Electrochemical fluorination using halogen mediators in ionic liquid hydrogen fluoride salt

Article abstract of DOI:10.1149/2.005307jes

In order to utilize ammomium halides (Et4NX, X=Cl, Br, I) as halogenmediator for electrocatalytic fluorination, cyclic voltammetry measurements of the halides were investigated. The catalytic current of the halides in the presence of a dithioacatal compound was observed and the macro-scale electrolysis of dithioacetals using the halogen mediator was also carried out in ionic liquid hydrogen fluoride (HF) salt to give the corresponding fluorinated products in excellent yields. The recycle use of the halogen mediator in the electrochemical fluorination was successfully demonstrated. More inexpensive halides such as potassium bromide and potassium iodide could be soluble in HF salt and worked well as halogen mediator for the electrocatalytic fluorination.

Full text of DOI:10.1149/2.005307jes

G3046
Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
0013-4651/2013/160(7)/G3046/7/$31.00 © The Electrochemical Society
JES FOCUS ISSUE ON ORGANIC AND BIOLOGICAL ELECTROCHEMISTRY
Electrochemical Fluorination Using Halogen Mediators in Ionic
Liquid Hydrogen Fluoride Salt
Kohta Takahashi, Shinsuke Inagi,∗ and Toshio Fuchigami∗∗,z
Department of Electronic Chemistry, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama 226-8502, Japan
In order to utilize ammomium halides (Et4NX, X = Cl, Br, I) as halogen mediator for electrocatalytic fluorination, cyclic voltammetry
measurements of the halides were investigated. The catalytic current of the halides in the presence of a dithioacatal compound was
observed and the macro-scale electrolysis of dithioacetals using the halogen mediator was also carried out in ionic liquid hydrogen
fluoride (HF) salt to give the corresponding fluorinated products in excellent yields. The recycle use of the halogen mediator in the
electrochemical fluorination was successfully demonstrated. More inexpensive halides such as potassium bromide and potassium
iodide could be soluble in HF salt and worked well as halogen mediator for the electrocatalytic fluorination.
© 2013 The Electrochemical Society. [DOI: 10.1149/2.005307jes] All rights reserved.
Manuscript submitted February 27, 2013; revised manuscript received April 8, 2013. Published April 18, 2013. This paper is part of
the JES Focus Issue on Organic and Biological Electrochemistry.
Electrochemical partial fluorination, i.e., anodic oxidation of or-
Experimental
ganic compound, followed by nucleophilic substitution with fluo-
ride anions, has been known as an environmentally friendly method
to provide fluorinated functional molecules.1,2 It is usually car-
ried out in organic solvent-based electrolytic solution containing
poly(hydrogen fluoride) complex of amine (Et3N-nHF) or ammonium
fluoride (Et4NF-nHF). Such hydrogen fluoride (HF) salts are so-called
ionic liquid with relatively low viscosity at room temperature; there-
fore, they are suitable solvent for electrochemical fluorination under
volatile organic solvent-free conditions.3
1
General.— H, 13C and 19F NMR spectra were determined on a
JEOL EX270 (1H: 270.05 MHz, 13C: 67.8 MHz 19F: 254.05 MHz)
spectrometer using CDCl3 as a solvent. The chemical shifts for 1H, 13
C
and 19F NMR were given in δ (ppm) from internal TMS, CDCl3, and
monofluorobenzene (−36.5 ppm), respectively. The product yields
were determined by 19F NMR using monofluorobenzene as an inter-
nal standard. Cyclic voltammetry (CV) measurements were carried
out using ALS 600A Electrochemical Analyzer. The preparative elec-
trolysis was performed using Metronix Corp. constant current power
supply model 5944 monitored with coulomb/amperehour meter HF-
201. High resolution mass spectra (HRMS) were recorded on JEOL
The MStaion JMS-700. Melting points were determined using Yanako
Micro Melting Point Apparatus MP-500P.
Recent development of mediators in electrochemical fluorina-
tion is of importance because they can suppress anode passivation,
which forms non-conducting polymer film on anode to be no longer
electroactive.4 In addition, they improve the reaction efficiency in
terms of their fast electron transfer rate compared to the direct elec-
trochemical fluorination. Several mediators with recycle ability have
been designed and resulted in the development of a “green” fluo-
rination method. Previously, we reported the synthesis of iodoarene
mediator5 and triarylamine mediator6 having an ionic-tag moiety. Both
of them showed effective mediatory behavior for the electrochemical
fluorodesulfurization of dithioacetal compounds, and were compatible
to ionic liquid HF salt; consequently the recycle use of mediator/ionic
liquid system was possible. Polymer-supported iodoarene was also de-
veloped as a recyclable mediator for electrochemical fluorination.7 To
drive this insoluble dispersed mediators, halide anion was employed
to mediate electron transfer between electrode and the solid iodoarene
mediator. Halide anion can be anodically oxidized to form cationic
halogen species or complex mixture, which can be used as oxidant
for organic reactions. Such halogen mediator system is very practi-
cal because; (i) halide anion is generally inexpensive, (ii) oxidation
power can be tuned by appropriate choice of halide, (c) both in-cell
and ex-cell8 methods are available depending on reactions used.
We previously found that bromine mediator (Et4NBr) was effec-
tive for fluorodesulfurization of a dithioacetal compound in Et3N-
3HF/CH2Cl2 (Scheme 1).9 The use of this mediator enabled the elec-
trolysis at a lower potential with a much smaller amount of charge
passed compared to direct electrochemical fluorination. Moreover,
mono-fluorinated product was selectively obtained by using the me-
diator, whereas the direct electrolysis gave di-fluorinated one.10
In this paper, we report the expansive study on the halogen media-
tory system for electrochemical fluorination. At first, electrochemical
measurements of a series of halide in HF salts are investigated, and
then macro-scale electrolysis of various dithioacetal, xanthate ester,
carbonate, and sugar derivatives is demonstrated.
Materials.— All reagents and chemicals were obtained from
commercial sources and used without further purification. Dry
solvents were perchased and used as received. Bis(4-chlorophenyl)-
bis(phenylthio)methane (1a),9 2,2-bis(4-chlorophenyl)-1,3-dithiolane
(5a),11 2,2-bis(4-fluorophenyl)-1,3-dithiolane (5b),11 2,2-diphenyl-
1,3-dithiolane (5c),11 2,2-bis(4-methylphenyl)-1,3-dithiolane (5d),11
2,2-bis(4-methoxylphenyl)-1,3-dithiolane
(5e),11
2,2-bis(4-
fluorophenyl)-1,3-dithiane (6b),12 4-phenylthio-1,3-dioxolane-2-one
(9),13 and phenyl 2,3,4,6-tetra-O-acetyl-1-thio-α-D-glucopyranoside
(11)14 were synthesized according to the literature methods. Xan-
thate esters 7 were prepared according to the literature method.15
Known compounds, O-benzyl-S-methyl dithiocarbonate (7a),
O-(4-chlorobenzyl)-S-methyl dithiocarbonate (7c), and O-(4-
bromobenzyl)-S-methyl dithiocarbonate (7d) were identified by
comparison with the spectral data of their authentic samples.16 The
spectral data for new compounds, 7b, 7e, and 7f are described as
follows.
O-(4-Fluorobenzyl)-S-methyl dithiocarbonate (7b).— Yield: 41%;
Colorless oil; 1H NMR (270.05 MHz, CDCl3): δ = 7.37 (dd, J = 8.6,
5.1 Hz, 2H), 7.04 (dd, J = 8.6, 8.6 Hz, 2H), 5.57 (s, 2H), 2.54 (s, 3H);
13C NMR (67.8 MHz, CDCl3): δ = 215.38, 164.45, 160.81, 130.40 (d,
J = 8.4 Hz), 115.56 (d, J = 21.7 Hz), 74.11, 19.00; 19F NMR (254.05
MHz, CDCl3): δ = −36.22 (tt, J = 8.6, 5.1 Hz 1F); HRMS Calcd for
C9H9FOS2: 216.0079, found: 216.0080.
O-(4-Cyanobenzyl)-S-methyl dithiocarbonate (7e).— Yield: 78%;
White solid, Mp 101.5–103.2◦C; 1H NMR (270.05 MHz, CDCl3): δ
= 7.68 (d, J = 8.3 Hz, 2H), 7.51 (d, J = 8.3 Hz, 2H), 5.69 (s, 2H), 2.59
(s, 3H); 13C NMR (67.8 MHz, CDCl3): δ = 215.51, 140.04, 132.38,
128.40, 118.44, 112.20, 73.16, 19.33; HRMS Calcd for C10H9NOS2:
223.0126, found: 223.0121.
∗Electrochemical Society Active Member.
∗∗Electrochemical Society Fellow.
zE-mail: fuchi@echem.titech.ac.jp
Downloaded on 2013-11-06 to IP 128.143.23.241 address. Redistribution subject to ECS license or copyright; see ecsdl.org/site/terms_use

Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
G3047
Scheme 1.
O-(4-Methylbenzyl)-S-methyl dithiocarbonate (7f).— Yield: 60%;
0.8 V vs. SCE (Figure 2). Upon this redox cycle, the reactions shown
in eqs. 3 and 4 seem to take place as reported previously.9 As shown
in Figure 2, the reversibility of the redox couple was changed de-
pending on the fluorine content in Et3N-nHF. Et3N-3HF contains free
triethylamine (Et3N)25 enough to interact with oxidized species of
Br−; consequently the redox couple became more irreversible. On the
other hand, Et3N was completely protonated in Et3N-5HF, resulted in
showing relatively reversible redox response.
White solid, Mp 58.3–58.7◦C; 1H NMR (270.05 MHz, CDCl3):
δ = 7.20 (d, J = 8.1 Hz, 2H), 7.01 (d, J = 8.1 Hz, 2H), 5.48
(s, 2H), 2.45 (s, 3H), 2.27 (s, 3H); 13C NMR (67.8 MHz, CDCl3):
δ = 215.56, 138.47, 131.57, 129.27, 128.65, 75.14, 21.17, 18.97;
HRMS Calcd for C10H12OS2: 212.0330, found: 212.0330.
Cyclic voltammetry.— CV measurement was carried out at 25◦C
in anhydrous MeCN containing 0.6 M tetraethylammonium perchlo-
rate (Et4NClO4, TEAP) using a platinum disk working electrode (φ
= 0.16 cm) and a platinum plate counter electrode (1 × 1 cm2).
Potentials were referred to a saturated calomel electrode (SCE) us-
ing saturated KCl salt bridge for the junction between SCE and the
MeCN solution. In the case of measurement in ionic liquid HF salt,
potentials were referred to Ag wire (pseudo reference) with external
ferrocene/ferrocenium (Fc/Fc+).
−2e
3Br−−→ Br−3
[1]
[2]
−2e
2Br−3 −→ 3Br2
−2e
2Br− + F−−→ Br2F−
[3]
[4]
Br2F− + F−
BrF− + Br−
ꢀ
ꢁ
2
The voltammograms of Et4NBr were also obtained in neat Et3N-
nHF using Ag wire as pseudo reference electrode and referred to
external ferrocene/ferrocenium (Fc/Fc+) (Figure 3). Similarly to the
results described above, the voltammogram became reversible as the
increase of HF content in Et3N-nHF. The small current observed in
General procedure for electrochemical fluorination using
Et4NX.— Constant current electrolysis was carried out with platinum
anode and cathode (1 × 1 cm2) in 1.1 M HF salt/CH2Cl2 (3.0 mL)
containing substrate (0.1 mmol) and Et4NX mediator (10 mol%) using
an undivided cell at room temperature. After the charge was passed,
the solution was passed through a short column filled with silica gel
using EtOAc as an eluent to remove salts. The yields of products were
determined by 19F NMR using monofluorobenzene (−36.50 ppm) as
an internal standard.
Fluorinated products, 2,9 3,17 8,18–20 10,21 and 1222 were all iden-
tified by comparison with the spectral data of their authentic samples.
Results and Discussion
Cyclic voltammetry measurement of Et4NX in the presence of Et3N-
nHF.— First, the cyclic voltammetry (CV) measurement of Et4NBr
was carried out in 0.6 M TEAP/MeCN at 25◦C (Figure 1). The ob-
served two irreversible oxidation peaks (Ep = 0.70 V and 1.00 V vs.
SCE) were assigned to the oxidation reactions of Br− as shown in
eqs. 1 and 2.23,24 When Et3N-nHF (n = 3–5) was used as supporting
electrolyte (0.6 M) in MeCN, one oxidation peak appeared at around
Figure 2. Cyclic voltammograms of Et4NBr (3 mM) at 25◦C at 100 mV/s in
0.6 M Et3N-nHF/MeCN; circle (◦): n = 3, square (ꢂ): n = 4, diamond (♦):
n = 5.
Figure 1. Cyclic voltammograms of Et4NBr (3 mM) in 0.6 M TEAP/MeCN
at 25◦C at various scan rates; solid circle (●): 10 mV/s, solid square (ꢀ): 50
mV/s, solid diamond (ꢁ): 100 mV/s, square (ꢂ): 300 mV/s, diamond (♦): 400
mV/s, cross (×): 1000 mV/s.
Figure 3. Cyclic voltammograms of Et4NBr (3 mM) at 25◦C at 100 mV/s in
Et3N-nHF; circle (◦): n = 3, square (ꢂ): n = 4, diamond (♦): n = 5.

G3048
Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
Figure 4. Cyclic voltammograms of Et4NCl (3 mM) at 25◦C at 100 mV/s (a) in 0.6 M Et3N-nHF/MeCN, (b) in Et3N-nHF; circle (◦): n = 3, square (ꢂ): n = 4,
diamond (♦): n = 5.
Figure 5. Cyclic voltammograms of Et4NI (3 mM) at 25◦C at 100 mV/s (a) in 0.6 M Et3N-nHF/MeCN, (b) in Et3N-nHF; circle (◦): n = 3, square (ꢂ): n = 4,
diamond (♦): n = 5.
the voltammogram of Et4NBr in Et3N-3HF seems to be due to the
relatively higher viscosity of the HF salt than that of the other HF
salts and MeCN.
ference of viscosity of those salts, namely, the viscosity of Et3N-5HF
is smaller than that of Et3N-4HF.29 From the CV analyzes, Et4NBr
and Et4NI were found to show reversible redox response in both Et3N-
nHF/MeCN and in Et3N-nHF, where they are expected to be mediator
for electrochemical fluorination.
Next, the cyclic voltammetry measurements of other halide, like
Et4NCl and Et4NI, were carried out in the presence of Et3N-nHF
(Figure 4 and 5). In the case of Et4NCl, the redox responses were
observed at a higher potential than those of Et4NBr according to
the sequence of oxidation potential of halide. Even when Et3N-5HF
was used as supporting salt, redox waves were not reversible; how-
ever, quasi-reversible redox waves were observed in neat Et3N-5HF
(Figure 4). The voltammogram of Et4NI in the presence of Et3N-nHF
showed two oxidation peaks, and the first redox couple appeared at a
potential much less positive than that of Et4NBr (Figure 5).
For the conditions showing reversible redox systems, the voltam-
metry of halides was measured by changing scan rates, and their
diffusion coefficients could be estimated from the gradient of the lin-
ear plot of anodic peak current vs. square root of scan rate based on
the Randles-Sevcik formula for a reversible process (eq. 5).26
Anodic fluorodesulfurization of dithioacetals using Et4NX as
mediator.— In order to survey the mediatory ability of Et4NX for
fluorodesulfurization of dithioacetal 1a in HF salt/CH2Cl2, the CV
measurements were carried out under various conditions. Figure 6a
shows the cyclic voltammograms of Et4NBr in Et3N-3HF/MeCN in
the absence and presence of substrate 1a. The observed irreversible
curve with a high current flow (catalytic current) suggested that the
redox of bromide catalyzed the oxidative fluorodesulfurization of 1a.
The more pronounced catalytic current was also observed in the cyclic
voltammograms of Et4NBr in Et3N-5HF/MeCN (Figure 6b).
Then the macro-scale constant-current electrolysis of 1a with a
catalytic amount of Et4NBr (10 mol%) was investigated in Et3N-
nHF/CH2Cl2 (Table II). When Et3N-3HF was used as fluorine
source, mono-fluorinated product 2a was yielded selectively with
small amount of electricity (1–1.5 F/mol) (Entries 1 and 2). The
1
2
ꢀ
ꢁ
nF
RT
1
2
1
IPa = 0.44nF
× AC0 D v 2
[5]
where Ipa is oxidation peak current [A], n is the number of reac-
tion electrons, A is the area of electrode [cm2], C0 is concentration
of substance [mol/cm3], D is diffusion coefficient [cm2/s] and v is
scan rate [mV/s]. The estimated diffusion coefficients of halides are
summarized in Table I. The diffusion coefficients estimated in Et3N-
nHF/MeCN were higher than those in neat Et3N-nHF. In general,
diffusion coefficient of molecules in ionic liquid is ca. 500 ∼ 1000
times lower than that in MeCN due to the highly viscous nature of
ionic liquid.27 However, the diffusion coefficients of such halides in
neat Et3N-nHF were little smaller than those in ordinary molecular
solvents like MeCN. This is because of low viscosity of Et3N-5HF
(1.30 mPa/s).28 The differences of the diffusion coefficient between
the media containing Et3N-4HF and Et3N-5HF were owing to the dif-
Table I. Diffusion coefficient (× 10−6 cm2/s) of Et4NX under
various conditions.
0.6 M Et3N-nHF/MeCN
Et3N-nHF
Mediator
n = 3
n = 4
–
n = 5
–
n = 3
n = 4
–
n = 5
–
a
a
a
a
a
a
Et4NCl
Et4NBr
Et4NI
–
–
–
–
a
a
8.03
9.30
1.83
4.24
a
a
a
a
–
–
–
–
1.97b
1.78b
aIrreversible process.
bCalculated from first peak current.

Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
G3049
Figure 6. Cyclic voltammograms of Et4NBr (3 mM) at 25◦C at 100 mV/s in 0.6 M Et3N-nHF/MeCN, (a) n = 3, (b) n = 5; circle (◦): in the absence of 1a, square
(ꢂ): in the presence of 1a (15 mM).
mechanism was discussed previously.9 A passage of 4 F/mol of elec-
tricity increased the yield of di-fluorinated product 3a (Entry 3).
Chloride and iodide also worked as mediator for the reaction and
the yield was higher than the case without mediator (Entries 4–6).
This product selectivity can be explained as follows. The oxidative
fluorodesulfurization of 1a proceeds via the electrogenerated cationic
bromine species to give mono-fluorinated product 2a, which has
slightly higher oxidation potential than substrate 1a owing to an
electron-withdrawing fluorine atom. Even when the large amount of
charge was passed, the oxidation of free Et3N contained in the elec-
trolyte may compete with the further oxidation of 2a, resulted in the
low yield of difluoro-product 3a.
In sharp contrast, the attempt using Et3N-5HF under similar
conditions afforded di-fluorinated product 3a in 45% yield without
formation of 2a (Entry 7). However, considerable amount of 4,4’-
dichlorobenzophenone (4a) was detected because the deprotection of
dithioacetal proceeded in the presence of a large amount of HF. To
avoid this acid-promoted deprotection and to complete the electrolysis
in a shorter time, the concentration of HF salt was decreased and the
current density was increased. Under the optimized conditions, the
yield of 3a reached 91% without side reaction (Entry 10). Although
the iodine mediator successfully provided 3a in high yield, the chlo-
rine mediator did not work well. The appropriate choice of halogen
mediator and HF salt afforded mono- and di-fluorinated products in a
selective manner.
Next we investigated the electrocatalytic fluorination of cyclic
dithioacetal 5b using halogen mediator by cyclic voltammetry
(Figure 7). The significant catalytic currents were observed in each
voltammogram.
Then we conducted the macro-scale fluorodesulfurization of 5 us-
ing Et4NBr mediator (Table III). The content of HF significantly influ-
enced on the yield of di-fluorinated product 3b in Et3N-nHF/CH2Cl2
(Entries 1–3). Although the use of other common solvents such as
MeCN and DME did not provide better results (Entries 4 and 5),
solvent-free condition (in neat Et3N-5HF) gave excellent yield with-
out side reaction (Entry 6). The high acidity of Et3N-5HF seemed to
promote effectively the protonation of contaminating water,30 resulted
in decreasing the nucleophilic reaction of water with the cationic inter-
mediate of 3b. This procedure was also successful for other substrates
except for 5e having the electron-donating methoxy groups (Entries
Table II. Selective mono- and di-fluorination of dithioacetal 1a using halogen mediators.
Yield (%)a
3a
Entry
X
HF salt
Concentration of HF salt (M)
Current density (mA/cm2)
Electricity (F/mol)
2a
1
2
3
4
5
6
7
8
Br
Br
Br
–
Cl
I
Et3N-3HF
Et3N-3HF
Et3N-3HF
Et3N-3HF
Et3N-3HF
Et3N-3HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
1.1
1.1
1.1
1.1
1.1
1.1
1.1
0.1
0.05
0.05
0.05
0.05
0.05
5
5
5
5
5
5
5
5
5
10
10
10
10
1
1.5
4
1.5
1.5
1
2
4
4
4
84
85
56
54
62
97
0
0
0
0
0
3
trace
26
7
1
0
b
Br
Br
Br
Br
45c
74
63
91
58
44
72
9
10
11
12
13
b
–
4
4
4
Cl
I
0
0
aDetermined by 19F-NMR.
bWithout mediator.
c4,4’-Dichlorobenzophenone (4a) was also obtained.

G3050
Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
Figure 7. Cyclic voltammograms of Et4NBr (3 mM) in 0.6 M Et3N-nHF/MeCN at 25◦C at 100 mV/s, (a) n = 3, (b) n = 4, (c) n = 5, circle (◦): in the absence of
5b, square (ꢂ): in the presence of 5b (15 mM).
7–10). Such electron-donating groups are known to stabilize the ben-
zylic cation once generated by oxidation, thus prevent the nucleophilic
attack of fluoride ion.10
In the mechanism for the oxidative fluorodesulfurization of
5, carbon-sulfur bond cleavage of 1,3-dithiolane moiety plays an
important role to generate benzylic cation intermediate. The elim-
inated moiety would form cyclic or linear disulfides. Considering
this mechanism, replacement of 1,3-dithiolane moiety to 1,3-dithiane
moiety seems to be effective to reduce oxidation potential of sub-
strates and promote elimination of cyclic disulfide, 1,2-dithiolane
Table III. Electrocatalytic fluorodesulfurization of cyclic dithioacetal 5 using Et4NBr as mediator.
Yield (%)a
Entry
HF salt
Solvent
Substrate
3
4
Recovery of 5
1
2
3
4
5
6
7
8
9
Et3N-3HF
Et3N-4HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
Et3N-5HF
CH2Cl2
CH2Cl2
CH2Cl2
MeCN
DME
5b (Y = F)
5b (Y = F)
5b (Y = F)
5b (Y = F)
5b (Y = F)
5b (Y = F)
5a (Y = Cl)
5c (Y = H)
5d (Y = Me)
5e (Y = OMe)
15
44
64
67
24
95
86
84
66
4
43
31
20
12
47
10
4
15
19
7
b
–
–
–
–
trace
trace
b
c
c
–
–
–
–
–
–
–
–
b
c
c
b
c
c
b
c
c
10
–
aDetermined by 19F-NMR.
bWithout solvent.
cNot determined.

Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
G3051
Scheme 2.
as shown in Scheme 2.31 The CV measurement of cyclic dithioac-
etal 6b was performed and the oxidation peak potential was ob-
served at around 1.1 V vs. SCE. The peak potential was remarkably
lower than that of 5b (2.3 V vs. SCE) due to the reasons mentioned
above.
Table V. Electrocatalytic fluorination of xanthate esters,
carbonate, and sugar derivatives using Et4NBr (10 mol%) as
mediator in Et3N-5HF.a
Substrate
Product
Yield (%)b
Then the electrocatalytic fluorodesulfurization of cyclic dithioac-
etal 6b was carried out in neat Et3N-5HF as shown in Table IV. The
Et4NX mediators used were all effective for the electrocatalytic reac-
tion of 6b to give di-fluorinated product 3b in good to excellent yield
(Entries 2–4), while the yield of 3b was moderate in the absence of the
mediators (Entry 1). Et3N-5HF can dissolve not only organic halide,
but also inorganic halide such as potassium halide. The inexpensive
KBr and KI were employed as halogen mediator in Et3N-5HF, and
worked very efficiently (Entries 5 and 6).
68c
81c
Recycle use of halogen mediators in neat HF salt.— The reusabil-
ity of the halogen mediator (Et4NBr, Et4NI) and Et3N-5HF was inves-
tigated after simple extraction of product 3b with hexane under the
similar conditions shown in Table IV (Entries 3 and 4). The recovered
Et3N-5HF still contained the halogen mediator and was reused for
subsequent runs, maintaining a good yield of 3b through the fifth run
(Et4NBr: 99–95%, Et4NI: 97–91%).
78 (58)d
86 (66)e
66d
Electrocatalytic fluorination of xanthate ester, carbonate and
sugar derivatives.— Finally, the halogen mediator/HF salt system
was extended to other substrate such as xanthate ester, carbonate, and
sugar derivatives. As shown in Table V, various xanthate esters pro-
vided the corresponding benzyl fluoride in moderate to good yields.16
In particular, substrates having weak electron-withdrawing group at
the para-position of the benzene ring showed good results. 7f having
electron-donating methyl group resulted in low yield due to the stabi-
lization effect of benzylic cation intermediate; in addition, the methyl
substituent was also fluorinated to give 1,4-bis(fluoromethyl)benzene.
The electrocatalytic fluorodesulfurization of carbonate derivative 9
and sugar derivative 11 proceeded well to afford the corresponding
fluorinated functional molecules 10 and 12, which are potential candi-
dates for organic electrolytic solvents or additives for rechargeable Li
batteries32 and useful biological probes and versatile chemical build-
ing blocks,22 respectively.
33e,f
78c
86c (α/β = 28/72)
aConditions: constant-current (5 mA/cm2), undivided cell.
bDetermined by 19F-NMR. Isolated yields were shown in parentheses.
c4 F/mol.
d3 F/mol.
e5 F/mol.
f1,4-Bis(fluoromethyl)benzene was detected as a by-product.
Table IV. Electrocatalytic fluorodesulfurization of cyclic
dithioacetal 6b using Et4NX and KX as mediator.
Conclusions
In conclusion, we have clarified redox behaviors of halides such
as Et4NX in HF salts by CV analysis and found the possibility to
use them for electrocatalytic fluorination reaction. The macro-scale
electrolysis of dithioacetals was successfully carried out to produce
the desired fluorinated products with the appropriate choice of halo-
gen mediator and HF salt. Moreover, HF salt could dissolve more
inexpensive halide like KBr and KI, and they also worked as halogen
mediator for electrocatalytic fluorination of dithioacetals. The halo-
gen mediator/HF salt system was found to be reusable for repeating
electrochemical fluorination after simple extraction of the product
with hexane. This halogen mediator system could be applied to the
electrochemical fluorination of xanthate ester, carbonate, and sugar
derivatives very efficiently.
Entry
Mediator
Yield (%)a
b
1
2
3
4
5
6
–
65
81
99 (91)
97
99
96
Et4NCl
Et4NBr
Et4NI
KBr
KI
aDetermined by 19F-NMR. Isolated yields were shown in parentheses.
bWithout mediator.

G3052
Journal of The Electrochemical Society, 160 (7) G3046-G3052 (2013)
References
15. K. Kanie, Y. Tanaka, K. Suzuki, M. Kuroboshi, and T. Hiyama, Bull. Chem. Soc.
Jpn., 73, 471 (2000).
1. T. Fuchigami, in: H. Lund and O. Hammerich (Eds.), Organic Electrochemistry, 4th
ed., Dekker, NY, (2001) Ch. 25.
2. T. Fuchigami and T. Tajima, in: V. Soloshonok, K. Mikami, T. Yamazaki, J. T. Welch,
and J. F. Honek (Eds.), Current Fluoroorganic Chemistry. New Synthetic Directions,
Technologies, Materials, and Biological Applications, ACS Symposium Series 949,
American Chemical Society, WA, (2007) Ch. 5.
3. T. Fuchigami and S. Inagi, Chem. Commun., 47, 10211 (2011).
4. W. V. Childs, L. Christensen, F. W. Klink, and C. F. Kolpin, in: H. Lund and
M. M. Baizer (Eds.), Organic Electrochemistry, 3rd ed., Dekker, NY, Ch. 26 (1991).
5. T. Sawamura, S. Kuribayashi, S. Inagi, and T. Fuchigami, Org. Lett., 12, 644
(2010).
6. K. Takahashi, T. Furusawa, T. Sawamura, S. Kuribayashi, S. Inagi, and T. Fuchigami,
Electrochim. Acta, 77, 47 (2012).
16. D. H. R. Barton and S. W. McCombie, J. Chem. Soc., Perkin Trans. I, 1574 (1975).
17. Y. Chang, A. Tewari, A. I. Adi, and C. Bae, Tetrahedron, 64, 9837 (2008).
18. K. Momota, K. Mukai, K. Kato, and M. Morita, Electrochim. Acta, 43, 2503 (1998).
19. E. Laurent, B. Marquet, and R. Tardivel, Tetrahedron, 47, 3969 (1991).
20. C. W. L. Bevan, J. Chem. Soc., 1347 (1960).
21. H. Ishii, N. Yamada, and T. Fuchigami, Chem. Commun., 1617 (2000).
22. S. Caddick, L. Gazzard, W. B. Motherwell, and J. A. Wilkinson, Tetrahedron, 52,
149 (1996).
23. T. Takiguchi and T. Nonaka, Bull. Chem. Soc. Jpn., 60, 3137 (1987).
24. W. K. Behl, J. Electrochem. Soc., 136, 2305 (1989).
25. S. Q. Chen, T. Hatakeyama, T. Fukuhara, S. Hara, and N. Yoneda, Electrochim. Acta,
42, 1951 (1997).
26. A. J. Bard and L. R. Faulkner, Electrochemical Methods. Fundamentals and Appli-
cations, 2nd ed., John Wiley & Sons, NY (2001).
27. B. K. Sweeny and D. G. Peters, Electrochem. Commun., 3, 712 (2001).
28. T. Isogai, T. Nakai, K. Nakanishi, M. Inaba, and A. Tasaka, Electrochemistry, 77,
713 (2009).
29. T. Isogai, T. Nakai, H. Inoue, K. Nakanishi, S. Kohtara, M. Saito, M. Inaba, and
A. Tasaka, J. Phys. Chem. B, 115, 9593 (2011).
30. Y. Kishi, H. Nagura, S. Inagi, and T. Fuchigami, Chem. Commun., 3876 (2008).
31. M. Platen and E. Steckhan, Tetrahadron Lett., 21, 511 (1980).
32. M. Hasegawa and T. Fuchigami, Electrochim. Acta, 49, 3367 (2004).
7. T. Sawamura, S. Kuribayashi, S. Inagi, and T. Fuchigami, Adv. Synth. Catal., 352,
2752 (2010).
8. B. Yin, S. Inagi, and T. Fuchigami, Synthesis, 2372 (2011).
9. T. Fuchigami and M. Sano, J. Electroanal. Chem., 414, 81 (1996).
10. T. Yoshiyama and T. Fuchigami, Chem. Lett., 1995 (1992).
11. S. C. Sondej and J. A. T. Katzenellenbogen, J. Org. Chem., 51, 3508 (1986).
12. W. B. Motherwell and J. A. Wilkinson, Synlett, 191 (1991).
13. H. Ishii, N. Yamada, and T. Fuchigami, Tetrahedron, 57, 9067 (2001).
14. A. J. Janczuk, Carbohydr. Res., 337, 1247 (2002).