzyloxycar-
bonyl-protected[16] piperidines 7a[17a] and 7b[17b] yielded the
expected carbon-allylation product 11 when treated with
allyltrimethylsilane. In the presence of 0.1 equivalents of the
chiral Lewis acid 3, the alkene (R)-11 was obtained with
56% ee from either N/O-acetal 7a or 7b.
Experimental Section
(S)-1-(2-Propenyl)-1-methylindane [(S)-2]: In a 50-mL two-necked
flask, equipped with a magnetic stirrer, a septum, and a connection to
a combined nitrogen/vacuum line, a solution of the titanium complex
3 (0.021 g, 0.035 mmol) in absolute CH2Cl2 (12 mL) was stirred at
À 788C under nitrogen. In a second 25-mL two-necked flask with the
same setup, siloxyindane 1b (0.077 g, 0.35 mmol) was dissolved in
absolute CH2Cl2 (3 mL) and subsequently added dropwise through a
cannula to the solution of the titanium complex 3. Allyltrimethylsi-
lane (0.048 g, 0.42 mmol) was then injected by syringe. The mixture
was allowed to warm to 08C over 12 h, and stirring was continued for
another 6 h at the same temperature. After the addition of a saturated
aqueous solution of NH4F (15 mL), the organic layer was separated,
and the aqueous phase was extracted with CH2Cl2 (3 ” 15 mL). The
combined organic layers were washed with a saturated aqueous
solution of NaF (10 mL) and dried with Na2SO4. The solvent was
removed on a rotary evaporator, and the oily residue was purified by
distillation to yield (S)-2 (0.058 g, 94%). B.p. 1188C/12 mbar; HPLC
(Chiracel OJ; Daicel Chemical Industries; n-hexane/2-propanol,
96:4): (S)-2: tR = 8.2 min, (R)-2: tR = 9.4 min; 98.9% ee; 1H NMR
(500 MHz, CDCl3, TMS): d = 1.24 (s, 3H; CH3), 1.81 (dt, Jd = 12.6 Hz,
Jt = 7.6 Hz, 1H; 2-Ha), 2.05 (dt, Jd = 12.6 Hz, Jt = 7.6 Hz, 1H; 2-Hb),
In all the carbon-allylation reactions described so far, the
titanium(iv) complex
3 was used in substoichiometric
amounts, usually 0.1 equivalents. However, when these con-
ditions were applied to the tetrahydropyranyl ether 8, the
allylation product was obtained in only 9%. It seems that the
chiral Lewis acid 3 is bound irreversibly to the tetrahydro-
pyran 8. To enforce the substoichiometric use of the
alkoxyimine titanium(iv) complex 3, a mixture of the Lewis
acid 3 and titanium tetrafluoride (1:9) was employed. Indeed,
these conditions, which aimed at an in situ regeneration of the
chiral titanium complex 3, led to the formation of the alkene
12 in 75% yield. The moderate enantioselectivity (54% ee)
could be enhanced to 79% ee when a larger amount of the
chiral Lewis acid 3 (2.0 equivalents) was used.
The ee values for the carbon-allylation products 2 and 9–
11 were determined in each case by HPLC on a chiral column
(chiracel OJ). The enantiomeric ratio of the tetrahydropyran
derivative 12 could only be determined by HPLC after its
conversion into the alcohol 13 (by means of hydroboration
=
2.29 (m, 2H; CH2CH ), 2.87 (t, J = 7.6 Hz, 2H; 3-H), 5.01 (d, Jcis
=
=
=
10.0 Hz, 1H; CHH CH), 5.01 (d, Jtrans = 17.0 Hz, 1H; CHH CH),
5.72 (m, 1H; CH2 CH), 7.15 ppm (m, 4H; aromatic H); 13C NMR
=
(125 MHz, CDCl3, TMS): d = 26.6, 30.1, 38.2, 45.7, 47.2, 117.1, 122.6,
124.3, 125.7, 126.1, 135.7, 143.2, 151.0 ppm; GC/MS (70 eV; tR =
9.28 min): m/z (%): 172 (7) [M+], 170 (38) [M+ À 2H], 155 (100)
516
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 514 –517

Angewandte
Chemie
C12H11; elemental analysis: calcd for C13H16 (172.27): C 90.64, H 9.36;
found C 90.62, H 9.38.
Angew. Chem. Int. Ed. Engl. 1987, 26, 477, and references
therein.
[16] A. Giovannini, D. Savoia, A. Umani-Ronchi, J. Org. Chem. 1989,
54, 228.
Received: June 12, 2003 [Z52128]
[17] a) O. Okitsu, R. Suzuki, S. Kobayashi, J. Org. Chem. 2001, 66,
809; b) prepared by silylation of N-Cbz-protected 2-hydroxypi-
peridine.[13]
[18] a) 10: L. Lardicci, P. Salvadori, A. M. Caporusso, R. Menicagli,
E. Belgodere, Gazz. Chim. Ital. 1972, 102, 64; b) 11: T. Yasuo, A.
Kumiko, T. Kentaro, A. Mitoshi, K. Hye-Sook, W. Yusuke, H.
Takahashi, Tetrahedron 2001, 57, 1213.
À
Keywords: allylation · asymmetriccatalysis · C C coupling ·
Lewis acids · nucleophilic substitution
.
[1] A. Fischli, Chimia 1976, 30, 4.
[2] Reviews: a) R. S. Ward, Tetrahedron: Asymmetry 1995, 6, 1475;
b) R. Noyori, M. Tokunaga, M. Kitamura, Bull. Chem. Soc. Jpn.
1995, 68, 36; c) H. Stecher, K. Faber, Synthesis 1997, 1; d) E. J.
Ebbers, G. J. A. Ariaans, J. P. M. Houbiers, A. Bruggink, B.
Zwanenburg, Tetrahedron 1997, 53, 9417; e) U. T. Strauss, U.
Felfer, K. Faber, Tetrahedron: Asymmetry 1999, 10, 107; f) S.
Caddick, K. Jenkins, Chem. Soc. Rev. 1996, 25, 447; g) F. F.
Huerta, A. B. E. Minidis, J.-E. Bäckvall, Chem. Soc. Rev. 2001,
30, 321.
[19] H. des Abbayes, M. A. Boudeville, Tetrahedron Lett. 1976, 2137.
[20] See: M. Ando, N. Kataoka, M. Yasunami, K. Takase, N. Hirata,
Y. Yanagi, J. Org. Chem. 1987, 52, 1429.
[3] Review: P. Beak, D. R. Anderson, M. D. Curtis, J. M. Laumer,
D. J. Pippel, G. A. Weisenburger, Acc. Chem. Res. 2000, 33, 715.
[4] P. M. Dinh, J. A. Howarth, A. R. Hudnott, J. M. J. Williams, W.
Harris, Tetrahedron Lett. 1996, 37, 7623; J. V. Allen, J. M. J.
Williams, Tetrahedron Lett. 1996, 37, 1859; M. T. Reetz, K.
Schimossek, Chimia 1996, 50, 668; A. L. E. Larsson, B. A.
Persson, J.-E. Bäckvall, Angew. Chem. 1997, 109, 1256; Angew.
Chem. Int. Ed. Engl. 1997, 36, 1211; R. Stürmer, Angew. Chem.
1997, 109, 1221; Angew. Chem. Int. Ed. Engl. 1997, 36, 1173, and
references therein.
[5] S. G. Davies, D. Middlemiss, A. Naylor, M. Wills, J. Chem. Soc.
Chem. Commun. 1990, 797; R. W. Hoffmann, T. Rühl, F.
Chemla, T. Zahneisen, Liebigs Ann. Chem. 1992, 719; D. S.
Matteson, H.-W. Man, J. Org. Chem. 1994, 59, 5734; D. Enders,
R. Maaßen, J. Runsink, Tetrahedron: Asymmetry 1998, 9, 2155;
R. N. Ben, T. Durst, J. Org. Chem. 1999, 64, 7700; A. Kubo, H.
Kubota, M Takahashi, K. Nunami, J. Org. Chem. 1997, 62, 5830;
P. N. Devine, U.-H. Dolling, R. M. Heid, Jr., D. M. Tschaen,
Tetrahedron Lett. 1996, 37, 2683; T. Rein, R. Kreuder, P.
von Zezschwitz, C. Wulff, O. Reiser, Angew. Chem. 1995, 107,
1099; Angew. Chem. Int. Ed. Engl. 1995, 34, 1023.
[6] T. Heinl, S. Retzow, D. Hoppe, G. Fraenkel, A. Chow, Chem.
Eur. J. 1999, 5, 3464.
[7] a) R. Noyori, T. Ohkuma, Angew. Chem. 2001, 113, 41; Angew.
Chem. Int. Ed. 2001, 40, 40; b) E. Coulon, M. C. C. de Andrade,
V. Ratovelomanana-Vidal, J.-P. GenÞt, Tetrahedron Lett. 1998,
39, 6467; c) J. M. Ready, E. N. Jacobsen, J. Am. Chem. Soc. 1999,
121, 6086; d) B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 1999,
121, 3543; e) G. C. Lloyd-Jones, C. P. Butts, Tetrahedron 1998, 54,
901; f) J. Liang, J. C. Ruble, G. C. Fu, J. Org. Chem. 1998, 63,
3154.
[8] Reviews: E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998,
110, 403; Angew. Chem. Int. Ed. 1998, 37, 388; J. Christoffers, A.
Mann, Angew. Chem. 2001, 113, 4725; Angew. Chem. Int. Ed.
2001, 40, 4591.
[9] Review: M. T. Reetz, Top. Curr. Chem. 1982, 106, 1.
[10] T. Seitz, Dissertation, Universität Marburg, 1988.
[11] E. C. Friedrich, D. B. Taggart, J. Org. Chem. 1978, 43, 805.
[12] R. Fleischer, H. Wunderlich, M. Braun, Eur. J. Org. Chem. 1998,
1063.
[13] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic
Synthesis, Wiley, New York 1991, p. 68 –73.
[14] See: H. Sakurai, Pure Appl. Chem. 1982, 54, 1; I. Fleming, J.
Dunoguꢀs, R. Smithers, Org. React. 1989, 37, 57.
[15] The downfield shift of more than 35 ppm measured in this case is
distinctly larger than that observed upon complexation of
ketones with TiCl4 or CH3TiCl3 (approximately 11 ppm); see:
M. T. Reetz, M. Hüllmann, T. Seitz, Angew. Chem. 1987, 99, 478;
Angew. Chem. Int. Ed. 2004, 43, 514 –517
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
517