A paradigm shift in rate determining step from single electron transfer between phenylsulfinylacetic acids and iron(III) polypyridyl complexes to nucleophilic attack of water to the produced sulfoxide radical cation: a non-linear Hammett

Article abstract of DOI:10.1002/poc.3571

Mechanism of oxidative decarboxylation of phenylsulfinylacetic acids (PSAAs) by iron(III) polypyridyl complexes in aqueous acetonitrile medium has been investigated spectrophotometrically. An initial intermediate formation between PSAA and [Fe(NN)₃]³⁺ is confirmed from the observed Michaelis–Menten kinetics and fractional order dependence on PSAA. Significant rate retardation with concentration of [Fe(NN)₃]³⁺ is rationalized on the basis of coordination of a water molecule at the carbon atom adjacent to the ring nitrogen of the metal polypyridyl complexes by nucleophilic attack at higher concentrations. Electron-withdrawing and electron-releasing substituents in PSAA facilitate the reaction and Hammett correlation gives an upward ‘V’ shaped curve. The apparent upward curvature is rationalized based on the change in the rate determining step from electron transfer to nucleophilic attack, by changing the substituents from electron-releasing to electron-withdrawing groups. Electron-releasing substituents in PSAA accelerate the electron transfer from PSAA to the complex and also stabilize the intermediate through resonance interaction leading to negative reaction constants (ρ). Conversely, electron-withdrawing groups, while retarding the electron transfer exert an accelerating effect on the nucleophilic attack of H₂O which leading to low magnitude of ρ⁺ compared to high ρ^? values of electron-releasing groups. Marcus theory is applied, and a fair agreement is seen with the experimental values. Copyright

Full text of DOI:10.1002/poc.3571

Research article
Received: 9 January 2016,
Revised: 11 March 2016,
Accepted: 18 March 2016,
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/poc.3571
A paradigm shift in rate determining step from
single electron transfer between
phenylsulfinylacetic acids and iron(III)
polypyridyl complexes to nucleophilic attack
of water to the produced sulfoxide radical
cation: a non-linear Hammett
Perumal Subramaniama*, Jebamoney Janet Sylvia Jaba Roseb and
Rajasingh Jeevi Esther Rathinakumarib
Mechanism of oxidative decarboxylation of phenylsulfinylacetic acids (PSAAs) by iron(III) polypyridyl complexes in aqueous
acetonitrile medium has been investigated spectrophotometrically. An initial intermediate formation between PSAA and
[Fe(NN)3]3+ is confirmed from the observed Michaelis–Menten kinetics and fractional order dependence on PSAA. Significant
rate retardation with concentration of [Fe(NN)3]3+ is rationalized on the basis of coordination of a water molecule at the car-
bon atom adjacent to the ring nitrogen of the metal polypyridyl complexes by nucleophilic attack at higher concentrations.
Electron-withdrawing and electron-releasing substituents in PSAA facilitate the reaction and Hammett correlation gives an
upward ‘V’ shaped curve. The apparent upward curvature is rationalized based on the change in the rate determining step
from electron transfer to nucleophilic attack, by changing the substituents from electron-releasing to electron-withdrawing
groups. Electron-releasing substituents in PSAA accelerate the electron transfer from PSAA to the complex and also stabilize
the intermediate through resonance interaction leading to negative reaction constants (ρ). Conversely, electron-withdrawing
groups, while retarding the electron transfer exert an accelerating effect on the nucleophilic attack of H2O which leading to
low magnitude of ρ+ compared to high ρꢀ values of electron-releasing groups. Marcus theory is applied, and a fair agreement
is seen with the experimental values. Copyright © 2016 John Wiley & Sons, Ltd.
Keywords: electron transfer reaction; iron(III) polypyridyl complex; Marcus theory; non-linear Hammett; phenylsulfinylacetic acid;
substituent effect
polypyridyl complexes are well-known one electron accep-
INTRODUCTION
tors,[13] and several interactions of these complexes with inor-
ganic and organic reductants have been reported.[6,13–17] Many
of the reactions are rationalized in terms of the outer-sphere
mechanism, supported by Marcus-type dependence of rate.
In recent years, much attention has been paid to the chemistry
of sulfur-centred radicals and radical cations because of their
importance as intermediates in organic synthesis, environmental
and biological studies. As these intermediates are reported in
Organo sulfur compounds, widely distributed in food,[1] are
functioning as antiseptics, antibiotics, antithrombotics, antioxi-
dants etc.[2] Sulfur containing compounds play an important role
in cellular biochemistry. Among the organic sulfur compounds,
organic sulfoxides are biphilic in nature, act either as electrophile
or nucleophile.[3] But still they resemble organic sulfides in many
oxidation reactions.[4–6] Some of the phenylsulfinyl compounds
have found to possess high therapeutic effects. They are used
for combating parasitic disorders with broad spectrum of action,
treating Alzheimer’s disease[7] and enhancing memory activities.
They also show antibacterial activities.[8]
* Correspondence to: Perumal Subramaniam, Research Department of Chemis-
try, Aditanar College of Arts and Science, Tiruchendur 628 216, Tamil Nadu,
India.
Iron is omnipresent on earth and is responsible for vital
biological reactions. The reduction of Fe(III) to Fe(II) by biological
reductants in biological systems is a well-known phenomenon.[6]
For clear understanding of these ET reactions, porphyrin and
polypyridyl complexes of Fe(III) have been synthesized as model
compounds and used as electron acceptors.[9–11] Recently iron
polypyridyl complexes have been identified as potential antican-
cer drug by inhibiting the cancer cell proliferation.[12] The iron(III)
E-mail: subramaniam.perumal@gmail.com
a P. Subramaniam
Research Department of Chemistry, Aditanar College of Arts and Science,
Tiruchendur 628 216, Tamil Nadu, India
b J. Janet Sylvia Jaba Rose, R. Jeevi Esther Rathinakumari
Department of Chemistry, Nazareth Margoschis College, Nazareth 628 617,
Tamil Nadu, India
J. Phys. Org. Chem. (2016)
Copyright © 2016 John Wiley & Sons, Ltd.

P. SUBRAMANIAM, J. JANET SYLVIA JABA ROSE AND R. JEEVI ESTHER RATHINAKUMARI
most of the sulfoxidation reactions, and there is no systematic
mechanistic study on the oxidation of phenylsulfinylacetic acid
(PSAA) in the literature except our recent publications,[18–21]
systematic mechanistic investigation on the reaction between
biologically active iron(III) polypyridyl complexes and therapeuti-
cally active PSAA is undertaken in the present work. On the basis
of the observed spectral, kinetic, thermodynamic and substitu-
ent effects a suitable mechanistic path is proposed.
coefficients of the order 1 × 104 Mꢀ1 cmꢀ1 and the corresponding
iron(III) complexes are transparent in the above absorption
regions,[6,22] the reactions of PSAAs with [Fe(NN)3]3+ complexes
were monitored by measuring the increase in absorbance with
time as shown in Fig. 2.
The kinetics of the reaction was followed spectrophotometri-
cally under pseudo first order conditions with excess of PSAA
over [Fe(NN)3]3+ in 50% aqueous acetonitrile (v/v) medium. The
absorption spectral studies were carried out on a double beam
BL 222 Elico UV–vis bio spectrophotometer with an inbuilt
thermostat. The pseudo-first-order rate constants were calcu-
lated from the linear plots of log(Aα ꢀ At) versus time by least
squares method,[6] where Aα is the maximum absorbance
obtained by the completion of the reaction and At is the absor-
bance at time t. The overall rate constants (kov) were calculated
from the relation, kov = k1/[PSAA]order. The error in the rate
constants is given according to 95% of the Student’s t-test.
a
EXPERIMENTAL SECTION
General
The ligands 2,2′-bipyridine (bpy), 4,4′-dimethyl-2,2′-bipyridine
(dmbpy), 1,10-phenanthroline (phen), 4,7-dimethyl-1,10-
phenanthroline (dmphen) and 5-chloro-1,10-phenanthroline
(Clphen) were obtained from Sigma-Aldrich and used as such.
Fe(III) polypyridyl complexes [Fe(NN)3]3+ were prepared by the
oxidation of corresponding Fe(II) tris(pyridyl) complexes with
lead dioxide in sulphuric acid medium.[6] The preparation of [Fe
(NN)3]3+ must be done in highly acidic medium in order to get
better yield. Finally, Fe(III) complexes were precipitated as per-
chlorate salts. The purity of the complexes was checked from
their IR and absorption spectra. The tris(pyridyl) complexes of
Fe(II) were obtained by known procedure.[22] The structure of
[Fe(NN)3]3+ complexes used in the present study and their
abbreviations are shown in Fig. 1.
Stock solutions of Fe(III) complexes were made up in concen-
trated perchloric acid and were diluted with aqueous acetonitrile
just before initiating the kinetic run. In order to avoid the decompo-
sition of complexes, stock solutions were kept in refrigerator and
no solution samples older than 12 h were used. CAUTION: As metal
perchlorates have a potential to explode, they should be handled
with utmost care, including working with as small amounts as
possible and preventing exposure to elevated temperatures.
PSAA, meta- and para-substituted PSAAs were prepared
from the corresponding phenylthioacetic acid (PTAA) by the
controlled oxidation with hydrogen peroxide.[17] The PTAAs were
prepared by condensing thiophenols and chloroacetic acid in
alkaline medium by known procedure.[23] PSAAs were purified
by recrystallization from ethyl acetate–benzene mixture and
their purities were checked by m.p. and LC-MS. The recrystallized
samples were stored in vacuum desiccator in order to avoid the
decomposition with moist air.
Product analysis
The reaction mixtures containing PSAA and [Fe(bpy)3]3+ / [Fe
(phen)3]3+ in 1:2 molar ratio under the experimental conditions
were kept aside for two days. After completion of the reaction
the organic solvent was removed under reduced pressure,
extracted with chloroform and dried over anhydrous sodium
sulfate. The sample obtained after the removal of the solvent
was characterized by GC-MS and FT-IR spectral methods and
the product was identified as diphenyl disulfone.
The FT-IR spectrum (Fig. 3) shows strong characteristic
bands at 1143 cmꢀ1 and 1294 cmꢀ1 which confirms the exis-
tence of iSO2 group in the product and the absorption bands
are assigned to symmetric and asymmetric stretching vibra-
tions of iSO2 respectively. The absorption bands at
2924 cmꢀ1 show aromaticity, 1570 cmꢀ1 show C¼C stretching,
737 cmꢀ1, 685 cmꢀ1 and 630 cmꢀ1 show C―H bending vibra-
tions of phenyl rings. GC-MS spectral analysis (Fig. 4) was done
using Thermo GC-Trace ultra ver : 5.0, Thermo MS DSQ II mass
spectrometer using DB 35-MS capillary standard non-polar col-
umn of length 30 m and internal diameter of 0.25 mm. The
sample was dissolved in benzene and 1 μL of the solution
was injected into the column. Helium gas was used as the car-
Kinetic measurements
The polypyridyl iron(II) complexes have absorbance at 522 nm for
[Fe(bpy)3]2+, 529nm for [Fe(dmbpy)3]2+, 510 nm for [Fe(phen)3]2+
,
513nm for [Fe(dmphen)3]2+ and 510 nm for [Fe(Clphen)3]2+. As
the iron(II) polypyridyl complexes have high molar extinction
Figure 2. Increase in the absorbance of [Fe(phen)3]2+ during the reac-
tion with the isobestic point
Figure 1. Structure of [Fe(NN)3]3+ complexes and their abbreviations
wileyonlinelibrary.com/journal/poc
Copyright © 2016 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2016)

SHIFT IN MECHANISM FROM SINGLE ELECTRON TRANSFER TO NUCLEOPHILIC ATTACK
calculated using the expression k2 = k1/[PSAA] for different
[PSAA] are not constant in a particular complex. Alternatively,
the relation k1/[PSAA]order gave a constant value. These show
fractional order dependence of rate on PSAA. The double recip-
rocal plots of k1 versus [PSAA] are linear with finite intercept
on the rate axis indicating an intermediate formation between
PSAA and [Fe(NN)3]3+ in the reaction mechanism before the rate
determining step, i.e. Michaelis–Menten kinetics.
Although the pseudo-first-order plots are found to be linear
up to 60% completion in all the experimental conditions the rate
of reaction decreases considerably with increase in [Fe(NN)33+
]
(Supporting Information Table S2). A possible explanation for
the observed decrease in rate would be coordination of a water
molecule by nucleophilic attack at the carbon atom adjacent to
the ring nitrogen of the iron(III) polypyridyl complexes (Eqn 1)
at higher concentrations making them inactive towards
oxidation.[24] The coordination of water to metal polypyridyl
complexes was proved by Burchett and Meloan[25] by infrared
studies and the possibility of attack of water at 2 and 9 positions
was shown by Schmid and Han[26] using CNDO calculations.
The formation of 1 during the course of the reaction in a
parallel reaction is ascertained by the tremendous magnitude
of rate retardation observed with complex 1e, containing
electron-withdrawing group (EWG), than other complexes. The
chloro substituent in 1e drains away the electron density from
the ring thus favouring the nucleophilic water attack and facili-
tating the formation of 1.
Figure 3. FT-IR spectrum of the product
Another possibility for the decrease in pseudo-first-order rate
constant with increase in [Fe(NN)33+] may be due to the
conversion of complex into oxo-bridged diiron complex (Fig. 5)
as proposed by Hey et al.[27,28] in the aqueous medium at higher
concentrations of [Fe(NN)3]3+. As the [Fe(NN)3]3+ increases the
rate of Eqn (1) and the formation of species (2) may be facili-
tated, resulting in decrease in stoichiometric concentration of
[Fe(NN)3]3+ in the reaction mixture. Thus the decrease in rate
constant with increase in [Fe(NN)3]3+ is explained on the basis
of deactivation of active species, [Fe(NN)3]3+ into inactive species
(1) and (2) in parallel reactions.
The increase in the concentration of acid favours the rate of
reaction (Supporting Information Table S3). The addition of bpy
increases the rate of reaction enormously in the case of iron(III)
bipyridyl complex (1a), whereas in iron(III) phenanthroline
complex (1c) the effect of phen is only marginal (Supporting
Information Table S4). The reason may be explained as follows:
In the absence of bpy, majority of [Fe(bpy)3]3+ may exist as less
active dimer form (2). As the [bpy] increases the dimer formation
Figure 4. GC-MS spectrum of the product
rier gas. The parent peak eluted at a retention time of
11.79 min at m/z = 282.3 in GC-MS confirms the formation of
diphenyl disulfone as the product in the reaction. The other
product was confirmed as [Fe(NN)3]2+ from the formation of
new peak in the absorption spectra where the characteristic
absorption increases with time during the course of the
reaction.
RESULTS AND DISCUSSION
The pseudo-first-order rate constants for all the reactions investi-
gated at various concentrations of PSAA with iron(III) polypyridyl
complexes (1a–e) increase progressively with [PSAA] but not in a
proportionate manner (Supporting Information—Table S1). The
order with respect to PSAA is determined from the slope value
of double logarithmic plot of k1 and [PSAA]. The plots of log k1
versus log [PSAA] for complexes (1a–e) gave a slope of non-
integral, fractional values of 0.890, 0.366, 0.661, 0.389 and 0.598
respectively. Further the second order rate constants, k2
Figure 5. Structure of oxo-bridged diiron complex
J. Phys. Org. Chem. (2016)
Copyright © 2016 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

P. SUBRAMANIAM, J. JANET SYLVIA JABA ROSE AND R. JEEVI ESTHER RATHINAKUMARI
is prevented followed by enhancement of rate. This may not be
constant (ρ = +0.351 to +0.836) except in complex 1a. The
unsubstituted PSAA has the least reactivity in this series. The
values of ρ+ and ρꢀ obtained for complexes 1a and 1c at differ-
ent temperatures are given in Table 2.
the case in [Fe(phen)3]3+ complex. This is also supported from
the oxidant variation studies, where appreciable rate retardation
is seen with increase in concentration of iron(III) bipyridyl com-
plexes compared to iron(III) phenanthroline complexes. The ionic
strength variation by the addition of NaClO4 has a considerable
influence on the rate of reaction. The rate increases steadily
upon increasing the ionic strength of the medium (Supporting
Information Table S4).
From these ρ values it is clear that the accelerating effect
shown by the electron releasing groups (ERG) is significantly
higher than that of the electron withdrawing groups. Such
type of high ρꢀ value is already reported in electron transfer
reaction between anilines and oxo(salen)-chromium(V) (ρ = ꢀ3.8),[29]
sulfoxidation of thioanisoles by CeIV (ρ = ꢀ3.3)[30] and
polypyridyl complexes (ρ = ꢀ3.2)[31] and the reactions of oxy-
gen atom transfer from aryl sulfoxides to alkyl sulfides
catalysed by rhenium(V) monoxo complex (ρ = ꢀ4.6).[32] High
ρ value is also observed in the oxidation reactions of organic
sulfides by chloramine-T (ρ = ꢀ4.25),[33,34] N-chlorosaccharin
(ρ = ꢀ3.33),[35] bromine (ρ = ꢀ3.2),[36] N-bromobenzamide
(ρ = ꢀ3.18)[37] and PTAAs by N-chlorosaccharin (ρ = ꢀ3.12)[38]
and ammonium meta vanadate (ρ = ꢀ3.64)[39] where electro-
philic attack of the oxidizing species on the sulfur centre
has been proposed as the rate-determining step.
In general, the non-linear Hammett correlation is diagnostic
of a change in the reaction mechanism or free radical mecha-
nism or a change in the rate determining step with change in
the nature of substituents. The involvement of free radical like
transition state in the oxidation of alcohols by ruthenate and
perchlorate ions,[40] the change in the relative importance of
bond formation and bond fission in the Cr(VI) oxidation of
benzylamines[41] and competition between the rates of com-
plex formation and its decomposition in the oxidation of
benzaldehydes by quinolinium chlorochromate[42] have been
given as explanation for the observed V-shaped Hammett
plots. V-shaped Hammett plots observed in manganese(V)
oxocorrolazine[43] and manganese(V) (imido)(corrole)[44] com-
plexes towards organic sulfur compounds were rationalized
by change in mechanism from electrophilic to nucleophilic
As an extension of the present investigation, the effect of
substituents on the reactivity has been studied at three temper-
atures ranging between 293 K and 313 K for complexes 1a and
1c and at 303 K for complexes 1b, 1d and 1e by utilizing several
meta- and para-substituted PSAAs. The overall rate constants
obtained with different complexes (1a–e) and PSAAs at 303 K
are given in Table 1. The rate data in Table 1 show that both
electron-withdrawing and electron-releasing substituents in the
phenyl ring of PSAA have accelerating effect on the rate. In
contradiction, the introduction of electron-withdrawing substitu-
ents in the polypyridyl complexes tremendously accelerates
the reaction while electron-releasing substituents retard the
rate significantly and a direct relation between reduction
potential of the complex and its reactivity is found.
Failure of the Hammett linear free energy relationship
On applying Hammett equation between substituent constant σ
and log kov of meta- and para-substituted PSAAs, a non-linear
correlation is observed. The Hammett plots on the reactivity for
the reactions of 1a–e with PSAAs exhibit concave upward curves
consists of two intersecting straight lines with a break point at
the unsubstituted PSAA (Fig. 6). The electron releasing substitu-
ents fall on one side of the curve with a large reaction constant
(ρ = ꢀ3.28 to ꢀ7.47) and the electron withdrawing substituents
fall on the other side of the curve with a small positive reaction
Table 1. Overall rate constants for the oxidation of para- and meta-substituted PSAAs by 1a–1e1
No.
X
102 kov (Mꢀ1sꢀ1
1c
)
1a
1b
1d
1e
1.
2.
3.
4.
5.
6.
7.
8.
p-F
7.43 0.05
14.9 0.03
18.0 0.01
29.4 0.09
34.7 0.04
38.5 0.03
6.03 0.01
12.2 0.01
56.9 0.01
93.6 0.01
163 0.01
338 0.03
485 0.03
2.18 0.11
0.995
0.144 0.01
0.171 0.01
0.189 0.12
0.212 0.02
0.218 0.01
0.222 0.01
0.137 0.01
0.209 0.06
0.348 0.01
0.525 0.02
0.711 0.01
1.13 0.03
1.63 0.03
0.583 0.06
0.982
1.76 0.01
2.12 0.01
2.24 0.01
2.59 0.01
2.74 0.01
2.92 0.01
1.49 0.01
2.37 0.01
6.73 0.03
9.63 0.03
11.9 0.03
19.2 0.02
23.9 0.01
0.652 0.05
0.989
0.168 0.01
0.198 0.01
0.219 0.01
0.243 0.01
0.246 0.01
0.249 0.01
0.158 0.01
0.343 0.04
0.424 0.01
0.586 0.01
0.722 0.01
1.02 0.06
1.39 0.02
0.534 0.06
0.978
6.47 0.04
7.24 0.06
7.57 0.02
7.98 0.09
8.74 0.11
8.65 0.08
6.09 0.02
22.7 0.22
54.7 0.29
73.8 0.38
107 0.01
172 0.12
195 0.23
0.391 0.04
0.975
p-Cl
p-Br
m-F
m-Cl
m-Br
H
m-Me
p-Et
p-Me
p-t.Bu
p-OEt
p-OMe
ρ+
9.
10.
11.
12.
13.
R
ρꢀ
7.47 0.34
0.995
4.02 0.32
0.985
4.71 0.22
0.995
3.28 0.25
0.986
5.58 0.33
0.992
R
1[H+] = 0.5 M, μ = 0.55 M, CH3CN–H2O = 50–50% (v/v), [X-PSAA] = 3 × 10ꢀ3 M, [1a] = [1b] = [1c] = [1d] = [1e] = 3 × 10ꢀ4 M, T = 303 K.
wileyonlinelibrary.com/journal/poc
Copyright © 2016 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2016)

SHIFT IN MECHANISM FROM SINGLE ELECTRON TRANSFER TO NUCLEOPHILIC ATTACK
Figure 6. Hammett plots for substituted PSAAs with 1b and 1c at 303 K
Table 2. Thermodynamic parameters for the oxidation of para- and meta-substituted PSAAs by 1a and 1c1
No.
X
1a
1c
102 k2 (Mꢀ1sꢀ1
293 K 313 K
)
Δ‡H kJ
ꢀΔ‡S J Kꢀ
102 k2 (Mꢀ1sꢀ1
)
Δ‡H kJ
ꢀΔ‡S J Kꢀ1
molꢀ1
1 molꢀ1
molꢀ1
molꢀ1
293 K
313 K
1.
2.
3.
4.
5.
6.
7.
8.
p-F
2.47 0.01 15.6 0.01 69.6 0.29 43.1 1.02
5.88 0.01 39.2 0.06 71.3 0.13 30.9 0.47
7.16 0.02 49.3 0.02 72.7 0.09 24.6 0.34
11.3 0.01 68.9 0.05 68.0 0.12 36.2 0.42
15.8 0.01 87.2 0.04 64.1 0.06 46.6 0.20
1.04 0.01
1.35 0.01
1.47 0.04
1.70 0.01
1.79 0.01
2.07 0.04
4.19 0.02 51.6 0.01
4.82 0.01 46.9 0.01
5.23 0.04 46.8 0.01
5.74 0.02 44.7 0.01
6.36 0.03 46.7 0.01
6.73 0.01 43.2 0.01
3.55 0.02 53.6 0.01
6.53 0.02 43.6 0.01
11.4 0.04 34.9 0.01
15.9 0.04 37.9 0.01
18.5 0.01 28.4 0.01
24.8 0.01 23.9 0.01
29.7 0.01 21.4 0.01
111 0.02
125 0.01
125 0.03
130 0.01
123 0.01
134 0.02
106 0.02
133 0.02
155 0.01
142 0.01
171 0.01
182 0.01
189 0.01
p-Cl
p-Br
m-F
m-Cl
m-Br
H
m-Me
p-Et
p-Me
p-t.Bu
p-OEt
p-OMe
ρ+
18.4 0.02
103 0.04 64.5 0.06 44.2 0.20
1.66 0.01 12.7 0.02 77.0 0.24 21.4 0.83 0.839 0.01
6.25 0.04 46.9 0.08 75.8 0.23 16.2 0.80
1.98 0.03
4.36 0.01
5.63 0.01
8.35 0.04
12.6 0.01
16.1 0.01
0.836 0.08 0.602 0.08
0.980
4.78 0.10
0.999
9.
25.1 0.01
39.5 0.03
56.6 0.05
119 0.04
169 0.05
162 0.06 70.1 0.02 23.0 0.08
216 0.03 63.9 0.03 39.6 0.09
261 0.07 57.3 0.03 57.8 0.11
629 0.06 62.6 0.01 34.3 0.05
869 0.03 61.5 0.01 34.8 0.04
10.
11.
12.
13.
2.57 0.15 2.31 0.17
0.987 0.991
7.51 0.19 6.79 0.18
0.999 0.996
r
0.969
3.46 0.13
0.997
ρꢀ
r
1[H+] = 0.5 M, μ = 0.55 M, CH3CN–H2O = 50–50% (v/v), [X-PSAA] = 3 × 10ꢀ3 M, [1a] = [1c] = 3 × 10ꢀ4 M
attack of complexes. Besides, similar V-shaped Hammett plots
were reported during the oxidative decarboxylation of PSAA
by oxo(salen)-chromium(V),[20] oxidation of trans cinnamic
acid by pyridinium chlorochromate[45] and chloramine-T,[46]
styrene derivatives by quarternary ammonium permanga-
nate[47] and sulfoxidation by titanium complex.[48]
The thermodynamic parameters were calculated from the lin-
ear Eyring’s plots[49] of log (kov /T) versus 1/T and the calculated
values are given in Table 2. The negative values of the entropy of
activation (Δ‡S) suggest extensive solvation and disorder
arrangement of the products over the reactants in the rate deter-
mining step. As no linear relationship exists between Δ‡H and
Δ‡S in the present series of reactions, a linear isokinetic relation-
ship is established from the rate constants at two different tem-
peratures as proposed by Exner[50] using the following equation.
log kovðT2Þ ¼ a þ b log kovðT1Þ
(2)
where T2 > T1
Excellent linear plots (r = 0.993 for 1a and r = 0.997 for 1c) of
log kov (313 K) versus log kov (293 K) not only prove a unified
mechanism[51] in all the PSAAs studied but also rule out the pos-
sibility of change in the reaction mechanism with substituent.
J. Phys. Org. Chem. (2016)
Copyright © 2016 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

P. SUBRAMANIAM, J. JANET SYLVIA JABA ROSE AND R. JEEVI ESTHER RATHINAKUMARI
Mechanism
has already been advocated by Rajagopal and co-workers[54] in
ruthenium(III) polypyridyl complexes and Adaikalasamy et al.[6]
When [Fe(NN)3]3+ is added to PSAA during the reaction, a sub-
stantial increase in the absorbance is noted in the reaction mix-
ture. This increase in absorbance is taken as spectral evidence for
the formation of new adduct (3) between PSAA and polypyridyl
complex in the initial stage of the mechanism.[52] It is also shown
that the increase in absorbance is more pronounced in
Phen complexes than in Bpy complexes. The observed
Michaelis–Menten kinetics[53] with PSAA can be taken as the
kinetic evidence for the formation of adduct (3). The strong
one electron nature of the oxidant [Fe(NN)3]3+ and formation
of a new peak characteristic of [Fe(NN)3]2+ as one of the products
of the reaction clearly confirm the formation of sulfoxide radical
cation (4) as the transient species by electron transfer from PSAA
to Fe(III). Similar type of sulfoxide radical cation formation
in iron(III) polypyridyl complexes. Based on these facts, a mecha-
nism involving sulfoxide radical cation in the rate determining
step (RDS) as shown in Scheme 1 has been proposed for the
reactions of PSAAs with [Fe(NN)3]3+
.
Electron-withdrawing substituents in the phenanthroline
ligand of [Fe(phen)3]3+ and electron-releasing substituents in
the phenyl ring of PSAA are found to enhance the rate of
reaction. These observations strongly favour the formation of
sulfoxide cation radical by internal transfer of an electron from
PSAA to [Fe(NN)3]3+ within the adduct (3) in the rate determining
step (Eqn 4). This is further supported by the rate retardation ef-
fect observed with electron-releasing substituents in [Fe(NN)3]3+
(Table 1). The electron-withdrawing chlorine in 1e decreases its
electron density at Fe(III) centre and favours the electron
Scheme 1. Mechanism for the reaction between PSAAs and [Fe(NN)3]3+
wileyonlinelibrary.com/journal/poc
Copyright © 2016 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2016)

SHIFT IN MECHANISM FROM SINGLE ELECTRON TRANSFER TO NUCLEOPHILIC ATTACK
accepting tendency from PSAA. The replacement of chlorine by
methyl makes the Fe(III) complex less electrophile and decreases
the electron accepting ability. Similarly the observed increase in
rate with acid concentration is explained on the basis of proton-
ation of [Fe(NN)3]3+ to form [HFe(NN)3]4+ as postulated by
Sutin[55] and Kmura[56] co-workers, which makes the oxidant
more electrophile thus favouring the formation of (3) and ET
from PSAA to the complex. In addition, H+ ions also stabilize
the sulfonium cation radical (4) formed in the rate determining
step. It has been shown that[57] the sulfur radical cation is more
stabilized by [H+]. Thus the rate enhancement by the addition
of acid gave additional evidence for the formation of sulfonium
cation radical. As charged species are more easily formed in high
ionic medium, an enhancement of rate observed with ionic
strength also supports the proposed mechanism. The
insignificant effect observed in the rate constant by the addition
of [Fe(NN)3]2+ to the reaction mixture indicates that the conver-
sion of Fe3+ to Fe2+ is not a reversible process. The existence of
isobestic point in the overlay spectrum of kinetic runs (Fig. 2)
during the reduction of [Fe(NN)3]3+ to [Fe(NN)3]2+ clearly demon-
strates that the conversion of Fe(III) to Fe(II) proceeds via simple
kinetics without any intermediate form.
The formation of sulfone as the final product of PSAA leads to
the conclusion that the major portion of sulfoxide cation radical
is consumed by the solvent, water. Thus formation of sulfoxide
radical (5) is proposed as a result of nucleophilic attack of water
on sulfonium cation intermediate (4) (Eqn 5) which then trans-
fers its electron to another [Fe(NN)3]3+ (Eqn 6) that leads to the
formation of sulfoxide cation. Such type of attack by water
followed by second electron transfer to another [Fe(NN)3]3+ is al-
ready shown in the oxidation of sulfur compounds by
polypyridyl complexes.[6,53] The sulfoxide cation then undergoes
rearrangement to form phenylsulfonyl free radical via cleavage
of a C-S bond β to the aromatic ring.[58] Srinivasan et al.[59]
showed a fairly intense peak corresponding to C6H5SO•2 radical
in the mass spectrum of phenylsulfonylacetic acid by the loss
of •CH2COOH. The same radical is also reported in the photolysis
study of PTAA by Filipiak et al.[60] Finally phenylsulfonyl free rad-
icals undergo dimerization leading to the formation of diphenyl
disulfone as the product, and such type of dimerization reaction
has been reported earlier by Green et al.[61] If the formation of
cation radical is the one and only rate determining step and
the electronic effect is the deciding factor, then the rate would
be decreased by EWGs on the phenyl ring of PSAA. In contradic-
tion, the observed rate acceleration with EWG points out that
electronic effect alone does not decide the reaction rate.
The strong evidence for the nucleophilic attack of water on inter-
mediate (4) (Eqn 5) as the rate determining step for EWG comes
from enormous increase in rate with increase in water content of
the medium for PSAA containing EWG. Interestingly no such sol-
vent effect is observed with PSAA and PSAA containing ERG
(Table 3) which indicates the non-involvement of water molecule
in these PSAAs. The observed low magnitude of ρ+ (Table 1 and
Fig. 6) with electron-withdrawing substituents compared to ρꢀ of
electron-releasing groups is because of opposite substituent ef-
fects of EWG on reaction rates, i.e. electron-withdrawing substitu-
ents simultaneously retard the electron transfer step (Eqn 4) and
exert accelerating effect in the nucleophilic attack of H2O (Eqn 5).
Thus, the observed non-linear Hammett behaviour with upward
curvature in the present reaction can be visualized to a change
in the rate determining step on changing the substituents in PSAA.
Application of Marcus theory of electron transfer
Marcus analysis[62] has been successfully applied to many ther-
mal, photochemical and electrochemical reactions where single
electron transfer takes place in the rate determining step. In
order to confirm the proposed single electron transfer in the
present reaction, it is subjected to the theoretical work of Marcus
cross-relation. According to this concept, the rate constant (k12)
for an ET reaction (Eqn 9) depends on intrinsic reactivity of the
redox couples involved, i.e. self-exchange rate constants at
zero driving force, k11 and k22 and the thermodynamics of
the couples.
(9)
(10)
(11)
The rate constant, k12 for an ET cross reaction obtained from
Marcus cross-relation in its simple form is given as
1=2
k12 ¼ ðk11k22K12f12
Þ
(12)
(13)
Â
Ã
2
ln f12 ¼ ½ lnðK12Þꢁ = 4 ln k11k22= z2
:
The apparent upward curvature in the Hammett plots (Fig. 6)
observed in the present case is rationalized based upon the
change in the rate determining step in the proposed mechanism
(Scheme 1) upon changing the substitutes in PSAA. Electron
releasing substituents in PSAA accelerate the electron transfer
from sulfur atom to iron in adduct (3) (Eqn 4) and also stabilize
the intermediate (4) through resonance interaction leading to
negative reaction constants (ρ). At the same time, they retard
the nucleophilic attack of OHꢀ on the sulfoxide cation radical
(Eqn 5). Conversely, the EWGs while retarding the electron trans-
fer (Eqn 4) not only exert an accelerating effect on the nucleo-
philic attack of H2O (Eqn 5) but also stabilize the intermediate
(5) in the mechanism. From the observed increase in rate with
EWG one could expect a change in rate determining step from
electron transfer to nucleophilic attack, by changing the substit-
uents from electron donating to electron withdrawing groups.
Table 3. Effect of water on the oxidation of PSAAs with 1c
CH3CN–H2O (v/v)
104 k (sꢀ1
)
1
p-Cl PSAA
PSAA
p-Et PSAA
10–90
20–80
30–70
50–50
60–40
80–20
315 0.07
283 0.08
248 0.02
174 0.07
121 0.08
33.4 0.05
6.88 0.04 69.8 0.05
8.53 0.05 71.4 0.01
7.83 0.02 64.2 0.04
6.21 0.03 68.6 0.09
7.18 0.05 67.9 0.02
8.12 0.06 62.3 0.03
[PSAA] = 1 × 10ꢀ2 M, [p-Cl PSAA] = [p-Et PSAA] = 3 × 10ꢀ2 M,
[1c] = 3 × 10ꢀ4 M, [H+] = 0.5 M, μ = 0.6 M, T = 303 K.
J. Phys. Org. Chem. (2016)
Copyright © 2016 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

P. SUBRAMANIAM, J. JANET SYLVIA JABA ROSE AND R. JEEVI ESTHER RATHINAKUMARI
Equations 12 and 13 have been successfully applied to a vari-
Acknowledgements
ety of inorganic, organic and organometallic and biochemical ET
reactions that have a wide range of structural types including
hetero atom substituted aromatics. In the above equations, K12
and z are the equilibrium constant for the cross reaction and
collision frequency for the uncharged reactant molecules in
solution respectively. The value of z is the pre exponential
JJSR thanks the UGC, SERO, Hyderabad (ETFTNMS137), the Man-
agement of Nazareth Margoschis College and Manonmaniam
Sundaranar University for the award of a fellowship under FDP.
The authors gratefully thank the Management, Aditanar College
of Arts and Science, Tiruchendur for providing Laboratory facili-
ties to do the research.
factor which is often taken as 1 × 1011 ꢀ1sꢀ1. The value of
m
self-exchange rate (k11) of [Fe(NN)3]3+/[Fe(NN)3]2+ couple is
taken from the previous studies of Sutin and co-workers[63] as
3.3 × 108 Mꢀ1 sꢀ1. The value of K12 is calculated from the redox
potential of the couples [Fe(NN)3]3+ / [Fe(NN)3]2+ and iSO•+/iSO
using the following equations
REFERENCES
[1] S. C. Sahu, Environ. Carcino. & Ecotox. Revs. 2002, C 20, 61–76.
[2] R. S. Rivlin, Historical perspective on the use of garlic. J. Nutr. 2001,
131, 951S–954S.
[3] A. M. I. Jeyaseeli, S. Rajagopal, J. Mol. Cat. A: Chemical. 2009,
ꢀ
ꢁ
309, 103–110.
K12 ¼ exp ꢀΔGO =RT
(14)
[4] D. Thenraja, P. Subramaniam, C. Srinivasan, Tetrahedron 2002,
58, 4283–4290.
[5] R. Suthakaran, S. Rajagopal, C. Srinivasan, Tetrahedron 2001,
57, 1369–1374.
where
[6] K. J. Adaikalasamy, N. S. Venkataramanan, S. Rajagopal, Tetrahedron.
2003, 59, 3613–3619.
[7] D. L. Bai, X. C. Tang, X. C. He, Curr. Top. Med. Chem. 2000, 7, 355–374.
[8] A. B. V. Kiran Kumar, K. S. V. Krishna Rao, M. Subhosh Chandra, M. C.
S. Subha, J. Yong Lark Choi, Korean Soc. Appl. Biol. Chem. 2009,
52(1), 34–39.
ꢀ
ꢁ
ΔGo ¼ nF EoS•þ
ꢀ EoFe3þ
:
(15)
=Fe2þ
O=SO
[9] S. Fukuzumi, I. Nakanishi, K. Tanak, T. Suenobu, A. Tabard, R. Guilard,
E. V. Caemelbecke, K. M. Kadish, J. Am. Chem. Soc. 1999,
121, 785–790.
The value of k22 for iSO•+/iSO couple can be determined from
an iterative procedure, i.e. a value of k22 is guessed and plugged
into Eqn 12 to calculate f. A plot of log k12 – 0.5 (log k11 – log f12)
versus log K12 is made, and from mean least-square calculations,
the intercept and slope of such plots are determined. From the
intercept a new estimated value of k22 is obtained, and this is
then used to calculate a new log f. The entire iterative process
was repeated until successive estimates of k22 differed by
less than 10%. The final result gives k22 = 1.3 × 107 Mꢀ1 sꢀ1 for
Fe(III) complexes as oxidant. It is pertinent to point out
that similar treatment of the oxidation of organic sulfoxides with
[Fe(NN)3]3+ has given the same value of k22. The value of k22 is
then used to get the rate constant for ET (k12) from PSAA to
[10] I. Hamachi, S. Tsukiji, S. Shinkai, S. Oishi, J. Am. Chem. Soc. 1999, 121,
5500–5506.
[11] I. Batinic-Haberle, I. Spasojevic, P. Hambright, L. Benov, A. L. Crumbliss,
I. Fridovich, Inorg. Chem. 1999, 38, 4011–4022.
[12] J. Chen, Z. Luo, Z. Zhao, L. Xie, W. Zheng, T. Chen, Biomaterials 2015,
71, 168–177.
[13] P. Balakumar, S. Balakumar, P. Subramaniam, Reac. Kinet. Mech. Cat.
2012, 107, 253–261.
[14] S. Balakumar, P. Thanasekaran, E. Rajkumar, K. J. Adaikalasamy, S.
Rajagopal, R. Ramaraj, T. Rajendran, B. Manimaran, K.-L. Lu, Org.
Biomol.Chem. 2006, 4, 352–358.
[15] T. V. N. P. Sarathi, A. K. Kumar, K. K. Kishore, P. Vani, J. Chem. Sci.
2005, 117, 329–332.
[16] P. Vani, K. K. Kishore, R. Rambabu, L. S. A. Dikshitulu, Proc. Indian.
Acad. Sci. (Chem. Sci). 2001, 113, 351–359.
[17] S. Deepalakshmi, A. Sivalingam, T. Kannadasan, P. Subramaniam, P.
Sivakumar, S. T. Brahadeesh, Spectrochim. Acta, A: Mol. Biomol.
Spectrosc. 2014, 124, 315–321.
[18] P. Subramaniam, N. Thamil Selvi, Am. J. Anal. Chem. 2013, 4, 20–29.
[19] P. Subramaniam, N. Thamil Selvi, S. Sugirtha Devi, S. J. Korean Chem,
Soc. 2014, 58, 17–24.
[Fe(bpy)3]3+ and [Fe(phen)3]3+
.
The calculated values,
12.3 × 10ꢀ4 sꢀ1 and 19.4 × 10ꢀ4 sꢀ1 are in fair agreement with
the experimental values 3.43 × 10ꢀ4 sꢀ1 and 3.24 × 10ꢀ4 sꢀ1. In
order to account for the dynamics of ET reactions it is essential
to include the solvation energy in the calculation. Thus we pre-
sume that if the solvation energy had been included, the agree-
ment between the experimental and calculated values would
have been better. The fair agreement between the experimental
and calculated rate constant values confirms the involvement of
[20] P. Subramaniam, S. Sugirtha Devi, S. Anbarasan, J. Mol. Cat. A: Chem-
ical. 2014, 390, 159–168.
[21] P. Subramaniam, N. Thamil Selvi, J. Serb. Chem. Soc. 2015,
80, 1019–1034.
[22] S. Balakumar, P. Thanasekaran, S. Rajagopal, R. Ramaraj, Tetrahedron.
single electron transfer from PSAA to [Fe(NN)3]3+
.
1995, 51, 4801–4818.
[23] C. Srinivasan, P. Subramaniam, J. Chem. Soc.Perkin Trans. 1990,
2, 1061–1279.
CONCLUSIONS
[24] R. D. Gillard, Coord. Chem. Rev. 1975, 16, 67–94.
[25] S. Burchett, C. E. Meloan, J. Inorg. Nucl. Chem. 1972, 34, 1207–1213.
[26] R. Schmid, L. Han, Inorg. Chim. Acta. 1983, 69, 127–134.
[27] M. H. Hey, Mineral Mag. 1982, 46, 111–118.
[28] M. H. Hey, Mineral Mag. 1982, 46, 512–513.
[29] S. Premsingh, N. S. Venkataramanan, S. Rajagopal, S. P. Mirza,
M. Vairamani, P. Rao, K. Velavan, Inorg. Chem. 2004, 43, 5744–5753.
[30] E. Baciocchi, D. Intini, A. Piermattei, C. Roh, R. Ruzziconi, Gazz. Chim.
Ital. 1989, 119, 649.
[31] C. G. Swain, W. P. Langsdorf, J. Am. Chem. Soc. 1951, 73, 2813–2819.
[32] J. Arias, C. R. Newlands, M. M. Abu-Omar, Inorg. Chem. 2001,
40, 2185–2192.
[33] F. Ruff, A. Kucsman, J. Chem. Soc. Perkin Trans. 1975, 2, 509–519.
[34] F. Ruff, A. Komoto, N. Furukawa, S. Oae, Tetrahedron 1976,
32, 2763–2767.
The electron transfer between substituted PSAAs and iron(III)
polypyridyl complexes was studied in aqueous acetonitrile me-
dium. PSAAs were converted to diphenyl disulfones as products.
The effect of acid concentration, ionic strength and solvent
variation on the rate of ET was studied. The observed substituent
effect and non-linear Hammett behaviour with upward curvature
are explained based upon the change in the rate determining
step in the mechanism from electron transfer to nucleophilic
attack of water upon changing the substitutes in PSAA. A satis-
factory mechanism incorporating all the effects was proposed.
Marcus relation has been successfully applied.
wileyonlinelibrary.com/journal/poc
Copyright © 2016 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2016)

SHIFT IN MECHANISM FROM SINGLE ELECTRON TRANSFER TO NUCLEOPHILIC ATTACK
[35] D. Thenraja, P. Subramaninam, C. Srinivasan, J. Chem. Soc. Perkin
Trans. 2002, 2, 2125–2129.
[52] N. S. Venkataramanan, S. Premsingh, S. Rajagopal, K. Pitchumani,
J. Org. Chem. 2003, 68, 7460–7470.
[36] U. Miotti, G. Modena, L. Sedea, J. Chem. Soc. B. 1970, 802–805.
[37] K. Chowdhury, K. K. Banerji, J. Org. Chem. 1990, 55, 5391–5393.
[38] N. M. I. Alhaji, A. M. U. Mohideen, S. S. L. Mary, E-J. Chem. 2011,
8, 159–166.
[53] M. A. Mansour, Transition Met. Chem. 2002, 27, 818–821.
[54] a) M. Ganesan, V. K. Sivasubramanian, S. Rajagopal, R. Ramraj, Tetra-
hedron 2004, 60, 1921–1929; b) D. Thiruppathi, P. Karuppasamy, M.
Ganesan, V. K. Sivasubramanian, T. Rajendran, S. Rajagopal,
J. Photochem. Photobiol., A Chem. 2014, 295, 70–78.
[55] N. Sutin, B. M. Gordon, J. Am. Chem. Soc. 1961, 83, 70–73.
[56] M. Kmura, G. Wada, Inorg. Chem. 1978, 17, 2239–2242.
[57] V. B. Gawandi, H. Mohan, J. P. Mittal, J. Phys. Chem. A. 2000,
104, 11877–11884.
[39] K. Karunakaran, R. Gurumurthy, K. P. Elango, Indian J. Chem. 1997,
36A, 984–986.
[40] D. G. Lee, L. N. Congson, Can. J. Chem. 1990, 68, 1774–1779.
[41] A. Thirumoorthi, D. S. Bhuvaneshwari, K. P. Elango, Int. J. Chem. Kinet.
2010, 42, 159–167.
[42] G. F. Jeyanthi, K. P. Elango, Int. J. Chem. Kinet. 2003, 35, 154–158.
[43] H. M. Neu, T. Yang, R. A. Baglia, T. H. Yosca, M. T. Green, M. G.
Quesne, S. P. de Visser, D. P. Goldberg, J. Am. Chem. Soc. 2014,
136, 13845–13852.
[58] B. Enrico, M. Bietti, O. Lanzalunga, Acc. Chem. Res. 2000,
33, 243–251.
[59] C. Srinivasan, P. K. Ganesan, Indian J. Chem. 1989, 28B, 141–145.
[60] P. Filipiak, G. L. Hug, B. Marciniak, J. Photochem. Photobiol., A Chem.
2006, 177, 295–306.
[44] M. J. Zdilla, J. L. Dexheimer, M. M. Abu-Omar, J. Am. Chem. Soc. 2007,
[61] T. R. Green, J. H. Fellman, Adv. Exp. Med. Biol. 1994, 359, 19–29.
[62] a) R. A. Marcus, N. Sutin, Inorg. Chem. 1975, 14, 213–216;
b) R. A. Marcus, Angew. Chem. Int. Ed. 1993, 32, 1111–1121.
[63] G. W. Brown, N. Sutin, J. Am. Chem. Soc. 1979, 101, 883–892.
129, 11505–11511.
[45] R. T. Sabapathy Mohan, M. Gopalakrishnan, M. Sekar, Tetrahedron
1994, 50, 10933–10944.
[46] R. T. Sabapathy Mohan, M. Gopalakrishnan, M. Sekar, Tetrahedron
1994, 50, 10945–10954.
[47] D. G. Lee, K. C. Brown, J. Am. Chem. Soc. 1982, 104, 5076–5081.
[48] G. Licini, M. Bonchio, G. Modena, W. A. Nugent, Pure Appl. Chem.
1999, 71, 463–472.
[49] G. A. Petersson, Theor. Chem. Acc. 2000, 103, 190–195.
[50] O. Exner, Collect. Czech.Chem. Commun. 1964, 29, 1094–1113.
[51] K. G. Sekar, C. L. Edison raj, J. Chem. Pharm. Res. 2011, 3, 596–601.
SUPPORTING INFORMATION
Additional supporting information can be found in the online
version of this article at the publisher’s website.
J. Phys. Org. Chem. (2016)
Copyright © 2016 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc