Regioselective 1,4- over 1,2-addition of 3,3-bis(silyl) allyloxy lithium to enals, enones and enoates. the remarkable α-effect of silicon

Article abstract of DOI:10.1039/c4ob00139g

A remarkable α-effect of silicon has been discovered that results in soft nucleophilicity at the Cγ of 3,3-bis(silyl) allyloxy lithium 1. The addition of 1 to α,β-unsaturated carbonyl compounds, including enals, proceeds in a 1,4- over 1,2-manner with medium to good regioselectivity, whereas the parent allyloxy lithium 4 undergoes complete 1,2-addition. The results from DFT calculations of HMPA-complexed 1 and 4 provide the rationale to explain this different regioselectivity.

Full text of DOI:10.1039/c4ob00139g

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Regioselective 1,4- over 1,2-addition of
3,3-bis(silyl) allyloxy lithium to enals, enones and
Cite this: Org. Biomol. Chem., 2014,
12, 3021
enoates. The remarkable α-effect of silicon†
Xincui Ye,a Xianwei Sun,a Zhenggang Huang,a Na Yang,b Zhishan Su,b Changwei Hub
and Zhenlei Song*a,c
Received 19th January 2014,
Accepted 6th March 2014
DOI: 10.1039/c4ob00139g
www.rsc.org/obc
A remarkable α-effect of silicon has been discovered that results in tivity from their parent carbanions. These differences are due
soft nucleophilicity at the Cγ of 3,3-bis(silyl) allyloxy lithium 1. The in part to the steric effect of the bulky silyl group, but primar-
addition of 1 to α,β-unsaturated carbonyl compounds, including ily they arise from the electronic effects of silicon. Silicon is
enals, proceeds in a 1,4- over 1,2-manner with medium to good thought to stabilize the α-carbanion through a p–d π-bonding
regioselectivity, whereas the parent allyloxy lithium 4 undergoes interaction or hyperconjugation, known as the α-effect of
complete 1,2-addition. The results from DFT calculations of HMPA- silicon.5 If this effect were doubled by incorporating two
complexed 1 and 4 provide the rationale to explain this different silyl groups in the same molecule, such as in the case of
regioselectivity.
geminal bis(silanes), would it alter the reactivity of carbanions
enough to shift an organolithium addition away from a
1,2-mechanism toward a 1,4-mechanism? Here we report that
the α-effect of silicon leads to soft nucleophilicity at the steri-
cally more accessible Cγ of 3,3-bis(triethylsilyl) allyloxy lithium
1. The addition of 1 to α,β-unsaturated carbonyl compounds,
including to highly reactive enals, proceeds in a predominant
1,4-manner to give 2 with medium to good regioselectivity
(Scheme 1).
3,3-Bis(triethylsilyl) allyloxy lithium 1 was generated from
the corresponding Z-benzyl enol ether 3 through sequential
regio-selective deprotonation and [1,5]-anion relay.6–8 The sub-
sequent addition to cinnamaldehyde proceeded at −78 °C pre-
dominantly in a 1,4-manner ([1,4] : [1,2] = 70 : 30), giving
aldehyde 2a in 64% yield with ≥95 : 5 diastereoselectivity
(Table 1, entry 1). Even though HMPA is believed to favor
solvent-separated ion pair (SSIP) formation and thereby
promote attack at the 4-position,2e increasing its loading
from 3.0 to 12.0 equiv. lowered the yield without altering
the product distribution (entry 2). Similar results were
Addition of organometals to enals and enones is one of the
most fundamental transformations in organic synthesis.1
Organolithium of typical reactivity normally undergoes
complete or predominant 1,2-addition over 1,4-addition.
To increase the synthetic usefulness of organolithium addi-
tion, extensive efforts2 have been made to reverse the regio-
selectivity in favor of 1,4-addition. Although these approaches
have allowed reasonably efficient addition of lithiodithianes
to enones, achieving selective 1,4-addition in favor of 1,2-
addition with more reactive enals remains a significant
challenge.
Recently, we launched a series of investigations into struc-
turally novel geminal bis(silanes).3 We wondered whether the
presence of two silicons might provide a path to shifting the
regioselectivity of organolithium addition to enals and enones.
Silicon-substituted carbanions4 usually possess different reac-
aKey laboratory of Drug-Targeting and Drug Delivery System of the Education
Ministry, West China School of Pharmacy, Sichuan University, Chengdu, 610041,
China. E-mail: zhenleisong@scu.edu.cn
bKey Laboratory of Green Chemistry & Technology, Ministry of Education,
College of Chemistry Sichuan University, Chengdu, 610041, China
cState Key Laboratory of Biotherapy, West China Hospital, Sichuan University,
Chengdu, 610041, China
†Electronic supplementary information (ESI) available: Experimental proce-
dures, characterisation data for new compounds. CCDC 974929 and 974934.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c4ob00139g
Scheme 1 Regioselective 1,4-addition over 1,2-addition of 3,3-bis-
(triethylsilyl) allyloxy lithium 1 to α,β-unsaturated carbonyl compounds.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3021–3025 | 3021

View Article Online
Communication
Organic & Biomolecular Chemistry
NBO analysis indicates that the negative charge is distributed
more towards Cα in 1-COM (Cα: −1.497 and Cγ: −0.073) than
in 4-COM (Cα: −0.853 and Cγ: −0.152). This most likely reflects
the α-effect of silicon, which means that Cγ accumulates less
electron density in 1-COM than in 4-COM, making it a softer
nucleophilic center.10 Based on the Pearson concept of hard
and soft acids and bases (HSAB)11 and the Klopman–Salem
concept of charge and orbital control of organic reactions,12
we predict that the HOMO of 1-COM is at higher energy than
that of 4-COM, favoring attack at the C4 of crotonaldehyde,
which is softer than the carbonyl C2 and has a larger LUMO
coefficient.13 This soft–soft interaction controlled by frontier
orbitals contrasts with the addition of 4-COM to C2 of croto-
naldehyde, which is probably favored by a charge-controlled
hard–hard interaction.
To provide a mechanistic basis for probing the stereo-
chemistry of this reaction, two “open” transition states 6a and
6b were proposed for the addition of 1 to cinnamaldehyde
(Scheme 2). We predict that 6a is favored over 6b, which
suffers a severe gauche interaction between geminal bis
(triethylsilyl) and phenyl groups, and that this preference for
6a leads to the observed 1,2-anti diastereoselectivity. On the
other hand, to interpret the 1,3-syn diastereoselectivity in the
addition with 2-methyl propenal, we initiated the reaction and
Table 1 Screening of reaction conditions
HMPA
(equiv.)
Yieldb
(2a)
Entry
T (°C)
drc
[1,4] : [1,2]d
1a
2
3
3.0
12.0
3.0
3.0
3.0
−78
−78
−98
−55
−78
64%
56%
52%
87%e
57%
≥95 : 5
≥95 : 5
≥95 : 5
—
70 : 30
70 : 30
68 : 32
—
4
5f
≥95 : 5
66 : 34
a Reaction conditions: 0.13 mmol of 3, 0.39 mmol of HMPA and
0.39 mmol of t-BuLi (1.3 M in pentane) in 1.0 mL of THF, −78 °C,
1.5 h; then 0.26 mmol of cinnamaldehyde, 10 min. b Isolated yields
after purification by silica gel column chromatography. c The anti/syn
ratios were determined using 1H NMR spectroscopy. The anti-
stereochemistry was assigned based on X-ray analysis of the di(3,5-
dinitro benzoate) of 2a. d [1,4] : [1,2] ratios were determined using
1H NMR spectroscopy. e Isolated yield of product generated by [1,2]-
Wittig rearrangement of 1. f 3.0 equiv. of CuCN was added after
generation of 1.
obtained at −98 °C, even though low temperature has been quenched it with Et3SiCl. E-Silyl enol ether 7 was obtained in
also proposed to favor 1,4-addition (entry 3).2h Nevertheless, 45% yield, suggesting that the lithium enolate that forms after
temperature did affect the stability of allyllithium 1: conduct- addition is in an E-configuration and probably adopts a
ing the reaction at −55 °C led to severe [1,2]-Wittig rearrange- gauche conformation as in 8, such that the bulky geminal bis-
ment before addition to the enal (entry 4). We were (triethylsilyl) group is antiperiplanar to the enolate in order to
unsuccessful in our attempt to convert 1 into the corres- minimize the nonbonded interaction. Protonation of 8 from
ponding lithium organocuprate: the reaction in the presence the sterically more accessible β-face would then give 2e with
of 3.0 CuCN gave a [1,4] : [1,2] ratio of 66 : 34, comparable to high 1,3-syn diastereoselectivity.
the ratio in entry 1 (entry 5).
In order to extend the synthetic usefulness of our addition
Next the scope of α,β-unsaturated compounds was tested approach, anionic silyl migration was utilized to functionalize
with 1. Reaction of 3-methyl-2-butenal, which shows increased the geminal bis(silyl) group in 2a (Scheme 3). Reduction and
steric hindrance at the 4-position (Table 2, entry 4), gave a deprotection of 2a gave rise to 1,4-diol 9 with an overall yield
lower yield and lower [1,4] : [1,2] ratio than did 4-mono-substi- of 65%. In the presence of CuCN and t-BuOLi, a [1,4]-Csp2 to
tuted aldehydes (entries 1–3). In contrast to the high 1,2-anti O silyl migration of 9 occurred regioselectively on the secondary
diastereoselectivity in entries 1 and 2, a more challenging 1,3- hydroxyl group to generate vinyl anion.14 Subsequent alkyl-
syn stereochemical control using 2-methyl propenal was ation with allyl and propargyl bromide provided trisubstituted
achieved to give aldehyde 2e in 56% yield with 95 : 5 diastereo- E-vinylsilanes 10a and 10b, respectively, in yields of 89% and
selectivity (entry 5). Switching from enals to less reactive 77%. In this way, the second electrophile was added to the
enones (entries 6–10) and enoates (entries 11 and 12) reduced α-position of benzyl enol ether 3.
diastereoselectivity, however, increased [1,4] : [1,2] selectivity in
In summary, we have described the 1,4- over 1,2-addition
most cases.
of 3,3-bis(triethylsilyl) allyloxy lithium 1 to α,β-unsaturated
Whereas the reaction of 1 with crotonaldehyde gave a carbonyl compounds, including highly reactive enals. Experi-
[1,4] : [1,2] ratio of 70 : 30, the reaction of the parent allyloxy mental and computational results suggest that the unusual
lithium 4 under the same reaction conditions led to the com- regioselectivity is because the α-effect of silicon makes 1 soft
plete 1,2-adduct 5 in 84% yield (Fig. 1).9 Apparently, the nature nucleophilic at the Cγ. Further studies into the mechanism of
of the organolithium plays a key role in determining the mode this unique α-effect and its synthetic applications are
of addition, in conjunction with several other factors that also underway.
influence regioselectivity, such as the metal counterion, temp-
We are grateful for financial support from the NSFC
erature, and solvents. In an attempt to get deeper insights to (21172150, 21321061, 21290180), the National Basic Research
how much the α-effect of silicon influences the negative Program of China (973 Program, 2010CB833200), the
charge distribution, we performed DFT calculations of HMPA- NCET (12SCU-NCET-12-03), and the Sichuan University 985
complexed allyloxy lithium 1 and 4 at the B3LYP/6-31G* level. project.
3022 | Org. Biomol. Chem., 2014, 12, 3021–3025
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Communication
Table 2 Scope of α,β-unsaturated carbonyl compounds
Entry
Electrophile
Product
Yielda
drb
[1,4] : [1,2]c
1
2
3
2a (R2 = Ph)
64%
42%
64%
≥95 : 5
88 : 12
90 : 10
70 : 30
72 : 28
70 : 30
2b (R2 = 4-ClC6H4)
2c (R2 = Me)
4
5
6
7
8
2d
2ed
2f
32%
56%
71%
75%
69%
—
45 : 55
72 : 28
≥95 : 5
90 : 10
≥95 : 5
95 : 5
—
2g
2h
53 : 47
≥95 : 5
9
10
2i (n = 0)
2je (n = 1)
50%
60%
65 : 35
82 : 18
≥95 : 5
83 : 17
11
12
2k
2l
70%
52%
50 : 50
88 : 12
≥95 : 5
≥95 : 5
a Isolated yields after purification by silica gel column chromatography. b The dr was determined using 1H NMR spectroscopy. c [1,4] : [1,2] ratios
were determined using 1H NMR spectroscopy. d 1,3-Syn stereochemistry was assigned based on NOE experiments with the γ-lactone of 2e. e 1,2-
Anti stereochemistry was assigned based on X-ray analysis of the di(3,5-dinitro benzoate) of 2j.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3021–3025 | 3023

View Article Online
Communication
Organic & Biomolecular Chemistry
3 For selected recent advances, see: (a) L. J. Yan, X. W. Sun,
Z. L. Song, H. Z. Li and Z. J. Liu, Org. Lett., 2013, 15, 1104;
(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. Gao, J. Lu,
Z. L. Song, X. L. Lin, Y. J. Xu and Z. P. Yin, Chem. Commun.,
2013, 49, 8963; (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;
(e) L. Gao, Y. B. Zhang and Z. L. Song, Synlett, 2013,
139.
4 For reviews, see: (a) A. Krief, Tetrahedron, 1980, 36, 2531;
(b) E. Colvin, Silicon, in Organic Synthesis, Butterworth,
London, 1981; (c) W. P. Weber, Silicon Reagents for Organic
Synthesis, Springer-Verlag, New York, 1983; (d) T. H. Chan
and D. Wang, Chem. Rev., 1995, 95, 1279; (e) I. Fleming,
A. Barbero and D. Walter, Chem. Rev., 1997, 97, 2063;
(f) L. F. van Staden, D. Gravestock and D. J. Ager,
Chem. Soc. Rev., 2002, 31, 195; (g) D. M. Hodgson and
E. Gras, Synthesis, 2002, 1625; (h) E. Schaumann and
A. Kirschning, Synlett, 2007, 177; (i) Z. F. Xi, Acc. Chem.
Res., 2010, 43, 1342; ( j) H. J. Reich, J. Org. Chem., 2012, 77,
5471.
Fig. 1 NBO analysis of allyloxy lithium-HMPA complexes 1-COM and
4-COM based on DFT calculations performed at the B3LYP/6-31G* level.
5 For
a review, see: J. S. Panek, Silicon Stabilization,
in Comprehensive Organic Synthesis, ed. B. M. Trost
and I. Fleming, Pergamon, New York, 1991, vol. 1,
p. 579.
Scheme 2 Model to explain 1,2-anti and 1,3-syn diastereoselectivity.
6 For reviews of anion relay chemistry (ARC) based on
anionic silyl migration, see: (a) A. G. Brook, Acc. Chem. Res.,
1974, 7, 77; (b) W. H. Moser, Tetrahedron, 2001, 57, 2065;
(c) A. B. Smith III and C. M. Adams, Acc. Chem. Res., 2004,
37, 365; (d) A. B. Smith III and W. M. Wuest, Chem.
Commun., 2008, 5883.
7 X. W. Sun, J. Lei, C. Z. Sun, Z. L. Song and L. J. Yan, Org.
Lett., 2012, 14, 1094. Reaction of 11 with allyl bromide
under the optimal conditions provided 12 in 99% yield
with ca. 75% of deuterium labeled at the benzylic position.
This result unambiguously confirmed that formation of 1
proceeds predominantly by [1,5]-anion relay based on an
intramolecular proton transfer.
Scheme 3 Anionic [1,4]-silyl migration of 9 to synthesize E-vinylsilanes
10a and 10b.
Notes and references
1 For reviews, see: (a) E. D. Bergman, D. Ginsburg and
R. Pappo, Org. React., 1959, 10, 179; (b) P. Perlmutter,
Conjugate Addition Reactions in Organic Synthesis, Pergamon,
Oxford, 1992; (c) P. Knochel, N. Millot, A. Rodriguez and
C. E. Tucker, Org. React., 2001, 58, 417.
2 (a) T. M. Dolak and T. A. Bryson, Tetrahedron Lett., 1977, 18,
1961; (b) D. Seebach and R. Locher, Angew. Chem., Int. Ed.
Engl., 1979, 18, 957; (c) W. Dumont, J. Lucchetti and
A. Krief, J. Chem. Soc., Chem. Commun., 1983, 66;
(d) D. J. Ager and M. B. East, J. Org. Chem., 1986, 51, 3983;
(e) T. Cohen, W. D. Abraham and M. Myers, J. Am. Chem.
Soc., 1987, 109, 7923; (f) K. Maruoka, M. Ito and
H. Yamamoto, J. Am. Chem. Soc., 1995, 117, 9091;
(g) T. Ooi, Y. Kondo and K. Maruoka, Angew. Chem., Int. Ed.
Engl., 1997, 36, 1183; (h) W. H. Sikorski and H. J. Reich,
J. Am. Chem. Soc., 2001, 123, 6527.
8 Deprotonation of 3 and the subsequent addition to D2O
and MeI, respectively, occurred exclusively at the α-position
to afford 11 and 13 in 82% and 67% yield with the exclu-
sive Z-configuration. These results combined with those in
Table 2 show a sterically-dependent shift in regioselectivity,
suggesting that the α-position of 1 has greater electron
density than the γ-position, but that the γ-position is steri-
cally more accessible than the α-position, which bears a
bulky geminal bis(triethylsilyl) group. Formation of Z-enol
ethers also suggests that allyllithium 1 adopts an endo-
orientation, probably promoted by coordination of an
3024 | Org. Biomol. Chem., 2014, 12, 3021–3025
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Communication
internal lithium ion with the OBn group. For references,
see: ref. 4d and 9.
45, 643 For the selected reviews, see: (f) R. G. Pearson,
Chemical Hardness, Wiley-VCH, Weinheim, 1997;
(g) T.-L. Ho, Chem. Rev., 1975, 75, 1; (h) H. Mayr,
M. Breugst and A. R. Ofial, Angew. Chem., Int. Ed., 2011, 50,
6470.
12 (a) G. Klopman, J. Am. Chem. Soc., 1968, 90, 223;
(b) L. Salem, J. Am. Chem. Soc., 1968, 90, 543;
(c) G. Klopman, Chemical Reactivity and Reaction Paths,
Wiley, 1974; (d) I. Fleming, Frontier Orbitals and Organic
Chemical Reactions, Wiley, Chichester, 1976; (e) I. Fleming,
Molecular Orbitals and Organic Chemical Reactions, Student
Edition, John Wiley & Sons, Chichester, 2009.
13 (a) O. Eisenstein, J. M. Lefour, C. Minot, N. T. Anh and
G. Soussan, C. R. Acad. Sci. Paris Ser. C, 1972, 1310;
(b) B. Deschamps, N. T. Anh and J. Seyden-Penne, Tetrahe-
dron Lett., 1973, 14, 521; (c) K. N. Houk and R. W. Strozier,
J. Am. Chem. Soc., 1973, 95, 4094; (d) J.-M. Lefour and
A. Loupy, Tetrahedron, 1978, 34, 2597.
9 W. C. Still and T. L. Macdonald, J. Org. Chem., 1976, 41,
3620. In addition, as previously reported, 4-COM reacts
with proton regioselectively at Cα. This result is consistent
with the present calculation that Cα accumulates more
electron density than Cγ.
10 The role of geminal bis(silane) is reminiscent of the 1,3-
dithiane group, which has been recognized to stabilize the
adjacent allyllithium, therefore creating a soft nucleophili-
city at Cγ to undergo 1,4-addition to enone. For references,
see: (a) F. E. Ziegler and J.-M. Fang, J. Org. Chem., 1981, 46,
825; (b) T. Cohen and M. Myers, J. Org. Chem., 1988, 53, 457.
11 For seminal publications on HSAB, see: (a) R. G. Pearson,
J. Am. Chem. Soc., 1963, 85, 3533; (b) R. G. Pearson, Science,
1966, 151, 172; (c) R. G. Pearson and J. Songstad, J. Am.
Chem. Soc., 1967, 89, 1827; (d) R. G. Pearson, J. Chem.
Educ., 1968, 45, 581; (e) R. G. Pearson, J. Chem. Educ., 1968,
14 (a) H. Taguchi, K. Ghoroku, M. Tadaki, A. Tsubouchi and
T. Takeda, Org. Lett., 2001, 3, 3811; (b) A. Tsubouchi,
S. Enatsu, R. Kanno and T. Takeda, Angew. Chem., Int. Ed.,
2010, 49, 7089.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3021–3025 | 3025