Library-friendly synthesis of fluorinated ketones through functionalized hydration of alkynes and investigation of the reaction mechanism

Article abstract of DOI:10.1016/j.jfluchem.2011.05.006

A library-friendly synthesis of α-substituted-α-fluoroketones through functionalized hydration of alkynes in one pot under mild conditions is reported. The ready availability of alkynes and organoboronic acids makes this reaction quite attractive. We also studied the key intermediate - a fluorinated cationic gold species - using in situ NMR spectroscopy and ESI-high resolution mass spectrometry.

Full text of DOI:10.1016/j.jfluchem.2011.05.006

Journal of Fluorine Chemistry 132 (2011) 804–810
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Library-friendly synthesis of fluorinated ketones through functionalized
hydration of alkynes and investigation of the reaction mechanism
Bo Xu 1, , Weibo Wang, Gerald B. Hammond
*
*
Department of Chemistry, University of Louisville, Louisville, KY 40292, USA
A R T I C L E I N F O
A B S T R A C T
A library-friendly synthesis of
Article history:
a
-substituted-a-fluoroketones through functionalized hydration of
Received 30 March 2011
Received in revised form 5 May 2011
Accepted 10 May 2011
Available online 17 May 2011
alkynes in one pot under mild conditions is reported. The ready availability of alkynes and organoboronic
acids makes this reaction quite attractive. We also studied the key intermediate—a fluorinated cationic
gold species—using in situ NMR spectroscopy and ESI-high resolution mass spectrometry.
ß 2011 Elsevier B.V. All rights reserved.
Keywords:
Cationic gold
Fluorination
Selectfluor
Fluoroketones
Alkynes
1. Introduction
KF (Halex reaction) [9] as well as the more recent transition-metal
promoted Ar–F bond formation with electrophilic ‘‘F+’’ reagents
Extensive research has established that the chemical reactivity
of fluorinated organic compounds is distinctly different from other
halide-containing analogues [1–7]. The small size of fluorine, its
high electronegativity, and its strong bond with carbon, contribute
to the fact that, once fluorine is introduced in an organic
compound—either in place of a hydrogen atom or another organic
functionality—it induces only minimal steric alterations but
profound electronic changes in the resulting compound. Conse-
quently, fluorine substitution can effectively modify the physico-
chemical properties of the molecule [1–7]. Until 1957, though, no
F-containing drugs had been developed. These days, fluorinated
drugs make up 20% of all pharmaceuticals sold worldwide, with
even higher figures for agrochemicals. Compared to other
heteroatoms, fluorine is not easy to introduce in an organic
compound, especially in non-aromatic systems. Today, the
synthesis of drugs and agrochemicals rely on commercially
available stocks of fluorinated aromatic substrates (e.g. fluorine
substituted benzaldehydes). Fluoroaromatics can be made by
traditional methods like conversion of aromatic amines via the
aryldiazonium salt (Balz–Schiemann reaction) [8] and the nucleo-
philic substitution of electron-poor bromo- or chloroarenes with
such as Selectfluor or N-fluoropyridinium salts [10–12]. A notable
recent breakthrough has been reported by Buchwald and co-
workers whereby palladium catalyzes the conversion of aryl
triflates to aryl fluorides [13], although this is not yet an
industrially feasible or medicinal chemistry-friendly process.
Compared to other heteroatoms, the use of fluorine as a tool in
organic synthesis has not come to full fruition because fluorine is
not an easy atom to introduce in an organic compound, especially
in an aliphatic system. By far, the most common method that
medicinal chemists use to make aliphatic fluorinated compounds
is through the conversion of alcohol, carbonyl compounds or alkyl
halides to fluorinated compounds by fluorination (e.g., DAST) [3].
But this protocol is not library-friendly, each fluorinated product
needs one specific starting material (Scheme 1, left). Because of
this, fluorine is not a user-friendly substituent to introduce,
especially when it comes to generate synthetic libraries using
combinatorial tools, so it comes as no surprise that the number of
drug candidates possessing aliphatic or cyclic fluorine substituent
is so dismally low.
We sought
a more efficient strategy to access aliphatic
fluorocompounds: using fluorine-generated cationic metal species
to mediate a tandem nucleophilic addition/coupling/fluorination
reaction, starting from readily available alkynes (Scheme 1, right).
One advantage of using alkynes is that they are ubiquitous in
synthesis and many of them are readily available [14,15]. Their
reaction with suitable coupling reagents (e.g. phenyl boronic acid)
and suitable nucleophiles in the presence of fluorination reagents
* Corresponding authors. Tel.: +1 502 852 5998.
E-mail address: gb.hammond@louisville.edu (G.B. Hammond)
URL: http://louisville.edu/chemistry/directory/faculty/gerald-b-hammond.html
1
Tel.: +1 502 852 5978.
0022-1139/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2011.05.006

B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
805
Scheme 1. Library-friendly synthesis of aliphatic fluorinated compounds library.
(e.g. Selectfluor) will generate a diversified library of aliphatic
fluorinated compounds (M Â N Â L members, Scheme 1, right). An
additional advantage is that alkynes are chemically inert to many
reaction conditions (like acidic or basic conditions); this is
especially important when using alkyne intermediates in late
steps of target syntheses. A proof of principle is the synthesis of
configurationally stable substituents for molecules containing a
tertiary chiral carbon atom next to a carbonyl group, an important
structural motif in medicinal chemistry [22]. Our one-pot tandem
addition/oxidative coupling/fluorination sequence—using readily
available starting materials (alkyne, water, organoboronic acid and
Selectfluor)—has clear advantages over literature methods, all of
which require multiple synthetic steps [17–20].
functionalized a-fluoroketones we recently discovered [16]. In this
paper, we will describe this process in more detail, especially our
mechanistic study of the key intermediate—fluorinated cationic
gold species using in situ NMR and ESI-high resolution mass
spectrometry.
A major shortcoming of a transition metal catalyzed hydration
is that it only adds the elements of H2O to an alkyne [17–20]. We
proposed that the vinyl metal complex hydration intermediate
2. Results and discussions
We examined the reaction of 1a with phenyl boronic acid
(Scheme 2). Ph3PAuCl could not catalyze the reaction of 1 with
phenyl boronic acid, but in combination with Selectfluor, we
obtained good yields of 3. Other gold(I) and gold(III) catalysts were
less effective and other transition metals like copper, silver or
palladium could not catalyze this transformation under similar
conditions. On exploring the scope of this novel functionalized
hydration, we found that functionalized and unfunctionalized
internal alkynes reacted with aryl boronic acid, giving very good
could react further during the hydration process to give an
a, a-
disubstituted ketone in a one-pot reaction. This process, that we
coined ‘functionalized hydration’ allowed, for example, the
synthesis of functionalized
and important synthetic intermediates because they are effective
mimics of -hydroxy ketones, useful probes in various biological
processes, and enzyme inhibitors [21,22]. They also provide
a-fluoroketones—well-known targets
a
yields of
a-substituted ketones, albeit with moderate regioselec-
tivity (Scheme 2). Steric and electronic effects determine the
Scheme 2. Scope of the functionalized hydrationa. a1 (0.4 mmol), 2 (0.8 mmol), Selectfluor (1.0 mmol), ClAuPPh3 (5%), 3 mL CH3CN/water (20:1), rt, 18–24 h. bNumbers in
parentheses correspond to the ratio of two regioisomers.

806
B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
Scheme 3. Proposed mechanism for the functionalized hydration.
regioselectivity [23,24]. For internal alkynes possessing a nucleo-
philic site nearby (e.g. the ester group in 1a), this site may influence
the regioselectivity through neighboring group participation [23–
25].
1c with phenylboronic acid gives the oxidative coupling product 6
[31]. The formation of 6 implies a tandem mechanism that involves
fluorination of the gold complex, transmetalation with boronic
acid or terminal alkyne, and reductive elimination, as shown in
Scheme 3. We have not been able to isolate 5 (Scheme 3), possibly
due to its fast conversion to 3. It is interesting to note that gold
catalysts undergo transmetalation and reductive elimination
readily, although they are not able to undergo oxidative inser-
tion—as Pd does in coupling reactions. A more detailed mechanism
(Scheme 5) can be used to rationalize the formation of 6.
The key step of this mechanism is the generation of cationic
gold species A by fluorination or oxidation (Scheme 3). The
oxidation state of gold changed from Au(I) to Au(III). Oxidation of
Au(I) to Au(III) by Selectfluor, while postulated as a reasonable
process [29,32,33], has never been confirmed experimentally. X-
ray photoelectron spectroscopy (XPS) is a quantitative spectro-
scopic technique used to measure the chemical state of the
elements [34]. For example, it can determine the chemical states of
supported gold catalysts [35]. The binding energy (BE) of Au 4f7/2
electron of each gold oxidation states is usually different enough to
be differentiated (for ClAuIPPh3, Au 4f7/2 = 85.7 eV, for NaAuIIICl4,
Au 4f7/2 = 87.6 eV) [34]. We used this technique to investigate the
valence change of gold in the reaction. First we tested our gold
standard samples (ClAuIPPh3 and NaAuIIICl4), the Au 4f7/2
photoelectron peak is located at a BE value at 85.7 and 87.5 eV
respectively, results that are quite consistent with the literature
[34]. We investigated our Selectfluor treated Au(I) samples using
the same conditions as for the standards. Our XPS measurements
confirmed the existence of gold(III) (BE of Au 4f7/2 = 87.6 eV) in the
reaction mixture of gold(I) catalyst and Selectfluor.
Our proposed mechanism is shown in Scheme 3. Initially, water
attacks the gold-activated alkyne to form a vinyl gold complex C
[26], which then reacts with a metal reagent RM (e.g., PhB(OH)2)
through a transmetalation process, to give intermediate D [27]. The
transmetalation of Au(I)–Cl complex with boronic acid has been
demonstrated by Hashmi and co-workers in a recent paper [28].
The transmetalation of Au–F with boronic acid has also been
proposed by Zhang and co-workers in their gold catalyzed cross-
coupling reaction of propargylic acetate [29]. We believe that the
strong B–F bond and the weak Au–F bond are the driving forces
behind this transmetalation. Reductive elimination of D gives E.
The reaction does not stop at this stage; instead E can be
fluorinated by Selectfluor to give the functionalized ketone 3 [30].
Because we had never isolated 5 in all cases, it is also possible that
intermediate D reacts with Selectfluor first to give F, and, following
reductive elimination, gives the final product 3.
To probe the proposed mechanism in Scheme 3 we conducted
two control experiments (Scheme 4). According to Scheme 3,
fluoroketone 4 could be formed in the absence of a coupling
partner R3M, in which case, the vinyl metal complex C undergoes
reductive elimination or fluorodemetalation [15] to give 4. This
was the case experimentally, although so far only moderate yields
have been obtained (Scheme 4a). The reaction of terminal alkyne
Our XPS measurements indicate the existence of gold(III)
species, but XPS experiments can only be conducted in high
Scheme 4. Synthesis of fluoroketones and oxidative coupling of alkynes.
Scheme 5. Putative mechanism for the formation of 6.

B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
807
Fig. 1. Monitor of reaction of AuCl with Selectfluor using 19F NMR (Ln = CH3CN).
vacuum as solid state. In order to investigate the reaction in
solution phase at room temperature, we monitored the progress of
the reaction using 19F NMR spectroscopy (Fig. 1). We mixed AuCl
and Selectfluor in CD3CN; after only 5 min (Figure 1a), a new peak
appeared (6% conversion, using BF4À peak as reference); after 2 h,
the new peak increased significantly (Fig. 1a, conversion 65%)
with phenylboronic acid, followed by reductive elimination to
give biphenyl (Fig. 1). This result indicates that Au(III)+ClFLn is a
highly reactive species: it undergoes transmetalation with
phenylboronic acid readily and its reductive elimination gives
biphenyl. The high reactivity seems to rule out the possibility
that it is a simple fluoride anion.
and the N+–F peak in Selectfluor decreased (
d
+50 ppm, not
Literature reports on 19F NMR of transition metal–fluorine
complexes, especially Au–F complexes are rare. Gray, Sadighi and
shown in Fig. 1). We assigned this peak (bs,
d
À184 ppm) to
correspond to Au(III)+ClFLn. And when we added PhB(OH)2 to the
system, and ran its 19F NMR spectrum immediately after, the
co-workers have reported
fluoride; its 19F NMR showed a single peak at
[36]. Grushin and co-workers have reported a PhPdF(Py)2 type
complex (19F NMR at
a
NHC-carbene-stabilized gold(I)
d
À247.0 ppm
Au(III)+ClFLn had disappeared and a new peak appeared (
d
À148 ppm, FB(OH)2). After the addition of PhB(OH)2, we also
detected the formation of biphenyl (Ph–Ph, confirmed by flash
chromatography of the reaction mixture). This result suggests
that the new peak represents a reactive metal fluoride species,
which we tentatively assigned as [AuIIIClFLn]+ (Ln = CH3CN). The
formation of biphenyl can be rationalized by transmetalation
d
À219.2 ppm). The fact that our Au–F signal
appeared downfield may be due to the higher electron-withdraw-
ing property of the cationic Au(III) metal center compared to Au(I)
and Pd(II) complexes.
Electrospray ionization mass spectrometry (ESI-MS) has
become an important tool in the identification of labile reaction
Scheme 6. Preparation of samples for ESI-MS study.

808
B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
Fig. 2. ESI-MS of sample A.
intermediates. An advantage of ESI-MS is that it allows direct
sampling from the reaction mixture, and because many of our gold
intermediates are charged species, ESI-MS could help us to detect
charged species. We conducted a high-resolution ESI-MS investi-
gation of the cationic or anionic species in the catalytic system
(Scheme 6). First, we checked the high-resolution ESI-MS spectra of
samples A and B but only Au(I) species (peaks c, d, f) and reduced
Selectfluor (peak b) were detected (Fig. 2). We speculated that a
Au(III) species would be difficult to detect under our ESI-MS
conditions due to possible multiple charge (z > 1) on the Au(III)
cation. Because a Au(III) complex has a square-planar geometry, a
bidentate ligand should greatly stabilize this Au(III) complex,
which in turn, could be easier to detect by ESI-MS spectrometry.
Hence, we added excess amounts of bipyridine to the reaction
mixture (Scheme 6, sample C). The high-resolution ESI-MS
spectrum is shown in Figs. 3 and 4.
In Fig. 3, various cationic Au(III) species were detected (peaks i,
k, l, m), but [Au(III)ClF]+ itself was not detected; this may be due to
the fact that metal–fluorine bonds tend to be labile and reactive.
Indeed, gold(I) fluoride, was once thought impossible to prepare.
The fluoride ion, according to hard/soft acid–base theory, is
mismatched with the cations formed by late transition metals
especially gold [36]. Its
p-donating ability, moreover, can lead to
destabilizing interactions with filled d orbitals of gold [36].
Compared to many other metal–fluorine bonds, Au–F is a weaker
bond. So, under the ESI-MS conditions (high temperature during
the electrospray ionization), [Au(III)ClF]+ (a) may lose fluorine and
pick up chloride from other gold complexes or OHÀ from trace
Fig. 3. ESI-MS of sample C (part 1, m/z 360–460).

B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
809
In summary, a potentially new role for fluorine in cationic gold
catalysis is proposed, an example of which is the hydration of
alkynes to give
mild conditions. The readyavailability of alkynes and organoboronic
acids, and the current interest in -fluoroketones makes this
a-substituted-a-fluoroketones, inone pot and under
a
reaction quite attractive. The broader implications of cationic metal
species enabled by fluorine are underway in our laboratory.
Acknowledgements
We are grateful to the National Science Foundation for financial
support (CHE-0809683). We acknowledge the support provided by
the CREAM Mass Spectrometry Facility (University of Louisville)
funded by NSF/EPSCoR grant # EPS-0447479.
References
[1] R.D. Chambers, Fluorine in Organic Chemistry, Blackwell Publishing Ltd./CRC
Press, Boca Raton, FL, 2004.
[2] T. Hiyama, Organofluorine Compounds, Chemistry and Applications, Springer-
Verlag, Berlin, 2000.
Fig. 4. ESI-MS of sample C (part 2, m/z from 50 to 360).
[3] P. Kirsch, Modern Fluoroorganic Chemistry, Wiley-VCH, Weinheim, 2004.
[4] K. Muller, C. Faeh, F. Diederich, Science 317 (2007) 1881–1886.
[5] M. Schlosser, Angew. Chem. Int. Ed. 37 (1998) 1496.
[6] V.A. Soloshonok, Fluorine-containing Synthons, ACS Symposium Series 911,
Oxford University Press, Washington, DC, 2005.
amounts of water in the system, because both chloride and
hydroxyl have much stronger affinity to gold (Scheme 7).
[7] K. Uneyama, Organofluorine Chemistry, Blackwell Publishing, Oxford, 2006.
[8] G. Balz, G. Schiemann, Ber. Dtsch. Chem. Ges. 60 (1927).
[9] G.C. Finger, C.W. Kruse, J. Am. Chem. Soc. 78 (1956) 6034–6037.
[10] N.D. Ball, M.S. Sanford, J. Am. Chem. Soc. 131 (2009) 3796–3797.
[11] K.L. Hull, W.Q. Anani, M.S. Sanford, J. Am. Chem. Soc. 128 (2006) 7134–7135.
[12] K.L. Hull, M.S. Sanford, J. Am. Chem. Soc. 131 (2009) 9651–9653.
[13] D.A. Watson, M. Su, G. Teverovskiy, Y. Zhang, J. Garcia-Fortanet, T. Kinzel, S.L.
Buchwald, Science 325 (2009) 1661–1664.
[14] F. Diederich, P. Stang, R.R. Tykwinski, Acetylene Chemistry: Chemistry, Biology,
and Material Science, Wiley-VCH, Weinheim, 2005.
[15] N. Krause, A.S. Hashmi, Modern Allene Chemistry, Wiley-VCH, Weinheim, 2004.
[16] W. Wang, J. Jasinski, G.B. Hammond, B. Xu, Angew. Chem. Int. Ed. 49 (2010) 7247–
7252.
The presence of Au(III)ClF+ (a) was further confirmed by its
conversion to Au(III)Br4À through the reaction with a large excess
of halide source (KBr). We prepared two samples (samples D and E)
using ClAuPPh3 and AuCl as starting materials, respectively
(Scheme 8). Because samples D and E give very similar mass
spectrum, only the spectrum for sample D is shown. The peak
À
corresponding to Au(III)Br4 was indeed detected by ESI-MS; the
exact mass and distinct bromine isotope pattern matched
theoretical calculations.
Scheme 7. Mechanism for the formation of gold(III) species in Fig. 3.
Scheme 8. Preparation of samples for ESI-MS study.

810
B. Xu et al. / Journal of Fluorine Chemistry 132 (2011) 804–810
[17] G.E. Heasley, C. Codding, J. Sheehy, K. Gering, V.L. Heasley, D.F. Shellhamer, T.
Rempel, J. Org. Chem. 50 (2002) 1773–1776.
[28] A.S.K. Hashmi, T.D. Ramamurthi, F. Rominger, J. Organomet. Chem. 694 (2009)
592–597.
[18] R.S. Neale, J. Am. Chem. Soc. 86 (2002) 5340–5342.
[19] L. Benati, P.C. Montevecchi, P. Spagnolo, Tetrahedron Lett. 29 (1988)
2381–2384.
[29] G. Zhang, Y. Peng, L. Cui, L. Zhang, Angew. Chem. Int. Ed. 48 (2009) 3112–3115.
[30] The fluorination of a ketone bearing aromatic ring has been reported: D. Enders,
M.R.M. Huttle, Synlett, (2005) 991.
[20] K. Tamao, K. Maeda, Tetrahedron Lett. 27 (1986) 65–68.
[21] Y. Guo, B. Twamley, J.n.M. Shreeve, Org. Biomol. Chem. 7 (2009) 1716–1722.
[22] G.K.S. Prakash, P. Beier, Angew. Chem. Int. E. 45 (2006) 2172–2174.
[23] N. Marion, R.S. Ramon, S.P. Nolan, J. Am. Chem. Soc. 131 (2009) 448–449.
[24] L. Hintermann, A. Labonne, Synthesis (2007) 1121–1150.
[25] W. Wang, B. Xu, G.B. Hammond, J. Org. Chem. 74 (2009) 1640–1643.
[26] L.-P. Liu, B. Xu, M.S. Mashuta, G.B. Hammond, J. Am. Chem. Soc. 130 (2008) 17642–
17643.
[31] F. Yang, Y. Wu, Eur. J. Org. Chem. 2007 (2007) 3476–3479.
[32] A.D. Melhado, W.E. Brenzovich, A.D. Lackner, F.D. Toste, J. Am. Chem. Soc. 132
(2010) 8885–8887.
[33] G. Zhang, L. Cui, Y. Wang, L. Zhang, J. Am. Chem. Soc. 132 (2010) 1474–1475.
[34] J.-H. Choy, Y.-I. Kim, J. Phys. Chem. B 107 (2003) 3348–3350.
[35] M.P. Casaletto, A. Longo, A. Martorana, A. Prestianni, A.M. Venezia, Surf. Interface
Anal. 38 (2006) 215–218.
[36] D.S. Laitar, P. Muller, T.G. Gray, J.P. Sadighi, Organometallics 24 (2005)
4503–4505.
[27] N.P. Mankad, F.D. Toste, J. Am. Chem. Soc. 132 (2010) 12859–12861.