Iron-Catalyzed Halogen Exchange of Trifluoromethyl Arenes**

Article abstract of DOI:10.1002/chem.202101324

The facile production of ArCF₂X and ArCX₃ from ArCF₃ using catalytic iron(III)halides is reported, which constitutes the first iron-catalyzed halogen exchange for non-aromatic C?F bonds. Theoretical calculations suggest direct activation of C?F bonds by iron coordination. ArCX₃ and ArCF₂X products of the reaction are synthetically valuable due to their diversification potential. In particular, chloro- and bromodifluoromethyl arenes (ArCF₂Cl, ArCF₂Br respectively) provide access to a myriad of difluoromethyl arene derivatives (ArCF₂R). To optimize for mono-halogen exchange, a statistical method called Design of Experiments was used. Optimized parameters were successfully applied to electron rich and electron deficient aromatic substrates, and to the late stage diversification of flufenoxuron, a commercial insecticide. These methods are highly practical, being run at convenient temperatures and using inexpensive common reagents.

Full text of DOI:10.1002/chem.202101324

Communication
doi.org/10.1002/chem.202101324
Chemistry—A European Journal
Iron-Catalyzed Halogen Exchange of Trifluoromethyl
Arenes**
Andreas Dorian,[a] Emily J. Landgreen,[b] Hayley R. Petras,[b] James J. Shepherd,[b] and
Florence J. Williams*[b]
These larger halides are versatile synthetic handles, with
heightened reactivity over the parent fluoride.
Abstract: The facile production of ArCF2X and ArCX3 from
ArCF3 using catalytic iron(III)halides is reported, which
Halex reactions on fluorinated substrates are typically
constitutes the first iron-catalyzed halogen exchange for
performed with stoichiometric early valent metal halides (TiX4,
non-aromatic CÀ F bonds. Theoretical calculations suggest
direct activation of CÀ F bonds by iron coordination. ArCX3
and ArCF2X products of the reaction are synthetically
valuable due to their diversification potential. In particular,
chloro- and bromodifluoromethyl arenes (ArCF2Cl, ArCF2Br
respectively) provide access to a myriad of difluoromethyl
arene derivatives (ArCF2R). To optimize for mono-halogen
exchange, a statistical method called Design of Experiments
was used. Optimized parameters were successfully applied
to electron rich and electron deficient aromatic substrates,
and to the late stage diversification of flufenoxuron, a
commercial insecticide. These methods are highly practical,
being run at convenient temperatures and using inexpen-
sive common reagents.
MoCl5), or p-block metalloid halides (BX3, AlX3, R3SiX).[7] These
halide reagents provide the necessary thermodynamic drive
through formation of
a strong fluoride bond (e.g. SiF4
166 kcalmolÀ 1, BF3 170 kcalmolÀ 1)[8] However, additional cata-
lysts are necessary to avoid high temperatures and long
reaction times.
Iron, being a low-cost high abundance metal, has been
surprisingly absent in catalytic defluorination development. The
potential of iron catalysis in this field is nevertheless exemplified
in enzymes such as horseradish peroxidase (Hrp), methane
monooxygenase (Mmo) and cytochrome P450s (Cyp), all
employing iron-based catalytic motifs, which perform catalytic
defluorination along with oxygenation (Figure 1a).[9]
To our knowledge only one precedent exists for iron-
catalyzed halogen exchange of CÀ F bonds: chlorination of
perfluorinated benzenes (Figure 1b).[10]
The following work constitutes the first example of iron-
catalyzed halogen exchange of non-aromatic CÀ F bonds, and to
our knowledge the first halogen-exchange with mechanistic
CÀ F bonds impart desirable features to a molecule, such as
chemical inertness and increased lipophilicity.[1] As such,
fluorinated moieties continue to play a critical role in the
modern pharmaceutical, agrochemical and materials chemistry
fields, and are of longstanding synthetic interest to organic
chemists.[2] A prominent example, the trifluoromethyl group
(CF3) is present in 24% of drugs globally approved in 2019.[3]
The selective conversion of CÀ F bonds to CÀ H or CÀ C bonds
in polyfluorinated molecules has enjoyed increased focus as a
strategy to access challenging fluorination patterns.[4] A related
transformation is the conversion to higher halide CÀ X bonds
(X=Cl, Br, I), through halogen-exchange (halex) reactions.[5,6]
[a] A. Dorian
Department of Chemistry
University of Alberta
Edmonton, Alberta, T6G 2G2 (Canada)
[b] E. J. Landgreen, H. R. Petras, Prof. J. J. Shepherd, Prof. F. J. Williams
Department of Chemistry
University of Iowa
Iowa City, Iowa, 52242 (USA)
E-mail: florence-williams@uiowa.edu
[**] A previous version of this manuscript has been deposited on a preprint
Figure 1. a) Cyp450-catalyzed defluorination and concurrent hydroxylation.[9b]
b) Prior iron-catalyzed aryl CÀ F halogen-exchange.[10] c) Iron-catalyzed
benzylic CÀ F halogen-exchange described herein.
server (https://doi.org/10.26434/chemrxiv.14058332.v1).
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202101324
Chem. Eur. J. 2021, 27, 1–6
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202101324
Chemistry—A European Journal
evidence for iron coordination and direct activation of a CÀ F
bond. This method was applied and optimized for medicinally
relevant ArCF3 substrates (Figure 1c).
We explored a variety of halophilic Lewis acidic transition
metal complexes in the presence of BBr3 and trifluoromethyl
(m-fluoro)benzene 1a. In the absence of a catalyst, the known
background reaction provides compound 2a at 7% conversion
(Table 1, entry 1).[6d] While silver(I), copper(I), iron(0) and iron(II)
showed no improvements on CÀ F exchange efficiencies (en-
tries 2–5), iron(III) and gallium(III) compounds accelerated the
halex process (entries 6–10), generating tribrominated product
4a at 83–91%. Interestingly, the identity of the halide on the
iron(III) center did not affect the reaction significantly.
Figure 2. Transformations of ArCF2X (X=Br, Cl) reported in the literature.[12]
Ensuing investigations were performed with iron(III)fluoride.
Solvent selection proved important (Table 1, entries 11–13),
possibly due to the varying solubility of the iron(III)fluoride
reagent. However, it should be noted that the iron complex
showed a distinct increase in solubility upon boryl halide
addition (see Figure S5, S15 in Supporting Information). Nitro-
methane coordination with BBr3 was also observed in the 11B
NMR spectrum, and this interaction is suspected to mitigate
reactivity.
Different boryl halides were also competent in the halex
reaction (entries 14–16), although these reactions were slow
relative to those with BBr3. When employing BCl3, higher
catalyst loadings enabled a high total CÀ F conversion and
generation of the trichloromethyl arene as the sole product
(entry 16). Tribromomethyl arene 4a could also be obtained as
a single product, using excess BBr3 and an extended reaction
time (entry 17).
Trihalomethyl arenes can be used in a number of synthetic
applications, including conversion to carboxylic acids/esters,
alkynes and alkenes, and to generate heterocycles.[11,6b] Accord-
ingly, we envision this operationally-simple method to be quite
useful, given the prevalence of trifluoromethyl groups in high-
profile compounds.
Table 1. Screening Investigations.
However, there remained potential to optimize a single
halogen-exchange, generating compound 2a. In addition to
unique intrinsic properties, such as increased capacity for
halogen bonding, ArCF2X compounds are common intermedi-
ates for redox and cross-coupling reactions that generate a
range of high value difluoro derivatives (ArCF2R, Figure 2).[12,13]
Optimizing a single halogen exchange on a trifluoromethyl
center is a challenge, as the CÀ F bond strength decreases with
each fluorine removal.[14] Encouragingly, lowering of the reac-
tion temperature increased conversion to ArCF2Br (2a) relative
to ArCBr3 (4a) (entry 18). This observation indicated that
appropriate reaction tuning could provide synthetically viable
amounts of ArCF2Br.
The kinetic complexity of the mono-halex reaction led us to
use the statistical method Design of Experiments (DOE) for
optimization. DOE uses regression analysis to generate a
mathematical model of a reaction outcome as a function of
defined reaction parameters (variables).[15] The model devel-
oped is tailored to the reaction space defined by the chosen
parameters. The relative importance of each parameter, as well
as interaction effects between parameters is determined by
analysis of variance (ANOVA).
We first performed a 5-variable fractional factorial DOE to
evaluate main linear effects, followed by a response surface
design, a more predictive model that accounts for non-linear
effects. Evaluation of data normalcy, R2 fit, R2predicted (a leave-one-
out cross-validation test for over-fitting), and analysis of error
°
Reactions performed at 1.0 mmol scale, 0.2 M molarity, 20 C, and 5 h.
Conversions determined by 19F NMR (relative to 4-fluorotoluene internal
standard). 3a was trace in all entries (<3% conversion). cat=catechol. nd.
=not detected. tr.=trace. [a] CÀ Ftotal refers to the molar fraction of CÀ F
bonds converted to CÀ X bonds. [b] 3 equiv. R2BBr used. [c] Chlorinated
products formed (ArCCl3 instead of 4a). Conditions: DCM/NO2Me (4:1),
°
0.6 equiv. FeCl3, 4 equiv. BCl3, 30 C, 16 h. [d] Conditions: 10 mol% FeF3,
°
1.5 equiv. BBr3, 14 h. [e] Performed at À 10 C.
Chem. Eur. J. 2021, 27, 1–6
www.chemeurj.org
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202101324
Chemistry—A European Journal
versus reaction order (test for systematic experimental error),
indicated a well-fitted model for ArCF2Br conversion. Additional
test points demonstrated experimental conversions comparable
to those predicted by the model (see Table S6 in Supporting
Information).
The significance of each variable in the DOE model is shown
in the Pareto chart depicted in Figure 3. Such information can
reflect mechanistic nuance. Temperature was found to have a
strong correlation with ArCF2Br conversion, as well as a
significant interaction effect with BBr3. Figure 4 shows the
parabolic shape of the temperature variable with a maximum at
°
À 3 C.
Surprisingly, time was not a highly significant variable to
the model, as shown by the shallow slope in Figure 4 and
Pareto values in Figure 3 (see Supporting Information for other
surface plots). These results suggest that optimal conversion is
more complex than identifying a time point where a statistical
mixture of products favors ArCF2Br.
The DOE model predicted conditions for a maximum
conversion to ArCF2Br 2a of 25%. Experimentally, these
conditions resulted in 27% conversion, slightly outperforming
the model (Scheme 1, 2a).
Given optimized reaction parameters for meta-fluoro
substrate 1a, we sought to apply conditions favoring ArCF2X
formation to a selection of substrates (Scheme 1). The
reaction was observed to progress more rapidly with
electron-rich aromatic complexes, which was easily compen-
sated for with shorter reaction times. Gratifyingly, similar
conversions to the desired ArCF2Br could be obtained for
most substrates examined. For compounds with Lewis basic
functionalities, additional equivalents of BBr3 were necessary
(2i, 2j, 2k). Substitution at ortho, meta, and para positions,
Figure 4. Surface plots from the DOE model. [Above] ArCF2Br conversion as a
function of BBr3 equivalents and temperature (constants: time=12 h,
concentration=0.2 M). [Below] ArCF2Br conversion as a function of time and
°
concentration (constants: BBr3 =1.2 equivalents, temperature=À 12.5 C).
Scheme 1. Scope investigations for the mono-halex of trifluoromethyl arenes
(0.40 mmol scale). Conversions determined by 19F NMR (relative to 4-
°
fluorotoluene), isolated yields in parentheses. [a] Performed at À 3 C. [b]
Figure 3. Pareto chart with p-values for a response surface model of 2a
conversion. The red dotted line is the standardized effect t-value determined
by significance level α=0.15, where α is the probability of obtaining a false
positive on a statistical test.
°
Conditions: 0.20 M NO2Me/DCM (2:1), 0.6 equiv. FeCl3, 1.7 equiv. BCl3, 10 C.
[c] Fe(OTf)3 used (in place of FeF3). [d] 1.45 equiv. BBr3 used.
Chem. Eur. J. 2021, 27, 1–6
www.chemeurj.org
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202101324
Chemistry—A European Journal
including halides, alkyl, and hydroxy groups were all well
tolerated. Chlorination was also achieved using similar
conditions (compound 2c). Isolation of the ArCF2X products
was performed using reverse-phase HPLC. Small molecule
substrates (for which high volatility affected isolated yields)
are reported as conversions, with additional characterization
by GC-MS (see Supporting Information).
To our delight, phosphine 1i was converted to desymme-
trized mono-brominated compound 2i in 17% isolated yield,
despite 9 fluorines which can participate in the halex reaction.
With substrate 1h, no evidence of styrene derivatives or
Friedel-Crafts alkylation arising from iron(III) activation of the
alkyl bromide side chain was observed. Further, a commercial
insecticide flufenoxuron (1k), was converted to mono-bromi-
nated derivative 2k in 21% conversion and 13% yield
(Scheme 2).
Under triple-halex conditions, both trichloro- (4a, 4d) and
tribromomethyl arenes (5a, 5d), as well as compound 6, were
generated from substrates 1a, 1d and 1l, respectively, with
high isolated yields (66–91%; Scheme 3). Compound 6 results
from rapid hydrolysis/methanolysis of the tribromomethyl
group during workup procedures. These substrates demon-
strate that electron-deficient and electron-rich arenes are
similarly competent in triple halogen-exchange reactions.
At the outset of our mechanistic investigations, both
heterogeneous and homogeneous catalysis appeared plausible
in this system. Theoretical calculations were used to evaluate
the viability of a homogeneous process, given the high degree
of solubilization of the iron complexes in the presence of the
boron trihalide reagents. Further, should a homogeneous
process appear reasonable, we wished to shed light on a
potential role of the iron catalyst in the mechanism. Using
unrestricted Kohn-Sham density functional theory B3LYP/6-
31G**,[16] coordination energies between PhCF3 1b, BBr3, and
FeF3 were determined in dimeric and trimeric orientations. An
analogous evaluation was performed with PhCF3, BCl3, and
FeCl3 for comparison (see Supporting Information).
The lowest dimeric coordination energy observed was
between PhCF3 and FeF3 at À 7.62 kcalmolÀ 1 (Figure 5). This
interaction is predicted to stretch the coordinated CÀ F bond
by approximately 0.1 Å. In contrast, the binding energy of
PhCF3 and BBr3 is predicted to be uphill by 7.01 kcalmolÀ 1 19F
.
NMR experiments support these calculations and provide
evidence for solution activation of the CÀ F bond. The fluorine
signal in PhCF3 is shifted downfield by 2.5 ppm in the presence
of FeCl3, whereas BCl3 only elicits a downfield shift of 0.2 ppm
(CD2Cl2).
The lowest energy trimers (PhCF3, BBr3 and FeF3) have
interactions dominated by PhCF3 and FeF3 binding, with the
lowest energy trimer at À 0.78 kcalmolÀ 1 relative to the
monomers. Therefore, while iron activation of the boron
tribromide remains a valid mechanistic possibility, calculations
and NMR experiments suggest direct interaction with PhCF3 is
energetically favorable and weakens the CÀ F bond.
Dimers of iron halides (Fe2F6 and Fe2Cl6) were also evaluated
for their coordinating affinity for either the PhCF3 substrate or
the boron trihalide (BBr3 and BCl3, respectively, see Supporting
Information).
Figure 6 shows plausible mechanisms based on these data.
The nucleophilic bromide is likely delivered from a boron halide
complex, as the identity of the boron halide reagent determines
the major halex product. Subsequent transfer of the fluoride
from the iron to the boron center is required for the process to
be exergonic overall (1st Br exchange: À 25.74 kcalmolÀ 1 B3LYP/
6-31G**),[17] and this is corroborated by the experimental
observations of gas evolution (BF3) and a disappearance of the
11B NMR signal. This also highlights that ligand exchange
between iron(III)fluoride and boron trihalides is facile under the
reaction conditions, which is further supported by 19F NMR
experiments (see Supporting Information), so heteroleptic iron
complexes likely play an important mechanistic role.
Scheme 2. Halogen exchange with a commercial insecticide flufenoxuron.
Figure 5. Computed geometries & bond lengths for FeF3 coordination to
PhCF3.
Scheme 3. Isolated yields of trihalomethyl arenes produced via triple-halex.
Chem. Eur. J. 2021, 27, 1–6
www.chemeurj.org
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202101324
Chemistry—A European Journal
[1] D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308–319.
[2] a) R. Berger, G. Resnati, P. Metrangolo, E. Weber, J. Hulliger, Chem. Soc.
Rev. 2011, 40, 3496–3508; b) E. P. Gillis, K. J. Eastman, M. D. Hill, D. J.
Donnelly, N. A. Meanwell, J. Med. Chem. 2015, 58, 8315–8359.
[3] a) M. Inoue, Y. Sumii, N. Shibata, ACS Omega 2020, 5, 10633–10640;
b) C. Alonso, E. Martinez de Marigorta, G. Rubiales, F. Palacios, Chem.
Rev. 2015, 115, 1847–1935.
[4] a) T. Ahrens, J. Kohlmann, M. Ahrens, T. Braun, Chem. Rev. 2015, 115,
931–972; b) T. Fujita, K. Fuchibe, J. Ichikawa, Angew. Chem. Int. Ed. 2019,
58, 390–402; Angew. Chem. 2019, 131, 396–408; c) J. D. Hamel, J. F.
Paquin, Chem. Commun. 2018, 54, 10224–10239; d) F. Jaroschik, Chem.
Eur. J. 2018, 24, 14572–14582; e) H. Amii, K. Uneyama, Chem. Rev. 2009,
109, 2119–2183.
[5] a) K. K. K. Goh, A. Sinha, C. Fraser, R. D. Young, RSC Adv. 2016, 6, 42708–
42712; b) M. Janjetovic, A. Ekebergh, A. M. Traff, G. Hilmersson, Org. Lett.
2016, 18, 2804–2807; c) Y. Mizukami, Z. Song, T. Takahashi, Org. Lett.
2015, 17, 5942–5945; d) K. Matsubara, T. Ishibashi, Y. Koga, Org. Lett.
2009; e) J. Terao, S. A. Begum, Y. Shinohara, M. Tomita, Y. Naitoh, N.
Kambe, Chem. Commun. 2007, 855–857; f) M. Namavari, N. Satyamurthy,
J. R. Barrio, J. F. Chem. 1995, 72, 89–93; g) G. A. Olah, S. C. Narang, L. D.
Field, J. Org. Chem. 1981, 46, 3727–3728; h) J. San Filippo Jr., A. F.
Sowinski, L. J. Romano, J. Org. Chem. 1975, 40, 3295–3296.
[6] a) D. Mandal, R. Gupta, A. K. Jaiswal, R. D. Young, J. Am. Chem. Soc.
2020, 142, 2572–2578; b) M. Ikeda, T. Matsuzawa, T. Morita, T. Hosoya, S.
Yoshida, Chem. Eur. J. 2020, 26, 12333–12337; c) F. Dankert, H. L.
Deubner, M. Müller, M. R. Buchner, F. Kraus, C. von Hänisch, Z. Anorg.
Allg. Chem. 2020, 646, 1501–1507; d) G. K. S. Prakash, J. Hu, J. Simon,
D. R. Bellew, G. A. Olah, J. F. Chem. 2004, 125, 595–601; e) R. K.
Ramchandani, R. D. Wakharkar, A. Sudalai, Tet. Lett. 1996, 37, 4063–
4064; f) J. Riera, J. Castaiier, J. Carilla, A. Robert, Tet. Lett. 1989, 30;
g) A. L. Henne, M. S. Newman, J. Am. Chem. Soc. 1938.
Figure 6. Plausible mechanisms for the halex reaction.
In conclusion, we have demonstrated that ferric halides
catalyze halogen exchange between trifluoromethyl arenes and
boron halides. While initially developed conditions favor full
exchange of benzylic fluorides, providing ArCX3 products, DOE
analyses provided conditions that maximize mono-exchange to
generate ArCF2X products in synthetically viable yields. Sub-
strate scope investigations demonstrated successful application
to electron-deficient and electron-rich arenes, with good func-
tional group tolerance. The reaction is eminently practical, using
inexpensive, readily available reagents and run at easily
maintained temperatures.
[7] T. Stahl, H. F. T. Klare, M. Oestreich, ACS Catal. 2013, 3, 1578–1587.
[8] a) D. J. Grant, D. A. Dixon, J. Phys. Chem. A 2009, 113, 777–787; b) R.
Walsh, J. Chem. Soc. Faraday Trans. 1 1983, 79, 2233–2248.
[9] a) Y. Wang, A. Liu, Chem. Soc. Rev. 2020, 49, 4906–4925; b) G. M. Amaya,
R. Durandis, D. S. Bourgeois, J. A. Perkins, A. A. Abouda, K. J. Wines, M.
Mohamud, S. A. Starks, R. N. Daniels, K. D. Jackson, Chem. Res. Toxicol.
2018, 31, 570–584.
Notably, mechanistic insight from computational analyses
suggests that ferric halides may be direct activators of the CÀ F
bond, which has important implications for future CÀ F
activation methodology development.
[10] A. I. Shipilov, L. A. Kolpashchikova, S. M. Igumnov, Russ. J. Org. Chem.
2003, 39, 975–978.
[11] a) Q. Sun, X. Yu, M. Bao, M. Liu, J. Pan, Z. Zha, L. Cai, H. Ma, C. Yuan, X.
Qiu, W. Xu, Angew. Chem. Int. Ed. 2018, 57, 4035–4038; Angew. Chem.
2018, 130, 4099–4102; b) H. Liu, S.-i. Kondo, N. Takeda, M. Unno, Eur. J.
Inorg. Chem. 2009, 1317–1319.
Acknowledgements
[12] a) M. Yoshida, M. Morinaga, M. Iyoda, J. F. Chem. 1994; b) S. Sumino, M.
Uno, T. Fukuyama, I. Ryu, M. Matsuura, A. Yamamoto, Y. Kishikawa, J.
Org. Chem. 2017, 82, 5469–5474; c) J. Guidotti, C. Wakselman, F. Metz,
M. Tordeux, Synlett 2004, 1759–1762; d) J. Guidotti, V. Schanen, M.
Tordeux, C. Wakselman, J. F. Chem. 2005, 126, 443–447; e) J.-W. Gu, W.-
H. Guo, X. Zhang, Org. Chem. Front. 2015, 2, 38–41; f) S. Verhoog, L.
Pfeifer, T. Khotavivattana, S. Calderwood, T. Collier, K. Wheelhouse, M.
Tredwell, V. Gouverneur, Synlett 2015, 27, 25–28.
[13] a) X. J. Tang, Z. Zhang, W. R. Dolbier, Jr., Chem. Eur. J. 2015, 21, 18961–
18965; b) Z. Zhang, X. Tang, W. R. Dolbier, Jr., Org. Lett. 2015, 17, 4401–
4403; c) H. Jiang, W. Lu, K. Yang, G. Ma, M. Xu, J. Li, J. Yao, W. Wan, H.
Deng, S. Wu, S. Zhu, J. Hao, Chem. Eur. J. 2014, 20, 10084–10092.
[14] G. Glockler, J. Phys. Chem. 1959, 63, 828–832.
The authors thank Dr. Pamela Tadross of Merck & Co. Inc. for
feedback and advice regarding DOE. This work was primarily
supported by NSERC RGPIN-2016-04843, the ACS Petroleum
Research Fund PRF 59191-ND1, and the Canadian Foundation
for Innovation IOF-34997 (FW and AD: experimental work and
DOE). Additional support (AD) is acknowledged from Dr. Ralf
Schirrmacher, through NSERC RGPIN-2019-04481. Electronic
structure calculations (JS, EL, and HRP) were supported by the
University of Iowa.
[15] S. A. Weissman, N. G. Anderson, Org. Process Res. Dev. 2015, 19, 1605–
1633.
[16] a) A. D. Becke, J. Chem. Phys. 1993, 98, 1372–1377; b) V. A. Rassolov,
M. A. Ratner, J. A. Pople, P. C. Redfern, L. A. Curtiss, J. Comput. Chem.
2001, 22, 976–984.
Conflict of Interest
[17] R. J. Bartlett, M. Musiał, Rev. Mod. Phys. 2007, 79, 291–352.
The authors declare no conflict of interest.
Keywords: CÀ F activation · defluorination · halogen exchange ·
iron catalysis · late-stage modification
Manuscript received: April 13, 2021
Version of record online: ■■■, ■■■■
Chem. Eur. J. 2021, 27, 1–6
www.chemeurj.org
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

COMMUNICATION
A. Dorian, E. J. Landgreen, H. R. Petras,
Prof. J. J. Shepherd, Prof. F. J. Williams*
1 – 6
Iron-Catalyzed Halogen Exchange of
Trifluoromethyl Arenes
The production of synthetically
valuable ArCF2X and ArCX3
optimize for mono-exchange, a statis-
tical analysis called Design of Experi-
ments was used. Optimized parame-
ters were successfully applied to both
electron-rich and electron-deficient
aromatic substrates, and to the late-
stage diversification of flufenoxuron, a
commercial insecticide.
compounds from ArCF3 using catalytic
iron(III)halides is described, which
constitutes the first iron-catalyzed
halogen exchange of non-aromatic
CÀ F bonds. Theoretical calculations
suggest direct activation of CÀ F
bonds by iron coordination. To