Catalytic enantioselective decarboxylative protonation

Article abstract of DOI:10.1021/ja063335a

We report a highly enantioselective, general catalytic system for the facile synthesis of tertiary stereocenters by protonation adjacent to cyclic ketones. The method relies on catalytic decarboxylative protonation of readily accessible racemic quaternary β-ketoesters. A range of substituted cycloalkanone compounds can be accessed through this process with high levels of enantioselectivity. Copyright

Full text of DOI:10.1021/ja063335a

Published on Web 08/15/2006
Catalytic Enantioselective Decarboxylative Protonation
Justin T. Mohr, Toyoki Nishimata, Douglas C. Behenna, and Brian M. Stoltz*
DiVision of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125
Received May 17, 2006; E-mail: stoltz@caltech.edu
Scheme 1
Enantioselective protonation represents a direct method for
generating tertiary carbon stereocenters from an achiral enolate or
an enol equivalent. Several distinct approaches toward effecting
this process have been reported, including the use of achiral enolates
with chiral Brønsted acids, chiral metal enolates with an achiral
proton source, and the combination of chiral enolates and chiral
Brønsted acids.1 Of the existing methods, most are limited in scope,
few are catalytic, and together they have not provided a general
solution to this deceptively simple goal.1d Herein, we report a highly
enantioselective, general catalytic system for the facile synthesis
of tertiary stereocenters by protonation adjacent to ketones.
Table 1. Optimization of Reaction Conditionsa
Recently, we disclosed a series of catalytic enantioselective
allylation reactions that deliver cyclic ketones bearing all-carbon
quaternary stereocenters at the R-position with high efficiency and
enantioselectivity.2,3 Crucial to the success of these transformations
was the use of catalysts derived from Pd(0) and a chiral phosphi-
nooxazoline (PHOX) supporting ligand (e.g., 1).4 Included in this
effort was our exploration of racemic allyl â-ketoesters as substrates
for a novel stereoablative enantioconvergent decarboxylative ally-
lation reaction (e.g., (()-2 f 4, Scheme 1).3 We believe that in
the course of this reaction a chiral Pd-enolate (3) likely is generated
in solution and that the high degree of organization about the
palladium center is responsible for the levels of enolate facial
selectivity observed in the alkylation. Interestingly, if this were the
case, the catalytic generation and utilization of such chiral enolate
complexes by this method would have the potential to be more
broadly applicable than we previously recognized. Thus, in an effort
to exploit this valuable chiral synthon, we chose to intercept this
intermediate with an alternative electrophile, namely, a proton.5,6
In our first attempt to achieve an enantioselective protonation,
racemic â-ketoester (()-2 was exposed to Pd(OAc)2 in the presence
of (S)-t-Bu-PHOX (1) with triethylamine and HCO2H, resulting in
smooth conversion to 2-methyl-1-tetralone with low, but measur-
able, enantiomeric excess (entry 1, Table 1).5 Although necessary
in a nonenantioselective version of the protonation reported by
Tsuji,5 removal of the amine base in our asymmetric system led to
improved, though still quite modest, enantiomeric excess (entry 2).
To sequester the small amount of water present in commercially
available formic acid, we added 3 Å molecular sieves (MS) to the
reaction mixture. Gratifyingly, we found that this additive provided
5 in dramatically increased enantiopurity (entry 3). A screen of
alternative Pd sources and solvents revealed the superiority of
Pd(OAc)2 to other Pd precursors and p-dioxane as the preferred
solvent (entries 3-6).7 Investigation of other potential drying agents
indicated their inferiority relative to MS, and, specifically, 4 Å MS
provided the highest enantiomeric excess of the observed products
(entries 6-8).7 At this point, we also examined the behavior of
other chiral ligands in this reaction. As in our earlier studies,2 we
found chelating P-N ligands to be the most effective, while
bisphosphine-type ligands provided only trace asymmetric induction
(entries 9-14). Finally, we found that when the solvent was freshly
distilled and the 4 Å MS were rigorously flame-dried immediately
entry
Pd source
ligand
additiveb
solvent
eec (%)
1
2
3
4
5
6
7
8
9
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
[Pd(allyl)Cl]2
Pd2(dba)3
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
1
1
1
1
1
1
1
1
Et3N (1 equiv) THF
none
7
24
72
41
49
79
88
85
-1
3
THF
THF
THF
THF
p-dioxane
p-dioxane
p-dioxane
p-dioxane
p-dioxane
p-dioxane
3 Å MS
3 Å MS
3 Å MS
3 Å MS
4 Å MS
5 Å MS
4 Å MS
4 Å MS
4 Å MS
(R)-BINAP
(R,R)-DIOP
(R,R)-Trost
ligand
(S)-QUINAP
(R)-Ph-PHOX 4 Å MS
(S)-i-Pr-PHOX 4 Å MS
10 Pd(OAc)2
11 Pd(OAc)2
3d
12 Pd(OAc)2
13 Pd(OAc)2
14 Pd(OAc)2
15e Pd(OAc)2
4 Å MS
p-dioxane -20
p-dioxane -65
p-dioxane
p-dioxane
87
92
1
4 Å MS
a Reactions performed with 0.1 mmol of (()-2 at 0.033 M in solvent
and run to complete consumption of (()-2. b Where indicated, 90 mg of
MS was used. c Determined by chiral HPLC. d Measured after 72 h at
approximately 60% conversion. e p-Dioxane was freshly distilled over Na
metal and the 4 Å MS were flame dried under vacuum prior to use. Reaction
performed with 0.2 mmol of (()-2 and 180 mg of 4 Å MS.
prior to use, the enantiomeric purity was further enhanced, providing
5 in 92% ee (entry 15).
Attempts to optimize the amount of HCO2H and 4 Å MS in the
reaction indicated that the balance of these two components was
intimately related to both enantio- and chemoselectivity (i.e., the
ratio of protonated to allylated products 5/4).7 In general, an excess
of HCO2H led to decreased enantioselectivity, while smaller
quantities afforded greater amounts of allylated product 4. Alter-
natively, small amounts of 4 Å MS produced 5 in decreased ee,
while large quantities led to increased allylation. For this particular
substrate, the optimal amount of HCO2H was found to be 6.0 equiv
with a 4 Å MS quantity of 1.80 g/mmol substrate. Under these
conditions, (S)-(-)-2-methyl-1-tetralone (5) was produced in 88%
yield and 94% ee with no observable allylation (Table 2, entry 1).8,9
Encouraged by these studies, we sought to explore the generality
and scope of this enantioselective reaction (Table 2). A variety of
9
11348
J. AM. CHEM. SOC. 2006, 128, 11348-11349
10.1021/ja063335a CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Enantioconvergent Decarboxylative Protonations
asymmetric alkylation methodology that delivers quaternary stereo-
centers from the same starting materials via catalytic enantiose-
lective allylation. Additional explorations of the scope, mechanism,
and applications of these technologies are currently underway.10
Acknowledgment. The authors thank Eli Lilly (predoctoral
fellowship to J.T.M.), Sankyo Co., Ltd. (financial support of T.N.),
The Fannie and John Hertz Foundation (predoctoral fellowship to
D.C.B.), Merck, Amgen, Johnson & Johnson, Bristol-Myers Squibb,
and the Dreyfus Foundation for generous funding.
Supporting Information Available: Experimental details. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) For reviews of enantioselective protonation, see: (a) Duhamel, L.;
Duhamel, P.; Plaquevent, J.-C. Tetrahedron: Asymmetry 2004, 15, 3653-
3691. (b) Yanagisawa, A.; Yamamoto, H. In ComprehensiVe Asymmetric
Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New
York, 1999; Vol. 3, pp 1295-1306. (c) Yanagisawa, A.; Yamamoto, H.
In ComprehensiVe Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer: New York, 2004; Supplement 2, pp 125-
132. (d) Fehr, C. Angew. Chem., Int. Ed. Engl. 1996, 35, 2566-2587.
(2) Behenna, D. C.; Stoltz, B. M. J. Am. Chem. Soc. 2004, 126, 15044-
15045.
a Isolated yield from the reaction of 0.3 mmol of substrate at 0.033 M
in p-dioxane with 10 mol % Pd(OAc)2, 12.5 mol % (S)-t-Bu-PHOX, 5-8
equiv HCO2H, and 405-810 mg of 4 Å MS at 40 °C (ref 7). b Determined
by chiral HPLC or GC; where noted, the absolute configuration was
determined by comparing the sign of optical rotation to literature values
(ref 7). c Reaction performed with 5 mol % Pd(OAc)2 and 6.25 mol %
(S)-t-Bu-PHOX. d GC yield using tridecane as internal standard. e Reaction
performed at 35 °C.
(3) Mohr, J. T.; Behenna, D. C.; Harned, A. M.; Stoltz, B. M. Angew. Chem.,
Int. Ed. 2005, 44, 6924-6927.
(4) For the development of phosphinooxazoline ligands, see: (a) Helmchen,
G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336-345. (b) Williams, J. M. J.
Synlett 1996, 705-710 and references therein.
substitutions is tolerated at the ketone R-position (entries 1-3) and
various positions about the aromatic ring (entries 4-8) of 1-tetralone
derivatives. Enantioenriched (S)-(+)-2-methyl-1-indanone can also
be produced from the corresponding â-ketoester (entry 9). Ad-
ditionally, monocyclic compounds (entries 10-13) and a hetero-
cycle (entry 14) were easily accessed under similar reaction
conditions. The absolute configuration of a number of products was
established by a comparison of the observed sign of optical rotation
to literature values (entries 1-3, 9, 11, and 12).7 Interestingly, fused
aromatic substrates (i.e., tetralones and indanones) lead to products
in the opposite enantiomeric series compared to that of the
cyclohexanone cases (cf. entries 1-3 and 9 to entries 11 and 12).
These results are in contrast to the consistent enantiofacial selec-
tivity observed across multiple substrate types in our asymmetric
allylation chemistry and suggest stark differences in their corre-
sponding mechanisms.2,3
(5) A nonenantioselective decarboxylative protonation has been reported by
Tsuji, see: (a) Tsuji, J.; Nisar, M.; Shimizu, I. J. Org. Chem. 1985, 50,
3416-3417. (b) Mandai, T.; Imaji, M.; Takada, H.; Kawata, M.; Nokami,
J.; Tsuji, J. J. Org. Chem. 1989, 54, 5395-5397.
(6) To our knowledge, there are three other groups that have reported related
Pd-catalyzed systems that produce similar enantioenriched products, see:
(a) He´nin, F.; Muzart, J. Tetrahedron: Asymmetry 1992, 3, 1161-1164.
(b) Aboulhoda, S. J.; He´nin, F.; Muzart, J.; Thorey, C.; Behnen, W.;
Martens, J.; Mehler, T. Tetrahedron: Asymmetry 1994, 5, 1321-1326.
(c) Aboulhoda, S. J.; Le´tinois, S.; Wilken, J.; Reiners, I.; He´nin, F.;
Martens, J.; Muzart, J. Tetrahedron: Asymmetry 1995, 6, 1865-1868.
(d) Baur, M. A.; Riahi, A.; He´nin, F.; Muzart, J. Tetrahedron: Asymmetry
2003, 14, 2755-2761. (e) Sugiura, M.; Nakai, T. Angew. Chem., Int. Ed.
Engl. 1997, 36, 2366-2368. (f) Hamashima, Y.; Somei, H.; Shimura,
Y.; Tamura, T.; Sodeoka, M. Org. Lett. 2004, 6, 1861-1864.
(7) See the Supporting Information for details.
(8) Despite extensive experimentation to verify the origin of the proton
observed in the products (i.e., 5), we have maximally observed 35%
D-incorporation when using HCO2D and rigorously dried 4 Å MS. By
contrast, under otherwise identical conditions, <1% D-incorporation was
observed when DCO2H was employed.7 The detailed mechanism of proton
incorporation (e.g., proton transfer, reductive elimination, or otherwise)
remains unclear and is under investigation.
(9) We were interested in whether the other enolate precursors we have
employed for enantioselective allylation chemistry would be competent
substrates for the protonation reaction. To investigate this possibility, allyl
enol carbonate i was subjected to our optimized reaction conditions for
the formation of 5. Contrasting the result when (()-2 was used (Table 2,
entry 1), in this case, 5 was produced in 74% ee with a 66/44 ratio of 5/4
on a 0.1 mmol scale (100% conversion). When enol silane ii was used,
low conversion (<5%) was observed, however, the ee of isolated 5 was
84%. While these results highlight the advantage of â-ketoester precursors
to the reactive enolate intermediate, it is uncertain why the reactivity and
selectivity of these substrates is so different. The mechanism of this process
is currently under investigation.
In conclusion, a novel system for the enantioconvergent decar-
boxylative protonation of racemic â-ketoesters has been developed.
The reaction tolerates a variety of substitution and functionality
and delivers products of high enantiopurity in excellent yield. The
enantioinduction in the observed protonated products is consistent
with the intermediacy of an enolate that is intimately associated to
the chiral Pd complex. This, in turn, substantiates our initial
hypothesis concerning the nature of the reactive intermediate 3 and
opens the door to further applications. The process capitalizes on
the availability and unique reactivity of racemic R-substituted allyl-
â-ketoesters, which are employed directly in the catalytic enantio-
selective process and deliver valuable tertiary-substituted products
in highly enantioenriched form. In general, the overall process (sub-
strate synthesis and use) represents a catalytic enantioselective vari-
ant of classic alkylation/decarboxylation sequences (e.g., acetoacetic
ester synthesis, cf. eqs 1 and 2). Furthermore, the asymmetric proto-
nation described here serves to complement our recently developed
(10) We recently completed the first asymmetric synthesis of (+)-dichroanone
using our catalytic enantioselective Tsuji allylation, see: McFadden, R.
M.; Stoltz, B. M. J. Am. Chem. Soc. 2006, 128, 7738-7739.
JA063335A
9
J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006 11349