[1,5]-Brook rearrangement: An overlooked but valuable silyl migration to synthesize configurationally defined vinylsilane. the unique steric and electronic effects of geminal bis(silane)

Article abstract of DOI:10.1039/c3cc45002c

An unusual [1,5]-Brook rearrangement of the lithium alkoxide of geminal bis(silyl) homoallylic alcohol is described. The unique steric and electronic effects of geminal bis(silane) were found to be crucial for promoting this long-range silyl migration, as

Full text of DOI:10.1039/c3cc45002c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
[1,5]-Brook rearrangement: an overlooked but
valuable silyl migration to synthesize configurationally
defined vinylsilane. The unique steric and electronic
effects of geminal bis(silane)†
Cite this: Chem. Commun., 2013,
49, 8961
Received 4th July 2013,
Accepted 31st July 2013
Lu Gao,a Ji Lu,a Zhenlei Song,*ab Xinglong Lin,a Yongjin Xua and Zhiping Yina
DOI: 10.1039/c3cc45002c
www.rsc.org/chemcomm
An unusual [1,5]-Brook rearrangement of the lithium alkoxide of geminal the norm for such reactions, particularly when lithium alkoxide is
bis(silyl) homoallylic alcohol is described. The unique steric and electronic used as an initiator. This inefficiency is also the main reason why the
effects of geminal bis(silane) were found to be crucial for promoting this [1,5]-Brook rearrangement has been studied to such a limited extent,
long-range silyl migration, as well as for facilitating the subsequent with very few applications in organic synthesis so far.
c/Z-selective addition of silyl allyllithium with carbonyl compounds
to synthesize diverse configurationally defined Z-vinylsilanes.
We recently launched a series of investigations on structurally novel
geminal bis(silanes) and have reported their unusual behaviors and
attractive bifunctionalities.4 Here we report that the lithium alkoxide
The [1,n]-Brook rearrangement1 describes the intramolecular silyl of geminal bis(silyl) homoallylic alcohol 1 can undergo an unusual
group migration from a carbon to an oxygen atom via the penta- [1,5]-Brook rearrangement (Scheme 1, below). The unique steric and
coordinated silicate species (Scheme 1, above). The relative ease of electronic effects of geminal bis(silane) are crucial for promoting such
silyl migration has been reported to be [1,2] > [1,3] c [1,4] > [1,5] > a long-range silyl migration. These effects are also pivotal for the
[1,6] based on the logic that shorter transfer distances are more subsequent g/Z-selective addition with carbonyl compounds to synthe-
favorable.2 Indeed, the typical reaction types (n = 2–4) are usually facile size diverse configurationally defined Z-vinylsilanes 2 (ref. 5).
and have found wide utilities in various transformations. In contrast,
The reaction was initially examined using 1a6 as a model scaffold,
long-range rearrangements (n Z 5) such as the [1,5]-Brook rearrange- and benzaldehyde as the electrophile. Without the polar solvent,
ment generally proceed with much greater difficulty.3 Inefficiency is HMPA, no C-to-O silyl migration would occur after deprotonation
of 1a with n-BuLi in Et2O at À78 1C (Table 1, entry 1). Using Et2O–
HMPA (3.7 : 1) as co-solvent led to only partial silyl migration, while
Table 1 Screening of [1,5]-Brook rearrangement–addition conditions
TMEDA
(equiv.)
Yieldd
(%)
Entry
Base
Solvents
1
2
3
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi/t-BuOK
Et2O/HMPA = 1.0/0
Et2O/HMPA = 3.7/1
Et2O/HMPA = 1.3/1
Et2O/HMPA = 0.6/1
Et2O/HMPA = 1.3/1
THF/HMPA = 1.3/1
Et2O/HMPA = 1.3/1
—
—
—
—
2.5
2.5
2.5
0
23
43
30
75
74
60
Scheme 1 General description of [1,n]-Brook rearrangement (above); [1,5]-
Brook rearrangement–addition involving geminal bis(silyl) homoallylic alcohol
and carbonyl compounds (below).
4
5a
6
7
a Key Laboratory of Drug-Targeting and Drug Delivery System of the Education
Ministry, Department of Medicinal Chemistry, West China School of Pharmacy,
Sichuan University, Chengdu, 610041, China. E-mail: zhenleisong@scu.edu.cn
b State Key Laboratory of Biotherapy, West China Hospital, Sichuan University,
Chengdu, 610041, China
a
Reaction conditions: 0.2 mmol of 1a, 0.5 mmol of TMEDA, and
0.22 mmol of n-BuLi (2.5 M in pentane) in 1.3 mL of Et2O, À78 1C to
À10 1C; 0.6 mmol of PhCHO and 1.0 mL HMPA; crude products were
b
treated with 1.0 mL of 10% aq. HCl. The Z-configuration was assigned
c
based on NOE experiments on 2j (Table 2). Ratios were determined by
1H NMR spectroscopy. Isolated yields after purification by silica gel
d
† Electronic supplementary information (ESI) available: Experimental procedures
and characterisation data for new compounds. See DOI: 10.1039/c3cc45002c
column chromatography.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 8961--8963 8961

View Article Online
Communication
ChemComm
Table
3
Scope of [1,5]-Brook rearrangement–addition of geminal bis(silyl)
the subsequent addition at À10 1C proceeded in a highly g-regio-
selective manner and with complete Z-configurational control to
form vinylsilane. After acid-promoted selective desilylation, the
desired 1,5-diol 2a was obtained in 23% yield (entry 2). Increasing
the loading of HMPA to an Et2O–HMPA ratio of 1.3 : 1 gave 2a in
higher 43% yield (entry 3); using even more HMPA, however,
appeared to lower the yield (entry 4). In our efforts to improve the
addition step, we found that using 2.5 equiv. of TMEDA as an
additive dramatically increased the yield to 75% without changing
the high g/Z-selectivity (entry 5). TMEDA probably stabilizes the
silyl allyllithium, the high basicity of which makes it prone to
protonation. THF–HMPA also proved to be suitable as a co-solvent,
and gave 2a in comparably good yield of 74% (entry 6). While the
transmetalation of the lithium alkoxide to potassium ions still led
to the smooth silyl migration, it made the addition less effective
(entry 7). The reduced efficiency may be due to the fact that the
carbon–potassium bond has less covalent character than a C–Li bond.
The reaction showed good applicability to a wide range
of aldehydes and ketones (Table 2). The electrophiles include
aryl and heterocyclic (entries 1–3), a-di- or trisubstituted alkyl
(entries 4–7), and a,b-unsaturated aldehydes (entry 8), and
benzophenone (entry 9). The desired 1,5-diols 2b–2j were gen-
erally obtained in good yields and with high g/Z-selectivity.
homoallylic alcohol 1 with benzophenone
The reaction in entry 4 gave a comparably lower yield of 45%.
This may be due to deprotonation of the a-position of isobutyr-
aldehyde by the basic silylallyl lithium. Reaction with the less
reactive benzyl bromide proved to be a challenge. The addition
proceeded slowly and with poor regioselectivity, giving 2k in
only 31% yield (entry 10).
The process was also compatible with diverse geminal
bis(silyl) homoallylic alcohols, in which the R groups featured
unbranched or branched chains, a cyclohexyl ring, and vinyl or
phenyl substituents. Varying the R group sterically and electro-
nically did not affect the high g/Z-selectivity, giving 1,5-diols
2l–2q in good to excellent yields (Table 3).
Table 2 Scope of [1,5]-Brook rearrangement–addition of 1a with electrophiles
A control experiment to determine whether the reaction
indeed proceeds via intramolecular [1,5]-silyl migration was
performed (Scheme 2, above). An equimolar mixture of 1a and
1a-Piv reacted with benzophenone to give 2a and 2a-SiMe3 in
overall 91% yield. The original 1a-Piv was recovered in 98% yield,
and no cross-product 2r was observed. These results indicate
that [1,5]-C-to-O silyl migration proceeds intramolecularly via a
pentacoordinated silicate species.3b The reaction of mono-SiMe3-
substituted homoallylic alcohol 3 was repeated (Scheme 2,
below) and appeared to be more complex than that of 1a
(Table 2, entry 9). The 1,5-diol 4 was produced in 70% yield,
Entry
E
Product
Yieldc (%)
1
2
3
4
5
p-MeO-PhChO
p-Cl-PhCHO
2-Thienyl-CHO
i-PrCHO
2b
2c
2d
2e
2f
88
60
78
45
44
t-BuCHO
6
7
2g
58
2h
2i
83
60
8
9
PhCOPh
BnBr
2j
97
31
10
2k
a
Entries 1–8 gave dr = 50 : 50 to 63 : 37. Ratios were determined by
1H NMR spectroscopy. The ratio of g : a = 50 : 50 in entry 10 was
b
Scheme 2 Control experiment to confirm the intramolecular [1,5]-silyl migra-
tion (above); [1,5]-Brook rearrangement–addition of mono-SiMe3-substituted
homoallylic alcohol 3 with benzophenone (below).
determined by 1H NMR spectroscopy. Isolated yields after purification
c
by silica gel column chromatography.
c
8962 Chem. Commun., 2013, 49, 8961--8963
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
in an overall 41% yield. Subsequent carbonylation to Z-methyl
enoate generated 12 in 72% yield; this compound is related to
the ring B of bryostatins.10
In summary, we have described a facile [1,5]-Brook rearran-
gement of geminal bis(silyl) homoallylic alcohols. The unique
steric and electronic effects of geminal bis(silane) were found to
be crucial for promoting this long-range silyl migration, as well
as for facilitating the subsequent g/Z-addition of silyl allyl-
lithium with carbonyl compounds. The reaction provides a
valuable protocol for synthesizing diverse Z-vinylsilanes.
Further applications of this methodology are underway.
We are grateful for financial support from the NSFC
(21172150, 21021001, 21290180), the National Basic Research
Program of China (973 Program, 2010CB833200), the NCET
(12SCU-NCET-12-03), and the Sichuan University 985 project.
Scheme 3 Model to explain the observed high g/Z-selectivity of [1,5]-Brook
rearrangement–addition reaction.
a sharp contrast to the 97% yield of 2j. Homoallylic alcohol 5
was also obtained in 20% yield from acidic hydrolysis of the
unmigrated 3 and O-SiMe3-substituted 5, which was generated
before work-up when the highly basic allyl anion 6 rapidly
abstracted a proton instead of adding to benzophenone.7 Such
a large difference in efficiency suggests that the geminal
bis(silyl) group plays a crucial role in both silyl migration and
the subsequent addition.
Notes and references
1 For reviews, see: (a) A. G. Brook, Acc. Chem. Res., 1974, 7, 77;
(b) M. Kira and T. Iwamoto, Silyl Migrations, in The Chemistry of
Organic Silicon Compounds, ed. Z. Rappoport and Y. Apeloig, John
Wiley & Sons, Ltd., 2001, vol. 3, pp. 853–948; (c) W. H. Moser,
Tetrahedron, 2001, 57, 2065; (d) A. B. Smith III and C. M. Adams, Acc.
Chem. Res., 2004, 37, 365; (e) E. Schaumann and A. Kirschning,
Synlett, 2007, 177. For the selected latest advances, see:
( f ) Z. L. Song, L. Z. Kui, X. W. Sun and L. J. Li, Org. Lett., 2011,
13, 1440; (g) M. Hayashi and S. Nakamura, Angew. Chem., Int. Ed.,
2011, 50, 2249; (h) H. Li, L. T. Liu, Z. T. Wang, F. Zhao, S. G. Zhang,
W. X. Zhang and Z. F. Xi, Chem.–Eur. J., 2011, 17, 7399; (i) M. Sasaki,
Y. Kondo, M. Kawahata, K. Yamaguchi and K. Takeda, Angew. Chem.,
Int. Ed., 2011, 50, 6375; ( j) D. B. C. Martin and C. D. Vanderwal,
Chem. Sci., 2011, 2, 649; (k) Y. Matsuya, A. Koiwai, D. Minato,
K. Sugimoto and N. Toyooka, Tetrahedron Lett., 2012, 53, 5955;
(l) A. M. Rouf, B. O. Jahn and H. Ottosson, Organometallics, 2013,
32, 16; (m) B. Melillo and A. B. Smith III, Org. Lett., 2013, 15, 2282.
2 (a) E. W. Colvin, Silicon in Organic Synthesis, Butterworths, London,
1981; (b) J. J. Eisch and M.-R. Tsai, J. Organomet. Chem., 1982, 225, 5;
(c) M. Lautens, P. H. M. Delanghe, J. B. Goh and C. H. Zhang, J. Org.
Chem., 1992, 57, 3270; (d) B. M. Comanita, S. Woo and A. G. Fallis,
Tetrahedron Lett., 1999, 40, 5283.
In the mechanism model, two chair-like transition states 7a
and 7b featuring pentacoordinated silicate were hypothesized
(Scheme 3). Transition state 7b, despite a 1,3-diaxial interaction
between SiMe3 and Ha, likely still be favored over 7a, which suffers
an even more severe A1,3 strain8 between SiMe3 and Hb. Thus,
starting from 7b, relieving the bulkiness of geminal bis(silane)
would drive the cleavage of the C–Si bond to give anion 8-exo. This
allyl anion would be further stabilized by the unmigrated SiMe3
through the p–d p-bonding interaction (a-silicon effect).9 It would
then undergo addition with an electrophile at the more accessible
g-position to generate 2-g/Z predominantly. In other words, the
steric effect of geminal bis(silane) would kinetically facilitate the
[1,5]-Brook rearrangement, and its electronic effect would thermo-
dynamically favor the subsequent addition. This mechanism
could also explain why the reaction of 3, which lacks the dual
effects of geminal bis(silane), proved to be less efficient in both
silyl migration and addition.
3 For previous studies of [1,5]-Brook rearrangement, see:
(a) M. M. Kabat and J. Wicha, Tetrahedron Lett., 1991, 32, 1073;
(b) A. B. Smith III, M. Xian, W.-S. Kim and D.-S. Kim, J. Am. Chem.
Soc., 2006, 128, 12368; (c) W. P. Weber, R. A. Felix and A. K. Willard,
J. Am. Chem. Soc., 1970, 92, 1420; (d) C. P. Casey, C. R. Jones and
H. Tukada, J. Org. Chem., 1981, 46, 2089; (e) M. C. Pirrung, L. Fallon,
J. Zhu and Y. R. Lee, J. Am. Chem. Soc., 2001, 123, 3638; ( f ) M. Saito,
A. Saito, Y. Ishikawa and M. Yoshioka, Org. Lett., 2005, 7, 3139.
4 For a review, see: (a) L. Gao, Y. B. Zhang and Z. L. Song, Synlett, 2013,
139. For the latest advances, see: (b) L. J. Li, X. C. Ye, Y. Wu, L. Gao,
Z. L. Song, Z. P. Yin and Y. J. Xu, Org. Lett., 2013, 15, 1068;
(c) L. J. Yan, X. W. Sun, Z. L. Song, H. Z. Li and Z. J. Liu, Org. Lett.,
2013, 15, 1104; (d) K. Groll, S. M. Manolikakes, X. M. du Jourdin,
M. Jaric, A. Bredihhin, K. Karaghiosoff, T. Carell and P. Knochel,
Angew. Chem., Int. Ed., 2013, 52, 6776.
The resulting Z-vinylsilane was subjected to further reactions in
order to demonstrate the bifunctionality of geminal bis(silane). Treat-
ing 2j with NIS in CH3CN gave Z-vinyl iodide 9 in 93% yield, which
was transformed into the corresponding Z-enyne 10 in 87% yield via
Sonogashira coupling with terminal alkyne (Scheme 4, eqn (1)).
In contrast, treatment of mono-SiEt3-protected 2j with bromine at
À78 1C led to an interesting bromination–cyclization process,
giving exo-cyclic Z-vinyl bromide substituted tetrahydropyran 11
5 To simplify the discussion, the resulting vinylsilane, in which the
silyl group lies on the same side as the hydroxyl group in the
substrate, is assigned to have a Z-configuration.
6 J. Lu, Z. L. Song, Y. B. Zhang, Z. B. Gan and H. Z. Li, Angew. Chem.,
Int. Ed., 2012, 51, 5367.
7 For a similar observation, see: A. B. Smith III and M. O. Duffey,
Synlett, 2004, 1363.
8 (a) F. Johnson, Chem. Rev., 1968, 68, 375; (b) R. W. Hoffmann, Chem.
Rev., 1989, 89, 1841.
9 For a review, see: T. H. Chan and D. Wang, Chem. Rev., 1995, 95, 1279.
10 For reviews on bryostatins, see: (a) K. J. Hale, M. G. Hummersone,
S. Manaviazar and M. Frigerio, Nat. Prod. Rep., 2002, 19, 413;
(b) K. J. Hale and S. Manaviazar, Chem.–Asian J., 2010, 5, 704.
For the latest total synthesis of bryostatins, see: (c) Y. Lu,
S. K. Woo and M. J. Krische, J. Am. Chem. Soc., 2011, 133, 13876,
and references therein.
Scheme 4 Iodination of 2j, and Sonogashira coupling of the formed Z-vinyl
iodide 9 with terminal alkyne to form Z-enyne 10 (eqn (1)); bromination–
cyclization of 2j, and carbonylation of the resulting exo-cyclic Z-vinyl bromide
11 to form Z-methyl enoate 12 (eqn (2)).
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 8961--8963 8963