Vicinal functionalization of propiolate esters via a tandem catalytic carbocupriation-Mukaiyama aldol reaction sequence

Article abstract of DOI:10.1021/ol800373g

The vicinal functionalization of propiolate esters via a tandem catalytic carbocupration-Mukaiyama aldol reaction sequence has been investigated. It has been shown that catalyst loadings as low as 8 mol % readily allow for good yields and excellent diastereoselectivities (>20:1) for a series of Grignard reagents and aldehydes.

Full text of DOI:10.1021/ol800373g

ORGANIC
LETTERS
2008
Vol. 10, No. 8
1649-1652
Vicinal Functionalization of Propiolate
Esters via a Tandem Catalytic
Carbocupration−Mukaiyama Aldol
Reaction Sequence
Amanda J. Mueller and Michael P. Jennings*
Department of Chemistry, The UniVersity of Alabama,
Tuscaloosa, Alabama 35487-0336
jenningm@bama.ua.edu
Received February 18, 2008
ABSTRACT
The vicinal functionalization of propiolate esters via a tandem catalytic carbocupration−Mukaiyama aldol reaction sequence has been investigated.
It has been shown that catalyst loadings as low as 8 mol % readily allow for good yields and excellent diastereoselectivities (>20:1) for a
series of Grignard reagents and aldehydes.
The advent of new catalytic reactions that allow for the
construction of two carbon-carbon bonds with high levels
of diastereoselectivity in a single flask are of great interest
to the synthetic organic field. Some of the more popular
approaches to multiple carbon-carbon bond-forming reac-
tions are deeply rooted in transition-metal-catalyzed pro-
cesses. For example, Negishi has reported on multiple
intramolecular Heck type processes en route to “zipping”
multiple rings together via Pd catalysis.1 Another approach
to multiple carbon-carbon bond-forming reaction processes
is vicinal functionalization of olefinic or acetylenic com-
pounds. One such thought along this line is a Michael
addition to an R,â-unsaturated carbonyl via a classical
Kharash or Gilman reagent followed by electrophilic quench-
ing of the resultant enolate. While this reaction process is
well-known, very few reports have been published regarding
a vicinal functionalization of R,â-acetylenic esters.2 As
reported by Li and co-workers in 1999, they observed that
a stoichiometric carbocupration under Corey’s conditions
with classical Gilman reagents followed by electrophilic
quenching of the metal allenolate intermediate with an
aldehyde led to product â-hydroxy-R-olefinic esters in good
yields and excellent dr.3 Based on our interest in developing
a diastereoselective, catalytic carbocupration of R,â-acetyl-
enic esters, we envisaged trapping an in situ generated silyl
allenolate with an electrophilic coupling partner.4 It has been
shown that Lewis acid additives such as TMSOTf not only
accelerate the carbocupration of alkynoates but also isomerize
(2) For some leading references, see: (a) Nakamura, E.; Mori, S. Angew.
Chem., Int. Ed. 2000, 39, 3750. (b) Woodward, S. Chem. Soc. ReV. 2000,
29, 393. (c) Lipshutz, B. H.; Sengupta, S. Org. React. 1992, 41, 135. (d)
Kennedy, J. W. J; Hall, D. G. J. Am. Chem. Soc. 2002, 124, 898. (e) Corey,
E. J.; Katzenellenbogen, J. A. J. Am. Chem. Soc. 1969, 91, 1851. (f) Chen,
D.; Guo, L.; Kotti, S. R. S. S.; Li, G. Tetrahedron: Asymmetry 2005, 16,
1757. (g) Wei, H.-X.; Hu, J.; Jasoni, R. L.; Li, G.; Pare, P. W. HelV. Chim.
Acta 2004, 87, 2359. (h) Wei, H.-X.; Timmons, C.; Farag, M. A.; Pare, P.
W.; Li, G. Org. Biomol. Chem. 2004, 2, 2893. (i) Wei, H.-X.; Jasoni, R.
L.; Hu, J.; Li, G.; Pare, P. W. Tetrahedron. 2004, 60, 10233. (j) Wei, H.-
X.; Chen, D.; Xu, X.n; Li, G.; Pare, P. W. Tetrahedron: Asymmetry 2003,
14, 971. (k) Wei, H.-X.; Kim, S. H.; Li, G. Org. Lett. 2002, 4, 3691. (l) Li,
G.; Wei, H.-X.; Phelps, B. S.; Purkiss, D. W.; Kim, S. H. Org. Lett. 2001,
3, 823. (m) Wei, H.-X.; Willis, S.; Li, G. Tetrahedron Lett. 1998, 39, 8203.
(3) Wei, H.-X.; Willis, S.; Li, G. Synth. Commun. 1999, 29, 2959.
(4) The electrophilic capture of silyl allenes derived from carbonyl
compounds have been reported, but not via a one-pot, intermolecular vicinal
functionaliztion. For some leading references, see: (a) Reynolds, T. E.;
Scheidt, K. A. Angew. Chem., Int. Ed. 2007, 46, 7806. (b) Reynolds, T. E.;
Stern, C. A.; Scheidt, K. A. Org. Lett. 2007, 9, 2581. (c) Reynolds, T. E.;
Bharadwaj, A. R.; Scheidt, K. A. J. Am. Chem. Soc. 2006, 128, 15382. (d)
Crimmins, M. T.; Nantermet, P. G. J. Org. Chem. 1990, 55, 4235.
(1) For a leading review in this area, see: Negishi, E.; Coperet, C.; Ma,
S.; Liou, S.-Y.; Liu, F. Chem. ReV. 1996, 96, 365.
10.1021/ol800373g CCC: $40.75
© 2008 American Chemical Society
Published on Web 03/19/2008

the intermediate vinyl cuprate to the TMS allenolate thus
releasing the bound organocuprate.5,6 As described in Figure
1, syn-carbocupration of ethyl propiolate (1a) with the dialkyl
Table 1. Vicinal Functionalization of 1a via a Catalytic
Carbocupration/Mukaiyama Aldol Reaction Sequence
no. mol % T (°C) time (h)
additive
yield (%)
Z/Ea
1
2
3
4
5
6
7
8b
9c
5
5
5
5
5
5
5
30
5
rt
rt
1
3
3
5
21
3
3
3
3
none
none
none
none
none
TiCl4
BF3‚OEt2
BF3‚OEt2
BF3‚OEt2
13
20
30
30
22
40
59
40
32
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
-78
-78
-78
-78
-78
-78
-78
a E/Z ratio determined by H NMR (360 or 500 MHz) from the crude
reaction mixture. b Reaction ran with TMSCl in place of TMSOTf.
c Reaction ran with 1.05 equiv of TMSOTf instead of 1.3 equiv as reported
in entries 1-7.
1
Figure 1. Catalytic cycle for carbocupration of 1a.
or aryl magnesiocuprate should furnish the vinyl cuprate
which would then be isomerized, by addition of TMSOTf,
to the TMS allenolate via release of the organocuprate.6
Presumably, the byproduct alkyl or aryl cuprate would then
be transformed into the corresponding dialkyl or aryl cuprate
by means of RMgBr addition and, in turn, complete the
catalytic cycle. Final electrophilic capture of the TMS
allenolate (after organocuprate consumption) by an aldehyde
should provide the desired aldol ester product.4 Herein, we
wish to report on the successful vicinal functionalization of
an R,â-acetylenic ester via a catalytic carbocupration/
Mukaiyama aldol reaction sequence.
As shown in Table 1, we initially observed that utilizing
5 mol % of CuI‚2LiCl for the initial carbocupration of 1a
with PhMgBr followed by the addition of PhCHO and
warming of the reaction to rt did indeed afford 2 with an
exceptional 20:1 dr for the (Z)-aldol product but with a low
yield of 13-20% over the two-step process (entries 1 and
2). Initially, we speculated that the 30 mol % excess of
TMSOTf was responsible (although we cannot rule out either
the Mg or Cu salts) for the activation of the aldehyde toward
nucleophilic addition of the intermediate TMS allenolate. In
our efforts to increase the overall yield, we allowed the
reaction to remain at -78 °C for 3 f 5 h and were provided
with an improved yield of 30%. Unfortunately, a longer
reaction time of 21 h (entry 5) did not improve the overall
yield but furnished a slightly inferior yield of 21%. Based
on these results, we surmised that the addition of a stoichio-
metric amount of Lewis acid would increase the electrophi-
licity of the aldehyde acceptor and ultimately improve the
overall yield via an enhanced capture of the nucleophilic
TMS allenolate. Much to our delight, the addition of TiCl4
or BF3‚OEt2 did advance the overall yield nearly 2-fold with
yields of 40 and 59%, respectively. As shown in entry 8,
increasing the catalyst loading to 30 mol % and alternatively
using TMSCl in place of TMSOTf furnished an overall lower
yield while maintaining a dr of 20:1 in favor of the Z product.
In addition, we also observed that using a lower equivalence
of TMSOTf (1.05 equiv) decreased the yield over the two-
step process from 59 to 32%. It should be noted that in all
cases the remaining material balance was a mixture of the
(Z)-R,â-unsaturated ester (arising from an incomplete aldol
reaction and proton quench of the allenolate) and starting
material.
Based on the optimized conditions in Table 1 (entry 7),
we decided to investigate the scope of the vicinal function-
alization of 1a with a series of aromatic Grignard reagents
and aldehydes. We initially observed that utilizing 4-fluo-
rophenyl MgBr for the carbocupration and benzaldehyde as
the electrophilic coupling partner provided product 3 with a
55% yield and an excellent dr of >20:1 for the Z product.
While pleased with the overall process, we decided to raise
the catalyst loading from 5 to 8 mol % and observed ∼10%
amplification in yield from 55% to 61%. Further increases
in the catalyst loading (>8%) unfortunately did not translate
into higher overall yields. With the new optimized conditions
in hand (8 mol % of CuI‚2LiCl and 1.3 equiv of TMSOTf),
we continued our investigation as previously indicated. As
delineated in Table 2, the vicinal functionalization of 1a with
o-tolyl MgBr and benzaldehyde proceeded to afford the
desired adduct 4 in good yield (58%) with an excellent dr
of 15:1 for the Z-aldol stereoisomer. The slight degradation
of dr from 20:1 Z/E for the aldol adduct 3 to 15:1 for
â-hydroxy-R-methylene ester 4 was attributed to the more
sterically hindered Grignard reagent. It is worth noting that
(5) (a) Nakamura, E.; Kuwajima, I. J. Am. Chem. Soc. 1984, 106, 3368.
(b) Nakamura, E.; Aoki, S.; Sekiya, K.; Oshino, H.; Kuwajima, I. J. Am.
Chem. Soc. 1987, 109, 8056. (c) Frantz, D. E.; Singleton, D. A. J. Am.
Chem. Soc. 2000, 122, 3288. (d) Nilsson, K.; Andersson, T.; Ullenius, C.
J. Organomet. Chem. 1997, 545-546, 591. (e) Nilsson, K.; Andersson, T.;
Ullenius, C.; Gerold, A.; Krause, N. Chem. Eur. J. 1998, 4, 2051. (f) Klein,
J.; Levene, R. J. Chem. Soc., Perkin Trans. 2 1973, 1971. (g) Marino, J.
P.; Linderman, R. J. J. Org. Chem. 1981, 46, 3696.
(6) (a) Mueller, A. J.; Jennings, M. P. Org. Lett. 2007, 9, 5327. (b)
Jennings, M. P.; Sawant, K. B. Eur. J. Org. Chem. 2004, 3201.
1650
Org. Lett., Vol. 10, No. 8, 2008

of 58% over the one-pot, two-step process. With the results
from Table 2 in hand, we sought to further investigate the
scope of the tandem catalytic carbocupration-Mukaiyama
aldol reaction of 1a with a variety of both aromatic and
aliphatic Grignard reagents and aldehydes.
Table 2. Vicinal Functionalization of 1a via a Catalytic
Carbocupration/Mukaiyama Aldol Reaction Sequence with
Various Aromatic Grignard Reagents and Aldehydesa,b
As reported in Table 3, the vicinal functionalization of 1a
Table 3. Vicinal Functionalization of 1a via a Catalytic
Carbocupration/Mukaiyama Aldol Reaction Sequence with
Various Grignard Reagents and Aldehydesa,b
1
a E/Z ratio determined by H NMR (360 or 500 MHz) from the crude
reaction mixture. b Yields are of the isolated, pure compounds.
when the electrophilic quench of the silyl allenolate derived
from the catalytic carbocupration of 1a with the o-tolyl
Grignard reagent was a proton, the dr was 11:1, which was
slightly lower than that of the Ph derived silyl allenolate (dr
of 13:1 for the Z product).6a We also observed a lower level
of dr when anisaldehyde was utilized as the electrophilic
quench of the TMS allenolate derived from the catalytic
carbocupration of 1a with PhMgBr. The aldol product 5 was
isolated as a 7:1 Z/E mixture with a combined yield of 53%.
Gratifyingly, the vicinal functionalization of 1a with
PhMgBr and the electron-poor 4-fluorobenzaldehyde pro-
vided the â-hydroxy-R-olefinic ester 6 in one of the best
yields in Table 2 of 63%, while re-establishing very high
levels of dr (>20:1 for the Z-product). Similarly, the catalytic
addition of the diphenyl cuprate to 1a followed by nucleo-
philic addition of the TMS allenolate to p-tolylaldehyde
afforded virtually an identical dr of >20:1 for the final
product 7 (compared with that of 6) in an acceptable yield
1
a E/Z ratio determined by H NMR (360 or 500 MHz) from the crude
reaction mixture. b Yields are of the isolated, pure compounds.
with the 4-fluorophenyl Grignard reagent readily provided
the TMS allenolate, and subsequent treatment with hexanal
and BF3‚OEt2 afforded the aldol adduct 8 in 62% yield with
an excellent dr of >20:1 for the Z-product. Similarly, initial
catalytic carbocupration of 1a with both o-tolylMgBr and
PhMgBr followed by an ensuing treatment of the allenolates
with hexanal afforded the â-hydroxy-R-olefinic esters 9 and
10 in practically identical yields of 54 and 55%, respectively,
while maintaining very high levels of dr (>20:1 for 9 and
10) for both of the Z-aldol adducts. Much to our delight,
exchanging from an aromatic Grignard reagent to an aliphatic
Org. Lett., Vol. 10, No. 8, 2008
1651

one had little effect on yield or diastereoselectivity. Thus,
catalytic carbocupration of 1a with hexylMgBr followed by
a subsequent quenching of the TMS allenolate with benzal-
dehyde led to the desired aldol adduct 11 in 54% yield with
an exceptional dr of 20:1, once again favoring the Z-product.
It is worth noting that when the electrophilic quench of the
silyl allenolate derived from the catalytic carbocupration of
1a with the hexyl Grignard reagent was a proton, the dr was
a trivial 2:1, which was considerably lower than that of the
aromatic-derived silyl allenolates (dr of 10 f 13:1 for the Z
product).6a Based on our observation, it appeared that the
nucleophilic addition of an aliphatic TMS allenolate is quite
dependent on the steric environment of the approaching
electrophile (aldehyde versus a proton). Last, vicinal func-
tionalization of 1a with hexylMgBr and hexanal selectively
furnished the â-hydroxy-R-olefinic ester 12 in an acceptable
yield of 48% coupled with an analogous dr of 20:1, as seen
with compounds 8-11.
Figure 2. Key NOE enhancements for 4, 13-E, and 13-Z.
chemistry of the remaining aldol adducts 2-3 and 5-12 were
assigned by analogy.
The relative stereochemical determination of 13 was more
challenging than that of 4 due to the initial difficulty in
assigning the vinyl methyl group versus that of the aromatic
1
methyl moiety. Much to our delight, the H-13C NMR
With the results from Tables 2 and 3 in hand, we decided
to investigate the vicinal functionalization of the â-substituted-
R,â-acetylenic ester 1b with the hope of affording the highly
stereodefined â-hydroxy-R- tetrasubstituted ester 13. Thus,
catalytic carbocupration of 1b with the o-tolyl Grignard
reagent under the standardized conditions of Tables 2 and 3
presumably furnished the tetra-substituted TMS allenolate.
Subsequent quenching of the nucleophilic allenolate with
BF3‚OEt2 coordinated benzaldehyde provided the aldol
adduct 13 in 61% overall yield, but with a disappointing dr
of 1.6:1 in favor the E-â-hydroxy-R-olefinic ester as
described in Scheme 1.
correlation spectroscopy (HMBC) furnished a four-bond
cross-correlation peak between the allylic methyl protons and
the carbonyl of the ester functional group. With the peak
assignments in hand, a strong NOE for both the methine aldol
and â-methyl protons provided unquestionable proof for the
assigned Z-olefin geometry. Likewise, a strong NOE between
the â-methyl and the methylene protons of the ethyl group
resident in the ester moiety provided convincing evidence
for the E-olefin isomer.
In conclusion, we have shown that vicinal functionalization
of propiolate esters via a tandem catalytic carbocupration-
Mukaiyama aldol reaction sequence with catalyst loadings
as low as 8 mol % readily allows for good yields and
excellent diastereoselectivities (>20:1) for a series of Grig-
nard reagents and aldehydes. Future directions of investiga-
tion will include further developments into the catalytic
vicinal functionalization of ynoates with other electrophiles
and chiral Lewis acid catalysts. Results from these studies
will be reported in due course.
Scheme 1
Acknowledgment. We thank The University of Alabama
for financial support of this work.
Supporting Information Available: The general reaction
procedure and NMR data (1H and 13C) for all of the products.
This material is available free of charge via the Internet at
http://pubs.acs.org.
As shown in Figure 2, the olefin geometry of 4 was
determined by means of H NMR. The strong 1D NOE for
both the methine allylic and â-vinylic protons provided
unquestionable proof for the assigned Z-olefin geometry. On
the basis of the NOE experiment results for 4, the stereo-
1
OL800373G
1652
Org. Lett., Vol. 10, No. 8, 2008