Catalytic C?H Trifluoromethoxylation of Arenes and Heteroarenes

Article abstract of DOI:10.1002/anie.201800598

The intermolecular C?H trifluoromethoxylation of arenes remains a long-standing and unsolved problem in organic synthesis. Herein, we report the first catalytic protocol employing a novel trifluoromethoxylating reagent and redox-active catalysts for the direct (hetero)aryl C?H trifluoromethoxylation. Our approach is operationally simple, proceeds at room temperature, uses easy-to-handle reagents, requires only 0.03 mol % of redox-active catalysts, does not need specialized reaction apparatus, and tolerates a wide variety of functional groups and complex structures such as sugars and natural product derivatives. Importantly, both ground-state and photoexcited redox-active catalysts are effective. Detailed computational and experimental studies suggest a unique reaction pathway where photoexcitation of the trifluoromethoxylating reagent releases the OCF₃ radical that is trapped by (hetero)arenes. The resulting cyclohexadienyl radicals are oxidized by redox-active catalysts and deprotonated to form the desired products of trifluoromethoxylation.

Full text of DOI:10.1002/anie.201800598

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201800598
German Edition: DOI: 10.1002/ange.201800598
Arene Trifluoromethoxylation Very Important Paper
À
Catalytic C H Trifluoromethoxylation of Arenes and Heteroarenes
Weijia Zheng, Cristian A. Morales-Rivera, Johnny W. Lee, Peng Liu,* and Ming-Yu Ngai*
À
Abstract: The intermolecular C H trifluoromethoxylation of
arenes remains a long-standing and unsolved problem in
organic synthesis. Herein, we report the first catalytic protocol
employing a novel trifluoromethoxylating reagent and redox-
as silver-mediated synthesis of trifluoromethoxylated
(hetero)arenes have advanced the state of the art
(Scheme 1a),[6b,7] a catalytic intermolecular C H trifluoro-
À
methoxylation of (hetero)arenes remains elusive.[8] Such an
approach is appealing because it precludes the need for the
pre-functionalization of aromatic compounds. In addition,
direct disconnection of the OCF3 group could be envisioned
anywhere onto the target and at any time of the synthesis,
which would allow the late-stage trifluoromethoxylation of
complex molecules.
À
active catalysts for the direct (hetero)aryl C H trifluorometh-
oxylation. Our approach is operationally simple, proceeds at
room temperature, uses easy-to-handle reagents, requires only
0.03 mol% of redox-active catalysts, does not need specialized
reaction apparatus, and tolerates a wide variety of functional
groups and complex structures such as sugars and natural
product derivatives. Importantly, both ground-state and photo-
excited redox-active catalysts are effective. Detailed computa-
tional and experimental studies suggest a unique reaction
pathway where photoexcitation of the trifluoromethoxylating
reagent releases the OCF3 radical that is trapped by (hetero)-
arenes. The resulting cyclohexadienyl radicals are oxidized by
redox-active catalysts and deprotonated to form the desired
products of trifluoromethoxylation.
T
he trifluoromethoxy (OCF3) group, which is found in more
than 350000 biologically active compounds according to
PubChem database as of October 2017, has a broad spectrum
of applications in pharmaceuticals.[1] The prevalence of the
OCF3 group in drugs can be attributed to its favorable
physicochemical properties such as outstanding electronega-
tivity (c = 3.7),[2] which improves molecular metabolic stabil-
ity, and excellent lipophilicity (Hansch parameter: Px =
1.04),[3] which enhances membrane permeability. Notably,
arenes bearing the OCF3 group have a distinct three-dimen-
sional scaffold where the plane containing the C-OCF3 group
is orthogonal to the plane of the aromatic ring.[4] Compounds
with such a structural architecture have been shown to
provide additional binding affinity to biological targets.[5]
Despite these attractive characteristics and the presence
of the OCF3 group in marketed drugs and agrochemicals, only
a handful of approaches have been reported for the synthesis
of trifluoromethoxylated arenes over the last 60 years.[6]
Although recent reports of several powerful strategies such
Scheme 1. Synthesis of trifluoromethoxylated (hetero)arenes.
À
Seeking to establish C H trifluoromethoxylation reac-
tions, we recently described an easy and robust synthesis of
a wide range of (hetero)aromatic hydroxylamides bearing the
N–OCF3 moiety, which undergo thermally induced hetero-
lytic cleavage of the N–OCF3 bond to form a nitrenium-
trifluoromethoxide ion pair.[9] Fast recombination of such an
ion pair affords the products of intramolecular (hetero)aryl
À
C H trifluoromethoxylation. We questioned whether the
photoexcitation of an appropriate N–OCF3 compound could
result in a homolytic cleavage of the N-OCF3 bond (BDE
ꢀ 50 kcalmolÀ1) and the release of the OCF3 radical, which
might be trapped intermolecularly by arenes to afford the
products of trifluoromethoxylation.[6k] Herein, we report the
successful development of a catalytic protocol utilizing an
unprecedented trifluoromethoxylating reagent and redox-
[*] W. Zheng, J. W. Lee, Prof. M.-Y. Ngai
Department of Chemistry
À
active catalysts for the direct intermolecular C H trifluoro-
methoxylation of various (hetero)arenes at room temperature
(Scheme 1b).
and Institute of Chemical Biology and Drug Discovery
Stony Brook University
Stony Brook, NY 11794 (USA)
A key to the success of the proposed transformation is the
development of easy-to-handle trifluoromethoxylating
reagents that release the OCF3 radical under mild reaction
conditions. After exploration of a wide array of reagents
bearing the N–OCF3 moiety, we were pleased to identify
OCF3-reagent 1 capable of direct trifluoromethoxylation of
arenes.[10] Upon exposing 1 (1 equiv) and benzene (10 equiv)
in acetonitrile (MeCN) to a 10 W light-emitting diode (LED,
lem = 402 nm, Figure S5) at room temperature for 16 hours,
E-mail: ming-yu.ngai@stonybrook.edu
C. A. Morales-Rivera, P. Liu
Department of Chemistry
University of Pittsburgh
Pittsburgh, PA 15260 (USA)
E-mail: pengliu@pitt.edu
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201800598.
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Selected examples of trifluoromethoxylation of (hetero)arenes.[a]
we observed 20% yield of the desired trifluoromethoxyben-
zene (3a) and 38% yield of N-arylation side product (3a’)
(Table 1, entry 1).[11] Interestingly, the addition of 20 mol% of
CuII or CuI salts shifted the product distributions and
Table 1: Selected optimization experiments.[a]
[a] Reactions were performed using 1 equivalent of 1 and 10 equivalents
of benzene. [b] Yields and selectivity were determined by 19F NMR
spectroscopy using trifluorotoluene as an internal standard. [c] One
equivalent of benzene was used. [d] Without light. [e] Air atmosphere.
BDE=bond dissociation energy.
increased the yield of 3a (entries 2 and 3), which indicated
that both CuII and CuI could alter the reaction pathway.
Notably, the replacement of the Cu salts with only 0.03 mol%
of Ru(bpy)3(PF6)2 afforded 3a in 70% yield (entry 4). The
transformation also worked with one equivalent of benzene,
albeit it gave the product with a lower yield (entry 5). Control
experiments showed that light and oxygen-free environment
are critical for the reaction to proceed with high efficiency
(entries 6 and 7).
With the optimized conditions in hand, we next sought to
examine the substrate scope. Gratifyingly, a wide range of
À
different functional groups were tolerated under the C H
trifluoromethoxylation conditions (Table 2).[12] Halide sub-
stituents (3b–3 f, 3q, 3r, 3t–3x) remained intact after the
reaction, providing an easy handle for further synthetic
elaborations. In addition, the mild reaction conditions were
compatible with protic functionalities such as carboxylic acid
(3g) and enolizable ketones (3h, 3x). Other functional groups
such as an ester (3j, 3k, 3n, 3p, 3s, 3v–3x), nitrile (3l),
carbonate (3m, 3n), ether (3s), and phosphine oxide (3o)
were also viable and afforded the desired products in
moderate to good yields. Moreover, substrates with benzylic
hydrogens, which are often prone to the hydrogen atom
abstraction in the presence of radical species, were tolerated
(3p, 3t, 3u). Heteroarenes (3q–3x), which are ubiquitous in
biologically active molecules, readily participated in the
coupling reaction as well. More importantly, compounds
with a more elaborate molecular architecture such as fructose
[a] Reactions were performed using 1 equivalent of 1 and 10 equivalents
of the (hetero)arene. Yields and regioselectivity were determined by 19F
NMR spectroscopy using trifluorotoluene as an internal standard.
[b] Reaction was performed at 408C. [c] Yield in parenthesis is the yield of
isolated product. [d] 1 equivalent of the substrate was used. [e] Yield of
isolated product based on the recovered starting material (BRSM).
(3v), (À)-menthol (3w), and trans-androsterone (3x) deriv-
atives were successfully trifluoromethoxylated using 1 equiv-
alent of substrate.
The regioselectivity of the reaction is guided by the
electronics of the substituent except in the case of the bulky
substituent. For example, substrates 3k and 3l, where the tert-
2
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 2. Mechanistic studies and proposed reaction mechanism. All energies are Gibbs free energies (kcalmolÀ1) with respect to the ground-
state reagent 1 and benzene calculated at the M06-2X/6–311+ +G(d,p)/SMD(MeCN)//M06-2X/6–31+G(d,p) level of theory. All the mechanistic
study experiments described in (a) and (b) used one equivalent of reagent 1 and ten equivalents of the arene.
butyl group blocks the ortho-position, afforded only one
regioisomer. Like in other radical-mediated aromatic sub-
stitution processes, the OCF3 radical adds to multiple sites of
arenes to form various regioisomers. This is beneficial in the
context of drug discovery as the isolation of the regioisomers
allows rapid biological-activity assays of trifluoromethoxy-
lated analogues.[13] Thus, this synthetic method is comple-
mentary to the existing site-selective strategies for the
synthesis of trifluoromethoxylated (hetero)arenes.[6b,7]
than that of CNR1R2 (Scheme 2b). This result corroborates the
experimental outcome where irradiation of a mixture of
reagent 1 and 1,3,5-trichlorobenzene in the absence of
a redox-active catalyst gave only the product of trifluorome-
thoxylation. Finally, redox-active catalysts play a critical role
in the final product distribution (3a vs. 3a’). In the absence of
a redox-active catalyst, a significant amount of N-arylation
side-product (3a’) was obtained (Table 1, entry 1). We spec-
ulate that once Ic and CNR1R2 are formed, they can undergo
either hydrogen-atom abstraction to give the desired product
3a or the radical coupling reaction to produce cyclohexadiene
If (see footnote).[17] The elimination of H-NR1R2 or H-OCF3
from If would afford 3a or 3a’, respectively. Computational
studies showed that the formation of 3a’ through the
elimination of H-OCF3 is both kinetically (DDG° = À8.7 kcal
molÀ1) and thermodynamically (DDG = À5.3 kcalmolÀ1, Fig-
ure S8) more favorable than that of H-NR1R2. These data
indicate that once If is formed, it will be exclusively converted
to N-arylation product 3a’. Thus, the low selectivity between
N- and O-arylation in the absence of a redox-active catalyst is
likely due to the competing pathways of radical coupling and
hydrogen-atom abstraction, which would form N- and O-
arylation products (3a’ and 3a), respectively. However, in the
presence of redox-active catalysts, the reaction favors the
formation of the desired product 3a. Presumably, redox-
active catalysts intervene in the reactions between Ia and Ic
through SET processes. Our experimental and computational
data (Figure S7) demonstrate that ground-state redox-active
catalysts are effective to promote the SET processes. With the
photoexcited redox-active catalyst, [Ru(bpy)3(PF6)2*], such
A series of experimental and computational studies were
conducted to shed light on the reaction mechanism
(Scheme 2).[14] We envisioned that COCF3 could be formed
À
by the homolytic cleavage of the N O bond through either
photoexcited 1 or the anion radical of 1 (1ÀC) generated from
the single-electron transfer (SET) between 1 and photoredox
catalysts.[15] To distinguish these two possible reaction path-
ways, we performed the trifluoromethoxylation reaction using
a bandpass filter (488 Æ 2 nm), where the photoexcitation of
the reagent 1 is not feasible but that of Ru(bpy)3(PF6)2 is
possible (Scheme 2a(i)). We recovered 93% of the starting
material 1, which implicates that the formation of COCF3
requires the photoexcitation of reagent 1. In addition, DFT
calculations showed that if the anion radical of 1 was formed,
À
the mesolytic cleavage of the N O bond favors the formation
of CNR1R2 instead of COCF3.[16] Thus, these collective results
suggest that the formation of COCF3 via the reduction of
reagent 1 by excited Ru(bpy)3(PF6)2 is unlikely.
We then examined the relative reactivity of COCF3 and
CNR1R2 towards an arene. DFT calculations showed the
addition of COCF3 to benzene is energetically more favorable
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
processes should be more efficient,[18] and thus only
0.03 mol% of Ru(bpy)3(PF6)2 is needed.
On the basis of these mechanistic studies, we envisage that
the photoexcitation of reagent 1 under the irradiation of 10 W
violet LED light (lem = 402 nm, Figure S5) forms excited 1,
[6] Synthesis of trifluoromethoxylated (hetero)aromatic com-
pounds: for reviews, see: a) F. R. Leroux, B. Manteau, J. P.
Vors, S. Pazenok, Beilstein J. Org. Chem. 2008, 4, 13; b) A. Tlili,
F. Toulgoat, T. Billard, Angew. Chem. Int. Ed. 2016, 55, 11726 –
11735; Angew. Chem. 2016, 128, 11900 – 11909; c) K. N. Lee,
J. W. Lee, M.-Y. Ngai, Synlett 2016, 27, 313 – 319; for a two-step
chlorination/chlorine – fluorine exchange reaction of electron-
deficient anisoles, see: d) L. M. Yagupolskii, Dokl. Akad. Nauk
SSSR 1955, 105, 100 – 102; for deoxyfluorination of fluorofor-
mates, see: e) W. A. Sheppard, J. Org. Chem. 1964, 29, 1 – 11; for
electrophilic trifluoromethylation of phenols, see: f) T. Ume-
moto, Chem. Rev. 1996, 96, 1757 – 1777; g) T. Umemoto, K.
Adachi, S. Ishihara, J. Org. Chem. 2007, 72, 6905 – 6917; h) K.
Stanek, R. Koller, A. Togni, J. Org. Chem. 2008, 73, 7678 – 7685;
for nucleophilic trifluoromethoxylation of arenes, see: i) M.
Nishida, A. Vij, R. L. Kirchmeier, J. M. Shreeve, Inorg. Chem.
1995, 34, 6085 – 6092; j) A. A. Kolomeitsev, M. Vorobyev, H.
Gillandt, Tetrahedron Lett. 2008, 49, 449 – 454; for radical
trifluoromethoxylation of arenes, see: k) F. Venturini, W.
Navarrini, A. Famulari, M. Sansotera, P. Dardani, V. Tortelli, J.
Fluorine Chem. 2012, 140, 43 – 48.
[7] a) C. Huang, T. Liang, S. Harada, E. Lee, T. Ritter, J. Am. Chem.
Soc. 2011, 133, 13308 – 13310; b) J. B. Liu, C. Chen, L. Chu, Z. H.
Chen, X. H. Xu, F. L. Qing, Angew. Chem. Int. Ed. 2015, 54,
11839 – 11842; Angew. Chem. 2015, 127, 12005 – 12008; c) T.
Khotavivattana, S. Verhoog, M. Tredwell, L. Pfeifer, S. Calder-
wood, K. Wheelhouse, T. L. Collier, V. Gouverneur, Angew.
Chem. Int. Ed. 2015, 54, 9991 – 9995; Angew. Chem. 2015, 127,
10129 – 10133; d) Q. W. Zhang, A. T. Brusoe, V. Mascitti, K. D.
Hesp, D. C. Blakemore, J. T. Kohrt, J. F. Hartwig, Angew. Chem.
Int. Ed. 2016, 55, 9758 – 9762; Angew. Chem. 2016, 128, 9910 –
9914.
À
which undergoes homolytic N O bond cleavage to generate
the N-centered radical (CNR1R2) and the OCF3 radical
(COCF3) (Scheme 2c). Once COCF3 is formed, it adds to an
arene to afford the cyclohexadienyl radical Ic. Redox-active
catalysts then mediate sequential single-electron transfer
(SET) processes between CNR1R2 and Ic to form ionic species
Id and Ie, respectively. Deprotonation of Ie restores the
aromaticity and gives the desired product of trifluoromethox-
ylation (3a).
In summary, we developed the first catalytic intermolec-
À
ular C H trifluoromethoxylation of (hetero)arenes employ-
ing redox-active catalysts and the trifluoromethoxylating
reagent 1 at ambient temperature. The mild reaction con-
ditions obviate the need for specialized reaction apparatus
and tolerate a wide array of functional groups. Mechanistic
studies showed that 1) photoexcitation of reagent 1 forms
COCF3 and 2) redox-active catalysts intervene in the radical
coupling reaction of Ia and Ic to favor the formation of the
desired product of trifluoromethoxylation.
Acknowledgments
Financial support for this work was provided by NIH
(R35GM119652 to M.-Y.N.), start-up funds from Stony
Brook University (to M.-Y.N.), and start-up funds from
University of Pittsburgh (to P.L.). We thank P. Zhao for help
with measurement of the emission spectrum of the 10 W LED
light (402 nm). We thank TOSOH F-Tech, Inc. for their gift of
TMSCF3 for the preparation of Togni reagents. Calculations
were performed at the Center for Research Computing at the
University of Pittsburgh and the Extreme Science and
Engineering Discovery Environment (XSEDE) supported
by the NSF (ACI-1053575).
[8] Catalytic trifluoromethoxylation of alkenes, see: a) C. H. Chen,
P. H. Chen, G. S. Liu, J. Am. Chem. Soc. 2015, 137, 15648 – 15651;
b) S. Guo, F. Cong, R. Guo, L. Wang, P. P. Tang, Nat. Chem. 2017,
9, 546 – 551; c) C. Chen, Y. Luo, L. Fu, P. Chen, Y. Lan, G. Liu, J.
Am. Chem. Soc. 2018, 140, 1207 – 1210; Catalytic decarboxyla-
tive fluorination of aryloxydifluorocarboxylic acids, see: d) M.
Zhou, C. F. Ni, Z. B. He, J. B. Hu, Org. Lett. 2016, 18, 3754 –
À
3757; Catalytic allylic C H trifluoromethoxylation, see: e) X.
Qi, P. Chen, G. Liu, Angew. Chem. Int. Ed. 2017, 56, 9517 – 9521;
Angew. Chem. 2017, 129, 9645 – 9649; f) C. Chen, Y. Luo, L. Fu,
P. Chen, Y. Lan, G. Liu, J. Am. Chem. Soc. 2018, 140, 1207 – 1210.
[9] a) K. N. Hojczyk, P. Feng, C. Zhan, M. -Y. Ngai, Angew. Chem.
Int. Ed. 2014, 53, 14559 – 14563; Angew. Chem. 2014, 126, 14787 –
14791; b) P. Feng, K. N. Lee, J. W. Lee, C. Zhan, M.-Y. Ngai,
Chem. Sci. 2016, 7, 424 – 429.
Conflict of interest
[10] See the Supporting Information for the discussion of reagent
design principles and the additional optimization data.
[11] For an example of photocatalyst-free photoreaction, see: L.
Buzzetti, A. Prieto, S. R. Roy, P. Melchiorre, Angew. Chem. Int.
Ed. 2017, 56, 15039 – 15043; Angew. Chem. 2017, 129, 15235 –
15239.
[12] Electron-rich arenes such as anisoles failed to provide the
desired products under the reaction conditions. In those cases,
the reaction mixture turned black, reagent 1 decomposed, and
fluorophosgene was detected. See the Supporting Information
for detailed information.
The authors declare no conflict of interest.
Keywords: arenes · photocatalysis · radicals ·
trifluoromethoxylation
[13] D. A. Nagib, D. W. C. MacMillan, Nature 2011, 480, 224 – 228.
[14] See the Supporting Information for additional mechanistic
details.
[1] P. Jeschke, E. Baston, F. R. Leroux, Mini-Rev. Med. Chem. 2007,
7, 1027 – 1034.
[2] M. A. McClinton, D. A. McClinton, Tetrahedron 1992, 48, 6555 –
6666.
[3] C. Hansch, A. Leo, Substituent Constants for Correlation
Analysis in Chemistry and Biology, Wiley, New York, 1979.
[4] D. Federsel, A. Herrmann, D. Christen, S. Sander, H. Willner, H.
Oberhammer, J. Mol. Struct. 2001, 567, 127 – 136.
[5] K. Mꢀller, C. Faeh, F. Diederich, Science 2007, 317, 1881 – 1886.
À
[15] Cleavage of the N O bond, see: a) C. R. Jamison, L. E. Over-
man, Acc. Chem. Res. 2016, 49, 1578 – 1586; Formation of O-
centered radical under visible light photoredox conditions, see:
b) J. J. Guo, A. H. Hu, Y. L. Chen, J. F. Sun, H. M. Tang, Z. W.
Zuo, Angew. Chem. Int. Ed. 2016, 55, 15319 – 15322; Angew.
Chem. 2016, 128, 15545 – 15548; c) K. F. Jia, F. Y. Zhang, H. C.
4
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Huang, Y. Y. Chen, J. Am. Chem. Soc. 2016, 138, 1514 – 1517;
d) C. Y. Wang, K. Harms, E. Meggers, Angew. Chem. Int. Ed.
2016, 55, 13495 – 13498; Angew. Chem. 2016, 128, 13693 – 13696;
e) H. G. Yayla, H. J. Wang, K. T. Tarantino, H. S. Orbe, R. R.
Knowles, J. Am. Chem. Soc. 2016, 138, 10794 – 10797; f) J. Zhang,
Y. Li, F. Y. Zhang, C. C. Hu, Y. Y. Chen, Angew. Chem. Int. Ed.
2016, 55, 1872 – 1875; Angew. Chem. 2016, 128, 1904 – 1907;
g) K. F. Jia, Y. Pan, Y. Y. Chen, Angew. Chem. Int. Ed. 2017, 56,
2478 – 2481; Angew. Chem. 2017, 129, 2518 – 2521.
[16] L. J. Allen, P. J. Cabrera, M. Lee, M. S. Sanford, J. Am. Chem.
Soc. 2014, 136, 5607 – 5610.
[17] Proposed mechanism in the absence of a redox-active catalyst:
[18] C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 2013,
113, 5322 – 5363.
Manuscript received: January 15, 2018
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Arene Trifluoromethoxylation
W. Zheng, C. A. Morales-Rivera, J. W. Lee,
P. Liu,* M.-Y. Ngai*
&&&& — &&&&
À
Catalytic C H Trifluoromethoxylation of
Arenes and Heteroarenes
Radical approach: Photoexcitation of
a novel trifluoromethoxylating reagent
bearing the N-OCF3 moiety results in
homolytic cleavage of the N–OCF3 bond
and releases the OCF3 radical at room
temperature. Trapping of the OCF3 radical
by (hetero)arenes in the presence of
redox-active catalysts leads to the first
example of intermolecular, catalytic (het-
À
ero)aryl C H trifluoromethoxylation.
6
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!