Novel Lanthanide Luminescent Materials Based on Complexes of 3-Hydroxypicolinic Acid and Silica Nanoparticles

Dynamic Article LinksC<Journal ofMaterials Chemistry

Cite this: J. Mater. Chem., 2011, 21, 15600

www.rsc.org/materials PAPER

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online / Journal Homepage / Table of Contents for this issue

Novel lanthanide luminescent materials based on multifunctional complexes of2-sulfanylpyridine-3-carboxylic acid and silica/titania hosts†

Lei Guo,a Lianshe Fu,b Rute A. S. Ferreira,b Luis D. Carlos,b Qiuping Lia and Bing Yan*a

Received 21st May 2011, Accepted 28th July 2011

DOI: 10.1039/c1jm12264a

Three different types of organic–inorganic hybrid materials formed by trivalent lanthanide (Ln3+ ¼Eu3+, Tb3+) complexes covalently grafted to silica-, titania-, or silica/titania-based hosts have been

prepared and fully characterized. Since the organic ligand 2-sulfanylpyridine-3-carboxylic acid (SPC),

a derivative of nicotinic acid, exhibits three potential binding sites (pyridine N, sulfhydryl S and

carboxylic O), the multifunctional precursor can be prepared through the reaction of the carboxylic

group with titanium alkoxide and the modification of the sulfhydryl group with silane crosslinking

reagents. Thus, the organic–inorganic hybrid materials covalently grafted with Eu3+ or Tb3+ complexes

are synthesized through coordination of the Ln3+ ions with the heterocyclic group in the

multifunctional precursor during the sol–gel process. The obtained hybrid materials were characterized

by X-ray diffraction (XRD), scanning electron microscopy (SEM), differential scanning calorimetry

(DSC), thermogravimetric analysis (TG), Fourier transform infrared (FTIR) spectroscopy, and

photoluminescence (PL). The detailed PL studies showed that, compared with the titania-based hybrid

materials (denoted as Ln–SPC–Ti), the silica- and silica/titania-based hybrid materials (denoted as Ln–

SPCSi and Ln–SPCSi–Ti, respectively) exhibited higher luminescence intensity and emission quantum

efficiency.

1. Introduction

Trivalent lanthanide (Ln3+) complexes are well-known molecular

luminescent materials, which are characterized by long-lived

excited-states and efficient narrow-width emission bands in the

visible and near infrared (NIR) regions.1 This is mainly because

the effective intramolecular energy transfer from the coordinated

ligands to the luminescent central Ln3+ ions, which in turn

undergoes the corresponding radiative emitting process (the so-

called ‘‘antenna effect’’).2 Therefore, they are expected to be

promising luminescent dopants for the preparation of organic–

inorganic hybrids with potential applications as phosphors, in

solid-state lighting, in integrated optics and optical telecommu-

nications, in solar cells, and in biomedicine. In recent years, there

has been a strong interest in Ln3+-containing organic–inorganic

hybrid materials.3 In these materials, the Ln3+ complexes are

entrapped in sol–gel-derived hosts, or alternatively, an

inorganic Ln3+ compound (like a polyoxometalate complex or an

aDepartment of Chemistry, Tongji University, Shanghai, 200092, P. R.China. E-mail: [email protected] of Physics, CICECO, University of Aveiro, 3810-193 Aveiro,Portugal

† Electronic supplementary information (ESI) available: Fig. S1: XRDpatterns of the (A) Eu–SPC–Ti, (B) Eu–SPCSi-1, and (C)Eu–SPCSi-1–Ti hybrid materials. Fig. S2: UV-visible diffuse reflectionabsorption spectra of Tb3+-containing hybrid materials. See DOI:10.1039/c1jm12264a

15600 | J. Mater. Chem., 2011, 21, 15600–15607

Ln3+-doped nanoparticle) is embedded in an organic polymer

matrix. In general, these hybrid materials have superior

mechanical properties and have better process abilities than the

pure Ln3+ complexes. Moreover, embedding an Ln3+ complex in

a hybrid matrix is also beneficial for its thermal stability and

luminescence output.4

Among the innumerous examples reported in the literature,

hybrid materials formed through the grafting of Ln3+ complexes

with b-diketones, aromatic carboxylic acids, and heterocyclic

ligands to the inorganic backbone (essentially a siloxane-based

skeleton) via a covalent bond have earned significant interest.5

The extension of the concept to other metal oxides or mixed-

metal oxides would then allow new interesting options for the

development of innovative Ln3+-containing organic–inorganic

hybrid materials. Titania is an essential functional material

because of its peculiar and fascinating physicochemical proper-

ties and a wide variety of potential use in diverse fields, including

solar-cells, energy conversion, environmental purification, and

photocatalysis.6 Thus, it would be highly attractive to incorpo-

rate Ln3+ complexes into titania- or silica/titania-based hosts and

to investigate their luminescence properties, comparing with

those of the analogous silica-based hybrids. Although lanthanide

luminescent complexes have been in situ synthesized in titania-

based host or were adsorbed on silica/titania-based host,7 only

weak interactions between inorganic and organic parts exist in

these hybrid materials and it is difficult to prevent clustering of

emitting centers and inhomogeneous dispersion of two phases.8

This journal is ª The Royal Society of Chemistry 2011

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

In general, the key to ‘‘design’’ the covalently bonded hybrid

materials is to tailor a functional bridge molecule (ligand) by the

alkylation reaction, which can have the double functions of

coordinating to Ln3+ ions and of allowing the sol–gel process to

build up the covalent Si–O network.9 However, Ln3+ complexes

cannot be linked to the titanium centres by a hydrocarbon spacer

like silicon due to the hydrolytic cleavage of Ti–C bonds.10

Therefore, a new strategy has to be developed for tethering Ln3+

complexes to titania materials. As we know, the reaction with

bidentate ligands is a common way for modifying the metal

alkoxides through sol–gel process, and carboxylates are mainly

used for this purpose, which can react with metal alkoxides to

form surface-functionalized metal oxo clusters.11

In the present work, the organic ligand 2-sulfanylpyridine-3-

carboxylic acid (SPC) was selected to construct the linkage

between inorganic matrix and Ln3+ ions. As we know, aromatic

carboxylic acids and heterocyclic ligands are important organic

ligands for lanthanide ions. SPC, a derivative of nicotinic acid,

exhibited three potential binding sites—pyridine N, thiol S, and

carboxylicO. It can reactwith titaniumalkoxide via the carboxylic

acid group, while the heterocyclic group can coordinate to Ln3+

ions, as well as sensitize their luminescence. Ln3+ complexes can

thus be anchored to the framework of organic–inorganic titania

materials. In addition, we can furthermodify the sulfhydryl group

of ligand SPCAwith the silane crosslinking reagents. On the basis

of the above work, we can further modify the sulfhydryl group of

ligand SPC with the two typical kinds of silane crosslinking

reagents (e.g. 3-aminopropyltrimethoxysilane, APTMS and

3-chloropropyltrimethoxysilane, CPTMS), which can introduce

the inorganic silica into hybrid materials with covalent bonding.

Then the Ln3+ luminescent organic–inorganic hybrid materials

with silica/titania-based host could be obtained, structurally

characterized and their luminescence properties studied in detail.

In addition, for comparison, the covalently bonded silica-based

hybrid materials were also synthesized.

2. Experimental section

2.1. Materials

Tetraisopropyl titanate (Ti(OCH(CH3)2)4), SPC, APTMS,

CPTMS and tetraethoxysilane (TEOS) were purchased from

Aldrich and used without further purification. Ln(NO3)3$6H2O

was obtained from their corresponding oxides in dilute nitric

acid.

2.2. Measurements

Fourier transform infrared (FTIR) spectra were recorded on

a Nicolet model 5SXC spectrometer in 4000–400 cm�1 region

using KBr disk. 29Si magic-angle spinning (MAS) NMR spectra

were obtained from a Bruker Avance 400 (9.4 T) spectrometer at

79.49 MHz. The X-ray diffraction (XRD) measurements were

carried out using powder samples in a BRUKER D8 diffrac-

tometer (40 mA/40 kV), using monochromated CuKa1 radiation

(l¼ 1.54�A) over the 2q range of 10� to 70�. Differential scanning

calorimetry (DSC) and thermogravimetric analysis (TG) were

performed on a NETZSCH STA 449C at a heating rate of 15 �Cmin�1 under a nitrogen atmosphere. Scanning electronic micro-

scope (SEM) images were obtained with a Philips XL-30. The

This journal is ª The Royal Society of Chemistry 2011

ultraviolet-visible diffuse reflection spectra of the powder

samples were recorded by a BWS003 spectrophotometer. The

photoluminescence spectra were recorded at room temperature

with a modular double grating excitation spectrofluorimeter with

a TRIAX 320 emission monochromator (Fluorolog-3�2-TRIAX, Horiba Scientific) with a reciprocal linear dispersion

density of 2.64 nm mm�1 coupled with a R928 Hamamatsu

photomultiplier, using the front face acquisition mode. The

emission slits were fixed at 0.3 mm, enabling a resolution of

0.7 nm in the emission spectra. The excitation source was

a 450 W Xe arc lamp. The emission spectra were corrected for

detection and optical spectral response of the spectrofluorimeter,

and the excitation spectra were corrected for the spectral distri-

bution of the lamp intensity using a photodiode reference

detector. The lifetime measurements were performed at room

temperature with the setup described for the luminescence

spectra using a pulsed Xe–Hg lamp (6 ms pulse at half width and

20–30 ms tail). The absolute emission quantum yields were

measured at room temperature using a quantum yield measure-

ment system C9920-02 from Hamamatsu with a 150 W Xenon

lamp coupled to a monochromator for wavelength discrimina-

tion, an integrating sphere as sample chamber and a multi

channel analyzer for signal detection. Three measurements were

made for each sample so that the average value is reported. The

method is accurate to within 10%.

2.3. Synthesis of hybrid materials Ln–SPC–Ti (Ln3+ ¼ Eu3+

and Tb3+)

A sample was prepared by adding 2 mmol (0.5685 g) of Ti(OCH

(CH3)2)4 to 50 mL of ethanol containing 2 mmol (0.3105 g) of

SPC under refluxing and stirring. The mixture was refluxed at

70 �C for 3 h. 0.5 mmol of Ln(NO3)3$6H2O (Ln3+ ¼ Eu3+ and

Tb3+) dissolved in ethanol was added dropwise into the clear

solution and 0.1 mL H2O was then added. The stirring was

continued for another 24 h to yield a precipitate, which was

recovered by centrifugation and dried for 24 h at 65 �C under

vacuum. The hybrid materials were denoted as Ln–SPC–Ti.

2.4. Synthesis of hybrid materials Ln–SPCSi (Ln3+ ¼ Eu3+ and

Tb3+)

The SPC ligand was modified by two different silane crosslinking

reagents, APTMS and CPTMS. The two organosilane precur-

sors were prepared by using a modified procedure based on the

methods described in ref. 12 and 13

Modified by 3-aminopropyltrimethoxysilane. 2 mmol (0.3105 g)

of SPC was first dissolved in dehydrate tetrahydrofuran (THF)

(20 mL) and then 2 mmol (0.3585 g) of APTMS was added

dropwise to the solution with stirring at 80 �C. The whole

mixture was refluxed at 80 �C for 24 h. The solution was

condensed to evaporate the solvent and then the residue was

dried on a vacuum line under argon atmosphere. A yellow oil

product denoted as SPCSi-1 was obtained.

Modified by 3-chloropropyltrimethoxysilane. 2 mmol (0.3105 g)

of SPC was first dissolved in THF (20 mL) and then 2 mmol

(0.3974 g) of CPTMS was added dropwise to the solution. An

J. Mater. Chem., 2011, 21, 15600–15607 | 15601

Scheme 1 Synthetic procedure and possible structure of hybrid mate-

rials Ln–SPCSi–Ti.

Fig. 1 FTIR spectra of the (A) Eu–SPC–Ti, (B) Eu–SPCSi-1, (C) Eu–

SPCSi-2, (D) Eu–SPCSi-1–Ti, and (E) Eu–SPCSi-2–Ti hybrid materials.

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

appropriate amount of K2CO3 was added as catalyst. The whole

mixture was refluxed at 80 �C for 3 h. After filtration, the solu-

tion was then condensed to evaporate the solvent. The residue

was dried on a vacuum line, and the resulting yellow oil was

denoted as SPCSi-2.

Synthesis of the hybrid materials Ln–SPCSi-1 and Ln–SPCSi-

2. The organosilane precursor, SPCSi-1 or SPCSi-2, was dis-

solved in N, N-dimethyl formamide solvent with stirring. Then,

a stoichiometric amount of the Ln(NO3)3$6H2O (Ln3+ ¼ Eu3+

and Tb3+) was added. After 3 h, TEOS and one drop of diluted

hydrochloric acid were added into the solution to promote the

hydrolysis and polycondensation reaction. The molar ratio of Ln

(NO3)3$6H2O : precursor : TEOS : H2O was 1 : 3 : 6 : 24. The

mixture was agitated magnetically to achieve a single phase

solution after two days, and then it was aged at 65 �C until the

onset of gelation in about one week. The gels were collected and

were ground as powders for the photophysical studies. The

resulting hybrid materials, denoted as Ln–SPCSi-1 and Ln–

SPCSi-2 (Ln3+ ¼ Eu3+ and Tb3+), were grounded as powders for

the photophysical studies.

2.5. Synthesis of hybrid materials Ln–SPCSi–Ti (Ln3+ ¼ Eu3+

and Tb3+)

The organosilane precursor, SPCSi-1 or SPCSi-2, was dissolved

in ethanol with stirring and an equimolar amount of Ti(OCH

(CH3)2)4 was added into the solution. The mixture was refluxed

at 70 �C for 3 h. Then, the Ln(NO3)3$6H2O (Ln3+ ¼ Eu3+ and

Tb3+) in ethanol was added dropwise into the clear solution and

0.1 mL H2O was added. The stirring was continued for another

24 h to yield a precipitate, which was recovered by centrifugation

and dried for 24 h at 65 �C under vacuum. The hybrid materials

were denoted as Ln–SPCSi-1–Ti and Ln–SPCSi-2–Ti (Ln3+ ¼Eu3+ and Tb3+). The general synthetic procedure and the possible

structures of the hybrids are displayed in Scheme 1.

3. Results and discussion

3.1. Structural characterization

Fig. 1 shows the FTIR spectra of obtained Eu3+-containing

hybrid materials in the 4000–400 cm�1 range. In the spectrum of

Eu–SPC–Ti (Fig. 1A), the typical antisymmetric and symmetric

stretching vibrations of carboxylate appear at about 1572 and

1491 cm�1. No absorption bands from the corresponding acid

can be observed, indicating that SPC was completely reacted

with Ti(OCH(CH3)2)4 and no dissolution occurred during the

sol–gel procedure. In addition, the quite broad band locating at

the range of 500 to 650 cm�1 is ascribed to the characteristic

absorption of Ti–O stretching and Ti–O–Ti bridging stretching

modes,14,15 which indicates the presence of TiO2. From the

spectra of B and C, the modifying reactions of SPC are evidenced

by the vanishing of the n(S–H) vibration at 2410 cm�1 and an

increase of n(C–S–C) vibration at about 700 cm�1. This is in good

agreement with the results reported previously.16 The bands at

1031 cm�1 and 1215 cm�1 are attributed to the stretching vibra-

tion n(Si–O) and the stretching vibration n(Si–C). The broad

absorption band n(Si–O–Si) at 1200–1100 cm�1 indicates the

formation of the Si–O–Si framework.

15602 | J. Mater. Chem., 2011, 21, 15600–15607

Similar to the spectra of Fig. 1B and C, the FTIR spectra for

all silica–titania based hybrid materials (Fig. 1D and E) clearly

show the bands at 1058, 1122, 774 and 914 cm�1. The former

three bands are ascribed to the characteristic Si–O–Si vibrations

and Si–O–H stretching vibration, while the weak peak at

914 cm�1 was assigned to the asymmetric stretching vibration of

Ti–O–Si.17 Meanwhile, the peaks at around 2410 cm�1 origi-

nating from the absorption of S–H groups disappeared, which

can prove the completion of the alkylation reactions. Besides, the

n(O–H) vibration at around 3380 cm�1 can also be observed,

indicating the presence of unsaturated Si–OH groups and/or

H2O molecules.

The 29Si MAS NMR spectra of Eu–SPCSi-1 and Eu–SPCSi-2

show the various organosiloxane Tm (Tm ¼ RSi(OSi)m(OH)3�m,

m ¼ 1–3) and siloxane Qn (Qn ¼ Si(OSi)n(OH)4�n, n ¼ 2–4)

species (Fig. 2). The organosiloxane Tm species result from the

hydrolysis and condensation of SPCSi-1 and SPCSi-2, whereas

the siloxane Qm species originate from the hydrolysis and

condensation of TEOS. The T3 and Q4 environments are clearly

dominant, suggesting that there are two main types of local

structures, RSi(OSi)3 and Si(OSi)4.18 On the other hand, the 29Si

This journal is ª The Royal Society of Chemistry 2011

Fig. 2 29Si MAS NMR spectra of (A) Eu–SPCSi-1, (B) Eu–SPCSi-2, (C)

Eu–SPCSi-1–Ti and (D) Eu–SPCSi-2–Ti.

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

MAS NMR spectra of Eu–SPCSi-1–Ti and Eu–SPCSi-2–Ti

exhibit characteristic signals of T1, T2 and T3 units, showing the

presence of two (for Eu–SPCSi-1–Ti) or three (for Eu–SPCSi-2–

Ti) main types of local structures, (SiO)Si(CH2)3(OH)2, (SiO)2Si

(CH2)3OH and (SiO)3Si(CH2)3. The absence of signals due to Qm

species between �90 and �120 ppm indicates that no cleavage of

the Si–C bonds occurred and that all the silicon atoms are

covalently connected to carbon atoms. The peak positions for

Eu–SPCSi-1, Eu–SPCSi-2, Eu–SPCSi-1–Ti and Eu–SPCSi-2–Ti,

and the relative populations of the various silicon sites (T and Q)

were quantitatively estimated and are listed in Table 1.

The evaluations of the condensation degree for T units (CT),

the condensation degree for Q units (CQ) and the global

condensation degree (C) of the materials have been performed by

applying the method outlined in ref. 19. For Eu–SPCSi-1, the

condensation degree of the T units (96.2%) is larger than that of

Q units (92.0%), showing that the T units are more effective than

that of Q units for the condensation. This behaviour can be

explained by the higher reactivity of organosilane precursors

than that of TEOS.20 For Eu–SPCSi-1–Ti, due to the absence of

Q units, the value of CT (95.9%) is the same as its global

condensation degree (C). The C for Eu–SPCSi-1–Ti (95.9%) is

larger than that of Eu–SPCSi-1 (93.6%), which further elucidates

the higher reactivity of organosilane precursors than that of

TEOS. However, the situation of the Eu–SPCSi-2 and Eu–

SPCSi-2–Ti is opposite. This observation can be ascribed to the

different structure of the organosilane precursor.

The powder XRD patterns of the Eu3+-containing hybrid

materials with different inorganic matrices are shown in Fig. S1

(see the ESI†). The XRD patterns of the samples are similar,

which showed only one broad band from 20� to 30� assigned to

an amorphous substance. From Fig. S1A† (Eu–SPC–Ti), no

characteristic peaks of anatase- or rutile-phase titania were

Table 1 29Si MAS NMR spectral chemical shifts (ppm), population of the d

Samples T1 (%) T2 (%) T3 (%) CT (%)

Eu–SPCSi-1 �47.1 (1.3) �57.4 (8.9) �67.5 (89.8) 96.2Eu–SPCSi-2 �49.3 (8.1) �57.8 (11.4) �66.5 (80.5) 90.8Eu–SPCSi-1–Ti �59.2 (12.2) �68.4 (87.8) 95.9Eu–SPCSi-2–Ti �54.3 (10.2) �58.9 (17.9) �67.6 (71.9) 87.2

This journal is ª The Royal Society of Chemistry 2011

observed, which is in good agreement with the results reported

previously.26,30 The absence of any crystalline regions in these

samples correlates well with the presence of organic chains in the

host inorganic framework. These phenomena lead us to conclude

that neither free europium/terbium nitrate salt, pure crystalline

organic ligand SPC, nor crystalline Ln3+ complexes occur

throughout the range of those hybrid materials.

The thermal stabilities of obtained hybrid materials were

demonstrated by TG and DSC measurements. Fig. 3 shows the

TG curves, derivative weight loss (DTG) curve and DSC curve

for Eu–SPC–Ti and Eu–SPCSi-1–Ti. For Eu–SPC–Ti, there are

two main weight loss regions: the initial stage ranging from 55 to

110 �C is attributed to the desorption of physically absorbed

water and the residuary solvent, while the second stage of the

peak at about 266 �C is due to the thermo-decomposition of the

organic Ln3+ complex. Correspondingly, two endothermic peaks

appear in the DSC curve around 94 and 264 �C, which corre-

spond to the two main weight loss of the DTG curve. Compared

with the TG curve of the hybrid material Eu–SPC–Ti, the TG

curve of Eu–SPCSi-1–Ti is similar, which also exhibits two main

weight loss stages. The slight difference in the weight loss

between them may be due to the modification of the silane

crosslinking reagents with the sulfhydryl group of organic ligand

SPC. On comparing these results with those achieved with the

hybrid materials containing silica as inorganic matrix,21 the

thermal stabilities of the obtained Ln3+-based hybrid materials

with different inorganic matrices are similar or even better.

Therefore, the choice of the inorganic matrix other than silica

becomes more extensive and thus more multifunctional lumi-

nescent hybrid materials could be obtained.

Surface morphology of obtained hybrid materials was studied

using SEM. Fig. 4 shows the SEM images of the hybrid materials

Eu–SPC–Ti (A), Eu–SPCSi-1 (B), Eu–SPCSi-1–Ti (C), and Eu–

SPCSi-2–Ti (D). From these images for the hybrid materials, the

shape of the particles shows much difference obviously. This

means that the different hybrid materials with different inorganic

matrices have an apparent influence on the morphology. The

hybrid material Eu–SPC–Ti exhibits many acicular clusters that

are composed of plentiful needlelike cuboid configurations with

widths of about 2–3 mm. The morphology of the hybrid material

with silica matrix (Fig. 4B) seems to be largely different from that

of Eu–SPC–Ti, forming the bulk materials with the columned

configurations. This can be ascribed to the different way of

coordination with the Ln3+ ions. In this kind of hybrid material,

the coordination behavior depends on the carboxylate group. As

we know, Ln3+ complexes of carboxylate derivatives readily form

the polymeric structure, and the carboxylate groups possess the

bridging coordination ability. This tendency will compete with

the construction of a polymeric network structure of Si–O–Si in

the hydrolysis and copolycondensation processes of silica. Thus,

ifferent T and Q species (%), CT, CQ and C (%)

Q2 (%) Q3 (%) Q4 (%) CQ (%) C (%)

�92.0 (3.4) �101.0 (25.4) �110.9 (71.2) 92.0 93.6�92.6 (1.6) �101.4 (24.6) �110.7 (73.8) 93.1 92.1

95.987.2

J. Mater. Chem., 2011, 21, 15600–15607 | 15603

Table 2 The I02/I01 intensity ratio and the energy (E00) and fwhm (fwhm00) values of the5D0 /

7F0 line of Eu3+-containing hybrids

Hybrids Eu–SPC–Ti Eu–SPCSi-1 Eu–SPCSi-2 Eu–SPCSi-1–Ti Eu–SPCSi-2–Ti

I02/I01 6.5 4.5 4.4 5.4 4.7E00/cm

�1 17 293.5 � 0.5 17 299.6 � 0.1 17 302.0 � 0.3 17 292.3 � 0.3 17 300.8 � 0.1fwhm00/cm

�1 26.5 � 0.3 25.1 � 0.2 27.0 � 0.6 36.3 � 0.6 30.4 � 0.2

Fig. 3 TG (TG1), DSC and DTG (----) curves of Eu–SPC–Ti and TG

(TG2) curve of Eu–SPCSi-1–Ti.

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

these hybrid materials retain the intensive tendency of growing

into a polymeric microstructure and reserve the coordinated

positions in the corresponding bulk materials.

Compared to the above-mentioned hybrid materials, the

morphologies of hybrid materials Eu–SPCSi-1–Ti and Eu-

SPCSi-2–Ti are aggregates of small particles and microspheres

with dimensions up to 0.2–0.5 mm, which can be seen from

Fig. 4C and D. It is speculated that the addition of the silica

inorganic matrix through the modified ligand SPC changes the

morphology of the titania based hybrid materials. From the

above discussion, we can see that the self-assembly process can be

influenced not only by the coordination behavior of the Ln3+

complexes but also by the different inorganic parts in the hybrid

materials. Here, it should be clarified that the hybrid materials

mentioned in our manuscript belong to the non-crystalline state

materials with amorphous nature. Therefore, there exist no exact

long-range ordered structure and microstructure like crystals or

crystalline materials. The SEM characterization only shows the

basic morphology of the materials but cannot give exact infor-

mation of the unit in the hybrid systems like crystalline materials.

Fig. 4 SEM images of (A) Eu–SPC–Ti, (B) Eu–SPCSi-1, (C) Eu–SPCSi-

1–Ti and (D) Eu–SPCSi-2–Ti hybrid materials.

15604 | J. Mater. Chem., 2011, 21, 15600–15607

We have tried to provide more information on the surface

morphology of them using TEM of SEM images with higher

resolution. But unfortunately, it cannot provide satisfactory and

clear results (see Fig. S2 and S4 in the ESI†). Besides, some

crystal-like microstructure particles may be derived from TiO2

from the sol–gel products of Ti(OCH(CH3)2)4 precursors and

excess Ln(NO3)3 salts, which are not the nanocrystals from the

covalently bonded hybrid system.

The UV-vis diffuse reflection absorption spectra were

measured on the powder materials for all the obtained hybrid

materials. The corresponding absorption spectra of Tb3+-con-

taining hybrid materials are shown in Fig. S2 (see the ESI†). All

of the spectra exhibit a broad absorption band in the range of

300–450 nm. It can be primarily predicted that in the hybrid

materials the central Ln3+

ions can be efficiently sensitized by the

ligand through the so-called ‘‘antenna effect’’. For the hybrid

materials with different inorganic matrices, an obvious red shift

and broadening of the absorption band are observed; the degree

of red shift and broadening increases in the order of silica

(329 nm) < silica/titania (389 nm) < titania (408 nm) based host

hybrid materials. This may be due to the different structures and

different ways of combination between the inorganic compo-

nents and Ln3+ complexes in the Ln3+ hybrid materials. Thus, the

different energy absorption in ultraviolet-visible, different

intramolecular energy transfer coefficients and different fluo-

rescence emission intensities can be observed. This will be further

discussed later in detail combined with the luminescence prop-

erties of these hybrid materials.

3.2. Photoluminescence

Fig. 5 shows the room temperature excitation spectra of Eu–

SPC–Ti, Eu–SPCSi-2, Eu–SPCSi-2–Ti, Tb–SPC–Ti, Tb–SPCSi-

1, and Tb–SPCSi-1–Ti, monitored within the 5D0 / 7F2

(614 nm) and 5D4 /7F5 (544 nm) transitions, respectively, for

Eu3+- and Tb3+-containing hybrids. For Eu3+-containing hybrid

materials, all the spectra are similar, showing a broad excitation

band between 300 and 350 nm, which partially overlap with the

UV-vis diffuse absorption spectra and can be assigned to the p–

p* states of the organic ligand.22,23 A peak at 393 nm is observed

due to the Eu3+ 7F0/5L6 f–f transition. This transition is weaker

than the absorption of the organic ligand, which proves that

luminescence sensitization via ligand excitation is much

more efficient than the direct excitation of the Eu3+ ion. For

Tb3+-containing hybrid materials, a similar broad band is also

detected and no apparent f–f transitions could be observed in the

spectra. For Tb–SPCSi-1 and Tb–SPCSi-1–Ti hybrid materials,

the broad band is blue-shifted compared with that of the Tb–

SPC–Ti (from 335 nm to 320 and 323 nm). Such a difference may

be attributed to the modification between the silane crosslinking

reagents and the sulfhydryl group of ligand SPC, and the change

This journal is ª The Royal Society of Chemistry 2011

Fig. 5 Excitation spectra of (A) Eu–SPC–Ti, (B) Eu–SPCSi-2, (C) Eu–

SPCSi-2–Ti, (D) Tb–SPC–Ti, (E) Tb–SPCSi-1, and (F) Tb–SPCSi-1–Ti

monitored at 614 nm for Eu and 544 nm for Tb, respectively. Fig. 7 Room temperature emission spectra of (A) Tb–SPC–Ti excited at

335 nm, (B) Tb–SPCSi-1 (solid line) and Tb–SPCSi-1–Ti (----) excited at

320 nm and 325 nm, (C) Tb–SPCSi-2 (solid line) and Tb–SPCSi-2–Ti (----)

excited at 320 nm and 325 nm. All spectra are normalized to a constant

intensity at the maximum.

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

in the polarity of the environment surrounding the Ln3+ complex

in the inorganic matrix.

Fig. 6 presents the normalized emission spectra at room

temperature for the obtained Eu3+-containing hybrid materials.

All the spectra exhibit the characteristic intra-4f 6 sharp lines

ascribed to the 5D0 /7F0–4 transitions. From the spectra of the

hybrids with the silica–titania based host, there is no broad band

emission observed in the blue and green spectral regions.

However, a broad emission is observed for the titania based

hybrid material (Fig. 6A). It is assumed that the difference in the

inorganic parts of the hybrid materials and the environment of

Ln3+ complexes can affect the energy transfer. Fig. 7 illustrates

the typical photoluminescence spectra of the Tb3+-containing

hybrid materials. The emission bands at 488, 543, 583, and

620 nm are recorded. These bands are attributed to the 5D4 /7F6,

5D4 /7F5,

5D4 /7F4, and

5D4 /7F3 transitions of Tb

3+

ions, respectively (Table 2).

Furthermore, the lifetime values of 5D0 (Eu3+) and 5D4 (Tb

3+)

excited states are estimated on the basis of the emission decay

curves monitored within the more intense Eu3+ (5D0 /7F2) and

Tb3+ (5D4 /7F5) transitions, respectively. All the curves reveal

Fig. 6 Room temperature emission spectra of (A) Eu–SPC–Ti excited at

325 nm, (B) Eu–SPCSi-1 (solid line) and Eu–SPCSi-1–Ti (----) excited at

328 nm, (C) Eu–SPCSi-2 (solid line) and Eu–SPCSi-2–Ti (----) excited

at 330 nm.

This journal is ª The Royal Society of Chemistry 2011

a single exponential behavior, which corroborates that all the

Ln3+ ions occupy the same average local environment within each

hybrid material. For the Eu3+-containing and Tb3+-containing

hybrid materials, the obtained lifetime values are given in Table 3

and Table 4, respectively. On the basis of the emission spectra

and 5D0 lifetimes, the emission quantum efficiency (h) of the 5D0

emitting level can be determined.24–27 Assuming that only non-

radiative and radiative processes are essentially involved in the

depopulation of the 5D0 state, h can be expressed as:

h ¼ Ar/(Ar + Anr) (1)

The radiative processes (Ar) are calculated from the relative

intensities of the 5D0 / 7F0–4 transitions (the 5D0 / 7F5,6

branching ratios are neglected because of their poor relative

intensity with respect to that of the remaining 5D0 /7F0–4 lines).

The 5D0 / 7F1 transition does not depend on the local ligand

field and thus may be used as a reference for the whole spectrum.

An effective refractive index of 1.5 was used leading to A01 z50 s�1,28 where A01 stands for Einstein’s coefficient of sponta-

neous emission between the 5D0 and the 7F1 Stark levels. Life-

time, radiative (Ar), and nonradiative (Anr) transition rates are

related through the following equation:

sexp ¼ (Ar + Anr)�1 (2)

On the basis of the above discussion, the parameters Ar, Anr

and the quantum efficiency values, h, for the 5D0 Eu3+ ion excited

state in the five hybrids can be obtained, as shown in Table 3.

Comparing the h values of the Eu–SPC–Ti hybrid materials with

those previously reported for the organic–inorganic titania

hybrid materials, the values in Table 3 are smaller. In spite of

higher Ar values, Eu–SPC–Ti displays much higher Anr values

relative to those known for the titania materials, namely, Ar and

Anr values of 0.336 and 2.68 for Ti–nit–Eu material.26 Comparing

the h values of the obtained hybrids with different inorganic

matrix, an increase in the h values is observed. For the case of

silica based hybrid materials Eu–SPCSi-1 and Eu–SPCSi-2, such

J. Mater. Chem., 2011, 21, 15600–15607 | 15605

Table 3 Lifetime (s), radiative (Ar) and nonradiative (Anr) transition probabilities, quantum efficiency of the 5D0 level (h), quantum yield (f), U2,4

intensity parameters (� 10�20 cm2) and number of coordinated water molecules (nw) for Eu3+-containing hybrid materials

Hybrids Eu–SPC–Ti Eu–SPCSi-1 Eu–SPCSi-2 Eu–SPCSi-1–Ti Eu–SPCSi-2–Ti

s/ms 0.202 � 0.003 0.489 � 0.003 0.367 � 0.007 0.283 � 0.002 0.358 � 0.003Ar/ms�1 0.483 0.329 0.304 0.350 0.381Anr/ms�1 4.467 1.716 2.421 3.184 2.412h (%) 9.8 16.1 11.2 9.9 13.7f (%) <1 2 1 <1 <1nw 4.6 1.6 2.3 3.2 2.3U2 9.8 6.7 6.6 7.1 8.0U4 5.0 2.5 1.1 2.9 2.4

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

an increase is due to both an increase in the lifetime value and

a decrease in the Anr value. For Eu–SPCSi-1–Ti and Eu–SPCSi-

2–Ti, after incorporation of the titania inorganic host, although

there is an increase in the radiative transition probability Ar,

there is a decrease in the lifetime value. In addition, the hybrid

materials modified with the different silane crosslinking reagents

have the different quantum efficiency values in different inor-

ganic hosts. In silica host hybrid materials, the quantum effi-

ciency of Eu–SPCSi-1 is higher than that of Eu–SPCSi-2.

However, the opposite result can be observed in the silica/titania-

based hybrid system. On the basis of these results, we may

presume that the modification with the titania matrix hybrids is

very helpful for improving the quantum efficiency. At the same

time, the different composition of the hybrid SPC materials may

have an influence on the luminescent lifetimes and quantum

efficiency. Moreover, the variations in h and Anr values may be

rationalized in terms of the number of water molecules coordi-

nated to the Eu3+ ions (nw) based on the empirical formula

reported by Supkowski and Horrocks:29

nw ¼ 1.11�(s�1 � Ar � 0.31) (3)

The results for obtained Eu3+-containing hybrid materials are

shown in Table 3. On the basis of the results, the different nwvalues of hybrid materials are possibly due to the fact that the

ways of the coordination between the Eu3+ ion and ligand SPC

are different. In addition, the decrease in the nw value is in good

agreement with the increase in the value of quantum efficiency.

The coordination water molecules produce the severe vibration

of the hydroxyl group, resulting in the large nonradiative tran-

sition and decreasing the luminescent efficiency.

In order to further quantify the different photoluminescence

features of the different hybrid materials, the absolute emission

quantum yields (f) were measured for all the materials (Tables 3

and 4). The maximum quantum yield values are found for the

Eu– and Tb–SPCSi-1 hybrids, whereas for some of the remaining

hybrids the values are lower than the detection limits of our

equipment (<0.01). According to ref. 1, the overall sensitization

efficiency, hsens, of the Eu3+-containing hybrids is defined as the

ratio between the quantum efficiency of the 5D0 level (h) and the

Table 4 Lifetime (s) and quantum yield (4) of Tb3+-containing hybrid mate

Hybrid materials Tb–SPC–Ti Tb–SPCSi-1

s/ms 0.364 � 0.003 0.516 � 0.0064 (%) <1 1

15606 | J. Mater. Chem., 2011, 21, 15600–15607

emission quantum yield f. A maximum value of hsens ¼ 0.12 is

estimated for Eu–SPCSi-1, indicating that the relatively low

quantum yield values of these hybrid materials could be

explained by an inefficient ligand sensitization, besides the

presence of a large number of coordinated water molecules.

In addition, we synthesized the pure Ln3+ (Ln ¼ Eu, Tb)

complexes and compared their luminescence property with

obtained hybrid materials. The results show that the obtained

hybrids possess longer luminescence lifetimes and higher

quantum efficiencies than that of a pure complex except for the

Eu–SPC–Ti hybrids which are slightly less than the pure

complex. But it is worthwhile pointing out that the effective

content of lanthanide (Eu, Tb) species in the hybrid materials is

much lower than the pure lanthanide complex, so in fact, the

luminescence of the hybrids are enhanced compared to the pure

complex (see Tables S1 and S2†).

The experimental intensity parameters (U2 and U4) were

determined from the emission spectra of Eu3+-containing hybrids

based on the 5D0 /7F2 and

5D0 /7F4 transitions and the 5D0

/ 7F1 magnetic dipole transition as the reference, and they are

estimated according to the following equation,23,24,27

Ul ¼ [3h/64p4 e2g3][9/n(n2 + 2)2][1/|h5D0||U(l)||7FJi|2] A0J (4)

where h is Planck’s constant, e is the electronic charge, g is the

angular frequency of the transition, the refraction index n ¼1.5, and h5D0||U

(l)||7FJih5D0||U(l)||7FJi values are the square

reduced matrix elements whose values are 0.0032 and 0.0023

for J ¼ 2 and 4, respectively.30,31 The U6 parameter was not

determined since the 5D0 / 7F6 transition could not be

experimentally detected.

The U2, U4 intensity parameters for the five hybrid materials

are shown in Table 3. A point to be noted in these results is the

relatively high value of the U2 intensity parameter for Eu–SPC–

Ti. This might be interpreted as a consequence of the hypersen-

sitive behavior of the 5D0 /7F2 transition, suggesting that the

dynamic coupling mechanism is quite operative and that the

chemical environment is highly polarizable. In addition,

the similar value of theU2 intensity parameter can be observed in

the hybrid materials with the same inorganic matrix.

rials

Tb–SPCSi-2 Tb–SPCSi-1–Ti Tb–SPCSi-2–Ti

0.715 � 0.008 0.417 � 0.008 0.447 � 0.0062 <1 <1

This journal is ª The Royal Society of Chemistry 2011

Dow

nloa

ded

by U

nive

rsid

ade

de A

veir

o (U

Ave

iro)

on

31 O

ctob

er 2

012

Publ

ishe

d on

26

Aug

ust 2

011

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C1J

M12

264A

View Online

4. Conclusions

In this work we have synthesized a series of new Ln3+ organic–

inorganic hybrid materials with different inorganic hosts. The

organic ligand SPC exhibited three potential binding sites—

pyridine N, sulfhydryl S and carboxylic O to construct the hybrid

materials. The structural characterizations, physical properties

and the photoluminescence properties were studied in detail. The

differences in the profiles of emission features, in lifetime values

and in quantum efficiency among all the synthesized materials

confirm that the different inorganic matrices of the hybrid

materials have an influence on the photoluminescence properties.

Moreover, compared with titania matrix hybrid materials, the

introduction of silica inorganic matrix through the modification

of the organic ligand results in a much higher quantum efficiency.

Therefore, this facile strategy to tether Ln3+ complexes to

organic–inorganic silica/titania hybrid materials can be conve-

niently applied to other hybrid material systems and the desired

properties can be tailored by an appropriate choice of the

precursors. In this way, numerous organic ligands are expected

to be introduced into the hybrid materials and more multifunc-

tional luminescent materials could be obtained.

Acknowledgements

This work was supported by the National Natural Science

Foundation of China (20971100), Program for New Century

Excellent Talents in University (NCET 2008-08-0398) and FCT

Project of Portugal (PTDC/CTM/108975/2008).

References

1 J.-C. G. B€unzli, Chem. Rev., 2010, 110, 2729; A. P. Bassett,S. W. Magennis, P. B. Glover, D. J. Lewis, N. Spencer, S. Parsons,R. M. Williams, L. De Cola and Z. Pikramenou, J. Am. Chem.Soc., 2004, 126, 9413; S. M. Bruno, R. A. S. Ferreira,F. A. Almeida Paz, L. D. Carlos, M. Pillinger, P. Ribeiro-Claroand I. S. Goncalves, Inorg. Chem., 2009, 48, 4882.

2 S. Sato and M. Wada, Bull. Chem. Soc. Jpn., 1970, 43, 1955;J. M. Lehn, Angew. Chem., Int. Ed. Engl., 1990, 29, 1304.

3 C. Sanchez, G. J. D. A. Sloer-Illia, F. Ribot, T. Lalot, C. R. Mayerand V. Cabuil, Chem. Mater., 2001, 13, 3061; P. Escribano,B. Juli�an-L�opez, J. Planelles-Arag�o, E. Cordoncillo, B. Viana andC. Sanchez, J. Mater. Chem., 2008, 18, 23; L. D. Carlos,R. A. S. Ferreira, V. D. Bermudez and S. J. L. Ribeiro, Adv. Funct.Mater., 2001, 11, 111; L. D. Carlos, R. A. S. Ferreira,R. N. Pereira, M. Assuncao and V. D. Bermudez, J. Phys. Chem.B, 2004, 108, 14924; L. S. Fu, R. A. S. Ferreira, S. S. Nobre,L. D. Carlos and J. Rocha, J. Lumin., 2007, 122, 265;K. Binnemans, Chem. Rev., 2009, 109, 4283; L. D. Carlos,R. A. S. Ferreira, V. D. Bermudez and S. J. L. Ribeiro, Adv.Mater., 2009, 21, 509; B. Yan and Q. M. Wang, Cryst. GrowthDes., 2008, 8, 1484.

4 M. Fernandes, M. C. Goncalves, V. D. Bermudeza, R. A. S. Ferreira,L. D. Carlos, A. Charas and J. Morgado, J. Alloys Compd., 2008, 451,201.

5 J. L. Liu and B. Yan, J. Phys. Chem. C, 2008, 112, 14168; X. M. Guo,H. D. Guo, L. S. Fu, L. D. Carlos, R. A. S. Ferreira, L. N. Sun,R. P. Deng and H. J. Zhang, J. Phys. Chem. C, 2009, 113, 12538;L. Guo and B. Yan, Eur. J. Inorg. Chem., 2010, 1267.

6 C. Sanchez, B. Julian, P. Belleville and M. Popall, J. Mater. Chem.,2005, 15, 3559; P. Wang, C. Klein, R. Humphry, S. Zakeeruddinand M. Gr€atzel, J. Am. Chem. Soc., 2005, 127, 808.

7 Y. G. Wang, L. Wang, H. R. Li, P. Liu, D. S. Qin, B. Y. Liu,W. J. Zhang, R. P. Deng and H. J. Zhang, J. Solid State Chem.,2008, 181, 562; R. F. de Farias, S. Alves, M. F. Belian and G. F. deS�a, J. Colloid Interface Sci., 2001, 243, 523.

This journal is ª The Royal Society of Chemistry 2011

8 C. Sanchez and F. Ribot, New J. Chem., 1994, 18, 1007.9 C. Y. Peng, H. J. Zhang, J. B. Yu, Q. G. Meng, L. S. Fu, H. R. Li,L. N. Sun and X. M. Guo, J. Phys. Chem. B, 2005, 109, 15278;X. F. Qiao and B. Yan, J. Phys. Chem. B, 2008, 112, 14742;J. L. Liu and B. Yan, J. Phys. Chem. B, 2008, 112, 10898;N. N. Lin, H. R. Li, Y. G. Wang, Y. Feng, D. S. Qin, Q. Y. Ganand S. D. Chen, Eur. J. Inorg. Chem., 2008, 4781; H. F. Lu, B. Yanand J. L. Liu, Inorg. Chem., 2009, 48, 3966.

10 P. Liu, H. R. Li, Y. G. Wang, B. Y. Liu, W. J. Zhang, Y. J. Wang,W. D. Yan, H. J. Zhang and U. Schubert, J. Mater. Chem., 2008,18, 735.

11 K. L. Frindell, M. H. Bartl, A. Popitsch and G. D. Stucky, Angew.Chem., Int. Ed., 2002, 41, 960; U. Schubert, J. Sol-Gel Sci. Technol.,2003, 26, 47; M. Puchberger, W. Rupp, U. Bauer and U. Schubert,New J. Chem., 2004, 28, 1289; H. R. Li, P. Liu, Y. G. Wang,L. Zhang, J. B. Yu, H. J. Zhang, B. Y. Liu and U. Schubert, J.Phys. Chem. C, 2009, 113, 3945.

12 B. Yan and H. F. Lu, Inorg. Chem., 2008, 47, 5601.13 B. Yan, K. Qian and H. F. Lu, J. Organomet. Chem., 2009, 694,

3160.14 M. Schraml-Marth, K. L. Walther, A. Wokaum, B. E. Handy and

A. Baiker, J. Non-Cryst. Solids, 1992, 143, 93.15 L. T�ellez, J. Rubio, F. Rubio, E. Morales and J. L. Oteo, Spectrosc.

Lett., 2004, 37, 11.16 S. P. Chen, Q. Yang and S. Gao, J. Therm. Anal. Calorim., 2009, 95,

685.17 D. C. M. Dutoit, M. Schneider and A. Baiker, J. Catal., 1995, 153,

165; R. Ryoo, C. H. Ko and R. F. Howe, Chem. Mater., 1997, 9,1607; N. Yamada, I. Yoshinaga and S. Katayama, J. Sol-Gel Sci.Technol., 2000, 17, 123.

18 M. C. Goncalves, V. D. Bermudez, R. A. S. Ferreira, L. D. Carlos,D. Ostrovskii and J. Rocha, Chem. Mater., 2004, 16, 2530;L. S. Fu, R. A. S. Ferreira, N. J. O. Silva, L. D. Carlos,V. D. Bermudez and J. Rocha, Chem. Mater., 2004, 16, 1507;S. S. Nobre, P. P. Lima, L. Mafra, R. A. S. Ferreira, R. O. Freire,L. S. Fu, U. Pischel, V. D. Bermudez, O. L. Malta andL. D. Carlos, J. Phys. Chem. C, 2007, 111, 3275.

19 S. Ungureanu, M. Birot, G. Laurent, H. Deleuze, O. Babot, B. Juli�an-L�opez, M. F. Achard, M. Ionel Popa, C. Sanchez and R. Backov,Chem. Mater., 2007, 19, 5786.

20 A. Van Blaaderen and A. P. M. Kentgens, J. Non-Cryst. Solids, 1992,149, 161.

21 B. Yan and X. F. Qiao, J. Phys. Chem. B, 2007, 111, 12362; K. Sheng,B. Yan, H. F. Lu and L. Guo, Eur. J. Inorg. Chem., 2010,3498.

22 H. R. Li, J. Lin, H. J. Zhang, L. S. Fu, Q. G. Meng and S. B. Wang,Chem. Mater., 2002, 14, 3651.

23 P. C. R. Soares-Santos, H. I. S. Nogueira, V. F�elix, M. G. B. Drew,R. A. S. Ferreira, L. D. Carlos and T. Trindade, Chem. Mater.,2003, 15, 100.

24 L. D. Carlos, R. A. S. Ferreira, V. D. Bermudez, B. Juli�an-L�opez andP. Escribano, Chem. Soc. Rev., 2011, 40, 536.

25 O. L. Malta, H. F. Brito, J. F. S. Menezes, F. R. G. E. Silva, S. Alves,F. S. Farias and A. V. M. deAndrade, J. Lumin., 1997, 75, 255.

26 O. L. Malta, H. J. Batista and L. D. Carlos, Chem. Phys., 2002, 282,21; L. D. Carlos, O. L. Malta and R. Q. Albuquerque, Chem. Phys.Lett., 2005, 416, 238; S. S. Nobre, X. Catto€en, R. A. S. Ferreira,C. Carcel, V. D. Bermudez, M. W. C. Man and L. D. Carlos,Chem. Mater., 2010, 22, 3599.

27 L. D. Carlos, Y. Messaddeq, H. F. Brito, R. A. S. Ferreira,V. D. Bermudez and S. J. L. Ribeiro, Adv. Mater., 2000, 12, 594;M. Fernandes, V. D. Bermudez, R. A. S. Ferreira, L. D. Carlos,A. Charas, J. Morgado, M. M. Silva and M. J. Smith, Chem.Mater., 2007, 19, 3892; S. S. Nobre, C. D. S. Brites,R. A. S. Ferreira, V. D. Bermudez, C. Carcel, J. J. E. Moreau,J. Rocha, M. Wong Chi Man and L. D. Carlos, J. Mater. Chem.,2008, 18, 4172.

28 J. C. Boyer, F. Vetrone, J. A. Capobianco, A. Speghini andM. Bettinelli, J. Phys. Chem. B, 2004, 108, 20137.

29 R. M. Supkowski and W. D. Horrocks, Inorg. Chim. Acta, 2002, 340,44.

30 W. D. Horrocks and D. R. Sudnick, J. Am. Chem. Soc., 1979, 101,334.

31 K. Binnemans, K. Van Herck and C. G€orller-Walrand, Chem. Phys.Lett., 1997, 266, 297.

J. Mater. Chem., 2011, 21, 15600–15607 | 15607