Zirconium organophosphonates as photoactive and hydrophobic host materials for sensitized luminescence of Eu(III), Tb(III), Sm(III) and Dy(III)

Article abstract of DOI:10.1039/b410222c

Novel luminescent materials based on lanthanides [Eu(III), Tb(III), Sm(III), and Dy(III)] and a mixed zirconium phenyl- and m-sulfophenyl phosphonate as the host matrix, have been prepared and photophysically characterized. Powder X-ray diffraction, ³¹P-MAS-NMR spectroscopy, IR spectroscopy, and diffuse-reflectance UV spectroscopy revealed the specificity of these materials such as a layered structure, the presence of phenyl groups in the galleries, and the ability to absorb lanthanide ions introduced by simple ion exchange. Characteristic line-shaped long-lived luminescence [ca. 0.27 ms for EU(III), 0.80 ms for Tb(III), and 0.05 ms for Sm(III)] was observed for different lanthanide ions, and was demonstrated to be generated by an antenna-induced energy transfer process to the metal. The overall luminescence quantum yield of europium- or terbium-loaded materials was measured to be ca. 0.3%. A detailed analysis of the EU(III) luminescence spectrum and H ₂O/D₂O exchange experiments indicated the presence of ca. 3 water molecules around each lanthanide. Based on the high maximum coordination numbers of lanthanides, up to 9 for free EU(III) in aqueous solutions, and the presence of only monodentate sulfonate binding sites, more water could be expected. This observation is explained by the rather hydrophobic microenvironment in the interlamellar space of the materials, due to pendant organic moieties, that is, the phenyl groups.

Full text of DOI:10.1039/b410222c

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
Zirconium organophosphonates as photoactive and hydrophobic
host materials for sensitized luminescence of Eu(III), Tb(III),
Sm(III) and Dy(III)w
b
a
c
c
Rita Ferreira,ab Preciosa Pires, Baltazar de Castro, Rute A. Sa Ferreira, Luıs D. Carlos
´
´
and Uwe Pischel*a
a
´
ˆ
REQUIMTE, Departamento de Quımica, Faculdade de Ciencias, Universidade do Porto,
R. Campo Alegre, 4169-007 Porto, Portugal. E-mail: upischel@fc.up.pt;
Fax: +35 1 22 608 2959; Tel: +35 1 22 608 2885
b
c
´
ˆ
ESTG, Instituto Politecnico de Viana do Castelo, Av. do Atlantico, 4900-348 Viana do
Castelo, Portugal
CICECO, Departamento de Fısica, Universidade de Aveiro, 3810-193 Aveiro, Portugal
´
Received (in Montpellier, France) 5th July 2004, Accepted 8th September 2004
First published as an Advance Article on the web 15th November 2004
Novel luminescent materials based on lanthanides [Eu(III), Tb(III), Sm(III), and Dy(III)] and a mixed
zirconium phenyl- and m-sulfophenyl phosphonate as the host matrix, have been prepared and
photophysically characterized. Powder X-ray diffraction, 31P-MAS-NMR spectroscopy, IR spectroscopy, and
diffuse-reflectance UV spectroscopy revealed the specificity of these materials such as a layered structure, the
presence of phenyl groups in the galleries, and the ability to absorb lanthanide ions introduced by simple ion
exchange. Characteristic line-shaped long-lived luminescence [ca. 0.27 ms for Eu(III), 0.80 ms for Tb(III), and
0.05 ms for Sm(III)] was observed for different lanthanide ions, and was demonstrated to be generated by an
antenna-induced energy transfer process to the metal. The overall luminescence quantum yield of europium-
or terbium-loaded materials was measured to be ca. 0.3%. A detailed analysis of the Eu(III) luminescence
spectrum and H2O/D2O exchange experiments indicated the presence of ca. 3 water molecules around each
lanthanide. Based on the high maximum coordination numbers of lanthanides, up to 9 for free Eu(III) in
aqueous solutions, and the presence of only monodentate sulfonate binding sites, more water could be
expected. This observation is explained by the rather hydrophobic microenvironment in the interlamellar
space of the materials, due to pendant organic moieties, that is, the phenyl groups.
assays (e.g., for enzyme activity or DNA hybridization),2
Introduction
electroluminescent devices4 or in optical telecommunication.11
However, the application of lanthanide-based luminescence
suffers from two serious drawbacks: (1) the aforehand-men-
tioned poor light absorption properties and (2) the efficient
non-radiative deactivation of their excited states by OH oscil-
lators such as water.5 In order to avoid these obvious
disadvantages for the application of lanthanides, a strategy
has been developed that involves the so-called antenna effect
(Scheme 1).12 In this approach the lanthanide ion is linked
via complexation with a ligand, which includes an orga-
nic chromophore capable of absorbing light energy with a
higher efficiency than the metal itself. In a subsequent energy
transfer process this chromophore acts as a sensitizer of
the lanthanide excited state, which can subsequently deacti-
vate through luminescence. Due to the large spectral gap
between antenna excitation (often UV light) and emitted
photons (visible to near IR), lanthanide-antenna conjugates
have been often referred to as ‘‘molecular devices for wave-
length conversion’’.12 Furthermore, complexation with a li-
gand provides lanthanides with a certain degree of protection
from surrounding water, hence increasing their luminescence
quantum yields.
The luminescence of lanthanides (Ln) has been a focal point of
photochemistry and its applications in recent years.1–4 Because
electronic transitions of lanthanides involve f orbitals, the
absorption of light energy as well as its opposite pathway,
the emission of photons, are parity-forbidden.1 This causes
very small molar extinction coefficients, generally lower than
1 Mꢀ1 cmꢀ1. On the other hand, due to the forbidden nature of
the involved electronic transitions, excited state lifetimes are
exceptionally long and generally in the hundreds of microse-
conds to several milliseconds range.5 Emission spectra of
lanthanides are line-shaped and cover a spectral range from
the visible [Eu(III), Tb(III), Sm(III), Dy(III)] to the near-infrared
region [Yb(III), Nd(III), Er(III)].1 The almost unprecedented
colour purity of light-emitting lanthanides, for example red
for Eu(III) and green for Tb(III), finds its explanation in this
special spectral feature.
The favourable luminescence properties of lanthanides have
fostered their application in the development of chemosen-
sors,6–8 supramolecular devices such as logic gates,9,10 bio-
Numerous lanthanide-antenna conjugates operating via such
an absorption–energy-transfer–emission sequence (A–EnT–E)
have been generated, most of them in solution.5,7–10,13,14 In
order to allow technological processing of lanthanide-based
luminescence, it is desirable to implement this sensitization
w Electronic supplementary information (ESI) available: experimental
details for the synthesis of phenyl phosphonic and m-sulfophenyl
phosphonic acids; FTIR spectrum of ZrSPP; X-ray diffraction patterns
of Na@ZrSPP and Ln@ZrSPP (Ln = Eu, Tb, Sm, Dy). See http://
www.rsc.org/suppdata/nj/b4/b410222c/
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
1506
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3

View Article Online
Results and discussion
Preparation of a mixed zirconium m-sulfophenyl/phenyl
phosphonate (ZrSPP)
A host material capable of sensitizing Ln luminescence via
energy transfer should fulfil two conditions: (a) present an
antenna chromophore and (b) present a lanthanide binding
site, preferably in close proximity to the antenna. ZrSPP
addresses both requirements by incorporating phenyl groups
as antenna chromophores and sulfonic acid groups as potential
ion exchange sites (Scheme 2).
Scheme 1 Sensitization of lanthanide excited states by energy transfer
from an antenna.
To prepare a mixed zirconium m-sulfophenyl/phenyl phos-
phonate (ZrSPP), two phosphonate sources were synthesized:
phenyl phosphonic acid and m-sulfophenyl phosphonic acid.
Phenyl phosphonic acid was obtained from the oxidation of
phenyl phosphinic acid by concentrated nitric acid. m-Sulfo-
phenyl phosphonic acid was synthesized by reaction of phenyl
phosphonic acid with chlorosulfonic acid. The preparation of
ZrSPP was performed in aqueous solution by dissolving the
corresponding phosphonic acid derivatives in water and add-
ing dropwise a solution of zirconyl chloride. The material was
formed immediately as a colourless precipitate. As mentioned
beforehand, water is able to deactivate lanthanide lumines-
cence through energy dissipation via OH vibrations. Therefore,
we prepared a mixed m-sulfophenyl/phenyl phosphonate, to
keep a moderate number of sulfonic acid groups and to provide
a hydrophobic surrounding constituted by organic moieties,
the phenyl groups, which should prevent the strong attraction
of water into the galleries of the material (see below). This
strategy has been successfully applied to synthesize hydropho-
bically shielded lanthanide complexes, where extensively phe-
nyl-substituted imidodiphosphinates were used as ligands.5
The resulting material, ZrSPP, was first ion-exchanged with
sodium cations from a solution of NaOH + NaCl, and
subsequently with the respective Ln ions from aqueous solu-
tions of LnCl3.
mechanism into solid state materials. Therefore, in recent
years several efforts have been made to immobilize lantha-
nide-antenna systems in zeolites, sol-gel materials, and meso-
porous MCM-41.15–23 In the present contribution, we wish to
report the synthesis of a layered material based on zirconium
organophosphonates, capable of binding lanthanides and pro-
viding antenna functionality. Generally, zirconium phosphates
and phosphonates form layered structures and by using con-
veniently modified organo-substituted phosph(on)ates as start-
ing materials, it is possible to obtain inorganic-organic hybrid
structures, which contain organic moieties in the galleries of
the material.24–27 The material’s layer-to-layer distance de-
pends on the size of the organic group, making it a key
structural element.
The high versatility of this approach towards functionalized
zirconium organophosph(on)ates has enabled the development
of a variety of novel materials. In this context various photo-
active components were integrated, such as for instance Os(II)
and Ru(II) bipyridyl complexes,28 fullerene or viologens as
electron acceptors,29,30 and aromatic chromophores (naphtha-
lene, anthracene).31 These materials and others, in which
organic guests have been intercalated in zirconium phosphate
phases,32–35 have been proven to be useful to study electron
transfer, energy transfer, and excimer formation in a solid host
matrix.
The first studies using layered zirconium phosphates or
phosphonates as host materials for lanthanides and their
complexes were reported only recently. Lanthanide com-
plexes were intercalated either in a-zirconium phosphate
phases or in pillared g-zirconium phosphates.36,37 In the
latter case polyethylenoxy chains served as pillars and
were demonstrated to enhance the luminescence of europium
and terbium via coordinative interactions.37 In another
report it was shown that uranyl ions served as sensitizers for
europium luminescence, both being coordinated to carboxylate
groups in the galleries of a zirconium carboxyethyl phospho-
nate.38
In this study we introduce a new luminescent zirconium
phosphonate based on phenyl chromophores as antennae
and several lanthanides [Eu(III), Tb(III), Sm(III), Dy(III)] as
energy acceptors and luminescent entities. The emphasis
of this contribution will lie on the exploration of the photo-
physical properties like luminescence lifetimes, quantum
yields and influence factors like luminescence quenching by
water. Several methods of structural characterization (PXRD,
31P-MAS-NMR) and photophysical measurements (diffuse-
reflectance UV spectroscopy, luminescence spectra, lifetimes,
and luminescence quantum yields) allow us to present a
comprehensive picture of lanthanide luminescence and its
sensitization by an organic chromophore integrated into an
inorganic architecture. Furthermore, our results point out the
role of the hydrophobic effects exerted by organic residues in
the galleries, which protect the lanthanides to a certain extent
from water quenching.
Scheme 2 Schematic representation of the layered structure of ZrSPP.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3
1507

View Article Online
Materials characterization
Parent ZrSPP and the ion-exchanged materials Na@ZrSPP
and Ln@ZrSPP (with Ln = Eu, Tb, Sm, Dy) were character-
ized by different techniques. Elemental analysis of the parent
ZrSPP showed carbon and sulfur as indicators of the presence
of the expected organic groups in the material. From the
carbon-to-sulfur ratio it can be estimated that every fifth
phenyl group carries one sulfonic acid function, leading to
the general formula Zr(O3PC6H5)1.6(O3PC6H4SO3H)0.4
ꢁ
nH2O. By thermogravimetric analysis, ca. 13 wt % water was
determined from the weight loss up to 150 1C. This corre-
sponds to n ca. 5.4. The thermal stability of the phenyl groups
was investigated with the same technique, which indicated no
appreciable decomposition of the organic content up to 450 1C.
Only between 450 and 600 1C was notable weight loss observed
due to degradation of the organic moieties.39
The lanthanide-ion-exchanged samples were analyzed by
ICP-AAS or ICP-OES (which gave the same results within
the error limits), to establish their Ln content. All materials
were loaded to the maximum limit by using Ln salts in a large
excess. Therefore, the final Ln content was practically the same
for all materials (0.42 ꢂ 0.02 mmol gꢀ1 material). Only the
Sm(III)-loaded sample had a slightly lower content with only
0.29 mmol Sm(III) gꢀ1 material. All samples showed a residual
content of unexchanged sodium (cf. Experimental). By com-
paring with the results obtained for sulfur by elemental analy-
sis, that is, 0.53 mmol gꢀ1 material, it can be estimated that
every Ln ion corresponds to one sulfonate group. The other
two positive charges of the trivalent Ln ions must be counter-
balanced by other anions, most likely chlorides.
In order to further characterize the materials with regards to
the presence of phenyl groups, the IR spectrum of ZrSPP was
recorded (see Electronic supplementary information, ESI).
Two regions were identified as belonging to the phosphonate
PO3 groups and to the phenyl rings. In the region between 900
and 1200 cmꢀ1, PO3 stretching vibrations of the inorganic
framework were detected. A strong band at 1149 cmꢀ1 was
ascribed to P–C6H5, based on the reported assignments for
similar metal(IV) phosphonates.40 Moreover, between 530 and
830 cmꢀ1 the characteristic vibrations of phenyl rings could be
observed and were assigned to mono-substituted (originating
from phenyl phosphonate) and di-substituted (originating
from m-sulfophenyl phosphonate) aromatic rings.
Three samples were investigated by 31P-MAS-NMR spectro-
scopy as shown in Fig. 1. The most intense band of the sodium-
loaded sample Na@ZrSPP was found at ꢀ4.9 ppm and
corresponds to phosphonate groups bound to three Zr(IV)
cations via non-protonated oxygen atoms.39–41 The signal is
well-shifted from that of free phenyl phosphonic acid, where
the phosphorous resonance appears at +17.3 ppm; this situa-
tion is typically encountered in layered metal(IV) phospho-
nates.41 Two smaller bands at 2.6 and 8.1 ppm were also
detected and more likely originate from phosphorous only
partly coordinated to Zr(IV) via some of the PO3 oxygen atoms.
Similar signals have been previously observed in the case of the
related zirconium p-methylphenyl phosphonate.42 In that case,
cross-polarization experiments supported the assignment of
these bands to phosphorous bound to free OH groups, which
were not integrated into the zirconium phosphonate layers. By
analogy, we suggest here the same assignment for the two
minor signals observed. The Ln-exchanged samples
Eu@ZrSPP and Tb@ZrSPP showed an equivalent signal at
ca. ꢀ4.9 ppm with strong line broadening. At first glance, the
absence of significant shifts of the signal is surprising, but it
could be masked by the observed line broadening. When the
fwhm of the signal was ca. 3 ppm in Na@ZrSPP, it was ca.
5 ppm in the Eu-loaded sample and even greater in the Tb-
loaded ZrSPP with a fwhm at ca. 32 ppm. Furthermore, for the
Ln-loaded materials, spinning side bands were more intense
Fig. 1 31P MAS-NMR spectra of (a) Na@ZrSPP, (b) Eu@ZrSPP,
and (c) Tb@ZrSPP. * marks spinning side bands.
than in Na@ZrSPP. These effects are typical of paramagnetic
lanthanides1 and give firm evidence of their presence in the
interlamellar space of the material.
The structures of Na@ZrSPP and the Ln-loaded ZrSPP
were studied by powder X-ray diffraction (PXRD; see ESI)
No sharp diffraction peaks were observed, which indicates the
amorphous character of the zirconium organophosphonates.
The layer-to-layer distances in the materials were estimated at
15–16 A from the first reflection peaks. These estimated dis-
tances are in agreement with those measured in similar zirco-
nium organophosphonates43 and indicate the presence of
space-filling phenyl groups in the galleries of the materials.
For comparison, the layer-to-layer distance of zirconium
phosphonate without organic groups is reported to be 5.6 A,
that is, lower than in the case of ZrSPP.25 The layer-to-layer
distance of Na@ZrSPP remains, in a first approximation,
unchanged upon loading the material with different Ln ions.
The diffuse reflectance UV spectrum of ZrSPP (see Fig. 2,
solid line) revealed two absorption bands at 221 and 265 nm,
which can be assigned to p,p* transitions in the phenyl
groups.44 These absorption bands are absent in phenyl-free
a-zirconium phosphate (a-ZrP) , which, instead, shows a much
weaker band at 242 nm (see Fig. 2, dotted line). The latter is
probably due to electronic defects in the structure as reported
for phosphate glasses.45
Europium(III) and terbium(III) luminescence
Excitation of the phenyl groups at lexc = 275 nm gave rise to
the typical luminescence spectra of lanthanides Eu(III) and
Tb(III), Fig. 3. In the case of Eu@ZrSPP, bands at 594, 617
and 655 nm were observed in the emission spectrum. These
bands have been previously assigned to 5D0 - 7FJ transitions,
with J = 1, 2 and 3, respectively (cf. Experimental).1 The
5
7
strong D0 - F2 transition at 617 nm is responsible for the
red colour of the luminescence. With Tb@ZrSPP, typical
1508
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3

View Article Online
piled in Table 1. Both samples have practically the same overall
luminescence quantum yield: Ftot C 0.3%. This value seems to
be quite low, but it must be kept in mind that the materials
were prepared from aqueous solutions and therefore contain
appreciable amounts of water, which can quench lanthanide
luminescence FLn to a large extent.1,5 However, other factors,
also contributing to the overall luminescence quantum yield,
are the efficiency of antenna intersystem crossing (ZISC) and the
efficiency of antenna-to-lanthanide energy transfer (ZEnT),
which is distance dependent.47,48 Eqn. (1) defines the overall
luminescence quantum yield (Ftot) of the system as a function
of these factors (with Fabs = quantum yield of light absorp-
tion):
Ftot = Fabs ꢃ ZISC ꢃ ZEnT ꢃ FLn
.
(1)
Furthermore, the intrinsic lanthanide luminescence quantum
yield FLn is defined in eqn. (2) by the competition between non-
radiative processes (knr), that is, mainly water quenching, and
radiative processes (kr):
Fig. 2 Diffuse reflectance UV spectra of solid parent ZrSPP (solid
line) and a-ZrP (dotted line).
kr
FLn
¼
ð2Þ
kr þ knr
In the case of europium luminescence the value of kr can be
estimated by spectral analysis with the help of eqn. (3):18,49
emission lines at 490, 547, 590 and 624 nm were detected. These
5
7
emission lines can be assigned to D4 - FJ transitions, with
J = 6, 5, 4 and 3, respectively. The strong band at 547 nm
(5D4
-
7F5) is the origin of the green light emission.1 All
4
X
A0ꢀ1ꢀho0ꢀ1
S0ꢀJ
spectra are characterized by very narrow bands (fwhm ca. 10
nm), which are typical for lanthanide luminescence.
kr ¼
ð3Þ
S0ꢀ1
J¼0 ꢀho0ꢀJ
In order to prove that the detected luminescence is caused by
energy transfer from excited phenyl antennae, excitation spec-
tra were recorded (cf. inset Fig. 3), monitoring the emission
wavelength at 617 nm for Eu(III) and 547 nm for Tb(III). The
characteristic phenyl absorption band at ca. 265 nm was clearly
identified. This demonstrates that the phenyl groups are able to
act as antennae and constitutes an interesting example of
energy transfer sensitization of Ln luminescence by an organic
chromophore, which is an integral and structure-forming part
of the inorganic-organic hybrid material. The excited triplet
state of the antenna is generally the one involved in the energy
transfer process.46 The excitation energy of this state is not
known for the phenyl groups in the material, but benzene is
certainly a reliable model to use. The excited triplet state
7
where J represents the final F0–6 levels, S is the integrated
intensity of the particular emission lines and io stands for the
corresponding transition energies. A0–1 is the Einstein coeffi-
cient of spontaneous emission between the 5D0 and the 7F1
5
7
Stark levels. The branching ratios for the D0 - F5,6 transi-
tions must be neglected as they are too weak to be observed
experimentally (cf. Experimental). Therefore, their influence
can be ignored in the depopulation of the 5D0 excited state. The
5D0 - 7F1 transition does not depend on the local ligand field
seen by Eu3+ ions and, thus, may be used as a reference for
the whole spectrum, in vacuo A0 – 1 = 14.65 sꢀ1 50
age refractive index equal to 1.5 was considered, leading to
.
An aver-
A0 ꢀ 1 C 50 sꢀ1 51
Appropriate analysis of the europium emission of
Eu@ZrSPP according to eqn. (3) yielded a kr value of 0.30
.
energy of benzene is 29500 cmꢀ1 44
well above the energy of
,
the luminescent 5D0 level of europium (17300 cmꢀ1) or the 5D4
level of terbium (20500 cmꢀ1). In other words, antenna-to-
lanthanide energy transfer is thermodynamically possible, as
confirmed by our experimental observations. The emission
characteristics of these samples were quantified and are com-
msꢀ1. With the experimental lifetime of the D0 state (texp
=
=
5
0.27 ms, cf. Table 1) the value of knr can be calculated (1/texp
kr + knr) to be 3.40 msꢀ1. Finally, with these rate constants,
eqn. (2) gives an intrinsic europium luminescence quantum
yield of FLn C 8.1%. This value compares very well with those
obtained for other europium-doped hybrid materials, for ex-
ample, FLn C 13% for ureasil-europium nanocomposites or
FLn C 13.5% for europium complexes of 3-hydroxypicolinic
acid.18,49 Instead of calculating kr from the luminescence
Table 1 Photophysical parameters of Eu@ZrSPP and Tb@ZrSPP
Eu@ZrSPP
Tb@ZrSPP
t
t
exp/msa
0.27
1.51
0.30c
3.40
3.9d
0.3
0.80
2.11
0.47
0.78
3.5d
0.3
D O/msb
2
kr/msꢀ1
knr/msꢀ1
nH O
2
Ftot
FLn
ZEnT
e
0.081
0.15
0.376
0.03
e
a
Lifetime of the solid material [lifetime as an H2O suspension: 0.23 ms
b
for Eu(III) and 0.72 ms for Tb(III)]. Measured as a suspension in
Fig. 3 Normalized emission spectra of solid Eu@ZrSPP (solid line)
and Tb@ZrSPP (dotted line) at lexc = 275 nm. Inset: excitation
spectrum of Tb@ZrSPP recorded at lem= 547 nm. The small peaks
at 460 and 540 nm in the emission spectrum of Tb@ZrSPP are
probably due to impurities.
c
D2O. Calculated with eqn. (3); value derived from the lifetime in
d
D2O: kr = 0.66 msꢀ1
ref. 55. See text for other values in the case of Eu(III). See text.
.
Calculated with eqn. (4) and parameters from
e
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3
1509

View Article Online
spectra, the reciprocal lifetime of D2O-exchanged samples can
be used [kr = 0.66 and 0.47 msꢀ1 for Eu(III) and Tb(III),
respectively; see below]. According to these values and the
reciprocal experimental lifetimes (1/texp), somewhat larger
lanthanide luminescence quantum yields are estimated for both
Eu@ZrSPP (FLn C 17.8%), and Tb@ZrSPP (FLn C 37.6%).
Now, eqn. (1) allows the calculation of the efficiency of
antenna-to-lanthanide energy transfer processes in Eu@ZrSPP
and Tb@ZrSPP. For this purpose the light absorption quan-
tum yield (Fabs) is assumed to be unity and the intersystem
crossing efficiency is taken from benzene as ZISC = 25%.52
Based on these assumptions, limiting values of ZEnT C 15% for
Eu(III) and 3% for Tb(III) for the energy transfer efficiency can
be estimated. It has to be emphasized that lanthanides can
exert a heavy atom effect, thereby increasing the intersystem
crossing efficiency of the antenna in our materials, resulting in
an even lower value for energy transfer efficiency. It is most
likely that lanthanides are only close to the phenyl rings that
present a binding sulfonate function (ca. 20% of all phenyl
rings; see above), but in the majority, the antennae are much
farther from the metals. This would naturally diminish the
efficiency of energy transfer, as a function of the donor-
acceptor distance,48 accounting for the rather low values.
The difference between Eu(III) and Tb(III) might be related to
different spectral overlap, which is directly proportional to the
energy transfer efficiency.48
From the above data (particularly the FLn and ZEnT values),
it can be concluded that the presence of water in our materials
contributes only to a certain extent to the diminution of the
overall luminescence quantum yield (Ftot) and that an energy
transfer of medium efficiency must be deemed responsible as
well. However, although small in absolute value, the overall
luminescence quantum yield of Eu@ZrSPP is comparable to
other europium complexes with classical ligand systems, espe-
cially when ligand-to-metal charge transfer (LMCT) processes
are possible.5 For instance, in europium cryptate [Eu
(III)C2.2.1], a total luminescence quantum yield comprised
between 0.1 and 0.3% was measured.53 In this context, it must
be mentioned that terbium is not involved in LMCT, since its
reduction potential does not favour charge transfer processes
in general. In principal, low-lying LMCT states could also
contribute to a diminution of the overall luminescence quan-
tum yield of Eu@ZrSPP, thus explaining the comparably
higher value of FLn in the case of Tb@ZrSPP (cf. Table 1).
Strikingly, the bare fact that luminescence is observed in our
materials is remarkable, since their preparation has been done
in aqueous solution and, furthermore, coordination of the
lanthanide ions to sulfonate groups is not expected to offer
sufficient protection from quenching by water molecules. Be-
side the estimation of the intrinsic luminescence quantum yield,
excited state lifetime measurements can be used to obtain
information on the presence of water in the coordination
sphere of lanthanide ions. For both Eu@ZrSPP and
Tb@ZrSPP materials, the decays, measured for the most
intense luminescence band (cf. Fig. 4), could be fitted by
mono-exponential functions. The lifetimes texp were measured
to be 0.27 ms for Eu@ZrSPP and 0.80 ms for Tb@ZrSPP.
These values are similar to those reported for several lantha-
nide complexes immobilized in sol-gel materials, zeolites,15,19
and lanthanide cryptates such as, for example, [Eu(III)C2.2.1],
which has a lifetime of 0.31 ms in the solid state and 0.22 ms in
water.53 However, stronger protecting ligand systems lead to
longer lifetimes, this accounting for the expected presence of
coordinated water in our materials.5 In order to estimate the
number of water molecules coordinated to each lanthanide ion,
we performed H2O/D2O exchange experiments. D2O ex-
changes water and acts as a much weaker quencher, owing to
the lower energy of the OD overtone vibrations compared to
OH oscillators. As expected, lifetimes are much longer in D2O
suspensions, with values approaching up to 2.11 ms for Tb
Fig. 4 Normalized luminescence decays of solid Eu@ZrSPP (full
circles) and Tb@ZrSPP (open circles) and their respective mono-
exponential fittings.
luminescence and 1.51 ms for Eu luminescence in Tb@ZrSPP
and Eu@ZrSPP. For comparison, in [Eu(III)C2.2.1] a much
shorter lifetime of 0.64 ms is obtained in D2O,53 pointing out
the reasonable protection of lanthanides in our materials,
albeit only a monodentate coordination with sulfonate groups
is possible. The application of the corrected Horrocks equa-
tion,54,55 eqn. (4):
nH O = C ꢃ (1/texp ꢀ kr + Dkcorr).
(4)
2
gives n ca. 3.5 ꢂ 0.5 water molecules coordinated to each
lanthanide ion in both materials. It is important to note that
this number corresponds to an average of coordinated water
molecules per lanthanide. It is expected that different lantha-
nide coordination modes may give rise to varying water access
to the individual metal centres. In eqn. (4), C denotes an
empirical constant [1.25 for Eu(III) and 4.90 for Tb(III)],55 texp
is the experimental lifetime of Ln luminescence, Dkcorr is a
correction term accounting for water quenching in the Ln
second coordination sphere [ꢀ0.25 for Eu(III) and ꢀ0.06 for
Tb(III)]55 and kr is the radiative rate constant for the decay of
the Ln excited state. kr can be obtained by two ways: (a) from
the analysis of the luminescence spectrum in the case of
Eu@ZrSPP [cf. eqn. (3)] or (b) from a measurement of the
lifetime in D2O, according to the relation kr = 1/tD O. Method
2
(b) has been often applied to lanthanide complexes in solution,
but is very seldom used for solid state materials. The good
agreement in the number of water molecules calculated by both
methods for Eu@ZrSPP, that is, 3.9 ꢂ 0.5 from the spectral
analysis and 3.5 ꢂ 0.5 from the H2O/D2O exchange, supports
the applicability of the latter technique to solid state materials.
Recently, Horrocks et al. reported another refined equation
for the calculation of Eu-coordinated water [C = 1.11 and
Dkcorr = ꢀ0.31 in eqn. (4)].56 This equation was shown to yield
results with smaller errors (ꢂ0.1 water molecules). Application
of this relationship to Eu-loaded Eu@ZrSPP leads to 3.0 ꢂ 0.1
and 3.4 ꢂ 0.1 water molecules, with kr obtained from D2O
exchange or spectral analysis, respectively. Based on the com-
bined results we anticipate the number of water molecules for
both Eu@ZrSPP and Tb@ZrSPP to be ca. 3.
This number is noteworthy, since the coordination of Eu(III)
and Tb(III) by a monodentate sulfonate and two other counter-
ions, possibly chlorides, does not offer a high degree of
protection. The values should be compared to the europium
cryptate [Eu(III)C2.2.1], where 3.2 water molecules are coordi-
nated (corrected to 3.1 using Horrocks’ refined equation56 and
data from ref. 53), although the ligand offers seven donor
binding sites.53 Furthermore, it is well-known that free Ln ions
1510
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3

View Article Online
in aqueous solutions can coordinate up to 9 water molecules,
for example, 9.2 water molecules in Eu(III)aq in 0.5 M Na-
ClO4.53 Therefore, this comparably low number of ca. 3 water
molecules determined for our materials is quite surprising. A
reasonable explanation is the limited water access in the
galleries of the material due to the presence of hydrophobic
phenyl rings. Even in a water suspension, the lifetimes are
virtually the same (ca. 10% quenching) as in the solid state (cf.
footnotes in Table 1). Hence, the lanthanide is protected by a
rather hydrophobic microenvironment, similar to the reported
cases of hydrophobic shielding by phenyl-substituted imidodi-
phosphinate ligands.5 In this regard, studies of enzyme uptake
by layered zirconium phenylphosphonate have provided ample
evidence of the hydrophobic nature of the interlamellar
space.57 Vice versa, lanthanide luminescence can be also con-
sidered as a probe signal of the interlamellar space hydro-
phobicity.
The higher sensitivity of samarium- and dysprosium-derived
luminescence towards water, in comparison to europium or
terbium luminescence, can be explained by the energy gap
between the excited luminescent states and the highest J levels
of the ground states of these lanthanides. This energy gap is C
7500 cmꢀ1 for Sm(III) and 7800 cmꢀ1 for Dy(III), that is,
considerably smaller than for Eu(III) and Tb(III) with values
around 12300 cmꢀ1 and 14800 cmꢀ1, respectively.5 Since OH
oscillators act independently (Evib = 3657 cmꢀ1; symmetric
OH stretching mode),58 it can be estimated that, in the case of
samarium and dysprosium, fewer water molecules are needed
to bridge the gap than in the case of europium and terbium (ca.
2 versus 3–4). This higher water sensitivity explains why life-
times of Dy(III) or Sm(III) luminescence are generally much
shorter than those observed with Eu(III) or Tb(III), that is, tens
of microseconds versus hundreds of microseconds to millise-
conds. We tried to measure the lifetime of the Sm(III)-loaded
ZrSPP with our setup, but it was apparently shorter than the
pulse duration of the excitation source (o0.01 ms). In D2O
suspension, a longer lifetime of 0.052 ms was obtained, which is
in the same range as for samarium complexes with imidodipho-
sphinate ligands in dry acetonitrile (0.15 ms)5 or with a m-ter-
phenyl derived tricarboxylate ligand measured in deuterated
methanol (0.090 ms).13
Samarium(III) and dysprosium(III) luminescence
In order to demonstrate the capability of antenna-modified
ZrSPP as sensitizers for lanthanides other than the commonly
used europium and terbium, we extended our work to Sm(III)
and Dy(III). Upon excitation at lexc = 275 nm, typical emission
spectra were detected of D2O suspensions of Sm@ZrSPP and
Dy@ZrSPP (Fig. 5). In the case of Sm@ZrSPP, this corre-
sponds to emission lines at 566, 602 and 647 nm, which are
assigned to 4G5/2 - 6HJ transitions, with J = 5/2, 7/2 and 9/2 ,
Conclusions
A zirconium organophosphonate was used as host material for
lanthanide absorption by simple ion exchange and allowed us
to prepare novel luminescent solid state materials. The re-
quired antennae used to sensitize the long-lived Ln luminescent
state are phenyl groups intimately implicated in the inorganic-
organic hybrid architecture. It has been demonstrated that
these aromatic rings are able to sensitize the excited lumines-
cent states of a variety of lanthanide ions [Eu(III), Tb(III),
Sm(III), and Dy(III)].
The Ln luminescence has been characterized by its overall
quantum yield [Ftot ca. 0.3% for Eu(III) and Tb(III)] and
lifetime [0.05 ms for Sm(III) suspended in D2O, 0.27 and
0.80 ms for Eu(III) and Tb(III) in the solid state, respectively].
Since the materials were prepared from water solutions, the
observation of long-lived lanthanide luminescence is striking.
A detailed analysis of the intrinsic lanthanide luminescence
quantum yield FLn (ca. 8–38%) and of the presence of co-
ordinated water (ca. 3 water molecules per Ln) in the case of
Eu@ZrSPP and Tb@ZrSPP points to a hydrophobic effect
caused by the presence of organic groups in the galleries of the
layered zirconium organophosphonate. Although the short
lifetimes of Sm@ZrSPP and Dy@ZrSPPs luminescence did
not allow us to measure them in the solid state, similar
conclusions can be drawn based on analogy.
4
6
respectively.5 The strongest peak is the G5/2 - H7/2 transi-
tion at 602 nm. The emission spectrum of Dy@ZrSPP is
composed of two intense lines at 484 and 577 nm, correspond-
4
ing to F9/2
-
6HJ transitions with J = 15/2 and 13/2,
respectively. The most intense emission line is at 484 nm.5
Generally, the luminescence spectra of Sm(III)- and Dy(III)-
loaded solids prepared from aqueous solutions were quite weak
and were significantly enhanced by recording them as D2O
suspensions. Quantum yields are expected to be very small
({0.3% measured for europium- and terbium-loaded materi-
als) and their measurements were therefore omitted. Occur-
rence of antenna-to-lanthanide energy transfer processes was
supported by the match between excitation spectra and phenyl
UV absorptions. As noted before for the energy transfer to
Eu(III) and Tb(III), the excited state energy of the triplet
antenna (29500 cmꢀ1) also lies well above the energy of the
luminescent 4G5/2 state of Sm(III) and the 4F9/2 state of Dy(III),
which are equal to 17900 and 21100 cmꢀ1, respectively.
The photophysical data obtained for Eu(III) and Tb(III) in
these materials compare very well with Ln ions complexed by
cryptands, which are intrinsically expected to offer a higher
protection of the Ln by providing more complexation sites.
Experimental
Chemicals
All chemicals used in the synthesis of the two organopho-
sphonic acids and the preparation of the parent material
(ZrSPP) were commercially available from Aldrich and Riedel
de Haen, and were used as received without further purifica-
tion. Lanthanide trichlorides in the hexahydrate form [Eu(III),
Tb(III)] or as anhydrous salts [Dy(III), Sm(III)] were purchased
from Aldrich (499.9% purity). Sodium hydroxide and sodium
chloride were purchased from Merck and used as received.
Deuterium oxide (499.5% D) was purchased from Aldrich.
Phenyl phosphonic acid, commercially available from Aldrich,
was synthesized in this work from phenyl phosphinic acid by
Fig. 5 Normalized emission spectra of Sm@ZrSPP (solid line) and
Dy@ZrSPP (dotted line), recorded at lexc = 275 nm on D2O suspen-
sions.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3
1511

View Article Online
oxidation with nitric acid (see ESI). m-Sulfophenyl phosphonic
acid was prepared by sulfonation of phenyl phosphonic acid
with chlorosulfonic acid in a standard procedure (see ESI).
This compound has already been described in the literature.59
Photophysical measurements
Room temperature emission and excitation photoluminescence
spectra and lifetime measurements were performed on Perkin
Elmer spectrometers (LS50 or LS50B), equipped with a flashed
xenon lamp (pulse width ca. 0.01 ms) and a 9781B model
photomultiplier from Electron Tubes Ltd., UK, working in the
Material characterization
range 165–680 nm. Therefore, the 5D0
-
7F4 transition of
Elemental analysis of the precursors and Na@ZrSPP was
performed by standard combustion techniques. Lanthanide
and sodium contents of the materials were determined by
induced coupled plasma–optical emission spectroscopy (ICP–
OES, Kingston Analytical Services, UK), or atomic absorption
europium at ca. 700 nm was not detected. For this reason,
quantum yields and a highly resolved luminescence spectrum
of europium were measured with a modular double grating
excitation spectrofluorimeter with a TRIAX 320 emission
monochromator (Fluorolog-3, Jobin Yvon-Spex) coupled to
a R928 Hamamatsu photomultiplier. However, even with this
spectroscopy (ICP–AAS, Laboratorio Central de Analises,
´ ´
fairly sensitive setup, the 5D0
-
7F5,6 transitions were not
Universidade de Aveiro, Portugal). Two different methods
were used to dissolve the samples leading, within error limits
(10%), to the same results: (a) fusion with lithium metaborate
followed by treatment with hydrofluoric acid and (b) digestion
with H2SO4. Thermogravimetric measurements (TG) were
detected. The spectra acquisition was done in front-face mode.
Quantum yield measurements were performed according to the
method by Wrighton et al.60 using magnesium oxide as the
reflecting standard. Spectral and lifetime measurements were
performed with solid samples or in suspensions in D2O or H2O,
as indicated in captions for figures and Table 1.
performed in the temperature range of 25–650 1C (5 1C minꢀ1
)
with a thermoanalyzer Netzsch STA 409 EP. FTIR spectra
were recorded on KBr pellets on a Jasco FTIR-460 spectro-
meter. Diffuse-reflectance UV/Vis (DR-UV/Vis) spectra were
recorded on a Shimadzu UV/3101PC spectrophotometer with
BaSO4 reflecting standard. XRD measurements of the solid
samples were performed on a Philips X’pert diffractometer,
using CuKa radiation (l = 1.540560 nm). 31P magic-angle
spinning (MAS) NMR spectra were recorded at 161.9 MHz, on
an Avance 400 spectrometer, using 2 ms (451) radio-frequency
pulses, 60 s recycle delays and 14 kHz spinning rates. Chemical
shifts are quoted in ppm and referred to 85% H3PO4.
Acknowledgements
We thank Prof. J. Rocha (Universidade de Aveiro, Portugal)
for recording the 31P-MAS-NMR spectra and Dr A. Labrincha
(Instituto Politecnico de Viana do Castelo, Portugal) for the
´
thermogravimetric measurement.
References
1
J.-C. G. Bunzli and G. R. Choppin, Lanthanide Probes in Life,
¨
Chemical and Earth Sciences – Theory and Practice, Elsevier,
Syntheses
Amsterdam, 1st edn., 1989.
H. Bazin, M. Preaudat, E. Trinquet and G. Mathis, Spectrochim.
´
Acta, Part A, 2001, 57, 2197–2211.
D. Parker, R. S. Dickins, H. Puschmann, C. Crossland and J. A.
K. Howard, Chem. Rev., 2002, 102, 1977–2010.
J. Kido and Y. Okamoto, Chem. Rev., 2002, 102, 2357–2368.
S. W. Magennis, S. Parsons and Z. Pikramenou, Chem.-Eur. J.,
2002, 8, 5761–5771.
A. P. de Silva, H. Q. Nimal Gunaratne, T. E. Rice and S. Stewart,
Chem. Commun., 1997, 1891–1892.
D. Parker, P. K. Senanayake and J. A. G. Williams, J. Chem. Soc.,
Perkin Trans. 2, 1998, 2129–2139.
Preparation of mixed zirconium m-sulfophenyl/phenyl phos-
phonate (ZrSPP). At room temperature, a solution of 9.03 g
(27.5 mmol) of zirconyl chloride octahydrate in 60 ml of water
was slowly added under stirring to a solution of 6.3 g (40.0
mmol) of phenyl phosphonic acid and 5.9 g (24.8 mmol) of
m-sulfophenyl phosphonic acid in 60 ml of water. Immediately,
a colourless precipitate was formed and, after centrifugation,
the resulting gel was isolated and dried at 80 1C under vacuum
for 72 h to yield 14.6 g of amorphous mixed zirconium m-
sulfophenyl/phenyl phosphonate (ZrSPP). Elemental analysis:
C 19.9%, S 1.7%. IR (KBr) in cmꢀ1: 1149 (P–C6H5), 1076
(PO3), 1025 (PO3), 804, 752, 728, 615, 560 (phenyl).
2
3
4
5
6
7
8
L. J. Charbonniere, R. Ziessel, M. Montalti, L. Prodi, N. Zac-
cheroni, C. Boehme and G. Wipff, J. Am. Chem. Soc., 2002, 124,
7779–7788.
9
´
T. Gunnlaugsson, D. A. Mac Donaill and D. Parker, J. Am.
Chem. Soc., 2001, 123, 12 866–12 876.
10 M. de Sousa, M. Kluciar, S. Abad, M. A. Miranda, B. de Castro
and U. Pischel, Photochem. Photobiol. Sci., 2004, 3, 639–642.
11 K. Kuriki, Y. Koike and Y. Okamoto, Chem. Rev., 2002, 102,
2347–2356.
12 J.-M. Lehn, Supramolecular Chemistry – Concepts and Perspec-
tives, VCH, Weinheim, 1st edn., 1995.
13 M. P. O. Wolbers, F. C. J. M. van Veggel, B. H. M. Snellink-Ruel,
J. W. Hofstraat, F. A. J. Geurts and D. N. Reinhoudt, J. Chem.
Soc., Perkin Trans. 2, 1998, 2141–2150.
Sodium ion exchange (Na@ZrSPP). ZrSPP (3 g) was added
to a solution of 60 mmol of sodium chloride and 21.6 mmol of
sodium hydroxide in 80 ml of water. After stirring the mixture
for 3 h, the solid was filtered off and washed with large volumes
of water until the filtrate presented a neutral pH. The solid was
finally dried at 80 1C.
Lanthanide ion exchange (Ln@ZrSPP). In
a typical
14 S. Petoud, S. M. Cohen, J.-C. G. Bunzli and K. N. Raymond,
¨
J. Am. Chem. Soc., 2003, 125, 13 324–13 325.
procedure, 0.7 mmol of lanthanide trichloride (EuCl3 ꢁ 6H2O,
TbCl3 ꢁ 6H2O, SmCl3 or DyCl3) in 20 ml of water was added to
500 mg of Na@ZrSPP. The slurry was stirred overnight and
thereafter the solid was filtered off, washed with large amounts
of water and dried at 80 1C.
Lanthanide and sodium contents per gram of material
(measurement technique in parenthesis): Eu@ZrSPP (ICP–
OES): 0.43 mmol Eu, 0.14 mmol Na; Tb@ZrSPP
(ICP–OES): 0.41 mmol Tb, 0.13 mmol Na; Tb@ZrSPP
(ICP–AAS): 0.44 mmol Tb; Sm@ZrSPP (ICP–AAS): 0.29
mmol Sm; 0.01 mmol Na; Dy@ZrSPP (ICP–AAS): 0.41 mmol
Dy, 0.04 mmol Na.
31P MAS-NMR (85% H3PO4): d for Na@ZrSPP: 8.1, 2.6,
ꢀ4.9; Eu@ZrSPP: ꢀ4.9; Dy@ZrSPP: ꢀ4.8.
´
15 M. Alvaro, V. Forne
´
s, S. Garcı
´
´
a, H. Garcıa and J. C. Scaiano,
J. Phys. Chem. B, 1998, 102, 8744–8750.
16 V. Bekiari, G. Pistolis and P. Lianos, Chem. Mater., 1999, 11,
3189–3195.
17 Y. Wada, T. Okubo, M. Ryo, T. Nakazawa, Y. Hasegawa and
S. Yanagida, J. Am. Chem. Soc., 2000, 122, 8583–8584.
´
18 R. A. Sa Ferreira, L. D. Carlos, R. R. Gonc¸ alves, S. J. L. Ribeiro
and V. de Zea Bermudez, Chem. Mater., 2001, 13, 2991–2998.
19 M. H. Bartl, B. J. Scott, H. C. Huang, G. Wirnsberger, A.
Popitsch, B. F. Chmelka and G. D. Stucky, Chem. Commun.,
2002, 2474–2475.
´
20 R. J. P. Corriu, F. Embert, Y. Guari, C. Reye and R. Guilard,
Chem.-Eur. J., 2002, 8, 5732–5741.
21 H. R. Li, J. Lin, L. S. Fu, J. F. Guo, Q. G. Meng, F. Y. Liu and
H. J. Zhang, Microporous Mesoporous Mater., 2002, 55, 103–107.
1512
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3

View Article Online
22 P. N. Minoofar, R. Hernandez, S. Chia, B. Dunn, J. I. Zink and
A.-C. Franville, J. Am. Chem. Soc., 2002, 124, 14 388–14 396.
23 S. Hashimoto, S. Kirikae and S. Tobita, Phys. Chem. Chem. Phys.,
2002, 4, 5856–5862.
44 H. G. O. Becker, Einfuhrung in die Photochemie, Deutscher Verlag
¨
der Wissenschaften, Berlin, 3rd edn., 1991.
45 D. Ehrt, P. Ebeling and U. Natura, J. Non-Cryst. Solids, 2000
263–264, 240–250.
24 G. Alberti, U. Costantino, S. Allulli and N. Tomassini, J. Inorg.
Nucl. Chem., 1978, 40, 1113–1117.
25 M. B. Dines and P. M. DiGiacomo, Inorg. Chem., 1981, 20, 92–97.
26 A. Clearfield and Z. K. Wang, J. Chem. Soc., Dalton Trans., 2002,
2937–2947.
46 Recently, a rare example has been reported, where the excited
singlet states of a dansyl chromophore and a lissamine dye
chromophore have been proven to sensitize the excited state of
neodymium(III). See: G.A. Hebbink, S. I. Klink, L. Grave, P. G. B.
Oude Alink and F. C. J. M. van Veggel, Chem. Phys. Chem., 2002,
3, 1014–1018.
47 Energy transfer from the excited triplet antenna to the lanthanide
is supposed to be proceeding via a Dexter mechanism, which
shows an eꢀr dependence on the distance r between energy donor
27 E. Brunet, M. J. de la Mata, O. Juanes and J. C. Rodrı
Angew. Chem., Int. Ed., 2004, 43, 619–621.
´
guez-Ubis,
28 F. Odobel, D. Massiot, B. S. Harrison and K. S. Schanze,
Langmuir, 2003, 19, 30–39.
29 E. Brunet, M. Alonso, M. J. de la Mata, S. Ferna
O. Chavanes and J.-C. Rodrıguez-Ubis, Chem. Mater., 2003, 15,
1232–1234.
´
ndez, O. Juanes,
and acceptor. Often, the involvement of Forster resonance energy
¨
´
transfer has been discussed, but it seems not very likely based
on the small molar extinction coefficients of lanthanides and
the resulting small spectral overlap integral of the Forster
¨
30 L. A. Vermeulen, J. L. Snover, L. S. Sapochak and M. E.
Thompson, J. Am. Chem. Soc., 1993, 115, 11 767–11 774.
31 J. C. Amicangelo and W. R. Leenstra, J. Am. Chem. Soc., 2003,
125, 14 698–14 699.
mechanism.
48 S. Speiser, Chem. Rev., 1996, 96, 1953–1976.
49 P. C. R. Soares-Santos, H. I. S. Nogueira, V. Fe
´
lix, M. G. B.
32 D. P. Vliers, R. A. Schoonheydt and F. C. de Schryver, J. Chem.
Soc., Faraday Trans. 1, 1985, 81, 2009–2019.
33 C. V. Kumar and A. Chaudhari, J. Am. Chem. Soc., 1994, 116,
403–404.
Drew, R. A. Sa Ferreira, L. D. Carlos and T. Trindade, Chem.
´
Mater., 2003, 15, 100–108.
50 M. H. V. Werts, R. T. F. Jukes and J. W. Verhoeven, Phys. Chem.
Chem. Phys., 2002, 4, 1542–1548.
34 C. V. Kumar, Z. J. Williams and R. S. Turner, J. Phys. Chem. A,
1998, 102, 5562–5568.
51 M. F. Hazenkamp and G. Blasse, Chem. Mater., 1990, 2,
105–110.
35 D. M. Kaschak, J. T. Lean, C. C. Waraksa, G. B. Saupe,
H. Usami and T. E. Mallouk, J. Am. Chem. Soc., 1999, 121,
3435–3445.
36 Q. Xu, L. Fu, L. Li, H. Zhang and R. Xu, J. Mater. Chem., 2000,
10, 2532–2536.
37 E. Brunet, M. J. de la Mata, O. Juanes and J. C. Rodrıguez-Ubis,
´
52 S. L. Murov, I. Carmichael and G. L. Hug, Handbook of Photo-
chemistry, Marcel Dekker, Inc., New York, 2nd edn., 1993.
53 N. Sabbatini, S. Dellonte, M. Ciano, A. Bonazzi and V. Balzani,
Chem. Phys. Lett., 1984, 107, 212–216.
54 W. D. Horrocks and D. R. Sudnick, Acc. Chem. Res., 1981, 14,
384–392.
Chem. Mater., 2004, 16, 1517–1522.
38 C. V. Kumar and A. Chaudhary, Microporous Mesoporous Ma-
ter., 1999, 32, 75–79.
39 G. Alberti, U. Costantino, M. Casciola, S. Ferroni, L. Massinelli
and P. Staiti, Solid State Ionics, 2001, 145, 249–255.
40 G. Guerrero, P. H. Mutin and A. Vioux, Chem. Mater., 2001, 13,
4367–4373.
41 A. Clearfield, D. M. Poojary, B. L. Zhang, B. Y. Zhao and
A. Derecskei-Kovacs, Chem. Mater., 2000, 12, 2745–2752.
42 P. Pires, Ph.D. Thesis, University of Liverpool, Liverpool, UK, 1995.
43 J. L. Colon, C. Y. Yang, A. Clearfield and C. R. Martin, J. Phys.
Chem., 1990, 94, 874–882.
55 R. S. Dickins, T. Gunnlaugsson, D. Parker and R. D. Peacock,
Chem. Commun., 1998, 1643–1644.
56 R. M. Supkowski and W. D. Horrocks, Inorg. Chim. Acta, 2002,
340, 44–48.
57 F. Bellezza, A. Cipiciani, U. Costantino and M. E. Negozio,
Langmuir, 2002, 18, 8737–8742.
58 CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC
Press, Boca Raton, FL, 80th edn., 1999.
59 G. Alberti, L. Boccali, M. Casciola, L. Massinelli and E. Mon-
toneri, Solid State Ionics, 1996, 84, 97–104.
60 M. S. Wrighton, D. S. Ginley and D. L. Morse, J. Phys. Chem,
1974, 78, 2229–2233.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 5 0 6 – 1 5 1 3
1513