Photophysics of aromatic thiourea derivatives and their complexes with anions. Fast and ultrafast spectroscopic investigations

Article abstract of DOI:10.1039/b927442a

In order to collect detailed information on the interaction mechanism between fluorescent thiourea derivatives and anions, 9-[4-(trifluoromethyl) phenylthioureidomethyl]anthracene (1) and the corresponding 10-cyanoanthracene derivative (2) were synthesized and investigated in DMSO and MeCN by using absorption and emission steady state techniques, both in the absence and in the presence of different anions (AcO^-, H₂PO₄ ^-, HSO₄^-, and Br^-). A wide examination of the mechanism of anion recognition was also performed by time-resolved transient absorption spectroscopy, with nano- and femto-second time resolution. A complete picture of the excited state deactivation pathways of 1 and 2, where the main operative processes were fluorescence, intersystem crossing and internal conversion, was obtained. Even if steady-state measurements suggest that 1 and 2 selectively interact in the ground state with the anions H₂PO₄^- and AcO^-, time-resolved investigations demonstrate that the substrates are able to complex all the four anions. The photophysics of such complexes was fully characterized. The anions mainly modify the lifetime of the lowest excited singlet state and, especially in the cases of H₂PO₄ ^- and AcO^-, the efficiencies of fluorescence emission and triplet formation. In particular, no evidence was found of further deactivation processes such as photoinduced electron transfer, photodissociation, and photoionization. the Owner Societies 2010.

Full text of DOI:10.1039/b927442a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Photophysics of aromatic thiourea derivatives and their complexes
with anions. Fast and ultrafast spectroscopic investigations
Tiziana Del Giacco, Benedetta Carlotti, Stefano De Solis,
Arianna Barbafina and Fausto Elisei*
Received 5th January 2010, Accepted 26th March 2010
First published as an Advance Article on the web 4th June 2010
DOI: 10.1039/b927442a
In order to collect detailed information on the interaction mechanism between fluorescent
thiourea derivatives and anions, 9-[4-(trifluoromethyl)phenylthioureidomethyl]anthracene (1) and
the corresponding 10-cyanoanthracene derivative (2) were synthesized and investigated in DMSO
and MeCN by using absorption and emission steady state techniques, both in the absence and in
the presence of different anions (AcOꢀ, H2PO4ꢀ, HSO4ꢀ, and Brꢀ). A wide examination of the
mechanism of anion recognition was also performed by time-resolved transient absorption
spectroscopy, with nano- and femto-second time resolution. A complete picture of the excited
state deactivation pathways of 1 and 2, where the main operative processes were fluorescence,
intersystem crossing and internal conversion, was obtained. Even if steady-state measurements
ꢀ
suggest that 1 and 2 selectively interact in the ground state with the anions H2PO4 and AcOꢀ,
time-resolved investigations demonstrate that the substrates are able to complex all the four
anions. The photophysics of such complexes was fully characterized. The anions mainly modify
ꢀ
the lifetime of the lowest excited singlet state and, especially in the cases of H2PO4 and AcOꢀ,
the efficiencies of fluorescence emission and triplet formation. In particular, no evidence was
found of further deactivation processes such as photoinduced electron transfer, photodissociation,
and photoionization.
The operating mechanism in these sensors, in competition
Introduction
with the fluorescent emission, has been suggested to be photo-
induced electron transfer (PET).23,24 This hypothesis is supported
considering that, upon anion recognition through hydrogen
bonding, the oxidation potential of the thiourea receptor decreases,
thus inducing a more efficient PET from the anion–receptor
unit to the excited anthracene moiety. Anyway, no direct
investigations have been yet performed to demonstrate the
effective mechanism of action.
Over the last years various fluorescent sensors for anions have
been synthesized,1–4 but many of them are based on charged
receptors5–11 that present the limit to make easier the competi-
tion between the anion and the solvent for binding sites. More
recently simple and efficient sensors have been developed by
using charge neutral receptors.
Among the reported uncharged receptors of anions, the aliphatic
and aromatic urea receptors have been widely investigated.12–21
Anion recognition occurs through hydrogen bonding inter-
actions between the urea protons and the anion forming anion–
receptor complexes. Gunnlaugsson et al. proposed fluorescent
anion sensors, based on the ‘‘fluorophore–spacer–anion receptor’’
system, employing anthracene as fluorophore and substituted
thioureas as receptor, bonded to the 9-position of the anthracene
via a methylene spacer. Different isothiocyanates were used in
order to monitor the acidity of the thiourea-based receptors,
which influences the stability of the anion–receptor complexes,
and the resulting affinity and selectivity. This class of sensors
demonstrated an ideal recognition of the anion: only the fluores-
cence lifetime and quantum yield would be influenced upon ion
detection. Therefore any further spectral changes, such as exciplex
or excimer emission,22 were not observed.
In this work we examined 9-[4-(trifluoromethyl)phenylthio-
ureidomethyl]anthracene (1) and the corresponding 10-cyano-
anthracene derivative (2) (Scheme 1). Compound 2 was simply
prepared following the synthesis already reported for 1,14
mixing 9-(aminomethyl)anthracenes with the commercial
4-(trifluoromethyl)phenyl isothiocyanate in opportune experi-
mental conditions. A first part of the study regarded the
investigation of ground state and excited state, by absorption
and emission measurements, both in the absence and the
presence of different anions (AcOꢀ, H2PO4ꢀ, HSO4ꢀ, and Brꢀ)
used as tetrabuthylammonium salts, in DMSO and MeCN.
An extensive photophysical study has already been reported
on 1 in DMSO,12,14 so in this solvent only some measurements
were carried out in order to compare the results with those
obtained with 2 in the same experimental conditions. A wide
examination of the mechanism of anion recognition, respon-
sible for the fluorescence quenching, was performed by time-
resolved spectroscopy, with nano- and femto-second time
resolution. This study represents, to the best of our knowledge,
one of the first attempts of an exhaustive investigation of
Chemistry Department, and Centro di Eccellenza sui Materiali
Innovativi Nanostrutturati (CEMIN), University of Perugia,
via Elce di Sotto 8, 06123 Perugia, Italy. E-mail: elisei@unipg.it;
Fax: +39-075-5855598; Tel: +39-075-5855588
ꢁc
This journal is the Owner Societies 2010
8062 | Phys. Chem. Chem. Phys., 2010, 12, 8062–8070

View Article Online
each wavelength. The experimental setup was calibrated by an
optically matched solution of benzophenone (fT = 1 and
DeT = 6500 Mꢀ1 cmꢀ1 at the corresponding absorp-
tion maximum) in MeCN.30 Triplet–triplet molar absorption
coefficients (eT), and then fT values, were determined by
energy transfer from benzophenone in MeCN. The experi-
mental errors on tT were estimated to be about ꢂ10% and on
eT and fT of about ꢂ15%. The quenching constants of the
triplets by molecular oxygen were evaluated by the slope of the
linear plots of triplet lifetime vs. different oxygen concentra-
tions. All measurements were carried out at room tempera-
ture; the solutions were saturated by bubbling with nitrogen.
The transient behaviour on the ultrafast time scale
was investigated using femtosecond pulses generated by an
amplified titanium–sapphire laser system (Spectra Physics,
Mountain View, CA), that produces 1 W pulses centred at
800 nm at a repetition rate of 1 kHz. Pump pulses centred at
400 nm were obtained by second harmonic generation in a
500 mm b-barium-borate crystal. In the transient absorption
set up (Helios, Ultrafast Systems) the pump pulses were passed
through a chopper which cut out every second pulse and
collimated to the sample in a 2 mm quartz cuvette. Probe
pulses for optical measurements were produced by passing a
small portion of 800-nm light through an optical delay line
(with a time window of 1600 ps) and focusing into a 2 mm
thick Ti:Sapphire window to generate a white-light continuum
in the 430–800 nm window. The white light was focused onto
the sample and the chirp inside the sample cell was determined
by measuring the laser-induced Kerr signal of the solvent.
The temporal resolution of our time resolved spectroscopic
technique is B150 fs while the spectral resolution is 1.5 nm.
All the measurements were carried out under magic angle in
order to avoid the rotational decay processes, in a 2 mm cell
and at an absorbance of about 0.5 at 400 nm. The solution was
stirred during the experiments to avoid photodegradation
processes. Transient absorption data were analyzed using the
Surface Xplorer PRO (Ultrafast Systems) software. It was
possible to visualize the 3D and 2D data surface, to simul-
taneously display and save multiple spectra and kinetics,
to perform temporal chirp corrections, to make time zero
adjustments, to fit the kinetics by exponential(s) including
instrumental response function deconvolution, to perform
Singular Value Decomposition31,32 of the 3D surface into
principal components (spectra and kinetics) and to perform
Global Analysis.
Scheme 1 Molecular structures of 1 (X = H) and 2 (X = CN).
the mechanism that is at the base of fluorescent sensor
working.
Experimental
Synthesis of 1 and 2
Compound 1 was synthesized in three steps, starting from
9-methylanthracene which is commercially available. Firstly,
9-(bromomethyl)anthracene was prepared by bromination of
9-methylanthracene with N-bromosuccinimide in carbon tetra-
chloride.25 This product was refluxed with hexamethylenetetra-
mine in anhydrous chloroform and the resulting product
(9-(aminomethyl)antracene) was dissolved in anhydrous dichloro-
methane with 4-(trifluoromethyl)phenyl isothiocyanate to give 1,
9-[4-(trifluoromethyl)phenyl thioureidomethyl]anthracene.14
The synthesis of 2 was carried out following the same
procedure reported for 1. In this case the starting reactant,
10-methyl-9-cyanoanthracene, was prepared from 10-methyl-
anthracene-9-carboxaldehyde as previously described.26
10-(Aminomethyl)-9-cyanoanthacene: 1H-NMR d (200 MHz,
CDCl3): 8.51–8.43 (m, Ar–1H, Ar–8H, Ar–4H, Ar–5H),
7.78–7.64 (m, Ar–2H, Ar–3H, Ar–6H, Ar–7H), 4.89
(s, CH2NH2); 2 1H-NMR d (400 MHz, CDCl3): 9.78
(s, NH), 8.69 (d, Ar–4H, Ar–5H), 8.63 (s, NH), 8.43
(d, Ar–1H, Ar–8H), 7.97 (t, Ar–3H, Ar–6H), 7.86 (t, Ar–2H,
Ar–7H), 7.79 (d, Ar–19H, Ar–20H), 7.65 (d, Ar–18H,
Ar–21H), 5.84 (s, CH2).
Photophysical measurements
Absorption spectra were recorded with a Perkin-Elmer Lambda
800 spectrophotometer. Fluorescence emission spectra were
measured with a Fluorolog-2 (Spex, F112AI) spectrophoto-
fluorimeter. Fluorescence quantum yields (experimental error
of ca. 7%, from three independent experiments) were deter-
mined from the emission spectra of the samples recorded at an
absorbance smaller than 0.1 at the excitation wavelength, to
avoid self-absorption effects, by use of anthracene in ethanol
as reference (fF = 0.22).27
Results and discussion
Triplet formation quantum yields and lifetimes were
measured by a nanosecond laser flash photolysis setup
previously described (Nd:YAG Continuum, Surelite II, third
harmonics, lexc = 355 nm, pulse width ca. 7 ns and energy r
1 mJ pulseꢀ1).28,29 First-order kinetics were observed for
the decay of the lowest triplet state (T–T annihilation was
prevented by the low excitation energy). The triplet lifetimes
were measured at an absorbance of ca. 0.2; the concentration
effect on triplet lifetime was not investigated. The transient
spectra were obtained by monitoring the change of absorbance
over the 300–900 nm range and averaging at least 10 decays at
Absorption and emission spectra
The S0 - Sn absorption spectra of 1 and 2 in MeCN
and DMSO are shown in Fig. 1. Both compounds present a
broad vibrationally structured band in the 320–450 nm range.
The relatively high values of the absorption coefficients
(7700–11 500 Mꢀ1 cmꢀ1) indicate the allowed transitions
(Table 1). These data are in good agreement with the absorp-
tion maxima and oscillator strengths obtained by quantum-
mechanical calculations by using the INDO/1-CI method
(data not shown).
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 8062–8070 | 8063

View Article Online
Fig. 2 Absorption spectra of 2 (left) and emission spectra of 1 (right)
in MeCN in the absence and in the presence of increasing amounts of
TBAH2PO4 (up to 1.6 ꢃ 10ꢀ3 M and 4 ꢃ 10ꢀ2 M respectively). Inset:
concentration effect of the salt on the absorbance at 400 nm (left) and
on the fluorescence intensity at 417 nm (right); the full line represents
the best titration fitting (see text).
Fig. 1 Absorption and emission spectra of 1 (a) and 2 (b) in MeCN
(full lines) and dimethylsulfoxide (dashed lines).
the receptor chromophore CQS absorbs. This experimental
evidence indicates that there is an interaction anion–substrate
which does not involve the anthracenic fluorophore. It is
feasible that the interaction takes place through hydrogen
bonding of the anion with the thiourea protons.34 This effect
was the only spectral change observed upon addition of the
anions Brꢀ and HSO4ꢀ to solutions of 1 and 2 in both MeCN
Table 1 Spectral properties of 1 and 2 together with the fluorescence
quantum yields
Solvent
lmax,abs/nm
e/Mꢀ1cmꢀ1a
lmax,em/nm
fF
1
2
MeCN 333,349,367,386 11 500
DMSO 336,353,370,390 8400
MeCN 353,369,388,409 8600
DMSO 353,371,394,416 7700
390,413,437,463 0.035
396,418,443,471 0.096
and DMSO. When H2PO4 and AcOꢀ were added to the
ꢀ
423,442
430,454
0.0013
0.0016
substrate 2, a bathochromic shift of the absorption bands
was observed increasing the anion concentration and several
isosbestic points appeared (Fig. 2A). This may be due to a
partial interaction of the two anions with the anthracenic
portion of 2 and it is more important in MeCN, a less polar
solvent with respect to DMSO and so less competitive for the
anion in binding the substrate.
a
Absorption coefficient at the maximum (in bold type).
With respect to the spectrum of anthracene in apolar solvent,33
the absorption spectrum of 1 is red shifted and exhibits
larger bands compared to those of the fluorophore itself. The
introduction of the cyano group in the anthracenic 10 position
(compound 2) shifts bathocromically the absorption spectrum
while the vibrational structure is less evident (larger bands and
less profound minima). A small blue shift was observed on
going from DMSO to MeCN.
Anion addition to 1 induced a fluorescence quenching: in
general the spectral shape was not modified by the presence of
the salts but there was an important variation in the fluores-
ꢀ
cence intensity especially when H2PO4 and AcOꢀ were used
(Fig. 2B and 3). The values of the fluorescence quantum yields
of 1 and 2 measured in the presence of a 0.04 M concentration
of the anions are reported in Table 2. At this anion concen-
tration the fluorescence intensity should have reached a
plateau limit value as Fig. 3 showsꢀ. The quantum yield values
confirmed the little effect of HSO4 e Brꢀ on the efficiency of
the fluorescence process of the substrates 1 and 2. A variation
of both spectral shape and fluorescence quantum yields was
observed upon complexation of 2 with H2PO4ꢀ and AcOꢀ. In
general, the effect on the fluorescence intensity significantly
depends on the nature of the anion according to the following
trend: HSO4ꢀ r Brꢀ { H2PO4ꢀ o AcOꢀ, in agreement with
the ability of the anion to form hydrogen bonds with the
thiourea protons. The stronger the bases the more efficiently
these anions interact with the two substrates (Ka = 1.8 ꢃ 10ꢀ5
for AcOH, Ka = 7.1 ꢃ 10ꢀ3 for H3PO4, Ka = 1 ꢃ 103 for
H2SO4, and Ka = 1 ꢃ 109 for HBr; where Ka is the acid
constant).
The fluorescence emission spectrum of 1 is characterized by
maxima centred at 396, 418, 443 and 471 nm in DMSO which
undergo a B5 nm blue shift in MeCN, similarly to what
already observed for the absorption spectra (Fig. 1). However,
a significant dependence of the fluorescence quantum yield on
the solvent was observed (Table 1). In fact, the fluorescence
quantum yield measured for 1 was drastically reduced switching
from DMSO to MeCN (fF = 0.096 and 0.035 in DMSO
and MeCN, respectively). Compared to the emission spectrum
of anthracene in non polar solvents,33 the spectra of 1 are
red shifted and less structured. The emission spectrum of 2
(Fig. 1) shows a less structured band centred at B440 and
450 nm in MeCN and DMSO, respectively. The fluorescence
quantum yields measured for 2 were very low (fF B 10ꢀ3
,
Table 1).
Upon addition of the tetrabutyl ammonium salts of four
different anions (Brꢀ, HSO4ꢀ, H2PO4ꢀ, AcOꢀ) the vibrational
structure of the absorption spectra is still present even at high
anion concentration (Table 2). The absorption spectra show
significant changes in shape in the region below 340 nm where
The absorbance of 2 recorded in the presence of an increasing
amount of H2PO4ꢀ and AcOꢀ was analyzed by use of eqn (1)35
ꢁc
This journal is the Owner Societies 2010
8064 | Phys. Chem. Chem. Phys., 2010, 12, 8062–8070

View Article Online
Table 2 Spectral properties of 1 and 2 in the presence of 0.04 M additives together with the fluorescence quantum yields
Solvent
MeCN
Additive (0.04 M)
l
max,abs/nm
l
max,em/nm
fF
1
TBABr
333,349,367,387
333,349,367,387
333,349,367,387
333,349,367,387
335,352,370,391
336,352,371,391
336,352,371,391
336,352,371,391
352,369,389,410
351,369,388,410
352,369,391,412
352,369,391,412
352,371,394,416
352,371,395,416
352,371,396,417
352,371,395,416
390,413,437,463
390,413,437,463
390,413,437,463
390,413,437,463
396,418,443,471
396,418,443,471
396,418,443,471
396,418,443,471
423,442
0.032
0.032
0.007
0.006
0.095
0.092
0.027
0.016
0.0011
0.0013
0.0005
0.0013
0.0017
0.0017
0.0009
0.0031
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
DMSO
MeCN
DMSO
2
423,442
417, 438, 451
416, 438
430,454
430,454
430, 444, 456
416, 443
of 1 and 2 and, therefore, the concentration of the complex in
solution was much smaller than that of the free anion.
The measured association constants (Table 3) are in the
300–46 000 Mꢀ1 range. They are larger for the complexes
with acetate with respect to those with dihydrogen phosphate,
in agreement with the relative basicity. Furthermore, in
the case of 1, the solvent effect on the association constant
(Kass,DMSO o Kass,MeCN) reflects the ability of DMSO to
compete with the anion in the complex formation, as also
suggested by the b-scale of hydrogen-bond acceptor basicities.37
Instead, higher Kass values were generally obtained in the case
of the more polar substrate 2.
Triplet properties
Fig. 3 Changes in the relative fluorescence intensity at 413 nm of 1
upon addition of TBABr (.), TBAHSO4 (m), TBAH2PO4 (’),
TBAOAc (K) in MeCN (lexc = 367 nm).
The study of the triplet state of 1 and 2 was performed in
MeCN and DMSO by laser flash photolysis. For both com-
pounds a single transient was observed in all cases. It was
assigned to the T1 - Tn absorption on the basis of the
following considerations: (1) it is formed during the laser pulse
(ca. 7 ns); (2) it decays by a first-order kinetics; (3) its lifetime is
quenched by molecular oxygen at an almost diffusional rate;
and (4) it is sensitized by higher-energy triplet donors, such as
benzophenone. The time resolved absorption spectra obtained
for 2 in MeCN are shown in Fig. 4 as an example, together
with the decay kinetics recorded at the absorption maximum
of ca. 430 nm (Fig. 4, inset); similar results were obtained in
the other experiments. In our experimental conditions at very
low laser intensity, the time-resolved absorption spectra and
the decay curves of 1 and 2 did not show any sign of
absorptions of cation radicals and free radicals produced by
photoionization and photodissociation, respectively.
to determine the values of the association constants for the
substrate–anion interaction (Fig. 2A, inset):
AS þ AC ꢃ Kass ꢃ ½anionꢄ
A ¼
ð1Þ
1 þ Kass ꢃ ½anionꢄ
where A is the absorbance at a wavelength were a great
variation is observed upon anion addition, [anion] is the anion
concentration in solution, AS and AC are the limit values of
absorbance of the free substrate and of the fully complexed
substrate respectively, Kass is the association constant of thꢀe
complex. Instead, in case of the interaction of 1 with H2PO4
and AcOꢀ, where the absorption changes are rather small, the
better information on the complexation was obtained from the
fluorescence titrations. The emission intensities were analyzed
by use of eqn (2)36 to determine the Kass values (Fig. 2B, inset):
Table 3 Association equilibrium constants (Kass) of 1 and 2
IF 1 þ DF ꢃ Kass ꢃ ½anionꢄ
¼
ð2Þ
Compd.
Solvent
MeCN
Additive
Kass/Mꢀ1
IFꢅ
1 þ Kass ꢃ ½anionꢄ
1
TBAH2PO4
TBAAcO
TBAH2PO4
TBAAcO
TBAH2PO4
TBAAcO
TBAH2PO4
TBAAcO
3000
45 000
300
1300
670
46 000
6200
32 000
where IF is the fluorescence intensity upon anion addition, I1F
is the fluorescence intensity of the free substrate, [anion] is the
anion concentration in solution, and DF is the fluorescence
intensity variation.
DMSO
MeCN
DMSO
2
It has to be noted that the anion concentrations used in the
above experiments were always significantly larger than those
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 8062–8070 | 8065

View Article Online
Table 5 Triplet properties of 1 and 2 in the presence of 0.04 M
additives (lexc = 355 nm)
kox
/
Additive
Solvent (0.04 M)
lmax/nm tT/ms fT
109 Mꢀ1 sꢀ1
1
MeCN
TBABr
TBAHSO4
TBAH2PO4 425
425
425
160
180
140
170
170
160
180
190
220
190
200
250
220
200
170
300
0.50 2.2
0.38 2.1
0.37 2.0
0.28 1.5
0.50 1.8
0.47 2.2
0.43 1.4
0.36 1.5
0.62 1.9
0.55 2.0
0.27 1.6
0.26 2.9
0.74 1.6
0.80 1.8
0.67 1.6
0.26 1.9
TBAAcO
DMSO TBABr
TBAHSO4
425
430
430
TBAH2PO4 430
TBAAcO
TBABr
TBAHSO4
430
430
430
2
MeCN
TBAH2PO4 430
TBAAcO
DMSO TBABr
TBAHSO4
430
435
435
Fig. 4 Triplet–triplet absorption spectra of 2 in MeCN recorded at
3 (&), 18 (J), 50 (n), 100 (,) and 150 ( v ) ms after the laser pulse
(lexc = 355 nm). Inset: decay kinetics recorded at 430 nm.
TBAH2PO4 435
TBAAcO 435
The triplet yields (Table 4) are rather high (ca. 0.4 and 0.7
for 1 and 2, respectively), indicating that the S1 - T1 inter-
system crossing is an efficient decay channel of the lowest
excited singlet state of these compounds, as already reported
for anthracene itself (fT = 0.66–0.71).27 It has to be noted
that the quantum yields were obtained by using in DMSO the
triplet–triplet absorption coefficients measured in MeCN.
The triplet lifetimes of 1 and 2, longer than 150 ms (Table 4),
represent lower limits for tT which is strongly affected by
molecular oxygen and excitation energy. Thus they were
measured at an excitation energy smaller than 1 mJ pulseꢀ1
after a vigorous bubbling with nitrogen.
ultrafast excitation pulses,38–41 the lowest vibrational levels of
the p,p* excited singlet state of anthracenic character were
populated upon excitation.
The time-resolved absorption spectra of the two substrates
in MeCN are shown in Fig. 5, together with the decay kinetics
recorded at two noteworthy wavelengths; similar spectra were
recorded in DMSO. A broad absorption band centred at
575–595 nm was detected just after the laser pulse. At longer
time delays this absorption was replaced by a rest absorption
with a maximum (lmax o 470 nm) located outside the investi-
gated spectral interval.
The transient absorption spectra in the presence of the salts
(Table 5) showed a sharp maximum at B430 nm which decays
with a time constant of hundreds of microseconds as in the
case of the free substrates. Thus, the spectral and kinetic
properties of the triplet state depend only on the fluorophore
nature and are not affected by the anion complexation. The
intersystem crossing quantum yields (fT) were calculated by
assuming the absorption coefficients of the triplet state of the
complexes to be equal to the correspondent one measured for
the uncomplexed substrate in MeCN and are shown in
Table 5. In general, the efficiency of production of the triplet
state resulted slightly influenced by the interaction with the
anion, being fT = 0.36–0.50 and 0.55–0.80 for complexes with
1 and 2, respectively. Only in the presence of the acetate anion
the triplet quantum yield was decreased to values of 0.26–0.28
for both the substrates in both the solvents.
The decay traces were analyzed at different wavelengths: the
best fittings were obtained by triexponential functions for
1 and biexponential functions for 2 (Table 6) convoluted
with the instrumental response profile with Gaussian shape.
Femtosecond absorption
The primary decay processes of 1 and 2 were investigated upon
ultrafast excitation at 400 nm. Taking into account the absorp-
tion spectra of the substrates and the spectral broadening of the
Table 4 Properties of the lowest triplet state of 1 and 2 (lexc = 355 nm)
eT/
kox/
109 Mꢀ1 sꢀ1
Fig. 5 Pump–probe time resolved absorption spectra of 1 (A)
recorded 0.2 (a), 5 (b), 20 (c), 60 (d), 200 (e), 500 (f) and 1020 (g) ps
and 2 (B) recorded 0.3 (a), 1.5 (b), 7 (c), 18 (d), 25 (e), 50 (f) and 1400
(g) ps after the laser pulse (lexc = 400 nm) in MeCN. Insets: decay
kinetics recorded at 595 (full lines) and 473 (dashed lines) nm for 1 (A)
and at 575 (full line) and 475 (dashed line) nm for 2 (B) in MeCN.
Solvent
lmax
tT/ms
Mꢀ1 cmꢀ1
fT
1
2
MeCN
DMSO
MeCN
DMSO
425
430
430
435
160
150
150
250
23 000
11 000
0.37
0.47
0.74
0.71
2.2
2.4
2.0
1.9
ꢁc
This journal is the Owner Societies 2010
8066 | Phys. Chem. Chem. Phys., 2010, 12, 8062–8070

View Article Online
S1 state which takes place in about 11 ps (Fig. 6A). The similar
results obtained in MeCN and DMSO indicate that the solvent
does not play an important role in determining the nature
of the excited states formed upon laser irradiation and their
lifetimes.
The short lifetimes of the S1 states, coupled with the
Stern–Volmer constants (KSV) obtained by the steady-state
fluorescence titrations of the substrates in both MeCN and
ꢀ
DMSO with the different anions (Brꢀ, HSO4ꢀ, H2PO4
,
AcOꢀ), allowed to obtain fluorescence quenching rate con-
stants (kq = KSV/tF) much larger than the diffusion limit of
1010
M .
ꢀ1 sꢀ1 42 This result is in agreement with an interaction
between the substrates and the ions which takes place in the
ground state and is not diffusion controlled.
The effect of the anions Brꢀ, HSO4ꢀ, H2PO4ꢀ, AcOꢀ as
tetrabutylammonium salts on the primary deactivation
processes of 1 and 2 was investigated both in MeCN and
DMSO. In particular, femtosecond transient absorption measure-
ments were carried out for the two substrates in the presence of
a 0.04 M concentration of the salts, value at which all the
substrate receptors in the sample are complexed with the
anions. Therefore, the deactivation pathways found to be
operative with this investigation could be considered charac-
teristic of the complex substrate–anion formed in the ground
state before laser excitation.
Fig. 6 Amplitudes of the decay components of 1 (A) and 2 (B) in
MeCN obtained by global analysis.
The global analysis of the decay kinetics also revealed the
presence of three and two exponential components (a short-
and a longer-lived one, S and L, respectively and a rest
absorption formed while L disappeared and that did not decay
during the investigated time interval, Fig. 6A).
The time resolved absorption spectra and the decay kinetics
recorded for the two substrates in the presence of the anions
were very similar to those recorded for the free molecules
(Fig. 7). The data analysis revealed also in these cases the
presence of two or three exponential components charac-
terized by spectral shapes analogous to the ones observed
for the substrates itself (Fig. 8 and Table 7).
In the case of 2 the data analysis revealed the presence of
two components: a transient characterized by a lifetime of
30 ps in MeCN and 27 ps in DMSO (L) and a rest absorption
(Fig. 6B). The L component shows a negative signal at
wavelengths shorter than 500 nm and a positive absorp-
tion with a maximum at 575 nm in both solvents; instead,
the rest transient shows a positive signal at wavelengths shorter
than 460 nm. This spectral properties are in agreement with a
precursor–successor scheme which involves the L - rest step.
Moreover the spectral shape obtained for L exhibits a negative
absorption signal in the region of the steady-state fluorescence
spectrum.
The L transient was assigned to the lowest excited singlet
state S1, while the rest component was attributed to the lowest
excited triplet state T1, formed upon the disappearance of S1
(Fig. 4).
For 1, the data analysis revealed the presence of an addi-
tional short-lived component, S, characterized by a lifetime of
11 ps in both MeCN and DMSO while the lifetimes associated
with the component L were 330 and 300 ps in MeCN and
DMSO, respectively. For the same reasons already discussed
for 2, L was assigned to the lowest excited singlet state S1 and
rest to the lowest excited triplet state T1. The S transient,
associated with a positive absorption band centered at 590 nm,
was due to the vibrational relaxation of the vibrationally hot
Fig. 7 Time resolved absorption spectra of 1 in MeCN (A) in the
presence of 0.04 M TBABr recorded 0.45 (a), 3 (b), 8 (c), 30 (d), 100
(e), 400 (f) and 1400 (g) ps after the laser pulse and (B) in the presence
of 0.04 M TBAAcO recorded 0.2 (a), 1.5 (b), 4 (c), 7 (d), 11 (e), 30 (f),
150 (g) and 500 (h) ps after the laser pulse (lexc = 400 nm). Insets:
decay kinetics recorded at 593 (full lines) and 435 (dashed lines) nm for
1 in MeCN in the presence of TBABr (A) and at 595 nm for 1 in
MeCN in the presence of TBAAcO (B).
Table 6 Time constants of 1 and 2 (lexc = 400 nm)
Solvent
lmax,1/nm t1/ps
lmax,2/nm t2/ps
lmax,3/nm t3/ps
1
2
MeCN 590
DMSO 450, 590
MeCN
11
11
595
590
575
575
330
300
30
o470
o435
o460
o460
Rest
Rest
Rest
Rest
DMSO
27
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 8062–8070 | 8067

View Article Online
the substrate alone (e.g. 1/AcOꢀ in MeCN and 2/H2PO4 in
ꢀ
both the solvents). In other cases, parallel behaviours were
found.
Interaction mechanism
The steady-state and the time-resolved measurements
provided a complete characterization of the lowest excited
states deactivation pathways of 1 and 2 in polar solvents. In
particular combining the data of the measured quantum yields
(fF, fT) it was possible to estimate also the efficiency of
internal conversion (fIC = 1 ꢀ fF ꢀ fT), since in our
experimental conditions there was no evidence of photoreac-
tions, photodissociations or photoionization. Using the values
of the lowest excited state lifetimes, the rate constants for three
deactivation processes observed (fluorescence, intersystem
crossing and internal conversion) were calculated for the free
substrates (Table 8).
It is worthy to note that the fluorescence rate constant (kF)
assumes values of 1.1 ꢃ 108 and 3.2 ꢃ 108 sꢀ1 for 1 in MeCN
and DMSO, respectively, confirming that the radiative transi-
tion S1 - S0 is allowed, while it is 4.3 ꢃ 107 and 5.9 ꢃ 107 sꢀ1
for 2 in MeCN and DMSO, thus indicating a slightly less
allowed transition. On the other hand, the high intersystem
crossing (ISC) rate constants of 1 and 2 (kISC E 109 sꢀ1 and
Fig. 8 Amplitudes of the decay components of 1 in MeCN in the
presence of 0.04 M TBABr (A) and 0.04 M TBAAcO (B) obtained by
global analysis (lexc = 400 nm).
Even if the shape of the time-resolved absorption spectra
was only slightly affected by the presence of the anions, the
lifetimes of the lowest excited singlet states of 1 and 2 were
significantly changed by the interaction with the anions
(Table 7). In general, after a fast vibrational relaxation which
takes place in less than 15 ps, the longer lived S1 state was
detected and characterized. The results of this analysis show
that the interaction between the substrates and the anions
cause effects much different than those recorded by steady-
state experiments. In particular, the effect on the lifetimes is
not parallel to that on the fluorescence quantum yields, thus
indicating that the photophysical behaviour of the systems
cannot be understood just taking into account of a further
decay pathway, such as the photoinduced electron transfer,
which takes place in the complexes. In particular, Brꢀ and
HSO4ꢀ, which change the fluorescence quantum yields only
slightly, are able to reduce significantly the S1 lifetimes of 1 in
MeCN (from 330 ps to 190 and 120 ps respectively). Instead,
in some cases where the fluorescence quantum yield is
quenched, the S1 lifetimes become even longer than those of
1010 ꢀ1, respectively) suggest the occurrence of an efficient
s
spin–orbit coupling between the lowest excited singlet and
triplet states which allows a very fast ISC process to be
operative. It has to be noted that in the case of 1, the solvent
effect on kF could reflect the influence of the different solvent
polarizability (0.212 and 0.284 for MeCN and DMSO, respec-
tively) on the relative position of the lowest excited singlet
states S1 and S2. In fact, for this molecule quantum mechanical
INDO1/CI calculations show the presence of close lying
states of p,p* anthracenic character with different oscillator
strengths (S1 forbidden and S2 allowed). The solvent could
affect the mixing between the two states and then the experi-
mental kF values.43,44
Irradiation at 400 nm generates vibrationally hot levels of
the lowest excited singlet state of the substrates 1 and 2 which
reach within about ten picoseconds the relaxed S1 state. This
state decays in about 300 ps and 30 ps for 1 and 2, respectively.
Table 7 Time constants of 1 and 2 in the presence of 0.04 M additives (lexc = 400 nm)
Solvent
MeCN
Additive (0.04 M)
l
max,1/nm
t1/ps
l
max,2/nm
t2/ps
l
max,3/nm
t3/ps
1
TBABr
590
590
585
590
600
600
600
6
4
4
8
10
14
11
590
595
590
600
595
595
590
585
570
570
565
580
565
570
575
650
190
120
35
400
250
250
80
25
36
26
100
400
19
24
280
130
o440
o470
o470
o440
o450
o470
o460
o470
o470
o460
o470
o470
o470
o470
o460
o440
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
Rest
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
TBABr
TBAHSO4
TBAH2PO4
TBAAcO
DMSO
MeCN
DMSO
2
570
4
570
570
4
3.5
570
565
6.6
4
ꢁc
This journal is the Owner Societies 2010
8068 | Phys. Chem. Chem. Phys., 2010, 12, 8062–8070

View Article Online
Table 8 Fluorescence, intersystem crossing and internal conversion
rate constants of 1 and 2
Table 9 Decay rate constants of 1 and 2 in the presence of 0.04 M
additives
Solvent
kF/109 sꢀ1
kISC/109 sꢀ1
kIC/109 sꢀ1
Solvent Additive (0.04 M) kF/109 sꢀ1 kISC/109 sꢀ1 kIC/109 sꢀ1
1
2
MeCN
DMSO
MeCN
DMSO
0.11
0.32
0.043
0.059
1.1
1.6
25
1.8
1.4
8.6
11
1
MeCN TBABr
TBAHSO4
TBAH2PO4
TBAAcO
DMSO TBABr
0.17
0.27
0.20
0.015
0.38
0.37
0.34
0.48
0.031
0.050
0.005
0.003
0.089
0.071
0.003
0.024
2.6
3.2
11
0.70
2.0
1.9
5.4
14
21
21
2.7
0.70
39
2.5
4.9
18
1.8
1.6
1.8
6.8
25
6.9
17
7.3
1.8
14
26
TBAHSO4
TBAH2PO4
TBAAcO
S1 is the fluorescent state and the precursor of the lowest
excited triplet state T1 whose formation could be followed by
the femtosecond resolved measurements and whose decay
was observed during the nanosecond laser flash photolysis
experiments (Scheme 2).
2
MeCN TBABr
TBAHSO4
TBAH2PO4
TBAAcO
DMSO TBABr
Upon addition of the tetrabutylammonium salts of the four
anions (Brꢀ, HSO4ꢀ, H2PO4ꢀ, AcOꢀ) the singlet-state life-
times were modified, together with the efficiency of the fluores-
cence process and formation of the triplet. The steady-state
and time-resolved measurements carried out on the anion–
substrate complexes allowed us to get a complete description
of the rate constants (Table 9) of the three main relaxation
pathways observed for the singlet state. In particular, the
TBAHSO4
TBAH2PO4
TBAAcO
33
2.4
2.0
8.3
1.2
5.7
induced only a small variation in the properties of the fluores-
cent lowest excited singlet state. The changes induced by the
two very basic H2PO4 and AcOꢀ species are remarkable, in
ꢀ
agreement with the values of the association constants between
the ground state of the substrates and these salts. This experi-
mental evidence confirmed the possibility of using the sub-
strates 1 and 2 as selective sensors of dihydrogen phosphate
and acetate even in small amounts.
ꢀ
complexes with the anions Brꢀ and HSO4 show rate con-
stants kF and kISC quite close of 1 and 2 (the difference is
smaller than the factor of 2) while those with the anions
H2PO4ꢀ and AcOꢀ has kF and kISC values of about one order
of magnitude smaller than those of the substrates alone. The
small effect of the heavy atom (Brꢀ) effect on the intersystem
crossing rate constant can be due to the high spin–orbit
coupling effect which is responsible of particularly high kISC
already in the substrates alone.
Conclusions
In this work compounds 1 and 2 were synthesized and
characterized by steady-state absorption and emission techniques,
both in the case of the free substrates and of their complexes
with different anions (Brꢀ, HSO4ꢀ, H2PO4 and AcOꢀ) in
Moreover the time resolved measurements did not show any
evidence of further deactivation pathways induced by the
presence of anions, such as photoinduced electron transfer
(PET). In fact, both the nanosecond and the femtosecond
measurements did not detect transients which could be formed
as a consequence of this process (for example the anion radicals
of the anthracene and cyano-anthracene fluorophores).45 The
excited state dynamics of the free (S) and complexed (C)
substrates are summarized in Scheme 2.
ꢀ
MeCN and DMSO. A complete investigation of the excited
state dynamics of 1 and 2 and of their complexes with the salts
was carried out by using time-resolved transient absorption
spectroscopy, with nano- and femto-second time resolution.
The observed singlet and triplet excited states were spectrally
and kinetically characterized and three main deactivation
pathways found to be operative (fluorescence, intersystem
crossing and internal conversion) were quantitatively described
in terms of their efficiencies and rate constants. In particular,
the substrate 1 is more fluorescent (fF = 0.035 in MeCN) and
shows a longer S1 state (tF = 330 ps in MeCN) than the
compound 2 (fF = 0.0013 and tF = 30 ps in MeCN). Instead,
the intersystem crossing is more efficient for 2 (fT = 0.74 in
MeCN) than for 1 (fF = 0.37 in MeCN).
The nature of the anion has an impact on the effect observed
both on the lifetimes and the quantum yields of the two sub-
strates. In fact ions like Brꢀ and HSO4ꢀ, which interact weakly
with the thiourea protons because of their low basicity,
The impact of Brꢀ and HSO4ꢀ on the photophysical proper-
ties of 1 and 2 was rather small; in fact the decay rate constants
of the lowest singlet state were changed by a factor of r2.
Instead, the substrates strongly bound the anions H2PO4ꢀ and
AcOꢀ in the ground state with association constants of
102–105
M
ꢀ1, depending on the substrate and the solvent.
The main effect of acetate and dihydrogen phosphate was on
the lifetime of the lowest excited singlet state and on the
fluorescence and triplet formation quantum yields.
The experiments carried out with nano- and femto-second
time resolution did not give evidence of further transients
ascribable to radical ions and radicals, which would be expected
Scheme 2 Deactivation pathways of the excited states of 1 and 2 (S)
and of their complexes with anions (C).
ꢁc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 8062–8070 | 8069

View Article Online
if a photoreaction process, such as photoinduced electron
transfer (PET), was operative both in the absence and presence
of anions.
20 R. M. Duke and T. Gunnlaugsson, Tetrahedron Lett., 2007, 48,
8043.
21 E. B. Veale and T. Gunnlaugsson, J. Org. Chem., 2008, 73,
8073.
22 S. Nishizawa, H. Kaneda, T. Uchida and N. Teramae, J. Chem.
Soc., Perkin Trans. 2, 1998, 2325.
23 R. A. Bissell, A. P. d. Silva, H. Q. N. Gunaratne, P. L. M. Lynch,
G. E. M. Maguire, C. P. McCoy and K. R. A. S. Sandanayake, in
Fluorescent PET (Photoinduced Electron Transfer) Sensors, Berlin,
1993.
24 A. P. d. Silva, D. B. Fox, T. S. Moody and S. M. Weir, Trends
Biotechnol., 2001, 19, 29.
25 S. Duan, J. Turk, J. Speigle, J. Corbin, J. Masnovi and R. J. Baker,
J. Org. Chem., 2000, 65, 3005.
Acknowledgements
The authors gratefully acknowledge the financial support of
the Ministero per l’Universita e la Ricerca Scientifica e Tecno-
logica (Rome, Italy), the University of Perugia and the
Fondazione Cassa di Risparmio di Perugia.
26 A. Torrado and B. Imperiali, J. Org. Chem., 1996, 61, 8940.
27 M. Montalti, A. Credi, L. Prodi and M. T. Gandolfi, Handbook of
Photochemistry, 3rd edn, CRC Press, 2006 (and references therein).
28 H. Gorner, F. Elisei and G. G. Aloisi, J. Chem. Soc., Faraday
Trans., 1992, 88, 29.
Notes and references
1 R. Martinez-Manez and F. Sancenon, Chem. Rev., 2003, 103, 4419.
2 R. Martı
3 P. A. Gale and R. Quesada, Coord. Chem. Rev., 2006, 250, 3219.
4 P. A. Gale, S. E. Garcıa-Garrido and J. Garric, Chem. Soc. Rev.,
2008, 37, 151.
nez-Manez and F. Sancenon, J. Fluoresc., 2005, 15, 267.
´ ´ ´
´
29 A. Romani, F. Elisei, F. Masetti and G. Favaro, J. Chem. Soc.,
Faraday Trans., 1992, 88, 2147.
5 D. H. Vance and A. W. Czarnik, J. Am. Chem. Soc., 1994, 116,
9397.
6 A. W. Czarnik, Acc. Chem. Res., 1994, 27, 302.
7 L. Fabbrizzi, M. Licchelli, G. Rabaioli and A. Taglietti, Coord.
Chem. Rev., 2000, 205, 85.
30 I. Carmichael and G. L. Hug, J. Chem. Phys., 1986, 1.
31 G. H. Golub and C. F. V. Loan, Matrix Computations, Johns
Hopkins University Press, 2nd edn, 1989.
32 G. Strang, Introduction to Linear Algebra, Wellesley-Cambridge
Press, 1998.
8 J. Yoon, S. K. Kim, N. J. Singh and K. S. Kim, Chem. Soc. Rev.,
2006, 35, 355.
33 I. B. Berlman, Handbook of Fluorescence Spectra of Aromatic
Molecules, Academic Press, 1971.
9 E. Garcı
´
a-Espana, P. Dı
´
az, J. M. Llinares and A. Bianchi, Coord.
34 T. Gunnlaugsson, M. Glynn, G. M. Tocci, P. E. Kruger and
F. M. Pfeffer, Coord. Chem. Rev., 2006, 250, 3094.
35 R. D. Braun, Introduction to Chemical Analysis, McGraw-Hill,
1982.
Chem. Rev., 2006, 250, 2952.
10 L. Fabbrizzi, M. Licchelli and A. Taglietti, Dalton Trans., 2003,
3471.
11 K. Ghosh, A. R. Sarkar and G. Masanta, Tetrahedron Lett., 2007,
48, 8725.
36 V. Thiagarajan and P. Ramamurthy, Spectrochim. Acta, Part A,
2007, 67, 772.
12 T. Gunnlaugsson, A. P. Davisab and M. Glynna, Chem. Commun.,
2001, 2556.
37 M. J. Kamlet, J. L. M. Abboud, M. H. Abraham and R. W. Taft,
J. Org. Chem., 1983, 48, 2877.
13 T. Gunnlaugsson, A. P. Davis, J. E. O’Brien and M. Glynn, Org.
Lett., 2002, 4, 2449.
14 T. Gunnlaugsson, A. P. Davis, G. M. Hussey, J. Tierney and
M. Glynn, Org. Biomol. Chem., 2004, 2, 1856.
15 T. Gunnlaugsson, A. P. Davis, J. E. O’Brien and M. Glynn, Org.
Biomol. Chem., 2005, 3, 48.
16 T. Gunnlaugsson, H. D. P. Ali, M. Glynn, P. E. Kruger,
G. M. Hussey, F. M. Pfeffer, C. a. M. G. d. Santos and
J. Tierney, J. Fluoresc., 2005, 15, 287.
17 T. Gunnlaugsson, P. E. Kruger, P. Jensen, J. Tierney, H. D. P. Ali
and G. M. Hussey, J. Org. Chem., 2005, 70, 10875.
18 T. Gunnlaugsson, P. E. Kruger, P. Jensen, F. M. Pfeffer and
G. M. Hussey, Tetrahedron Lett., 2003, 44, 8909.
19 F. M. Pfeffer, M. Seter, N. Lewcenko and N. W. Barnett,
Tetrahedron Lett., 2006, 47, 5241.
38 H. Wang, H. Zhang, O. K. Abou-Zied, C. Yu, F. E. Romesberg
and M. Glasbeek, Chem. Phys. Lett., 2003, 367, 599.
39 D. Zhong, A. Douhal and A. H. Zewail, Proc. Natl. Acad. Sci.
U. S. A., 2000, 97, 14056.
40 P. T. Chou, Y. C. Chen, W. S. Yu, Y. H. Chou, C. Y. Wei and
Y. M. Cheng, J. Phys. Chem. A, 2001, 105, 1731.
41 O. K. Abou-Zied, R. Jimenez, E. H. Z. Thompson, D. P. Millar
and F. E. Romesberg, J. Phys. Chem. A, 2002, 106, 3665.
42 S. L. Murov, Handbook of Photochemistry, M. Dekker, 1973.
43 A. Spalletti and G. Bartocci, Phys. Chem. Chem. Phys., 1999, 1,
5623.
44 E. Marri, G. Galiazzo, U. Mazzucato and A. Spalletti, Chem.
Phys., 2000, 260, 383.
45 C. Zimmermann, M. Mohr, H. Zipse, R. Eichberger and
W. Schnabel, J. Photochem. Photobiol., A, 1999, 125, 47.
ꢁc
This journal is the Owner Societies 2010
8070 | Phys. Chem. Chem. Phys., 2010, 12, 8062–8070