Selective electrochemical synthesis and reactivity of functional benzylic fluorosilylsynthons

Article abstract of DOI:10.1016/S0022-1139(00)00373-0

Electrochemical reductive silylation of meta-(trifluoromethyl)arenes by the sacrificial anode technique selectively led to meta-trimethylsilyldifluoromethylarenes (ArCF₂TMS), in the presence of an excess of TMSCl and in a THF/cosolvent mixture

Full text of DOI:10.1016/S0022-1139(00)00373-0

Journal of Fluorine Chemistry 107 (2001) 301±310
Selective electrochemical synthesis and reactivity of
functional benzylic ¯uorosilylsynthons
Philippe Clavela, Guillaume Lessenea, Claude Birana, Michel Bordeaua,*,
b
Nicolas Roques , Stephane Trevin , Dominique de Montauzon
c
d
Â
L.C.O.O., Universite Bordeaux 1, 351 cours de la liberation, 33405 Talence Cedex, France
Â
a
Â
Â
b
Á
Rhodia Recherches 85, avenue des freres Perret BP 62, 69192 Saint-Fons, France
 Á
Electricite de France-DER-Les Renardieres, Route de Sens, BP No. 1, 77818 Moret sur Loing, France
c
d
Â
L.C.C., Universite Paul Sabatier, UPR 8241, Route de Narbone, 31077 Toulouse, France
Received 25 April 2000; accepted 11 May 2000
Abstract
Electrochemical reductive silylation of meta-(tri¯uoromethyl)arenes by the sacri®cial anode technique selectively led to meta-
trimethylsilyldi¯uoromethylarenes (ArCF2TMS), in the presence of an excess of TMSCl and in a THF/cosolvent mixture
(cosolvent ˆ DMPU or HMPA). In the case of meta-(trimethylsilyldi¯uoromethyl)tri¯uoromethylbenzene, the in¯uence of the cosolvent
on the silylation selectivity was studied. A cyclic voltammetry study allowed an explanation of the difference in the results obtained
between the tri¯uoromethylbenzene and meta-bistri¯uoromethylbenzene series. ArCF2TMS (Ar ˆ Ph, m-CF3C6H4) species were found
ef®cient for ArCF2-group transfer to diverse electrophiles under Fuchigami's conditions (KF catalysis in DMF). # 2001 Elsevier Science
B.V. All rights reserved.
Keywords: Fluoro-benzylsilanes; Electrosynthesis; Sacri®cial aluminium anode; Cyclic voltammetry; Solvent effect; Di¯uorobenzylation
1. Introduction
is described here as well as the reactivity of 2 towards
benzaldehyde.
The interest of introducing a ¯uorinated group into a
molecular structure has clearly been demonstrated, particu-
larly concerning the modi®cation of its biological properties
[1,2]. Consequently, synthons allowing selective access to
¯uorinated compounds are actively researched as shown, for
example, by the numerous papers relative to the use of
Ruppert's reagent [3] in anionic tri¯uoromethylation reac-
tions. As far as we were concerned, we focused our interest
on RCF2-building block synthons; we therefore, reported a
molar scale electrosynthesis of PhCF2TMS (TMS ˆ SiMe3)
[4] and TMSCF2COOEt [5]. We demonstrated the ability of
these synthons to transfer the RCF2-group (R ˆ Ph, COOEt)
to carbonyl compounds. We report here the electrosynthesis
of meta-(di¯uorotrimethylsilylmethyl)tri¯uoromethylben-
zene (2), meta-(di¯uorotrimethylsilylmethyl)phenoxytri-
methylsilane (3), the corresponding deprotected phenol
(30), meta-(di¯uorotrimethylsilylmethyl)-N-trimethylsilyla-
niline (4) and the corresponding deprotected aniline (40). A
wide range of anionic di¯uorobenzylation with PhCF2TMS
2. Results and discussion
2.1. Electrosilylation of m-bis(trifluoromethyl)benzene (1)
In order to prepare a synthon with two different ¯uori-
nated groups (CF3- and CF2-) on the same aromatic ring, we
found the conditions for the selective reduction of commer-
cially available m-bis(tri¯uoromethyl)benzene, m-BTFMB
(1), in the presence of an excess of trimethylchlorosilane,
TMSCl, which led to synthon 2. To our knowledge, the only
present chemical route to 2 implies a photochemical reaction
between 1 and hexamethyldisilane [6]; but this reaction was
de®nitely not selective (the benzenic ring being also sily-
lated) and compound 2 was obtained in low yield (26%)
Scheme 1.
The remarkable chemo- and regioselectivities of electro-
synthesis in organosilicon series1 encouraged us to apply
this technique to the preparation of 2.
* Corresponding author. Tel.: ‡33-556-846281; fax: ‡33-556-846646.
E-mail address: m.bordeau@lcoo.u-bordeaux.fr (M. Bordeau).
1 To be published in Rhodia Publications.
0022-1139/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 0 0 ) 0 0 3 7 3 - 0

302
P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
Scheme 1. Photochemical silylation of m-BTFMB.
Table 1
Trimethylsilylation of m-BTFMB with HMPA as the cosolvent
comparison, those of tri¯uoromethylbenzene (TFMB) and
its monosilylated derivative were measured in several mix-
tures of THF and cosolvents [7]. As we aimed at ®nding a
way to carry out a selective monosilylation, we report the
sole ®rst reduction peak values and the difference between
the reduction potentials of the unsilylated substrates and the
corresponding monosilylated compounds in four aprotic
solvents and solvent mixtures (Tables 2 and 3).
Current passed
1
1 (% GC)
2 (% GC)
2a (% GC) 2b (% GC)
(F mol
)
0.3
0.6
2.0
3.6
90
80
40
0
9
14
18
21
1
6
0
0
0
0
40
77
The values found for TFMB were very close to those
reported in the literature under different conditions [8,9], the
difference E1 E10 is at least 100 mV in all solvents. We had
previously shown [10] that the selective monosilylation of o-
dichlorobenzene required a 200 mV difference. In addition,
the preparative silylation of TFMB was performed with
HMPA, N,N0-dimethylpropyleneurea (DMPU) or dimethyl-
formamide (DMF) as cosolvent and a good selectivity
was observed for the monosilylation versus disilylation
[4]: in this case, 100 mV are suf®cient to reach the required
selectivity.
As we generally proceed for each new reaction, the
electrolysis of 1 was ®rst performed with hexamethylpho-
sphoramide (HMPA) as the cosolvent according to
Scheme 2.
In Table 1 it is reported that the balance of the reaction
1
according to the current passed, 2 F mol being theoreti-
cally necessary for the reduction of a C±F bond.
These results lead to two remarks:
Comparing Tables 2 and 3, we can see that the reduction
of a C±F bond in 1 or 2 is easier (by about 400 mV) than that
observed in the TFMB series. This observation is consistent
with an activation due to the second tri¯uoromethyl group
decreasing the LUMO energy of the compound and then
making the bis(tri¯uoromethylated) molecule more easily
reducible than the mono substituted one. Furthermore, in
THF/HMPA (9:1), as far as compounds 1 and 2 were
concerned, an unexpected small difference between E1
and E10 (70 mV) was observed: this value could explain
the loss of selectivity for the monosilylation of m-BTFMB
when electrosynthesis was performed in the presence of this
cosolvent. The difference E1 E10 being higher with DMPU
(140 mV), an electrolysis was performed with DMPU and
under these conditions, 2 was obtained with a good con-
version (90%) and a reasonable yield (72%) (Table 4).
ꢀ monosilylation was not selective under these conditions:
only 60% of 1 were converted when 2 F mol±1 were
passed whereas disilylation reached 40%;
ꢀ disilylation occurred on the benzylic carbon concerned in
the first reductive silylation and the gem-disilylation
leading exclusively to 2a largely prevailed after the
1
passage of 3.6 F mol (77 versus 21% selectivity for
the monosilylation).
To understand these unexpected results, a cyclic voltam-
metry study was necessary. Indeed, the chemoselectivity of
electrosynthesis reactions, even under the intensiostatic
mode, depends on the value of the relative reduction poten-
tial of each species present in the reaction mixture
(the less cathodic compound being the ®rst reduced). So,
the successive reduction peak potentials of 1, 2 and for
Scheme 2. Electrochemical silylation of m-BTFMB.

P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
303
Table 2
First reduction potential of TFMB and its monosilylated derivative
Solvent
E1 (V/SCE)a Æ0.01
E10 (V/SCE)a Æ0.01
E1 E10 (mV)
Monosilylated/disilylated
1
after 2 F mol
THF
2.68
2.53
2.60
2.50
2.68
2.68
2.70
2.63
0
150
100
130
Low conduction
THF/HMPA (9/1)
THF/DMPU (9/1)
DMF
92/8
90/5
95/5
a Measured at a 125 mm diameter gold working microelectrode, in a 0.1 M NBu4PF6 solution.
Table 3
First reduction potential of 1 and 2
Solvent
E1 (V/SCE)a Æ0.01
E10 (V/SCE)a Æ0.01
E1 E10 (mV)
THF
2.16
2.17
2.10
2.07
2.28
2.24
2.24
2.22
120
70
THF/HMPA (9/1)
THF/DMPU (9/1)
DMF
140
150
a Measured at a 125 mm diameter gold working microelectrode, in a 0.1 M NBu4PF6 solution.
Taking into account the gap of 150 mV found between E1
and E10 in DMF, it could have been thought that DMF was
suitable to perform the reductive silylation. However, in the
case of TFMB, we demonstrated [4,7] that electrolysis in
DMF led to the monosilylated compound with excellent
selectivity (see Table 2) but with a low conversion because
of the reduction of the amide itself, favoured by the electro-
philic assistance of TMSCl. For this reason, DMF was not
used for the synthesis of 2.
The electrochemical analytical study points out that the
nature of the cosolvent is of great importance to reach the
desired selectivity and that DMPU was found to be the best
cosolvent for the monosilylation of m-BTFMB. On the
contrary, HMPA appears to be the most suitable cosolvent
to perform the disilylation; it must be noted, that only the
®rst reduction potential value is signi®cantly modi®ed when
changing the cosolvent. However, the interpretation of these
results is not easy; indeed, until now, the reduction mechan-
ism of benzylic ¯uorocompounds has not been speci®cally
studied. According to Saveant's paper [11] concerning the
electrochemical reductive cleavage of arylmethyl chlorides
and bromides, two hypothesis are to be considered: (i) with a
stepwise mechanism, an intermediate radical-anion would
be formed, followed by carbon±¯uorine bond breaking,
favoured by strong withdrawing groups such as the nitro
group; (ii) with a concerted electron transfer-bond breaking
mechanism, the radical would be formed directly as in the
case of unsubstituted benzyl chloride and bromide [11].
Thus, the reduction of TFMB would probably take place via
the latter mechanism. As far as compound 1 is concerned,
we think that the reduction occurs following a stepwise
mechanism because the presence of the second CF3 group,
Table 4
Monosilylation of m-BTFMB with DMPU as the cosolvent
1
)
Current passed (F mol
(% GC)
(% GC)
(% GC)
(% GC)
1.4
2.4
40
10
53
72
3
0
0
13

304
P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
Scheme 3. Regioselectivity of the disilylation.
lowering the energy of the pà orbital of the aromatic moiety,
stabilises the radical-anion. In this case, as shown by Saveant
[11], the main factor governing the thermodynamics and the
kinetics of the reductive cleavage is then the dissociation
energy of the bond being broken. Two other groups [12,13]
have shown that the solvent nature could modify the clea-
vage reaction rate; for our part, researches are in progress to
verify our hypothesis concerning the reduction mechanism
of 1. The solvent effect could then be interpreted taking into
account the additional stabilisation brought by the Mg2‡
cation associated with the radical-anion, ion pairs being
tighter in DMPU than in HMPA.
From 2, the stabilisation of radicals [14±16] and anions
[17,18] by the a-silyl group (Scheme 3) could be a more
important factor than the solvent effect for the cleavage of
the C±F bond of the radical-anion; thus, differences between
reduction potentials measured in HMPA or DMPU would be
minimised. When disilylation occurs, only the gem-disily-
lated product is obtained. The stabilisation of the intermedi-
ates (a-silylradical 2a0 and a-silylanion 2a00) by the
trimethylsilyl group, compared to the destabilisation of
2b00 by the ‡I p effect [19] of the two ¯uorine atoms could
explain the exclusive breaking of the C±F bond of the
di¯uorotrimethylsilylmethyl group (Scheme 3).
[7]; we report here new results (Scheme 4 and Table 5)
concerning the transfer of di¯uoromethylarene groups
from PhCF2TMS and from 2 under Fuchigami's conditions
[19].
If wanted, all the alcohols can be isolated in their
protected form; most of them have never been described
previously. With the less reactive electrophiles (pheny-
lethylketone and cyclohexanone) the yields remain low
but could be increased using an excess of electrophile
[20]. The reactivity of 2 was only tested towards benzalde-
hyde; compared to PhCF2SiMe3, the presence of an addi-
tional CF3 group in 2 enhances the reactivity of this
compound, the reaction rate being, under the same condi-
tions, four times greater with an excellent isolated yield
(80%).
2.3. Electrosilylation of m-(trifluoromethyl)phenol and m-
(trifluoromethyl)aniline
The reaction was performed in HMPA according to
Scheme 5.
As both compounds include an acidic hydrogen, the ®rst
electrochemical reaction which occurs is the reduction (with
H2 evolution) of HCl previously formed in the O- or N-
silylation step. The obtained results are summarised in
1
2.2. Reactivity of benzylic silyldifluorosynthons
Table 6, after having passed 3.2 F mol
.
It must be noted that the hydrolysis of the reaction mixture
by a saturated NaHCO3 solution allowed stabilisation of the
O- or N-trimethylsilyl derivatives which are much more
stable than the corresponding phenol or aniline.
On the other hand, replacing HMPA by DMPU led to
non-reproducible results with the amine and to moderate
We previously reported the reactivity of PhCF2TMS
towards carbonyl compounds using a F catalysis, with
tetrabutylammonium ¯uoride (TBAF) [4]. However, the
yields were medium to low, due to the faster hydrolysis
of the silylsynthon by the water contained in TBAF solution
Scheme 4. Reactivity of benzylic silyldifluorosynthons.

P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
305
Table 5
Transfer of ArCF2 moieties to aldehydes and ketones
Electrophile
ArCF2TMS
Alcohol
Yield %
73a
80a
60a
60b
33b
24a
a Isolated yield.
b Determined by 19F NMR.
yields with m-tri¯uoromethyl phenol after having passed
1
3.8 F mol conversion was only 48% but with a good
potential due to the presence of a strong donating group
on the aromatic ring. In such a case, HMPA is a better
cosolvent than DMPU because of its wider cathodic poten-
tial window.
chemoselectivity (94%); the low faradaic yield can be
explained by the very cathodic value of the reduction
Scheme 5. Electrosilylation of m-(trifluoromethyl)phenol and aniline.

306
P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
Table 6
1
Electrosilylation of m-trifluoromethylphenol and aniline after 3.2 F mol
Y
(% GC)
(% GC)
(% GC)
(% GC)
O
0
0
5
2
89
93
5
4
NH
3. Conclusion
Solvents, PhCF3, m-F3CC6H4CF3, KF were purchased from
Aldrich and SiO2 (9385) from Merck. m-F3CC6H4NH2 and
m-F3CC6H4OH were provided by Rhodia Chimie.
These results constitute a new example of the potentiality
of electrosynthesis for regioselective reactions from poly-
functional compounds; moreover, we found an unexpected
solvent effect on the selectivity of the silylation which could
be correlated to the modi®cation of the reduction potential
values according to the solvent mixture. We also report
optimised conditions for transferring di¯uoro building
blocks to electrophiles.
4.2. Electrosynthesis of trimethylsilyldifluoromethylarenes,
ArCF2TMS, in a 70 ml laboratory cell
The electrolysis of magnetically stirred solutions was
performed under nitrogen, in a previously described undi-
vided cell ®tted with an aluminium rod as the anode and a
concentric cylindrical stainless steel grid as the cathode.
These two electrodes were previously chemically scored by
a 10% HCl solution, then rinsed out several times with
distilled water and with acetone. The oven dried cell con-
taining the supporting electrolyte, NBu4Br (0.25 g,
0.8 mmol) was deaerated twice under vacuum and then with
dry nitrogen gas. THF (55 ml), DMPU (1.8 ml, 14.9 mmol)
or HMPA (3.2 ml, 14.9 mmol) and TMSCl (9 ml,
68.5 mmol) were introduced under a low stream of nitrogen.
HCl resulting from the reaction between TMSCl and the
residual water was removed by pre-electrolysing the solu-
tion (i ˆ 0:1 A; j ˆ 0:4 A dm 2). The other hydrolysis pro-
duct, Me6Si2O, remains electrochemically inert. When
evolution of H2 ceased, the substrate ArCF3 (13.7 mmol)
was introduced through a septum by syringe. The electro-
lysis was then performed (i ˆ 0:1 A; j ˆ 0:4 A dm 2) over
9 h, until the required charge (2.4 F mol 1) has been passed.
The progress of the reaction was monitored by gas chro-
matography. At the end of the electrolysis, the mixture was
poured into 250 ml of cold water. The organic layer was
extracted three times with Et2O (100 ml) and washed twice
with cold water (100 ml). After drying over MgSO4, Et2O
was evaporated off. Fractional distillation over a Vigreux
column was then performed to recover the pure product.
4. Experimental
4.1. Materials
For electrolysis in a 70 ml cell, THF (SDS) was distilled
over sodium-benzophenone ketyl. The cosolvents HMPA
(Lancaster), DMPU (Fluka), TDA-1 (Aldrich) were used
without any treatment. The supporting electrolytes were
pumped off during 48 h at room temperature. Trimethyl-
chlorosilane was distilled over Mg powder just before use.
Gas chromatography was performed with a temperature-
programmable Hewlett-Packard 5890A apparatus equipped
1
with a 25 m  0:25 mm CP-Sil 5CB capillary column. H
NMR spectra were recorded in CDCl3 at 250 MHz with a
Brucker AC 250 spectrometer, using residual CHCl3
(d ˆ 7:27 ppm) as the internal standard. 13C NMR spectra
were obtained at 62.86 MHz with a Brucker AC 250 using
CDCl3 (d ˆ 77:70 ppm) as the internal standard. The signals
for 1H and 13C NMR are designated s (singlet), d (doublet), t
(triplet), q (quartet) and m (multiplet). 29Si NMR spectra
were recorded in CDCl3 at 39.73 MHz with a Brucker AC
200 spectrometer and were referenced to TMS. 19F NMR
spectra were recorded in CDCl3 at 282 MHz with a Brucker
AC 200 spectrometer and were referenced to CFCl3 using
PhOCF3 (d ˆ 58:3 ppm) as the internal standard. Electron
impact mass spectra were recorded at an ionisation voltage
of 70 eV with a VG Micromass 16F spectrometer coupled
with a gas chromatograph equipped with a 25 m  0:25 mm
CP-Sil capillary column. IR spectra were recorded with a
Perkin±Elmer 1420 spectrophotometer in pure liquid ®lms
(NaCl discs). Elementary microanalysis were performed by
the ``Service Central de Microanalyses'' of CNRS (France).
4.3. Trimethylsilyldifluoromethylbenzene
CAS number: 149021-01-2; M ˆ 200:31 g mol 1; col-
ourless liquid; Eb20 ˆ 808C.

P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
307
1H NMR: dH 0.21 ppm (s, 9H, TMS), dH 7.26±7.45 ppm
1253, 1165, 1128, 1075; MS: m/z (relative intensity): 322
‡
(m, 5H, Ph); 13C NMR: dC 4.9 ppm (s, TMS), dC3 C7
,
(M) (<1), 303 (M 15, 12), 230 (77), 215 (13), 203 (48), 189
3
124.7 ppm (t, JC ˆ 8:0 Hz), d 128.3 ppm (s), dC4,C6
(13), 188 (14), 153 (36), 133 (71), 115 (10), 101 (8), 77
(100), 73 (81), 59 (19), 45 (26); anal. calcd (found) for
C14H22F4Si2: C% ˆ 52:14 (51.48), H% ˆ 6:88 (6.94),
F% ˆ 24:19 (23.87).
F
C5
128.8 ppm (t, 4JC ˆ 2:6 Hz), dC1 134.5 ppm (t,
F
1JC ˆ 265:0 Hz), d 138.3 ppm (t, 2JC ˆ 20:4 Hz);
F
F
C2
19F NMR: dF 112.5 ppm (s); 29Si NMR: dSi 4.3 ppm (t,
2JSi ˆ 34:7 Hz); IR: 2965, 1954, 1878, 1812, 1754, 1449,
F
‡
1253, 1234, 1080, 989; MS: m/z: 185 (M 15) , 127
(M 73) , 108 (M 92) , 93 (M 107) , 77 (SiFMe2 ),
4.6. m-Trimethylsilyldifluoromethylphenol
‡
‡
‡
‡
‡
73 (SiMe3 ); anal. calcd (found) for C10H14SiF2:
C% ˆ 59:96 (58.12), H% ˆ 7:04 (7.04), F% ˆ 18:97
(19.00).
4.4. m-(Trimethylsilyldifluoromethyl)trifluoromethyl-
benzene
1
M ˆ 216:31 g mol ; Eb5 ˆ 118ꢁC (phenol ‡ phenoxy
TMS mixture, the isolated phenol decomposed at 458C).
1H NMR: dH 0.14 ppm (s, 9H, TMS), dH 5.80 ppm (s,
OH), dH 6.83±7.30 ppm (m, 4H, Ar); 13C NMR: dC 4.8
ppm (t, 3JC ˆ 1:9 Hz, TMS), dC3 111.7 ppm (t,
F
3JC ˆ 8:1 Hz), d 115.9 ppm (t, JC ˆ 2:4 Hz), d
4
F
F
C4
C7
1
M ˆ 268:31 g mol ; colourless liquid; Eb5 ˆ 60ꢁC.
117.1 ppm (t, 3JC ˆ 7:6 Hz), dC5 129.7 ppm (s), dC1
F
130.2 ppm (t, 1JC ˆ 265:0 Hz), dC2 140.0 ppm (t,
1H NMR: dH 0.16 ppm (s, 9H, TMS), dH 7.53±7.67 ppm
F
2JC ˆ 20:0 Hz), dC6 155.6 ppm (s); 19F NMR: dF
3
(m, 4H, Ar); 13C NMR: dC 5.2 ppm (t, JC ˆ 1:9 Hz,
F
F
TMS), dC7 121.6 ppm (m), dC5 125.6 ppm (m), dC8
112.1 ppm (s); 29Si NMR: dSi 4.4 ppm (t,
125.4 ppm (q, 1JC ˆ 263:9 Hz), d
126.1 ppm (t,
2JSi ˆ 38:2 Hz); IR: 3379, 2960, 1578, 1449, 1332,
F
F
C1
‡
1JC ˆ 264:2 Hz), d
128.1 ppm (t, 3JC ˆ 7:6 Hz),
1253, 1172, 1079; MS: m/z (relative intensity): 216 (M ,
37), 143 (M 73, 10), 124 (92), 103 (20), 96 (56), 77 (42), 73
(100), 45 (23).
F
F
C3
dC4 128.9 ppm (s), dC6 131.0 ppm (q, 2JC ˆ 32:4 Hz),
F
dC2 139.4 ppm (t, 2JC ˆ 21:0 Hz); 19F NMR: dF1
F
112.7 ppm (s, 2F), dF8 62.7 ppm (s, 3F); 29Si NMR:
2
4.7. m-Trimethylsilyldifluoromethylaniline
dSi 5.2 ppm (t, JSi ˆ 37:1 Hz); IR: 2964, 1445, 1335,
F
1255, 1219, 1170, 1132, 1104, 1076; MS: m/z (relative
intensity): 249 (M 19, 7), 195 (M 73, 7), 176 (M 92,
100), 153 (16), 126 (12), 107 (8), 81 (14), 77 (14), 73 (68),
45 (22); anal. calcd (found) for C11H13SiF5: C% ˆ 49:24
(49.18), H% ˆ 4:88 (5.12), F% ˆ 35:41 (33.62).
1
4.5. m-(Bistrimethylsilylfluoromethyl)trifluoromethyl-
benzene
M ˆ 215:327 g mol
.
1H NMR: dH 0.15 ppm (s, 9H, TMS), dH 5.60 ppm (s, 2H,
NH2), dH 6.65±7.27 ppm (m, 4H, Ar); 13C NMR: dC
4.8 ppm (t, 3JC ˆ 1:9 Hz, TMS), dC3 111.1 ppm (t,
F
3JC ˆ 8:6 Hz), d 115.6 ppm (t, JC ˆ 2:9 Hz), d
4
F
F
C4
C7
117.2 ppm (t, 3JC ˆ 7:2 Hz), d 129.7 ppm (s), dC1
F
C5
128.4 ppm (t, 1JC ˆ 265:1 Hz), dC2 139.4 ppm (t,
F
2JC ˆ 20:0 Hz), d
146.4 ppm (s); 19F NMR: dF
F
C6
1
M ˆ 322:51 g mol ; colourless liquid; Eb5 ˆ 102ꢁC.
112.2 ppm (s); 29Si NMR: dSi 4.4 ppm (t,
1H NMR: dH 0.09 ppm (s, 18H, TMS), dH 7.34±7.49 ppm
2JSi ˆ 38:5 Hz); IR: 3368, 2961, 1624, 1594, 1495,
F
3
(m, 4H, Ar); 13C NMR: dC 2.6 ppm (d, JC ˆ 3:8 Hz,
1468, 1457, 1341, 1253, 1124, 1077; MS: m/z (relative
intensity): 215 (M , 53), 142 (M SiMe3, 12), 123 (79),
122 (88), 96 (23), 77 (29), 73 (100), 65 (12), 45 (21).
F
‡
TMS), dC1 99.7 ppm (d, 1JC ˆ 158:3 Hz), dC7 117.8 ppm
F
(m), dC5 120.8 ppm (q, 3JC ˆ 3:8 Hz), dC8 124.4 ppm (q,
F
1JC ˆ 271:8 Hz), d 124.8 ppm (d, 3JC ˆ 3:6 Hz), d
F
F
C3
C4
128.3 ppm (s), dC6 130.2 ppm (q, 2JC ˆ 31:0 Hz), d
4.8. Cyclic voltammetry procedure
F
C2
145.8 ppm (d, 2JC ˆ 13:4 Hz); 19F NMR: dF1
F
216.9 ppm (s, 1F), dF8 63.0 ppm (s, 3F); 29Si NMR:
Measurements were carried out with an Electrokemat
potentiostat using the interrupt method to minimise the
2
dSi 4.74 ppm (d, JSi ˆ 23:1 Hz); IR: 2960, 1427, 1329,
F

308
P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
uncompensated resistance (IR) drop. Electrochemical
experiments were performed at room temperature in an
airtight three-electrode cell connected to vacuum argon/
N2 line. The reference electrode was a saturated calomel
electrode (SCE) separated from the non-aqueous solution by
a bridge compartment. The counter electrode was a Pt wire
spiral of ca. 1 cm2 apparent surface area, 8 cm long and
0.5 mm diameter. The working electrode was an Au disk of
125 mm diameter. THF (Riedel de Haehn, extra pure) was
puri®ed by the previously described method and was stored
4.10. 1,2-Diphenyl-2,2-difluoroethanol
1
CAS number: 1494-20-8; C14H12OF2; M ˆ 234 g mol
white solid; F ˆ 90ꢁC.
;
1H NMR: dH 2.83 ppm (d, 1H, JH
ˆ 10:0 Hz, OH),
H
OH
dH8 5.07 ppm (td, 1H, 3JH ˆ 10:0 Hz), d 7.25±7.48 ppm
Ê
over 4 A molecular sieves before use. The supporting elec-
F
(m, 10H, Ph); 13C NMR: dC8 76.9 ppm (t, 2JC ˆ 30:9 Hz),
trolyte was Bu4NPF6 (Fluka electrochemical grade), used as
received. All solutions measured were 0.6±1.10 3 M in the
organic compound and 0.1 M in supporting electrolyte.
Under the same conditions, ferrocene was oxidised at
Eo ˆ 0:51 V.
F
1
dC1 121.1 ppm (t, JC ˆ 248:3 Hz), d
126.3 ppm (t,
F
C3,C7
3JC ˆ 6:5 Hz), d 127.79 ppm (s), dC4,C6 127.87 ppm
F
C5
(s), dC11,C13 127.92 ppm (s), dC12 128.6 ppm (s), dC10,C14
4
2
130.0 ppm (t, JC ˆ 1:5 Hz), d 134.3 ppm (t, JC
ˆ
F
F
C2
25:9 Hz), dC9 135.8 ppm (t, 3JC ˆ 2:3 Hz); 19F NMR: dF
F
106.7 ppm (d, 3JH ˆ 9:5 Hz); IR: 3431, 2931, 2852,
F
4.9. General procedure for anionic aryldifluoro-
methylation
1945, 1891, 1823, 1764, 1450, 1254, 1160, 1062; MS: m/z
(relative intensity): 234 (M , 5), 214 (M HF, <1), 127
‡
(PhCF2, 26), 107 (Ph-CHOH, M 127, 100), 105 (PhCO,
8), 90 (PhCH, 4), 79 (CF2CHO, 54), 77 (Ph, 39), 51 (CF2H,
16); high resolution MS: M ˆ 234:085622 (234.085737,
0.5 ppm).
KF (30 mg, 0.05 equivalent), DMF (5 ml) and the elec-
trophile (10 mmol) were charged, in that order, in a 25 ml
round bottomed ¯ask, under nitrogen. ArCF2TMS
(10 mmol) was then added in one portion at room tempera-
ture. The reaction mixture was stirred over a period of either
18 to 24 h at room temperature or 2 to 6 h at 70 to 1008C for
hindered electrophiles. The reaction mixture was ®nally
poured into 20 ml of aqueous acidic solution (HCl, 1%)
and left under ef®cient agitation for 15 min. The resulting
solution was extracted three times with diethyl ether
(3 Â 10 ml) and the organic phase was washed with brine
(10 ml), with an iced saturated bicarbonate aqueous solution
(10 ml) and last with brine (10 ml). The ethereal phase was
dried with MgSO4 and concentrated under vacuum. The
crude product was ®nally obtained as a mixture of two major
products, the O-silylated alcohol and the deprotected alco-
hol, in the minimum ratio of 9/1 in favour of the protected
alcohol.
For the deprotection reaction, ethanol (10 ml) and HCl
35% (0.2 ml) were charged in a 25 ml round bottomed
¯ask. The O-silylated alcohol was then slowly added
and a strong stirring was maintained over a period of 1 h.
In the case of tertiary alcohols, we increased the reaction
time up to 10 h. More ef®cient deprotection conditions
like a stoichiometric amount of ¯uoride (TBAF) in THF
or citric acid in methanol could also be used. At the end of
the stirring period, water was added (20 ml). The resulting
medium was extracted three times with diethyl ether
(3 Â 10 ml) and the organic phase was washed with brine
(10 ml), with an iced saturated bicarbonate aqueous solution
(10 ml) and last with brine (10 ml). The ethereal phase was
dried with MgSO4 and concentrated under vacuum. The
product was isolated by column chromatography on silica
gel eluting with light petroleum ether and ethyl acetate (95/5
to 90/10).
4.11. 3-Methyl-1-phenyl-1,1-difluorobutan-2-ol
C11H14OF2; M ˆ 200:23 g mol 1; colourless oil.
1H NMR: dH11 0.96 ppm (d, 3H), dH10 1.01 ppm (d, 3H),
d
H9 1.78±1.90 ppm (hd, 1H), dH 2.21 ppm (s, 1H, OH), dH8
3.79 ppm (td, 1H, 3JH ˆ 11:5 Hz), d 7.43±7.56 ppm (m,
F
H
5H, Ph); 13C NMR: dC11 16.5 ppm (s), dC10 20.7 ppm (s),
dC9 28.4 ppm (t, 3JC ˆ 1:9 Hz), d
78.3 ppm (t,
F
C8
2JC ˆ 28:0 Hz), d 122.2 ppm (t, 1JC ˆ 247:0 Hz),
F
F
C1
dC3,C7 125.7 ppm (t, 3JC ˆ 6:3 Hz), d
128.4 ppm
F
C4,C6
(s), dC5 130.0 ppm (s), dC2 135.2 ppm (t, 2JC
ˆ
F
F
25:3 Hz); 19F NMR: dF 105.3 ppm (d, 2F, 3JH
ˆ
11:5 Hz); IR: 3425, 2966, 2879, 1959, 1887, 1813, 1712,
1452, 1261, 1187, 1116, 1054, 1031; MS: m/z (relative
intensity): 200 (M , 3), 138 (7), 129 (11), 128 (84), 127
‡
(58), 110 (11), 109 (21), 78 (8), 77 (21), 73 (M 127, 100),
72 (13), 55 (48), 43 (30); high resolution MS:
M ˆ 200:101272 (200.101335, 0.3 ppm).
4.12. 3-Methyl-1-phenyl-1,1-difluorobut-3-en-2-ol
C11H12OF2; M ˆ 198:22 g mol 1; colourless oil.

P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
309
‡
1H NMR: dH10 1.72 ppm (s, 3H), dH 2.42 ppm (s, 1H,
OH), dH8 4.48 ppm (t, 1H), dH11trans 4.92 ppm (s, 1H),
dH11cis 5.02 ppm (s, 1H), dH 7.33±7.56 ppm (m, 5H, Ph);
13C NMR: dC10 19.1 ppm (s), dC8 77.8 ppm (t,
intensity): 226 (M , <1), 127 (PhCF2, 23), 99 (C6H10-
OH, M 127, 100), 81 (C6H9, 44); high resolution MS:
M ˆ 226:117561 (226.116922, 2.8 ppm).
2JC ˆ 29:5 Hz), d
116.6 ppm (s), dC1 121.2 ppm (t,
4.15. 1-Phenyl-2-(m-trifluoromethylphenyl)-2,2-
difluoroethanol
F
C11
1JC ˆ 247:2 Hz), d
126.1 ppm (t, JC ˆ 6:2 Hz),
3
F
F
C3,C7
dC4,C6 128.1 ppm (s), dC5 130.2 ppm (s), dC2 134.6 ppm (t,
3
2JC ˆ 22:9 Hz), d 141.6 ppm (t, JC ˆ 1:9 Hz); 19F
F
F
C9
3
NMR: dF 106.2 ppm (d, 2F, JH ˆ 11:5 Hz); IR: 3431,
F
3068, 3038, 2976, 2930, 1959, 1891, 1813, 1716, 1651,
1608, 1452, 1267, 1168, 1057; MS: m/z (relative intensity):
‡
178 (M , 49), 163 (M 15, 6), 127 (90), 109 (6), 77 (21), 71
(M 127, 100).
C15H11OF5; M ˆ 302:26 g mol 1; colourless oil.
1H NMR: dH 2.76 ppm (s, 1H, OH), dH8 5.10 ppm (t, 1H,
3JH ˆ 9:2 Hz), d 7.20±7.73 ppm (m, 9H, Ar); 13C NMR:
4.13. 1,2-Diphenyl-1,1-difluorobutan-2-ol
F
H
dC8 76.6 ppm (t, 2JC-F 31.0 Hz), dC1 120.6 ppm (t, 1JC
ˆ
F
248:9 Hz), dC15 123.7 ppm (q, 1JC ˆ 271:8 Hz), d
F
C3
123.5 ppm (m), dC5 126.7 ppm (m), dC6 127.6 ppm (s),
dC11,C13 128.1 ppm (s), dC12 128.4 ppm (s), dC10,C14
129.0 ppm (s), dC7 129.9 ppm (t, 3JC ˆ 5:7 Hz), d
F
C4
130.4 ppm (q, 2JC ˆ 32:4 Hz), dC2 134.6 ppm (t,
F
1
C16H16OF2; M ˆ 262:31 g mol
.
3
2JC ˆ 26:7 Hz), d 135.4 ppm (t, JC ˆ 2:9 Hz); 19F
F
F
C9
1H NMR: dH10 0.77 ppm (t, 3H), dH 2.10 ppm (m, 1H,
OH), dH9 2.40 ppm (m, 2H), dH 7.27±7.59 ppm (m, 10H,
Ph); 13C NMR: dC9 7.1 ppm (s), dC10 27.3 ppm (t,
NMR: dF15 63.0 ppm (s), dFA1 105.7 ppm (dd, 1F,
2
3JH
(dd, 1F, JH
ˆ 8:9 Hz, JFA
ˆ 249:6 Hz), d
107.4 ppm
ˆ 249:6 Hz); IR: 3382,
FA
FB
FB1
3
2
ˆ 8:9 Hz, JFA
FB
FB
3JC ˆ 1:9 Hz), d 79.9 ppm (t, JC ˆ 28:1 Hz), d
2
F
F
C8
C1
2977, 1958, 1908, 1813, 1738, 1694, 1622, 1495, 1454,
1337, 1247, 1170, 1131, 1079; MS: m/z (relative intensity):
122.6 ppm (t, 1JC ˆ 253:1 Hz), d
127.1 ppm (t,
F
C3,C7
3JC ˆ 6:7 Hz), d 127.2 ppm (s), dC4,C6 127.3 ppm (s),
‡
F
C5
302 (M , <1), 283 (7), 195 (11), 145 (5), 107 (100), 79 (58),
77 (34), 51 (10); high resolution MS: M ˆ 302:073006
dC13,C15 128.0 ppm (s), dC14 128.6 ppm (s), dC12,C16
129.5 ppm (t, 4JC ˆ 1:9 Hz), dC2 134.2 ppm (t,
F
(302.071471, ‡5.1 ppm).
2JC ˆ 26:6 Hz), d
138.0 ppm (t, 3JC ˆ 3:1 Hz);
F
F
C11
19F NMR: dFA 106.1 ppm (d, 1F, JFA
ˆ 245:3 Hz),
2
FB
2
dFB 108.4 ppm (d, 1F, JFA ˆ 245:3 Hz); IR: 3484,
2977, 1963, 1900, 1816, 1449, 1276, 1221, 1053; MS: m/z
(relative intensity): 262 (M , <1), 135 (M 127, 100), 127
(14), 105 (13), 91 (5), 77 (21), 57 (55), 29 (21); high
FB
Acknowledgements
‡
Â
The ®nancial support by Rhodia Recherche, Electricite de
France, CNRS, Ministere EN and Region Aquitaine
Á
Â
resolution MS: M ˆ 262:116922 (262.111389, ‡21.1 ppm).
(France) is gratefully acknowledged.
4.14. 2-Phenyldifluoromethylcyclohexanol
References
[1] J.T. Welch, S. Eswarakrishnan, Fluorine in Bioorganic Chemistry,
Wiley, New York, 1991.
[2] I. Ojima, J.R. McCarthy, J.T. Welch (Eds.), Biochemical Frontiers of
Fluorine Chemistry, ACS symposium series, Vol. 639, Washington
DC, 1996.
1
C13H16OF2; M ˆ226:27 g mol ; white solid; Fˆ76ꢁC.
[3] G.K.S. Prakash, Chem. Rev. 97 (1997) 757.
1H NMR: dH9±13 1.62 ppm (m, 10H), dH 2.02 ppm (s, 1H,
OH), dH 7.39±7.52 ppm (m, 5H, Ph); 19C NMR: dC12,C10
20.7 ppm (s), dC11 25.6 ppm (s), dC9,C13 30.3 ppm (t,
[4] P. Clavel, M.-P. Leger-lambert, C. Biran, F. Serein-Spirau, M.
Bordeau, N. Roques, H. Marzouk, Synthesis 5 (1999) 829.
[5] P. Clavel, C. Biran, J. Dunogues, N. Roques, Patent Rhodia chimie,
1999 (registered under R99034 (1/4)/99.
3JC ˆ 2:2 Hz), d 74.4 ppm (t, JC ˆ 27:7 Hz), d
2
F
F
C8
C1
122.9 ppm (t, 1JC ˆ 249:4 Hz), dC3,C7 127.1 ppm (t,
[6] M. Kako, T. Morita, T. Totihara, Y. Nakadaira, J. Chem. Soc., Chem.
Commun. (1993) 678.
F
3JC ˆ 6:7 Hz), d
127.3 ppm (s), dC5 129.6 ppm (s),
F
C4,C6
[7] P. Clavel, Thesis of Bordeaux 1 University, No. 2118.
[8] C. Saboureau, M. Troupel, S. Sibille, J. Perichon, J. Chem. Soc.,
Chem. Commun. (1989) 1138.
dC2 134.2 ppm (t, 3JC ˆ 26:7 Hz); 19F NMR: dF
F
111.4 ppm (s); IR: 3421, 2940, 1960, 1886, 1808, 1769,
1450, 1278, 1185, 1146, 1043, 984; MS: m/z (relative
[9] H. Lund, N.J. Jensen, Chem. Acta Scand. B28 (1974) 263.

310
P. Clavel et al. / Journal of Fluorine Chemistry 107 (2001) 301±310
[10] D. Deffieux, M. Bordeau, C. Biran, J. Dunogues, Organometallics 13
(1994) 2415.
[16] J.R. Hwu, B.L. Chen, S.S. Shiao, J. Org. Chem. 60 (1995)
2448.
[11] C.P. Andrieux, A. Le Gorande, J.M. Saveant, J. Am. Chem. Soc. 114
(1992) 6892.
[17] A.R. Bassindale, P.G. Taylor, in: S. Patai, Zvi. Rappoport (Eds.), The
Chemistry of Organic Silicon Compounds, Wiley, New York, 1989,
Chapter 14, p. 893.
[12] J.S. Jaworski, Tetrahedron Lett. 40 (1999) 5771.
[13] H. Jensen, K. Daasbjerg, Acta Chem. Scand. 52 (1998) 1151.
[14] A.M. Doncaster, R.J. Walsh, J. Chem. Soc., Faraday Trans.1 72
(1976) 2908.
[18] C. Biran, N. Duffaut, J. Dunogues, R. Calas, J. Organomet. Chem. 85
(1975) 149.
[19] T. Hagiwara, T. Fuchigami, Synth. lett. (1995) 717.
[20] M. Yoshida, D. Suzuki, M. Iyoda, J. Chem. Soc., Perkin Trans. 1
(1997) 643.
[15] I.M.T. Davidson, T.J. Barton, K.J. Hugues, S. Ijadi-Maghsoodi, A.
Ravis, G.C. Paul, Organometallics 6 (1987) 644.