Aromatic C-H amination in hexafluoroisopropanol

Article abstract of DOI:10.1039/C8SC04966A

We report a direct radical aromatic amination reaction that provides unprotected anilines with an improvement in the substrate scope compared to prior art. Hydrogen bonding by the solvent hexafluoroisopropanol to anions of cationic species is responsible for increased reactivity and can rationalize the enhancement in substrate scope. Our findings may have bearings on radical additions to arenes for direct C-H functionalization in general.

Full text of DOI:10.1039/C8SC04966A

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Aromatic C–H amination in
hexafluoroisopropanol†
Cite this: DOI: 10.1039/c8sc04966a
b
ab
Erica M. D'Amato,a Jonas Borgel
and Tobias Ritter
*
All publication charges for this article
have been paid for by the Royal Society
of Chemistry
¨
We report a direct radical aromatic amination reaction that provides unprotected anilines with an
improvement in the substrate scope compared to prior art. Hydrogen bonding by the solvent
hexafluoroisopropanol to anions of cationic species is responsible for increased reactivity and can
rationalize the enhancement in substrate scope. Our findings may have bearings on radical additions to
arenes for direct C–H functionalization in general.
Received 7th November 2018
Accepted 21st December 2018
DOI: 10.1039/c8sc04966a
rsc.li/chemical-science
Aryl amines are broadly useful in the pharmaceutical, agro-
chemical and material science elds and have been tradition-
Introduction
ally synthesized on scale using an electrophilic nitration/
reduction sequence.11–14 While most modern aromatic C–H
amination methods15–27 have failed to match the scope of elec-
trophilic nitration, reaction development has succeeded in
increasing functional group tolerance, reducing the two-step
sequence to a one-step process, and improving the safety of
the reaction. To date, only the addition of pyridinium radicals to
arenes features a broad substrate scope of anilines in a 2-step
Radical addition to arenes is an attractive strategy for aromatic
C–H functionalization because it avoids the difficult C–H met-
alation step that is common to several C–H functionalization
processes.1 Instead, radical addition can be followed by a facile
C–H deprotonation as the nal step. Radical addition to arenes
has been used to install a variety of functional groups onto
aromatic rings, including aryl, alkyl, hydroxyl and amino
groups, and oen delivers multiple isomeric products, which
can be useful for small molecule diversication.2–6 However, the
radical must be matched in polarity with the arene for
a productive reaction.7–10 Therefore, the substrate scope of any
particular radical addition is typically small, with electrophilic
radicals reacting with nucleophilic arenes and vice versa.
Herein, we demonstrate and rationalize the previously unap-
preciated, scope-expanding effect of the solvent hexa-
uoroisopropanol (HFIP) on a radical aromatic C–H amination
that provides free anilines in a single step. Our reaction protocol
enables efficient amination of electron-poor arenes, such as
nitrobenzene (Fig. 1a). We discuss how ion pair disruption
through specic hydrogen bonding interactions with HFIP can
expand the substrate scope of a radical C–H functionalization
and can obviate the need for the metal catalyst used in
conventional reactions. Our hypothesis may provide a concep-
tual framework for the development of other, new radical
addition reactions, and in the expansion of their scope.
aDepartment of Chemistry and Chemical Biology, Harvard University, 12 Oxford
Street, Cambridge, Massachusetts 02138, USA
Fig. 1 (a) [Fe]-catalyzed amination of nitrobenzene. (b). Substrate
scope of this work compared to prior art. aCombined yield of isolated
analytically pure individual isomers. Ratio A : B : C ¼ 2.4 : 1.0 : 1.0.
bWhile s values cannot be used to compare reactions proceeding
through different mechanisms, they do provide a semi-quantitative
measure of arene electron density.
b
¨
¨
Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mulheim
an der Ruhr, Germany. E-mail: ritter@mpi-muelheim.mpg.de
† Electronic supplementary information (ESI) available: Detailed experimental
procedures and spectroscopic characterization for all new compounds. CCDC
1545194. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c8sc04966a
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 2 (a) Mechanistic hypothesis for aromatic C–H amination. (b) X-ray crystal structure of 1 crystallized from HFIP. (c) Calculated structure of 1
with explicit HFIP molecules. (d) Comparison of [Fe]-catalyzed and metal-free reactions. aStructure shown with 50% probability ellipsoids.‡
sequence including electron-poor arenes.28,29 Advances in C–H and HFIP as solvent affords unprotected anilines from aromatic
amination to provide unprotected anilines have been reported C–H bonds across an electronic range of arenes broader in
by Nicewicz,30 under whose conditions ammonium carbamate scope than any reported modern aromatic C–H amination
¨
traps an arene radical cation intermediate, and by Kurti and reaction. The reaction is characterized by a simple setup that
Falck,31 under whose conditions a rhodium nitrene interme- does not require any special precautions to exclude air or
diate is proposed to insert into a C–H bond. A third approach to moisture, and by reaction times shorter than 2 h. In addition,
direct aromatic C–H amination was pioneered by Minisci in the multiple iron sources, including ferrocene and FeSO4$7H2O, are
1960s, in which hydroxylamine-O-sulfonic acid (HOSA) is used competent for the reaction. For example, on a 2.0 g scale,
as an ammoniumyl radical precursor in the presence of iron(II) nitrobenzene is aminated in 86% yield within 45 min with
salts.2,32 In 2016, Morandi revised the Minisci protocol with the 1.0 mol% iron loading (Fig. 1a). Each constitutional isomer was
use of the reagent [MsO–NH3]OTf (1).33 Moreover, in 2017, Jiao isolated as individual, analytically pure sample.
has demonstrated that ammoniumyl radicals, which were
The expansion of the scope is attributed to the unique
generated from different [RCO2–NH3]OTf reagents, add to properties of the solvent HFIP, including high polarity, low
a variety of electron-rich arenes in TFE/H2O.34 However, all re- nucleophilicity, and strong hydrogen bond-donating ability.35,36
ported modern amination methods to make anilines in a one- As a result, the use of HFIP has been shown to have a great effect
step procedure break down if an electron-poor arene is used on reactivity and/or selectivity in a number of reactions.36–42 We
as a substrate (Fig. 1b). For example, no reaction has been re- propose that HFIP increases the electrophilicity of several
ported to afford more than 5% conversion with benzonitrile as cationic species in our reaction through hydrogen-bonding with
a substrate, and most reactions do not afford synthetically their anionic counterions, which in turn leads to effective
useful yields for arenes less electron-rich than bromobenzene. amination of more electron-poor arenes. Such an increase in
reactivity through hydrogen bonding has not been demon-
strated previously for aromatic amination. Based on our
experiments, we propose the mechanism shown in Fig. 2a: an
Results and discussion
ammoniumyl radical, generated either through N–O bond
Herein, we show that the combination of the easy-to-handle
homolysis or iron-mediated single electron reduction, adds to
hydroxylamine-derived reagent 1, 1.0 mol% iron(II) catalyst,
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
Table 1 Aromatic C–H amination in hexafluoroisopropanol
localized cation on nitrogen. A second hydrogen bond can occur
between an oxygen of the mesyl group and HFIP. Consequently,
the cation of 1 is more electrophilic when it is dissolved in
HFIP.§ DFT calculations with explicit HFIP solvent molecules
support our hypothesis and show that the LUMO of reagent 1 is
7.7 kcal molꢀ1 lower in energy when both hydrogen-bonding
interactions are present (Fig. 2c). The increased reactivity of
reagent 1 in HFIP is conrmed by its reduction potential, which
we measured by cyclic voltammetry in both HFIP (ꢀ0.77 V vs.
Fc/Fc+) and MeCN (ꢀ1.28 V vs. Fc/Fc+) (see ESI†). The difference
in the reduction potential of reagent 1 in the two solvents is
ꢁ0.5 V and indicates that the reagent is a notably stronger
oxidant in HFIP, which supports our hydrogen-bonding
hypothesis. In addition, attempts to synthesize derivatives of 1
that contain counterions less capable of hydrogen bonding (e.g.
PF6ꢀ) were unsuccessful, which suggests that the hydrogen
bond between the triate counterion and the reagent is an
important stabilizing factor for 1 before it is activated by
dissolution in HFIP (see ESI†).
HFIP may not only have an effect on the reactivity of 1 but
also may increase the reactivity of cationic reaction intermedi-
ates, specically the ammoniumyl radical and intermediate A
(Fig. 2a), through hydrogen-bonding interactions with their
triate counterions. Decreased ion pairing and the lack of
a hydrogen-bond accepting solvent would lower the LUMO of
the ammoniumyl radical derived from the cation of 1 and would
enable addition to more electron-poor arenes. Similarly,
increased reactivity of the cationic cyclohexadienyl radical A
would enable more efficient oxidation to the nal product. The
overall increased electrophilicity of 1, the ammoniumyl radical,
and A would synergize to give the much improved substrate
scope presented herein.
HFIP also enables a metal-free reaction (<1 ppb Fe detected)
to occur (Fig. 2d). Metal-free activation does not occur under
any previously reported conditions for ammoniumyl radical
addition to arenes but is viable due to the activating properties
of HFIP, albeit with longer reaction times. Such a background
reaction pathway is a common feature of radical chain reac-
tions.51 Under metal-free conditions, we identied N–O bond
homolysis as a likely initiation step to generate the ammo-
niumyl radical. The N–O bond homolysis energy was calculated
using DFT (uB97XD) to be 35 kcal molꢀ1 (Fig. 2c). N–O bond
homolysis can therefore be considered feasible under the
reaction conditions, regardless of the presence of an iron salt.
Rearomatization of cyclohexadienyl radical A could then occur
either by aerobic oxidation or chain propagation.
a
c
b
Performed at 40 ꢂC. Performed under an atmosphere of oxygen.
TfOH (1.00 equiv.) added.
an arene to generate a putative cationic cyclohexadienyl radical
(A). Intermediate A then rearomatizes to the aniline product
through one of three pathways: single electron oxidation by
iron(III),43–45 aerobic oxidation46–50 or chain propagation.51
We identied a hydrogen bond in reagent 1 between one
N–H and the triate counterion in the solid state (Fig. 2b). HFIP
likely disrupts the internal hydrogen-bonding and ion pairing
of 1, because it functions as a hydrogen-bond donor to the tri-
ate counterion but cannot function as a hydrogen-bond
acceptor to [MsO–NH3]+.52,53 The result of the HFIP–triate
hydrogen bond is a less associated ion pair with a more
When FeSO4$7H2O is present in the reaction, formation of
the ammoniumyl radical can occur through single electron
reduction of reagent 1. A third pathway for rearomatization then
becomes available – single electron oxidation of A by iron(III)
generated in the reduction of 1. Iron(III) has been shown capable
of oxidizing cyclohexadienyl radicals,43–45 but whether the
iron(II) salt in our reaction is turning over as a catalyst or acting
as a radical chain initiator cannot be discerned from our data.
Based on our discovery of the benecial effects of HFIP on
radical C–H amination, we synthesized a number of anilines
utilizing our new protocol with a primary focus on electron-
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
decient arenes (Table 1). While previous methods demonstrate
efficient amination of arenes no more electron poor than bro-
mobenzene to provide unprotected anilines, our method is
suitable for the amination of arenes such as nitrobenzene (2),
methyl phenyl sulfone (3), and benzonitrile (4). Reactivity is
maintained with electron-rich arenes as well (see ESI†). Most
halides are tolerated (5, 10, 11, 13), as are tertiary amines (11)
and benzylic C–H bonds (7, 13). Amination can occur on ve-
membered heterocycles (6) and on benzofused ve- and six-
membered heterocycles (7, 9). However, no amination has
been observed on six-membered heterocycles. While esters (6,
9), amides (11, 13), nitriles (10, 4), and sulfonamides (8) are
suitable substrates, aldehydes, ketones, and alkenes typically
undergo side reactions without appreciable ring amination.
While for some densely functionalized substrates, no or low
2 F. Minisci, Synthesis, 1973, 1–24.
3 R. Bolton and G. H. Williams, Chem. Soc. Rev., 1986, 15, 261–
289.
4 A. Studer, Angew. Chem., Int. Ed., 2012, 51, 8950–8958.
´
5 R. Rossi, M. E. Buden and J. F. Guastavino, in Arene
Chemistry: Reaction Mechanisms and Methods for Aromatic
Compounds, ed. J. Mortier, John Wiley & Sons, Hoboken,
2016, pp. 219–242.
¨
6 J. Borgel, L. Tanwar, F. Berger and T. Ritter, J. Am. Chem. Soc.,
2018, 140, 16026–16031.
7 A. Clerici, F. Minisci, M. Perchinunno and O. Porta, J. Chem.
Soc., Perkin Trans. 2, 1974, 416–419.
8 M. Tiecco, Pure Appl. Chem., 1981, 53, 239–358.
9 J. Fossey, D. Lefort and J. Sorba, Free Radicals in Organic
Chemistry, John Wiley & Sons, New York, 1995.
conversion of starting material was observed (see Table S6†), we 10 A. Bravo, H.-R. Bjørsvik, F. Fontana, L. Liguori, A. Mele and
demonstrated the utility of our method to late-stage function- F. Minisci, J. Org. Chem., 1997, 62, 7128–7136.
alization of drug molecules such as moclobemide and runa- 11 J. G. Hoggett, R. B. Moodie, J. R. Penton and K. Schoeld,
mide, which were aminated to give derivatives 11 and 13,
respectively.
Nitration and aromatic reactivity, Cambridge University
Press, New York, 1971.
12 G. A. Olah, R. Malhotra and S. C. Narang, Nitration Methods
and Mechanism, VCH Publishers, Inc., New York, 1989.
13 H.-U. Blaser, H. Steiner and M. Studer, ChemCatChem, 2009,
1, 210–221.
14 L. Lu, H. Liu and R. Hua, Org. Lett., 2018, 20, 3197–3201.
15 M.-L. Louillat and F. W. Patureau, Chem. Soc. Rev., 2014, 43,
901–910.
16 J. Jiao, K. Murakami and K. Itami, ACS Catal., 2016, 6, 610–
633.
17 H. Kim and S. Chang, ACS Catal., 2016, 6, 2341–2351.
18 L. Legnani, B. N. Bhawal and B. Morandi, Synthesis, 2017, 49,
776–789.
Conclusions
We present a practical aromatic C–H amination reaction and
provide a mechanistic framework for understanding the effect
of the solvent HFIP on the reaction. Though aminiumyl radical
additions have been known for over half a century, the mech-
anistic insight presented herein has resulted in a previously
unrealized reaction utility. HFIP is proposed to comprise
a unique solvent environment that increases the electrophilicity
of multiple cationic species in the reaction to provide a drasti-
cally expanded substrate scope. We anticipate that our ndings
will inform further investigation and development of radical
addition reactions for aromatic C–H functionalization.
19 Y. Park, Y. Kim and S. Chang, Chem. Rev., 2017, 117, 9247–
9301.
20 G. B. Boursalian, M.-Y. Ngai, K. N. Hojczyk and T. Ritter,
J. Am. Chem. Soc., 2013, 135, 13278–13281.
21 L. J. Allen, P. J. Cabrera, M. Lee and M. S. Sanford, J. Am.
Chem. Soc., 2014, 136, 5607–5610.
Conflicts of interest
There are no conicts to declare.
´
22 K. Foo, E. Sella, I. Thome, M. D. Eastgate and P. S. Baran,
J. Am. Chem. Soc., 2014, 136, 5279–5282.
23 H. Kim, T. Kim, D. G. Lee, S. W. Roh and C. Lee, Chem.
Commun., 2014, 50, 9273–9276.
24 L. Song, L. Zhang, S. Luo and J.-P. Cheng, Chem.–Eur. J.,
2014, 20, 14231–14234.
Acknowledgements
We thank Claudia Kleinlein for assistance with electro-
chemistry, and Dr Shao-Liang Zheng and Claudia Kleinlein
for assistance with X-ray crystallography. J. B. acknowledges the
Fond der Chemischen Industrie for funding. We acknowledge
UCB Pharma for funding.
25 T. W. Greulich, C. G. Daniliuc and A. Studer, Org. Lett., 2015,
17, 254–257.
26 T. Kawakami, K. Murakami and K. Itami, J. Am. Chem. Soc.,
2015, 137, 2460–2463.
27 G. B. Boursalian, W. S. Ham, A. R. Mazzotti and T. Ritter,
Nat. Chem., 2016, 8, 810–815.
28 W. S. Ham, J. Hillenbrand, J. Jacq, C. Genicot and T. Ritter,
Angew. Chem., Int. Ed., 2019, 58, 532–536.
Notes and references
‡ Cambridge Crystallographic Data Centre (CCDC) number of compound 1:
1545194.
§ A mixture of MeCN and H2O, as was used in ref. 33, would not have the same
effect on 1, as the solvent can both donate and accept hydrogen bonds, and the
zwitterionic nature of the reagent (HOSA) used by Minisci (ref. 32) mitigates any
such disruption of ion pairing.
¨
29 S. Rossler, B. Jelier, P. Tripet, A. Shemet, G. Jeschke, A. Togni
and E. M. Carreira, Angew. Chem., Int. Ed., 2019, 58, 526–531.
30 N. A. Romero, K. A. Margrey, N. E. Tay and D. A. Nicewicz,
Science, 2015, 349, 1326.
1 N. Kuhl, M. N. Hopkinson, J. Wencel-Delord and F. Glorius,
Angew. Chem., Int. Ed., 2012, 51, 10236–10254.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
31 M. P. Paudyal, A. M. Adebesin, S. R. Burt, D. H. Ess, Z. Ma, 43 K. Bahtia and R. H. Schuler, J. Phys. Chem., 1974, 78, 2335–
¨
L. Kurti and J. R. Falck, Science, 2016, 353, 1144–1147.
2338.
32 A. Citterio, A. Gentile, F. Minisci, V. Navarrini, M. Serravalle 44 R. F. Anderson, Radiat. Phys. Chem., 1979, 13, 155–157.
and S. Ventura, J. Org. Chem., 1984, 49, 4479–4482.
33 L. Legnani, G. Prina Cerai and B. Morandi, ACS Catal., 2016,
6, 8162–8165.
34 J. Liu, K. Wu, T. Shen, Y. Liang, M. Zou, Y. Zhu, X. Li, X. Li
and N. Jiao, Chem.–Eur. J., 2017, 23, 563–567.
35 L. Eberson, M. P. Hartshorn, O. Persson and F. Radner,
Chem. Commun., 1996, 2105–2112.
45 G. V. Buxton, J. R. Langan and J. R. L. Smith, J. Phys. Chem.,
1986, 90, 6309–6313.
46 D. G. Hendry and D. Schuetzle, J. Am. Chem. Soc., 1975, 97,
7123–7127.
47 X. Fang, X. Pan, A. Rahmann, H.-P. Schuchmann and C. von
Sonntag, Chem.–Eur. J., 1995, 1, 423–429.
˜´
48 E. Estupinan, E. Villenave, S. Raoult, J. C. Rayez, M. T. Rayez
¨
36 I. A. Shuklov, N. V. Dubrovina and A. Borner, Synthesis, 2007,
and R. Lesclaux, Phys. Chem. Chem. Phys., 2003, 5, 4840–
2925–2943.
4845.
37 Y. Kita, H. Tohma, K. Hatanaka, T. Takada, S. Fujita, 49 M. Eberhardt and E. L. Eliel, J. Org. Chem., 1962, 27, 2289–
S. Mitoh, H. Sakurai and S. Oka, J. Am. Chem. Soc., 1994,
116, 3684–3691.
2290.
50 D. P. Curran and A. I. Keller, J. Am. Chem. Soc., 2006, 128,
13706–13707.
¨
¨
38 A. Berkessel, J. A. Adrio, D. Huttenhain and J. M. Neudor, J.
Am. Chem. Soc., 2006, 128, 8421–8426.
39 A. Berkessel and J. A. Adrio, J. Am. Chem. Soc., 2006, 128,
13412–13420.
40 C. Yuan, Y. Liang, T. Hernandez, A. Berriochoa, K. N. Houk
and D. Siegel, Nature, 2013, 499, 192–196.
41 A. Libman, H. Shalit, Y. Vainer, S. Narute, S. Kozuch and
D. Pappo, J. Am. Chem. Soc., 2015, 137, 11453–11460.
42 I. Colomer, C. Batchelor-McAuley, B. Odell, T. J. Donohoe
and R. G. Compton, J. Am. Chem. Soc., 2016, 138, 8855–8861.
51 A. Studer and D. P. Curran, Angew. Chem., Int. Ed., 2016, 55,
58–102.
52 M. A. Matesich, J. Knoefel, H. Feldman and D. F. Evans,
J. Phys. Chem., 1973, 77, 366–369.
53 D. F. Evans and M. A. Matesich, in The Physical Chemistry of
Aqueous System, ed. R. L. Kay, Springer US, Boston, MA, 1973,
pp. 95–118.
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.