Practical synthesis of allylic silanes from allylic esters and carbamates by stereoselective copper-catalyzed allylic substitution reactions

Article abstract of DOI:10.1002/adsc.200404381

The first copper-catalyzed allylic substitution reactions of allylic acetates and carbamates employing a bis(triorganosilyl)zinc reagent are reported. This novel procedure avoids the use of stoichiometric amounts of copper salts which are usually mandatory with this chemistry. Application of this methodology to standard model substrates substantiates a high diastereoselectivity for the double bond geometry (E:Z) as well as the relative configuration (syn:anti).

Full text of DOI:10.1002/adsc.200404381

COMMUNICATIONS
Practical Synthesis of Allylic Silanes from Allylic Esters and
Carbamates by Stereoselective Copper-Catalyzed Allylic
Substitution Reactions
Martin Oestreich,* Gertrud Auer
Institut fꢀr Organische Chemie und Biochemie, Albert-Ludwigs-Universitꢁt Freiburg, Albertstrasse 21,
79104 Freiburg im Breisgau, Germany
Fax: (þ49)-761-203-6100, e-mail: martin.oestreich@orgmail.chemie.uni-freiburg.de
Received: December 14, 2004; Accepted: February 21, 2005
Abstract: The first copper-catalyzed allylic substitu-
tion reactions of allylic acetates and carbamates em-
ploying a bis(triorganosilyl)zinc reagent are report-
ed. This novel procedure avoids the use of stoichio-
metric amounts of copper salts which are usually
mandatory with this chemistry. Application of this
methodology to standard model substrates substanti-
ates a high diastereoselectivity for the double bond
geometry (E:Z) as well as the relative configuration
(syn:anti).
1a–9a and carbamates 1b, 4b–9b under copper catalysis
(Scheme 1). Our novel procedure avoids the use of equi-
molar amounts of copper (based on transferred silyl
moiety) while showing similar diastereoselectivities
and yields as the stoichiometric process developed by
Fleming.[5]
Acetates 1a–9a[11a] as well as carbamates 1b, 4b–
9b[11b] investigated in this survey were synthesized in
good yields from the corresponding alcohols using
standard procedures except for cyclic acetates 2a and
3a, which were directly prepared from alkenes accord-
ing to a procedure by ꢂkermark.[11c]
Keywords: allylic compounds; copper; cuprates; sili-
con; zinc
Allylic silanes are important silicon-based reagents,
which have found broadest application in synthetic
methodology as well as numerous complex molecule
syntheses.[1–3] Several reliable reaction protocols have
been developed for the preparation of allylic silanes.[3]
Among these, the allylic substitution of allylic esters
and carbamates employing silyl cuprates is well estab-
lished.[4–8] Even so, the generation of silicon-copper re-
agents such as (PhMe2Si)2Cu(CN)Li2 requires stoichio-
metric amounts of copper as they are accessed from
the requisite silyllithium (2.0 equivs.) and a copper salt
(1.0 equiv.). This quantitative transmetallation from
lithium to copper appears indispensable since the silyl-
lithium alone is a hard and basic nucleophile.
Based on an almost unnoticed report by Nozaki,[9] we
have recently introduced soft bis(triorganosilyl)zinc
species, such as (PhMe2Si)2Zn, which display decreased
nucleophilicity and basicity when compared with the
corresponding silyllithium.[10] Therefore, these mild
zinc reagents are well suited for copper-catalyzed reac-
tions. In this context, we have realized a facile copper-
catalyzed conjugate addition of (PhMe2Si)2Zn to a,b-
unsaturated carbonyl compounds.[10] In this communica-
tion, we report that bis(triorganosilyl)zinc species are
also applicable to allylic substitution reactions of esters Scheme 1. Copper-catalyzed allylic substitution (SN or SNꢃ).
Adv. Synth. Catal. 2005, 347, 637–640
DOI: 10.1002/adsc.200404381
ꢄ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
637

Martin Oestreich, Gertrud Auer
Table 1. Copper-catalyzed allylic substitution of allylic acetates 1a–9a and carbamates 1b, 4b–9b.[a]
COMMUNICATIONS
Entry
Allylic Substrate
a (R¼Ac) or
E:Z of
Substrate
Allylic Silane
E:Z of Silane
Yield [%]
Yield [%]
(Acetate)[b]
(Carbamate)[b]
b [R¼C(O)NHPh]
1
2
3
4
5
–
–
62
68
82
75
83
95
–
–
–
–
–
–
99: 1[c]
99: 1[d]
99: 1[c]
99: 1[d]
70
65
6
1: 99[d]
1: 99[d]
1: 99[d]
1: 99[d]
1: 99[d]
1: 99[d]
96
98
97
90
97
86
7[e]
8[f]
9
–
50: 50[d]
90
82
10
11
–
–
50: 50[c, g]
85: 15[c]
74
66
68
83
12
–
–
74
83
[a]
All reactions were conducted with a substrate concentration of 0.14 M on a 1.00-mmol scale using 5.0 mol % of CuI (see
Representative Procedure for details).
[b]
[c]
[d]
Yield of analytically pure product after flash chromatography on silica gel.
Determined from the 1H NMR spectra of the crude reaction mixture by integration of baseline separated resonance signals.
Determined from the 13C NMR spectra of the crude reaction mixture since the 1H NMR spectra of the double bond isomers
are almost identical; a single isomer was detected for (E)-14 and (Z)-14, respectively.
Reactions were performed on a 25.0-mmol scale using 5.0 mol % of CuI.
[e]
[f]
[g]
Reactions were performed on a 5.00-mmol scale using 1.0 mol % of CuI.
Ratios ranged from 40: 60 to 60: 40.
638
ꢄ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 637–640

Synthesis of Allylic Silanes from Allylic Esters and Carbamates
COMMUNICATIONS
Simple prenyl acetate 1a and carbamate 1b underwent
clean allylic substitution (SN) when added to a solution
of (PhMe2Si)2Zn pre-treated with catalytic amounts
(5 mol %) of copper iodide but with a higher yield for
the latter substrate (Table 1, Entry 1); likewise, good
yields were obtained for cyclic secondary acetates 2a
and 3a (Table 1, Entries 2 and 3). As expected, allylic
substitution of cinnamic ester (E)-4a and carbamate
(E)-4b proceeded with complete preservation of the
double bond geometry in good yields (Table 1, Entry
4). In order to further verify the stereoselectivity of
the allylic substitution, we independently subjected iso-
merically pure geranyl acetate (E)-5a and neryl acetate
(Z)-5a to these reaction conditions. We were pleased to
find that allylic silanes (E)-14 and (Z)-14, respectively,
were formed with excellent diastereoselectivity in high
yields. The corresponding carbamates (E)-5b and (Z)-
5b performed equally well (Table 1, Entries 5 and 6).
In contrast, allylic substitution (SNꢃ) of linalyl acetate
6a and carbamate 6b provided allylic silane 14 with no
stereocontrol at all (Table 1, Entry 9). The same was ob-
served for secondary acetate 7a and carbamate 7b (Ta-
ble 1, Entry 10); only 8a and 8b showed some diastereo-
selectivity albeit not in synthetical useful E:Z ratios (Ta-
ble 1, Entry 11). These results are in accordance with re-
ported data.[4,5] Tertiary derivatives 9a and 9b furnished
allylic silane 16 in high yields (Table 1, Entry 12).
Scheme 2. Diastereoselective copper-catalyzed allylic substi-
tution (anti-SNꢃ).
As exemplified for the nerol-derived substrates (Z)-5a
and (Z)-5b, the high yields remained completely unaf-
fected on a 25.0-mmol scale (Table 1, Entry 7). Impor-
Upon quantitative deprotonation of the carbamate
tantly, lowering the amount of catalyst from 5.0 mol % moiety followed by the addition of stoichiometric
to 1.0 mol % on a 5.00-mmol scale was also met with suc- amounts of copper iodide and then the silyllithium,
cess (Table 1, Entry 8). Both large-scale reactions and this group is transformed into a reagent-directing
catalyst loadings as low as 1.0 mol % clearly underline group[12] thereby switching the anti-SNꢃ into a syn-SNꢃ
the practicability of this procedure. It is noteworthy that process.[5c] Not unexpectedly, this elegant diastereodi-
all reactions were conducted without using an excess of vergent substitution of allylic carbamates 17b failed
the zinc reagent; these yields were achieved with equimo- for our catalysis! Again, clean anti-selectivity and com-
lar quantitiesof(PhMe2Si)2Znand substrate. We routine- parable yields (78%) were detected for the deprotonat-
ly used copper iodide as the catalyst but other copper ed carbamate.
sources such as copper cyanide also work.
In summary, we have elaborated a practical reaction
Fleming investigated the stereoselectivity of such al- protocol for the first preparation of allylic silanes by a
lylic substitution reactions starting from disubstituted copper-catalyzed allylic substitution of allylic esters
cyclic substrates syn-17 and anti-17,[5c] which are accessi- and carbamates. Apart from good to excellent yields,
ble in diastereomerically enriched form by syn-selective (1) allylic silanes are formed highly diastereoselectively
reduction of the corresponding carbonyl compound[11d] (E:Z) when starting from primary allylic esters 4a–5a
and subsequent Mitsunobu inversion[5d] as an entry and carbamates 4b–5b (Table 1, Entries 4–8) and (2)
into the anti relative configuration. As previously shown high diastereospecifity (anti-SNꢃ) is seen for the model
using stoichiometric amounts of pre-formed silyl cupra- substrates syn-17 and anti-17 (Scheme 2).
te,[5c] allylic substitution of these substrates proceeded
with high diastereoselectivity following an anti-SNꢃ
mechanism. We were able to confirm these observations
for our catalytic procedure (Scheme 2). Both benzoate
Experimental Section
syn-17c and carbamate syn-17b provided allylic silane
Bis(dimethylphenylsilyl)zinc
anti-18 with complete preservation of the (inverted)
stereochemical information (syn-17 ! anti-18). Accord-
ingly, substrates anti-17c and anti-17b with opposite rel-
ative configuration afforded syn-18 (anti-17 ! syn-18).
Dimethylphenylsilyl chloride (427 mg, 2.50 mmol, 2.50
equivs.) was maintained with freshly cut lithium (large excess)
in THF (4 mL) at ꢀ88C under an argon atmosphere for 18 h.
Adv. Synth. Catal. 2005, 347, 637–640
asc.wiley-vch.de
ꢄ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
639

Martin Oestreich, Gertrud Auer
COMMUNICATIONS
In order to separate the formed dimethylphenylsilyllithium
(2.00 mmol, ~80% conversion) from unreacted lithium metal,
the resulting dark red solution was transferred to another flask
via a double-ended cannula. At 08C, ZnCl2 (1.00 mL,
1.00 mmol, 1.00 equiv., 1 M in Et2O) was added accompanied
by a color change from red to yellowish brown. The reaction
mixture was maintained at this temperature for further
15 min and was then ready to use.
Acknowledgements
The research was supported by the Deutsche Forschungsge-
meinschaft (Emmy Noether Fellowship, Oe 249/2–3), the Fonds
der Chemischen Industrie, and the Wissenschaftliche Gesell-
schaft, Freiburg im Breisgau. The authors thank Barbara Wein-
er for preliminary studies and Ilona Hauser for the preparation
of PhMe2SiCl. M. O. is indebted to Professor Reinhard Brꢀck-
ner for his continuous support.
References and Notes
[1] For a review on recent developments, see: L. Chabaud, P.
James, Y. Landais, Eur. J. Org. Chem. 2004, 3173–3199.
[2] a) E. Langkopf, D. Schinzer, Chem. Rev. 1995, 95, 1375–
1408; b) C. E. Masse, J. S. Panek, Chem. Rev. 1995, 95,
1293–1316; c) T. H. Chan, D. Wang, Chem. Rev. 1995,
Representative Procedure for the Copper-Catalyzed
Allylic Substitution
A suspension of CuI (9.6 mg, 0.050 mmol, 5.0 mol %) and THF
(1 mL) was pre-cooled to ꢀ788C and treated with bis(dime-
thylphenylsilyl)zinc (1.00 mmol, 1.00 equiv.) via syringe. The
auburn reaction mixture was allowed to warm to 08C and
maintained at this temperature for 20 min. Addition of either
geranyl/neryl acetate (E)-5a/(Z)-5a (196 mg, 1.00 mmol, 1.00
equiv.) or geranyl/neryl N-phenyl carbamate (E)-5b/(Z)-5b
(273 mg, 1.00 mmol, 1.00 equiv.) in THF (1 mL) was followed
by stirring for 1 h at 08C. Upon completion of the reaction,
the mixture was poured into saturated aqueous NH4Cl
(5 mL) and the flask was rinsed with methyl tert-butyl ether
(10 mL). The aqueous phase was separated and extracted
with methyl tert-butyl ether (3ꢁ10 mL). The combined organ-
ic phases were extracted with brine (10 mL). After drying
(Na2SO4), the solvents were evaporated under reduced pres-
sure and the crude product, (E)-14 or (Z)-14 respectively,
was purified by flash chromatography on silica gel using cyclo-
hexane as solvent.
`
95, 1279–1292; d) I. Fleming, J. Dunogues, R. Smithers,
Org. React. 1989, 37, 57–575.
[3] For an excellent summary, see: T. K. Sarkar, in: Science
of Synthesis, Vol. 4, Georg Thieme Verlag, Stuttgart,
2002, pp. 837–925 and references cited therein.
[4] a) R. K. Dieter, in: Modern Organocopper Chemistry,
(Ed.: N. Krause), Wiley-VCH, Weinheim, 2002, pp. 79–
144; b) I. Fleming, in: Organocopper Reagents. A Practi-
cal Approach, (Ed.: R. J. K. Taylor), Oxford Academic
Press, New York, 1994, pp. 257–292.
[5] a) I. Fleming, T. W. Newton, J. Chem. Soc. Perkin Trans.
1 1984, 1805–1808; b) I. Fleming, A. P. Thomas, J. Chem.
Soc. Chem. Commun. 1985, 411–413; c) I. Fleming, A. P.
Thomas, J. Chem. Soc. Chem. Commun. 1986, 1456–
1457; d) I. Fleming, D. Higgins, N. J. Lawrence, A. P.
Thomas, J. Chem. Soc. Perkin Trans. 1 1992, 3331–3349.
[6] For Me2Cu(CN)Li2-catalyzed allylic substitution of allyl-
ic epoxides using mixed silyl zincates, see: B. H. Lipshutz,
J. A. Sclafani, T. Takanami, J. Am. Chem. Soc. 1998, 120,
4021–4022.
(E)-(3,7-Dimethyl-2,6-octadienyl)dimethylphenylsilane
[(E)-14] (Table 1, Entry 5): Colorless liquid; yield: 83% [from
(E)-5a] and 65% [from (E)-5b]; Rf (cyclohexane)¼0.48; IR
˜
(cuvette): n¼3055 (s), 2985 (s), 2686 (w), 2361 (m), 2307 (s),
2254 (m), 1549 (w), 1427 (s), 1264 (s), 1155 (m), 1113 (m),
1019 (m) cmꢀ1; 1H NMR (CDCl3, 400 MHz): d¼0.26 (s, 6H),
1.50 (s, 3H), 1.61 (s, 3H), 1.63 (d, J¼8.6 Hz, 2H), 1.68 (s, 3H),
1.98 (m, 4H), 5.09 (tt, J¼6.6 Hz, J¼1.3 Hz, 1H), 5.18 (tq, J¼
8.6 Hz, J¼1.3 Hz, 1H), 7.34 (m, 3H), 7.50 (m, 2H); 13C NMR
(CDCl3, 100 MHz): d¼ ꢀ3.2, 15.9, 17.7, 17.8, 25.8, 27.0, 40.1,
119.7, 124.7, 127.8, 128.9, 131.2, 133.3, 133.7, 134.3, 139.5; LR-
MS (CI/NH3): m/z¼290 [(MþNH4)þ], 273 [(MþH)þ]; anal.
calcd. for C18H28Si (272.50): C 79.34, H 10.36; found: C 79.13,
H 10.22.
(Z)-(3,7-Dimethyl-2,6-octadienyl)dimethylphenylsilane
[(Z)-14] (Table 1, Entries 6–8): Colorless liquid; yield: 96–
98% [from (Z)-5a] and 86–97% [from (Z)-5b]; Rf (cyclo-
hexane)¼0.48; 1H NMR (CDCl3, 400 MHz): d¼0.28 (s, 6H),
1.62 (s, 3H), 1.67 (d, J¼8.6 Hz, 2H), 1.70 (s, 6H), 1.95 (m,
4H), 5.11–5.14 (m, 1H), 5.17 (t, J¼8.6 Hz, 1H), 7.35 (m, 3H),
7.51 (m, 2H); 13C NMR (CDCl3, 100 MHz): d¼ ꢀ3.1, 17.4,
17.7, 23.4, 25.8, 26.5, 31.8, 119.9, 124.7, 127.8, 128.9, 129.2,
131.4, 133.7, 134.3, 139.4.
[7] For the preparation of allylic silanes from 1,3-dienes us-
ing silyl cuprates, see: V. Liepins, J.-E. Bꢁckvall, Eur. J.
Org. Chem. 2002, 3527–3535.
[8] For the stereoselective preparation of allylic silanes, see:
J. H. Smitrovich, K. A. Woerpel, J. Org. Chem. 2000, 65,
1601–1614.
[9] Y. Morizawa, H. Oda, K. Oshima, H. Nozaki, Tetrahe-
dron Lett. 1984, 25, 1163–1166.
[10] M. Oestreich, B. Weiner, Synlett 2004, 2139–2142.
[11] a) Ac2O, Et3N, 4-N,N-dimethylaminopyridine (5.0
mol %), CH2Cl2, rt, 59–92%; b) phenyl isocyanate, Et3
N, 4-N,N-dimethylaminopyridine (50 mol %), CH2Cl2, rt,
45–90%; c) S. Hansson, A. Heumann, T. Rein, B. ꢂker-
mark, J. Org. Chem. 1990, 55, 975–984; d) D. Liotta, G.
Zima, M. Saindane, J. Org. Chem. 1982, 47, 1258–1267.
[12] B. Breit, P. Demel, in: Modern Organocopper Chemistry,
(Ed.: N. Krause), Wiley-VCH, Weinheim, 2002, pp. 188–
223.
640
ꢄ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 637–640