Kinetic and redox characteristics of semireduced species derived from phenosafranine in homogeneous aqueous and sodium dodecyl sulfate micellar media

Article abstract of DOI:10.1002/kin.10024

One-electron reduction of phenosafranine (PS⁺, 3,7-diamino-5-phenylphenazinium chloride), a phenazinium dye, has been studied in homogeneous aqueous and sodium dodecyl sulfate (SDS) micellar media, using the pulse radiolysis technique. The various reducing radicals employed for the study in homogeneous aqueous medium were e_aq^-, H^., CO₂^.-, and isopropyl ketyl radicals (CH₃)₂^.COH. Semireduced species generated by these reactions have been characterized by their absorption spectra, decay kinetics, and pK_a. The one-electron reduction potential of PS⁺ was determined at pH 7 in homogeneous aqueous solution employing nitrobenzene (NB/NB^.-) as the standard couple. One-electron reduction in SDS micellar medium and a detailed spectrophotometric investigation of the parent dye in this surfactant system was carried out in order to understand the dye-surfactant interactions in the micellar and premicellar media.

Full text of DOI:10.1002/kin.10024

Kinetic and Redox
Characteristics of
Semireduced Species
Derived from
Phenosafranine in
Homogeneous Aqueous
and Sodium Dodecyl
Sulfate Micellar Media
V. B. GAWANDI, S. N. GUHA, H. MOHAN, J. P. MITTAL
Radiation Chemistry & Chemical Dynamics Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
Received 11 January 2001; accepted 2 August 2001
ABSTRACT: One-electron reduction of phenosafranine (PS+, 3,7-diamino-5-phenylphenazi-
nium chloride), a phenazinium dye, has been studied in homogeneous aqueous and sodium
dodecyl sulfate (SDS) micellar media, using the pulse radiolysis technique. The various reduc-
ing radicals employed for the study in homogeneous aqueous medium were eaq , H , CO2
,
and isopropyl ketyl radicals (CH3)2 COH. Semireduced species generated by these reactions
have been characterized by their absorption spectra, decay kinetics, and pKa. The one-electron
reduction potential of PS+ was determined at pH 7 in homogeneous aqueous solution employ-
ing nitrobenzene (NB/NB ) as the standard couple. One-electron reduction in SDS micellar
medium and a detailed spectrophotometric investigation of the parent dye in this surfactant
system was carried out in order to understand the dye–surfactant interactions in the micellar
C
and premicellar media. 2001 John Wiley & Sons, Inc. Int J Chem Kinet 34: 56–66, 2002
INTRODUCTION
Phenosafranine, a cationic dye (PS+, 3,7-diamino-5-
Correspondence to: J. P. Mittal; e-mail: mittaljp@magnum.barc.
phenylphenazinium chloride; Scheme 1) belonging to
phenazinium class, finds application in diverse areas
[1–3] wherein its redox chemistry plays an important
role. Redox reactions of this class of compounds are
ernet.in.
J. P. Mittal is also an Honorary Professor at the Jawaharlal Nehru
Center for Advanced Scientific Research, Bangalore, India.
c
2001 John Wiley & Sons, Inc.
DOI 10.1002/kin.10024

KINETIC AND REDOX CHARACTERISTICS OF SEMIREDUCED SPECIES
57
clean method where the concomitant oxidation prod-
uct is water, which is not expected to complicate either
the decay kinetics or the transient absorption spectrum
of the semireduced species. In their flash photolysis
experiments, Gopidas and Kamat [14] have generated
semireduced PS+ by quenching of the dye triplet by
amines in acetonitrile solution in which the transient
is expected to show a different absorption property as
compared to neat aqueous medium. In this paper ,we re-
port pulse radiolytic generation and characterization of
semireduced phenosafranine radical in homogeneous
aqueous medium. As practical systems thiazines and
phenazines suffer from very poor power conversion ef-
ficiency. One of the factors has been the slower dis-
charge of the photoproduced reduced dye species at
the photoanodes as compared to its homogeneous re-
combination in the bulk of the solution. In these sys-
tems, photoinduced reduction leads primarily to unsta-
ble semireduced dye species, which disproportionate to
give further reduced leuco dye and the parent molecule.
Any system parameters that can affect this dispropor-
tionation step can have considerable influence on the
photogalvanic effect particularly the power conversion
efficiency. The importance of micelles in controlling
the course of chemical reaction particularly those of
relevance to the problem of high energy conversion
has been recognized in the past [16]. The present study
was extended to find out if the fast disproportionation
of semireduced species can be slowed down in suitable
micellar medium. Since phenosafranine is a cationic
dye we have employed anionic sodium dodecyl sul-
phate (SDS) surfactants for detailed pulse radiolysis
study to see the behaviour of semireduced species in
this micellar medium. Before studying the redox reac-
tions of phenosafranine in this micellar medium, we
considered it worthwhile to investigate its location and
interaction in this surfactant system. Although some
work on the dye–surfactant interaction in premicellar
[17a] and micellar [18] media employing this dye has
been reported in the literature, no work on the redox re-
actions of PS+ is reported so far in this micellar system.
Further, we have used an improved iteration method
to compute the binding constant, and on the basis of
medium polarity effect, we have suggested the loca-
tion of the dye inside the SDS micelle.
Scheme 1
important in view of their usefulness as sensitizers in
photogalvanic cells for the conversion of light to elec-
trical energy [1]. Thus, for example, the phenazine
dye, Safranine T (SF+) has been considered to be a
well-known sensitizer and in order to explain the
photopotential of a photogalvanic cell employing the
SF+–EDTA system [1], the semireduced safranine rad-
ical was postulated as the electroactive species. For-
mation of semireduced species of phenazinium dyes
has been demonstrated in flash photolysis experiments
[4–7]. Also, in the case of thiazine dyes the forma-
tion of the semireduced species was observed [8–12]
in both flash photolysis and pulse radiolysis experi-
ments. Generally in the photoreduction of dyes an in-
termediate semireduced radical is formed through the
excited states of the dye in the presence of a suitable
reducing agent [6]. Chlorophyll-sensitized reduction
of safranine has been investigated in the past [3], in
connection with the elucidation of the mechanism of
photosynthesis in the plant. Formation of the excited
states species of a simple member of this class vizPS+
has been demonstrated in laser flash photolysis experi-
ments [13,14]. Similar studies were reported with other
members (e.g. Safranine T and Safranine O) using flash
photolysis technique [4–7]. The semireduced safranine
species cannot exclusively be produced by flash pho-
tolysis of the parent dye molecule without the presence
of dye triplet and also the radicals derived from exter-
nally added reducing agents such as EDTA, ascorbic
acid, etc [6]. These radicals and the triplet present in
the systemmight possiblyinterfere withthe absorbance
of the semireduced species. In the case, where ground
state molecule itself has been found to quench its triplet
reductively and no external reductant is required to
generate semireduced species, a concomitant semiox-
idised radical is invariably formed interfering with the
absorbance of semireduced species and also with its
decay kinetics. Baumgartner et al. [6] have in fact ex-
perienced these difficulties and discussed the uncer-
tainties of decay kinetics arising because of the pres-
ence of externally added reducing agent. On the other
hand, generation of semireduced species by the reac-
tion of hydrated electron produced on pulse radiolysis
of an aqueous solution of a suitable solution [15] is a
EXPERIMENTAL
Phenosafranine (PS+, 3,7-diamino-5-phenylphenaz-
inium chloride) was a Sigma product and was recrys-
tallized twice from water–ethanol mixture before use.
The pH dependence of the absorption spectrum of PS+
showed that there is no pKa of the dye in the pH

58
GAWANDI ET AL.
range of 2–11. Sodium dodecyl sulfate (SDS) was ex-
trapure obtained from Sigma and was used as such. All
other chemicals were of the highest available purity.
Solutions were prepared in nano pure water obtained
by passing demineralized water through a Barnsted
Nanopure water system. Nitrogen and N2O gases used
for purging the solutions were Iolar grade from BOC
India Ltd, Mumbai. UV–Visible double beam record-
ing spectrophotometer (Hitachi model 330) was used
to record absorption spectra. The pHs of the solu-
tions were adjusted using H2SO4, Na2HPO4 2H2O,
KH2PO4, and NaOH.
ficients, formation rate constants and redox potential
values absorbed doses employed were 8 Gy.
The extinction coefficients of the product tran-
sient species (εT) were evaluated at any wavelength
␭ employing corrections for the parent dye absorption
using Eq. (5).
[AT]. G
. ε
(SCN)2
(SCN)2
εT = εP
+
(5)
GT
A
(SCN)2
1
where εp is the extinction coefficient in dm3 mol
cm 1 for the parent dye molecule at wavelength , GT
is the G-value of the product transient species. AT and
For studying the reactions of eaq and H atom, tert-
butyl alcohol ( 0.2 mol dm 3) was used as the OH
radical scavenger as shown in Eq. (1).
A(SCN) are the observed absorbances of the transient
2
dye at and of the (SCN)2 radical at 500 nm respec-
tively under isodose conditions.
(CH3)3COH + OH → CH2(CH3)2COH + H2O
(1)
RESULTS AND DISCUSSION
A solution containing 2-propanol (1 mol dm 3) was
used at all the pHs to study the reactions of isopropyl
ketyl radicals, which were generated as follows.
One-Electron Reduction Studies
in Homogeneous Aqueous Medium
One-electron reduction of PS+ was carried out in aque-
ous solution employing eaq , H , (CH3)2 COH, and
CO2 radicals, in appropriate solutions using the tech-
nique of nano-second pulse radiolysis.
(CH3)2CHOH + OH( H)
CH3 CH2CHOH + H2O(H2) (2a)
I
␤-hydroxy radical( 15%)
Reactions of Hydrated Electron.eaq . The absorption
(CH3)2 COH + H2O(H2)
␣-hydroxy radical ( 85%)
(2b)
I
of eaq , monitored at its
tert-butyl alcohol solution, was found to decay faster
= 720 nm in N2 saturated
max
5
in the presence of PS+(8 10 mol dm 3, Fig. 1a
Sodium formate was added to the N2O saturated solu-
tions to generate CO2 radicals according to Eqs. (3)
and (4)
eaq + N2O → OH + OH + N2 (3)
HCOO + OH(H ) → CO2 + H2O(H2) (4)
5
3
All the above solutions contained 8 10 mol dm
PS+ and were taken for the pulse radiolysis studies.
Details of the pulse radiolysis set up have been
described elsewhere [8]. Single pulses of 7 MeV elec-
trons of 50 ns duration with typical doses of 10 Gy
(1 Gy = 1 J kg 1) were used.
The absorbed dose was determined by monitoring
the transient absorbance of (SCN)2 formed by puls-
3
ing aerated solutions of 0.01 mol dm KSCN. G ε
1
for (SCN)2 = 21,520 dm3 mol cm 1/100 eV at
500 nm [19], where G is radiation chemical yield
and is defined as the number of molecules formed
per 100 eV of energy absorbed (G = 1 corresponds
to 0.1036 ␮mol J 1) and ε is the molar extinction co-
efficient at 500 nm. For evaluation of extinction coef-
FIGURE 1 Kinetic traces at 720 nm for a N2-bubbled so-
lution (pH = 10.5) containing tert-butyl alcohol (0.05 mol
3
5
dm 3) and PS+ (a) 0.0 mol dm and (b) 8.0 10 mol
3
dm
.

KINETIC AND REDOX CHARACTERISTICS OF SEMIREDUCED SPECIES
59
and 1b.). Also considerable bleaching of the dye ab-
sorption band at 520 nm was observed (see parent dye
absorption spectrum in Fig. 6a). Both these observa-
tions clearly indicate the high reactivity of eaq with
the dye. The transient absorption spectrum recorded in
the 300–800 nm region is shown in Fig. 2a. Two well
defined absorption bands with max at 410 and 650 nm
were observed. Both the bands were found to grow
with time and reached maximum in 4 ␮s. The de-
cay of the transient absorbance in the 410 and 650 nm
bands followed second order kinetics with 2k close to
each other and the average value being 9.6 108 dm3
1
mol
s
1. These two bands can therefore be inferred
to be due to a single transient species i.e. semireduced
PS+ (PS , Scheme 2) formed from the one-electron
reduction of the dye. The formation traces were found
to fit psuedo-first-order kinetics with PS+ concentra-
tion. The bimolecular rate constant for the reaction
of eaq with PS+ evaluated from these traces at pH 7
FIGURE 2 Transient optical absorption spectrum (end of
50 ns electron pulse) obtained on reaction of eaq with PS+
at pH (a) 7 and (b) 9.6. Inset shows variation of absorbance
vs. pH at wavelength (c) 410 nm and (d) 650 nm. Dose =
11.0 Gy/pulse.
1
was 2.1 1010 dm3 mol
s
1. The molar extinction
coefficients evaluated using the relationship (Eq. (5))
1
1
were 12,600 100 dm3 mol cm and 11,300
1
1
50 dm3 mol cm at 410 and 650 nm, respectively.
Thenatureoftheabsorptionspectrumrecordedatlower
pH 5 was exactly similar to the one observed at neu-
tral pH. However, the transient spectrum observed at
alkaline pH 9.6 was quite different and indicated a
distinct pH effect (Fig. 2b). Thus, the 410 nm band ex-
hibited shift in the absorption maxima and the 650 nm
band diminished considerably in intensity showing a
big change in extinction coefficient. This effect can be
attributed to the existence of a pKa of the semireduced
species. Plots of absorbance (1A) vs pH recorded at
both the wavelengths 410 and 650 nm are shown in
inset of Fig. 2. The inflexion points of both the curves
revealed pKa of 8.3 for the semireduced PS+ species.
Since the dye exists as PS+ over the entire pH, its re-
duction with eaq can be represented as
eaq + PS+ → PS
(6)
and the protolytic equilibrium of the semireduced
species can be represented as shown in Eq. (7) and
Scheme 2.
pKa=8.3
+
PSH + + H2O (
PS + H3O+
(7)
Reactions of Isopropyl Ketyl (CH3)2 COH Radicals.
(CH3)2 COH is well-known reducing radical and are
known to react with many dyes by electron transfer
process giving rise to semireduced species. The re-
action of this radical was studied in N2O-saturated
5
3
aqueous solution containing 8 10 mol dm PS+
3
and 1 mol dm 2-propanol at pH 7. Electron pulse
irradiation of this solution gave rise to transient ab-
sorption spectrum with two well defined bands having
at 410 and 650 nm identical to the one obtained
max
by eaq reaction at this pH. The absorbances at these
wavelengths reached maxima at 11 ␮s and from
the formation traces for different PS+ concentrations
the bimolecular rate constant for the reaction of iso-
propyl ketyl radical with PS+ was computed to be 3.3
1
109 dm3 mol
s
1. The transient decayed by second-
order kinetics with 2k value comparing closely with
that obtained in tert-butyl alcohol solution. From the
Scheme 2

60
GAWANDI ET AL.
PS+ by electron transfer and gives rise to semireduced
species. The transient semireduced PS+ yield was al-
most 100% and the extinction coefficient was same
in this case also. The rate constant for the reaction of
spectral similarity and also the closeness of decay
kinetics in the two solutions, it is inferred that
(CH3)2 COH also brings about one-electron reduction
ofPS+ leadingtotheformationofsemireducedspecies.
The reaction of (CH3)2 COH was studied over entire
pH from 2 to 11, and was observed that while the
transient absorption spectrum at lower pH 2.5 re-
mained the same as observed in pH 7, the nature of
the spectrum observed at pH 10.0 was considerably
different, and was exactly similar to the one observed
in the case of eaq reaction at the higher pH. It has
been reported in the past [20] that among the radicals
formed by reactions (2a) and (2b) 15% are unreac-
tive ␤-hydroxy alkyl radicals (CH3) CH2 COH and the
rest 85% are ␣-hydroxy alkyl radical (CH3)2 COH
which are the strong reducing species. In computing
the yield of semireduced PS+, the G-value has been
assumed to be 5.1 i.e. 15% lower than the total radical
yield with G-value of 6.0. With this assumption and
the corrected band intensity, we computed the yield of
semireduced PS+ to be same as obtained in the reac-
tion of eaq , with experimental uncertainty of 5%.
It may be noted here that although (CH3)2 COH rad-
icals are reducing in nature and bring about electron
transfer reactions, they are known to undergo addition
as well as abstraction reactions. Since in the present
case isopropyl ketyl radical gives rise to 100% yield of
semireduced species, it is apparent that the other modes
of reactions such as abstraction and additions are totally
absent. In this system also we attempted to determine
pKa using both the bands over the entire pH range from
pH 2 to 11. The advantage in this case, in addition to a
high G-value, is that a single solution can be employed
over the entire pH range. In order to determine pKa of
semireduced PS+, a single solution consisting of N2O
CO2 with PS+ as evaluated from formation traces
1
was 4.1 109 dm3 mol 1 s
.
Reaction of H atom. The spectrum of the transient
species formed by reaction of H. atoms at pH 2.6
is shown in Fig. 3a. The solution used for the reac-
tion was N2-saturated 0.5 mol dm 3 tert-butyl alcohol
5
3
containing 8 10 mol dm of PS+. For compari-
son, the normalized spectrum (normalized to the same
dose and G-value) of the semireduced species, formed
byreactionofthe(CH3)2 COHradicalatthispH, isalso
given in Fig. 3b. It is seen that the nature of the 650 nm
band is the same except that its intensity is lower in
the case of H -atom reaction products. The reducing
ability of of H atom is expected to be higher than that
of (CH3)2 COH on the basis of the one-electron re-
duction potentials of these two species. Moreover, the
rate constant for the reaction of H -atom with PS+, de-
termined from the pseudo-first order rate for build-up
of transient absorption was found to be higher (4.5
1
109 dm3 mol
s
1) than that for reaction of (CH3)2
COH with PS+ (3.3
109 dm3 mol
s
1). There-
1
fore, the decrease in the intensity is obviously not
due to inefficiency of the hydrogen atom reaction with
PS+. Since H atoms are known to react by different
modes, we infer that H atoms formed in the system
react to give other species in addition to semireduced
PS+. Evidence for the formation of such species was
obtained from comparison of the shorter wavelength
band in the spectra of Fig. 3. It is seen that there
is an appreciable shift of the 410-nm band and also
a change in intensity. In the resulting spectrum ob-
tained by subtracting the absorbance of the two spec-
saturated 8 10 mol dm PS+ and 1 mol dm 2-
propanol was employed throughout the pH study. The
optical density of transient measured at 410 and 650 nm
as a function of pH gave curves (figure not shown) with
inflexion points corresponding to pKa of 8.3 same as in
thecaseofeaq . ThespectrarecordedatpH2.5and10.0
may be assigned to the two different forms of semire-
duced species PSH + and PS respectively as shown in
Scheme 2.
5
3
3
tra, we clearly see a new band with
at 370 nm
max
(Fig. 3c). This band could be attributed to the new
species formed by H -atom reaction. In the pulse ra-
diolysis work on nicotinic acid Solar et al. [21] ob-
served different modes of H-atom reaction includ-
ing H-adduct with different absorption band. In the
case of thionine such a new band has been assigned
to H-adduct [12]. It is reasonable to assume in the
Reaction of CO2 . Reaction of CO2 radical in N2O-
present case that the new band with
at 370 nm
max
5
saturated aqueous solutions containing 8 10 mol
is also due to H-adduct. It is supported by the obser-
vation that the decay of the transients formed by iso-
propyl ketyl radical and H -atom reactions at 370 nm
are entirely different (inset A of Fig. 3). The decay
of 650 nm band generated by H -atom reaction fol-
lowed second order kinetics with 2k/εl close to that
observed for semireduced dye species. The nature of
the decay traces monitored at 650 nm in the case of
3
3
dm PS+ and 0.1 mol dm sodium formate were
studied at pH 2.8 and 9.6. The radical absorption
spectra were similar to those observed in eaq and
(CH3)2 COH reactions at respective pHs. The transient
decay in this case also followed a second order kinet-
ics with 2k almost same as in the case of eaq and
(CH3)2 COH. It is inferred that CO2 also reacts with

KINETIC AND REDOX CHARACTERISTICS OF SEMIREDUCED SPECIES
61
FIGURE 3 Absorption spectra (end of 50 ns electron pulse) of transient semireduced phenosafranine species formed by
reaction of PS+ (a) with H atoms at pH 2.6, (b) with isopropyl ketyl radicals at pH 2.5, and (c) difference in the absorption
spectra of (b) and (a). Inset (A) shows absorption time profiles of radical decay formed on reaction with isopropyl ketyl radical
(d) and with H -atom (e) at 380 nm. Inset (B) shows absorption time profiles of radical decay formed on reaction with isopropyl
ketyl radical (f) and with H -atom (g) at 650 nm.
isopropyl ketyl radical and H -atom reactions are seen
to be similar in inset B of Fig. 3. It is therefore in-
ferred that the band observed at 650 nm in the H -atom
reaction is exclusively due to semireduced species.
Since the isopropyl ketyl radical reacts with PS+ by
electron transfer with 100% efficiency, and using the
absorbances of 650 nm band, we compute the yield
of semireduced species formed on reaction with H
atom to be 58% and the remaining 42% is attributed
as due to H-adduct. The following Scheme 3 suggests
the two different modes of H atom reactions with
PS+.
One-Electron Reduction Potential
Of Phenosafranine (PS+)
Among the various standard couples (Duroquinone,
DQ/DQ ; Anthraquinone disulfonate, AQDS/
AQDS ; Nitrobenzene, NB/NB ) employed to de-
termine one-electron reduction potential of PS+, we
have found that the NB/NB couple establishes an
equilibrium as shown below:
K
PS+(H+) + NB
PSH + + NB
(8)
where K is equilibrium constant and was evaluated
using both kinetic and yield measurement methods.
Kinetic Method. In this method equilibrium constant
K is related to kf and kb as K = kf/kb, where kf and kb
are the rate constants for the forward and the back reac-
tions respectively. The observed rate constant kobs de-
termined at different NB concentrations can be shown
to follow the relationship.
kobsd = kf[PS+] + kb[NB]
(9)
kobsd
kf[PS+]
or
=
+ kb
(10)
[NB]
[NB]
On plotting kobsd/[NB] vs [PS+]/[NB] we computed kf
from the slope and kb from the intercept of the plot of
Scheme 3

62
GAWANDI ET AL.
Fig. 4A and finally we calculated equilibrium constant
K as kf/kb to be 293. By substituting this value of K
in the Nernst’s Eq. (11)
length (1 cm in the present case). Plot of 1/1A vs.
[PS+]/[NB] would be linear with slope/intercept = K.
From the linear plot [Fig. 4B], slope and intercept was
computed to be 41743 and 140, respectively, and the K
was evaluated to be 298, close to the value obtained in
the previous case. The one-electron reduction potential
of PS+ at pH 7 was then computed by substituting the
above value of K in Eq. (11) and was found to be close
to the one obtained by kinetic method.
E1(PS+/PS ) = Eo(NB/NB ) + 59.16 logK (11)
the one-electron reduction potential E1 (mV) of PS+
was computed to be 340 mV vs. NHE (Normal
Hydrogen Electrode) at pH 7 by using Eo(NB/NB ) =
486 mV vs. NHE [22].
Optical Absorption Studies in Presence
of Anionic Surfactant SDS
Yield Measurement Methods. The equilibrium con-
stant K can be shown to be related to 1A as
The effect of surfactant SDS on the absorption prop-
erties of the dye can be seen in Fig. 5. It is seen that
as the SDS concentration is increased the absorbance
of the dye at 520 nm decreases until around 0.8
1
K[PS+]
∈ l[R].[NB]. ∈ l[R]
1
=
+
(12)
1A
3
3
˜
10 moldm and on further increase of SDS con-
where [R] is the total radical concentration (at given
dose and G-value), ε is the molar extinction coeffi-
cient of semireduced species and l is the optical path
centration the absorption rises between 0.8 10 3 and
2
3
10 mol dm 3. This rise of dye absorbance in
3
the region 0.8 10
3
10 2 mol dm 3 SDS is re-
lated to the formation of surfactant micelles in which
the dye is incorporated. From the inflexion point of this
curve the CMC (critical micelle concentration) of SDS
3
is evaluated to be 2.6 10 mol dm 3. The CMC of
SDS evaluated from the inflexion point of curve (Fig. 5)
is lower than the ones obtained by light scattering and
conductivity measurements. Lowering of CMC values
in dye–surfactant systems has also been observed in
the past by Mukherjee and Mysels [23]. Even in the
case of thiazine dye (i.e. the thionine and SDS sys-
tem) also the CMC of SDS evaluated from the inflexion
point of similar curve was lower than the ones obtained
by other methods. It is reported in the work [24] that
the observed CMC values increased and approached
the latter with increasing thionine concentrations. The
FIGURE 4 Plots of (A) kobs/[NB] vs. [PS+]/[NB] at pH 7
and (B) 1/1A650 nm vs. [PS+]/[NB] in N2-saturated 2 mol
dm 3 tert-butyl alcohol solution containing varying concen-
trations of NB.
FIGURE 5 Effect of SDS on the absorbance of PS+
5
(1.8 10 mol dm 3) solutions at 520 nm.

KINETIC AND REDOX CHARACTERISTICS OF SEMIREDUCED SPECIES
63
present CMC value in fact agree closely with the one
reported for the thionine–SDS system for a comparable
dye concentration.
Interactions with Anionic Micelles. As stated before
PS+ is present as complex (PSS) in the premicellar
region of the SDS. As the surfactant concentration in-
creases beyond the CMC the main dye band appears
and the complex band at 500 nm disappears. This indi-
cates that the dye surfactant complex is unstable in the
micellar region, where the surfactant molecules tend to
aggregate to form the micelle. The equilibrium for the
incorportion of the dye into the micelles (M) can be
represented as
The decrease in absorbance of the dye at 520 nm,
below the CMC (premicellar region), can be attributed
to the formation of a dye–surfactant complex. In neat
aqueous solution the dye (PS+) exhibits a pronounced
band with max = 520 nm (ε = 36,000 100 dm3
1
mol cm 1) (Fig. 6a). At SDS concentrations in the
premicellar region the absorbance of the 520 nm dye
band decreases and a new band with
= 500 nm
max
KB
4
appears, which becomes prominent at 8 10 mol
PSS + M
PSM
(14)
3
dm SDS (Fig. 6b). Although the new band appears
PSS, M, and PSM represent dye–surfactant complex,
micelle, and micellized dye species respectively. The
measured absorbance (A) of the dye at different SDS
concentrations can be used to compute the dye–micelle
binding constant KB. Thus, Aobs = fc [PS+]Tεc l + fm
[PS+]T εm l, where εc and εm are molar absorption
coefficients of complexed and micellised dye at a given
wavelength, respectively, fc and fm are the fractions of
the complexed and micellized dye respectively.
close to the dye band, the nature of the two spectra are
quite different. The band at 500 nm in the premicellar
region is broader and the intensity of the absorption
band is reduced considerably for the same dye concen-
tration. Similar changes in absorption spectra of PS+
in the presence of oppositely charged SDS molecules
below the CMC have been reported in the past [17a]
and have been attributed to the formation of a dye–
surfactant complex. The authors have determined the
equilibrium constant for interaction of PS+ with S
(SDS) in the premicellar region and concluded that the
hydrophobicity of SDS plays an important role in the
ion-pair formation. In their review on dye–surfactant
interactions, Diaz Garcia and Sanz-Medel [17b] also
suggested the existence of ion-association complexes
formed at below CMC between ionic surfactants and
dyes of opposite charge. In the present case, we infer
that the dye–surfactant interaction leads to the forma-
tion of ion-pair complex of the following type:
[M] is given by [M] = [M]T – fm [PS+]T, where
[PS+]T and [M]T are the total dye and micelle con-
centrations respectively. [M]T is given by
([S]T CMC)
[M]T =
(15)
N
where N is the aggregation number and is defined as
the number of surfactant molecules involved in the for-
mation of micelle. For our calculation N for SDS has
been taken to be 62 [25]. The binding constant, KB,
can be shown to be related to measured absorbance of
the dye as
PS+ + S
PSS
(13)
1
1
{A/[PS+]Tl εc} = (εm εc) + {(εm εc)
KB([M]T fm[PS+]T)}
1
(16)
where l is the optical path length, and a plot
of {A/[PS+]Tl εc}
vs. {([M]T fm[PS+]T)}
1
1
should be linear with binding constant KB(KB = in-
tercept/slope). As fm is not known, an initial plot was
drawn with fm = 0 (Fig. 7a). The approximate value
of KB obtained was used to get the value of fm from
Eq. (17), which is quadratic in fm.
KB[PS+]T fm2 + { (KB[M]T + KB[PS+]T + 1)} fm
+ KB[M]T = 0
(17)
FIGURE6 Absorptionspectra(endof50nselectronpulse)
of PS+ (1.8 10 5 mol dm 3) (a) in neat aqueous medium,
4
(b) in premicellar (SDS = 8 10 mol dm 3), and (c) in
From the above quadratic equation the root with a value
<1 is taken, as fm is a fraction. A more accurate plot is
micellar (SDS = 0.04 mol dm 3) region.

64
GAWANDI ET AL.
from high to low polar solvents the dye band inten-
sity increases with progressive red shift of the band.
The change in extinction coefficient of PS+ with the
medium polarity and also the dielectric constant gave
a linear plot (Fig. 8A and 8B). From the change in ex-
tinction coefficient in SDS micelle, and using the above
plot, the dielectric constant experienced by PS+ was
read off and the value was estimated to be 55. Similarly,
the polarities of the environment of PS+ in this micelle
was inferred form the above plot to correspond to ET
(solvent polarity parameter) value [26] lying between
that of water and glycerol [Fig. 8A]. Similarly, the red
shift of the PS+ absorption maximum also shows a cor-
relation with dielectric constant or the solvent polarity
parameter, but since the shifts are rather small, accurate
evaluation would be difficult and hence no attempt has
been made to estimate using this property.
FIGURE 7 Plot of {A/[PS+]Tl εc}
vs {([M]T
1
fm[PS+]T)} with (a) fm = 0 and (b) fm = 0.545.
1
now constructed using these fm values and the succes-
sive approximation procedure is repeated until KB and
fm become invariant. The binding constant KB so eval-
1
uated (Fig. 7b) was found to be 261 20 dm3 mol
.
The dye band at 520 nm was found to be red shifted by
11 nm with increase in the extinction coefficient of
the dye in micellar medium. It may be noted that our
value of KB is lower than that reported by Bhat et al.
at room temperature [17a]. In their method for evalu-
ating KB, the authors have considered the equilibrium
[18] involving free dye rather than dye–surfactant com-
plex in the premicellar region, although they have sug-
gested [17a] formation of such a complex. In the itera-
tion method adopted by us for computation of KB, we
have considered this aspect in the equilibrium (Eq. 14).
We feel that a lower value of KB in fact is in agreement
with our observation that the CMC curve (Fig. 4) is not
sharp but is spread over a wide concentration range of
SDS to reach the plateau.
Medium Polarity Effect. As mentioned earlier, there is
an appreciable red shift of the dye band in the anionic
micellar media followed by the increase in extinction
3
coefficient, for example, in 0.04 mol dm SDS the
of the dye band is red shifted by 11nm and the
max
extinction coefficient in this micellar medium is about
9% higher as compared to the homogeneous aqueous
medium. The red shift and increase in extinction coef-
ficient both indicate a decrease in the polarity or the di-
electric constant of the medium surrounding the probe
molecule. Thus, for example, in water–alcohol mix-
tures of different dielectric constants or in the presence
of solvents of different polarity, similar changes in ab-
sorption properties were also observed [24]. It is seen
that as the dielectric constant of the solution in water–
alcohol mixture decreases or the medium is changed
FIGURE 8 Relative increase in the extinction coefficient
(1ε) of PS+ (8 10 5 mol dm 3) in ethanol–water mixtures
of different compositions expressed in V % of ethanol.(A) and
the dielectric constant of different solvents.

KINETIC AND REDOX CHARACTERISTICS OF SEMIREDUCED SPECIES
65
explained assuming a finite concentration of the solute
in the bulk aqueous phase. Under the conditions of our
present work, SDS undergoes micelle formation above
One-Electron Reduction by eaq in Micellar Medium.
The transient absorption spectrum of the semireduced
PS+ recorded in SDS micellar medium is shown in
Fig. 9b. For a ready comparison absorption spectrum
of the transient semireduced species recorded in neat
aqueous solution is also shown in the Fig. 9a. It is seen
that the transient band is red shifted by 30 nm indicat-
ing incorporation of the transient into the micelle. The
rate constant for the reaction of eaq with PS+, which
3
3
about 2.6 10 mol dm concentration. In the mi-
cellar medium at surfactant concentration 0.04 mol
dm 3, where most of the dye is incorporated into the
micelle, a small but finite dye concentration would ex-
ist in the bulk aqueous phase as the complex (Eq. (13)).
Because of the existence of an equilibrium (Eq. (13)),
it is expected that both free and complexed dye species
would be present in the bulk aqueous phase and would
undergo fast reaction with eaq . However, reaction of
eaq with the micellized dye is to be considered to ac-
count for the slowness of the reaction. The observed
rate constant therefore would reflect on the reactions
of eaq with the bulk as well as micellised dye species.
As regard to the decay of the semireduced PS+, the
2k/ε l value observed in the homogeneous aqueous
medium was as high as 8.5 104 s 1, but the decay
in SDS micellar system was found to be almost ar-
rested at least in the millisecond time scale (inset of
Fig. 9). Because of the low absorption cross section
of the transients, it is difficult to carry out the experi-
ments at still low doses because of poor signal-to-noise
ratio. The absorption–time profile of the transient in the
1
1
was 2.1 1010 dm3 mol
s
in homogeneous aque-
ous medium decreased to a value of 9.4 109 dm3
mol
s
1, assuming the presence of entire dye in the
1
micelle. In the case of SDS, the negatively charged
micelle surface hinders like charged eaq to approach
because of electrostatic repulsion and therefore reac-
tion of eaq with the micellised PS+ is expected to be
slow. The rate constant for the reaction of eaq with so-
lute like pyrene, in similar micellar systems, has been
found to be lower by several orders of magnitude (from
1
1010 to less than 107 dm3 mol
s
1). But in this sys-
tem surprisingly we found that the rate constant is only
nominally lower and cannot be explicable on the basis
of the extent of micellisation of the dye into the SDS
micelles [27]. In the case of a phenothiazine dye Safra-
nine T similar behavior was observed [28], which was
FIGURE 9 Transient absorption spectra obtained on pulse radiolysis of N2-saturated aqueous (pH = 7) solution of PS+
(8 10 5 mol dm 3) containing tert-butyl alcohol (0.3 mol dm 3) in absence of SDS (a) (9 ␮s after the pulse) and in presence
of SDS (0.03 mol dm 3) (b) (21 ␮s after the pulse). Inset show absorption time profiles of radical decay in absence (c) and
presence of SDS (d) at 650 nm.

66
GAWANDI ET AL.
presence and absence of SDS are shown in the inset of
Fig. 9.
9. Mahadevan, J.; Guha, S. N.; Kishore, K.; Moorthy, P. N.
Proc Indian Acad Sci 1989, 101,43.
10. Kishore, K.; Guha, S. N.; Mahadevan, J.; Moorthy, P. N.;
Mittal, J. P. Radiat Phys Chem 1989, 34, 721.
11. Kayser, R. H.; Young, R. H. Photochem Photobiol 1976,
24, 395.
12. Solar, S.; Solar, W.; Getoff, N. Radiat Phys Chem 1982,
20, 165.
13. Jockusch, S.; Timpe, H. J.; Schnabel, W.; Turro, N. J. J
Phys Chem A 1997, 101, 440.
14. Gopidas, K. R.; Kamat, P. V. J Photochem Photobiol A
1989, 48, 291.
15. Guha, S. N.; Mittal, J. P. J Chem Soc, Faraday Trans
1997, 93(20), 3647.
16. Kiwi, J.; Kalyanasundaram; Gratzel, M. Struct Bonding
1982, 49, 37.
17. (a) Dutta, R. K.; Bhat, S. N. Bull Chem Soc Jpn 1992, 65,
1089; (b) Diaz Garcia, M. E.; Sanz-Medel, A. Talanta
1986, 33(3), 255.
CONCLUSION
The reactivity of phenosafranine PS+ dye towards
pulse radiolytically generated reducing radicals is very
high and the rate constant is close to the diffusion con-
trolled value. of H atoms react with PS+ by both elec-
tron transfer and addition.
The present study clearly demonstrates that the
spectral properties of PS+, a cationic dye are affected in
anionic micellar media. From the data of solvent effect
on PS+, it isevident that environment aroundPS+ inthe
anionic micelles of SDS is highly polar. Electrostatic
attraction between the PS+ dye (cationic dye) and SDS
(anionic micelle) would favour location of PS+ close
to the head groups. The results of the pulse radioly-
sis experiments on one-electron reduction in micellar
system clearly indicate that the decay kinetics of the
semireduced PS+ can be profoundly altered, leading to
the stabilization of transient in the micelles.
18. Dutta, R. K.; Bhat, S. N. Colloids Surf A 1996, 106, 127.
19. Fielden, E. M. In The Study of Fast Processes and Tran-
sient Species by Electron Pulse Radiolysis,.. Baxendale,
J. H.; Busi, F. (Eds.); D. Riedel: Dordrecht, Holland,
1982; p. 49.
20. Asmus, K. D.; Mackel, H.; Henglein, A. J Phys Chem
1969, 73, 2573.
21. Solar, S.; Solar, W.; Getoff, N.; Holcman, J.; Sehested,
K. Radiat Phys Chem 1988, 32, 585.
BIBLIOGRAPHY
1. Kaneko, M.; Yamada, A. J Phys Chem 1977, 81,
1213.
2. Jana, A. K.; Roy, S.; Bhowmik, B. B. Energy (Oxford)
1988, 13, 161.
3. Krasnovskii, A. A.; Drozdova, N. N. Biokhimiya 1961,
26, 859.
4. Gopidas, K. R.; Kamat, P. V. Langmuir 1989, 5, 22.
5. Chibisov, A. K.; Karyakin, A. V.; Skortsov, B. V.;
Shvindt, N. V. Khim Vys Energy 1968, 2, 3.
6. Baumgartner, C. E.; Richtol, H. H.; Aikens, D. A. Pho-
tochem Photobiol 1981, 34, 17.
7. Neumann, M. G. J Photochem 1986, 32, 379.
8. Guha, S. N.; Moorthy, P. N.; Kishore, K.; Naik, D. B.;
Rao, K. N. Proc Indian Acad Sci 1987, 99, 261.
22. Meisel, D.; Neta, P. J Am Chem Soc 1975, 97, 5198.
23. Mukerjee, P.; Mysels, K. J. J Am Chem Soc 1955, 77,
2937.
24. Guha, S. N.; Moorthy, P. N.; Rao, K. N. Proc Indian
Acad Sci 1982, 91, 73.
25. Fendler, J. H., Fendler, E. J. In Catalysis in Micellar and
Macromolecular Systems, Academic Press: New York,
1975.
26. Reichardt, C. Angew Chem, Int Ed Engl 1965, 4, 29.
27. Thomas, J. K. In Radiation Chemistry: Principles and
Applications; Farhataziz; Rodgers, M. A. J. (Eds.); VCH
Publishers: New York, 1987; p. 377.
28. Guha, S. N.; Mittal, J. P. Proc Indian Acad Sci (Chem
Sci) 1992, 104, 497.