Journal of Colloid and Interface Science 246, 122128 (2002)

doi:10.1006/jcis.2001.8030, available online at http://www.idealibrary.com on

Luminescent Properties of Tris(2,20 -bipyridine)ruthenium(II)

in Sol-Gel-Processed Dip-Coated Thin Films
S. Lopez, A. Senz, and H. E. Gsponer1
Departamento de Qumica y Fsica, Facultad de Ciencias Exactas, Fisicoqumicas y Naturales, Universidad Nacional de Ro Cuarto,
Ruta 8 Km 601, 5800 Ro Cuarto, Argentina
Received June 7, 2001; accepted October 5, 2001

The luminescence decay and spectral behavior of ruthenium(II)

tris-1,2-bipyridine dichloride dissolved in different organically
modified silicate gel matrixes were investigated. Dip-coated thin
films were synthesized from tetraethoxysilane (TEOS), methyltriethoxysilane (MTEOS), ethyltriethoxysilane (ETEOS), and methyltrimethoxysilane (MTMOS). A blue shift in the ruthenium complex
emission spectrum with respect to the aqueous solution was observed for all the films on the sol to gel conversion. This spectral
shift was slightly dependent on the precursor used to obtain the films
and independent of the reaction pH to prepare the sol. In the data
treatment of the time-resolved luminescence measurements, it was
assumed that the distribution of the luminophore in the films was
nonhomogeneous. The analysis of the luminescence decay profiles
was based on a multisite model. All decay curves are best described
by a double-exponential model. The parameters of the decay components depended principally on the thermal treatment used in
the processing of the films. The lifetimes decreased and the emission espectra showed a red shift with the increase in the drying
temperature. A luminescence quenching of the ruthenium complex
in the films by dissolved oxygen in aqueous solution was also observed. The quenching rate constant obtained from the preexponential amplitude-weighted mean lifetimes (M ) was in the order of
109 M1 s1 . When a phenolic derivative was used as quencher the
process rate was greatly reduced compared to the quenching in water. It would seem that the metallic complex sequestered within the
film is placed either into a higher microviscosity microenvironment
or in a location which the phenolic quencher cannot access. In both
cases, the quenching plot based on Mo /M could be fitted satisfactorily by a sum of two terms of Stern-Volmer. This fact is indicative of
the matrix microheterogeneity for the films and is fully consistent
with the biexponential nature of the luminescence intensity decay
profiles. C 2002 Elsevier Science
Key Words: thin films; sol-gel; tris(2,20 -bipyridine)ruthenium(II);
luminescence; lifetime; quenching.

INTRODUCTION

Polypyridinic ruthenium(II) complexes, particularly tris(2,20 bipyridine)ruthenium(II), Ru(bpy)2+

3 , are organometallic com1

To whom correspondence should be addressed. Fax: 54-358-4676233.

E-mail: Hgsponer@exa.unrc.edu.ar.
0021-9797/02 $35.00

C 2002 Elsevier Science

All rights reserved.

pounds which have received much attention in the literature due

to their unique properties, such as strong luminescence, moderate excited-state lifetime, appreciable reactivity of the excited
state in electron and energy transfer reactions, and relatively high
thermal and chemical stability (1). The luminescent excited state
is assigned to a metal to ligand charge-transfer (MLCT) state.
Then, the luminescent properties of these complexes are very
sensitive to the polarity and viscosity of the environment. The
dynamic properties of the excited state of Ru(bpy)2+
3 and related complexes adsorbed on semiconductors are also widely
studied (24). Apart from those on semiconductors, there is an
important number of studies on Ru(bpy)2+
3 adsorbed on nonconducting materials such as SiO2 , which can actually modify
photochemical reactions (58).
Sol-gel technology is a versatile process for making ceramic
and glass materials and it is carried out in a liquid solution.
In general, the sol-gel process involves the transition of a system from a liquid sol (moisturized colloid) into a solid gel
phase. The starting materials used in the sol preparation are
usually inorganic oxides or organometalic compounds such as
metal alkoxides. In a typical sol-gel process, the precursor is subjected to a series of hydrolysis and condensation/polymerization
reactions to form a colloidal suspension or a sol. Further processing of the sol enables us to make ceramic materials with
different properties on the basis of the processing used. Thin
films can be produced on a piece of substrate by spin coating or
dip coating. The materials are optically transparent, though they
contain a high volume of interconnected and unstable porosity.
The porous matrixes may be impregnated with a second organic
or inorganic active optically material to provide a new type of
sensors.
Entrapping ruthenium complexes in solid matrixes, particularly in thin films, allows the preparation of devices for many
applications.
Many of the previous works on doped materials obtained
by the sol-gel process have demonstrated that the dopants are
present as individual molecules or complexes in the inorganic
matrixes, and that the properties in the matrixes are similar to
those of the liquid solution. In other words, the functionality
of the dopants is also available in the solid state, allowing the
doped material to be used as a device.

122

LUMINESCENT PROPERTIES OF Ru(bpy)2+

3

Phenol-derived compounds, especially chlorate derivatives,

are environmental pollutants, which may be found dissolved in
aqueous solutions.
Therefore, and on the basis of our previous knowledge on
the luminescence quenching of the MLCT state of Ru(bpy)2+
3
by different phenol derivatives in homogeneous, microheterogeneous and heterogeneous environments (9), we decided to study
the luminescent properties of this complex in dip-coated thin
films and to investigate the possibility of obtaining a luminescent sensor to detect the presence of phenol derivatives dissolved
in aqueous solutions in contact with the doped material.
EXPERIMENTAL

All the products used were of the highest commercially available quality, and they were used with no previous treatment.
Aqueous solutions were prepared with water purified by reverse
osmosis, followed by ion exchange and activated charcoal (its
typical conductivity was 18 M1 ).
Sol preparation. The preparation with tetraethoxysilane
(TEOS) was done at pH 1 and a reaction time of 18 h at
70 C. The molar ratio water/precursor (Rw ) was 4/1 and the
one ethanol/water was 1/1. The same procedure was used
for methyltriethoxysilane (MTEOS) but the molar ratio water/precursor was 2/1.
Two different procedures were used for methyltrimethoxysilane (MTMOS). For one of them the pH was 8, Rw 2/1 and
methanol/water was 1/1. The reaction time was 18 h at 70 C.
For the other procedure, the molar ratios were the same, but the
pH was 4 and the reaction time was 1 week at room temperature.
The last procedure, pH 4 and reaction at room temperature,
was also carried out using ethyltriethoxysilane (ETEOS) as precursor and ethanol as cosolvent.
In all cases the Ru(bpy)2+
3 concentration used in the sol was
2 103 M.
Film processing. The sols were allowed to stand for at least
1 h at room temperature before deposition. The glass slides
were prepared for film deposition by soaking them in 1 M HCl
(aq) for 1 h, rinsing them with copious amounts of water, and
drying them in an oven at 150 C.
Coating films were prepared from the sol by dip-coating on
cleaned glass slide substrate at three different withdrawing rates
but constant in each experiment. Withdrawal speeds were 0.85,
2.12, and 4.25 mm/s. In a few cases Mylar or polypropylene was
used as substrate of the films. The substrates of the films were
cut to fit the diagonal length of a standard square cell.
Following the deposition step, some films were dried in an
oven for 18 h at 70 C ((O) films) and others were dried in the
dark under ambient atmosphere at room temperature (22 2 C)
for at least 72 h ((R) films). In some cases, individual films were
thermally dried in an oven for 1 h at either 80, 103, or 150 C.
Measurements. The luminescence emission spectra were
obtained by a Spex Fluoromax DM3000 fluorescence spectrophotometer. The excitation wavelength was 450 nm. Due to

123

interference from the excitation source, the sample was placed

in the fluorescence cell in such a way that the light scattered or
reflected by the film coming from the excitation beam diminished at the entrance of the detector. The excitation radiation was
impinged on the film-coated side of the glass substrates at an incidence angle of 45 with a 90 angle maintained between the excitation beam trajectory and the emission collection optics. In
the quenching experiments, emission intensities were recorded
at the wavelength maximum of the noncorrected spectra.
A pulsed nitrogen laser with FWHM 4 ns and wavelength
337.1 nm pulses was used as the excitation source to perform
the time-resolved intensity decay experiments. The luminescence emission was monitored through a 582-nm cutoff filter
to eliminate stray room and pump light. The signal from the
photomultiplier was relayed through a 50 termination plug
that was passed into digitizing oscilloscope (Hewlett-Packard
HP54200A). The oscilloscope was interfaced to a personal computer to resolve the luminescence intensity decay profiles by
using a nonlinear least-squares software.
The absorption spectra were recorded with a Hewlett-Packard
8453 spectrophotometer.
RESULTS AND DISCUSSION

Figure 1 shows the Ru(bpy)2+

3 emission espectrum in TEOSderived sol-gel-processed dip-coated thin (O) films on different
classes of substrates. The luminescence of the metallic complex
arises from metal to ligand charge-transfer excited state. The
emission shows significant alterations with changes in solvent
polarity or donor/acceptor properties of the environment due
to the fact that the emitting state involves large radial changes
in charge distribution. It can be observed from Fig. 1 that the
emission spectrum (max 593 nm) was practically independent
of the material of the substrate used for each film. There is no

FIG. 1. Emission spectra of Ru(bpy)2+

in TEOS (pH 1)-derived sol3
gel-processed dip-coated thin (O) films on different substrates: A, mylar; B,
polypropylene; C, glass slide.

124

LOPEZ,
SENZ, AND GSPONER

change in the wavelength of the emission maximum between

the different supports of the films. This could be attributed
to the fact that the complex of Ru(II) does not interact with
the substrate material of the films or to the fact that the complex interacts specifically with the SiO groups of the silicate
network (10).
In the MTEOS-derived films at pH 1 and the MTMOS-derived
ones at pH 4 and 8, the emission maximum was at 589 nm in
(O) films and at 596 nm in (R) films. This difference in the
emission maximum wavelengths could be due to a high content
of residual water in the (R) films dried in ambient atmosphere at
room temperature. However, for the (O) and (R) films obtained
from whichever precursor, a blue spectral shift with respect to the
aqueous solution (max in water 609 nm) was always observed.
Furthermore, the absorption spectra for the (O) and (R) films
did not show a measurable shift at the 450-nm bandassigned
to a MLCT transitionwith respect to the observed spectrum
for Ru(bpy)2+
3 in aqueous solution. Then, this indicates that the
evolution of the media affects the MLCT excited state without
perturbing the excited-state Franck-Condon level.
A reasonable explanation for these data can be given in
terms of solute-solvent interactions between the metallic complex and its environment. The solvent molecules surrounding the
Ru(bpy)2+
3 complex are not fully free to move or rotate because
of their possible interaction with the silanol groups (SiOH) of
the gel network. This restriction in the freedom of movement
of the solvent molecules would cause the rigidochromism (11)
of the Ru(bpy)2+
3 complex. The excited state is not completely
stabilized or relaxed within its lifetime, and hence, the emission
occurs from a higher energy level than that in a fluid solution.
This leads to a blue shift on the emission peak on sol to gel film
conversion. The environment of the Ru(bpy)2+
3 complex can be
said to experience the fluid to glass transition on sol-gel film
conversion.
This blue shift on the emission spectra of ruthenium(II) complexes has also been observed by other authors in similar systems
(8, 12, 13).
The above blue spectral shift in the emission max was
observed for all the dip-coated thin films independently of the
reaction pH to prepare the sol. Then, it would seem that the
pH dictates the kinetics throughout the sol-gel process, but
the final environment sensed by the Ru(bpy)2+
3 probe is independent of the initial pH. However, the slight blue spectral shift
of 593 nm in the TEOS-derived (O) films to 589 nm in the
MTEOS-derived and MTMOS-derived (O) films could be due
to an effect of polarity on the stabilization of the excited state
MLCT. The presence of the nonpolar group SiCH3 in the gels
of MTEOS and MTMOS would produce a lower stabilization of
the MLCT state and, consequently, a blue shift in the emission
spectrum.
On the other hand, the stability of the films was assayed by
leaving a film on glass substrate in a fluorescence cell with water
for 48 h. In this time, the spectra and the luminescence intensities
showed small changes which were within the experimental error.

In the experiments of luminescence intensity measurements

of Ru(bpy)2+
3 performed with a film in the presence of a phenolic compound in aqueous solution, the reproductibility between the measurements lowered as the quencher concentration
increased. We attributed this lack of precision in the measurements to a low reproductibility in the position of the cell or of
the film-cell pair in the cell holder after each addition of the
quencher solution. However, a low quenching efficiency could
be appreciated for an efficient quencher in aqueous solution such
as 2,4-dichlorophenol (14).
A purely dynamic luminescence quenching obeys the classic
form of the Stern-Volmer relationship
I o /I = o / = 1 + K SV [Q] = 1 + kq o [Q],

[1]

where I o and I are the steady-state luminescence intensities

in the absence and presence of quencher (Q), o and are the
excited-state luminescence lifetimes in the same conditions, [Q]
is the quencher concentration, K SV is the dynamic Stern-Volmer
quenching constant, and kq is the bimolecular rate constant describing the efficiency of the collisional encounters between the
luminophore and the quencher. In our quenching experiments,
[Q] is the concentration of the quencher dissolved in aqueous
solutions in contact with the film.
We observed the Stern-Volmer plots with a downward curvature instead of a linear Stern-Volmer plot as it is predicted
by Eq. [1]. This nonlinearity of the plot has been commonly
observed in Stern-Volmer plots for quenching of species in microheterogeneous media (1517). In this case, the overall SternVolmer expression is simply the superposition of Stern-Volmer
expressions resulting in a summation over all sites and Eq. [1]
can be recast as
"
#1
m
X
o
o
f 0 i /(1 + K SVi [Q])
,
[2]
I /I = / =
i

where the f 0 i variables are the fraction of the total emission from
each luminophore located at a site type i under unquenched conditions and the K SVi variables are the associated Stern-Volmer
quenching constants for each site.
These types of quenching profiles are indicative of matrix
microheterogeneity and are fully consistent with the mutiexponential nature of the luminescence intensity decay profiles. In a
multisite model, the luminophore time-resolved intensity decay
kinetics is best described by a sum of decay processes given by
I (t) =

m
X

i et/ i ,

[3]

where i is the preexponential amplitude associated with the

site and/or domain i within the host matrix, and i is the corresponding excited-state luminescence lifetime.
In a subsequent stage, we carried out measurements of
time-resolved luminescence intensities to obtain the Ru(bpy)2+
3

LUMINESCENT PROPERTIES OF Ru(bpy)2+

3

125

FIG. 2. Luminescence decay profiles of Ru(bpy)2+

3 in a MTMOS (pH 8)-derived sol-gel-processed dip-coated thin (R) film and residuals: (A) fitted by a
single-exponential curve, (B) fitted by a double-exponential curve.

excited-state lifetimes for different films in aerated aqueous solutions. In all studied cases, the luminescence decay profile was
properly fitted by a sum of two exponential decays instead of a
single-exponential one as judged by the appearance of the plot
residuals (Fig. 2). This kinetics result of m = 2 in Eq. [3] for
all our luminescence decay measurements may be considered a
realistic representation for this work, given the fact that previous
studies reported heterogeneity at the molecular level in this type
of material (1820).
The decay lifetimes are tabulated in Table 1, together with the
precursors and the methods of preparation/drying used for each
film, and the relative contribution to the luminescence
decay of
P
each component calculated as %i = (i i / i i ) 100.
From Table 1, it can be observed that the lifetimes for the
films dried in oven (O) are slightly dependent on the preparation conditions and the time elapsed until the luminescence
measurement was made (tL ). These lifetimes ranged from 260
to 460 ns for the short lifetime and from 1220 to 1350 ns for the
long lifetime. The last component was the most important in all
the cases contributing 71 to 95%. Here we should be reminded
that the change in the lifetime looks like that in the emission
max , which was explained by the immobilization of Ru(bpy)2+
3
in sol to gel film conversion. In general, except for the case of
the MTMOS-derived films at pH 4, the contribution of the long
lifetime component increases over time tL .
The short lifetime value was in the order of the Ru(bpy)2+
3
luminescence lifetime in aerated aqueous solution (400 ns),
but the long lifetime was far above this value, over 1 s in all
cases. The short lifetime component values for the (O) films
MTEOS derived and MTMOS derived, calculated as an average
from data in Table 1, were 270, 340, and 395 ns at pH 1, 4, and 8

respectively. Then, the lifetime of this component increases with

the increase of the reaction pH to prepare the sol. A reasonable
explanation for this pH effect could be given in terms of the
adsorption of Ru(bpy)2+
3 on the silica surface, the charge of
which is determined by pH. At pH 4 and 8 the silica has a
superficial excess of negative charge (point of zero charge is at
pH 2.5 (21)). Then, at these pH the amount of metallic complex
adsorbed and immobilized is higher than pH 1, because of the
electrostatic interaction between the metallic complex and the
TABLE 1
Effects of Precursor of the Sol-Gel Process, pH, and Drying Method of the Dip-Coated Thin Films on the MLCT State
Luminescence Intensity Decay Kinetics for Ru(bpy)2+
3
Precursor

Ph

Drieda

1 /ns (%)b

2 /ns (%)b

tL c

TEOS
MTEOS
MTEOS
MTMOS
MTMOS
MTMOS
MTMOS
MTMOS
MTMOS
MTMOS

1
1
1
8
8
8
4
4
4
4

O
O
O
R
O
O
R
O
O
O

460 (20)
280 (29)
260 (7)
970 (42)
440 (19)
350 (8)
710 (50)
310 (5)
420 (12)
290 (5)

1310 (80)
1350 (71)
1220 (93)
1980 (58)
1280 (81)
1330 (92)
2560 (50)
1350 (95)
1340 (88)
1310 (95)

6 months
4 months
8 months
1 month
3 days
1 month
13 days
6 days
15 days
1 month

a (O) films dried 18 h in oven at 70 C; (R) films dried at room temperature

(22 2 C) for at least 72 h.
b Estimated uncertainties 25%. The relative contribution to the decay kinetics
P
of each component is given in brackets and was calculated as (i i / i i )
100.
c Measurements were performed after the time t .
L


LOPEZ,
SENZ, AND GSPONER

126

negatively charged surface of the siloxane polymers. Therefore,

the lifetime of the component of short lifetime increases with
the increase of the pH.
The difference observed for the (R) films is that both lifetimes
(see Table 1) were much longer than the ones for the (O) films.
Furthermore, the relative contributions to the luminescence decay of each component were very similar, 42 and 58% for the
lifetimes 970 and 1980 ns, respectively, for MTMOS-derived
films at pH 8, and 50% for both components with lifetimes 710
and 2560 ns for MTMOS-derived films at pH 4.
A reasonable explanation for this effect of the drying temperature on the lifetimes for the dip-coated thin films, i.e., (O) and
(R) might be the following. It is well known that aging and thermal treatment under subsintering conditions result in increased
pore size within these types of glasses (22). This increase in pore
size may account for the observed increase in oxygen quenching and/or permeability. In consequence, this fact can explain
the decrease of the lifetimes in the (O) films and the minor contribution of the short lifetime component (1 ) since this lifetime
would correspond to the Ru(bpy)2+
3 complex located in sites
which were more accessible to oxygen.
To confirm this effect of drying temperature of the films on the
luminescence lifetimes 1 and 2 we carried out measurements
of luminescence intensity decay with individual films thermally
dried 1 h in oven at temperatures of 80, 103, and 150 C. The
dip-coated thin films used were MTMOS and ETEOS derived at
a molar ratio water/precursor of 2 and pH 4. The results obtained
are shown in Table 2.
The lifetimes decreased with the increase in the drying temperature of the films, showing at the same time a red shift in the
emission spectrum (see Table 2). The emission max underwent
a shift from 589 to 608 nm for the MTMOS precursor as the film
drying temperature was increased from 80 to 150 C; when using
ETEOS precursor, the resulting values were 599 to 606 nm when
the drying temperature varied from 80 to 103 C. This spectral

shift has also been suggested by Avnir et al. for rhodamine 6G

in silica gels (23).
Drying at room temperature cannot fully release the solvent
molecules from gels, especially those chemically bonded to
silanol groups. This can only cause the blue shift under drying as it was discussed above. In contrast, because of the drying at higher temperatures, even the chemically bonded solvent
molecules gradually evaporate, leaving the complex in relatively
vacant gel pores. In these pores the rotational movements of the
excited Ru(bpy)2+
3 complex are also expected, allowing the excited state to be relaxed and resulting in the red shift on the
luminescence peak and a decrease in the lifetimes. These results
confirm the differences observed in the lifetimes according to
the drying temperature of the (O) and (R) films shown in Table 1.
In a recent study, Yoko and colleagues (12) have observed an
inverse effect of drying temperature on the luminescence lifetime of Ru(bpy)2+
3 in the films. They found an increase in the
lifetime on drying temperature in TEOS-derived films. However, the results in that paper are not largely comparable with
our results. Some differences to be taken into account are the
following. First, the time used to dry the films were 10 min in
Ref. (12) and 60 min in our work. Secondly, in both works the
relative contributionsP
to the decay kinetics of each component
calculated as (i i / i i ) 100were also very different.
For example, 55 and 45% for the short and long lifetimes, respectively, in films dried at 100 C in Ref. (12); vs 710 and
9390% for the short and long lifetimes, respectively, in films
of MTMOSETEOS dried at 103 C in this work (see Table 2).
Then, we carried out quenching experiments by using the
luminescence intensity decay measurements in the presence
of a phenolic quencher. In these experiments we used a very
efficient quencher in aqueous solution such as 4-bromo-2,6dimehylphenol (24).
The treatment of the results was carried out calculating the
so-called preexponential weighted mean lifetime (M ) from the
parameters i and i according to

TABLE 2
Effects of Precursor of the Sol-Gel Process and Drying Temperature of the Dip-Coated Thin Films on the MLCT State Luminescence Intensity Decay Kinetics and Emission Spectrum for
Ru(bpy)23+

X
m
m
X
(i i )
i .
M =

Precursora

Drying T b
( C)

1 (%)c
ns

2 (%)c
ns

tL d
months

max
nm

MTMOS
MTMOS
MTMOS
ETEOS
ETEOS

80
103
150
80
103

475 (10)
330 (7)
200 (9)
375 (10)
275 (10)

1190 (90)
1050 (93)
755 (91)
825 (90)
710 (90)

1
1
1
1
1

589
599
608
599
606

Sol preparation at Rw 2 and pH 4.

The drying time was 1 h.
c Estimated uncertainties 25%. The relative contribution to the decay kinetics
P
of each component is given in brackets and was calculated as (i i / i i )
100.
d Measurements were performed after the time t .
L
b

[4]

It is important to note that in this form of calculation of M , the

preexponential factors are normalized and therefore, the average
lifetime obtained by Eq. [4] takes into account only the effect of
the dynamic processes on the luminescence intensity decay.
The M values were calculated in the presence and in the absence of quencher for various Q concentrations. The plot of the
o
/M vs [Q], where [Q] is the concentration of
data in the form M
4-bromo-2,6-dimehylphenol in the aqueous solution in contact
with the film, was fitted by using Eq. [2] but applied to lifetimes.
For example, the obtained values of K SV1 and f 01 were 240 M1
and 0.06, respectively for a MTMOS-derived film at pH 8 and
dried at 70 C, while K SV2 and f 02 were 0 M1 and 0.94, respectively (see Fig. 3). The last data, f 02 , indicate either that

LUMINESCENT PROPERTIES OF Ru(bpy)2+

3

FIG. 3. Stern-Volmer plot for the luminescence quenching of Ru(bpy)2+

3 in
a MTMOS (pH 8)-derived sol-gel-processed dip-coated thin (O) film by 4-Br2,6-dimethylphenol in aqueous solution at pH 12.

94% of the excited Ru(bpy)2+

3 complex is not accessible to the
quencher or that the quenching process has a very low efficiency
for this site or domain within the host matrix.
It can be observed from Tables 1 and 2 that a contribution
in the order of a 90% would correspond to the species of long
lifetime. Therefore, this species is also the least accessible to
quenching by oxygen, since the data in these tables were obtained from measurements in aerated aqueous solutions. From
these data, it would seem that the accessibility to the quencher
in the films prepared in this work is practically the same for
oxygen than for 4-bromo-2,6-dimethylphenol. This is an unexpected result and requires further confirmation. However, similar
results were obtained in the study of dynamic electron transfer in
aquo- and alco-SiO2 gels using tetramethoxysilane as precursor
(25). Only the fast component, which has a minor contribution
to the decay, was dynamically quenched by methyl viologen
adsorbed in Ru(bpy)2+
3 -doped aquogels.
On the other hand, the value of the K SV1 constant was lower
than the K SV corresponding to the luminescence quenching of
Ru(bpy)2+
3 by 4-bromo-2,6-dimethylphenol in aqueous solution
at pH 12 (2250 M1 ) (24). Then, these results show a significantly lower quenching of the metallic complex sequestered
within the film. This implies that in our case the probe is placed
either into a higher microviscosity microenvironment or in a location which the quencher cannot access. However, unlike the luminescence intensity quenching experiments, the data obtained
from measurements of luminescence intensity decay quenching
o
/M
were more precise. The estimated uncertainties on the M
were 25% based on the precision of the individual measurements of I (t) vs t. These uncertainties were also of the same
order than the residuals obtained by fitting of the luminescence
intensity decay measurements with the model of m = 2 in Eq. [3]
(see Fig. 2).
In order to probe the behavior of our films in the quenching
by oxygen, we carried out measurements of luminescence inten-

127

sity decay in water saturated with either argon, air or, oxygen.
Oxygen is frequently used in an important class of sensors
based on the decrease of luminescence intensity and lifetime of
transition-metal complexes. The experiments were performed
with MTMOS-derived sol-gel-processed dip-coated thin films
obtained at pH 4 and dried at 70 or 150 C.
o
/M vs [oxygen], the oxygen conThe data were plotted as M
centration dissolved in aqueous solutions in contact with the
film. Again, the data in the plot were well fitted by Eq. [2]. The
Stern-Volmer plot can be described satisfactorily by a sum of
two terms for the films dried a 70 C. The constants K SV1 and
K SV2 were 6000 and 0 M1 with contributions in the order
of 30 and 70%, respectively. Both constants were practically
equal to 4500 M1 with contributions in the order of 30 and
70% for the films dried at 150 C. Consequently, only one term
is more suitable for describing the Stern-Volmer plot for the
films dried at 150 C. For the K SV constants different from zero,
the obtained bimolecular quenching constants were in the order of (46) 109 M1 s1 , a diffusional rate constant for the
quenching by oxygen. Similar results, different values for the
K SV constants at low drying temperatures and very similar values for both K SV constants at high drying temperatures, were
obtained by other authors using a complex of ruthenium(II)
diphenylphenantroline in TEOS-derived sol-gel-processed films
(20).
In summary, a blue shift in the emission spectrum of the
MLCT excited state of Ru(bpy)2+
3 with respect to the aqueous
solution was observed for all the dip-coated thin films on the sol
to gel conversion, regardless of the reaction pH used to prepare
the sol. In all cases, the analysis of the luminescence intensity decay profiles was best described by a double-exponential
model. The parameters of the decay components depended principally on the thermal treatment used in the processing of the
films. The lifetimes decreased with the increase in the drying
temperature, at the same time the emission spectra showed a
red shift. A luminescence quenching of the ruthenium complex
in the films by dissolved oxygen in aqueous solution was also
observed. When a phenolic derivative was used as quencher,
the process was far less efficient. This means that the metallic complex sequestered within the film is placed either into a
higher microviscosity microenvironment or in a location which
the phenolic quencher cannot access. For both quenchers, the
o
/M vs [Q]) obtained from the luminescence
quenching plot (M
decay measurements could be fitted satisfactorily by a sum of
two terms of Stern-Volmer. This fact is indicative of the matrix microheterogeneity for the films and is fully consistent with
the biexponential nature of the luminescence intensity decay
profiles.

ACKNOWLEDGMENTS
Thanks are given to Consejo de Investigaciones Cientficas y Tecnologicas de
la Provincia de Cordoba (CONICOR) and SECYT de la Universidad Nacional
de Ro Cuarto.


LOPEZ,
SENZ, AND GSPONER

128
REFERENCES

1. Kalyanasundaram, K., Photochemistry of Polypiridines and Porphyrin

Complexes. Academic Press, London, 1992.
2. Desilvestro, J., Gratzel, M., Kavan, L., Moser, J., and Augustynski, J.,
J. Amer. Chem. Soc. 107, 2988 (1985).
3. OReagan, B., Moser, J., Anderson, M., and Gratzel, M., J. Phys. Chem.
94, 8720 (1990).
4. Hashimoto, K., Hiramoto, M., Kajiwara, T., and Sakata, T., J. Phys. Chem.
92, 4636 (1988).
5. Thomas, J. K., and Wheeler, J., J. Photochem. 28, 285 (1985).
6. Kamat, P. V., and Ford, W. E., J. Phys. Chem. 93, 1405 (1989).
7. Fan, J., Shi, W., Tysoe, S., Strekas, T. C., and Gafney, H. D., J. Phys. Chem.
93, 373 (1989).
8. Matsui, K., and Momose, F., Chem. Mater. 9, 2588 (1997).
9. Vera, D. M., Arguello, G. A., and Arguello, G. A., and Gsponer, H. E.,
J. Photochem. Photobiol. A 76, 13 (1993); Senz, A., and Gsponer, H. E.,
J. Colloid Interface Sci. 165, 60 (1994); Senz, A., and Gsponer, H. E.,
J. Phys. Org. Chem. 8, 706 (1995); Senz, A., and Gsponer, H. E., J. Colloid.
Interface Sci. 180, 193 (1996); Senz, A., and Gsponer, H. E., J. Colloid
Interface Sci. 195, 94 (1997); Lopez, S., Senz, A., and Gsponer, H. E.,
J. Colloid. Interface Sci. 224, 126 (2000).
10. Hashimoto, K., Hiramoto, M., Lever, A. B. P., and Sakata, T., J. Phys. Chem.
92, 1016 (1988).

11. Milder, S. J., Gold, J. S., and Kliger, D. S., J. Phys. Chem. 90, 548
(1986).
12. Innocenzi, P., Kozuka, H., and Yoko, T., J. Phys. Chem. B 101, 2285
(1997).
13. Castellano, F. N., Heimer, T. A., Tandhasetti, M. T., and Meyer, G. J., Chem.
Mat. 6, 1041 (1994).
14. Gsponer, H. E., J. Photochem. Photobiol. A 55, 233 (1990).
15. Carraway, E. R., Demas, J. N., DeGraff, B. A., and Bacon, J. R., Anal Chem.
63, 337 (1991).
16. Mills, A., and Thomas, M., Analyst 122, 63 (1997).
17. Demas, J. N., DeGraff, B. A., and Xu, W., Anal. Chem. 67, 1377 (1995).
18. Narang, U., Jordan, J. D., Bright, F. V., and Prasad, P. N., J. Phys. Chem.
98, 8101 (1994).
19. Dunbar, R. A., Jordan, J. D., and Bright, F. V., Anal. Chem. 68, 604
(1996).
20. Baker, G. A., Wenner, B. R., Watkins, A. N., and Bright, F. V., J. Sol-Gel
Sci. Technol. 17, 71 (2000).
21. Pierre, A. C., Introduction to Sol-Gel Processing. Kluwer Academic,
Publishers, Boston, 1998.
22. Brinker, C. J., and Scherer, G. W., Sol Gel Science. Academic Press,
New York, 1989.
23. Avnir, D., Levy, D., and Reisfield, R., J. Phys. Chem. 88, 5956 (1984).
24. Senz, A., Chemistry Doctor thesis, Ro Cuarto University, 2000.
25. Castellano, F. N., and Meyer, G. J., J. Phys. Chem. 99, 14742 (1995).