Studies of solvation of ketocyanine dyes in homogeneous and heterogeneous media by UV/Vis spectroscopic method

Article abstract of DOI:10.1016/S1386-1425(00)00321-8

Solvation characteristics of ketocyanine dyes (I-VI) have been investigated in pure solvents and heterogeneous media by absorption and fluorescence studies. The dyes are good reporters of solvent polarity. In protic solvents they exist as equilibrium mixt

Full text of DOI:10.1016/S1386-1425(00)00321-8

Spectrochimica Acta Part A 56 (2000) 2763–2773
www.elsevier.nl/locate/saa
Studies of solvation of ketocyanine dyes in homogeneous
and heterogeneous media by UV/Vis spectroscopic method
Parimal Kumar Das, Ramkrishna Pramanik, Debashis Banerjee 1,
Sanjib Bagchi *
Department of Chemistry, Burdwan Uni6ersity, Burdwan 713 104, India
Received 15 May 2000; accepted 25 May 2000
Abstract
Solvation characteristics of ketocyanine dyes (I–VI) have been investigated in pure solvents and heterogeneous
media by absorption and fluorescence studies. The dyes are good reporters of solvent polarity. In protic solvents they
exist as equilibrium mixtures of bare and hydrogen-bonded form in the ground state (S0), the latter being the emitting
species. In aprotic solvents of low polarity association of the S1 state of the dye takes place. In aqueous micellar media
the dye resides at the micelle–water interface. The binding constant for dye–micelle interaction has been determined.
Fluorescence data in b-cyclodextrine solution resemble that for that neutral micellar solution indicating that the
interaction between the -OH group of the heterogeneous part (micelle/cyclodextrine cavity) and the carbonyl oxygen
of the dye is important in both the cases. © 2000 Elsevier Science B.V. All rights reserved.
1. Introduction
cent probes because of their widespread applica-
tion in biology [6]. In our laboratory
symmetrically substituted ketocyanine dyes I & II
(Fig. 1), which show unique solvatochromic prop-
erties in both absorption and emission, have been
used extensively for studying solvation interaction
[7–9]. It was found that in homogeneous media
the dyes are good reporters of hydrogen-bond
donating ability of protic solvents. The micropo-
larity and dielectric constant at the micelle–water
interface may also be obtained from the study of
the fluorescence characteristics of the dyes [10].
Solvent effect on the photophysics of two other
structurally similar dyes III & IV of which III is
symmetrically substituted have also been studied
by other workers [11,12]. In the present paper we
Optical response of molecules showing an in-
tramolecular charge transfer (ICT) has been
found to be dependent on the immediate environ-
ment [1]. As such these molecules act as micropo-
larity reporters probing the local environments in
simple solutions or complex biomimicing system
[2–5]. Of these, the fluorescent polarity probes are
more advantageous compared to the non-fluores-
* Corresponding author. Tel.: +91-342-60810; fax: +91-
342-64452.
E-mail
address:
burchdsa@cal.vsnl.net.in
bsan-
jibb@yahoo.com (S. Bagchi).
1 Present address: Department of Chemistry, Haldia Gov-
ernment College, Haldia, Midnapore, India.
1386-1425/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S1386-1425(00)00321-8

2764
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
wards solvation interaction has been addressed.
Comparison among the dyes (I–VI) has been
made. Studies have also been made on the spec-
tral characteristics of the dyes III–VI in several
heterogeneous aqueous media, viz., cationic, an-
ionic and neutral micelles and aqueous b-cy-
clodextrine solution. Semiempirical molecular
calculation at AM1 level have been carried out to
provide the optimised geometry, charge distribu-
present extensive studies on absorption and emis-
sion spectral characteristics of two other related
dyes V & VI along with the dyes III & IV in
various homogeneous media. All these compound
show ICT transition between aniline nitrogen and
the carbonyl oxygen. The possible ground state
complexation in alcohol solvents, aggregation of
the solute in non-polar media and contribution of
various modes of solute–solvent interaction to-
Fig. 1. Dyes used in the present investigation. The charges on N and O atom have been indicated. The dipole moment of the dye
in the S0 state as calculated by AM1 method is also given.

P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
2765
tion and the dipole moment in the ground state
of the dye molecules.
fluorescence spectra (emission & excitation) were
taken on a HITACHI F4500 fluorescence spec-
trophotometer equipped with polarisation acces-
sory and a temperature controlled cell holder.
For measurement at different temperatures wa-
ter was made to circulate through the cell
2. Materials and methods
Ketocyanine dyes (I & II) were synthesised as
describe in the literature [13]. Purity of the pre-
pared dyes were checked by IR, UV/Vis absorp-
tion, and fluorescence spectral data. The
compound III was synthesised from cyclopen-
tanone and 4-(dimethylamino)-benzaldehyde. Cy-
clopentanone was dissolved in ethanol and
sodium ethoxide was added to it. 4-(Dimethy-
lamino)benzaldehyde was added to the above
mixture and refluxed for 1 h. It was then cooled
and filtered. The product was recrystallised from
ethanol. Compounds IV, V, and VI were synthe-
sised similarly by the condensation of indanone
acetophenone, and a-tetralone, respectively, with
4-(dimethylamino)benzaldehyde. The compounds
were recrystallised from dry ethanol. Purity of
the prepared compounds was checked by IR
and C, H, N analysis.
All the solvents used were the best grade
available commercially. They were distilled from
calcium hydride immediately before use to en-
sure the absence of peroxides and oxidising
agents. Mixed solvents were prepared by care-
fully mixing the components so as to minimise
contamination by moisture. Sodium dodecyl sul-
phate (SDS), cetyl–trimethylammonium bromide
(CTAB) and Triton X-100 were received from
Sigma and used without purification. For
holder from
(HETO HOLTEN, temperature range −30–
100°C).
The emission anisotropy at a wavelength u
was calculated from the following relationship
[4],
a
constant temperature bath
rem(u)=[IVV(u)−G(u)IVH(u)]
/[IVV(u)+2G(u)IVH(u)]
(1)
where I(u) denotes the fluorescence intensity at
the wavelength u and the first and the second
subscripts H and V respectively refer to the
horizontal and vertical setting of excitation and
emission polarisers. G(u) is an instrumental fac-
tor representing the polarisation characteristics
of the photometric system and is given by,
G(u)=IHV(u)/IHH(u), the subscripts having the
same meaning. Excitation anisotropy (rex) were
calculated similarly.
Dimroth–Reichardt empirical solvent polarity
parameters ET(30) [14] for the neat solvents
were taken from the literature [15]. Some of the
values were determined from the wavelength of
maximum absorption (umax) of the indicator dye,
viz., 2,6-diphenyl-4-(2,4,6,-triphenyl-1-pyridino)
phenolate, in the solvent according to the rela-
tion,
preparing
a
micellar medium the required
ET(30) (kcal mol−1)=28 590/umax (nm)
amount of the surfactant was added to triply
distilled water, and the contents kept in a ultra-
sound sonicator (JENCONS, U.K., Model T 80)
for a considerable period. Solutions of the so-
lutes in the micellar media were prepared by
adding the dye to the micellar media with
proper sonication. b-cyclodextrine (b-CD) was
received from Aldrich and was used without
purification.
The UV/Vis absorption spectra were measured
on a SHIMADZU UV 1020 PC spectrophoto-
meter fitted with a temperature controlled unit
(Model TB-85 Thermobath, SHIMADZU). The
2.1. Theoretical calculations
Semi empirical MO calculations at the AM1
[16,17] level using the MOPAC PC program
were carried out for the dyes III–IV on a PC.
The initial optimisation of the ground state ge-
ometries of the dye molecules was achieved us-
ing a molecular mechanics program followed by
unrestricted geometry optimisation. Fully opti-
mised geometry, charge distribution, and the
dipole moment for the dye molecules in the S0
state are shown in the Fig. 1.

2766
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
Fig. 2. Representative modified absorption spectrum [m(w¯)/w¯ vs. w¯] and modified fluorescence spectrum [I(w¯)/w¯ 3 vs. w¯]. (a) Dye VI in
benzene; (b) dye VI in acetonitrile; (c) dye V in 1,2-dichloroethane, and (d) dye V in ethanol.
3. Results and discussions
and the indicator dye for the ET(30)scale is the
same. Thus, the longest wavelength band of the
dyes are of intramolecular charge transfer (ICT)
origin. In this context it may be noted that the Vis
absorption band for dyes III and IV has been
assigned by other workers [11,12] as due to an
ICT from the nitrogen of the dimethylamino
group to the carbonyl oxygen.
3.1. Pure sol6ents
The absorption and fluorescence band maxima
in various pure solvents have been summarised in
Table 1. There is in general an overall red shift of
the Vis absorption band as the polarity of the
solvent increases. Fig. 2 shows some representa-
tive absorption spectra. The absorption spectra in
all the aprotic solvents show similar structures
(1700–1900 cm−1), which loose their prominence
as the polarity of the solvent is increased. The
polar solvents are expected to interact with the
carbonyl oxygen, the negative end of the dipolar
dye molecule and the structure may be explained
as due to carbonyl vibration. Plots of the maxi-
mum energy of absorption, E(A), versus the po-
larity parameter ET(30) is shown in the Fig. 3.
The E(A) values are rather insensitive towards a
change in ET(30) values. A linear correlation with
the ET(30)scale of solvent polarity indicates that
the nature of transition for the dyes under study
A red shift (2–4 kcal mol−1) is observed as we
go from aprotic to alcohol solvents. Moreover,
the band shifts dramatically due to an addition of
a very small amount of alcohol. In some cases an
isosbestic point is clearly observed. Study of ab-
sorption band in n-hexane+ethanol mixed sol-
vent (1:1 mole ratio) at different temperatures also
indicate the existence of isosbestic point. All these
facts point to exothermic formation of a ground–
state complex of the solute with alcohols possibly
through hydrogen bond formation with the car-
bonyl oxygen. The DH values are in the order of
dye V : dye VI (:1.2 kcal mol−1) and dye III
: dye IV (:0.9 kcal mol−1). This parallels the
AM1 charge densities on the carbonyl oxygen.

P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
2767

2768
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
For benzene, toluene, 1,4-dioxane, and tetrahy-
moment on excitation. The value of v1, the dipole
moment in the S1 state, may be obtained from the
solvatochromism of the absorption and the emis-
sion band. Using the formalism developed by
Marcus [18–21] one can partition E(A) and E(F)
as follows
drofuran the maximum energy of absorption
comes up at a lower value although these solvents
are characterised by lower polarity in the ET(30)
scale. This suggests that the nature of interaction
of the dyes with these solvents is somewhat differ-
ent from that in the case of indicator solute for
the ET(30) scale. It appears that these solvents
goes into a stronger interaction with the dyes than
predicted by the polarity parameters. This is also
reflected by the high solubility of these dyes in
these solvents.
The emission band may be characterised by
S1“S0 transition. For most of the solvents the
bands appear as structureless at 298 K. The
fluorescence band is roughly symmetrical with
respect to the absorption band (Fig. 2), the ab-
sorption band being wider in all the cases. The
fluorescence maximum shows strong sensitivity
towards a change in solvent polarity, the band
shifting to the red as the polarity increases. This is
intelligible in terms of an increase in the dipole
E(A)=DG(solv)+u1
E(F)=DG(solv)−u0
(2)
(3)
Where DG(solv) is the difference in free energy of
the ground and equilibrium (relaxed) excited state
in a given solvent and u represents the reorganisa-
tion energy. The suffixes ‘0’ and ‘1’ denote the
ground and excited state, respectively. Under the
condition that u0:u1:u one gets [22]
E(A)+E(F)=2DG(solv); E(A)−E(F)=2u
(4)
DG(solv) and u may be calculated using the On-
sager reaction field model. Thus, the final equa-
tion is as follows.
Fig. 3. Plots of E(A) and E(F) vs. ET(30). (a) Dye III; (b) dye IV; (c) dye V, and (d) dye VI.

P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
2769
ratio of the dipole moments in the S1 and S0 state
(v1/v0) may be obtained from the ratio of the
slopes of such plots. From this one can calculate
the v1 using the v0 values obtained from the
semiempirical AM1 calculation. Note that this
method does not require the value of the cavity
radius, ‘a’. The calculated v1 values for the dyes
are 9.5, 10.2, 9.9, and 9.9 D for dye III, IV, V,
and VI, respectively.
Fig. 3 shows also plots of E(F), the energy of
fluorescence maximum, versus the ET(30) parame-
ter. A double linear correlation is found for all
the dyes. Alcohol solvents fall on a separate line
(having a lower slope) indicating that the nature
of the emitting state is different for these class of
solvents. The enhanced solvent sensitivity of E(F)
compared to E(A) may be explained in terms of
increased solute–solvent interaction in the excited
state due to an increased in the dipole moment
upon excitation. For alcohols, apart from dipolar
interaction tighter hydrogen bonding is expected
in the excited state of the solvent as a result of
increased charge density on the carbonyl oxygen
on excitation. Thus, the emitting state in the case
of protic solvent is a strongly hydrogen-bonded
solvated complex. For dye IV a proton transfer
across the hydrogen bond has been suggested
[12]. The stronger solvent–solute interaction for
alcohols may also be inferred from the anisotropy
(r) of fluorescence. A higher value for alcohol
solvents (Table 1) indicate a slower rotational
motion of the transition dipole in an alcohol
cage.
In order to investigate the role of non-specific
dipolarity interaction and specific hydrogen
bonding ability of solvents a multiple linear re-
gression analysis [23] was made. A linear correla-
tion of E(F) was sought with the Kamlet–Taft
parameter [24] y*, h, and i representing the
dipolarity–polarisability interaction, the hydro-
gen–bond donating, and acceptance ability of
solvents. The following regression equations have
been obtained for the dyes.
Fig. 4. Representative plot of (a) E(A)+E(F) and (b) E(A)−
E(F) as a function of dielectric function for dye V. f(m)=
[2(m−1)/(2m+1)] and f(m, n2)=[2(m−1)/(2m+1)−2(n2−1)/
(2n2+1).
E(A)+E(F)=[(v20−v21)/a3][2(m−1)/(2m+1)]
+2DG(gas)+D(sp)
(5)
and
E(A)−E(F)
=[(v1−v0)2/a3][2(m−1)/(2m+1)
−2(n2−1)/(2n2+1)]
(6)
where m and n are the dielectric constant and the
refractive index of the solvent, DG(gas) is the
value and DG in the gas phase and a is the cavity
radius. These equations are valid for aprotic sol-
vents where specific interactions are absent. Fig. 4
shows representative plots of E(A)9E(F) versus
the appropriate dielectric function. Note that the
For dye III:
E(F)=63.99−5.29h−5.00i−11.25y*,
n=24, r=0.96.

2770
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
For dye IV:
E(F)=64.50−3.24h−4.57i−9.19y*, n=20,
r=0.94.
leads to an extension of conjugation modifying
the energy levels as pointed out by Kessler and
Wolfbeis [13]. The constant term in the above
equation represent the E(F) value in a non-polar
media, e.g. n-hexane where solute–solvent inter-
action is not present. For all the dyes the value
appears at :64 kcal mol−1, which is in good
agreement with the experimental value (Table 2).
The dyes are practically insoluble in water and so
the direct measurement of E(F) for water is not
possible. However, when water is added to an
ethanol solution of the dye the fluorescence maxi-
mum shifts to higher wavelength. Plot of the E(F)
values against the mole fraction of water gives a
straight line from which the E(F) value for pure
water may be found out by extrapolation. The
values thus obtained have been tabulated in the
Table 1.
For dye V:
E(F)=64.30−3.05h−3.71i−9.36y*, n=20,
r=0.90.
For dye VI:
E(F)=64.30−3.60h−5.50i−8.50y*, n=20,
r=0.94.
For all the dyes the coefficient of the y* term is
greater than that of the h or i term, indicating
that the dipolarity/polarisability interaction is
more important for the dyes especially for the
unsymmetrical dyes IV–VI. This behaviour is dif-
ferent for the dyes I and II where an opposite
trend is observed [7,8]. For these dyes an increase
in the electron density on the carbonyl oxygen
In aromatic hydrocarbons (benzene, toluene)
and cyclic ethers (1,4-dioxane, tetrahydrofuran,
2-methyltetrahydrofuran) the fluorescence band is
found to be structured. When n-hexane is added
to these solvents a gradual blue shift is observed
and the structures become more prominent. Fluo-
rescence in these solvents are also concentration
dependent, the higher wavelength band is more
prominent at higher concentration. Fig. 5 shows a
representative plot. For example, in a dilute solu-
tion (:10−6 M) the two bands around u=525
and 545 nm were obtained for the dye I, the
relative intensities of the bands do not change
significantly as the temperature is changed in the
range 288–318 K. But in concentrated solution
the relative intensity of the 550 nm band increases
at lower temperature. Difference of the spectra in
concentrated and dilute solution indicates a broad
band with maxima at :550 nm. The overall
intensity of the band increases as the temperature
Fig. 5. Fluorescence spectrum of dye I in (a) dilute solution
(+10−6 mol l−1) and (b) concentrated (H10−5 mol l−1
)
solution at 288 K. (c) Fluorescence spectrum of the concen-
trated solution at 308 K; (d) and (e) are the difference of the
spectrum in concentrated and dilute solution at 288 and 308 K
respectively.
Table 2
Fluorescence characteristics of the dyes in heterogeneous media at 298 K
Environment
Dyes
III
IV
V
VI
E(F)
Anisotropy
(r)
K
E(F)
Anisotropy
(r)
K
E(F)
Anisotropy
(r)
K
E(F)
Anisotropy
(r)
K
(kcal mol−1
)
(mol l−1
)
(kcal mol−1
)
(mol l−1
)
(kcal mol−1
)
(mol l−1
)
(kcal mol−1
)
(mol l−1
)
SDS
CTAB
TX-100
49.55
49.29
49.38
0.12
0.25
0.21
0.33
–
–
–
–
52.46
52.95
54.46
54.25
0.30
0.28
0.28
0.27
1860
5000
51 300
–
52.95
52.95
54.46
54.88
0.32
0.28
0.24
0.30
732
20 000
129 000
–
51.98
52.95
53.95
54.98
–
340
0.30
0.23
0.31
–
26 000
–
b-cyclodextrine 51.89

2771
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
Fig. 6. Fluorescence spectra of dye VI in CTAB (a); SDS (b) and TX-100 (c) micelles (conc. of micelle 4 CMC). The dotted line
indicates the spectrum when the surfactant concentration is 0.25 CMC. (d) Representative plot of the Eq. (8) for dye VI in CTAB.
intensity of the band increases as the temperature
is lowered. No such change, however, has been
observed in the excitation or the absorption spec-
trum. These facts points to an aggregation of the
dye in the excited state in a concentrated solution
(concentration ]10−5) in these non-polar sol-
vents. In a polar solvent the chromophore individ-
ually interacts with solvent dipoles and are no
longer free to form aggregates.
in micellar environment, meaning a significant
dye–micelle interaction. Fig. 6 is a representative
plot, showing the fluorescence characterstics of
the dyes in a solution of a surfactant. Below the
critical micelle concentration (CMC) of the sur-
factant the intensity of fluorescence is small and
the band shows a structure. The structure be-
comes less prominent as the concentration of the
surfactant is increased beyond the CMC value
and practically vanishes at higher CMC. The
structure is possibly indicative of the carbonyl
vibration, which smoothes out as the dye–micelle
interaction is increased. For fixed concentration
of dye the intensity of maximum emission in-
creases as the concentration of micelle is increased
and reaches a limiting value at a very high con-
centration of micelle (when micellisation of the
dye is complete). For a particular concentration
of the micelle the intensity of fluorescence in-
creases as the temperature is decreased. All the
observations point to the existence of the
equilibrium
3.2. Heterogeneous media
The absorption and fluorescence characteristics
of the dyes III–VI in heterogeneous media have
been summarised in Table 2. The dyes are only
slightly soluble in water. This is perhaps due to
the presence of hydrophobic part of the dye. The
solubility of the dye III in micellar media is also
small, but for the other dyes solubilisation is
higher. Increased solubility in micellar media indi-
cates dye–micelle interaction. Measurements of
emission polarisation gives high values of fluores-
cence anisotropy (Table 2). This indicates a slower
rate of rotational motion of the transition dipole
D+M ? D…M
(7)

2772
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
Where D and M represent the dye and the micelle
respectively and D…M represent the dye bound
with micelle.
respectively. Thus, the dye–micelle interaction is
similar to the dye–alcohol interaction for these
micelles.
The fluorescence band maximum of the dye
bound to the ionic micelles appear below the
extrapolated E(A) value for water and is in the
order SDS\CTAB. For the neutral micelle Tri-
ton X 100 (TX-100) the maximum of emission
appear around u=525–530 nm for all the dyes
that are similar to the E(A) value in alcohol
solvent. These observations suggest that the im-
mediate environment of the dye is a polar one.
This rules out the possibility of the dye being
located inside the micelle. It has been observed by
earlier workers [24–26] that a part from the head
group and counter ions only water and alkyl
chains contribute to the polarity of the micelle–
water interface and as such alcoholic solvents are
better model solvents to describe the region. Thus,
we infer that the dyes are located in the micelle–
water interface.
The negative charge density of the carbonyl
group of the dye in the S1 state will be involved in
the electrostatic interaction with the polar head
group of the ionic micelles. Besides this, hydro-
phobic interaction of the hydrocarbon like wing
of the dye with the micelle also plays a part as
reported by other workers [27] for different dyes.
In the cationic micelle (CTAB), the positively
charged head group will attract the negative
charge density of the carbonyl oxygen and the dye
will be in a more hydrophobic environment. This
is particularly more important in the S1 state of
the dye. Thus, the possibility of hydrogen bonding
interaction will be reduced. On the other hand,
for anionic micelles (SDS) the chromophoric part
of the dye will be more exposed to hydrophilic
environment and more effective hydrogen bond-
ing will take place. This is also reflected in the
fluorescence band maximum of the dyes in the
micelles as discussed earlier. The non-ionic surfac-
tant TX-100 consists of bulky phenyl head groups
and a long polyoxyethylene chain, which termi-
nates in an -OH group. The fluorescence maxi-
mum in TX-100 micelle appear around the same
region as in alcohols. This is true for other non-
ionic micelles. For example, in brij-35 and tween
80 the band for V appears at u=530 and 525 nm,
A quantitative estimate of the binding constant
(K), i.e. the equilibrium of the process Eq. (7), the
equation proposed by Almgren et al. [28,29] can
be used. Thus,
(I8 −I0)/(I−I0)=1+(K[M])−1
(8)
where I8, I0, and I are the fluorescence intensities
under complete micellisation of the dye, in ab-
sence of micellisation, and at any intermediate
micelle concentration, respectively. [M] represents
the concentration of the micelle, which is given by
[M]=([Surf]−CMC)/N and [Surf] represents the
surfactant concentration and N is the aggregation
number. For the calculation of M the values of N
are 62, 60, and 143 for SDS, CTAB, and TX-100,
respectively. The measured K values (910%) are
given in the Table 2. DH values may also be
obtained similarly. Our experimental results indi-
cate that ꢀDHꢀ values (1–2 kcal mol−1) are in the
order CTAB : TX-100\SDS. For the cationic
micelle (CTAB) the electrostatic interaction be-
tween the carbonyl oxygen and the positive
charged micelle is the major contributing factors
towards DH values. In SDS, however, the positive
end of the dye (S1 state), i.e. the N atom of the
dimethylamino group goes into interaction with
the micelle. But the charge of N atom is some-
what shielded by the methyl group. Thus, the
electrostatic interaction is less. There is however a
hydrogen bonding interaction between the car-
bonyl oxygen and water molecules. The results
indicate that for SDS the overall interaction en-
ergy becomes less than that for the CTAB micelle.
It is important to note that the dye–micelle inter-
action for symmetric ketocyanine dyes (I–III) is
less than for the unsymmetric dyes (IV–VI). For
unsymmetric dyes, one end of the dipolar dye
(positive or negative) may remain attached to the
micelle interface keeping the other end (negative
or positive) part away from the micelle, thus
minimising the electrostatic interaction. But this is
not possible for the symmetric structure, making
the dye–micelle interaction weaker.
Table 2 also lists the spectral features in
aqueous b-cyclodextrine (b-CD) solution. Strong

2773
P.K. Das et al. / Spectrochimica Acta Part A 56 (2000) 2763–2773
[6] B. Valeur, in: S.G. Schulman (Ed.), Molecular Lumines-
cence Spectroscopy, Chemical analysis series, vol. 77,
Wiley, New York, 1993 part 3.
[7] D. Banerjee, A.K. Laha, S. Bagchi, Indian J. Chem. Sect
A 34 (1995) 94.
dye–b-CD interaction is indicated by high values
of fluorescence anisotropy. It appears that the
fluorescence characteristics in the medium is very
similar to those in the neutral micelles, e.g. TX-
100, brij 35, or tween 80, pointing to the fact that
interaction of the dyes in these media are very
similar. Thus, the dye is located near the b-CD
bucket and the interaction of the carbonyl part of
the dye with the hydroxyl groups in b-CD are
important.
[8] D. Banerjee, A.K. Laha, S. Bagchi, J. Photochem. Photo-
biol. A Chem. 85 (1995) 153.
[9] D. Banerjee, S. Mondal, S. Ghosh, S. Bagchi, J. Pho-
tochem. Photobiol. A Chem. 90 (1995) 171.
[10] D. Banerjee, P.K. Das, S. Mondal, S. Ghosh, S. Bagchi,
J. Photochem. Photobiol. A Chem. 98 (1996) 183.
[11] M.V. Barnabas, A. Liu, A.D. Trifunac, V.V. Krongauz,
C.T. Chang, J. Phys. Chem. 96 (1992) 212.
[12] Y. Wang, J. Phys. Chem. 89 (1985) 3799.
[13] M.A. Kessler, O.S. Wolfbeis, Spectrochim. Acta A 47
(1991) 187.
4. Conclusion
[14] K. Dimroth, C. Reichardt, T. Siepmann, F. Bohlmann,
Liebigs Ann. Chem. 661 (1963) 1.
[15] C. Reichardt, Solvents and Solvent Effects in Organic
Chemistry, second ed., VCH, Weinheim, 1988.
[16] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart,
J. Am. Chem. Soc. 107 (1985) 3902.
Ketocyanine dyes are good indicators of solvent
polarity. In protic solvents they exist as an equi-
librium mixture of bare and hydrogen bonded
form, the later being the emitting species. In
aprotic solvents of lower polarity the S1 state of
the dye exists in associated form in equilibrium
with the solvated monomer. The unsymmetric
dyes show greater micelle–dye interaction. The
binding constant and the heat of binding can also
be determined from fluorimetric studies.
[17] M.J.S. Dewar, K.M. Dieter, J. Am. Chem. Soc. 108
(1986) 8075.
[18] R.A. Marcus, J. Chem. Phys. 38 (1963) 1858.
[19] R.A. Marcus, J. Chem. Phys. 38 (1963) 1335.
[20] R.A. Marcus, J. Chem. Phys. 39 (1963) 460.
[21] R.A. Marcus, J. Chem. Phys. 43 (1965) 1261.
[22] B.S. Brunschwig, S. Ehrenson, N. Sutin, J. Phys. Chem.
91 (1987) 4714.
[23] H. Abraham, P.L. Grellier, J.-L.M. Abboud, R.M. Do-
herty, R.W. Taft, Can. J. Chem. 66 (1988) 2673.
[24] J. Kamlet, J.-L.M. Abboud, M.H. Abraham, R.W. Taft,
J. Org. Chem. 48 (1983) 2877.
References
[25] K.A. Zachariasse, N.V. Phuc, B. Kozankiewicz, J. Phys.
Chem. 85 (1981) 2676.
[1] A.M. Kajer, J. Ulstrup, J. Am. Chem. Soc. 109 (1987)
1934.
[26] P. Mukherjee, J.R. Cardinal, J. Phys. Chem. 82 (1978)
1620.
[2] K. Bhattacharya, M. Chowdhury, Chem. Rev. 93 (1993)
507.
[27] S.B. Savvin, Rev. Anal. Chem. 8 (1979) 55.
[28] M. Almgren, F. Griesser, J.K. Thomas, J. Am. Chem.
Soc. 101 (1979) 279.
[3] C. Reichardt, Chem. Rev. 94 (1994) 2319.
[4] J.R. Lakowicz, Principles of Fluorescence Spectroscopy,
Plenum, New York, 1983.
[29] G. Saroja, B. Ramachandram, S. Saha, A. Samanta, J.
Phys. Chem. B 103 (1999) 2906.
[5] K. Kalyanasundaram, Photochemistry in Microheteroge-
neous Systems, New York, Academic Press, 1987.
.