Catalytic SNAr Hydroxylation and Alkoxylation of Aryl Fluorides

Article abstract of DOI:10.1002/anie.202106440

Nucleophilic aromatic substitution (S_NAr) is a powerful strategy for incorporating a heteroatom into an aromatic ring by displacement of a leaving group with a nucleophile, but this method is limited to electron-deficient arenes. We have now established a reliable method for accessing phenols and phenyl alkyl ethers via catalytic S_NAr reactions. The method is applicable to a broad array of electron-rich and neutral aryl fluorides, which are inert under classical S_NAr conditions. Although the mechanism of S_NAr reactions involving metal arene complexes is hypothesized to involve a stepwise pathway (addition followed by elimination), experimental data that support this hypothesis is still under exploration. Mechanistic studies and DFT calculations suggest either a stepwise or stepwise-like energy profile. Notably, we isolated a rhodium η⁵-cyclohexadienyl complex intermediate with an sp³-hybridized carbon bearing both a nucleophile and a leaving group.

Full text of DOI:10.1002/anie.202106440

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Catalytic SNAr Hydroxylation and Alkoxylation of Aryl Fluorides
Authors: Qi-Kai Kang, Yunzhi Lin, Yuntong Li, Lun Xu, Ke Li, and Hang
Shi
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202106440
Link to VoR: https://doi.org/10.1002/anie.202106440

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Catalytic SNAr Hydroxylation and Alkoxylation of Aryl Fluorides
Qi-Kai Kang[a],[b],+, Yunzhi Lin[a],+, Yuntong Li[a], Lun Xu[a], Ke Li[a],[b], and Hang Shi*,[a],[b]
[a]
Dr. Q.-K. Kang[+], Y. Lin[+], Y. Li, L. Xu, Dr. K. Li, Prof. Dr. H. Shi
Key Laboratory of Precise Synthesis of Functional Molecules of Zhejiang Province, School of Science
Westlake University
18 Shilongshan Road, Hangzhou 310024, Zhejiang Province, China
shihang@westlake.edu.cn
[b]
[+]
Dr. Q.-K. Kang[+], Dr. K. Li, Prof. Dr. H. Shi
Institute of Natural Sciences
Westlake Institute for Advanced Study
18 Shilongshan Road, Hangzhou 310024, Zhejiang Province, China
These authors contributed equally to this work
Supporting information for this article is given via a link at the end of the document.
Abstract: Nucleophilic aromatic substitution (SNAr) is a powerful
strategy for incorporating a heteroatom into an aromatic ring by
displacement of a leaving group with a nucleophile. However, this
method is generally limited to electron-deficient arenes. We have
now established a reliable method for accessing phenols, as well as
phenyl alkyl ethers via catalytic SNAr reactions. Specifically, by using
a rhodium catalyst in the absence of a base, we achieved both
hydroxylation and alkoxylation of aryl fluorides, which are inert under
classical SNAr conditions. The method was applicable to a broad
array of electron-rich and neutral aryl fluorides, which were activated
product, p-cresol (Scheme 1b).[7] Hydroxylation of electron-rich
and neutral aryl fluorides remains unprecedented. We
envisioned that a catalytic umpolung aromatic substitution
strategy might allow us to develop a reliable protocol for the
synthesis of phenols from the corresponding aryl fluorides.
via η6-coordination with
a cyclopentadienyl rhodium species.
Although the mechanism of SNAr reactions involving metal arene
complexes is hypothesized to involve a stepwise pathway (addition
followed by elimination), experimental data that support this
hypothesis is still under exploration. We performed mechanistic
studies and carried out density functional theory (DFT) calculations,
which suggested either a stepwise or stepwise-like energy profile.
Importantly, we isolated a rhodium η5-cyclohexadienyl complex
intermediate with an sp3-hybridized carbon bearing both the
nucleophile and the leaving group.
Scheme 1. Approaches to Phenols from Aryl Halides.
Introduction
Phenols, as well as phenyl alkyl ethers, are found in numerous
in agrochemicals, pharmaceuticals, and natural products, and
thus the development of efficient, reliable methods for aryl C–O
bond formation is a topic of long-standing interest in synthetic
chemistry.[1] Traditionally, phenols have been prepared by
Sandmeyer reactions of aryldiazonium salts[2] or by direct
oxidation of arenes.[1a,3] These methods have limited substrate
scopes because the former requires an additional step to
prepare explosive diazo intermediates, and the latter requires
strong oxidants. Recently, hydroxylation of aryl halides with
catalysis by a transition-metal (TM) such as palladium or copper
has been used by a number of researchers to forge C(sp2)–O
bonds (Scheme 1a).[4] However, aryl fluorides are unsuitable
coupling partners for these reactions due to the difficulty of
oxidative addition of aryl C−F bonds to transition-metals.[5]
Aryl fluorides bearing strongly electron-withdrawing groups can
be hydroxylated straightforwardly and effectively by nucleophilic
aromatic substitution reactions with external nucleophiles,[6] but
this method is incompatible with electron-rich and neutral aryl
fluorides. For instance, heating 4-fluorotoluene with NaOH at
309 °C for 80 min gives only a 2% yield of the hydroxylation
Scheme 2. Rh-Catalyzed SNAr Hydroxylation of Aryl Fluorides.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
1
Since the first chromium(0) η6-arene complex was reported by
Fischer and Hafner in 1955,[8] chemists have made considerable
progress in SNAr reactions of aromatic compounds, that is,
umpolung aromatic substitution, by taking advantage of the
electron-withdrawing effect of η6-coordination between metals
and arenes.[9] In standard SNAr reactions, substitution reactions
of η6-coordinated arenes are widely accepted as proceeding
product 2a was obtained in 42% yield by H NMR spectroscopy.
Moreover, increasing the reaction temperature enhanced the
yield to 65%. To elucidate the active form of the catalyst, we
synthesized [Cp*Rh(CH3CN)3]2+ from cat.
7 by chloride
abstraction and used the resulting complex in the hydroxylation;
this experiment gave a comparable yield. In contrast, catalysts
with less electron-donating ligands (cat. 8, 9) gave lower yields.
stepwise[9d,9e,9j,9n] via Meisenheimer-type intermediates. However, Further optimization revealed that in addition to ethers, acetone
experimental evidence for such intermediates has yet to be
obtained; instead, η5-complexes bearing a nucleophile and a
leaving group at different positions have been reported.[10] With a
few exceptions that generally require a large excess of aromatic
compounds to drive arene-exchange in catalysis,[11] umpolung
aromatic substitutions have relied on the use of stoichiometric
η6-arene complexes as substrates, which are preformed and
require additional steps to liberate free arenes.[9] One obstacle to
realizing catalytic reactions is that some of the conditions used
for liberation of the arene from the metal center, such as
oxidation,[9f] photolysis,[9p] and ligand exchange,[9j,9l] are
incompatible with the substitutions. Specifically, η6-phenol
complexes readily undergo deprotonation under the basic
conditions typically required for nucleophilic addition; the
resulting η5-phenoxo isomers do not undergo arene
dissociation.[9j,9l,11e,12] In comparison, phenyl alkyl ethers are
unable to tautomerize into phenoxo isomers, and their use would
thus overcome the obstacle to realization of catalytic SNAr
reactions.[13] Herein, we report that we have developed a
protocol for Rh-catalyzed SNAr hydroxylation of inert aryl
fluorides without the need for a base (Scheme 2).
(a polar aprotic solvent) was also suitable, providing a slightly
lower yield (for details, see Supporting Information).
Results and Discussion
Reaction Development
To explore the feasibility of our SNAr hydroxylation strategy, we
selected 4-fluorotoluene (1a) as the limiting reagent (Scheme
3a). We began by using a Ru(II) catalyst bearing a hemilabile
phosphine ligand (cat. 1), which we recently proved to be
effective for SNAr amination reactions.[11g] However, after trying a
number of oxygen nucleophiles, such as water and hydroxide,
we failed to detect desired phenol product 2a by 1H NMR
spectroscopy. Moreover, screening of Ru catalysts bearing a
monodentate (cat. 2), tridentate PCP (cat. 3),[14] or
cyclopentadienyl (Cp) (cat. 4) ligand failed to improve the
results; 1a was recovered almost quantitatively. Because the
catalytic cycle of the envisioned process consists of arene
exchange and aromatic substitution, we tested preformed Ru(II)
η6-fluorobenzene complex 3 to gain additional insight, and found
that water was not nucleophilic enough to attack the complex;
and although hydroxide (Me4NOH) displaced the fluoride, the η5-
phenoxo complex (4) generated under the basic conditions was
too stable to liberate the phenol.
[a] 1a (0.20 mmol), nucleophile (0.60 mmol), Ru catalyst (0.01 mmol), THF
(0.20 mL), N2, 140 °C. [b] 1a (0.20 mmol), H2O (0.60 mmol), catalyst dimer
(0.005 mmol) or monomer (0.01 mmol), AgOTf (0.02 mmol), THF (0.20 mL),
N2, 140 °C. [c] 1a (0.80 mmol), H2O (2.4 mmol), cat. 7 (0.02 mmol), AgOTf
(0.08 mmol), THF (0.20 mL), N2, 150 °C. Yields were determined by 1H NMR
spectroscopy. NR, no reaction.
Scheme 3. Exploration of Reaction Conditions.
Substrate Scope of the Hydroxylations and Alkoxylations
With the optimized reaction conditions in hand, we evaluated the
scope of the catalytic SNAr hydroxylation with respect to the aryl
fluoride (Table 1). The reaction was undisturbed by steric effects,
as indicated by the fact that ortho-substituted substrates gave
the corresponding products (2c and 2d) in yields similar to that
of 2b. A broad range of meta substituents, including methoxyl
(2e), hydroxyl (2f), methyl (2g), and benzyl (2h) groups, were
tolerated. Moreover, reactive functionality such as carbonyl (2i),
carboxyl (2j), and alkenyl (2k) groups survived the reaction
conditions. Para-substituted aryl fluorides showed good
reactivity, affording 2l–2p; and multisubstituted aryl fluorides
also gave the desired products (2q–2s) in moderate yields. To
demonstrate the synthetic utility of the protocol, we tested 1s, a
We then switched to tuning the electrophilicity of the fluoroarene
complex (Scheme 3b). Specifically, we wondered whether a
metal in a higher oxidation state (e.g., Co(III), Ir(III), or Rh(III);
cat. 5–7) would facilitate nucleophilic addition of water under
nonbasic conditions required for dissociation of the product
phenols. To our delight, we found that when we used Cp*Rh
complex cat. 7, which has a rigid, facial structure, hydroxylation
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
liquid crystal compound,[15] as a substrate to obtain 2t (52%),
and synthesized the bioactive molecule estrone (2u, 50%) from
the corresponding fluoride. Heterocycles such as pyridine are
not suitable due to the coordination effect that disturbs arene
exchange (for details, see Supporting Information). Notably, the
reaction was chemoselective: the C–F bond was preferentially
cleaved over C–Cl, C–Br, C–I, and C–OTf bonds (2v–2y, >20/1
selectivity), which are more reactive in conventional TM-
Table 2. Selective Examples in Catalytic SNAr Alkoxylationa.
catalyzed
cross-coupling
reactions,
including
hydroxylation.[4e,4g,16] Because the [Cp*Rh]2+ prefers ligating with
an electron-rich aromatic ring, reaction of substrate 1z, which
has two fluorinated aromatic units, proceeded predominantly on
the more electron-rich one to afford 2z. This selectivity is
opposite that observed under classical SNAr conditions, which
give 2z.[17]
Table 1. Substrate Scope of Catalytic SNAr Hydroxylationa.
[a] Conditions: 1 (0.40 mmol), [Cp*RhCl2]2 (0.01 mmol), AgOTf (0.04 mmol),
alcohol (1.2 mmol), solvent (0.20 mL), N2, 120 °C. Isolated yields are reported.
[b] MeOH (2.0 mmol).
We also explored the application of the protocol for reactions of
aryl fluorides with alcohols to generate phenyl alkyl ethers,
which are building blocks for many natural products and
biologically active molecules. Several studies toward SNAr
alkoxylation of aryl fluorides have been reported, though the
scope of these methods remains to be extended.[13,18] By
changing the solvent, we could methoxylate aryl fluorides
bearing a wide variety of substituents (selected examples are
presented in Table 2; see Table S11 in SI for a global scope).
Notably, reactive functionality such as alkenyl (5a), amide (5b),
and imide (5c) groups were well tolerated. Multisubstituted (5e–
5f) and polycyclic (5g–5h) aromatic fluorides also afforded the
corresponding products in reasonable yields. The reaction was
chemoselective for fluorine atoms over other halogen atoms (5i–
5k). In addition, we successfully employed the protocol to
prepare aryl ethers with a deuterated methoxy group (5l, 5m).
We then examined the scope of the reaction with respect to the
alcohol by using 3-fluorophenol. Both β-branched and cyclic
primary alcohols proved to be suitable nucleophiles, affording
alkyl aryl ethers 5n–5s in moderate yields. Benzylic, secondary,
and tertiary alcohols were unsuitable owing to the possibility of
β-H elimination and acid-triggered side reactions; for example,
tetrahydro-4-pyranol gave only a 28% yield (see Table S11).
[a] Conditions: 1 (0.80 mmol), H2O (2.4 mmol), [Cp*RhCl2]2 (0.02 mmol),
AgOTf (0.08 mmol), THF (0.20 mL), N2, 150 °C. Isolated yields are reported.
[b] [Cp*RhCl2]2 (0.04 mmol), AgOTf (0.16 mmol). [c] 140 °C. [d]
[Cp*Rh(CH3CN)3](OTf)2 (0.04mmol).
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Mechanistic Insight of the Catalytic Cycle
To rationalize the activating effect of the rhodium catalyst in
SNAr reactions, we calculated the condensed local
electrophilicity indices[21] of fluorobenzene, 4-nitrofluorobenzene,
and fluorobenzene η6-complexes (Scheme 5a). Compared with
[Cp*Ru(II)]+ or nitro group, the doubly cationic Cp*Rh(III) species
serves as a much stronger π–acid, reducing the electron density
at the carbon adjacent to the fluorine dramatically. In addition,
the rhodium was found to be the predominant contributor to the
LUMO and LUMO+1 of the Cp*Rh(III) fluorobenzene complex
(Scheme 5b, over 40% versus <10% for each of other atoms).[23]
Orbital composition analysis via NAO (Natural Atomic Orbitals)
method[24] indicated that the calculated LUMO is primarily of dzx
parentage with respect to rhodium, while for LUMO+1, the main
2
2
.
contributors are dxy and dx -y
[a] η5-phenoxo Rh complex was detected in quantitative.
Scheme 4. Mechanistic Insight of the Catalytic Cycle.
[a] Condensed local electrophilicity indices were calculated on the basis of
conceptual DFT[21] with the Multiwfn program.[22a] [b] The isosurfaces of the
LUMO and LUMO+1 were obtained by analyzing the wave functions with
Multiwfn,[22a,22b] and they were drawn with the VMD program.[22c] Atom
contributions to the orbitals were calculated by means of the Hirshfield method
in Multiwfn.[22a,22d]
A simplified catalytic cycle of our SNAr hydroxylation reaction
involves two units: arene exchange and nucleophilic aromatic
substitution on η6-fluoroarene complexes (Scheme 4a). We
monitored the reaction process (Scheme 4b), and found that
when the hydroxylation of fluoromesitylene was terminated in a
shorter reaction time, the substitution product phenol together
with a significant amount of the corresponding phenol η6-
complex 6, which might be the resting state of the Rh catalyst,
were detected.[19] This observation suggested that displacement
of the substitution product (phenol) by the substrate
(fluorobenzene) on the Rh(III) center dramatically influenced the
catalysis efficiency. Moreover, an arene exchange test on Rh(III)
phenol complex 6 supported the above hypothesis (Scheme 4c).
After a mixture of 6 and excess mesitylene was heated for 2 h,
only a tiny amount of mesitylene complex 7 was detected. The
addition of a Brønsted or Lewis acid slightly promoted the
exchange,[20] whereas the addition of a base was deleterious,
owing to deprotonation to afford a more stable η5-phenoxo
complex.
Scheme 5. Activating Effect of the Rhodium Catalyst.
Mechanistic Studies of the SNAr Process
Activating Effect of the Rh Catalyst
Scheme 6. Different pathways of SNAr reaction.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Moreover, we conducted the hydroxylation under basic
conditions with hydroxide, but failed to observe any intermediate
(Figure 1b). This observation may be attributable to rapid
elimination of fluoride from the substitution intermediate, which
would be facilitated by deprotonation under the basic conditions,
leading to a dramatic reduction of the energy barrier.
Then, we moved to investigate the mechanism of SNAr process.
Nucleophilic substitutions on aromatic rings can proceed via a
concerted mechanism[25,26] or
a stepwise mechanism that
involves an intermediate (benzyne or Meisenheimer adduct,
Scheme 6).[6,27] Although SNAr reactions of metal η6-arene
complexes are thought to proceed via an addition-elimination
process (pathway c),[9d,9e,9j,9n] definitive experimental evidence
for the proposed reaction intermediate, a Meisenheimer-type
adduct, is still lacking.
Figure 1. Mechanistic studies of hydroxylation: (a) tracing the substitution
process with water as the nucleophile; (b) hydroxylation under basic conditions.
To gain insight into the substitution mechanism, we conducted
the SNAr hydroxylation of a preformed Cp*Rh fluoroarene
complex (Figure 1a, 8). At room temperature, 8 disappeared
rapidly in the presence of water as a nucleophile, as indicated by
1H NMR spectroscopy.[28] This result excludes a pathway
involving a benzyne intermediate (Scheme 5b, path b), the
formation of which requires a hydrogen next to the fluorine atom.
At a lower temperature (−40 °C) in the presence of a large
excess of water that mimics the ratio between Rh complex and
nucleophile in the catalytic hydroxylation, we observed very
weak new signals that disappeared immediately when the
reaction mixture was warmed to room temperature. Additional
Figure 2. Mechanistic studies of methoxylation: (a) tracing the substitution
process; (b) characterization of the reaction intermediate.
1
NMR analyses, including H–19F HOESY, failed to provide any
precise structural information due to its low concentration.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. Kinetics studies: (a) reaction rate dependence; (b) KIE studies.
Figure 4. Reaction coordinate diagram for SNAr methoxylation of Rh complex calculated by DFT: (a) calculated reaction coordinate diagram with two molecules of
MeOH; (b) calculated reaction coordinate diagram with one molecule of MeOH; (c) calculated reaction coordinate diagram for elimination step with MeONa.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Next, we examined substitution reactions of 8 with methanol
(Figure 2a). As was the case for the reaction with water, when
the reaction with MeOH was carried out at −30 °C, both an
intermediate and product 10 were observed. Moreover, when we
treated 8 with NaOMe, a stronger nucleophile, we observed only
the intermediate (11) that was generated in the reaction with
methanol at −30 °C. Likewise, the signals for this intermediate
disappeared at room temperature, and formation of the product
was observed (for details, see Supporting Information).
On the basis of the results of the above-described kinetics
experiments, we speculated that hydrogen bonding played a
crucial role in attack of the nucleophile and/or the loss of fluoride.
To better understand how the substitution reaction proceeds, we
performed density functional theory (DFT) calculations at the
PWPB95-D3(BJ)/ma-Def2-QZVPP//B3LYP-D3(BJ)/ma-Def2
TZVP(-f) level[29a–g] with the SMD implicit solvent model.[29h]
We began by modeling the reaction coordinate diagram for the
substitution of
a doubly cationic Cp*Rh(III) fluorobenzene
To determine the structure of 11, we isolated it with a
considerable effort, and characterized it as being a Rh η5-
cyclohexadienyl complex with an sp3-hybridized carbon bearing
a OMe group (the nucleophile) and a fluoride (the leaving group)
(Figure 2b; see Supporting Information for more spectra).
Moreover, the structure of 11 was further confirmed by high-
complex with two molecules of MeOH (Figure 4a). The reaction
commences with nucleophilic addition of one methanol with
assistance from the other molecule by means of hydrogen
bonding, which generates Rh η5-cyclohexadienyl complex Int1-1.
This intermediate does not undergo elimination directly; instead,
the O···H bond breaks and a F···H bond forms (conversion of
Int1-1 to Int1-2), as indicated by ab initio molecular dynamics
calculations (see Supporting Information for details). Then the
Int1-2 liberates fluoride to yield the substitution product P1 via
an early transition state (TS1-2) bearing an approximately sp3
hybridized carbon (in the transition state, ∠O–C1–C2 = 115.9°,
and in the precursor, ∠O–C1–C2 = 113.1°). We also evaluated a
reaction pathway involving a single methanol molecule (Figure
4b) and observed only a concerted pathway with a much higher
energy barrier (ΔG‡ = 47.6 kcal/mol). In addition, the effect of
hydrogen bonding in the elimination step was further
emphasized by comparison of this reaction profile (2.7 kcal/mol
energy barrier from Int1-2 to TS1-2) with that of a reaction with
methoxide (NaOMe) as the nucleophile (Figure 4c). In the
pathway for reaction of NaOMe, a higher energy barrier to the
elimination step was obtained (9.5 kcal/mol), probably owing to a
weaker interaction between the fluoride anion and the sodium
cation in the transition state.[26] This result is consistent with the
fact that the Meisenheimer-type adduct 11 that formed in the
resolution
mass
spectrometry
([(η5-F-OMe-
trimethylcyclohexadienyl)Rh(III)Cp*]+; calcd m/z = 407.1258;
found 407.1271). The khaki complex (11) is not stable at
ambient temperature, degrading to the complex 10 gradually. To
our knowledge, this is the first time that a Meisenheimer-type
intermediate of a transition-metal-mediated SNAr reaction has
been characterized.
In addition, we carried out kinetics studies to obtain further
insight into the mechanism. First, we measured the rate
dependence on the nucleophile concentration by conducting a
substitution reaction of 8 in acetone, and we found that the rate
dependence was second order for both water and methanol
(Figure 3a). Second, we observed kinetic isotope effects of 1.44
and 2.65 for D2O and MeOD, respectively (Figure 3b).
DFT Calculations of the SNAr Process
methoxylation with NaOMe could be isolated at
temperature.
a low
Figure 5. Reaction coordinate diagram for SNAr hydroxylation of Rh complex calculated by DFT: (a) calculated intrinsic reaction coordinate with two molecules of
H2O; (b) Mayer bond order analysis for C-O bond formation and C-F bond cleavage.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
[1]
a) Z. Rappoport, The Chemistry of Phenols; Wiley-VCH Verlag GmbH
& Co. KGaA: Weinheim, 2003; b) J. H. P. Tyman, Synthetic and
Natural Phenols; Elsevier: New York, 1996; c) S. D. Roughley, A. M.
Jordan, J. Med. Chem. 2011, 54, 3451−3479; d) W. Schutyser, T.
Renders, S. Vanden Bosch, S. F. Koelewijn, G. T. Beckham, B. F. Sels,
Chem. Soc. Rev. 2018, 47, 852−908.
Next, we carried out calculations for the hydroxylation of the
same complex, and found that as opposed to methoxylation, no
obvious Meisenheimer-type intermediate could be located on the
intrinsic reaction coordinate (Figure 5a), implying a possibility for
concerted mechanism. A concerted SNAr mechanism should
include a transition state in which both a nucleophile and a
leaving group are attached to the arene by partial rather than full
bonds.[25,26] However, Mayer bond order analysis[22a,30] of the
reaction (Figure 5b) revealed a stepwise-like mechanism in
which the C–O bond (black line) forms prior to cleavage of the
C–F bond (red line). In 2018, Jacobsen et al. reported similar
results in their study of SNAr fluorination of 2,4-
dinitrochlorobenzene that kinetic isotope effect analysis by
12C/13C NMR spectroscopy, as well as computational analyses,
supported an addition–elimination pathway somewhere between
stepwise and concerted.[25g] This behavior is observed because
the putative intermediate is stabilized by a strongly electron-
withdrawing group but destabilized by the presence of a good
leaving group. In our case, the liberation of fluoride is facilitated
by hydrogen bonding between fluoride and water (Figure 5b,
blue line).
[2]
[3]
a) T. Sandmeyer, Ber. Dtsch. Chem. Ges. 1884, 17, 1633–1635; b) T.
Sandmeyer, Ber. Dtsch. Chem. Ges. 1884, 17, 2650–2653.
For recent examples of oxidation of arenes, see: a) P. Borah, X. Ma,
K. T. Nguyen, Y. Zhao, Angew. Chem. Int. Ed. 2012, 51, 7756–7761;
b) K. Kamata, T. Yamaura, N. Mizuno, Angew. Chem. Int. Ed. 2012, 51,
7275–7278; c) K. Ohkubo, A. Fujimoto, S. Fukuzumi, J. Am. Chem.
Soc. 2013, 135, 5368−5371; d) O. Shoji, T. Kunimatsu, N. Kawakami,
Y. Watanabe, Angew. Chem. Int. Ed. 2013, 52, 6606–6610; e) M.
Yamada, K.D. Karlin, S. Fukuzumi, Chem. Sci., 2016, 7, 2856–2863; f)
H. Huang, T. H. Lambert, Angew. Chem. Int. Ed. 2021, 60, 11163–
11167.
[4]
For reviews of metal-catalyzed hydroxylation of (hetero)aryl
halides, see: a) S. Enthaler, A. Company, Chem. Soc. Rev. 2011, 40,
4912−4924; b) A. Tlili, M. Taillefer, In Copper-Mediated Cross-Coupling
Reactions; Evano, E., Blanchard, N., Eds.; John Wiley & Sons Inc.:
Hoboken, NJ, USA, 2013; pp 41−91. For selected Pd-catalyzed
hydroxylation of (hetero)aryl halides, see: c) K. W. Anderson, T.
Ikawa, R. E. Tundel, S. L. Buchwald, J. Am. Chem. Soc. 2006, 128,
10694−10695; d) . chulz, C. orborg, . chaffner, . Huang, A.
apf, R. adyrov, A. rner, M. Beller, Angew. Chem., Int. Ed. 2009,
48, 918−921. For selected Cu-catalyzed hydroxylation of
(hetero)aryl halides, see: e) A. Tlili, N. Xia, F. Monnier, M. Taillefer,
Angew. Chem., Int. Ed. 2009, 48, 8725−8728; f) D. Zhao, N. Wu, S.
Zhang, P. Xi, X. Su, J. Lan, J. You, Angew. Chem., Int. Ed. 2009, 48,
8729−8732; (g) S. Xia, L. Gan, K. Wang, Z. Li, D. Ma, J. Am. Chem.
Soc. 2016, 138, 13493−13496.
Conclusion
In conclusion, we have developed a protocol for Rh-catalyzed
SNAr hydroxylation and alkoxylation reactions of aryl fluorides as
the limiting reagents. Experimental studies and DFT calculations
provided detailed information about the mechanism of these
[5]
[6]
a) P. Fitton, E. Rick. J. Organomet. Chem. 1971, 28, 287–291; b) M .F.
Semmelhack, L. Ryono. Tetrahedron Lett. 1973, 31, 2967–2970; c) J. K.
Stille, K. S. Y. Lau. Acc. Chem. Res. 1977, 10, 434–442.
substitution
reactions.
Notably,
a
Meisenheimer-type
intermediate was isolated and well-characterized, providing solid
evidence for the proposed mechanism. Further work aimed at
extending this protocol to related transformations is currently
underway in our laboratory.
Terrier, F. Modern Nucleophilic Aromatic Substitution. WileyVCH,
Weinheim, 2013, e-book ISBN: 9783527656141.
[7]
[8]
[9]
M. Zoratti, J. F. Bunnett, J. Org. Chem. 1980, 45, 1769-1776.
E. O. Fischer, W. Z. Hafner, Naturforsch. 1955, 10b, 665.
For selected reviews, see: a) R. D. Pike, D. A. Sweigart, Coordination
Chemistry Reviews 1999, 187, 183-222; b) A. R. Pape, K. P. Kaliappan,
E. P. Kündig, Chem. Rev. 2000, 100, 2917−2940; c) F. C. Pigge, J. J.
Coniglio, Current Organic Chemistry 2001, 5, 757−784; d) M. F.
Semmelhack, A. Chlenov, Top. Organomet. Chem. 2004, 7, 43. For
selected examples, see: e) M. F. Semmelhack, G. Hilt, J. H. Colley,
Tetrahedron Lett. 1998, 39, 7683−7686; f) K. Kamikawa, S. Kinoshita,
M. Furusyo, S. Takemoto, H. Matsuzaka, M. Uemura, J. Org. Chem.
2007, 72, 3394−3402; g) W. Braun, B. Calmuschi-Cula, U. Englert, K.
Höfener, E. Alberico, A. Salzer, Eur. J. Org. Chem. 2008, 2065–2074;
h) G. I. McGrew, J. Temaismithi, P. J. Carroll, P. J. Walsh, Angew.
Chem. Int. Ed. 2010, 49, 5541–5544; i) G. I. McGrew, C. Stanciu, J.
Zhang, P. J. Carroll, S. D. Dreher, P. J. Walsh, Angew. Chem. Int. Ed.
2012, 51, 11510–11513; j) A. J. M. Miller, W. Kaminsky, K. I. Goldberg,
Organometallics 2014, 33, 1245−1252; k) S. Shirakawa, K. Yamamoto,
K. Maruoka, Angew. Chem. Int. Ed. 2015, 54, 838–840; l) E. M.
D’Amato, C. N. Neumann, . Ritter, Organometallics 2015, 34,
4626−4631; m) J. Mao, J. Zhang, H. Jiang, A. Bellomo, M. Zhang, Z.
Gao, S. D. Dreher, P. J. Walsh, Angew. Chem. Int. Ed. 2016, 55, 2526–
2530; n) M. H. Beyzavi, D. Mandal, M. G. Strebl, C. N. Neumann, E. M.
D’Amato, . Chen, . M. Hooker, . Ritter, ACS Cent. Sci. 2017, 3,
944−948; o) J. A. Pike, J. W. Walton, Chem. Commun. 2017, 53,
9858−9861; p) L. A. Wilkinson, J. A. Pike, J. W. Walton,
Organometallics 2017, 36, 4376−4381.
Acknowledgements
We are grateful for financial support from the National Natural
Science Foundation of China (22071198), the Leading
Innovative and Entrepreneur Team Introduction Program of
Zhejiang (2020R01004), the China Postdoctoral Science
Foundation
Postdoctoral
(2019M662118,
projects funded
2019M662119),
by Zhejiang
and
Province
the
(G02146521901). We thank Instrumentation and Service Center
for Molecular Sciences at Westlake University for the assistance
work in measurement/data interpretation. We thank Dr. Xiaohuo
Shi, Dr. Xingyu Lu and Dr. Yinjuan Chen for the assistance work
in NMR and HRMS measurement/data interpretation of
intermediate 11, and Prof. Xin Hong (Zhejiang University) for
helpful discussion.
Conflict of interest
[10] L. I. Goryunov, N. M. Romanova, G. S. Zhilovskii, V. D. Shteingarts,
Russ. J. Org. Chem., 2007, 43, 1765−1771.
The authors declare no conflict of interest.
[11] For catalytic SNAr reactions via η6-coordination, see: a) L. A.
Wilkinson, J. W. Walton, Organomet. Chem. 2019, 42, 125–171; b) S.
Takemoto, H. Matsuzaka, Tetrahedron Letters 2018, 59, 697–703; c) M.
Otsuka, K. Endo, T. Shibata, Chem. Commun., 2010, 46, 336–338; d)
Keywords: nucleophilic aromatic substitution, Meisenheimer-
type intermediate, η6-coordination, rhodium, C-O bond formation
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
M. Otsuka, H. Yokoyama, K. Endo, T. Shibata, Synlett 2010, 17, 2601–
2606; e) A. I. Konovalov, E. O. Gorbacheva, F. M. Miloserdov, V. V.
Grushin, Chem. Commun., 2015, 51, 13527–13530; f) J. W. Walton, J.
M. J. Williams, Chem. Commun., 2015, 51, 2786−2789; g) Q.-K. Kang,
Y. Lin, Y. Li, H. Shi, J. Am. Chem. Soc. 2020, 142, 3706−3711.
[12] A. Igau, Coordination Chemistry Reviews 2017, 344, 299–322.
[13] a) R. P. Houghton, M. Voyle, R. Price, J. Chem. Soc., Perkin Trans. I
1984, 925; b) L. I. Goryunov, V. V. Litvak, V. D. Shteingarts, Zhurn. Org.
Khim. 1987, 23, 1230−1237.
J Comput Chem, 2011, 32, 1456–1465; e) F. Weigend, R. Ahlrichs,
Phys. Chem. Chem. Phys. 2005, 7, 3297–3305; f) L. Goerigk, S.
Grimme, J. Chem. Theory Comput. 2011, 7, 291–309; g) J. Zheng, X.
Xu, D. G. Truhlar, Theor. Chem. Acc. 2011, 128, 295–305; h) A. V.
Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B, 2009, 113,
6378–6396.
[30] a) I. Mayer, Chem. Phys. Lett. 1983, 97, 270−274; b) Software for
extracting each frame of the intrinsic reaction coordinate (SplitIRC) was
developed by T. Lu and is available at http://sobereva.com/200.
[14] J. Takaya, J. F. Hartwig, J. Am. Chem. Soc., 2005, 127, 5756–5757.
[15] P. Krisch, M. Bremer, Angew. Chem. Int. Ed. 2000, 39, 4216–4235.
[16] K. Daasbjerg, J. Chem. Soc., Perkin Trans. 2 1994, 1275−1277.
[17] P. S. Fier, K. M. Maloney, Org. Lett. 2016, 18, 2244−2247.
[18] a) J. Su, Q. Chen, L. Lu, Y. Ma, G. H. L. Auyoung, R. Hua, Tetrahedron
2018, 74, 303−307; b) J. R. Rodriguez, J. Agejas, A. B. Bueno,
Tetrahedron Lett. 2006, 47, 5661−5663.
[19] Fluoromesitylene η6-complex was not observed due to its high reactivity
towards H2O (see Figure 1a).
[20] Adding Brønsted or Lewis acid did not improve the catalytic reaction.
[21] The local electrophilicity index contains information obtainable from
hitherto known different global and local reactivity and selectivity
descriptors,
in
addition
to
the
information
regarding
electrophilic/nucleophilic power of a given atomic site in a molecule,
see: a) R. K. Roy, S. Krishnamurti, P. Geerlings, S. Pal, J. Phys. Chem.
A 1998, 102, 3746–3755; b) P. K. Chattaraj, B. Maiti, U. Sarkar, J. Phys.
Chem. A, 2003, 107, 4973–4975. For selective applications, see: c) P.
Sharma, A. Kumar, V. Sahu. J. Phys. Chem. A 2010, 114, 1032–1038;
d) M. Ahuja, G. Kour Reen, A. Kumar, P. Sharma, Chem. Lett. 2016, 45,
752–754; e) E. Chamorro, M. Duque-Noreña, M. Ríos-Gutiérrez, L. R.
Domingo, P. Pérez. Theor. Chem. Acc. 2016, 135, 202–214; f) A.
Makhloufi, O. Belhadad, R. Ghemit, M. Baitiche, M. Merbah, DJ.
Benachour. J. Mol. Struc. 2018, 1152, 248–256.
[22] a) T. Lu, F. Chen, J. Comput. Chem. 2012, 33, 580–592; b) T. Lu, F.
Chen, J. Mol. Graph. Model., 2012, 38, 314–323; c) W. Humphrey, A.
Dalke, K. Schulten, J. Molec. Graphics 1996, 14, 33–38; (d) T. Lu, F.
Chen, Acta Chim. Sinica, 2011, 69, 2393–2406.
[23] The LUMO and LUMO+1 might be viewed as degenerate orbitals owing
to the tiny energy gap [0.018eV].
[24] a) F. Weinhold, In Encyclopedia of Computational Chemistry, Vol. 2,
Ed.: Schleyer, P. v. R., John Wiley & Sons, West Sussex, 1998, 1792-
1811; b) T. Lu, F.-W. Chen, Acta Chim. Sinica, 2011, 69, 2393-2406.
[25] For selected reviews, see: a) C. N. Neumann, T. Ritter, Acc. Chem.
Res. 2017, 50, 2822−2833; b) S. Rohrbach, A. J. Smith, J. H. Pang, D.
L. Poole, T. Tuttle, S. Chiba, J. A. Murphy, Angew. Chem. Int. Ed. 2019,
58, 16368–16388. For selected examples, see: c) C. N. Neumann, J.
M. Hooker, T. Ritter, Nature 2016, 534, 369−373; d) S. D. Schimler, M.
A. Cismesia, P. S. Hanley, R. D. J. Froese, M. J. Jansma, D. C. Bland,
M. S. Sanford, J. Am. Chem. Soc. 2017, 139, 1452−1455; e) D. Y. Ong,
C. Tejo, K. Xu, H. Hirao, S. Chiba, Angew. Chem. Int. Ed. 2017, 56,
1840–1844; f) A. Kaga, H. Hayashi, H. Hakamata, M. Oi, M. Uchiyama,
R. Takita, S. Chiba, Angew. Chem. Int. Ed. 2017, 56, 11807 –11811; g)
E. E. wanꢀ, Y. eng, H. A. esser, E. N. acobsenꢀ, Nat. Chem. 2018,
10, 917−923; h) S. Mallick, P. Xu, E.-U. Würthwein, A. Studer, Angew.
Chem. Int. Ed. 2019, 58, 283–287; i) X.-W. Liu, C. Zarate, R. Martin,
Angew. Chem. Int. Ed. 2019, 58, 2064–2068; i) Z. You, K. Higashida, T.
Iwai, M. Sawamura, Angew. Chem. Int. Ed. 2021, 60, 5778–5782.
[26] S. Rohrbach, J. A. Murphy, T. Tuttle, J. Am. Chem. Soc. 2020, 142,
14871–14876.
[27] a) J. A. Garcia-López, M. F. Greaney, Chem. Soc. Rev. 2016, 45,
6766–6798; b) H. Takikawa, A. Nishii, T. Sakai, K. Suzuki, Chem. Soc.
Rev. 2018, 47, 8030–8056.
[28] When the reaction proceeded in the presence of a large amount of
water (100 equiv.), η5-phenoxo complex 9 was detected instead of
phenol complex 6. This phenomenon may be attributed either to
hydrogen bonding by excess water as a donor or to solvent effects.
[29] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) P. J. Stephens,
F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98,
11623–11627; c) S. Grimme, J. Antony, S. Ehrlich, H. A. Krieg, J.
Chem. Phys. 2010, 132, 154104; d) S. Grimme, S. Ehrlich, L. Goerigk,
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202106440
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
A rhodium catalyzed hydroxylation of aryl fluorides via η6-coordination was achieved. This method was also applicable in the
alkoxylation. A Meisenheimer-type intermediate was successfully isolated and well characterized. Density functional theory
calculations and mechanistic studies suggested a stepwise or stepwise-like energy profile for the SNAr reaction.
10
This article is protected by copyright. All rights reserved.