Palladium-catalysed electrophilic aromatic C-H fluorination

Article abstract of DOI:10.1038/nature25749

Aryl fluorides are widely used in the pharmaceutical and agrochemical industries, and recent advances have enabled their synthesis through the conversion of various functional groups. However, there is a lack of general methods for direct aromatic carbon-hydrogen (C-H) fluorination. Conventional methods require the use of either strong fluorinating reagents, which are often unselective and difficult to handle, such as elemental fluorine, or less reactive reagents that attack only the most activated arenes, which reduces the substrate scope. A method for the direct fluorination of aromatic C-H bonds could facilitate access to fluorinated derivatives of functional molecules that would otherwise be difficult to produce. For example, drug candidates with improved properties, such as increased metabolic stability or better blood-brain-barrier penetration, may become available. Here we describe an approach to catalysis and the resulting development of an undirected, palladium-catalysed method for aromatic C-H fluorination using mild electrophilic fluorinating reagents. The reaction involves a mode of catalysis that is unusual in aromatic C-H functionalization because no organometallic intermediate is formed; instead, a reactive transition-metal-fluoride electrophile is generated catalytically for the fluorination of arenes that do not otherwise react with mild fluorinating reagents. The scope and functional-group tolerance of this reaction could provide access to functional fluorinated molecules in pharmaceutical and agrochemical development that would otherwise not be readily accessible.

Full text of DOI:10.1038/nature25749

LeTTer
doi:10.1038/nature25749
Palladium-catalysed electrophilic aromatic
C–H fluorination
Kumiko yamamoto1,2*, Jiakun Li1*, Jeffrey a. O. Garber1,2, Julian D. rolfes1,2, Gregory b. boursalian1,2, Jannik C. borghs2,
Christophe Genicot3, Jérôme Jacq3, Maurice van Gastel1, frank Neese1 & Tobias ritter1,2
Aryl fluorides are widely used in the pharmaceutical and is electrophilic at fluorine and capable of oxidative fluorine transfer to
agrochemical industries1,2, and recent advances have enabled their arenes. We designed the Pd(ii) complex 1, ligated simultaneously by a
synthesis through the conversion of various functional groups. tridentate (terpyridine, terpy) and a bidentate (2-chloro-1,10-phenan-
However, there is a lack of general methods for direct aromatic throline, 2-Cl-phen) ligand, which would be oxidized by electrophilic
carbon–hydrogen (C–H) fluorination3. Conventional methods fluorinating reagents to yield the desired Pd(iv)–F complex 2 (Fig. 1b).
require the use of either strong fluorinating reagents, which are often The oxidation of doubly cationic 1 is promoted by a destabilizing inter-
2
unselective and difficult to handle, such as elemental fluorine, or less action between the lone pair of the apical donor atom and the filled dz
reactive reagents that attack only the most activated arenes, which orbital on Pd(ii), which is readily apparent in the highest occupied
reduces the substrate scope. A method for the direct fluorination molecular orbital of 1 as calculated by density functional theory (DFT)
of aromatic C–H bonds could facilitate access to fluorinated (Fig. 1c). X-ray diffraction corroborates the apical interaction: the
derivatives of functional molecules that would otherwise be difficult
to produce. For example, drug candidates with improved properties,
F
a
such as increased metabolic stability or better blood–brain-barrier
Pd catalyst 1
penetration, may become available. Here we describe an approach to
CN
CN
*
catalysis and the resulting development of an undirected, palladium-
catalysed method for aromatic C–H fluorination using mild
electrophilic fluorinating reagents. The reaction involves a mode of
catalysis that is unusual in aromatic C–H functionalization because
no organometallic intermediate is formed; instead, a reactive
transition-metal-fluoride electrophile is generated catalytically
for the fluorination of arenes that do not otherwise react with mild
fluorinating reagents. The scope and functional-group tolerance
of this reaction could provide access to functional fluorinated
molecules in pharmaceutical and agrochemical development that
would otherwise not be readily accessible.
2 equiv. NFSI
MeCN, 25 °C
20 h
3a,1 equiv.
15 mmol (2.7 g) scale
4a, 61% isolated yield
69:31 ortho:para
F
b
“F+”
L
L
L
L
L
Pd
Pd
L
L
L
L
L
Pd(IV)
Pd(II)
Octahedral
Electrophilic ꢀuorinating species
Square planar
2+
3+
F
Conventional methods for aromatic fluorination require elemental
fluorine or similarly reactive reagents, which are unselective and
require specialized equipment to handle safely4. Bench-stable electro-
philic fluorinating reagents—such as N-fluoropyridinium salts,
N-fluorobenzenesulfonimide (NFSI) and Selectfluor—are easier
to handle but less reactive, and require either very electron-rich
arenes or multiple equivalents of the arene to accomplish direct C–H
fluorination5,6. Catalysis of aromatic C–H fluorination reactions has
been reported using coordination-assistance to promote fluorination
proximal to Lewis-basic functional groups, but such approaches are
limited in scope to those substrates containing the required directing
groups7–10. Advances in aliphatic C–H fluorination have been made11,
but currently there is no method for direct aromatic C–H fluorination
with broad scope.
N
N
N
N
“F+”
PdII
PdIV
N
N
N
N
N
Cl
N
Cl
1
2
c
d
Figure 1 | Aromatic fluorination catalysed by 1. a, Palladium catalyst 1
enables the direct, non-chelation-assisted fluorination of 4-cyanobiphenyl.
b, Oxidation of Pd(ii) complex 1, assisted by the ligand combination of
terpy and 2-Cl-phen, yields the triply cationic Pd(iv)–F electrophile 2.
c, The highest occupied molecular orbital of 1, as calculated by DFT,
showing destabilizing orbital interaction. Hydrogen atoms are omitted
for clarity. Calculations at the CPCM(MeCN)TPSS0 D3/def2-QZVP//
PBE0 D3/def2-TZVP level of theory; iso =0.05. d, X-ray crystal structure
of 1, shown with 50%-probability ellipsoids. Hydrogen atoms and solvent
In our investigation into the catalysis of aromatic C–H fluorination
reactions, we sought an approach that was distinct from the common
C–H activation sequence in which C–H metalation precedes
functionalization; with few exceptions12–14, the conventional approach15
requires multiple equivalents of the arene substrate to promote C–H
metalation in the absence of a coordinating directing group. Instead,
we sought to design catalysts with ancillary ligands that would favour
the oxidation of the complex before any interaction with the substrate,
giving rise to a reactive, high-valent metal-fluoride intermediate that molecules are omitted for clarity.
1Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mülheim an der Ruhr, Germany. 2Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street,
Cambridge, Massachusetts 02138, USA. 3Global Chemistry, UCB NewMedicines, UCB Biopharma, 1420 Braine-L’Alleud, Belgium.
*These authors contributed equally to this work.
2 2 f e b r u a r y 2 0 1 8
| V O L 5 5 4 | N a T u r e | 5 1 1
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
2+
–
2
BF4
N
PdII
N
5 mol% 1
O
S
O
S
H
F
Cl
Ph
O
Ph
O
2 equiv. N–F reagent
N
N
N
R
R
N
F
N
MeCN, 25 °C
8–36 h
–
2
BF4
F
N
3
4
Cl
1
1 equiv.
Selectꢀuor
NFSI
F
F
F3C
F3C
F
F
F
Br
Br
Cl
Cl
F
N
N
*
*
N
F
4ba
4bb
4ca
4cb
49%, 52:48 (4ba:4bb)†‡
51%, 59:41 (4ca:4cb)†‡
4d, 85%, 69:31†‡
4e, 55%, 77:23†‡
4f, 46%, 67:33†‡
NO2
NO2
O
O
S
O
N
S
Me
N
O
Me
N
O
S
F
Cl
Cl
F
F
O
*
F
O
N
N
*
N
H
NHAc
O
*
OMe
O
F
4ja
4jb
4g, 45%, 69:31†‡
4h, 59%, 70:30§||
4i, 52%, 86:14§
60%, 51:49 (4ja:4jb)†‡
O
O
O
O
F
Cl
Cl
Cl
O
O
Me
Me
Me
Me
Me
N
F
N
O
N
N
F
N
N
N
N
Ac
F
*
Cl
O
O
Ac2N
F
4na
4nb
F
4ka
4kb
F-procymidone
35%, 56:44 (4na:4nb)§Ì††
4m, 54%†#
56%, 53:47 (4ka:4kb)†¶
4l, 71%, 72:28†¶
F
Me
O
Me
O
*
F
O
F
O
N
O
N
O
O
OnBu
OEt
N
H
N
H
Cl
Cl
N
H
O
Me Me
OH
OH
F
Me
Me
Me
4qa
Cl
4qb
Cl
Me
F-ethyl nateglinide
4o, 39%, 65:35†
F-butyl ciproꢀbrate
4p, 30%§Ì
F-tolvaptan
56%, 51:49 (4qa:4qb)‡§Ì
Figure 2 | Substrate scope of the Pd-catalysed fluorination of arenes.
Reaction conditions: arene, 1 mmol; catalyst 1, 5 mol%; Selectfluor
or NFSI, 2 equiv.; MeCN, 0.1 M. All isomers of non-volatile products
have been isolated and characterized as analytically pure samples. The
asterisk denotes the site of fluorination of the constitutional isomer that
is not shown. Overall yields and the ratio of the constitutional isomers
are based on 19F NMR integration of reaction mixtures with internal
standard. †Reaction performed with Selectfluor. ‡Reaction performed
at 50 °C. §Reaction performed with NFSI. Reaction performed with
1,2-dichloroethane and MeCN (1:1, 0.1 M). ¶Reaction performed at 0 °C.

#10 mol% catalyst 1 was used. 7.5 mol% catalyst 1 was used. ††Reaction
performed at 80 °C.
Pd–N distance in the X-ray structure of 1 (Fig. 1d) is 2.6Å, which is 1.2Å Structurally complex substrates—such the pesticide procymidone (3n),
shorter than the sum of the van der Waals radii of Pd and N (ref. 16). the type-2 diabetes drug nateglinide (3o), the lipid-lowering agent
Complex 1 is a competent catalyst for the fluorination of various ciprofibrate (3p) and the hyponatremia drug tolvaptan (3q)—were
arenes by either Selectfluor or NFSI. The substrates shown in Fig. 2 fluorinated directly via catalysis with 1. Although fluorine can impart
underwent fluorination in the presence of 1 at room temperature or desirable properties on pharmaceuticals and agrochemicals, fluorinated
at 50°C, but little background reactivity was observed in the absence analogues of structurally complex molecules can currently be diffi-
of 1 under otherwise identical conditions (maximum <1% yield) or cult to access; conventional fluorination methods failed to provide the
even under reflux in acetonitrile (maximum 21% yield). In our studies, fluorinated products shown in Fig. 2.
catalyst 1 was formed in situ from [Pd(terpy)(MeCN)](BF4)2 and
In most cases, the fluorination reaction affords mixtures of at least
2-Cl-phen, but it can also be generated by combining the commercially two constitutional isomers, resulting from similar rates of fluorination
available palladium source [Pd(MeCN)4](BF4)2 and the ligands before at the positions ortho and para to the aromatic substituents. Purification
the addition of the reactants. Compatible functional groups include of aryl fluoride products from mixtures of their constitutional isomers
nitriles (3a), aryl bromides (3b), chlorides (3c, 3g, 3n, 3p, 3q), certain and the starting material is often challenging17,18. However, the isolation
heterocycles (3e–3g, 3m), sulfonamides (3h, 3j), ketones (3h, 3k), and characterization of all the non-volatile products obtained here has
amides (3i, 3l–3o, 3q), esters (3i, 3o, 3p), carbamates (3l), ethers (3p) been achieved, although optimization of the separation protocol was
and free hydroxy groups (3q). Five-membered heteroarenes containing required for each substrate. For example, the ortho- and para-fluorinated
nitrogen (3m) can be tolerated, but oxidatively labile functional groups products of the gram-scale fluorination of 4-cyanobiphenyl have been
such as amines and thiols cannot, owing to their general incompatibility separated in 61% isolated yield (Fig. 1a). Although high positional selec-
with electrophilic fluorinating reagents. Electron-deficient arenes tivity is generally desired in C–H functionalization reactions, mixtures
such as 3b–3d are not successfully fluorinated through conventional of constitutional isomers can be advantageous for some applications. For
methodologies, but are suitable substrates for fluorination via catalysis example, in the late-stage derivatization of drug candidates, each product
with 1. However, more electron-deficient arenes, such as methyl isomer is an additional derivative that can be obtained without the need
benzoate, are insufficiently reactive and undergo little or no conversion. for costly and laborious de novo synthesis19,20. Fluorinated tolvaptan
5 1 2
| N a T u r e | V O L 5 5 4 | 2 2 f e b r u a r y 2 0 1 8
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
a
Cl
Cl
we hypothesize that the fluorination mechanism proceeds through a
single transition state via fluoride-coupled electron transfer (Fig. 3a, b).
Selectfluor fluorinates only electron-rich arenes such as anisole23;
complex 1, conversely, is able to catalyse the fluorination of elec-
tron-deficient arenes such as chlorobenzene. DFT calculations suggest
that Pd(iv)–F 2 has a higher single-electron reduction potential
than that of Selectfluor, although both compounds have a similar
thermodynamic driving force for electrophilic fluorination. The tran-
sition state TS of the fluorination of chlorobenzene with Pd(iv)–F 2
shows high spin-density on the Pd as well as on the aryl carbon atoms.
As such, the transition state is most appropriately characterized as a
singlet diradical; two subsequent fluoride-coupled electron trans-
fers occur asynchronously as the reaction proceeds through a single
transition state. The mechanism is reminiscent of that previously
reported for the fluorination of enamines and organometallic reagents
with an isolated Pd(iv)–F (ref. 24). The calculated energy barrier for
electrophilic fluorination of 21.8 kcal mol−1 (Fig. 3b) is in agreement
with the observed reaction time and the temperature of the reaction
(as discussed below, see Supplementary Information for details).
We sought to produce Pd(iv)–F complex 2, to verify our design
principle as well as to investigate the reactivity of 2 with arenes.
–H+
+
H
F
F
2+
4c
5c
N
N
II
Cl
3+
Pd
N
N
Cl
+
H
N
N
N
N
F
Cl
+
–
F
2 BF4
N
PdIII
N
1
N
N
Cl
N
+
3+
Cl
F
N
N
PdIV
N
–
BF4
TS
N
N
N
Cl
Cl
2
3c
Treatment of 1 with XeF2 in the presence of LiBF4 produced a 19
F
TS
21.8
b
d
NMR signal attributable to 2 at δ=−258.6 p.p.m.; however, complex
2 is reduced to 1 in acetonitrile solution, even in the absence of sub-
strate, which hindered our attempts at isolation. Treatment of 1 with
Selectfluor over a range of temperatures (−40°C to 25°C) resulted in
the desired reduction of Selectfluor, but did not produce a 19F NMR sig-
nal that could be attributed to a Pd(iv)–F species, presumably because 2
is reduced faster than it is formed under these conditions, and therefore
does not accumulate in observable quantities.
ΔG
(kcal mol–1
)
3c
1
Cl
+
N
N
Complex 1′, in which 2-Cl-phen is replaced with unsubstituted
+
F
2
phenanthroline (phen), is a competent catalyst for aromatic fluorina
-
1
1.2
tion, although not as effective as catalyst 1. When 1′ was treated with
Selectfluor in acetonitrile at room temperature and then allowed to
stand at −35°C, the Pd(iv)–F complex 2′ (19F NMR −259.5p.p.m.)
precipitated in 73% yield (Fig. 3c). Complex 2′ is sufficiently stable at low
temperature to enable characterization and reactivity studies. The higher
stability of 2′ as compared to 2 is understood to result from the greater
electron-donating ability of phen relative to that of 2-Cl-phen. Likewise,
the greater reactivity of 2 may explain why the 2-Cl-phen-ligated com-
plex 1 outperforms the phen-ligated 1′ in terms of catalytic ability.
Pd(iv)–F complex 2′ reacts with arenes to yield fluorinated products
(Fig. 4). For example, 4-cyanobiphenyl (3a), when treated with 2′ in
acetonitrile, yielded a 66:34 ratio of ortho- and para-fluoro isomers in
63% overall yield. The positional selectivity of fluorination by 2′ is similar
to that observed in fluorination catalysed by 1′ (69:31 ortho:para), con-
sistent with 2′ being the C–F-bond-forming species in catalysis by 1′.
These selectivity ratios are, in turn, similar to that observed upon fluori-
nation with the optimal 2-Cl-phen-ligated catalyst 1 (71:29 ortho:para).
We cannot rule out the involvement of a different C–F bond-forming
pathway in catalysis by 1 (for example, through a Pd(iii) species);
however, the similar positional selectivities observed in fluorination
catalysed by 1, fluorination catalysed by 1′, and stoichiometric fluorina-
tion by 2′ are consistent with fluorination by similar Pd(iv)–F species in
all three cases, corresponding to 2 in the case of catalysis by 1.
0
Cl
+
N
N
5c
–7.1
c
2+
3+
F
2 equiv.
Selectꢀuor
N
PdII
N
N
N
PdIV
N
N
N
N
MeCN
73%
N
N
–
2
BF4
–
3
BF4
1′
2′
Figure 3 | Mechanism of fluorination catalysed by 1. a, The proposed
catalytic cycle for the fluorination of chlorobenzene (3c). b, Energy-
level diagram of the proposed catalytic cycle with chlorobenzene (3c);
energies calculated by DFT. c, Synthesis of Pd(iv)–F complex 2′ via the
oxidation of 1′ by Selectfluor. d, X-ray crystal structure of 2′, shown with
50%-probability ellipsoids. Pd–F bond length: measured, 1.9120(7) Å;
calculated, 1.89 Å.
(4q), for example, would be challenging and time-consuming to pre-
pare with conventional chemistry through de novo syntheses. Late-stage
fluorination, even with the requirement for a custom-made separation
protocol, can conveniently produce promising new candidates that may
have never been evaluated otherwise.
The proposed mode of action of 1 is highly unusual in the catalysis
of aromatic oxidation reactions: conceptually, an activated electrophile
is generated in situ from 1, in the form of Pd(iv)–F intermediate 2. The
activated Pd(iv)–F electrophile 2 would therefore be capable of electro-
philic fluorination of weakly nucleophilic arenes that cannot be fluori-
nated directly by Selectfluor and NFSI (Fig. 3a)21,22. A DFT analysis of
the aryl fluorination reaction suggests a mechanism that is accessible to
Pd(iv)–F 2 but not to Selectfluor or NFSI. On the basis of these results,
Aromatic C–H oxidations catalysed by transition metals generally
proceed via C–H metalation followed by oxidation of the resulting
organometallic intermediate, with product formation ensuing through
reductive elimination. The aromatic fluorination catalysed by 1
presented here is an unusual example of an alternative mode of catalysis
for aromatic C–H oxidation, in which a transition-metal catalyst is oxi-
dized to a high-valent intermediate, which in turn oxidizes the substrate
by group transfer of a ligand. Such an ‘oxidation-first’ mechanism is rem-
iniscent of various transition-metal-catalysed aliphatic C–H oxidations,
such as hydroxylations25–27 and halogenations28 through high-
valent metal–oxo complexes, and aminations29 via metal–nitrenoid
2 2 f e b r u a r y 2 0 1 8
| V O L 5 5 4 | N a T u r e | 5 1 3
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
+
9. Wang, X., Mei, T.-S. & Yu, J.-Q. Versatile Pd(OTf)2·2H2O-catalyzed ortho-fluorination
using NMP as a promoter. J. Am. Chem. Soc. 131, 7520–7521 (2009).
10. Hull, K. L., Anani, W. Q. & Sanford, M. S. Palladium-catalyzed fluorination of
carbon–hydrogen bonds. J. Am. Chem. Soc. 128, 7134–7135 (2006).
11. Liu, W. et al. Oxidative aliphatic C–H fluorination with fluoride ion catalyzed by
a manganese porphyrin. Science 337, 1322–1325 (2012).
3
F
CN
CN
N
N
PdIV
N
N
CD3CN
+
N
12. Mkhalid, I. A. I., Barnard, J. H., Marder, T. B., Murphy, J. M. & Hartwig, J. F. C–H
activation for the construction of C–B bonds. Chem. Rev. 110, 890–931 (2010).
13. Cheng, C. & Hartwig, J. F. Rhodium-catalyzed intermolecular C–H silylation of
arenes with high steric regiocontrol. Science 343, 853–857 (2014).
14. Wang, P. et al. Ligand-accelerated non-directed C–H functionalization of arenes.
Nature 551, 489–493 (2017).
F
–
3
BF4
*
3a
5 equiv.
2′
1 equiv.
4a
63%, 66:34
+
2
15. Kuhl, N., Hopkinson, M. N., Wencel-Delord, J. & Glorius, F. Beyond directing
groups: transition-metal-catalyzed C–H activation of simple arenes. Angew.
Chem. Int. Ed. 51, 10236–10254 (2012).
16. Alvarez, S. A cartography of the van der Waals territories. Dalton Trans. 42,
8617–8636 (2013).
17. Regalado, E. L., Makarov, A. A., McClain, R., Przybyciel, M. & Welch, C. J. Search
for improved fluorinated stationary phases for separation of fluorine-
containing pharmaceuticals from their desfluoro analogs. J. Chromatogr. A
1380, 45–54 (2015).
18. Regalado, E. L. et al. Support of academic synthetic chemistry using separation
technologies from the pharmaceutical industry. Org. Biomol. Chem. 12,
2161–2166 (2014).
19. Hyohdoh, I. et al. Fluorine scanning by nonselective fluorination: enhancing
Raf/MEK inhibition while keeping physicochemical properties. ACS Med. Chem.
Lett. 4, 1059–1063 (2013).
20. Cernak, T., Dykstra, K. D., Tyagarajan, S., Vachal, P. & Krska, S. W. The medicinal
chemist’s toolbox for late stage functionalization of drug-like molecules.
Chem. Soc. Rev. 45, 546–576 (2016); erratum 46, 1760 (2017).
21. McCall, A. S. & Kraft, S. Pyridine-assisted chlorinations and oxidations by
palladium(iv). Organometallics 31, 3527–3538 (2012).
N
N
N
PdII
N
CN
CN
N
–
2
BF4
X
Cl
N
5 mol%
MeCN
+
N
F
F
–
2
BF4
*
3a
1 equiv.
4a
Selectꢀuor
2 equiv.
X = H (1′): 40%, 69:31
X = Cl (1): 80%, 71:29†
Figure 4 | Comparison of the positional selectivity of stoichiometric
and catalytic fluorinations using 2′. Top, stoichiometric fluorination;
bottom, catalytic fluorination. †5 equiv. of 4-cyanobiphenyl (3a) and
1 equiv. of Selectfluor were used.
22. McCall, A. S., Wang, H., Desper, J. M. & Kraft, S. Bis-N-heterocyclic carbene
palladium(iv) tetrachloride complexes: synthesis, reactivity, and mechanisms
of direct chlorinations and oxidations of organic substrates. J. Am. Chem. Soc.
133, 1832–1848 (2011).
23. Geng, C., Du, L., Liu, F., Zhu, R. & Liu, C. Theoretical study on the mechanism of
selective fluorination of aromatic compounds with Selectfluor. RSC Adv. 5,
33385–33391 (2015).
24. Brandt, J. R., Lee, E., Boursalian, G. B. & Ritter, T. Mechanism of electrophilic
fluorination with Pd(iv): fluoride capture and subsequent oxidative fluoride
transfer. Chem. Sci. 5, 169–179 (2014).
25. Groves, J. T. High-valent iron in chemical and biological oxidations. J. Inorg.
Biochem. 100, 434–447 (2006).
species. Reported examples of such a mechanism for aromatic C–H
oxidation, however, are rare, and are proposed to involve metal-
nitrene or aminyl-radical transfer to the arene, although evidence for
the proposed modes of action in these cases is indirect30,31. To the best
of our knowledge, the aromatic fluorination reaction reported here
is the only example so far of a synthetic method for aromatic C–H
functionalization in which oxidation reactivity between a high-valent
catalytic intermediate and arenes has been directly scrutinized.
The other examples of aromatic and aliphatic C–H oxidation reac
-
26. McNeill, E. & Du Bois, J. Ruthenium-catalyzed hydroxylation of unactivated
tertiary C–H bonds. J. Am. Chem. Soc. 132, 10202–10204 (2010).
tions proceeding through ‘oxidation-first’ mechanisms mentioned
above function because the high-valent intermediate provides access 27. Chen, M. S. & White, M. C. A predictably selective aliphatic C–H oxidation
reaction for complex molecule synthesis. Science 318, 783–787 (2007).
to a mechanism of oxidation that is not available to the starting
28. Liu, W. & Groves, J. T. Manganese catalyzed C–H halogenation. Acc. Chem. Res.
48, 1727–1735 (2015).
29. Du Bois, J. Rhodium-catalyzed C–H amination. An enabling method for
chemical synthesis. Org. Process Res. Dev. 15, 758–762 (2011).
30. Boursalian, G. B., Ngai, M.-Y., Hojczyk, K. N. & Ritter, T. Pd-catalyzed aryl C–H
imidation with arene as the limiting reagent. J. Am. Chem. Soc. 135,
13278–13281 (2013).
31. Paudyal, M. P. et al. Dirhodium-catalyzed C–H arene amination using
reagents, such as a radical-rebound mechanism in the case of hydrox-
ylation through metal–oxo intermediates. We have shown here data
that support the interpretation that catalyst 2 provides access to a
fluoride-coupled electron-transfer mechanism that is not accessible to
electrophilic fluorinating reagents such as Selectfluor.
We anticipate that the direct electrophilic C–H fluorination of arenes
hydroxylamines. Science 353, 1144–1147 (2016).
reported here will be a useful tool in medicinal chemistry; indeed,
the reaction has already found use in the late-stage derivatization of
Supplementary Information is available in the online version of the paper.
drug molecules. Furthermore, the unusual mechanism of catalysis by
Acknowledgements We thank L. Gitlin for HPLC purification, R. Goddard,
S. Palm and J. Rust for X-ray crystallographic analysis, C. Farès for NMR
spectroscopy, and M. Blumenthal, D. Kampen and S. Marcus for mass
spectrometry. We thank UCB Biopharma for funding and compound separation,
complex 1, in which a high-valent transition-metal intermediate under-
goes group transfer to arenes, may become the basis for a new approach
to the catalysis of C–H functionalization reactions.
the Japan Society for the Promotion of Science and L’Oréal-UNESCO Japan for
Data Availability Data are available from the corresponding author on
reasonable request.
graduate fellowships to K.Y., the Fonds der Chemischen Industrie for a graduate
fellowship for J.D.R., and the German Academic Exchange Service, DAAD for an
Otto-Bayer fellowship to J.C.B.
received 21 March; accepted 17 December 2017.
Author Contributions K.Y. designed catalyst 1 and optimized the fluorination
reaction. K.Y. and J.D.R. conducted mechanistic studies. K.Y. and G.B.B.
developed the conceptual approach to the project. J.L., J.A.O.G., J.C.B., K.Y. and
J.D.R. explored the substrate scope. J.D.R. performed DFT calculations with
input from M.v.G. and F.N. G.B.B. wrote the manuscript with input from all other
authors. C.G. and J.J. supported development towards useful examples. K.Y., J.L.,
J.A.O.G., G.B.B., J.D.R. and T.R. analysed the data. T.R. directed the project.
1. Müller, K., Faeh, C. & Diederich, F. Fluorine in pharmaceuticals: looking beyond
intuition. Science 317, 1881–1886 (2007).
2. Gillis, E. P., Eastman, K. J., Hill, M. D., Donnelly, D. J. & Meanwell, N. A. Applications
of fluorine in medicinal chemistry. J. Med. Chem. 58, 8315–8359 (2015).
3. Campbell, M. G. & Ritter, T. Modern carbon–fluorine bond forming reactions for
aryl fluoride synthesis. Chem. Rev. 115, 612–633 (2015).
4. Sandford, G. Elemental fluorine in organic chemistry (1997–2006). J. Fluor.
Chem. 128, 90–104 (2007).
5. Taylor, S. D., Kotoris, C. C. & Hum, G. Recent advances in electrophilic
fluorination. Tetrahedron 55, 12431–12477 (1999).
6. Lal, G. S., Pez, G. P. & Syvret, R. G. Electrophilic NF fluorinating agents.
Chem. Rev. 96, 1737–1756 (1996).
7. Fier, P. S. & Hartwig, J. F. Selective C–H fluorination of pyridines and diazines
inspired by a classic amination reaction. Science 342, 956–960 (2013).
8. Chan, K. S. L., Wasa, M., Wang, X. & Yu, J.-Q. Palladium(ii)-catalyzed selective
monofluorination of benzoic acids using a practical auxiliary: a weak-
coordination approach. Angew. Chem. Int. Ed. 50, 9081–9084 (2011).
Author Information Reprints and permissions information is available
at www.nature.com/reprints. The authors declare no competing financial
interests. Readers are welcome to comment on the online version of the
paper. Publisher’s note: Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
Correspondence and requests for materials should be addressed to
T.R. (ritter@kofo.mpg.de).
reviewer Information Nature thanks J. Groves and the other anonymous
reviewer(s) for their contribution to the peer review of this work.
5 1 4
| N a T u r e | V O L 5 5 4 | 2 2 f e b r u a r y 2 0 1 8
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.