Time resolved fluorescence spectroscopy of quercetin and morin complexes with Al3+

Article abstract of DOI:10.1016/S1386-1425(01)00515-7

The association process of Al³⁺ with quercetin and morin in methanol was studied by electronic absorption and emission spectroscopies. The number of species in solution with different absorption spectra were determined by the method of Rank ana

Full text of DOI:10.1016/S1386-1425(01)00515-7

Spectrochimica Acta Part A 58 (2002) 83–89
www.elsevier.com/locate/saa
Time resolved fluorescence spectroscopy of quercetin and
morin complexes with Al3+
Amanda C. Gutierrez, Marcelo H. Gehlen *
Instituto de Qu´ımica de Sa˜o Carlos, Uni6ersidade de Sa˜o Paulo, Caixa Postal 780 13560-970, Sao Carlos, SP, Brazil
Received 9 March 2001; accepted 26 April 2001
Abstract
The association process of Al3+ with quercetin and morin in methanol was studied by electronic absorption and
emission spectroscopies. The number of species in solution with different absorption spectra were determined by the
method of Rank analysis of the absorbance matrix, and the stoichiometries of the complexes were evaluated using the
Job method. The number of fluorescent species in solution were calculated from the Rank analysis method of the time
resolved emission spectra (TRES), and compared with a global analysis of the decay surface using a proper
multi-exponential decay model. The association of Al3+ with morin gives rise to two complexes with 1:1 and 2:1
(morin: Al3+) stoichiometries, but in both species the association of the cation involves the carbonyl and 3-hydroxyl
groups of the pyrone ring. The fluorescence decay surface of this system is biexponential and the lifetimes of the 1:1
and 2:1 complexes are 4.3 and 2.0 ns, respectively. The association of Al3+ with quercetin forms preferentially two
complexes with 1:1 and 1:2 (quercetin: Al3+) stoichiometries where the first cation binds to the site of the pyrone ring
but the second one is bound to the cathecol group of the molecule. However, the multichelation character of the
quercetin ligand allows larger aggregates to be formed, thereby the species Al2Q3 is also detected in methanol. The
lifetime of the 1:1 complex is about 2.7 ns, while for 1:2 and 3:2 complexes the lifetimes measured are 3.5 and 1.8 ns,
respectively. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Flavonoids; Aluminum (III); Chelation; Fluorescence lifetimes; TRES and Rank analysis
several cases are highly fluorescent, a property
which has been explored in analytical methods of
metal and ligand identification [1–10]. For in-
stance, quercetin and morin are well known
reagents for determination of aluminum(III)
traces in water and in biological samples. Besides
these analytical applications, the chelation of
mono and polyhydroxy flavones with cations is an
important factor in their bioactivity as carriers
and regulators of metal concentration. Free or
1. Introduction
Quercetin and morin are phenolic compounds
derived from hydroxyl substitutions on the
flavone chromophore. They complexate with
metal cations to form stable products which in
* Corresponding author. Tel.: +55-16-273-9951; fax: +55-
16-273-9952.
E-mail address: marcelog@iqsc.sc.usp.br (M.H. Gehlen).
1386-1425/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1386-1425(02)00515-7

84
A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
complexated, these compounds are also anti-oxi-
dants and free radical scavengers in biological
systems [5,11–13].
addition of small amounts of its stock solution,
via a microsyringe, followed by a stirring period
to allow the equilibrium condition of the com-
plexation process to be reached. The systems were
also investigated in acidic methanol solution con-
taining 0.01 M of nitric acid.
Absorption measurements were performed on a
Hitachi U-2000 spectrophotometer. Corrected
steady-state fluorescence spectra were recorded on
a CD-900 Edinburgh spectrofluorimeter. For
measurements, air equilibrated samples in 1 cm
quartz cuvettes were thermostated at 298 K. Fluo-
rescence quantum yields were determined by using
as standard the dye acridine orange, assuming a
value of 0.4 in methanol. Fluorescence decay sur-
faces were measured by time correlated single-
The enhancement of the fluorescence signal
upon chelation of flavones with a nonparamag-
netic metal is related to the inhibition of the
excited state intramolecular proton transfer
(ESPT) process [14–21] between hydroxyl and
4-keto groups of the cromone ring. The ESPT
mechanism, which occurs in several hydroxyl sub-
stituted flavones, give rises to a fast excited state
equilibrium between the normal and tautomeric
forms, and therefore to dual fluorescence usually
with low emission quantum yields at room
temperature.
Most of the studies of quercetin and morin
association with metals have focused attention on
the investigation of the stoichiometry of the com-
plexes and determination of possible sites of bind-
ing [4–10]. In the present work, these points are
also addressed, but are discussed together with the
electronic excited state properties of quercetin and
morin complexes with aluminum (III) in methanol
solution.
The number of species in solution with different
absorption spectra is determined by the Rank
analysis method of the absorbance matrix, and
the stoichiometry of the complexes is evaluated
using the Job method. The number of fluorescent
species in solution is defined by the Rank analysis
method of the time resolved emission spectra
(TRES) matrix. The lifetime of the complexes in
solution is then determined from a global analysis
of the decay surface using a proper multi-expo-
nential decay model.
photon counting technique using
a CD-900
Edinburgh spectrometer operating with a hydro-
gen-filled nanosecond flash lamp at 40 kHz pulse
frequency. The data were analyzed using global
multiexponential and time resolved emission spec-
tra (TRES) routines of the Edinburgh Instru-
ments Level 2 software.
In the Rank analysis of the absorbance ma-
trices, a modified version of the program Triang
by Hartley, Burgess and Alcock was used [22]. In
the case of the TRES data, the number of counts
of the traces were divided by the maximum value
to form a scaled emission spectra matrix for Rank
analysis. The Rank or the number of species in
solution with distinct spectra corresponds to the
number of diagonal elements of the trigonal ma-
trix which are larger in module than three times
the corresponding value of the reduced error ma-
trix. The Rank is determined as a function of the
error assumed in the measurements.
2. Experimental
3. Results and discussion
The flavonoids quercetin and morin (Aldrich)
were
recrystallized
from
ethanol,
and
3.1. Absorption spectroscopy
Al(NO3)3.9H2O (synth. \99.5%) was used as re-
ceived. Stock solutions of the reagents were pre-
pared in methanol at 1 mM concentration. In all
samples for absorption and emission spectroscopy
measurements, the flavonoid concentration was 10
mM obtained from dilution of the stock solutions.
The concentration of Al(III) was varied by the
The addition of Al3+ to a quercetin solution in
methanol results in significant change of the ab-
sorbance spectrum of the flavonoid solution, with
the appearance of a new band centered on 430 nm
with a bathochromic shift of about 58 nm from
the original band in absence of the metal. A

A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
85
typical example of the spectral changes observed
upon addition of Al3+ to quercetin solution is
illustrated in Fig. 1.
In the case of morin solution, the addition of
Al3+ gives first a hipsochromic shift with a
change of the maximum from 380 nm to 355 nm,
and the development of a shoulder at around 420
nm. Further addition of the metal leads to the
growth of a new band centered on 420 nm, with
an isosbestic point at 380 nm. Fig. 2 shows the
two stage change of the absorption spectrum of
the Morin/Al system. When the spectra of the
flavonoids are taken in acidic methanol solution,
the absorption bands observed are practically the
same as before, but the growth of the new band in
the red region of the spectrum is reduced in both
cases indicating that the amount of the new spe-
cies formed by metal complexation has changed.
For the morin/Al system however, the initial be-
havior observed in methanol disappeared in acid
medium. This result suggests that the hip-
sochromic shift of the absorption spectrum of
morin upon addition of a small amount of
Al(NO3)3 may be related to a change in the
dissociation equilibrium of the 4%–hydroxyl group
of morin (pKa=9.0) when adding the acid salt of
Al.
Fig. 2. Electronic absorption spectra of morin (10 mM) in
methanol during titration with Al(III). Concentration range:
0–4 mM (top); 4–10 mM (bottom).
All the spectral data obtained can be further
analysed by looking at the number of species in
solution with a distinct absorbance spectrum,
which corresponds to determination of the Rank
of the absorbance matrix for a given error in the
measurement. A typical plot of the number of
species in solution as a function of the error in the
measurement, taking the case of the quercetin/Al
system, is illustrated in Fig. 3. For absorbance
error in the range of 0.01 to 0.002, which corre-
sponds to the longer flat line of the plot, the
number of species for quercetin/Al is three and
two in neutral and acidic methanol solution, re-
spectively. For morin/Al system, the number of
species depends on concentration of the added
metal. From 0–10 mM of Al3+, the number of
species found in methanol is three. However, if
the spectral analysis is performed excluding the
data of the first additions of the metal, for in-
Fig. 1. Electronic absorption spectra of quercetin (10 mM) in
methanol during titration with Al(III), concentration from
0–10 mM, T=298 K.

86
A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
stance using only the spectral data for [Al]=0.4
to 10 mM, the number of species is reduced to
two. In acidic methanol solution, the number of
species in solution with distinct absorption spectra
is two, and it does not depend on the concentra-
tion range of added metal.
The stoichiometry of the complexation was in-
vestigated by using the Job method [22]. Consid-
ering a global equilibrium of Al (III) and n
ligands (L) on the form,
Al+nL=AlLn,
(1)
n is determined from the plot of the absorbance
as a function of the mole fraction x of the added
ligand. In the absorbance maximum,
Fig. 4. Job plot of the Morin/Al(III) system. Absorbance
measurements at 420 nm. (“) methanol solution; (ꢀ)
methanol+0.01 M HNO3.
3.2. Fluorescence spectroscopy
xmax
n=
(2)
1−xmax
The addition of Al to quercetin and morin
solutions gives rise in both cases to an intense
fluorescence signal which increases with metal
concentration. Excitation and emission spectra for
the quercetin/Al system are shown in Fig. 6 for
illustration. However, no time-resolved data have
been reported and analyzed with discussion of the
excited state properties in correlation with the
number of complexes formed and their stoi-
chiometries. With this aim, the decay surface of
each system is analyzed by the global method [23].
In this procedure, multiexponential decay func-
tion with linked lifetimes at different emission
A typical plot according to the Job method is
presented in Fig. 4 for the Al/morin system in
methanol and in acidic methanol. In methanol,
the most probable stoichiometry is n=2, while in
acid medium, n=1.
In the case of quercetin, the stoichiometry is
well defined in acidic methanol where n=1, but
in methanol solution, more than one type of
complex is formed, giving rise to species like
Al2Q, AlQ and Al2Q3. The Job plots for
quercetin/Al system in methanol are presented in
Fig. 5.
Fig. 3. The number of species in solution (Rank) of Quercetin/
Al(III) as a function of the error in absorbance measurements.
(“) methanol solution; (ꢀ) methanol+0.01 M HNO3.
Fig. 5. Job plot of the Quercetin/Al(III) system. (“) methanol
solution, and (ꢀ) methanol+0.01 M HNO3 with absorbance
measurements at 430 nm; (ꢁ) in methanol at 475 nm.

A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
87
Fig. 6. Normalized excitation and emission spectra of
Quercetin/Al(III) in methanol with both species at 10 mM.
Fig. 7. Scaled TRES data obtained at 3.0, 4.5, 6.0, 7.5, and 9.0
ns after the excitation pulse. [Quercetin]=[Al(III)]=10 mM,
uexc=430 nm, T=298 K.
wavelengths is used as the fitting condition. The
results found indicate that decay surface is biexpo-
nential in most of the systems and only monoexpo-
nential in the case of morin/Al in acidic methanol.
The lifetimes measured are listed in Table 1.
Time resolved emission spectra (TRES) may be
obtained by slicing the decay surface at different
times, and the whole set of spectra may be further
investigated by Rank analysis of the emission data.
Typical scaled TRES data are shown in Fig. 7. The
corresponding number of fluorescent species as a
function of the error in the relative intensity is
plotted in Fig. 8. The longer flat region indicates
the presence of two fluorescent species in the
Al/quercetin system, in accordance with the global
biexponential fitting and best chi-square value. The
Rank of the scaled TRES data for the other systems
are reported in Table 1. The result indicates a good
correlation between the Rank of the TRES matrix
and the number of exponential terms of the fluores-
cence decay process.
Concerning the results obtained, some useful
generalizations may be written. In the morin/Al
system in acidic methanol solution, the Rank of the
absorbance matrix is 2, the main stoichiometry is
1:1, and the decay is monoexponential. It means
that there are two molecular species in solution, the
morin (M) and its fluorescence complex AlM with
a lifetime of about 4.3 ns. In methanol, however,
AlM2 is formed preferentially but a small fraction
of the complex AlM should be present because the
Rank of the TRES matrix is 2 and the decay is
biexponential. The recovered lifetimes have the
values of 2.0 and 4.3 ns indicating that the short
lived component may be ascribed to the AlM2
species.
The association of quercetin with Al depends
more critically on concentration of metal and
solvent medium. In methanol/0.01 M HNO3, the
species AlQ is formed preferentially. The presence
Table 1
Stoichiometry of the complexes of Al(III) with Quercetin and Morin and their respective lifetimes from global analysis
2
global
Solvent
Stoichiometry
Lifetimesa (ns)

Rank of TRES
Methanol
Al2Q; Al2Q3
AlM; AlM2
AlQb
3.5; 1.8
4.3; 2.0
2.7; 4.9
4.3
1.090
1.024
1.102
1.126
2
2
2
1
Methanol+0.01 M HNO3
AlM
a The standard deviations of the lifetimes are about 0.3 ns.
b The major species formed. The biexpoential behavior is ascribed to a minor presence of the Al2Q complex.

88
A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
larger amounts. The decay surface is biexponen-
tial with components of 1.8 and 3.5 ns. With
addition of more Al, the decay becomes slow (see
Fig. 9). Global analysis of a decay surface, ob-
tained by measuring the fluorescence signal at
different concentrations of Al(III) but at a fixed
emission wavelength, indicates that the long lived
component with a decay time of 3.5 ns governs
the fluorescence deactivation at a high concentra-
tion of metal (see inset in Fig. 9). Thus the
lifetime of 3.5 ns should be ascribed to Al2Q
formed preferentially when more metal is added.
This result is in agreement with the increase in the
fluorescence quantum yield (ƒF) from 0.08 up to
0.12 with concentration of Al(III).
2
Fig. 8. Rank analysis of the TRES matrix of Fig. 7.  is the
value obtained from global biexponential and monoexponen-
tial analysis.
The major species in acid medium are the com-
plexes AlQ and AlM, and their ƒ values mea-
F
sured were 0.3 and 0.6, respectively. For the
Al/morin system in methanol, the ƒF values of the
AlM and AlM2 complexes, measured in 5:1 and
1:5 metal:ligand concentration ratios, were 0.5
and 0.2, respectively. Note that the quantum
yields measured are correlated with the lifetimes
of the complexes (see Table 1). Since ƒF=kF~,
of an intense fluorescence indicates that the bind-
ing site in AlQ is the 3-hydroxy and 4-keto groups
of the pyrone ring. The decay is biexponential,
but the main component with about 90% weight
and lifetime of 2.7 ns should correspond to the
AlQ complex. In methanol, the Rank of ab-
sorbance matrix is 3 indicating that two types of
complex plus the free quercetin are in solution.
The complexes Al2Q and Al2Q3 are formed in
the species with higher ƒ would have longer
F
lifetimes if the radiative rate constant kF remained
constant.
The main sites of Al(III) binding to morin and
quercetin are shown in the molecular models of
Fig. 10. It should be emphasized that in all sys-
tems investigated, the p*“p singlet emission ap-
pears because the binding of Al3+ to the carbonyl
and 3-hydroxyl groups of the pyrone ring inhibits
the intramolecular proton transfer in this site,
which forms a nonemissive or low emission pho-
totautomer at room temperature. In the samples
investigated the fluorescence occurs always from
the anion form of the 7-hydroxy group of the
pyrone ring because the pKa of the excited state is
−2.2, the pKa of the ground state is 7.4, and the
photodissociation is a very rapid process accord-
ing to results by Wolfbeis at al. [24].
Fig. 9. Fluorescence decays of Quercetin/Al(III) system in
methanol. [Quercetin]=30 mM, [Al(III)]=5, 10, 20, 80 mM
from bottom to top. uexc=430 nm, uem=495 nm, T=298 K.
Inset: normalized amplitudes of the biexponential global fitting
as a function of metal concentration. 2=1.072; (ꢁ) ~1=1.8
ns; (“) ~2=3.5 ns.
4. Conclusions
In this contribution, several spectroscopy meth-
ods, including Rank analysis of absorbance and

A.C. Gutierrez, M.H. Gehlen / Spectrochimica Acta Part A 58 (2002) 83–89
89
Fig. 10. Molecular models of Al(III) complexation with Morin and Quercetin.
[4] M.J. Ahmed, J. Hossan, Talanta 42 (1995) 1135.
[5] K. Robards, M. Antolovich, Analyst 122 (1997) 11R.
[6] P.C.H. Hollman, J.M.P. van Trijp, M.N.C.P. Buysman,
Anal. Chem. 68 (1996) 3511.
[7] L.J. Porter, K.R. Markham, J. Chem. Soc. C 9 (1970)
1309.
[8] J. Pusz, M. Kopacz, Polish J. Chem. 66 (1992) 1935.
[9] J. Pusz, B. Nitka, Microchem. J. 56 (1997) 373.
[10] H. Deng, G.J. Van Bekel, J. Mass Spectrom. 33 (1998)
1080.
[11] B. Havsteen, Biochem. Pharmacol. 32 (1983) 1141.
[12] J.E. Brown, H. Khodr, R.C. Hider, C.A. Rice-Evans,
Biochem. J. 330 (1998) 1173.
[13] M. Yoschino, K. Murakami, Anal. Biochem. 257 (1998)
40.
[14] P.K. Sengupta, M. Kasha, Chem. Phys. Lett. 68 (1979)
382.
[15] D. McMorrow, M. Kasha, J. Phys. Chem. 88 (1984)
2235.
TRES matrixes, stoichiometry by Job method,
and global fitting with exponential decay func-
tions, are used to determine the properties of the
complexes formed by quercetin and morin with
the cation Al(III) in methanol solution. The re-
sults obtained here are in agreement with previous
studies [4–10] confirming that morin forms AlM
and AlM2 complexes, while quercetin is more
likely to generate the AlQ and Al2Q species, but
larger aggregates, like the Al2Q3 complex, are not
ruled out. The lifetimes of these species are deter-
mined. In general, the complexes with more than
one ligand have a shorter lifetime than the mono-
ligand complex, and therefore they contribute less
to the fluorescence quantum yield in solution.
[16] A.J.G. Strandjord, P.F. Barbara, J. Phys. Chem. 89
(1985) 2355.
[17] O.S. Wolfbeis, A. Knierzinger, R. Schipfer, J. Photochem.
21 (1983) 67.
Acknowledgements
Financial support by FAPESP (Brazil) is grate-
fully acknowledged. ACG thanks CNPq for a
graduate fellowship.
[18] M. Sarkar, J. Guharay, P.K. Sengupta, Spectrochim.
Acta Part A 52 (1996) 275.
[19] J. Guharay, R. Chaudhuri, A. Chakrabarti, P.K. Sen-
gupta, Spectrochim. Acta Part A 53 (1997) 457.
[20] J. Guharay, P.K. Sengupta, Spectrochim. Acta Part A 53
(1997) 905.
References
[21] G.J. Smith, K.R. Markham, J. Photochem. Photobiol. A:
Chem. 118 (1998) 99.
[22] F.R. Hartley, C. Burgess, R.M. Alcock, Solution Equi-
libria, Ellis Horwood Limited, New York, 1980.
[23] M.H. Gehlen, F.C. De Schryver, Chem. Rev. 93 (1993)
199.
[24] O.S. Wolfbeis, M. Leiner, P. Hochmuth, Ber. Bunsenges.
Phys. Chem. 88 (1984) 759.
[1] K.R. Markham, Techniques of Flavonoids Identification,
Academic Press, London, 1982.
[2] G.G. Guilbault, Practical Fluorescence, Theory, Methods
and Techniques, Marcel Dekker, New York, 1973, p. 221.
[3] O.S. Wolfbeis, M. Begum, Z. Naturforsch. 39b (1983)
231.