Direct photocatalytic fluorination of benzylic C-H bonds with N-fluorobenzenesulfonimide

Article abstract of DOI:10.1039/c5cc04058b

The late-stage fluorination of common synthetic building blocks and drug leads is an appealing reaction for medicinal chemistry. In particular, fluorination of benzylic C-H bonds provides a means to attenuate drug metabolism at this metabolically labile position. Here we report two complimentary strategies for the direct fluorination of benzylic C-H bonds using N-fluorobenzenesulfonimide and either a decatungstate photocatalyst or AIBN-initiation.

Full text of DOI:10.1039/c5cc04058b

ChemComm
View Article Online
View Journal
COMMUNICATION
Direct photocatalytic fluorination of benzylic C–H
bonds with N-fluorobenzenesulfonimide†
Cite this:DOI: 10.1039/c5cc04058b
Matthew B. Nodwell,a Abhimanyu Bagai,a Shira D. Halperin,a Rainer E. Martin,b
Henner Knustc and Robert Britton*a
Received 16th May 2015,
Accepted 15th June 2015
DOI: 10.1039/c5cc04058b
www.rsc.org/chemcomm
The late-stage fluorination of common synthetic building blocks
and drug leads is an appealing reaction for medicinal chemistry. In
particular, fluorination of benzylic C–H bonds provides a means to
attenuate drug metabolism at this metabolically labile position.
Here we report two complimentary strategies for the direct fluori-
nation of benzylic C–H bonds using N-fluorobenzenesulfonimide
and either a decatungstate photocatalyst or AIBN-initiation.
Fig. 1 Beneficial properties of benzylfluorides in lead optimization.15,16
The incorporation of one or more fluorine atoms into a small
molecule is of great interest to medicinal chemistry.1 For example,
replacement of a hydroxyl group or hydrogen atom with a fluorine
Recently, Lectka described the mechanism of C–H fluorination
is a common tactic used to improve metabolic stability, cell with Selectfluor13 and proposed that fluorine transfer from Select-
permeability, or ligand-target interactions (Fig. 1).2 Consequently, fluor (BDENF B260 kJ molÀ1 14a) to a carbon radical generates a
fluorine is present in a large and growing number of approved cationic nitrogen-centered radical that is responsible for hydrogen
drugs.3 Owing to the unfavorable properties of F2 gas, however, abstraction. Curiously, the related fluorine transfer reagent14
the wide spread incorporation of organofluorines in medicinal N-fluorobenzenesulfonimide (7) (NFSI, BDENF B266 kJ molÀ1 14a
)
chemistry campaigns has relied on the development of electro- was unable to propagate aliphatic C–H fluorination.11d Moreover,
philic and nucleophilic fluorination reagents and reactions.4 To among the growing number of C–H fluorination reactions, few are
compliment these processes, there has recently been considerable operative7,9,11d and only our recently reported decatungstate-
interest in the direct fluorination of unactivated C–H bonds.5,6 catalyzed C–H fluorination17 is optimal with NFSI. This unique
Such late-stage fluorination strategies are especially attractive photocatalytic fluorination reaction presumably involves C–H
to drug discovery programs, where the direct fluorination abstraction by the photoexcited decatungstate18 followed by
of metabolically labile C–H bonds would obviate extensive fluorine transfer14 from NFSI (7) (see inset, Scheme 1). Here, we
re-synthesis campaigns. In particular, the fluorination of benzylic describe the development of a complimentary benzylic C–H
C–H7–11 bonds provides a means to attenuate metabolic degrada- fluorination reaction using decatungstate photocatalysis (TBADT =
tion at these so-called ‘‘hot spots’’.12
tetrabutylammonium decatungstate). Moreover, we demonstrate
that NFSI is capable of propagating the radical fluorination of
benzylic C–H bonds and that benzylic fluorination with NFSI can
be initiated by AIBN. Notably, these unique NFSI-based C–H
fluorination reactions provide differing selectivities in the fluori-
nation of substrates with multiple benzylic C–H bonds, and thus
provide new opportunities for site-specific fluorination.
As depicted in Scheme 1, we first investigated the fluorina-
tion of ethyl benzene (10) and the p-acetoxy derivative 11 using
conditions developed previously by us for aliphatic C–H fluori-
nations.17 We were surprised that a major product of both
reactions were the acetamides 12 and 13. While decatungstate
oxidation of an intermediate benzylic radical to a carbocation19
a Department of Chemistry, Simon Fraser University, Burnaby, British Columbia,
V5A1S6, Canada. E-mail: rbritton@sfu.ca
b Molecular Design and Chemical Biology, Roche Pharma Research and Early
Development, Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd,
Grenzacherstrasse 124, CH-4070 Basel, Switzerland
c Medicinal Chemistry, Roche Pharma Research and Early Development,
Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd, Grenzacherstrasse 124,
CH-4070 Basel, Switzerland
† Electronic supplementary information (ESI) available: Experimental procedures,
reaction optimization data, and characterization data for all new compounds. See
DOI: 10.1039/c5cc04058b
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
Scheme 1 Decatungstate-catalyzed fluorination of unactivated C–H
a
bonds and the fluorination of ethyl benzenes 10 and 11. Yield of the
volatile benzyl fluoride 15 based on analysis of 1H NMR spectra with
b
internal standard. Accompanied by 42% recovered 11.
followed by reaction with acetonitrile would explain the for-
mation of these Ritter-type products, Brønsted acid activation20
of initially formed benzylic C–F bonds by dibenzenesulfonamide
(9, NHSI) may also be operative. Thus, we evaluated the effect of
various bases as additives to the fluorination of 10. Unsurprisingly,
soluble amine bases such as pyridine and Et3N were not compa-
tible with this process, neither were a small collection of inorganic
bases (e.g., Cs2CO3, Na2CO3, K2CO3, LiOH). Fortunately, when
either NaHCO3 or Li2CO3 was added to the reaction mixture the
desired benzyl fluoride 15 was formed as the exclusive product in
good yield. Likewise, employing these reaction conditions with the
ethyl benzene 11, the fluoroethyl benzene 16 was produced in 75%
yield. To gain further insight into the formation of acetamides 12
and 13, the benzyl fluorides 15 and 16 were treated independently
Fig. 2 Decatungstate-catalyzed and AIBN-initiated fluorination of
benzylic C–H bonds. Conditions A: NFSI (3 equiv.), TBADT (2 mol%),
CH3CN (0.6 M), NaHCO3 (1 equiv.), 365 nm lamp, 16–48 h. Conditions
B: NFSI (3.0 equiv.), AIBN (5 mol%), CH3CN (0.6 M), Li2CO3 (1 equiv.), 75 1C,
16 h. a Isolated yield. b Yield based on analysis of 1H NMR spectra recorded
on the crude reaction mixture with an internal standard. c Accompanied by
the isomeric a-fluoroketone.
with a catalytic amount of NHSI (9) in CH3CN, which led to and electron poor (e.g., 17–19) alkylbenzenes are competent
mixtures containing predominantly the acetamides 12 and 13 at substrates for this reaction. The fluorination of isobutylbenzenes
temperatures as low as 35 1C. When these reactions were repeated proceeded smoothly (e.g., 20 and 23), however, more sterically
without NHSI, the benzyl fluorides were recovered unreacted. This encumbered isopropyl benzenes proved difficult to fluorinate
NHSI-promoted fluoride substitution reaction represents a unique (e.g., 22) and proceeded in modest yield only with the electron
example of Brønsted acid catalyzed C–F activation and may prove rich p-acetoxy derivative. In this later case, the desired fluoro-
itself to be a useful transformation with further examination.
isopropyl adduct 28 was produced in 50% yield. In all cases,
In order to gain insight into the scope of the decatungstate- deoxygenation of the acetonitrile solvent was critical to avoid
catalyzed benzylic C–H fluorination reaction, we evaluated its competing decatungstate-catalyzed benzylic oxidation.19 In
effectiveness on a range of substrates with differing electronic contrast to all other benzylic C–H fluorination reactions,7–11
and steric features. As summarized in Fig. 2 (conditions A), we replacing NFSI with Selectfluor resulted in significantly lower
were pleased to find that this reaction is general and cleanly yields. Interestingly, despite the necessary addition of base to
provides a range of benzyl fluorides 17–30 in modest to good these reactions (vide supra), benzylic C–H fluorination is com-
yield. It is notable that the discrepancies between conversion petitive with a-fluorination of ketone-containing substrates,
(in parentheses) and isolated yields reflect the difficulty in iso- and provides access to the unusual g-fluorotetralone 25 and
lating and separating the benzyl fluoride products from the b-fluoroindanones 27 and 29.
parent hydrocarbon, a challenge common to late-stage C–H fluori-
In an effort to gain further insight into the mechanism of
nation strategies.5a Importantly, both electron rich (e.g., 16 and 28) decatungstate-catalyzed benzylic fluorination, we examined the
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
fluorination of ethyl benzene (i.e., 10 - 15, Scheme 1) without and suggested a preference for hydrogen abstraction from the
the photocatalyst and observed no formation of the benzyl- methyl group in 31 based on steric considerations.22 In the present
fluoride 15 (with or without irradiation or heating). However, case, it is important to note that NFSI is not completely dissolved
when a catalytic amount of the radical initiator AIBN was added in the decatungstate-catalyzed reactions at room temperature but
and the reaction mixture was heated (75 1C), we observed clean is fully solublilized at 75 1C in the AIBN-initiated reactions. Thus,
conversion to the benzylfluoride 15, albeit in modest yield we speculated that the unique fluorination selectivities may in fact
(B20%, eqn (1)). We were able to improve the yield of this result from a higher concentration of NFSI in the heated reactions
AIBN-initiated reaction to 70% by simply replacing NaHCO3 and the consequent trapping of an initially formed methyl radical.
with Li2CO3. This dramatic base effect may be understood by Conversely, at lower concentrations of NFSI (decatungstate-
21
considering that NaHCO3 slowly decomposes to Na2CO3 at catalyzed reaction), equilibration between benzylic radicals23
the elevated temperatures required for the AIBN-initiated reaction, may be responsible for the preferred formation of the fluoro-
and that Na2CO3 is incompatible with the NFSI fluorination ethyl products 32 and 35. To probe this hypothesis, the AIBN-
reaction (vide supra). It is notable that several aliphatic substrates initiated fluorination of 31 (conditions B) was repeated with
failed to fluorinate using these reaction conditions (i.e., AIBN, decreasing amounts of NFSI to assess the effect of NFSI
NFSI, Li2CO3) suggesting that unlike Selectfluor,11,13 NFSI is only concentration. When substoichiometric amounts (0.5 equiv.)
capable of propagating the radical fluorination of relatively weak of NFSI were added to this reaction, the fluoroethyl product 32
C–H bonds (BDECH benzylic B377 kJ molÀ1). As summarized in was also observed as a minor product (33 : 32 = 1.4 : 1). This
Fig. 2 (conditions B), these reaction conditions proved to be less result indicates that the equilibration of benzylic radicals is
optimal and general than the decatungstate-catalyzed fluorination likely responsible for the differing selectivities in the fluorina-
(conditions A), and several substrates containing benzylic C–H tion of ethyltoluenes, and provides a useful tool for imparting
bonds failed to provide any fluorinated products (e.g., 22–24).
or attenuating selectivity in the fluorination of substrates with
multiple benzylic C–H bonds.
Finally, in an effort to decrease the reaction time for the
decatungstate-catalyzed fluorination, we examined the fluori-
nation of ibuprofen methyl ester (37) in flow.10,24 As depicted in
Scheme 3, irradiation of the reaction mixture in FEP tubing
(3 m length, 1.4 mL total volume) wrapped around a blacklight
blue lamp (BLB lamp, l = 365 nm) reduced the reaction time
from 24 h (batch) to 5 h (flow) without impacting the yield.
Considering the potential importance of late-stage fluorination
to medicinal chemistry we were also interested in gauging the
effect of benzylic fluorination on basic pharmacokinetic pro-
perties of ibuprofen. Thus, the ester 20 was hydrolyzed to
afford the fluoroibuprofen 38. Not surprisingly, fluorination
of ibuprofen resulted in a modest decrease in pKa (from 4.35
(ibuprofen) to 4.23 (38)). Fluorination of ibuprofen also slightly
improved metabolic stability in both human and rat microsomes,
leading to a decrease in clearance from 19 to 12 mg minÀ1 mgÀ1
protein in human microsomes and 71 to 39 mg minÀ1 mgÀ1
protein in rat microsomes.
(1)
During our examination of both the decatungstate-catalyzed
and AIBN-initiated fluorination reactions, we evaluated the
fluorination of p-ethyl toluene (31) (Scheme 2). Surprisingly,
the decatungstate-catalyzed reaction selectively fluorinated the
ethyl group in 31, while the AIBN-initiated fluorination provided
the fluoromethyl product 33 exclusively. A similar trend was
observed in the fluorination of m-ethyltoluene (34). Ni has reported
analogous selectivity in the amination of p-ethyltoluene with NFSI,
Scheme 2 Differing selectivities in the fluorination of ethyltoluenes using
a
AIBN-initiated or decatungstate-catalyzed processes.
Accompanied
by 7% of 33. Yield based on analysis of 1H NMR spectra with internal
Scheme 3 Photochemical fluorination of ibuprofen methyl ester (37) in
flow, and synthesis of fluoroibuprofen 38. (a) LiOH, MeOH, THF, H2O, 24 h.
b
standard.
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
10 D. Cantillo, O. de Frutos, J. A. Rincon, C. Mateos and O. C. Kappe,
J. Org. Chem., 2014, 79, 8486.
In summary, we have developed two complimentary methods
for benzylic fluorination that rely on the fluorine transfer agent
NFSI in combination with either a decatungstate photocatalyst
or radical initiator (AIBN). These processes tolerate a range of
functional groups in providing benzyl fluorides in modest to
excellent yield. Furthermore, the photocatalyic fluorination can
be adapted to continuous flow without compromising yield.25
That this process is uniquely effective with NFSI and not Select-
fluor differentiates it from existing and complimentary benzylic
fluorination strategies and provides new opportunities for late
stage fluorination in drug discovery.
This work was supported by an NSERC Discovery Grant to
R.B., a MSFHR Career Investigator Award to R.B., an NSERC
Postgraduate Scholarship for S.D.H., and a Hoffmann-La Roche
PostDoc Fellowship (RPF) for M.N. The authors would like to
thank Mr. Severin Wendelspiess for pKa determinations,
Dr. Andrew Lewis for assistance with the acquisition of NMR
spectra, and Mrs. Isabelle Walter for microsomal stability data.
11 (a) S. Bloom, C. R. Pitts, R. Woltornist, A. Griswold, M. G. Holl and
T. Lectka, Org. Lett., 2013, 15, 1722; (b) S. Bloom, M. McCann and
T. Lectka, Org. Lett., 2014, 16, 6338; (c) S. Bloom, C. R. Pitts,
D. C. Miller, N. Haselton, M. G. Holl, E. Urheim and T. Lectka,
Angew. Chem., Int. Ed., 2012, 51, 10580; (d) C. R. Pitts, B. Ling,
R. Woltornist, R. Liu and T. Lectka, J. Org. Chem., 2014, 79, 8895;
(e) S. Bloom, S. A. Sharber, M. G. Holl, J. L. Knippel and T. Lectka,
J. Org. Chem., 2013, 78, 11082.
12 (a) B. R. Baer, R. K. DeLisle and A. Allen, Chem. Res. Toxicol., 2009,
22, 1298; (b) S. B. Rosenblum, T. Huynh, A. Afonso, H. R. Davis Jr,
N. Yumibe, J. W. Clader and D. A. Burnett, J. Med. Chem., 1998,
41, 973.
13 C. R. Pitts, S. Bloom, R. Woltornist, D. J. Auvenshine, L. R. Ryzhkov,
M. A. Siegler and T. Lectka, J. Am. Chem. Soc., 2014, 136, 9780.
14 (a) M. Rueda-Becerril, C. C. Sazepin, J. C. Leung, T. Okbinoglu,
P. Kennepohl, J. F. Paquin and G. M. Sammis, J. Am. Chem. Soc.,
2012, 134, 4026; (b) J. C. T. Leung, C. Chatalova-Sazepin, J. G. West,
M. Rueda-Becerril, J.-F. Paquin and G. M. Sammis, Angew. Chem.,
Int. Ed., 2012, 51, 10804.
15 W. J. Thompson, P. S. Anderson, S. F. Britcher, T. A. Lyle, J. E. Thies,
C. A. Magill, S. L. Varga, J. E. Schwering, P. A. Lyle and M. E. Christy,
et al., J. Med. Chem., 1990, 33, 789.
16 J. I. Andres, J. M. Alonso, A. Diaz, J. Fernandez, L. Iturrino,
P. Martinez, E. Matesanz, E. J. Freyne, F. Deroose, G. Boeckx,
D. Petit, G. Diels, A. Megens, M. Somers, J. Van Wauwe, P. Stoppie,
M. Cools, F. De Clerck, D. Peeters and D. de Chaffoy, Bioorg. Med.
Chem. Lett., 2002, 12, 653.
Notes and references
1 (a) J. Wang, M. Sanchez-Rosello, J. L. Acena, C. del Pozo, A. E.
Sorochinsky, S. Fustero, V. A. Soloshonok and H. Liu, Chem. Rev., 17 S. D. Halperin, H. Fan, S. Chang, R. E. Martin and R. Britton, Angew.
2014, 114, 2432; (b) K. Mu¨ller, C. Faeh and F. Diederich, Science,
2007, 317, 1881.
2 (a) S. Purser, P. R. Moore, S. Swallow and V. Gouverneur, Chem. Soc.
Rev., 2008, 37, 320; (b) W. K. Hagmann, J. Med. Chem., 2008,
51, 4359; (c) K. L. Kirk, Org. Process Res. Dev., 2008, 12, 305.
3 D. O’Hagan, J. Fluorine Chem., 2010, 131, 1071.
Chem., Int. Ed., 2014, 53, 4690.
18 (a) M. D. Tzirakis, I. N. Lykakis and M. Orfanopoulos, Chem. Soc.
Rev., 2009, 38, 2609; (b) S. Protti, D. Ravelli, M. Fagnoni and
A. Albini, Chem. Commun., 2009, 7351; (c) D. Dondi, A. M.
Cardarelli, M. Fagnoni and A. Albini, Tetrahedron, 2006, 62, 5527;
(d) L. A. Combswalker and C. L. Hill, J. Am. Chem. Soc., 1992,
114, 938; (e) R. F. Renneke, M. Pasquali and C. L. Hill, J. Am. Chem.
Soc., 1990, 112, 6585.
4 (a) T. Liang, C. N. Neumann and T. Ritter, Angew. Chem., Int. Ed.,
2013, 52, 8214; (b) R. P. Singh and J. M. Shreeve, Synthesis, 2002,
2561; (c) M. G. Campbell and T. Ritter, Chem. Rev., 2015, 115, 612; 19 S. F. Jen, A. B. Anderson and C. L. Hill, J. Phys. Chem., 1992, 96, 5658.
(d) M. G. Campbell and T. Ritter, Org. Process Res. Dev., 2014, 20 (a) P. A. Champagne, J. Pomarole, M. E. Therien, Y. Benhassine,
18, 474; (e) P. T. Nyffeler, S. G. Duron, M. D. Burkart, S. P. Vincent
and C. H. Wong, Angew. Chem., Int. Ed., 2004, 44, 192; ( f ) J. Wu,
S. Beaulieu, C. Y. Legault and J.-F. Paquin, Org. Lett., 2013, 15, 2210;
(b) H. Amii and K. Uneyama, Chem. Rev., 2009, 109, 2119.
Tetrahedron Lett., 2014, 55, 4289; (g) P. A. Champagne, J. Desroches, 21 M. Hartman, K. Svoboda, M. Pohorely and M. Syc, Ind. Eng. Chem.
J.-D. Hamel, M. Vandamme and J.-F. Paquin, Chem. Rev., 2015, DOI:
10.1021/cr500706a.
5 (a) C. N. Neumann and T. Ritter, Angew. Chem., Int. Ed., 2015,
54, 3216; (b) J.-A. Ma and S. Li, Org. Chem. Front., 2014, 1, 712.
6 (a) W. Liu, X. Y. Huang, M. J. Cheng, R. J. Nielsen, W. A. Goddard
Res., 2013, 52, 10619.
22 Z. Ni, Q. Zhang, T. Xiong, Y. Zheng, Y. Li, H. Zhang, J. Zhang and
Q. Liu, Angew. Chem., Int. Ed., 2012, 51, 1244.
23 R. D. Guthrie, B. C. Shi, V. Rajagopal, S. Ramakrishnan and
B. H. Davis, J. Org. Chem., 1994, 59, 7426.
and J. T. Groves, Science, 2012, 337, 1322; (b) S. Bloom, J. L. Knippel 24 J. P. Knowles, L. D. Elliott and K. I. Booker-Milburn, Beilstein J. Org.
and T. Lectka, Chem. Sci., 2014, 5, 1175.
Chem., 2012, 8, 2025.
7 J. B. Xia, C. Zhu and C. Chen, J. Am. Chem. Soc., 2013, 135, 17494.
8 X. Huang, W. Liu, H. Ren, R. Neelamegam, J. M. Hooker and
J. T. Groves, J. Am. Chem. Soc., 2014, 136, 6842.
25 TBADT-catalyzed fluorination of 4-ethylphenyl acetate and 4-ethyl-
acetophenone in continuous flow provided the corresponding
benzyl fluorides 16 (66%, 2 h) and 17 (34%, 16 h) in yields similar
to those achieved in batch reactions (see Fig. 2).
9 Y. Amaoka, M. Nagatomo and M. Inoue, Org. Lett., 2013, 15, 2160.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015