Additions of functionalized α-substituted allylboronates to aldehydes under the novel Lewis and Br?nsted acid catalyzed manifolds

Article abstract of DOI:10.1016/j.tetlet.2005.10.115

The stereo- and chemoselectivity in the additions of four model α-substituted allylboronates to benzaldehyde was examined under the standard thermal (uncatalyzed) conditions and the novel Lewis and Br?nsted acid-catalyzed conditions. With either of Sc(OTf

Full text of DOI:10.1016/j.tetlet.2005.10.115

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 8981–8985
Additions of functionalized a-substituted allylboronates
to aldehydes under the novel Lewis and Brønsted acid
catalyzed manifolds
Lisa Carosi, Hugo Lachance and Dennis G. Hall*
Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2
Received 27 September 2005; revised 20 October 2005; accepted 21 October 2005
Available online 9 November 2005
Abstract—The stereo- and chemoselectivity in the additions of four model a-substituted allylboronates to benzaldehyde was exam-
ined under the standard thermal (uncatalyzed) conditions and the novel Lewis and Brønsted acid-catalyzed conditions. With either
of Sc(OTf) or triflic acid as catalysts, an a-ethyl allylboronate, 1a, led to a surprising inversion of stereoselectivity that can be ten-
3
tatively rationalized through subtle differences in the geometry of the allylboration transition state between uncatalyzed and cata-
lyzed pathways. It was also found that the chemoselectivity of a-silyl reagents (1c and 1d) can be reversed upon use of a catalyst,
providing allylsilation products instead of the allylboration product obtained from the thermal uncatalyzed reaction.
Ó 2005 Elsevier Ltd. All rights reserved.
1
. Introduction
tive of this study with racemic a-substituted allylboro-
nates is to provide additional mechanistic insight on
the novel catalytic allylboration manifolds, and to help
identify the most promising combination of substituents
and catalytic conditions in view of using optically pure
a-chiral reagents in the future.
1
Since their discovery, uncatalyzed additions of allylbor-
onates to aldehydes have evolved into one of the most
popular methods for stereoselective C–C bond forma-
tion. Compared to dialkyl allylic boranes, allylic boro-
2
nic esters are often more advantageous as a class of
reagents due to their increased air-stability. Two strate-
gies have been developed for controlling enantiofacial
selectivity in the additions of allylic boronates to achiral
aldehydes: (1) the use of a chiral diol or a diamine deriv-
ative as boronÕs two non-allylic substituents (B-chiral
The preparation of a-chiral allylboronates 1 and their
additions to aldehydes were pioneered by Hoffmann.
4
These reagent-controlled additions proceed with almost
perfect transfer of chirality to give two diastereomeric
products 4 and 5 (Fig. 1). These Z and E allylic alcohol
products are epimeric, and their ratio is highly depen-
3
allylboronates); and (2) the use of optically pure, a-
1
substituted reagents (C-chiral, or a-chiral allylboro-
dent upon the nature of the a-substituent (R ) and the
4
4
nates). The recent discovery of Lewis acid catalyzed
nature of the boronic ester. The ratio of 4 and 5 can
5
6
additions of allylboronates, by our group and others,
be explained in terms of steric and dipolar effects on
opened doors to new and highly stereoselective aldehyde
allylation methods using B-chiral allylboronates.
the two competing chairlike transition structures 2 and
3. With a non-polar alkyl substituent R , steric interac-
7
1
Whereas the scope of aldehyde substrates in Lewis
acid-catalyzed allylboration has been thoroughly inves-
tigated, little is known on the scope of functionalized
allylboronates. Here, we present an investigation on
the scope and stereoselectivity of additions of a-substi-
tuted allylboronates promoted by the low-temperature
tions play a dominant role on the relative energies of
these putative transition structures. Transition structure
3 is destabilized by a steric interaction between the boro-
nic ester and the pseudo-equatorial a-substituent R . On
the other hand, transition structure 2 features unfavor-
1
able allylic interactions due to the pseudo-axial position
8
1
Lewis and Brønsted acid-catalyzed variants. The objec-
of the R substituent. The use of a hindered ester, such
1
as pinacolate, aggravates interactions between R and
the pinacol methyls in structure 3, and tends to favor
transition structure 2 leading to the Z-configured prod-
uct 4. With a polar R substituent (halogen, alkoxy),
*
Corresponding author. Tel.: +1 780 492 3141; fax: +1 780 492
231; e-mail: dennis.hall@ualberta.ca
9
1
8
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.10.115

8982
L. Carosi et al. / Tetrahedron Letters 46 (2005) 8981–8985
O
B
HO
O
B
O
B
EtMgCl
O
(1)
(2)
R2
Cl
O
B
O
O
H
O
THF, -15 °C
O
R1
Et
R1
R2 R1
6
H
4
1a (89%)
1
+
2
O
B
O
O
MgCl
Br
B
O
HO
O
O
R2
THF, -78 °C to rt
R1
R2
B
Br
O
O
R2
Br
R1
7
1b (65%)
5
3
Cl
SiMe3
O
B
O
B
Figure 1. Postulated competing transition structures in the allylation
of aldehydes with a-substituted allylboronates 1. Note: A stereodefined
reagent is shown in order to emphasize the stereochemistry of the
reaction and the epimeric relationship between products 4 and 5.
Li
(3)
(4)
O
O
THF, -78 °C to rt
SiMe3
8
1
c (41%)
O
B
dipolar effects tend to dominate and further favor tran-
sition structure 2 with the pseudo-axial C–R bond anti
to the axial B–O bond.
1
PhMe2SiLi
6
O
THF, -15 °C to rt
SiMe2Ph
1d (60%)
2
. Results and discussion
Scheme 1.
Here, a small panel of racemic a-substituted allylboro-
nates of type 1 representing a variety of substituent types
0
selective S 2 addition on 6, using dimethylphenylsilyl
N
1
5
12
(
alkyl, silyl, halogen) were prepared. Pinacol allylic bor-
lithium (Eq. 4).
onates were chosen due to their ease of handling and
their stability to chromatographic conditions and mois-
ture, and also to help optimize the Z/E ratio of addition
products to favor isomer 4. Thus, a-ethyl, a-bromo, a-
trimethylsilyl, and a-dimethylphenylsilyl reagents 1a–d
The allylation of benzaldehyde with a-ethyl reagent 1a
was examined under different conditions (Table 1).
The thermal (non-catalyzed) reaction demanded to be
performed neat at a temperature of 0 °C to provide a
good yield favoring the Z-isomer 4a over 5a in a 2:1
9
were prepared using known methods shown in Scheme
1
0
16
5,7
1
. Reagent 1a was synthesized through a selective
ratio (entry 1). In contrast, both Sc(OTf) -catalyzed
3
0
8
S 2 addition (or vinylogous Matteson rearrangement)
and TfOH-catalyzed reactions (entries 2 and 3) pro-
ceeded at ꢀ78 °C to give a comparable ratio favoring
E-isomer 5a. The TfOH-catalyzed reaction is particu-
N
of ethylmagnesium chloride to the known 3-chloroprop-
1
1
12
enylboronate 6 (Eq. 1). a-Bromo reagent 1b was
accessed using a reverse Matteson homologation of
dibromomethyl boronate 7 with vinylmagnesium chlo-
1
7
larly efficient, providing a quantitative yield.
1
3
ride (Eq. 2). Likewise, a-trimethylsilyl reagent 1c was
To the best of our knowledge, such an inversion of selec-
tivity is unprecedented and cannot be explained by a
temperature effect. It is likely that subtle differences in
made using a Matteson homologation on vinylboronate
1
4
8
(Eq. 3). Like 1a, reagent 1d was synthesized by a
a
Table 1. Allylboration of a-ethyl and a-bromo allylboronates 1a and 1b with benzaldehyde
O
O
HO
Ph
HO
Ph
B
R1
Ph
H
O
+
R1
CH2Cl2
4 Å m.s.
R1
1
1
a (R = Et)
b (R = Br)
4
5
b
c
Entry
Boronate
Catalyst
T (°C)
Time (h)
Products
Ratio
Yield (%)
1
2
3
4
5
6
1a (R = Et)
1a
None
Sc(OTf)
TfOH
None
Sc(OTf)
TfOH
0
ꢀ78
ꢀ78
0
15
40
40
25
40
40
4a, 5a
4a, 5a
4a, 5a
4b, 5b
4b, 5b
4b, 5b
2:1
75
63
99
75
50
24
3
1:1.3
1:1.7
13:1
10:1
9:1
1a
1b (R = Br)
1b
1b
3
ꢀ78
ꢀ78
Note: Although racemic 1a/1b were used, stereodefined reagents are shown in order to emphasize the epimeric relationship between products 4 and 5.
a
Typical reaction scale: approx. 0.7–1.0 mmol, 0.2 M in dichloromethane.
Ratio of 4 to 5 as determined by H NMR analysis of crude reaction mixtures.
Unoptimized yields of pure products isolated after flash chromatography. m.s. = molecular sieves.
b
c
1

L. Carosi et al. / Tetrahedron Letters 46 (2005) 8981–8985
8983
predominance of the Z-isomer 4b for reactions of a-
bromo reagent 1b can be explained by the particular
strength of the dipolar effect described above.
O
B
O
B
4
a
a
5a
5a
O
O
Ph
Ph
O
O
Et
O
Ph Et
H
H
The a-silyl substituted reagents 1c and 1d presented fur-
ther interesting opportunities to gain insight into the
mechanism of Lewis and Brønsted acid-catalyzed allyl-
boration. Indeed, as observed by Hiyama and co-work-
ers, addition of these reagents to aldehydes can lead
either to alkenylsilane products 13c/13d via allylbora-
tion, or to the common alkenylboronate 14 via allylsila-
1
0
9
O
B
O
B
4
O
Et
O
O
Ph Et
L.A.
+
L.A.
+
(H )
(H )
1
2
19
1
1
tion (Table 2). The nature of products observed would
reveal information on the preferred mode of activation
by the catalyst, that is, aldehyde activation in the allyl-
silation versus boronate activation in the allylboration.
As can be seen with the modest yields of products, these
allylations are difficult (Table 2). However, aside from
products, starting materials were recovered with very lit-
tle degradation. With both reagents 1c and 1d, the ther-
mal reaction led to the expected allylboration to give the
Z alkenylsilanes 13c and 13d as major products (entries
1 and 3).
Figure 2. Postulated competing transition structures in the allylation
of benzaldehyde with a-ethyl allylboronate (1a). L.A. = Lewis acid.
the geometry of the transition structures between the
uncatalyzed and catalyzed reaction manifolds are
responsible for the inversion of selectivity. Our previous
mechanistic study on Lewis acid-catalyzed allylboration
supported an electrophilic boronate activation mode
through coordination of a boronate oxygen, a process
that increases boronÕs electrophilicity and strengthens
1
8
the B–O(aldehyde) interaction. In the thermal reac-
tion, the preference for 4a over 5a can be explained by
the presence of severe non-bonded interactions between
the pseudo-equatorial ethyl substituent and the large
pinacolate group in transition structure 10 leading to
The predominance of the Z-isomers of 13c/13d can be
explained by the large size of the silyl group, which
would aggravate the interactions with the pinacolate
group in transition structure 3. To our initial surprise,
both Sc(OTf) and TfOH-catalyzed reactions provided
3
5
a (Fig. 2).
the E-alkenylboronate 14 as major product of what
appears to be acid-catalyzed allylsilations (entries 2, 3,
5, and 6). No allylboration products were observed. This
interesting switch of chemoselectivity can be explained
by a preference for an aldehyde activation mode (i.e.,
transition state assembly 16) over the electrophilic boro-
nate activation mode (15) expected in the absence of an
a-silyl substituent (Fig. 3).
To explain the inversion of selectivity, we hypothesize
that the acid-catalyzed processes feature more advanced
transition states with a shorter B–O(aldehyde) bond and
a longer B–C bond. Thus, in structure 12 such a longer
bond alleviates the non-bonded interactions between the
ethyl substituent and the pinacolate, which would be-
come less important than the allylic interaction caused
by the pseudo-axial ethyl substituent in structure 11.
Hence, a switch of selectivity from the non-catalyzed
reaction occurs and product 5a becomes predominant.
The a-bromo reagent 5b provided comparable selectivi-
ties invariably of conditions employed, thermal, or acid
catalyzed (entries 4–6). The yield of product, however,
was acceptable only in the uncatalyzed reaction. The
3. Conclusion
This study on the allylation of aldehydes using a-substi-
tuted allylboronates provides additional insight into
the mechanism of Lewis and Brønsted acid-catalyzed
allylboration manifolds. Compared to the thermal
a
Table 2. Allylboration of a-silyl allylboronates 1c (R = SiMe
3
) and 1d (R = SiMe
2
Ph) with benzaldehyde
O
O
B
HO
Ph
HO
Ph
O
H
+
Bpin
Ph
SiR3
2
CH Cl2
SiR3
13
14
1
c (R = SiMe3)
1
d (R = SiMe Ph)
2
b
Entry
Boronate
Catalyst
None
T (°C)
Time (h)
Product
E/Z
Yield (%)
1
2
3
4
5
6
1c
1c
1c
1d
1d
1d
0
ꢀ78
ꢀ78
0
15
40
40
15
40
40
13c
14
1:6.7
4:1
3:1
1:6.2
4:1
4:1
46
40
43
22
11
26
Sc(OTf)
TfOH
3
14
None
13d
14
Sc(OTf)
TfOH
3
ꢀ78
ꢀ78
14
a
Typical reaction scale: approx. 0.3 mmol, 0.2 M in dichloromethane.
Unoptimized yields of pure products isolated after flash chromatography.
b

8984
L. Carosi et al. / Tetrahedron Letters 46 (2005) 8981–8985
6
348; (b) Corey, E. J.; Yu, C.-M.; Kim, S. S. J. Am. Chem.
O
B
O
B
HO
Ph
Soc. 1989, 111, 5495–5496.
4. (a) Hoffmann, R. W. Pure Appl. Chem. 1988, 60, 123; (b)
Hoffmann, R. W.; Neil, G.; Schlapbach, A. Pure Appl.
Chem. 1990, 62, 1993; (c) Hoffmann, R. W.; Wolff, J. J.
Chem. Ber. 1991, 124, 563–569.
5. (a) Kennedy, J. W. J.; Hall, D. G. J. Am. Chem. Soc. 2002,
124, 11586–11587; (b) Kennedy, J. W. J.; Hall, D. G. J.
Org. Chem. 2004, 69, 4412–4428.
. Ishiyama, T.; Ahiko, T.-a.; Miyaura, N. J. Am. Chem.
Soc. 2002, 124, 12414–12415.
7. (a) Lachance, H.; Lu, X.; Gravel, M.; Hall, D. G. J. Am.
Chem. Soc. 2003, 125, 10160–10161; (b) Gravel, M.;
Lachance, H.; Lu, X.; Hall, D. G. Synthesis 2004, 1290–
O
O
SiR3
O
L.A. (or H+)
Ph
Ph
SiR3
SiR3
1
3
1
c/d
1
5
L.A.
or
H
+
L.A. (or H+)
O
+
O
HO
Ph
6
Ph
H
Bpin
H
O
B
O
14
SiR3
1
6
1
302.
Figure 3. Postulated competing mechanisms in the allylation of
benzaldehyde with a-silyl allylboronates 1c and 1d. Although racemic
8. Yu, S. H.; Hall, D. G. J. Am. Chem. Soc. 2005, 127,
12808–12809.
1
c/1d were employed, stereodefined reagents are shown in order to
9. Hoffmann, R. W.; Weidmann, U. J. Organomet. Chem.
1980, 195, 137–146.
emphasize the stereospecificity in the formation of 13. L.A. = Lewis
acid.
10. Preparation of allylboronate 1a and spectral data:
Typical procedure to make boronic ester 1a: In a flame
dried flask under inert atmosphere, alkenylboronic ester 6
uncatalyzed reaction of 1a the inversion of stereoselec-
tivity in the formation of homoallylic alcohols under
acid catalysis is suggestive of a more advanced transition
structure with a shorter B–O(aldehyde) bond and a
longer B–C bond. The case of a-silyl reagents also led
to an interesting switch of selectivity, one of chemoselec-
tivity that provides information on the relative easiness
between boronate and aldehyde activation pathways.
Overall, our results with reagent 1a suggest that careful
optimization of the nature of the alkyl group and boro-
nateÕs diol unit in an optically pure a-chiral allylboro-
nate could provide a stereodivergent method to access
both enantiomers of homoallylic alcohols by a simple
choice of thermal or acid-catalyzed allylboration
conditions.
(
630 mg, 3.1 mmol, 1.0 equiv) was charged, diluted with
THF (5 mL), and cooled to ꢀ15 °C in an ethylene glycol,
dry ice bath. A solution of ethylmagnesium chloride 2.0 M
in THF (2.0 mL, 4.0 mmol, 1.3 equiv) was added dropwise
and the mixture was stirred for 30 min at ꢀ15 °C. The
reaction mixture was quenched with an aqueous solution
of saturated ammonium chloride and brought to room
temperature. The phases were then separated, the aqueous
phase was extracted three times using dichloromethane.
The combined organic phases were washed once with
brine, dried over anhydrous sodium sulfate, filtered, and
concentrated in vacuo to afford 562 mg (92% yield) of
boronic ester 1a as a colorless oil (>90% purity). Reagent
1
a was used without further purification.
Characterization for 1a: IR (dichloromethane cast film)
ꢀ1
1
2
4
1
978, 1632, 1360, 1319, 1143 cm ; H NMR (CDCl ,
3
00 MHz) d 5.79 (ddd, 1H, J = 17.1 Hz, J = 10.2 Hz, J =
.2 Hz), 4.99 (ddd, 1H, J = 17.1 Hz, J = 2.0 Hz, J =
Acknowledgments
1.2 Hz), 4.94 (ddd, 1H, J = 10.2 Hz, J = 2.0 Hz,
J = 0.9 Hz), 1.75 (ddd, 1H, J = 7.8 Hz, J = 7.7 Hz, J =
7.6 Hz), 1.60 (ddq, 1H, J = 13.3 Hz, J = 7.3 Hz,
J = 7.3 Hz), 1.46 (ddq, 1H, J = 13.3 Hz, J = 7.4 Hz, J =
This research work was supported by the Natural Sci-
ences and Engineering Research Council (NSERC) of
Canada and the University of Alberta. L.C. thanks the
University of Alberta for a Province of Alberta Gradu-
ate Scholarship. H.L. thanks NSERC and the Alberta
Ingenuity Fund for graduate scholarships.
7
7
2
.4 Hz), 1.25 (s, 12H), 0.92 (dd, 3H, J = 7.4 Hz, J =
13
.3 Hz); C NMR (CDCl
3
, 100 MHz) d 139.5, 113.5, 83.1,
1
1
4.7, 24.6, 23.4, 13.5; B NMR (CDCl , 128.3 MHz) d
3
3
3.2; HRMS (EI) m/z calcd for C H BO : 196.1635.
1
1
21
2
found: 196.1627.
Spectral data for 1b can be found in Ref. 13.
Spectral data for 1c can be found in Ref. 14.
Spectral data for 1d can be found in Ref. 19.
References and notes
1
1. Gravel, M.; Tour e´ , B. B.; Hall, D. G. Org. Prep. Proc. Int.
2004, 36, 573–579.
12. Lombardo, M.; Morganti, S.; Tozzi, M.; Trombini, L.
Eur. J. Org. Chem. 2002, 2823–2830.
13. Hoffmann, R. W.; Landmann, B. Chem. Ber. 1986, 119,
1039–1053.
14. Matteson, D. S.; Majumdar, D. Organometallics 1983, 2,
236–241.
1
2
. Blais, J.; LÕHonor e´ , A.; Souli e´ , J.; Cadiot, P. J. Organo-
met. Chem. 1974, 78, 323–337.
. (a) Roush, W. R. In Houben-Weyl, 4th ed.; Stereoselective
Synthesis; Thieme: Stuttgart, Germany, 1995; Vol. E21b,
Chapter 1.3.3.3.3; (b) Denmark, S. E.; Almstead, N. G. In
Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-VCH:
Weinheim, Germany, 2000, Chapter 10, pp 299–402; (c)
Chemler, S. R.; Roush, W. R. In Modern Carbonyl
Chemistry; Otera, J., Ed.; Wiley-VCH: Weinheim, Ger-
many, 2000, Chapter 11, pp 403–490; (d) Kennedy, J. W.
J.; Hall, D. G. In Boronic Acids: Preparation and Appli-
cations in Organic Synthesis and Medicine; Hall, D. G.,
Ed.; Wiley-VCH: Weinheim, Germany, 2005, Chapter 6,
pp 241–277.
15. Suginome, M.; Matsuda, T.; Ito, Y. Organometallics 2000,
19, 4647–4649.
16. Spectral data for 4a: Nokami, J.; Nomiyama, K.; Shafi,
M. S.; Kataoka, K. Org. Lett. 2004, 6, 1261–1264; Spectral
data for 5a: Nokami, J.; Nomiyama, K.; Matsuda, S.;
Imai, N.; Kataoka, K. Angew. Chem., Int. Ed. 2003, 42,
1273–1276.
3
. (a) Roush, W. R.; Ando, K.; Powers, D. B.; Palkowitz, A.
D.; Halterman, R. L. J. Am. Chem. Soc. 1990, 112, 6339–
17. Typical procedure for allylboration under triflic acid
˚
catalysis: In a flame dried flask, 4 A molecular sieves

L. Carosi et al. / Tetrahedron Letters 46 (2005) 8981–8985
8985
(
240 mg) were charged. The flask was purged three times by
with dichloromethane. The combined organic phases were
washed once with brine, dried over anhydrous sodium
sulfate, filtered, and concentrated in vacuo to afford 513 mg
of a crude mixture. The ratio of compounds 4a to 5a was
vacuum–argon exchange. Dichloromethane (4.5 mL) was
added followed by triflic acid (8 lL, 0.096 mmol,
1
0 mol %). The mixture was cooled to ꢀ78 °C and freshly
1
distilled benzaldehyde (145 lL, 1.4 mmol, 1.5 equiv) was
added. In another flame dried flask, a solution of boronic
ester 1a (188 mg, 0.96 mmol, 1.0 equiv) in 1.3 mL dichlo-
romethane was prepared and added dropwise to the flask
containing benzaldehyde at ꢀ78 °C. The reaction mixture
was stirred at ꢀ78 °C for 40 h, then quenched at ꢀ78 °C
with 1.5 mL of aqueous saturated sodium bicarbonate and
brought to room temperature over 3 h. The phases were
separated and the aqueous phase was extracted three times
determined by H NMR of the crude mixture. The crude
mixture was then purified by flash chromatography on
silica gel using 30% hexanes–70% dichloromethane as
eluent and afforded 172 mg (quantitative yield) of a mixture
f
of compounds 4a and 5a (R : 0.26, 100% dichloromethane).
18. Rauniyar, V.; Hall, D. G. J. Am. Chem. Soc. 2004, 126,
4518–4519.
19. Shimizu, M.; Kitagawa, H.; Kurahashi, T.; Hiyama, T.
Angew. Chem., Int. Ed. 2001, 40, 4283–4286.