Adsorption effects on the selective electrofluorination of α-(phenylthio)acetamides and α-(phenylthio)acetates in Et3N·3HF/CH3CN

Article abstract of DOI:10.1016/S0022-1139(98)00279-6

Cyclic voltammetric behaviour of α-(phenylthio)acetamides and α-(phenylthio)acetates were compared in CH₃CN/tetrabutylammonium perchlorate (TBAP) and CH₃CN/Et₃N·3HF to assess the influence of fluoride ion on the oxidative

Full text of DOI:10.1016/S0022-1139(98)00279-6

Journal of Fluorine Chemistry 93 (1999) 53±59
Adsorption effects on the selective electro¯uorination of
a-(phenylthio)acetamides and a-(phenylthio)acetates
in Et3NÁ3HF/CH3CN
V. Suryanarayanan, S. Chellammal, M. Noel*
Central Electrochemical Research Institute, Karaikudi, 630 006, India
Received 28 May 1998; accepted 11 September 1998
Abstract
Cyclic voltammetric behaviour of a-(phenylthio)acetamides and a-(phenylthio)acetates were compared in CH3CN/tetrabutylammonium
perchlorate (TBAP) and CH3CN/Et3NÁ3HF to assess the in¯uence of ¯uoride ion on the oxidative pathway. These studies support the
Pummerer mechanistic pathway involving an initial attack at sulfur in selective electro¯uorination (SEF). A linear correlation between the
peak current values in cyclic voltammetry and overall synthetic yield in preparative electrolysis was obtained suggesting the importance of
weak blocking or a reactive adsorption effect on the overall synthetic processes. Other factors in¯uencing the adsorption and insulation of
electrode surfaces during the SEF are also discussed. # 1999 Elsevier Science S.A. All rights reserved.
Keywords: Selective electro¯uorination; Adsorption effects; a-(phenylthio)acetamides; a-(phenylthio)acetates
1. Introduction
the evidence for this mechanism [6]. However, it would
be worthwhile to look for further evidence in support of
this pathway from voltammetric [13,14] as well as prepara-
tive electrolytic studies. This is the objective of the present
work.
In contrast to earlier attempts towards nuclear ¯uorination
[1], current efforts were mainly oriented towards side chain
¯uorination [2,3]. Among the active methylene groups,
those containing ±S±CH2±Z were found to undergo very
facile selective electro¯uorination (SEF) where Z is an elec-
tron withdrawing substituent [4±6] or a nitrogen heterocyclic
ring [7,8]. The selective electrochemical ¯uorination (SEF)
of ethyl a-(phenylthio)acetate (PhSCH2COOEt) was studied
as a model compound in greater detail using a ¯ow cell and
different anode materials [9] and T. Fuchigami has recently
reviewed this [10]. The effect of substituents on the overall
ef®ciency of SEF has also been recently evaluated [11,12].
The mechanism of SEF was initially assumed to be of
an ECEC type involving electron transfer, proton release,
electron transfer and nucleophilic ¯uoride attack; in this
sequence.
2. Experimental
The purity of the samples was checked by HPLC (Shi-
madzau LC 8-A) with a Shim-pack (4.0 mm  25 cm) ODS
column and 100% methanol as eluent. FT IR spectra of the
starting materials were taken with a Perkin-Elmer 783
1
spectrometer. H NMR spectra were taken with a Brucker
NMR spectrometer at 90 MHz with tetramethylsilane as
internal standard and CDCl3 as solvent and 19F NMR spectra
with a Brucker WP 80 CY spectrometer at 75.4 MHz with
CFCl3 as standard and CDCl3 as solvent. Positive shifts were
designated as negative.
Acetonitrile (CH3CN, HPLC grade) was obtained from
SRL. The electrolyte Et3NÁ3HF was prepared by mixing
Et3N and anhydrous HF (AHF) at low temperature and
evaluating the HF content by titration. Tetrabutylammonium
perchlorate (TBAP) was dried in a vaccum dessicator and
used. A series of substituted a-(phenylthio)acetamides
PhSCH2CONHR0 where
However, Fuchigami and his co-workers, have proposed a
Pummerer mechanism involving initial oxidation at sulfur to
explain the facile nature of this class of compounds [6]
(Scheme 1).
The difference in the ef®ciencies of anodic ¯uorination
and methoxylation under identical conditions was cited as
*Corresponding author. Tel.: +91-4565-23213; fax: +91-4565-22088.
0022-1139/99/$ ± see front matter # 1999 Elsevier Science S.A. All rights reserved.
P I I : S 0 0 2 2 - 1 1 3 9 ( 9 8 ) 0 0 2 7 9 - 6

54
V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
lysed at 3.0 V until the current dropped to a constant value
of10 mA. The 5 mmol of compound was added and elec-
trolysed with the polarities of the electrode being reversed at
interval of 10 s.
After passage of charge for (2 ‡ 0.4F/mol) reaction,
Scheme 1. Pummerer Mechanism
mono as well as di¯uoro products were obtained.
PhSCH2R Et NÁ3HF=CH CN
!
PhSCHFR ‡ PhSCF2R
ꢀ1b 8b†
3
3
R0 ˆ C6H4 Cl p ꢀ3a†
R0 ˆ H ꢀ5a†
ꢀ1a 8a†
ꢀ1c 8c†
The solution was neutralised with aqueous ammonia and
acetonitrile was evaporated. The liquid content was poured
into 100 ml of water and extracted with ether. The ether
layer was washed with brine solution (200 ml), the ether
phase was dried over anhydrous MgSO4. HPLC was mainly
used to ®nd the product distribution after each experiment.
Under optimum conditions, prolonged preparative electro-
lysis was carried out, the ¯uorinated product was isolated
and its structure was con®rmed by 1H NMR and 19F NMR.
R0 ˆ C6H4 NO2 p ꢀ6a†
R0 ˆ C6H5 ꢀ7a†
and a-(phenylthio)acetates PhSCH2COOR00 where
R00 ˆ CH3 ꢀ1a†
R00 ˆ C6H4 Cl p ꢀ2a†
R00 ˆ C6H4 NO2 p ꢀ4a†
R00 ˆ C6H5 ꢀ8a†
were prepared and characterised by measuring the m.p. or
b.p., IR and NMR [15].
3. Results and discussion
Cyclic voltammetry was performed as previously
described [16]. Preparative electrolysis was carried out at
Pt anode and cathode (area 9 cm2) at 258C Æ 18C under a
dry N2 atmosphere. The electrolyte 0.3 M Et3NÁ3HF/
CH3CN was placed in the undivided cell and pre-electro-
3.1. Cyclic voltammetry
Typical cyclic voltammetry (CV) responses for com-
pound 1a at different sweep rates in CH3CN/TBAP are
presented in Fig. 1(A). Three distinct anodic peaks are
Fig. 1. Cyclic voltammograms for the oxidation of 4 mM of C6H5SCH2COOCH3 on Pt in CH3CN (A) 0.1 M TBAP and (B) Et3NÁ3HF at different scanning
speeds (v) (mV s 1). (a) 20, (b) 40, (c) 80, (d) 160, (e) 320, and (f) 640.

V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
55
noticed. All the anodic peaks are found to increase with
sweep rates (Fig. 1(A)) and concentration of reactant. In all
respects, the CV responses obtained for 1a are similar to
voltammetric responses obtained earlier for arylsul®des
[13,14]. In CH3CN/TBAP other compounds (2a±8a) also
showed similar behaviour. Hence, the general oxidative
mechanistic pathway discussed for arylsul®des [13,14] also
should hold good for the arylsul®des containing active
methylene groups currently investigated here. This is con-
®rmed by the present voltammetric investigation.
give higher yields in SEF (see later). The individual ®rst
anodic peak current values for the four compounds inves-
tigated here do not show a linear dependence on molecular
size or molecular weight (Table 1, Fig. 2) as would be
expected for a simple diffusion controlled process. This
suggests a predominant role for adsorption or relative rate of
chemical steps.
Compounds 8a is an exception to the general trends
outlined as shown in Fig. 3. For this compound, ipal in
0
CH3CN/TBAP is always considerably higher than the ipal
Compared to the CH3CN/TBAP system, the CH3CN/
Et3NÁ3HF system gives slightly lower anodic limit. Sig-
ni®cant background oxidation current is always noticed
beyond 2 V in CH3CN/Et3NÁ3HF medium. Typical CV
for compound 1a in Et3NÁ3HF medium are presented in
Fig. 1(B). The anodic peak current is slightly higher in this
medium (compare Fig. 1(A) and Fig. 1(B)). The oxidation
peak potential is also shifted to a slightly less positive
region. Similar patterns in the CV responses were also
noted for compounds 2a±8a in CH3CN/Et3NÁ3HF. In this
medium since the back-ground itself increases beyond 2 V,
the third anodic peak observed in CH3CN/TBAP is not
distinctly visible (compare Fig. 1(A) and Fig. 1(B)).
The anodic peak potentials and peak current values for all
the eight compounds 1a±8a are summarised in Table 1. The
®rst anodic peak potential is generally found to shift to less
positive values indicating that the oxidation process pro-
ceeds more easily in CH3CN/Et3NÁ3HF and the peak current
is found to be higher in CH3CN/Et3NÁ3HF when compared
to CH3CN/TBAP (Table 1). The general voltammetric pat-
terns are also similar. Hence, one may presume that the
active site for the initial oxidation of the reactant is the same
in CH3CN/TBAP as well as CH3CN/Et3NÁ3HF media,
namely the ±S± atom. This observation supports the Pum-
merer mechanistic pathway (Scheme 1) suggested earlier
for similar compounds.
value in CH3CN/Et3NÁ3HF medium (Fig. 3). The com-
pound exhibits signi®cant adsorption effects in CV mea-
surements. With increase in concentration, for example the
CV shows a sharp drop beyond the anodic peak potential
value in the later medium alone (Fig. 4). At 10 mM con-
centration levels, for example, in multisweep, voltammo-
grams, the anodic peak current decreases signi®cantly in
second and subsequent sweeps. The electrode requires
repeated cleaning after each voltammetry recording to
ensure reproducibility.
3.2. Preparative electrolysis
For compounds which do not show blocking type of
adsorption during constant current electrolysis when the
overall cell voltage would remain fairly stable except for a
small increase due to concentration over-voltage. A typical
cell voltage vs. time curve for PhSCH2COOMe is shown in
Fig. 5 (curve a). In cases where blocking type adsorption is
predominant, the active surface area (1 ꢀ) continuously
decreases as surface coverage due to insulating ®lm (ꢀ)
develops. The cell voltage-time response for compound 8a
(PhSCH2COOPh) under identical conditions for example,
increases from 3 V to nearly 7 V during electrolysis (Fig. 5
Curve b).
Under ideal conditions, passage of charge for a
2F ‡ 0.4F/mol reaction should lead to the mono¯uoro
derivative. But, HPLC data shows that in most cases, two
product peaks correspond to mono¯uoro as well as di¯uoro
derivative. For compound 8a, HPLC shows multi peaks with
poor reproducibility. The molar yield of mono¯uoro deri-
vative, di¯uoro derivative and total yield for all the eight
Generally the peak current in CH3CN/TBAP as well as in
CH3CN/Et3NÁ3HF is found to increase with increase in
concentration. The relative variation in the peak current
values depends signi®cantly on the nature of individual
0
compounds. The compounds for which the ipal values in
CH3CN/Et3NÁ3HF medium are signi®cantly higher also
Table 1
Oxidation peak potentials and peak currents of sulfides containing active methylene group, PhSCH2R
2
ipal (mA cm )
No.
Sub. no.
Substrate
[R]
Epal (V)
TBAP/CH3CN
Et3N.3HF/CH3CN
TBAP/CH3CN
Et3N.3HF/CH3CN
1.
2.
3.
4.
5.
6.
7.
8.
1a
2a
3a
4a
5a
6a
7a
8a
COOMe
1.62
1.84
1.63
1.76
1.71
1.69
1.71
1.69
1.46
1.72
1.63
1.62
1.57
1.52
1.48
1.04
3.27
2.71
1.50
2.78
1.64
1.58
1.48
3.69
4.43
3.75
3.06
2.94
2.17
2.11
2.46
1.07
COOC6H4-Cl-P
CONCH6H4-Cl-p
COOC6H4-NO2-p
CONH2
CONHC6H4-NO2-p
CONHC6H5
COOC6H5
Ipal ± first anodic peak current, Epal ± first anodic peak potential, S.R. ˆ 40 mV/sec, Concentration ˆ 10 mM.

56
V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
0
Fig. 2. Plots of ip vs. C of sulfides containing active methylene group [1a±8a] where ipal is peak current in 0.1 M TBAP/CH3CN and ipal in 0.1 M Et3N.3HF/
CH3CN.
compounds are summarise in Table 2 and their 19F char-
acteristics in Table 3.
1a, di¯uoro derivative (corresponding to 4F/mol reaction) is
found to be the predominant product. This may be due to the
blocking type of adsorption and/or polymerisation by a
competitive pathway. The concentration of reactant avail-
Suf®cient charge for a 2F ‡ 0.4F/mol reaction was
passed for all compounds. However, except for compound
Table 2
Selective electrofluorination of sulfides containing active methylene groups
No.
Sub. No.
Sulfides
Average cell voltage (V)
Product yielda
Total yield
Unreacted substrate
b
c
b ‡ c
1.
2.
3.
4.
5.
6.
7.
8.
1a
2a
3a
4a
5a
6a
7a
8a
COOMe
3.80
5.10
5.20
5.00
4.70
5.10
5.80
7.00
72
12
7
5
77
68
63
56
49
42
34
27b
2
±
COOC6H4-Cl-p
CONHC6H4-Cl-p
COOC6H4-NO2-p
CONH2
56
56
50
27
36
28
10
±
6
14
22
5
22
6
CONHC6H4-NO2-p
CONHC6H5
6
10
14
COOC6H5
17
a
± Yield from HPLC.
b
± Multiple peaks.
PhSCH2R Et NÁ3HF
!
PhSCHFR ‡ PhSCF2R
ꢀ1b 8b†
ꢀ1c 8c†
3
CH3CN
ꢀ1a 8a†

V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
57
Fig. 5. Plots of cell voltage vs. time of (a) C6H5SCH2COOCH3 and (b)
C6H5SCH2COOC6H5 during the constant current electolysis in 0.3 M
Et3N.3HF/CH3CN.
Fig. 3. Cyclic voltammograms for the oxidation of 4 mM of
C6H5SCH2COOC6H5 on Pt in CH3CN (A) 0.1 M TBAP and (B) Et3N.3HF
at different v (mV s 1) (a) 20, (b) 40, (c) 80, (d) 160, (e) 320, and (f) 640.
Table 3
19F NMR characteristics of products
19F NMR of products (ppm)
Sub. no. (a)
Monofluoro (b)
Difluoro (c)
1a
2a
3a
4a
5a
6a
7a
8a
160.7 (d, 1F, J ˆ 52 Hz)
158.7 (d, 1F, J ˆ 52 Hz)
157.6 (d, 1F, J ˆ 52 Hz)
161.2 (d, 1F, J ˆ 52 Hz)
160.3 (d, 1F, J ˆ 52 Hz)
161.7 (d, 1F, J ˆ 52 Hz)
158.3 (d, 1F, J ˆ 52 Hz)
±
83.3 (s, 2F)
84.0 (s, 2F)
82.9 (s, 2F)
84.1 (s, 2F)
83.4 (s, 2F)
83.6 (s, 2F)
82.3 (s, 2F)
±
(Table 2) show that a adsorption effect can also be lower for
some compounds containing more than one phenyl group.
Fig. 4. Cyclic voltammograms for the oxidation of C6H5SCH2COOC6H5
at Pt in CH3CN containing 0.1 M Et3N.3HF at different concentrations
(mM), v ˆ 40 mV s 1. (a) 2, (b) 4, (c) 6, (d) 8, and (e) 10.
3.3. Comparison of results of cyclic voltammetry (CV) and
preparative electrolysis
Cyclic voltammetry (CV) data presented in Table 1
clearly suggest that the peak currents are not related to
molecular size and molecular weight as noted above
(Table 2). The anodic peak potential in presence of
CH3CN/Et3NÁ3HF varies between 1.45 and 1.72 V except
for compound 8a where the ®rst anodic peak is noted around
1.0 V (Table 1). The compounds such as PhSCH2COOMe
and PhSCH2COOPh which give the highest yield and lowest
yield, respectively (Table 1) also show signi®cant devia-
tions from the linear relationship (Fig. 6).
able for ¯uorination would be considerably lower than the
initial concentration taken and the charge passed per actual
mole of reactant available for ¯uorination is considerably
higher than the charge required for a 2F reaction leading to
the formation of di¯uoro derivative. T. Fuchigami et al. also
suggested an in¯uence of insulating polymer ®lm on the
product distribution pattern [17] and J. Simonet et al. have
indicated that this blocking type of adsorption is absent for
a-(phenylthio)derivatives containing alkyl groups attached
to methylene and electron withdrawing substituents in the
phenyl ring [11,12]. The results of the present study
Quite interestingly, much better correlation between the
®rst anodic peak current and the total molar yield was found

58
V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
Fig. 6. Plots of Ep (10 mM, 40 mV s 1) vs. molar yields of PhSCH2CONHR and PhSCH2COOR where R ˆ [1a±8a].
Fig. 7. Plots of ip (10 mM, 40 mV s 1) vs. molar yields of PhSCH2CONHR and PhSCH2COOR where R ˆ [1a±8a].
for all the eight compounds (1a±8a as shown in Fig. 7). The
higher the ®rst anodic peak current, the better the observed
yield. This con®rms the view that the formation of insulating
®lm by adsorption decreases the anodic peak current in CV
and the overall yield in preparative electrolysis. The exact
cause for the adsorption effect or the correlation between the
molecular structure and adsorption is not clear.
Two observations on the structure-synthetic yield corre-
lation may be made. In general, a-(phenylthio)acetates
undergo more ef®cient ¯uorination when compared to a-

V. Suryanarayanan et al. / Journal of Fluorine Chemistry 93 (1999) 53±59
59
(phenylthio)acetamide derivatives. Compared to the substi-
tuted phenyl rings, unsubstituted phenyl rings in the acetates
as well as amide derivatives exhibit strong inhibitive adsorp-
tion effects and hence very low overall yield due to their
large molecular size and molecular weight.
CSIR, New Delhi for the award of fellowship. Thanks are
also due to Mr. S. Govindu for analytical assistance.
References
[1] I.N. Rozhkov, Russ. Chem. Rev. 45 (1976) 615.
[2] M. Noel, V. Suryanarayanan, S. Chellammal, J. Fluorine Chem. 83
(1997) 31.
4. Conclusions
[3] S.M. Lee, J.M. Roseman, C.B. Zartman, E.P. Morrison, S.J.
Harrison, C.A. Stankeiwicz, W.J. Middleton, J. Fluorine Chem. 77
(1996) 65.
Cyclic voltammetric responses of sul®des containing
active methylene groups in CH3CN/Et3NÁ3HF are similar
to those obtained in conventional CH3CN/TBAP medium
suggesting that the SEF also proceeds through the sul®de
cation radical route termed as the Pummerer mechanism.
However, adsorption properties of individual compounds
in¯uence the overall yield ef®ciency of SEF signi®cantly.
Reactants which show little blocking type adsorption beha-
viour in voltammetric studies, invariably lead to high yields
of mono¯uoro derivatives. In preparative electrolysis, wher-
ever blocking type of adsorption predominates, the overall
molar yield was found to be low and di¯uoro derivatives
were predominant. An interesting direct correlation between
cyclic voltammetric peak currents and overall molar yields
of products was obtained for all the compounds involved.
This suggests the possibility of ascertaining the synthetic
potentiality of reaction from preliminary cyclic voltam-
metric results.
[4] T. Fuchigami, M. Shimojo, A. Konno, K. Nakagawa, J. Org. Chem.
55 (1990) 6074.
[5] T. Brigaud, E. Laurent, Tetrahedron Lett. 31(16) (1990) 2287.
[6] T. Fuchigami, A. Konno, K. Nakagawa, M. Shimoja, J. Org. Chem.
59 (1994) 5937.
[7] A.W. Erian, A. Konno, T. Fuchigami, Tetrahedron Lett. 35 (1994)
7245.
[8] T. Fuchigami, M. Shimojo, A. Konno, J. Org. Chem. 60(11) (1995)
3459.
[9] A. Konno, T. Fuchigami, J. Appl. Electrochem. 25 (1995) 173.
[10] T. Fuchigami, Rev. Heteroatom chem. 10 (1994) 155.
[11] P. Baroux, R. Tardivel, J. Simonet, Tetrahedron Lett. 36 (1995)
3851.
[12] P. Baroux, R. Tardivel, J. Simonet, J. Electrochem. Soc. 144 (1997)
274.
[13] S. Chellammal, M. Noel, P.N. Anantharaman, Ind. J. Chem. Technol.
3 (1996) 274.
[14] M.N. Elinson, J. Simonet, L. Toupet, J. Electroanal. Chem. 350
(1993) 117.
[15] J.I.G. Cadogen, S.V. Ley, G. Pattenden, R.A. Raphael, C.W. Rees,
Dictionary of Organic Compound, 6th ed., vol 5, P. 5343 and
references cited.
Acknowledgements
[16] M. Noel, V. Suryanarayanan, S. Krishnamoorthy, J. Fluorine Chem.
74 (1995) 241.
The authors thank Volkswagen Stiftung, Germany, for
®nancial support of this work. One of thte author V.S. thanks
[17] T. Fuchigami, M. Sano, J. Electroanal. Chem. 414 (1996) 81.