Unusual carbon-carbon bond formations between allylboronates and acetals or ketals catalyzed by a peculiar indium(I) lewis acid

Article abstract of DOI:10.1021/ol100450s

In^IOTf has been uncovered as an effective Lewis acid catalyst for unprecedented nucleophilic substitution of acetals or ketals with allylboronates. A transmetalative S_N1 mechanism is proposed in which a single In^I center acts as a dual catalyst to activate both reagents sequentially. Contrary to the classic γ-selectivity of allylsilanes (Hosomi-Sakurai reaction), this In^I-catalyzed borono variant displays distinct α-selectivity. Substrate scope and functional group tolerance proved to be excellent.

Full text of DOI:10.1021/ol100450s

ORGANIC
LETTERS
2010
Vol. 12, No. 11
2488-2491
Unusual Carbon-Carbon Bond
Formations between Allylboronates and
Acetals or Ketals Catalyzed by a
Peculiar Indium(I) Lewis Acid
Uwe Schneider, Hai T. Dao, and Shu¯ Kobayashi*
Department of Chemistry, School of Science and Graduate School of Pharmaceutical
Sciences, The UniVersity of Tokyo, The HFRE DiVision, ERATO, JST, Hongo,
Bunkyo-ku, Tokyo 113-0033, Japan
shu_kobayashi@chem.s.u-tokyo.ac.jp
Received March 10, 2010
ABSTRACT
InIOTf has been uncovered as an effective Lewis acid catalyst for unprecedented nucleophilic substitution of acetals or ketals with allylboronates.
A transmetalative SN1 mechanism is proposed in which a single InI center acts as a dual catalyst to activate both reagents sequentially.
Contrary to the classic γ-selectivity of allylsilanes (Hosomi-Sakurai reaction), this InI-catalyzed borono variant displays distinct r-selectivity.
Substrate scope and functional group tolerance proved to be excellent.
Innovative metal catalysis for selective bond formation plays
a vital role in chemistry.1 After pioneering reports,2,3
allylboronates have been established among the most im-
portant nontoxic reagents for C-C coupling.4 Originally,
these species have been used for uncatalyzed 1,2-additions
to aldehydes, which proceed via internal activation (CdO
f B) in a cyclic transition state (γ-selectivity).2-4 Following
seminal metal-catalyzed aldehyde allylborations,5 catalytic
activation of allylboronates for 1,2-additions to ketones6 and
imines7 has been explored.8 Regarding indium(I) (InI)
catalysis we developed,6c,7c,d it is noted that this main group
metal is appealing because In-based compounds have low
toxicity, are safe, inexpensive, selective, and tolerant toward
various functional groups.9 Contrary to commonly employed
InIII Lewis acids,9 InI catalysis is in its infancy.6c,7c,d,10
(6) (a) Wada, R.; Oisaki, K.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 2004, 126, 8910. (b) Lou, S.; Moquist, P. N.; Schaus, S. E. J. Am.
Chem. Soc. 2006, 128, 12660. (c) Schneider, U.; Kobayashi, S. Angew.
Chem., Int. Ed. 2007, 46, 5909. (d) Schneider, U.; Ueno, M.; Kobayashi,
S. J. Am. Chem. Soc. 2008, 130, 13824. (e) Barker, T. J.; Jarvo, E. R. Org.
Lett. 2009, 11, 1047. (f) Barnett, D. S.; Moquist, P. N.; Schaus, S. E. Angew.
Chem., Int. Ed. 2009, 48, 8679.
(1) ComprehensiVe Organometallic Chemistry III; Crabtree, R., Mingos,
M., Eds.; Elsevier Ltd.: Amsterdam, 2006.
(7) (a) Wada, R.; Shibuguchi, T.; Makino, S.; Oisaki, K.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2006, 128, 7687. (b) Lou, S.; Moquist,
P. N.; Schaus, S. E. J. Am. Chem. Soc. 2007, 129, 15398. (c) Schneider,
U.; Chen, I.-H.; Kobayashi, S. Org. Lett. 2008, 10, 737. (d) Kobayashi, S.;
Konishi, H.; Schneider, U. Chem. Commun. 2008, 2313. (e) Fujita, M.;
Nagano, T.; Schneider, U.; Hamada, T.; Ogawa, C.; Kobayashi, S. J. Am.
Chem. Soc. 2008, 130, 2914.
(2) (a) Mikhailov, B. M.; Bubnov, Y. N. IzV. Akad. Nauk. SSSR, Ser.
Khim. 1964, 1874; Chem. Abstr. 1965, 62, 11840e. (b) Favre, E.; Gaudemar,
M. C. C. R. Seances Acad. Sci., Ser. C 1966, 263, 1543
(3) (a) Herold, T.; Hoffmann, R. W. Angew. Chem., Int. Ed. 1978, 17,
768. (b) Hoffmann, R. W.; Zeiss, H.-J. Angew. Chem., Int. Ed. 1979, 18,
.
306
.
(8) Transition metal-catalyzed use of allylboronates for C-C couplings:
(a) Sieber, J. D.; Morken, J. P. J. Am. Chem. Soc. 2008, 130, 4978. (b)
Shaghafi, M. B.; Kohn, B. L.; Jarvo, E. R. Org. Lett. 2008, 10, 4743. (c)
Zhang, P.; Morken, J. P. J. Am. Chem. Soc. 2009, 131, 12550. (d) Ferrer
Flegeau, E.; Schneider, U.; Kobayashi, S. Chem.sEur. J. 2009, 15, 12247.
(4) Hall, D. G. Synlett 2007, 1644.
(5) (a) Kennedy, J. W. J.; Hall, D. G. J. Am. Chem. Soc. 2002, 124,
11586. (b) Ishiyama, T.; Ahiko, T.-a.; Miyaura, N. J. Am. Chem. Soc. 2002,
124, 12414.
10.1021/ol100450s  2010 American Chemical Society
Published on Web 05/12/2010

Acetals play a key role in nature11 and chemistry;12 acetal
allylation is a useful carbon-carbon bond formation that can
provide valuable homoallyl ethers.13 Typically, this chal-
lenging C-C coupling proceeds via Lewis acid activation
to form an oxocarbenium ion that can react with nucleophilic
allylsilanes.14 Lewis or Brønsted acid-catalyzed variants15
have been developed,16 but only some catalytic methods were
found to be truly effective, practical, and general. Allylbo-
ronates are, in the absence of transition metals,8 unreactive
toward sp3-type electrophiles (noncarbonyls); these boron
reagents have been neglected, although they may offer
significant advantages such as easier access, superior stability,
and unique reactivity, and selectivity. In the quest for new
electrophiles compatible with our InI catalysis, we envisioned
acetals for challenging nucleophilic substitution with allyl-
boronates, an unprecedented process that would require
Lewis acid and Lewis base activation (Scheme 1). We sought
use of indium(I) iodide6c,7c,d or other halides proved to be
ineffective (Table 1, entries 1-3). To our surprise, when
Table 1. Screening of Lewis Acid Catalysts
entry Lewis acid catalyst (mol %) solvent conversiona (%)
1
2
3
4
5
6
toluene
toluene
toluene
toluene
toluene
nr
trace
nr
InII (20)
InIBr (20) or InICl (20)
InIOTf (20)
>99 (95)b
nd (12)b
trace
InIII(OTf)3 (20)
GaIII(OTf)3 or AlIII(OTf)3 or toluene
CuIOTf (20)
7
8
9
10
11
12
13
14
ScIII(OTf)3 (20)
CuII(OTf)2 (20)
AgIOTf or ZnII(OTf)2 (20)
InIOTf (5)
toluene
toluene
toluene
toluene
n-hexane
DCM
2
5
nr
>99
>99
>95
20
Scheme 1. Nucleophilic Substitution with an Allylboronate?
InIOTf (5)
InIOTf (5)
InIOTf (5)
THF
toluene
InIOTf (1)
>99 (91)b
a Conversions of 1a to 3a determined by 1H NMR spectroscopic analysis
of aliquots of the reaction mixtures. b Isolated yields of homoallyl ether 3a
after purification on silica gel (PTLC). nr ) no reaction; nd ) not determined
(due to the formation of byproduct).
a single catalyst capable of activating both reagents and report
here our preliminary results.
In initial experiments, the uncatalyzed reaction between
acetal 1a and allylboronate 2 did not occur, and catalytic
InIOTf17 (20 mol %) was employed, this C-C bond
formation proceeded smoothly to provide homoallyl ether
3a in 95% yield (entry 4). Strikingly, various other metal
triflates including InIII(OTf)3 were found to be inefficient
(entries 5-9). This indicated that, contrary to classic allyl-
silanes, a strong Lewis acid, for the activation of 1a, is not
sufficient to promote this C-C coupling with 2. Rather, the
ability to activate both 1a and 2 seems to be crucial. Note
that (1) toluene was shown to be the best solvent of those
examined (entries 10-13) and (2) the InI catalyst loading
could be reduced to 1 mol % (entry 14; 91% yield).
Next, we investigated the substrate scope (Figure 1). This
reaction proceeded smoothly with acyclic or cyclic aromatic,
heteroaromatic, and aliphatic acetals or ketals. It is noted
that readily cleavable ethers such as product 3b (R ) Bn)
are accessible and that the mild conditions tolerate reactive
functional groups such as unprotected O-H and aliphatic
C-Hal bonds (products 3c, 3s, and 3t), which reveals the
high appeal of this InI catalysis for synthesis.
(9) Recent review on the use of indium for synthesis: Auge´, J.; Lubin-
Germain, N.; Uziel, J. Synthesis 2007, 1739.
(10) Selected stoichiometric examples for indiumI-mediated C-C bond
formation: (a) Araki, S.; Ito, H.; Katsumura, N.; Butsugan, Y. J. Organomet.
Chem. 1989, 369, 291. (b) Chan, T. H.; Yang, Y. J. Am. Chem. Soc. 1999,
121, 3228. (c) Yang, Y.; Chan, T. H. J. Am. Chem. Soc. 2000, 122, 402.
(d) Araki, S.; Kamei, T.; Hirashita, T.; Yamamura, H.; Kawai, M. Org.
Lett. 2000, 2, 847. (e) Babu, S. A.; Yasuda, M.; Shibata, I.; Baba, A. Org.
Lett. 2004, 6, 4475.
(11) Selected examples: (a) Vinogradov, E.; Bock, K. Angew. Chem.,
Int. Ed. 1999, 38, 671. (b) Milroy, L.-G.; Zinzalla, G.; Loiseau, F.; Qian,
Z.; Prencipe, G.; Pepper, C.; Fegan, C.; Ley, S. V. ChemMedChem 2008,
3, 1922.
(12) Selected examples: (a) Nicolaou, K. C.; Seitz, S. P.; Papahatjis,
D. P. J. Am. Chem. Soc. 1983, 105, 2430. (b) Dixon, D. J.; Foster, A. C.;
Ley, S. V.; Reynolds, D. J. J. Chem. Soc., Perkin Trans. 1 1999, 1635. (c)
See also ref 11b.
(13) Selected examples: (a) Margot, C.; Rizzolio, M.; Schlosser, M.
Tetrahedron 1990, 46, 2411. (b) Carey, J. S.; Thomas, E. J. Synlett 1992,
585. (c) Higashino, T.; Sakaguchi, S.; Ishii, Y. Org. Lett. 2000, 2, 4193.
(d) Le Noˆtre, J.; Brissieux, L.; Se´meril, D.; Bruneau, C.; Dixneuf, P. H.
Chem. Commun. 2002, 1772. (e) Cui, Y.-M.; Huang, Q.-Q.; Xu, J.; Chen,
L.-L.; Li, J.-Y.; Ye, Q.-Z.; Li, J.; Nan, F.-J. Bioorg. Med. Chem. Lett. 2005,
15, 4130.
We then studied the mechanism by employing substituted
allyl reagents (Scheme 2). As expected, use of R-methyla-
llylsilane 4 provided the classic γ-product 7a. In sharp
contrast, use of R-methylallylboronate 5 resulted in the
exclusive formation of the rarely obtained R-product 8a in
83% yield. It should be noted that both crotylboronates (E)-6
and (Z)-6 gave an identical result with respect to regio- and
(14) Hosomi, A.; Endo, M.; Sakurai, H. Chem. Lett. 1976, 941.
(15) Selected examples for catalytic Hosomi-Sakurai reactions: Lewis
acids: (a) Sakurai, H.; Sasaki, K.; Hosomi, A. Tetrahedron Lett. 1981, 22,
745. (b) Yadav, J. S.; Reddy, B. V. S.; Srihari, P. Synlett 2001, 673. (c)
Wieland, L. C.; Zerth, H. B.; Mohan, R. S. Tetrahedron Lett. 2002, 43,
4597. (d) Braun, M.; Kotter, W. Angew. Chem., Int. Ed. 2004, 43, 514. (e)
Ooi, T.; Takahashi, M.; Yamada, M.; Tayama, E.; Omoto, K.; Maruoka,
K. J. Am. Chem. Soc. 2004, 126, 1150. (f) Spafford, M. J.; Anderson, E. D.;
Lacey, J. R.; Palma, A. C.; Mohan, R. S. Tetrahedron. Lett. 2007, 48, 8665.
(g) Brønsted acid: Kampen, D.; List, B. Synlett 2006, 2589.
(16) Allylation of acetals with nucleophilic allyl trialkylborates (sto-
ichiometric SiIV): Hunter, R.; Michael, J. P.; Tomlinson, G. D. Tetrahedron
1994, 50, 871.
(17) Preparation of InIOTf as a soluble indium(I) source: Macdonald,
C. L. B.; Corrente, A. M.; Andrews, C. G.; Taylor, A.; Ellis, B. D. Chem.
Commun. 2004, 250.
Org. Lett., Vol. 12, No. 11, 2010
2489

Scheme 3. Deuterium Labeling Experiment
We then examined this B-In system by NMR spectro-
scopic analyses, which revealed that boronate 2 is stable in
toluene-d8 at rt in the presence of InIOTf; no B-to-In
transmetalation was detected even after 12 h. However, upon
addition of acetal 1a, the C-C coupling proceeded smoothly
to form product 3a in >99% NMR yield. In a preliminary
kinetic study, a first-order dependence in catalyst (InIOTf)
was determined.21 At present, we propose a transmetalative
SN1 mechanism in which InI acts as a unique dual catalyst
(Scheme 4). InIOTf as a Lewis acid may activate 1 to form,
Figure 1.
Scope for InIOTf-catalyzed C-C bond formation. (a)
Reaction conditions: InIOTf (1-10 mol %), toluene or hexane,
25-30 °C, 14-40 h; isolated yields of homoallyl ethers 3b-u after
purification on silica gel (PTLC). (b) R ) Bn. (c) R ) Me. (d)
Substrate 1i is the corresponding methoxy acetal.
diastereoselectivity compared with 5, suggesting the same
reactive intermediate for all three boron reagents.18 The
R-selectivity with 5 may indicate catalytic B-to-In trans-
metalation, while the lower reactivity of (E)-6 and (Z)-6 may
be explained with slower transmetalation due to steric
demand at the γ-position.
Scheme 4. Proposed Transmetalative SN1 Mechanism
To test the B-to-In transmetalation hypothesis, the deu-
terated allylboronate deuterio-28a,b was used (Scheme 3). If
InIOTf acts solely as a conventional Lewis acid, the reaction
would proceed regiospecifically between 1a and deuterio-2
to form homoallyl ether deuterio-3a (acyclic transition
state).19 However, we observed the clean formation of an
equimolar mixture of both regioisomers deuterio-3a and
deuterio-3′a. This result supports the suggestion that trans-
metalation generates both nucleophilic allyl InI isomers A
and B (fast equilibrium),20 thereby scrambling the deuterium
label. Both isomers may react with the electrophile to provide
the observed product mixture.
via C, oxonium ion D and InIOR (R ) Me, Bn). This
alkoxide shuttle delivers the required Lewis base to activate
the boronates 2, 5, or 6 to generate, via E, the nucleophilic
InI species F or G (fast equilibrium).20 The more stable
reagent F (if R6 ) Me) may react with D to provide, via H,
homoallyl ethers 3 or 8. The proposed SN1 mechanism,
including transmetalation and C-C coupling via an acyclic
transition state, is consistent with both the R-selectivity of 5
Scheme 2. Mechanistic Experiments
(18) Isomerization of allylboronates 5, (E)-6, and (Z)-6 or of products
7a and 8a via InIOTf catalysis has not been observed.
(19) In the absence of acetal 1a, allylboronate deuterio-2 proved to be
stable under the reaction consitions; isomerization and/or decomposition
did not occur.
a Isolated yields of homoallyl ethers 7a or 8a after purification on silica
gel (PTLC). b Conditions: (E)-6 (1.5 equiv), 40 °C, 50 h. c Conditions: (Z)-6
(1.5 equiv), rt, 50 h. M ) B(pin) or SiMe3; R1, R2, R3 ) H or Me.
(20) Isaac, M. B.; Chan, T. H. Tetrahedron Lett. 1995, 36, 8957.
(21) See the Supporting Information for details.
2490
Org. Lett., Vol. 12, No. 11, 2010

and diastereoselectivity; the almost identical result with (E)-6
and (Z)-6 may be explained with the more stable InI species
F (R6 ) Me) being the real nucleophile in all three cases.18,20
The proposed mechanism is fully consistent with the fact
that the weaker Lewis acid InIOTf is substantially more active
than InIII(OTf)3. Indeed, although InI is significantly larger
than InIII,22 its acidity is sufficient to activate 1. On the other
hand, InI may be more effective than InIII for several reasons:
(1) The formed InI-O bond within InI-OR may be longer
than in case of InIII; thus, the O-Lewis basicity is stronger,
which accounts for an easier activation of the acidic boron
atom (hard-hard). (2) The larger InI is more suitable for
transmetalation at the CdC double bond than InIII (soft-soft).
(3) The generated InI-C bond within the allyl InI reagent is
longer than in case of InIII; this nucleophile is more reactive.
result may also be ascribed to catalytic B-to-In transmeta-
lation to form propargyl and allenyl InI intermediates I and
J (fast equilibrium). The more stable allenyl species J24 may
act as the real nucleophile, which may account for the
regioselectivity in favor of 10a.
This report introducing a new catalyst (InIOTf) represents
the first main group metal-catalyzed activation of allylbor-
onates8 for unprecedented carbon-carbon bond formation
with noncarbonyls. The present catalytic indium(I) study is
(1) distinct from earlier work with allylboronates and
carbonyls2-7 and (2) compares well with a related stoichio-
metric boron(III) protocol.25 We propose a transmetalative
SN1 mechanism in which a single InI center acts as a dual
catalyst for sequential activation of both reagents. Contrary
to the classic Hosomi-Sakurai reaction,14,15 the present InI-
catalyzed borono variant displays distinct R-selectivity. This
reaction proved to be tolerant toward various functional
groups and is applicable to propargylation. Being aware of
the possibilities opened up by this unusual InI dual catalysis,
we are currently investigating an asymmetric version, C-
glycosylation, and other new transformations with nontoxic
boron reagents.
Scheme 5. Regiospecific Propargylation of 1a with 9
Acknowledgment. This work was partially supported by
a Grant-in-Aid for Scientific Research from the Japan Society
for the Promotion of Science (JSPS) and Global COE
Program (Chemistry Innovation through Cooperation of
Science and Engineering), The University of Tokyo, MEXT,
Japan.
Finally, allenylboronate 9 was converted regiospecifically
into homopropargyl ether 10a23 (Scheme 5). This excellent
Supporting Information Available: Full experimental
details and NMR spectral reproductions for all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) Pauling ionic radii have been reported to be 104 pm (InI) and 81
pm (InIII): http://www.webelements.com.
(23) This result constitutes a substantial advance compared with the best
result obtained for an allenylsilane in a classic Hosomi-Sakurai reaction
(3% yield for 10a): Niimi, L.; Hiraoka, S.; Yokozawa, T. Tetrahedron 2002,
58, 245.
OL100450S
(24) Allenylindium species are more stable than propargylindium species:
Miao, W.; Chung, L. W.; Wu, Y.-D.; Chan, T. H. J. Am. Chem. Soc. 2004,
126, 13326.
(25) During the preparation of our manuscript, a related stoichiometric
boron(III) method was reported: Mitchell, T. A.; Bode, J. W. J. Am. Chem.
Soc. 2009, 131, 18057.
Org. Lett., Vol. 12, No. 11, 2010
2491