Direct Trifluoromethoxylation without OCF3-Carrier through In Situ Generation of Fluorophosgene

Article abstract of DOI:10.1002/ejoc.202100429

Owing to the high interest in the OCF₃ group for pharmaceutical and agrochemical applications, trifluoromethoxylation received great attention in the last years with several new methods for this approach towards OCF₃-comprising compounds. Yet, it most often requires the beforehand preparation of specific F₃CO^? transfer reagents, which can be toxic, expensive, unstable, and/or generate undesired side-products upon consumption. To circumvent this, the in-situ generation of gaseous fluorophosgene from triphosgene, its conversion by fluoride into the OCF₃ anion, and the direct use of the latter in nucleophilic substitutions is an appealing strategy, which, although recently approached, has not been fully exploited. We disclose herein our efforts towards this aim.

Full text of DOI:10.1002/ejoc.202100429

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: Direct trifluoromethoxylation without OCF3 -carrier via in-situ
generation of fluorophosgene
Authors: Jérémy Saiter, Thomas Guérin, Morgan Donnard, Armen
Panossian, Gilles Hanquet, and Frédéric R. Leroux
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: Eur. J. Org. Chem. 10.1002/ejoc.202100429
Link to VoR: https://doi.org/10.1002/ejoc.202100429
01/2020

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
Direct trifluoromethoxylation without OCF3-carrier via in-situ
generation of fluorophosgene
Jérémy Saiter, Thomas Guérin, Morgan Donnard, Armen Panossian, Gilles Hanquet* and Frédéric R.
Leroux*
[a]
Jérémy Saiter, Dr. Thomas Guérin, Dr. Morgan Donnard, Dr. Armen Panossian, Dr. Gilles Hanquet* and Dr. Frédéric R. Leroux*
Université de Strasbourg, Université de Haute-Alsace, CNRS, UMR 7042-LIMA, ECPM, 67000 Strasbourg, France
E-mail: frederic.leroux@unistra.fr, ghanquet@unistra.fr ; homepage URL: http://coha.unistra.fr
Supporting information for this article is given via a link at the end of the document
Abstract: Owing to the high interest into the OCF3 group for
pharmaceutical and agrochemical applications, trifluoromethoxylation
received great attention in the last years with several new methods to
perform it. Yet, it most often requires the beforehand preparation of
specific F3CO– transfer reagents, which can be toxic, expensive,
unstable and/or which generate undesired side-products upon
consumption. To circumvent this, the in-situ generation of gaseous
fluorophosgene from triphosgene, its conversion by fluoride into the
OCF3 anion and the direct use of the latter in nucleophilic substitutions
is an appealing strategy, which, although it has been approached
recently, has not been fully exploited. We disclose herein our efforts
towards this aim.
structures thanks to trifluoromethoxylating reagents especially
designed for this reaction such as trifluoromethyl
trifluoromethanesulfonate (TFMT) whose major drawbacks are
the low boiling point and high cost,[16–20] trifluoromethyl
arylsulfonates (TFMS) which has to be synthesized from the
corresponding sulfonic acids and 1-(trifluoromethyl)-1,2-
benziodoxol-3(1H)-one
dinitrotrifluoromethoxybenzene
(Togni’s
reagent)[21,22]
which
or
2,4-
afford
trifluoromethoxylated products in low yield.[23] All these reagents
release F3CO– upon activation by fluoride anions. Very recently,
trifluoromethyl methyl ether, a gas of comparable price with
regard to TFMT, was used as trifluoromethoxide source by
treatment with tertiary amine nucleophiles,[25] and O-
trifluoromethyl benzaldoximes (TFBO) were shown to liberate
F3CO– in the presence of cesium carbonate as Brønsted base
(Figure 1).[26]
Introduction
Sometimes called “super-halogen”, the trifluoromethoxy
group shows remarkable electronic properties. It is slightly more
electron withdrawing than fluorine by induction (OCF3: c=3.7, σI=
+0.50 to +0.55/ F: c=3.98, σI= +0.51 to +0.54) and also less
electron-donating by resonance (OCF3: σR= −0.13 to −0.18/ F:
σR= −0.32 to −0.40).[1–5] One reason for these special properties
is the hyperconjugation of the non-bonding p orbitals of the
oxygen atom with the antibonding σ* orbitals of the C-F bonds. As
a result, in the case of trifluoromethoxylated aromatics, the
trifluoromethoxy group lies in an orthogonal plane with regard to
the aromatic ring, unlike its non-fluorinated analogue (OMe).[5–7]
The lipophilicity of this group is also noticeable since it is one of
the most hydrophobic moieties (Hansch lipophilicity parameter π
= +1.04).[8,9] Such original characteristics make understandable
its presence in pharmaceutical and agrochemical bioactive
ingredients. Among the general strategies to access the C–OCF3
motif, the direct introduction of the already constructed OCF3
group onto molecules is arguably the most appealing strategy, as
it allows functionalization at any stage of a synthesis. Whereas
radical trifluoromethoxylation has been successfully used only on
(hetero)arenes so far, anionic trifluoromethoxylation is more
general and has been more studied, although it remains
challenging.[8,10–14] One of the reasons is the easy decomposition
Figure 1: Trifluoromethoxide transfer reagents
However, using a trifluoromethoxylating reagent, besides adding
at least one step to the global procedure, leads upon generation
of the F3CO– anion to the presence of a side-product in the
reaction mixture (sulfonyl-, aryl- or acyl fluoride for example). An
alternate strategy would be to prepare in situ and stabilize the
trifluoromethoxide anion and to react it directly with the substrate.
Early work on this strategy consisted in forming, from
fluorophosgene and a fluoride source, a trifluoromethoxide salt,
which could be used in nucleophilic substitution.[27–29] However,
the use of COF2 gas is quite hazardous and represents a serious
limitation. Recently, an interesting approach for the in situ
generation and use of trifluoromethoxide was reported by Xiao et
of the trifluoromethoxy anion into fluorophosgene and fluoride[15]
,
and the poor nucleophilicity of trifluoromethoxide. Nevertheless,
last decades witnessed an increasing number of reports
concerning the introduction of trifluoromethanolate onto different
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
Halide trifluoromethoxylation process using TFBz (Hu et al.)
Tube 1
Tube 2
1) KF (1.5 equiv.)
18-crown-6 (1.5 equiv.)
2) Benzoyl bromide (1 equiv.)
AgF
R-X
KF (45 equiv.)
O
O
18-crown-6 (1.35 equiv.)
O
OCF3
OCF3
R
Cl3C
CCl3
F
F
X = Br, I
MeCN, -78 - 80 °C
MeCN, -78 - -30 °C
2 examples
78 - 95%
(5 equiv.)
70%
· Important quantities of reagent needed · Involves uncommon reagent (spray-dried KF) · Glovebox required ·
Halide trifluoromethoxylation process using AgOCF3 stock solution (Schoenebeck et al.)
AgF (3 equiv.)
O
R-X
Filtration
OCF3
+
AgOCF3
AgOCF3
AgCl
R
Cl3C
CCl3
MeCN, 20 °C
X = Br, I
(0.33 equiv.)
stock
solution
5 examples
82 - 98%
· Glovebox required ·
This work
O
Tube 1
Tube 2
1) AgF (3 equiv.)
2) R-I (1 equiv.)
NEt3·3HF (9 equiv.)
O
OCF3
R
Cl3C
(2 equiv.)
· Involves common reagents · No glovebox needed · Safe process ·
CCl3
F
F
MeCN, -78 - 50 °C
MeCN, -25 - 20 °C
10 examples
10 - 99%
Figure 2 Halide trifluoromethoxylation processes involving fluorophosgene generation from triphosgene
al., where difluorocarbene is oxidized, forming COF2, which
subsequently reacts with AgF to form the targeted F3CO- that can
al., where difluorocarbene is oxidized, forming COF2, which
subsequently reacts with AgF to form the targeted F3CO- that can
be engaged in nucleophilic substitutions with various C(sp3)
halides.[30] Notably, this approach generates several side-
products coming from the carbene source and the oxidant.
Another user-friendly procedure was described by Hu and
coworkers in their preparation of trifluoromethyl benzoate (TFBz),
a trifluoromethoxylating reagents effective in various kinds of
The approach described by Hu and co-workers[24] seemed
to us the best starting-point as it should be transposable to a direct
process not involving an intermediate trifluoromethoxylation
reagent (TFBz) by just replacing benzoyl bromide with a series of
substrates. In their paper, the authors described a screening of
fluoride sources and conditions in THF for the conversion of
fluorophosgene into F3CO–, as well as a study of the relative
amounts of fluoride sources and 18-C-6 crown ether in both
stages —i.e. F2CO generation and conversion to F3CO–.
We started our study by choosing α-iodoacetophenone 1a
as a model substrate and using conditions similar to the ones
employed by Hu et al. except for the temperature in Tube 1 —
where fluorophosgene is generated—, which was maintained at
50 °C instead of 80 °C to prevent condensation of acetonitrile in
the second tube, and for temperature in Tube 2 —where the
F3CO– anion is formed—, which was initially of -25 °C instead of -
78 °C for compatibility with more solvents. However, in our case,
no desired trifluoromethoxylation product (2a) was observed
(Table 1, entry 1). As we observed a gas generation in the first
tube, we believed the fluorophosgene generation to be effective.
Consequently, we first focused on the conditions in the second
tube. Therefore, we moved to silver fluoride in the absence of 18-
trifluoromethoxylation reactions.[24]
.
These authors treated
triphosgene (5 equiv.) with large amounts of spray-dried
potassium fluoride (45 equiv.), leading in situ to the generation of
an overpressure of gaseous fluorophosgene. The latter was
trapped after its transfer onto a fluoride suspension, affording the
trifluoromethoxide anion, which was able to perform a nucleophilic
substitution on benzoyl bromide to finally provide TFBz (vide
supra).[24] Finally, while our present manuscript was in preparation,
Schoenebeck et al. reported, in line with Hu’s work, the ex situ
preparation of AgOCF3 by mixing triphosgene and silver fluoride.
Interestingly, the final stock solution of this salt was shown to be
very stable and could be used in trifluoromethoxylations and other
reactions; however, a filtration of insoluble silver salts in a
glovebox was required.[31]
C-6 (entry 2), which provided
a mixture of 1a, 2a and
monofluorinated product 3a in a 27/37/36 ratio. When the solvent
was switched to acetonitrile and using the KF/18-C-6 combination
again (entry 3), we were pleased to observe a 42/58 ratio between
1a and 2a and no monofluorinated 3a. Finally, when used in
acetonitrile, AgF gave a full conversion of the starting material into
the trifluoromethoxylated product (entry 4) and the following
studies were conducted using it as fluoride source in Tube 2.
At this point, we noticed that the overpressure of toxic
fluorophosgene needed before its transfer onto the fluoride
suspension could be a safety issue. To avoid this problem, we
used a slightly different method for our following experiments,
Consequently, an efficient process combining the formation
of both COF2 and then F3CO–, and the direct use of the latter in
trifluoromethoxylation, altogether in situ without the intermediate
formation and isolation of a trifluoromethoxylating reagent, would
still be highly valuable. We describe herein our efforts towards this
aim (Figure 2).
Results and Discussion
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
which consists in allowing the transfer of the generated gas onto
the fluoride salt (here, AgF) suspension as soon as it is generated.
This allows the pressure to be distributed in the whole reaction
set-up and gratifyingly gave the same results (Table 1, entry 5).
We therefore pursued our study according to this technical
modification (Table 1).
entries 7 vs. 8), probably indicating an inefficient generation of
COF2. Accordingly, we maintained a constant 1/9/0.27 ratio of
triphosgene/KF/18-C-6 and decreased the global number of
equivalents, with reactions involving 1.5 and 1 equivalent of
triphosgene and the proportional amounts of other reagents in
Tube 1 (Table 2, entries 9–10). In these experiments, the
monofluorinated product 3a was formed in significant amounts,
highlighting the lower limits for the quantities of reagents for
fluorophosgene generation (Table 2).
Table 1: Modifications of Hu’s conditions for the trifluoromethoxylation of α-
iodoacetophenone
Table 2: Study of reactions parameters used in Tube 1 for fluorophosgene
generation
Entry
Solvent
THF
F- source
KF/18-C-6 1:1
AgF
1a/2a/3a ratio[a],[b]
100/nd/nd
27/37/36
1a/2a/3a
ratio[a],[b]
Entry
X
Y
Z
Solvent
1
2
THF
1
5
45
32
32
32
32
27
18
12
13.5
9
1.35
1,35
1
MeCN
MeCN
MeCN
MeCN
Diglyme
MeCN
MeCN
MeCN
MeCN
MeCN
0/100/0
1/99/0
3
MeCN
MeCN
MeCN
KF/18-C-6 1:1
AgF
42/58/0
2
3
4
0/100/0
3
3
3/97/0
5[c]
AgF
0/100/0
4
3
0
0/76/24
29/38/33
5/95/0
Conditions A : 1) KF (45 equiv.), 18-crown-6 (1.35 equiv.), MeCN, -78 °C.
2) MeCN, 50 °C, 1h. 3) Connection to Tube 2, MeCN, 50 °C, 2h [a] 1H
NMR ratio in the crude mixture using the methylene signal of each
compound as characteristic signal. [b] nd = not detected. [c] Gas transfer
during the whole process
5
3
0
6
3
0.8
0.6
0.36
0.4
0.26
7
2
0/100/0
11/77/12
0/78/22
11/12/77
We decided therefore to decrease these amounts to the
most atom-economical point without affecting yields. When we
reduced the equivalents of triphosgene and KF in Tube 1 to
respectively 3 and 32 equivalents with 1.35 equivalents of crown
ether, the results remained unchanged (Table 2, entries 1 and 2).
We were also interested in decreasing the quantity of crown ether
used. Consequently, we carried out an experiment with
respectively 3, 32 and 1 equivalents of triphosgene, KF and 18-
C-6 (Table 2, entry 3), which gave similar results. Performing the
reaction in the absence of crown ether proved detrimental, as a
significant amount of monofluorinated product 3a was formed
(Table 2, entry 4). Following this idea, instead of using 18-C-6 as
additive to solubilize KF and enhance its reactivity in acetonitrile,
we wondered if diglyme as solvent would be suitable for the
reaction in Tube 1; unfortunately, the conversion of the starting
material was incomplete under these conditions (Table 2, entry 5).
Hence, we turned back to the use of acetonitrile as solvent and
18-C-6 as additive. We carried out reactions involving
respectively, 3 and 27 equivalents (Table 2, entry 6), 2 and 18
equivalents (entry 7), 2 and 12 equivalents (entry 8) of
triphosgene and KF, always with a proportional amount of crown
ether with regard to KF. The two first experiments gave an
unchanged yield (Table 2, entries 5–6). With the last ratio
(triphosgene/KF = 1/6), which is theoretically the lowest possible
amount of KF relative to triphosgene to achieve full conversion of
the latter into COF2, unreacted substrate 1a was observed and
the desired product 2a was formed along undesired 3a (Table 2,
8
2
9
1.5
1
10
Conditions B : 1) AgF (1.5 equiv.), MeCN, -25 °C. 2) Connection to Tube
1, MeCN, -25 °C, 2h. 3) 1a (1 equiv.), MeCN, 20 °C, 16h. [a] 1H NMR ratio
in the crude mixture using the methylene signal of each compound as
characteristic signal. [b] nd = not detected
We then considered increasing the fluorophosgene
pressure by reducing the free volume of the reaction, i.e. reducing
the glassware volume. By this means, we would possibly be able
to keep decreasing the amounts of reagents. Unfortunately,
despite an important search for the good system, the
reproducibility of the obtained results was not satisfying. Finally,
the reaction conditions demanding 2 equivalents of triphosgene,
18 equivalents of potassium fluoride and 0.6 equivalent of crown
ether were selected for the generation of fluorophosgene (Tube
1).
The next parameter to be examined was the fluoride source
in the second tube, for the generation of the trifluoromethoxide
anion.
Tetrabutylammonium
fluoride
(TBAF),
tetramethylammonium fluoride (TMAF), tetrabutylammonium
difluorotriphenylsilicate (TBAT) and cesium fluoride were
assessed as potential replacements for AgF; however, they all
proved inefficient (Table 3). Lastly, the amount of silver fluoride
employed in Tube 2 was investigated. When the reaction was
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
carried out with only one equivalent of the metal salt, the amount
of trifluoromethoxylated product 2a in the crude material
decreased to 73% (Table 3, entry 7).
Trifluoromethoxylation of 3-iodoprop-1-ynylbenzene 1h indicates
that the presence of an alkyne does not perturbate the reaction
(Table 4, entry 8). Reactions with secondary iodides 1i-k were
also carried out, however, lower yields were obtained (Table 4,
entries 9–11). As often when it comes to trifluoromethoxylated
products, isolated yields are in some cases much lower than NMR
yields because of the difficult purification from starting materials
and monofluorinated side-products (Table 4).
Table 3: Screening of fluoride sources in Tube 2 for the generation of F3CO–
Table 4: Scope of the reaction
Entry
F– sources
AgF
1a/2a/3a ratio[a],[b]
0/100/0
1
2
3
4
5
6
7
1H NMR ratio[a],[b]
19F
yield[c]
NMR
TBAF[c]
TMAF
90/0/10
Entry
R-X (1)
(isolated
yield)
(1)
(2)
(3)
74/0/26
TBAT
94/6/0
1
2
-
100
-
93% (92%)
90% (94%)
CsF
100/0/0
1a
Flamed CsF
AgF[d]
95/4/1
-
100
-
22/73/5
1b
Conditions C : 1) KF (18 equiv.), 18-crown-6 (0.6 equiv.), MeCN, -78 °C.
2) Connection to Tube 2, MeCN, 50 °C, 2h. [a] 1H NMR ratio in the crude
mixture using the methylene signal of each compound as characteristic
signal. [b] nd = not detected. [c] 1 M in THF. [d] 1 equiv of AgF was used
3
4
5
6
-
-
-
-
100
100
100
90
-
-
-
(87%)
1c
70% (74%)
1d
Unfortunately, while studying the scope of the reaction, we
observed some reproducibility issues due to the quality of the
commercial spray-dried potassium fluoride which must be further
dried and handled very carefully. Additionally, its cost and its low
availability pushed us to envision another fluoride source in Tube
1. To this end, a reaction involving NEt3●3HF instead of KF/18-C-
6 to generate fluorophosgene was carried out. Triethylamine
trihydrofluoride being less expensive than potassium fluoride, it
could even be used in larger amount to promote more efficiently
the generation of fluorophosgene gas. Also, the use of NEt3●3HF
makes the mixture of Tube 1 homogenous, which would be
suitable for an industrial application. After a quick optimization of
the reaction conditions to the employment of NEt3●3HF, the use
of 3 equivalents of triphosgene, 9 equivalents of NEt3●3HF in
Tube 1 and 3 equivalents of AgF in Tube 2 was determined as the
best conditions.
Quant.
(99%)
1e
traces 94% (16%)
1f
7
10 39
51
10%
1g
8
-
-
98
39
traces 87% (56%)
1h
37%
61
9
(27%)[d]
1i
1j
10
19 81
-
72%
Starting with benzyl-2-iodacetate 1b as an α-iodoester, the
trifluoromethoxylated product was obtained in excellent yield.
(Table 4, entry 2) Benzylic iodides 4-(iodomethyl)benzonitrile 1c
and 1-bromo-4-(iodomethyl)benzene 1d provided good results
(Table 4, entries 3–4). An excellent yield was also obtained for the
reaction using iodoundecane 1e as starting material (Table 4,
entry 5). When submitted to the reaction conditions, 1-iodoethyl-
2-benzene 1f also provided good results although purification
proved difficult (Table 4, entry 6). Unsurprisingly, reactions
involving bromides as substrates provided lower yields as shown
by the reaction with bromoacetophenone 1g (Table 4, entry 7).
11
72% (19%)
1k
Conditions D : 1) NEt3●3HF (9 equiv.), MeCN, -78 °C. 2) MeCN, 50 °C, 2h.
Conditions E : 1) AgF (3 équiv.), MeCN, -25 °C. 2) MeCN, -25 °C, 2h.
3) 1 (1 equiv.), MeCN, 20 °C, 16h. [a] 1H NMR ratio in the crude mixture
using the methylene signal of each compound as characteristic signal. [b]
nd = not detected. [c] Determined using fluorobenzene as internal standard.
[d] isolated as 8:2 mixture with 1j
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
CDCl3): δ = 3.0, 111.7, 118.6, 129.6, 132.7, 144.8. In accordance with
previous reports.[34]
1-Bromo-4-(iodomethyl)benzene 1d 1H NMR (400 MHz, CDCl3) δ = 4.40
(s, 2H), 7.25 (d, 3J = 8.5 Hz, 2H), 7.42 (d, 3J = 8.5 Hz, 2H). 13C NMR (400
MHz, CDCl3): δ = 4.3, 121.9, 130.5, 132.1, 138.5. In accordance with
previous reports.[35]
Conclusion
In conclusion, the method presented in this article allows the
preparation of a range of trifluoromethoxylated structures with no
use of an intermediate trifluoromethoxylating reagent, but in situ
construction and use of the F3CO– anion as a salt, avoiding the
presence of a useless side-product resulting from activation of an
OCF3 carrier. Product separation is thus facilitated. This method
relies on the in-situ generation of gaseous fluorophosgene which
we believe to be the most atom-economical manner to generate
the OCF3 anion.
(3-Iodoprop-1-ynyl)benzene 1g 1H NMR (400 MHz, CDCl3) δ 3.97 (s, 2H),
7.28 – 7.36 (m, 3H), 7.37 – 7.48 (m, 2H). 13C NMR (101 MHz, CDCl3) δ -
17.3, 85.4, 86.4, 122.6, 128.4, 128.8, 131.8. In accordance with previous
reports.[36]
2-Iodo-1-phenyl-1-propanone 1i 1H NMR (400 MHz, CDCl3) δ 2.08 (d, 3J
= 6.7 Hz, 3H), 5.50 (q, 3J = 6.7 Hz, 1H), 7.43 – 7.52 (m, 2H), 7.54 – 7.62
(m, 1H), 7.98 – 8.05 (m, 2H). 13C NMR (101 MHz, CDCl3) δ 18.2, 22.1,
128.8, 128.8, 133.6, 133.8, 194.9. In accordance with previous reports.[37]
Experimental Section
3
b
-Iodopregn-5-en-20-one 1k 1H NMR (400 MHz, CDCl3) δ 0.63 (s, 3H),
0.91 – 1.03 (m, 1H), 1.03 – 1.20 (m, 3H), 1.08 – 1.32 (m, 3H), 1.36 – 1.79
(m, 7H), 1.94 – 2.08 (m, 2H), 2.12 (s, 3H), 2.11 – 2.36 (m, 3H), 2.52 (t, 3J
= 8.9 Hz, 1H), 2.68 (ddd, 2J = 13.8, 3J = 4.5, 2.2 Hz, 1H), 2.93 (m, 1H),
3.81 – 4.21 (m, 1H), 5.34 (m, 1H). 13C NMR (101 MHz, CDCl3) δ 13.3, 19.4,
20.9, 23.0, 24.6, 30.2, 31.7, 31.7, 36.6, 36.7, 38.9, 42.0, 44.1, 46.5, 50.4,
57.0, 63.8, 121.5, 142.9, 209.6. In accordance with previous reports.[38]
General information: Starting materials, if commercially available, were
purchased from standard suppliers (Sigma-Aldrich, Fluorochem, ABCR,
Acros, Alfa-Aesar or Apollo Scientific) and used as such, provided that
adequate checks by NMR analysis had confirmed the claimed purity. THF
was dried by distillation over sodium/benzophenone and used within the
day. Acetonitrile was firstly distilled on P2O5 and then on CaH2 and used
within the day. Air- and moisture-sensitive materials were stored and
handled under an atmosphere of argon. Reactions were carried out under
an atmosphere of argon when needed using classic Schlenk-line
techniques. Reactions were monitored by using thin-layer chromatography
with precoated silica on aluminum foils (0.25 mm, Merck silica-gel (60-
F254)). Flash column chromatography was performed on VWR silica gel
(40–63 μm) using the indicated solvents, the solvent systems being
indicated in v/v. Spectroscopic NMR and MS data were obtained using
chromatographically homogeneous samples. 1H NMR (400 or 500 MHz),
19F NMR (376 or 472 MHz) and 13C NMR (101 or 126 MHz) spectra were
recorded in CDCl3 on a Bruker Avance III HD 400 MHz or 500 MHz
instruments respectively, unless noted otherwise. Chemical shifts are
reported in parts per million (ppm) and are referred to partially deuterated
chloroform (δ[1H] = 7.26 ppm and δ[13C] = 77.16 ppm). Multiplicities were
abbreviated as br s (broad signal), s (singlet), d (doublet), t (triplet), q
(quartet), p (pintuplet) m (multiplet), td (triplet of doublets), dd (doublet of
doublets). Coupling constants J were given in Hz. Spectra were processed
with the program MestReNova v14.1.0-24037
General procedure for trifluoromethoxylation via in-situ generation of
fluorophosgene: A H-shaped tube (see figure below)1 was used for this
reaction. In tube 1 was added NEt3●3HF (6 eq.) and freshly distilled
acetonitrile (20 mL) under argon. In tube 2 was added AgF (3 eq) and
freshly distilled acetonitrile (3.94 mL) under Ar and the tube was covered
with aluminum foil to be protected from light. Tube 2 was sealed and tube
1 was cooled to -78 °C. When the solution in tube 1 was frozen,
triphosgene (3 eq.) was added to tube 1, which was then sealed and the
cooling bath was removed. Tube 2 was cooled to -30 °C. Tube 1 was
allowed to warm gently to room temperature and was then heated to 50 °C
for 2 h. The substrate 1 (1 eq., 2 mmol) was syringed (using a minimal
amount of solvent to solubilize it when it is a solid) into tube 2 through the
PTFE coated septum. The mixtures in tubes 1 and 2 were then stirred
overnight at room temperature. The reaction mixture in tube 2 was then
filtered through Celite® and the solvent was evaporated under reduced
pressure. The crude material was then dissolved in the minimal amount of
DCM and washed with brine. The aqueous layer was extracted three times
with DCM. The combined organic phases were then dried over Na2SO4
and the solvent was evaporated under reduced pressure. The crude
mixture was purified by flash column chromatography on silica gel.
General procedure for the synthesis of starting alkyl, benzyl or
propargyl iodides: All non-commercial iodides used as starting materials
were synthesized by a Finkelstein reaction. To a solution of Alkyl, benzyl
or propargyl bromide in acetone (1 eq., 0.2M) was added a solution of NaI
in acetone (1.1 eq., 0.2M). The reaction was monitored by 1H NMR. If
necessary, more of the NaI solution could be added to obtain a full
conversion of the starting material. The mixture was then filtered through
a pad of Celite® and the solvent removed under reduced pressure. DCM
was added to reach a concentration of 0.2M and the mixture was filtered
again and the filtrate was concentrated under reduced pressure. The
desired iodides obtained were used without further purification.
Tube 1
Tube 2
α-Iodoacetophenone 1a 1H NMR (400 MHz, CDCl3) δ = 4.37 (s, 2H), 7.49
(dd, 3J = 8.0, 7.5 Hz, 2H), 7.60 (dd, 3J= 8.0, 7.5 Hz, 1H), 7.99 (d, 3J= 7.5
Hz, 2H). 13C NMR (400 MHz, CDCl3): δ = 2.0, 128.9, 129.1, 133.2, 133.9,
193.0. In accordance with previous reports.[32]
Benzyl 2-iodoacetate 1b 1H NMR (400 MHz, CDCl3) δ = 3.74 (s, 2H), 5.18
(s, 2H), 7.30 – 7.45 (m, 5H). 13C NMR (400 MHz, CDCl3): δ = -5.4, 67.9,
128.4, 128.6, 128.7, 135.2, 168.7. In accordance with previous reports.[33]
4-(Iodomethyl)benzonitrile 1c 1H NMR (400 MHz, CDCl3) δ = 4.31 (s,
2H), 7.34 (d, 3J = 8.3 Hz, 2H), 7.43 (d, 3J = 8.4 Hz, 2H) 13C NMR (400 MHz,
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
Obtained from the general procedure as a yellow oil (226 mg, 1.13 mmol,
56%). Purified by flash column chromatography (heptane).
Tube 1
Tube 2
1H NMR (400 MHz, CDCl3) δ 4.83 (s, 2H), 7.28 – 7.43 (m, 4H), 7.43 – 7.52
(m, 2H). 13C NMR (101 MHz, CDCl3) δ 56.1 (q, 3JC-F = 4.5 Hz), 80.9, 88.3,
121.7, 121.8 (q, 1JC-F = 257.2 Hz), 128.5, 129.3, 132.0. 19F NMR (377 MHz,
CDCl3) δ -60.69 (s, 3F). In accordance with previous reports.[26]
2-Trifluoromethoxy-1-phenyl-1-propanone 2i
Obtained from the general procedure as a yellow oil (218.18 mg, 0.54
mmol, 27%). Purified by flash column chromatography (pentane:EtOAc
10:1).
1H NMR (400 MHz, CDCl3) δ 1.66 (dd, 3J = 6.9, 0.8 Hz, 3H), 5.49 (q, 3J =
6.9 Hz, 1H), 7.44 – 7.55 (m, 2H), 7.58 – 7.67 (m, 1H), 7.93 – 8.02 (m, 2H).
13C NMR (101 MHz, CDCl3) δ 18.8, 75.5 (q, 3JC-F = 2.5 Hz), 121.6 (q, 1JC-
1
Attention must be paid to the distance between the tubes which must
allow the easy handle of hot and cold bathes. In our case, a distance of 23
cm was used.
α-Trifluoromethoxyacetophenone 2a
Obtained from the general procedure as a light yellow oil (375 mg, 1.84
mmol, 92%).
= 256.6 Hz), 128.9, 129.1, 133.7, 134.2, 195.1. 19F NMR (377 MHz,
F
CDCl3) δ -58.78 (s,3F). In accordance with previous reports.[41]
1H NMR (400 MHz, CDCl3) δ 5.18 (s, 2H), 7.56 – 7.47 (m, 2H), 7.68 – 7.59
(m, 1H), 7.94 – 7.87 (m, 2H). 13C NMR (400 MHz, CDCl3): δ 68.4 (q, 3JC-F
= 2.9 Hz), 120.5 (q, 1JC-F = 256.7 Hz), 128.0, 129.2, 133.9, 134.5, 190.3.
19F NMR (400 MHz, CDCl3) δ -61.00 (s, 3F). In accordance with previous
reports.[17]
Ethyl 2-(trifluoromethoxy)propanoate 2j
Obtained from the general procedure in mixture with the corresponding
iodide (8/2)
1H NMR (400 MHz, CDCl3) δ 1.30 (t, 3J = 7.1 Hz, 3H), 1.52 – 1.66 (m, 4H),
4.25 (qd, 3J = 7.1, 0.8 Hz, 2H), 4.68 (q, 3J = 6.9 Hz, 1H). 19F NMR (377
MHz, CDCl3) δ -59.65 (s,3F). In accordance with previous reports.[42]
Benzyl 2-(trifluoromethoxy)acetate 2b
Obtained from the general procedure as a light yellow oil (440 mg, 1.88
mmol, 94%).
1H NMR (400 MHz, CDCl3) δ 4.52 (s, 2H), 5.25 (s, 2H), 7.30 – 7.44 (m,
5H). 13C NMR (400 MHz, CDCl3): δ 63.2 (q, 3JC-F = 3.6 Hz), 67.7, 121.6 (d,
1JC-F = 256.9 Hz), 128.7, 128.9, 128.9, 134.8, 166.0. 19F NMR (400 MHz,
CDCl3) δ -61.42 (s, 3F). In accordance with previous reports.[17]
3 -(Trifluoromethoxy)pregn-5-en-20-one 2k
b
Obtained from the general procedure as a white solid (148 mg, 0.38 mmol,
19%). Purified by flash column chromatography (pentane:EtOAc 10:0.5).
1H NMR (400 MHz, CDCl3) δ 0.64 (s, 3H), 0.67 (t, 3J = 7.0 Hz, 1H), 0.86 –
1.01 (m, 1H), 1.02 (s, 3H), 1.06 – 1.31 (m, 3H), 1.39 – 1.81 (m, 7H), 1.87
– 2.09 (m, 4H), 2.13 (s, 3H), 2.14 – 2.27 (m, 1H), 2.38 – 2.57 (m, 3H), 3.88
– 4.26 (m, 1H), 5.41 (m, 1H). 13C NMR (101 MHz, CDCl3) δ 13.3, 19.3,
21.1, 22.9, 24.6, 28.7, 31.6, 31.9, 36.5, 36.9, 38.9, 39.0, 44.1, 50.0, 56.9,
4-((Trifluoromethoxy)methyl)benzonitrile 2c
Obtained from the general procedure a yellow oil (350 mg, 1.74 mmol,
87%).
3
1
63.8, 78.6 (q, JC-F = 1.9 Hz), 121.7 (q, JC-F = 254.2 Hz), 123.1, 139.2,
209.5.19F NMR (377 MHz, CDCl3) δ -57.54 (s,3F). HR-MS m/z: Calcd for
C22H32F3O2 [M+H] 385.2352 found 385.2349. IR (neat, cm-1) n 2938, 2851,
1701, 1469, 1357, 1274, 1205, 1194, 1128, 1012, 870, 662 mp = 85-90 °C
1H NMR (400 MHz, CDCl3) δ 5.05 (s, 2H), 7.48 (d, 3J = 8.0 Hz, 2H), 7.70
(d, 3J = 8.4 Hz, 2H). 13C NMR (400 MHz, CDCl3): δ 67.8 (q, 3JC-F = 3.6 Hz),
1
112.9, 118.4, 121.7 (q, JC-F = 256.3 Hz), 128.1, 132.7, 139.1. 19F NMR
(400 MHz, CDCl3) δ -60.75 (s,3F). In accordance with previous reports.[39]
1-Bromo-4-((trifluoromethoxy)methyl)benzene 2d
Obtained from the general procedure a light yellow oil (377 mg, 1.48 mmol,
Acknowledgements
74%).
1H NMR (400 MHz, CDCl3) δ 4.94 (s, 2H), 7.24 (d, 3J = 8.5 Hz, 2H), 7.53
(d, 3J = 8.8 Hz, 2H). 13C NMR (400 MHz, CDCl3): δ 68.4 (q, 3JC-F = 3.5 Hz),
121.8 (d, 1JC-F = 255.8 Hz), 123.2, 129.7, 132.1, 133.0. 19F NMR (400 MHz,
CDCl3) δ -60.45 (s, 3F). In accordance with previous reports.[39]
This work was supported by the French Centre National de la
Recherche Scientifique (CNRS). We thank the French Agence
Nationale pour la Recherche (ANR) (grant number ANR-17-
CE07-0008-01, DEFIS) for financial support. The French Fluorine
Network (GIS CNRS Fluor) is also acknowledged.
1-(Trifluoromethoxy)undecane 2e
Obtained from the general procedure a colorless oil (475.81 mg, 1.98
mmol, 99%).
Keywords: Fluorine • Trifluoromethoxylation •
Fluorophosgene • Nucleophilic substitution
1H NMR (400 MHz, CDCl3) δ 0.80 – 1.01 (m, 2H), 1.20 – 1.44 (m, 16H),
1.68 (p, 3J = 6.7 Hz, 2H), 3.95 (t, 3J = 6.6 Hz, 2H). 13C NMR (101 MHz,
CDCl3) δ 14.1, 22.7, 25.4, 28.7, 29.1, 29.3, 29.4, 29.5, 29.7, 31.9, 67.5,
67.5, 67.5, 67.6, 120.4, 123.0. 19F NMR (377 MHz, CDCl3) δ -60.64 (s, 3F).
GC-MS: [M]+ 240.200, [M-OCF3]+ 155.100, [M-CH2OCF3]+ 141.100, [M-
CH3CH2]+ 211.200, [M-CH3CH2CH2]+ 197.100. IR (neat, cm-1) n 2925,
2856, 1467, 1408, 1263, 1136, 1219, 1045
[1]
[2]
[3]
R. W. Taft, E. Price, I. R. Fox, I. C. Lewis, K. K. Andersen, G. T.
Davis, J. Am. Chem. Soc. 1963, 85, 3146–3156.
R. W. Taft, E. Price, I. R. Fox, I. C. Lewis, K. K. Andersen, G. T.
Davis, J. Am. Chem. Soc. 1963, 85, 709–724.
I. W. Serfaty, T. Hodgins, E. T. McBee, J. Org. Chem. 1972, 37,
2651–2655.
[2-(Trifluoromethoxy)ethyl]benzene 2f
Obtained from the general procedure as a yellow oil (60.5 mg, 0.32 mmol,
16%).
1H NMR (400 MHz, CDCl3) δ 3.01 (t, 3J = 7.2 Hz, 2H), 4.16 (t, 3J = 7.2 Hz,
2H), 7.20 – 7.29 (m, 3H), 7.30 – 7.38 (m, 2H). 13C NMR (101 MHz, CDCl3)
δ 35.5, 68.0 (q, 3JC-F = 3.0 Hz), 121.9 (d, 1JC-F = 254.3 Hz), 127.2, 128.9,
129.1, 136.8. 19F NMR (377 MHz, CDCl3) δ -60.69 (s,3F). In accordance
with previous reports.[40]
[4]
[5]
[6]
[7]
W. A. Sheppard, J. Am. Chem. Soc. 1961, 83, 4860–4861.
W. A. Sheppard, J. Am. Chem. Soc. 1963, 85, 1314–1318.
F. E. Herkes, J. Fluorine Chem. 1977, 9, 113–126.
F. Rose-Munch, R. Khourzom, J. P. Djukic, E. Rose, B. Langlois, J.
Vaisserman, J. Organomet. Chem. 1994, 470, 131–135.
F. R. Leroux, B. Manteau, J. Vors, S. Pazenok, Beilstein J. Org.
Chem. 2008, 4.
[8]
[9]
A. Leo, P. Y. C. Jow, C. Silipo, C. Hansch, J. Med. Chem. 1975, 18,
865–868.
[10]
[11]
A. Tlili, F. Toulgoat, T. Billard, Angew. Chem. Int. Ed. 2016, 55,
11726–11735.
T. Besset, P. Jubault, X. Pannecoucke, T. Poisson, Org. Chem.
(3-Trifluoromethoxyprop-1-ynyl)benzene 2h
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
Front. 2016, 3, 1004–1010.
[12]
[13]
F. Toulgoat, T. Billard, Chem 2017, 2, 327–329.
M. A. Hardy, H. Chachignon, D. Cahard, Asian J. Org. Chem. 2019,
8, 591–609.
[14]
[15]
[16]
[17]
K. N. Lee, J. W. Lee, M. Y. Ngai, Tetrahedron 2018, 74, 7127–7135.
M. E. Redwood, C. J. Willis, Can. J. Chem. 1965, 43, 1893–1898.
R. E. Noftle, G. H. Cady, Inorg. Chem. 1965, 4, 1010–1012.
O. Marrec, T. Billard, J. Vors, S. Pazenok, J. Fluorine Chem. 2010,
131, 200–207.
[18]
[19]
[20]
Q. Zhang, J. F. Hartwig, Chem. Commun. 2018, 54, 10124–10127.
D. Chen, L. Lu, Q. Shen, Org. Chem. Front. 2019, 6, 1801–1806.
C. Chen, P. Chen, G. Liu, J. Am. Chem. Soc. 2015, 137, 15648–
15651.
[21]
[22]
[23]
[24]
[25]
R. Koller, Q. Huchet, P. Battaglia, J. M. Welch, A. Togni, Chem.
Commun. 2009, 5993–5995.
S. Guo, F. Cong, R. Guo, L. Wang, P. Tang, Nat. Chem. 2017, 546–
551.
O. Marrec, T. Billard, J. Vors, S. Pazenok, B. R. Langlois, Adv.
Synth. Catal. 2010, 352, 2831–2837.
M. Zhou, C. Ni, Y. Zeng, J. Hu, J. Am. Chem. Soc. 2018, 140,
6801–6805.
J. J. Newton, B. J. Jelier, M. Meanwell, R. E. Martin, R. Britton, C.
M. Friesen, Org. Lett. 2020, 22, 1785–1790.
Y. Li, Y. Yang, J. Xin, P. Tang, Nat. Commun. 2020, 11, 755.
D. D. DesMarteau, Inorg. Chem. 1970, 9, 2179–2181.
Q. Huang, D. D. DesMarteau, Chem. Commun. 1999, 1671–1672.
Q. Huang, D. D. DesMarteau, Inorg. Chem. 2000, 39, 4670–4672.
J. Yu, J. H. Lin, D. Yu, R. Du, J. C. Xiao, Nat. Commun. 2019, 10,
5362.
[26]
[27]
[28]
[29]
[30]
[31]
A. Turksoy, T. Scattolin, S. Bouayad-Gervais, F. Schoenebeck,
Chem. Eur. J. 2020, 26, 2183–2186.
[32]
[33]
P. Goswami, S. Ali, M. M. Khan, A. T. Khan, Arkivoc 2007, 82–89.
M. Yasuda, K. Kiyokawa, K. Osaki, A. Baba, Organometallics 2009,
28, 132–139.
[34]
[35]
[36]
[37]
[38]
[39]
B. Lecea, J. M. Aizpurua, C. P. Palomo, Tetrahedron 1985, 41,
4657–4665.
A. R. Hajipour, A. Zarei, A. E. Ruoho, Synth. Commun. 2006, 36,
1039–1050.
S. Jammi, D. Mouysset, D. Siri, M. P. Bertrand, L. Feray, J. Org.
Chem. 2013, 78, 1589–1603.
T. J. Donohoe, M. A. Kabeshov, A. H. Rathi, I. E. D. Smith, Synlett
2010, 2956–2958.
S. Nagasawa, Y. Sasano, Y. Iwabuchi, Chem. Eur. J. 2017, 23,
10276–10279.
H. Yang, F. Wang, X. Jiang, Y. Zhou, X. Xu, P. Tang, Angew.
Chem. Int. Ed. 2018, 57, 13266–13270.
[40]
[41]
R. Zriba, E. Magnier, J. C. Blazejewski, Synlett 2009, 1131–1135.
H. Kondo, M. Maeno, K. Hirano, N. Shibata, Chem. Commun. 2018,
54, 5522–5525.
[42]
A. A. Kolomeitsev, M. Vorobyev, H. Gillandt, Tetrahedron Lett.
2008, 49, 449–454.
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100429
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
An efficient atom economical trifluoromethoxylation method has been developed with no need for trifluoromethoxylation reagent. This
method relies on the in-situ generation of fluorophosgene and its conversion into OCF3 anion to achieve nucleophilic substitution on a
series of alkyl iodides.
Institute Twitter usernames: @Leroux_group
Keywords: Fluorine • Trifluoromethoxylation • Fluorophosgene • Nucleophilic substitution
8
This article is protected by copyright. All rights reserved.