Photochemical behaviour of a new 1,2,3,4-tetrahydroxanthylium fluorescent dye with "rhodamine-like" structure in liquid media and adsorbed onto a TiO2 photo-responsive substrate

Article abstract of DOI:10.1016/j.dyepig.2016.01.030

A new "rhodamine-like" fluorescent NIR dye based on the 9-(2′-carboxyphenyl)-6-(N,N-diethylamino)-1,2,3,4-tetrahydroxanthylium system was synthesized. Its spectral-luminescent properties were investigated both in solution and adsorbed onto a TiO₂

Full text of DOI:10.1016/j.dyepig.2016.01.030

Dyes and Pigments 128 (2016) 279e288
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Photochemical behaviour of a new 1,2,3,4-tetrahydroxanthylium
fluorescent dye with “rhodamine-like” structure in liquid media and
adsorbed onto a TiO2 photo-responsive substrate
a
a
b
b
a, *
~
D.S. Conceiçao , D.P. Ferreira , Y. Prostota , P.F. Santos , L.F. Vieira Ferreira
a
ꢀ
Centro de Química-Física Molecular and IN-Institute of Nanoscience and Nanotechnology Instituto Superior Tecnico, Universidade de Lisboa, Av. Rovisco
Pais, 1049-001 Lisboa, Portugal
b
ꢀ
Centro de Química, Vila Real Universidade de Tras-os-Montes e Alto Douro, 5001-801 Vila Real, Portugal
a r t i c l e i n f o
a b s t r a c t
A new “rhodamine-like” fluorescent NIR dye based on the 9-(20-carboxyphenyl)-6-(N,N-diethylamino)-
1,2,3,4-tetrahydroxanthylium system was synthesized. Its spectral-luminescent properties were inves-
tigated both in solution and adsorbed onto a TiO2 photoactive solid substrate. Regarding liquid samples,
the highest fluorescent quantum yield and fluorescence lifetime were obtained in dichloromethane
(øF ¼ 0.28, tF ¼ 2.67 ns) while in ethanol and acetonitrile we observed the largest Stokes Shift, indicating
a twisted intramolecular charge transfer mechanism. Additionally, the behaviour of the dye adsorbed
onto TiO2 revealed some interesting features regarding the formation of the radical derived from the dye
cation, following the electron injection in the conduction band of the semiconductor. Laser flash
photolysis and ground state diffuse reflectance were used to evaluate and discriminate the mechanism of
photoexcitation presented, as well as to characterize the photostability of the molecule. In this case, two
different scenarios could be recognized: the self-photosensitization pathway and the photocatalytic
mechanism.
Article history:
Received 25 November 2015
Received in revised form
27 January 2016
Accepted 28 January 2016
Available online xxx
Keywords:
Rhodamines
1,2,3,4-tetrahydroxanthylium
Luminescence
Titanium oxide
Radical cation
Transient absorption
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
the latter case, dye-TiO2 interactions are the primary and most
important event of the device. This is due to the injection of an
Rhodamine dyes are one of the most explored families of mol-
ecules regarding fluorescent applications, mostly due to their high
fluorescence quantum yields, high photostability, high extinction
coefficients and low intersystem crossing quantum yield. Classic
rhodamines such as Rhodamine 6G, Rhodamine B and Rhodamine
101 are generally employed as laser dyes and fluorescence stan-
dards [1e5] but the demand for new rhodamine dyes is still sig-
nificant, considering the extensive range of different systems in
which fluorescent dyes can be applied to. For instance, near-
infrared (NIR) responsive dyes with high solubility in aqueous so-
lutions, for biological applications [6,7], and visible responsive dyes
with high stability, together with an appropriate redox potential
and a low quantum yield of triplet formation, in the case of
photovoltaic applications, specifically, in dye sensitized solar cells
(DSSCs), which are titanium oxide (TiO2) based systems [8e11]. In
electron from the dye excited state into the conduction band (CB) of
the TiO2, which initiates the dye sensitization process in this kind of
photovoltaic systems. The respective recombination between
injected electrons and oxidized dyes, which occurs after the elec-
tron injection, is also entirely dependent on the dye-TiO2
interactions.
The well-known photocatalytic properties of TiO2 are derived
from the formation of photogenerated charge carriers (hole and
electron) which occurs upon the absorption of ultraviolet (UV) light
corresponding to the band gap energy of the semiconductor. Thus,
as electrons are promoted from the valence band to the conduction
band, they can generate electronehole pairs as shown in Equation
(1):
TiO2 þ hv ðl < 390 nmÞ/eꢀ þ hþ
(1)
Valence band (hþ) potential is positive enough to generate hy-
droxyl radicals (HOꢁ) at TiO2 surface and the conduction band (ee)
* Corresponding author.
E-mail address: LuisFilipeVF@ist.utl.pt (L.F. Vieira Ferreira).
http://dx.doi.org/10.1016/j.dyepig.2016.01.030
0143-7208/© 2016 Elsevier Ltd. All rights reserved.

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
280
potential is negative enough to reduce molecular oxygen into
2. Experimental section
superoxide ion radical (Oꢁ2ꢀ) e Equations (2) and (3), respectively.
2.1. General
H2O þ hþ/OH þ Hþ
O2 þ eꢀ/Oꢁ2ꢀ
(2)
Titanium (IV) oxide (nanopowder, <25 nm particle size) was
obtained from SigmaeAldrich and microcrystalline cellulose
(powder) was obtained from Fluka. Solvents used in the measure-
ment of spectral-luminescent properties were of spectroscopic
grade. Reagents and solvents used in the synthesis of the dye were
of commercial origin and, unless otherwise stated, used as received.
All reactions were monitored by thin-layer chromatography on
aluminium plates pre-coated with Merck silica gel 60 F254
(0.25 mm). 9-(20-carboxyphenyl)-6-(N,N-diethylamino)-1,2,3,4-
tetrahydroxanthylium perchlorate (3) [25] and 3-methylimidazo
[1,5-a]pyridine (5) [26] were prepared as previously described. 1H
and 13C NMR spectra were recorded on a Bruker ARX400 spec-
trometer (at 400.13 and 100.62 MHz, respectively); d in ppm rela-
tive to residual solvent signals and J in Hz. IR spectra were obtained
on a Unicam Research Series FT-IR spectrometer; ymax in cmꢀ1. High
resolution electrospray ionization time-of-flight (ESI-TOF) mass
spectra were measured with a VG AutoSpec M spectrometer.
Melting points were determined in open capillary tubes in a Buchi
535 melting point apparatus and are uncorrected.
(3)
These radicals will participate in the degradation of the dye.
However, two different mechanisms can be mentioned regarding
the dyeeTiO2 interaction and the consequent formation of radicals
ꢁ
such as HOꢁ, Oꢀ2 , hydroxylated intermediates and the dye radical
cation. As already well stated in the literature [12,13], these two
pathways can be defined as the photocatalytic mechanism, and the
self-photosensitized pathway, depending of the excitation source,
UV irradiation or visible light illumination, respectively. One of the
intermediates formed in these reactions is H2O2, which is also
photoactive and can be split in HOꢁ and OHꢀ after irradiation
[14,15].
Oꢁ2ꢀ þ Hþ/ꢁO2H
(4)
(5)
(6)
(7)
(8)
2ꢁO2H/H2O2 þ O2
2.2. Synthesis of the dye
H2O2 þ Oꢁ2ꢀ/ꢁOH þ O2 þ OHꢀ
H2O2 þ eꢀ/ꢁOH þ OHꢀ
H2O2 þ hþ þ OHꢀ/H2O þ O2H
2.2.1. 3-Methylimidazo[1,5-a]pyridine-1-carbaldehyde (6)
To a solution of 3-methylimidazo[1,5-a]pyridine (5) (1.50 g,
11.4 mmol) in anhydrous dimethylformamide (10 mL), under
vigorous stirring at 0e5 ꢂC, was added dropwise freshly distilled
phosphorus oxychloride (1.10 mL, 11.8 mmol) during 30 min. Once
the addition was complete, the reaction mixture was kept at room
temperature for 30 min, and then heated to 100 ꢂC for additional
3 h. After cooling to room temperature, the reaction mixture was
poured into water (100 mL) and made alkaline with 20% aqueous
sodium hydroxide. The resulting mixture was extracted with
dichloromethane (3 ꢃ 20 mL), washed with water (50 mL), dried
over anhydrous sodium sulfate and evaporated to dryness under
reduced pressure. The resulting solid residue was recrystallized
from propan-2-ol to afford 2 (1.30 g, 71%) as yellow flakes. Mp
132e134 ꢂC. IR (KBr) ymax: 755, 830, 1027, 1295, 1484, 1640, 1871,
Therefore, it can be used as an electron scavenger, to mimic
some of the effects presented in the dye-TiO2 system, namely the
possible formation of a radical cation.
The technique of laser flash photolysis has been extensively
addressed in the past to characterized TiO2 based materials, in what
concerns the kinetics of the electronehole separation and recom-
bination [16e18]. In this way, it is possible to identify, until several
nanoseconds after laser excitation, the trapped electron and the
trapped hole, at the surface of the semiconductor. Following that, it
is also expected and possible to obtain, by means of this technique,
data that supports the interaction of the dye with the semi-
conductor and the intermediates formed after irradiation.
2773. 1H NMR (CDCl3, 298 K)
(1H, t, J ¼ 7.8), 7.85 (1H, d, J ¼ 6.9), 8.18 (1H, d, J ¼ 8.9), 10.00 (1H, s).
13C NMR (CDCl3, 298 K)
: 12.5,115.1,119.4,121.8,125.7,129.2,133.9,
d: 2.66 (3H, s), 6.88 (1H, t, J ¼ 6.7), 7.18
d
It is assumed from previous studies regarding rhodamines and
rhodamine-like molecules that the molecular structure and nature
of the solvent play an important role in the mechanisms that
control the fluorescence yields of rhodamine dyes. Specifically, it
was found that deactivation mechanisms of the first excited singlet
state involving intramolecular charge transfer (ICT) interactions
were directly correlated with structural modifications that enabled
a larger Stokes shift [19,20]. Taking all these features into account, a
new dye (7) derived from the 9-(20-carboxyphenyl)-6-(N,N-dieth-
137.2, 185.1. HRMS (ESI-TOF) m/z: 160.06391 (Mþ; calc. for
C9H8N2O: 160.06374).
2.2.2. 9-(20-carboxyphenyl)-6-(N,N-diethylamino)-4-[(3-
methylimidazo[1,5-a]pyridin-1-yl)methylidenyl]-1,2,3,4-
tetrahydroxanthylium perchlorate (7)
A solution of 2 (0.07 g, 0.43 mmol) and 3 (0.20 g, 0.42 mmol) in
freshly distilled acetic anhydride (5 mL) was heated under reflux for
5 min. The reaction mixture was cooled to room temperature and
the residue formed was filtered off, and washed with ethyl acetate
(10 mL) and diethyl ether (5 mL) to afford chromatographically
pure 4 (0.17 g, 65%) as a green solid with metallic lustre. Mp
246e248 ꢂC (dec.). IR (KBr) ymax: 706, 754, 841, 1059, 1176, 1254,
ylamino)-1,2,3,4-tetrahydroxanthylium system possessing
a
“rhodamine-like” structure was synthesized (Scheme 1) and was
characterized in terms of its behaviour in liquid and solid samples.
In the first case, three different solvents were used e dichloro-
methane, ethanol and acetonitrile - and a complete photochemical
evaluation was performed. In solid samples, the main goal was to
study the behaviour of the dye adsorbed onto a photoactive ma-
terial such as TiO2, in the presence of light. Cellulose was used as
the inert standard material, for comparison purposes. The scope of
this work targets a better understanding of the mechanisms
involved in the photochemical behaviour of this new synthesized
dye, in substantially different environments.
1319,1437,1532,1581,1628,1729. 1H NMR (DMSO-d6, 323 K)
d: 1.26
(6H, t, J ¼ 7.0), 1.73e1.76 (2H, m), 2.28e2.37 (2H, m), 2.75 (3H, s),
3.34e3.43 (2H, m), 3.64 (4H, q, J ¼ 7.0), 6.86 (1H, d, J ¼ 9.4), 7.09
(1H, dd, J ¼ 2.4, 9.4), 7.17 (1H, t, J ¼ 6.8), 7.30 (1H, d, J ¼ 2.4), 7.38
(1H, d, J ¼ 7.5), 7.49 (1H, m), 7.74 (1H, t, J ¼ 7.4), 7.84 (1H, t, J ¼ 7.5),
8.19 (1H, d, J ¼ 7.4), 8.42e8.48 (3H, m). 13C NMR (DMSO-d6, 298 K)
d: 12.5, 20.5, 20.9, 25.9, 26.5, 44.8, 95.7, 114.8, 115.4, 155.7, 118.0,
119.9, 121.1, 124.9, 126.5, 127.3, 128.6, 129.0, 129.7, 129.8, 130.8,

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
281
Scheme 1. Synthesis of dye 7.
132.9,134.6,137.8, 141.6, 153.5,156.8, 163.6, 166.4, 171.9. HRMS (ESI-
recorded continuously in specific time intervals: 1 min, 5 min,
TOF) m/z: 518.24278 (Mþ; calc. for C33H32N3O3: 518.24381).
10 min, 20 min, 30 min, 40 min, 50 min and 60 min.
2.5. Laser-induced luminescence (LIL): fluorescence emission
2.3. Sample preparation
quantum yield determinations
Liquid samples were prepared by diluting the dye in the three
solvents used in this work: acetonitrile, chloroform and dichloro-
methane. The optical density was adjusted to 0.6, in all cases. Solid
samples were prepared by solvent evaporation under constant
stirring, followed by an overnight vacuum drying cycle. In that case,
a solution of the dye in acetonitrile was used, with a concentration
of 1.0 ꢃ 10ꢀ4 M, followed by its adsorption on 200 mg of solid
substrate. Then, the solvent was allowed to evaporate until the
powder remained on the flask. For solid samples, concerning TiO2,
all steps were prepared in the dark, to avoid immediate dye
degradation. TiO2 dispersions in acetonitrile solutions of the dye
were prepared with a dye concentration of 1.0 ꢃ 10ꢀ4 M, mixing
40 mg of TiO2 in 4 mL of solution.
Schematic diagrams of the LIL system are also presented in
Ref. [21]. A N2 laser (PTI model 2000, ca. 600 ps of full width at half
maximum (FWHM), ~1.0 mJ per pulse, 5 Hz), was used in the laser-
induced luminescence experiments, the excitation wavelength
being 337 nm for the liquid samples. In that case, the optical density
at the excitation wavelength (337 nm from the N2 laser) used for
both the unknown and standard samples was 0.6, following the
methodology of a previous publication [22]. In our set-up the errors
in the determination of øF are 3%. The fluorescence quantum
yields of dye 7 in acetonitrile, chloroform and dichloromethane
were calculated relative to the standard from their respective
average fluorescence peak areas, and the published quantum yield
of the standard, which in this case was Rhodamine 101 (øF¼ 0.98 in
ethanol). For solid samples, in the case of the dye's adsorption onto
TiO2, its fluorescence emission was acquired by means of laser
excitation at 500 nm, and by using a cut-off filter of 590 nm,
together with an optical density filter of 2.2%.
2.4. UVevisible absorption spectra and ground state diffuse
reflectance spectra (GSDR)
Steady-state absorption spectra were recorded with the use of a
Camspec M501 single beam scanning UV/Visible spectrophotom-
eter at room temperature in the spectral range from 190 to
1100 nm. The optical densities were measured using a UV quartz
cuvette (1 cm path length). Ground-state absorption studies were
performed using a homemade diffuse reflectance laser flash
photolysis setup, with a powerful 150 W tungsten-halogen lamp as
monitoring lamp, triggering the system in the normal way but
without the laser fire, and in this way recording the lamp profile for
all samples under study and also for two standards, barium sulfate
and magnesium oxide powders. A fixed monochromator coupled to
an ICCD with time gate capabilities was used for detecting the
reflectance signals. The reflectance, R, from each sample was ob-
tained in the UVeViseNIR spectral regions and the remission
function, F(R), was calculated using the KubelkaeMunk equation
for optically thick samples. The remission function is
F(R) ¼ (1 ꢀ R)2/2R. Details regarding the data treatment can be
found in Ref. [21] and references quoted therein. For the photo-
catalytic evaluation of the dye onto TiO2, the GSDR spectra were
2.6. Fluorescence lifetimes determination
Fluorescence lifetimes were determined, in the case of liquid
samples, using Easylife V™ equipment from OBB (Lifetime range
from 90 ps to 3 ms). This technique uses pulsed light sources from
different LEDs (630 nm in this case) and measures fluorescence
intensity at different time delays after the excitation pulse. In this
case a 695 nm cut-off filter was used. The instrument response
function was measured using a Ludox scattering solution. FelixGX
software from OBB was used for fitting and analysis of the decay
dynamics.
2.7. Laser flash photolysis: tripletetriplet transient absorption
spectra
The schematic diagrams for the laser flash photolysis technique
and the used methodology are presented in Refs. [21,23,24]. For the

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
282
laser flash spectra shown, the fourth harmonic of a Nd:YAG laser
(266 nm, ca. 6 ns full-width half maximum, FWHM) from B. M.
Industries (Thomson-CSF, model Saga 12-10, France) was used, by
employing the transmission mode, in the case of liquid samples,
and reflectance mode, in the case of solid substrates. In the first
case, the monitoring lamp used was a 150 W quartz tungsten-
halogen in an Oriel housing, whereas in the latter case a 450 W
xenon lamp was applied to the geometry. The light arising from the
irradiation of the samples by the laser pulse, with approximately
60 mJ per pulse, was collected by a collimating beam probe coupled
to an optical fibre (fused silica) and detected by a gated intensified
charge coupled device Oriel model Instaspec V, (Andor ICCD, based
on the Hamamatsu S57 69-0907). The ICCD was coupled to a fixed
compact imaging spectrograph (Oriel, model FICS 77441). The
system can be used either by capturing all light emitted by the
sample or in a time-resolved mode by using a delay box (Stanford
Research Systems, model D6535), whenever needed. The ICCD has
high speed gating electronics (about 2.3 ns) and intensifier,
covering at least the 250e900 nm wavelength range. Liquid sam-
ples were prepared with an optical density of 0.9 at the excitation
wavelength, and oxygen removal was achieved by argon bubbling
for about 15 min. Transient absorption data are reported as change
Fig. 1. Absorption spectra of dye 7 in acetonitrile (blue), dichloromethane (red), and
ethanol (green). (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
Ethanol and acetonitrile are the more polar solvents, with
Dimroth's ET and Kosower's Z solvent parameters of 51.9 and 79.6,
in the case of ethanol and 46 and 71.3, in the case of acetonitrile.
Dicloromethane possesses an ET parameter of 41.1 and a Z-value of
64.2 and within the scope of this work, it is the solvent with the
lowest polarity. The solvent polarity parameters can be directly
correlated to Stokes shift (the shift between the ground state and
excited state transitions of dyes), i.e. the large Stokes shift is also
indicative of the charge transfer transition. The large magnitude of
the Stokes shift indicates that the excited state geometry could be
different from that of the ground state. Generally speaking, in-
crements in the Stokes shift value are a function of an increase in
solvent polarity, which indicates an increase in the dipole moment
of the excited state. In such cases, the relaxed excited singlet-state
(S1) will be energetically stabilized compared to the ground state
(S0) and a significant red shift of fluorescence will be observed. The
Stokes Shift values will be presented and discussed in the following
section. The absorption bands of the dye in both acetonitrile and
dichloromethane are red-shifted, compared to the most polar sol-
vent, ethanol. This evidences that the observed lmax is, as expected
for this type of molecules, a complex function of both polarity and
polarisability of the solvent. The same behaviour is also witnessed
for the fluorescence emission'. This trend is in accordance with the
solvatochromic behaviour of rhodamine B and rhodamine 6G,
already studied and reported in the literature [27]. Thus, it can be
expected that the solvent polarity features present the main
contribution in solvatochromism of rhodamine and rhodamine-like
dyes.
of optical density
Abs ¼ 100
DOD or percentage of absorption defined by %
D
Jt/Jo ¼ 100 (1 ꢀ Jt/Jo), where Jo and Jt are the trans-
mitted light from sample before exposure to the exciting laser pulse
and at time t after excitation, respectively.
3. Results and discussion
3.1. Synthesis
The target “rhodamine-like” dye 7 was synthesized in good yield
via simple condensation of 9-(20-carboxyphenyl)-6-(N,N-dieth-
ylamino)-1,2,3,4-tetrahydroxanthylium perchlorate (2) with 3-
methylimidazo[1,5-a]pyridine-1-carbaldehyde (5), in acetic anhy-
dride (Scheme 1). The synthesis of the starting 1,2,3,4-
tetrahydroxanthylium salt 3 was carried out by an improved
method developed by one of us involving the cyclocondensation of
2-[4-(N,N-diethylamino)-2-hydroxybenzoyl]benzoic acid (1) with
cyclohexanone (2) [25]. Aldehyde 6 was obtained through Vils-
meiereHaack formylation of 3-methylimidazo[1,5-a]pyridine (5),
readily prepared by cyclocondensation of commercially available 2-
picolylamine (4) with acetic anhydride, in the presence of an
equimolar amount of p-toluenesulfonic acid monohydrate [26].
3.2. Ground state UVevisible spectra of dye 7, in solution
3.3. Fluorescence emission spectra and fluorescence lifetime decays
The absorption profile of the dye studied in this work, in three
different solvents, acetonitrile, ethanol and dichloromethane, is
presented in Fig. 1. It is well establish that the solvatochromic
behaviour of rhodamine dyes and their derivatives arise from
different types and extent of solute and solvent interactions in the
solution media. These interactions mainly include non-specific
(dielectric enrichment) or specific interactions, such as hydrogen
bonding, proton transfer, and intermolecular charge transfer (ICT).
In fact, there are competition mechanisms between the sol-
uteesolute and soluteesolvent interactions that determine the
nature of solvatochromic behaviour of the solute molecules [27,28].
When compared with the classic Rhodamine B, it can immediately
be seen the large bathochromic deviation of the absorption max-
ima, which can imply the importance of the fused imidazole-
pyridine rings in the increase of the resonance of the molecule
(typically, the absorption maxima of rhodamine B is approximately
543 nm, depending on the solvent).
The laser induced fluorescence emission of dye 7 in the three
solvents studied in this work is presented in Fig. 2. The spectra
show a significant emission in the range of 650e900 nm, pre-
senting a huge displacement when compared with the classical
rhodamine
B (with maxima at approximately 560e575 nm,
depending on the solvent). This deviation is indeed higher than the
one observed in the absorption bands, pointing out to the occur-
rence of intramolecular charge transfer mechanisms of deactivation
of the excited state. This feature is also complementary to the large
Stokes shift that is usually present in dyes that exhibit this
behaviour.
Table 1 presents the most important parameters obtained for
the dye under study. Together with internal conversion and inter-
system crossing deactivation pathways, ICT can also play an
important role in the deactivation of the lowest singlet state,
contributing to the reduction of the quantum yield for the

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
283
Dye 7 in
Dichloromethane
a)
b)
1
0.8
0.6
0.4
0.2
0
Ludox
2000
Fit Curve
200
20
600
700
800
900
1000
55
60
65
70
Wavelength (nm)
Lifetime (ns)
Fig. 2. a) Fluorescence emission spectra of dye 7 in acetonitrile (blue), dichloromethane (red) and ethanol (green). b) Fluorescence decay curve, instrument responsive curve (IR),
and fitting of the dye, in dichloromethane. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Table 1
0.08
0.06
0.04
0.02
0
Spectral data of dye 7 in acetonitrile, dichloromethane and ethanol.
3Dye*
labs (nm) lemi (nm) øF
tF (ns) kF
Stokes shift (nm)
CH3CN 695
CH2Cl2 699
EtOH 680
741
735
725
0.16 1.67
0.28 2.67
0.20 1.98
1.0 ꢃ 108 46
3Dye*
1.1 ꢃ 108 36
1.0 ꢃ 108 45
Dyedepl
fluorescence emission, øF, and for the increase of the rate constant
associated with the non-radiative path of deactivation. This effect is
more significant in polar solvents, and, indeed, that seems to be the
case, when analyzing the fluorescence quantum yields and fluo-
rescence lifetimes. Dichloromethane exhibits the highest value of
øF and tF, 0.28 and 1.98 ns, respectively. Its fluorescence lifetime
profile is also shown in Fig. 2. In acetonitrile, the lowest values of
fluorescence quantum yield and fluorescence lifetime were ob-
tained, 0.16 and 1.67 ns, respectively.
-0.02
-0.04
Dyedepl
300
400
500
600
700
800
Wavelength (nm)
Fig. 3. Transient absorption spectrum of dye 7, in acetonitrile, in an argon purged
sample.
dye molecule, previously mentioned. The same pattern, of tri-
pletetriplet absorption and ground state depletion is already well
established in the literature, for instance in the cases of rhodamine
101, rhodamine 6G, rhodamine 3G and sulforhodamines 101 and B
[30,31]. No significant information was observed for air equili-
brated samples.
3.4. Transient absorption analysis
The transient absorption measurements, obtained by means of
laser flash photolysis in transmittance mode, enables the moni-
toring of transient species that can be formed after laser excitation.
In this case, the dye was excited at 266 nm, in acetonitrile. Aceto-
nitrile was selected as the solvent of use due to the lower fluores-
cence emission of the dye in it, and because the solid samples
(cellulose and TiO2 substrates) were also prepared with this sol-
vent. It is well known that titanol groups act as surface traps for the
photogenerated hole, preventing the recombination rate and,
theoretically, increasing the photo-activity of the semiconductor
[29]. These titanol groups are created from the residual water that
exists at the surface. Every solvent has a residual content of water,
and therefore, to minimize this effect at the solvent evaporation
step, acetonitrile was the solvent of choice since it is the one with
the smallest amount of water, when compared with ethanol and
dichloromethane (Acetonitrile: 0.007%; Ethanol e 0.2%; Dichloro-
methane e 0.02%). Fig. 3 displays the results obtained by this
technique in an argon purged sample with an acquisition gate
3.5. Ground state diffuse reflectance (GSDR) spectra of dye 7,
adsorbed onto cellulose and TiO2
The GSDR spectra of the dye 7 were initially obtained in cellu-
lose, the inert substrate. Samples of the dye adsorbed onto micro-
crystalline cellulose were prepared with five different
concentrations, as depicted in Fig. 4. The first spectra shown (Fig. 4
e Top) are quite similar to the one obtained in solution, with a
rather well defined vibronic structure, for all five concentrations,
and a maximum absorption wavelength peaking at 680 nm. A
crucial characteristic for efficient dyes, used in photochemical ap-
plications such as photovoltaic devices, is the presence of suitable
functional groups in their structure which can strongly bind to the
surface of semiconducting metal oxides. It is well known that when
considering mainly covalent attachment brought about by directly
linking groups of interest or via linking agents, groups such as
silanyl, amide, carboxyl and phosphonate have been shown to form
stable linkages. In this case, the anchoring group of the molecule,
which reacted with surface hydroxyl groups of the metal oxide
surface e titanol groups e is the carboxylic acid. Compared with
cellulose, the spectra of the dye adsorbed onto TiO2 nanopowder
(Fig. 4 e bottom), is completely different. Besides the typical UV
absorption band of TiO2 with maximum at approximately 350 nm,
the plot shows that the vibronic definition of the dye is absent, for
the five concentrations used. The effect is most probably due to the
width of 2 ms, and immediately after the laser pulse. The spectrum
show two important features: the first one is the ground state
depletion, which assumes negative values since it is the initial
species presented in the sample, before excitation. The minimum
for the depletion is located at approximately 695 nm, equivalently
to the UVevisible absorption spectra presented previously. In
addition to that, the minima peaking at approximately 490 nm and
330 nm can also be identified, Dyedepl, with lower intensities, when
compared with the main depletion; the second important feature is
the tripletetriplet absorption band, 3Dye*, located between 300
and 585 nm, only interrupted by the two small depletions of the

~
284
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
0.02 umol/g
0.1 umol/g
0.25 umol/g
0.5 umol/g
1.0 umol/g
0 min
1 min
5 min
10 min
20 min
30 min
40 min
50 min
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
a)
0.05
0.04
0.03
0.02
0.01
0
300 400 500 600 700 800
400 450 500 550 600 650 700 750
Wavelength (nm)
Wavelength (nm)
Not adsorbed
0.02 umol/g
0.1 umol/g
0.25 umol/g
0.5 umol/g
1.0 umol/g
0 min
1 min
b)
0.2
0.15
0.1
5 min
0.1
0.08
0.06
0.04
0.02
0
10 min
20 min
30 min
40 min
50 min
60 min
0.05
0
300 400 500 600 700 800
400 450 500 550 600 650 700 750
Wavelength (nm)
Wavelength (nm)
Fig. 4. GSDR spectra of dye 7 adsorbed onto Cellulose (Top) and TiO2 (Bottom).
100
80
60
40
20
0
c)
Photocatalytic mechanism
immediate interaction of the dye-TiO2 system with the light from
the excitation lamp (inherent to the technique of GSDR). In that
case, the dye radical cation is instantly formed, leading to the loss of
conjugation by the molecule and, consequently, the disappearance
of its previously well defined absorption bands, with a shift to
lower wavelengths. Following that, only a single absorption band
could be detected, peaking at approximately 590 nm. On the other
hand, in the case of a fresh TiO2 dispersion in a solution of the dye,
the immediate monitoring of the GSDR depicts the same absorption
profile found in cellulose, apart from the typical TiO2 absorption
peaking on the UV range. Within the range of concentrations pre-
Self-photosensitized pathway
Dye 7 onto Cellulose
0
20
40
60
Time (min)
sented, the sample of 0.5 mmol/g was chosen as the best one to
Fig. 5. Photodegradation of dye 7, adsorbed onto TiO2. a) Photocatalytic
mechanism Self-photosensitized pathway; b) Self-photosensitized pathway; c)
perform the remaining photochemical characterization, without
attaining aggregation phenomena. To confirm that, the fluores-
cence emission of the dye adsorbed onto cellulose using all five
concentrations was monitored. The maximum fluorescence emis-
þ
Degradation kinetics, including the photodegradation effect of the Xe lamp, on a dye e
Cellulose sample.
sion intensity was obtained for the concentration of 0.5
(data not shown).
mmol/g
being excited, enabling both the photocatalytic mechanism and
the self-photosensitized pathway. The plot shows that, as stated
before, only the band peaking at approximately 590 nm can be
seen. However, after approximately 1 min of exposure, a hyp-
sochromic shift can be observed, of approximately 65 nm (to
~525 nm). This effect was surely due to the presence of another
neo-formed species. Regarding Fig. 5b), it presents the dye
photodegradation profile when enabling only the self-
photosensitized pathway. For that purpose, a cut-off filter of
414 nm was used, therefore preventing the direct excitation of
the TiO2 substrate. In general, the features presented by Fig. 5a
and b could mean that, almost immediately after light absorp-
tion, the dye could be promoted to its excited state and would
transfer an electron to the conduction band of the semiconductor
leading to the formation of the dye radical cation, as suggested
by Equations (9) and (10). Therefore, the first curve of both
spectra (t ¼ 0) traduces the initial contact between the sample
and the excitation light, in which there is still some ground state
absorption profile from the original state of the dye, balanced
3.6. Photocatalytic degradation of the dye
The photocatalytic degradation of dye 7, adsorbed onto TiO2,
was monitored, as mentioned in the experimental section, by
means of a 150 W tungsten-halogen lamp, which constantly
irradiated the sample, for 60 min. As already stated in the
introductory section of this work, two different mechanisms and
pathways can be presented in this types of systems, where an
organic dye is adsorbed onto TiO2. Depending if the excitation
source is targeting directly the dye or the semiconductor,
different kinetics can be obtained for the degradation of the
molecule, as presented in Fig. 5. Fig. 5a) shows the F(R) function
of the sample prepared with a concentration of 0.5
mmol/g,
monitored for 60 min of lamp exposure, with no cut-off filters.
The lamp irradiation profile applies in the range of approxi-
mately 250e1000 nm, and this feature means that within this
range of wavelengths, both the semiconductor and the dye are

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
285
with the instantly generation of the radical cation. The proposed
mechanism is further explored in this section.
transfer from the dye to the TiO2 material. In what concerns the
luminescence spectrum, the same dual-band profile remains, with
a maximum at around 735 nm, relative to the fluorescence emis-
sion of the dye, and another located at approximately 600 nm,
concerning the radical cation, Dyeꢁþ. The absorption and emission
band placed at 655 nm and 735 nm, respectively, that endured
minutes of lamp excitation and stirring, are certainly related not
only with the molecules that did not undergo through electron
transfer processes but also with the molecules in the dispersion
that were not adsorbed onto the powdered TiO2 material. The
macroscopic behaviour of the dispersion is illustrated in Fig. 6b).
Immediately after lamp excitation (a few seconds), the pattern is
the one presented: a violet color starts to appear from the top of the
flask, towards the semiconductor powder, implying that the dye is
transferring and injecting e-to the TiO2 material.
To properly ensure that this phenomena is due to electron
transfer, and that, possibly, the radical cation is formed, a solution
of the dye of interest, in acetonitrile, together with hydrogen
peroxide (H2O2, 1:5 v/v), was prepared, and submitted to lamp
irradiation. It is well known and it was already mentioned in the
introductory section of this work that H2O2 can trap the photoin-
duced eꢀ originated from TiO2 to stabilize the paired eꢀ-hþ [14]. In
a similar way, H2O2 can also accept an electron from the dye,
originating not only the dye's radical cation, but HOꢁ. In addition to
that, the photodissociation of H2O2 also creates additional HOꢁ.
These mechanisms can be defined by the following equations:
DyehvDye*
(9)
!
Dye* þ TiO2ðCBÞ/Dyeꢁþ þ TiO2ðCBÞꢀ
(10)
The kinetics of the photodegradation effect of the TiO2,
following both the photocatalytic mechanism and the self-
photosensitized pathway, and only the latter one, on the dye, is
evidenced in Fig. 5c). In the first case, after 60 min, only 8% of the
molecules were still intact at the surface of TiO2, showing the
strong photocatalytic capability of the semiconductor. Following
this pathway and after the generation of the electronehole pair
there is a significant production of radical species that degrade the
molecule, either at the valence band with the production of OH
radicals from the adsorbed H2O at the surface of TiO2, or at the
conduction band, yielding Oꢁ2ꢀ, which participate in the production
of HOOꢁ, H2O2 and, ultimately, HOꢁ. However, the dye would also
present a significant photodegradation after being exposed to the
lamp. Therefore, to evidence the photodegradation capability of
TiO2 on this molecule, this study was also performed with dye 7
adsorbed onto cellulose. The results showed that there is a 3-fold
increase in the photocatalytic capability of the system when us-
ing TiO2, compared with the inert substrate.
To understand the nature of the new species, formed almost
immediately after irradiation, onto TiO2, a different strategy was
employed, by dispersing the commercial TiO2 nanopowder in a dye
solution (1.0 ꢃ 10e4 M), in acetonitrile, and using a powerful 450 W
Xe lamp. A key point of the strategy was that the output of the lamp
was cut off bellow 414 nm, by means of a cut-off filter. In this way,
only the self-photosensitized pathway would be privileged and
excessive photodegradation prevented. For reaching a faster and
better understanding of the behaviour of the dye in the presence of
TiO2, the monitoring of the GSDR and luminescence spectra of the
dispersion was performed. After lamp excitation for approximately
2 min, the dispersion was continuously stirred until reaching an
acceptable degree of homogeneity. The obtained results are pre-
sented in Fig. 6a). Regarding the GSDR, it shows three important
features: the characteristic absorption band of TiO2, in the
240e400 nm range, a first absorption peaking at approximately
525e530 nm, attributed to the dye radical cation, Dyeꢁþ, and
another absorption with a maximum located at 655 nm, attributed
to the original absorption of the dye, already deviated from its
initial absorption maximum, presumably due to the immediate
formation of the radical cation. Despite that fact, the same hyp-
sochromic shift previously observed (~65 nm) remains, showing
that the mechanism underlined here is most probably an electron
H2O2 þ hv/2HOꢁ
(11)
(12)
Dye* þ H2O2/Dyeꢁþ þ H2O2 þ eꢀ
Adding to the already presented Equation (7),
H2O2 þ eꢀ/HOꢁ þ OHꢀ
and finally,
(13)
H2O2 þ HOꢁ/H2O þ ꢁO2H
Ultimately, the interaction of the dye with H2O2 will result in the
dye's radical cation and hydroxylated intermediates which, at the
end, will contribute for the degraded products of the original dye:
(14)
ꢁ
ꢁþ
Dye* þ fH2O2; HO or ꢁO2Hg/Dye
þ hydroxylated intermediates//degraded products
(15)
Fig. 7 validates the assumptions presented in this work,
regarding the formation of the radical cation, since it displays a new
band arising at approximately 540 nm after photoexciation and in
(a)
(b)
(c)
(d)
(e)
1.2
1
GSDR Spectrum
a)
b)
1Dye 7*
Luminescence Spectrum
1
0.8
0.6
0.4
0.2
0
0.8
0.6
0.4
0.2
0
Dye.+ *
350
450
550
650
750
250
350
450
550
650
750
850
Wavelength (nm)
Wavelength (nm)
Fig. 6. a) GSDR and Luminescence spectrum of a previously excited TiO2 dispersion, in
a dye solution, by means of a Xe lamp. b) Macroscopic behavior of the dispersion, after
a few seconds of excitation.
Fig. 7. Absorption spectra of (a) dye 7 in CH3CN, (b) Photodegradation of the dye in
acetonitrile, (c) The dye in CH3CN:H2O2 (5:1 v/v), (d) Photodegradation of the dye in
CH3CN:H2O2 (5:1 v/v) after 5 min of excitation, (e) Dark recovery.

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
286
the presence of H2O2 (d). Besides that fact, this new band can also
be qualified as reversible, since it disappears in the dark (e). This is
consistent with an excitation/deexcitation mechanism which cor-
roborates the sensitization behaviour of the dye, in the presence or
absence of light, respectively.
presented in the commercial nanopowder, and locates in the
450e750 nm range. This could be attributed to the generated holes
entrapped at the surface of TiO2, hþtrapped. This effect is directly
related to the fact that the dye-TiO2 sample does not possess this
type of feature. Instead, a significant band ascends in the range of
600e750 nm. It was already mentioned that the molecule donates
an electron after photoexcitation, when adsorbed onto TiO2.
Therefore, the trend suggests that the electrons donated by the
molecules adsorbed onto the semiconductor surface will recom-
bine with the superficial trapped holes, leading to the disappear-
ance of their featuring band and to the significant grow of the
characteristic band of the trapped electrons, e-trapped, which, as
stated before, is located in the 700e800 nm range.
The powdered dye-TiO2 system did not allow a complete eval-
uation of the behaviour of the dye in this environment, by means of
laser flash photolysis, probably due to the thermal effect observed
before, which instantly degrades the sample. Therefore, colloidal
dispersions of TiO2 in acetonitrile solutions of the dye were pre-
pared and studied, in transmittance mode. Fig. 9 shows the tran-
sient absorption spectra of the regular, deaerated dye solution, in
acetonitrile, and of a colloidal dispersion of TiO2 in that same so-
lution, without stirring, and in an air equilibrated environment. The
assay was immediately performed after adding TiO2 in the solution
of the dye, to prevent significant dye adsorption and consequent
generation of the dye radical cation. In a prepared solution of the
dye, the normal pattern, previously presented, was obtained.
However, for the colloidal dispersion, some interesting features
were observed, starting from the fact that the tripletetriplet ab-
sorption band could be easily visible, in the same range as the one
presented in solution e from 300 to approximately 585 nm. Sur-
prisingly, this band was obtained in an air equilibrated sample, and
was equivalent to the signal obtained in an argon purged sample
(data not shown). Thus, the results suggest that, in the presence of
TiO2, there could be oxygen consumption from the semiconductor,
which enables the tripletetriplet absorption monitoring. As pre-
viously stated, at the conduction band of the semiconductor, the
oxygen presented in the neighbouring enviroꢁnꢀment is consumed to
generate the anion superoxide radical O2 (Equation (3)). As
opposed to the trend observed on the TiO2 nanopowder, with a
reflectance geometry, no thermal photoinduced effect could be
seen in the case of TiO2 dispersions. Following that, an air equili-
brated colloidal dispersion of TiO2 in a dye‘s solution was prepared
and photoexcited, by means of a 450 W Xe lamp, to generate the
radical cation of the dye. After obtaining a homogeneous pink
3.7. Laser flash photolysis applied to the system dye-TiO2
Laser flash photolyis has been an important tool to study the
excited state carriers originated from the TiO2 semiconductor. In
the literature, several reports point out to the presence of the
excited charge carrier species [32e34], namely trapped holes and
trapped electrons, which would absorb light at around 500 nm and
700 nm, respectively. These species would also be generated ac-
cording with the interfacial photochemical reactions described
below. From photoexcitation, mentioned earlier on equation (1), it
follows:
Charge carrier trapping : eꢀðCBÞ/etꢀrapped
(16)
Charge carrier trapping : hþðVBÞ/htþrapped
(17)
In the presence of an electron donor, such as the dye studied in
this work, the following reaction can occur:
Electron-hole recombination:
eꢀdye þ htþrapped/TiO2 ½eꢀhꢄ þ non
ꢀ radiative ðheatÞ or radiative emission ðluminescenceÞ
(18)
The equation shows the electronehole recombination that can
occur between the electron donated by compound 7, and the hole
entrapped by surficial defects such as titanol groups, commonly
found at the semiconductor surface. That recombination generates
the initial unexcited TiO2 state, via non-radiative pathway (with
heat emission), or via radiative pathways (luminescence emission).
Fig. 8 shows the transient absorption spectra, in reflectance
mode, of the commercial TiO2 nanopowder, and of a dye-TiO2
sample (0.5 mmol/g). The plot shows three distinct features: a first
absorption located at approximately 380 nm, which could be
attributed to a deviation of the band edge of TiO2 following the
laser excitation, due to a rapid photoinduced thermal phenomenon,
TiO2th-depl. The trend was already observed in the past by Wilkinson
and Willsher [35], where the transient absorption spectrum was
red shifted from the ground state absorption edge of TiO2, in the
case of an anatase powder; the second important signal is
25
3Dye*
3Dye*
14
12
10
8
TiO2 th-depl
20
15
10
5
e-trapped
h+
trapped
6
4
0
Dyedepl
2
-5
Dyedepl
0
-10
300 350 400 450 500 550 600 650 700 750
300
400
500
600
700
800
Wavelength (nm)
Wavelength (nm)
Fig. 8. Transient absorption spectra of commercial TiO2 nanopowder (blue) and dye 7-
mol/g (red), excited at 266 nm. Both spectra were obtained
TiO2 sample, c ¼ 0.5
m
Fig. 9. Transient absorption spectra of dye 7 in acetonitrile (blue), and of TiO2
dispersion in dye‘s solution (red). (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)
immediately after laser pulse. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
287
20
15
10
5
Acknowledgements
e-trapped
Thanks are due to FCT (Portugal's Foundation for Science and
Technology), for the funding projects ERA-MNT/0003/2009 and
Pest-OE/CTM/LA0024/2013 and for the PhD fellowships (SFRH/BD/
3Dye*
~
95358/2013 and SFRH/BD/95359/2013) of David S. Conceiçao and
Diana P. Ferreira, respectively.
0
-5
References
.
Dye +
Dyedepl
depl
-10
-15
[1] Mulders JJL, Steenhuysen LWG. The performance of dyes and dye mixtures of
rhodamine 6G/rhodamine 560 and of styryl 8/styryl 9 for single mode cw ring
laser operation. Opt Commun 1985;54(5):295e8.
[2] Faraggi M, Peretz P, Rosenthal I, Weinraub D. Solution properties of dye lasers.
Rhodamine B in alcohols. Chem Phys Lett 1984;103(4):310e4.
TiO2 depl
300 350 400 450 500 550 600 650 700 750 800 850
Wavelength (nm)
ꢀ
[3] Reisfeld R, Grinberg M, Levchenko V, Kuklinski B, Mahlik S, Magdassi S, et al.
Solegel glasses with enhanced luminescence of laser dye Rhodamine B due to
plasmonic coupling by copper nanoparticles. Opt Mater 2014;36(10):1611e5.
[4] Zehentbauer FM, Moretto C, Stephen R, Thevar T, Gilchrist JR, Pokrajac D, et al.
Fluorescence spectroscopy of Rhodamine 6G: concentration and solvent ef-
fects. Spectrochim Acta Part A Mol Biomol Spectrosc 2014;121:147e51.
[5] Barigelletti F. Effect of temperature on the photophysics of rhodamine 101 in a
polar solvent. Chem Phys Lett 1987;140(6):603e6.
Fig. 10. Transient absorption spectra of a previously irradiated TiO2 dispersion in a dye
solution.
colloidal dispersion, the sample was monitored by laser flash
photolysis, in reflectance mode, with constant stirring. The ob-
tained transient absorption spectrum is presented in Fig. 10. In this
case, the ground state depletion of the original dye and also of the
radical cation can be observed, at 685 nm and 525 nm respectively.
Adding to that, the same featuring pattern of the trapped electron is
presented, again in the range of 700e850 nm. The tripletetriplet
absorption band can also be visible, at around 430 nm, together
with the usual TiO2 ground state depletion, peaking at approxi-
mately 330 nm. All the characteristic features of the original dye
presented in this spectrum reflect the molecules from the disper-
sion that did not completely adsorb onto the TiO2.
~
[6] Conceiçao DS, Ferreira DP, Vieira Ferreira LF. Photochemistry and cytotoxicity
evaluation of heptamethinecyanine near infrared (NIR) dyes. Int J Mol Sci
2013;14(9):18557.
~
[7] Conceiçao DS, Ferreira DP, Graça VC, Silva CR, Santos PF, Vieira Ferreira LF.
Photochemical studies of new benzothiazole- and benzoselenazole-derived
aminosquarylium dyes. Tetrahedron 2015;71(6):967e76.
[8] Chen C, Qi X, Zhou B. Photosensitization of colloidal TiO2 with a cyanine dye.
J Photochem Photobiol A Chem 1997;109(2):155e8.
[9] Long J, Liu X, Guo H, Zhao B, Tan S. Effect of conjugated side groups on the
photovoltaic performances of triphenylamine-based dyes sensitized solar
cells. Dyes Pigments 2016;124:222e31.
[10] Ye S, Kathiravan A, Hayashi H, Tong Y, Infahsaeng Y, Chabera P, et al. Role of
adsorption structures of Zn-Porphyrin on TiO2 in dye-sensitized solar cells
studied by sum frequency generation vibrational spectroscopy and ultrafast
spectroscopy. J Phys Chem C 2013;117(12):6066e80.
4. Conclusions
[11] Tsai H-HG, Tan C-J, Tseng W-H. Electron transfer of squaraine-derived dyes
adsorbed on TiO2 clusters in dye-sensitized solar cells: a density functional
theory investigation. J Phys Chem C 2015;119(9):4431e43.
[12] Wu T, Liu G, Zhao J, Hidaka H, Serpone N. Photoassisted degradation of dye
Pollutants. V. Self-Photosensitized oxidative transformation of rhodamine B
under visible light irradiation in aqueous TiO2 dispersions. J Phys Chem B
1998;102(30):5845e51.
[13] Chen F, Zhao J, Hidaka H. Highly selective deethylation of rhodamine B:
adsorption and photooxidation pathways of the dye on the TiO2SiO2 com-
posite photocatalyst. Int J Photoenergy 2003;5(4).
[14] Tseng D-H, Juang L-C, Huang H-H. Effect of oxygen and hydrogen peroxide on
the photocatalytic degradation of monochlorobenzene in aqueous suspension.
Int J Photoenergy 2012;2012:9.
[15] Chu L, Anastasio C. Formation of hydroxyl radical from the photolysis of
frozen hydrogen peroxide. J Phys Chem A 2005;109(28):6264e71.
[16] Tamaki Y, Furube A, Katoh R, Murai M, Hara K, Arakawa H, et al. Trapping
dynamics of electrons and holes in a nanocrystalline TiO2 film revealed by
femtosecond visible/near-infrared transient absorption spectroscopy. Comp-
tes Rendus Chim 2006;9(2):268e74.
[17] Tojo S, Tachikawa T, Fujitsuka M, Majima T. Direct observation of one-electron
oxidation processes of aromatic sulfides in TiO2 colloidal solution by laser
flash photolysis. Phys Chem Chem Phys 2004;6(5):960e4.
In
this
report
a
new
9-(20-carboxyphenyl)-1,2,3,4-
tetrahydroxanthylium fluorescent dye with a “rhodamine-like”
structure was studied in both liquid media and adsorbed onto a
photo-active substrate, TiO2. Regarding the liquid samples, it was
found that the highest fluorescent quantum yield and fluorescence
lifetimes was obtained in a solution of dichloromethane, whereas
larger Stokes shifts were found in acetonitrile and ethanol, pointing
out to the presence of ICT mechanisms as the main deactivation
pathways, in these solvents. The tripletetriplet absorption spectra
were also monitored, in acetonitrile, showing a huge band in the
range of 350e580 nm, in an argon purged sample. When adsorbed
onto the photoactive material, the molecule presented some
interesting features, the most important one being the generation
of the dye radical cation, after photoexcitation, indicating the in-
jection of electrons into the conduction band of the semiconductor.
The photodegradation kinetics of the dye-TiO2 system was inves-
tigated, by enabling both the self-photosensitized pathway and the
photocatalytic mechanism, reaching to a 3-fold increase in the
photocatalytic capability, when compared with the non-
photoactive substrate, cellulose. By means of laser flash photol-
ysis, the powdered dye-TiO2 sample was monitored and it was
found that, in the presence of the dye, there was an electronehole
recombination that made the characteristic band of hþ to disap-
pear, whereas an excessive population of entrapped eꢀ could be
observed, when compared with the pure TiO2 nanopowder. Ulti-
mately, the work reflects the photochemical characterization of a
new organic fluorescent NIR dye which possesses a huge capability
to interact with the TiO2 photoactive substrate, in the presence of
light, although its stability still offers some drawbacks when con-
cerning future usage in photovoltaic and photochemical
applications.
[18] Furube A, Asahi T, Masuhara H, Yamashita H, Anpo M. Charge carrier dy-
namics of standard TiO2 catalysts revealed by femtosecond diffuse reflectance
spectroscopy. J Phys Chem B 1999;103(16):3120e7.
~
[19] Ferreira DP, Conceiçao DS, Prostota Y, Santos PF, Vieira Ferreira LF. Fluorescent
“rhodamine-like” hemicyanines derived from the 6-(N,N-diethylamino)-
1,2,3,4-tetrahydroxanthylium system. Dyes Pigments 2015;112:73e80.
[20] Tian Z, Tian B, Zhang J. Short communication. Dyes Pigments 2013;99(3):
1132e6.
[21] Vieira Ferreira LF, Ferreira Machado IL. Surface Photochemistry: organic
molecules within nanocavities of calixarenes. Curr Drug Discov Technol
2007;4(4):229e45.
[22] Vieira Ferreira LF, Costa SMB. Fluorescence quantum yield evaluation of
strongly absorbing dye solutions as a function of the dye concentration.
J Luminescence 1991;48:395e9.
[23] Botelho do Rego AM, Vieira Ferreira LF. 4. Photonic and electronic spectros-
copies for the characterization of organic surfaces and organic molecules
adsorbed on surfaces. In: Hari Singh N, editor. Experimental methods in the
physical sciences. Academic Press; 2001. p. 269e354.
[24] Ferreira DP, Conceicao DS, Ferreira VRA, Graca VC, Santos PF, Vieira
Ferreira LF. Photochemical properties of squarylium cyanine dyes. Photochem
Photobiol Sci 2013;12(11):1948e59.

~
D.S. Conceiçao et al. / Dyes and Pigments 128 (2016) 279e288
288
[25] Prostota YA. Convenient method for synthesis of 6-(N,N-Diethylamino)-9-(2-
carboxyphenyl)-1,2,3,4-tetrahydroxanthylium perchlorate. Chem Heterocycl
Compd 2004;40(1):116e7.
[26] Prostota Y, Kachkovsky OD, Reis LV, Santos PF. New unsymmetrical squaraine dyes
derived from imidazo[1,5-a]pyridine. Dyes Pigments 2013;96(2):554e62.
[27] Hinckley DA, Seybold PG, Borris DP. Solvatochromism and thermochromism of
rhodamine solutions. Spectrochim Acta Part A Mol Spectrosc 1986;42(6):747e54.
[28] Tian Z, Tian B, Zhang J. Synthesis and characterization of new rhodamine dyes
with large Stokes shift. Dyes Pigments 2013;99(3):1132e6.
[31] Beaumont PC, Johnson DG, Parsons BJ. Photophysical properties of laser dyes:
picosecond laser flash photolysis studies of Rhodamine 6G, Rhodamine B and
Rhodamine 101. J Chem Soc Faraday Trans 1993;89(23):4185e91.
[32] Schneider J, Matsuoka M, Takeuchi M, Zhang J, Horiuchi Y, Anpo M, et al.
Understanding TiO2 Photocatalysis: mechanisms and materials. Chem Rev
2014;114(19):9919e86.
[33] Jing L, Zhou J, Durrant JR, Tang J, Liu D, Fu H. Dynamics of photogenerated
charges in the phosphate modified TiO2 and the enhanced activity for pho-
toelectrochemical water splitting. Energy Environ Sci 2012;5(4):6552e8.
[34] Tachikawa T, Tojo S, Kawai K, Endo M, Fujitsuka M, Ohno T, et al. Photo-
catalytic oxidation reactivity of holes in the sulfur- and carbon-doped TiO2
powders studied by time-resolved diffuse reflectance spectroscopy. J Phys
Chem B 2004;108(50):19299e306.
[29] Folli A, Pochard I, Nonat A, Jakobsen UH, Shepherd AM, Macphee DE. Engi-
neering photocatalytic cements: understanding TiO2 surface chemistry to
control and modulate photocatalytic performances.
J Am Ceram Soc
2010;93(10):3360e9.
[30] Beaumont PC, Johnson DG, Parsons BJ. Excited state and free radical properties
of rhodamine dyes in aqueous solution: a laser flash photolysis and pulse
radiolysis study. J Photochem Photobiol A Chem 1997;107(1e3):175e83.
[35] Wilkinson F, Willsher CJ, Uhl S, Honnen W, Oelkrug D. Optical detection of a
photoinduced thermal transient in titanium dioxide powder by diffuse
reflectance laser flash photolysis. J Photochem 1986;33(3):273e8.