Cleavage of carbon-carbon bonds through the mild release of trifluoroacetate: Generation of α,α-difluoroenolates for aldol reactions

Article abstract of DOI:10.1021/ja202213f

The selective cleavage of carbon-carbon bonds is a significant challenge in synthetic chemistry, yet this strategy can be a powerful way to generate reactive intermediates. We have discovered that, through the facile release of trifluoroacetate which occurs by C-C bond scission, difluoroenolates can be generated under very mild reaction conditions. Unlike existing reactions, this method is not limited to a small group of fluorinated building blocks. We have applied this process to the aldol reaction to install difluoromethylene groups.

Full text of DOI:10.1021/ja202213f

COMMUNICATION
pubs.acs.org/JACS
Cleavage of CarbonÀCarbon Bonds through the Mild Release of
Trifluoroacetate: Generation of r,r-Difluoroenolates for Aldol
Reactions
Changho Han,† Eun Hoo Kim,† and David A. Colby*,†,‡
†Department of Medicinal Chemistry and Molecular Pharmacology and ‡Department of Chemistry, Purdue University, West Lafayette,
Indiana 47907, United States
S Supporting Information
b
Scheme 1
ABSTRACT: The selective cleavage of carbonÀcarbon
bonds is a significant challenge in synthetic chemistry, yet
this strategy can be a powerful way to generate reactive
intermediates. We have discovered that, through the facile
release of trifluoroacetate which occurs by CÀC bond
scission, difluoroenolates can be generated under very mild
reaction conditions. Unlike existing reactions, this method is
not limited to a small group of fluorinated building blocks.
We have applied this process to the aldol reaction to install
difluoromethylene groups.
blocks (e.g., ethyl bromodifluoroacetate) and/or harsh reac-
tion conditions, which further limit the scope of the substrates
for the process.12 Herein, we describe the application of the
he rapid and efficient synthesis of complex structures has
trifluoroacetate-release strategy to cleave a CÀC bond and
been a clear goal of synthetic organic and medicinal chem-
T
generate an R,R-difluoroenolate. This method uses exceed-
ingly mild reaction conditions, is not limited to a small group
of fluorinated building blocks, and is compatible with many
substrates.
istry, and accordingly, many processes to form new chemical
bonds have been developed. However, the inverse process of
selectively cleaving bonds poses a significant synthetic challenge.
In the case of cleaving a CÀC bond, very few procedures are
available.1À4 One of the most useful of these methods is
decarboxylation, because it is a powerful, yet mild process for
generating a reactive intermediates.4À8 Recent reports by
Shibasaki,5 Trost,6 Stoltz,7 Tunge,8 and others9 have demon-
strated the power of decarboxylation. Despite the tremendous
synthetic utility of these bond-cleaving processes, they are
powerful, underdeveloped methods for generating reactive pre-
cursors by releasing labile groups under mild reaction conditions.
Trifluoroacetate is a ubiquitous counterion in many reactions,
and we hypothesized that a release of trifluoroacetate would be
analogous to decarboxylation; however, a gem-diol of a triflu-
oromethyl ketone would be the requisite precursor. Indeed, a
mild and rapid CÀC bond fragmentation of hexafluoroacetone
hydrate has been observed after the addition of 2 equiv of base;
trifluoroacetate is released from this hydrate fragmentation
process (Scheme 1).10 We are unaware of any synthetic methods
that exploit this potentially powerful fragmentation.11 Thus, we
envisioned that this trifluoroacetate-release strategy could be
readily adapted to generate R,R-difluoroenolates by replacing
one fluorine atom on hexafluoroacetone hydrate with a carbonyl
group to stabilize the anion as an enolate (Scheme 1). The
generation of R,R-difluoroenolates has attracted significant
attention,12 because it is a reliable method to install difluoro-
methylene groups.13 The usual synthetic protocols to prepare R,
R-difluoroenolates rely on the few available fluorinated building
To implement this strategy, our initial studies were designed
to find a versatile route to gem-diols with R-fluorination to test
the trifluoroacetate-release strategy. We found that the requisite
substrates can be easily prepared under neutral conditions using
Selectfluor and 1,1,1-trifluoro-2,4-diones 1À9 (Table 1).14 Some
1,1,1-trifluoro-2,4-diones are commercially available, and others
are readily accessed by trifluoroacetylation of methyl ketones.15
The 1,1,1-trifluoro-2,4-diones 1À9 exist exclusively in the enol
form;16 therefore this transformation is markedly efficient and is
devoid of byproduct from partial fluorination.17 Additionally,
silica chromatography was not necessary for the purification of
products 10À12, 14, and 15. The crystalline compound 14
provided further structural data following X-ray analysis, and the
data supported the assignment of a gem-diol rather than a ketone
with a molecule of water.18
With substrates 10À18 in hand, trifluoroacetate release was
found to be promoted rapidly using LiBr and Et3N. Specifically,
treatment of substrate 10 with LiBr/Et3N at room temperature
produced almost instantaneously the difluorinated compound 19
as well as the self-adduct 20 (eq 1). This example provided
evidence that trifluoroacetate release occurs rapidly and that the
difluoroenolate can react with electrophiles. Also, this selective
Received: March 10, 2011
Published: March 28, 2011
r
2011 American Chemical Society
5802
dx.doi.org/10.1021/ja202213f J. Am. Chem. Soc. 2011, 133, 5802–5805
|

Journal of the American Chemical Society
COMMUNICATION
Table 1. Syntheses of Highly r-Fluorinated gem-Diolsa
Table 2. Aldol Reactions with r,r-Difluoroenolates from
Trifluoroacetate Releasea
a Unless otherwise noted, reactions were performed with 2.5 equiv of
c
Selectfluor. b Isolated yields. Two-step yield from the parent methyl
ketone.
CÀC bond cleavage to release trifluoroacetate is a striking
contrast to previous data.19
Next, we explored aldol reactions with substrates 10À18 and
aldehydes 21À31 using the trifluoroacetate-release strategy
(Table 2). The significance of developing new mild enolization
protocols for aldol reactions has been recently demonstrated by
Coltart and co-workers.20 For trifluoroacetate release, the scope
was equally impressive for the highly fluorinated substrates
10À18, because each readily participated in the process. Also,
trifluoroacetate release is extremely fast, and the reactions were
typically completed after 3 min at room temperature or 30 min at
À78 °C. Good to excellent yields (64À95%) of the aldol
adduct(s) were routinely isolated. Both aryl and alkyl aldehydes,
as well as the R,β-unsaturated aldehyde 22 and the sensitive
dienal 23, were fully compatible with this process. Other sensitive
aldehydes also smoothly reacted with the R,R-difluoroenolates;
β-elimination was not observed with 25, and little R-epimeriza-
tion (i.e., 5%) was observed with 31 after 19F NMR analysis of the
crude reaction mixture. Also, the reaction with 31 displayed the
unique tendency of product 47 to form a hemiacetal with an
a Reactions were typically performed with 3 equiv of LiBr and 1 equiv of
Et3N; see Supporting Information for details. b Isolated yields and diaster-
eomeric ratios (dr) based on 19F NMR analysis of unpurified reaction
products. c Absolute stereochemistry determined after desilylation and
cyclization. d Absolute stereochemistry determined after MPA analysis.
e Ratio of diastereomers is 45:45:46:46. f Yield determined after acid-
promoted hydrolysis of hemiacetal; see Supporting Information for details.
g Diastereomeric ratio is (R,R)-isomer/(S,R)-isomer/(S,S)-isomer.
additional aldehyde (this side product was readily hydrolyzed
with aqueous acid).18 In the case of the highly sensitive
5803
dx.doi.org/10.1021/ja202213f |J. Am. Chem. Soc. 2011, 133, 5802–5805

Journal of the American Chemical Society
COMMUNICATION
cis-myrtenol-derived substrate 17 (the trans-myrtenol derivative
18 is not susceptible to R-epimerization but provided adduct 46
to contrast with 45),21 15% of the product mixture was the trans-
isomer. Even though R-deprotonation was observed with ketone
17, the two fluorines adjacent to the ketone significantly enhance
the irreversible tendency for the cis-myrtenol to convert to the
trans-myrtenol. Therefore, our conditions were compatible with
this highly sensitive substrate. This strategy is quite chemoselec-
tive for aldehydes, because the keto-aldehyde 28 provided only
the product 43 from aldehyde addition. Also, all of the products
32À47 have a ketone that is additionally activated by two
R-fluorines, yet these aldol-adducts do not out-compete the
aldehydes. Further evidence of the mild reaction conditions is
available from the triene 39 that was isolated in good yield (78%)
from the unsaturated starting material 13. Assignment of the
absolute stereochemical configuration of 47 using the traditional
Mosher method (i.e., with MTPA esters) was not clear. Existing
literature precedents suggest such an analysis on some fluori-
nated substrates is not straightforward.22 However, using the
recent comparisons of Hoye and co-workers,23 preparation of
the MPA-derivatives of 47 allowed assignment of the absolute
configuration.18
To probe the mechanism of the trifluoroacetate-release strat-
egy, we performed a series of control reactions to verify that the
aldol reaction is not a stepwise process in which trifluoroacetate
release occurs, followed by deprotonation of the R,R-difluor-
oketone, and addition to the aldehyde. Indeed, the R,R-difluor-
oketone 19 did not react with aldehyde 21 using the trifluoroacetate-
release conditions (LiBr/Et3N) or other combinations of reagents
(eq 2). Also, we tested if a retro-aldol process occurs under the
reaction conditions and the aldol adduct 32 was recovered
without the presence of any products from a retro-aldol reaction
with benzaldehyde (eq 3). These mechanistic data further high-
light the powerful, yet mild process of trifluoroacetate release to
generate reactive intermediates.
solution to the existing synthetic limitations of preparing R,R-
difluoroenolates. We demonstrated the broad scope of substrates
and aldehydes that participate in this mild process and provided
key mechanistic details. Studies to apply our trifluoroacetate-
release strategy to generate reactive intermediates for other
processes are currently underway as well as additional mechan-
istic studies.
’ ASSOCIATED CONTENT
S
Supporting Information. Full experimental details, spec-
b
troscopic data, and X-ray data. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
dcolby@purdue.edu
’ ACKNOWLEDGMENT
We are grateful to Purdue University for funding. We thank
Huaping Mo for assistance in 19F NMR analysis and Karl V.
Wood for assistance in HRMS acquisition and analysis. We
acknowledge Phillip E. Fanwick and the X-ray Crystallography
Center at Purdue University.
’ REFERENCES
(1) Payette, J. N.; Yamamoto, H. J. Am. Chem. Soc. 2008, 130,
12276–12278.
(2) Draghici, C.; Brewer, M. J. Am. Chem. Soc. 2008, 130,
3766–3767.
(3) Kamijo, S.; Dudley, G. B. J. Am. Chem. Soc. 2005, 127,
5028–5029.
(4) Pohlhaus, P. D.; Bowman, R. K.; Johnson, J. S. J. Am. Chem. Soc.
2004, 126, 2294–2295.
(5) Yin, L.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2009,
131, 9610–9611.
(6) Trost, B. M.; Xu, J.; Schmidt, T. J. Am. Chem. Soc. 2008,
130, 11852–11853.
(7) Keith, J. A.; Behenna, D. C.; Mohr, J. T.; Ma, S.; Marinescu, S. C.;
Oxgaard, J.; Stoltz, B. M.; Goddard, W. A., III. J. Am. Chem. Soc. 2007,
129, 11876–11877.
(8) Waetzig, S. R.; Tunge, J. A. J. Am. Chem. Soc. 2007,
129, 14860–14861.
As a demonstration of the potential broad compatibility of this
trifluoroacatete-release process to other substrates, we mono-
brominated24 the trifluoroacetylated ketone 1, stirred over water,
and then immediately treated the crude reaction mixture with
LiBr and Et3N in the presence of aldehyde 30 (eq 4). The
reaction was completed in 10 min at room temperature, and the
Darzens product 48 was isolated in good 65% yield (from 1) as a
single diastereomer.25 The implication of this finding is that this
process is not limited to the generation of R,R-difluoroenolates
and the production of R,R-difluoroketones but presents addi-
tional synthetic opportunities.
(9) Goossen, L. J.; Deng, G.; Levy, L. M. Science 2006, 313, 662–664.
(10) Prager, J. H.; Ogden, P. H. J. Org. Chem. 1968, 33, 2100–2102.
(11) Reports of trifluoroacetate release are very rare in the literature
(for an example, see: Fioravanti, S.; Pellacani, L.; Ramadori, F.; Tardella,
P. A. Tetrahedron Lett. 2007, 48, 7821–7824 and references therein).
(12) (a) Takikawa, G.; Toma, K.; Uneyama, K. Tetrahedron Lett.
2006, 47, 6509–6511. (b) Kloetzing, R. J.; Thaler, T.; Knochel, P. Org.
Lett. 2006, 8, 1125–1128. (c) Yang, Y.-Y.; Meng, W.-D.; Qing, F.-L. Org.
Lett. 2004, 6, 4257–4259. (d) Amii, H.; Kobayashi, T.; Terasawa, H.;
Uneyama, K. Org. Lett. 2001, 3, 3103–3105. (e) DeBoos, G. A.; Full-
brook, J. J.; Percy, J. M. Org. Lett. 2001, 3, 2859–2861. (f) Weigel, J. A.
J. Org. Chem. 1997, 62, 6108–6109. (g) Howarth, J. A.; Owton, W. M.;
Percy, J. M. J. Chem. Soc., Chem. Commun. 1995, 757–758.
(13) (a) Ni, C.; Liu, J.; Zhang, L.; Hu, J. Angew. Chem., Int. Ed. 2007,
46, 786–789. (b) Prakash, G. K. S.; Hu, J.; Mathew, T.; Olah, G. A.
Angew. Chem., Int. Ed. 2003, 42, 5216–5219.
In summary, we have discovered a novel trifluoroacetate-
release strategy to generate reactive intermediates using mild
conditions to cleave a CÀC bond and concurrently provided a
(14) For the difluorination of substrate 1 with Selectfluor, see:
Baumann, M.; Baxendale, I. R.; Martin, L. J.; Ley, S. V. Tetrahedron
2009, 65, 6611–6625.
5804
dx.doi.org/10.1021/ja202213f |J. Am. Chem. Soc. 2011, 133, 5802–5805

Journal of the American Chemical Society
COMMUNICATION
(15) Danheiser, R. L.; Miller, R. F.; Brisbois, R. G.; Park, S. Z. J. Org.
Chem. 1990, 55, 1959–1964.
(16) Sloop, J. C.; Bumgardner, C. L.; Washington, G.; Loehle, W. D.;
Sankar, S. S.; Lewis, A. B. J. Fluorine Chem. 2006, 127, 780–786.
(17) No byproduct from partial fluorination was observed. If 1 equiv
of Selectfluor is used, only starting material and the product with two
additional fluorines are isolated.
(18) See Supporting Information.
(19) Olivella, S.; Sole, A.; Jimenez, O.; Bosch, M. P.; Guerrero, A.
J. Am. Chem. Soc. 2005, 127, 2620–2627.
(20) Sauer, S. J.; Garnsey, M. R.; Coltart, D. M. J. Am. Chem. Soc.
2010, 132, 13997–13999.
(21) Peterson, P. E.; Grant, G. J. Org. Chem. 1991, 56, 16–20.
(22) (a) Seco, J. M.; Quinoa, E.; Riguera, R. Chem. Rev. 2004,
104, 17–117. (b) Takagi, Y.; Nakatani, T.; Itoh, T.; Oshiki, T. Tetra-
hedron Lett. 2000, 41, 7889–7892.
(23) Hoye, T. R.; Erickson, S. E.; Erickson-Birkedahl, S. L.; Hale,
C. R. H.; Izgu, E. C.; Mayer, M. J.; Notz, P. K.; Renner, M. K. Org. Lett.
2010, 12, 1768–1771.
(24) Le, Q. T. H.; Umetani, S.; Suzuki, M.; Matsui, M. J. Chem. Soc.,
Dalton Trans. 1997, 643–647.
(25) High diastereoselectivity is observed during the addition of
R-bromoenolates to aldehydes: Barluenga, J.; Baragana, B.; Concellon,
J. M.; Pinera-Nicolas, A.; Diaz, M. R.; Garcia-Granda, S. J. Org. Chem.
1999, 64, 5048–5052.
5805
dx.doi.org/10.1021/ja202213f |J. Am. Chem. Soc. 2011, 133, 5802–5805