280
A. KUMAR AND S. KUMAR
of CdS in the visible region nor caused the development
of any new peak. It eliminated the possibility of any
chemical interaction between the two. However, the
scanning of the absorption in the UV region showed a
decrease in absorbance where 2-MI had absorption. This
suggested that 2-MI is physisorbed on the surface of the
particles. From the Langmuir plot of the adsorption data,
its binding constant was found to be 0.6 Â 103 dm3 molꢀ1
(Fig. 2). For other indoles, namely indole, 3-MI14,
2,3-dimethylindole15 and tryptophan, also no chemical
interaction between the substrate and CdS has been
reported and their intensities of adsorption were found to
be 2.0 Â 103, 2.1 Â 103, 3.3 Â 103 and 0.6 Â 103 dm3
molꢀ1, respectively. These values are of similar order of
magnitude to that of 2-MI. These investigations reveal
that the photogenerated holes possibly intercept the
surface-bound substrate.
Surface interaction of the substrate could also be
probed by exploiting the luminescence from colloidal
CdS. If indeed the surface of the particles is modified by
adsorption of 2-MI, the luminescence of colloidal CdS
should change in its presence. The luminescence spectra
of CdS in the absence and presence of 2-MI are shown in
Fig. 3. It can be seen that the addition of 2-MI produced a
new emission band at 530 nm and the red emission due to
CdS was simultaneously quenched. The 530 nm band is
different to the band gap emission of CdS, which lies
around 490 nm, and corresponded to a quantum effi-
ciency of 5 Â 10ꢀ3. The green emission of used CdS
particles is weak and has been assigned earlier to the
recombination of free charge carriers.9,10b,16 At higher
[2-MI], the 530 nm band shifts slightly to the red. Any
contribution to the observed emission on account of 2-MI
can be neglected as 400 nm light was used for excitation;
2-MI does not have any absorption at this wavelength.
Since the absorption spectrum of CdS remains unchanged
in the presence of 2-MI, it shows the absence of any
complexation in the ground state between the two. The
surface of colloidal CdS particles is known to have many
defects and traps. Binding of 2-MI to these sites may,
thereby, occur through physical adsorption. Hence the
appearance of the 530 nm band can be attributed to the
emission from the exciplex formed between the excited
CdS and the adsorbed 2-MI. In contrast to indole, no
isoemissive point is observed with 2-MI and a bath-
ochromic shift of the green band is noted at high [2-MI].
This is possibly due to the formation of intermediate
complexes of varied stoichiometry at various [2-MI]
which may have different emission characteristics.
Similar green emission bands at different wavelengths
have also been noted for other substituted indoles. The
wavelengths of the green band in comparison with that of
indole follow the trend indole < tryptophan < 2-MI < 3-
MI and are observed at 508, 520, 530 and 540 nm,
respectively. A change in the wavelength of the green
band for different indoles evidently rules out the
possiblity of it being the band gap emission due to
CdS. This behaviour is understood in terms of the varied
oxidation potential of these substrates which affects the
extent of charge transfer in the exciplex formed between
excited CdS and the respective indole. The order of
decreasing oxidation potential17 of the indoles studied is
indole < tryptophan < 2-MI < 3-MI, which is the same
as the order in which the bathochromic shift of the
wavelength of the green emission band changes. The
enhancement of the luminescence of colloidal CdS and
Cd3As2 particles by binding of tertiary amines and
aliphatic thiols8 can also be argued along similar lines.
The reactivity of photogenerated trapped holes towards
2-MI was checked by measuring the emission lifetime of
CdS in the absence and presence of 2-MI. In these
experiments CdS particles were excited by 400 nm light,
where 2-MI does not show any absorption. The emission
was followed at 600 nm. The decay curve of CdS
emission could be fitted in a three exponential decay
program. In the presence of 2 mM of 2-MI the average
lifetime of CdS emission decreased from 8.0 to 1.2 ns.
The decrease in the lifetime confirms the interception of
holes by 2-MI following dynamic quenching.
In the light of the above results, the reaction scheme for
the formation of 2 and 3 in the CdS-mediated oxidation of
2-MIꢀcan be outlined as shown in Scheme 1.
O2 formed in the cathodic reaction (ii) may
eventually produce H2O2.18 The photochemical and
thermal reaction of H2O2 with 2-MI under similar
experimental conditions did not produce any of the
above - identified products. These experiments support
ꢀ
the participation of O2 depicted in step (vi). Indolyl
ꢀ
radical is known to couple with O2 to yield hydroper-
oxide.19 Hence the initial step in CdS-sensitized reaction
of 2-MI may consist of its oxidation to produce 2-
methylindolyl radical cation. The pKa of this radical
cation is 5.7 Æ 0.1,20 and at pH 10.5 it will be converted
largely into 2-methylindolyl radical.
In contrast to the CdS-sensitized reaction of indole,9b
O2 and O2ꢀ react with 2-methylindolyl radical to produce
the corresponding peroxy radical and hydroperoxide in
steps (v) and (vi). These intermediates subsequently
disproportionate and decompose to yield 2 and 3 in steps
(vii) and (viii), respectively. The higher reactivity of 2-
methylindolyl radical with molecular oxygen in compar-
ison with that of indole radical can be explained by the
difference in their reduction potentials.17,19 In case of
indole the initially produced radical cation mainly forms
an adduct with OHꢀ at the electron-deficient C-2
position,9b in contrast to 2-MI, in which the C-2 position
is blocked by a methyl group. The OH adduct is then
oxidized by the photogenerated hole to give 3-hydro-
xyindole, which is known to autooxidize readily to indigo
in the presence of oxygen.21
Hence in CdS-sensitized reactions of indole and 2-MI,
the difference in reactivity of radical intermediates
ꢀ
formed with O2/O2 is responsible for the formation of
different products of oxidation in these reactions.
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 277–282 (1998)