TiCl4 mediated preparation of (E)-non-conjugated homoallylic alcohols with α-substituted allylsilanes

Article abstract of DOI:10.1016/j.tetlet.2013.06.047

The allylation of various aldehydes with α-substituted allylsilanes in the presence of TiCl₄ has been investigated. It has been shown that these reagents readily allow for good yields and high to excellent diastereoselectivities (up to >20:1) for a series of aldehydes, thereby providing a means of preparing non-conjugated (E)-homoallylic alcohols in a single step.

Full text of DOI:10.1016/j.tetlet.2013.06.047

Tetrahedron Letters 54 (2013) 4487–4490
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
TiCl4 mediated preparation of (E)-non-conjugated homoallylic
alcohols with
a-substituted allylsilanes
⇑
Aymara M. M. Albury, Michael P. Jennings
Department of Chemistry, The University of Alabama, 250 Hackberry Lane, Tuscaloosa, AL 35487-0336, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The allylation of various aldehydes with
a-substituted allylsilanes in the presence of TiCl4 has been
Received 17 May 2013
Revised 5 June 2013
Accepted 11 June 2013
Available online 18 June 2013
investigated. It has been shown that these reagents readily allow for good yields and high to excellent
diastereoselectivities (up to >20:1) for a series of aldehydes, thereby providing a means of preparing
non-conjugated (E)-homoallylic alcohols in a single step.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Allylation
Allylsilane
Olefination
Natural products
Diastereoselective synthesis
´
The Lewis acid mediated reaction of allylsilanes with carbonyl
compounds has provided the synthetic chemist with a remarkable
tool for the regio- and stereospecific preparation of homoallylic
alcohols.1 In 1976, Sakurai and co-workers reported that two
cleavage of the terminal alkene) via a Julia–Kocienski protocol.
Unfortunately, both of these mentioned processes have significant
disadvantages.
For example, the terminal olefin moiety of a homoallylic alcohol
typically undergoes a non-chemo and diastereoselective cross-
metathesis reaction with another type I alkene to afford a low
yield of product with modest E/Z selectivities.9 Likewise, the
a-substituted allylsilanes reacted with an aliphatic aldehyde under
Lewis acidic conditions to afford the -substituted linear homoal-
c
lylic alcohols as a non-defined cis- and trans-mixture of olefin
geometry.2 Later, Kumada investigated the addition of the chiral
Julia–Kocienski olefination would require a minimum of four reac-
´
phenyl
a
-substituted allylsilane to a variety of aliphatic aldehydes
tion processes commencing with protection of the alcohol moiety,
oxidative cleavage of the terminal alkene, olefination and a final
deprotection to unmask the alcohol functional group.10 Herein,
we wish to report on a systematic study leading to the successful
direct preparation of non-conjugated (E)-homoallylic alcohols in
one step from the parent aldehyde via a Lewis acid mediated addi-
and observed solely the E stereochemistry of the conjugated olefin
product.3,4 Subsequently, Panek and Miyashita have shown that
substituted chiral crotylsilanes react with acetals to provide high
levels of dr and er for the newly formed stereocenters coupled with
selective (E)-olefin geometry formation of the a-substituted homo-
allylic ether products.5,6
tion of a-substituted allylsilanes.
While the synthesis of (E)-conjugated
alcohols (or ethers) has been disclosed by means of substituted
allylsilanes, surprisingly the synthesis of non-conjugated
a
-substituted homoallylic
As shown in Scheme 2, preparation of the substituted allylsi-
lanes 5a and 5b utilized vinyl silanes 1a and 1b, which were syn-
thesized based on our previous report.11 Thus, treatment of 1a
and 1b with Pearlman’s catalyst [Pd(OH2)] under an atmosphere
(E)-homoallylic alcohols derived from a-substituted allylsilane re-
agents has yet to be fully investigated as described in Scheme 1.7
In order to obtain such said products prior to this investigation
utilizing allylsilanes, further functionalization of the given terminal
alkene resident in an unsubstituted homoallylic alcohol would have
to be conducted.8 A couple of options for this additional functional-
ization include a cross metathesis with another type I olefin or
olefination of the resultant aldehyde (by means of oxidative
of H2 in EtOH readily provided the saturated a-silyl esters 2a and
2b in 89% and 83% yields, respectively.12 Initially, we had hoped
to partially reduce the ester moieties of 2a and 2b with DIBAL to
the corresponding aldehydes. However, we consistently observed
over reduction of the carbonyls and after reaction optimization ob-
tained alcohols 3a and 3b in yields of >80%. An ensuing oxidation of
3a and 3b with Dess–Martin Periodinane (4) furnished the desired,
yet chromatography unstable
a-silyl aldehydes which were used
directly for the subsequent Wittig olefination. Accordingly, the
crude aldehydes were added to the preformed methylene ylide
⇑
Corresponding author. Tel.: +1 (205) 348 0351; fax: +1 (205) 348 9104.
E-mail address: jenningm@bama.ua.edu (M.P. Jennings).
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.06.047

4488
A. M. M. Albury, M. P. Jennings / Tetrahedron Letters 54 (2013) 4487–4490
Table 1
OH
Allylation of propanal with allylsilane 5a in the presence of different lewis acids and
Me3Si
R'CHO
TiCl4
R
solventsa
R'
R
OH
Me3Si
"cis and trans mixture"
Lewis Acid
R = Me,Ph
R
Sakurai1976
O
6a
5a
H
OH
Me3Si
R'CHO
TiCl4
No.
Lewis acid
Solvent
Yield (%)
E/Zb
R'
1
2
3
4
5
6
7
8
9
10
TiCl4
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Toluene
Cl(CH2)2Cl
Hexane
66
7
23
—
—
—
11/1
5/1
10/1
—
—
—
R
BF3ÁOEt2
TMSOTf
SnCl4
Kumada1982,1983
R = Me, H
c
In(OTf)3
OMe
c
R'CHO, TMSOMe
TMSOTf
Sc(OTf)3
PhMe2Si
PhMe2Si
PhMe2Si
c
R
Er(OTf)3
—
—
R'
TiCl4
TiCl4
TiCl4
39
17
7
10/1
11/1
9/1
R
Panek1991,2010
R = OTBS, COOMe
OR
OR
a
Reaction ran with 1 equiv of 5a, 1.3 equiv of Lewis acid, and 1.3 equiv of
R'
OAc
propanal for 3 h at À78 °C.
OBn
R'
TiCl3(OiPr)
E/Z ratio determined by 1H NMR (500 MHz) from the purified reaction mixture.
Lewis acid was dissolved in THF before being added to the reaction.
b
c
Miyashita 2004
OBn
OMe
R'CHO, TMSOMe
TMSOTf
solution of 5a and propanal (1.3 equiv) in CH2Cl2 at À78 °C for
3 h provided homoallylic alcohol 6a in 66% yield with an E/Z ratio
of 11/1. Exchanging the Lewis acid TiCl4 for BF3ÁOEt2 dramatically
reduced both the yield of 6a to 7% and diastereoselectivity (5/1)
of the newly formed olefin moiety. Similarly, the usage of TMSOTf
provided a low 23% yield of 6a, however the E/Z ratio was restored
to ꢀ10/1 favoring the E-alkene. It is worth noting that under both
reaction conditions utilizing BF3ÁOEt2 and TMSOTF, the major prod-
uct was the substituted THF ring as reported by Panek, Woerpel
and Roush.15 Surprisingly, when we examined SnCl4 in place of
TiCl4 little to no homoallylic alcohol 6a was isolated, but complete
decomposition of allylsilane 5a was observed. Correspondingly, the
utilization of rare-Earth triflate salts [In(OTf)3, Sc(OTf)3 and
Er(OTf)3] as Lewis acids led to limited product formation (<5%)
and quantitative re-isolation of starting material 5a.
Armed with the knowledge that TiCl4 was the optimal Lewis
acid and provided 6a with the greatest yield (66%) and E/Z selectiv-
ity (11/1), we sought to further define the reaction scope by exam-
ining a potential solvent effect. We exchanged CH2Cl2 for a variety
of other non-polar solvents (toluene, hexane, and ClCH2CH2Cl) and
observed significantly reduced yields (7–39%) while maintaining
the olefin diastereoselectivity of P9/1 for the E-alkene. It was a lit-
tle unexpected that ClCH2CH2Cl afforded such a low yield of 17%
compared to that of CH2Cl2 (66%). With the reaction conditions
(1.3 equiv of TiCl4) in hand as described in Table 1, we desired to
further investigate the scope and limitations of both 5a and 5b
as allylating reagents with an assortment of aldehydes.
COOMe
R'
COOMe
Panek2010
Scheme 1.
O
R
O
a) H2, Pd(OH)2
Me3Si
Me3Si
OEt
OEt
EtOH
R
2a = 89%
2b = 83%
1a: R = Ph
1b: R = C6H13
O
O
b) DIBAL
toluene
I
OAc
OAc
4
AcO
c) 4, NaHCO3
Me3Si
Me3Si
OH
CH2Cl2
d) Ph3P=CH2
THF
R
R
3a = 85%
3b = 81%
5a = 48% over 2 steps
5b = 45%over 2 steps
Scheme 2.
(generated from Ph3P+-CH3 BrÀ and nBuLi) and afforded allylsilanes
5a and 5b in modest yields of 48% and 45% over two steps from 3a
and 3b. It is worth noting that the elimination of Ph3P@O was pref-
erential to that of a Peterson type olefination with a loss of
TMSOH.13
With the desired allylsilanes in hand, we proceeded to investi-
gate the Lewis acid-mediated addition of 5a to propanal under a
variety of reaction conditions as shown in Table 1. We initiated
the inquiry with a couple of objectives in mind. First, we wanted
a process that would provide solely the linear homoallylic alcohol
and not the ether product (via the acetal or intramolecular cycliza-
tion to afford the substituted THF ring) as developed by Panek.14
Secondly, the yield and E/Z selectivity must mirror or be greater
than that of any multi-step approach. With these goals in mind,
we initiated our examination by scanning a variety of Lewis Acids.
As shown in Table 1, the slow addition of 1.3 equiv of TiCl4 to a
Initial treatment of the TBDPS protected aldehyde16 (derived
from 1,3-propanediol) with 5a in the presence of 1.3 equiv of TiCl4
provided alcohol 7a in ꢀ50% yield with an E/Z ratio of 13/1 as deter-
mined by 1H NMR while a significant amount of aldehyde remained
unreacted (ꢀ35–40%). Thus, addition of another 0.2 equiv of TiCl4
and a second equiv of silane 5a readily promoted the complete con-
sumption of the aldehyde and afforded 7a in 82% yield with an E/Z
ratio of 16/1 as described in Table 2. Based on this improvement in
yield and diastereoselectivity of the newly formed alkene, we car-
ried out the remaining allylations with both 5a and 5b (2 equiv) uti-
lizing 1.5 equiv of TiCl4. Hence, treatment of propanal and 5a with
1.5 equiv of TiCl4 furnished 6a in a greater yield of 77% with an in-
creased E/Z ratio of 12/1. Under identical reaction conditions, the
addition of 5b to propanal afforded homoallylic alcohol 6b with

A. M. M. Albury, M. P. Jennings / Tetrahedron Letters 54 (2013) 4487–4490
4489
Table 2
As noted above, attempted TiCl4-mediated allylation of p-tolyl-
aldehyde and p-anisaldehyde with 5a under the standardized reac-
tion conditions from Table 2 did not afford the desired homoallylic
alcohols. As shown in Scheme 3, the addition of 2 equiv of 5a to p-
tolylaldehyde (with 1.5 equiv of TiCl4) afforded the diallylated
compound 10a in 46% yield with an E/Z ratio of 9/1.17 Likewise,
treatment of p-anisaldehyde under identical conditions as noted
furnished the diallylated compound 11a in 61% yield and a slightly
decreased E/Z ratio of 8/1.
Allylation of various aldehydes with allylsilane 5a and 5ba,b,c
Me3Si
TiCl4
OH
R'CHO
R'
R
R
5a R = Ph
6 - 9
5b R = C6H13
Aldehyde
Product yield (E/Z)
O
OH
As reported by Reetz and Keck, allylsilanes and stannanes typi-
cally react via a chelation controlled addition to a b-ether carbonyl
to furnish the 1,3-anti diol product.18,19 With this in mind, we
elected to investigate the allylation of the benzyl protected b-hy-
droxy aldehyde 12 under the previously described reaction condi-
tions with silane 5a. As described in Scheme 4, treatment of 1220
with 1.5 equiv of TiCl4 presumably formed the six-membered che-
lated intermediate coupled with the addition of the allylsilane re-
agent 5a led to the formation of the 1,3-homoallylic alcohol 13
with a diastereomeric ratio of >20:1 with the presumed anti-ste-
reochemistry (based on literature precedent) in 81% yield. In addi-
tion, the E/Z ratio of the newly formed alkene was determined to be
>20/1 by 1H NMR. In order to unequivocally determine the diol ste-
reochemistry resident in 13 we subsequently performed a concom-
itant hydrogenation/hydrogenolysis of both the olefin and benzyl
ether with 1 atm of H2 and 10% Pd(OH)2 in EtOH to afford diol 14
in 57% yield. Final acetonide formation of 14 under the standard
reaction conditions of 2,2-dimethoxypropane and CSA provided
acetal 15 in 76% yield.
With 15 in hand, we initially hoped to determine the anti-ste-
reochemistry via 1D NOE spectroscopy. Unfortunately, non-re-
solved ether methine signals at ꢀ3.7 ppm in the 1H NMR
spectrum (500 MHz) did not permit for the preferred NOE experi-
ment. However, close inspection of the 13C NMR did provide the
necessary information for the positive confirmation of the pro-
posed anti-stereochemistry. As noted by Rychnovsky, the 13C
NMR chemical shifts of the two geminal methyl groups of an
anti-acetonide moiety should exhibit pseudo-equivalent signals
H
6a
77%, E/ Z:12/1
OH
5
6b
84%, E/ Z:11/1
O
OH
H
TBDPSO
TBDPSO
7a
82%, E/Z:16/1
OH
TBDPSO
5
7b
85%, E/Z:16/1
O
OH
H
8a
F3C
F3C
F3C
51%, E/Z:8/1
OH
5
8b
31%, E/Z:8/1
O
OH
H
9a
69%, E/Z:17/1
Key nOe
Enhancements
OH
H
H
H
H
TBDPSO
O
Ph
H
H
TiCl4
H
7a
Ph
5a
a
R
Reaction ran with 2 equiv of 5a/5b, 1.5 equiv of TiCl4, and 1.0 equiv of aldehyde
R
for 3 h at À78 °C.
b
E/Z ratio determined by 1H NMR (500 MHz) from the purified reaction mixture.
Yields are of the isolated, pure compounds.
R = OMe, Me
10a: R = Me, 46%, E/Z = 9/1
11a
c
: R = OMe, 61%, E/Z =8/1
Scheme 3.
84% yield coupled with an 11/1:E/Z ratio. Likewise, allylation of the
TBDPS protected aldehyde with 5b provided alcohol 7b with a very
high diastereoselectivity of 16/1 favoring the E isomer in 85% yield.
Similar to that of both of the other aliphatic aldehydes, allylation of
3-phenyl-1-propanal with 5a readily proceeded to furnish the cor-
responding homoallylic alcohol 9a in 69% yield and an E/Z ratio of
17/1. Interestingly, attempted allylation of electron neutral and rich
aromatic aldehydes with both 5a and 5b failed to provide the de-
sired homoallylic alcohols, vide infra. However, the electron defi-
cient p-CF3-benzaldehyde underwent allylation with both 5a and
5b to provide the corresponding homoallylic alcohols 8a and 8b
in 51% and 35% yields, respectively. However, the E/Z ratios for both
8a and 8b (8/1 favoring the E isomer) were diminished to that of
their aliphatic counterparts. As delineated in Table 2, the olefin
geometry of 7a was determined by means of 1H NMR. The strong
1D NOE for both the methylene allylic and vinylic protons provided
convincing evidence for the assigned E-olefin geometry. Based on
the NOE experimental results for 7a, the stereochemistry of the
remaining homoallylic adducts was assigned by analogy.
O
OH
O
O
O
5a
a) , TiCl4
81%
Ph
H
5
5
13
E/Z >20:1
12
anti/syn >20:1
b) H2/Pd(OH)2
EtOH
57%
OH OH
O
c) DMP, CSA
Ph
Ph
5
5
4
4
76%
15
14
Scheme 4.

4490
A. M. M. Albury, M. P. Jennings / Tetrahedron Letters 54 (2013) 4487–4490
due to the twist boat conformation at ꢀ25 ppm.21 In our specific
case, the methyl signals for acetonide 15 overlap at 24.8 ppm (as
determined by HSQC) in the 13C NMR. Based on literature prece-
dence, we felt confident that the stereochemistry of the 1,3-diol
subunit was indeed anti.21
In conclusion, we have shown, by means of a systematic study, a
successful direct preparation of non-conjugated (E)-homoallylic
alcohols in one step from the parent aldehyde via a Lewis acid
2. Hosomi, A.; Sakurai, H. Tetrahedron Lett. 1976, 17, 1295.
3. (a) Hayashi, T.; Konishi, M.; Ito, H.; Kumada, M. J. Am. Chem. Soc. 1982, 104,
4962; (b) Hayashi, T.; Konishi, M.; Kumada, M. J. Am. Chem. Soc. 1982, 104,
4963.
4. Hayashi, T.; Konishi, M.; Kumada, M. J. Org. Chem. 1983, 48, 281.
5. Panek, J. S.; Yang, M. J. Org. Chem. 1991, 56, 5755.
6. (a) Wu, J.; Chen, Y.; Panek, J. S. Org. Lett. 2010, 12, 2112; (b) Komatsu, K.;
Tanino, K.; Miyashita, M. Angew. Chem., Int. Ed. 2004, 43, 4341.
7. For an excellent review, see (a) Masse, C. E.; Panek, J. S. Chem. Rev. 1995, 95,
1293; (b) Osumi, K.; Sugimura, H. Tetrahedron Lett. 1995, 36, 5789; (c) Brown,
M. J.; Harrison, T.; Overman, L. E. J. Am. Chem. Soc. 1991, 113, 5378; (d) Grese, T.
A.; Hutchinson, K. D.; Overman, L. E. J. Org. Chem. 1993, 58, 2468.
mediated addition of
a-substituted allylsilanes. Future directions
of investigation will include further developments of novel chiral
allyl silane reagents derived from our previously reported vinyl
silanes and their utilization in target driven synthesis. Results from
these studies will be reported in due course.
8. For reports on using a-substituted allylboranes for the synthesis of homoallylic
alcohols with selective E olefin geometry, see (a) Peng, F.; Hall, D. G. J. Am.
Chem. Soc. 2007, 129, 3070; (b) Hall, D. G. Pure Appl. Chem. 2008, 80, 913; (c)
Chen, M.; Roush, W. R. Org. Lett. 2010, 12, 2706.
9. (a) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2010, 110, 1746; (b)
Chatterjee, A. K.; Choi, T. L.; Sanders, D. P.; Grubbs, R. H. J. Am. Chem. Soc. 2003,
125, 11360.
10. (a) Blakemore, P. R.; Cole, W. J.; Kocien´ ski, P. J.; Morley, A. Synlett. 1998, 26; (b)
Blakemore, P. R. J. Chem. Soc., Perkin Trans. 1 2002, 2563.
11. Mueller Hendrix, A. J.; Jennings, M. P. Org. Lett. 2010, 12, 2750.
12. Pearlman, W. M. Tetrahedron Lett. 1967, 8, 1663.
13. Peterson, D. J. J. Org. Chem. 1968, 33, 780.
14. (a) Panek, J.; Yang, M. J. Am. Chem. Soc. 1991, 113, 9868; (b) Panek, J.; Beresis, R.
J. Org. Chem. 1993, 58, 809.
Acknowledgments
Support for this project was provided by the University of
Alabama and in part the National Science Foundation CAREER
program under CHE-0845011.
15. (a) Peng, Z. H.; Woerpel, K. A. Org. Lett. 2002, 4, 2945; (b) Peng, Z. H.; Woerpel,
K. A. J. Am. Chem. Soc. 2003, 125, 6018; (c) Heitzman, C. L.; Lambert, W. T.;
Mertz, E.; Shotwell, J. B.; Tinsley, J. M.; Va, P.; Roush, W. R. Org. Lett. 2005, 7,
2405.
16. Druais, V.; Hall, M. J.; Corsi, C.; Wendeborn, S. V.; Meyer, C.; Cossy, J. Org. Lett.
2009, 11, 935.
17. (a) Albaugh-Robertson, P.; Katzenellenbogen, J. A. J. Org. Chem. 1983, 48, 5288;
(b) Durand, A.-C.; Brahmi, L.; Lahrech, M.; Hacini, S.; Santelli, M. Synth.
Commun. 1825, 2005, 35.
18. (a) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556; (b) Reetz, M. T. Acc.
Chem. Res. 1993, 26, 462.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2013.06.
047.
References and notes
1. (a) Fleming, I. Org. React. 1989, 37, 57; (b) G. Majetich, in: Organic Synthesis:
Theory and Application, vol. 1, 1989, p. 173.; (c) Yamamoto, Y.; Asao, N. Chem.
Rev. 1993, 93, 2239; (d) Denmark, S. E.; Almstead, N. G. J. Mex. Chem. Soc. 2009,
53, 174.
19. Keck, G. E.; Castellino, S.; Wiley, M. R. J. Org. Chem. 1986, 51, 5478.
20. Kurosu, M.; Lorca, M. Synlett 2005, 1109.
21. Rychnovsky, S. D.; Rogers, B.; Yang, G. J. Org. Chem. 1993, 58, 3511.