Amidinate salt of hexafluoroacetone hydrate for the preparation of fluorinated compounds by the release of trifluoroacetate

Article abstract of DOI:10.1021/ol303291x

A powerful, new reagent, an amidinate salt of hexafluoroacetone hydrate, is an air-stable salt that can be used for the preparation of fluorinated organic molecules. Nucleophilic trifluoromethylation reactions are demonstrated following the base-promoted release of trifluoroacetate. This reagent is soluble in many polar organic solvents and produces fluoroform, following the release of trifluoroacetate. Reactions with this reagent and common electrophiles provide excellent yields of trifluoromethylated products.

Full text of DOI:10.1021/ol303291x

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Amidinate Salt of Hexafluoroacetone
Hydrate for the Preparation of Fluorinated
Compounds by the Release of
Trifluoroacetate
Mark V. Riofski,† Allison D. Hart,† and David A. Colby*,†,‡
Department of Medicinal Chemistry and Molecular Pharmacology and Department of
Chemistry, Purdue University, West Lafayette, Indiana 47907, United States
dcolby@purdue.edu
Received November 30, 2012
ABSTRACT
A powerful, new reagent, an amidinate salt of hexafluoroacetone hydrate, is an air-stable salt that can be used for the preparation of fluorinated
organic molecules. Nucleophilic trifluoromethylation reactions are demonstrated following the base-promoted release of trifluoroacetate. This
reagent is soluble in many polar organic solvents and produces fluoroform, following the release of trifluoroacetate. Reactions with this reagent
and common electrophiles provide excellent yields of trifluoromethylated products.
The incorporation offluorine into organic compoundsis
a powerful strategy to attenuate the biological properties
of candidate molecules in the pharmaceutical industry.1,2
Accordingly, there has been a substantial effort to develop
novel reagents and strategies that can install a trifluoro-
methyl group using nucleophilic,3ꢀ5 electrophilic,6ꢀ8 radical,9
and metal-mediated approaches.10ꢀ12 Among these meth-
ods, nucleophilic trifluoromethylation has been exten-
sively studied, but the incorporation of a CF3 group
remains limited by the high instability of the CF3 anion
by R-elimination to the difluorocarbene.4 The most com-
mon trifluoromethylating reagent, the Ruppert-Prakash
reagent (TMSCF3) stabilizes the CF3 anion through a
SiꢀCF3 bond (Figure 1).3,5 Although Prakash et al.
additionally developed phenyl trifluoromethyl sulfone
to accomplish nucleophilic trifluoromethylations,13,14
fluoroform also is a major option for installing trifluo-
romethyl groups by nucleophilic addition or metal-
mediated coupling.11 The primary drawback that hinders
its widespread use in many synthetic laboratories is that
fluoroform is a gas. Therefore, a reagent that could
† Department of Medicinal Chemistry and Molecular Pharmacology.
‡ Department of Chemistry.
(1) Hagmann, W. K. J. Med. Chem. 2008, 51, 4359–4369.
€
(2) Muller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881–1886.
(3) Kawai, H.; Kusuda, A.; Nakamura, S.; Shiro, M.; Shibata, N.
Angew. Chem., Int. Ed. 2009, 48, 6324–6327.
(4) Joubert, J.; Roussel, S.; Christophe, C.; Billard, T.; Langlois,
B. R.; Vidal, T. Angew. Chem., Int. Ed. 2003, 42, 3133–3136.
(5) Prakash, G. K. S.; Mandal, M. J. Am. Chem. Soc. 2002, 124,
6538–6539.
(6) Matsnev, A.; Noritake, S.; Nomura, Y.; Tokunaga, E.; Nakamura,
S.; Shibata, N. Angew. Chem., Int. Ed. 2010, 49, 572–576.
(7) Koller, R.; Stanek, K.; Stolz, D.; Aardoom, R.; Niedermann, K.;
Togni, A. Angew. Chem., Int. Ed. 2009, 48, 4332–4336.
(8) Magnier, E.; Blazejewski, J.-C.; Tordeux, M.; Wakselman, C.
Angew. Chem., Int. Ed. 2006, 45, 1279–1282.
(11) Zanardi, A.; Novikov, M. A.; Martin, E.; Benet-Buchholz, J.;
Grushin, V. V. J. Am. Chem. Soc. 2011, 133, 20901–20913.
(12) Cho, E. J.; Senecal, T. D.; Kinzel, T.; Zhang, Y.; Watson, D. A.;
Buchwald, S. L. Science 2010, 328, 1679–1681.
(13) Prakash, G. K. S.; Wang, Y.; Mogi, R.; Hu, J.; Mathew, T.;
Olah, G. A. Org. Lett. 2010, 12, 2932–2935.
(9) Ji, Y.; Brueckl, T.; Baxter, R. D.; Fujiwara, Y.; Seiple, I. B.; Su, S.;
Blackmond, D. G.; Baran, P. S. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
14411–14415.
(10) Morimoto, H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F.
Angew. Chem., Int. Ed. 2011, 50, 3793–3798.
(14) Prakash, G. K. S.; Hu, J.; Olah, G. A. Org. Lett. 2003, 5, 3253–
3256.
r
10.1021/ol303291x
XXXX American Chemical Society

release fluoroform into a reaction would be highly ad-
vantageous. Herein, we describe a novel amidinate salt of
hexafluoroacetone hydrate that releases fluoroform fol-
lowing the release of trifluoroacetate and demonstrate its
application in executing nucleophilic trifluoromethyla-
tions for the synthesis of fluorinated organic molecules.
salt 1 upon addition of DBU (Figure 2). Other common
organic bases participate in the process, but the DBU-
derived salt provided the optimum physical properties (see
below).21 The hexafluoroacetone hydrate amidinate salt 1
is a stable, anhydrous solid that can be freely weighed in
air. NMR studies, combustion analysis, and X-ray crystal-
lography verified that the salt 1 does not contain water,
unlike the parent hexafluoroacetone trihydrate.22 Addi-
tionally, this salt 1 is not hygroscopic, even after multiple
exposures to air for 3 months. It displays high solubility in
DMF, DMSO, EtOAc, and CH3CN and partial solubility
in toluene and THF. The salt 1 readily participates in base-
promoted trifluoroacetate release to generate fluoroform
(Figure 3). Specifically, the fragmentation of 1 in the pre-
sence of heat and DBU (1 equiv) were recorded by
19F NMR, and complete conversion of salt 1 into trifluoro-
acetate and fluoroform was observed.
Figure 1. Common nucleophilic trifluoromethylation reagents.
Recently, our laboratory has described the synthetic
utility of releasing trifluoroacetate to generate reactive
intermediates.15ꢀ17 These studies were based on a report
in 1968 from Prager and Ogden that hexafluoroacetone
trihydrate fragments in the presence of basic hydroxide
into trifluoroacetate and fluoroform (Scheme 1).18 We
hypothesized that this process could be exploited for
trifluoromethylations if the CF3 anion was produced in
an anhydrous environment to prevent immediate conver-
sion to fluoroform by the presence of water. Accordingly,
the first two goals were to determine an alternative to the
hydroxide bases and to generate an anhydrous form of
hexafluoroacetone hydrate. Our initial experiments deter-
mined that hexafluoroacetone trihydrate fragments in
polar organic solvents, such as DMF, DMSO, and
CH3CN, and that metal hydroxides can be replaced by
organic bases, such as Et3N and DBU. Unfortunately, the
CF3 anion cannot be trapped by an appropriate electro-
phile, such as an aldehyde, due to the immediate conver-
sion to fluoroform in the 19F NMR (data not shown) by
the presence of water. Therefore, the need for an anhy-
drous form of hexafluoroacetone hydrate was paramount.
Figure 2. Invention of a new trifluoromethylation reagent,
hexafluoroacetone hydrate amidinate salt 1.
Scheme 1. Fragmentation of Hexafluoroacetone Hydrate
The anhydrous reagent was developed using an acid
ꢀbase process, because the protons of the gem-diols of
hexafluoroacetone hydrate are highly acidic (pKa1 = 6.76
and pKa2 = 13.53)19 compared to fluoroform (pKa =
30.5).20 It was found that an ethereal solution of hexa-
fluoroacetone trihydrate would precipitate the amidinate
Figure 3. 19F NMR studies of fragmentation of salt 1.
Fluoroform is the most intrinsic source of a trifluoro-
methyl group and is available as a side product of the
industrial synthesis of Teflon.23 Shono,24 Normant,23,25
Roques,20 and Langlois26,27 have reported synthetic meth-
ods for the incorporation of the CF3 anion from
(15) Han, C.; Kim, E. H.; Colby, D. A. J. Am. Chem. Soc. 2011, 133,
5802–5805.
(16) Riofski, M. V.; John, J. P.; Zheng, M. M.; Kirshner, J.; Colby,
D. A. J. Org. Chem. 2011, 76, 3676–3683.
(17) John, J. P.; Colby, D. A. J. Org. Chem. 2011, 76, 9163–9168.
(18) Prager, J. H.; Ogden, P. H. J. Org. Chem. 1968, 33, 2100–2102.
(19) Hine, J.; Flachskam, N. W. J. Org. Chem. 1977, 42, 1979–1981.
(20) Russell, J.; Roques, N. Tetrahedron 1998, 54, 13771–13782.
(21) Colby, D. A.; Riofski, M. V.; Han, C. U.S. Patent Application
No. PCT/US12/30089, filed March 22, 2012.
(22) See Supporting Information for details.
B
Org. Lett., Vol. XX, No. XX, XXXX

fluoroform into organic molecules. Typically, potassium
bases, such as KHMDS, KH, t-BuOK, and dimsyl-K, are
optimal for the deprotonation of fluoroform.25 However,
the decomposition of the CF3 anion, generated from
fluoroform, can occur with metals other than potassium.23
Also, DMF is the solvent of choice, because the CF3 anion
can be stabilized by solvent trapping to form the DMF-
derived hemiaminal.23,25 Moreover, the hemiaminal trap
has been designed into reagents for trifluoromethyla-
tion.4,28 The role of the fluoroform for the synthesis of
fluorinated organic molecules has been recently expanded
by Grushin in the direct cupration of fluoroform through
the use of a potassium base.11 Therefore, we subjected salt
1 to these potassium bases in DMF at ꢀ30 °C to accom-
plish the trifluoromethylation of para-anisaldehyde (Table 1).
When the hexafluoroacetone hydrate salt 1 was reacted
with KHMDS, no product was observed (entry 1); how-
ever, when 18-crown-6 was added to the reaction mixture,
the desired product was obtained, albeit in a very low yield
of 2%, as determined by 19F NMR (entry 2). Exchanging
the base for KH increased the yield of product to 17%,
and comparably, the use of dimsyl-K, as described by
Normant,25 provided a similar conversion. The use of
t-BuOK substantially increased the yield (i.e., 52%) when
18-crown-6 was included (entry 7). Further optimizations
were hindered by competing byproducts from Cannizzaro
reactions. The final optimization that produced superior
yields of the trifluoromethylated product came from ex-
changing 18-crown-6 with tetrabutylammonium chloride
(entry 9). This approach was based on reports from Shono
and co-workers that tetraalkylamines are superior coun-
terions when Cannizzaro reactions are observed from
potassiumcounterionsduring reactions with fluoroform.24
Thusly, promoting the reaction with a combination of
tetrabutylammonium chloride and t-BuOK gave an ex-
cellent 82% yield of the trifluoromethylated product. The
counterions are likely exchanged, and potassium chloride
is generated from the mixture. Indeed, a precipitate is
observed upon addition of n-Bu4NCl to t-BuOK in the
reaction mixture. The presumed base ist-BuONBu4, which
has been reported to have distinct properties compared to
t-BuOK.29,30 The ratio of base to the salt 1 (see entry 9)
correlated well with the prior work with fluoroform25 and
phenyl trifluoromethyl sulfone,14 because 2 equiv of base
were required for each of the exchangeable protons in 1.
Yields decreased dramatically when fewer equivalents of
base were used (entry 8). We presume that the combination
of 1 under the optimized basic conditions promotes the gen-
eration of the CF3 anion that adds across the CdO bond.
Table 1. Trifluoromethylations of p-Anisaldehyde with Salt 1
entry
base (equiv)
additive (equiv)
yielda (%)
1
2
3
4
5
6
7
8
9
KHMDS (4.0)
KHMDS (4.0)
KH (2.0)
none
0
18-crown-6 (1.0)
none
2
7
KH (4.4)
18-crown-6 (1.0)
none
17
15
10
52
7
dimsyl-K
t-BuOK (4.0)
t-BuOK (4.0)
t-BuOK (3.0)
t-BuOK (4.4)
none
18-crown-6 (2.0)
n-Bu4NCl (3.0)
n-Bu4NCl (4.4)
82
a Yields based on 19F NMR (CF3C6H5 is the internal standard).
With the optimized procedure, a series of aldehydes and
ketones were converted to their trifluoromethyl alcohols in
good to excellent (78ꢀ96%) isolated yields (Table 2). Aryl
aldehydes substituted with electron-donating groups or
halogens are fully compatible to trifluoromethylation with
salt 1. Similarly, diaryl ketones can also be trifluoromethy-
lated using this method to provide excellent yields of
trifluoromethyl alcohols (Table 2, entries 7ꢀ10). The
workup procedure for this synthetic method is notable,
because the byproducts of the reaction (i.e., t-BuOH,
tetrabutylammonium salts, and trifluoroacetate) are easily
removed by an aqueous wash. High-yielding trifluoro-
methylations with salt 1 can be executed on the milligram
scale and up to the gram scale. For example, an 87%
isolated yield was obtained with para-dimethylaminoben-
zaldehyde (15 mg, 0.1 mmol) and 1 (Table 2, entry 5), and
upon scale-up to 1 g (6.7 mmol), a 75% yield of the
trifluoromethylated alcohol was procured. Additionally,
phenyl disulfide, a precursor to the phenyl trifluoromethyl
sulfone reagent, is trifluoromethylated with salt 1 (eq 1).
Overall, the scope of the substrates that participate in this
synthetic method with salt 1 is nearly identical to that for
the novel phenyl trifluoromethyl sulfone developed by
Prakash et al.13 This reagent provides a substantial im-
provement in yields when compared to the deprotonation
of fluoroform, and only 1.2 equiv of 1 is required. Also,
releasing trifluoroacetate into the reaction media is not a
concern, because it is already widely present as a counter-
ion in many synthetic reactions and is also nontoxic.31
Lastly, salt 1 can be prepared in one synthetic step up to a
multigram scale without any purification.
ꢀ
(23) Folleas, B.; Marek, I.; Normant, J.-F.; Jalmes, L. S. Tetrahedron
Lett. 1998, 39, 2973–2976.
(24) Shono, T.; Ishifune, M.; Okada, T.; Kashimura, S. J. Org. Chem.
1991, 56, 2–4.
ꢀ
(25) Folleas, B.; Marek, I.; Normant, J.-F.; Saint-Jalmes, L. Tetra-
hedron 2000, 56, 275–283.
(26) Large, S.; Roques, N.; Langlois, B. R. J. Org. Chem. 2000, 65,
8848–8856.
(27) Billard, T.; Bruns, S.; Langlois, B. R. Org. Lett. 2000, 2, 2101–
2103.
(28) Billard, T.; Langlois, B. R.; Blond, G. Eur. J. Org. Chem. 2001,
1467–1471.
(29) Matsumoto, M.; Yamada, K.; Ishikawa, H.; Hoshiya, N.;
Watanabe, N.; Ijuin, H. K. Tetrahedron Lett. 2006, 47, 8407–8411.
(30) Norris, A. R. Can. J. Chem. 1980, 58, 2178–2182.
salt 1, t-BuOK, n-Bu4NCl
Ph ꢀ S ꢀ S ꢀ Ph
Ph ꢀ S ꢀ CF
f
3
DMF, ꢀ 30 °C, 73%
ð1Þ
The role of fluoroform in the synthesis of fluorinated
organic compounds has expanded substantially following
Org. Lett., Vol. XX, No. XX, XXXX
C

1 participates in a similar difluoromethylation process
by adapting the protocol of Mikami. Using the lithium
derived enolate of the methyl ester of ibuprofen, difluoro-
methylation succeeded by bubbling fluoroform generated
from 1 through the reaction media (eq 2). The product was
observed in a 60% yield as determined by 19F NMR and
obtained in 50% yield after isolation. These modest yields
are quite comparable to the previously reported yields (i.e.,
33%ꢀ82%),33 yet represent another powerful use of this
novel amidinate salt of hexafluoroacetone hydrate (1) and
the release of trifluoroacetate.
Table 2. Scope of Trifluoromethylations with Salt 1
In conclusion, we have discovered an amidinate salt of
hexafluoroacetone hydrate and DBU that generates
fluoroform following the base-promoted release of tri-
fluoroacetate. This stable reagent is not hygroscopic, can
be routinely weighed in air, and will participate as a
nucleophilic trifluoromethylation reagent. It can be pre-
pared in one step without any purification (up to a multi-
gram scale) and is not a fluoro-halocarbon-based chem-
ical, which is potentially ozone-depleting. The addition
of a trifluoromethyl group occurs from 1 equiv of reagent
and in good to excellent yields on carbonyl and disulfide
electrophiles. A reaction with this reagent can be executed
on a milligram scale and up to a gram scale. This reagent is
a dynamic source of fluoroform, as it can either directly
release fluoroform into an organic solvent or generate it
for bubbling through a reaction mixture. Moreover, this
reagent will drive the discovery of new approaches for the
incorporation of a trifluoromethyl group that use fluoro-
form. Such methods have garnered increasing utility from
the recent elegant work with fluoroform by Grushin11 and
Mikami.33 We have additionally demonstrated that this
reagent can also prepare a difluoromethylated molecule
using a modified version of the latter method. The bypro-
ducts from this reagent are innocuous (i.e., trifluoroacetate
and DBU). Future studies will describe new avenues of
inducing trifluoroacetate release and extending the use of
this salt in trifluoromethylation reactions as well as other
preparations of fluorinated organic structures.34
a Isolated yields
standard)].
[
19F NMR yields (CF3C6H5 is the internal
Acknowledgment. We acknowledge the Trask Innova-
tion Fund from the Purdue Research Foundation and
Purdue University for funding this work. Also, we ac-
knowledge Philip E. Fanwick and the X-ray Crystallogra-
phy Center at Purdue University.
the direct cupration methods reported by Grushin in
201111 and 2012.32 Additionally in 2012, Mikami et al.
have demonstrated a striking protocol for the CꢀF activa-
tion of fluoroform with lithium enolates for difluoro-
methylation.33 Accordingly, we aimed to determine if salt
Supporting Information Available. Full experimental
details, spectroscopic data, and X-ray data. This material
is available free of charge via the Internet at http://pubs.
acs.org.
(31) Han, C.; Kim, E. H.; Colby, D. A. Synlett 2012, 23, 1559–1563.
ꢀ
(32) Novak, P.; Lishchynskyi, A.; Grushin, V. V. Angew. Chem., Int.
Ed. 2012, 51, 7767–7770.
(33) Iida, T.; Hashimoto, R.; Aikawa, K.; Ito, S.; Mikami, K. Angew.
Chem., Int. Ed. 2012, 51, 9535–9538.
(34) The reagent 1 has recently been commercialized by Sigma-
Aldrich (product #L511315).
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX