Catechol derivatives as fluorescent chemosensors for wide-range pH detection

Article abstract of DOI:10.1002/chem.200800722

The synthesis and characterization of a new family of catechol derivatives designed to behave as fluorescent chemosensors for wide-range pH detection has been described. These compounds were prepared by covalently coupling a catechol unit with other aromatic rings, thus obtaining π-delocalized systems with both pH-responsive groups and fluorescent behavior. In the case of a pyridine-catechol derivative, this leads to up to three different protonation states with distinct optical properties in organic media, as corroborated by density functional theory calculations. By applying dualwavelength detection techniques, this compound shows complementary "off-on-off" and "on-off-on" emission profiles upon pH variation, a behavior that can be exploited to perform acidity detection over a broad pH range.

Full text of DOI:10.1002/chem.200800722

DOI: 10.1002/chem.200800722
Catechol Derivatives as Fluorescent Chemosensors for Wide-Range pH
Detection
Emilia Evangelio,[a] Jordi Hernando,[b] Inhar Imaz,[c] Gisela G. Bardají,[b]
Ramon AlibØs,[b] FØlix BusquØ,*[b] and Daniel Ruiz-Molina*[d]
Abstract: The synthesis and characteri-
zation of a new family of catechol de-
rivatives designed to behave as fluores-
cent chemosensors for wide-range pH
detection has been described. These
compounds were prepared by covalent-
ly coupling a catechol unit with other
aromatic rings, thus obtaining p-delo-
calized systems with both pH-respon-
sive groups and fluorescent behavior.
In the case of a pyridine–catechol de-
rivative, this leads to up to three differ-
ent protonation states with distinct op-
tical properties in organic media, as
corroborated by density functional
theory calculations. By applying dual-
wavelength detection techniques, this
compound shows complementary “off–
on–off” and “on–off–on” emission pro-
files upon pH variation, a behavior
that can be exploited to perform acidi-
ty detection over a broad pH range.
Keywords: catechols
· chemosen-
sors · fluorescent probes · pH detec-
tion · switches
Introduction
creasingly interesting owing to the need for monitoring pH
values in many chemical and biological processes, clinical
analysis, and environmental-protection analyses.[5] In addi-
tion, pH-responsive molecular systems displaying optical ac-
tivity have been proposed as logical operators to perform
molecular computing.[6]
In the last few years, the design, characterization, and
device implementation of chemosensors has emerged as an
active area of research owing to their wide-ranging applica-
tions in various fields of science and industry.[1] Because of
its high sensitivity and selectivity as well as its noninvasive
character, fluorescence is very often selected as the trans-
duction signal for chemical sensing events.[2] Thus, great
effort has been devoted to developing new materials for the
luminescent detection of chemically and biologically rele-
vant ions[3] and molecules.[4] Among them, fluorescent che-
mosensors that report on acidity changes are becoming in-
In contrast to glass electrodes, fluorescent chemosensors
are usually not suitable for measurements over wide pH
ranges, but they allow acidity detection in narrow pH-value
windows.[5e–g] In fact, the dynamic range of most common
chemosensors, which present only a single pH-responsive
group leading to one fluorescent and one non-fluorescent
state upon pH variation (i.e. “on–off” or “off–on” probes),
is limited to approximately two pH units.[5f] Several strat-
egies can be followed to broaden the sensing range of fluo-
rescent pH chemosensors. On the one hand, this can be ach-
ieved by mixing different systems that respond to distinct
pH values.[5f,g] More interestingly, pH detection over large
intervals can also be attained with specific chemosensors,
provided that they possess multiple pH-responsive groups
that give rise to several protonation states with distinct opti-
cal properties.[5e,7] Multistate molecular systems displaying
pH-induced “off–on–off”[7e,8] or “on–off–on”[9] fluorescence
profiles are particularly relevant examples of this latter case.
For “off–on–off” fluorescent pH chemosensors, minimal
emission is detected in strongly acid and basic media, where-
as maximal fluorescence is emitted at intermediate pH
values. On the contrary, “on–off–on” systems become in-
creasingly fluorescent at extreme pH values.
[a] E. Evangelio
Institut de Ci›ncia de Materials de Barcelona (CSIC)
Esfera UAB, 08193, Cerdanyola del Vall›s (Spain)
[b] Dr. J. Hernando, G. G. Bardají, Dr. R. AlibØs, Dr. F. BusquØ
Departament de Química, Universitat Autònoma de Barcelona
08193 Cerdanyola del Vall›s (Spain)
Fax : (+34)93-581-1265
E-mail: felix.busque@uab.es
[c] Dr. I. Imaz
Institut Català de Nanotecnologia, Edifici CM7 08193
Cerdanyola del Vall›s (Spain)
[d] Dr. D. Ruiz-Molina
Centro de Investigación on Nanociencia y Nanotecnología
Edifici CM7, Campus UAB 08193, Cerdanyola del Vall›s (Spain)
Fax : (+34)93-580-1853
E-mail: druiz@cin2.es
9754
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9754 – 9763

FULL PAPER
The vast majority of multistate fluorescent pH chemosen-
sors so far reported are complex molecular systems com-
posed of a fluorophore and electronically uncoupled pH-
sensitive units capable of modulating its luminescent proper-
ties. The mechanisms for such modulation are various, in-
cluding photoinduced electron transfer and electronic
energy transfer, among others.[5e,7–9] Alternatively, another
approach has been the development of intrinsic fluorescent
probes[2d] for wide-range pH detection, that is, chemosensors
in which multiple pH-responsive groups are part of the p sy-
stem of the fluorophore, whose inherent emissive properties
therefore vary with pH. This allows a relatively simple syn-
thesis of the chemosensors as well as an easier manipulation
of the resulting structures.
Inspired by the behavior of well-known pH-responsive
fluorescent probes such as fluorescein,[6a,10] our approach has
consisted of the synthesis of new fluorophores that contain
several groups with acid–base activity. Specifically, we have
focused our attention on catechol derivatives because of its
intrinsic acid–base properties (pKa =9.2 and 13.0 for pure
catechol in water)[11] and their other interesting features. For
instance, catechol derivatives have been successfully used in
the design of molecular switches[12] and have significant im-
plications in relevant biological processes such as the activa-
tion of small molecules like O2 and N2.[13] Herein, we de-
scribe the synthesis and the experimental and theoretical in-
vestigation of the optical properties of the catechol-derived
compounds 1 and 2. Compounds 1 and 2 have been pre-
pared by covalently coupling a catechol unit with phenyl
and pyridine moieties, respectively (see Scheme 1). p-Elec-
tron delocalization along both constituent groups is expect-
ed to lead to fluorescence emission in the visible region of
the spectrum with a significant pH sensitivity thanks to the
hydroxy groups of the catechol unit. More importantly, the
additional acid–base activity of pyridine (pKa =5.2 for the
pyridinium ion in water[14]) should make compound 2 a suit-
able fluorescent probe for acidity detection over a broad pH
interval, as recently assessed.[15]
derivatives was accomplished by using TBDPSCl and anhy-
drous 1,8-diazabicycloACHTRE[UGN 5.4.0]undec-7-ene (DBU) in a mix-
ture of THF and N,N-dimethylformamide (DMF), affording
compound 4 in 58% yield along with the monoprotected
product 5 in 7% yield. This reaction failed when bases other
than anhydrous DBU were used. A Wittig reaction of alde-
hyde 4 with benzyl-triphenylphosphonium chloride by using
tBuOK as the base yielded an inseparable 1:2 mixture of the
E and Z isomers of olefin 6 in 75% yield. This mixture was
treated with triethylamine trihydrofluoride, affording a 1:2
mixture of the neutral form of the E and Z isomers of 1 in
78% yield (n1; Scheme 2). These isomers were separated by
means of successive column chromatography. The isolated
E isomer proved to be stable enough to be used for further
experiments. On the contrary, the Z isomer isomerized back
in a few days to the original (Z)- and (E)-1 mixture even
upon storage in the dark and at 48C.
Scheme 2. Synthesis of compound n1: a) AlCl3, pyridine, CH2Cl2, 24 h,
reflux, 77%; b) TBDPSCl, DBU, THF/DMF, RT, 3 h, 58%; c) (Ph)3P+
CH2-Ph ClÀ, tBuOK, THF, RT, 2 h, 75%; d) TEA/3HF, THF, RT, 2 h
78%.
Wittig reaction of intermediate 4 and triphenyl-(2-pyridyl-
methyl)phosphonium chloride hydrochloride by using
tBuOK as the base afforded (E)-olefin 7 in 83% yield. Final
removal of the protecting silyl groups of 7 was achieved by
using triethylamine trihydrofluoride, affording, after addi-
tion of 35% HCl, the hydrochloride salt of 2 in 65% yield
(c2; Scheme 3). This yield decreased when tetrabutylammo-
nium fluoride (TBAF) was used as a fluoride source for this
reaction.
Scheme 1. Chemical structures of compounds 1 and 2.
Results and Discussion
X-ray structure: All the attempts to obtain single crystals of
1 were unsuccessful in spite of the recurrent use of different
crystallization techniques and solvents. More successful
were the attempts to crystallize compound 2. Yellow needles
suitable for X-ray diffraction were obtained by slow evapo-
ration of a solution of 2 in dichloromethane. Compound 2
crystallizes as the hydrochloride salt with one molecule in
the asymmetric unit. Figure 1 shows the atom labeling and
Synthesis: The sequence of reactions leading to target com-
pounds began with commercially available 5-bromovanillin.
Thus, cleavage of the methyl ether group of 5-bromovanillin
by using anhydrous AlCl3 in the presence of pyridine fur-
nished catechol 3[16] in 77% yield. Next, protection of the
hydroxy groups of 3 as their tert-butyldiphenylsilyl (TBDPS)
Chem. Eur. J. 2008, 14, 9754 – 9763
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9755

F. BusquØ, D. Ruiz-Molina et al.
Figure 2. Variation of the absorption spectrum of 1 upon pH increase in
acetonitrile (c1 =1”10À5 m). The spectra shown were measured at pH
values of 17.0, 17.7, 19.0, 20.5, 21.9, 22.4, 22.6, 23.1, 23.2, 23.5, and 24.0.
Scheme 3. Synthesis of compounds c2 and a2: a) (Ph)3P+CH2-Py-HCl
ClÀ, tBuOK, THF, RT, 2 h, 83%; b) i) TEA-3HF, THF, RT, 2 h; ii) HCl
35%, 65%; c) 1m NaOH, CH2Cl2, 94%.
Table 1. Experimental and TD-DFT energies and intensities of the UV/
Vis absorption bands of the different protonation states of 1 and 2.
State
Solvent
Experimental[a]
labs [nm]
eabs [LmolÀ1 cmÀ1
TD-DFT[b,c]
labs [nm]
]
n1
THF
acetonitrile
DMSO
325
319
329
21259
25098
20711
346
406
408
350
422
a1
c2
n2
a2
THF
acetonitrile
DMSO
398
388
399
24621
30385
21274
THF
acetonitrile
DMSO
379
371
382
22248
24558
22164
THF
acetonitrile
DMSO
334
327
338
25095
26773
24178
Figure 1. ORTEP view of the asymmetric unit of 2. Hydrogen atoms,
with the exception of the hydrogen atom that is implicated in the hydro-
gen bond with Cl1, are omitted for clarity.
THF
acetonitrile
DMSO
424
406
421
32169
32362
28564
the anisotropic displacement ellipsoids of 2. The resolved
structure is planar (C5-C6-C7-C8 dihedral angle 1738), and
the intramolecular bond lengths and angles agree with those
expected. Hydrochloric acid is connected to the molecule
through a very strong hydrogen bond (N····Cl: 2.218 Š
NHCl: 178.3 8). Although the molecular packing shows that
the different molecules are organized in parallel rows, no
p–p interactions are expected between them owing to the
long intermolecular separation distance observed (5.583 Š).
We ascribe this result to the presence of the voluminous
halogenated atoms.
[a] Measured at the absorption maxima. [b] B3LYP/6–311+GACHTREUNG(d,p) level
and accounting for acetonitrile solvent. [c] The ratios of the computed os-
cillator strengths (f) are: fa1/fn1 =1.0, fc2/fn2 =0.8, and fa2/fn2 =1.0.
(a1), as previously reported for monohydroxystilbenes.,[17,18]
Importantly, this process can be fully reverted back upon
acid addition (hydrochloric acid), thus recovering the ab-
sorption spectrum of n1.
The acid–base equilibrium between n1 and a1 is depicted
in Scheme 4. Deprotonation to yield a1 has been tentatively
assigned to take place on the OH moiety at the para posi-
tion of the catechol group according to inductive and conju-
gation considerations. This fact was confirmed by quantum-
chemical calculations (see below). From the variations of
the absorption spectra with pH,[19] an acidity constant value
of pKa =22.6 has been determined for this process in aceto-
nitrile, which is in good agreement with that reported for 2-
bromophenol in the same solvent (pKa =23.9).[14] As the au-
toprotolysis constant of acetonitrile is pKACN ꢀ33,[20] n1 be-
haves as a weak acid in this solvent. Further basification of
a1 to yield the double phenolate form of 1 (pH>25 in ace-
tonitrile) was unsuccessful and resulted in irreversible degra-
Acid–base equilibria: The acid–base activity of the catechol
and phenyl/pyridine moieties in compounds 1 and 2 was in-
vestigated by UV/Vis absorption spectroscopy. Figure 2 dis-
plays the changes measured in the absorption spectrum of
an acetonitrile solution of n1 (c1 =1”10À5 m) upon base ad-
dition (tetrabutylammonium hydroxide; TBAH). Clearly, a
new band at labs =388 nm appears with increasing pH,
whereas the intensity of the initial peak at labs =319 nm de-
creases. Similar results are obtained in other solvents
(Table 1). The absorption red-shift encountered upon base
addition is consistent with the deprotonation of a hydroxy
group in the neutral form n1 to yield an anionic species
9756
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9754 – 9763

Fluorescent Chemosensors from Catechol Derivatives
FULL PAPER
ready reported for related styrylpyridine compounds.[22–23]
Moreover, from the variations of the absorption spectra
with pH, an acidity constant value of pKa =12.5 has been
determined for the deprotonation of the pyridinium group
in acetonitrile. This result nearly matches the acidity con-
stant reported for the pyridinium ion in acetonitrile (pKa =
12.3–12.6) and it demonstrates the behavior of c2 as a weak
acid in this solvent.[14]
Subsequent deprotonation of n2 to yield the anionic
monophenolate state a2 leads to additional variations in the
absorption bands, as depicted in Figure 3. Such variations
are indeed in agreement with those previously found for
compound 1, that is, red-shifted absorption upon base addi-
tion. This confirms deprotonation of the hydroxy group at
the para position of the catechol moiety of 2 at larger pH
values (pKa =21.8 in acetonitrile). Remarkably, this second
acid–base equilibrium is also fully reversible, that is, the a2
species can be reverted back to the n2 species, making feasi-
ble the three-state device shown in Scheme 4. However, as
already described for compound 1, further basification to
deprotonate the second -OH group of 2 leads to a degrada-
tion process (at pH>25 in acetonitrile), thereby preventing
reversible conversion of a2 into the double anionic dipheno-
late form of this compound. Similar results are obtained in
other solvents different from acetonitrile as summarized in
Table 1 in which the experimental energies and intensities of
the UV/Vis absorption bands of the different protonation
states of 2 are shown.
Scheme 4. Molecular structures of the neutral (n1) and anionic (a1)
states of 1 as well as of the cationic (c2), neutral (n2), and anionic (a2)
states of 2. The acidity constants determined spectrophotometrically for
the reversible acid–base equilibria between these states in acetonitrile
are indicated.
dation of the compound, which we ascribe to either (photo-
induced) addition of water to the ethylenic double bond[21]
or a polymerization process.
A more complex acid–base activity was encountered for
compound 2 (see Scheme 4). The absorption spectrum of a
1”10À5 m solution of c2 in acetonitrile displays two different
bands at labs =327 and 371 nm. The relative intensities of
these bands varies with the addition of HCl and TBA with
the peaks at 327 and 371 nm becoming predominant at
pH 14.1 and 8.5, respectively (Figure 3). This indicates that
even though complex 2 was originally in its cationic form,
an equilibrium between the cationic form c2 (labs =371) and
its neutral state in which the pyridinium group is deproton-
ated (n2, labs =327) occurs in solution in acetonitrile. Inter-
estingly, this process can be controlled and reversibly dis-
placed forward or backwards by acid–base titration, as al-
Quantum-chemical calculations: Theoretical calculations
were performed to further investigate the correspondence
between the different absorption bands found in our spec-
trophotometric acid–base titration experiments and the dis-
tinct states of compounds 1 and 2. Density functional theory
(DFT) calculations employing the B3LYP hybrid functional
and the 6–311+GCAHTRE(UNG d,p) basis set and accounting for acetoni-
trile solvent were carried out to obtain the ground electronic
state energies and optimal geometries of the different proto-
nation states of 1 and 2. Initially, computations of the mono-
phenolate forms a1 and a2 in which the deprotonated hy-
droxy group was systematically modified to lay at either the
para or meta position of the catechol moiety were per-
formed. Noticeably, the ground electronic state energy of
the former molecules (phenolate at the para position) was
found to be approximately 10 kcalmolÀ1 lower in energy
than that of the meta position. These results confirm the as-
signment of a1 and a2 to the structures shown in Scheme 4,
thus demonstrating the larger acidity of the OH substituent
at the para position of the catechol ring.
As is well-known for stilbene derivatives,[24] each one of
the protonation states of 1 and 2 can present several differ-
ent conformations associated with the rotation of the cate-
Figure 3. Variation of the absorption spectrum of 2 upon pH increase in
acetonitrile (c2 =1”10À5 m). Left: The changes arising from the acid–base
equilibrium between c2 and n2, which were measured at pH values of
8.5, 9.9, 10.7, 11.4, 11.8, 12.4, 12.5, 12.6, 12.9, 13.0, 13.7, and 14.1. Right:
The acid–base equilibrium between n2 and a2, which were registered at
pH 14.1, 15.0, 16.9, 18.5, 19.8, 20.4, 20.9, 21.5, 22.2, 22.8, 23.1, 23.9, 24.2,
and 24.8.
À
À
chol and pyridine rings around the C1 C2 and C3 C4
bonds, respectively. This fact is illustrated in Scheme 5, in
which the optimal geometries computed for the four possi-
ble rotamers of c2 are shown (a, b, c, and d) as a representa-
tive example.
Chem. Eur. J. 2008, 14, 9754 – 9763
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9757

F. BusquØ, D. Ruiz-Molina et al.
units) fall below 10% in all cases. We partly ascribe these
differences to the polarizable continuum model employed to
account for acetonitrile solvent. Although this method
allows accurate description of the electrostatic solvent ef-
fects, it does not consider the microscopic interactions be-
tween the solute and solvent molecules, such as hydrogen-
bond formation.[25] Proper evaluation of all these solvent ef-
fects would require computationally demanding calculations
in which both continuum and discrete solvation models
were considered.[25]
On the other hand, TD-DFT calculations also reveal that
the absorption transitions measured occur between the
HOMO and LUMO states of all investigated molecules.
These frontier orbitals are delocalized over both the cate-
chol and phenyl/pyridine rings, thereby explaining why the
absorption spectrum changes upon protonation/deprotona-
tion of substituents located in any of the two constituent
units of compounds 1 and 2.
Scheme 5. Optimal geometries computed for the different rotamers of c2
at the B3LYP/6–311+G(d,p) level and accounting for acetonitrile sol-
AHCTREUNG
vent. The ground-electronic-state energies of these conformers relative to
that of a are 0 (a), 1.3 (b), À1.4 (c), and À0.1 kcalmolÀ1 (d).
Similar planar structures were calculated for n2 and a2.
For n1 and a1, only two different conformers are possible,
which also present planar geometries. In all cases, our calcu-
lations yield very similar ground-electronic-state energies
for all possible rotamers of a given protonation state of 1
and 2 (less than 2.5 kcalmolÀ1 of difference). Thus, very
small energy differences were computed for the four differ-
ent rotamers of c2 shown in Scheme 5, which fall below the
accuracy of the calculation method employed. In spite of
this, the occurrence of an equilibrium mixture of different
rotamers in acetonitrile solution is anticipated, as previously
observed for other stilbene derivatives.[24] This has been con-
firmed by means of NMR spectroscopy measurements.
ROESY spectroscopy experiments of c2 in deuterated di-
methyl sulfoxide (DMSO) show a nuclear Overhauser effect
(NOE) between both ethylenic protons and all the hydrogen
atoms at the ortho positions of the pyridine and catechol
rings. Together with the presence of only one series of
pH-modulated fluorescence: The emission properties of the
different protonation states of compounds 1 and 2 have
been evaluated. According to their stilbene character, the
different protonation states derived from them exhibit fluo-
rescence. The fluorescence spectral maxima (lf) and quan-
tum yields (Ff) measured on different solvents are shown in
Table 2, whereas the corresponding fluorescence spectra in
acetonitrile are given in Figure 4. Clearly, n1 and a1 as well
as c2, n2, and a2 show distinct emissive behaviors, which
confirm their potential use as fluorescent pH chemosensors.
Both the neutral forms of 1 and 2 display a large red-shift
in fluorescence spectra upon deprotonation (ꢁ100 nm), as
previously observed for related monohydroxystilbenes[17,18]
and in the absorption measurements. Similarly, protonation
1
proton signals in the H NMR spectrum, this indicates the
occurrence of a dynamic equilibrium between different rota-
tional conformers in solution and at room temperature. An
average behavior is therefore measured in our spectroscopic
experiments.
Finally, time-dependent functional theory (TD-DFT) cal-
culations at the B3LYP/6–311+GACTHRE(UNG d,p) level and accounting
Table 2. Fluorescence properties of the different protonation states of 1
and 2.
State
Solvent[a]
lf [nm][b]
Ff
n1
THF
acetonitrile
DMSO
388
392
414
0.065
0.031
0.072
for acetonitrile solvent were carried out to compute the ex-
citation energies and oscillator strengths of the absorption
bands of the different protonation states of 1 and 2. To ac-
count for conformational equilibria, TD-DFT calculations
on the ground-electronic-state optimal geometries of all pos-
sible rotamers were performed. Although minor differences
of their spectroscopic parameters were observed for a given
protonation state, the final values of absorption energies
and oscillator strengths were averaged over all possible ro-
tamers by taking into account the Boltzmann distribution of
conformer populations at room temperature. The resulting
values are compared with the experimental ones in Table 1.
As can be seen, a good agreement between quantum-chemi-
cal calculations and spectrophotometric measurements is ob-
served. Thus, our computations reproduce, to a good extent,
the spectral shifts obtained experimentally upon pH varia-
tion as the deviations between the calculated excitation en-
ergies and the experimental absorption maxima (in energy
a1
c2
n2
a2
THF
acetonitrile
DMSO
488
484
501
0.036
0.066
0.054
THF
acetonitrile
DMSO
500
511
501
0.19
0.069
<0.01
THF
acetonitrile
DMSO
406
409
442
0.097
0.053
0.34
THF
acetonitrile
DMSO
523
524
532
0.20
0.22
0.36
[a] Solvent dielectric constants are eTHF =7.5, eACN =36.6, and eDMSO
=
47.2, whereas the Kamlet–Taft b factors accounting for their hydrogen-
bond acceptor capacity are bTHF =0.55, bACN =0.31, and bDMSO =0.76.[26]
[b] Measured at the emission maxima.
9758
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9754 – 9763

Fluorescent Chemosensors from Catechol Derivatives
FULL PAPER
effect of the nitrogen atom and the substitution of the hy-
droxyl group must account for the larger values of Ff that
are typically obtained in this work for 2 with respect to 1
and stilbene (Ff =0.04 in hexane[27]).
All protonation states of compound 2 display significant
solvent effects on their fluorescence quantum yields. In the
case of c2, a Ff value as high as 0.19 is determined in THF.
Nevertheless, increasing solvent polarity results in a dramat-
ic decrease in fluorescence quantum yield, which leads to
nearly complete suppression of emission for polar solvents
such as DMSO (see Table 2), DMF, ethanol, and water
(data not shown). A different behavior is encountered for
the neutral state n2, whose fluorescence quantum yield rises
with increasing hydrogen-bond accepting capacity of the sol-
vent, as also observed for n1. Unfortunately, such large
emission efficiencies of n2 in organic media are lost in aque-
ous solutions, as has previously been reported for other hy-
droxystilbenes.[17] For the anionic form a2, solvent effects on
the fluorescence quantum yield seem less important because
large and similar Ff values are measured in THF, acetoni-
trile, and DMSO. In this case, however, major effects are en-
countered when a2 is dissolved in solvents with hydrogen-
bond donor capacity that can interact with the deprotonated
hydroxy group of the phenolate. This results in very low Ff
values (data not shown). Thus, for instance, aqueous solu-
tions of a2 in water are nearly not fluorescent, as also ob-
served for c2 and n2. In spite of this, the large emission effi-
ciencies and different spectral properties observed for these
species in several other solvents allows us to envisage com-
pound 2 as a potential fluorescent pH chemosensor in or-
ganic media.
Figure 4. Fluorescence emission spectra of the different protonation
states of compounds 1 (lexc =280 nm, c1 =1”10À5 m) and 2 (lexc =310 nm,
c2 =1”10À5 m) in acetonitrile. Each spectrum is normalized with respect
to the absorption at the excitation wavelength. Acid (HCl) or base
(TBA) was added to adjust the pH of the solution to the values at which
full conversion into the desired protonation state was achieved in each
case: pH 17.0 (n1), 24.0 (a1), 8.5 (c2), 14.5 (n2), and 24.0 (a2).
of n2 to yield c2 not only leads to bathochromic absorption,
but additionally results in red-shifted fluorescence, which is
in agreement with previous reports on the luminescence of
related styrylpyridine compounds.[22,23] Solvatochromic ef-
fects on the fluorescence spectra of compounds 1 and 2 are
also observed, which we ascribe to both changes in solvent
polarity as well as in solvent hydrogen-bond accepting ca-
pacity, as quantified by means of the Kamlet–Taft
b
factor.[26] Thus, similar fluorescence maxima are, in general,
observed in THF and acetonitrile as the higher hydrogen-
bond accepting capacity of the former (bTHF =0.55, bACN
=
0.31) is counterbalanced by the larger polarity of the latter
(eTHF =7.5, eACN =36.6). In case of DMSO as the solvent, its
higher e (eDMSO =47.2) and b (bDMSO =0.76) values lead to
red-shifted emission for n1, a1, n2, and a2. Opposite behav-
ior is found for c2, for which larger hydrogen-bond interac-
tion between the pyridinium group and THF or DMSO mol-
ecules results in hypsochromic displacement of the emission
spectrum. Our experimental data show that fluorescence
quantum yields also vary for different solvents and protona-
tion states of 1 and 2 (see Table 2).
Fluorescence pH-sensing: If a wide-range pH sensor based
on compound 2 is to be achieved, the capacity of this species
to univocally respond to acidity changes over a large inter-
val of pH values must be proved. Herein we discuss how the
pH-sensitive fluorescence properties of 2 can be exploited
to ensure the accomplishment of this condition.
In the previous section, we have shown that the fluores-
cence spectrum of the neutral form n2 is approximately
100 nm blue-shifted with respect to c2 and a2, whereas these
two ionic species emit in the same spectral region but with
different quantum yields (see Figure 4). As a consequence,
if selective detection at the maximum of the emission spec-
trum of n2 is performed (ꢁ400 nm), the fluorescence signal
measured will be maximal at intermediate pH values and it
will decrease for increasingly acid or basic media. This
means that compound 2 behaves as an “off–on–off” pH
sensor under such detection conditions,[7e,8] as is depicted in
Figure 5a. On the contrary, if the fluorescence detection
window is centred on the maximum of emission of c2 and a2
(ꢁ500 nm), the response of the system upon pH variation
becomes of the “on–off–on” type.[9] Therefore, maximal
fluorescence signal is detected at extreme values of pH in
this case, as also illustrated in Figure 5a. In Figure 5a, the
fluorescent “off–on–off” and “on–off–on” behaviors of 2 in
acetonitrile at different detection conditions are compared
In the case of compound 1, both n1 and a1 states present
relatively low Ff values within the 0.03–0.07 range. Howev-
er, while larger fluorescence quantum yields are obtained
for n1 in solvents with high hydrogen-bond accepting capaci-
ty (THF and DMSO), an opposite trend is observed for a1.
Most importantly, larger Ff values are determined for 2
under selected conditions, which is a relevant feature for po-
tential fluorescent-sensing applications. This contrasts with
previous studies on the effect of introducing a nitrogen
atom in the styryl system, in which a decrease in fluores-
cence emission from stilbene to 2-styrylpyridine was ob-
served.[27] Most probably, this difference arises from the
effect of the hydroxyl group and bromo substituents of the
catechol ring. Indeed, a tenfold increase in Ff has been re-
ported for 3-hydroxystilbene with respect to stilbene in or-
ganic solvents,[17] thus demonstrating the critical effect of
meta substitution with OH moieties on the photochemical
properties of styryl systems. Consequently, the combined
Chem. Eur. J. 2008, 14, 9754 – 9763
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9759

F. BusquØ, D. Ruiz-Molina et al.
probe as this compound shows both “off–on–off” and “on–
off–on” responses upon pH variation when dual-wavelength
detection is applied. However, this condition is not sufficient
for univocal pH sensing. To be capable of discerning low
from high pH values at which the fluorescence of the system
simultaneously turns “off” (at lem ꢁ400 nm) or “on” (at lem
ꢁ500 nm), the compound must display an asymmetric “on–
off–on” profile, that is, it must show different luminescent
properties at those two pH ranges. As mentioned above, this
is accomplished by 2 in some organic media owing to the
different fluorescence quantum yields of the c2 and a2
states. Consequently, each pH value will result in an univo-
cal set of emission intensities at the two monitored wave-
lengths. This is demonstrated by the 3D plot in Figure 5b,
which shows the correspondence between pH values and the
fluorescence intensities measured for 2 in acetonitrile at two
different emission wavelengths, lem =409 nm (F409
) and
524 nm (F524). Remarkably, a single value of pH can be re-
covered from a given pair of F409 and F524 values measured
by means of dual-wavelength detection. Therefore, we envis-
age compound 2 as a potential acidity chemosensor over
large pH intervals in organic media. Furthermore, because
of the fluorescent nature of all protonation states of this
species, future application of compound 2 for pH detection
could benefit from the use of ratiometric methods, thus pre-
venting the need for prior calibration of the sensing sys-
tem.[5d,f,7d]
Figure 5. a) pH dependence of the fluorescence intensity (F) of 2 in ace-
tonitrile (lexc =310 nm, c2 =1”10À5 m) at lem =409 nm (black) and lem
=
524 nm (gray). Experimentally measured intensities are represented as
points shown on the graphs. The solid lines plot the simulated emission
intensities that were computed by using the pKa constants, absorption ex-
tinction coefficients, and fluorescence quantum yields determined for 2 in
acetonitrile. b) 3D plot showing the correspondence between pH and the
fluorescence intensities measured for 2 in acetonitrile by means of dual-
wavelength detection at lem =409 and 524 nm (F409 and F524).
Conclusion
Herein we describe the synthesis and characterization of a
new chemosensor for a wide-range pH detection based on a
fluorescent system decorated with several pH-responsive
groups. With this aim, two catechol derivatives were pre-
pared by covalently coupling a catechol unit to phenyl (1)
and pyridine (2) rings. This led to p-delocalized systems that
display pH-sensitive fluorescence emission in the visible
region. Although 1 presents two stable protonation states
with distinct optical properties, both experimental measure-
ments and DFT calculations reveal that compound 2 shows
up to three different luminescent protonation states in or-
ganic media owing to the combined acid–base activity of its
constituent catechol and pyridine units. Importantly, the op-
tical behavior of these states is such that compound 2 dis-
plays complementary “off–on–off” and “on–off–on” emis-
sion profiles upon pH variation by applying dual-wavelength
detection techniques. This not only allows application of ra-
tiometric methods, but, more importantly, it can be exploit-
ed to perform acidity detection over a broad pH range in or-
ganic solvents.
with simulations performed by using the values of pKa, ab -
sorption extinction coefficients, and fluorescence quantum
yields that were previously determined for this species.
Clearly, a good agreement exists between the simulated and
experimental data. Thus, the dependence of fluorescence in-
tensity at lem =409 nm is rather symmetric around pH 17, as
expected from the calculations. More importantly, an asym-
metric curve describes the variation of the emission intensity
with pH at lem =524 nm, which arises from the different Ff
values of c2 and a2. This situation contrasts with the behav-
ior observed for previously reported fluorescence “on–off–
on” systems.[9]
The coexistence of the fluorescent “off–on–off” and “on–
off–on” behaviors of 2 in acetonitrile at different detection
conditions is an excellent scenario for the development of
the sensing routine. Fluorescent molecular chemosensors ex-
clusively displaying “off–on–off” or “on–off–on” profiles are
mainly suited for the detection of pH windows. Unfortu-
nately, they can hardly provide any information on the pH
values at which point the system is “off”, thus preventing
the device from sensing over larger pH intervals. This draw-
back can be overcome by the use of 2 as the fluorescent pH
Experimental Section
General procedures: Commercially available reagents were used as re-
ceived. The solvents were dried by distillation over the appropriate
9760
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9754 – 9763

Fluorescent Chemosensors from Catechol Derivatives
FULL PAPER
drying agents. All reactions were performed avoiding moisture by stan-
dard procedures and under nitrogen atmosphere and monitored by ana-
lytical thin-layer chromatography (TLC) by using silica gel 60 F254 pre-
coated aluminum plates (0.25 mm thickness). Flash column chromatogra-
phy was performed by using silica gel 60 Š, particle size 35–70 mm. NMR
spectroscopy experiments were performed at the Servei de Ressonància
Magn›tica Nuclear of the Universitat Autònoma de Barcelona. 1H NMR
spectra were recorded on Bruker DPX250 (250 MHz) and Bruker
DPX360 (360 MHz) spectrometers. Proton chemical shifts are reported
in ppm (d) (CDCl3, d=7.26 ppm or [D6]DMSO, d=2.50 ppm). 13C NMR
spectra were recorded on Bruker DPX250 (62.5 MHz) and Bruker
DPX360 (90 MHz) spectrometers with complete proton decoupling.
Carbon chemical shifts are reported in ppm (d) (CDCl3, d=77.2 ppm or
[D6]DMSO, d=39.5 ppm). NMR signals were assigned with the help of
COSY, HSQC, HMBC, and NOESY experiments. Infrared spectra were
recorded on a Sapphire-ATR spectrophotometer; peaks are reported in
cmÀ1. High-resolution mass spectra (HRMS) were recorded at Micro-
mass-AutoSpec by using (ESI+ or ESIÀ).
5.89 (d, J=12.2 Hz, Z) (1H), 1.09 (s, E) + 1.07 (s, Z) (9H), 0.76 (s, E) +
0.73 ppm (s, Z) (9H); 13C NMR (90 MHz, CDCl3): d=147.2, 147.1, 143.4,
143.0, 137.3, 136.7, 135.6, 135.5, 135.4, 133.7, 133.01, 132.8, 130.9, 130.8,
130.0, 129.9, 129.8, 129.7, 129.7, 128.9, 128.7, 128.4, 128.2, 128.1, 127.9,
127.8, 127.7, 127.6, 127.2, 126.8, 126.4, 125.9, 124.8, 121.7, 118.3, 116.0,
115.7 ppm; HRMS (ESI+): m/z: [M+Na]+ calcd for C46H47BrNaO2Si2
791.2180; found: 791.2168.
(E)-3-Bromo-5-(2-phenyl-1-ethenyl)-1,2-benzenediol (n1): Triethylamine
trihydrofluoride (0.63 mL, 3.92 mmol) was added dropwise at 08C to a
solution of 6 (1.50 g, 1.96 mmol) in dry THF (20 mL). The mixture was
stirred for 2 h at room temperature, quenched with brine (0.3 mL), dilut-
ed with ethyl ether (20 mL), and filtered. The filtrate was concentrated
under vacuum and the resulting crude material was purified by chroma-
tography on silica gel (gradient, hexanes/CH2Cl2 5:1 to 3:1) to afford a
1:2 mixture of E and Z isomers of 3-bromo-5-(2-phenyl-1-ethenyl)-1,2-
benzenediol (7) as a solid (0.443 g, 1.52 mmol; 78%). Repeated column
chromatography allowed isolation of the pure E isomer and a Z-enriched
mixture (4:1) of isomers. (E)-8: m.p. 95–968C (ethyl ether); 1H NMR
(250 MHz, CDCl3): d=7.50–7–39 (m, 2H), 7.39–7.28 (m, 2H), 7.28–7.18
(m, 1H), 7.14 (d, J=2.7 Hz, 1H), 7.03 (d, J=2.7 Hz, 1H), 6.93 (d, J=
16.3 Hz, 1H), 6.86 (d, J=16.3 Hz, 1H), 5.52 ppm (br s, 2H); 13C NMR
(62.5 MHz, CDCl3): d=144.7, 139.9, 137.2, 132.2, 128.9, 128.6, 127.9,
127.0, 126.6, 121.9, 112.8, 109.9 ppm; IR (ATR): n˜ =3422, 3356, 1590,
3-Bromo-4,5-dihydroxybenzaldehyde (3):[16] Anhydrous aluminum chlo-
ride (3.2 g, 23.8 mmol) was suspended in a solution of 5-bromovanillin
(5.0 g, 21.6 mmol) in CH2Cl2 (50 mL). While cooling to maintain the tem-
perature at 30–358C, pyridine (7.7 mL, 95.2 mmol) was added slowly. The
resulting clear solution was heated to reflux for 24 h, allowed to warm to
room temperature, quenched by the slow addition of 10% HCl solution
(60 mL), and further diluted with CH2Cl2 (30 mL). The resulting suspen-
sion was filtered to afford catechol 3 as a white solid (3.60 g, 16.6 mmol;
77%). M.p. 225–2278C.
1521, 1427, 1291, 959, 858, 830, 750, 692 cmÀ1
; HRMS (ESI): m/z:
[MÀH+]À calcd for C14H10BrO2 288.9859; found: 288.9861. Mixture (4:1)
1
of (Z)- and (E)-7 isomers: H NMR (360 MHz, CDCl3): d=7.50–7.43 (m,
E) + 7.39–7.31 (m, E) + 7.30–7.15 (m) (5H), 7.17 (d, J=1.8 Hz, E) +
7.06 (d, J=1.8 Hz, E) + 6.75 (d, J=1.6 Hz, Z), + 6.92 (d, J=1.6 Hz, Z)
(2H), 6.96 (d, J=16.2 Hz, E) + 6.90 (d, J=16.2 Hz, E) + 6.55 (d, J=
12.0 Hz, Z) + 6.40 (d, J=12.0 Hz, Z) (2H), 5.52 (br s, E) + 5.48 (br s,
Z) + 5.39 (br s, Z) + 5.29 ppm (br s, E) (2H); 13C NMR (90 MHz,
CDCl3): d=144.7, 144.1, 139.9, 139.5, 137.2, 137.0, 132.2, 131.6, 130.6,
129.0, 128.9, 128.6, 128.5, 128.5, 127.9, 127.5, 127.0, 126.6, 124.1, 121.9,
115.5, 112.8, 109.9, 109.3 ppm.
3-Bromo-4,5-bis(tert-butyldiphenylsilyloxy)benzaldehyde
(4):
DBU
(6.5 mL, 43.8 mmol) was added dropwise over 20 min to a stirred solution
of 3 (3.8 g, 17.5 mmol) and TBDPSCl (10.5 mL, 40.3 mmol) in a mixture
of dry THF (100 mL) and anhydrous DMF (20 mL) at 08C. After the
mixture had been stirred at room temperature for 3 h, ethyl ether
(100 mL) was added and the resulting suspension filtered. The filtrate
was concentrated under vacuum and the resulting yellowish wax was pu-
rified by column chromatography on silica gel (gradient, hexanes/CH2Cl2
4:1 to 1:1) to afford the following fractions: i) 3-bromo-4,5-bis(tert-butyl-
diphenylsilyloxy)benzaldehyde (4) as a solid (7.10 g, 10.2 mmol; 58%);
and ii) 3-bromo À5-(tert-butyldiphenylsilyloxy)-4-hidroxybenzaldehyde
(5) as a solid (0.56 g, 1.20 mmol; 7%). 4: m.p. 151–1538C (ethyl ether);
1H NMR (250 MHz, CDCl3): d=9.29 (s, 1H), 7.83–7.72 (m, 4H), 7.61 (d,
J=1.9 Hz), 7.53–7.20 (m, 16H), 6.86 (d, J=1.9 Hz), 1.13 (s, 9H),
0.76 ppm (s, 9H); 13C NMR (62.5 MHz, CDCl3): d=189.4, 149.3, 147.6,
135.5, 135.3, 133.0, 132.0, 130.1, 130.0, 129.9, 127.9, 127.8, 127.3, 122.2,
117.0, 27.1, 26.4, 20.7, 19.3 ppm; IR (ATR): n˜ =3267, 2930, 2856, 1669,
1428, 1314, 1300, 1100, 885, 701 cmÀ1; HRMS (ESI): m/z: [M+2+Na+]+
calcd for C39H41BrNaO3Si2 717.1657; found: 717.1640. 5: m.p. 126–1278C
(E)-2-[3-Bromo-4,5-di(tert-butyldiphenylsilyloxy)phenyl]-1-ethenyl]pyri-
dine (7): tBuOK (1.11 g, 9.94 mmol) was added portionwise at 08C to a
solution of triphenyl-(2-pyridylmethyl)phosphonium chloride hydrochlo-
ride (1.84 g, 4.32 mmol) in dry THF (40 mL). After the mixture had been
stirred for 30 min at room temperature, a solution of 4 (3.0 g, 4.32 mmol)
in dry THF (10 mL) was added. The mixture was stirred for 2 h at room
temperature, quenched with brine (0.5 mL), diluted with ethyl ether
(30 mL), and filtered. The filtrate was concentrated under vacuum and
the resulting crude material was purified by flash chromatography on
silica gel (gradient, hexanes/EtOAc 6:1 to 4:1) to afford (E)-2-[3-bromo-
4,5-di(tert-butyldiphenylsilyloxy)phenyl]-1-ethenyl]pyridine (7) as a solid
(2.76 g, 3.58 mmol; 83%). M.p. 182–1858C (ethyl ether); 1H NMR
(360 MHz, CDCl3): d=8.52–8.47 (m, 1H), 7.81–7.75 (m, 4H), 7.56 (dt,
J=7.9 Hz, J=1.8 Hz, 1H), 7.45–7.20 (m, 16H), 7.22 (d, J=2.2 Hz, 1H),
7.07 (d, J=15.8 Hz, 1H), 7.09–7.02 (m, 1H), 7.02 (d, J=7.9 Hz, 1H), 6.60
(d, J=2.2 Hz, 1 h), 6.21 (d, J=15.8 Hz, 1H), 1.10 (s, 9H), 0.75 ppm (s,
9H); 13C NMR (90 MHz, CDCl3): d=155.4, 149.7, 147.2, 144.0, 136.6,
135.5, 135.4, 133.6, 132.8, 130.9, 130.3, 129.9, 129.7, 127.9, 127.7, 127.1,
125.1, 122.6, 122.0, 119.2, 116.1, 27.1, 26.5, 20.7, 19.3 ppm; IR (ATR): n˜ =
1
(ethyl ether); H NMR (250 MHz, CDCl3): d=9.42 (s, 1H), 7.75–7.64 (m,
4H), 7.60 (d, J=1.7 Hz, 1H), 7.53–7.34 (m, 6H), 6.84 (d, J=1.7 Hz, 1H),
6.57 (s, 1H), 1.15 ppm (s, 9H); 13C NMR (62.5 MHz, CDCl3): d=189.3,
150.3, 143.5, 135.5, 130.9, 129.8, 128.5, 128.4, 117.9, 109.2, 26.8, 19.7 ppm;
IR (ATR): n˜ =2960, 2856, 1696, 1424, 1310, 1107, 917, 870, 697 cmÀ1
;
HRMS (ESI+): m/z: [M+2+Na]+ calcd for C23H23BrNaO3Si 479.0475;
found: 479.0464.
2932, 2858, 1584, 1563, 1547, 1477, 1333, 1307, 1265, 1107, 929, 701 cmÀ1
;
(E)-3-Bromo-5-(2-phenyl-1-ethenyl)-1,2-di(tert-butyldiphenylsilyloxy)-
benzene (6): tBuOK (0.419 g, 3.75 mmol) was added portionwise at 08C
to a solution of benzyltriphenylphosphonium chloride (1.23 g, 3.17 mmol)
in dry THF (25 mL). After the mixture had been stirred for 30 min at
room temperature, a solution of 4 (2.0 g, 2.88 mmol) in dry THF (5 mL)
was added. The mixture was stirred for 2 h at room temperature,
quenched with brine (0.3 mL), diluted with ethyl ether (20 mL), and fil-
tered. The filtrate was concentrated under vacuum and the resulting
crude material was purified by column chromatography on silica gel (gra-
dient, hexanes/CH2Cl2 5:1 to 3:1) to afford a 1:2 mixture of E and
Z isomers of 3-bromo-5-(2-phenyl-1-ethenyl)-1,2-di(tert-butyldiphenylsilyl-
oxy)benzene (6) as a wax (1.65 g, 2.15 mmol; 75%). 1H NMR (360 MHz,
CDCl3): d=7.88–7.70 (m, 4H), 7.50–6.90 (m) + 6.84 (d, J=1.8 Hz, Z)
(22H), 6.56 (d, J=1.8 Hz, E) + 6.26 (d, J=1.8 Hz, Z) (1H), 6.49 (d, J=
16.2 Hz, E) + 6.22 (d, J=12.2 Hz, Z) (1H), 6.05 (d, J=16.2 Hz, E) +
HRMS (ESI+): m/z: [M+H]+ calcd for C45H47BrNO2Si2 770.2313;
found: 770.2289.
(E)-2-[(3-bromo-4,5-dihydroxyphenyl)-1-ethenyl]pyridinium
chloride
(c2): Triethylamine trihydrofluoride (0.38 mL, 2.34 mmol) was added
dropwise at 08C to a solution of 7 (0.60 g, 0.78 mmol) in dry THF
(10 mL). After the mixture had been stirred for 2 h at room temperature,
35% HCl was added (0.5 mL) and the resulting suspension was stirred
for 10 min at room temperature and then filtered. The solid was washed
with ethyl ether (2”5mL) and dried under vacuum to afford (E)-2-[(3-
bromo-4,5-dihydroxyphenyl)-1-ethenyl]pyridinium chloride (c2) as
a
yellow solid (0.166 g, 0.51 mmol; 65%). M.p. 210–2128C (acetone);
1H NMR (360 MHz, [D6]DMSO): d=10.44 (br s, 1H), 9.93 (br s, 1H),
8.70 (d, J=5.0 Hz, 1H), 8.43 (t, J=7.7 Hz, 1H), 8.26 (d, J=8.3 Hz, 1H),
7.96 (d, J=16.6 Hz, 1H), 7.75 (t, J=6.8 Hz, 1H), 7.31 (d, J=2.0 Hz, 1H),
Chem. Eur. J. 2008, 14, 9754 – 9763
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9761

F. BusquØ, D. Ruiz-Molina et al.
7.19 ppm (d, J=2.0 Hz, 1H); 13C NMR (90 MHz, [D6]DMSO): d=150.5,
146.6, 145.6, 144.8, 141.4, 139.2, 127.2, 124.0, 123.7, 123.5, 116.9, 113.2,
110.0 ppm; IR (ATR): n˜ =3049, 2811, 1637, 1614, 1593, 1429, 1296, 1163,
Basabe-Desmonts, D. N. Reinhoudt, M. Crego-Calama, Chem. Soc.
Rev. 2007, 36, 993–1017.
[3] a) B. Valeur, I. Leray, Coord. Chem. Rev. 2000, 205, 3–40; b ) P. D.
Beer, P. A. Gale, Angew. Chem. Int. Ed. 2001, 40, 487–516; c) R.
Martínez-MµÇez, F. Sancenón, Chem. Rev. 2003, 103, 4419–4476;
d) L. Fabbrizzi, M. Licehelli, A. Taglietti, Dalton Trans. 2003, 3471–
3479; e) G. W. Gokel, W. M. Leevy, M. E. Weber, Chem. Rev. 2004,
104, 2723–2750.
964, 830, 760 cmÀ1
;
HRMS (ESIÀ): m/z: [MÀHClÀH+]À calcd for
C13H9BrNO2 289.9822; found: 289.9817.
Crystal data: Owing to the low diffraction intensity of the crystals, the
diffraction data were collected under synchrotron radiation (l=0.977 Š)
at 150 K with a CCD detector in the BM16 Spanish beamline at the
ESRF (Grenoble). The crystal was coated in paratone. The structure was
solved by direct methods and refined by full-matrix least-squares tech-
niques on F2 with SHELX-97.[30] Crystal adsorption was corrected by
Scalepack.[31] Non-hydrogen atoms were refined anisotropically. Crystal-
lographic details for 2: Mr =441.84, crystal size 0.3”0.2”0.03, monoclinic
phase, space group P21/c, a=9.975(2), b=20.793(3), c=16.760(2) Š, V=
[4] a) J. W. Bell, N. M. Hext, Chem. Soc. Rev. 2004, 33, 589–598; b) R.
Jelinek, S. Kolusheva, Chem. Rev. 2004, 104, 5987–6016.
[5] a) A. P. de Silva, R. A. D. D. Rupasinghe, J. Chem. Soc. Chem.
Commun. 1985, 1669–1670; b) S. G. Schulman, S. Chen, F. Bai,
M. J. P. Leiner, L. Weis, O. S. Wolfbeis, Anal. Chim. Acta 1995, 304,
165–170; c) L. M. Daffy, A. P. de Silva, H. Q. N. Gunaratne, C.
Huber, P. L. M. Lynch, T. Werner, O. S. Wolfbeis, Chem. Eur. J.
1998, 4, 1810–1815; d) S. Charier, O. Ruel, J.-B. Baudin, D. Alcor,
J.-F. Allemand, A. Meglio, L. Jullien, Angew. Chem. 2004, 116,
4889–4892; Angew. Chem. Int. Ed. 2004, 43, 4785–4788; e) G. Nishi-
mura, Y. Shiraishi, T. Hirai, Chem. Commun. 2005, 5313–5315;
f) A. S. Vasylevska, A. A. Karasyov, S. M. Borisov, C. Krause, Anal.
Bional. Chem. 2007, 387, 2131–2141; g) A. P. de Silva, S. S. K. de Sil-
va, N. C. W. Goonesekera, H. Q. N. Gunaratne, P. L. Lynch, R. K.
Nesbitt, S. T. Patuwathavithana, N. L. D. Ramyalai, J. Am. Chem.
Soc. 2007, 129, 3050–3051; h) B. Tang, X. Liu, K. Xu, H. Huang, G.
Yang, L. An, Chem. Commun. 2007, 3726–3728.
3578(5) Š3, Z=8, 1calcd =1.078 gcmÀ3
,
F
A
mACHTER(UNG MoKa)=
0.642 mmÀ1. A total of 17129 reflections were measured in the range of
5.66ꢂ2qꢂ56.788, of which 13560 were unique (Rint =0.0235). Final R in-
dices: R1=0.0746 [I/2s(I)], wR2=0.1821 (all data) for 682 parameters;
À3
max./min. residual electron density 0.733/À1.130 eŠ
.
Steady-state absorption and fluorescence spectroscopy: UV/Vis spectra
were recorded at room temperature by using a Varian Cary05e spectro-
photometer. Fluorescence spectra were recorded at room temperature in
a Perkin Elemer LS 45 fluorescence spectrophotometer. The fluores-
cence quantum yields of all solutions investigated were determined rela-
tive to N,N’-bis(1-hexylheptyl)perylene-3,4,9,10 in acetonitrile (Ff =1).[28]
[6] a) D. Margulies, G. Melman, A. Shanzer, Nat. Mater. 2005, 4, 768–
771; b) A. P. de Silva, S. Uchiyama, Nat. Nanotechnol. 2007, 2, 399–
410.
pH measurements: pH measurements upon acid–base addition to aceto-
nitrile solutions of 1 and 2 were performed at room temperature with a
Crison 52–01 pH electrode in a Crison BASIC 20 potentiometer. pH
values are given relative to the acetonitrile solvent (sspH scale).[29] To cal-
ibrate the electrode system, we used reference buffer solutios in acetoni-
trile (pyridine–pyridinium bromide and phenol–sodium phenolate),
whose sspH can be derived from the Henderson–Hasselbach equation by
using the pKa values in acetonitrile reported for these systems.[29]
[7] a) T. Gunnlaugsson, D. Parker, Chem. Commun. 1998, 511–512;
b) M. Su, H. Ma, Q. Ma, Z. Wang, J. Yang, M. Wang, Chem.
Commun. 2001, 960–961; c) F. Pina, J. C. Lima, C. Lodeiro, J. S.
de Melo, P. Díaz, M. T. Albelda, E. García-EspaÇa, J. Phys. Chem.
A 2002, 106, 8207–8212; d) V. F. Valuk, G. Duportail, V. G. Pivovar-
enko, J. Photochem. Photobiol. A 2005, 175, 226–231; e) Y. Shir-
aishi, Y. Tokitoh, G. Nishimura, T. Hirai, J. Phys. Chem. B 2007,
111, 5090–5100.
[8] a) A. P. de Silva, H. Q. N. Gunaratne, C. P. McCoy, Chem. Commun.
1996, 2399–2400; b) T. Gunnlaugsson, J. P. Leonard, K. SØnØchal,
A. J. Harte, J. Am. Chem. Soc. 2003, 125, 12062–12063; c) S. A. de
Silva, K. C. Loo, B. Amorelli, S. L. Pathirana, M. Nyakirangµni, M.
Dharmasena, S. Demarais, B. Dorcley, P. Pullay, Y. A. Salih, J.
Mater. Chem. 2005, 15, 2791–2795; d) Y. Diaz-Fernandez, F. Fori, C.
Mangano, P. Pallavicini, S. Patroni, A. Perez-Gramatges, S. Rodri-
guez-Calvo, Chem. Eur. J. 2006, 12, 921–930.
Quantum-chemical calculations: Quantum-chemical calculations were
performed by employing the Gaussian 03 package of programs[30] on a
32-bit multiprocessor computer. DFT geometrical optimization of ground
electronic states of all investigated molecules was carried out at the
B3LYP hybrid functional level with the 6–311+GACHTRE(UNG d,p) basis set. To ac-
count for solvent polarity effects, calculations were performed in acetoni-
trile solvent by means of the polarizable continuum model (PCM, e=
36.64). The resulting geometries and molecular orbitals were plotted by
using Molden. The excitation energies and oscillator strengths (f) of the
UV/Vis absorption transition bands were then computed by means of
time-dependent DFT at the B3LYP/6–311G+(d,p) level on the B3LYP/
6–311+GACHTREUNG(d,p) ground-state geometries and accounting for acetonitrile
[9] a) P. Pallavicini, V. Amendola, C. Massera, E. Mundum, A. Taglietti,
Chem. Commun. 2002, 2452–2453; b) V. Amendola, L. Fabrrizzi, C.
Mangano, H. Miller, P. Pallavicini, A. Parotti, A. Taglietti, Angew.
Chem. 2002, 114, 2665–2668; Angew. Chem. Int. Ed. 2002, 41, 2553–
2556; c) Z. Wang, G. Zheng, P. Lu, Org. Lett. 2005, 7, 3669–3672;
d) G. Zheng, Z. Wang, L. Tang, P. Lu, W. P. Weber, Sens. Actuators
B 2007, 122, 389–394.
solvent (PCM). For each protonation state of 1 and 2, the spectroscopic
properties of all possible rotamers were calculated and the final average
values were obtained by using Boltzmann coefficients at room tempera-
ture (exp(ÀErotamer/kBT)/ꢀ exp(ÀErotamer/kBT)) as weighting factors.
[10] R. Sjçback, J. Nygren, M. Kubista, Spectrochim. Acta Part A 1995,
51, L7L21.
[11] R. J. Motekaitis, A. E. Martell, Inorg. Chem. 1984, 23, 18–23.
[12] E. Evangelio, D. Ruiz-Molina, Eur. J. Inorg. Chem. 2005, 15, 2957–
2971.
[13] C. G. Pierpont, C. W. Lange, Prog. Inorg. Chem. 1994, 41, 331–443.
[14] S. Espinosa, E. Bosch, M. RosØs, J. Chromatogr. A 2002, 964, 55–66.
[15] As a proof-of-concept, the implementation of 2 in a nanostructured
pH sensor and the performance of the resulting device has also al-
ready been reported in a previous communication: A. Martinez-
Otero, E. Evangelio, R. Alibes, J. L. Bourdelande, D. Ruiz-Molina,
F. Busque, J. Hernando, Langmuir 2008, 24, 2963–2966.
[16] R. G. Lange, J. Org. Chem. 1962, 27, 2037–2039.
Acknowledgements
We acknowledge the financial support of the Ministerio de Educación y
Ciencia (MEC) through projects MAT2006–13765-C02–01, CTQ2004–
02539, and CTQ2006–01040. F.B. and J.H. thank MCyT-ERDF for their
“Ramon y Cajal” contracts, E.E. thanks the MEC for a predoctoral
grant, and G.G.B. thanks DURSI and FSE for a predoctoral grant.
[1] B. R. Eggins, Chemical Senors and Biosensors (Analytical Tech-
niques in the Sciences), Wiley-VCH, Weinheim, 2002.
[2] a) Fluorescent Chemosensors for Ion and Molecule Recognition
(Ed.: A. W. Czarnik), ACS Books, Washington, 1993; b) A. P. de Sil-
va, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P.
McCoy, J. T. Rademacher, T. E. Rice, Chem. Rev. 1997, 97, 1515–
1566; c) O. S. Wolfbeis, J. Mater. Chem. 2005, 15, 2657–2669; d) L.
[17] F. D. Lewis, E. M. Crompton, J. Am. Chem. Soc. 2003, 125, 4044–
4045.
[18] E. M. Crompton, F. D. Lewis, Photochem. Photobiol. Sci. 2004, 3,
660–668.
9762
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9754 – 9763

Fluorescent Chemosensors from Catechol Derivatives
FULL PAPER
[19] H. Ilhmels, A. Meiswinkel, C. J. Mohrschladt, D. Otto, M. Waidelich,
M. Towler, R. White, M. Albrecht, A. Schnurpfeil. J. Org. Chem.
2005, 70, 3929–3938.
[20] I. M. Kolthoff, M.K. Chantoon, J. Phys. Chem. 1968, 72, 2270–2272.
[21] T. Murohoshi, K. Kaneda, M. Ikegami, T. Arai, Photochem. Photo-
biol. Sci. 2003, 2, 1247–1249.
[22] G. Favaro, U. Mazzucato, F. Masetti, J. Phys. Chem. 1973, 77, 601–
604.
[23] G. Ginocchietti, U. Mazzucato, A. Spalletti, Int. J. Photoenergy
2004, 6, 241–250.
[24] U. Mazzucato, F. Momicchioli, Chem. Rev. 1991, 91, 1679–1719.
[25] a) B. Mennucci, J. Am. Chem. Soc. 2002, 124, 1506–1515; b) J.
Tomasi, R. Cammi, B. Mennucci, C. Cappell, S. Corni, Phys. Chem.
Chem. Phys. 2002, 4, 5697–5712.
[26] M. J. Kamlet, J. L. M. Abboud, M. H. Abraham and R. W. Taft, J.
Org. Chem. 1983, 48, 2982–2996.
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[27] G. Marconi, G. Bartocci, U. Mazzucato, A. Spalletti, F. Abbate, L.
Angeloni, E. Castellucci, Chem. Phys. 1995, 196, 383–393.
[28] T. Kircher, H.-G. Lçhmannsrçben, Phys. Chem. Chem. Phys. 1999,
1, 3987–3992.
[31] G. M. Sheldrick, Program for the Refinement of Crystal Structures;
University of Gçttingen, Germany, 1997.
[32] Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307–326.
[29] S. Espinosa, E. Bosch, M. RosØs, Anal. Chem. 2000, 72, 5193–5200.
[30] Gaussian 03 (Revision B.04), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
Received: April 15, 2008
Revised: July 10, 2008
Published online: September 11, 2008
Chem. Eur. J. 2008, 14, 9754 – 9763
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9763